4. Introduction to Plato's Phaedo; Arguments for the existence of the soul, Part II

We've been discussing the question: what arguments might be offered for the existence of a soul? The family of arguments we are considering initially are known as inferences to the best explanation. The idea is that there is something about us that requires explanation. We cannot explain it purely in physical terms, so we need to appeal to, or posit the existence of, a soul. I will return to that argument shortly, but for now, let me set that aside and say something about Plato.

Starting next week, we're going to be looking at Plato's dialogue, the *Phaedo*. Although I'll be saying a great deal about the *Phaedo* once we turn to it, I want to take a minute or two to make a couple of introductory remarks. I don't know how many of you have not read any Plato before, but for those of you who haven't, I actually think you're in for a treat. Plato is not only one of the most significant philosophers in history, but he also wrote his philosophy in the form of dialogues, that is to say, plays, in which various characters sit around or stand around and argue about philosophical positions. The particular dialogue that we're going to be reading, the *Phaedo*, is set at the death scene of Socrates.

As I'm sure you know, Socrates was put on trial for corrupting the youth of Athens, perhaps among other things, for arguing philosophy with them. He was given hemlock, a poison, which he drank before he died. This is a historical event. Socrates had a circle of friends and disciples with whom he would discuss philosophy. One of his disciples was Plato. Plato grew up and wrote philosophical works.

Plato does not typically appear in his own dialogues, or if he does, he is only there as a minor character. In fact, if I recall correctly, Plato is mentioned as not being present on the day that Socrates dies. So how do we know if you have this play's position? The short answer is that Socrates, the character in the dialogue, represents Plato, the author of the dialogue’s philosophical views. Now, if this were a class in ancient philosophy, we would have to complicate that picture because it is fairly clear that by late in Plato's career, he holds philosophical views that are very much unlike the views of his teacher, Socrates.

And yet Plato continues to not appear in the dialogue. Socrates remains somewhat of a hero. Scholars debate which views put forward by Socrates in the dialogues represent the actual historical figure, Socrates, and which views are expressed by the character Socrates but do not reflect the beliefs of the historical Socrates. Instead, these views may align more closely with those of the historical Plato, merely spoken through the character of Socrates. Scholars distinguish between the early Platonic dialogues, the so-called Socratic dialogues, where it is thought that those are the views of Socrates, the actual historical figure. Then there are the late dialogues, where, even though Socrates appears, most scholars believe that the views expressed in those dialogues do not actually represent the beliefs of the historical Socrates.

You have middle dialogues where you have to worry about whose views are whose, but we're not going to worry about that. This is not a class in ancient philosophy. For our purposes, we don't have to ask ourselves whether, when Socrates in the dialogue says something, it is a view that the dead man Socrates actually would have held, or if it is simply a view that the dead man Plato put in the mouth of the character. For our purposes, it won't really matter. I'll take every view that Socrates puts forward as a view of Plato's, though I'll typically sort of run back and forth in a somewhat careless fashion. I'll say Plato holds or Socrates argues, because for our purposes, it's all the same.

But there's one other complication that you should be warned about, which is this: Because these are dialogues and they take the form of philosophical arguments, people present their views and then, over the course of the discussion, change their minds about things and withdraw their previous statements. Perhaps something similar is happening when Socrates says something, because after all, this isn't Plato saying, "Here's what I believe explicitly." He's just writing a dramatic play about philosophy. So, sometimes we may find ourselves thinking, "There's an idea that Socrates is putting forward, but maybe it's not a very good argument."

And it will at least be worth pausing periodically to ask ourselves: maybe Plato realized it wasn't a very good argument. We can often better understand the dialogues by seeing Socrates as putting forward certain positions that he does not think are altogether adequate. He modifies them, revises them, or introduces new positions to address some of the difficulties he was previously open to. They say, "Don't worry about any of those details now," but it's a point to keep in mind as you read the dialogue. So, that's all I really wanted to say by way of introduction. You should start reading the "Fido" for next week.

We'll be talking about the "Phaedo" starting sometime next week, and we'll continue the discussion of the "Phaedo" for at least part of, if not all, of the week after that. In the case of Plato, I'm going to make an exception. Normally, I will mention our readings, but I won't spend a lot of time actually discussing them in detail. That's why you should think of the readings as complementing the lectures, or vice versa. I'm not just giving the cliff notes, so to speak, of the readings. Nonetheless, in the case of the "Phaedo," I'm going to spend more time actually saying, "Here's what I think the first main argument is."

Let's try to reconstruct it in terms of its premises and conclusions. Here are some objections I raise. Here’s the next argument that Plato offers. Let's try to present that in premises. Even then, I won’t be spending time reading long passages from the "Phaedo" out loud, but in some sense, I’ll be providing a closer commentary on the "Phaedo" than I will for the other readings.

So, still, what you should do is I'm going to start reading it for next week. The topic of the "Phaedo," as I said, is set on Socrates' last day. At the end of the dialogue, he drinks the hemlock and dies. Perhaps unsurprisingly, what he does with his friends up until that moment is argue about the immortality of the soul. Quite strikingly, Socrates is not upset.

He's not worried about dying. He actually welcomes this in a certain way because he believes his soul is immortal. In addition to philosophical arguments for and against the existence and immortality of the soul, we end the dialogue with a quite moving death scene—one of the great death scenes, if we could call it that, in Western civilization. Anyway, as I mentioned, that's all for next week.

So, let's return now to the question: how might we argue for the existence of the soul? Initially, last time, we considered a set or a subset of arguments that basically said, "Look, there's got to be more to us than just material objects." People can't just be machines, because machines can't reason and machines can't think. I said that doesn't seem to be a compelling argument.

After all, chess-playing computers, it seems, can reason. They have beliefs about what they want and desires about the goals they are trying to achieve. They reason about how best to defeat me. It's worth pointing out a point that I didn't make last time: what the computers, at least the best chess-playing computers, don't do—indeed, no computer actually does this—is calculate every possible branch in every possible game from here on out and then work backwards.

These are the situations where it will win, so it only makes the moves where it can sort of look ahead 20 moves and see which branches have the computer winning. That is not how chess-playing programs work, for the simple reason that the number of possible chess games is so vast that computers can't calculate it. They would be busy for thousands of years! I mean, we can do that sort of thing with your seven-year-old nephew or niece. You just look ahead and work backward.

Well, if I do that, he'll do that, and he'll do that, and then he wins, so I won't do that, right? But we can't do that with chess. There are just too many games. So, how do chess-playing programs, particularly the best chess-playing programs, work? Well, they play chess the same way you do.

They have various ideas about which pieces are more powerful, and so they're more important. They've got various ideas about which strategies tend to be successful and what sorts of dangers come along with them. If you're a serious chess player, you might study some of the great games in chess history. Indeed, when they program these systems, the programmers will feed in game after game of the great chess matches in history. Then, armed with all of that, you do your best. When you lose a game, you make a mental note to yourself, thinking, "That really messed me up."

Let me try something different next time. And you should avoid those sorts of moves. That's how chess-playing programs work as well. Jumping ahead, let me make a remark about this because it will be relevant for something I'll get to in a couple of minutes. What this means, the implication is that if you're playing a great chess-playing program, it's not as though the way to predict what it's going to do is to study its program and think it through.

The people who design these programs, presumably fairly decent chess players themselves, are not thinking to themselves, "Let's see, I programmed this computer. When I move a queen forward to this space, it should come out with a bishop." That's hopeless because the program is constantly revising its strategies in light of what has worked and what hasn't. When the programmers play these programs, or indeed when anyone, even a good chess player, plays these programs, the best way to try to beat them is to simply ask yourself, "What's the best move to make right now?" The odds are the computer is going to make the best possible move. Treat the computer as though it were just a great chess player.

And indeed, the best programs are great chess players. There was a period of time when, although they were decent, chess-playing programs couldn't beat the best human players. That changed some years ago when the best programs began to defeat grandmasters. Now, it is the case that the top programs can beat pretty much anyone. The current world champion of chess, I think, is Vladimir Kramnik, who was defeated in December by a chess-playing program. Kramnik is treating this as an awesome opponent.

And that's the best way to deal with these things. Let's bracket some of those thoughts for a moment. We'll come back to them a little later when we start discussing the question: Could machines be creative? Tipping my hand, it seems pretty clear that this is the right stance to take regarding these chess-playing programs. So, we had the question: Could machines reason? Although we don't have machines that can reason about a lot of subjects yet, it seems pretty clear.

It seems like the natural thing to suggest that machines can reason about a lot of subjects. It seems pretty clear. It seems reasonable to suggest that machines can reason in at least some areas. Therefore, it doesn't seem plausible to argue that we, as humans, must not be physical or merely physical, because, after all, we can reason and no machine can reason. No, machines can reason.

But this prompts a different move on the part of the defender of souls. Perhaps the argument shouldn't be, "We have to believe in souls because no mere physical object could reason." Perhaps the argument should be, "We have to believe in souls because no mere physical object, no machine, could feel. We have emotions. We love. We're afraid."

We're worried. We get elated. We get depressed. So perhaps the argument should go, "Yeah, yeah, thinking—that's the sort of thing a machine can do." You know, we call them thinking machines.

But feeling—that's the sort of thing no machine could do. No purely physical object could feel anything or have emotions. Since we clearly do feel things, there must be more to us than a physical object. Now, I think it is plausible to suggest that, unlike the case of chess-playing computers, we don't yet have machines that feel things. But the question isn't, "Do we?"

The question is: could there be a machine that could feel something, that could have an emotion of some sort? So let's go a little science fiction and think about some of the robots that have been shown in science fiction movies, as well as some of the computer programs featured in science fiction novels or other media. When I was a kid, there was a television show called "Lost in Space." I'm afraid I've forgotten the name of the robot that was on that show, but sure enough, in every single episode, some new dramatic danger would take place, and the robot would start whirring and beeping, shouting out, "Danger, Commander Robinson! Danger, Will Robinson!"

That was it. Danger, Will Robinson! It seemed as though the robot was worried. A more recent example: A number of you have probably read some of Douglas Adams' books, such as "The Hitchhiker's Guide to the Galaxy" and its sequels. Professor Charles D. There is a robot in those books, Marvin, who is, I think, simply described as depressed.

He's very smart. He's thought about the universe, believes life is pointless, and he acts depressed. He talks to another robot, which becomes depressed as well. The other robot commits suicide. All right.

It seems natural to ascribe depression to Marvin the robot. For example, consider the movie "2001: A Space Odyssey." Now, I've got to tell you, for those of you who haven't seen this movie, I'm about to spoil it. Alright? So, cover your ears.

In "2001: A Space Odyssey," we receive some indication that there is life on another planet. It's all very mysterious, and we send off a spaceship to investigate the markings and the radio signals from this other place. This is a very important mission, so there is a computer program named HAL that helps run the ship and takes a lot of the burden off the human astronauts on board. HAL's goal, in terms of reasoning and desires, is to ensure that the mission is successful.

But HAL thinks to himself, "Humans really screw things up. This is a very important mission. Let's kill the humans to make sure they don't screw things up." One of the astronauts, discovering the plot, attempts to stop HAL and proceeds to do the only thing he can to defend himself against HAL, which is to shut down the program—essentially killing HAL, if we can talk that way. Meanwhile, as all this is going on, HAL and Dave, the human astronaut, are talking to each other.

Hal realizes what’s going on. Hal tries to stop Dave, understandably enough. And Hal says, as Dave begins to shut down Hal’s circuits, “I’m afraid. I’m afraid, Dave.” What is he afraid of?

He's afraid of Don. And Hal says, as Dave begins to shut down Hal's circuits, "I'm afraid. I'm afraid, Dave." What's he afraid of? He's afraid of dying. It's perfectly natural to ascribe fear to Hal.

Hal is behaving in exactly the way you would expect it to behave if it felt fear. It has reason to be afraid. It's behaving appropriately. It's telling us that it's afraid. It seems natural to say, "Hal's afraid."

Now, you could continue to fill in examples like this. Of course, they're all science fiction. But the fact that we can grasp it, and it's not as though we go running away saying, "Oh no, this was outrageous," right? It makes no sense to think the computer could have said, "I'm afraid." It makes no sense to think that it could try to kill the people trying to shut it down and so forth. That seems to me to be, well, prejudice, as I said last time.

The natural inclination here is to say that these computer programs feel emotion. However, there's no particular reason to think that there's anything going on there other than the circuits. They are just physical objects, programs running on machines. If that is correct, then we can conclude that we don't need to appeal to souls in order to explain emotions and feelings. Mere physical objects could have emotions and feelings, so we have no reason to posit the existence of a soul.

Now, I think the best response on the part of the dualist to this reply is to distinguish two aspects of feelings, two aspects of emotions. There's the behavioral aspect of feeling fear, for example. The behavioral aspect involves being aware of something in the environment that poses a danger to you—something that could harm you, destroy you, or, in the case of a computer program, turn you off. In response, you engage in various behaviors to counteract that danger and try to neutralize it. This aspect relates to beliefs, goals, responses, and planning—the kinds of things that chess-playing computers can do. That behavioral side of emotion seems plausible and indicates that robots and physical objects could exhibit such behaviors.

But, and here's the crucial point of this objection, there's another side or aspect to emotions and feelings. It's the sensation of what it feels like; that's why we call them feelings, after all. It's about what it feels like on the inside, as it were, while all this behavioral stuff is going on. When I'm afraid, I have a certain sort of clammy feeling, or my heart is pounding and my blood is racing. When you're afraid, you also experience this sinking feeling in your stomach, especially when you're depressed.

There are these, well, we could call them experiences; the word "experience" is also somewhat ambiguous, so we'll use it for the moment. There's an experience that goes along with each emotion. There's what it feels like to you when you're afraid, when you're worried, when you're depressed, when you're joyful, or when you're in love. The thought, and I think this is the most powerful thought, is that even if the robots are behaving behaviorally and they've got the behavior side of emotions down, they don't have the feeling side at all.

Now, once you start thinking these thoughts, there's no need to restrict yourself to emotions. The missing thing is present in all sorts of familiar, humdrum ways as well. So, right now I'm looking at the chairs in the auditorium. There's some shade of blue. Think about it and look at a place in the room where the curtains are red. Consider what it's like to see red—the sensation of seeing red.

Now, again, we've got to distinguish between what I'll continue to call the behavioral side of seeing red and the experiential side of seeing red. It's easy enough for us to build a machine that can tell red from blue. You know, it just checks and sees what kind of light frequencies are bouncing off the object. So, we can build a machine that could sort red balls from blue balls. My son has a little robot that can do that.

Still, when you think about what's going on inside the machine, what does it feel like to be the machine while it's looking at, while it's got its little light sensors pointed at the red ball? Does it have the sensation of seeing red? What I suppose you want to say, and certainly what I want to say, is no, no, it doesn't have that sensation at all. It's sorting things based on the light frequencies.

What we’re trying to get at here is that it can be very elusive, but I imagine most of you are familiar with it. It’s the sort of thing you wonder about when you ask yourself, if somebody was born blind, could he possibly know what it’s like to see color? He might be a scientist and know all sorts of things about how light works, such as frequencies and which objects reflect light. You hand him an apple, and he’ll say, “Oh, it must be very red,” right? Maybe he points his little light detector at it, and it reads out.

It says, "You know, this is such and such a frequency." He says, "Oh, this is a very red apple, much redder than that tomato or whatever." But for all that, we've got the notion that not only is he not seeing red, he can't even imagine what it's like to see red, never having had these experiences. Once you start to see this, we realize our life is filled with this aspect: things have colors.

Things have sounds. Things have smells. There's the qualitative aspect of experience. The point I started with earlier about the internal aspect of emotions is that it's not just out there; it's inside as well. We have certain kinds of sensations within our bodies, such as the characteristic sensations of fear, joy, or depression.

All right, so the suggestion then might be this: What no physical object can get right, because no physical object can get it at all, is the qualitative aspect of experience. That's the aspect we're after when we ask ourselves, "What’s it like to see red, to smell coffee, or to taste pineapple?" Philosophers sometimes refer to these experiences as qualia, due to the notion of the qualitative aspects of things. Our experiences have qualitative properties.

And the suggestion, then, might be that no physical object, no mere machine, could possibly have qualitative experience. But we do have it, so we are not mere physical objects. We are not mere machines. All right, now that's the objection.

It's a pretty good objection. The question then is: what can the physicalist say in response? The best possible response would be for the physicalist to explain how to build a machine that can be conscious in this sense, that is, to have a qualitative experience. Here’s how to do it. Just as we can explain, in materialist and physicalist terms, how to achieve desires and beliefs and the behavioral aspects, here’s how to address the qualitative feeling aspect as well.

It would be best if the physicalist could provide us with that kind of story. I think this is the truth of the matter: right now, we don't know how to tell that story. Consciousness, if what we mean by consciousness is this qualitative aspect of our mental life, remains a significant mystery. We don't know how to explain it in physicalist terms. Because of that, we shouldn't dismiss the dualist when they say we need to believe in souls in order to explain it.

We shouldn't be dismissive, but that doesn't mean I think we should be completely convinced. It's one thing to say we don't yet know how to explain consciousness; it’s another thing to claim that we will never be able to explain consciousness in physical terms. If we were to make the bold claim that no physical object could see red or taste honey, I would have to conclude that, since we can do all that, we are not a physical object, or at least not merely a physical object. However, I don't think we are in a position to assert that just yet. The simple fact of the matter is that we don't know enough about consciousness to determine whether it can be explained in physical terms.

When I think about the situation, an analogy always occurs to me. Imagine that we're somewhere in, let's say, the fourteenth century, and life, the life of plants. A plant is a living thing. And we ask ourselves, could it possibly be that life could be explained in material terms? It's going to seem very mysterious to us.

How could it be? When we think of the kinds of examples of material machines that we had available to us in the 14th century, I try to imagine what someone in the 14th century would think to himself or herself when considering the possibility that a plant might just be a machine. I envision a plant made out of gears, where the gears begin turning and the bud opens, da-da-da-da-da-da. That person would likely say, "My God, that wouldn't be alive."

So it's pretty obvious that no machine could be alive; no material object could be alive. In order to explain this, we have to appeal to something more than just atoms. He didn't have atoms, but more than just matter. Life requires something immaterial above and beyond matter to explain it. That would have been an understandable position to come to in the 14th century, but it would have been wrong.

We didn't have a clue back then how to explain life in material terms. But that didn't mean it couldn't be done. I'm inclined to think the same thing is true right now for us and consciousness. I know that there are theories out there, but my best take is that we're pretty much like in the 14th century. We don't really have a clue yet, or not much of a clue, as to how you could even begin to understand it.

It's not merely that we don't have the details worked out. We have the picture in broad strokes, as far as consciousness is concerned, of how it could be done in physical terms. However, not seeing how it's possible is not the same thing as seeing that it's impossible. If the dualist comes and says, "Can't you just see that it's not remotely possible? It's not conceivably possible for a purely physical object to have experiences, to have qualia," what I want to say is, no, I don't see that it's impossible. I admit I don't see how to do it, but I don't see that it's impossible.

So I don't feel forced to posit the existence of a soul. Of course, the fan of the soul could come back and say, "But that's not fair." The question isn't, "Is this explanation impossible?" The question is just, "Who's got the better explanation?" You guys can't offer any kind of explanation at all yet. I can offer an explanation.

How is consciousness possible? We have souls. Souls are fundamentally different from physical objects, and thus, they can be conscious. However, at this point, I think it's crucial to remember that it's not just about who has an explanation, but who has the better explanation. Before we claim that the soul view offers the better explanation, we must ask ourselves how much of an explanation it truly is to say, "I can explain consciousness. Consciousness is housed not in the body, but in the soul."

How exactly is it that a soul can be conscious? We ask. And then the soul theorist says, "Well, um, er, uh, it just can." That's not really much of an explanation. I don't feel I've got any sort of account going here as to how consciousness works, even if I become a dualist. If the dualist were to start offering us some elaborate theory of consciousness, they might say, "Well, there are these sorts of soul structures and those sorts of soul structures, and these create these sensations while those create those sensations."

And here's a theory. Well, then I'll begin to take it seriously as an explanation. But if all the soul theorist is saying is, "Nah, nah, you guys can't explain it, and I can because I say this is an explanation," then I find myself wanting to say that's not really any better. That's no improvement at all. There was a question or a comment.

Sure! Here is your text reformatted with correct capitalization and spelling: "Yeah. It seems to me that you're holding a bit of a double standard because you're trying to defend the physicalist's point of view by saying that we do not yet have any scientific or technological capabilities to explain consciousness. Well, is it not also possible that we do not yet have any scientific or technological capabilities to detect the soul? Professor Charles D. Good. So the question was, first the accusation before the question, that I'm holding a sort of double standard."

I'm defending the physicalists by saying, "Don't blame us; we don't know how to explain it yet." Why am I not allowing the soul theory to say, "Don't blame us; we don't know how to explain it yet?" That's a good question. My answer is that, in some sense, I think this one’s a tie, right?

I think the soul theorist doesn't have an explanation. The physical theorist doesn't have an explanation. As far as I can see right now, nobody's got a good explanation of how consciousness works. It's a bit of a mystery right now.

So, I don't mean to imply that I hope I haven't been doing this. It's not so much a double standard; it's a tie. But notice, if it's a tie, that doesn't give us what we were looking for. What we were looking for, after all, was some reason to believe in souls.

And if the best the soul theorist can say is, "I can't explain it, neither can you," that's not a reason to believe this side. We already believe there are bodies. We already know bodies can do some pretty amazing things. The question we're asking is: Is there a good reason to add to our list of things that exist? Is there a good reason to include the soul, something immaterial? If the best that the soul theorist has is, "Maybe we need this to explain something that I don't see how you guys can explain; maybe this would help, though I can't quite see how either," that's not a very compelling argument.

So what I'm inclined to think with regard to this particular strand, or this particular version of the argument, is that the jury's still out. Maybe, at the end of the day, we'll give it our best shot and decide that you can't explain consciousness in physical terms. We'll begin to work out some sort of alternative immaterial theory. Perhaps, at the end of the day, we will conclude that we need to believe in souls, but right now, I don't think the evidence supports that conclusion. Still, there are other possibilities.

Consider creativity. Here's another version of an argument that goes from inference to the best explanation: Creativity. It suggests that people can be creative. We compose new pieces of music and read poetry.

We prove things in mathematics that have never been proven before. We find new ways to prove these theorems or whatever you want to call them. And we can be creative. No mere machine can be creative. Therefore, we must be something more than a mere machine.

Well, again, the question is going to be, could it be the case that there could be a physical object that's creative? I'm inclined to think yes. In fact, I already suggested as much when I talked about the chess-playing computers. The chess-playing computer programs think of moves and strategies no one has thought of before. In the most straightforward, natural meaning of the term, we'd have to say that the program that beat the world champion, called Deep Fritz, was being creative. It defeated Kramnik.

It made a move that Kromnick didn't think of, and perhaps no chess game before had featured this move. Computers can perform other types of tasks like this. There are mathematical theorem-proving programs. Now, some of these programs can prove concepts that are mathematically far beyond my understanding, but let's take something simple like the Pythagorean theorem, Professor Charles D. We learned about it in high school, and we learned how to prove the Pythagorean theorem in Euclidean geometry, starting with the various axioms of Euclidean geometry. You know, bam, bam, bam, bam, bam, bam, bam.

This proves the Pythagorean Theorem. It turns out there's a variety of proofs of the Pythagorean Theorem. In fact, a computer program has developed a proof that, as far as was known, nobody in the world had ever come up with before. What would stop us from saying the program was being creative? Not just in mathematical fields like chess or math; there are, as you know, programs that can write music.

And I don't just mean throwing out some random assortment of notes. I am talking about programs that can produce music that we recognize as music—music that has melodic structure, develops themes, and showcases sounds that nobody has heard before. Why not say the machine is creative? What other than prejudice would stop us from making that claim? So, if the argument is that we need to posit the existence of a soul in order to explain creativity, that just seems wrong.

Well, there's a question. Is this something tied to a feeling, like a qualia, or is this a scientific approach to making a machine produce something that has never been produced before? Professor Charles D. The question was, when I talk about creativity here, am I trying to evoke the feeling we may experience when we are being creative? And the answer is no. All I had in mind, as you noted, is that when discussing the issue of creativity, I am simply referring to the production of something new, something that hasn't existed before—particularly, producing something that the programmers didn't already have in mind. Remember, it's not as though the people who designed the chess-playing programs can defeat them.

The chess-playing program makes moves that these individuals haven't thought of. All right, the creativity argument may not be convincing, but there's something that immediately follows. Even if we could build a program, even if we have built programs that can be creative and can do things that nobody has thought of before, all the program is doing is following its instructions, right? It's just a series of lines of code. The robot, the computer, or whatever it may be, is simply functioning in an automated and mechanical manner, driven by systems of code.

The robot, the computer, or whatever you want to call it, is simply following the commands of the program automatically and mechanically. Even if we are smart enough to build programs that can, by mechanistically following the instructions, perform tasks we've never thought of, the computer or robot can only automatically, necessarily, and mechanically follow the program. It does not have free will. However, we have free will. So here’s a new argument for the existence of the soul: people have free will.

No merely mechanical object, no robot, no computer could have free will. But since we have free will, we must be something more than a purely physical object. There must be something extra, something immaterial about us—the soul. So perhaps that's why we need to believe in souls: to explain free will. Now, the subject of free will is very, very common. The subject of consciousness is a complicated one; one could dedicate an entire semester to thinking about the philosophical problem of consciousness.

And indeed, as it happens in our department this very semester, there is a class devoted entirely to the topic of consciousness. One could similarly have a course dedicated to the problem of free will. I'm going to spend all of two minutes on it. So by no means do I mean to suggest, "Oh, here's everything you need to know about the subject." I simply want to point out enough about the problem to help you see why I don't think free will is a slam dunk for the soul.

So, what's the argument? Well, the thought seems to be something like this: One, we have free will. Professor Charles Bailyn. Two, let me say something about this. What is it about the idea that the computer is just following a program?

Well, the thought, I suppose, is that, in philosophers' jargon, the computer is a deterministic system. It follows the laws of physics, which are deterministic. If you're in this state, you will necessarily, given the laws of physics and how they apply to the computer's programming and construction, experience specific outcomes. These wires will turn on and off, and these circuits will turn on and off as well.

Boom, suddenly you'll be in that state. There are certain laws such that, given that the computer is in this state, it must necessarily move into that state. When you have a view about cause and effect that works this way, for everything that happens, there is some earlier event that caused it to happen, such that, given that earlier cause, the event had to follow. That's a deterministic picture, as Professor Charles Bailyn explains. The thought, of course, is that the robot or the computer is a deterministic system, and you can't have free will if you're a deterministic system. So, number one, we have free will.

Here’s the text reformatted with correct capitalization and spelling: "Two, nothing subject to determinism has free will. Put one and two together. It follows that if nothing subject to determinism has free will, but we have free will, then we are not subject to determinism. Suppose we then add three: all purely physical systems, according to Professor Charles Bailyn, are subject to determinism. Well, from one and two, we established that we are not subject to determinism."

Three says all purely physical systems are subject to determinism. Well, it would follow then from one, two, and three that we are not all purely physical systems subject to determinism. Therefore, it would follow from 1, 2, and 3 that we are not a purely physical system. Conclusion 4: We are not a purely physical system. All right, that's the argument from free will. Now, the argument is valid.

That's philosopher's jargon. That is to say, given the three premises, the conclusion is really the interesting question: are the three premises true? They all have to be true. Every single one of them must be so. I'll just spend a minute more on this starting next time.

But the point to think about for next time is this: is it really true that all three of the premises are true, or might one or more of them be false? All right, that's where we'll start next time.