Transcript Reading: Kant Part 1

What's on the docket for today is what I said. We're going to take up Kant. We're going to look at what's wrong with the analytic-synthetic distinction. I'm going to explain that, and after I do, we'll have a few comments about where Hume comes out in all of this, because that's what we were doing—we were winding up Hume. And then I'll get to Kant and his reaction to all this. For Hume, this is the clinching argument: if all statements have to be either analytic or synthetic, and he's correctly described each, then we can't get any legitimate law statement. We can't know anything other than our own perceptions. So, we don't know there's an external world. We also lose the concept of a self.

Hume said that whenever he looked most intimately into himself, he only found some present impression or an idea, a copy of a past impression. He didn't find anything called the self. It's really like saying, “I pointed the camera at this object and that, but none of them gave me a picture of the camera.” In a way, it's a little silly, okay, but now he's got the analytic-synthetic distinction. These are the big guns, and it wasn't until past the middle of the 20th century that people began to catch on that this was a red herring. It was wrong, and it really didn’t do any of the things he said it would. To show you why, I want to set this up again and want you to look at the board.

Analytic statements are defined as statements in which the meaning of the subject—that is, the concept of the subject—involves whatever the predicate asserts of it. So if the subject is "bachelor" and the predicate is "unmarried," then sooner or later, in analyzing (that’s why Kant used the term "analytic"), we come across the predicate "unmarried."

That's why we get the old statement, and this is supposed to explain where we get the idea of a necessary truth, of a law. So when we make a statement such as "All water is H₂O," that's because H₂O is part of the meaning of water, and it can't be false. It's a law. Okay? Is that a good explanation for where we got the idea of a necessary truth? And the answer, I think, is no, for this reason: if the meaning of the predicate term of a sentence is not part of the meaning of the subject, it's not because the sentence isn't necessary, it's because the sentence isn't true.

It's truth and falsity that are explained here, and Hume doesn't realize it. Any statement that asserts something of a subject, and this predicate is not part of the concept of the subject, is because it's not true of the subject. Meaning containment is a condition for truth. "Today is Thursday" looks like a synthetic proposition, and it is. We bring together "Thursday" and "today," but that's true if, and only if, today is Thursday. If it really has the property, that doesn't make it a necessary truth. It's not a necessary truth that today is Thursday—today might be Tuesday or Friday, and I'm mixed up. So, the definition of a synthetic statement, as Hume gives it, where the meaning of the predicate is not in the concept of the subject, is because the subject doesn't have that predicate, and the sentence is false.

We're not looking at two kinds of true statements. It's not that analytic ones have meaning containment, and synthetic ones don't—no containment. If there's no containment of P in S, it's because P isn't one of the properties combined in the concept of S, and if it isn't, it's because it's not true of S. So, take the example I gave you before: "My house is brick." What defenders of this definition often try to do is give sentences like this as an example of a statement in which the meaning of the predicate is not contained in the meaning of the subject. But, ladies and gentlemen, that’s equivalent to the three-card trick, you know, where you put your money down and it disappears.

Look at it again. They say the subject of the sentence is "house," and a house doesn’t have to be brick. But the subject of the sentence isn’t "house"—it’s "my house," and my house is brick. So, if you had the proper concept of "my house," one of the things you'd come across is that it's made of brick, and that’s a condition for this being a true statement, not a law. There’s no law of nature that says my house has to be brick. That’s ridiculous. But it is, in fact, brick, and the sentence that asserts that it is brick has the meaning of the predicate as part of the meaning of the subject. That’s what made it true. But it’s still contingent—it’s not necessary.

So, meaning containment doesn’t distinguish necessary truths from contingent truths. It distinguishes true statements from false statements and leaves us unsatisfied and unanswered as to where we got the idea of a law. If the only things that come into our minds are first from the senses, then we can't have an idea of law, but we bloody well do. Read Galileo and Newton—we have laws proposed all over the place and experiments that confirm them. So, this is a howler. This is a really bad mistake that fooled a lot of people for a long time. This is why there’s no proof of God. There's no proof of God because if the statement is analytic, it's true only by definition—it depends on how we define the subject and the predicate. If it's not just true by definition, if it’s really a matter of fact, well, then it’s only contingent.

You see, in the rationalist tradition, they tried to prove that God's existence was a necessary truth, that existence was part of the definition of God. Some of the most famous arguments for the existence of God argued that way. It may sound silly to you, but they did. They said God is defined as the being with all perfections and only perfections; therefore, He has to have perfect existence.

Does that convince you? I’m sorry—it doesn’t convince me. I think Hume is right. You can’t just define something as existing, and therefore it exists. If it were that easy, I could get ten million dollars by defining it as being in my bank account, and then it would have to be there, because having money in my bank account is one of the perfections—it’s the chief perfection. Well, anyway, I think you see the problems with this, and Kant misses it.

Kant doesn’t say that the analytic-synthetic distinction is, in fact, a distinction between true and false statements. It doesn’t tell us why the ones that appear to be necessarily true are, and the ones that appear not necessarily true, really aren’t. It doesn’t straighten all that out, and it doesn’t account for where we got the idea of a law in the first place. So, look at Hume’s record here before we go on to how Kant deals with it.

Hume wants to say we can’t know there are external objects, an external world. And why? Because there's nothing in the mind not first in the senses. But then, how would we know this? It's not a set of percepts that come into the mind. There's nothing in the mind not first in the senses, yes, but there’s the idea that there's nothing in the mind not first in the senses, which doesn’t come through the senses. It's one of those things you couldn't know if you were right. If it's right, how would it get into your mind?

The same thing holds for his other points. We have no knowledge of anything called the self—the one who perceives, the one who thinks. Remember, whenever I look most intimately into myself, I never come across anything but some present perception or impression, or the copies of past impressions. And Hume tells us that a woman wrote him a letter and said, “Then, what's the ‘I’ that looks into myself?”

And he just says, "I don’t know." So, something is looking into himself, but since he doesn't find that to be a perception, he dismisses it, because otherwise it would violate the rule that there's nothing in the mind not first in the senses. That's also the basis on which he rejects the external world.

So, I would say, since this is false, that is false; and since that is false, this is. It's called modus tollens in logic. See, he's sweeping the board. He's cleaning house, man. There is no self, there are no external objects, there’s no certain ethics. You can’t prove that God exists—not the God of the Bible. You might prove that some kind of intelligence designed something or other, maybe—that's the best you can get out of that. But you certainly can't show there's only one, that it's also the creator, or that it offers you a covenant of love, forgiveness, and everlasting life. None of that comes along.

So, kind of like the spirit of the world, self, others, God, science—Hume’s like the drunk who starts the fight in the bar and wrecks the place. It’s all wiped out. And that’s how Kant sees it. Kant said he was "stirred from his dogmatic slumber." He had been just like someone dreaming, and he accepted the self, the world, God, ethics—right and wrong. Then Hume came along and swept them all away, saying, “You can't know them. You can't know any of them.”

So Kant just couldn't let him get away with that. In 1781, Kant published The Critique of Pure Reason, and he says something now that Locke had said, but had kind of gotten ignored by Berkeley and Hume.

The way to start out is to analyze how the mind works before we conclude what does or doesn't exist. We have to figure out what the mind can know and what it can't, and to do that, we have to introspect and find out as much as we can about how the mind operates—how it operates in ordinary experience, how it operates in perception, how it operates in abstraction, and how it forms concepts, and so on.

Kant says, "That's what I'm going to start with, and I'm going to explain to you"—this is a very famous statement—"how a priori synthetic judgments are possible." I remember that in the chart, it's supposed to be that synthetic statements are about matters of fact, and they're not necessary. It's only analytic ones that are necessary, and they're not about any facts.

They don't have to be, but I'm going to show you how there are statements that are both synthetic (they're about reality), and they're a priori, which means they're known from the meaning of the terms to be true. They're necessarily true, but they're about reality, and that's what laws are. So, for example, every action has an equal and opposite reaction—this is supposed to be a synthetic judgment about the world and, like analytic judgments, necessarily true. The first thing I want to show you is how those statements are possible. What he's saying is that he's going to show how, besides analytic and synthetic, there’s another class of statements—a priori synthetic—and they exist too, and here’s how and why. I’ll show you now.

I’m about to come to that, but you realize then what he's doing is accepting the analytic-synthetic distinction, which I recommended we not do because it's false. That's why. But he’s saying, "I can show you there’s a third." So, yes, Hume, you're right—there’s analytic and synthetic—but I’m going to do an end run around that because I’m going to say there’s also a third kind, and the third kind is the laws. And now I’ve saved science. There are two things Kant will not let go of, and one is science—by that he means astronomy and physics and other emerging branches of science. In his day, they mostly fell under physics. They were things such as electrical theory, theories of vision, light transfer, and so on. But he expected more sciences to emerge—just after that, geology emerged, and there were a number of others. But anyway, they all require laws, and so he wants to save that.

The other thing he thinks he must save is ethics. He’s going to show you that there are ethical laws and that these ethical laws are necessary truths, just like the laws of science. See, there are many things about which he doesn't think there are necessary truths. Anything that tries to prove the existence of God just falls flat. It can’t be saved by his move here, he thinks. But it will save science and it will save ethics because it will show why there is necessary truth in science, necessary truths in ethics—the ethical statements that capture these laws—and he's going to show you how you know that they’re necessary truths and what they are, therefore, things that must never be done. So, he lays out ethics in a system as close to mathematics as he can get it.

"Don't lie" was on his list. It’s one of the first examples he gives in his theory of ethics, and he’s going to say that turns out to be one of the necessary truths—it’s always wrong to lie. I don’t think so. I gave you an example already: the man who was hiding Jewish families from the Nazis lied, and I think that was the right thing to do because it was the loving thing to do. But Kant doesn’t bring ethics down to "love your neighbor as yourself," you see. He brings it down to a formula for testing a statement to see whether it has certain logical properties, and if it has them, it’s right; if it lacks them, it’s wrong. So he equates ethical right and wrong with having certain logical properties in the statements. And I don’t think that’s what logic is about, but he did—that was his proposal. He thinks he can get up from that, for example, "It’s wrong to lie." Don't lie. And people who read it say, "You can't really mean that. You can't mean that you wouldn’t lie to the Gestapo to save innocent families." So Kant wrote a little tract, a little pamphlet, a kind of popular work, and in it is this line: "I would not lie, no, not to save mankind."

And that has led some people to say that Kant wrote about ethics but was himself evil because if he wouldn’t tell a lie that could save mankind, then he can just step off the stage and not talk to me at all. That’s a little preview also.

So, how does Kant start out tackling this if he misses that the analytic actually shows how a statement is true, and synthetic would make a statement false? If the meaning of what’s asserted of something is a characteristic the subject doesn’t really have, then the assertion that it has it is wrong. So he missed that fact—that might be a good way to introduce him. Let’s see what else he missed. I think we can do that.

Kant misses the mistake in the analytic-synthetic distinction.

That’s one big mistake. An even bigger one, which I guess I should have listed first, but I didn’t say I would list them in order of importance, is even more important. He misses the mistake that the given in experience—that is, what comes to us, whether we will it or not. When I look over here, I see a clock on the wall, not because I want to, but because there’s a clock there. What is it that just comes to us? Presents itself to us? That’s been called the given in experience. It’s just handed to us. We don’t call it up. It doesn’t result from anything we’re doing. If I open my eyes and look in that direction, I’ll see a clock. I can not look, or I can close my eyes, I don’t have to see a clock. But if I look at that wall, I will. This given in experience is sensations—that’s what comes into the mind: sights, touches, smells, tastes, and sounds.

They all missed that. I mean, that is just a given in modern philosophy. Descartes starts with that. He wants to believe there are external objects that have only physical properties, but they impact the mind, which is a rational, thinking thing, by producing perceptions in the mind that are copies of the external objects, right? Remember, the trouble is, how can you have a purely sensory picture of a purely physical thing? The physical thing has no sensory properties in that theory, right? A rock outside Descartes' window has no color, makes no sound, has no feel to it, doesn’t smell in any particular way. Those are all things that register in minds. So they’re not true of the rock. The rock only has physical properties. It only occupies space, moves, has weight, mass, density, and so on. But if it has no sensory properties, a sensory picture can’t copy it. All it has are physical properties, and they’re not in minds. Minds are completely non-physical. So Descartes divides reality in two, and when he gets all done, he can’t put Humpty Dumpty back together again. But that didn’t lead any of the people who followed him to say, "Whoa, wait a minute. The given in experience isn’t perceptions. It isn’t the sights, tastes, touches, smells, and sounds." It isn’t things that have only color, tactile sensations, smells, or tastes true of them. What’s given to our experience are things like walls, lights, clocks, books over my shoulder, bookcases, and cars in the parking lot. That’s what’s given. What comes to us, whether we want it or not, if we look in a certain direction, are the things, events, states of affairs, people, and relationships that constitute the world around us. And we abstract from them. We single out logical properties from the rest, sensory properties from the rest, physical properties from the rest. And we may focus on a particular kind of property, but in science, we never deal exclusively with what we’ve picked to focus on because there is nothing that’s purely physical or purely sensory or purely rational. There isn’t anything that’s purely anything. Nobody ever stopped to think of that—Dewey River did in the 20th century—but these guys were convinced of what we’ll call the empiricist dogma.

The empiricist dogma is that what’s given to us are colors, sounds, tactile feels, tastes, and smells. That’s it. In fact, the theory is that they come in separately into the mind, and the mind combines them into objects. If that’s right, then there isn’t anything we experience that’s really given—they’re all artifacts that our minds put together. So, if you’re willing to buy that, I guess it’s no big deal to say, "And we don’t know that these reflect an outside world at all. Maybe there is nothing called the external world. Or if there is, anyway, we don’t know anything about it. So why would it have to be like the copy of what occurs in our minds?" Right?

There are causes that produce effects that are nothing like the causes. If I take five pounds of plutonium and five pounds of plutonium and, in the right circumstances, smack them together and set off dynamite, I get one hell of a mushroom cloud. Okay, but that’s nothing like a lump of metal and another lump of metal and the explosion of a dynamite stick—one’s not a copy of the other. So we don’t think that causes always have to produce effects that copy them—that’s not true. We want to say that some of them do, like in normal sense perception, but I don’t think normal sense perception gives us copies.

Now, what other mistakes? Okay, let’s say here’s the argument given by Berkeley and repeated by Hume as to why our perceptions are not the same as the external world. I’ll call it counter-examples to realistic perception. We’ve covered this before, but it doesn’t matter if it’s a mistake—we’re covering it because it’s one that Kant makes too. He assumes this from his predecessors.

Counter-examples to realistic perception include things like putting a stick in water and having it appear bent. The bent part of the stick is an optical illusion. We say the real stick is still straight, but in that case, the appearance of the bent stick has to be something different from the stick itself because the stick itself doesn’t bend when we put it in the water. There are other things like this—railroad tracks. We’re walking along the railroad tracks, and as we look ahead toward the horizon, the railroad tracks appear to converge, but we know they don’t. We can walk right down to the spot that looks like a convergence, and they’re still the same width apart, with the steel rails and wood ties still just as wide apart as ever. So, there have to be two things: the real track that doesn’t converge and our picture of it that does. That’s the argument. There has to be another thing inside our minds that converges, and it isn’t really representing the truth about the object.

There are lots of other examples like that—the same thing, like all things appearing to get smaller the further away they are. Now, infants make out the truth of this and aren’t fooled by these things. I’m saying that the theory that there are two things involved, where one is supposed to be the copy of the other but manages in some serious way not to be, is not the best explanation—or the simplest. The simplest explanation says that, for example, if I paint a wall white and later, after it dries, shine a certain shade of purple light onto the wall and the wall appears lavender, is that because there are two different things—the wall, which is really white, and what appears lavender in my mind? Or is this a better explanation: the white surface of the wall has the disposition to appear white to normal vision in normal light but purple in the presence of this kind of light? In that case, why can’t we say the same thing about the converging railroad tracks? The railroad tracks have, among other properties, the property of appearing to converge at the horizon (or at some distance) to normal vision in normal light. Similarly, with the stick: the stick not only is straight but also has the disposition to appear bent in water.

Now, all those dispositions I just talked about—the tendency, the potential property—are sensory properties. It has the sensory dispositional property to appear bent in water, or for the tracks to appear converged at the horizon, or for the wall to appear purple under a black light.

And the reason we speak of white as being the real color of the wall is because that's what normal perception sees in normal light. That black light here that I took out and plugged in shines on the wall. That's not normal light. These are normal perceptions—the tracks and the stick—but they are not being perceived under optimal conditions. The optimal condition for the stick is for it to be out of the water. Put it in water, and we don't have optimal conditions anymore, and the disposition to appear bent is activated. Same thing for the tracks. Standing that low to the ground and looking down the tracks is not optimal perception. Being up in a helicopter and seeing the tracks spread across the land below me, they don't appear to converge. That's the optimal position for observing them. So, the optimal conditions for that are not to be standing at my height or yours and looking down the railroad track. Why isn't that a more obvious and simpler explanation of what's going on? The fact is, these things all have sensory properties. They're passive to our perceptive abilities, and their potential is only activated when observed. But we don't have to say there are two things. We don't have to multiply entities needlessly—this rings a bell, it's called Occam's law. It says the better theory always guesses at fewer things. If we can explain something with three theories and everything else takes more, we prefer the one with three. Every other theory takes more—four, five, six, or ten—then the one with three is the one we take if they all have the same explanatory power. And I'm saying that what I'm suggesting to you has the same explanatory power and doesn't need to postulate a second set of realities inside our heads, only to be startled by the conclusion that’s all we really know. So, we don't know the world? There's a world. Do you realize that this is pretty extreme, right? These are things that can't be dismissed.

The Enlightenment is often praised as a period in which people threw off superstition, followed reason, developed science, and supposedly came out of the Dark Ages—the Middle Ages, yeah? And what do we end up with? Well, we're not sure there's a world. For many years, I told my students: If you think that's nuts, you're right. If there's anything you know, it's that you're not the only thing in the universe. You're in a world. And there's a world containing things, events, states of affairs, relations, and other people. Of course, there is. Because normal sense perception shows you those things to be self-evidently real. Now, the Enlightenment gives itself that name. Kant wrote an essay called What is Enlightenment, and he says, "Dare to think, dare to postulate what things are really like. Be bold. Strike out." Sounds like the rationalists, right? They thought that was the way to go, and they ended up inventing the weirdest stuff. Like, "Well, reality is really monads, and monads have physical power, and they’re all conscious, so the dust under your feet knows you're stepping on it." What? This is serious? Yeah. And they said they were a period of enlightenment, as compared to what—the Middle Ages? Well, the Middle Ages were dominated by philosophers who were Christians, Jews, or Muslims, and they all took normal sense perception to be reliable. That’s one thing you can know—things that are self-evident. The evidence of the senses was proven. They all held that. And these guys—Descartes, Spinoza, Leibniz, Locke, Berkeley, Hume, and now Kant—are going to say, "No, we don't. We can't trust normal sense perception because it has these dispositions to appear other than the way things really are, provided it's not optimal conditions for observing them or our observational equipment isn't functioning right." So, they think they're enlightened because the Middle Ages were "dumb enough" to think that ordinary sense perception reveals the world as it is.

Huh? That's not progress; it's regress. So let's look at what the Enlightenment bequeaths to us: There's a real world independent of us. Can we know that? No. Our self is active in gaining knowledge of the world, but does this mean we know the real world? The answer is no. And we don't know much about the self, do we?

Uh, so I'm trying to be fair with this. No, we don't know there is a self. Or if there is, we don't know much about it. Now, it’s hard to think of anything sillier than saying there are no human selves, no minds, no such thing as consciousness. This is what Hume came to: There are just perceptions. Oh yeah? What are they? How can something be a perception unless somebody perceives it? There has to be a perceiver, right? And Hume says, "No, whenever I look into myself, I don't find something called a self. I find some present impression or some past one." But why the heck would you expect to find yourself as one of the perceptions yourself has? That’s like saying, every time a camera takes a picture, it itself should show up in the picture. I mean, these are their assumptions—that minds operate like cameras. I'm not saying they do. I'm saying, given their own assumptions, these are silly conclusions to draw. There is a self, and there's certainly no truth to their certainty about God.

Now, to be fair about the idea of God, I want to say that Descartes tried to prove God existed. So did Locke, so did Berkeley, so did Spinoza, and so did Leibniz. It’s Hume and Kant who don't, but they don't leave the matter entirely in the dark. Eventually, according to Hume—and I could read this passage to you, but it's impressive—Hume says that the arguments fail, but that should drive a person to consult revelation. And by "revelation," he means the Scriptures.

And that's a very remarkable statement from somebody you wouldn't expect it from. You'd never think Hume would say a thing like that. Look, you go back and forth and back and forth. You're agonized. You look at all the arguments; they're all inconclusive. Maybe it's trying to show you that if you're wise, you go back to Revelation, which is the real source of the knowledge of God, not our own reasoning. Now, I can applaud that one. I think he's right—that's Hume. Unfortunately, it is not Kant. For Kant, God is an idea that cannot be proven but is useful in life, and that's where he's going to leave it. The idea of God has heuristic value. That means we live as though there were a God, but we don't know there is. That's what he's recommending.

In the first place, I don't think it's possible to do that, and I don't think it's desirable. You know God, or you don't. And if you haven't met God in your own experience—especially through Scripture and His Word, reading His Word—there's no other way you're going to know God and know that God is real. It only comes about that way, the way Hume recommended at the end. So, next time, we will launch into Kant proper, and we will see what he does with the analytic-synthetic distinction, how he tries to show that there's a third kind of statement: synthetic a priori. Then, he wants to go from there—if there are law statements, statements about reality, that means the sensations in our heads, remember—that are necessary truths by which we can get science.

I'm going to close with a sort of amusing story that really happened. When I taught at Rutgers, there was a history professor who had an office not too far from mine, and he told me this. He said, "All through my undergraduate years and then again in graduate school, I found that when I met the people working in philosophy, they were all these super-bright types. I was really impressed, and I wondered why it was so. So, I wondered if there must be something to philosophy. I'll find out." So he made an experiment for himself. His experiment was to go to the library, pick out a major work in philosophy, and read it cover to cover to see if there was anything in it that was interesting and important. And what did he pick? He picked Kant's Critique of Pure Reason, one of the most difficult things to read that has ever been penned on this planet. And of course, he didn't know what Hume had already done or said. And he comes across a sentence like this: "I'm writing this book," Kant says, "to show how a priori synthetic judgments are possible." And this poor guy goes, "What?" And he tries another sentence, and another sentence, and it's all gibberish. So he said to me, "So I put the book back, and I never thought about it again. There's nothing."