I have long cited David Kelley’s great 1980s lectures, the Foundations of Knowledge lecture series. It used to be sold on audiocassette by Laissez-Faire Books, as I recall; now on Youtube. These were recorded at various conferences in the 1980s, e.g. The Jefferson School Conference, San Diego, 1985; Portland Institute Conference, 1986. They include:
1. The Primacy of Existence
2. The Epistemology of Perception
3 and 4. Universals and Induction
5. Skepticism
6. The Nature of Free Will
Their transcripts, for reference and searching, are below.
Introduction
0:00
[Music] Welcome to the Principal Source audio presentation of The Primacy of Existence by David Kelley. A cardinal principle of Objectivist metaphysics is the primacy of existence, which asserts that there is one objective world which exists independently of consciousness. This fact is irreducible and does not have to be proved, because existence is the fundamental self-evident truth that lies at the foundation of all knowledge. The primacy of existence is a corollary of three basic axioms of Objectivist metaphysics and epistemology: the axiom of existence, the axiom of identity, and the axiom of consciousness. Objectivism holds that each of these axioms is implicit in any act of knowledge. Thus the primacy of existence is a fact contained within the truths of these philosophical axioms.
1:00
Citing Galt’s speech in Atlas Shrugged and the title essay in For the New Intellectual, Dr. Kelley explains why many philosophers have denied the primacy of existence and embraced the primacy of consciousness in what he terms the diaphanous model, and he points out the essential fallacies in this model of awareness. A well-known philosopher, teacher and writer, David Kelley earned his Ph.D. in philosophy from Princeton University in 1975. He has since taught philosophy, cognitive sciences and other courses at Vassar College and Brandeis University. Dr. Kelley’s books include The Evidence of the Senses, The Art of Reasoning, and A Life of One’s Own: Individual Rights and the Welfare State. He was recently featured on John Stossel’s ABC TV news special Greed and is a popular lecturer and talk-show guest. Dr. Kelley is founder and, since 1990, the executive director of the Institute for Objectivist Studies in Poughkeepsie, New York. This lecture was given in 1985 at the Jefferson School Conference, where Dr. Kelley presented a lucid defense of Objectivism’s standard of the primacy of existence, thus clearing the way for a rational theory of knowledge.
Walking into the Classroom
2:24
At some point within the last 15 minutes or so, probably the last five, every one of you walked into the classroom. You probably took a glance around to orient yourself. You saw a room with the podium in front. You saw seats arranged in rows. You saw the blackboards, the walls, the ceiling with all the junk on it. You saw people. You recognized some of them. And finally you chose a seat and went and sat down in it. Now you probably did all this pretty much without thinking about it. You didn’t really need to think about it. This is not a high-demand cognitive task. You’ve done it a thousand times before. When you saw someone you recognized, the recognition was automatic and immediate, and you assumed it was accurate. You probably did not entertain the hypothesis that the person you saw was a robot or an impostor, and you didn’t ask yourself what reason you had for thinking that such a hypothesis might be false.
3:27
When you took in the room at a glance you probably were not consumed by anxiety about whether your senses were reliable. You didn’t ask yourself whether the objects in your visual field—the whole array of tables, chairs, podium, etc.—whether that whole array might be just subjective appearances. You did not consider in your minds any of the following hypotheses: that you were really home in bed dreaming; that you were having a psychotic hallucination; that you were the victim of an evil demon; or that you were really a brain in a vat detached from the rest of your body, with electrodes stimulating your sensory cortex and scientists manipulating the electrodes to give you the experience of being in a room. Maybe you were thinking that. I can’t look inside your mind, but I would put a certain amount of money on the proposition that you weren’t.
4:21
When you took your seats a few minutes ago you probably looked briefly at the chair to make sure there were no visible defects and you sat down. You did it on the assumption that this chair would behave as other chairs behave—other seats in this case, since they’re connected—that it would sit there quietly supporting your weight. You didn’t worry that your past experience with chairs might be no guide to the present or the future. You did not entertain the hypothesis that causality might be an arbitrary human construct that we impose on reality and that the causal relation between the chair’s nature and its actions might change randomly and unpredictably. So you sat in perfect confidence that the seat would support you. You were not worried that it might do something radically different instead: that it might put its arms around you, or rise up ten feet in the air, or suddenly start complaining that its feet were getting tired.
5:17
In short, when you entered the room you assumed that your faculties were functioning as tools of cognition and that they were doing business with an objective reality in which objects exist and have an identity subject to causal law independently of your perception of them. In this respect you’re perfectly normal.
What Most Philosophers Say
5:37
Most philosophers would say you’re wrong. It’s not that the philosophers believe that our senses always deceive us or that there is no objective reality and the world we experience is only an illusion. Some philosophers believe that, but I don’t think it’s the majority view today. For one thing it’s too systematic and principled. So it’s not that they believe you were wrong to believe what you did. Okay? They think you were wrong to take it for granted as something not requiring proof. Most philosophers today hold that the existence of an objective world and the reliability of our senses are problematic. They are conclusions that have to be established by proof or argument. All the hypotheses that I mentioned, from the brain in the vat to the chair rising up ten feet in the air—all of these are considered to be logical possibilities that you must consider and refute before you can rationally have confidence in your senses. In other words, most philosophers today are not skeptics themselves, but they view skepticism as a logically coherent position that has to be considered and refuted by offering evidence against it.
The Primacy of Existence
6:50
Now Objectivism rejects this approach entirely. A cardinal principle of Objectivist metaphysics and epistemology is the primacy of existence. The primacy of existence says, in effect: you can take it for granted. There is an objective world. Our cognitive faculties are capable of identifying this world. And these are axiomatic truths. The primacy of existence cannot be validated by proof because it lies at the foundation of all proof, and it does not have to be validated by proof because it is self-evident.
7:22
This afternoon I’m going to talk about the primacy of existence: what it means and how it is validated. Next time, on Friday, I’ll talk about some more specific applications to sense perception. Now I assume that most of you have some familiarity with the basic concept of the primacy of existence from Ayn Rand’s writings or from Dr. Peikoff’s lectures, perhaps also from an essay I wrote several years ago in The Objectivist Forum. So while I am going to cover the basics I’m also going to discuss some of the further implications and refinements and historical connections and so on. I also assume that I don’t have to convince this audience how important the primacy of existence is and how pertinent it is to a conference on the intellectual foundations of a free society. Every assault on freedom has begun with an assault on reason, and every assault on reason has begun by attacking man’s confidence that his mind, operating on the basis of sense perception, is capable of grasping reality objectively. If a culture abandons the primacy of existence as an intellectual foundation it will also in due course lose its freedom too. So without any further preface let’s get to work.
Two Components of the Thesis
8:40
First, what does the primacy of existence mean? What does it say as a philosophical thesis? What it says as a philosophical thesis can be broken down into two main components. I’m going to formulate these as separate assertions, but I want to stress at the outset that these separate assertions cannot be divorced. There are really two aspects or two perspectives on the same basic fact. And the first assertion is that existence exists independently of consciousness. There’s a world out there. It exists and it is what it is regardless of our thoughts, our wishes or our fears about it, regardless of our minds. Okay? Facts are facts. Facts are not malleable. They don’t depend on our recognition or our approval.
9:33
The second point is that consciousness is not independent of existence. It’s dependent. Consciousness is primarily an outward focus on the world. All the contents of our minds—our perceptual experience, our thoughts and ideas—all of these derive in one way or another from some cognitive contact that we’ve had with reality. Consciousness is outward-directed. It always involves a relationship to what is external to itself. Okay? So those are the two main points.
Objection: Self-Awareness
10:10
Now right away I hear you murmuring to yourselves with objections, so let me address those. Objection number one: what about self-awareness? When I introspect, my awareness is inner-directed, not outer-directed. Well that’s true enough, and the primacy of existence does not deny that. But let’s look at what happens when you introspect. Suppose you’re monitoring an emotion, say a feeling of anger. Well if you’re angry you must be angry at someone or about something. That is, your anger is outer-directed. Or suppose you are introspectively examining your beliefs. You’re asking yourself what you really believe. A belief has to be a belief about something—about some fact or issue or person or whatever. In other words, introspection is inner-directed, but it involves a focus on conscious states that are themselves directed outward at reality. And without those primary states directed outward there would be nothing to introspect on. You would look inward and find nothing. Okay? So self-consciousness presupposes some prior consciousness of the world, and that’s all that the primacy of existence says.
Objection: Error, Fantasy, and Delusion
11:17
Objection number two: how can I say that consciousness depends on reality for its contents when the contents of some people’s minds—and let’s face it, our own sometimes—are out of touch with reality? Okay? There are fantasies and delusions and self-deceptions. There are false beliefs. And even beyond falsity there’s a kind of mental mush, conceptual hash, that we observe when politicians open their mouths. All this is true again. Consciousness is active in processing the contents that it derives from reality. In this respect it can misfunction. It can make mistakes. It can make a hash out of things. We can organize and rearrange the material provided by our senses just as in the material realm technology can structure matter to serve our purposes. But just as technology can’t create matter out of nothing, consciousness cannot originate its own contents out of nothing. What the primacy of existence says is that at a fundamental level the mind is metaphysically passive. Until it responds to what is outside it, it has no contents to work with.
12:34
So the primacy of existence says that existence exists independently of consciousness. Consciousness is identification. Its function is to identify what exists outside it. In the traditional terminology of the subject–object relationship, which the philosophers in the room will be familiar with, we can put it this way: the relation between the objects of knowledge and the conscious subject of knowledge is not a relation of equals. Okay? The object is primary. It sets the terms of cognition, and the role of the subject is to make his thoughts conform to the object. So for example a belief is true if it identifies a fact that exists out in the world.
13:16
The primacy of existence also means that at a fundamental level there can be no question or doubt about the validity of consciousness. Now it’s true, as I said, that in the conceptual realm we can make mistakes. We can arrive at wrong conclusions. That’s why we need epistemology: to give us some guideline for avoiding such conclusions. But because consciousness is metaphysically passive it is not possible for it to be fundamentally out of touch with the world. In particular, as we’ll see next time in some detail, there can be no such thing as an invalid or false perception. So if the primacy of existence is true then we can dismiss the skeptical notion that perceptual experience might be a complete illusion. That notion is a metaphysical impossibility if the primacy of existence is true. Remember I’m just presenting to you the thesis now. I’m not validating it yet.
Realism versus Idealism
14:07
In philosophy, people who defend the primacy of existence are known as realists, for obvious reasons: they believe in reality. Although you have to be somewhat careful with this term because it’s not used consistently. But when I use it I mean those who adopt the primacy of existence. And the classical example of a realist would be Aristotle. By way of contrast let’s look now at the opposing doctrine, which is called the primacy of consciousness.
14:38
The primacy of consciousness can be formulated in terms of the same two issues that I used to define the primacy of existence. First it says that reality is dependent on the mind. The world is created, or at least it’s shaped and given identity, by consciousness. The objects of knowledge depend on the subject. They depend on our thoughts, on our language, on us in one way or another. And secondly the primacy of consciousness says that consciousness is not dependent on anything outside itself. It is metaphysically active. It is capable of originating all its own contents. So on this view it’s consciousness that has the upper hand in the subject–object relationship. Consciousness sets the terms for cognition and the objects have to conform to consciousness. Proponents of this view in philosophy are called idealists, again for obvious reasons: the primacy of ideas.
15:41
Now most people find it hard to believe that anyone actually believes the primacy of consciousness. I mean the whole idea is outrageous. We can all see how someone might embrace the doctrine implicitly. After all, if someone evades an unpleasant fact, what he’s doing implicitly is saying the fact won’t exist if I don’t recognize it. But who would defend the doctrine explicitly? Well, just to convince you that I’m not making this up, let me quote from the most profound and consistent advocate of the primacy of consciousness, who was Immanuel Kant.
16:15
“Hitherto it has been supposed,” Kant says in his major work, “that all our knowledge must conform to the objects.” But he argues that under that supposition every effort to establish the validity of consciousness has failed. So, quote: “The experiment therefore ought to be made whether we should not succeed better with the problems of metaphysics by assuming that the objects must conform to our mode of cognition.” End quote. Kant in fact is the father of idealism, because although there were other philosophers who adopted the idea in a partial or implicit way before Kant, Kant adopted the idea wholesale. He advocated the primacy of consciousness as a clear and explicit doctrine. And in the two centuries since he wrote, his followers have adopted the idea that the objects conform to the mode of cognition in a thousand different forms and varieties. Today, for example, I would say that the most prevalent version is cultural relativism, which says that people in different cultures live in different realities. And they don’t mean by different realities like different realities we sometimes say. They mean literally different realities—realities that are shaped by the languages they speak or the customs, mores, their conceptual schemes or worldview or whatever.
Skepticism versus Idealism
17:47
Now at several points along the way I have mentioned skepticism. Where does the skeptic fit in this picture? Doesn’t the skeptic endorse the primacy of consciousness? So what’s the difference between skepticism and idealism? Okay, it’s a good question. Well there is a difference, and to see what it is consider again the two aspects of the primacy of existence that I used in defining it. A realist says, first of all, that existence, the object of knowledge, exists independently of the mind. That’s point one. Point two is that consciousness is metaphysically passive. It can’t create its own contents and can’t be fundamentally out of touch with reality. The primacy of consciousness denies both of those points. So the idealist is a polar opposite of the realist. But the skeptic doesn’t go that far. The skeptic denies point two—that is, he says consciousness could create all of its own contents. And all the skeptical hypotheses that I mentioned earlier—that you could be dreaming, that you could be hallucinating, that the world might not be the way it seems—all of these are based on the assumption that consciousness could have originated the contents of your experience out of itself. Right? But the skeptic does not go on, as the idealist does, to deny point one. On point one the skeptic sides with realism. Okay? This is the external independent existence of a world. Of course the skeptic claims that we don’t know and can’t know for sure whether there is an independent objective reality out there. Okay? Which is point one. He says we can’t know that for sure. But he assumes that if consciousness were to be valid it would have to be capable of grasping such an independent reality. In other words he accepts the realist standard for the validity of consciousness even though he doesn’t think consciousness can meet that standard.
19:38
Okay? So is this distinction clear? The skeptic is trying to split the difference between primacy of existence and primacy of consciousness, and therefore the skeptic is not quite the same thing as the idealist. However, having said that I have to say also—remember I said earlier you can’t divorce the two points; you can’t, not really. Skepticism always does lead eventually to idealism. For if people believe that on the primacy-of-existence standard knowledge is impossible, then they will be ripe for a philosopher who urges them to reject that standard and embrace the primacy of consciousness all the way. Right? And if that—there’s a certain reasonableness to doing so. Once your standards get out of kilter with your belief about consciousness, it’s not unreasonable to change the standards. So in the history of philosophy the primary example of this is that David Hume, the skeptic of the 18th century, prepared the way for Kant the idealist.
The Driving Analogy
20:45
Maybe to help keep some of these relationships straight it would help if I give you an analogy. Suppose there are three people driving a car down a road. Okay? One’s a realist, one’s a skeptic, one’s an idealist. The realist is driving, and he’s driving the way you and I drive. He looks at the road and he steers accordingly. But the skeptic is worried. “What if the road made a turn and you didn’t see it?” he says. “How can you be sure you’re steering the car right? People have accidents, you know. How can you be sure we won’t go in the ditch? How can you be sure we haven’t already gone in a ditch and we’re all just hallucinating?” Then the idealist joins the conversation. His name is Kant, and he says: “Hitherto it has been supposed that our steering must conform to the road. But on this supposition it has proved impossible to establish the validity of our steering. The experiment therefore ought to be made whether we should not have more success with the problems of driving by assuming that the road must conform to our steering.”
21:53
Thank you.
Validation: Axioms and Corollary
22:00
Now despite my recent dig at the skeptics and idealists I haven’t really refuted their position, nor have I validated the primacy of existence. What I’ve done is to explain what the doctrines say. So let’s turn now to the issue of validation. I said before that the primacy of existence is axiomatic. It’s a fundamental self-evident truth that lies at the base of all our knowledge. But it’s not strictly speaking an axiom itself. It’s a corollary of the three basic axioms of Objectivist metaphysics and epistemology. And these are—I hope these are familiar to you—first, the axiom of existence: that something exists, that existence exists, that what is, is. These are different ways of formulating the primary basic unanalyzable fact of existence. Second, the axiom of identity: that whatever is, is something specific. A thing has an identity, a definite determinate identity. It is what it is. A is A. And third, the axiom of consciousness: that consciousness exists and that one is conscious.
23:18
Now Objectivism holds that each of these axioms is implicit in any act of knowledge, from the simplest sensation to the theory of relativity. There’s always something that we know—that is, something must exist to be known. There must be something to be known about it—that is, it must have some identity. And of course our knowing it reflects the fact that we’re conscious. The primacy of existence is a corollary of these facts. It’s not a separate fact that is derived from them by any process of proof. This is not a deductive system like in geometry here that we’re talking about. The primacy of existence is a restatement of what is implicit in these facts, the axiomatic facts, when we examine them as a totality, when we take them as a single package.
24:07
Consider the two facts of consciousness and of existence. To be conscious is to be conscious of something. Consciousness is inherently relational. It’s always directed on something that is external to itself. Okay? Despite the claims of Transcendental Meditation, there’s no such thing as pure consciousness devoid of all content, and to be conscious of nothing would be to be unconscious. Okay? All of this is implicit in the axiom of consciousness. It’s part of what we grasp when we grasp that we are conscious. What are we conscious of then? Another state of consciousness? Well then that state must be the consciousness of something, as in introspection. At some point consciousness must be conscious of something outside itself. You can’t have an infinite regress of consciousness of consciousness of consciousness, etc. Okay? In this respect then the primacy of existence expresses the relationship between the primary facts of existence and consciousness.
25:09
Furthermore, something which exists must have some identity. It has to be something specific. So the awareness of something which exists must be the awareness of at least some aspect of its identity. Now suppose someone agreed that a thing’s existence does not depend on consciousness but claimed that consciousness is the source of its identity. That notion would violate the axiom of identity. A thing is what it is, not what consciousness makes out of it. If our minds created the identity of their objects, then those objects would not have an identity prior to our being conscious of them. They would exist—remember we’ve agreed to that, they exist—but they would not be anything in particular. And that’s impossible. To be is to be something. So in this respect the primacy of existence reflects the relationship between the primary facts of consciousness and identity, and it can be expressed in the statement: consciousness is identification. Okay? In short, if we understand the axioms we can see that the primacy of existence is not a new fact but something already implicit in them or contained in them. If we know that the axioms are true, therefore, then we’ll know that the primacy of existence is true. So the logical question is: how do we know that the axioms are true?
Fundamentaity of Axioms
26:35
In this context it’s important to grasp two features of a philosophical axiom: first that it identifies a fundamental fact, and second that it is self-evident. I want to look at each of these features in turn: fundamentality and self-evidence. Okay. Fundamentality. An axiom identifies a fundamental fact. The facts of existence, identity and consciousness are primary. They’re not analyzable in terms of other facts or other concepts. They are the epistemological bedrock. They lie at the base of all our knowledge. They are implicit in and presupposed by all the rest of our knowledge. So they can’t be derived from prior knowledge by any process of proof, because there is no prior knowledge. There is no knowledge prior to them, and any effort to prove them would be circular.
27:26
How, for example, would you prove that something exists? What evidence would you cite? Well, whatever it is, it would have to be evidence that exists. It would have to be something that exists. Otherwise it wouldn’t be valid evidence. Okay? So you’d be begging the question by trying to prove that something exists. How would you prove that you’re conscious? What prior knowledge would you appeal to in support of the conclusion that you’re conscious? Well, prior knowledge is a form of consciousness, and therefore prior knowledge would already contain the fact of consciousness, and you’d be presupposing what you set out to prove. Okay? Something’s got to be at the base of knowledge, and this is it.
28:04
Now the same is true of the primacy of existence. Because it is implicit in the axioms it too lies at the base of all knowledge and therefore can’t be derived by proof from any prior knowledge. Now in a moment I’ll show you how we can establish the axioms and the primacy of existence as true. But I do want to pause for a moment on this property of fundamentality: the fact that the primacy of existence is implicit in and presupposed by everything else that we know. It’s implicit in any activity of consciousness whatever. It is inescapable.
Inescapability: Error and Self-Refutation
28:41
Now I want to illustrate this inescapable character to you. Obviously I can’t survey each and every act of consciousness that has ever occurred in anyone’s mind, but I do want to take two cases where you might think the primacy of existence is least likely to be implicit. So if I can convince you that it’s implicit there, then you can see that it will be implicit across the board. Okay? First case: cognitive error, in the case of a false belief that somebody has. In any such case it might seem consciousness has departed from reality and originated its own content. For example, let’s take an actual example of a false belief. Suppose someone believed that Benito Mussolini was a pitcher for the Dodgers. This is not true. There’s a certain failure of correspondence here between consciousness and reality. So in what way is the primacy of existence implicit in that belief? Well, to start with, the name Benito Mussolini refers to an actually existing person, historically existing person, and the name Dodgers to an actually existing baseball team. So the person has not made up the content of his consciousness completely. He has mistakenly combined contents that he derived from some kind of contact with reality. He hasn’t made up the content of his consciousness out of whole cloth, so to speak.
30:15
Now try to imagine a statement that did not contain, did not reflect any contact with reality whatever. It couldn’t contain any familiar words, because those express concepts that we derive from our perception of the world, that we derived by somebody’s perception. Such a statement could not contain invented words that had any private meaning for the person in terms of his own perception, right, or in terms of memories, images or even fantasies that were derived from his perceptions. Right? If you follow this through you’d end up with a statement that had no meaning whatever, in which case we’re not really dealing with a statement or anything cognitive. We’re dealing with physical noise.
30:59
Secondly—even though this is not the second example, it’s the second point about this Benito Mussolini belief—even though this person’s belief is false, he asserts it as something that is true independently of him. He was trying to grasp Mussolini’s real objective identity even though he didn’t succeed. Okay? In general any statement, any belief or assertion, is intended as an identification of a fact in reality, a fact that exists independently of the speaker. Implicit in any statement whatever is the statement: this is how things are, really.
31:39
And as a kind of ultimate example of this fact—and this is the second illustration—we should notice that the primacy of existence is implicit even in the attempt to deny it. That is, anyone who denied the primacy of existence, anyone who asserted the primacy of consciousness, would contradict himself. How? Well let’s work this through. An idealist claims that what is known depends on our knowledge of it. Does he claim to know that? Of course. That means that from his own view the truth of his thesis would depend on his believing in it. Okay? But that’s clearly not what he means. That’s not what idealists are trying to say. When idealists claim that the objects of knowledge depend on consciousness, they’re asserting this as a fact about consciousness and its objects. They’re asserting it as something which is true because of the objective nature of consciousness, not because they happen to believe it. Otherwise they’d have to allow that the primacy of existence is true for those who believe in that. Realism is true for realists. But of course Kant and his fellow idealists don’t want to allow for that. They claim that the primacy of existence is false. Okay? And when they say it’s false they mean it doesn’t correspond to the fact, the real nature of consciousness and its objects. So someone who asserts the primacy of consciousness is asserting that as a fact he did not create, even though what he’s asserting is there are no such facts. So he’s contradicting himself.
33:14
Now it’s important to emphasize that the argument I just gave is not a refutation of the primacy of consciousness, nor is it a proof of the primacy of existence. Why not? Well because I was assuming the primacy of existence myself. It was implicit in every one of my assertions just as much as it was in every one of theirs. And so if I offered this as an argument for the primacy of existence someone could accuse me of, and reasonably accuse me of, arguing in a circle. For example, to make this a little more concrete, what I just did was to show you that it is self-contradictory to assert the primacy of consciousness. But so what? Okay? A contradiction is not objectionable unless we already accept the primacy of existence and the law of identity. Okay? I want to stress the point: the primacy of existence cannot be validated by proof or argument, including the kind of argument that I just gave, the sort of self-refutation argument. The argument does not show that the primacy of existence is true. It only shows that it is fundamental. It’s implicit in all the rest of our knowledge.
Self-Evidence versus Proof
34:28
Well, having stressed all of that, I’m sure the question you want to ask now is: if the axioms can’t be proved, then how do we know that they’re true? You may have noticed that I keep putting this question off. That’s a standard philosophical maneuver, but I don’t want to keep you in suspense any longer. Proof is only one form of validation. Validation is the broader concept; that’s the genus, and proof is one species of it. There are others. Okay? The axioms can be validated, which brings us to the second of the two features that I mentioned earlier. The first was fundamentality. The second is self-evidence. The axioms are self-evident, which means we are directly aware—we directly perceive—the facts which the axioms identify. We directly perceive the facts of existence, identity and consciousness.
Rationalism versus Perception
35:24
Now a brief historical digression. This view of self-evidence, that it is based on perception, is distinctive to the Aristotelian tradition in philosophy. It was rejected by the modern philosophers from Descartes to Kant, especially by the rationalists, and it’s almost completely unknown to contemporary philosophers. If you suggested this to a contemporary philosopher, his eyes would boggle. He wouldn’t know what you’re talking about. The rationalists believe that there are axioms such as the law of identity, and they believe that they’re self-evident. But a self-evident truth for the rationalist was a self-contained proposition that made some kind of direct appeal to the intellect, an appeal that didn’t go through the senses at all, was not based on perceptual experience. In the ancient world, for example, Plato, who was a rationalist—Plato thought that we grasped these propositions by some sort of mystical intuition or revelation. The modern philosophers such as Descartes typically held that knowledge of the self-evident truths was innate in the mind at birth. There was no revelation. It was just there, stocked in the mind from the outset. Okay? In either case, whether mystical revelation or innate ideas, in either case rationalists considered a self-evident truth to be a priori, which is a Latin phrase that means it’s known by the mind prior to any perceptual contact with reality, prior to and independent of any perception. And this is still what contemporary philosophers think of when they hear the term “self-evident,” and that’s one of the reasons they don’t think there are any self-evident truths, at least most people don’t.
37:10
Now Objectivism rejects this whole approach, the whole rationalist approach. The intellect, that is reason, cannot know anything prior to perception. Every attempt to deny this leads ultimately to mysticism on the one hand or to subjectivism on the other hand. Let me just briefly indicate why. If the axioms are not known through perception then they must be known through some nonsensory mystical form of contact with reality, some nonsensory perception or extrasensory perception. Now that was Plato’s type of approach. Or the other alternative is to say that they are subjective human constructs with no basis in reality. Okay? They’re just—one example of this view would be the innate-ideas idea: that they’re there not because of any contact with reality but just because they’re there. Another example would be they’re implicit in our language and they’re subjective social conventions. So that’s the subjective side. So you’re stuck with either mysticism or subjectivism if you deny that the axioms are validated through perception. Okay? So the Objectivist view is that the axioms are validated through perception.
How Perception Validates Axioms
38:24
But the question is how. When we perceive we’re aware of a concrete specific object, a single individual object, whereas the axioms are highly abstract. The concepts of existence and identity embrace everything, and the concept of consciousness embraces every type of consciousness there is. Now this is part of the reason for the rationalist view of axioms. The perception of a single object on the one hand, like this podium, and the grasp of a principle subsuming everything in the universe seem to lie at opposite ends of the cognitive scale. So how can such a principle be validated merely by perceiving a single object? To put it crudely it seems like a wild overgeneralization. Okay? Well, to answer the question, to deal with the problem, let’s see how it works by looking at a specific example of perception.
39:22
Objectivism holds that the axioms identify facts that are implicit in any percept, any perception. Everything there is to be known about these facts—about existence, identity and consciousness—is available in a single glance at reality. So let’s all glance at reality. Look at the podium. Well actually consider all the things you’re seeing, because that’s part of the point. Okay? The people in front of you, the podium, me, the walls, the ceiling—everything. Now what you see is a lot of different objects, and you can identify the specific nature and properties of what you see in terms of concepts for perceptual concretes. Okay? Your visual awareness tells you that there are people here, that the podium is rectangular in shape, that the wall is brown, I guess. These are called perceptual judgments. The judgments I just expressed—these judgments express in conceptual form the content of perception, and they’re self-evident. You know that they’re true not by inference or proof but by direct observation of the facts they identify. If someone doesn’t understand why you believe that the wall is brown, there’s only a limited extent to which you can argue with him or provide evidence. At some point you have to say, well look, just look. I can’t give you my perception. You have to see for yourself, and then you will see that it’s brown.
40:51
Okay. Now these judgments identify aspects of the things you see in regard to which those things differ from other objects in the world. Okay? When you observe that the wall is brown you’re distinguishing it from walls of any other color, and when you identify it as a wall in the first place you’re distinguishing it from any other kind of object. So the perceptual judgments you can make in any given context depend on the specific nature of the objects you’re perceiving in that context. But there are other aspects of what you see that would not be different in a different context. They would be the same no matter what you’re perceiving.
41:22
Right now look at the podium again. This podium differs in a thousand ways from other objects that you might be perceiving. Okay? But one way in which it doesn’t differ is that it’s here. It’s a presence before you. It’s in your visual field. There’s no way to express it except to say it’s there. It exists. Okay? I could take it away and put something else in its place and you would have a different perception, but whatever I put in its place it too would have to be there. It would have to exist. Whatever you perceive, whether it’s a solid object like this podium or a nonsolid object like cigarette smoke, or an event like my raising my arm, or a relationship like the fact that I’m taller than the podium—no matter what kind of existent you perceive, you can say of it: it is. And that single word “is” describes the common invariant fact of existence. And that fact is available in any perception.
Identity in Perception
42:28
Now let’s turn to another aspect of what you see. There are many perceptual judgments you could make about this podium. Okay? That it’s tan, that it’s rectangular in shape, that indeed it’s a podium. If I took it away and put some other object in its place you would make a different set of perceptual judgments. But now suppose that I put an object here about which you could not make any perceptual judgment whatever. Okay? It isn’t a podium, but you can’t identify it as anything else either. And I don’t just mean it’s an unfamiliar object for which you don’t have a concept. I mean—because in that case you could at least notice some of its properties—I mean something that has no properties. It isn’t tan but it isn’t any other color either. It isn’t rectangular but it doesn’t have any other shape. It just doesn’t have a shape. You can see that this is impossible. Whatever the specific properties of the thing you perceive, it must have some properties. Okay? It must be something. And this is the fact of identity. And this fact too is available in any instance of perception.
43:33
Let me digress for a moment. There’s another way that you might get at this point about identity. Whenever you perceive—and I would say in any act of consciousness whatever—you are differentiating something from something else. In this case you’re differentiating the podium from its background. I’m going to talk a lot next time about the nature of that differentiation. But we can see already that any awareness involves the awareness of difference. But for there to be differences there must be two things that are different—that is, that differ in identity. So again, because perception always involves differentiation or discrimination, it always involves the awareness of identity.
Consciousness in Perception
44:17
Finally turn your attention to your own awareness of these objects—that is, shift from the objects to your perceptual experience of them. Certain aspects of your awareness are specific to this particular context. Those of you who are nearsighted, for example, may be perceiving with less acuity than those with 20/20 vision. You’re aware of less. You’re resolving less detail. Those of you on the right have a different perspective on the podium from those of you on the left. But whatever the specific nature of your awareness, you are aware. There’s a you in there which you identify by saying “I.” If you’re conscious you can all say “I.” There’s a you in there. There’s an it out here, whatever it might be. And there’s a relationship between you and it. You’re aware of it. That would be true in any perceptual context, no matter whether your vision is 20/20 or something else, no matter how clear or blurry your vision, no matter what the specific properties of your awareness are. There’s a you and it and you’re aware of it. That’s the fact of consciousness.
45:27
Now you see what we’ve done. In each case we started with perceptual judgments about these specific facts of the situation. These judgments were self-evident because we directly perceive the facts, like the color of the podium. Then we notice that contained within those facts are certain other ones. And these facts—that something exists, that it has identity, that you’re conscious—are embodied in every other fact about the situation, and they would be present in any other situation as well. Precisely for that reason everything there is to be known about them, in a sense, is available in this situation. You can imagine a circumstance in which you might be mistaken that this is a podium. I mean you all have adequate evidence now. There’s no reason to doubt your senses, as I’ll prove that to you next time. But you could all imagine a circumstance in which you might have some reason to doubt whether this is a podium. But there’s no chance that you would ever have a reason to doubt that something exists, or that what exists has identity, or that you’re conscious, because any ground for doubting any of those would itself have to be something that exists, as an identity, and it would have to convince you as a conscious subject. Okay? If you did have reason to doubt that this is a podium there are ways you could acquire further evidence. You could come up here and look, check it out, you could rap it and so forth. But there’s no further evidence to be had that something exists, that what exists has identity, or that you are conscious. So the judgments that identify those facts are supported fully by a single perception. These judgments are the axioms, and they’re self-evident.
The Primacy of Existence as Self-Evident
47:07
All right. Now let’s consider the primacy of existence. It’s just a matter of spelling out what I’ve been saying so far. Because it’s a corollary of the axioms it too is available in any perceptual context and it’s self-evident. If you examine your perception of the room you’re aware of your perception as metaphysically passive. It reveals the things you see. It doesn’t create them. The podium, the walls, the other people, the blackboard—all of these exist in a world that you are in, not a world that is in you, in your mind. Okay? When you turn your head to look at something and you start to have a perception, you don’t experience the onset of your awareness as the onset of the object. You don’t experience the object as jumping into existence. Okay? Try that with a person sitting next to you. Everything you perceive has some identity. It is what it is, not what you make of it. It’s not like imagination, which you can shape to suit your own fancy.
48:05
Now suppose the skeptic comes along and says that perhaps the things you see have no independent existence. Suppose, says the skeptic, that the things you see exist in your own consciousness. This idea is literally unintelligible if we have grasped the fact of consciousness. Why? Well because we grasp this fact by attending to our conscious awareness of the objects in the room. Right? From a first-person internal perspective we focus on our own perceptions. We isolate the fact of consciousness. We isolate our awareness by distinguishing our awareness from the objects that we’re aware of. The awareness is a relationship between ourselves and those objects. Consciousness is that which is not the object but my awareness of the object. Okay? So someone who says that the objects are in consciousness is destroying the very contrast that we used in order to form the concept of consciousness, to grasp the fact of consciousness. Okay? The skeptic is stealing the concept of consciousness by using it to negate the very basis on which we form the concept, namely the recognition of an object as distinct from our awareness of it. Okay? It’s as if we explain the word “female” to someone by contrasting male and female and then that person said, “Yeah but couldn’t males all really be females?” Well no they couldn’t. If you think that you haven’t got—you don’t understand what the two things are. Similarly with consciousness and its objects.
49:45
Okay. The primacy of existence therefore is not only fundamental and inescapable. It is self-evident. I’ve established that it’s true now not by proving it but by directing your attention to the facts which make it true. So we’ve now validated the primacy of existence and we’ve refuted the skeptics, the idealists and all that group.
Why Philosophers Deny It
50:09
Now ordinarily I would consider that a full day’s work. But I’m going to ask you to give me a few more minutes to give you another chapter in this story. It’s a shorter one, I promise. But there’s another point about consciousness that we have to grasp in order to appreciate fully what the primacy of existence says, and also to make any further progress next time in epistemology. Okay. Let’s begin by asking the question: if the primacy of existence is self-evident and if it can’t be denied without self-contradiction, then why has it been denied by so many philosophers? Well the answer lies partly outside epistemology. There are as many motives for denying the primacy of existence as there are for evasion, subjectivism, for any effort to escape the absolutism of reason. And I think one of Ayn Rand’s great achievements was her analysis of those motives in Galt’s speech, in For the New Intellectual, and other writings. Okay? So to a large extent opposition to the primacy of existence comes from outside epistemology. But it also does come to some extent from within epistemology. There’s a certain patch of epistemological fog that allows the opponents of reason to cover their tracks, and unfortunately it also has misled some defenders of the primacy of existence. So it’s especially important that we understand this.
Two Perspectives on Awareness
51:44
Let me start by drawing your attention to a certain phenomenon of consciousness: the phenomenon namely that there are two perspectives we can take in regard to any act of conscious awareness. Let’s take a person. Imagine we’re in a psych class and we’ve got our experimental subject here. Subject is looking at a tree. You have to imagine the tree. Okay? We’re the external observers of this person engaged in perceiving the tree. As external observers we can study certain aspects of his perception. We can study the process by which he perceives. We can follow the causal chain from the reflection of light off the tree to its action on his visual receptors to the impulses traveling to his visual cortex and so on. From this perspective we can study the person’s consciousness in the sense of the capacities, the conscious faculties that give rise to his conscious awareness—in this case the visual capacity. And it’s perfectly clear to us that this capacity has a definite identity and that it operates in a specific manner which is determined by its nature.
52:51
But there’s another perspective: that of the person himself, our subject. As a conscious subject he directly experiences his perceptual awareness of the tree. From his perspective the tree is present to him. It’s revealed as it is. Okay? From his perspective—this internal or so-called first-person perspective—his perception is just the revelation of the tree. Okay? It’s a focus or direction upon the tree as the object of his attention. Okay? Now it’s from this perspective, as conscious subjects ourselves, that we grasp the basic fact of consciousness. Okay? And therefore the most, in a sense the most accurate formulation of the axiom of consciousness would be: I am conscious.
53:39
Now these two perspectives are quite different. In particular neither one allows any direct access to the fact that is grasped by the other. Let me explain what I mean. From the external perspective we have no direct access to the perceiver’s conscious awareness. We cannot literally look into his mind. Okay? No matter how thoroughly we understand the nature and operations of his visual apparatus—his eyes, his visual cortex—we don’t directly observe the awareness that it produces. Of course we can know, I can know that someone else is conscious, but I can’t know it directly by looking into his mind. Now I have to infer it from the fact that he’s a human being, that he doesn’t appear to be in a coma, and so forth. It’s only in my own case that I can grasp the fact of consciousness directly. So from the external perspective we can study the faculty of consciousness and how it works, but not the conscious awareness it produces. So we can study the process but not the product.
54:44
On the other hand, in the internal perspective the situation is exactly the opposite. The perceiver has no introspective access to the causal process by which he perceives. He sees the tree. He doesn’t see the means by which he sees the tree. He doesn’t see his own eyes, much less his visual cortex. He can observe his conscious awareness, but not directly the faculty by which it’s produced. So he can study the product but not the process of awareness. Of course he will know abstractly, if he knows anything, that his visual system does have a certain identity and that it’s operating in a certain way. You know your eyes are working. Right? But you don’t directly perceive your visual—you’re not directly aware of your visual system in operation in the way that you are directly aware of the objects that you’re looking at, and our subject here is aware of the tree.
The Diaphanous Model
55:40
Now so far what we have is an interesting phenomenon. It’s no big problem as far as I can see. These are perfectly consistent perspectives. But this is the point at which the confusion begins. Okay? This is where the fog starts rolling in. From the internal perspective it seems as if conscious awareness is transparent or diaphanous. Philosophers have taken the sense of transparency and erected an entire theory of consciousness—or rather a theory of what consciousness must be if it is to be valid, if it is to give us the awareness of objects outside consciousness, outside the mind. We experience the perceptual awareness of an object as a kind of revelation of the object. It reveals the object. Philosophers have therefore assumed that to be valid consciousness must literally be a revelation in which the object alone determines the way it appears to us and our faculty of perception contributes nothing. From this internal perspective we do not observe the process by which our awareness is produced. Therefore philosophers have assumed that to be valid consciousness cannot be produced by any definite process. From the internal perspective we do not directly observe the identity of our conscious faculties—not in the case of perception anyway. Philosophers have therefore concluded that to be valid consciousness cannot have any definite identity.
57:11
Now these inferences are all mistaken, as we’ll see. But in the history of philosophy they’ve been extremely powerful. Most philosophers have accepted them in one form or another in regard to one type of knowledge or another. The result is a certain view of consciousness which I call the diaphanous model, after the notion of transparency or diaphanousness. The view is that in order to be valid consciousness has to be a kind of perfectly reflecting mirror that reflects or reproduces the objects without distortion. Okay? Or to use another metaphor that’s pretty common: it’s the view that to be valid consciousness must be a transparent diaphanous medium, a kind of cognitive see-through fabric. Okay? A fabric that does not in any way affect the appearance of objects.
58:06
What this view comes to in non-metaphorical terms is that the conscious awareness of objects cannot be produced by any process whose nature affects the way objects appear. Okay? That is, the process cannot affect the form or the manner in which we are aware of those objects. And this means that the faculties of consciousness which produce this conscious awareness—the faculties such as the sense organs and the conceptual faculty—cannot have any identity of their own. For if a faculty of consciousness had a specific identity then it would give rise to conscious awareness by means of some specific type of interaction with the object, with the world outside, and the specific nature of that interaction would affect the resulting conscious awareness, the perceptual experience. So if consciousness has a definite nature—in other words I’m still expounding the model here—if consciousness has a definite nature, that nature would be like bubbles in the mirror which distort the reflection. Or to use the other metaphor: if consciousness had a nature then it wouldn’t be a diaphanous medium anymore. It would be a translucent medium, and therefore things would not appear the way they really are.
59:15
This whole view of consciousness can be summed up in a single principle. The principle is: if consciousness has an identity of its own, then it cannot grasp the identities of things outside it. If consciousness has an identity of its own, then it cannot grasp the identities of things outside it, and therefore it can’t function as identification.
The Dilemma: Naïve Realism versus Representationalism
59:41
Now Objectivism rejects this principle at its root and in every manifestation. But before we turn to the Objectivist view let me just indicate briefly why it’s so important to reject it. It’s important because if we accept this principle I just gave you then we’re faced with an immediate dilemma. On the one hand suppose we want to acknowledge the primacy of existence, that consciousness can grasp the identities of things in reality. Then according to the principle we have to deny that consciousness itself has an identity or that it operates by any specific means. So what we’d have to do in perception is to adopt the position that is known as naïve realism, which holds that every feature of an object as it appears in the visual field or the auditory field or whatever—every feature of an object as it appears—belongs to that object intrinsically, apart from us, apart from any interaction with our senses. If an object appears a certain way then it is that way, period. Consciousness contributes nothing. If the sugar tastes sweet, the sweetness is in the sugar as an intrinsic property. If the stick that is half in water looks bent, then it is bent.
1:00:58
Now the last example especially I think suggests that this is not a very attractive view. It’s not one that we’d want to hold. So but we’re forced to it by the principle of the diaphanous model of awareness on the first side of this dilemma that I’m presenting. If we want to preserve the primacy of existence we have to deny that consciousness has any identity of its own that contributes in any way to our awareness of objects. On the other side of the dilemma, if we want to recognize that consciousness does have an identity, then the principle would force us to abandon the primacy of existence. If our visual faculty, for example, does affect the way things look, then we would have to conclude, if we accept this diaphanous model, that we do not really see external things. What we see is a world of images inside our own minds created by consciousness. Okay? Those images may or may not correspond to things outside, but as far as we’re concerned they are equivalent to the images in a dream. This is a view which has a historical name: it’s representationalism. You don’t have to write that down. We’ll talk about it in detail next time.
1:02:11
Okay. In other words the diaphanous model of awareness says we have to make a choice. We can preserve the primacy of existence or we can recognize that consciousness has an identity, but we can’t have both. We cannot have both. Now this is not an acceptable choice. This is not one of life’s little trade-offs like buying a VCR or buying a new stereo, or having dessert or keeping your figure. We can’t deny the primacy of existence and we can’t deny that consciousness does have an identity. And we’ve already seen why the primacy of existence is true. And as for the fact that consciousness has an identity: well, the law of identity says that everything does. Consciousness can’t be exempted from the law of identity. And the law of causality tells us that an effect such as our perceptual awareness of an object must have a cause such as our perceptual faculty, and that the nature of the effect depends on the nature of the cause. Right?
Objectivism: Identity and Activity of Consciousness
1:03:11
So Objectivism rejects this diaphanous model of consciousness and all the mistaken inferences on which it rests. Objectivism holds that consciousness does have an identity like everything else that exists. It’s a faculty possessed by living organisms. It has a specific determinate nature and operates in specific determinate ways like any of their other biological systems. Objectivity begins, as Ayn Rand has said, with the realization that man—including his every attribute and faculty, including his consciousness—is an entity of a specific nature who must act accordingly, and that there is no escape from the law of identity. So consciousness is not a passive mirror of the world. It’s an active process of integration and differentiation. At the perceptual level the process is performed by the brain and it’s automatic. At the conceptual level the process to a large extent is conscious and voluntary. In either case consciousness is epistemologically active, not passive.
1:04:12
Now this doesn’t mean—I’ve used the word passivity before, so let’s get clear exactly what we’re saying. It doesn’t mean that consciousness is metaphysically active. It doesn’t mean consciousness can originate its own contents out of nothing. It only means that our faculties are active in processing that content and that they do so in accordance with their own nature. So there’s no conflict between metaphysical passivity—that is, the primacy of existence—and epistemological activity, the fact that consciousness operates by specific means. To put it in a slightly different way: what we are aware of is determined by reality. There’s nothing else to be aware of. But how we’re aware is determined by our means of awareness. And there’s no conflict between the what and the how. How could there be? It’s true that as conscious subjects we’re not directly aware of the means by which we perceive. But to infer from that that there are no means is to believe in magic. It says the perception occurs by magic. In any case it is a complete non sequitur. If we’re going to perceive it has to be by some definite means. If there’s going to be a what there has to be a how.
1:05:34
Now because we perceive by specific means, by the interaction of our perceptual systems with external objects in certain highly specific ways, an object’s appearance is affected by the nature of that interaction. The way a thing looks is affected by the way our senses perform. We perceive an object in a form that depends on the nature of our own perceptual apparatus, and an organism with a different kind of perceptual apparatus might well perceive the same object in a different form. But that doesn’t alter the fact that it’s the object that you perceive. That’s what you perceive. When you came into the lecture hall this afternoon after being out in the bright sunlight, the room probably looked darker to you than it does now when your eyes have adapted. Okay? So when you came into the room you were perceiving the room in a different form from the one in which you’re perceiving it now. But you were still perceiving the room. Right? The difference in apparent brightness is an aspect of the form of perception. In my next lecture I’m going to talk about this form of perception, and you will learn more than you want to know about it. But for now the important point is that the form pertains to the how. It reflects the nature of our perceptual faculties and is thus an aspect of the way we perceive.
Conclusion: Arbitrary Standard of Validity
1:06:54
Okay. Let me offer one kind of final summary expression of all this by way of conclusion. The diaphanous model of awareness, given everything we’ve said, you can see is analogous to the altruist code in ethics. Altruists invent an arbitrary standard of morality, a standard that is incompatible with man’s nature, and then they denounce man as being sinful and evil because he doesn’t live by that standard. In exactly the same way the diaphanous model has invented an arbitrary standard for the validity of consciousness, a standard that is not compatible with the nature of consciousness. It’s a standard that refuses to acknowledge even that consciousness has a nature, and then it condemns man’s consciousness as invalid and delusory because it doesn’t conform to that standard. In ethics Ayn Rand identified the essential fallacy of altruism and cleared the way for a rational code of ethics. I consider it an equivalent achievement in epistemology that she identified this essential fallacy of the diaphanous model of consciousness. She rejected it at its root in all its forms, and she cleared the way for a rational theory of knowledge and a complete consistent defense of reason. She showed us the road to an objective epistemology. Now next time we’ll take a few steps down that road by looking at some issues in the epistemology of perception. But that’s all for today.
Question Period: Book Publication
1:08:33
Okay. Questions. Yes. I’ve heard many conflicting—when’s your book coming out? Did you tell us? All right. I was going to say something about this next time. As some of you know I’ve written a book called The Evidence of the Senses. It is being published by Louisiana State University Press. It has been copyedited, so most of what can go wrong already has and that’s been taken care of. It is now scheduled for the spring of 1986—that is, next spring—which could mean anywhere from Christmas to Easter, to pick two traditional dates. So that’s about as definite as I can tell you. I will be finding out shortly what the exact schedule is, and perhaps we can publicize it in some forum like The Objectivist Forum. Okay.
Acting on the Primacy of Consciousness
1:09:26
Yeah. Other questions. Yes. Dr. was talking in the previous lecture about the importance of acting on our belief, that action was follow-through. Okay. The question is whether someone could consistently act on the basis of the primacy of consciousness, or whether those who profess that view don’t always really act as realists. Well yes, the point of your question is certainly true. Like any profoundly false philosophical doctrine, it would be impossible I think to act with complete consistency on the primacy of consciousness, because pretty soon, it being false and there being a reality out there, reality would catch up with you. There are however people who go pretty far. Right? In regard to the whole inner realm of one’s emotions and feelings and thoughts, beliefs, character, people can get away with the primacy of consciousness pretty much all the way. In other words the primacy of existence means not just that there are physical objects out here. It means your emotions are what they are whether you’re willing to recognize them for what they are. People can get away with self-denial and believing what they want to believe a long ways there. I grew—when I was in college I was part of the ’60s generation, and there were a lot of my peers who went pretty far even in the external world on the primacy of consciousness. It was a drug culture. People were dancing around the Pentagon chanting to stop war, trying to levitate the Pentagon. I know that’s true. You remember this. And I assume that they believed it. Now if you stopped eating or walked out a window on the assumption that the law of gravity would suspend itself if you stopped believing in it—no, no one could do that. But don’t underestimate the ingenuity of people in finding ways to act on this.
Reality versus Existence
1:11:26
Okay. Let me take a question from here. Yes. What’s the distinction between the two terms “reality” and “existence”? Is reality a more concrete term that refers to like human perspective? Well I should actually ask you to ask Dr. Peikoff about this because I think he’s expressed it more clearly than I can. But existence as it’s normally used means what’s there, what’s out there, what exists, with no reference to us as being aware of it. The term “reality” tends to mean existence insofar as we have grasped or identified it. Right? So in the vernacular people say that I kind of knew it was true but it wasn’t real to me. Right? That doesn’t necessarily—it’s not necessarily an expression of the primacy of consciousness. It may just mean I was aware that it existed but I wasn’t fully aware of its existence. It wasn’t fully real. So but in most contexts I use them interchangeably.
Primary and Secondary Qualities
1:12:34
Yes. Do you believe there’s a distinction between primary and secondary qualities, or there’s no distinction? This is an extremely detailed question about a point of detail in the theory of perception. There’s a traditional distinction in philosophy between so-called primary qualities like shape, size, number—how many things are there—motion through space on the one hand, and secondary qualities which would be colors, taste sensations, feeling of warmth and coolness, auditory sensations like pitch. I think there is a certain distinction you can draw, but it’s not anything like the traditional one. The traditional distinction was that primary qualities are out there in reality and secondary qualities are in the mind. And I think, as I will explain in some detail next time, Objectivism would reject that whole idea. In my book I talk about this issue, and there’s a specific point that involves a relation between sensation and perception. Secondary qualities tend to be qualities that could be present in a single sensation—okay?—sensation of warmth, an isolated quality. The primary qualities of shape and size tend to be qualities that we grasp when we grasp objects as wholes. Right? Because you’ve got to see this as an entity to see the size or the shape. And that leads to certain differences in how our senses interact with it. I think beyond that I’d have to go into too much physiological detail, but I’d be glad to talk about it further with you later.
Other Minds and Androids
1:14:31
Okay. Yes. At the beginning of your speech you mentioned something about how we see other people around us we assume that they’re not androids. I was thinking in relationship to what you were talking about in terms of one mind to existence: what about the relationship between minds? How can you tell, for example, if you’re trying to deduce that somebody is not an android, assuming you had an android that exactly like a human being—is there any kind of direct inductive means you can use to tell? Well it’s true that we don’t directly perceive each other’s—the question is how do we know. This is the classic other-minds problem. How do we know that other people are conscious, assuming that we know from internal introspection that we are conscious? Well it’s by induction or analogy. I mean I know that other people are conscious because they’re people and they act in such a way that the only reasonable hypothesis is that they are experiencing conscious feelings and thoughts and have purposes and goals and possess a conceptual faculty and certain kinds of knowledge and so forth. Now the philosophers—this is another version of skepticism—philosophers always say, well yeah but all that could be false. Isn’t it logically possible that since you don’t directly perceive other people’s awarenesses that all of you are in fact cleverly disguised robots or androids that some maniacal computer programmer has put together on the basis of a whole set of futuristic technologies? Well, as in response to any other kind of skepticism, the answer to that would have to be: before you can say—I don’t agree that it’s possible until you’ve provided me some evidence for that hypothesis. If it’s an arbitrary speculation on your part then I won’t even entertain it as a hypothesis. In order to say that a certain proposition is possible, that it’s possible that it’s true, you’ve got to provide some minimal amount of evidence, and there isn’t a shred of evidence for it.
1:16:49
Reframe the question a bit. How about if you know ahead of time, or begging the question, you assume you know that the person is in fact an android but behaving exactly like a person. Would you be able to tell them in some sense that it might be conscious anyway? If it’s just an android I mean could an artificial entity like a robot or android be conscious? Well, is that your question more or less? Yes. Okay. I don’t know how I would answer that as a philosopher in my armchair. Consciousness is a natural phenomenon. We know it exists because we experience it. But what its causes are—we know it’s based on the brain, but what its neurological bases are we don’t really know very much about yet. We have a lot of clues but we don’t know fully. So whether you could design something to reproduce those causes but in an artificial form, I don’t know. I don’t think you could do it in any of the standard artificial-intelligence ways that people have tried, where people tried to replicate intelligent behavior in computer programs. But this is all getting us into speculation.
Could Diaphanousness Be Consciousness’s Identity?
1:18:08
Yes. I think that you said in the diaphanous model of awareness that they say that if a consciousness has an identity then it distorts. Why couldn’t they define consciousness as its identity being diaphanous? Why do they necessarily assume that the consciousness would distort? Why can’t they just define the specific identity of consciousness as a diaphanous capacity and they wouldn’t have that distortion? Well because what a diaphanous capacity would be is a capacity with no identity that had no effect on the result. Okay. This is a good question. Why couldn’t someone who holds this diaphanous view say that well that’s the identity of consciousness: consciousness just is diaphanous? The issue of clarity has nothing to do with diaphanousness. We use these metaphors of looking through a clear pane of glass or whatever, but clarity is a term drawn from our visual awareness and it’s based on the nature of our visual perception. It’s not that the Objectivist position is that consciousness is not clear. There are clear thoughts and there are unclear thoughts. There are clear perceptions and there are not so clear perceptions. The diaphanous view uses the metaphor of clarity in the case of vision as a metaphor for the notion that there is no identity standing between the object and our awareness of it, including no identity of our own conscious faculties. I mean so there’s a limit to which you can take the metaphor of clarity literally.
What Is the Nature of Consciousness?
1:20:18
Follow-up question to that would be: you mention the Objectivist view then was that consciousness has an identity, has a specific nature. Going beyond that can you say anything more—what its specific nature and identity is? What would you say? You can’t go beyond just saying the contents of what it is? Right. When you say consciousness has a specific nature and identity, and what if someone said to you okay what is it, what would you say? What I would say is it’s a faculty for grasping, for identifying objects in reality. Now here again we might want to distinguish the faculty of consciousness—meaning the way the brain works, and the subconscious, in such a way as to produce a given kind of awareness. So if you ask me to say something about the faculty, say of perception, I would say talk to a neurophysiologist and talk to a psychologist who would tell you how the eye works, the ear works and so on. Or in the case of the conceptual level I would say, well, here there’s much less knowledge. I would say read Ayn Rand’s Introduction to Objectivist Epistemology on the process by which concepts are formed, because in this case the process is a conscious one. So in terms of the faculty and how it operates to produce conscious awareness there’s a lot we can say. If you mean the conscious awareness itself, then no. When you try to describe it you always end up describing the content of your awareness and defining your awareness against that. It’s just—it’s not the content but my awareness of it, and that’s about as far as you can go. And that’s why so many philosophers were led to think well there’s nothing to consciousness. They took what we experience about it from the internal perspective and applied it to the faculty, and that’s where the mistake occurs. That’s at least a primary form of a mistake.
Rationalist Concept of the Self-Evident
1:22:34
Okay. Yes. Could you review the rationalist concept of the self-evident? Okay. The rationalists held that a self-evident proposition is known—the term is a priori—known prior to any sensory experience. It’s known by the intellect without the aid of the senses at all. So now that’s the essence of the view. Then you ask well all right how is that possible, and the rationalists will say—usually they fall into two camps: the mystics who hold that these things are known by revelation, and the other camp which says they’re known because they are self-evident or in some other way are innate in our mind or our language.
Animal Conceptual Faculties
1:23:21
Yes in the middle here. Man has evolved from—is there any possibility that the higher primates have a rudimentary form of conceptual faculty, or must man’s conceptual faculty have arisen by a great leap? Once again this is primarily a scientific question. I don’t know of any philosophical principles on which you could deduce the number of species that have a conceptual faculty. And it’s not, as it happens, an area of science that I’m an expert on. But let me just say a word or two about it. It’s an interesting question. There’s been a lot of talk in the press lately about chimpanzees that seem to be acquiring some kind of language, or the gorillas like Koko, or just recently one, the pygmy chimpanzee, who apparently they say has acquired a vocabulary of I don’t know a couple hundred words. The problem with all of these experiments and observations is animals clearly learn lots of things, and they can learn what I hate to use these words but stimulus–response connections. I mean they learn that if they do certain things certain other things will result. You know dogs learn to beg and learn the connection between the master returning and their food. Chimpanzees clearly have a much higher capacity for this sort of thing, and maybe some other species do too. But in order to believe that they really are functioning at a conceptual level I would want to be shown, I would want to be convinced of a couple of things. First that they were doing this sort of thing spontaneously without being drilled and trained. Okay? In most of these experiments what they do is they take an animal and condition it endlessly. I mean it just takes hours and hours and hours to get the thing to push a button for “cat” when a cat strolls by. And that’s quite different from the way a human infant learns. Right? You do that with a human for a while and you’re frustrated and then all of a sudden the kid starts learning quickly. All you have to do is say once “cat”—oh, cat. What’s that? Dog. Dog. So that kind of spontaneity is an indication that there’s a genuine faculty. Another thing connected with that is that they don’t do it in the wild. We don’t know of any other species that actually clearly uses a language in the wild. So it may be—it seems to me it’s a scientific question. I don’t see any philosophical reason for doubting that maybe you could, by conditioning and learning in special environments, get some kind of animal to the early stages or a certain way up the conceptual scale. But I haven’t seen any really convincing evidence of it yet.
1:26:27
Okay. You want to follow up on that? You say this question is a problem of science, but Rand wrote in some of her essays rather matter-of-factly that animals are only perceptual and that man is the only organism, the only living entity that has the conceptual. Well I think based on the evidence so far that’s probably true. It’s not a philosophical question though. It’s a scientific question anyway. Right? As far as I can see it has to be—I don’t see any philosophical objection, for example, to the notion that life occurred on other planets, and if it did there may well be other intelligent species that possess reason. You can’t deduce from philosophical assumptions that that is true or false. It’s just there’s no evidence for it yet. And I would say the same thing about other species speaking. Okay? But I should also say when I say that I’m speaking—I’ve gotten up out of my armchair and I’m speaking on the basis of, you know, not as an expert about science. You’re speaking from your present context of knowledge. Right.
Infinite Regress
1:27:40
Yes in the back. Your first attempt to validate the primacy of existence you have used the problem of infinite regression. Well this question concerns the explanation I gave of why the primacy of existence is implicit in the act of consciousness of consciousness and existence, and what I was pointing out was that if consciousness is consciousness of something then it might be consciousness of some other state of consciousness, but then that state has to be consciousness of something and you can’t have an infinite regress. Now you’re asking about the status of that assumption that you can’t have an infinite regress. Well it’s an interesting question. I think what I’d say is that it’s simply implicit in grasping that consciousness is consciousness of something, that you can see that you cannot repeat this—you can repeat any number of steps but you can’t go on forever because you’re left with some end point. Maybe that’s—I’m not sure. I guess I would have to think about this as to whether that’s a separate principle that’s self-evident like from mathematics. Okay. We are over time, so let’s—
2. The Epistemology of Perception
The Epistemology of Perception
David Kelley
Jefferson School Conference, 1985
(Corrected transcript)
Introduction and speaker background
0:04
Welcome to the Principal Source audio presentation of The Epistemology of Perception by David Kelley. Perception is the foundation of knowledge and the source of everything we know. Epistemology is the branch of philosophy that studies the nature, acquisition, and validation of knowledge. To make the connection between perception and the rest of our knowledge—for example, to form concepts—we need a set of guidelines to understand the nature of this integration. The Epistemology of Perception fills this gap.
In Part One of this talk, Dr. Kelley tackles the questions that skeptics have used to challenge the reliability of the senses, such as illusions and other specific phenomena, by distinguishing between what we perceive and how we perceive—between the object and the form in which we are aware of the object. Part Two takes up another important distinction in epistemology: the distinction between sensation and perception.
A well-known philosopher, teacher, and writer, David Kelley earned his Ph.D. in philosophy from Princeton University in 1975. He has since taught philosophy, cognitive science, and other courses at Vassar College and Brandeis University. Dr. Kelley’s books include The Evidence of the Senses, The Art of Reasoning, and A Life of One’s Own: Individual Rights and the Welfare State. He was recently featured on John Stossel’s ABC TV news special Greed and is a popular lecturer and talk-show guest. Dr. Kelley is founder and, since 1990, the executive director of the Institute for Objectivist Studies in Poughkeepsie, New York.
This lecture was given in 1985 at the Jefferson School conference. There Dr. Kelley presented an illuminating exposition of perception as a basic level of cognition that stands between conceptual knowledge and sensations and which is a cardinal principle in the Objectivist theory of epistemology.
Topic and relation to prior lecture
2:16
The topic for today is the epistemology of perception. Now, last time I talked about the foundations of epistemology. We talked about certain fundamental principles such as the primacy of existence or the fact that consciousness has an identity, and these apply to all areas of epistemology. They apply to any issue you might want to raise.
Today we’re going to look a little more closely, specifically at perception. These topics that I’m going to talk about certainly won’t cover the whole of perceptual epistemology. They’re drawn from my book The Evidence of the Senses, so this in a way will be kind of a preview of the book. It’s been approved for general audiences, so if there are any children here you needn’t take them out.
This outline here will give you kind of a road map of the lecture. That’s your physical map; this is your philosophical map, and I will try to indicate where we are, roughly, as we’re going along.
Why there is an epistemology of perception
3:18
Now, before I begin on the issues of perceptual epistemology I want to talk about a prior question, which is: Why is there such an area to begin with? Why are there any epistemological issues about perception?
Epistemology is a branch of philosophy that studies the nature, the acquisition, and the validation of knowledge. So epistemology has two broad goals. One is to understand the basic relationships between knowledge and reality. The fundamental relationship is the primacy of existence, but there are many questions of a more specific nature. For example: How are concepts formed? How do they relate to objects in the world? What exactly in reality do concepts stand for? Or how do the procedures of inductive inference identify the causal relations between objects and actions? Or how do we grasp axiomatic facts? This is a branch of epistemology we had to use last time. So the first goal is one of understanding.
The second goal of epistemology is to offer guidelines for acquiring and validating knowledge. As we saw last time, consciousness is epistemologically active. It’s an active process that integrates and differentiates the content that we derive from our senses. Now at the conceptual level these processes are largely conscious and voluntary. Conceptual thinking is self-initiated and it’s self-directed, and as a result we can make mistakes. There’s no giant textbook of life with all the answers in the back so that we can see whether we did the problems right. The conclusions we arrive at are only as good as the processes by which we arrived at them.
Now, since we’re responsible for that activity we need standards—standards, for example, for defining concepts, standards for forming new concepts, standards for making valid inferences and testing scientific hypotheses, and so on. We need epistemology to provide those standards. In this respect epistemology is a normative discipline. It gives us rules, guidelines, ways of telling better from worse ways of functioning.
Perception is direct and automatic
5:35
Okay, now let’s turn to perception. Perception involves the direct awareness of objects in the environment, and by “direct” I mean it is not based on any prior awareness of those objects. Perception is not derived from a prior state of awareness by any conscious process. And perception is automatic. It happens automatically. We can choose whether or not to look in a given direction; we can choose whether to open our eyes or not; but once we’ve opened our eyes in a certain direction we don’t get to choose what we’re going to see.
And now I should stress that I’m talking here about perception, not about perceptual judgments that we might make on the basis of perception. For example, I’m talking about your actual sensory experience of this room, not your conceptual identification of it as a room, as a lecture room, as a place you must want to be right now, as a place you least like to be right now, or whatever. Conceptual identifications of that sort are not direct or automatic, not in the way perception is.
And because perception is direct and automatic it may seem that there are no epistemological questions to be raised about it. Why not? Well, for one thing there’s no particular problem about the relationship between perception and its objects. In the area of concepts we have to do some thinking to figure out what it is that concepts stand for. The concept “man” is an abstraction. What is it in reality—since we don’t see any abstractions—that the concept stands for? But in the case of perception the object is right in front of your nose. It’s whatever you’re seeing. There’s the object and there’s you and you perceive it. End of discussion. Time for the beach, and we could just close up shop right there.
Now of course the awareness is produced by an active process. It’s an extremely delicate and complex process that’s performed by the nervous system. But philosophers have no business speculating from their armchairs about how the nervous system does this. That’s a scientific issue.
Second, perception is automatic. It’s the inexorable effect of a causal interaction between the environment and the nervous system. So we have no choices to make. Once you open your eyes, what you see is determined by the way your visual system responds to the light. So in perception we don’t need any guidelines or standards. We don’t have to be taught how to do it.
If you’re sitting in your living room and you observe a water stain on the ceiling and you’re wondering where it came from, well then logic can help you. It will tell you that since water runs downhill you might begin your investigation upstairs. But if you’re having trouble seeing the water stain, then logic isn’t going to help you. What you need is a doctor.
Finally, because perception is direct and automatic there’s no further question of validation. Perception is the foundation of knowledge. It’s the source of everything that we know. So any attempt to validate it in terms of other knowledge would be circular. And as we saw last time in talking about the primacy of existence, perception doesn’t need to be validated in that way. Its validity is self-evident.
Okay, well, for all these reasons it may seem that epistemologists can take perception pretty much for granted and that a book called The Evidence of the Senses written by an epistemologist would be a pretty slim volume—which, by the way, it isn’t.
Why the field exists: connections to the rest of knowledge
9:05
So why is there such a field as the epistemology of perception? Well, the answer lies in the connections between perception and all the rest of our knowledge. If we were restricted to the perceptual level as the animals are, then we would indeed have no need for epistemology—even if by some cognitive miracle we were capable of discussing epistemology without concepts.
But because perception is the foundation for the whole structure of conceptual knowledge there are two kinds of issues that epistemology has to consider about perception.
First of all, the validity of all our knowledge rests on the validity of perception. We’ve seen there’s no fundamental question whether perception is valid, but there are a host of specific phenomena such as optical illusions that skeptics have used to challenge the reliability of the senses. All of these challenges can be answered by distinguishing between what we perceive and how we perceive—by introducing the distinction between the object of awareness and the form in which we’re aware of it. Those are distinctions that I mentioned in my last lecture, but we need to understand in more detail how they apply to the actual phenomena, and that’s the subject I’m going to talk about in the first part of the lecture—Part One.
Secondly, all of our knowledge is based on the integration of perceptual data. For example, forming concepts. Now this integration is a conscious process, so epistemology has to study the integration of perceptual data. But in order to understand the integration we have to understand the nature of the data that are being integrated. So in the second part of the lecture I’m going to talk about one of the issues that fall under that heading—namely, that perception is the awareness of entities. Unlike a fleeting sensation, perception is the discrimination of an object as a unitary whole having a specific identity and enduring over time. So in the second part of the lecture we’ll examine the distinction between sensation and perception.
Narrower issues still matter
11:07
Now I should say that we’re going to be moving at a somewhat more concrete level today than we did on Monday. We’ll be looking at some actual features of experiences—perceptual experiences that we all have in common. We’ll even be doing one or two simple experiments. I’m going to ask you to indulge me in a few of these. They won’t be very complex. You can get an idea of their complexity from the experimental apparatus that I have here.
I should also say that we’re going to be dealing with a somewhat narrower range of issues, or rather issues of a somewhat narrower interest. The level we were talking on last Monday was the level of broad philosophical fundamentals in epistemology, and that’s the level at which epistemology affects areas outside philosophy such as journalism or education—issues that have come up in the last few days. These topics about perception tend to be of somewhat narrower interest to philosophers, to epistemologists. However, I want to make a case to you for paying some attention to them.
First, because while all you really need, in a sense, is the basic distinction between what we perceive and how we perceive—and correspondingly between the object of perception and the form of perception—in order to possess those concepts you need to see how to apply them in some concrete cases. And that’s what we’re going to do today.
Secondly, while the fundamental level of epistemology is the one that affects the broader political and cultural trends, you have to understand that some of these narrower issues are the ones that lead philosophers to one or the other side of the basic issues—primacy of existence versus primacy of consciousness—and that of course affects the kind of culture and political system that we live in. If we are dissatisfied by the fact that we are now living in a culture, a political system, that no longer recognizes John Locke’s principles of individual rights, the reason is in part because of Kant and his espousal of the primacy of consciousness. But that followed in turn from the inability of philosophers—and John Locke was preeminent among them—the failure of those philosophers to deal with some of the puzzles and problems of perception.
Okay, let’s then get started.
Part One: Form and object
13:44
The operation of our senses can affect the way things appear to us—the way they look, sound, feel, taste, and smell. You’re all familiar with many, many examples of this. When you go out into the sunlight you experience intense and often uncomfortable brightness, but gradually your eyes adapt and things seem less bright than at first even though there’s been no change in the objective intensity of the light. You’ll experience that in about an hour and a quarter. In the same way, when you come inside things seem unusually dim until your eyes adjust to that level of illumination. You probably had that experience a few minutes ago.
Or again, let’s look at my little experimental apparatus here. If you look at this, what you see is a circle. This is very close to being a perfect circle because I drew it with a little string. This is a perfect circle. Those of you, however, who are sitting to one or the other side of this, if you focus on this as an item in your visual field, if you disregard the various cues that tell you that you’re looking at it at a slant and just focus the way an artist would, what you’ll see is something that looks like an ellipse. Those of you in the center will be seeing it in a more circular form, but if I now tilt it, even those of you in the center will be able to see this as it will look somewhat like an ellipse. Unless you can project your astral body about 20 feet in the air, no one in this room now can see this as a real circle—although you can see that it is a circle. I’m not denying that. But it will look somewhat like an ellipse. And of course as you moved and changed your position the contours of the ellipse, or the shape of the ellipse, would somewhat change.
These are examples of what philosophers call the relativity of perception. I’m here talking about the facts of perceptual relativity. This relativity is not limited to vision. When a car passes you with its horn blaring it often sounds as if the pitch of the horn is changing even though the vibrator is going at the same rate. If you’ve just had pancakes with syrup, milk will taste sour. I include that because that was my first experience of perceptual relativity as a kid. If you’ve just been washing dishes in hot water, then warm water will feel cool on your hands.
Perceptual relativity occurs in every sense modality and with regard to every attribute of objects that we can perceive through any of the sense modalities. The relativity consists in the fact that the same identical object may appear differently in different circumstances. Two people looking at the same identical object may see it differently, or the same person may see it differently at different times.
Everyday relativity and illusions
16:38
Now the cases I’ve mentioned so far have been everyday familiar types of perceptual relativity. There are also more radical cases that are known as illusions. For example—and this is the classical example—a stick that is half submerged in water will look bent even though in fact it’s straight. Or you can take a flat disc of white paper, paint a certain design on the surface using only black, and then if you spin it at the right rate you’ll see colors. This is called a Benham disc, and for those of you who are going to San Diego on your free day you’ll find one at the science museum in Balboa Park along with a number of other optical illusions.
Now there’s no mystery why perceptual relativity occurs. The appearance of an object is the result of a causal interaction between the object and our sense faculties. In the case of vision and hearing there’s also a medium—light and the sound waves. The outcome of that interaction is determined by all three participants: the object, the medium, and the sensory systems. So even if the object doesn’t change, changes in the other elements can alter the appearance. No mystery. And scientists in fact know a fair amount about how that works in specific cases. For example, the bent look of the stick, as I’m sure you know, results from the optical phenomenon of refraction. Or again, when you come in out of the sunlight and the room looks dimmer than usual, the reason is that when you were outside your visual receptors were firing extremely rapidly. That depleted a visual pigment called rhodopsin, and rhodopsin is in a sense the fuel that receptors use when they fire in visual receptors. And so when you came inside you had a lower level of rhodopsin and your senses couldn’t respond adequately to the new lower level of illumination.
Now of course I’m oversimplifying like crazy, and there are some phenomena for which scientists haven’t yet found the exact causal basis. But it’s clear in principle that these phenomena result from the interaction between our senses and the environment. There’s no mystery about why they occur.
The argument from perceptual relativity
18:54
However, the fact that they occur has seemed to most philosophers incompatible with the validity of perception. And to express this felt contradiction that most philosophers have seen between the facts of perceptual relativity and the validity of perception, I’m going to present to you a classical argument which is called the argument from perceptual relativity—an ancient argument that goes back to the Greeks. It goes like this:
How can we see the object as it really is if our perception of it is affected by factors that are extraneous to the object? If a thing’s appearance is affected by the nature of our own sensory systems, then it doesn’t appear to us the way it really is. There’s a gulf between appearance and reality. So what we perceive is only the appearance, not the real external object.
Now let’s take this a step further. If we don’t perceive the external object according to this argument, then what do we perceive? Well, we perceive an internal object.
Philosophers who accept this argument—the argument from perceptual relativity—are known as representationalists. It’s a mouthful, but there’s no simpler way to express it, or a better way to express it. They are representationalists. They hold that an appearance is a subjective state. An appearance is an effect of the object on our consciousness. This subjective state, they say, exists in consciousness. It’s a kind of mental image. When you look at this circle—what you’re seeing in fact, especially if I tilt it to make it clear—what you see, what you’re seeing, is really an elliptical appearance, not an actual physical circle. And this image has the same status as the images in a dream or an hallucination. The image represents the external object in the way that a portrait represents a person. That’s where the name comes from. But a portrait is a distinct object from the person who’s represented in that portrait, and so if you’re looking at a portrait you’re not literally seeing the person. In the same way, representationalists say we do not literally see the external object. We only see an image that is a distinct entity from that external object.
The mind, according to the representationalist, is a kind of inner theater, a kind of inner movie house where we watch images that play across an inner screen. We have no direct contact with reality outside. Our only contact is that the images may correspond to objects in reality—but on the other hand they may not.
Implicit in this view, in representationalism, is the assumption that the mind could have generated all of these images out of itself. The external world is a possible source of these images, but it’s not a necessary source of the images. Consciousness could originate all of its own perceptual contents, as it seems to do in a dream, they say.
Representationalism as a middle position
22:14
Now let’s step back for a moment and look at what’s going on here. First of all, notice that representationalism accepts one of the main tenets of the primacy of consciousness that we talked about last time. It holds that consciousness can originate its own contents independently of reality. Consciousness is metaphysically active. But the representationalists do not accept the other main tenet of the primacy of consciousness. They don’t think that the objects—the ultimate objects of knowledge—depend on the subject. They think there is a real world out there independent of us. It’s just that consciousness is not inherently dependent on that world for its contents. So representationalism in effect is an attempt to split the difference between the primacy of existence and the primacy of consciousness.
Let me give you an analogy. Just as liberals in foreign policy think that neither the U.S. nor the Soviet Union should have military superiority, representationalists think that neither consciousness nor existence is primary. They think that existence and consciousness are two separate but autonomous realms that exist in a kind of metaphysical detente.
Now you may recall that on Monday I gave the same analysis of skepticism that I just gave you for representationalism as a kind of middle position. The skeptic believes that consciousness could originate all of its own contents, as we saw, and to that extent he holds the primacy of consciousness. But the skeptic holds not exactly that there is a world out there, but that consciousness, to be valid, would have to be capable of grasping such a world. So the skeptic accepts the primacy-of-consciousness view that the mind is active, but he also holds the realist, the primacy-of-existence standard for validity. So the only difference between skeptics and representationalists is that representationalists think we can have knowledge of the external world even though we don’t directly perceive objects in that world. We can infer the existence of such objects from their effects on our consciousness. Whereas skeptics feel that such inferences are always unsound; they’re never valid; and that the existence of any objects outside the mind is always going to be in doubt. So the skeptic in effect is a representationalist who’s having an anxiety attack.
Now this anxiety is quite well deserved. It’s very appropriate. If it were true that we’re trapped in an inner theater, directly aware only of images that our consciousness itself originated, then there would indeed be no ground for inferring the existence of external objects. So the representationalism that was maintained by philosophers such as René Descartes and—I’m sorry to say—John Locke in the 17th century, that representationalism led to the skepticism of David Hume a century later in the 18th century, and that in turn, as we saw, led to the full-blown idealism of Immanuel Kant. So you really can’t split the difference between the primacy of existence and the primacy of consciousness. Once you step off the shore of realism, so to speak, you’re going to be sucked out to sea. Ultimately you’ll be drawn into representationalism and the skeptics will then eat at you until you have an anxiety attack and then you’ll be blown all the way out.
On the other hand, though, there’s no basis for skepticism until we have adopted the representationalist position. Without representationalism the skeptical arguments can’t even get off the ground, because from a realist point of view, if we’re directly aware of external objects we don’t have to infer their existence. We’re directly aware of them. Their existence is self-evident.
So it matters a good deal. This traditional argument for representationalism, which is this argument from perceptual relativity, has been an important basis for anti-realism.
The diaphanous model in the argument
26:07
Let’s look at that argument. The argument says that to perceive an object as it is, our perceptual faculties cannot affect the way the object appears. The angle of observation must not affect the apparent shape of the object, so a circle must occupy a circular portion of our visual field no matter where we’re standing. If consciousness is going to be valid, the rhodopsin level in our visual receptors must not affect the apparent brightness of the illumination. This is what the argument is saying. The prior state of adaptation of our temperature receptors may not affect the experience of warmth and coolness.
Now isn’t this impossible, you may ask? Is it not in effect a demand that consciousness violate the laws of physics and biology? Would it not in fact violate the law of identity by demanding that the faculty of consciousness produce an effect that doesn’t depend on the nature of consciousness? Yes, says the argument. All these consequences follow. If consciousness cannot meet this standard, well then so much the worse for consciousness.
You can see therefore that this argument incorporates the diaphanous model of awareness that I talked about last time. It assumes that we can perceive objects and their attributes only if perception is a form of revelation—a state of awareness that is not produced by any specific means. It assumes that if a faculty of perceptual awareness has any identity, then it cannot grasp the identities of objects outside of it.
The argument then goes on to say that our perceptual faculties do have an identity. They do interact with objects in specific ways, and this interaction does affect the way objects appear. Therefore we don’t really perceive those objects outside the mind but only their effects on our consciousness.
So in this respect a representationalist is a kind of frustrated naive realist. As soon as he realizes that he can’t have a diaphanous view of external objects he kind of invents an inner realm of inner objects that he thinks he can be diaphanously aware of.
But we know from last time that the diaphanous model is false. That whole picture of how the mind works is false. So we can dispense with the argument from perceptual relativity. But of course we can’t dispense with the facts of perceptual relativity, because they are facts. We still need to see how to interpret those facts properly. And let’s now turn to the Objectivist theory.
Appearance is a relationship, not an entity
28:46
The first step is to recognize that an object’s appearance is not an entity. It’s not a thing distinct from the object. The concept of appearance designates a relationship between the object and the perceiver.
Now look at this sheet of paper here. The fact that the paper appears to you, that it looks a certain way to you, and the fact that you’re aware of the paper—those are the same fact. The concept of appearance and the concept of awareness or consciousness in this case describe the same relationship between you and the paper. The only difference—if I can speak like a linguistic analyst for a moment—the only difference is that the two concepts describe this relationship between you and the paper from opposite ends, so to speak. We say that you are aware of the paper, but we say that the paper appears to you. Same relationship, just looking at it from the two different ends.
So the statement that we perceive appearances—that’s a hopeless confusion. It’s equivalent to saying we perceive our perception. That’s nonsense.
The concept of form
30:04
Secondly, certain aspects of the object’s appearance depend on the nature of your sense organs and the way they interact with the object and the medium. Look at the circle again. Let me tilt it a bit for the full effect. If you focus on it as an item in your visual field, if you focus on it in the way that an artist would, you’ll notice that the surface does look like an ellipse in a certain respect. It’s not that you’re actually seeing an ellipse. There’s no ellipse here. There’s a circle. There’s no actual ellipse anywhere in this room, at least not that I’m aware of. But the circle in a certain respect does look the way an actual ellipse would look if it were in a plane perpendicular to your line of sight.
This aspect of the circle’s appearance—the respect in which it looks like an ellipse—depends on the way your visual system works. It depends on the fact that an object in three-dimensional space is being projected onto a two-dimensional retina in your eye.
To designate any such aspect of an object’s appearance, Objectivism uses the concept of form. A form is an aspect of an object’s appearance that depends on the way our senses respond to that object in the conditions at hand.
So when I drew your attention to the apparent shape of the circle in your visual field I was drawing your attention to the form in which you perceive its real shape.
Let me give you the definition of form again. This is important. A form is an aspect of an object’s appearance that depends on the way our senses respond to that object in the conditions at hand.
A form is an aspect of an object’s appearance that depends on the way our senses respond to that object in the conditions at hand. I’m not sure I’d want to be held to that as a definition, but we can pursue that more in the question period.
So the elliptical appearance of that circle is an aspect of the form. When you buy a tie or a scarf inside a store and then take it outside and it looks different in the sunlight from the way it looked in the store, what’s happening is that the difference in illumination is causing you to perceive its color in a different form.
In general, because we perceive every attribute of an object by some specific process, we perceive every attribute in some specific form.
Now this is the essential point to remember: a form is an aspect of appearance that can vary with variations in the means of perception. It’s an aspect of appearance that can vary even if the object is unchanging but only the state of the medium or your own sensory systems are changing. So that’s our test for whether something is part of the form of perception.
The form is a consequence, at the level of conscious experience, of the physiological fact that our senses operate in a definite physical way.
Four points about form
33:31
Now there’s a great deal we could say about the form of perception. I know this because I have said a great deal about it elsewhere. But for our purposes I want to limit myself to four main points.
First, the form is not an image or a copy of the object. The form is not an entity in its own right distinct from the object. A form is an aspect of an object’s appearance. That’s what I’ve just been saying. Without the object there’s no appearance, and without the appearance there are no aspects of appearance. We don’t perceive the form as distinct from the object. We perceive an object in a form. When you look at the circle, the patch in your visual field that looks like an ellipse—that’s not an image or a veil that stands between you and the real circle. It is the circle. The patch in your visual field is this very circle as seen by you from there.
Secondly, the form is not in your mind. Contrary to representationalism, it is not subjective. Why not? Well, it’s a product of an interaction between the object and the senses. Therefore, to be sure, it’s not in the object as an intrinsic property. This is not literally elliptical. But it’s not in your senses either, nor is it in consciousness, nor is it anywhere in between. Precisely because it’s a product of an interaction it cannot be located in any of the things that interact.
Let me give you an analogy. Imagine two cars have a violent collision and then someone comes along and says, “Well there’s violence here. I wonder which car was violent, the Ford or the Buick.” Well that’s absurd. It’s the collision that’s violent. It’s the interaction that’s violent, not any one of the things that interact. And the same is true for the form of perception. The form is an aspect of the way we interact with the object. If you want to know what the form is a property of—something—what is it a property of? It’s a property of the object as perceived by us.
Third, notice that because the form depends on the way our senses respond to an object, it’s possible to perceive the same object in different forms. You can see this in a simple case whenever the conditions of perception change. If you walk around the room the form in which you perceive this circle would alter as you moved.
Let’s look at a slightly more interesting case. Suppose our sense organs were different. We didn’t just change the conditions of perception but we actually altered our sense organs. Imagine, for example—well I’ll give you an actual example. When we look at an object such as a pen—which I don’t have here; I have my lighter—I can look at this lighter and see that it’s roughly kind of a squashed cylinder and kind of long. I can see that, and I see it in a visual form that consists of colors and a certain shape in the visual field, etc. I can also feel the shape of it in my hand even if I’m not looking at it. I’m still perceiving the shape. I can still tell what shape it is, but I’m perceiving that in a quite different form—a whole set of tactile sensations. There’s no colors in my hands.
Let’s finally consider one even further-out example. As I’m sure you all know, bats perceive objects by means of echoes and echolocation. Their auditory system is so highly developed that they can detect the size and the shape of objects and even in some cases the texture of an object by means of hearing—by the way they reflect sound. Bats literally hear the size, the shape, and even the texture of objects. Now we have no idea—at least I have no idea—what it would be like to experience objects in that form. But what they perceive, those properties of shape, size, and texture—those are the same properties that we can perceive by means of vision and touch.
Fourth and finally, there can be no such thing as a false or invalid perception. This is a crucial point. A false perception would have to mean perceiving an object in the wrong form. But there’s no such thing as the wrong form to perceive an object in. Why not? Well, the form is a causal product of a physical interaction among the object, the medium, and the sense organs. In any given set of circumstances there’s only one form that can be produced by that interaction, and that form faithfully reflects all the factors that are operating. So long as a form allows us to discriminate an object or to discriminate a given attribute, so long as that’s possible, then it is a form in which we perceive that object or that attribute.
Remember that we have rejected the diaphanous model of perception. That means there’s no right way for an object to appear apart from its interaction with sense organs of a particular type. You can’t step out from behind your mind, kind of put your mind there and look out at reality to see, and then say, “Oh that’s how objects really look. See, they don’t look that way from behind my eyes.” That would be a contradiction. If you step out from behind your senses, objects don’t look any way at all because you can’t obviously perceive them.
Now there are illusions. There are such things as the bent stick or the Benham disc, etc. But these are not perceptual phenomena. What’s illusory about the bent stick is not its appearance per se but the fact that we’re likely to make the wrong judgment about it. We’re likely to misidentify it conceptually. We’ll assimilate it to the concept “bent” instead of the concept “straight.” In and of itself the appearance is not false or illusory. It’s the only possible form in which we could perceive the stick given that the light it reflects is being refracted. The problem is that that’s an unusual circumstance in which to perceive an object and therefore we’re perceiving it in an unusual form, and our concepts for perceptual properties are tied to the usual forms in which we perceive them. So the error here occurs at the conceptual level. In effect we’re assuming that conditions are normal when they’re not—I mean if you judge that the stick is bent.
So in summary: a form of perception is not an entity and it’s not in the mind—the first two points. The same object can be perceived in different forms—point three. And there are no wrong or illusory forms, just mistaken judgments—point four.
Sensory qualities as forms
40:39
Okay, now let’s apply the concept of form to one last issue: the issue of sensory qualities. The examples of perceptual relativity that I’ve talked about so far are familiar everyday phenomena, but the concept of form is not restricted to these phenomena, to these cases. It’s an open-ended concept that we use to identify any sort of relativity as we acquire evidence for it.
Now common sense gives us all the evidence we need about the familiar cases, but there are other, deeper levels of relativity that have been discovered by science, and I want to look at this deeper level which involves sensory qualities.
We all know that an object’s particular shade of color seems to vary with changes in the illumination. I mentioned earlier the example of the scarf that looks different in daylight from the way it looked in the store. If the scarf is red, say, the exact shade of red varies with the conditions of perception, with the nature of the illumination, and this changing shade of red pertains to the form in which we perceive the color.
But science tells us that the very color itself—the very quality red, not just the exact shade but red as such, the quality that we describe as a warm color—that quality also depends on our perceptual apparatus. It isn’t out there in the objects apart from us. Our experience of that red color is a response by our visual system to light of a certain wavelength, and a physical object reflects light of that wavelength essentially because of the atomic structure of its surface, of the atoms in its surface.
So red is not an intrinsic property of the object itself. It’s a product of the interaction between the object with its atomic structure and the surface, the light it reflects, and the way our eyes respond to that light. And so red is the form in which we perceive—well, in which we perceive what? Well, the property of the object to which our senses in fact are responding is the atomic structure of its surface. That’s why it’s reflecting the light. So red is the form in which we perceive that atomic structure.
Now the same is true for other sensory qualities in the other sense modalities. Warmth and coolness, for example, don’t exist in objects outside of us—not felt warmth and coolness. Sensations of warmth and coolness are the form in which we perceive the kinetic energy of molecules in the surface of the object that we touch with our hands. That molecular energy level is what’s intrinsically in the object. That’s what a thermometer measures, and that’s what our touch receptors are responding to when we feel warmth and coolness.
Sweetness and other sensory qualities of taste and smell are the forms in which we perceive certain chemical properties of substances.
And in the case of hearing, I’m sure you’ve all heard the old conundrum: If a tree falls in the forest when no one is there, does it make a sound? Well we can now answer that question. If by “sound” we mean a certain pattern of vibrations in the air, well then of course yes, that does occur. The tree makes that pattern of vibrations in the air. But if we mean the roaring, whooshing noise that has a certain sensory pitch and loudness, well then no, that’s not there. That’s the form in which a perceiver, if one were present, would respond to and perceive the pattern of air vibrations.
So this is a more radical kind of relativity. And I want to emphasize that the reason for accepting it as a fact of relativity is purely scientific. This is not a philosophical issue; it’s a scientific one. Nevertheless the concept of a form applies to this level, and everything I said before about the forms of perception is true for colors, tastes, and other sensory qualities.
A person whose visual system were wired up differently from ours, for example, might perceive objects in the form of different colors. What we see as green he might see as red. Such a person would still be perceiving the same surface properties that we do. He would classify them in the same way on the basis of their similarities and he would learn to measure them as we do in terms of the wavelengths of light. It’s just that he’d perceive those surface properties in a different form.
And you know, to a certain extent this is what happens in certain forms of color blindness. People with certain types of color blindness can discriminate the differences among objects pretty much in the same way we do, although they’re experiencing them in a different form. It’s just that, well, for example, it often happens that a color-blind child will grow up and be able to discriminate the different colors. He’ll learn to call a tomato red, grass green, milk white, and so on. In other words he learns to classify objects under the color concepts that his parents are teaching him and using the color words that they’re using. And neither he nor anyone else knows that he’s experiencing them in a somewhat different form until he’s given more sophisticated tests by optometrists.
Also, because color is the product of an interaction between the object and our sensory systems, the question “Where is red?” is not a valid one. Color is not in the mind. It’s not in consciousness. Consciousness is not red. But it’s not in the object either. As a product of an interaction, as I said before, it can’t be located in that sense.
Two objections about color
46:30
Finally let me end this part of the lecture by considering briefly two objections—the two objections that in my experience have been raised most often against this view of sensory qualities.
Objection number one: How can I say that we actually perceive atomic structure? Come on, my students say, what we see just doesn’t look like an atomic structure.
But then I would have to ask in reply: What is an atomic structure supposed to look like? We all have a picture in our minds that we got from our physics textbooks or from Scientific American or from Time magazine for that matter—a picture of atomic structure as a kind of swirl of little gray particles. Atomic structure might look that way to atomic-size perceivers. That would be merely another way that a certain attribute could appear to perceivers with certain types of sense organs, and their philosophers—the atomic-size philosophers—would ask the same questions we do: Where’s the gray? How can we see the real shape of atoms if the apparent shape depends on our angle of vision? They’d be perceiving in a form just the way we are. It’s just a different form.
Obviously this objection isn’t an attempt to step out from behind our senses to see what atomic structure looks like diaphanously, without affecting anyone’s sense organs. And of course that’s a contradiction. A thing cannot look any way without appearing in a certain way to a certain perceiver. Color is the way atomic structure appears to perceivers like us.
Objection number two: Colors. This is the objection. Colors are not intrinsic properties of objects. They’re relational qualities that result from an interaction. That’s what we’ve been saying. But that’s not how we perceive them, says the argument, says the objection. We experience a color as external to the mind. What I’ve been telling you about colors should come as a surprise if you haven’t heard it before, because to us colors look as if they’re out there in the objects, in the surface of the objects, as intrinsic properties of that surface. And that again is true.
Well, so goes the objection: Isn’t that a kind of illusion? Philosophy tells us that color is not intrinsic in the object. Our senses tell us that color is out there, intrinsic in the object. So aren’t our senses deceiving us?
Now I think if we can answer this question we will pride ourselves on thoroughly understanding the concept of form.
To answer the question, the objection, let’s take the concept of deception literally. Let’s ask what would have to be the case for our senses to be convicted of deceiving us. They would have to tell us something that isn’t true. All right, what’s the truth here? The truth is that there are two properties. There’s the atomic structure of the surface and there’s the sensory quality red. The atomic structure as distinct from the sensory quality is an intrinsic property of the surface. It’s in the surface. The sensory quality as distinct from the atomic structure is not intrinsic to the surface. It’s a relational property that depends on the perceiver.
But that distinction between the two qualities can be drawn only at the conceptual level, only on the basis of scientific knowledge. It cannot be drawn at the perceptual level. I don’t see two separate properties—the atomic structure and the sensory quality red. For me to do that, each would have to appear to me in a certain form. But color is the form in which the atomic structure is appearing to me. So you can’t ask the senses to perform the function of the conceptual level.
So when I experience the red color as external to my mind, as a property of the surface, my senses are not telling me that the red as distinct from the atomic structure is intrinsic to the surface. That would be a kind of deception. What my senses are saying is: some property that I experience in the form red is an intrinsic property of the surface. That’s what it means to experience the red as being in the surface. It means something that I’m perceiving in the form red is an intrinsic property of the surface. And that’s true. My eyes do not deceive me, because I am experiencing something—namely the atomic structure—that is intrinsic to that surface. Experiencing the red as external to my mind is the only way that I could perceive the externality of the atomic structure.
So given the fact that we have to perceive by specific means and therefore in a specific form, we perceive color as being in exactly the right place: the place of the object, of the attribute that we’re discriminating by means of color.
All right, I know that was pretty fast and furious. I’ll be happy to answer questions about it in the question period. But to conclude this part of the lecture, because we are running out of time: once we’ve grasped the concept of form and once we’ve thoroughly rejected the diaphanous model, we can see that the facts of perceptual relativity do not pose any threat to the validity of perception, the reliability of the senses.
Part Two: Sensation versus perception
51:56
Which brings us to Part Two. I want to change gears now and take up the separate topic: the distinction between sensation and perception.
Now as I mentioned at the outset of the lecture, this is a very important distinction in epistemology because it makes a big difference—in ways that I’m going to explain later—which of these two levels of awareness represents the given, which of them represents the foundations of knowledge. To put the importance in a word at the outset: if the level of sensations were the given, we would not be able to reach the conceptual level, which would be a fairly large deficit in our cognitive life.
So let’s talk about sensations and perceptions.
Molyneux’s question and newly sighted patients
52:47
Three centuries ago a man named William Molyneux wrote to the philosopher John Locke with an interesting question. Suppose, he said, that a man who had been blind from birth had grown up, had learned to find his way around in the world, had learned how to speak, and then his sight were restored to him by some operation. Would he see the world the way we do? If he was shown an object that he was familiar with by touch, could he identify it just by looking? Would he even see objects as such?
Well there is in fact an operation that can restore sight in certain types of blindness where a lens has been clouded or occluded, and so Molyneux’s question can be answered experimentally. As it happens, however, there’s no single answer. Some patients have a lot less difficulty than others in learning how to use their new sensory capacity. But a large number have tremendous difficulty, and their first experience of sight after the operation is radically different from our own experience of sight. They do not see objects as such. What they see are vague patches of color. They don’t see depth. They see the color patches as lying in a two-dimensional plane before their eyes—except even that is too definite a description because they don’t experience that two-dimensional plane as having any definite distance from them. It just kind of floats somewhere in an indeterminate way.
One doctor described it this way: they see colors much as we smell an odor of peat or varnish which unfolds and intrudes upon us. Imagine seeing the world in that form. They can see; they can see the differences between the colors, but apparently most of them cannot focus on the shapes or sizes of the color patches. When an object moves across their visual field they don’t follow the object as a single thing moving. What they see is an interlude of light and then dark and then light—a kind of kaleidoscopic change as the object passes in front of them.
Now this is not the way we see the world. The study of people who have just gained their sight gives us some inkling of what it would be like to experience the world in the form of sensations. But this mode of experience is really quite foreign to us. Our own experience takes the form of perception. I’m using “form” here not in the technical sense, just in a more general way. We perceive objects. We perceive them as wholes, as unitary existents. We discriminate them as units from the background of other objects. In vision we see depth, the third dimension, as easily and automatically as we perceive the two other dimensions. We perceive an object as having a definite identity as a unit—an identity that remains the same even as we change our perspective on it. When an object moves we can follow it over time.
So the essential characteristic of perception is that it involves the awareness of things as such. It involves the discrimination of individual particular things as units from their background. It involves the awareness of entities.
Discrimination is not the difference
56:10
Now let’s look at this aspect of perception a little more closely. The difference between sensation and perception does not lie in the fact that perception involves discrimination. In the broadest sense, discrimination means differentiation—that is, the awareness of difference—and that is present at every level of cognition, from sensation through all of our concepts.
Even at the level of sensation there has to be some contrast present. There’s a famous experiment in which they tried to present a subject with a completely undifferentiated visual field. The one I’m familiar with: they took a ping-pong ball, cut it in half, put the two halves over the eyes, and then put the subject in a room where there was just diffuse green light with no changes in the intensity of the light. As a result the subjects reported that they experienced an undifferentiated spread of green—no patterns, no borders, just this purely green fog. That’s what their visual field consisted of.
Well of course at first they had sensations of green, but these gradually faded and they were replaced by what’s called brain gray, which is the physiologist’s term for the experience we have in the dark with our eyes adapted. When you’re lying awake at night you can’t get to sleep and your eyes are closed, you’re experiencing brain gray. It’s not pure black but it’s just a few random firings. It’s the absence of visual experience in effect. That’s what happened to these subjects even though the light falling on their eyes was quite intense.
So even a simple sensation requires some kind of contrast, some discrimination, some awareness of difference.
So the difference between sensation and perception doesn’t lie in discrimination per se. It lies in what is discriminated. Sensation is the awareness of individual qualities as more or less self-contained units—qualities as self-contained units—whereas perception is the awareness of entities, objects possessing a unified nature that includes and embraces all of their qualities. So we have sensations of red or sweet or middle C, but we perceive an object, a chunk of food, or a sound.
This is not a definition—this one is for sure not a definition—but it’s a description. A sensation would be the awareness of an individual quality, an isolated quality, whereas perception involves the awareness of a whole thing which of course has many qualities, and we perceive those qualities as going together.
Gestalt, figure-ground, and entities as primary
59:04
For example—and I think this will spell it out a little further—in psychology the Gestalt school, despite its other faults, I think identified this aspect of perception clearly, although their explanation wasn’t too hot. They pointed out that the perceptual field is segmented. We don’t experience a welter of unrelated qualities. That would be sensations. No, we experience the qualities as going together, as belonging together in certain groups, they said.
Now there’s some truth to that way of putting it. Now if you look at a person you see various—there are various patches of color there: the clothing, the face, the hair. You don’t experience those as distinct objects, distinct entities. They’re all parts of the same entity. The border between a person’s hair and the background is a contour between one entity and other entities, but the contour between a person’s, say, face and his shirt, which you can all see, is not experienced as a contour between two different entities.
We can also take a negative example to see a case where the contours are not visible. Imagine a soldier wearing camouflage fatigues is standing at the edge of the woods. Now if the camouflage works, the borders between the patches of color on his fatigues are more prominent than the border between him and the woods. That means you don’t perceive the soldier as such even if you can see every part of his facing surface. The whole point of camouflage is to throw off our perceptual faculty. At the level of sensations camouflage would be pointless because there would be nothing to disguise. We wouldn’t see the soldier anyway. We’d always see the patches of color. He could be wearing day-glow blue and all we would see is a patch of day-glow blue.
Now in vision this aspect of perception is called the figure-ground phenomenon. It’s a term some of you may have heard. We discriminate a figure as a whole from its background. But the other sense modalities work the same way. In hearing, for example, there’s a so-called cocktail-party phenomenon. You know, you’re standing in the midst of a bunch of people, all of them are babbling away, and all those voices are arriving at your ear as a single complicated wave train. But your auditory system allows you to pick out one of those wave trains—the voice of a single person—and listen to it as a unit. Even though that voice is not literally an entity, you’re perceiving it as a unified whole.
And remember it’s not just qualities within a single sense modality that we group together perceptually. When you hear the telephone ring you turn to look at it and you reach out to pick up the receiver. You experience it as a single object that you hear, see, and touch. Or when you eat a candy bar you perceive the flavor by taste and the odor by smell and the texture of it by touch receptors, but you don’t experience them as isolated qualities. You experience them as aspects of a single unified, delectable whole.
However, it’s dangerous to describe this as the Gestalt psychologists did—to say that what we’re doing is grouping together qualities—because that suggests that the groupings maybe are kind of loose and they’re kind of arbitrary. It presupposes in the way it’s formulated that we start with the awareness of qualities and then put them together. But that’s not what happens. We start out with the awareness of entities as unified wholes and then we can focus selectively on individual qualities. At the perceptual level entities are the primary objects and the awareness of their particular qualities comes later. It’s a derivative ability.
Moreover, even when we do focus selectively on a single quality in perception such as color or shape of an object, we don’t experience it as a self-contained unit. We perceive it as the color or shape of a particular entity. We perceive the quality as belonging to a certain entity even if we’re not paying much attention to the entity.
Now this is a subtle point but it’s an important one. You know, when Aristotle laid out his basic categories—the basic kinds of existence that there are, such as entities, actions, attributes, relationships—Aristotle pointed out that entities are the basic existents. An attribute is always an attribute of an entity. An action is always performed by some entity. A relationship is always a relationship among entities, and so on. Well my point now is that this fact is available to us in perception. We are aware of qualities as qualities of entities. We perceive them that way.
For example, let’s take the figure-ground phenomenon. The figure is distinguished from the ground by means of some contour. This line here is the upper edge of the podium. It’s the contour that separates the podium as a figure from its background—namely me. Now if you were functioning at the level of sensations you might experience this line here merely as a border between two patches of color. You might experience this line as belonging equally to this patch of color and to this patch of color. But I hope that’s not how you perceive it. You perceive the line as an attribute of the podium, not as an attribute of me. You perceive it as a contour where the podium stops, as an edge where the podium stops, not as an edge where I stop. Thank God.
So in that sense we perceive the attribute as an attribute of an entity.
Finally, the awareness of entities means the awareness of objects that persist and retain their identities over time. A sensation is a momentary experience. At the level of sensation every moment is a new experience replacing the one before and replaced in turn by the one that comes after. But an entity is not a momentary existent. You couldn’t have the experience of entities in a moment. It requires integration over time.
So when a baseball player—you will observe next week—when a baseball player runs to catch a fly ball he sees it as a single object that is moving and changing its position relative to him, not as a kaleidoscopic change in his visual field and not as a series of individual frames as in a slow-moving animated cartoon.
To put all this in a nutshell: perception is the awareness of things—things that possess various attributes and relationships and that move and change in time while maintaining their identities. Perception tells us not just that something exists, as we saw last time, but that entities exist.
Why this matters for concepts and causality
1:06:08
Now this is a crucial fact in epistemology because all the rest of our knowledge depends on it. When we form concepts, for example, we do so on the basis of perceived similarities. But similarity is a relationship among entities. When we explore those similarities we distinguish the attributes that entities have in common from the attributes that distinguish them. So we couldn’t begin to form concepts unless we could perceive entities as such. You couldn’t get to the conceptual level if you had to start with sensations as your epistemological given.
Again, a lot of our knowledge is concerned with causality, but causality is a relationship between a thing and the way it acts. If we didn’t see things as such we couldn’t form the concept of causality.
So perception is a distinct and a crucially important level of cognition. It stands between sensations on the one hand and conceptual knowledge on the other.
Now our conceptual knowledge derives from perceptions by means of a conscious process of integration. Does perception in turn derive from sensations by means of a conscious process of integration? No. Perception is the basic level of cognition. Perception is the given.
Now this is a crucial point, as I’ve said, and I want to wrap up our discussion by trying to emphasize what the issue is here.
At one time most psychologists believed that infants start out at the level of sensations. The experience of blind people who gained their sight later in life was taken as an indication of what the world would look like to an infant who just opened his eyes. In a famous passage that you may have heard before, William James described the experience of the infant as a blooming, buzzing confusion.
Now more recently a number of psychologists have found evidence that at least I think indicates that we begin our lives already at the perceptual level, and that perceptual learning involves fine-tuning a set of abilities at the perceptual level, not of putting sensations together. For example, newborn infants perceive depth in their visual fields, and there have been a number of different kinds of experiments that test that, and they’ve convinced me anyway. When newborn infants smell an odor or hear a sound, at least according to some experimenters, they can tell—although not with much accuracy—what direction that smell or sound is coming from because they’ll turn with their eyes to look at it. So what they’re doing is integrating data from different sense modalities automatically. They can track an object with their eyes although it’s pretty jerky at first, and they’re surprised if an object breaks up and starts moving and the parts start moving in separate directions.
Now all of these are characteristically perceptual abilities. But regardless of whether we’re born with the ability to perceive or whether it develops later as a result of the maturation of the nervous system, adults are perceivers. We don’t experience sensations as such except perhaps in extremely rare circumstances or as a result of damage to our nervous systems. For the normal adult, perception is the given.
So the awareness of entities is direct. We do not have to infer that there are entities in the world by putting together sensations or constructing entities out of our sensations by any conscious process. On the contrary, the process by which we perceive is an automatic process of integration that’s performed by the brain.
You can think of the point this way. You could take two subjects, two different perceivers, and present them with the same stimuli, and one would experience sensations and the other would experience perceptions due to differences in the degree of integration their brains were capable of performing on that perceptual data. But a brain that is capable of performing the perceptual level of integration will do it automatically. It’s not a conscious process or a consciously directed process.
I was going to give you just a simple example to illustrate what kind of integration this is, but I’m going to put that off to the question period. If there’s any unclarity about this point, go ahead and ask me.
Sensationalism and Hume
1:10:52
This view of perception as the basic level of cognition is contrasted with a view called sensationalism. Unfortunately sensationalism is a dominant view and has been for a long time in psychology and also to an extent in philosophy. Sensationalism holds that sensations are the given and that we reach the perceptual level through some kind of process of inference or construction or hypothesizing, calculation—different types of theories—but what they have in common is the notion of a conscious process.
Now in my book The Evidence of the Senses I examine this whole doctrine in a lot of its different forms, and I can’t even begin to summarize that material here. But let me by way of conclusion just give you one example of what sensationalism leads to.
David Hume, among his other faults, was a sensationalist. He held that the epistemological given consisted of isolated, detached qualities. So he asked: Why do we believe that there are entities in the world that possess those qualities as aspects of their identity? Well Hume denied that we’re directly aware of the entities as such, and try as he might at the level of sensations he could not find anything else over and above the qualities that he could use as kind of metaphysical glue to hold the qualities together. All he could find were more qualities. So he concluded that the notion of an entity is a subjective human construct with no basis in experience or reality.
Further, because he thought we don’t have any basis for believing in entities, he saw no basis for believing that an entity’s nature governs its actions—that is, he saw no basis for believing in causality. That too he came to regard as a subjective human construct.
So this is yet another way in which Hume paved the way for Kant. Hume’s sensationalism helped make it possible for Kant to get away with the notion that the world of entities acting according to causal law is a product of consciousness.
And such are the consequences of a mistaken view of perception.
End of Part Two. End of lecture. End of this mini-course.
Question period
1:13:36
Thank you very much. Thank you. I’m sorry that I ran over, but we still have some time for questions. So why don’t we start here in the blue.
Construct vs. concept
Well, the question is: I used the term—I said Hume thought that the entities, the notion of entities and of causality, were subjective human constructs. How does the notion of a construct differ from the notion of a concept?
“Construct” means it’s constructed by us rather than being something in reality that we have identified. And that’s a common usage in philosophy. When you hear the word “construct” you should smell a faint odor of Kantianism and explore further. It’s a typical Kantian term.
Is perception learned? Newly sighted adults
Yes. Okay, the question is: I said that perception is direct and automatic and yet I talked about the case with the blind people who gain their sight later in life and have a different form of experience from ours. Doesn’t that mean that, or indicate perhaps that, perception is after all learned?
Before I answer I should say that for those of you familiar with the questioner this was not a planted question.
No, I don’t think so. The problem is that these are people who have matured. Their nervous systems have matured. First of all they’ve passed crucial stages in maturation without any visual input at all, and we know physiologically that the optical system in particular does develop—neural connections are made between cells in the visual part of the brain at crucial stages of maturation during the early years. These people didn’t have a chance for that to happen. So gaining sight as adults puts them in a different position from that of an infant.
Secondly it’s likely that there has been some atrophy—that their visual systems have atrophied somewhat as a result of the lack of stimulus—and there may be other kinds of damage also. And one very sad aspect of this is that many of them never do learn how to see. So in that respect they’re clearly unlike infants. They just don’t. Many of them give up and wish they hadn’t had the operation and continue their lives essentially as blind people although they simply can’t integrate this new experience into their systems. So again that indicates that they’re quite different from infants.
Aspects of appearance and the definition of form
Right. Yes it does. The question is: I defined a form, or I gave you as a description of a form that it’s an aspect of an object’s appearance that depends on the way our senses respond to the object. Does that mean that there are aspects of the form that do not depend on the way our senses respond to the object?
That would be slightly bizarre because how can there be any aspect of appearance that doesn’t depend somehow on our senses? This is why I hesitated to call this a definition.
The point is this: in a sense every aspect of an appearance depends on the way both on the object and on our senses—depends on the whole interaction. It’s an integrated holistic phenomenon. However we can isolate by holding the object constant and varying the conditions of perception, or by varying the object and holding the conditions. We can hold the object constant and then vary the conditions of perception, or we can hold the conditions constant and vary the object.
In the first case, where the object is constant, typically we perceive a constant object but the appearance varies, as when I rotate this. You all know that in fact this ellipse is not changing. Whereas if we held the conditions of perception constant and you suddenly notice this shrinking, you would know that the object is in fact changing.
So we can say that although in a general way every aspect of appearance depends both on the object and on the perceiver, certain aspects depend primarily on—they’re primarily due to—the one or the other, and we can say at least that they vary in certain ways in accordance with the following sorts of variations.
Radical empiricism and perception as judgment
Yes they were led to that conclusion. The question is that there was a movement called radical empiricism in philosophy of which Hume—well in a sense Hume was a radical empiricist; it usually refers to William James, who is a follower of Hume. This movement said that the given was sensations and they were led to hold that perception is a kind of judgment based on sensations. What do I think of this? Do I think perception is a judgment?
No, and I think this is a crucial point that I would argue as against sensationalism. If you think it through this way: a judgment is a conceptual thing. It involves saying that something has some property—that the lectern is brown, the lectern is whatever. I talked about this a lot last time. Now that means that if you say perception involves judgment you’re saying perception already involves the conceptual level. You couldn’t perceive until you had concepts. Well then where would you get those concepts? That leaves you with only two alternatives: either we form concepts on the basis of sensations—and then you’re driven to the kind of Humean position that our concepts are arbitrary constructs—or you say that our concepts are innate. You’re driven to the Kantian position. And that in fact—so in fact in psychology you have a battle between what’s called empiricism and nativism. Often the empiricists are the Humeans, the first side of that dilemma, and the nativists are the Kantians. But either way it’s hopeless because you’re left without any foundation in reality for the conceptual level. So that’s why I think this is a real fighting issue.
Does perception depend on prior sensation?
I regard the question as: Do I regard perception as dependent on prior sensation, prior experience of sensations?
I consider that a scientific issue. It seems to me that there are two possibilities. One which I was suggesting seems to be supported increasingly by evidence is that there is, for most normal people who matured normally, no prior sensational-level experience. We start at the perceptual level and then we refine it and acquire new perceptual skills.
However other psychologists hold rather that we do start out at the sensational level in the way that William James thought, but that perception evolves out of them. In that case you could say in a sense that perception depends on prior experience of sensations. But it’s crucial to remember that if so, the dependence is a causal one. It’s that you’ve got at one stage in your life you perceive sensations; that does something to your nervous system which allows you then to perceive. But the perception is automatic. The crucial thing you have to be careful of is the claim that perception depends epistemologically on sensation—that you have to make certain kinds of inferences to get from sensation to perception. That I think is the pit, the trap to avoid.
Validating particular modes of awareness
Although you don’t validate sense perception by proving it, you do have to prove the validity of particular cases—for instance sight, or the bats’ method of perceiving objects, as opposed to clairvoyance or sixth sense.
Good question. We don’t validate perception per se, I was arguing, but don’t we have to validate that a certain mode of experience is a form of awareness or is perception? And for example we have to make the judgment that clairvoyance or ESP are not valid modes of experience.
Yes. From the inside, when we look at the world, the fact that we are aware of objects is self-evident to us. But now if someone comes along and says, “Well in addition to the five senses that you have I have this sixth sense which allows me to perceive God,” or whatever I perceive that you cannot perceive, how am I supposed to interpret that? Since I can’t look in his mind I’ve got to ask myself: What evidence is there that this person has this faculty?
Well I would ask: Can he make predictions that I can then confirm in terms of my own senses? And if you had sight in a land where everyone else was blind and you said, “Look I can see things you guys can’t,” and they said, “Why should we believe that?” you could say, “Well there’s something approaching you and you’re going to feel it in a minute.” And you do that over and over again until they have enough confirmation.
Secondly, you could specify what the causal basis for the sensory thing was. Since the faculty of sense perception has to have a causal basis, if someone comes along and says, “Last night I saw the face of God in a vision and it was a true vision,” well you have to ask: Well why? I don’t have that faculty, sorry, but I’m not sure you do either. Can you show me what the causal basis is and where are the receptors, how is it happening? And if you can’t answer that then the person is left in effect saying it will happen by magic, and that we know on first principles is not.
Representationalist image vs. Objectivist form
The question was: Could I elaborate on the distinction between the representationalist view and the Objectivist view?
Okay, let me do this at two levels. At the fundamental level the representationalist view says consciousness is metaphysically active. It could originate all of its own contents. Consciousness and existence are two separate spheres that exist in this kind of metaphysical detente. Objectivism denies that. It says that the primacy of consciousness is false. Consciousness is metaphysically passive.
Now the second level which I would hope to answer would be specifically in terms of perception: How can I distinguish the image that representationalists talk about from the form that Objectivists talk about?
Well think of it this way. Here’s the subject. Behind here is the mind’s eye. This is the subject of awareness, whoever the perceiver.
Representationalism says what we’re directly aware of is an image, and then we can say that this image corresponds to a certain object. But we don’t perceive the object. We see this thing, this image that’s in the mind, which is a distinct entity in the mind standing between us and the object.
Whereas the Objectivist view is: we perceive the object. There’s no image standing between us and the object. It’s just that—and I don’t know exactly how to represent this in a drawing—we perceive the object but in a form. However be careful: don’t think that the form is someplace between, is at this very location. The form really is a property of the entire interaction. So I hope that helps.
Color-blindness tests
Let me repeat the question. What are these optometrist tests that allow them to distinguish between those who are color-blind and those who have normal sight in those subtle cases where they have learned the same words and they don’t know that they’re perceiving abnormally?
The tests are ones I’m sure you’ve all seen. They involve a whole pattern of dots in a little circle and they’ve got numbers in a different colored dots, and the two colors are chosen in such a way that they’re just slightly different or they’re matchings that would appear differently to someone who’s got the color blindness. But they require kind of—the differences are so small or something that normally you wouldn’t notice it in real life.
3 and 4. Universals and Induction
Universals and Induction
David Kelley
Portland Institute Conference, 1988
(Corrected transcript)
Introduction and overview
0:04
Welcome to the Principal Source audio presentation of Universals and Induction by David Kelley.
Unlike other species, man has the ability to reorganize his perceptual world in such a way as to discover new relationships. The essence of this ability is his capacity to generalize and to extract universal truths that are applicable to things he does not perceive. The two cognitive processes that are fundamental to this capacity are concept formation and induction.
In Part One of this talk, Dr. Kelley applies Ayn Rand’s theory of concepts, set forth in her Introduction to Objectivist Epistemology, to the important philosophical problem of universals. Since all of our knowledge depends on our reasoning and all of our reasoning depends on our concepts, philosophers must consider what concepts stand for in reality. Conceptually, how do we take things apart into their component attributes, and by what process of integration do we put them together again into categories? The two important conceptual features that define this process are the universality and abstractness of concepts.
Part Two concerns itself with the question of justifying inductive inference, specifically the type of cognitive operation known as inductive generalization.
A well-known philosopher, teacher, and writer, David Kelley earned his Ph.D. in philosophy from Princeton University in 1975. He has since taught philosophy, cognitive sciences, and other courses at Vassar College and Brandeis University. Dr. Kelley’s books include The Evidence of the Senses, The Art of Reasoning, and A Life of One’s Own: Individual Rights and the Welfare State. He was recently featured on John Stossel’s ABC TV news special Greed and is a popular lecturer and talk-show guest. Dr. Kelley is founder and, since 1990, the Executive Director of the Institute for Objectivist Studies in Poughkeepsie, New York.
This lecture was given in 1988 at the Portland Institute conference, where Dr. Kelley presented Ayn Rand’s theory of concept formation and why he regards it as a truly major contribution to philosophy: a solution to the problem of universals.
You know, it’s often said that man differs from the animals in his fundamental relationship to his environment. Other species have to adapt themselves to their environment; man can adapt his environment to himself. That’s a kind of truism, but it’s true also of man’s form of cognition. The other species are limited cognitively to what is given to their senses. They have no ability to integrate what they perceive in such a way as to extract information that is not directly given to them. Their perceptual world is organized in terms of concrete, specific relationships, mostly spatial relationships among objects. They cannot reorganize their perceptual world in such a way as to discover new relationships. But man has that ability, and the essence of the ability is his capacity to generalize: to notice common properties and ways of acting, to classify things accordingly, and to extract universal principles and truths that are applicable to things that we do not perceive.
Now there are two cognitive processes that are fundamental to this capacity: those are concept formation and induction. I’m going to talk about concept formation today and induction tomorrow. We’ll see that these two capacities are very tightly related.
Now my text for today’s lecture on concept formation is Ayn Rand’s work Introduction to Objectivist Epistemology. Ayn Rand always said that her theory of concepts was one of her major contributions to philosophy. I certainly agree with that assessment, and my goal today is to explain why I think that assessment is accurate.
Now when I first read this book it all seemed pretty clear and straightforward to me—or, to be honest, most of it seemed pretty clear and straightforward after the second or third reading. But not having studied any philosophy at the time, I really didn’t understand its significance. And I found that this has been a common experience with many Objectivists: it’s clear what the answers in the book are, but it’s not so clear what the questions were. In fact her theory is addressed to a profound and profoundly important problem in philosophy: the problem of universals. Philosophers have been wrestling with this problem since the time of Plato and Aristotle, and I believe she essentially solved it. That’s the significance of her theory, and that’s what I’m going to try to convey today.
So what I’m going to do first is describe in some detail what the problem of universals is, and second to describe the main historical theories of universals so that we can see what problems philosophers have gotten themselves into. And having done that, having established that context, I think we can see very clearly why Ayn Rand’s work is so significant.
The conceptual level of awareness
5:26
So let us begin by talking about the conceptual level of awareness just in order to formulate the problem. Let’s distinguish the conceptual level from the perceptual. Sense perception gives us the awareness of particular concrete objects and events in our environment. We’re aware of their specific attributes and relationships. We see their specific actions. We can remember some of them and recall them later, recognize them when we see them again. All of these are perceptual-level abilities.
But human beings also have the capacity to classify things together into groups, the categories, to form concepts for general types of things and for the abstract properties that they share. The obvious reflection of this is language. Most of the words—individual words in a language—reflect concepts in one way or another. They stand for general classes or kinds of things or attributes, actions, relationships, etc. Our language embodies a vast and elaborate conceptual framework, a system of classification, and we apply this system of classification automatically to all the things that we perceive without even having to think about it.
When you take an apple from the lunch table there’s a very good chance—I’d say the probability is on the order of 100 percent—that you’ve never seen that exact apple before, that particular apple, and yet you recognize it immediately as an apple and you apply to it everything you know about apples. You know on the one hand that it’s edible and on the other hand that you can’t have a romantic relationship with it.
Now that was an example of deductive reasoning. Okay: all apples are edible; this is an apple; therefore this is edible. All of our deductive reasoning revolves around our concepts. It’s an application of concepts to particulars. But concepts are also involved in the other broad category of human reasoning, namely induction. A child burns himself by putting his hand in a candle flame and draws a generalization: flames burn, so he won’t do it again. A crucial step in the inference is his classification of the thing that burned him as a flame. If instead he had classified it as a yellow object, then he would have drawn a different generalization—that yellow objects burn—and he would have begun acting very peculiarly in the presence of buttered popcorn, blondes, gold coins, which would be very sad.
Okay, the problem arises when we ask what concepts stand for in reality. In perception, if we ask what is it in reality that we’re aware of, well the answer is pretty simple: it’s this. The podium that we all see exists in reality apart from us as a single unit. Now there are philosophical questions about the validity of our awareness of it, and I don’t want to diminish the nature of the problem—I wrote a whole book about it—but really this is a problem only philosophers have. It’s perfectly clear what the object of perception is. But this is not so clear in regard to concepts.
A concept like “dog” stands for an indefinitely large collection of animals of a certain type. Now these objects—the dog kingdom, so to speak—never stand before us in their totality, nor do individual dogs wear signs indicating their category membership. We have to assemble them into a category by some process of integration. And the same is true of a concept like “red,” which identifies a particular attribute. These concepts also stand for any number of things, any number of red things. But here we have an additional problem. When I describe a car as red I’m not describing the whole of its identity, only one aspect: its color. But a thing’s identity does not appear before us in perception all divided up into individual aspects. A thing’s identity is not like a bag of marbles—a color marble, the shape marble, etc.—which we can just pull out one at a time and examine individually. Some process of selective attention is required in order to individuate individual aspects.
Okay, and when we think about that, the same is true really for a concept for a type or kind of thing like “dog.” Even though the concept like “dog” stands for the whole animal and not just one of its attributes, still not every aspect of its identity is relevant to its membership in the category dogs. So once again there’s some active process of selective attention involved.
So the question is: how do we do this? How do we take things apart into their component attributes and how do we put them together again into categories? And what justification do we have for doing it? What basis do we have in reality? Is there something in the objects themselves that requires us to do this, or at least justifies us in doing it? Perhaps there is no basis. The doctrine of conceptual relativism says that concepts are subjective devices. Different individuals or different cultures divide up the world and classify things differently, and no one system is objectively better than any other system.
Now this is an issue of profound importance in philosophy as in science and indeed in all of human knowledge. All of our knowledge depends on our reasoning, and as we just saw all of our reasoning depends on our concepts. Any doubt about the objectivity of concepts will cast doubt on the objectivity of knowledge in general.
Universality and abstractness
11:00
Okay, now before we turn to the various theories that address the problem I’ve just outlined, I want to take a deeper look at the problem itself. There are two important features of concepts that define the problem, and I want to talk a little bit about the nature of those two features. I’m referring to the universality of concepts and their abstractness.
A concept is universal whereas its instances or members are particular, and a concept is abstract whereas its instances or members are concrete, specific, determinate. Let’s make sure we understand what these terms mean.
And let’s begin with universality. Imagine a child learning a language, his first language. Even when he gets the idea that the sounds he hears his parents making have the remarkable property of referring to objects—and this is a major achievement on his part—even when he gets that idea he still has to learn that they don’t refer to objects in the same way. If the child points to an animal wandering around the house and indicates that he wants to know what it’s called, he might get one of two answers. He might be told “That’s Spot,” or he might be told “That’s a dog.” He hasn’t understood these answers until he understands that “Spot” is a name for this particular animal, whereas “dog” is a name not only for this animal but for any other animal of the same type in the neighborhood, any other animal of that type that he will ever encounter, any other animal of that type regardless of whether he ever encounters it at all.
So the first and obvious point about a concept then is that it is universal. It refers indifferently to a number of different individual particular objects.
Now these objects constitute a set, and if it were just a matter of putting things together into sets there would be no particular problem. We can group things together to some extent even at the perceptual level. We can attend to the forest as well as to the trees. But unlike a perceptual grouping, a concept is open-ended. It includes not just the particular instances that one happens to have encountered but all objects of a certain type. So we can’t learn a concept by memorizing its instances. We have to learn some kind of rule or something like a rule for determining what is included and what is excluded from the concept.
Moreover we are not usually concerned with those objects as a set in their totality. When I describe something as a dog I don’t care how many other dogs there are. I certainly don’t know how many other dogs there are. What I’m thinking about is this dog, and what I’m doing is identifying its nature. But the point is that what I’m describing also seems to be present in any number of other animals. Okay, in describing any of these other animals as a dog I’m attributing to them the same nature, the same characteristics. The question is: what is this nature? What is this thing that many different objects can have in common?
And that brings us to the next point: the abstractness of concepts.
The idea that the instances of a concept have something in common suggests that they are identical in some way. But they’re not identical. When we form a concept putting objects together into groups, the objects are similar but they’re not identical. The concept “red,” for example, includes a range of perceptibly different shades of red. The concept “dog” includes animals that differ in every respect. Obviously they differ in attributes that are not relevant to their biological identity—attributes like whether they’re owned by somebody or not, whether they know how to fetch a stick. But all dogs differ also in those attributes that are relevant to their biological category. All dogs have hair, but dogs differ in the shape and color of their hair. Dogs are social animals, but as you know if you’ve ever had a paper route or done door-to-door work, dogs differ a good deal in their sociability. Dogs have presumably a common DNA structure which biologists will someday tell us about, but even so no two dogs except identical twins have exactly the same DNA.
So in general the instances of a concept are not identical; they are merely similar. A concept for an attribute designates a particular dimension of similarity, and a concept for a type of entity like dogs identifies a number of dimensions of similarity: hair, shape, etc. Nevertheless, despite the fact that the instances are merely similar, we treat them as if they were identical. When I say that dogs bark, I’m treating all dogs as if they were identical. I’m ignoring all the differences among them. I’m treating them as interchangeable units. And when I say that they all bark, using the concept “bark,” I’m ignoring all the differences among the sounds that they make.
In order to do that, however, I obviously have to ignore the differences among these animals—the differences that are perfectly obvious to me at the perceptual level. I believe this is one of the reasons—this fact of ignoring differences is one of the reasons—why people sometimes resent being classified. People sometimes say “Don’t pigeonhole me” when they’re described under some category such as conservative or housewife or engineer or athlete or whatever. The classification presupposes that, or seems to convey that, the person is identical with anyone else in that category and ignores the unique character of the person. But be that as it may, the fact that we can integrate individual objects into a single category and treat them as identical allows us to present in the form of a single proposition—to use the same example, “Dogs bark”—a fact that would otherwise require many different propositions: poodles yap, boxers growl, Dobermans snarl, etc. And even that is stated conceptually; I use the narrower concepts poodle, Doberman, etc. Ultimately, to avoid concepts I would have to say this dog makes this noise, that dog makes that noise, and so on for all four, seven, 47 billion or however many dogs there are in the world. By treating instances as identical a concept allows us a tremendous cognitive economy. It allows us to reduce massive information to a single unit.
Now let’s pause here to make sure we understand the difference between universality and abstractness, and I think the clearest way to summarize the difference is in terms of a philosophical distinction between what’s called numerical difference and qualitative difference. If I take two coins of the same denomination—these happen to be American quarters—these are numerically different; they’re two objects, but they’re qualitatively identical, at least as far as perception goes. They’re exact duplicates of each other, right? Whereas a quarter and a nickel or any other coin would be qualitatively different as well as numerically different.
Okay, now a concept is universal because it includes an open-ended range of numerically different objects, numerically distinct objects. A concept is abstract because it includes an open-ended range of qualitatively distinct objects, objects that are merely similar but not qualitatively identical. A concept is universal because it unites a numerical diversity. A concept is abstract because it unites a qualitative diversity. And I think you can see that abstractness is therefore the more fundamental of the two features. The reason that we can integrate numerically different objects into a single mental structure is that we can ignore their qualitative differences—that is, we can unite their qualitatively different things. So the concept is universal in virtue of being abstract.
So the problem of universals in my opinion really should have been called the problem of abstractions.
Now this problem of abstraction has two components: a metaphysical and an epistemological aspect.
Metaphysically, as we saw, the instances of a concept are similar but they’re not identical. But what is similarity? Ayn Rand describes it—and this is a fairly standard description—as: similarity is the relationship between things that have the same attribute in different degree. But what is this attribute that things can have in different degree? Two different shades of red possess the same attribute of redness, but what kind of thing is this common color, redness?
Suppose I said I just bought a car and it’s red, and you said “Oh, what shade?” I said “No shade, no particular shade, it’s just red, it’s bare red.” Or suppose I said I have a dog and you said “What breed?” and I said “No breed,” and you say “Oh, you mean it’s mixed,” and I said “No, it’s neither purebred nor mixed breed; it’s just a dog, standard-issue dog.” Well clearly this is impossible. An attribute such as red or a general nature such as dog has to exist in some particular concrete determinate form. So then the question is: what’s the metaphysical status of the abstract attribute or kind or essence such as “dog,” and what relationship does it bear to the particular concrete forms in which it exists?
The epistemological problem is to explain how, by what process, we ignore the differences among things in order to form an abstract concept. If you show me two dogs and tell me to ignore all their differences, all the respects in which they differ, well I’m going to have to ignore their shape, their color, their size, the length of hair, their way of behaving, their gender, the degree of docility, their preferences in food—in short I’m going to have to ignore all of this, and what’s left? Nothing. I’ve thrown out every attribute that they have. If I have to ignore all their characteristics in order to form a concept then the concept will be completely empty. So a theory of concepts has to solve this epistemological problem: explaining how it is that we can ignore the differences without ignoring the object as a whole.
Realism
21:24
All right then, let’s turn now to the classical theories that try to solve these problems, and I’m going to start with the theory of realism.
Realists hold that our concepts mirror the structure of attributes in reality. If our concepts are universal and abstract, that’s because something out there in the world is universal or at least abstract. Concepts refer to real universals: Redness, Humanity. Okay, these are always spelled with a capital, always capitalized to indicate that they have a special metaphysical status.
Now there’s an obvious problem here in saying that there are these real universals. If Humanity, for example, exists as something outside the mind, it must be some definite particular thing. How then can it be shared in common by all the particular human beings? Okay, if humanity is what’s essential to you and me and all the other human beings and it’s a single thing that we both have, how can you and I and all the other individuals be different individuals? So perhaps we all share a slice of the universal. This is a theory that Plato once proposed or entertained. Okay, you might have rationality, I might have animality, right? But no, that doesn’t work. All of us are fully human; we’ve got the whole shot, right? So how can a single entity be shared, literally shared, by many different particular entities?
Well this problem led realists in two different directions. Extreme realists, of which Plato is the best example, said that the universal Humanity is not in any of the particular human beings. It’s a separate entity existing in a realm beyond particulars. Okay, the world he called the world of Forms. Particulars that have the same attribute such as red, or belong to the same type such as human beings, do so because they relate to the same universal—Redness or Humanity—existing in a different world. The universal is shared by all instances in the sense that they all bear the same relationship to it, in the way that many mirrors arranged around a candle would all have an image of the candle in them; they would all reflect the same one entity.
Now the other approach, taken by philosophers who were unwilling to posit a supernatural realm, is called moderate realism, and the best example of this is Aristotle—at least Aristotle is normally interpreted. Moderate realists deny that universals exist in some separate realm. Indeed moderate realists deny that anything outside the mind is literally universal at all. But they say that particulars do contain, as intrinsic elements in their identity, abstract essences and attributes. My essence, my humanity, is distinct from yours, but because our essences consist of abstract properties they are qualitatively identical. My humanity and your humanity are like the two quarters that I held up before: numerically distinct but exactly alike in the two of us. So in that sense it’s the same attribute that we share.
And to get a better feel for the metaphysics here let me give you an analogy. Imagine several cars of the same make as they leave the factory. Now to all appearances these cars are identical, but once they’re purchased and driven for a while they each acquire their little individuating elements. Okay, one gets souped up, another one gets fuzzy dice hanging off the mirror, another one gets a bumper sticker, and so forth; they get customized. Well, for the realist, when two objects of the same type come off the metaphysical production line, so to speak, they are identical; they’re just red or just human. But then they get particularized to a specific concrete determinate form of the attribute when they’re embodied in matter, when they come to exist as material objects.
Okay, so if we take two shades of red they differ in their specific shade. To perception they are merely similar, not identical, and yet the similarity is a product of two factors. We can analyze it into two factors: there’s an element of identity in that—the presence of the same abstract feature, redness, in the two of them—and there’s an element of difference, an individuating element that makes the two shades different.
Okay, the point to stress here is that the abstract feature exists as an intrinsic element in the identity of the color, and in the same way my humanity, the abstract essence, exists as an intrinsic feature of me, and same with you.
Okay, now realists explain the cognitive process of concept formation in terms of these abstract properties and essences. The properties exist out there in reality as intrinsic components of objects, so concept formation is like perception. We are directly aware of what is there before us. In perception it’s the whole object; in concept formation it’s the abstract property or the essence. Now some preparatory process of sifting and comparing similar objects may be necessary before we can isolate the abstract feature that they have in common. We may have to focus our attention in a certain way in order to perceive, in order to be in a position to see the abstract essence or property, just as at the perceptual level you may have to train your attention a certain way in order to see a bone fracture in an X-ray. But once you’ve trained your attention and focused your attention in the right way, then you just see it. And the same, says the realist, is true of concepts. Once you have trained your attention in a certain way you just open your mind and there it is: the abstract property that the concept refers to.
Okay, so in this way, as you can see, the realist can explain why the instances of a concept are treated as identical: it’s because they are identical. They all share the same identical property and we are directly aware of that property.
Unfortunately this is exactly what makes realism untenable. The realist doctrine of intrinsically abstract attributes runs into all sorts of problems as soon as you start exploring it in depth.
One classical problem is the problem of borderline cases. On the color spectrum red shades into yellow. Now if what is common to all shades of red is the feature Redness and what’s common to all shades of yellow is the abstract feature Yellowness, then at some definite specific point redness must end and yellowness must begin. But it seems arbitrary to stipulate any one point where that happens. And if you try to solve the problem by saying well there’s orange in between—okay, so there’s red, orange, and yellow—but then we have the problem about red and orange. I mean you’re not going to solve it by getting narrower concepts.
Okay, second problem: how many distinct attributes does an object have? Just staying with the realm of color, is an object’s color a single abstract attribute, redness, or is it the three attributes that psychologists study—hue, brightness, and saturation? Does it have three abstract attributes, or does it have the many elements of color that a painter can recognize and reproduce, including all the surface texture qualities, sheen, texture, and so forth?
It’s clear that the realist has no independent metaphysical reason for deciding how many attributes there are. We can’t see them. What we see are the specific determinate qualities. The abstract attributes are posited merely for the sake of explaining our possession of abstract concepts. The realist is saying in effect: our concepts are abstract; in order to be objective they must correspond to something in reality that is intrinsically abstract; so for each concept there must be a separate abstract property.
Now this is the essence of what Ayn Rand called intrinsicism: the idea that to be valid our concepts must reflect in a passive way a structure of properties that is out there and that can be grasped by a perceptual-type mode of awareness. Realism is the quintessential form of intrinsicism.
Now I’m not going to go into everything that’s wrong with that, but this is one of the fundamental points that Objectivism insists on: that intrinsicism is wrong. Concepts can be objective; knowledge can be objective without being passive, without passively mirroring what’s out there. I’ll be happy to expand on the reasons for that in the question period if anyone wants to pursue it.
Nominalism
30:28
All right, let me turn now to nominalism, the other classical alternative in philosophy.
Nominalists deny that abstractions exist outside the mind. They deny that any realm of entities exists beyond the world of particulars—in other words they deny extreme realism—and they deny that a particular thing’s nature is bifurcated into a concrete and an abstract constituent. For the nominalists a thing is determinate through and through.
The 18th-century British empiricist philosopher David Hume, who was a nominalist, put it this way: the precise degree of a line—that is, a precise length of a line—is not different nor distinguishable from the line itself, nor the precise degree of any quality from the quality itself. That is, a quality just is the particular concrete degree of the quality. If you ignore its specific character there would be nothing left.
So the nominalist does not analyze similarity into an element of identity and an element of difference. Two objects may happen to have exactly the same determinate quality, the exact same shade of red, but that would be rare. In the more common case where the two objects are similar but not identical, they are just similar, period, and similarity is a metaphysical primary. It cannot be analyzed or broken down any further.
Well, if there’s nothing universal or abstract in the world outside the mind, then concept formation—the epistemological process—cannot be a matter of direct awareness. The universality of concepts and the abstractness of concepts must be produced by some cognitive process. As John Locke, one of Hume’s predecessors, put it: the features of general or abstract and universal belong not to the real existence of things; they’re not out there in the world, but they are the inventions and creatures of the understanding, made by it for its own use.
Well, how does the mind do this? By what process does a nominalist think we generate abstractness and universality?
Well, as with realism, nominalism divides into two camps, one more extreme than the other.
Extreme nominalists regard concept formation as a completely arbitrary process. We divide a thing’s nature into attributes and group things together into categories any way we wish, and any way is just as good as any other. There are no objective constraints. I could group together a tree, a truck, and the city of Toronto and call them all “grucks,” and I would have a concept. Now of course the only thing that those three items have in common is that I chose to call them by the same name. But that, says a nominalist, is true of all of our concepts. The only thing that red things have in common is that we chose to call them by the same name, “red.”
Now the obvious problem with this view is that it makes it impossible to identify new instances of a concept. Suppose I held up a new item and said “Is that a gruck?” I just taught you this new concept; now is this an instance of the concept or not? Obviously you can’t tell, since I put together the concept by whim. You have no idea what my whim is going to deliver in the way of a judgment about this object, nor do I. It’s a completely arbitrary process.
Okay, and the same would be true for all concepts on the nominalist view. Extreme nominalism is a clear case of what Objectivism calls the primacy-of-consciousness view. It regards concepts as wholly subjective, and since all of our thinking depends on our concepts the implication would be that thought as such is subjective. We’re going to deal with a particularly striking case of this tomorrow in regard to induction. But let me leave the extreme subjectivist in his subjectivism for the moment, in his little world, and turn to the more reasonable version of nominalism, which is called resemblance theory.
Resemblance theory says that we group objects together on the basis of similarity and we use a common name for the group. We identify new instances by determining whether they are similar to the instances that we know already. So the moderate nominalist, or the resemblance theorist, is trying to provide some kind of objective basis for our concepts.
The classic example of this view is the one put forward by the third in the classical empiricist trio, Bishop Berkeley, who came in time between John Locke and David Hume. Berkeley said, using the concept of triangle as his example, that when we employ the concept and reason about triangles in geometry we are not thinking about some abstract essence, triangularity. What we’re thinking about is a particular triangle such as the one that we draw on the blackboard to serve our demonstration. It’s just that we realize that the conclusions we reach about that particular object apply to a range of similar objects, and we extend the same conclusion to other particulars but without our ever having in our mind or seeing in reality some abstract essence, triangularity.
And another version of this view has been recently developed in psychology by psychologists who have studied a phenomenon they call typicality or prototypicality. What they found is that if you take a normal concept that people use for a biological type or a type of artifact—table, chair, etc.—some instances are more typical or representative of the concept than others. Now this may not seem like a very surprising fact to spend all that research money on, but there are some interesting results from it, such as that if you show people objects and ask them to classify them they can classify the typical ones faster than the atypical ones. They can classify a beagle as a dog faster than they can classify or recognize a Pekingese as a dog. And there are various other effects that this seems to have on aspects of cognitive processing. So it’s a real, as far as I can tell, a very real and somewhat interesting phenomenon.
The theory that it has generated, however, is very much like Bishop Berkeley’s. The idea is that a concept is organized around a prototype or a prototypical instance plus a rule of similarity that says: include anything similar to that prototype. So the concept is this thing, this prototype, this concrete instance, and everything that’s similar to it.
Now the problem with this theory emerges as soon as we look more closely at the role of similarity.
First of all, it’s not enough that an object be similar to the prototype in order to be included in the concept; it has to be sufficiently similar. If you showed me a demonstration in geometry regarding a particular triangle drawn on the board and said “And this is true also of all similar objects,” well a square is somewhat similar to a triangle but it’s not similar enough to fit the theorems. So we need some way of delimiting the degree of similarity. An object has to have a certain degree of similarity to the prototype to count as an instance of a concept. This is not an insuperable problem; various theories have been proposed for how we could mark off the degree of similarity. But it is a problem.
A more important problem is that when two things are similar they are similar in some particular respect. If we don’t specify the respect in which objects have to resemble the prototype then almost anything might be included in the concept. Suppose I try to teach you the concept of red by showing you a fire engine—at least where I live the fire engines are red—and I said okay, that’s a prototypical red thing; now include everything that’s similar. All right, you go out and you see a garbage truck and identify it as a red object because it is certainly quite similar to the prototype that I showed you. Obviously I haven’t given you the right information. I haven’t conveyed to you that you’re to include everything that is similar to this object in the respect of color.
But the problem is: how do we specify that respect? How do we represent the particular respect in which things have to be similar in the concept? I conveyed it to you by using the concept of color. Okay, so we use concepts to identify a particular respect in which things might be similar. But when we’re forming concepts obviously we don’t have concepts like “color” available to us. So a theory has to be able to explain how someone who doesn’t yet possess concepts can individuate and isolate a single dimension of similarity.
Okay, so a theory of concepts cannot merely invoke similarity to explain how we group things together. It has to explain how the mind isolates a particular dimension of similarity and a particular degree of similarity along that dimension. Because the resemblance theory treats similarity as a primary, something not subject to further analysis, it does not provide those explanations. And so the resemblance theory, despite its good intentions, tends to collapse into the more extreme subjectivist type of nominalism.
Analog and digital aspects of concepts
40:33
Now before I leave the subject of nominalism I want to comment on one final point, one final issue which I call the analog/digital issue.
Now for those of you who have worked with computers I know these will be familiar terms, but let me explain them anyway. Think of the difference between an analog and a digital measuring instrument such as a bathroom scale. On the analog scale weight is measured by a needle that moves across a set of numbers. Okay, on a digital scale the weight is represented by a number that appears and you read off the screen. On the digital scale any weight that falls between, say, 164.5 and 165.5 will generate, will be represented by, the same number 165, assuming this measures only to the accuracy of a pound. On the analog scale, however, if you are exactly 165 pounds the needle will be right on that mark; if you’re 164 and a half it will be slightly to one side, and same on the other side. So an analog representation varies; it represents its different instances in a varied way. It varies along with the instances in correspondence to the things that it represents, whereas a digital representation uses a single item—in this case a number—to stand for all the differences. It ignores all the differences among the things that it stands for.
Okay, now concepts I believe have both an analog and a digital aspect. On the analog side we do make use of typical instances. We do distinguish between the things that fall under a concept between the typical and the borderline cases, and this has various consequences in our reasoning. But concepts also have a digital aspect. A Pekingese may be an atypical dog but it’s still a dog. The statement “This is a dog” is completely true; it’s 100 percent true, same as it is for a beagle.
Okay, as we saw earlier a crucial part of concepts is that we treat the instances as identical. So a theory of concepts has to explain both features.
Let me give you a little example of both features that work in a genus-species hierarchy. If I showed you various pictures of robins and I asked you which is a typical robin, you would have no trouble picking it out, and I think we’d all agree the albino one is not typical, the one that’s two feet high is not typical; the typical one is little robin redbreast. Okay, so in that respect we are alert to differences among robins. But now suppose I ask you which is a more typical bird, a robin or a penguin. Well again you’d have no trouble answering; you’d say it’s a robin. But notice now you are treating the category of robins as a single unit. You are no longer taking account of differences among robins; you’re treating them as all identical interchangeable instances of that category.
Okay, a more interesting example of the role of these two aspects of concepts, and it shows a little more why I’m drawing your attention to it, is in ethics. Consider the virtue of independence. Now I could define this virtue in the abstract. I could state principles about the type of actions that it entails, and all of this is very important. But think how valuable it is to have in addition a clear example, an ideal case of independence, a prototype—a Howard Roark. There is a kind of thinking we can do, an application of the principle to our own behavior, that it helps us to have this sort of prototype for. Indeed that is one of the functions of art: to embody abstractions in concrete forms. Okay, of course Howard Roark is not an average instance; he’s an ideal instance, but nevertheless it’s still a prototype.
Okay, so that’s a way in which the analog aspect of concepts is very important in ethics. On the other hand it’s not hard to find examples of the danger of thinking exclusively in terms of typical cases or stereotypes: racial and ethnic prejudice, for example, or tribalism—what Ayn Rand calls tribalism in general. I think as an example of this it’s a phenomenon whose basis is at least partly epistemological, in a kind of—well she calls it the anti-conceptual mentality. It reflects the way of thinking in which we see things as typical; we regard, we focus on what’s typical and don’t grasp what is common. We don’t fully see the underlying identical principle that all human beings share, for example. We don’t understand—a typical tribalist doesn’t understand—that people have rights qua individuals, human beings, okay, and that regardless of racial or tribal or any other specific characteristics, individuals, all individuals, have the same rights. So in that respect it is crucial to think in terms of, to ignore all the differences among people, to focus just on what is in common.
Okay, now as I said a theory of concepts has to be able to account for both aspects of concepts, and that I think is one of the great virtues of Ayn Rand’s theory, to which let us now turn.
Ayn Rand’s theory of concepts
46:04
Ayn Rand would agree with realists that our concepts have a basis in reality. She would agree with nominalists that what exists outside the mind is fully concrete, specific, and determinate. And she would agree with nominalism that concepts are formed by noticing patterns of similarity among objects. But unlike nominalists she does not treat similarity as a primary, as a cognitive starting point. The starting point is difference.
In perception, where concept formation begins, we discriminate objects from their backgrounds. Discrimination means the awareness of difference. In vision when we see a figure we differentiate it from its background on the basis of differences in color, depth, and other attributes. So if we were to observe two chairs, for example, our perceptual attention would be drawn to their differences. Each would be part of the background from which the other one is discriminated.
The awareness of similarity arises when we perceive two chairs alongside something else such as a table. Each of the chairs differs from the table, the salient difference being in shape. Each of the chairs also differs from the other chair in shape, but the shape difference between the two chairs is not as large as the difference between either one and the table. It is this less-than-complete difference between the two chairs which is brought out by the greater difference between either one of them and the table that provides the foundation for seeing the two chairs as similar. It is that that’s the context in which we grasp the similarity between the two chairs.
Now notice, without going any farther, that this solves the two problems of similarity that I mentioned before. First there’s no problem about determining how similar the two chairs have to be. They have to be more similar than either one is to the table. It depends on the context, but the context guarantees that there will be a certain limit within which—it guarantees, it gives us a standard for telling what is a sufficient degree of similarity.
Okay, and secondly there’s no problem of isolating the respect in which the two chairs are similar. The respect in which they are similar, the respect in which they’re perceived as similar, is determined by the respect in which they are differentiated from the tables. Okay, it’s because we distinguish the chair from the table in shape that it is similarity of shape that our attention is drawn to between the two chairs.
Now when we see the two chairs as similar it’s because the difference between them is less than the difference between either one and the table. The two chairs are seen as differing quantitatively from each other, whereas either one, either chair, differs qualitatively from the table. If you look at the two chairs you could see how one could be transformed into the other by merely quantitative changes: lengthening the back a little bit, lowering the seat or whatever. Whereas neither one could be transformed into the table by merely quantitative changes; it would take a qualitative change.
Okay, similarity is therefore a quantitative relationship. It’s a difference in degree, in specific measurement along some dimension. Okay, at the perceptual level the dimension could be shape or color, tone in hearing, warmth in the sense of touch, or whatever. But at higher levels of abstraction the same principle applies. In forming the concept of government, for example, we notice that certain social institutions differ qualitatively from certain other ones such as the family, trade union, church or whatever, in that they use force. Force is the means by which they cohere, as opposed to custom, tradition, contract, persuasion, and other types of interaction. Okay, but within the category of governments they differ merely quantitatively in the degree to which they use force, the scope to which their use of force intrudes into individual lives, the number of people who have control over the disposition of that force, etc. That is, they vary along quantitative dimensions.
The awareness of objects as similar then is the awareness of them as commensurable. Each could serve as a unit by which the others are measured, in the way a specific length such as an inch or a foot is a unit for measuring all lengths. The measurements are relationships among the objects, quantitative relationships, each of which possesses and is seen as possessing its own specific degree on that dimension. And that is why she uses the term “units” to describe members of a group of objects put together on the basis of their similarity.
Okay, it’s as if we perceived this group of similar objects, or rather the relationships among these objects, in the form of a simple equation: b equals k a, where b and a are two units, two members of a group of similar objects, and k indicates the quantitative relationship. For example if this is a length of one foot and this is a length of one inch, k is twelve. Now yet at the perceptual level we’re not aware of the actual cardinal numbers. Okay, and indeed there are many dimensions along which we don’t yet know how to apply a cardinal number system. We have to deal with ordinal numbers first, second, third; we’re ranking them just in a certain order of priority. But that is still a quantitative set of relationships. And so this little example illustrates the phenomenon; that is the way of symbolizing the form in which we grasp the relationships between similar objects.
All right, now so far what we have is a group of similar objects. We don’t yet have a concept. The awareness of the units does not yet involve anything universal. We’re dealing with a finite set of particular objects that happen to be before us perceptually or that we can recall from memory. And we don’t yet have anything that is abstract. We’re aware of each object as having a specific characteristic, and we’re aware of the specific set of characteristics, a specific set of relationships among objects in virtue of those characteristics. But those relationships that I’ve symbolized that way are the stepping stone to a concept.
We form a concept when we realize that any number of other objects could have any number of other specific measurements in relationship to these particular objects. We realize that the specific chairs can be ordered along a certain dimension, a shape dimension, and that other items could occupy other positions on that dimension. In effect we notice that the specific quantitative relationships that we’ve grasped in this form are simply special cases of a more general set of relationships that we can represent this way.
Now my apologies to the mathematicians for the bizarre look of this equation, but there’s a method in my madness here. When we form a concept we realize that this particular quantitative relationship between the two units that we happen to perceive is only an instance of a whole pattern of possible quantitative relationships that objects could have to a given unit. Right, that is, if this is again length, we realize that an object could have any length and still be similar to the things that we have before us. And because an object could bear any quantitative relationship to a given object, any number of objects could be so related.
So we realize that we form a concept by expanding the set of concrete relationships that we grasp, unifying it into the idea of a single dimension on which any number of objects could have any number of places or specific measurements.
Okay, again I want to insert, I want to remind you that I don’t intend this to be taken literally—that the child is thinking in terms of arithmetic and algebra. These are simply ways of representing the essence of the process. In Ayn Rand’s words, I quote: “If a child considers a match, a pencil and a stick, he observes that length is the attribute they have in common but their specific lengths differ. The difference is one of measurement. In order to form the concept ‘length,’ the child’s mind retains the attribute and omits its specific measurements. Or more precisely, if the process were identified in words it would consist of the following: length must exist in some quantity but may exist in any quantity. I shall designate as ‘length’ that attribute of any existent possessing it which can be quantitatively related to a unit of length without specifying the measurement.” Right, without specifying the quantity.
Another particular example of length in her view sets the pattern for all concept formation. We realize that membership in the similarity class, the class of units, can be expanded indefinitely by including anything that can be quantitatively related to units already included. Okay, on the principle that all members of the class must possess some particular measurement but may possess any. We recognize the commensurability of units without specifying the particular measurements, and the result of that process is a concept.
That process of measurement omission is a concept. A concept is a mental integration of two or more units into a new mental structure that allows us to treat those units as identical. Now the specific units are still similar; they’re not literally identical as the realist holds. But because their differences are differences in measurement, omitting their measurements is a way of ignoring their differences.
Now we don’t ignore them entirely. We certainly don’t treat them as if they didn’t exist. Okay, on the contrary the specific measurements do exist and we have to be aware of them in order for us to grasp the existence of a dimension along which things can be ranked. Okay, the point is simply that we grasp the dimension as something on which something can have any number of places. Okay, indeed since we are aware of the specific measurements and because of that it’s natural for us to choose a central or prototypical instance to serve as a representative of all the members. So in this respect her theory I think does account for what I’ve called the analog aspect of concepts. But it does also account for the digital aspect. We omit measurements on the principle that they have to exist in some degree but may exist in any, and that is true equally and identically of all the units.
Now once a concept is formed it serves as a new mental unit represented by a word. The concept functions like a mental file folder in which we retain information about things of a certain type. That’s what allows the cognitive economy that I mentioned earlier. We reduce an unlimited number of numerically and qualitatively distinct objects to a single unit, and we can go on to group these new mental units together into higher-order concepts. For example once we have formed the concept “chair” by omitting the specific measurements of chairs we can go on to form the concept “furniture.” We notice that the differences among tables, chairs, beds, etc., which we first perceived as qualitative differences, at this higher level of abstraction can be regarded as quantitative differences. Okay, as against the qualitative difference all of them have to non-furniture items like silverware, lamp fixtures, wall fixtures, etc. So what was first perceived as a qualitative difference between tables and chairs now is seen as a mere quantitative difference in relationship to the qualitative difference between furniture and other artifacts.
Okay, and in that way as we expand the scope of our concepts and reach increasingly high levels of abstraction we are reducing qualitative to quantitative differences.
Okay, now in her book Introduction to Objectivist Epistemology Ayn Rand goes on to derive a number of further consequences from her view of concept formation: implications about definitions, about the contextual theory of knowledge, about axiomatic concepts, and many other issues. And to my mind as a philosopher the appeal of her theory lies partly in its fruitfulness, its fertility in suggesting all these further implications. But I focused only on the foundations of the theory because that’s the first and the crucial test that a theory of abstraction, of concepts, has to pass. It has to deal with the basic problems that arise from the fact that concepts are universal and abstract.
Okay, so by way of summary and conclusion let’s compare her view with the two classical alternatives that I mentioned at the outset.
A concept on her view represents an abstract attribute such as length or a general type of thing such as chairs. The attribute or the type does not exist as such, as abstract, independently of the cognitive process of measurement omission. What exists outside the mind are the particular objects with their specific relationships, qualitative and quantitative, and the potential that they have to be ranked in order along various dimensions.
So Ayn Rand’s approach does not treat abstractions as intrinsic features of objects as the realist does, but she doesn’t treat abstractions as a subjective phenomenon either. Okay, a result of the mind grouping things together more or less arbitrarily on the basis of an undifferentiated awareness of similarity as in nominalism. Abstractness is objective. It’s a product of two facts: one about the world and one about the human mind. It’s a product of the metaphysical fact that the particular natures of things are quantitatively related along various dimensions, and it’s a product of the epistemological fact that the human mind has the capacity to identify those dimensions by omitting specific measurements. And that’s why, as I said, I regard it as a truly major contribution to philosophy: a solution to the problem of universals.
Thank you.
Questions and answers, part one
1:01:45
Okay, you all appear still to be awake and we have some time left for questions, so I expect you have some.
Yes.
No, in the sense in which you intended I think the answer is no. I would regard difference as a metaphysical primary—that is, the existence in reality of a relationship of difference. That’s a primary. Now particular differences might be broken down into other differences. You know, the difference between capitalism and collectivism can be broken down into a number of component dimensions. But the phenomenon of difference as such I think has to be a metaphysical primary.
I think the reason I think that is that it’s clear to me that the law of non-contradiction has to be axiomatic, that it can’t be a derivative conclusion, because it is presupposed by any process of reasoning that might generate a conclusion. The law of non-contradiction has to be implicit in the law of identity, and it has to be grounded somehow in the identities of things. But what non-contradiction says is that a thing cannot be A and not-A at the same time in the same respect. So the implication to me is that the phenomenon of difference is implicit in the very existence of identities in the world. That part of what it is for things to have identities—which is the bedrock inescapable fact—part of what it is for things to have identities is for there to be different identities in the world. Part of what it means, the essence of what it means to be something, is to be not other things. So difference is implicit there and so I would regard it as—well in fact it’s an axiomatic concept.
Now I want to emphasize that I’m not saying that we can deduce similarity from difference. The relationship is one of the order in which they have to be grasped. Similarity has a positive nature. It’s not simply the lack of a larger difference. You know, I said we grasp similarity as the smaller difference against the larger one, but similarity doesn’t mean just the lack of a larger difference. It’s a positive relationship of commensurability. It’s an actual relation, a numerical quantitative relationship between two specific characteristics. It’s just that in order to grasp that the cognitive context has to be prepared by first noticing certain differences.
Okay, other questions.
Yes. In pre-conceptual cognition or at the perceptual level, how is it that the essential defining characteristics of existence are grasped in order to form the concept or the beginnings of a concept? How is it the essence of something, that the essential defining characteristics of things are grasped without the concepts themselves?
Well typically they aren’t. That is, we normally when we start forming concepts we do so by grasping the more superficial similarities among things that are going to be instances of the concept, and then as we have the concept and learn more and more we learn more and more fundamental aspects of those objects, and it’s those that we then use to define, that we treat as essential and use to define the concept. For example, to use Ayn Rand’s example, a child forming the concept of man, of human beings, persons, would not begin by noticing rationality versus other modes of cognitive functioning. He would begin by noticing shape and certain sounds they make, a certain way of interacting with the child, and so forth. It’s only at a higher level of knowledge that the child can grasp a feature like reason.
Now does that answer your question? Because it sort of does, but it leads into something related in terms of if the child is forming the concept of man by the perceptual evidence and yet what it is to be man then it isn’t really a perceptual thing. I mean rationality as you say, if that’s man’s defining characteristic and that’s not perceptually obvious yet, that’s really the key to forming or to holding the concept of man. So that when the child is recognizing that man is this sort of thing as opposed to the dog or whatever or the table and so on, there’s something about these beings or existence that make them particular. It’s not qua man that he’s understanding them to be related. So I’m trying to—how is he forming the concept of man if it relies on a non-perceptual similarity?
Okay, well be careful. This is a crucial point about concepts. Be careful not to identify a concept with the essential property that we use in defining the concept. The essential property is only one of all the properties common to the members of a concept which we select for the cognitive purpose of giving a definition, of reducing a mass of information yet further to a single proposition. Okay, but that’s not to say that the other attributes are unreal or are not part of the concept. They are part of the concept. So what it means to be human is not just merely to be rational but rationality plus animality. It’s to have a language, to invent, to use technology, to survive by producing rather than finding or taking what you need, to need art, okay, to be capable of laughter. All of these features that human beings have in common are part of the concept of man. It’s just that because reason is what underlies all of them it’s the best one to use as a definition so that we can summarize all that information and condense it in the statement “Man is a rational animal.” But that does not exhaust the content of the concept. The content really is all the objects out there with all of their properties.
So there are many concepts where the essential feature is one that can’t be perceived. I mean rationality is one; names of diseases—many of those are imperceptible, at least essential properties. The symptoms that may be perceptible but the cause isn’t. And when we get to social and economic systems, you know, we’re way beyond the range of what can be grasped perceptually.
Okay.
Yes. Dr. Kelley, Ayn Rand said that in developing a concept we isolate certain attributes as fixed across the particulars of the concept and allow other attributes to vary. You gave examples where length was allowed to vary. Are there not also other concepts where length is the fixed attribute and it’s other things like shape that vary? However, say in a clothing store for example, everything that is on the rack that is called “small,” but the length is the fixed attribute and its colors, shapes and other attributes—
Okay, that’s an interesting point. When we classify objects we normally try to classify them by essential principles. So in clothing, I mean if I said classify clothing, since clothing is a human artifact the essential property is going to be its function. So you would classify it by the different functions they perform such as covering what part of the body and then what design intended for what circumstances. So you classify into coats, jackets, trousers, blouses, skirts, etc. But of course you also can classify by other attributes such as color, length, style, and a clothing store, to serve the function of serving its customers, would very naturally classify by what in a generalized context would be non-essential but in this context is essential, namely length, because a given shopper doesn’t care about the other lengths.
But notice that classification is not the same thing as concept formation. Concept formation begins by classifying things together into a group, but it then has the other step of measurement omission that produces a new concept that is a new mental unit. Now once we have a stock of concepts we can group things together conceptually and not just by perceptual similarity. We use concepts to classify things into various categories. But unless the category is of fairly significant importance to us we don’t form a new concept for it because we don’t have a concept for clothes that are 42 regular, right? We don’t need it. That’s a description. Well this is what is sometimes called Rand’s razor: don’t form, don’t multiply concepts beyond necessity. There are many classification tasks that you can perform to serve the needs of a moment or a specialized context where you don’t need a new concept and therefore you don’t form one. Forming a new concept is a major epistemological event and it’s like having kids: it’s not to be done lightly.
Okay, so it’s perfectly rational to proceed as you say, but I would say not to form a concept. But there may be a case where length is the uniform—
Oh yes, yeah certainly.
Okay. Alan, yeah.
Is there a criterion that can easily be stated for when the concept has been formed as against a description? I know one condition would be that you actually condense the description into a mental unit that’s usable. So it occurs to me suppose it’s just a trousers shop, not to introduce the complexes and the people in the shop and in the business have reduced the description since you know 30 whatever x deploy has the smalls and the small trousers are over there, you know the smalls are over there. Now there is a mental unit which is certainly condensed for economy purposes but—
Well it seems to me that it wouldn’t, because even though you have reduced a description to the single word “small” you’ve been able to do that only because the context makes it clear that what you’re talking about are small trousers. If I came in and I said “This is small,” right, this is small even though I can’t put it on, right, so this is not an instance of the single-word term there. In other words what you’re doing is taking a word that has a concept you’ve already formed and simply—this isn’t it clearly I think an implicit description, and the evidence for that is that the term “small” is still used, still available for outside that specialized context. I think you’d really need a new word, vocabulary, because I’m assuming that if there is a concept at all the subdivision of the concept trousers—
And the question I suppose is with regard to subdivision concepts where you’ve got a structure generated by the concept itself. You’re bound—I mean I’m sure that many subdivision concepts began as that happens. So the fact even would be conclusive.
Well all right, I mean if we actually got into some of the etymological questions, yeah surely there are cases where—I mean they’re all obvious. We should also be able to think of lots of cases where a word that was first generalized, named for an attribute and like an adjective, gets nominalized and becomes a noun designating a specific type of thing. I don’t know why I can’t think of an example. All right sure. That seems to be sort of like a borderline case because it still seems to me pretty clearly short for a blonde person or even more normally blind woman. But yeah there certainly are cases like that and I’m not sure that there is any hard and fast criterion. You just have to look at how the whole context of a person’s knowledge and to see whether how fully this has crystallized into a new unit. And I think there would probably be some perceptual tasks that would allow you to do some experimental work on it. For example there are experimental findings that suggest that people have access to a concept faster if you prime them by using other concepts in the same area, the same genus for example, or giving them the genus name. Okay, so I’m going to have the word “beagle” come faster to my lips if you know and ready to identify objects if you’ve already primed me for the category “dog.” So you might use that kind of experiment to decide whether—you know, give the word “person,” a word “trousers,” and then see whether “small” comes faster in a way that is different from “small” in other applications.
Okay.
That would be the phenomenon we were just talking about. Yeah. Yes it would be right. Abstractions from abstractions refer to all those concepts and which includes in fact the vast majority of them which are formed on the basis of grasping similarity in a conceptual rather than a purely perceptual form, whether they are wider integrations, narrower differentiations, extended special-case concepts or whatever. So yeah that would be an abstraction from an abstraction.
Okay. Yeah. What is more fundamental, induction or concept formation—process of induction or concept—
Well if by induction we mean the specific process of supporting a general proposition “All S are P” or “No S are P,” the process I’m going to talk about tomorrow, I would say concepts, concept formation, is more fundamental because you need the concept to indicate the category of things you’re talking about and the attribute that you are generalizing that they all have.
Now there are processes that we could call broadly inductive that work along—in fact concept formation itself could be described, Ayn Rand herself describes it as in essence an inductive process. That’s why induction and conceptualization are so tightly tied together. They’re both processes of starting from particulars and extracting from them a general rule or principle. And so in that more generalized sense induction is something that happens along with perception, and I would even say in that broader sense if you try to explain animal functioning there are various processes of association that they have that might in a sense be called induction, at least for animals that are capable of the more sophisticated types of learning such as the primates and so forth. But in the narrower logical sense of an inference to support a proposition like “All S are P” you need concepts first.
Okay. This side. Yeah.
I still fully understand your answer to this question because where is the difference between someone saying that there’s a big difference between a robin and a penguin but what about blue jay? Well they’re both typical. They both have—people actually do studies where they come up with a number system for ranking typicality. Blue jays and robins both are real high. I mean well I know blue jays may be higher. So yeah. But you know it’s actually interesting to list from blue jays and robins down to wrens and—tough to tip mice and I don’t know owls maybe and then down to you know the really weird things like hawks and dodos and ostriches and so forth, penguins. What these are, these are compiled from judgments that subjects make and what you do is to take out all the answers and compile them. It’s like you know it’s like a certain voting system where you decide who wins by compiling all the first and order first, second, third order preferences. That’s okay but the voting—I still kind of agree with this point that voting would be different. I think maybe if you lived—I think that’s true. I think that would be a factor. Yes I certainly agree and I think there’s evidence that familiarity and frequency of experience with something has an effect on typicality. I’m just not sure it has the only effect.
Yeah there’s a problem. How do you know things are similar together in the same context as a result of the scenarios just popped out between the contents like chairs with shape? Let’s do furniture. If you present these ideas to resemblance theorists who say aha you’ve just merely shifted upon maintaining a similar—
Yeah I’m going to have to reformulate that to see, make just make sure I understand it. I claim that Ayn Rand’s theory avoids a problem that the prototype theory has, namely she explains the prototype theory has a problem of saying when two objects are sufficiently similar to be counted, when an object is sufficiently similar to the prototype to be counted in the concept, and she solves that problem because you’re always dealing with a contrast object like a table. Now here’s why I’m not sure I get you. Couldn’t the prototype theorists say well that just shifts the problem a little bit because now you have to explain when two contexts are sufficiently similar to be counted as two contrast objects, and the context they determine are sufficiently similar for two people who have been considered as forming the same concept. Is that a fair—
Okay.
Well in a way that’s—I would not regard that as a fair question because it’s comparing two questions at two different levels. One is what process do we think the child is following, and that’s where the objection to the prototype theory is located, and the other one is how do I as an epistemologist defend my own conceptual practice in saying that two children are forming the same concept. That’s a meta-level question which presupposes that we already have an enormously complicated context which gives us resources to answer that, such as we know that we fully possess the concept “chair” so we know what the borderlines are. We know that concepts involve definitions. So if you ask me how would I establish that two children are forming the same concept I would look at a much wider range of issues. In other words I would ask, I would test them on things: do they recognize this as a chair even though it’s not rectangular? Can they give me any kind of definition of a chair? Can they locate it in the right higher-order concept as furniture? And I would expect that there would be some differences between the two children in the particular contrast object that they might have used at first in order to form the concept. Indeed I would expect that learning concepts, conceptual learning, is largely a process of where people start at somewhat different levels. People including a concept somewhat different ranges of things and then gradually through learning the concepts become, as they especially as they learn language and talk to people, concepts come closer and closer to having the same content. Children notoriously overextend their concepts. The child looks at the moon and says “That’s a ball.” Okay well he doesn’t have the concept of brown so what else can he call it? But eventually he’ll learn that that’s actually not a ball and so on.
All right we have to end here so I noticed Bill is getting very antsy.
The problem of induction
1:23:50
The topic is the problem of induction or the problem of justifying inductive inference, which of course is a notorious problem in philosophy. And to start, state the problem we have to start by defining induction.
Now this term can be used in a very broad sense to describe various types of inferences and even more broadly various types of cognitive operations. But today I’m going to be concerned with a specific type known as inductive generalization.
Inductive generalization means inferring that something is true of an entire class of things because it has been found to be true of certain members of that class. We observe that certain members of the class S have a certain property P and we draw the conclusion that all S’s are P. Copper melts at 1083 degrees centigrade—that’s true of all copper. All men are mortal. In both cases we support these propositions by the observation of some particular pieces of copper and some particular human beings—all the ones who have lived so far. In the language of statistics inductive generalization is drawing an inference from a sample to a population as a whole, from some S’s to all S’s.
Now at its root induction relies on causal connections. What justifies us in assuming that all copper melts at 1083 degrees centigrade is the existence of a causal link between the molecular structure of copper which makes it an instance of copper in the first place and the temperature at which it changes its state. Whenever we draw an inference that all S are P based on a sample we are relying on the assumption that S and P are connected and that connection in one way or another is a causal connection.
Now here’s the problem. When I generalize I start with information about certain members of a class and I draw a conclusion about all members, most of which I have never observed and never will. How can I be sure that what’s true of some members is true of all the others, the ones I haven’t observed? Am I not going beyond my evidence? Isn’t there a leap of faith here? If I say that the inference is supported by a causal connection, well that only shifts the question: how do I know that there is a causal connection between S and P? Even if S and P have always been connected in the past, how can I be sure that that connection will persist in the future? The sun has risen every morning in recorded history but how can I be sure it’s going to rise tomorrow?
Now you can see this is a serious problem. If we can’t rely on induction then we can’t employ deductive reasoning either. Deductive reasoning is the application of a general principle to an instance that falls under it: all men are mortal; Socrates is a man; therefore Socrates is mortal. But such an inference is only as good as its premises, and if the premise is a generalization then unless we can support that inductively the rest of the inference is useless. In fact if we cannot generalize inductively from our experience then there’s no point in forming concepts in the first place. The point of forming concepts, of classifying things, is to discover the attributes and ways of acting common to all the members of a category so that we can be prepared to deal with new instances as we encounter them, and we can’t do that without inductive generalization.
My goal today is to give an answer to the problem of induction. Now this answer is not original with me. When some people when they heard that I was going to be giving these lectures said “Oh are you going to solve the problem of induction? That would be nice,” and I felt a little embarrassed because I thought it had already been solved. The problem is not even the answer. Anyway it’s not even really original with Ayn Rand. I believe it’s implicit in a broader Aristotelian tradition in philosophy, although we’re going to see that her theory of concepts allows a much stronger formulation of certain points, a much more impregnable defense of the Aristotelian position than the actual Aristotelians could offer.
Now we need to look at this problem at two different levels. First we need to ask how to validate the law of causality, the fundamental principle that says reality is governed by causality in the first place. The law of causality as we’ll see is an indispensable metaphysical basis of induction. Secondly we’re going to examine certain questions about how we establish specific causal connections such as the ones I’ve used as examples, melting point of copper and so forth. These are questions about the methodology, the epistemology, the logic of induction. Okay, and we’ll see that at this level particularly a good theory of concepts is indispensable.
Hume’s argument
1:28:43
Right now I want to begin by taking a closer look at the problem, and I want to do that by reviewing for you the arguments of the man who first posed this problem in its most powerful and dramatic way. I’m referring to David Hume, the 18th-century British empiricist.
Hume believed—these are his conclusions—that there are no inherent connections between causes and effects, between events that we see to follow one another. In reality they do follow one another but there is no inner connection between them. There’s no necessity. In his own words you can put it this way: all events seem entirely loose and separate. One event follows another but we can never observe any tie between them. They seem conjoined but never connected. Therefore we have no rational justification for generalizing beyond the range of our immediate experience. The fact that we’ve seen the sun rise every day up to this point gives us no rational basis for expecting it to rise tomorrow. The fact that bread has nourished us in the past gives us no reason for thinking that it will not be toxic the next time we have a slice. The regularities that we observe represent a kind of metaphysical coincidence, a coincidence like the fact that for the last 70 years Russian agricultural areas have had unusually bad weather. But because there are no inherent connections among events there is no reason to expect this coincidence to continue.
Now the problem is not that some unpredictable factor might interfere. The problem with knowing that the sun will rise tomorrow is not that there may be some comet that’s going to collide with the earth and disintegrate it or at least stop its rotation. No, even without any interfering events we can’t know that the sun will rise tomorrow because we can’t know that it is in and of the nature of the earth in its present state to continue rotating even if no further unpredictable event occurs. We don’t know that causality will continue in its normal course.
Now of course we do; we expect that it will. We expect things to continue acting as they have in the past. We can’t help ourselves. But our basis is not reason. This expectation is a kind of psychological habit that has been induced by the regularities that we’ve observed. We’re like superstitious people who have observed some chance coincidence—breaking a mirror and having a run of bad luck—and we expect it to continue. All of our beliefs about causality, says Hume, have the logical status of superstitions. They’re real enough as psychological phenomena but they have no objective basis.
Now why would anyone believe this? Well, says Hume, imagine you had never seen water before. Could you tell just by looking at the water, observing its fluid and transparent character, that you could not breathe under it, that it would suffocate you if you held your head under it? If you had never seen fire, could you tell just from the observable color and shape of the flames that it would burn you? No. Cause and effect are distinct and observing one by itself will not allow you to deduce the other.
Very well, you might say, but once I have observed what effect follows from a given cause I can see that they’re connected and then I can generalize that other instances of that cause will have the same effect. Once I observe that the flame burns then I know that future flames will burn.
But wait a minute, says Hume. Can you really see the connection? You see the flickering yellow flames of the fire; you feel the pain when you put your hand in it. But do you in addition see a connection, a relationship of necessity between cause and effect? No. Again a sensation of yellow is followed by a sensation of pain but all there is is the temporal sequence in your experience. You observe a billiard ball colliding with another one; you observe the second billiard ball moving off. But again all you observe are the two events. You don’t observe over and above those two events a relationship of necessity connecting them.
Now to reinforce the point Hume offers us two additional arguments.
First of all even after you have seen the second billiard ball roll off in a certain direction couldn’t you just as easily imagine it rolling off in a different direction or jumping off the table or collapsing into a puddle? In his own words, even after we have observed an effect, quote, “the conjunction of it with the cause must appear arbitrary since there are always many other effects which to reason must seem fully as consistent and natural.”
Secondly, if you did observe some real connection between cause and effect then you would be able to generalize from a single instance. After all in mathematics and geometry if you observe a relationship, if you follow a proof from a single illustration on the board that the internal angles of a triangle equal 180 degrees, then you can see by the internal necessity of that demonstration that the same would be true of all triangles. Okay, by parity of reasoning if there’s a necessity between cause and effect then observing one instance of that relationship should allow you to form a generalization without the need for repeating the experiment in any way. And yet Hume says it is a characteristic of induction that we do have to repeat and we have to gather a series of examples of the same thing in order to support a generalization.
Now please remember this point because I’m going to come back to it later on.
Now at this point you might try a different tack. You might appeal to the law of causality, and for our purposes we can think of the law of causality as having two basic elements, two clauses which I have stated on the handout.
The first clause says that everything that happens has a cause. Events do not occur randomly. The occurrence of an event and its particular nature are determined by the circumstances in which the event occurs, including the nature of the entities that act, the preceding events, and so forth.
Secondly the law of causality says that the same cause has the same effect. Once we have discovered the factor responsible for some event in a particular case we can be sure that the same factor will have the same consequence in any other case so long as the other circumstances that are relevant have not changed.
So you might answer Hume as follows. True enough, you might say, I don’t actually perceive causal necessity. I see the fire; I feel the pain; but I don’t literally perceive a relationship of causal necessity between them. The law of causality however tells me that such a relationship exists. The first part of the law says that the pain must have some cause, and once I’ve isolated the fire as the causal agent the second part of the law tells me that the fire will have the same effect in any other instance.
But now Hume will ask the obvious question: what justification do you have for believing in the law of causality? How would you validate the law of causality? You can’t establish it by inductive reasoning, you would say, because that would be circular. Inductive reasoning presupposes the law of causality so you can’t turn around and use inductive reasoning to prove the law of causality. That would be circular. So the law of causality would have to be a self-evident axiom or at least a principle that could be deduced from some axiom.
But, says Hume, the mark of an axiom of a self-evident truth is that we cannot deny it without contradicting ourselves, and the test of whether a proposition is contradictory is whether we can imagine it. Okay, we cannot imagine a case in which A equals not-A. That’s why the law of identity, A is A, is self-evident. It cannot be denied without contradiction. We cannot imagine it being false. But we can imagine that fire soothes on Monday, Wednesday and Friday and burns on Tuesday, Thursday and Saturday, Sunday being a day of metaphysical rest. The law of causality can be denied without self-contradiction so it cannot be an axiom. And even if it were an axiom, finally that wouldn’t do us any good. Axioms, says Hume, are merely relations among our own ideas. They are arbitrary constructs that we make true by the way we define our terms, and therefore they don’t tell us anything about reality.
So Hume’s conclusion to summarize is that all events are loose and separate, to use his phrase. They follow one another in a temporal sequence but they are not connected. Any event could be followed by any other. We may notice certain patterns of regularity but these don’t reflect any interconnection between one event and another or between an entity and the way it acts. So when I turn on the stovetop burner on my stove I expect gas to come out the jets. It’s done that a thousand times before in the past but it would be mere coincidence if that’s what happened this time. It’s equally possible and indeed equally likely that some other, any number of other events could have happened. The stove could explode; it could rise up off the floor; an army of ants could come streaming out marching in unison and singing Deutschland über Alles. Anything is equally possible.
Now this is a bizarre view of reality. It’s a kind of nightmare vision. I have my image for it is like spending your whole life driving on ice, which we’ve all had some experience with lately, and frankly it’s incredible. But it’s also been very influential in philosophy. So let’s turn now to the opposing case and see how to answer Hume’s argument. How can we validate the law of causality in the face of his assault?
Validating the law of causality
1:38:45
Well Hume was correct first of all in saying that the law of causality cannot be established by induction—that is, we cannot proceed by proving first that water obeys the law, then that fire obeys the law, then that the economy obeys the law, and then generalizing that all phenomena obey the law. Induction presupposes the law. We need the law in order to establish any specific conclusion. So the law of causality must be axiomatic.
Now most philosophers have held that if a statement does not rest on induction then it does not derive in any way from the evidence of the senses. In their view induction is the only way to bring empirical evidence to bear on a general statement such as the law of causality. The implication is that an axiom must be a statement adopted by reason operating in some way independently of the senses. It must be a priori.
And Hume like many other although not all philosophers drew the further implication that axioms must be subjective. They must be arbitrary. We adopt them by convention rather than by discovering the facts. After all if perception is our only contact with reality, as the empiricists rightly believed, and if induction is the only way to extract information from perception, then a principle not based on induction cannot be true by virtue of some contact with reality. It must be true merely by convention. It must be arbitrary at its root.
Now Objectivism, along with the entire Aristotelian tradition I might add, does not accept this argument. An axiom is a self-evident statement, and a self-evident statement is one whose truth can be observed directly. Such a statement formulates in explicit terms what is implicit in perceptual awareness. Inductive generalization is not the only way to extract information from the senses.
So let’s turn our attention to perception and see how we can try to extract the law of causality from what is given.
Okay, the essence of perception is the awareness of entities—that is, of objects as wholes, discriminated from their background, existing and acting as units. In this respect perception is different from sensation, which is a more primitive level of awareness. A sensation is a momentary awareness of an isolated sensory quality such as a patch of color, a feeling of pressure on the skin, a taste of salt. A perception by contrast is the awareness of an entity as a whole, of a constellation of qualities which are integrated. Now of course we are aware of qualities in perception but we’re not aware of them as isolated atomistic units, nor do we perceive the entity as a bare framework on which these qualities are hung like ornaments on a Christmas tree. When we see an apple we’re aware of its color, its shape, its size, its texture and all its other qualities as integrated elements of a single identity. We perceive the object as a unified nature, and the object is its nature.
This is the basis for the law of identity, the axiom which says that an existent is something specific. Whatever it is that we perceive, it is something specific. It has an identity. It is its identity. It is what it is.
Now we can also perceive the actions of entities, the way they move and change, and in that respect too perception differs from sensation. Sensation is a momentary experience. Change at that level is experienced as the substitution of one quality for another, as in a kaleidoscope. But perception does not consist of isolated snapshots in that way. It’s the continuous awareness of objects over time. A hunter follows the prey as it moves across the field. A doctor listens to his patient’s heart as it beats. Both of them are continuously aware of a single entity as a single entity and aware of its actions, and they’re aware of the actions as actions of the entity. Whenever we perceive an action we perceive it as the action performed by something, just as we always perceive attributes as attributes of something. There cannot be motion without a moving object. There can’t be a dance without a dancer.
Now when we form a concept for a type of action of course the entities involved become omitted measurements. Now a waltz is a waltz no matter who’s doing it. But the measurements must exist in some form. Without something that acts there cannot be any action.
Now all of this I claim is implicit in perception, and the law of causality is simply a way of making it explicit, putting it into explicit conceptual form. And we can do this in three steps, although I want to emphasize as strongly as I can that these are not steps of an argument, since what I’m presenting to you here is axiomatic. I would be undercutting my own case if I presented it to you as a proof. It’s simply a way of drawing your attention to what is implicit in your own perception so that you can grasp for yourselves the connection. In psychology I’ve heard references to something called guided fantasies. Well this is sort of the opposite: this is a guided perception.
All right. Step one. The first part of the law says that all actions have causes. To say that an action has a cause is to say that it is produced by something, that it is a dependent phenomenon, not a self-sufficient existence, not a primary. And that is precisely what perception tells us about actions: they do not occur without entities. It may be as Hume says that we can imagine an action as a self-sufficient existent occurring without a cause. Okay, but if so we can do that only by dropping the context of what we observe perceptually. If you examine your perception, if you are anything like me, you cannot imagine an action occurring without an entity as long as you hold the perceptual context. You literally cannot imagine it. It’s like you can omit the measurements of course when you form a concept, but just as you can’t imagine a color that is red but not any specific shade of red, it’s impossible to imagine yourself actually observing, so long as you’re in touch with your perceptions, an action occurring with no particular thing that acts.
Okay, so the first part of the law is simply a way of identifying the observable fact that an action is the action of some entity. It identifies the dependent nature of the action, the fact that actions are not primary metaphysically.
Step two. If actions depend on entities then an action must depend on the nature of the entity that acts. A thing is its nature. If we try to imagine an action that depends on the entity but not on its nature we have to imagine the entity as something distinct from its nature. We have to drive a kind of metaphysical wedge between a thing and what it is. But a thing is what it is. There’s no such gap.
So now we know that an action must depend on the nature of the entity and we are ready for step three, the final step, which is to realize that because an action depends on the nature of the entity that acts, an entity with the same nature operating in the same circumstances must act in the same way. The same cause must have the same effect, which is part two of the law of causality.
Now it’s true as Hume said that we can’t imagine the same cause giving rise to different effects, the same entity acting differently, but we can do this once again only by dropping context—in this case the context of what we have just acknowledged in step two: that the action depends on the nature of the entity. Uniformity is implicit in the idea of an action depending on the attributes of the entity or of those attributes producing the action. If you take away uniformity you take away any concept of production or of dependence.
Suppose I said the motion of a billiard ball when it is struck is produced by its rigidity. It’s that rigidity that makes the ball move when the first ball hits it. But now suppose I say that another ball with the same rigidity, or the same ball at a later time, might stand still when it is struck or collapse on the table like a rotten mushroom or recite the Gettysburg Address, and that in all those cases it would still be the rigidity that is causing the action. Well that’s incoherent. By imagining these different effects I have severed any connection between the rigidity of the ball and the way it acts. I have therefore withdrawn the admission I made previously that the action depends on the nature of the ball. A cause that does not act uniformly is not a cause. We’ve replaced a connection between the entity and its actions by a random relationship. A random relationship means there is no relationship.
Okay, well that’s it. That’s the validation of the law. That’s like our guided perception.
To summarize what we’ve seen let me quote from a work by an Aristotelian logician, H. W. B. Joseph, who was an Oxford philosopher writing in the earlier part of the century. The passage I’m going to read you is from a work called An Introduction to Logic, a chapter called “Presuppositions of Induction.” Joseph says: if a thing A under conditions C produces a change X in a subject S—it gets easier from here on—the way in which it, A, the way in which it acts must be regarded as a partial expression of what it is. It could only act differently if it were different. To say that the same thing acting on the same thing under the same conditions might yet produce a different effect is to say that a thing need not be what it is. But this is in flat conflict with the law of identity. A thing to be at all must be something and can only be what it is. To assert a causal connection between A and X implies that A acts as it does because it is what it is, because in fact it is A. So long therefore as it is A it must act thus, and to assert that it may act otherwise on a subsequent occasion is to assert that what is A is something else than the A which it is declared to be.
In short the law of causality is the law of identity applied to action. Events are not loose and separate. They’re tightly constrained by the entities that act, and once we have identified specific causal connections our confidence that things will continue to act accordingly in the future is not groundless or subjective. It’s a recognition of an objective fact.
So we have answered Hume’s argument. We’ve rejected his assumption that axioms are arbitrary conventions; they’re grounded in perception. We’ve rejected his assumption that what we can imagine is a valid test of what is possible. We can imagine violations of the law of causality only by dropping the context of what is given in perception. Our imagination may be interesting from a psychological standpoint but it has no epistemological significance.
There’s a third point that I didn’t emphasize but I think it deserves some mention. Hume regarded causality as a relationship between events, whereas Aristotelians regard causality as primarily a relationship between an entity and its actions. Okay, this is a metaphysical difference in viewpoint, but it has an epistemological basis implicit in what we have just been over. The basis is that Hume did not see perception as the foundation of knowledge. The epistemological starting point for him, the starting point, was sensation. What is given in his view are merely isolated qualities in momentary acts of awareness. Perception, said Hume, is the result of putting those sensations together by some process of association, and the entity that we perceive is simply a set of qualities. But we don’t see any metaphysical glue holding those qualities together. The entity is a constructed invention of the mind. So we have no reason for believing that all attributes are attributes of entities nor that all actions are actions of entities. So Hume deprived himself of a crucial point on which the law of causality depends and therefore could not see any reason why events could not be primary and autonomous types of existence. The dance is given but not the dancer.
Now of course you recognize that in some cases we do associate an effect with an attribute of an object as well as with a prior event. Okay, we associate the motion of a billiard ball with its rigidity as well as with the prior event of the collision. Okay, so Hume acknowledged the facts that Aristotelians would cite in supporting their view of causality, but the form in which we’re aware of those facts, says Hume, is a form of sensations. So the rigidity of the billiard ball is not as it’s given in experience an enduring property, a part of a nature; it’s the momentary content of a momentary act of awareness. Even an attribute at the level of sensations is experienced as an event. It’s a momentary content of awareness to be replaced a moment later by something else. Okay, it’s the occurrence of a certain quality at a time and place. So you see the metaphysical difference is not fundamental. It’s the epistemological difference—starting with sensations versus perceptions—that is crucial here.
Mill’s methods of induction
1:53:21
Okay, well let me turn now to the second major topic to be covered today which is the method of induction. I think it’s clear that the Aristotelian and the Humean philosophies differ radically in their basic view about causality and the law of causality, and the same is true about their views of the procedure by which we establish particular causal connections. Okay, so let’s turn to the second broad topic and let me start with a few remarks about Hume and the Humean tradition in inductive logic.
Hume and his followers as we saw do not see any interconnection between cause and effect. There are observable regularities. An antecedent factor A may always be followed by a certain type of effect E. They may be constantly conjoined, in Hume’s phrase, but there’s no reason why A must produce E. So observing a cause of regularity on that view is like drawing colored balls from an urn, to use a favorite example of statisticians. If we draw a series of balls and all of them are black we may begin to suspect that all the balls in the urn are black, but since there’s no reason why they must be black we are limited to a claim of probability. We can’t be sure that the next one we’re going to draw is black. The more balls we draw the more probable it becomes that they’re all black, but we can’t be certain. The probability never reaches 100 percent until we’ve drawn all the balls out.
Okay, now in the same way when we put our finger near a flame and feel heat we observe an association between flame and heat, but because there is no inherent reason why flames must be hot a single instance is not very good evidence for a generalization that all flames are hot. We have to repeat the experiment over and over getting more and more confirming instances. The more instances we have the higher the probability rises that all flames are hot, although in this case since we cannot ever observe all flames, it being an open-ended concept, the probability never reaches 100 percent. So in this way Humeans typically regard inductive logic as a branch of probability theory and they hold that the strength of a generalization depends in an essential way on the number of confirming instances.
The Aristotelian approach is quite different. The law of causality tells us that the same cause has the same effect, so once we have isolated the cause of a given effect in a given case we can generalize immediately. We know that that cause will always have that effect.
The problem is to isolate the relevant causal factor in a particular case. When we observe an action or an event we know that it depends causally on the natures of the entities that are present in that situation, but we don’t know which aspects of their identities are relevant. We need to discover which particular attribute of a thing gives it the capacity to act as it does. The fact that salt dissolves in water tells us that salt is soluble, that solubility is part of its nature, but we don’t know what particular attribute of salt makes it soluble, gives it that capacity for action. This is what we need a method of induction for.
And that method, which was first proposed in an early version by Francis Bacon and later refined and systematized by John Stuart Mill, is the familiar method of systematically varying the factors experimentally. So let’s look briefly at three of Mill’s methods. These are called Mill’s methods of induction. He named five but these are the three primary ones and I’ve illustrated them on your sheets on the handout.
You’ll notice that in all three cases—the methods of agreement, method of difference, method of concomitant variations—on the right-hand side is the letter E which stands for the effect. Preceding E there ought to be an arrow. I didn’t think—my printer doesn’t do this kind of thing and I didn’t think to put it in before I started xeroxing so I left it to you to do. Please put in arrows, arrowheads on the right-hand side of all the dashes.
In the method of agreement we look for a unique factor that is common to all the cases in which the effect occurs. We hold one factor A constant and we vary all the others. For example we want to explain why all the people who ate at a restaurant on a given night got sick and we identify, we find that the common factor is that they all had the Brussels sprouts. So we’ve isolated factor A as the cause because it is the only factor that satisfies the law of causality, in particular which says if a factor is sufficient to produce a given effect it must produce that effect even if other non-relevant factors are altered. Okay, if in case number three under the method of agreement we had not obtained the effect then we would know that A was not the cause because the same cause must have the same effect.
Let’s look at the method of difference. In this method we keep all the factors constant except for the one that we are testing, and if the effect occurs in the presence of A but does not occur in its absence we know that A is at least a necessary condition for the effect and that none of the other factors are sufficient. The law of causality tells us that if B were the cause of the effect, that if B were the cause of E in case number one then B would have to be the cause of E in case number two as well. But in case number two E does not occur. You should—it should be a little slash across the arrow there to indicate that in this case E does not occur. My printer does not do slashes either.
This of course is the pattern of controlled experiments in science where you have an experimental case and then a control case, holding all other factors other than the one you’re testing constant. And again if in case number two here the effect E had occurred then we would know that A was not a necessary condition and that we would suspect that B, C or the combination were the real cause.
Finally let’s look at the method of concomitant variations, and this is like the method of difference except that instead of contrasting the presence or absence of A we vary its magnitude and look for variations in the magnitude of the effect. Okay, so here all the arrows are positive arrows, no slash marks. This is a very powerful method of establishing causality especially given what we learned yesterday about the relationship between an attribute and its instances. Okay, the various instances of the attribute A are related quantitatively and the attribute is nothing more than the dimension along which they are related. So to understand a causal relationship fully, a causal relationship between A and E, we want to know the systematic relationship between their specific measurements, and the method of concomitant variations is especially well suited to give us that because we correlate quantitative variations in A with variations in E.
More on this point in a moment. I’m going to be coming back to it.
All right, these—I don’t want to insult your intelligence. I’m sure you’ve all either used these methods—I know you’ve all used these methods or you couldn’t do business in reality. Even simple generalizations rely implicitly on these things. But you probably also learned them in explicit logic or methodology courses. John Stuart Mill by the way was not himself an Aristotelian. He was an empiricist in Hume’s tradition. But I think you can see that his methods of induction were a real contribution to the Aristotelian approach to causality. These methods are procedures we use to ensure that our reasoning conforms to the law of causality, in just the way that the Aristotelian syllogism in deductive logic is a set of procedures for ensuring that our reasoning conforms to the law of non-contradiction.
Now notice that in all three cases we had to examine more than—all three methods we had to examine more than one case. These cases differ however, and the differences between them—the presence or absence of A, the variation in the other factors—those differences are crucial to the conclusions that we draw. So we are not simply gathering confirming instances as in Hume’s approach. We are looking in a very structured way at a single set of instances.
Okay, now remember Hume’s argument that I gave you earlier: that if there were a necessary connection between cause and effect we could generalize from a single instance. We wouldn’t need to repeat the observation or the experiment. Well that’s true. There is a necessary connection between cause and effect, and since the cases that we compare in using any of the methods constitute a single connected set of observations we do in a sense generalize from a single instance. At any rate we do not rely on repetition. We don’t have to collect identical confirming instances.
Now if I could digress slightly, although I believe this is really drawing out the broader implication of what I just said: repetition plays no essential role in knowledge at all—not in induction, not in concept formation, not in any reasoning process. Whenever we reach certain conclusions about a given phenomenon by observation or by inference, the occurrence of an exact repetition of that phenomenon does not allow us to draw any new conclusions except the obvious conclusion that this has happened before. Repetition as such, the sheer fact of repetition, is epistemologically barren.
In concept formation for example we could not form the concept “red” by observing two identical shades of red. We form the concept by omitting measurements, but in order to do that we have to have some measurements to omit. So we have to have shades that differ quantitatively so that we can grasp the measurement relationship between them. And once we have two different shades that differ quantitatively and we notice that quantitative relationship we have all that we need to form the concept. Further instances add nothing essential, although they may facilitate the process psychologically.
And the same is true of induction. Observing the same action occurring in two or more identical situations would not allow us to isolate the relevant causal factor. We need some variation among cases in order to use any of Mill’s methods. But once we have the kind of variation required by a given method we have all we need in order to generalize. Further experiments that simply reproduce the cases add nothing essential.
Now it is true—I hear the question springing to your lips—it is true that the replication of experiments is an important part of science. But that’s because replication is a way of checking to see that the initial experiment was not biased, that it didn’t omit any relevant factors in subtle, in all the many subtle ways that can happen, and that is of true fundamental importance. But the replication does not strictly speaking add new evidence. It’s only a test to make sure that the evidence provided by the initial experiment was valid in the first place. It’s like corroborating one witness’s account of a crime with another one. It’s a corroboration of the evidence but not a new piece of evidence.
Okay, now so once we have used Mill’s methods to isolate the cause of an effect in a single case or a single set of related cases we can generalize immediately because the law of causality ensures that the same cause has the same effect. But when we refer to the same cause we raise once again the problem of universals.
Concepts, induction, and Goodman’s “grue”
2:06:00
In the method of agreement for example we look for a factor that is common to all the cases, the fact that I’ve labeled A. A is a concept designating a feature whose instances in the different cases are not exactly alike in specific degree. For example a college student trying to explain why he liked certain courses that he took might find that the common factor was that the instructor made the subject matter seem interesting and exciting. That’s the cause of his enjoyment. But of course the instructors do this in different ways. The subject matter differs from one course to another. The student has to abstract the common factor by omitting all those measurements. So the ability to apply Mill’s methods depends crucially on our ability to form concepts, and the validity of our concepts is a crucial part of the process.
So I want to show as a final point in our discussion today the consequences of adopting a bad theory of concepts. I want to look at what effects nominalism and realism would have on the field of induction and thereby show again one of the virtues of Objectivism.
Okay, let’s start with nominalism because nominalism especially in its extreme form has an especially destructive effect on induction. As we saw yesterday, extreme nominalism says that a concept represents an arbitrary grouping of things that have nothing in common except that we call them by the same name. If that were true we would have no reason whatever for thinking that everything we arbitrarily group under the concept A would have one of the effects that we arbitrarily group under the concept E.
Here’s an example, and I hasten to add this is not my example. I doubt I have the imagination to have come up with anything as bizarre as what I’m about to tell you. This is an example put forward by a philosopher named Nelson Goodman and it has been discussed at an amazing length in the philosophical literature.
Suppose we have observed that emeralds are green and we have observed them under enough conditions, enough varying conditions that we’re sure the color is in effect of the essential nature of the gem. So we generalize that all emeralds will have the same color. But what counts as the same color? Let me form a new concept, “grue,” spelled this way.
What is grue? To be a member of this color category an object must be green if it is examined prior to some arbitrary time, let’s say 1990. If it is not examined prior to 1990 it’s blue. Okay, so this concept designates a color, a set of specific determinate shades of color which we’ve separated into two subsets: those that have been examined prior to 1990 are green; those that have not been examined prior to 1990, which obviously includes all colors that come into existence after that time, are blue.
Okay, now all the emeralds—some of you are looking at me in amazement; I kid you not, this is a real example—all the emeralds we’ve examined so far are green. All those emeralds have also been grue. They fit the definition of grue. If we can generalize that all emeralds in the future will continue to be green, why can’t we with equal justification generalize that all emeralds in the future will continue to be grue? But that would mean that they will all be blue after 1990. That is, the same evidence that all emeralds up to this point have been green supports two generalizations that have conflicting predictions after 1990.
Why about it? You could do this obviously with any concept.
Now you might object that grue is an invalid concept because it implies that an emerald has to change color without cause on New Year’s Day 1990. That’s true, but that’s not the fundamental answer. Goodman would reply that the emerald changes color only if we think in terms of green and blue as our color concepts. If we think in terms of grue then an emerald that remained green on New Year’s Day would have changed color without cause from grue to non-grue.
The fundamental objection to this argument is that it rests on extreme nominalism. Goodman assumes that any concept is as valid as any other. We can group things together any way we want. We can define predicates in his language arbitrarily without any constraint from reality, any constraint by similarity or any other objective factor. But on the Objectivist theory by contrast, as we saw yesterday, concepts must be formed in accordance with the constraints set by the patterns of quantitative and qualitative differences, and the concept grue violates that constraint. When we compare two shades of green to something that is blue the difference between green and blue is necessarily perceived as qualitative in comparison with the quantitative relationship between the two shades of green. The two shades of green are therefore necessarily seen as more similar than either one is to the blue, and that’s a perceptual fact which the mere passage of time cannot affect.
Okay, now if a nominalist theory of concepts would invalidate induction, what about a realist theory? Most Aristotelian philosophers who have written about induction—and this includes H. W. B. Joseph whom I cited earlier—have been moderate realists, Aristotelian realists. So let’s take a look at realism.
Realism and induction
2:11:40
Now the major problem that realism causes in the field of induction is that it isn’t true. This is a serious defect if you are trying to validate a procedure to rely on a premise that isn’t true. If the only way to show that induction is objective is to bring ourselves to believe in intrinsic essences and attributes that exist intrinsically as abstract, then we have a very tough selling job to do both to ourselves and to other people. And in fact I believe this is one of the reasons why Goodman’s argument and other Humean approaches have been so persuasive: that most philosophers can’t accept that and so feel driven into nominalism.
Okay, however I want to focus or turn to some somewhat more delimited problems that realism causes in regard to induction because I think they bring out once again further virtues of Ayn Rand’s theory of concepts.
When a realist says that the same cause has the same effect, the word “same” means exact identity. Okay, the realist means the possession of the same identical abstract attribute or essence. Suppose we observe that asphalt pavement becomes hot in the summer sun and after we compare different substances we isolate the black color of the pavement as at least a contributing cause. So we generalize that black absorbs more heat from sunlight than white objects. What this means to the realist is that objects absorb heat in virtue of their blackness, an intrinsically abstract attribute that all black objects share. Two objects may differ in the particular shade or degree of blackness; they may be merely similar. But behind that similarity, behind the concrete determinate colors that we observe, is the abstract attribute that they share, and that abstract attribute as distinct from the individuating elements that make them individual shades is the cause of the heat absorption.
For the realist the fact that the determinate degrees of any attribute are quantitatively related, the fact that they can be ranked in order along a dimension of measurement, is not relevant to the objectivity of concepts or to the process of concept formation as we saw yesterday, nor is it fundamentally relevant for the realist to causality or induction.
Now one consequence is that realists view abstractions as the real causal agents in nature. An entity does not act in virtue of its specific determinate nature; it acts as an embodiment of an abstract attribute that it shares with other entities. And another consequence is that realists gravitate toward a qualitative understanding of causality. I think this is most easily visible in the science of the period prior to Galileo and Galileo’s insight that mathematics is the language of nature.
The Aristotelian scientists—and this by the way is not—I’m not charging them with not grasping something that they really weren’t in a position to grasp yet—but the fact is they typically sought to establish qualitative laws such as heavy objects fall toward earth, species reproduce after their kind, fire burns, water flows. The specific quantitative measurements of an attribute did not play any major role in the theories. But those specific measurements are crucial. Massive objects fall to earth as a result of a specific degree of gravitational force that is mathematically related to the specific mass of the object. In general scientists look for causal relationships which show not just that one variable or one attribute X depends on Y—that’s a qualitative causal relationship—they look for functional relationships of that form which relate the two variables X and Y in a way that reveals the quantitative relationship, the systematic mathematical relationship between the particular values of each variable, the specific measurements.
Okay, you know so for example the law of gravitation, if I remember my physics correctly, it looks something like that. Here the gravitational attraction between two objects is related by this function to the masses of the objects and the distance between them. Or in economics some of you heard lectures on the equation of exchange which is that money supply times the velocity equals the price level times the number of transactions. Okay, again—and I’m sure there’s a constant in here somewhere—again what we’re doing is saying there’s a systematic relationship between the specific measurements of those attributes.
Now the Objectivist theory of concepts is ideally suited to explain this aspect of scientific inquiry. We form concepts by noticing quantitative relationships among similar objects, relationships that allow us to rank them on a common dimension and measure them by a common unit. Induction is then a matter of correlating two such dimensions. We don’t need to attribute causal efficacy to abstract attributes that lurk behind the specific measurements. And yet the classification of similar objects and the omission of their measurements is governed by the objective existence of their quantitative relationships, so it is not an arbitrary process as it is in a nominalist theory. And we are therefore immune from the skeptical conclusions that nominalists are forced to draw about causality and induction.
Okay, now there are many other questions about induction that I have not addressed today. There are questions of detail about the application of Mill’s methods, questions about how you know when you have varied all the factors, and so forth. There are also questions about how we formulate and test theories that are posited to explain causal connections. These are all important questions in the philosophy of science. But the point of departure for any proper theory, philosophy of science, in my view is the Aristotelian view of causality and Ayn Rand’s theory of concepts. Together these provide a validation of the law of causality and of the basic method of induction which are the foundations for everything else.
Thank you.
Questions and answers, part two
2:19:00
Thank you very much. I think we have almost as much time as I was supposed to leave you for questions so—and I’m sure there are some so fire away. Who wants to start?
Yes. One little one and a bigger one. When you said that the way the Humean people would say the grue/non-grue type of thing in 1990 I didn’t—I understood what I’d say to that but I didn’t understand what their opinion was. He said that—
Okay what would you say? I’d say you’re asking if from the laws of reality you’re going to say it’s going to remain green the entire time and you’re almost playing with semantics now. I wouldn’t say that’s the essence of what I’d be thinking. It’s going to be green; it’s going to be what it is; it won’t change in 1990. So it’s just a matter of what you’re going to call it and I’d go from there.
Okay. Goodman’s claim in answer to that is that as I said you think in terms of green and blue. That’s your parochial Western assumption. I think in terms of grue. In order to remain grue however the thing has to act a little differently than it does if it has to remain green. If it stays green in your terms then that means it changes color in my terms from grue to not—actually he defined “bleen” as the exact opposite of grue. So if an object stays green past that point T, 1990, then in Goodman’s framework it changes from grue to bleen.
Now I don’t want to defend him too far. Clearly where we’re talking about a single object with a single specific color, if you were to observe that continuously from prior to—you know over New Year’s Eve on 1990—okay one of two things, well a number of things might happen and we know what would happen: it would stay the same color. But Goodman would say we could observe then whether it remains grue or remains green. Now it does seem to me that the whole argument presupposes that we can tell the difference between blue and green; otherwise there’s no difference between green and grue after 1990. So given that we can tell the difference, on the stroke of midnight 1990 we will be able to observe that we will be able to discriminate the two cases of it remaining green and it remaining grue, and we will be able to compare the color that it is right after midnight with the color that it is just prior. So that Goodman himself, by assuming there is an observable difference between these two colors in particular at the level of determinate colors, is committed to our being able to tell. It implies that there is an objective difference between it staying green and staying grue because since we can tell the difference between green and blue we can tell if it changed from one to the other at that point. That’s a real change whereas remaining green is not a change. So in that sense there’s an asymmetry between the two cases.
But more generally the problem here that grue introduces is not concerned so much with the law of causality as applied to specific determinate measurements but rather to types of measurements. It’s addressed to the problem: how can we generalize from this specific shade of green to all shades of green, or from this specific mass to all measurements of the variable mass? Clearly in doing that we are forming a concept and we are putting together things that differ. Okay, so now we’re dealing with two shades of green that differ. When we say that all emeralds are green we are treating, putting, lumping together two shades that in fact differ at the level of determinate concrete shades. What justifies putting those two together as opposed to the blue and the green?
And there the key question is: is there an objective similarity basis for saying that the two shades of green are more similar than the two shades of—than the blue and the green? Goodman as an extreme nominalist says no and the Objectivist theory says yes.
Goodman by the way believes that any two things are exactly as similar to one another as any other two things. Do you know why? Here’s the argument. Take any two things and then take the entire universe of things which let’s say it’s finite. Things can—all the objects in the universe can be grouped into sets. Okay, and by some mathematical law involving the power set or the exclamation mark we can know the total number of sets that is possible to construct out of all the objects in the universe. Okay, now two objects are similar if they can be described by the same term. Similarity, says Goodman, is relative to our conceptual framework and a concept is just an arbitrarily collected set of objects. So any two objects are members—sorry, two objects are similar if they’re members of the same set and any two objects are members of exactly the same number of sets. Okay, so any two pairs of objects are exactly similar. My watch and your watch are as similar as my watch and the Empire State Building. That’s primacy of consciousness with a vengeance.
Okay. Yeah let’s let me take another one, another question.
Yes. You said that axioms were arbitrary but he still didn’t accept accidents. I have trouble understanding that because I would say to Hume well if axioms are arbitrary then if you don’t assume A is A I don’t understand the word you’re saying because we can’t speak a language without assuming A is A.
Well Hume would probably say fine that’s true but that only—that’s a fact about language; it’s not about reality. I mean for Hume that whole approach: axioms are conventional and they may be necessary to the practice of speaking a language. Okay, but after all that’s a human activity which we construct on our own. I mean suppose they’re like the rules of a game, to use Wittgenstein’s way of putting it. Suppose you slid into second base and the second baseman was waiting there with the baseball and tagged you and the umpire said you’re out. You said wait a minute, prove to me that being tagged before you hit the base objectively makes me out. What’s the basis for that? How can you validate this rule? Well that would be an absurd question. It’s just part of—that’s part of the game. It’s arbitrary. But if you don’t want to observe it then don’t play the game. Hume and that tradition says the law of identity is one of the ways we organize our language. True enough and we couldn’t speak without it. But we couldn’t—maybe someone some other language with a different set of rules could. But that only tells us something about our language not about reality.
Now you’re right though to push the question because a key part of saying that this is conventional is that we could—there could be another language structured around a different set of principles, and in fact that really can’t be done. Unlike baseball where you can imagine other games you can imagine with different sets of rules or you can imagine a similar game like baseball only you get to second base free if you’re—regardless it’s a wild base. But we can’t imagine a wild language where the law of identity doesn’t apply. And it’s not just a fact of imagination. I mean you can show that even someone who tries to assert the possibility of such a language has to assume the law of identity and thereby reaffirm it in the very act of denial.
Okay. Yes. Just your point about the corroborative nature of inductions to ascribe lesser importance to induction in terms of gaining knowledge. The point about repetition right that you don’t gain any new knowledge by new instance of some evidence.
No I’m not saying no repetition. I’m saying repetition as such is not a part of the inductive process. The inductive process is a crucial—it’s a foundation of all our knowledge, all of our general knowledge. But the point is simply that we don’t rely on repetition as such when we use that process. So I’m not impugning the importance of induction. I’m just saying that it doesn’t involve any reliance on repetition.
Let me see if I can get another.
Yes. Yeah I’ve always had a little bit of a problem with the phrase “it is in the nature of an entity to do something under certain conditions.” I’ve always felt it’s in the nature of the laws of the universe for an entity to act a certain way under certain conditions. For example if I throw anything it will continue forward because of the law of momentum. And so it is not in the nature of a ball to carry forward; it is in the nature of the law of momentum to carry any object forward.
Yes right, and this is a good point and it brings out once again the relationship between concepts and induction in the following way. It is not in virtue of being a ball that an object continues in motion unless a force acts on it, and we can prove that by the use of Mill’s method of difference. Okay, we take something that is not a ball and it continues to move forward; it has the same effect; it continues to move forward if there’s no interfering force. So it’s not qua rounded rubbery object that it has that. And yet it has to be some aspect of reality that is—the law is not something that’s inscribed on our minds or on the skies that we read and that things obey in the way that we obey legal laws out of fear of the consequences. Things obey laws because their natures require them to. In this case it is an aspect of the thing’s nature, namely its mass, which happens to be something that all physical objects have in common. So we have to reach the appropriate level of generality in order to identify the relevant causal factor in the thing’s nature. We have to reach the relevant level of abstraction which in this case is as high as you can go in science anyway, as far as we know. But it’s still an aspect of the thing’s nature. It has to be existence in some form that makes things act the way they do.
Yes. Magician notion of relevance because if you have a true description such as all—also white—you know what is the difference between that one or more what is the—
Okay yeah this is a classic problem. Oh the question is determining what factors are relevant and the example that Patricia mentioned was the generalization “All swans are white,” okay, which has always seemed to be a shaky generalization even though we have thousands of, you know, countless confirming instances of it with all kinds of variations of species and circumstances and so forth. And this is one of the kinds of problems that I alluded to at the end of my talk as something that a philosophy of science has to provide some specific guidelines on.
There is tremendous variation in the amount of structured inquiry structured by Mill’s methods that you need to undertake to support different kinds of generalizations. The classic opposition is between finding the melting point of a metal where all you have to do is melt one sample and you know that any metal will melt at that temperature again, and this particular exact case of “all swans are white” where you know there was abundant evidence and yet it finally—it always seemed shaky and it eventually did turn out that there are black swans. You know I believe in Australia. I hope that’s where they were. That’s what I said in my book.
But I think the key factor here is that in the case of the metal and the melting point we have a good understanding of the mechanism by which a metal melts, and that understanding of the mechanism shows us that it’s in the atomic structure of it, the thing that defines it as that kind of metal, that governs the way it changes state. So all we have to do is find what the melting point is for one particular instance and we can generalize immediately. That’s because we have an enormous context of prior established theories about the melting points of metals. But prior to that point for example before the elements had been isolated where you dealt with metals in different ores, different ores do melt and behave differently at different temperatures. So you would have had to do a lot more varying of factors. A lot of varying of factors was necessary in order to extract conceptually as well as technologically the metal from the ore.
In the case of white swans it’s not clear what the biological basis for the coloration is or it’s not clear why all species that—all birds that are members of this species using the standard characteristics that seem essential to species membership here—how those relate to the coloration. So we are left in the position of having us knowing that there are—precisely because we don’t know the mechanism by which these things are related we don’t know, we do know that we haven’t identified yet all the factors and so we have to make an explicit proviso on that generalization that this is subject to our current context of knowledge.
That’s the best answer I can give you at this point.
Okay thank you and let’s call it.
5. Skepticism
Chapter 1: Introduction to skepticism
0:03
Welcome to the Principal Source audio presentation of Skepticism by David Kelley. Skepticism is the belief that knowledge of reality is impossible and that man lacks any rational basis for certainty or confidence in what he thinks he knows. By undercutting the efficacy of reason, skepticism forces us to fall back on emotion, tradition, the authority of the church, mystical revelation, and non-rational philosophies in order to make our personal and political choices. For this reason it is important to answer the attacks that skeptics have raised against the perceptual foundations of our knowledge. Concentrating on four of the classical skeptical arguments—illusions, dreams, the malevolent God or diabolical scientist, and the brain-in-a-vat—Dr. Kelley identifies the pattern that all these arguments take and exposes their underlying motive: the desire to exempt consciousness from the law of identity.
A well-known philosopher, teacher, and writer, David Kelley earned his PhD in philosophy from Princeton University in 1975. He has since taught philosophy, cognitive science, and other courses at Vassar College and Brandeis University. Dr. Kelley’s books include The Evidence of the Senses, The Art of Reasoning, and A Life of One’s Own: Individual Rights and the Welfare State. He was recently featured on John Stossel’s ABC TV news special Greed and is a popular lecturer and talk-show guest. Dr. Kelley is founder and, since 1990, the executive director of the Institute for Objectivist Studies in Poughkeepsie, New York. This lecture was given in 1987 at York University.
Dr. Kelley able answered skepticism’s traditional arguments by submitting them to objective validation and a process of proof which, in the end, shows that despite the skeptic’s claim to be cautious and rigorous in his approach to knowledge, he in fact is prepared to believe that anything is possible.
Thank you. I am an epistemologist; that is my field of knowledge. Epistemology is the branch of philosophy that studies knowledge—that studies the nature and validation of knowledge.
Chapter 2: Defining skepticism
2:23
Now in most disciplines, whatever their internal disagreements, there is general agreement on the existence of their topic or subject matter. But over the centuries epistemologists have spent an enormous amount of time and energy either asserting or trying to disprove the thesis that we don’t have any knowledge, that knowledge is impossible. This is the thesis of skepticism, and that has been especially the case since modern philosophy began with the work of René Descartes in the early 1600s, when I would say—I think most philosophers would agree—epistemology came into its own as a distinct branch of philosophy. Of course there were philosophers who discussed knowledge before that time; there were skeptics before that time. In ancient Greece there was a Sophist named Gorgias who, according to legend, wrote a book maintaining three theses: (a) that nothing exists; (b) that if anything did exist we couldn’t know it; and (c) that if we could know it we could not communicate it in language. He was a skeptic. Skepticism also flourished for a time during the Hellenistic period, the few centuries just before the birth of Christ, and many of the skeptical arguments that are now part of the common repertoire of philosophers date from that time. But as a general and fundamental problem in philosophy, as something that absorbs epistemologists, skepticism is a modern phenomenon.
Now in general skepticism means the belief that we lack knowledge or that we lack any rational basis for certainty or confidence in what we think we know. There are many forms of skepticism, therefore, depending on what type of knowledge is involved. For example, there is skepticism about the senses: How do I know that the world is the way that it appears? How do I know that the world as I experienced it is not a complete illusion? How, for that matter, do I know there is a world out there? Or there’s skepticism about induction and causal laws: How do I know that things will continue to behave in the way that I’ve observed them to behave in the past? How do I know the sun will rise tomorrow? Well, there’s skepticism about the existence of other minds: How do I know that what I take to be other people are not inanimate bodies, mindless machines without any consciousness present?
Now of all of these, skepticism about the senses is basic; that’s the fundamental one, because the senses provide us with our basic cognitive contact with reality. So all other knowledge that might be called into question depends on that sensory knowledge. If I can’t even know that the sun rose yesterday, then the question doesn’t arise whether it’s going to rise tomorrow. If I can’t even trust my perception that there are other bodies in the room, then it’s premature to ask how I know that there are other minds animating those bodies. So my subject today is skepticism with regard to the senses. As Jeff mentioned, I’ve recently published a book called The Evidence of the Senses in which I dealt somewhat briefly and in passing with skeptical arguments because I was defending a thesis called realism, which says that we do perceive the world and perceive it as it is, and the world is out there independent of us. Tonight I want to look at these arguments in a little more detail. I want to consider the classical arguments that skeptics have raised against the perceptual foundations of our knowledge.
Chapter 3: The impact of skepticism
5:54
Now these arguments are extremely abstract, theoretical, hypothetical. Some of them are going to sound, especially to those of you who may not be in philosophy or haven’t taken Locke—haven’t taken philosophy courses—what sounded like parlor games. And some of them are. No one is a skeptic in daily life. Skepticism as a doctrine is not the kind of global philosophy that people fight and die for. No one goes to the ramparts to defend skepticism. Even the great 18th-century skeptic David Hume admitted that when he left his study he went about his business the way everyone else did, on the assumption that the world out there exists and acts according to regular causal laws, even though he felt that he had proved that there’s no rational basis for that assumption. But that’s precisely why skepticism is important beyond the confines of epistemology. To believe that the senses cannot be trusted is to undercut any belief we might have in the efficacy of reason. So in the entire sphere of life outside epistemology—and there is life outside of epistemology, outside even the academic kinds of questions that we’re going to look at tonight—in the entire sphere of choosing values, making moral decisions at the individual level and then at the social level, engaging in politics, trying to structure a society that is suitable to man’s nature, in all of these areas skepticism leaves us with no reason to rely on reason. But we need some principles to use in making these choices, so we fall back on emotion, tradition, the authority of the church, mystical revelation—any one of any number of non-rational bases for knowledge. We can’t escape the need for some kind of philosophical outlook. Skepticism says we can’t rely on reason and therefore opens the door to non-rational philosophies.
There’s a personal incident that may encapsulate this connection I’m trying to draw. When I was a college student I remember one night at dinner in the dining hall overhearing a conversation—eavesdropping on a conversation—between someone who was apparently a conservative of the religious type, someone who believed that politics should be founded on traditional religious views, and someone else who appeared to be a critic of that view. Anyway the critic at one point took the issue right down to fundamentals and questioned the conservative’s belief in God, and he said, “How can you be so sure of it? Philosophers have poked holes in all the arguments for God’s existence.” And the conservative replied, “Why should I care whether philosophers can’t prove the existence of God? They can’t even prove the existence of physical objects. They can’t prove there’s a table here.” All right, so obviously their arguments are useless, and he felt justified in falling back on tradition or whatever it was that he relied on. And historically we see the same connection on a much wider scale throughout the history of Western philosophy. The periods such as the Hellenistic period that were dominated by skeptical views tended to be followed by the rise of non-rational views such as the religious cults that sprang up during that period in ancient Greece, of which of course the one that triumphed was Christianity. So the issues that I’m going to discuss today, no matter how abstruse they may seem, in fact have a vital significance. They are not simply a philosophical parlor game.
Chapter 4: Classical skeptical arguments
10:04
Okay, but with that context now let’s enter the game. I’m going to lay before you some of the skeptical arguments, and I’m going to do this quite quickly because there is a pattern. So first I want to go through the arguments, identify the pattern, then we’ll come back and examine the arguments in a little more detail. And the place to begin here is with René Descartes, a work of his called the Meditations. Descartes himself was not a skeptic; he laid out a series of skeptical arguments for the purpose of trying to answer them. But his text is regarded as a kind of locus classicus for skeptical arguments because he presented them so concisely and powerfully. So it’s a good text to work from.
Okay, four arguments. First argument is an argument from illusions, and we’re all familiar with it: typical introductory-psychology-type illusion, the bent stick in the water. If I put the chalk in the water, which I will not do, it would look bent, but in fact it would be straight because of the refraction of light. Descartes’ example was a tower that in fact is round, cylindrical in shape, but from a distance, because your depth perception is too weak, it looks square. What you see is as if it were a solid flat plane, so you think it’s square when in fact it’s round. Now Descartes concludes: “What I have so far accepted as true par excellence I have got either from my senses or by means of my senses. Now I have sometimes caught my senses deceiving me, and the wise man never entirely trusts those who have once cheated him.” So the argument here is: we know illusions are possible; how then do we know that all of our experience is not illusory? That’s the first argument.
Second argument turns on the phenomenon of dreaming. An illusion is a case in which we allegedly misperceive an object; the object appears to be something other than what it is, but the object is still there. In a dream there’s a more radical gulf, it seems, between our experience and the world, because the objects that we experience in dreams aren’t there. We dream about objects that don’t exist, that aren’t there where we are asleep, and events that are not in fact occurring; it’s all made up. So we have a full and vivid set of experiences in the dream of a world. So no matter how full and vivid our experience of what we call the real world might be, how do we know that that isn’t all a dream? How can we prove that we are not all asleep right now? You can see that this is a more radical argument because it calls into question the very existence of the objects that we think we’re perceiving. Some philosophers have referred to hallucinations in this regard, because a hallucination is an experience in which there is a more radical disruption of your perceptual experience than an illusion: you seem to see and hear things that aren’t there. But let’s stick with dreams.
Okay, the third argument. This is a little more sweeping. Suppose that there is a being who is like God only is malicious instead of benevolent. This being, a demon, an evil spirit, exists for the sole purpose of fooling me. That’s his only goal in life: to deceive me. And he has made me believe that I have a body, that I exist in a world of other physical objects, but it’s all make-believe; none of it’s true. Descartes says all of these things that I believe are merely elusive dreams by means of which he, the evil demon, lays snares for my credulity. In fact Descartes’ hypothesis about this demon is so sweeping that he felt he could not even trust his abstract capacity for reasoning. It might be that even though two plus three seems to him obviously to add up to five, maybe the demon is fooling him about that. Maybe the demon has got a grip on his mathematical reasoning abilities so that in fact two plus three equals something else; it just seems to him that it equals five. Now you can see that this is a very sweeping hypothesis because it rules out not just the sensory level of knowledge and our confidence that we’re perceiving a world, but any trust we might place in our cognitive faculties. But we’ll use it tonight; we’ll focus tonight just on its relationship to our sensory evidence. Okay, that’s the third argument.
The last one is really just a contemporary, vaguely scientific version of the same thing. Suppose, goes the argument, that I have been anesthetized completely. I’ve been subjected to an operation in which my entire brain is removed from my body—a complete neuroectomy (not complete because they leave the nerves in, but the brain)—and they have preserved my brain, put it into a vat of preserving fluid that keeps its biological functions going. They have taken the nerves that lead into my brain—the sensory nerves, the optic nerve, auditory nerve, etc.—hooked them up to electrodes which in turn are hooked to a computer which has been programmed to stimulate those electrodes so that the signals coming into my brain give rise to exactly the kind of experience that I’m having, so that I have the experience of waking up every day, having my coffee, walking the dogs, sitting down at the word processor to write an essay, for example an essay saying how ridiculous skepticism is. But in fact I’m not in the room; I don’t have a dog; I am a brain in a vat. Sometimes this is embellished for the science-fiction buffs by saying that instead of human scientists these are aliens who have secretly transported me to a planet circling Arcturus. But science fiction is my thing, so I will leave that embellishment out.
All right, now there’s clearly a pattern here to these arguments. In fact there’s a single argument that is being presented in different forms, and I want to identify the pattern.
Chapter 5: The burden of proof
16:33
When we perceive an object we perceive it as something existing outside our minds. We perceive it as something existing independently of us in a physical world. The experience purports to be an experience of reality. If it isn’t, then it’s invalid; it’s misleading; it’s non-veridical (the term philosophers use for it: non-veridical). Well, the skeptical arguments are just different ways of saying that perhaps all our experience is non-veridical. Whenever we think we are perceiving something, say the objects in this room, the argument tells us that we could be having exactly the same experience even if these things were not there. The different cases—the dreams or hallucinations or diabolical scientists—these are just different ways of trying to make graphic how such a non-veridical, a complete systematic but non-veridical experience could allegedly occur. Each one describes an experience, a type of experience that is said to be indistinguishable from our actual perception of a real world that in fact is non-veridical.
Okay, the argument then says we can’t prove that our experience is not of that non-veridical type; therefore we can’t trust what our experience tells us about the presence of objects and the nature of the objects that we see. All right, let’s look a little more closely at the presuppositions of the argument. It’s clear that these alternative types of experience, the ones that are non-veridical, they must have two characteristics. First of all they have to be indistinguishable from actual perceptions. If we could distinguish these non-veridical ones from actual veridical perceptions, then there wouldn’t be a problem: we just look and see and say, “Oh, now I’m having a non-veridical experience so I won’t trust it, and now I’m having a veridical one so I will trust them.” No, they have to be indistinguishable. Secondly these non-veridical experiences have to be possible; they have to be a kind of experience that actually could occur. If they were impossible, then once again we wouldn’t have to bother trying to rule them out; there would be no reason why we should have to prove that they’re not occurring. They can occur.
Now the key question in this regard is: who has the burden of proof? Does the skeptic have to prove it’s possible that we are having such an experience, or do we have to prove that we’re not having that experience? Who has to make the first move here? Does the skeptic have to come along and actually establish that such experiences could occur, or do we have to take it for granted that they could and prove that they’re not happening? Now I’m going to argue that the skeptic is the one who has the burden of proof and that he cannot discharge it. Any attempt to show that such experiences are possible in fact presupposes the existence of physical objects and some reliability of sense perception. So that’s my basic strategy.
Okay, let’s start by examining the context in which any such argument must be presented. The context is our actual experience. The skeptic says that we have to start from what’s given to us in our direct experience; we can’t rely on any other knowledge that we have. For example we can’t rely on evolution and say that we were selected for—our senses were selected for their ability to represent the world accurately—because the skeptic will say, “How do you know about evolution? The basis, evidence for evolution all traces back to sensory evidence at some point.” So we have to start just with what’s given to us in direct perception. We have to start with what is self-evident. All right, that’s fair enough. I think we should start there too. Let’s examine what is given and self-evident to us in our perceptual experience of the world, and we’ll consider it in and of itself as the skeptic requires. We will try to put ourselves in the position of children looking at the world without any sophisticated knowledge about how the world works, how their senses work, or any of that. We’ll just examine our experience from the inside as conscious subjects who are capable of having these experiences.
Chapter 6: The primacy of existence
20:45
The skeptic’s conclusion is that our experience, say of an object like the podium, might be invalid; there might be no podium here. Now if he just asserted this without any argument, his assertion would have to be dismissed out of hand. We would have to say that’s impossible; it’s not possible that the podium isn’t there. Why do I say that? Well, think about it. Examine your perception. You’re aware; it’s a form of awareness, and to be aware you must be aware of something. What is that something? Well, as far as you can tell it’s this object here. When you reflect on your experience of these objects—you’re aware of it as an experience of certain objects. Take away the objects and there’s nothing left; it all goes dark, and you would no longer be aware. To be aware is to be aware of something. And this something more we can observe: we experience our perceptual awareness as metaphysically passive—that is, we don’t experience it as creating the objects but as revealing them. The podium, the walls, the people in the room—all of these exist in a world that you are in, not a world that is in any way in you. Your attention to a given object as you turn your head to it and then turn away—your attention may come and go, but you don’t experience the onset of your attention to an object as the object springing into existence. As you turn to look at an object you experience that object as something there which comes increasingly into your visual field, something that was there for you to explore and then you chose to explore it and see what it was. Moreover the things that you perceive have certain characteristics. The podium has a shape, a color, size; it has relationships to other things, spatial relationships to the other objects on the table; it engages in certain actions. It’s not very active right now, but there are certain things that it can do and certain things that it won’t do. You observe that as well. These constitute its identity, and its identity is something that your experience reveals to you. It’s not that the identity of this object depends on you; the thing is what it is. That’s not what you make of it.
Now we can summarize all of this in terms of what I call—this is Ayn Rand’s term actually—the primacy of existence. Primacy of existence says perceptual awareness is inherently a relationship between you and a world of objects; that to be aware is to be aware of something. It’s a relationship between subject (you, the subject of consciousness) and an object, between consciousness and existence, and existence is the primary term in that relationship. Consciousness doesn’t create what it is conscious of; it reveals it.
Now I’m not dismissing skepticism yet. I’m simply trying to describe the way things seem to me as I examine my own experience from the inside, and I assume that you would find the same things if you examine your experience from the inside. Now maybe this perspective that we’re adopting here is wrong in some way; we’ll leave it open for the skeptic to try to prove that it is. But if he’s asking us to start from what is self-evident, we have to examine what is given, and I would say that this observation is in fact given.
Now so let’s consider how a skeptic might proceed. Suppose that a skeptic came along and said, “Maybe you’re not conscious; maybe you’re not aware.” Well, you know that’s wrong; it’s obviously—well of course you’re aware; you can observe your own awareness; it’s self-evident. In fact you’d have to be aware; you have to be conscious even to understand this supposition. If the skeptic did manage to create any doubt in your mind as to whether you are conscious, the very doubt would itself be a conscious state. So you can see there’s no way for the skeptic to call into question the fact that you are conscious. This is something Descartes himself identified; it’s the principle of the cogito, the famous argument cogito ergo sum: I think therefore I am. This is the one thing he felt that the demon couldn’t fool him about: that he was conscious. Okay, now that’s something that most skeptics would agree with. They say we can’t be skeptics about the existence of our own consciousness. Fair enough.
Well now let’s suppose a skeptic comes along and says that perhaps the things that you see, the objects of your awareness, perhaps they have no independent existence; they exist only in your own consciousness. Well this supposition is also unintelligible, and it’s unintelligible in the same way as the supposition that we’re not conscious. Why? We can isolate our own awareness when we examine perceptual experience; we can isolate the element of awareness only by contrasting it with the objects that we’re aware of. We look at this experience; we said, well all I see there’s an object and I’m aware of it. My awareness is that which isn’t the object but a certain relationship between me and the object. So someone who suggests that the objects are in our consciousness is undercutting the very contrast between object and awareness that was necessary in order for us to form the concept of our own awareness. The skeptic is, so to speak, stealing the concept of consciousness from the context in which it is possible to form that concept. It’s as if someone tried to explain the meaning of the term “female” by contrasting female and male and then said, “Oh but you know it could be that all males are female.” Well no it couldn’t be, or if it is then I don’t understand what “female” means anymore because you told me what it means only by contrasting it with “male.” You see you can’t use a concept to undercut the cognitive basis of that concept; that drains it of meaning, and that’s what the skeptic is doing so far as we can tell from this perspective that we’ve adopted on our own experience. So the skeptic cannot simply assert that the objects we perceive may not exist apart from us. He has to offer some argument; he has to give us some reason to think that the perspective we’ve just adopted of examining our experience from the inside is not the whole story; it’s naive in some way; it does not reveal all there is to be known about consciousness, awareness, or the way it works. He has to find some way to drive a wedge between the objects that we think we can see perfectly well and ourselves as conscious subjects. Now it is possible to do this to some extent and in a certain respect. This perspective of examining our own experience from the inside does not tell the whole story about consciousness; that’s true. And this is where the skeptical arguments come in. So let’s turn now to those arguments and see whether they can drive a large enough wedge between the objects that we’re aware of and ourselves to make us doubt the validity of our senses.
Chapter 7: Refuting illusion arguments
29:01
First of all the skeptic says that illusions occur, and so let’s ask what an illusion is actually. That is a rather long story in my view anyway, but the key point here is illusions are possible: the bent stick in water and the others. How are they possible? Because the properties that I’m aware of in objects, such as the shape or color in some cases, these properties are things that I attribute to the object as intrinsic properties. Okay, when you look at the podium everything you see—the color, the shape, the size—you attribute those properties to this as intrinsic properties of the podium itself, right? Everything is out there as you experience it. But in fact not everything is out there in that sense, because we know that some properties depend in some way on our own perceptual faculties or on the causal chain that lies between the object and ourselves. That’s the case for the bent stick. The reason the stick looks bent even though it’s straight is that light is refracted by the passage from water to air. Okay, but the question now is: how am I going to come to know that? How am I going to come to know that illusions are possible? How, for that matter, does a skeptic know they’re possible? Well let’s look at an example. Let’s take a case of an illusion that might fool anyone. You walk into a room; you look at the curtains; they’re rose-colored. But in fact the curtains are beige; they’re just being illuminated by the setting sun so they only look rose-colored. Well this is a case in which if you didn’t know what was going on you might well make a mistake; you’d say these are rose-colored curtains; the rose color is really there in the curtain. That’s not the case. But how can I come to know that it’s not the case, and how can the skeptic come to know that? Well obviously the skeptic knows that the curtains are not in fact rose-colored because the skeptic knows that they are beige; he’s been in the room before and he saw that they were beige. Also he knows that the setting of the sun cannot possibly change the actual color of the curtain. All right, maybe over years and years it will, but not at a moment, and sunlight just isn’t that strong. Now all of this is knowledge about the world; all of this is knowledge about physical objects, the nature of light, and involves actually a lot of causal knowledge. Where did the skeptic get that? Obviously through his senses. How’s he going to convince me of it? By getting me to trust my senses at least far enough to see what curtains look like when normal light is falling on them. So however you put it, the skeptic knows that illusions are possible because they actually occur, and he knows that illusions actually occur only because he knows that some of the apparent properties of objects are incompatible with their real properties. The apparent rose-color is not compatible with what he knows to be the real color of the curtains. But of course that’s knowledge of reality. He’s relying on his knowledge of the way objects really are in order to discover that illusions are possible. So you cannot then turn around and use the existence of illusions to call into question any capacity to know what real objects are like.
Now in general let me try to formulate the general philosophical point here. It’s true that not all of the attributes that we experience objects as having are intrinsic features of those objects. Some of them are relational properties; they result from the interaction between the object and our sense organs in the medium of perception. That’s because the senses work in a certain definite way; they’re not exempt from the law of causality. They operate according to their own nature by responding to light or sound waves or chemicals in the air or whatever it might be, and responding in certain definite ways, and that is bound to affect the way objects look. Okay, but we discover that, and that is a fact that’s not available to us; it’s not visible to us if we look at our experience only from the inside. From the inside I don’t know that I have visual receptors or a visual nerve leading to the brain or, for that matter, a brain. I don’t see any of that. What I see is the object. It’s only by pursuing a certain course of scientific inquiry that we learn that these things are possible. But that is inquiry about the world that we’re perceiving, and therefore any doubt about the validity of our initial perceptions would call into question the knowledge that illusions are possible or that objects may not always be what they seem. So the skeptic is trying to undercut the very knowledge that he depends on in mounting his argument. So in fact he’s not calling the primacy of existence into question; he’s reaffirming it.
Chapter 8: Refuting dream arguments
34:23
All right, now let’s turn to dreaming. What I’ve laid out for you is a general pattern of an answer, and I’m just going to be applying the same pattern in all the different cases, but each one raises slightly different questions. Let’s look at dreaming now. Once again dreams are a real phenomenon, but once again when the skeptic appeals to this phenomenon he’s reaffirming the primacy of existence. Okay, the obvious point about the argument from dreams is that dreams are supposed to be indistinguishable from waking experience, and the obvious problem is that each of us does in fact distinguish dreaming from waking. We do it roughly once a day, usually in the morning; some of us do it more often, and we do this throughout the whole course of our lives. No matter how good a dream may be, when we wake up that dream pales beside the kind of contact with reality that we have when we’re awake. By the light of day a dream takes on what we call a dreamlike quality. They’re often incoherent and contradictory, and even if they’re internally consistent they don’t fit into the pattern of our waking experience. We dream we are in Tahiti and we wake up somewhere else, and we could not have gotten from Tahiti there in the time it took us to wake up.
Okay, now skeptics usually reply that in a dream we may have the experience of thinking that we’re awake. Many people have had such dreams: you are in a dream and you wonder whether you’re dreaming or awake and you decide that you’re awake. Now philosophers have a specialized version of this, very common among philosophers; you usually have it in graduate school. I mean I had this dream myself when I was studying these issues: you dream that you’re in a philosophy seminar arguing with a professor who is presenting the skeptical arguments, and you say, “No I can’t be dreaming; I know I’m not dreaming because my dream wouldn’t be this clear and vivid and consistent; it wouldn’t fit into the whole pattern of my life the way this does”—and then you wake up and feel like a jerk. But of course you feel that way only because you then know that you were only dreaming. But in any case this particular fact—that we cannot distinguish dreams from waking when we’re asleep—doesn’t prove anything about what we can do when we’re awake. The fact that our judgment is sufficiently deranged while we are sleeping that we cannot make this distinction doesn’t prove that we are deranged when we are awake. But in any case the skeptic himself presupposes that we can distinguish dreaming from waking. Why? Because he assumes that dreams are non-veridical, that the events don’t really occur, the objects don’t really exist. That’s implicit in the very concept of a dream. Okay, so how does the skeptic know this? How does he know that the events that he dreamed of didn’t really occur? Well because he formed the concept in the same way we did: he wakes up every day, knows that what he was just experiencing was only a dream and therefore wasn’t real. If we really could not distinguish dreams from waking experiences, then we would have to assume that all our dream experiences were true, were veridical. I dreamed I was chased by tigers last night. Now if I couldn’t distinguish dreaming from waking I would have to assume that I was chased by tigers last night. All my experience would be of the same type and I would say it must all be true. It’s only because I know that there’s a difference between dreaming and waking that I can say the dream experiences aren’t real, and it’s only because I can say they aren’t real that I can raise any question about veridicality. So the skeptic, in using dreams in his argument to try to undercut waking experiences, in fact is assuming that he can make that distinction. So once again he’s presupposing knowledge of reality in the attempt to argue that we can’t trust any knowledge of reality that we get through our senses.
Chapter 9: Refuting demon fantasies
38:35
Okay, let’s turn now to the evil demon, which raises a slightly different question, because here there’s no need or possibility of arguing that an experience given to us by this demon would have to be distinguishable from a real perception. That’s the argument we made in dreams, but we can’t make it here. Why? Well the demon is Descartes’ invention; this is Descartes’ story he’s telling. Okay, he’s attributing to this demon godlike powers: omniscience, omnipotence, at least sufficient power to do whatever he wants with our minds. Okay, the demon can control our reasoning so we get mixed up about simple arithmetic, so why couldn’t it give us an experience that is just like an ordinary perception, a systematic delusion? This is Descartes’ own personal fantasy; he can fantasize whatever he wants and he can attribute to the demon whatever power he wants. But the question is: how can Descartes show us that such a thing is possible, that such a demon could exist and could give rise to the kind of experience that he is talking about? Obviously he’s not going to establish that this is possible by giving us an actual example, as in the case of illusions and dreams. One of the significant things about demons is that there aren’t any, so there are no examples to be given. So is Descartes’ capacity to imagine the demon a proof that it’s possible? No. To say that something is possible is to make a statement about reality, and nothing follows about the nature of the world from the fact that Descartes is capable of having these arbitrary subjective fantasies except that Descartes himself is capable of having these arbitrary subjective fantasies. Okay, for example it’s perfectly possible to imagine a contradiction. I see no reason why one can’t imagine the internal angles of a triangle equaling something other than two right angles, even though we all learned in high school a proof that that would entail a contradiction and therefore the two internal angles of a triangle must equal two right angles. But I can imagine them equaling 179 degrees, 181 degrees—why not? Descartes would argue that the hypothesis of the evil demon is logically possible, that there’s no contradiction involved. But remember that the context in which he is trying to persuade us of this is the naive perspective that we’ve adopted in which we’re considering only what is given and self-evident from the standpoint of our looking at our experience from the inside. Okay, in that context it’s a straightforward contradiction to say that the objects we’re aware of might not actually be there. That’s a contradiction. If they weren’t there then there would be nothing for us to be aware of. If the idea is that they might just be images in our consciousness, then again from this perspective there’s no way we can make sense of that; that’s unintelligible. To say that they’re in consciousness, for the reason I outlined before: we form the concept of consciousness only by contrast with objects outside it. So either way there’s no basis for believing that this demon and his power of deceiving us in this way is even logically possible; there’s no way we can make sense of it.
Chapter 10: Refuting the brain-in-a-vat
42:09
Okay, let’s turn to the final argument now: the brain in the vat. This combines elements of both of the preceding types of cases that I talked about. Unlike Descartes’ demon this hypothesis has a sort of a vaguely scientific air; it bears some relation to what we know about the way the senses work, and in that respect we have to treat it along the lines of illusions and dreams. But on the other hand there’s a sense in which when push comes to shove in the argument it turns out that the diabolical scientists are being attributed in effect omnipotent powers, which makes them very much like Descartes’ demon but in slightly more modern dress. All right, at a fundamental level I think the answer that I would give to this argument is clear. The story is about my senses, stimulation of my sense organs, of my sensory receptors and the fibers leading to the brain by electrodes which are wired up to computers. Somehow I have to be told; I have to be given now to make this intelligible to me. To make this useful, to call into any doubt what I derive from my experience, you have to explain to me that there are electrodes, that there are computers, that I have sense organs, that there’s an optic fiber leading to my brain. This is knowledge of reality. Clearly this supposition presupposes a whole raft and background of knowledge about the world. So we can’t then use that knowledge any more than the other cases we’ve talked about to call into question the basis on which we know it, namely our sense perception. If we couldn’t trust our senses then we could not have acquired the knowledge that there are electrodes, that the senses work this way. Okay, so in general the argument presupposes that a physical world exists and that it operates in certain specific ways, so we can’t then turn around and use that knowledge to cut off its own foundation.
But let’s play along with the argument a little further. Let’s suppose that we’re not the victims of the experiment; we know that we’re perceiving, but some other poor victim has been chosen and his brain excised and he’s been hooked up. Yeah, is it possible in that case, and how would he know? Would he be able to know whether he is a brain-in-a-vat or a live, normal functioning human being? Well I don’t think that’s clear actually, whether that is possible. It’s not clear whether the brain would continue functioning under those circumstances or whether the computer programs, the electrodes, could provide a sufficiently rich array of stimulus information to completely duplicate ordinary perceptual experience. I’m skeptical about that. But suppose it is possible. At the very least we know that the illusion would have to be total, and that would mean that not only the incoming nerve fibers would have to be programmed but the outgoing motor nerves that control—or used to control—the guy’s eyes and arms would have to be hooked up to the computer as well, because our perceptual experience takes account of our own motion. As we turn our eyes we integrate the information that our eyes are turning with the change in the scene that we’re seeing; that’s how we know that we’re perceiving a stable world even though our eyes are moving. So we would have to wire up this brain so that it not only could perceive but could act; they could initiate action. So now we have a subject who is getting reliable information about something outside him, namely the computer, and he is capable of acting on that, and his own actions will affect the nature of the stimulation he’s getting. I would say in that case he is perceiving a physical world—the computer, the world of the computer, depending a little bit on how this story was told—but he’s perceiving something. Okay, and he could eventually discover that the something he’s perceiving is different from the something that we perceive by normal scientific inquiry. It would take him some time, but eventually he should be able to discover the underlying nature of what it is that he perceives. It’s just that the outcome in his case of this kind of inquiry is different from the outcome in our case. We discovered that what we perceived, its underlying nature is atoms with properties that allow them to bind into macroscopic objects and reflect light of a certain wavelength. He would discover that the underlying nature of what he perceives is certain electronic states of the computer. But there’s no reason why he could not discover that. After all the computer is supposed to be something different from what we’re perceiving, so eventually that difference would have to lead to some observable consequence—unless, and this is what the argument really presupposes, the scientists have the power no matter what the subject does, no matter what test he proposes or experiment he performs, to suspend or to step in between the underlying nature of the computer and the consequence the subject might observe and fool him, throw him a curve. But you have to assume that they could do that for any causal connection whatever, which means they would have in effect to be omnipotent. They would have to be able to anticipate every conceivable thought that the subject might have, but that in effect means they have to be omniscient, at least have a form of cognition that goes so far beyond the human that we have to call its very possibility into question in the same way that we question the possibility of a demon. So this is the respect in which really we’re not talking about scientists or a clearly possible situation; we’re talking about someone’s imagining in effect a god-like being.
So my final conclusion I draw about this is that the problem with the skeptic is really that he’s too credulous. He claims to be extremely cautious and rigorous in his approach to knowledge, but in fact he’s prepared to believe that anything is possible. He’s prepared to believe this to an extent that would put to shame the readers of a supermarket tabloid, the kind with headlines like “Medical Breakthrough: New Hope for the Dead.” That’s child’s play compared to evil demons and these diabolical scientists. So in that respect there is a certain similarity between the mystic and the skeptic, and you remember at the beginning of the lecture I said there is historically a connection between those two approaches to knowledge. Given what we’ve just seen I think that connection shouldn’t be surprising.
Thank you.
Chapter 11: Q&A and conclusion
49:29
Okay, let’s get started. Hi. Before I begin with the answers I’d like to give to the written questions and then open the floor to York—there’s spontaneous questions—let me pass along a request in regard to the taping: if you could speak clearly and distinctly, to use Descartes’ phrase, so that the tape can pick it up, and try to make the question as concise as possible. I don’t want to discourage discussion but make it as concise as possible so that if I need to repeat it I can hold it in my own consciousness.
Okay, let me start with a couple of written questions that I got. One asks: could I define knowledge? I didn’t define knowledge during the talk and yet I was talking about its possibility. This actually is a somewhat difficult question because it really depends on the context in which we are functioning and the need we have for the concept. In a broad sense knowledge is just another name for awareness or cognition, any grasp of reality, and that I would say that concept is not definable because it’s a primary; it is an axiomatic concept. That’s why, because awareness of reality is so general, all we can do is point to it, grasp it by direct observation and from the inside because we are conscious, and we can give synonyms: we can say just to be aware, to be conscious, to grasp the fact, to cognize the fact, but all of these mean the same thing. In this sense the concept of consciousness is like the concept of existence. There’s no way you could define existence; all I can do is give synonyms. Are we okay?
However there is a need for a narrower concept of conceptual-level cognition, conceptual-level grasp of reality. This is what most philosophers in epistemology today mean by the term “knowledge,” so they define it in terms of the conceptual knowledge that we formulate in propositions as a complete sentence. And in that case then I would say knowledge is the conceptual grasp of reality, and because of the way concepts are formed that implies that it is based on sensory perception but represents a higher level of integration that involves abstracting common properties from similarities and differences that we form concepts like table, chair, force, etc., and formulating them according to and establishing their truth according to the rules of logic.
Okay, let me turn to another question. All right, yes I mentioned an example of the sun rising tomorrow. Why isn’t that—why is the skeptic not right? How do I know the sun will rise tomorrow? Well let me reformulate the question. The sun—at least I subscribe to the heliocentric astronomy—so the question is: how do I know that the earth will keep turning so that the sun, as we say, comes up? Why does—how do I know the earth will keep turning? Because—well I’m not a scientist and I think I’m in a room full of many scientists so I hesitate to lay out the physics of it, but it has to do with the mass of the earth and the fact that it is something about conservation of angular momentum; it couldn’t just stop without a cause. Right now how do we know that? Well we know it because the fundamental laws about the behavior of massive objects. You know, how do we know that these laws are true? Well because we have identified it’s part of the identity of an object to act in a certain way; that is part of its identity, and its identity is its identity; it won’t change without some cause. Things act the way they do because they are the way they are. This is an Aristotelian view of causality, and it’s what it means is that the law of causality is a consequent, is an implication or part of the law of identity. It says the thing has a certain identity, so that’s all I can say. We have discovered that the identity of rotating objects is that they continue in rotation until they’re stopped. We have abundant inductive evidence for that. Now there are very specific technical questions about how that’s a justification of induction that arise, but the basic idea of things suddenly beginning to act in some way other than the way its nature has—the way we have observed it acting in the past—without any change in its nature means that actions are detached from the identity of an object, that actions are free-floating, random, inexplicable phenomena, and that gives a metaphysical primacy to action that it doesn’t have. An action is an action of an entity, so the type of action an entity performs must be consistent with its identity. Okay, that is, as I said, that’s a fundamental question, and I’m sorry if I got the physics wrong. Okay, basic idea is right.
Okay, now here’s a question about the evil-demon example. I think it’s about the evil demon: a comparison between a benevolent and a malicious God. One can argue that a benevolent God through the design created world number one; in this creation our perceptions and illusions are self-consistent and preordained. Okay, and the inhabitants—I’m not sure what this means—the inhabitants have no way of exploring outside the limits that this God has ordained. We can—the God could equally well throw the dice and create world number two that has a separate set of rules and perception of reality. This is a many-worlds point of view. Okay, well I think that’s the essence of the question: many—there are many worlds that are possible; how do we know which one we’re in? Okay, well I don’t think there are many worlds; there’s reality. That’s what you mean by the world. I don’t think that you can start with the idea of possibility and project a whole range of worlds, realities that are all possible, and then say well we live in one that’s actual but you know maybe one of the others could be the case and we can’t prove that it isn’t. The concept of possibility has to be derived somehow. That concept doesn’t—concepts don’t spring into existence as cognitive goods; they have to be produced. The only way to produce them is by integrating sensory material. So I would say you have to—for the only way to form the concept of possibility is to start where we started tonight, looking at the world, the objects you see, noticing that they can act in certain ways and not in other ways, forming the concept of possibility and using it accordingly. But you cannot reverse the process and start by saying this, that, or the other is possible just because it can enter your mind, just because you can conceive of it. So I’m not sure I have answered what the questioner was getting at, but I think the idea in surprise—the idea turns on the question turns on the idea of possible worlds and a whole raft of possible worlds that God could have created. I don’t accept the concept of possible worlds in that sense. Also I don’t believe in God, and this is relevant here because again to show me that it’s possible for God to have done something the first thing would have to be to show me that there’s a God, and again I start where we started tonight with what is self-evident to me, and if anyone can offer such an argument I’ll be happy to entertain it, but I’ve never heard an argument that starts from actual observable facts and shows that God is even the possibility that a God exists: a pure consciousness detached from any physical being which has omnipotent powers and omniscient means of knowledge. So okay.
Now let’s take any questions from the floor.
And when have something—yes. Say you might be wrong. I believe you or whatever, and it seemed that some people who discount accept any notion of certainty, and there are few six, seven, eight, nine, seventy other than walking away. That’s great. Dr. Hume in some period of time—have you found a useful technique for convincing people like this that certainty is possible? Have I found one? Yes. No. Yeah, it’s—at some point it is, because if you have explained why the possibility—the fact that it’s possible for you to make a mistake does not mean that you’re actually making a mistake in this case—unless you can be given some evidence. Or what if you’ve gone through that and explained it, then the argument just goes in circles; there’s no further point to take it. Or in the case if it’s for some skepticism about the senses, if you go through the arguments and the person keeps saying well but it’s possible there’s this demon or possible scientist or whatever, and you ask how you know it’s possible and the person still says well but it is, then the argument has to stop. Now I think that because Ayn Rand had an analysis of skepticism of what the motivation behind it was, and since I’m confident that it’s false I will entertain hypotheses about what the motivation there is other than, you know, reasonable misunderstanding. And but I think it really—it’s useful; I mean it’s a true analysis and fundamental: it’s that the skeptic shares the same ideal of knowledge that the mystic has, namely effortless, automatic, infallible knowledge, knowledge that entails no risk or requires no effort. The skeptic says there isn’t any such knowledge so there isn’t knowledge at all, and the mystic says oh yes there is; no such—like that I have it because God spoke to me or whatever. And if that is the motive at work there’s nothing—no argument you will give to persuade someone who simply doesn’t want to accept the idea of knowledge that requires an active process that takes effort and entails possibility of error. You just stop.
Yes. When does the line get crossed between dreams spoken of skeptically and dreams spoken of mystically as far as senses and perception are concerned? Could you tell me what you mean by dreams spoken of mystically? Dreams pursue dreams skeptically: one speaks of having dreams and thereby how can we know that we really are awake or not; then dreams almost become perception to mystics. I see, I see. Okay, the question is: how do we distinguish—for the tape—how do we distinguish dreams, the skeptic’s interpretation of dreams from the mystic’s? And that’s an interesting point. The mystic is someone who believes that his dreams are true, right, that he actually was visited by an angel last night. Zack had mentioned one particular dream of some significance in Western culture. The line that I wouldn’t say there’s a line between the two types of dreams because I think there’s only one type. The skeptic is right about dreams: the objects don’t occur. But as I say, because of the connection between mysticism and skepticism in the concept of knowledge, it’s not surprise. Dream in effect is the ideal if only it could be true: it’s effortless cognition. Yes.
Okay. In mathematics there’s the need to go, it seems, beyond the immediate sense perceptions and forming concepts and basic things in geometry like point, line—terms like that that I can see how they could be formed from pure sense experience. How would you form sort of the concepts of infinity and all these useful things in mathematics that don’t have the same immediacy? That’s like you know James. Okay, well I think they all can be formed by a process of abstraction from sense perception. There are a lot of technical questions about how it’s done in the philosophy of mathematics, but in essence I would say you form the concept of a solid object by identifying the three dimensions of physical space and then you can abstract a two-dimensional object from that. Okay, the thing has a surface; you can form a concept of a surface because there are such things, and then you simply disregard certain aspects of the surface like that it has a depth and consider it only insofar as it has these two dimensions, and then you can consider a line, and even though the line has some breadth and depth you disregard those and you focus only on the length as on that dimension. But you do that with the understanding that you’re not going to pay attention to it but it does exist; any actual line would have some breadth and depth. And finally with the point it’s the intersection of the lines considered as having no length at all but simply as a location, but you realize that any actual location would have some size and therefore some measurement in all three dimensions. Infinity I would say we form the concept by the fact that there are certain processes that we perform which like adding to—adding we know one, two, or counting rather one two three four five—which we can in any actual case we count only up to a certain number, but we can see that there is no particular need to stop at that number; we always could go further, and so we abstract; we just disregard the fact that we always have to stop at some point and focus only on the fact that there is the absence of any necessity for stopping at a particular point. Now I mean this is very rough, but that’s how you—you know I think you can form the concepts that are used in mathematics without—on the basis of sense perception even though there are no such things as actual points or infinity. Okay.
Yes. They read about causality; they usually describe in different terms. One who civic that actually can’t happen before its commerce we can’t go back in future, and another one is that if you set up some system and you establish its initial conditions they’ll go through the same identical process over again. Some of these settlements same conditions. How do you like these them two common expressions that comes out I’ll leave normally using. Well the basic idea of causality that I have, as I said, was an Aristotelian one in which—which forms the concept of causality in the country by observing that entities act. Now entities act in certain conditions, and the way that they act, the particular action that an entity will take, depends partly on its own nature and partly on the conditions which are impinging on it. But the way it—the primary term here is the entity’s own nature, because the way it will respond to those conditions depends on its nature. You know if I have a—I think here and I’ve you know run the billiard ball to collide with it, the collision is the initial condition, but what happens will depend on the nature of the object. Care if it’s another billiard ball it will roll off; if it’s a glob of shaving cream it will disperse; if it’s a negative rake him all over the table. So the deputy understanding of the concept of causality that you gave was the relationship between the conditions in which an entity acts and the action it performed as a result, and that’s certainly a real relationship, but it is mediated by the relationship exists because of the nature of the entity which is acting. So I would—to give a full account of causality you have to include both elements. Does that answer your question? Well I would say that the fact that a cause can’t—I mean in effect cannot occur before its effect—is an implication of the very concept of the way we formed the concept of causality. We observe that actions happen and actions are not autonomous free-floating existence; they are actions of entities so they can be explained. But by the very nature of time—maybe this is a self-evident truth—by the very nature of time that which explains an action must be something occurring simultaneously with it or prior, certainly not after it. But I’d still say that’s a derivative; that’s an implication or a corollary of the basic idea of causality.
Okay, other—yes. The question is couldn’t the scientists in this example distort his perception sufficiently to prevent it from doing that? Well certainly they could program the computer, assuming this is even possible in the first place to preserve the brain that way; they could stimulate those electrodes in such a way that his experience is totally chaotic, in which case he could not even form perception of solid enduring physical objects. But in that case then that would clearly be a distinguishable form of experience from the one we have. Remember the examples—all these arguments depend on the assumption that the experience that the victim gets is indistinguishable from the one that we have and therefore must have the feature that we can act on our world and observe consequences. Okay, we can observe we pass our hand in front of something and find it occluded visually; we move around and notice perspective changing. Now if the scientists don’t want to do that then they’re not representing the experience, and so they have no—the case is not described in such a way that it provides an argument for this. So I was assuming that they would somehow stimulate these electrodes in such a way as not to distort the perception in that way. Now okay I think that’s the other questions.
Yes. Dreamy and historical overview you gave: the scepticism in the Hellenistic period was followed by period of irrationalism. The dart babies then you said today’s skepticisms very active. Desert doesn’t focus on the future test-drive the connection between skeptical views and like what are the consequences. Well the basic connection is that the one I mentioned is if the prevalence of skepticism puts reason out of commission as a source of moral, political guidance, puts it out of commission as a source of any philosophical outlook, so people look to other sources. Today skepticism is—why I would say is widespread in a certain sense, usual more often implicit than explicit, but the fact is these arguments are quite common and it’s commonly said that we have—there’s no basis for certainty, and in fact what we observe is people turning to other types of philosophy: this new rise of religion including fundamentalist types of religion, tremendous Easter interest in Eastern religions especially true during the 70s, but this is all part of a cultural phenomenon now. So I would say the extent that skepticism is prevalent it’s not a good time, since I believe that a healthy culture and society is one that is confident in reason. I think that’s the full answer today.
Yes. So you said that we can perceive reality therefore we can say that exist. He’s also saying that the reality that I’ve received is the same reality. Okay there were a lot of elements in your question; let me try to deal with them one at a time. And the main question is how can I be sure that we’re all perceiving the same living in the same world given that we in various respects perceive it differently. Well one element in what you said was some people do not have all the senses that others have, or for example Helen Keller was blind and deaf, or some people range of sensory experience cover is wider than ours; some people well take species: dogs hear higher silent at higher frequencies than we do. Well no one would claim I certainly would not claim that we have direct perceptual contact with every aspect of reality. No, we have protective contact only with a tiny slice of it, with a certain band of electromagnetic radiation, different band of auditory frequencies, objects of a certain size no bigger no smaller and so forth. The rest we have to learn by extracting information from what we can perceive and extrapolating through forming concepts and scientific inquiry so that we discover that the range of colors we see, the visible spectrum, is part of the myth; it’s on a dimension that extends in both directions, and same with the other senses. Someone who had who say perceived in the x-ray range and not the visual color range would see the world quite differently from us; that is, would see a different aspect of it. But once that person discovered that his narrow slice was part of a longer wider continuum he would identify that there are frequencies, the frequencies that give rise to visible that we call visible light that have certain properties, and even if he took it far enough he could realize that that was stimulating certain types of sense organs, the ones we have, and so although we would start in different places we would arrive at the same view of the world. Now it’s also in the nearsighted-farsighted case and the case of people who have color blindness, whatever species that perceive apparently quite differently from us—all these cases I have to make an additional point which is the way in which we perceive the world does depend on our own sense origin. My claim is just that it is a world that we perceive, and we can’t untangle from just by examining from the inside like we can’t untangle which properties are intrinsic in the object and which are partly the result of the way our senses respond to them. We have to discover that by inquiry, and we have discovered some things; as I’m sure there’s more to be discovered. The common world we live in is the one of objects with their intrinsic properties, and to the extent that we have all are of the same species and have sense organs of the same type we do ten mostly ninety percent say perceive the world the same way—not only the same world but in the same way. We all see colors and your tone and so forth, although there are small variations. So I’m not claiming that we all perceive the world identically, just that we perceived in the same world out there. And when you get to the case of the poets and the physics these are forms of knowledge at such high levels of abstraction that are the result of so many lifelong choices of what is important and what to focus on that it’s perfectly possible for people to see the world in that sense quite differently, but it’s the same world that they’re seeing; they’re just seeing different aspects of it.
About that you’re not receiving it all and captivating their world that he dipped it without talking total awareness which is ugly but they calculate that accumulated up total awareness is to be—no no I would reject that. The way you cannot make as a condition of knowledge something that is impossible, and omniscience is not possible. So any standard that says to know reality we have to know everything and we can’t know everything therefore we don’t know reality is relying on a false premise that knowledge—that to know reality we have to know everything. There’s no basis for that precisely because we can’t; that’s not how the mind works. We’re not omniscient. They work and omniscience nowhere would have to be something that knew by magic. Does any specific process of knowledge would have to be finite and operate to provide specific finite knowledge. We will continually grow into by a real and the negative physics our future generations will discover something else that you know knowledge that the progress of knowledge won’t stop until we run out of things to know, and I don’t see any reason to think that we ever will, not because the universe is infinite but just because there are infinite—there are there is no limit to the kinds of integrations we can make among all the facts there are. But the fact that there are things yet to be discovered doesn’t mean but in fact presupposes that it doesn’t mean that we haven’t discovered anything yet. The fact that science will make advances on what we now know doesn’t mean that we don’t know what we now know. There’s a difference between a part of something and an illusion. What we see is true even if it’s not the whole story. Let me stop here because we’re starting to go around in circles a little bit. Enough I want to do want to see if there any other questions. Yes.
Yeah I don’t know who the first person to put it that way is. It would it would have been during the 19th century sometime. Isn’t in a sense you can attribute that to Kant by implication because Kant used as a criterion is whether criterions for what forms were structuring the mind what was what was common to the way different people perceived the world so that he he said one in one place in the Prolegomena they’re familiar with active he says there’s a passage in the effect the mark of objectivity is common a common universal agreement that a judgment is objective if it’s not idiosyncratic but would be agreed to by everyone therefore common agreement is objectivity. That’s not a verbatim quote like that something like that. So I mean that in effect is the idea of intersubjectivity as the definition of the objective, but I’m not sure which of the later philosophers who have followed him actually made that formulation.
Anyone else any new questioners? Yeah yeah computing in self-exclusion but he’s admitted himself when he said that particular yes if someone says straight-out I know nothing we have to ask how do you know or do you know that. Well perhaps we should pause here. I thank you.
Introduction
0:04
Welcome to the Principal Source audio presentation of The Nature of Free Will by David Kelley.
0:12
Free will is a fundamental question in philosophy because it is the foundation of ethics.
0:17
If man has no free choice in his actions then he does not need a moral code of how to choose his actions from a range of alternatives. In its broadest sense the idea of free will centers on the question
0:30
of what is the place of the mind in nature. In Part One of this talk Dr. Kelley focuses on three basic issues: a formulation of the free will theory as opposed to the determinist position,
0:44
Ayn Rand’s theory of free will, and the argument that the existence of free will would violate the law of causality. Part Two examines volition from a biological perspective
0:55
and broadly outlines why the development and expansion of the brain which led to man’s conceptual form of cognition could also have been expected to give rise to volition as well.
1:07
A well-known philosopher, teacher and writer, David Kelley earned his Ph.D. in philosophy from Princeton University in 1975. He has since taught philosophy, cognitive sciences and other courses at Vassar College and Brandeis University.
1:24
Dr. Kelley’s books include The Evidence of the Senses, The Art of Reasoning, and A Life of One’s Own: Individual Rights and the Welfare State. He was recently featured on John Stossel’s ABC TV news special Greed and is a popular lecturer and talk-show guest.
1:43
Dr. Kelley is founder and, since 1990, the executive director of the Institute for Objectivist Studies in Poughkeepsie, New York. This lecture was given in 1986 at the Portland Institute conference. Dr. Kelley presented a fascinating extension of Ayn Rand’s ideas on free will
2:03
to show that man’s volitional control over his consciousness is not only entirely consistent with the law of causality but also acts as a regulator or governing mechanism over the contents of his consciousness.
The Nature of Free Will
2:20
My topic for today and next time is the nature of free will. Now free will is a fundamental question in philosophy. It’s a question of crucial importance. If you drew up a list of the ten top questions in philosophy, free will would certainly be on it. It’s one of the questions that any system of philosophy would have to have a position on to count as a significant philosophical system.
2:50
And the chief reason for its importance is that it’s the foundation of ethics. If we have no free choice over our actions, if we have no ability to choose freely which action we will perform from among a range of alternatives, then we hardly need a code, a moral code, to tell us how to choose actions from a range of alternatives.
3:08
If we’re not morally responsible for our actions then we can hardly be blamed for them or praised for them in moral terms. If we have no choice about the goals that we pursue then we don’t need a code of values to tell us what goals to pursue. So ethics is the primary reason that the issue of free will is so important.
3:32
But today I’m going to focus on free will from a somewhat different perspective. Actually my two lectures together are going to take a somewhat different perspective on the question. For a number of years I’ve been interested in what in philosophy is called the mind-body problem. As you know, in Objectivism the mind-body dichotomy is an important part of Ayn Rand’s analysis for a number of philosophical trends that she opposes, such as Platonic idealism, the idea of Platonic love versus sex or of art versus entertainment, or theory versus practice. All of these represent a dichotomy between the mind and the body.
4:23
But in philosophy the mind-body problem has a somewhat narrower focus. It means it’s a question of what the relation is between the mind, consciousness, and the physical basis for consciousness, the brain in particular. And a little more broadly it’s the question of what is the place of the mind in nature: how is it that nature exhibits such a phenomenon as consciousness, thinking, the mind. That’s the context in which I’m going to approach the question of free will.
4:54
You know it’s often said that free will may seem plausible to us from the inside when we reflect, when we introspect on the way our minds work, but that we have to abandon the idea of free will when we look at ourselves from the outside as beings in nature subject to nature’s laws. This is the idea that free will is not compatible with a scientific orientation, that it’s a relic of a religious outlook, a product of wishful thinking, the kind of thing only a soft-headed person in the humanities could believe. That’s the idea that someday when we know more about how the brain works we’ll look back on the idea of free will the way we now look back on the idea that the stars move on crystalline spheres pushed along by angels.
5:36
I think this idea is completely false. The more I think about the biological role of consciousness and man’s form of consciousness in particular, and the more I learn about the way the brain functions, the more natural it seems to me that we have free choice and more clearly it seems to me free will fits perfectly into a coherent scientific view of nature. So that’s my focus in these two lectures.
6:00
Today, however, we have to start by laying out some fundamentals, and this is going to mean covering some ground that I think some of you are probably basically familiar with. I want to today, in today’s lecture, I want to cover three basic topics. First of all we need to formulate the issue precisely. We need to find out what it is exactly that determinists say and what it is that a free-will theory says. Secondly I want to go over Ayn Rand’s theory of free will which holds that the essential free-will choice is the choice to focus, the choice to think. Thirdly and finally I want to raise the problem of causality, the alleged contradiction between free will and causality, the idea that the existence of volition would violate the law of causality. I’ll state that in general and then that will be our platform for going into more detail next time.
Choice as a Neutral Term
7:06
So let me begin by formulating the issue, and I need to start with a few comments on the concept of choice. Now we all know in our lives that we face choice. We face alternatives. We do this all the time. We choose whether to go to a certain school or college versus another one. We choose which job to accept. We face decisions as to whether to leave a job. We face decisions in our personal relationships, whether to continue a relationship or to end it, whether to start one. We choose whether to go out on Saturday night or to stay home. This is obviously a real phenomenon. We all know if we look inside our minds that there are occasions when we say “I could do this or I could do that” and we have to, and we think about which one to do. No one could deny that that phenomenon occurs. Even a determinist who says that the outcome is preordained by a whole set of factors still has to recognize that phenomenon of deliberation and choice.
8:06
So today when I use the term “choice” I want to use it in a neutral sense, not as implying volition. Okay? And therefore it has something as a term that the determinist can use to refer to a real psychological phenomenon. When I want to refer specifically to a choice regarded as free I will talk about a free choice or about volition. Okay? I need to make this terminological point for reasons we’ll get to in a minute when I define determinism and free will.
Determinism
8:36
All right. What does determinism say? That at any given moment there is only one action an individual can take and that action is determined by the sum total of causes operating on the individual up to that point in time. That total, some total of causes may include his personality, his beliefs, values, physical condition, his genes, his parents, his education, his physical and social environment.
9:08
A determinist says that your mental processes, your cognitive processes, all function like perception. There may be a train of conscious thoughts going through your mind but the product is inexorably determined by causal factors in just the way that perception is and is therefore in a sense automatic. You have no choice. In the same way the determinist says that your actions are inexorably determined in the same way that a reflex is. Again, in an action there may be a sequence of considerations and feelings and desires that lie between some stimulus and some response, but the link between them is as fully necessitated as it is in the case of the knee-jerk reflex when the doctor strikes your knee. You know you are no more in control of what you feel, what you want, what you do than you are in control of your knee when it jerks.
10:07
So we can define determinism as the view that all actions, thoughts and choices are necessitated by antecedent factors. And that’s the definition that I’ve put down on the handout. I should make sure by the way that you all have a copy of this to follow along. Okay? If you don’t there must be copies in the back.
10:30
Determinism is the thesis that all actions, thoughts and choices are necessitated by antecedent factors, and these antecedent factors of course are themselves the effects of prior causes which are the effects of prior causes going back in time. Ultimately the whole sequence is outside of our control. That’s determinism.
Free Will
10:42
What about free will? Well a free-will theory holds the opposite position. If you study Aristotelian logic it’s the exact contradictory position. Free will says that some action or choice or thought is not necessitated by antecedent factors. Some actions, perhaps not all but at least some, are under direct volitional control. In regard to these actions we face genuine alternatives. At that particular moment there is more than one action possible to us in the next instant. Okay? The sum total of causes operating at a given moment leaves an individual with more than one alternative, and when someone makes a choice and performs an action we can say he could have done otherwise; he could have acted in some other way.
11:39
Now these are general definitions. They state the principles that all determinists on the one hand and all free-will theories on the other hand would agree about. But within each camp there are various different versions of each general thesis, and I’ll just, to give you an idea of the range of possibility here, I want to cover, say something about the varieties of determinist theory and the varieties of free-will theory.
Varieties of Determinism
12:17
Determinists agree that all actions are necessitated by antecedent factors, so an obvious next question is: which ones? Which factors? Which type of factor is the most important? And we can classify determinists by the way that they answer that question. So let me go over briefly three different categories of determinist viewpoints. I call them environmental, psychological and physiological determinism.
Environmental Determinism
12:36
Environmental determinism says that the primary determinants of action are causes in the external environment, particularly the social environment. This view says that man’s nature is essentially plastic; we are malleable; we have an innate nature that is susceptible to complete influence and domination by forces acting on us. Okay? Now the best-known example of this kind of determinism would be probably behaviorism, the psychological theory associated with B. F. Skinner. Skinner says that what we do is the result of the way we’ve been conditioned throughout the whole course of our lives. Throughout our lives we’ve been rewarded for some actions by parents, by our teachers, employers, friends, etc., and we’ve been punished for others; that is, we’ve acted and we’ve had a negative consequence. Our actions then reflect the sum of these external influences. We tend to act in a way that is consistent with a history of our conditioning.
13:48
Another influential version of environmental determinism is Marx’s view. Marx held, as you know, that the entire realm of personality, what he called the realm of consciousness, including ideas, values, feelings—all of that is a causal byproduct of the real underlying forces which are economic, such as the mode of production, the class system of a society, the distribution of property. That’s what moves history because that’s what moves people, and people’s individual thoughts are simply byproducts; they are epiphenomena, things like steam thrown off by the steam engine that does no further work. Marx often said consciousness is a social product. Okay? And in one form or another this view is very common to people on the political left. Virtually any, in fact from liberals on left, there’s almost always some tendrils of a theory of environmental determinism at work, as you can see most clearly if you look at their theories of crime.
Psychological Determinism
14:53
All right. Psychological determinism says that the primary determinants of action are internal psychological factors such as thoughts, feelings and desires. These come together in a certain way and we tend to act on the basis of our strongest desire about which we have no choice. It’s particularly clear in the case of Freud’s theory, which is perhaps the best-known version of psychological determinism, that it has to be deterministic because Freud holds that the conscious thoughts and feelings are products of some subconscious factors over which of course we have no control. As one expositor liked to put it, for Freud the conscious mind is a mouthpiece for the subconscious.
Physiological Determinism
15:36
Okay. And finally physiological determinism holds that the primary determinants of action are neural events in the brain. These are physical events: the firing of individual cells, circuits and feedback loops and everything else that goes on within clusters of cells, cell assemblies in the brain, the flow of electrical energy, electrical impulses through the brain and the electrical fields that are generated by the combined action of nerve cells. All of these physical events are what govern action. This view also holds that consciousness as such is an epiphenomenon, something thrown off perhaps by neural activity but not itself doing any further work, not giving us any kind of control over the stuff going on in our brain. And in fact some people who hold this view hold consciousness is an illusion; there just isn’t any such thing. I find it possible to take this view seriously since I don’t know how there could be an illusion if there’s no consciousness to be deceived about it. But nevertheless this is, I would say, the most common view among philosophers today in the area of philosophy of mind, at least in British and American philosophy.
Varieties of Free Will
16:51
Okay. Those are versions of determinism. What about free will? Well free will says that some actions, choices and thoughts are not necessitated by antecedent factors; they’re under direct volitional control. So the key question here is: which choice, which action, which thought is under direct volitional control? Okay? And we can classify theories of free will on the basis of how they answer that question.
17:22
Now in the extreme case, existentialism, at least in the form presented by Jean-Paul Sartre, says that we have total control. At any given moment we can choose our actions, our emotions, our personality, our values, our goals. In short, for Sartre the entire mental realm is volitional. There’s no causal relation between one moment and the next; there is no causal relation between one mental state and another. Now the basis for this view is an elaborate ethereal Sartre part which I’m not going to try to recapitulate, to the effect that consciousness in a sense cannot have an identity; it’s not like objects in nature; it’s radically divorced from anything that you find in nature; it does not have a nature and therefore cannot be subject to causal constraints. Needless to say this is not the Objectivist position.
18:18
Most free-will theories don’t try to exempt consciousness from causality. They acknowledge that at any given moment one does not have direct volitional control over all the contents and processes in one’s mind. The question is in what specific areas of mental functioning do we find volition. Objectivism belongs to a school of thought that locates volition in the area of attention, in the direction of the mind to specific contents. We are capable of self-awareness; we are capable of monitoring the contents of our minds, and we select the contents, the thoughts and the feelings that we attend to and the degree of attention that we pay to them. That’s the basic choice; that’s the realm over which we have direct volitional control. Okay? By means of those kinds of choices we have indirect control over our actions because our actions flow from the contents of our consciousness; our actions flow from the products of our thought. And in the long run we also have control over the basic premises and values that we accept and incorporate into our hierarchy of values, our character, our personality. Those too are the products of thinking but over a much longer time frame. So by making such choices, the choice to think, we do have indirect control over our values and feelings, our personality as well. But the primary choice is the choice to think. All of the other control that we have over our minds and our actions traces back to that one.
Ayn Rand’s Theory of Free Will
19:57
Let me give you Ayn Rand’s summary of her view from her essay “The Objectivist Ethics.” I quote: “Man’s particular distinction from all other living species is the fact that his consciousness is volitional. Reason is the faculty that perceives, identifies and integrates the material provided by man’s senses. It is a faculty that man has to exercise by choice. Thinking is not an automatic function. In any hour and issue of his life man is free to think or to evade that effort. Thinking requires a state of full, focused awareness. The act of focusing one’s consciousness is volitional. Man can focus his mind to a full, active, purposefully directed awareness of reality, or he can unfocus it and let himself drift in a semi-conscious daze.” End quote.
20:47
Now it’s clear that the central concept here is the concept of focusing one’s mind. The essence of free will is a choice to focus. So let’s start with the question: what does that mean? Let’s look a little more closely at this act of focusing, and I want to start with an analogy to a phenomenon at the perceptual level. Let’s call it perceptual attention.
Perceptual Attention Analogy
21:14
Suppose you’re alone at night in your home. You’re reading a book. You’re dimly aware of your surroundings, the furniture around you, the customary noises from the street, or if you live in the country as I do the crickets chirping and birds etc., the feel of the chair and the feel of the book in your hands, and you’re dimly aware of the visual layout of objects around you. But you’re only dimly aware of all that because you’re focused on the book; you’re absorbed in what you’re reading. And then you hear a scratching noise at the door.
21:51
Now if you’re anything like me this would have an effect on your perceptual awareness. Your attention would be drawn to the sound, but in addition and more importantly the level or intensity of your perceptual awareness would go up. Okay? You would not be dimly aware of that sound; you would not remain dimly aware of that sound. Your attention would be riveted on it. Your senses would go on red alert, sensitive to every nuance and aspect of that sound. Of course now your mind would be racing too; your mind would be racing thinking “What is it? Is it my dog? Is it the neighbor? Is it Anthony Perkins?” you know, like from Psycho. Since I’m making up the example and I’m believing in the benevolent universe, let’s say it’s your dog. But leave that aside. I’m not concerned with your conceptual functioning, your thinking. I’m just focused, concentrate on your attention here. It’s clear that something would happen to the character of your perceptual awareness.
22:49
Okay. If we think of the perceptual awareness as a kind of spotlight then two things have happened here. One, the spotlight has turned to a new item in your environment, but also the level or intensity of the light has gone up; you’ve turned up the wattage on the light. Okay? Awareness is always awareness of something, but that relationship comes in degrees. You can be more or less fully and clearly aware of it. Okay? Now the act of focusing one’s mind is analogous. Focusing one’s mind means raising the level of one’s awareness. Focus is to the conceptual level what the phenomenon I just described is to the perceptual level.
23:33
However there is a tremendous difference between the perceptual level and the conceptual level. The perceptual level is the ground floor of human awareness. It’s the given; it’s direct; it’s automatic; it’s effortless. So at the perceptual level it’s especially easy to see this variation that I’m talking about, and that’s why I used it as an analogy. But the conceptual level is not given; it’s not direct; it’s not automatic; it’s not effortless. Conceptual awareness, like perceptual, means the grasp of reality, but we have to achieve that grasp by means of a conscious process. Conceptual awareness like perceptual comes in degrees, and the act of focusing is the act of raising that degree of awareness. But to understand what that means we have to understand something about the nature of the conscious processes involved in conceptual awareness.
Ayn Rand’s Advance on Attention
24:25
Now in this respect, if I may digress for a moment, I think this is one of the signal advances of Ayn Rand’s theory of free will over any predecessor. As I indicated before she is not the first philosopher or thinker to suspect that free will had something to do with attention, but most other philosophers just left the matter there. Ayn Rand went on to identify the choice to devote attention and raise the degree of attention as the choice to think, as the choice to exercise one’s conceptual faculty. She then went on to integrate that very tightly with her ethics, in which the primary virtue is rationality, choosing to think, and all of her ethics then, all the virtues, are simply amplifications of that. And on the other hand in epistemology she didn’t just say the choice to focus is the choice to think; she went on and developed an entire epistemological theory about what it means to think, what the conceptual level is. And in a sense if I were to try to explain to you fully and completely what’s involved in the choice to think I would have to review the entire Objectivist epistemology from concept formation to induction to certainty and everything else, which of course I can’t do here today. I can try to summarize the point so that we understand what you’re choosing when you choose to focus by bringing the relevant aspects of this choice under two broad headings or categories: integration and objectivity. Integration and objectivity. And we’ll start with integration.
Integration
26:07
All thinking, all human conceptual functioning, involves the identification of relationships, ultimately relationships between perceptual concretes. In one form or another that’s what we’re doing when we think, and the identification of those relationships is what I mean in the broadest and most fundamental sense by the term integration. When you see an animal in a zoo and you recognize it as a horse, what you’re doing is relating that animal to a certain class of other animals that you have identified and subsumed under a concept. Of course to form that concept in the first place what you had to do was to bring together, notice that certain animals had certain similarities and that they differed in certain ways from other animals. Similarity and difference are relationships. You put them together on the basis of similarity to form the concept, and a new mental unit emerged in that process, namely the concept “horse.” So we are always, our thought is always either engaged in forming new mental units on the basis of relationships that we perceive in reality or it’s taking those mental units and assimilating new concretes to them, identifying new objects as instances of a concept or of a general principle.
27:31
Every operation of conceptual thought from classification to reasoning to explanation involves the awareness of relationships among objects in reality: similarity and difference, cause and effect, genus and species, premise and conclusion, etc. So in that respect being in focus means acting on the policy of integrating, acting on a commitment to integrate. It means not treating every object as a self-contained unit as something detached from any context but comparing and contrasting, noticing its similarities and differences with other things, identifying its causes, integrating it into a single unified framework of knowledge.
28:24
For example, suppose you’re in a history class in college and the professor is talking about how the industrialists of the late 19th century were robber barons. All right. Yesterday in your economics class the topic was how entrepreneurs exploit workers. The day before that the political science teacher was going on and on about how wealthy businessmen corrupt politicians and have legislation written to suit their purposes. Now you could treat each of these as an isolated instance, take each assertion at face value and leave it at that—and I’m afraid that’s how a great many students function—or you could integrate them. You could say, don’t these different views have something in common? Isn’t there some kind of common doctrine that I’m being indoctrinated by?
29:07
Or suppose I’m feeling an emotion. Okay? And I want to identify it. Now at a fairly low level of integration I might say my thought process might amount to something like this: I’m feeling kind of down; I’ve got like the blahs. That might be my first step toward identifying; it’s a step in the right direction but it doesn’t go very far. At a higher level of integration and precision I would have to ask: what exactly am I feeling? Is it sadness? Then what am I sad about? What specific thing? Or am I depressed in general, and if so what might have put me in that kind of mood? Or am I just tired? Right? So in that respect choosing to focus means taking an integrative approach to knowledge.
Choosing to Focus
29:50
A non-passive approach. It means taking charge, operating on the mental policy of identifying new objects and new concretes in conceptual terms, assimilating them into the context of knowledge that you already have, and expanding that context by being on the lookout for new patterns, wider integrations, finer distinctions. So the choice to focus then is first and I would say foremost the choice to integrate, to put two and two together.
Objectivity
30:22
Let’s turn now to objectivity, the second major heading. Perceptual awareness by its very nature is in contact with reality. It results from the interaction of external objects with our sense organs, and there’s no such thing as a false or invalid percept. There are degrees of awareness as in our example, but even dim awareness is still awareness. At the conceptual level, however, our conclusions are not produced in that way; they’re not automatically true or valid. They result from integrative processes that are subject to error. When we classify things we could, it’s possible for us to do it on the basis of non-essential similarities and end up with an invalid concept as a result. The history of philosophy is riddled with these things. When we draw conclusions from premises, when we weigh the evidence for and against a given statement, we can commit errors of logic. Any conclusion we reach, any piece of conceptual knowledge that we have, is only as good as the process by which it was reached, and that process is subject to error. We can make mistakes; we can overlook relevant information; we can be affected by bias or emotion.
31:34
To be in focus then it’s not enough that we commit ourselves to the policy of integrating; we have to take responsibility for the accuracy and objectivity of that process. We have to take responsibility for checking the steps by which we arrived at a conclusion, for making sure that we’ve taken account of all the relevant evidence and that we’ve excluded any irrelevant considerations that might otherwise have affected our judgment. So in this context focus means a commitment to objectivity. Raising the level of one’s conceptual awareness means raising the accuracy and the commitment to accuracy in your thought processes. Okay? Means proceeding at a commitment to proceed on the basis of the facts and only the facts.
32:17
Okay. Being out of focus by contrast would mean operating haphazardly, jumping to conclusions, letting emotions sway your judgment, ignoring facts that are unpleasant or threatening, failing to re-examine a conclusion in light of new evidence, adopting one explanation for something without considering any alternatives, looking only at one piece of the picture and ignoring the full context. All of these violate the principles of method that keep our conclusions in contact with reality; they undercut our grasp of reality. So choosing to think, to be aware of reality, means being objective.
33:00
So to summarize: the choice to focus or not is the choice to seek and maintain awareness of reality at the conceptual level, the level that’s appropriate to a human mind, and awareness of reality at that level requires a process of integration and a commitment to objectivity.
33:20
Now I want to emphasize that people differ in their intellectual skills, in their ability to integrate and to evaluate the process by which they reach conclusions. How far you can go in integrating depends on how much knowledge you have in a given area and what thinking skills you’ve acquired. In our example above the person who identified his feeling just as “well I’m feeling kind of down”—that might be as far as someone could get if he hadn’t developed his introspective skills any further. And same with objectivity: we’re not born knowing the laws of logic; we have to learn them; we have to make them automatic through practice. And here again people differ in their abilities. So this degree of skill is not subject to volitional choice, not at a given instant, although over the long term we choose how far we want to go. So what is subject to volition and what’s meant by the choice to focus is the choice to use what you’ve got. It means a commitment to the policy of integration and objectivity using the best approach and all the skills that you have.
Focus vs. Concentration
34:26
We should also recognize that focus comes in degrees. At one extreme would be a state of complete passivity. To take the emotion, in the case of the emotion, being completely out of focus would mean just being immersed in the feeling and drifting with it without any effort whatever to identify what’s going on.
34:44
And finally one last point: focus is not the same thing as concentration. This is an important distinction. Concentration is a state of attending to some specific mental content to the exclusion of everything else, as in solving a problem or reading a book or whatever it might be. Concentration implies a high level of focus, but you can be in focus without concentrating. You can be relaxing, taking a walk, doing any number of things. At the extreme I would say you can even daydream in focus. In fact that can be a very important part of the creative process, letting your subconscious run more or less free for a while, but you have to stay in focus to observe what comes out of it. Okay? This is pretty important because that’s how I write. I don’t mean I’m out of focus; I’m daydreaming all the time, but certain key steps occur in that context.
Evasion
35:40
All right. Now so far I’ve been talking about the level of focus as an overall state of mind regardless of what specific things we’re thinking about. I’ve been talking about the level or intensity level of intensity of one’s mental spotlight regardless of where that spotlight is directed. If we choose to operate at a lower level of focus the intensity goes down across the board no matter what we happen to be thinking about. But it’s also possible for us to go out of focus in regard to a specific content, to evade a specific fact or feeling or issue. Going out of focus in general means ceasing to put forth a certain effort to focus the mind, but evasion means actively pushing something out of mind, usually because it’s threatening in some way. The classic example outside Ayn Rand’s own works of this is Scarlett O’Hara who dealt with unpleasant facts by saying “Well I’ll think about that tomorrow,” and of course what she meant was, since tomorrow never came, what she meant was “I won’t think about it; it won’t be there if I don’t think about it.”
36:42
Now in order to evade something it has to be present to consciousness in some form, usually as a dim awareness which we push back into the darkness instead of bringing out into full awareness. For example I’m driving along in my car and I run out of gas. Okay? This is frustrating. I’d like to have someone to be angry at. My wife is a likely candidate, and so I start feeling “Why did you tell me the tank was empty? Why didn’t you fill the tank?” and I had the satisfaction of not being at fault myself for what was a stupid action, and it was very satisfying. But unless I am completely out of control the thought is going to occur to me at some point: wait a minute, why did you have to fill the tank? Why did you have to tell me about the gas gauge when I can look at it myself? I have visual contact with reality; it’s right there.
37:34
And at this point I have a choice. I can take that dim awareness, that nagging thought in the back of my mind of being unfair to her, and I can say “Well yeah that’s true; this is I’m being silly,” and the anger goes away because I focus on the injustice of what I was starting to do. Or I can just push that out and say “No I want to be angry now and this thought is getting in the way of that, so goodbye.”
38:02
Now sometimes evasion does not involve pushing something out of consciousness entirely but only keeping it at a sufficiently low level of awareness that we are not aware of its implications or its full meaning for what we’re doing. This happens for example when we evaluate people. Someone you like, for example, does something shady or hurtful or whatever and we can hardly ignore the act, but we can keep it at a sufficiently low level of focus that we think we just don’t see; we don’t think about what that might mean about this person or our relationship with them. Or conversely if there’s someone you don’t like who does something good, which can be extremely irritating, you might say “Well that wasn’t that good an insight” or “That was probably a bad motive behind it” or just let it hover there with some vague rationalization surrounding it so that you don’t have to appreciate its implications. Okay? And of course we often, especially when feelings are running high, can evade the need to be objective. We have a dim inkling somewhere in our mind that what we’re doing, we’re moving along, we’re jumping to conclusions, we’re engaged in shoddy epistemological practice and we better stop and check it out, but we say “Well no I like this conclusion when I got to it and I’ve got lots of evidence in it; I won’t worry about it.”
The Choice to Focus Governs Action
39:21
Okay. Now I hope it’s clear now at least in a broad way what we mean by the choice to focus, the choice to think. I’ve tried to isolate a particular specific mental action that we perform, extracting it from the flow of other actions that occur in our minds, from the flow of thoughts and feelings and desires, from the so-called stream of consciousness. There’s this one specific act which is focusing. But now having isolated it we need to put it back into its context and see how the choice to focus affects our psychological processes involved in acting. Okay? We want to see how the choice to focus gives us control over our actions when we have to make a decision in reality and act to take some existential action. And I think the best way to do this, because it’s an enormous subject, is to take an example, a sufficiently rich example that we can see the many, at least get a feel for some of the issues here. And the example I’ve chosen is from Ayn Rand’s novel The Fountainhead, the example of Roark’s decision to turn down the assignment to do the bank building, I believe it was called the Manhattan Bank Company building.
Roark’s Decision in The Fountainhead
40:35
As you remember in The Fountainhead, just to give you a little context, Roark—this is at a stage where Roark just opened his architectural office. He’d done I think one office building or department store and a few houses, but his business had really not started; in fact he was running out of money and was facing the need, the situation where he was going to have to close down his office. And at that point a director for this bank company who had seen one of his houses and liked it came to him and said he’d talk the board of directors into letting Roark submit a plan, which he does, and now he’s waiting to hear. He finally hears that the board has accepted his plan; they’ve chosen him as the architect for the building. But there’s one condition: he has to accept the classical motif on the building. He had designed it according to his own architectural standards including his standards of ornament, facades and so forth. They wanted a classical motif to appease the traditionalists, and Roark tries to explain to them why that would destroy the integrity of the building, why it would ruin it, but the board’s decision was final, and Roark has to take it or leave it.
41:47
And now from the novel you understand the situation. “Mr. Roark,” says the chairman of the bank. “Yes,” said Roark. His eyes were lowered; he was looking down at the drawings. Well Roark did not answer. “Yes or no, Mr. Roark?” Roark’s head leaned back; he closed his eyes. “No,” said Roark.
42:16
Now this is a portrait of a man making a decision, a decision in reality. He’s clearly faced with this. What’s going on inside his mind is an example of what I described at the outset of choice. He’s thinking about the alternatives. We see him from the outside but let’s try to read between the lines and think about what’s going through his mind. He’s got a clear-cut decision to make and his choice, whichever one he makes, is going to be based on his values and on his judgment about what will best serve those values.
Motives
43:09
On the one hand he has a very powerful motive for accepting the commission. This is a bigger project than anything he’s done and it’s going to advance his career. If he doesn’t, he’s at the end of his money, so turning down this commission is going to mean closing his office and leaving architecture for he doesn’t know how long. He’s passionately committed to his work. Saying no is going to mean abandoning for he doesn’t know how long the thing that he loves most.
43:15
On the other hand his belief in the integrity of a building’s design and appearance is absolutely essential to his love for architecture. He builds in order to bring that value into existence. Putting up a flawed, compromised building would not serve that purpose. So he also has a powerful motive for turning down the commission. So either way he’s got a motive: a motive for accepting, a motive for rejecting the bank’s deal. If he accepts it that action will be caused by one set of values and beliefs—okay?—the value of his career to him and the recognition that this would be a step forward. If he rejects it that action too will be caused by another set of values and beliefs: the value of integrity and the recognition that the building is flawed.
43:55
Moreover, just to enlarge the picture, those motives that are there present at the moment in Roark’s mind—those motives are the product of his character, his values, his knowledge, his principles. Okay? At the moment we see him making the choice these are not subject to his direct volition; they are products of choices he’s made over the entire course of his life. The feelings that are passing through his mind, his desire to get the commission, his anguish at the thought of losing it and even worse at the thought of leaving architecture, his disgust for this compromise design, the disappointment and contempt and anger that he must feel toward the men who want him to produce this thing, this monstrosity—these emotions would be automatic responses based on his values.
44:36
So where does free will occur in this event, in this decision? Well it occurs in his choice as to which set of causes will govern his action. He’s got two there; his action could potentially be caused by two different sets of motives. The choice is in effect which ones are going to govern his action, and he makes that choice by exercising his capacity to focus. Because he chose to stay in focus he realized that putting up a building without integrity would not in fact satisfy his desire to build; it would not; it would represent an inversion of his hierarchy of values. Because what’s at issue here, what’s the issue is whether he’s going to be able to get the commission, do the plans, oversee the construction, make some money—all the things that are involved in putting up a building. But those are only means to his fundamental goal which is to embody an architectural vision in reality. Accepting the commission would be elevating the means over the end, and by choosing to stay in focus he realizes that. He must also realize that whatever advantage it would give his career in the short run to do this building, whatever prestige he might earn, it would not be a step forward in the long run. The building would stand as a constant reproach to himself and a constant reminder to others and warning that Roark does not stand by the principles that he espouses. So in the long run it would not help his career. But that too is a fact that he acts on only because he chooses to keep it in focus.
46:00
Because he focuses on these facts the motive that would otherwise cause him to accept the commission is renounced; it’s disconnected; it has no causal efficacy over his action. It’s detached as if he had shifted the gear lever and those particular motives can no longer drive his wheels. His action is governed instead by the considerations that he has chosen to keep in focus. But he could have evaded those considerations. He could have put the long run out of his mind and dwelled only on the short run. He could have evaded his principles of integrity and his hierarchy of values and let his mind be dominated solely by the anguish at the thought of losing the project. Or he could have recognized that this would compromise his integrity but he could have rationalized it away. He could have thought “Well the building won’t be perfect but it’ll be pretty good and next time I can do it my way.”
46:51
Now the same basic pattern is involved in any voluntary choice, voluntary action. There is always some motivation behind an action, and with hindsight looking back we can trace the action back to some specific set of some specific motivation which would include a set of feelings and desires and beliefs, and we can trace those in turn back to a person’s values and choices he’s made through the whole course of his life. So in general it’s valid to say that actions are caused by the contents of our mind. But the choice to focus gives us control over the contents of our mind by choosing the degree to which we seek conceptual-level awareness of reality, by choosing whether to identify or evade specific facts. We choose the particular contents that are going to govern our actions.
Validation of Free Will
47:40
Okay. So I’m running a little behind and I don’t want to elaborate any further on the issue of how the choice to focus governs action. We can turn more to that in the question period if you like. I want to turn now to the validation of the free-will theory. I’ve described what I’ve been describing so far simply a description. I’m giving you a description of what occurs in the mind. But a determinist could accept everything I’ve said so far and simply say “Well the choice to focus is necessitated by antecedent factors.” Okay? So we need to ask what’s wrong with that position. How do I know that the determinist is wrong?
48:21
Well I’m going to proceed very quickly now, and in a sense I feel entitled to it because he is, I think he is self-evidently wrong. The basis for knowing that we are, that we have free will, that we are volitional beings, is our own capacity to observe volition in our minds. The existence of volition can be established by direct observation, observation of the way our minds work. And once we’ve isolated the action of focusing the mind we can see that the choice to focus is up to us. We can observe that we are in control. We observe that focusing takes a certain effort. It’s possible for us to exert that effort and it’s possible for us to withhold the effort. Either way it won’t just happen. If I’m out of focus just drifting along out of control you can yell at me all you want; I can yell at myself, you know “Come on, wake up, wake up and smell the coffee, focus, get a grip on yourself.” But I can also, no matter how much I do that, I can tune out that little voice in my mind or I can tune you out even more easily and choose not to take account of the considerations that make it reasonable for me to choose to focus in that context.
49:35
A determinist would say that the choice is necessitated by antecedent factors, but when I’m in focus they say there’s a motive: I want to understand something and I realize that I have to be in focus in order to understand. But that motive is itself something that I could evade. The desire to understand something will lead me to focus only if I hold in mind and focus on the fact that understanding requires focus, requires integration and objectivity. I could evade that fact. I could try to have knowledge without causes, knowledge without effort. A large number of people have spent a large part of their lives trying just that kind of—there are all kinds of get-smart schemes analogous to the get-rich schemes that flourish in the same way.
50:22
A determinist would say that when I go out of focus or evade there’s also a motive: I resent the effort it would take to focus or some fact is too unpleasant for me to deal with. But in that case I’m going to be aware however dimly that I’m not going to get anywhere without exerting the effort, or I’m going to be aware dimly that the facts won’t go away just because I push it out of my mind. I can choose whether to make that dim awareness more fully real or to suppress it and put the awareness out of mind. So I believe it is self-evident that the choice to focus is a free one, and this fact is so fundamental that it is inescapable. Even a determinist who tries to deny it will implicitly rely on this fact. So it’s not only self-evident; it’s fundamental to an extent that it is inescapable.
The Determinist’s Contradiction
51:14
Why? Well a determinist who denies the existence of free choice is offering his position as something he knows to be true. By his own theory, however, it must also be something he was causally determined to believe. To be consistent he must regard his own belief in determinism as something necessitated by antecedent factors. And there’s a contradiction here. If he regards himself as knowing that determinism is true he must believe that he arrived at his position on the basis of evidence and proof and not on the basis of some irrational motive such as the desire to escape responsibility for his actions. He has to admit that such irrational functioning is possible to him; he’s not infallible. But his own theory says it was not up to him whether he would proceed on a rational or an irrational basis. He had no choice as to whether he would be objective in his thinking about the issue of determinism. In short he is saying that his belief in determinism is based on the evidence—that’s how he knows it—but he’s also saying that he has no control over whether any of his beliefs are based on evidence.
52:18
Moreover a determinist is trying to persuade us to accept his position. He thinks that anyone who believes in free will is mistaken, that free will is a soft-headed piece of wishful thinking which we should grow up and get over. He wants to persuade us to stop indulging this kind of wishful thinking, to wake up and focus on the facts that make determinism true. But all that presupposes that we can choose to be objective, that objectivity is up to us. If we have no choice in the matter then it would be irrelevant for the determinist to give us arguments, to offer evidence for his position, since neither he nor his opponents have any free choice as to whether they’re going to proceed on the basis of evidence or on the basis of subjective emotion or desire or the conventional beliefs of their society or whatever. So to weigh the evidence for or against any position we must do it in accordance with standards of logic and proof, and the use of standards in epistemology presupposes choice, free choice, in just the same way and for the same reason that the use of standards in ethics presupposes free choice.
53:23
The existence of volition is therefore an axiom, or it is axiomatic. It lies at the base of any proof, any conceptual-level knowledge. So on the one hand we can’t prove it because any proof would presuppose it, but on the other hand it doesn’t need proof; it’s self-evident and inescapable.
Free Will and Causality
53:45
Okay. Now I’m going to try to leave some time for questions because I know people have a lot of questions about free will. I’m sure some of you will have questions about the argument that I just gave. But I want to go on and at least introduce the final topic for today which is free will and causality, and I will try to keep this to five or ten minutes.
54:10
I would say that the chief objection to free will, the chief philosophical reason why people deny free will—there are many psychological reasons—but the chief philosophical reason is the view that volition would violate the law of causality. The law of causality says that an entity’s actions are determined by its nature. Actions do not occur at random. We don’t live in a world of flux where things just happen without cause, unpredictably. There is no slippage or free play in nature. Right? Things are governed by causes. The action of an entity is produced by its nature and the way it responds given that nature to external circumstances. And the same cause must have the same effect. Two identical things in identical circumstances must act in the same way.
54:56
Now determinists argue that volition would violate that principle. When a person acts—and this includes the act of focusing his mind—he is acting in some particular set of circumstances and he has a particular nature or identity. Both the circumstances and his nature are antecedent factors; they don’t spring into being at the moment of choice. If the choice is not necessitated by antecedent factors then how can we say it has a cause? And if the act of focusing does not have a cause then it would be random and unpredictable; it would be something that just happens, in which case we are not in control of it. Okay? So there’s a dilemma here. If the act of focusing is necessitated by antecedent factors then we’re not in control of it. If it is not necessitated by antecedent factors then it doesn’t have a cause and we’re not in control of it either; it just happens. I just sometimes I’m in focus and sometimes I’m not but I’m not in any control over it. So either way we are not in control of the act of focusing.
56:03
Now this is an objection, this is an argument that we need to be able to answer. We can’t dismiss the law of causality. The law of causality is itself axiomatic; it has the same level of fundamentality in human knowledge as the existence of free will. So we’re faced with a situation where there’s an apparent conflict between two fundamental bases of our knowledge. That’s not acceptable. If a thing… so we have to see what’s wrong in the determinist assumption that for an action to have a cause it must be necessitated by antecedent factors. In other words we have to see what’s wrong in the way determinists apply the law of causality to the example or the specific concrete of the act of focusing.
Hume versus Aristotle
56:51
And in this respect the point I want to conclude on today is to deal with what I think is one essential part of the problem. In fact it’s really the essence of it. And this is the item on the handout: Hume versus Aristotle. It is a conflict between, it’s a question of choosing the appropriate view of causality, about the nature of causality.
57:25
Part of the reason for the determinist argument that I summarized before is the view that causality is a relationship between events. Okay? An event at one moment causes an event at the next moment. This view derives from David Hume, although he had predecessors in it, but he’s the one who emphasized it, and it’s held by many scientists and philosophers today. It’s certainly the most common view. This is sometimes called the billiard-ball theory of causality because the classical example is billiards: you strike the cue ball, cue ball rolls and hits the object ball, the object ball rolls off—a sequence of events each causing the next.
The Billiard-Ball Theory
58:05
On this theory, by definition the only way for an action to have a cause is for it to be caused by prior events because that’s what causes are: events. So the only way for an action to have a cause would be for it to be necessitated by antecedent factors, and determinism would have to be true.
58:19
Now Objectivism rejects that view of causality in favor of a view that’s associated with Aristotle. Not everything I’m going to say is something Aristotle himself said, but this is in the Aristotelian tradition and spirit. On this view events per se are not causes. An event is an action of an entity or the combined actions of a set of entities. So an event is never a cause; it’s always an effect. Events, actions, are not their own causes and they don’t cause other actions. Entities cause actions because it’s entities that act. The cause of actions is always the entity or the entities which act. So when you focus your mind the act certainly does have a cause regardless of whether it’s voluntary or not. The cause is the entity which acts, namely you. You are the agent; you are the entity acting in that context. The act of focus is your act; you make it happen. It’s not a free-floating occurrence. So it certainly has a cause.
59:27
But there’s a second and equally important point. Hume’s view of causality tends to put all the emphasis on the external circumstances in which an entity acts, the forces that are acting on the object, the way it is affected by other objects, the way it responds to its environment. Hume and his followers in other words tend to treat all causality as being of the billiard-ball type. And I should say that even in the case of the billiard balls the Aristotelian theory says it’s the billiard balls that are causing the action in accordance with their nature. But Aristotle noted that external factors, the forces acting on an object, are only one factor involved in causing it to act the way it does. Okay? And they’re not the most important factor. Since an entity is the cause of its actions, the most important factor is the nature of the entity itself, the identity of the thing which acts. This might include its matter, what Aristotle called the material cause; it includes the form in which that matter is organized, the internal structure, a set of relations that the components exhibit when they form that structure; it includes the properties and the potentialities of the object; it includes its entire identity. It is that identity in conjunction with the external circumstances that explains the way the entity acts. The entity acts in accordance with its nature, and its nature determines the way it will respond to external circumstances. It will determine which external influences can affect the object and which not.
1:01:08
Okay. For example, the price of eggs in Providence is an external factor governing, surrounding, present in the context of a hen laying an egg in Providence, but the hen is not responding to the price of eggs in Providence. It is not saying “Well demand’s up, I’m going to take advantage of this.” This is not something you can just respond to because the nature doesn’t permit it. Okay?
1:01:33
What this view allows is a much more sophisticated view of causality than the billiard-ball type. The Aristotelian view which says that causality is a matter of the nature or identity of the objects which act allows Aristotle or an Aristotelian to recognize that there are maybe many different modes of causality in nature, that there may be many different types of specific relationships between the nature of an entity and the external circumstances that it acts in. Okay? And in fact I will argue now in brief and elaborate in more detail next time I think this is exactly what we observe in nature.
1:02:16
To anticipate a little bit: inanimate objects like billiard balls respond in an essentially passive way to the forces acting on them. Living organisms such as plants are goal-directed; the source of their actions comes from within in a sense that I’ll try to define next time. Animals going a step further: animals act on the basis of conscious awareness and desires; they exhibit psychological causation, and one aspect of this is that the locus of control over the act has moved still further inward. And man finally has the ability to reason, to introspect, to focus on his awareness. It’s natural to see that as the culmination of this trend toward increasingly internal control over action. And if our nature gives us complete control, if our nature makes it possible for us to focus freely by exercising the capacity to focus, that would not violate the law of causality. In exercising the choice freely we would be acting completely in accordance with our nature. Volition then would not violate the law of causality; it would be an instance of the law of causality; it would be an instance of an entity acting in accordance with its nature.
1:03:36
Now I believe that what I have just given you is a philosophical answer to the problem of causality or the objection that free will would violate causality. And as a philosophical answer I think it’s adequate. But next time I want to elaborate more fully on this. I want to take the little sketch that I just gave you, the view of nature and of the increasingly internal control exhibited by objects as we move along that scale—this view of nature that Aristotle’s theory of causality makes possible. Next time I want to try to make it actual by presenting it in some detail. But for today we’ll pause here. Thank you.
Question Period
1:04:30
Hi. Thank you. And let’s open the floor to questions now. We have about 25 minutes.
Libet Experiments
1:04:38
Yes. I have two problems in this area. I don’t know if you’re really talented but I’ll let you decide. The first one: I read reports not too long ago that I would see it in the case an individual chooses generally freely to do a certain action that there is actually some physiological manifestation of the choice before he actually becomes aware of it, as you say.
1:05:15
I believe that what you’re referring to is a set of experiments done by a neurophysiologist at San Francisco State called Benjamin Libet who found—he studied—he had people doing certain voluntary actions and he was measuring the electrical potential on their, from recording up from the scalp and found a certain specific potential connected with action which occurred prior to their conscious awareness of the urge to act. Does that sound like what you’re—yeah that’s the only real clear physiological evidence of that kind that I know about.
1:05:51
All right. Look, I had a feeling this question might come up because I read Science News and this article was in Science; I think it was even in the New York Times. Let me just briefly explain what that experiment was because you need to see the full context.
1:06:09
What Libet did was to have people perform an action, clenching their fists, and he told them to do it without any planning. He says don’t plan it; they just sit in a room and whenever they have the whim to close their hand he says do it. Don’t plan it; don’t try to do it any specific time; just but when it happens, when you experience the urge to close your hand, make a note of when that happens. And he allowed them to—he said notice where the position of the hand on a clock is. Okay? So he could tell when in objective time they were first aware of the impulse. Okay? And what he found also he was recording from their scalps, recording something called a readiness potential which is a physiological trend in a—I have to draw it. There’s a—if you record from the scalp what you get is something like this because what you’re getting is a record of electrical activity being summed over the entire or at least over a large area of the brain, and it’s regularly been found—I take the scientists’ word for this—that there is something here which I can never see when I look at their little graphs but a little dip which precedes—if the action occurs at this point in time it occurs about half a second before the action. Okay? And what Libet found was that the people’s awareness of their impulse occurred two-tenths of a second, only about two-tenths of a second before the act occurred. So this readiness potential which they think comes from the motor cortex and is therefore a source of the act preceded the conscious awareness of any impulse to act, and therefore the conclusion is the action was already on its way physiologically to being performed before the person had any clue that he was doing it. So how could he have conscious control?
1:08:12
Now let me say first of all that Libet himself doesn’t draw the conclusion that we don’t have free will. His conclusion was free will consists in the ability to have veto power over the act. In other words the act starts in motion but you can stop it. And he went on to do some further experiments, but I don’t know how people perform this act of introspection because we’re talking about milliseconds, tenths—we’re talking about tenths of a second here prior to an act. He says when the impulse starts stop it. Okay? And he found same readiness potential. Okay? Awareness three, three would be three-tenths of a second later but then the act didn’t occur. Okay? So Libet himself thinks it’s free will but he thinks it’s only veto power over impulses that are generated by the subconscious.
Audience Comments on Libet
1:08:59
Okay. Now let me before I give you my answer let me ask what does anyone have any objection to the character of this experiment. Yes. I think that that reading that starts just before the guy wants to clench his fist, his part is already following his decision to do that, whether how conscious that decision was because it’s an easy action to do. I think that’s part of the whole action, not something that precedes the man wanting to take the action but it’s part of the decision to take the action. But you said the decision was not conscious and so Libet could say okay call it a decision if you want but what that means is that you’re not consciously the source of the act of conscience. It is conscious but not very—you’re not focused on it very sharply because it’s such a simple action, less a degree of consciousness. Okay. I will certainly accept the point about the simplicity of the action and the character of the act being important here.
1:10:03
Let me go and see what other answers people come up with. Yes. The experiment requires a decision to note a time and that decision can only occur after the original decisions, i.e. two-tenths of a second later. You pay attention to the time but you only do that after you’ve moved your hand. But they were—and there’s even a timing involved between the eyes and the brain which you’ll lose some time in.
1:10:40
There are some questions that have been raised about the validity of this measurement technique, but what they do is you know there’s a hand, a big hand you know like a child’s clock moving along and you note the time when it is to—you remember where the hand was when you make the choice to or feel the experience the impulse to act, your first conscious awareness of any impulse to act. I don’t think they correct it for any time lag between the transmission from the eye to the brain. I believe they corrected for that; that some effect remains after you correct for that.
1:11:28
Okay. Let me get one more. And yes. Do you not make the choice when you’re told to cut your hand? You’ve made a choice in a sense to program your subconscious that you’re going to do that at some point. So the original choices made before you even got into the room and then when you do it you’re sort of previously aware that you’re going to.
1:12:00
Okay. I would say that in combination with the other comments that that’s exactly right. This act is something you’re told to do. You’re not told to consciously plan when you’re going to do it but you’re told to set your subconscious to clench your hand. We do not go around clenching our hands like this unless we have some problem. You have to, you do have to give your subconscious an order, generate some whims, generate some impulses. Okay? And then it would make perfect sense to me that you would not become aware of it until the action is already underway. This is in this context taking an act that was consciously that was specifically intended to have no specific meaning for the person, to be completely arbitrary random act, would not require any focus. It’s something that your subconscious can do without much trouble. This particular, this—it would not—the whole context, the whole situation is one which because of the nature of the act involved wouldn’t bear particularly on the choice to focus.
1:13:07
Ed, I haven’t read this but I read the same thing you did. I’m willing to bet you it’s an artifact measurement, the timing, because you look at introspectively in a situation like that how are you going to decide what to do, what you’re going to say to yourself now. Okay? But you have the duration of the percept thing to clock; you have the duration of the thought; you have the duration of reporting the thought as somebody said; and how can he measure, and the decision might not take place instantaneously. How can he measure all the processes properly and just a little error in any of those could completely throw off the data. I know I’m willing to bet you to conscious and he just measured—he didn’t measure all the steps involved just preventing that. If somebody redoes it right it was subconscious it would be more like it seems to be like when you see the red light do it and you program it without even thinking then it might be, but in a case like this you’d have to say to yourself now and I bet you if you measure the lifetime that that takes and the duration of the first half to see the clock that he didn’t include all of those things in the measurement. I’m not even sure.
1:14:20
That’s certainly possible and in that case then the experiment would have zero significance. I read his summary in a not the original research papers but the interdisciplinary journal kind of summary of it and I thought he had some way of correcting or taking account at least of the perceptual transmission process. Recording itself is a lag much less deciding than to act on that perception and how can he measure the duration of all of those accurately or did—that’s a good question. And that I couldn’t really give you an answer on that.
Same Cause, Same Effect
1:15:05
Okay. Yes. Mike, your talk when you introduce the term causality one of the terms, one of the things you said about it was that when an entity the same entity in the same circumstance would produce the same action. Presumably then that doesn’t—is that that truth because out in general was that an example or a force of causality? I presume that doesn’t apply to human beings then.
1:15:22
I believe it does apply to human beings, but you have to understand properly what it means to have the same cause in that context. That’s something I’m going to talk about next time. Next time I’ll try to show you that it is true that if you had the same cause, same person in exactly the same circumstances in all relevant respects then you would get the same result, but that is not a violation of volition. Okay? So that’ll be at the end of my talk next time.
Focus and Simple Actions
1:15:54
Okay. Yes. This experiment if I did not focus on the object of experiment I would never clench my fist.
1:16:04
Well right but remember what these subjects are being asked to do. They would have to be in a high state of focus I think. I don’t know if I could do it. They have to stay in touch with what with this impulse so that they’re—but they’re focusing as observers on their own introspective state. Okay? It’s not like they’re focusing on some reality-life decision “Shall I clench or shall I not” and what reasons are relevant and which ones shall I focus on and how much attention do I need to pay to all these factors. They were intended, they were instructed to just let it happen. Now maybe as Dr. Locke says there’s no way to make it a fully automatic process but the focus they would exert I would think would be mainly directed at trying to see where the clock, the hand of the clock is at the very moment that they’re feeling this impulse.
1:17:10
Yes. Did they attempt at all to try and find out if there was any delay between the electrical impulse and if someone was trying to consciously do something? You know when does the readiness potential occur in volitional action? That’s a good question. I think they found that it occurred somewhat earlier in that case. Right. No, in the case where the individual was trying to plan to do it at some specific instant it would occur on the order of one second before the action. But I mean that’s understandable. You would be watching the clock now and you have to exert control in that sense.
Do We Choose to Choose?
1:18:09
Yes. I think you had your—yeah I’m trying to understand here. Am I correct understanding that we do not choose to choose and yet this is my big problem with free will and yet trying to avoid that seemingly infinite regress that we don’t choose to focus and I may be misunderstanding you know the situation here but that’s my problem.
1:18:27
Do you make sure that I understand the problem? Yeah trying to choose to focus like what sets in your mind how do you sort of initiate the process or whatever.
1:18:40
The things I can say about what it means to choose to focus, but precisely because it is an act that is under our fundamental control I can’t give you an answer in terms of well in order to focus you do go through procedures A, B and C because then procedures A, B and C would be the ones that you’re in direct volitional control of. I mean volition is a primary, a causal primary as well as an epistemological primary. It’s not something that you can break down further into sub-causes, some sequence of operations that you perform in the course of doing it. You just do it. So the only way to explain, to answer your question I think would be to give you a fuller description perhaps of what it is that you’re choosing if you choose to focus or what is your choosing if you choose not to focus. But the choice itself I don’t think can be any more fully elaborated.
1:19:33
I think my problem originated once when I heard Harry Binswanger just trying to elaborate on free will and he suggested that he gave the example of when you say wake up in the morning or you’re asleep and you wake up and you’re in a subconscious state while you’re asleep and you sort of wake up and sort of shake your head and he sort of gave that as an example of in a sense you know using your free will or sort of waking it up or turning it on or whatever and that’s where the problem came from is where does this come from sort of turning on your consciousness or whatever. I don’t know.
1:20:12
Well that would be one context in which—it’s kind of a specialized context but usually the choice to focus you’re not you know in a completely unconscious state at the outset. And if you are certain physiological activity has to occur before you’re in a position to choose to exert choice. I mean you have to be conscious for one thing. And so in that context the choice to focus would be something that happens at the last stage where you say you know you’re conscious and you’re just kind of you know rolling around and you decide to take control as opposed to just you know drifting and dazing. But it’s no matter what context it is it’s a choice that you exercise directly. You know if you ask can you choose to choose to focus, well in a sense no that choice is imposed on you by the nature of your brain. You are at every moment in a position of choosing what level of mental effort you’re going to exert. Okay?
Readiness Potential and Training
1:21:14
Yes. Seems to me that that could be similar to a validation of free will. It could certainly fit in those experimental data if you had something like a race-car driver. Well he in fact would want to shorten that gap between almost to make it a reflex. You want to shorten the veto period of his consciously you know if he sees something coming up he wants to shorten the time between this the dip you called it the readiness reflex. You would want to shorten that time so but that would still be a result of his prior to this considering this thing.
1:21:52
Well I don’t know. I’m not a neuroscientist. I don’t know whether that is subject to variation or whether it’s part of some sort of fixed reaction-time phenomenon or what. I don’t know whether the readiness potential varies. Perhaps I don’t know if anyone here knows, may know this.
Volition as Causal Primary
1:22:14
It’s almost like asking why existence exists to ask for antecedent causes to volition.
1:22:36
Yes and no. I think you’re insofar as volition is a causal primary that is not further reducible to some sequence of operations that’s true. It is and as with other causal primaries you can’t ask all right what’s the explanation of this. But in another sense it is valid so we’re talking but that there we’re talking about the actual choice as made by you at a given instant. I do think it is possible and this is one respect in which volition differs from something like existence: it is possible to explain where the capacity comes from, oh something has given rise to it. Right, right. And that’s the material that I’m going to try to cover next time. But volition is a specific capacitive phenomenon in nature and it arose somehow through the course of evolution and so in that sense the capacity will have a cause and I assume will be explainable. I’m even going to try to offer at least the broad outline of an explanation. But you’re right that once we have the capacity whether we exercise it or not is primary and not subject to explanation.
Meditation and Quiescence
1:23:53
Yes. Choice to focus as a choice to think. Before I become while—
1:24:16
Well that’s a good, that’s an interesting question. You know the phenomenon—it’s part of many Eastern religions of putting the mind into a state of quiescence of some kind by chanting a mantra or focusing on a visual, an imaginary visual object and trying to exclude all of the thoughts and sort of damp down the mind. This raises a question. It’s very much like the question of how do we fall asleep because in fact the brain state they achieve is very much like sleep I would imagine. And this is a speculation that there is that in a sense they do begin by exercising a certain kind of focus of not of trying to integrate in this case by trying to focus and concentrate the mind in a contemplative way on just a single object. But I mean what the mind does is it thinks; it integrates; it goes through a conscious process. So if you use the capacity to focus to prevent the mind from doing what it’s designed to do—and I think this is the intent—it shuts down the mind and you—it’s not surprising to me that that can produce a state of passivity. And in fact you know when sometimes one way to get to sleep to try to you know calm down the what the TM people call the chatter in your mind, you know to slow down your racing thoughts is to concentrate on you know sensations of your in your muscles or whatever and sometimes that—but the product there is not—I mean it’s a kind of a paradoxical case because you focus in order to put you out of the need of having to focus; I put you asleep. So I—
Subconscious Programming
1:26:06
Yes in the back. As far as this guy’s experiment goes and it seems like what he’s doing is a kind of post-hypnotic suggestion that these things are emanating from the person, some of these choices to clench or emanating from the person’s subconscious, and if the signal comes out of the person’s subconscious and goes to their muscles to clench, their conscious awareness of that impulse has to come after the subconscious decision. So it seems to me as a result perfectly reasonable but it doesn’t have really anything to do with free will because the actual choice to program your subconscious this way has come many many cycles previous to the actual clenching and the blip that he should be looking for is really as he’s explaining the experiment.
1:26:51
Yeah I agree. I agree. I think that’s a very good statement of the hypothesis that I would adopt. It’s a little different from Dr. Locke’s and I wouldn’t begin to say that this experiment tells us which of the two are true but yeah I think that that’s probably is what happening is what is happening.
Choosing Content vs. Level of Focus
1:27:15
Okay. Is anyone else before I call him Michael? Have I called on you yet? Yes. Well all right anyone else. I want to give everyone a chance. All right you’re on. You have them. All right. Well we can choose to focus, increase the level of our awareness, but to what extent can we choose the content of that focus? Are there two choices now? It’s just one and then we sort of get the content given to us by the subconscious.
1:27:38
Well I would say that primarily we choose the level of awareness and the contents are supplied by whatever purposes we have, whatever standing orders we’ve given to our subconscious in effect. For example you are now in a state of focus. I’m sure if I say who’s the first president of the U.S., okay a certain mental content will swim into your minds, right, George Washington. But—and that that’s right isn’t it? Yeah look at this puzzle looks. I would say in that case it’s a clear example of you—I’ve in that case I’ve given a standing order to your subconscious and I’ve gotten an answer. But within the realm of conscious awareness there are going to be things that are central and things that are peripheral, things you’re dimly aware of, things in the back of your mind, on the tip of your tongue, that this phenomenon, and within that you choose focus. That’s probably choosing where which things, which contents that you there are in some sense in your consciousness to bring into the forefront. And so in that sense it is a selective process as well as an increasing intensity or level of awareness.
Concept-Stealing
1:28:47
Yeah. Can you re-explain why it’s not concept-stealing on the part of the proponent of free will when he makes free-will arguments and the determinist can just say that those arguments are determined, the free-will arguments themselves? Why that isn’t concept-stealing on the part of the free-will—
1:29:10
I think it is concept-stealing. Yeah. He’s stealing the concept of logic, evidence, standards from the fact that is the basis for the possibility of those concepts which is volition. So you could say that the problem is he’s stealing the concept. It’s like a determinist who had a passionate moral view or elaborated a code of ethics. He’d say you’re stealing the concept of ethics from one of its preconditions. So I would say it is concept-stealing and that’s a good identification.
1:29:42
Okay. Thank you.
