a) A Correct Analysis of the Cartesian
Quandary

To study Reid after having read Descartes, Locke, Berkeley, and Hume, is to step out of Plato’s cave into a bright spring morning. On page after page, Reid illuminates the origins of the Cartesian predicament and castigates the follies of its idealistic consequences. He clearly sees that it is the principle of immanence that pushes the Cartesian into skepticism. And he never tires of ridiculing the idealistic claim that we perceive nothing but our own ideas (sensations, impressions). What we perceive are the ordinary objects around us: trees and flowers, houses and mountains, people and animals, clouds and rainbows. No philosophical argument, no matter how sophisticated or how convincing it may appear at first, can possibly be sound, if it denies this truism. This is the spirit of Reid’s philosophy. It is, in the words which he uses to describe the Cartesians, “a more important acquisition to mankind than any of its particular tenets” (T. Reid, Essays on the Intellectual Powers of Man, p. 150), and for exercising this spirit so zealously, Reid deserves our grateful attention.

But what Reid observes about Descartes, namely, that he rejected only a part of the ancient theory concerning the perception of external objects, holds just as well for Reid himself. While Reid rejects the principle of immanence, he does not escape from the argument based on the relativity of sensing. Even more devastating, in my eyes, is his acceptance of the distinction between primary and secondary qualities: It forces him to the very edge of idealism. Reid fully realizes the shortcomings of the Cartesian system. And he also resists the temptations of idealism. But he does not find his way out of the maze created by Descartes. He cannot find his way out because he accepts both the argument from physics and the argument from the relativity of sensing. It is a measure of the strength of these two arguments that not even Reid dared to reject them. Small wonder, therefore, that so many recent and contemporary philosophers surrender to them.

Descartes, according to Reid, shed only part of the Aristotelian theory. This theory, he explains, “may be divided into two parts: the first, that images, species, or forms of external objects, come from the object, and enter by avenue of the senses to the mind; the second part is, that the external object itself is not perceived, but only the species or image of it in the mind” (Reid, p. 150). Descartes rejects the first part but accepts the second. But Reid not only sees that it is the principle of immanence that gives Descartes trouble, but he also uncovers the reasons for accepting it. There are, he says, two such reasons or, as he puts it, “prejudices” (Reid, 406). The first is the conviction that the mind can only act on what is part of it; there can be no interaction between what is outside and what is inside the mind. The second is the belief that if a mind has an object, this object must exist; the mind cannot be about things which are not.

I must confess that I am not sure what to make of the first of these two prejudices. Reid quotes a Dr. Porterfield in this connection: “How body acts upon mind, or mind upon body, I know not; but this I am very certain of, that nothing can act, or be acted upon, where it is not; and therefore, our mind can never perceive any thing but its own proper modifications, . . .” (Reid, p. 217; my italics). What are we to make of this “where it is not”? When I perceive a tree in the distance, I am not where the tree is. Neither my body nor my mind is in the proximity of the tree. It does not even make sense to think of my mind as being near to the tree or far from it. My mental act of seeing the tree in the distance, though it causally depends on my brain which is spatially located, is not at any particular place. It is neither where the tree is, nor is it at a certain distance from the tree. Is the objection that interaction presupposes spatial contiguity? But how, if this is the point, can the mind be said to perceive its own ideas, since neither the idea nor its perception is spatially located? And how can any interaction between body and mind then be possible?

I understand much better the second reason for the principle of immanence. This reason also buttresses the so-called argument from hallucination. The mind, one maintains, can only be related to things which exist. If you see a pink elephant in one of your hallucinations, a pink elephant cannot really be the object of your act of seeing, for there exists no such thing in this situation. Something else must truly be before your mind, something which does exist. One view has it that this something is a sense-datum (a bundle of sensations). Thus, what we perceive in hallucinations are not perceptual objects but sensations. But this conclusion proves that the argument cannot be sound. Whatever we may be said to perceive in hallucinations, it is certainly not our sensations. Shall we then give up the idea that the mind can only relate to existents? But how can we? How can there possibly exist a relation between something (the mind, the mental act) and nothing?

The Cartesians may have added a third reason for accepting the principle of immanence: The mind only knows what is in it because the mind can know for certain only what is in it. Knowledge, in the Cartesian spirit, is identified with certitude. Nothing less deserves to be called knowledge. And then one argues that one can have certain knowledge only of what is in the mind.

At any rate, whatever the reasons for accepting the principle of immanence may be, Reid clearly saw that the principle leads to idealism.

b) Tricks of the Idealistic Trade

A moment ago, I spoke of the conclusion that what we perceive are always our own ideas (sensations, sense-impressions). But this is usually presented as a self-evident axiom of the idealistic system. Berkeley, you may recall, starts out by claiming that it is evident to anyone who takes a survey of the objects of human knowledge that they are ideas of three different kinds. Or take the first sentence of Hume’s Treatise: “All the perceptions of the human mind resolve themselves into two distinct heads, which I shall call impressions and ideas.” How can philosophers get away with starting their inquiries with such an obvious falsehood? You mean to say, George Berkeley, that when we study molecular bonding, we study our ideas? You hold that when we study the causes of earthquakes, we study ideas? You maintain that when we study the feeding habits of orangutans, we study ideas? What is it that prevents the idealist from realizing the absurdity of his position? What is it that protects his view from the ridicule which it deserves? It is a way with words! Words serve as a smokescreen to hide the clash with common sense. Reid has no equal when it comes to blowing away this smokescreen. His criticisms of the terminological contortions of Locke, Berkeley, and Hume are masterpieces.

Speaking very generally, the idealist plays with the meaning of words like ‘idea’, ‘sensation’, ‘object’, etc. Locke, of course, is famous for using ‘idea’ for anything under the sun. He apologizes for his lack of precision, but this very apology perpetuates the idealistic sleight of hand:

It [idea] being that term which, I think, serves best to stand for whatsoever is the object of the understanding, when a man thinks, I have used it to express whatever is meant by phantasm, notion, species, or whatever it is which the mind can be employed about in thinking; and I could not avoid frequently using it.

I presume it will be easily granted me, that there are such ideas in men’s minds: every one is conscious of them in himself; and men’s words and actions will satisfy him that they are in others.

(J. Locke, An Essay Concerning Human Understanding. Introduction, p. 8)

We cannot but grant Locke “that there are such ideas in men’s minds.” But these ideas, we protest, are not the only objects of the understanding. Every idealist, I submit, has to perform this particular trick: he has to convince himself and us that it is really ideas which we perceive, think of, understand, or compare, when we perceive, think of, understand, or compare the things in the world around us. The price for failing the performance is an application of modus tollens to his theory.

Berkeley, too, is aware that he uses the word ‘idea’ in a very peculiar sense: “But, say you, it sounds very harsh to say we eat and drink ideas, and are clothed with ideas. I acknowledge it does so—the word idea not being used in common discourse to signify the several combinations of sensible qualities which are called things; . . .” (G. Berkeley, A Treatise Concerning the Principles of Human Knowledge, par. 38). But then he goes on to defend this unfortunate choice of terminology: “. . . but this does not concern the truth of the proposition which, in other words, is no more than to say we are fed and clothed with those things which we perceive immediately by our senses” (Berkeley, 1957, par. 38). Again, we cannot but grant Berkeley that we are fed and clothed with those “things which we perceive immediately by our senses,” but those things are certainly not ideas.

We have just come across another trick of the idealist, a trick most masterfully employed by Berkeley and greatly perfected by contemporary phenomenalists and other “reductionists.” Whenever we are outraged by the idealist’s fast and loose way with words, he assures us that our indignation is unjustified. We are told that the idealist really does not wish to deny anything we believe; that he is completely on the side of common sense or, as Berkeley puts it, on the side of the vulgar. Berkeley is a master of this ploy. He is so successful that there are philosophers who believe that he is a common-sense realist! Please, have a look at the third dialogue of the Three Dialogues between Hylas and Philonous. Philonous there assures Hylas that his philosophical (idealistic) view comes down to this:

If by material substance is meant only sensible body, that which is seen and felt, and the unphilosophical part of the world, I dare say, means no more, then I am more certain of matter’s existence than you or any other philosopher pretend to be. If there be any thing which makes the generality of mankind averse from the notions I espouse, it is a misapprehension that I deny the reality of sensible things; . . .

(Berkeley, Three Dialogues between Hylas and Philonous)

Did Berkeley really believe that his view that an apple is nothing but a collection of ideas in a mind is in accordance with the universal sense of mankind? Or was he merely trying to convince himself and others not to apply modus tollens to his view? We must never underestimate the stubbornness of philosophers. In this respect, philosophers are not so different from the vulgar: Having finally, after much toil and trouble, arrived at their views, they are loath to give them up, no matter how silly they are. But in addition philosophers have a more perverse streak: They delight in shocking accepted opinion. Some philosophers, it seems, are only happy if they have arrived at absurd conclusions. Berkeley, at any rate, tried to conceal the clash between his conclusions and common sense.

Things go from bad to worse when we turn from Berkeley to Hume. Look at this famous pronouncement:

We may divide all the perception of the human mind into two classes or species, which are distinguished by their different degrees of force and vivacity. The less lively and forcible are commonly dominated thoughts or ideas. The other species want a name in our language, and in most others; let us therefore use a little freedom, and call them impressions. By this term, impression, then, I mean all our more lively perceptions, when we hear, or see, or feel, or love, or hate, or desire, or will.

(Quoted from Reid, p. 11)

Reid, analyzing this quotation, points out the many confusions it contains. One of these deserves special attention. Hume’s list of so-called impressions appears to be a list of mental acts: An act of hearing, an act of seeing, of feeling, of loving, of hating, etc. But it could also be a list of the objects of such acts: of what is seen, of what is heard, what is loved, what is hated, etc. Nowhere in Hume does this most important distinction between a mental act and its object receive the attention that it deserves. Reid complains: “And when I read all that he has written on that subject, I find this word impression sometimes used to signify an operation of the mind, sometimes the object of the operation; but for the most part, it is a vague and indetermined word that signifies both” (Reid, p. 11).

Contemporary phenomenalists continue this tradition of trickery. The terminology has changed. Ideas are supplanted by sense-impressions. But the claim is the same: Even though sense-impressions are mental things, it is not absurd, but perfectly compatible with common sense, to believe that ordinary perceptual objects consist of them. One pretends that these mental sensations may quite naturally be thought of as “parts of the surfaces of perceptual objects.” But we must object: If by “sense-impression” we mean a mental sensation, something that would not exist if there were no minds, then it is the received opinion of all mankind that perceptual objects do not, in any sense, shape, or form, consist of sense-impressions. On the other hand, if we mean by this expression something that could conceivably be a part of a perceptual object, then a sense-impression could not possibly be mental. Sensations are sensations, perceptual objects are perceptual objects, and there is no reason at all to confuse the two.

c) Reid’s First Mistake: The Nature of Sensations

It is one thing to recognize the absurdity of idealism, but quite another to dismantle the idealistic arguments. Reid was never taken in by the idealistic legerdemain. But he did not know, alas, how to break off the point of the idealistic spear. He reminds one of G. E. Moore, whose commitment to realism never wavered, but who never found a convincing answer to the idealistic arguments. The crucial test for Reid, as for any realist, is his treatment of the arguments from physics and from the relativity of sensing. He must fail because he accepts the argument from physics and, therefore, the fatal distinction between primary and secondary qualities. Let us look at this part of his philosophy first, and at the argument from the relativity of sensing in the next section.

Reid does not distinguish between a sensation and the act of experiencing the sensation. This is one of the very few parts of his analysis of the contents of the mind with which I disagree. It becomes important for our present concern because this mistake prevents him from realizing how futile it is to distinguish between primary and secondary qualities. Reid, of course, is not the only philosopher who refuses to make the act—object distinction for sensations. But this refusal has always been rather poorly motivated. When you see a tiger, everyone admits, there is the seeing and there is the object seen, the tiger. And the seeing is not the same as its object; the seeing is not the tiger. But when you feel a pain, some say, there exists no corresponding distinction between the feeling and the pain. Why not? It seems to me to be quite obvious that just as the seeing is not the tiger, so the feeling is not the pain. May I not also feel something else, something other than a pain? And if so, does it not follow that the mental act cannot be identical with a pain? Perhaps the word “feeling” is to be blamed, since we have to speak of feeling a feeling. But this can be avoided if we agree to speak of experiencing a pain, a desire, a memory. I hold, in accordance with this terminology, that there exists a certain kind of mental act, experiences, which has objects of different sorts. Of course, what I experience when I experience a pain or when I experience a desire is just as mental as the act of experiencing itself is. This marks an important difference between these examples and our earlier example of seeing a tiger. But I fail to see how this important difference makes a difference in regard to the act-object distinction. Why, then, do so many philosophers insist that there is no act-object distinction for sensations? Why, in particular, does Reid reject the distinction? Here is what he says:

Sensation is a name given by philosophers to an act of mind, which may be distinguished from all others by this, that it hath no object distinct from the act itself. Pain of every kind is an uneasy sensation. When I am pained, I cannot say that the pain I feel is one thing, and that my feeling it is another thing. They are one and the same thing, and cannot be disjoined, even in imagination. Pain, when it is not felt, has no existence. It can be neither greater nor less in degree or duration, nor any thing else in kind, than it is felt to be. It cannot exist by itself, nor in any subject, but in a sentient being. No quality of an inanimate insentient being can have the least resemblance to it.

(Reid, p. 27)

Notice that Reid says that sensations are acts, but acts without objects. This is a truly amazing remark. A mental act without an object, according to Brentano’s celebrated thesis, is an ontological absurdity. But perhaps Reid means to say, not that the act of feeling (a pain!) has no object, but rather that it is its own object. But this assertion is no less absurd: how could a mental act have no other object than itself? We must keep in mind that this is not Brentano’s view, according to which every act has two objects, a primary object (its ordinary object, so to speak) and, as its secondary object, itself. According to Reid’s radical view, there is no primary object at all. But this is not all. Even more amazing than the claim that the feeling of a pain is an act without an object, is Reid’s identification of the pain with the feeling of the pain. According to his view, therefore, it would be more nearly correct to say that there are no pains but only feelings of pain than to maintain that there are no feelings of pain but only pains. The act-object distinction disappears for pain, not because there is no act in this case, but because there is no object.

Let us turn to Reid’s arguments. He says, firstly, that a pain has no existence when it is not felt. I agree with him completely: Just as a desire which is not experienced does not exist, so does a pain which is not felt (experienced) does not exist. In general, it holds for all mental things, with one exception, that to be experienced and to exist are equivalent. The exception is the act of experiencing itself: It exists, even when it is not itself experienced. Notice also that I said that to be experienced is equivalent to existence, not that it is the same thing. Berkeley’s slogan thus is true for all mental things save one: To be is to be experienced. But it is most certainly false for perceptual objects and their perception: To be is not to be perceived. But how does the agreed upon equivalence show that there is no distinction between the act of experiencing, on the one hand, and what is experienced, on the other? Surely, from the fact that the pain exists if and only if its experience exists it does not follow that the pain is the experience or that the experience is the pain.

I shall also agree with Reid that a pain cannot be other than what it is felt to be. But here, too, I must add a caveat. I do not mean to endorse the common view that knowledge of sensations (and other mental things) is infallible, absolutely certain, indubitable. Rather, I agree with Reid, not because I think that we cannot be mistaken about pains (and other mental things), but because experience is the kind of act which is neither correct nor incorrect. This is a longer story, and we shall come back to it later. For the moment, let us set aside certain questions and reservations and simply agree that pains are what they are experienced to be. But again, I do not see how this fact proves Reid’s contention that a pain is the same thing as its experience. Nor, finally, does this contention follow from the fact that a pain cannot exist by itself, or from the fact that it must exist in a mind.

Turn from pain to a “color sensation.” If Reid is right, then the “sensation olive green” is a mental act! Now, if there is one thing that rubs common sense the wrong way, then it is certainly the notion that the color of the olive in my Martini is a mental act. But this does not prevent Reid from exploiting his identification of act and object by treating sensations sometimes as acts and sometimes as objects, that is, as sensible properties. And this brings us to Reid’s second fundamental mistake, his acceptance of the distinction between primary and secondary qualities.

d) Reid’s Second Mistake: Secondary Qualities

When I read chapter XVII of Reid’s Essays, I am reminded of the uncomfortable truth that philosophical catastrophe lies in wait for even the most reasonable, most thoughtful, and most careful of philosophers. Extension, according to Reid, is a primary quality of bodies. Color, on the other hand, is a secondary quality. Let us compare the shape square with the color shade olive green. According to Reid, our senses give us a direct and distinct notion of the primary quality squareness, but of the secondary quality olive green, our senses give us only a relative and obscure notion. However, this is misleading, for it is not really olive green which is the secondary quality of which we are said to have a relative and obscure notion, but rather a property, not further specified, which causes us to have the sensation olive green. It would be more correct to describe Reid’s view by saying that perceptual objects have two kinds of quality: certain primary qualities like shape, and certain secondary qualities which are known only as the causes of certain sensations. It is not really olive green, therefore, which is the secondary quality, but rather some unknown quality which causes the sensation olive green. We are dealing with three things: a primary quality (squareness), a secondary quality (the cause of the sensation olive green), and the sensation olive green. We shall see later on how this distinction plays havoc with Kant’s philosophy.

I see no reason, other than the argument from physics, why shape should be favored in this manner over color. May it not just as well be asserted that there is a quality of the perceptual object, otherwise unknown, which causes the sensation square? Conversely, may one not with equal justification hold that the color is a primary quality, directly and distinctly known? Of course a shape is not a color, but I know both for what they are and equally directly. Reid tries desperately to convince us that color is not a property of the perceptual object. His example is the smell of a rose rather than its color. Our position is firm and clear, backed by the awesome power of common sense: the rose has a certain smell, this smell exists independently of any person or mind that notices the smell, and we know this smell “directly and distinctly.” Against this, Reid remarks that he is at a loss to say what that quality is in the rose which we call its smell. How quaint! Surely, that quality is the smell of the rose. We happen to have few words for distinct smells, so that we describe a smell usually as the smell of a certain kind of thing. But this does not negate the fact that we know smells directly by acquaintance, and not just by description, as the smells of certain objects. A person without the sense of smell may know the smell of a rose only as a property which the rose has, but we are acquainted with that smell.

Reid continues by claiming that he has a distinct notion of the sensation which the unknown quality of the rose, its smell, produces in his mind. We cannot quarrel with this assertion as far as it goes. But then comes the punch line: There can be nothing like this sensation in the rose, he says, because the rose is insentient. Here is the heart of his argument. Here, his contention that sensations are mental acts pays off. For, if the sensation is a mental act, then it cannot exist in the rose, since the rose is not a mind. From our point of view, Reid has committed two mistakes. Firstly, he has mistakenly identified a property (the smell) with an individual thing (the sensation). Secondly, he has mistakenly identified this sensation with the act of experiencing it. By the same path, one can of course easily prove that color cannot be a property of the rose, cannot be “in” the rose: The color is a sensation in the mind; this sensation is a mental act; and since mental acts only exist in minds, not in roses, the color cannot exist in the rose. But the smell of the rose is not a sensation. It is a property of the rose and also a property of the sensation. It exists “in” the rose and “in” the sensation. But even if we follow the pernicious custom of thinking of the property as a sensation, it does not follow that this sensation (since it is the property) cannot exist in the rose. This is why Reid needs his peculiar assumption that sensations are mental acts.

This assumption also allows him to pretend that his view does not really clash with common sense. Reid is aware that the clash exists: “But there seems to be a contradiction between the vulgar and the philosopher upon this subject, and each charges the other with a gross absurdity. The vulgar say that fire is hot, and snow cold, and sugar sweet; and to deny this is a gross absurdity and contradicts the testimony of our senses” (Reid, p. 258). Is it not ironic that the champion of common sense is forced to take the side of the philosopher against the side of the vulgar on this matter? But Reid manages to convince himself, by practicing the familiar philosophical self-deception, that the contradiction is “more apparent than real.” As usual, the self-deception requires a verbal bridge. Like some philosophers before him and many more after him, Reid claims at this point that when we say so-and-so, we do not really mean what we say, but mean such-and-such instead. When the philosopher says that there is no heat in the fire, he means, according to Reid, that the fire has no sensation of heat, that is, that there does not occur a mental act of a certain sort in the fire. And since he means precisely this and nothing more by saying that heat is not in the fire, the vulgar will agree with him as soon as his meaning becomes clear to them. On the other hand, the philosopher agrees with the vulgar that heat is in the fire, if by “heat” is meant that unknown quality which causes the sensation in us. Thus the vulgar and the philosopher really agree, and the former do not contradict the latter.

What a piece of sophistry! We should expect this kind of assurance from Berkeley, not from Reid. It should be clear to everyone that Reid does not resolve the contradiction between the opinion of common sense and the theory of the philosopher. Of course, the fire does not contain a mental act of experiencing (sensing) heat. The philosopher has common sense on his side with this remark. But why does he insist on expressing this truism by saying that the fire is not hot (that “heat is not in the fire”)? Surely this is not what we ordinarily mean by saying that fire is not hot. Obviously, what we do mean to assert by saying that fire is hot is that it has a certain quality of hotness; that it has this property; that it has this characteristic. In short, we ascribe a certain property to the fire, not a mental act. The issue is not whether or not fire feels heat, but whether or not it is hot. The philosopher, according to Reid, also agrees with us vulgar people that there is really a property hotness in the fire, that the fire really has this property. But it is clear that the property which the philosopher has then in mind is not the property which we have in mind. When we speak of heat, we mean heat, not some quality, we know not what, which causes us to have certain sensations. We mean a property with which we are acquainted. Again, there really is no agreement between the vulgar and the philosopher, contrary to Reid’s claim. And this shows us, in the end, that the philosopher must be mistaken.

If squareness, unlike color, is a true property of bodies, we may ask, what is the corresponding sensation? According to Reid, there is such a sensation and it is a mental act. But what is this sensation? I cannot think of a plausible answer. Reid, however, assures me that I am hard pressed for an answer only because I pay no attention to this sensation: “When a primary quality is perceived, the sensation immediately leads our thought to the quality signified by it, and is itself forgotten. We have no occasion afterward to reflect upon it; and so we come to be as little acquainted with it, as if we had never felt it” (Reid, p. 257).

If Reid were correct, the color olive green would be comparable to the feeling of a toothache. One could also say that it is comparable to a toothache; for the feeling of a toothache and the toothache are, according to Reid, one and the same thing. At any rate, do we not say that we feel a toothache, but see the color of the rose? If color really were a sensation like a pain, would we not say that we feel the color of the rose? Reid invokes the familiar ploy according to which what is not there, is really there, but not attended to. When we feel a toothache, he says, we also perceive a disorder in the tooth. Thus we find sensation and perception conjoined both in the case of color and in the case of pain. There is this important difference between the two cases: “But, in the toothache, the sensation being very painful, engrosses the attention; and therefore we speak of it, as if it were felt only, and not perceived; whereas, in seeing a colored body, the sensation is indifferent, and draws no attention. The quality in the body, which we call its color, is the only object of attention; and therefore we speak of it, as if it were perceived, and not felt” (Reid, p. 268). According to Reid, in the case of pain, we pay attention to the pain, this sensation, but do not pay attention to something else that occurs simultaneously with it, namely, a perception of some disorder or other. Well, I do not think that any such perception occurs when I have a toothache. There is no seeing, hearing, smelling, etc. of any disorder in the tooth. But this is not the strangest part of Reid’s view. As you just read, he goes on to say that when we perceive a colored body, we pay no attention to its color; the color, this sensation, we are told is indifferent. But surely, if there is one thing to which we pay attention when we see a colored body, it is its color. Of course, we must not be misled by Reid’s phrase “the quality in the body, which we call its color.” We saw earlier that this quality is not what we would call its color. If we called it anything, we would speak of “that property of the surface of the body which is responsible for our experience of a sensation of such-and-such a color” or of something similar.

We feel a pain in a certain tooth. But if the pain is the feeling of the pain, if it is a mental act, then it cannot possibly be in the tooth, and common sense must again be mistaken. Here, then, is another problem for Reid. This time, he grants that we say something that is neither obscure nor false when we say that the ache is in the tooth. But he claims that it is really the disorder which is in the tooth and that the pain is, properly speaking, nowhere. Common sense simply confuses the perceived cause of the pain with the pain itself. But, we are immediately reassured, common sense is nevertheless not mistaken, because it never makes the distinction and means by ‘pain’ both the sensation and its cause. Speaking once again on behalf of the vulgar, I categorically deny that we confuse the pain with its cause and that we thus mean both by “pain.” Since we do not confuse the pain with the decay in the tooth, we do not mean to say that that particular tooth is decayed when we say that it hurts. Once again it is obvious that Reid’s view cannot be reconciled with common sense, no matter how hard he tries.

Why is it, we may want to know at this point, that philosophers continuously tell us that we cannot mean what we say, but must mean something else? And why is it that they keep insisting that we must mean what they say we mean, rather than what we know we mean? Should we not be the best judges in these cases? And why do they constantly claim that we would cheerfully admit that their absurdities are our truisms, if only we would make the proper distinctions, when in reality we make precisely these distinctions? Let it be granted, once and for all, that we mean by ‘olive green’ a certain color shade and not an unknown quality; that we mean by ‘toothache’ a pain and not the decay in the tooth; and that we mean by ‘feeling’ a pain the feeling of a pain and not the pain itself.

So much for Reid’s contention that color is a sensation, while shape is a property of bodies. The reason for this contention, we already know, does not come from philosophy proper. There is no phenomenological argument to the effect that color is in the mind while shape is in the world. No, the contention comes from misunderstood physics. Reid alludes to this background in connection with a remark about the atomists: “That the atoms, which they held to be the first principles of things, were extended, solid, figured, and moveable, there was no doubt; but the question was, whether they had smell, taste, and color . . . (Reid, p. 259). So-called primary qualities are simply those properties, not to be discovered by philosophical reflection, but by physics, which the ultimate particles have; secondary qualities, then, are all of the rest of the properties which perceptual objects seem to have. “Out there,” so the reasoning goes, only elementary particles and their properties exist. If there are any other properties at all, they must then be “in the mind.” Hence they must be sensations. In this manner is born the absurd view that a color like olive green is some kind of mental phenomenon. Reid, we have just seen, holds that it is a mental act. Brentano, we shall see later, denies that it is a mental phenomenon, precisely because it is not a mental act. But he agrees with the received opinion of almost all philosophers that it is not a physical phenomenon either. Thus he comes to the conclusion that it does not exist at all!

Reid faces still another problem. Secondary qualities, precisely speaking, are properties in bodies which are known only as the causes of certain sensations. But this cannot really be true, according to Reid’s theory. “It has been discovered,” he says, “that the sensation of smell is occasioned by the effluvia of bodies; that of sound by their vibrations. The disposition of bodies to reflect a particular kind of light occasions the sensation of color” (Reid, p. 256). Thus it seems that we know these secondary qualities after all. We can say a number of positive things about them. We know them clearly and distinctly, just as we know the so-called primary qualities of figure, motion, number, etc.

e) Reid’s Third Mistake: Acquired Perception

One of the great virtues of Reid’s philosophy is that he admits the existence of mental acts of perception. And it is to his everlasting credit that he clearly sees that perception is not a matter of inference. Yet, when he analyzes the crucial argument from the relativity of sensing, he cannot shake off traditional philosophical wisdom and escape from error. He possesses the means to counter the argument, but does not, at the crucial moment, employ them effectively.

Perception, according to Reid, has three important ingredients (Reid, p. 111). Firstly, it involves a conception or notion of the perceived object. Secondly, it contains a strong and irresistible conviction and belief in the existence of the object. Thirdly, this belief in the existence of the object is immediate and not a matter of reasoning. In short, an act of perception, according to Reid, is an act of believing in the existence of a certain perceptual object, but this belief is not arrived at by inference. I hold that perception is propositional. One perceives, for example, that (the apple) Oscar is green. I am therefore willing to grant without further ado that perception involves a conception of the perceived object. It is the second feature of perception, pointed out by Reid, which is of concern to me.

I am not sure whether Reid holds that one cannot perceive what does not exist, or merely that one is always convinced that what one perceives exists. At one point, he says, “I acknowledge that a man cannot perceive an object that does not exist; . . .” (Reid, p. 419). But at other places he merely states, “When I perceive any external object, my belief of the real existence of the object is irresistible” (Reid, p. 404). Of course, there is no conflict between these two views. Reid may have held both. But for my purposes it is necessary to distinguish between them. I think it is clearly false that all objects of perception exist; just as clearly as that all beliefs are true or that all desires are fulfilled. I take it to be a matter of common sense that one can see a pink elephant in one’s hallucination, or that one can hear the devil whisper obscenities, while smelling his hellish odor. Under certain abnormal circumstances, there occur perceptual acts whose objects do not exist. These acts, and this is the crucial point, are perfectly good acts of perception: the seeing of the pink elephant is in precisely the same way a seeing as the seeing of a real elephant in a circus; it is not just a mere judging, or believing, or surmising. What is abnormal about such situations is precisely this, that a perfectly “normal” act of perception has a nonexistent object. This is in my opinion, as we shall see later, the proper response to the argument from hallucination. You can see how important it is for me to distance myself from Reid’s view, if his view implies that all objects of perception exist.

What about the second view, the view that there is always an irresistible belief in the existence of the perceived object? It is true that one may believe in the existence of an object, even if that object does not exist. Why, otherwise, would a person follow the advice of the devil who speaks to him? But does one always believe in the existence of the perceived object? Can there be no perception with the simultaneous conviction that what one perceives does not exist? Is it not possible to know that one is hallucinating and, therefore, not to believe in the existence of the perceived objects? I think that it is. But we must be careful at this point. When we believe in the existence of the perceived object, or when we doubt its existence, the belief or doubt is not a constituent of the perceptual act. In my view, perception does not consist, even in part, of believing. Perception is not a form of believing. An act of perception, that is, an act of seeing, or of hearing, etc., is an indefinable, unanalyzable mental act, as different from an act of believing as it is from an act of desiring or an act of remembering. The adage “Seeing is believing” does not mean that seeing is the same thing as believing. Rather, it expresses the truth that one believes what one has seen with one’s own eyes. Belief, we may perhaps say, follows in the wake of perception. It is based on perception. It accompanies perception. But it is not the same mental phenomenon as perception.

Reid’s third characteristic of perception is the crucial one for our immediate purpose. Our conviction of the existence of the perceived object, Reid tells us, is immediate; it is not derived from a train of reasoning and argumentation (Reid, p. 116). We do not infer that the perceived apple exists, but perceive it. With this insight, Reid breaks the spell of the principle of immanence. Let me describe the situation in my own words. When you see that Oscar is green, you neither infer that there is an apple before you nor do you infer that it is green. In particular, we do not infer these things from our visual sensations. Perception is not sensation plus inference. As Reid clearly sees, it is precisely this assumption that perception somehow consists of sensation plus inference that the skeptic needs for his argument (Reid, p. 666). Such an inference must be based on knowledge of a specific connection between certain sensations and certain perceptual objects. But this knowledge can be gained only if we can be acquainted with perceptual objects independently of sensations. Such acquaintance, however, is impossible according to the principle of immanence.

According to one accepted view, when you see that Oscar is green, you experience certain visual sensations, and you infer from these sensations that there is a green apple in front of you. By means of this inference, as it is sometimes put, you acquire the belief that there is a green apple in front of you. It is this acquired belief which is your perception. Or perhaps it is the acquisition of the belief (together with the acquired belief?), that is, the process of inference, which is the perception. In either case, you reason from the experienced sensations, according to certain principles, to the existence and nature of a perceptual object. The skeptic argues that you could not possibly have acquired the necessary principles of inference. Assuming the truth of the principle of immanence, as I have said before, he is right. Well, what is the correct analysis of the perceptual situation? One thing is certain: In perception, no inference takes place. There occur certain sensations and there occurs an act of perceiving. But no mental process of inference takes place, either consciously or unconsciously, either explicitly or implicitly, either deliberately or habitually. What, then, is the connection between the experienced sensations and the act of perception?

According to Reid, the sensation is a natural sign which signifies the property of the external body. But this connection between sign and property is not itself perceived. Thus we cannot infer the property from the sign on the basis of our knowledge of the connection. As Reid describes the situation, “When I grasp an ivory ball in my hand, I feel a certain sensation of touch. In the sensation, there is nothing external, nothing corporeal. The sensation is neither round nor hard; it is an act of feeling of the mind, from which I cannot, by reasoning, infer the existence of any body. But, by the constitution of my nature, the sensation carries along with it the conception and belief of a round hard body really existing in my hand” (Reid, p. 638). I disagree with Reid, of course, about the nature of the sensation: It is not a mental act. And though it is true that there is “nothing external, nothing corporeal” in it, it is not true that the sensation is not round and hard. But these disagreements do not matter for the moment. What matters, rather, is the following agreement. Reid says that by the constitution of his nature the sensation carries along with it the relevant perception. When he has the sensation, he also has the perception. I think that this is essentially correct, but I would express it in a different way. Roughly, the experienced sensations are a (partial) cause of the act of perception. Since you experience certain sensations, and given that certain other conditions are fulfilled, you also experience a certain act of perception, that is, you see that Oscar is green. Thus the act of perceiving does not occur as a result of a process of inference. It is not caused by a thought or reasoning process, but is caused directly, as it were, by the sensations.

Human nature is such that we have certain perceptions whenever we experience certain sensations (and certain other conditions are fulfilled). This bond is a matter of our constitution. In principle, an organism could have all of our sensations, but none of our perceptions. An organism could even experience acts of perception without the corresponding sensations! Reid sees this too: “Nor can we perceive any necessary connection between sensation and the conception and belief of an external object. For anything we can discover, we might have been so framed as to have all the sensations we now have by our senses, without any impressions on our organs and without any conception of any external object. For any thing we know, we might have been so made as to perceive external objects, without any impressions on bodily organs, and without any of those sensations which invariably accompany perception in our present frame” (Reid, p. 289). Needless to say, the causal relationships between sensations and corresponding perceptions are extremely complicated. And I must emphasize once again that many other factors determine just which perceptions occur. Expectations play a role, and so do feelings and emotions, past experiences, and learning. Most importantly, the perceptual objects themselves influence what we perceive. This is such an important fact because it clashes with the customary picture of the causal chain in perception.

According to this customary picture, certain physiological processes occur in the sense-organs, the nervous system, and in the brain, as a consequence of certain external stimuli (the perceptual object, etc.). These physiological processes eventually cause certain sensations to be experienced. The mind somehow goes to work on these sensations and, as a result, arrives at a certain belief about a perceptual object. The mind somehow reasons from the sensations to what there is in front of one’s nose. Sensations furnish the only basis for one’s perceptual beliefs. The way to the perceptual belief leads causally as well as inferentially through the sensations. Now, Reid does away with the notion that the perceptual belief is arrived at by means of an inference from the sensations. But this leaves us still with the causal chain. What I wish to point out is that this connection as well is tentative. We must change our preliminary characterization of the relationship between sensations and perception. The physiological processes which take place in the brain as a consequence of the stimulation of the sense-organs, and of past experience, expectations, etc., cause both the experience of certain sensations and the experience of certain acts of perception. The occurrence of a certain act of perception, in other words, does not causally depend on the sensations alone, although it may look that way, since the two usually occur together. Rather, we cannot perceive without sensations because the processes in the brain which cause us to perceive, inevitably (as a matter of fact) also cause us to have certain sensations. Can we have sensations without perceptions? The traditional answer assumes that the dependency has one direction: perceptions are caused by sensations, but sensations can occur without perceptions. According to the customary picture, one somehow learns to interpret one’s sensations. These interpretations are one’s perceptions. A baby merely experiences sensations; a buzzing, booming, confusion of sensations. Only later does it learn slowly, by inference, what there is in front of its eyes and mouth. Or perhaps this learning takes place rather quickly, once the baby has grasped the basic rules of inductive and deductive logic. What really happens is of course quite different from this philosophical picture. At a very early age, the baby perceives perceptual objects and experiences sensations. If there ever is a time when it merely experiences sensations and does not perceive, then this is undoubtedly due to the fact that its nervous system has not yet reached a certain degree of maturity.

We are finally ready to evaluate Reid’s reply to the argument from the relativity of sensing. He considers Hume’s version of the argument: “The table, which we see, seems to diminish as we remove further from it; but the real table, which exists independent of us, suffers no alterations. It was therefore nothing but its image which was present to the mind” (Hume, as quoted by Reid, p. 221). The crucial assumption is that we see a smaller table, as we move away from it. Since the real table does not get smaller, so Hume argues, what we see cannot be the actual size of the table, but only, as he puts it, an “image.” Here is Reid’s response:

We learn by experience to judge of the distance of a body from the eye within certain limits; and from its distance and apparent magnitude taken together, we learn to judge of its real magnitude. And this kind of judgment, by being repeated every hour, and almost every minute of our lives, becomes when we are grown up, so ready and so habitual, that it very much resembles the original perceptions of our senses, and may not improperly be called acquired perception.

(Reid, p. 225)

It seems to me that we have here the precise kind of picture of how we “learn how to perceive on the basis of our sensations” which I criticized earlier. I say cautiously “it seems to me,” because the case is not as clear for magnitude (size) as it is for other properties of perceptual objects. But what Reid says about size in this passage, he also seems to assert, for example, for shape: “. . . open your eyes and you shall see a table precisely of that apparent magnitude, and that apparent figure, which the real table must have in that distance, and in that position” (Reid, p. 227). Allow me therefore to switch from size to shape, and let us consider the shape of the table top. Hume’s argument then becomes: The shape of the table top which you see when you approach the table changes from trapezoidal to rectangular. But the actual shape of the table top does not change. Hence, what you see when you approach the table is not the actual shape of the table top, but only the shape of an impression in your mind.

Berkeley already noted in the Treatise (para. 15) that this argument is not valid. In truth, the argument does not show that we cannot see the shape of the table top, but only, assuming that the premises are true, that we cannot know which of the seen shapes is the shape of the table top. Granted that what we see from different angles are different shapes, the question arises: Which of the seen shapes is the real shape of the table top? Is it trapezoidal, for example, or is it rectangular? The skeptic claims that there is no reason why we should prefer one of the seen shapes to all the others.

We shall return to the argument and give it a more precise formulation. However, the gist of it is clear enough that we can evaluate Reid’s response. Reid says that we learn to judge the true shape of the table top by taking into consideration our distance from the table, our elevation relative to it, etc. This learning takes place in childhood. After a while, the inference becomes so “ready and so habitual” that we no longer notice it. It then resembles ordinary perception which, as we saw, is not a matter of inference. I think that there are two things wrong with this account. First of all, it is simply not true, as experience plainly shows, that children infer the shapes of things in this manner. It is absurd to believe that they use the rather complicated geometric laws required for such inferences. Anyone who has tried to teach geometry to older children can only scoff at the thought that tiny tykes should constantly and accurately deduce the shapes of things by means of the laws of geometry. Nor is there any introspective evidence for the existence of such a process of inference. I am convinced that nobody can recall practicing in early childhood inferences about the true colors of things (with the help of the laws of optics?), about their true shapes, etc. Or do these inferences lie so far back that no memories can reach them? But then how plausible is the claim that even small babies use the laws of geometry and optics?

Nor, secondly, does Reid’s explanation really come to grips with the skeptic’s argument. The skeptic wants to know how we can possibly select the real shape of the table top from all the shapes we see when we walk around the table, when none of them wears a label saying “I am the true shape.” Obviously, we need some kind of principle of selection, for example, the truth that if, from a certain definite point of view, you see the table top with such-and-such a shape, then it actually is rectangular. But we can never discover such a principle, according to the skeptic, for we could only discover it if we could compare the seen shape with the actual shape; and this comparison is only possible if either you can directly see the actual shape or else infer what it is by means of a principle which is different from the one we are trying to discover. The first possibility is excluded by the nature of the situation: it is assumed that none of the shapes which you directly see stands out as the shape of the table. The second possibility is also eliminated because it would involve us in a vicious infinite regress. It follows that we cannot compare the apparent with the real shape. But since we cannot compare them, we cannot discover the required principle of inference. And without such a principle, we cannot know the actual shape of the table top.

I believe that the skeptic’s line of reasoning, given the premises which Reid grants him, is sound and convincing. If we accept Reid’s view that the shape of the table top can only be known through an inference, then it follows that it cannot be known at all. But it is rather astonishing that Reid adopts this view, for he could hold instead that we know the shape directly through perception. Perception, we noted, is for him not a matter of inference. However, he would then have to reject Hume’s first premise, namely, that we see different shapes from different points of view. Perhaps he was under the spell of the traditional picture according to which babies learn to infer the properties of the things around them from their sensations.

Be that as it may, I think that Reid’s failure to overcome idealism and his capitulation to skepticism are not so much signs of a lack of philosophical acumen, as they are proof of the power of the distinction between primary and secondary qualities and of the principle of immanence. I know of no philosopher from Descartes to Moore who could withstand this power.