bles for that judgment? To be in error about the application of a symbol, you must have a symbol that symbolizes something. But in so far as the thing symbolized is not known through the symbol, how is it symbolized by that symbol? Is it not, like the Unknowable, once for all out of the thought, so that one cannot just then be thinking about it at all, and 402 THE RELIGIOUS ASPECT OF PHILOSOPHY so cannot, in this thought at least, be making blun¬ ders about it? But in so far as the thing symbolized is, through the symbol, in one’s thought, why is it not known, and so correctly judged? All this in¬ volves that old question of the nature of symbols. They are to mean for us more than we know that they mean. How can that be? No doubt all that is really possible, but how?
IV.
IV.
We follow our difficulty into another department. Let us attempt a sort of provisional psychological description of a judgment as a state of mind. So regarded, a judgment is simply a fact that occurs in somebody’s thought. If we try to describe it as an occurrence, without asking whence it came, we shall perhaps find in it three elements, — elements which are in some fashion described in Ueberweg’s well- known definition of a judgment as the “ Conscious¬ ness about the objective validity of a subjective union of ideas.” Our interpretation of them shall be this: The elements are: The Stibject, with the ac¬ companying shade of curiosity about it; the Pred¬ icate, with the accompanying sense of its worth in satisfying a part of our curiosity about the subject; and the Sense of Dependence, whereby we feel the value of this act to lie, not in itself, but in its agree¬ ment with a vaguely felt Beyond, that stands out there as Object.
Now this analysis of the elements of a judgment is no explanation of our difficulties; and in fact for THE POSSIBILITY OF ERROR the moment only embarrasses us more. But the na¬ ture of the difficulty may come home to us somewhat more clearly, if we try to follow the thread of this analysis a little further. Even if it is a very imper¬ fect account, it may serve to lead us up to the true insight that we seek into the nature of error. Let us make the analysis a little more detailed.
In its typical form then, the judgment as a mental state seems to us to begin with a relatively incom¬ plete or unstable or disconnected mass of conscious¬ ness, which we have called the Subject, as it first be¬ gins to be present to us. This subject-idea is at¬ tended by some degree of effort, namely, of atten¬ tion, whose tendency is to complete this incomplete subject by bringing it into closer connection with more familiar mental life. This more familiar life is represented by the predicate-idea. If the effort is successful, the subject has new elements united to it, assumes in consciousness a definiteness, a coherency with other states, a familiarity, which it lacked at the outset of the act of judgment; and this coherency it gets through its union with the predicate. All this is accompanied further by what one for short may call a sense of dependence. The judgment feels it¬ self not alone, but looks to a somewhat indefinite ob¬ ject as the model after which the present union of ideas is to be fashioned. And in this way we ex¬ plain how the judgment is, in those words of Ueber- weg’s definition, “ the consciousness about the objec¬ tive validity of a subjective union of ideas.
Now as a mere completion of subject-idea through the addition of a predicate-idea, the judgment is sim- 404 THE RELIGIOUS ASPECT OF PHILOSOPHY ply a mental phenomenon, having interest only to the person that experiences it, and to a psychologist. But as true or as false the judgment must be viewed in respect to the indefinite object of what we have called the sense of dependence, whereby the judg¬ ment is accompanied. Seldom in any ordinary judg¬ ment does this object become perfectly full and clear; for to make it so would often require many, perhaps an infinite, series of judgments. Yet, for the one judgment, the object, whether full and clear or not, exists as object only in so far forth as the sense of dependence has defined it. And the judgment is true or false only with reference to this undefined ob¬ ject. The intention to agree with the object is con¬ tained in the sense of dependence upon the object, and remains for this judgment incomplete, like the object itself. Somewhat vaguely this single act in¬ tends to agree with this vague object.
Such being the case, how can the judgment, as thus described, fairly be called false? As mere psy¬ chological combination of ideas it is neither true nor false. As accompanied by the sense of dependence upon an object, it would be false if it disagreed with its imperfectly defined object. But, as described, the only object that the judgment has is this imper¬ fectly defined one. With this, in so far as it is for the moment defined, the judgment must needs agree. In so far as it is not defined, it is however not object for this judgment at all, but for some other one. What the imperfect sense of dependence would fur¬ ther imply if it existed in a complete instead pf in an incomplete state, nobody can tell, any more than THE POSSIBILITY OF ERROR one can tell what towns would grow up by a given rain-pool, if it were no pool, but a great lake. The object of a single judgment, being what it is, namely, a vaguely defined object, present to this judgment, is just what it is for this judgment, and the judgment seems once for all to be true, in case it is sincere.
Some one may here at once answer that we neg¬ lect in this description the close interdependence of various judgments. Thought, some one may say, is an organic unity. Separated from all else but its own incompletely defined object, a single judgment cannot be erroneous. Only in the organic unity of a series of judgments, having a common object, is the error of one of them possible. W e reply that all this will turn out to be just our result. But the usual supposition at the outset is that any judgment has by itself its own object, so that thereby alone, apart from other judgments, it stands or falls. And thus far we have tried to show that this natural sup¬ position leads us into difficulty. We cannot see how a single sincere judgment should possibly fail to agree with its own chosen object. But enough of our problem in general. We must consider certain classes of errors more in detail. Let us see how, in these special classes of cases, we shall succeed in ver¬ ifying the natural presupposition of common sense, which regards error as possible only when our object is not wholly present to mind, and which assumes that a judgment can have an object that is yet only partially present to mind. In choosing the classes of cases, we shall first follow common sense as to their definition. We shall take just the assumptions of 406 THE RELIGIOUS ASPECT OF PHILOSOPHY daily life, and shall show that they lead us into diffi¬ culty. We are not for the first bound to explain why these assumptions are made. That common sense makes them is enough.
But let the reader remember: The whole value of our argument lies in its perfect generality. How¬ ever much we dwell on particular classes of errors, we care nothing for the proof that just those errors are inexplicable, but only for the fact that they il¬ lustrate how, without some entirely new hypothesis, absolutely all error becomes impossible. This or that class of judgments may be one in which all the judgments are relative, but the total relativity of our thought implies an incomprehensible and contra¬ dictory state of things. Any hypothesis about error that makes total relativity the only admissible view, must therefore give place to some new hypothesis. And our illustrations in the following are intended to show that just what constitutes the difficulty in respect of these illustrations, makes the existence of any error inexplicable without some new hypothesis.
y.
y.
The class of errors that we shall first take seems, to common sense, common enough. It is the class known as errors about our neighbor’s states of mind. Let us then, for argument’s sake, assume without proof that our neighbors do exist. For we are not here concerned to answer Solipsism, but merely to exemplify the difficulties about the nature of error. If our neighbors did not exist, then the nature of THE POSSIBILITY OF ERROR the error that would lie in saying that they do exist would present almost exactly the same difficulties. W e prefer, however, to begin with the common-sense assumption about ourselves and our neighbors as separate individuals, and to ask how error can then arise in judging of our neighbors’ minds.
In the first place then: Who is my neighbor? Surely, on the assumptions that we all make, and that we made all through the ethical part of our dis¬ cussion, he is no one of my thoughts, nor is any part of him ever any part of my thought. He is not my object, but, in Professor Clifford’s phrase, an “ eject,” wholly outside of my ideas. He is no “ thing in my dream,” just as I am not in his dream.
Yet I make judgments about him, and he makes them about me. And when I make judgments about him, I do so by having in my thought some set of my own ideas that, although not himself, do yet, as I say, represent him. A kind of dummy, a symbol, a graven image of my own thought s creation, a phan¬ tom of mine, stands there in me as the representa¬ tive of his mind; and all I say about my neighbor’s inner life refers directly to this representative. The Scottish philosophy has had much to say to the world about what it calls direct or presentative, as opposed to representative, knowledge of objects. But surely the most obstinate Scottish philosopher that ever ate oatmeal cannot hold so tenaciously by his national doctrine as to say that I have, according to common sense, anything but a representative knowledge of my neighbor’s thoughts and feelings. That is the only sort of knowledge that common sense will re- 408 THE RELIGIOUS ASPECT OF PHILOSOPHY gard as possible to me, if so much as that is possible. But how I can know about this outside being is not now our concern. We notice only that our difficulty about error conies back to us in a new form. For how can I err about my neighbor, since, for this com¬ mon-sense view, he is not even partly in my thoughts? How can I intend that as the object of my thought which never can be object for me at all?
But not everybody will at once feel the force of this question. We must be more explicit. Let us take the now so familiar suggestion of our great hu¬ morist about the six people that take part in every conversation between two persons. If John and Thomas are talking together, then the real John and Thomas, their respective ideas of themselves, and their ideas of each other, are all parties to the con¬ versation. Let us consider four of these persons, namely, the real John, the real Thomas, John as Thomas conceives him, and Thomas as John con¬ ceives him. When John judges, of whom does he think? Plainly of that which can be an object to his thoughts, namely, of his Thomas. About whom then can he err? About his Thomas? No, for he knows him too well. His conception of Thomas is his conception, and what he asserts it to be, that it is for him. About the real Thomas? No, for it should seem, according to common sense, that he has nothing to do with the real Thomas in his thought, since that Thomas never becomes any part of his thought at all. u But,” says one, “ there must be some fallacy here, since we are sure that John can err about the real Thomas.” Indeed he can, say THE POSSIBILITY OF ERROR we; but ours is not this fallacy. Common sense has made it. Common sense has said: “ Thomas never is in John’s thought, and yet John can blunder about Thomas.” How shall we unravel the knot?
One way suggests itself. Mayhap we have been too narrow in our definition of object. Common sense surely insists that objects axe outside of our thought. If, then, I have a judgment, and another being sees both my judgment and some outside ob¬ ject that was not in my thought, and sees how that thought is unlike the object in some critical respect, this being could say that my assertion was an error. So then with John and Thomas. If Thomas could know John's thoughts about him, then Thomas could possibly see John’s error. That is what is meant by the error in John’s thought.
But mere disagreement of a thought with any ran¬ dom object does not make the thought erroneous. The judgment must disagree with its chosen object. If John never has Thomas in thought at all, how can John choose the real Thomas as his object? If I judge about a penholder that is in this room, and if the next room is in all respects like this, save for a penholder in it, with which my assertion does not agree, who, looking at that penholder in that other room, can say that my judgment is false? For I meant not that penholder when I spoke, but this one. I knew perhaps nothing about that one, had it not in mind, and so could not err about it. Even so, suppose that outside of J ohn there is a real Thomas, similar, as it happens, to John s ideal Thomas, but lacking some thought or affection that 410 THE KELIGIOUS ASPECT OF PHILOSOPHY John attributes to his ideal Thomas. Does that make John’s notion an error? No, for he spoke and could speak only of his ideal Thomas. The real Thomas was the other room, that he knew not of, the other side of the shield, that he never could con¬ ceive. His Thomas was his phantom Thomas. This phantom it is that he judges and thinks about, and his thoughts may have their own consistency or in¬ consistency. But with the real other person they have nothing to do. The real other is not his ob¬ ject, and how can he err about what is not object for him?
Absurd, indeed, some one will reply to us. John and Thomas have to deal with representative phan¬ toms of each other, to be sure; but that only makes each more apt to err about the real other. And the test that they can err is a very simple one. Suppose a spectator, a third person, to whom John and Thomas were both somehow directly present, so that he as it were included both of them. Then John’s judgment of his phantom Thomas would be by this spectator at once compared with the real Thomas, and even so would Thomas’s judgment of John be treated. If now John’s phantom Thomas agreed with the real Thomas, then John’s ideas would be declared in so far truthful; otherwise they would be erroneous. And this explains what is meant by John’s power to err about Thomas.
The explanation is fair enough for its own pur¬ pose, and we shall need it again before long. But just now we cannot be content with it. Foe what we want to know is not what the judgment of a THE POSSIBILITY OP ERBOK third thinker would be in case these two were some¬ how not independent beings at all, but things in this third being’s thought. For we have started out with the supposition of common sense that John and Thomas are not dreams or thoughts of some higher third i>eing, but that they are independent beings by themselves. Our supposition may have to be given up hereafter, but for the present we want to hold fast to it. And so John’s judgment, which we had supposed to be about the independ¬ ently existing Thomas, has now turned out to be only a judgment about John’s idea of Thomas. But judgments are false only in case they disagree with their intended objects. What, however, is the ob¬ ject of John’s judgment when he thinks about Thomas? Not the real Thomas, who could not pos¬ sibly be an object in another man’s thoughts. John’s real object being an ideal Thomas, he cannot, if sin¬ cere, and if fully conscious of what he means by Thomas, fail to agree in his statements with his own ideal. In short, on this our original supposition, John and Thomas are independent entities, each of which cannot possibly enter in real person into the thoughts of the other. Each may be somehow rep¬ resented in the other’s thoughts by a phantom, and only this phantom can be intended by the other when he judges about the first. For unless one talks nonsense, it should seem as if one could mean only what one has in mind.
Thus, like the characters in a certain Bab ballad, real John, real Thomas, the people in this simple tale, are total strangers to each other. You might 412 THE RELIGIOUS ASPECT OF PHILOSOPHY as well ask a blind man to make true or false judg¬ ments about the real effects of certain combinations of colors, as to ask either John or Thomas, defined as common sense defines them, to make any judg¬ ments about each other. Common sense will assert that a blind man can learn and repeat verbally cor¬ rect statements about color, or verbally false state¬ ments about color, but, according to the common- sense view, in no case can he err about color-ideas as such, which are never present to him. You will be quite ready to say that a dog can make mistakes about the odors of the numberless tracks on the highway. You will assure us, however, that you cannot make mistakes about them because these odors do not exist for you. According to the com¬ mon-sense view, a mathematician can make blunders in demonstrating the properties of equations. A Bushman cannot, for he can have no ideas correspond¬ ing to equations. But how then can John or Thomas make errors about each other, when neither is more present to the other than is color to the blind man, the odor of the tracks on the highway to the dog’s master, or the idea of an equation to a Bushman? Here common sense forsakes us, assuring us that there is such error, but refusing to define it.
The inconsistency involved in all this common- sense view, and the consequences of the inconsistency, will appear yet better with yet further illustration. A dream is false in so far as it contains the judgment that such and such things exist apart from us; but at least in so far as we merely assert in our dreams about the objects as we conceive them, we make true THE POSSIBILITY OF ERROR 413 assertions. But is not our actual life of assertions about actual fellow-beings much like a dream to which there should happen to correspond some real scene or event in the world? Such correspondence would not make the dream really “ true,” nor yet false. It would be a coincidence, remarkable for an outside observer, but none the less would the dreamer be thinking in his dream not about external objects, but about the things in his dream. But is not our supposed Thomas so and only so in the thought of John as he would be if John chanced to dream of a Thomas that was, to an external spectator, like the real one? Is not then the phantom Thomas, John’s only direct object, actually a thing in John’s thought? Is then the independent Thomas an object for John in any sense?
Yet again. Let us suppose that two men are shut up, each in a closed room by himself, and for his whole life; and let us suppose that by a lantern con¬ trivance each of them is able at times to produce on the wall of the other’s room a series of pictures. But neither of them can ever know what pictures he produces in the other’s room, and neither can know anything of the other’s room, as such, but only of the pictures. Let the two remain forever in this re¬ lation. One of them, A, sees on his wall pictures, which resemble more or less what he has seen in his own room at other times. Yet he perceives these to be only pictures, and he supposes them to represent what goes on in another room, which he conceives as like his own. He is interested, he examines the phe¬ nomena, he predicts their future changes, he passes 414 THE RELIGIOUS ASPECT OF PHILOSOPHY judgment upon them. He may, if you like to con¬ tinue the hypothesis, find some way of affecting them, by himself acting in a way mysterious to him¬ self so as to produce changes in B’s actual room, which again affect the pictures that the real B pro¬ duces in A’s room. Thus A might hold what he would call communication with his phantom room. Even so, B lives with pictures before him that are produced from A’s room. Now one more supposi¬ tion, namely, that A and B have absolutely no other means of communication, that both are shut up alto¬ gether and always have been, that neither has any objects before him but his own thoughts and the changing pictures on the wall of his room. In this case what difference does it make whether or no the pictures in A’s room are actually like the things that could be seen in B’s room? Will that make A’s judgments either true or false? Even if A, acting by means that he himself cannot understand, is able to control the pictures on his wall by some alteration that he unconsciously produces in B’s room and its pictures, still A cannot be said to have any knowledge of the real B and his room at all. And, for the same reason, A cannot make mistakes about the real room of B, for he will never even think of that real room. He will, like a man in a dream, think and be able to think only of the pic¬ tures on his wall. And when he refers them to an outside cause, he does not mean by this cause the real B and his real room, for he has never dreamed of the real B, but only of the pictures and of his own inter¬ pretation of them. He can therefore make no false THE POSSIBILITY OF ERROR judgments about B’s room, any more than a Bushman can make false judgments about the integral calculus.
If to our present world there does correspond a second world somewhere off in space, a world exactly like this, where just the same events at every instant do actually take place, still the judgments that we make about our world are not actually true or false with reference to that world, for we mean this world, not that one, when we judge. Why are not John’s Thomas and the real Thomas related like this world and that second world in distant space? Why are not both like the relation of A’s conceived phantom room and B’s real room? Nothing of either real room is ever present to the other. Each prisoner can make true or false judgments if at all, then, only about the pictures on his wall; but neither has even the suggestion that could lead him to make a blunder about the other’s real room, of which he has and can have not the faintest idea.
One reason why we fail to see at once this fact lies in the constant tendency to regard the matter from the point of view of a third person, instead of from the point of view that we still implicitly attrib¬ ute to A and B themselves. If A could get outside of his room once and see B’s room, then he could say: “ My picture was a good one,” or the reverse. But, in the supposed case, he not only never sees B’s room, but he never sees anything but his own pic¬ tures, never gets out of his room at all for any pur¬ pose. Hence, his sole objects of assertion being his pictures, he is innocent of any power to err about B’s room as it is in itself, even as the man born blind 416 THE religious aspect of philosophy is innocent of any power to err about the relations of colors.
Now this relation of A and B, as they were sup¬ posed to dwell in their perpetual imprisonment, is essentially like the relation that we previously pos¬ tulated between two independent subjects. If I can¬ not have you in my thought at all, but only a picture produced by you, I am in respect to you like A con¬ fined to the pictures produced from B’s room. How¬ ever much I may fancy that I am talking of you, I am really talking about my idea of you, which for me can have no relation whatever to the real you. And so John and Thomas remain shut up in their prisons. Each thinks of his phantom of the other. Only a third person, who included them both, who in fact treated them as, in the Faust-Epilogue, the Pater Seraphicus treats the selige Knaben (Er nimmt sie in sich, says the stage direction) — only such an inclusive thought could compare the phan¬ toms with the real, and only in him, not in them¬ selves, would John and Thomas have any ideas of each other at all, true or false.
This result is foreign to our every-day thought, be¬ cause this every-day thought really makes innocent use of two contradictory views of the relations of conscious beings. On the one hand we regard them as utterly remote from one another, as what Pro¬ fessor Clifford called ejects; and then we speak of them as if the thoughts of one could as such become thoughts of the other, or even as if one of them could as an independent being still become object in the thought of the other. No wonder that, with THE POSSIBILITY OF ERROR such contradictory assumptions as to the nature of our relations to our neighbors, we find it very easy to make absurd statements about the meaning of o error. The contradiction of common sense has in fact just here much to do with the ethical illusion that we called the illusion of selfishness. To clear up this point will be useful to us, therefore, in more ways than one.
VI.