If, then, we ask the question: What is the real external world? it is clear that we cannot answer it satisfactorily without taking into consideration the somewhat shifting senses of the word "real." What is the real external world to the plain man? It is the world of touch things, of objects upon which he can lay his hands. What is the real external world to the man of science? It is the world of atoms and molecules, of minuter touch things that he cannot actually touch, but which he conceives as though he could touch them.

It should be observed that the man of science has no right to deny the real world which is revealed in the experience of the plain man. In all his dealings with the things which interest him in common life, he refers to this world just as the plain man does. He sees a tree and walks towards it, and distinguishes between its real and its apparent color, its real and its apparent size. He talks about seeing things as they are, or not seeing things as they are. These distinctions in his experience of things remain even after he has come to believe in atoms and molecules.

Thus, the touch object, the tree as he feels it under his hand, may come to be regarded as the sign of the presence of those entities that science seems, at present, to regard as ultimate. Does this prevent it from being the object which has stood as the interpreter of all those diverse visual sensations that we have called different views of the tree? They are still the appearances, and it, relatively to them, is the reality. Now we find that it, in its turn, can be used as a sign of something else, can be regarded as an appearance of a reality more ultimate. It is clear, then, that the same thing may be regarded both as appearance and as reality—appearance as contrasted with one thing, and reality as contrasted with another.

But suppose one says: I do not want to know what the real external world is to this man or to that man; I want to know what the real external world is. What shall we say to such a demand?

There is a sense in which such a demand is not purely meaningless, though it may not be a very sensible demand to make. We have seen that an increase of knowledge about things compels a man to pass from the real things of common life to the real things of science, and to look upon the former as appearance. Now, a man may arbitrarily decide that he will use the word "reality" to indicate only that which can never in its turn be regarded as appearance, a reality which must remain an ultimate reality; and he may insist upon our telling him about that. How a man not a soothsayer can tell when he has come to ultimate reality, it is not easy to see.

Suppose, however, that we could give any one such information. We should then be telling him about things as they are, it is true, but his knowledge of things would not be different in kind from what it was before. The only difference between such a knowledge of things and a knowledge of things not known to be ultimate would be that, in the former case, it would be recognized that no further extension of knowledge was possible. The distinction between appearance and reality would remain just what it was in the experience of the plain man.

22. THE BUGBEAR OF THE "UNKNOWABLE."—It is very important to recognize that we must not go on talking about appearance and reality, as if our words really meant something, when we have quite turned our backs upon our experience of appearances and the realities which they represent.

That appearances and realities are connected we know very well, for we perceive them to be connected. What we see, we can touch. And we not only know that appearances and realities are connected, but we know with much detail what appearances are to be taken as signs of what realities. The visual experience which I call the house as seen from a distance I never think of taking for a representative of the hat which I hold in my hand. This visual experience I refer to its own appropriate touch thing, and not to another. If what looks like a beefsteak could really be a fork or a mountain or a kitten indifferently,—but I must not even finish the sentence, for the words "look like" and "could really be" lose all significance when we loosen the bond between appearances and the realities to which they are properly referred.

Each appearance, then, must be referred to some particular real thing and not to any other. This is true of the appearances which we recognize as such in common life, and it is equally true of the appearances recognized as such in science. The pen which I feel between my fingers I may regard as appearance and refer to a swarm of moving atoms. But it would be silly for me to refer it to atoms "in general." The reality to which I refer the appearance in question is a particular group of atoms existing at a particular point in space. The chemist never supposes that the atoms within the walls of his test-tube are identical with those in the vial on the shelf. Neither in common life nor in science would the distinction between appearances and real things be of the smallest service were it not possible to distinguish between this appearance and that, and this reality and that, and to refer each appearance to its appropriate reality. Indeed, it is inconceivable that, under such circumstances, the distinction should have been drawn at all.

These points ought to be strongly insisted upon, for we find certain philosophic writers falling constantly into a very curious abuse of the distinction and making much capital of it. It is argued that what we see, what we touch, what we conceive as a result of scientific observation and reflection—all is, in the last analysis, material which is given us in sensation. The various senses furnish us with different classes of sensations; we work these up into certain complexes. But sensations are only the impressions which something outside of us makes upon us. Hence, although we seem to ourselves to know the external world as it is, our knowledge can never extend beyond the impressions made upon us. Thus, we are absolutely shut up to appearances, and can know nothing about the reality to which they must be referred.