The continued use of the distinction gets its plausibility from the fact that the terms Ilse't have clear meanings within the original limits of their application. A thinker is therefore tempted to assume that the terms still have defi nite significance in their new sphere of applica tion. An example of the change of meaning in the word 'really' as we pass 'from popular to chemical usage will show that it is unsafe to treat the word 'real' as if it expressed an unchanging. definite positive conception whenever used. The ordinary man of affairs would say unhesitatingly that charcoals and diamonds are really different. The analytical chemist would he tempted to say that they are really identical. There is no contra diction in these two because 'really' for the layman means for the ordinary pur poses of life,' and for the chemist it means 'from the point of view of chemical analysis.' Now. if the word 'really' can change so much when we are still within the limits of experience, is it not well to ask ourselves what it. means when we go beyond the limits of actual and possible experi ence and say that. apparently things are as we tied them in our experience, but what they really are we never shall know? 'Really' here ha- no positive meaning; at least it has no practical meaning. It seems merely to serve notice of an attempt to carry out a practical distinction when there is no longer any possibility of any practical advantage from it. But is there not a theoretical advantage? Does it not clarify thought to have it thrust upon our attention that we can know only what experience reveals? Undoubtedly; but does it not serve to confuse thought to designate what we cannot know as 'reality' and to call what we can know 'mere appearance'? Does not this terminology assume that there is something which we cannot know? What warrants this assump tion? To say that we cannot know reality would of cour,c be justifiable if we knew that there is such a reality. and yet knew that we never can know anything about it except that it exists. But if we never can know anything whatever about it, why call it `rcality'—a term which con notes value? If all that is meant is that if there is anything which can never present itself in ex perience, we can never know it, then this meaning is not clearly expressed by saying that we can know appearance, but not reality. For whether we will or not, the popular idea of reality as something better, more secure, more valuable than appearance clings to the word, and by con trast depreciates the conception of appearance. But we may go further, and say that there is no justification in assuming the possibility of such an inexperienceable object. The assumption has no meaning. All of our conceptions are obtained from experience. and all of our words are words that have significance only when applied to ob jects of experience or to objects conceived of as experienceable. For instance, the word 'exist ence' has a very definite nicAing. To exist is to be part of experience, or to be in some way or other related to a part. of experience. Horses exist as parts of our experience. Any as yet invisible star exists as related in sonic way to what we do know. And 'to be related' is also a term which has meaning only when used of cer tain connections in experience. The relation of cause and effect, for example. is a definite sort of conne•tion recognized as existing between one experienced object and another. We can extend it, to connections between objects not yet experienced, provided we think of them as possible objects of experience. But when one of the terms of the predicated relation is asserted to be inexperienceable, then the relation itself has no meaning.
Thus we see that to assert existence of any thing that is at the same time said to be not a possible ohjeet of experience is simply to use words without meaning. Agnosticism, or the doctrine that we cannot know reality, and skep ticism, or the doctrine that we must ever doubt whether we can know- reality, are therefore both meaningless doctrines, because Ter/My as used in these doctrines has no meaning. The theory of phenomenalism is open to the same objection if it is an assertion that knowledge is limited to phenomena or appearances, with an implication that it cannot comprehend reality. But if all that phenomenalism stands for is the view that all reality is to be defined as either actual or possible presentation to consciousness, or any discoverable relation between such presentations, and that beyond such presentation there is noth ing conceivable except possible presentations. then
phenomenalism is tenable.
Finding, then, that a reality opposed to all ap• pearanee is unthinkable, let us examine the view that all knowledge is of appearances. Atay there not be a skepticism pos-sihle lucre? The argument for it could be stated as follows: "Every sense may deceive; thought may be fallacious; but as sense and thought arc our instruments of knowledge, we can never know- anything." It takes but little logical acumen to detect the fal lacy of such an argument. A sense deceives us when it misleads us in our expectations with regard to other senses. We can say that it misleads only when we have a test for ity. accu racy or inaccuracy; this test, we have in the con sistency of our sense-perceptions. it is not justi fiable to say that each sense may deceive, and that therefore all, when consistent and harmoni ous in their presentations, may deceive. This is an example of the fallacy of composition. (See FALLACY.) This is what in substance was argued by the tropes of Agrippa, mentioned above. Against such sophistry we must say that before we can impeach any perception or judgment we must know that it fails to conform to the stand ard. Thus knowledge is presupposed in all acknowledged ignorance, and skepticism is an acknowledged ignorance. Thus far we have shown that a thinker must know at least something in order to indulge himself in a reasonable doubt, and that what he knows is a part at least of what finds itself in his experience. Universal skepticism and agnosticism have been shown to be untenable. Experience there undoubtedly is, and within that experience there undoubtedly is knowledge of reality, if reality is to have any conceivable meaning. Now the business of the epistemologist is to investigate this knowledge that actually exists.
The epistemologist can do this successfully only by following scientific method. This means that lie must compare one act of knowledge with an other, in order to discover what knowledge really is. He must not start out with smile theory of his own and make the facts adjust themselves to this theory.
Now, one fact that is characteristic of all valu able knowledge is the fact of generalization. Every mare thinker is constantly generalizing his knowledge. The same is true of scientific thinkers. No science is merely descriptive in the sense that it confines itself to the ascertainment and the bare statement of facts and relations actually experienced at any one time. \Vhether justifiably or not. the scientist always assumes that if he can only state facts as they really are at any one time. lie has a statement that is valid for all time when like conditions prevail. For in stance, the discovery that under a definite atmos pheric pressure pure water freezes when the ther mometer stands at 32° F. is, for plain man and scientist alike, of significance not only as reveal ing an isolated fact of experience true at the time of the discovery, but as giving a 'law.' This law is the further fact that under like atmospheric pressure equally pure water freezes when in a similarly constructed thermometer the mercury registers 32'. Every statement of known fact can thus be generalized, and if when generalized it provas untrue, then the assumption of knowl edge at the outset was unjustifiable. This is why earful scientists demand that an experi ment should be repeated for verification. To verify an experiment means to ascertain its cor rectness. Obviously, however, one cannot go back in time and live over again the identical experi ence in order to verify the descriptive statement of it. All that can be done is to put one's self under like conditions and then to see whether like statements hold good of like facts. Even if it should be tound that like statements are true of the second experience, the previous statement is not verified unless it contained more than a description of the previous experience as an iso lated fact. This surplusage of content in a state ment beyond bare description of an isolated fact is technically called its universality.