KNOWLEDGE. 566 KNOWLEDGE. his own experience has furnished him with this knowledge. Onee question or deny the reason assigned, and the skeptical conclusion can no longer be maintained bv the skeptic without dog- matic assertion of the reason as self-evident or without the jiroduction of a new reason, which, when disputed, shares the fate of its predecessor. In other words. ui)on examination it is clear that skepticism cannot recommend itself as an escape from the danger of dogmatic assertion or of an infinite regress of proofs: for we cannot accept skepticism without proof, and if the skeptic is right, proof involves either unwarranted dog- matism or impossible regress of argument 1o infinity. But it may >e said that this inconsistency and illogicality of skepticism is no refutation of skep- ticism. Perhaps not : but it makes it necessary for the skeptic to bring forward other reasons for the mifaith that is in liim; and when he does, these new reasons can be met in the same way. The skeptic nuist assume the validity of his argu- ment in order by argument to convict his oppo- nent of error or of unreasonableness. Hut. again, it may be asked: Why may not the skeptic hold to his skepticism without attempting to establish his position or to dislodge his opponent? The answer is that to cherish any view without proof is to be dogmatic, and tliis is what the skeptic is. above all things, anxious to avoid. But still, again, it may be urged that instead of believing in the impossibility of knowledge, one may mere- ly doubt its possibility. It is often said by writers against ske])ticism that a skeptical skep- ticism is unassailable. In one sense it is. If a man will persist in listening to no reason and in observing no laws of thou;;ht. then no argu- ment directed against him will reach him. so long as he takes refuge beliind the bulwark of unthink- ingness. But what we can do is to convince oiir- selrrs by valid reasoning that skepticism, wheth- er dogmatic or doubting, is untenable by a rea- sonable being. That is, if we follow the laws of tlioight we cannot become ske])tical with re- gard to all our knowledge, however skeptical we may be on .some of its details. To sunnnarize our argument so far. let us say that we cannot reasonably doubt anvthing without good reason. and good reason is known reason, and known rea- son imjdies the validity of knowledge. The insight into this truth enables us to ap- preciate the skeptic's rationality perhaps l]ctter than he does himself. Whenever we find a skep- tic in philosophy we find a man who is tr'ing to reason consistently, and who in this endeavor to be consistent even goes so far as to say that all knowledge is impossible. In such a case it is better worth our while to see what his funda- mental assumptions are than to grieve or be merry over his conclusions. Now, one assumption is made by every skeptic, as by every other thinker — viz. that a Hunker must make the ob- jects of his fhoiifiht coiisisteiit. It is the demand for consistency which drives all thinkers on to their conclusions, whatever these may be. I^et us now examine this logical demand for consistency, and see what it is and what it involves. This logical demand for consistency involves, as a presupposition, that there be some object-matter in dealing with which a thinker is to be consistent. In other words, consistency is a relation, thought thus to obtain between objects. To think con- sistently is to think objects into consistency. The law of consistency, like all other laws of thought (see Louie), is a law directing the disposition which thought is to make of its objects. It is true that we maj' demand consistency in our pro- ces.ses of thought as well as in the disposition of objects. But this is because we objectify our thinking jjrocesscs. We may then compare one process with another, or one part of a process with another part, and nuiy find that they arc not consistent. We may then come to <iemand order in our. tliinking as well as order in the things we think about. But it must be under- stood that this is possible only when our own thinking is a thing we think about; i.e. only when thought becomes an object of thought. We may therefore safely say that consistency as a logical law is a demand that the objects of thoiight be made to accord with each other. It is this demand which impels the child to constitute his sensations into an ordered world of experi- ence. He tries to think them into some consistent relation with each other. Such an attempt to a large extent succeeds, but the success is slow. The work is done by means of categories (q.v. ), or principles of connectiim which the thinker rec- ognizes as obtaining between dillVrent objects. One may say tlnit the reduction of experience into consistency is nothing l)ut the ascertainment of such relations, or principles of connection, be- tween ol)jccts as make it |M3ssible to think them together without having tlicm confiict with eacli other as they appear in thought. For instance, in a certain dish on a stove 1 see one minute a clear, transparent, solid cube; the next minute in the same pan I see a clear, transparent liquid. In order to be able to think these two objects as one. which the continuity of the objects under my gaze prompts me to do. 1 must be able to think of the cold, solid cube as clianging into the liquid. Change is thus .a category, or a prin- ciple of connection recognized as existing between two objects and making them consistent with each other. In the case of our example there are other categories; e.g. those of attribute and substance: and. provided T think of the experience as my experience, those of subject and object. Xow. one pair of correlative categories that ap- pears (piite early in the development of thought is that of reality and illusion. . careful study of these will enable ns to understand the skep- tic's position. A changing object docs not de- ceive >is, provided we know enough about the order of the changes. The trouble comes when an unexpected change occurs which throws us off our calculation. For example, we are accustomed to see broken sticks, and by some sort of opera- tion we may be able to put the two broken ends together and practically make a whole stick again. But this requires time and work. We are not accistomed to seeing broken sticks straighten themselves of their own accord when they are moved. But now suppose we come upon a stick projecting out of a clear pool of water. It looks broken; i.e. there is a break in the outline of the object as seen. Sippo.se now we lift it out of the water. It becomes straight in the operation. We didn't expect this change, and we are fooled. We find, however, that by passing our hand along the stick it feels straight; that is. the character and order of our tactual sensations in this case are what they usually are when objects look straight. We dip the stick back into the water. It looks broken again. Still keeping it under water, we
