202 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



temperature rather than at another, the fact of expansion 

 upon solidification in certain substances and contraction in 

 others, and so on) which are not explicable by the theory of 

 molecules; hence the explanatory conception accounts for 

 only a part of the known facts. But the theory enables one 

 to predict certain phenomena of diffusion, surface tension, 

 and solution, which are later verified ; hence the explanatory 

 conception accounts for more than the known facts. Conse- 

 quently one is obliged to say that the observational and the 

 imaginative partly overlap, each extending beyond the 

 other. Qualitatively, the observationally given would seem 

 to have an essential advantage over the imaginatively deter- 

 mined; the facts of perception have a brute character which 

 compels one's acceptance of them. Yet the imaginatively 

 determined has a vividness which cannot be denied. The 

 suggestion has already been made that the anticipations 

 which one makes of nature are founded upon analogies drawn 

 from the character of nature as known; hence the imaginative 

 retains some of the forcefulness of the perceptual. But, 

 what is more important, the imaginative is almost always 

 vivified through the extensive use of iconic symbols; mole- 

 cules are thought of as small billiard balls, the ether as a 

 jelly, and forces as elastic tubes. This gives a clarity to 

 these notions which would be lacking from a less pictorial 

 symbolism. That the clarity may be of a pseudo-sort, and 

 hence misleading, is not to be denied. But something of this 

 kind seems to be the interpretation of the fact that the 

 explanatory entity is ordinarily more clearly known than 

 that which is to be explained. 



The emphasis upon familiarity as an attribute of ex- 

 planatory entities leads readily to the conclusion that ex- 

 planation is merely verbal, or linguistic. To explain a 

 proposition, according to this conception, is to express 

 another proposition to which the first can be reduced by 

 certain rules of transformation. Propositions are equivalent 

 if they may be substituted for one another in all contexts in 

 which they occur. Explanation is therefore essentially the 



