Page:Popular Science Monthly Volume 68.djvu/229

Rh only those interest us which are repeated. This arbitrary though fit selection creates the impression that these are the only recurring combinations, or that, in other words, the law of casuality or of type rules unqualifiedly. How universal or how limited is the application of these laws is therefore rather a question of our skill in selecting the constant combinations from among all those occurring than a question of objective natural phenomena.

Thus we observe the development and practise of all sciences progressing in this manner by the discovery, on the one hand, of ever more numerous individual constant combinations, and, on the other, of more and more universal laws by means of which components are brought into relation with one another, which formerly no one had attempted to bring together. Thus sciences grow in that they become complex and at the same time unified.

If now we consider the development and course of the various sciences from this point of view we shall attain a rational classification of all science by an inquiry into the extent and complexity of the combinations or complexes which they treat. Both characteristics are in a sense antagonistic. The simpler a complex is, that is, the fewer the components united in it, the more frequently will it occur; and vice versa. It will therefore be possible to classify all the sciences in this fashion by beginning with the minimum of complexity and the maximum of extent and ending with the maximum of complexity and the minimum of extent. The first science will include the most general and therefore the poorest and most meager concepts; the last the most specialized and therefore the richest.

What then are these limiting concepts? The most universal is the thing, any fraction of experience arbitrarily selected from the stream of our experiences and capable of repetition. The most specialized and the richest is the concept of human society. Between the study of things and the study of human society all the remaining sciences may be interpolated in an orderly series. The attempt to follow out this scheme leads to the following table:

This table contains an arbitrary element inasmuch as the steps assumed in it may be multiplied. Thus mechanics and physics might