which was to be predicted. But because we can never get at all of the factors operating in the mind of any individual, both of these methods fail. Bergson, then—and in a broader sense Bradley also—states that prediction of conduct means a performance by proxy of the act to be prophesied. Bergson is pessimistic as to the feasibility of this method, but the writer, assuming the idealist's attitude, believes that this method is practical in view of the fact that the universe is one, and specifically, because similar ideas are commonly of simultaneous occurrence. Moreover, if the general conditions and elements of character are known in any individual case, one ought to be able to predict the drift of action under definite circumstances.
A scientific philosophy has a task beyond the mere recording and classifying of data furnished by the special sciences. It must frame and hand over to the sciences for further investigation certain general questions or hypotheses, suggested by universal, practical demands or values, and thus affect somewhat the course of scientific procedure. To accomplish this end philosophy requires methods and criteria by which to single out the significant facts; it must have facts, and not abstract notions, as its principles, and must offer suggestions as to the utilization of these facts. The method suggested is that of material abstraction in contrast to mere ideal abstraction. To get at the individuals composing a system, elements which are actualities, and not mere conceptual devices, must be separated from each other. For this work, the logical method of difference must be supplemented by such auxiliary scientific methods as segregation, neutralization, differentiation, etc. Since the elements are thus independent and individual, the construction of the system consists in an identification of these elements. Intuitions, that is, the universally accepted facts of experience, must be used as principles rather than conceptions, both because of the demand of induction for the individual, or particular fact, and of deduction for a progression from the known to the unknown. Systematization, then, consists in the identification of intuitive elements of representation. Granted that this is a good theory, can science take it and use it? The fact of sensorial symbolism makes this possible. Every sense organ has a system of symbolisms, its own particular set of representations for describing the world. Are there any of these symbolic systems which can be substituted for others? The kinæsthetic and affective systems can be thus substituted, and they are the most useful because they alone are susceptible of quantitative expression. The intuitive element of the first system is the material point, the least conceivable center of strain or movement, with its two simple functions, those of gravity and of electricity. The intuitive elements of the second system are the elementary tendencies to fuse and to discriminate. These elements can be substituted for the conventionally understood elements of the various sciences. For instance, in biology, the cell can be expressed as a unit
