But if we wish to work in a really psychological way, we shall want to know the _meaning_ of psychological phenomena. After learning the kinds of steel the various parts of a locomotive are made of, and from what ironworks and mines they come, we do not really know anything about the locomotive's _function_, that is to say, its _meaning_. But "function" as conceived by modern science is by no means solely a causal concept; it is especially a final or "teleological" one. For it is utterly impossible to consider the soul from the causal view-point only; we are obliged to consider it also from the final point of view. As Dr. Mitchell also points out, it is impossible for us to think of the causal determination conjointly with a final connection. That would be an obvious contradiction. But our theory of cognition does not need to remain on a pre-Kantian level. It is well known that Kant showed very clearly that the mechanistic and the teleological view-points are not _constituent_ (objective) principles, in some degree qualities of the object, but that they are purely _regulative_ (subjective) principles of thought, and as such they are not mutually inconsistent. I can, for example, easily conceive the following thesis and antithesis:-- _Thesis_: Everything came into existence according to mechanistic laws. _Antithesis_: Some things did not come into existence according to mechanistic laws only. Kant says to this: Reason cannot prove either of these principles, because _a priori_ purely empirical laws of nature cannot give us a determinative principle regarding the potentiality of things. As a matter of fact, modern physics has necessarily been converted from the idea of pure mechanism to the final concept of the conservation of energy, because the mechanistic explanation only recognises reversible processes, whereas the actual truth is that the process of nature is irreversible. This fact led to the concept of an energy that tends towards relief of tension, and therewith also towards a definite final state. Obviously, I consider both these points of view necessary, the causal as well as the final, but would at the same time lay stress upon the fact that since Kant's time we have come to know that the two view-points are not antagonistic if they are regarded as regulative principles of thought, and not as constituent principles of the process of nature itself.