On the other hand, the aims and problems arising out of the philosophical disciplines help to give direction to psychology and thus influence it. The interest in logical, ethical, epistemological, and metaphysical problems arouses interest in certain phases of mind and leads to a psychological study of the same. (Perhaps we can partly explain the trend away from psychophysics in this way also.) It fastens the attention on processes of mind which the natural scientist is apt to ignore because he can find no physical antecedents for them. Such philosophical study also acts as a safeguard against a false mental atomism and tends to keep in view the unity of mind. As these remarks apply with even greater force to metaphysics, the especial bugbear of some scientists, it may not be out of place to discuss this point a little further.
The relation of psychology to metaphysics is not to be conceived in the old-fashioned sense of an a priori construction of the facts of psychology from metaphysical principles. If the thing could be done, if the facts discovered by empirical psychology could be deduced from a few fundamental principles, without any regard to experience, there would be no objection whatever to doing it. But no system of metaphysics exists that can shake out of its sleeve all the mental phenomena with which we become acquainted through observation, and so far as I know no system has ever attempted such a thing. But if dependence on metaphysics means that psychology must start out with some broad assumptions or general principles, then psychology, like every other science, is metaphysical. To refuse to start out with any epistemological and metaphysical assumptions is not to start out at all. The only question here is with what assumptions to start out, and most of the trouble is due to the fact that one man's assumptions are gall and wormwood to another. And often the psychologist is not conscious of having any assumptions, or his assumptions seem so self-evident to him that he takes them for what he is pleased to call facts, while his colleague's presuppositions strike him as unwarranted metaphysical fictions. So enamored are we of our own pet notions! If, finally, the introduction of hypotheses makes psychology metaphysical, psychology cannot escape metaphysics; indeed, no
