Page:Philosophical Review Volume 1.djvu/47

Rh Helmholtz's Physiologische Optik and Tonempfindungen, nearly all (indeed all, but the worthless conjectural part) of the monographs on psychometry, must — if we adhere strictly to this definition be ruled out of the domain of scientific psychology. To be sure, these treatises establish uniform relations between thoughts and feelings, as such, and observed facts of the application of stimuli to the periphery of the human body. But they are too far from the sacred centre, where the "explosions" and the "overlappings," and the "central adjustments" take place, to be entitled to the name, "natural science," in the highest meaning of the phrase. Or, at least, their value is to be estimated by the inducement they afford the psychologist to frame diagrammatic, or verbal, representations of precisely what it is, in which such "explosions," "adjustments," and "overlappings" consist.

Let it not be supposed that Professor James himself remains faithful to this extremely contracted conception of psychology as a natural science. On the contrary, by far the larger and better portions of his two large volumes are those, where he feels irresistibly impelled to introduce some metaphysics, or to leave out nearly all the conjectural cerebral physiology. For example, the chapters in the second volume which treat of Perception by the Senses are, from the scientific point of view, much the most valuable part of the entire work. Here the author is at his best. He introduces freely his wide acquaintance with modern analytic and experimental psychology. His descriptive science is here — as everywhere — admirable. He indulges, but not immoderately, in speculation — not omitting all help from semi-metaphysical assumptions. And, above all, the amount of cerebral physiology, which is introduced to explain the rise and change of the states of consciousness, is relatively so small that it really is not worth taking into the account.

The attempt to establish psychology as a natural science upon such an extremely tenuous and cloudy foundation as our present or prospective knowledge of cerebral "explosions" and "overlappings" is doomed to failure from the very