WUNDT AND 'PURE SELF-OBSEEVATION '. WHEN a science is advancing with leaps and bounds, the problem of the methods to be employed in that science cannot be regarded as a burning question ; but when there are many workers, as in the psychological field, and when the advance of psychology appears still to be a matter of individual opinion, the question of method deserves the gravest and most minute attention. The appearance of the first volume of the translation of Wundt's Principles of Physiological Psychology (fifth edition) is, therefore, a fitting oc- casion for reviewing the problem. The "subjective" method, Wundt holds, has entirely failed because while " in natural science it is possible, under favourable conditions, to make an accurate observation without recourse to experiment, there is no such possibility in psychology" (p. 4). It is, therefore, not the speculative method, but scientific self-observa- tion two very different things which is condemned by Wundt, for, as he says, " accurate observation implies that the object of observation can be held fast by the attention " (p. 5), and that is not possible in self-observation where the contents " are at the opposite pole from permanent objects " (p. 4). Now it is abstractly possible that Wundt and others have made determined efforts at systematic self -observation and that they have, as a result, come to the conclusion that systematic self-obser- vation is impossible; but it is also abstractly possible that, since " psychological inquiries have, up to the most recent times, been undertaken solely in the interest of philosophy " (p. 2), only the speculative method has been applied, the method par excellence of philosophy and philosophers. Accordingly, it is at least abstractly possible that the introspective method, scientifically directed and applied, may prove a method which is eminently useful in psychology. With the great German thinker, and with all non-introspection- ists, I would, after some years of laborious inquiry, plead somewhat as follows. In all experiment something must be observed if that something is to be known. In all scientific work, observation is implied. All observation is subject to mental exhaustion and illusion. In all observation, too, especially if not of the simplest kind, there is in- volved memory with all its drawbacks. Now where we have merely to jot down a single figure, mal-observation and mal-recollection have little scope ; but in ordinary scientific work, even of an
