V. CRITICAL NOTICES. Volkerpsychologie. Eine Untersuchung der Entwicklungsgesetze von Sprache, Mythus und Sitte. Von WILHELM WUNDT. Erster Band: Die Sprache. ASSUREDLY the theme of this first volume of Prof. Wundt's monumental work is profoundly interesting. We commenced the study of it with a very real enthusiasm. Here at last was a systematic treatment of linguistic material from the psycho- logical point of view. Here we should find a vast array of facts, countless and diverse, culled from all possible sources, com- pared with one another, articulated upon the continuous thread of psycho-genetic explanation. Instead of mere disjecta membra, viewed from the outside, instead of the empirical classifications of the comparative philologist, we should have exhibited to us the internal mechanism, the causal connexions ; linguistic forms would be shown to be the results of mental process ; a new insight into mental process would be gained from the comprehensive study of linguistic forms. Nothing less, we imagined, could satisfy the psychologist bent on giving a systematic account of the evolution of language than a survey of all possible means of expression, including the language of signs. And since the evidence for psycho-genetic theory is largely to be found in comparative philo- logy, the marshalling of linguistic facts should go hand in hand with theoretical exposition. We expected that Prof. Wundt's work would be, in a very real sense, at once a philological treatise for psychologists, and a psychological treatise for students of linguistics. Such a book would be the labour of a life-time. Wundt's work is but an incident in one of the busiest learned careers on record. There is far too much theory, and too little fact to please us. The facts are quoted merely as illustrations of theories, not as proofs of them, and no one but a competent philo- logist could judge whether the illustrations are fairly chosen or not. The same instances from the same languages are apt to recur wearisomely often. The references to primitive languages are much too scanty and vague. For the partial disappointment we are bound to confess to, we may be to blame. We may have pitched our expectations too high ; and assuredly an author has a
