victory to initiative, to aggression, to assertion of the will, is a tacit recognition of the efficacy of power. It is important, then, as evidence that power is the pre-supposition which underlies, not merely the wonder-working of the sorcerer, but all sympathetic magic. But the cogency of this kind of evidence may perhaps be called in question,—though not, I believe, with reason,—as being of too fanciful a nature. To regard magic as a formal abstract science was a mistake that has led us far astray. It must always be remembered that in magic, as in religion, we have persons acting under the stress of passion, or in a highly strung, tense, emotional state. Like religion, magic is the field, not of rational consideration, but of belief or faith. Its forms must not be mistaken for the content of its efficacy. It is to the emotions with which its formulæ are regarded by agent and victim that we must have recourse in order to understand the belief in its efficacy. The mediaeval scientist, for example, was feared by the ignorant as a sorcerer precisely because it was believed that he was able to violate the laws of causality by some mysterious power of his own, or with the aid of devils, not because the categories of similarity and identity were confused. It is not the possession of knowledge, but the supposed character of the knowledge he possesses, that clothes the sorcerer with awe and fear.
The apotheosis of ritual at the hands of anthropologists has not been altogether fortunate in its results. It is true that the recognition of the value of ritual as the most concrete kind of evidence at the disposal of the student of religions was a valuable discovery, but the consequent neglect of the psychology of the persons for whom ritual was but an instrument, and in the last resort but a distinctive mode for the adequate expression of their emotions and purposes, has created many difficulties and misunderstandings. For, indeed, to attribute the basis
