Magic, until recently, was somewhat neglected by those who treated of savage ideas. In Sir E. B. Tylor’s _Primitive Culture_ only one chapter is given to Magic, against seven to Animism (belief in the agency of spirits). In Spencer’s _Sociology_, Part I. is almost wholly devoted to the genesis and development of Animism, without a single chapter on Magic. The importance of early Animism became such an obsession, that travellers observed and reported upon it wherever they went, making only casual references to Magic—much to our loss. The tradition of this way of thinking seems to run back to Hume’s _Natural History of Religion_, where he traces the development of religion from a primitive belief that all natural activities are like our own, and that everything is possessed and actuated by a spirit. This idea was adopted by Comte, and elaborated in his celebrated law of the three stages of the explanation of Nature as determining the growth of human culture: Fetichism, which ascribes all causation to the particular will of each object, and which by generalisation leads through polytheism to monotheism; Metaphysics, which, giving up the notion of personal will, attributes the activities of things to abstract forces; and Positivism, which, discarding the variety of forces that can never be known, turns to the exact description of the order of phenomena. Mill, in turn, adopted these doctrines from Comte, and gave them currency amongst us, as part of the extraordinary influence which for many years he exerted upon all our thoughts. Hence the priority of Animism to every other theory of things seemed at that time a matter of course. Recently there have been signs that this conception of primitive thought is giving way to another, namely, that Animism was preceded by Magic. Sir J. G. Frazer puts it, that the idea of Magic is simpler than that of Animism; that Magic is found in full force amongst people whose Animism is feebly developed, and that its beliefs are more uniform throughout the world.[101] Substantially—with qualifications that will appear hereafter—this view of the matter is here accepted. When one comes to argue it, to produce the primitive facts is, of course, impossible. It must even be admitted that such evidence as we have amongst the few facts collected of late years concerning ancient races of men, gives the earlier date to animistic ideas; for if some of the cave-paintings of Aurignacian origin, in which, for example, wild cattle are shown pierced with arrows, may be interpreted as of magical significance; on the other hand, the burial at Le Moustier, still more ancient, shows a regard for the corpse, in the disposal of it and in the things left with it, such as usually goes with the ghost-theory. These discoveries take us back some thousands of years before Menes; but probably leave us far from the beginning of human ideas concerning the supernatural.