is overlooked we are liable to fall into the error of imagining that there was a time when man did not distinguish between the natural and the supernatural. This error may take the form of saying either that to primitive man nothing was supernatural or that everything was supernatural. Nothing, it may be said, was supernatural, for, as in a dream the most incongruous and impossible incidents are accepted by the dreamer as perfectly natural, and are only recognised as surprising and impossible when we wake and reflect on them, so events which are seen by civilised man to be incredible and impossible are to primitive man matters of everyday occurrence, and are perfectly natural. On the other hand,

it is said that, when no natural laws are known there can be no natural and necessary sequences of events, and everything therefore is supernatural. According to this view, primitive man lived in a state of perpetual surprise: he marvelled every time he found that water was wet, he was racked with anxiety every time he went to bed lest the sun should not rise the next day, and he was filled with grateful astonishment when he found that it did rise. But this view, sufficiently improbable in itself, must be rejected for two reasons: first, the very animals have, for instance, their lairs and their customary drinking-places to which they resort in full confidence that they will find them where they were before; and we cannot rate the intelligence of primitive man so far below that of the animals, as to imagine that he was ever in doubt whether, for instance, water would slake his thirst, or food appease his appetite. Next, it is a fact of psychology that the native tendency of the human mind to believe that what has once happened will happen again is so strong that, until experience has corrected it, a single occurrence is sufficient to create an expectation of recurrence: the child to whom you have given sweetmeats once, fully expects sweetmeats from you at your next meeting.

We may then regard it as certain that from the beginning there were some sequences of phenomena, some laws which man had observed, and the occurrence of which he took as a matter of course and regarded as natural. Or putting ourselves at the practical point of view-the only point of view which could exist for primitive man in his

strenuous and unrelaxing struggle for existence-we may say that he discovered early how to set going certain portions of the mechanism of nature to further his own private ends, and that he felt neither surprise nor gratitude when the machinery produced its usual results. It was when the machinery did not produce its usual results that he was astonished-when it produced nothing or produced something the opposite of what he expected-when, for instance, the cool water which aforetimes had refreshed his limbs gave him, in his heated condition, erysipelas. And as at the present day man takes to himself the credit of his good actions and throws the blame of the bad on circumstances over which he had no control-so we may be sure that primitive man took to himself the credit of his successful attempts to work the mechanism of nature for his own advantage, but when the machinery did not work he ascribed the fault to some overruling, supernatural power. In fine, where the natural ended, the supernatural began. Laws on which man could count and sequences which he habitually initiated and controlled were natural. It was the violation of these sequences and the frustration of his expectations by which the belief in supernatural power was not created but was first called forth.1 That this was the first and earliest way in which man's attention was directed to the supernatural is probable, because his earliest inductions were necessarily framed on a narrow basis of experience, and consequently must soon have broken down. He must therefore from the beginning have been brought to confront a mysterious power which was beyond both his calculation and his control. In the next place, the shock of surprise with which he witnessed the violation of his expectations

1 Since writing the above, I find Waitz says (Introduction to Anthropology, p. 368) "that which regularly and periodically recurs passes by unheeded, because, being expected and anticipated, he (primitive man) is not obstructed in his path"; and that Major Ellis (Tshi-speaking Peoples, p. 21), quoting this passage from Waitz, says: "Hence the rising and the setting of the sun and moon, the periodical recurrence of the latter, the succession of day by night, etc., have excited no speculation in the mind of the Negro of the Gold Coast. None of the heavenly bodies are worshipped; they are too distant to be selected as objects of veneration; and the very regularity of their appearance impresses him less than the evidences of power and motion exhibited by rivers, the sea, storms, landslips, etc."

was as great as that with which civilised man would witness the unaccountable suspension or inversion of what he considered a law of nature; for the tenacity with which a belief is held does not vary with the reasonableness of the belief or the amount of evidence for it; but, on the contrary, those people are usually most confident in their opinions who have the least reason to be so. Again, it will hardly be doubted that, when primitive man found his most reasonable and justifiable expectations (as they appeared to him) frustrated in a manner for which he could not account or find any assignable cause, the feeling thus aroused in him would be that which men have always experienced when they have found themselves confronted by what they deemed to be supernatural. At all times the supernatural has been the miraculous, and the essence of miracle has been thought to be the violation of natural law. Even where there is no violation of natural laws, men may be profoundly impressed with the conviction that they are in the hands of an inscrutable, overruling, and supernatural power. To awaken this conviction it is only necessary that their "reasonable " expectations should be disappointed in some striking way, as, for instance, by the triumph of the ungodly or the undeserved suffering of the innocent. In fine, to be convinced of the existence of the supernatural, it is sufficient that man should realise his helplessness.

When, however, primitive man realised that he was in the hands, at anyrate occasionally, of a mysterious and supernatural power, it was inevitable that he should cast about for some means of entering into satisfactory relations with that power. We shall have to consider hereafter what were the conditions which governed and directed his first attempts; here, however, we may note two things. The first is, that it is not always necessarily to the disadvantage but sometimes to the advantage of man that his reasonable expectations may be miraculously disappointed-in other words, the belief in the supernatural is not necessarily or exclusively the outcome of fear. Thus "tradition says that the people of Cape Coast first discovered the existence of Djwi-j'ahnu [the local deity of Connor's Hill] from the great loss which the Ashantis experienced at this spot

during their attack on Cape Coast on the 11th of July 1824. The slaughter was so great and the repulse of the Ashantis so complete, that the Fantis, accustomed to see their foes carry everything before them, attributed the unusual result of the engagement to the assistance of a powerful local god," and they set up a cult accordingly.1 The Kaffirs of Natal make thankofferings and express gratitude to the spirits for blessings received thus: "This kraal of yours is good; you have made it great. I see around me many children; you have given me them. You have given me many cattle. You have blessed me greatly. Every year I wish to be thus blessed. Make right everything at the kraal. I do not wish any omens to come. Grant that no one may be sick all the year." In fine, as Mr. Clodd says, in primitive religion there is "an adoration of the great and bountiful as well as a sense of the maleficent and fateful." 3

The second thing to notice is that, as it was owing to man's physical helplessness in his competition with his animal rivals that he was compelled to exercise his intellect in order to survive in the struggle for existence, so it was his intellectual helplessness in grappling with the forces of nature which led him into the way of religion; and as it was his intellectual faculties which gave him the victory over his animal competitors, so it was the strength drawn by him from his religious beliefs that gave him the courage to face and conquer the mysterious forces which beset him.

Assuming, then, that from the beginning man was compelled from time to time to recognise the existence of a supernatural power intervening unaccountably in his affairs and exercising a mysterious control over his destinies, we have yet to inquire how he came to ascribe this supernatural power to a spirit having affinity with his own. Now, savages all the world over believe that not only animals and plants but inanimate things also possess life; and the inference that whatever moves has life, though mistaken, is so natural, that we have no difficulty in understanding how the gliding stream and the leaping flame may be considered to be veritably living things. But savages also regard motionless 1 Ellis, Tshi-speaking Peoples, 40. 2 Shooter, Kafirs of Natal, 166. 3 Clodd, Myths and Dreams, 114.

:

objects as possessing life; and this, too, is not hard to understand the savage who falls and cuts himself on a jagged rock ascribes the wound to the action of the rock, which he therefore regards as a living thing. In this case there is actual physical motion, though the motion is the man's. In other cases the mere movement of attention" by which an object was brought within the field of consciousness would suffice to lend the thing that appearance of activity which alone was required to make it a thing of life. Then, by a later process of reasoning, all things would be credited with life; we talk of a rock "growing" (i.e. projecting) out of the ground, the peasant believes that stones actually "grow" (i.e. increase), and as it is from the earth that all things proceed, the earth must be the source of all life, and therefore herself the living mother. In fine, all changes whatever in the universe may be divided into two classes, those which are initiated by man and those which are not; and it was inevitable from the first that man should believe the source and cause of the one class to be Will, as he knew it to be the cause and source of the other class of changes.

All the many movements, then, and changes which are perpetually taking place in the world of things, were explained by primitive man on the theory that every object which had activity enough to affect him in any way was animated by a life and will like his own-in a word (Dr. Tylor's word), on the theory of animism. But the activity of natural phenomena as thus explained neither proceeds from nor implies nor accounts for belief in the supernatural. This may easily be made clear. Primitive man's theory, his animism, consists of two parts: the facts explained and the explanation given-and in neither is anything supernatural involved. Not in the facts explained, for the never-hasting, neverresting flow of the stream, for instance, was just as familiar and must have seemed just as "natural" to primitive as to civilised man: there was nothing supernatural in such activity. But neither was the cause to which he ascribed this activity supernatural; for the cause assigned was a will which, being exactly like his own, had nothing unusual, mysterious, or supernatural about it; for we must remember two things, first, that for the average mind "explanation"