oses
means which any one could use if he would, and ends by being credited
with a power peculiar to himself of working impossibilities. I now
wish to point out that a process exactly parallel is simultaneously
carried on by which arts beneficent to society are supposed to be
evolved. Rain-making may be taken as an art socially beneficial. The
_modus operandi_ of rain-making appears in all cases to be based on the
principle that like produces like; and to be in its {88} nature a
process which any one can carry out and which requires no mysterious
art to effect and no mysterious personal power to produce. At the same
time, as it is a proceeding which is beneficial to the tribe as a
whole, it is one in which the whole tribe, and no one tribesman in
particular, is interested. It must be carried out in the interest of
the tribe and by some one who in carrying it out acts for the tribe.
The natural representative of the tribe is the head-man of the tribe;
and, though any one might perform the simple actions necessary, and
could perform them just as well as the head-man, they tend to fall into
the hands of the head-man; and in any case the person who performs them
performs them as the representative of the tribe. The natural
inference comes in course of time to be drawn that he who alone
performs them is the man who alone can perform them; and when that
inference is drawn it becomes obvious that his personality, or the
power peculiar to him personally, is necessary if rain is to be made,
and that the acts and ceremonies through which he goes and through
which any one could go would not be efficacious, or not as efficacious,
without his personal agency and mysterious power. Hence the man who
works {89} wonders for his tribe or in the interests of his tribe, in
virtue of his personal power, does things which are impossible for the
ordinary member of the tribe.

Up to this point, in tracing the evolution of magic, we have not found
it once necessary to bring in or even to refer to any belief in the
existence of spiritual beings of any kind. So far as the necessities
of the argument are concerned, the belief in magic might have
originated in the way I have described and might have developed on the
lines suggested, in a tribe which had never so much as heard of
spirits. Of course, as a matter of fact, every tribe in which the
belief in magic is found does also believe in the existence of spirits;
animism is a stage of bel