from these to build up, as it were, a revised version. It is, perhaps,
only by studying those rudimentary forms of art that are closely akin to
ritual that we come to see how utterly wrong-headed is this conception.

Take the representations of Osiris that we have just described--the
mummy rising bit by bit from his bier. Can any one maintain that art is
here a copy or imitation of reality? However "realistic" the painting,
it represents a thing imagined not actual. There never was any such
person as Osiris, and if there had been, he would certainly never, once
mummified, have risen from his tomb. There is no question of fact, and
the copy of fact, in the matter. Moreover, had there been, why should
anyone desire to make a copy of natural fact? The whole "imitation"
theory, to which, and to the element of truth it contains, we shall
later have occasion to return, errs, in fact, through supplying no
adequate motive for a widespread human energy. It is probably this lack
of motive that has led other theorizers to adopt the view that art is
idealization. Man with pardonable optimism desires, it is thought, to
improve on Nature.

* * * * *

Modern science, confronted with a problem like that of the rise of art,
no longer casts about to conjecture how art _might_ have arisen, she
examines how it actually _did_ arise. Abundant material has now been
collected from among savage peoples of an art so primitive that we
hesitate to call it art at all, and it is in these inchoate efforts
that we are able to track the secret motive springs that move the artist
now as then.

Among the Huichol Indians,[4] if the people fear a drought from the