w." That is, we must note, it is a system which is
spurious in our eyes, but which, to those who believed in it, was "a
statement of the rules which determine the sequence of events
throughout the world--a set of precepts which human beings observe in
order to compare their ends" (_ib._, p. 39).

The point, then, from which I wish to start is that {72} magic, as it
is now viewed by students of the science of religion, on the one hand
is a spurious system of natural law or science, and on the other a
spurious system of religion.

Our next point is that magic could not be spurious for those who
believed in it: they held that they knew some things and could do
things which ordinary people did not know and could not do; and,
whether their knowledge was of the secrets of nature or of the spirit
world, it was not in their eyes spurious.

Our third point is more difficult to explain, though it will appear not
merely obvious, but self-evident, if I succeed in explaining it. It
will facilitate the work of explanation, if you will for the moment
suppose--without considering whether the supposition is true or
not--that there was a time when no one had heard that there was such a
thing as magic. Let us further suppose that at that time man had
observed such facts as that heat produces warmth, that the young of
animals and man resemble their parents: in a word, that he had attained
more or less consciously to the idea, as a matter of observation, that
like produces like, and as a matter of practice that like may be
produced by like. Having attained to that practical idea, he will of
{73} course work it not only for all that it is worth, but for more.
That is indeed the only way he has of finding out how much it is good
for; and it is only repeated failure which will convince him that here
at length he has reached the limit, that in this particular point
things do not realise his expectations, that in this instance his
anticipation of nature has been "too previous." Until that fact has
been hammered into him, he will go on expecting and believing that in
this instance also like will produce like, when he sets it to work; and
he will be perfectly convinced that he is employing the natural and
reasonable means for attaining his end. As a matter of fact, however,
as we with our superior knowledge can see, in the first place those
means never can produce the desired effect; and next, the idea that
they can, as it withers and bef