theory namely that "among primitive peoples religion reflects
collective feeling and collective thinking." Dionysius, the god of the
Greek mysteries, is according to her interpretation a product of the
group consciousness.

The mystery-god arises out of those instincts, emotions,
desires which attend and express life; but these emotions,
desires, instincts, in so far as they are religious, are at the
outset rather of a group than of individual consciousness....
It is a necessary and most important corollary to this
doctrine, that the form taken by the divinity reflects the
social structure of the group to which the divinity belongs.
Dionysius is the Son of his Mother because he issues from a
matrilinear group.[19]

This whole study is, in fact, merely an application of Durkheim's
conception of "collective representations."

Robert H. Lowie, in his recent volume, _Primitive Society_, refers to
"ethnologists and other historians," but at the same time asks: "What
kind of an historian shall the ethnologist be?"

He answers the question by saying that, "If there are laws of social
evolution, he [the ethnologist] must assuredly discover them," but at
any rate, and first of all, "his duty is to ascertain the course
civilization has _actually_ followed.... To strive for the ideals of
another branch of knowledge may be positively pernicious, for it can
easily lead to that factitious simplification which means
falsification."

In other words, ethnology, like history, seeks to tell what actually
happened. It is bound to avoid abstraction, "over-simplification," and
formulae, and these are the ideals of another kind of scientific
procedure. As a matter of fact, however, ethnology, even when it has
attempted nothing more than a description of the existing cultures of
primitive peoples, their present distribution and the order of their
succession, has not freed itself wholly from the influence of abstract
considerations. Theoretical problems inevitably arise for the solution
of which it is necessary to go to psychology and sociology. One of the
questions that has arisen in the study, particularly the comparative
study, of cultures is: how far any existing cultural trait is borrowed
and how far it is to be regarded as of independent origin.

In the historical reconstruction of culture the phenomena of
distribution play, indeed, an extraordinary part. If a trait
occu