remarkable evidence in support of Herbert Spencer's exposition of the
law of religious development. To comprehend this general law, we
must, however, go back to the origin of religious beliefs. One should
bear in mind that, from a sociological point of view, it is no more
correct to speak of the existing ancestor-cult in Japan as
"primitive," than it would be to speak of the domestic cult of the
Athenians in the time of Pericles as "primitive." No persistent form
of ancestor-worship is primitive; and every established domestic cult
has been developed out of some irregular and non-domestic
family-cult, which, again, must have grown out of still more ancient
funeral-rites.

Our knowledge of ancestor-worship, as regards the early European
civilizations, cannot be said to extend to the primitive form of the
cult. In the case of the Greeks and the Romans, our knowledge of the
subject dates from a period at which a domestic religion had long
been established; and we have documentary evidence as to the
character of that religion. But of the earlier cult that must have
preceded the home-worship, we have little testimony; and we can
surmise its nature only by study of the natural history of
ancestor-worship among peoples not yet arrived at a state of
civilization. The true domestic cult begins with a settled
civilization. Now when the Japanese race first established itself in
Japan, it does not appear to have [24] brought with it any
civilization of the kind which we would call settled, nor any
well-developed ancestor-cult. The cult certainly existed; but its
ceremonies would seem to have been irregularly performed at graves
only. The domestic cult proper may not have been established until
about the eighth century, when the spirit-tablet is supposed to have
been introduced from China. The earliest ancestor-cult, as we shall