I think I have defined totemism, {71} and the reader may consult Mr.
Frazer's work on the subject, or Mr. MacLennan's essays, or 'Totemism' in
the Encyclopaedia Britannica. However, I shall define totemism once
more. It is a state of society and cult, found most fully developed in
Australia and North America, in which sets of persons, believing
themselves to be akin by blood, call each such set by the name of some
plant, beast, or other class of objects in nature. One kin may be
wolves, another bears, another cranes, and so on. Each kin derives its
kin-name from its beast, plant, or what not; pays to it more or less
respect, usually abstains from killing, eating, or using it (except in
occasional sacrifices); is apt to claim descent from or relationship with
it, and sometimes uses its effigy on memorial pillars, carved pillars
outside huts, tattooed on the skin, and perhaps in other ways not known
to me. In Australia and North America, where rules are strict, a man may
not marry a woman of his own totem; and kinship is counted through
mothers in many, but not in all, cases. Where all these notes are
combined we have totemism. It is plain that two or three notes of it may
survive where the others have perished; may survive in ritual and
sacrifice, {72a} and in bestial or semi-bestial gods of certain nomes, or
districts, in ancient Egypt; {72b} in Pictish names; {72c} in claims of
descent from beasts, or gods in the shape of beasts; in the animals
sacred to gods, as Apollo or Artemis, and so on. Such survivals are
possible enough in evolution, but the evidence needs careful examination.
Animal attributes and symbols and names in religion are not necessarily
totemistic. Mr. Max Muller asks if 'any Egyptologists have adopted' the
totem theory. He is apparently oblivious of Professor Sayce's reference
to a prehistoric age, 'when the religious creed of Egypt was still
totemism.'
Dr. Codrington is next cited for the apparent absence of totemism in the
Solomon Islands and Polynesia, and Professor Oldenberg as denying that
'animal names of persons and clans [necessarily?] imply totemism.' Who
says that they do? 'Clan Chattan,' with its cat crest, may be based, not
on a totem, but on a popular etymology. Animal names of _individuals_
have nothing to do with totems. A man has no business to write on
totemism if he does not know these facts.