574 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1918. 



he was asked. "Certainly," he replied, "for, look you, in this place 

 is mentioned smoke, here they speak of fire, and here again they sing 

 of a car, and what is a locomotive save a car with fire and smoke?" 

 So, to prove the existence of totemism, it is not enough to point to 

 descent from a lion or to an individual name. In Africa clan-totem- 

 ism often reverts to animal names given to one chief in flattery, "O 

 thou elephant," "O thou lion among men." 



Totem is said to mean "token," implying group relationship; but 

 not blood relationship, since this would exclude plant totems, unless 

 these are all secondary. But at present there is a tendency to deprive 

 the word totem of every meaning it ever had. The totem of British 

 Columbia is a protective spirit (often not animal) seen in a vision 

 and has no relation to relationship; it is individual, not clannish. 

 An African chief, on dying, said that he would become a butterfly. 

 Straightway the butterfly became the "totem" of his clan (i. e., they 

 would not kill it). And what shall we say of totems defined as "odds 

 and ends" and "knots" (in Samoa), or the "heart of all animals" and 

 "intestines" (African Kiziba "totems") ? What is the use of calling 

 these totemic phenomena? Each is simply a case of taboo; to one 

 clan "intestines," qua taboo, became sacred ; but that is not a totem. 

 So sex totems, honorific totems, color totems, cloud totems (Austra- 

 lian) , twins as totems (Bantu Bahima) — are these totems at all ? Or 

 shall we say with Doctor Goldenweiser that, since every characteris- 

 tic of totemism is negligible, 1 there remains as totemism nothing save 

 a vague tendency for social groups to become associated "with objects 

 and symbols of emotional value," and that totemism is merely a 

 "specific socialization of emotional values" ? Would not this tenuous 

 definition apply to a Baptist church as well as to a totemic clan? 



It may not be superfluous to remind the general student that totem- 

 ism as the foundation of religion is only one of many suggested 

 foundations, not one of which by itself will uphold the burden 

 placed upon it. It was thought to be fundamental because it was 

 said to be universal. But despite Robertson Smith's great work it has 

 not been proved to be Semitic. 2 Nor has it been found among the 

 Aryans, where even in the Lupercalia it cannot be discovered. 3 In 

 Africa what is called totemism is not religious and is usually derived 



1 The " invariable characteristics " of totemism are supposed to be exogamy, taboo, re- 

 ligious veneration (totem worship), name, and descent. But none of these is a necessary 

 factor in totemism. Dr. River's " three essentials " are in typical form exogamy, descent, 

 and taboo (of totem flesh), whereas totemism may exist without any of these character- 

 istics and essentials. See " Totemism, an Analytical Study," by A. A. Goldenweiser, 

 Journal of American Folklore, 23 (1910), p. 182, 2G6, 275. 



2 What Dr. Robertson Smith showed to exist among the Semites were elements of a pos- 

 sible totemism ; but he could not show their combination. See his Religion of the Semites, 

 p. 42 f. and 287 ; and (opposed) Lyall, in JRAS, 1904, p. 589. 



3 See L. Deubner in the Archiv fiir Religionswissenschaft, 11)10, p. 481 f. For other 

 Aryan fields, see Saussaye, The Religion of the Teutons, p. 74, 98 ; and A. B. Keitb. JRAS, 

 1907, p. 939. 



