THE CANNIBAL IN EVOLUTION 165 



would necessarily alter cannibalism between the allied 

 groups, while necessity, combined with customary habit, 

 would end in the practice of eating each other's dead, 

 rather than in a continuance of former customs. On this 

 hypothesis such habits as those of the Binbinga are per- 

 fectly intelligible ; it is seen how they came about, and, 

 by customary inertia, were continued. Although there is 

 no necessity to enter into the vexed question of totemism, 

 it may be assumed that in each familial group there was 

 already some such name badge, whatever its origin and 

 whatever it may have developed into later, when magic, 

 myth, and tradition moulded and welded the tribe into its 

 later form. There would be then a tribe of two classes, 

 mutually exogamous, each class with a totem. And if 

 such a hypothesis is admitted as possible, then on further 

 pressure of war, induced as war practically always is by 

 economic conditions, however much disguised in modern 

 times, the transition from a two-class tribe to a four- or 

 eight-class tribe actually explains itself. There would be 

 in the latter case eight totem classes, whose ancient law 

 sanctioned intermarriage with one other totem class only, 

 while any intercourse between the rest of the classes, though 

 at first conditioned mainly by jealousy, would gradually 

 become tribal morality, and a safeguard to tribal unity. 



On this hypothesis there is no compulsion to posit 

 advanced abstract notions as a driving force, a conception 

 contrary to logic and the nature of language as well as to 

 the established fact that progression is from the concrete 

 to the abstract, and not vice versa. Moreover such a theory, 

 for if it unravels the meaning of so difficult a case as that 

 of the Binbinga it becomes more than a hypothesis, is in 

 accordance with processes to-day, while it explains modern 

 sex morality as the last, but assuredly not enduring, 



