CONVERGENT EVOLUTION IN ETHNOLOGY 



By Robert II. Lowir 



THE (loctrint- of evolution us commonly understood meiins the diver- 

 f^ence of forms from a common point of origin, similarities of 

 structure between different species being accepted as an indication 

 of blood relationship. Biologists however ha\-e long ago abandoned the 

 notion that all resemblances must be due to a conunon origin. Thus, 

 Lankester has distinguished homologies that depend simply on the inheri- 

 tance of a common part from homologies that depend on the action of like 

 conditions on such homogenous parts, or on " parts whic-h for other reasons 

 offer a likeness of material to begin with." And Osborn has illustrated the 

 point by the example of niammalian teeth, the primiti\e three cusps common 

 to all mammals exemplifying the inheritance of common i)arts (homogeny) 

 and the eight cusps that develop or may suljsequently develop in different 

 orders in a uniform manner exemplifying Lankester's second process (homo- 

 plasy). In addition of course, there may also be convergence toward a 

 similar structure on the part of urn-elated animals, or mere analogy 

 of function without far-reaching morphological resemblances. 



In ethnology, in the study of the development of human culture, the 

 emphasis has hitherto been almost entirely on di\ergent evolution and 

 resemblances have l)een interpreted as due either to a psycholog\- common to 

 all humanity or to borrowing; that is to say, in either case resemblances 

 have been conceived as genuine homologies due to a common cause. In 

 recent years however, several distinguished students, borrowing the concept 

 of convergent evolution from the biologists, have shown that the older 

 mode of reasoning should be subjected to revision. A few concrete examples 

 will make their position somewhat clearer. 



One of the most widely discussed of ethnological problems is that of 

 exogamy — the fact that in many primitive communities the tril)e is sub- 

 divided into social groups, members of any one of which are forbidden to 

 intermarry. Transgressions are usually regarded in the light of incest and 

 may be punished with death even where no blood relationship obtains 

 between fellow-clansfolk. In other words, the kinship within these " clans" 

 or "gentes" is in a, measure fictitious from our point of view. Moreover 

 as membership in a group is inherited either through the father or the 

 mother, the clan or gens bond is a one-sided relationship that neglects to 

 take into account either the maternal or the paternal side. 



Now this type of clan or gens afhliation has been commonly supposed 



139 



