The Evolution of Morality. 213 



A new theory may of course be formed of the 

 order of their connection, or sequence ; but it 

 is the indisputable merit of McLennan to have 

 shown the existence and prevalence of the phe 

 nomena themselves. One could almost wish that 

 so keen an observer had contented himself with 

 collecting and grouping facts of savage life, an 

 increase of which would scarcely have failed to 

 sober his speculations. For nothing is more 

 striking in his work than the disproportion be 

 tween the vastness of his hypothesis and the 

 comparative scantiness of the facts adduced to 

 support it. It does not appear unreasonable to 

 suppose that among savages who generally mar 

 ried within their own tribe wives should, when 

 opportunity offered, have been stolen from other 

 tribes; and even descent through females may 

 always, as it does to-day, coexist with descent 

 through males. In any case, we shall require a 

 much larger collection of evidence than has yet 

 appeared to convince us that every branch of the 

 human family has gone through precisely the 

 same course of development. Yet this supposi 

 tion seems to underlie current investigation into 

 the history of family relations. The a priori fal 

 lacy would seem to have arisen from confound 

 ing facts with the mind s method of apprehend- 



