240 Robertson SmitJis Logic. 



bility, of the same phenomenon having different 

 causes. The &quot; necessary and sufficient proof &quot; must 

 show, not only (1) the prevalence of the effects, 

 and (2) the adequacy of a certain antecedent to 

 produce them, but also (3) the impossibility of 

 their being produced by any other antecedent or 

 antecedents. This last all-essential link in the 

 demonstration is what is wanting in current 

 theories of the development of the family. And 

 with the omission of it goes a corresponding 

 neglect of the environment and circumstances, 

 physical, social, and especially historical, in which 

 any particular form of marriage appears. Iso 

 lating the various conjugal relations from their 

 historic settings, in which alone an explanation of 

 each is to be found, the theorist generally puts 

 them in an arbitrary row, as one might string 

 beads, and then asseverates that this linear ar 

 rangement of contemporaneous phenomena in 

 space corresponds to the successive order of their 

 evolution in time ! Meanwhile, no one knows that 

 there has been such a universal development ; or 

 that there ever was a time when all the forms of 

 the family did not coexist as they do to-day. 



It would seem, therefore, that even the most 

 conservative school of moralists need sacrifice 

 nothing to the current theory of the evolution of 



