ON THE EVOLUTION OF THE FAMILY. 135 



belonging to higher types, and while in some cases citing confirmatory 

 evidence furnished by certain barbarous peoples of lower types, he 

 has practically disregarded the great mass of the uncivilized, and ig- 

 nored the vast array of facts they present at variance with his theory. 

 Though criticisms have led him somewhat to qualify the sweeping gen- 

 eralizations set forth in his "Ancient Law;" though, in the preface 

 to its later editions, he refers to his subsequent work on " Villasje 

 Communities " as indicating some qualifications yet the qualifications 

 are but small, and in great measure hypothetical. He makes light of 

 9,ViCh. adverse evidence as Mr. McLennan and Sir John Lubbock give, 

 on the ground that the part of it he deems most trustworthy is sup- 

 plied by Indian Hill-tribes, which have, he thinks, been led into abnor- 

 mal usages by the influences invading races have subjected them to. 

 And, though in his " Early Institutions " he goes so far as to say that 

 " all branches of human society may or may not have been developed 

 from joint families which arose out of an original patriarchal cell," he 

 clearly, by this form of expression, declines to admit that in many 

 cases they have not been thus developed. 



He rightly blames earlier writers for not exploring a sufficiently 

 wide area of induction. But he has himself not made the area of in- 

 duction wide enough ; and that substitution of hypothesis for observed 

 fact which he ascribes to his predecessors is, as a consequence, observ- 

 able in his own work. Respecting the evidence available for framing 

 generalizations, he says : 



" The rudiments of the social state, so far as they are known to us at all, are 

 known through testimony of three sorts accounts by contemporary observers 

 of civilizations less advanced than their own, the records which particular races 

 have preserved concerning their primitive history, and ancient law." 



And since, as exemplifying the " accounts by contemporary ob- 

 servers of civilizations less advanced than their own," he names the 

 account TacituS gives of the Germans, and does not name the accounts 

 modern travelers give of uncivilized races at large, he clearly does 

 not include as evidence the statements made by these.* Let me name 

 here two instances of the way in which this limitation leads to the 

 substitution of hypothesis for observation. 



Assuming that the patriarchal state is the earliest, Sir Henry 

 Maine says that " the implicit obedience of rude men to their parent 



* He does, indeed, at page 17 of his " Village Communities," deliberately discredit 

 this evidence speaking of it as " the slippery testimony concerning savages which is 

 gathered from travelers' tales." I am aware that, in the eyes of most, antiquity gives sa- 

 creduess to testimony ; and that so what were " travelers' tales " when they were written 

 in Roman days have come, in our days, to be regarded as of higher authority than like 

 tales written by recent or living travelers. I see, however, no reason to ascribe to Taci- 

 tus a trustworthiness which I do not ascribe to modern explorers, many of them scien- 

 tifically educated Barrow, Barth, Galton, Burton, Livingstone, Seeman, Darwin, Wal- 

 lace, Humboldt, Burckhardt, and others too numerous to set down. 



