February i6, 1922] 



NATURE 



203 



corroboration, although the author only claims to 

 have " got it up " since the idea came to him. 

 The revolution in thought, if the theory is substan- 

 tiated, may be expected to resemble the change in 

 astronomical ideas at the time of Copernicus. It is 

 ^o be hoped that an English edition will -soon 

 jpear. 



The Earliest Forms of Society. 



l) Primitive Society: The Beginnings of the 



Family and the Reckoning of Descent. By Dr. 



E. S. Hartland. Pp. v+i8o. (London: 

 [Methuen and Co., Ltd., 192 1.) 6s. net. 



Primitive Society. By Dr. R. H. Lowie. Pp. 



mi + 453. (London : George Routledge and 



Jons, Ltd., 192 1.) 215. net. 



IT is interesting to place these two books side 

 by side in order to contrast the methods of 

 attacking the problems involved in the study of 

 primitive society vs^hich have been adopted by the 

 respective schools to which the authors belong. 

 (i) Dr. Hartland is one of the leading exponents 

 of the view that there is a reasonable presumption 

 that in the evolution of society wherever the patri- 

 archal system now exists it has been preceded by 

 the matriarchate. In the volume under notice 

 he restates this view and summarises the evidence 

 on which it is based in popular form. (2) Dr. 

 Lowie, however, maintains that this theory is 

 based upon an a priori assumption, and that 

 Morgan and his followers, in their desire to formu- 

 late a logical scheme of social evolution, have dis- 

 torted the facts by confining their attention to a 

 single group of data. Pouring scorn on the heads 

 of "the older school of anthropologists," he in- 

 sists upon the empirical character of the evidence, 

 and would have each case taken on its merits, 

 subjected to intensive study, and treated as a 

 whole. 



After a review of the evidence on these lines. 

 Dr. Lowie concludes that the theory of unilinear 

 development is entirely fallacious and unwar- 

 ranted. So far from the group organisation of 

 the sib or clan being the foundation of primitive 

 society, it is only one, and that frequently not 

 the most important, of a number of forms of 

 organisation to which the individual may belong. 

 While he is prepared to allow that duplication of 

 conditions may produce duplication of a sequence, 

 as in the relation of polyandry and female infanti- 

 cide, he formally abjures independent reproduction 

 of the same series of "stages." He goes so far 

 as to say that he is "not convinced of the reality 

 of the totemic phenomenon," and for him the 

 NO. 2729, VOL. 109] 



problem of totemism resolves itself into a "series 

 of specific problems not related to one another." 

 If, however, he believes in independent develop- 

 ment only in the very limited degree indicated, 

 neither is he a whole-hearted supporter of dif- 

 fusion; while attaching full weight to diffusion, 

 particularly in continuous areas, he recognises that 

 it does not necessarily preclude independent in- 

 vention within a limited scope. 



It must be acknowledged that if Dr. Lowie's 

 argument in favour of empiricism fails to carry 

 conviction, he has done good service in empha- 

 sising the necessity for intensive study of all the 

 facts of a given area as a whole. By concentra- 

 tion on the group organisation of the kin, the 

 supporters of the evolutionary theory have some- 

 times been led astray. The existence of the family 

 as a social unit at an early stage has been obscured 

 by the view that the family emerged from the 

 group. Dr. Hartland, indeed, speaks of "sexual 

 promiscuity — relieved perhaps by temporary 

 unions in the nature of monogamy." At the same 

 time, owing to his preoccupation with kin organ- 

 isation, he is unable "to bring Andamanese society 

 within any category at present known." This fact 

 does not, however, suggest to Dr. Hartland a 

 modification of his conclusions, as might perhaps 

 be expected ; he prefers to await further evidence. 



Our Bookshelf. 



The Calendar: Its History, Structttre, and Improve- 

 ment. By Alexander Philip. Pp. xii4-io4. 

 (Cambridge: At the University Press, 1921.) 

 75. 6d. net. 

 This is not the kind of work that we expect from 

 the Cambridge University Press. It contains 

 numerous historical errors, and is not free from 

 astronomical errors also. The author has endea- 

 voured to guard against criticism of the 

 latter by stating in his preface that his 

 astronomical facts have been derived from the 

 commonly available sources, and that he has dis- 

 regarded " qualifying refinements known to modern 

 astronomy but irrelevant to a calendrial purpose." 

 This ambition has not prevented him, however, from 

 stating the length of the tropical year to hundredths 

 of a second, or the length of 4000 tropical years to 

 an exact number of minutes. The introduction of 

 these refinements, " irrelevant to a calendrial pur- 

 pose," might have been pardoned, if they were 

 accurate, which, unfortunately, they are not. But 

 it is in the history of the calendar that the defects of 

 the book are particularly displayed. The author 

 ignores the two most valuable treatises on the sub- 

 ject, Ideler's " Handbuch der Mathematischen und 

 Technischen Chronologic," and Ginzel's work 

 which bears the same title. He writes in an easy 

 way of Egyptian, Chaldean, and Chinese calendars ; 



