.504 



NATURE 



[Ma)'ch 31, i88x 



reserving all speculation for a time when these facts will 

 have been accumulated in sufficient number to afford a 

 sound basis for more general inductions. '' The book is 

 a body of directions for collectors " (Preface vi.). It is 

 divided into three chapters, one "On the Alphabet," 

 another containing " Hints and Explanations," and a 

 third supplying a large number of forms or " Schedules " 

 to be filled up by the collector. The chapter on the 

 Alphabet aims at establishing some uniform system of 

 spelling for all the native tongues, and puts forth a com- 

 prehensive scheme embodying many useful suggestions 

 .well deser\ing the attention of our " spelling reformers." 

 These are summed up in a few fundamental rules, the 

 chief of which are the e.xclusion of all characters and 

 diacritical marks except those found in ordinary English 

 printing offices, and the restriction of each sign to a single 

 sound. The difficulty of adapting the Roman system to 

 the Indian tongues will be understood when it is stated 

 that " there are probably sounds in each which do not 

 appear in the English or any other civilised tongue ; and 

 perhaps sounds in each which do not appear in any of the 

 others, and further, that there are perhaps sounds in each 

 of such a character, or made with so much uncertainty, 

 ithat the ear is unable to clearly determine what these 

 sounds are, even after many years of effort " (p. 2). 

 Nevertheless the difficulty is manfully faced and largely 

 overcome by the scheme here adopted, which is founded 

 on one originally proposed by Prof. J. D. Whitney, and 

 which is consequently at once scholarly, simple, and 

 comprehensive. A few improvements might here and 

 ^there be suggested, but on the whole there is little to 

 complain of, except perhaps the use of the circumflex ( •• ), 

 to mark both a long a sound, as in all, and a short ii 

 ■sound as in htt. Some contuiion is caused by an 

 awkward misprint at p. 5, where this ii appears instead of 

 the German ii. It might also perhaps be better to indicate 

 •excessive vowel length by doubling the vowel as in Dutch, 

 than by the clumsy addition of the sign +. Thus mium 

 rather than 711a + n. 



Chapter II. contains a number of well-digested and 

 tersely-expressed remarks on dress, ornaments, dwellings, 

 implements, food, colours, plants, animals, medicine, 

 social organisation, kinship, government, and many otlier 

 topics, which at first sight seem to have little connection 

 with the subject of American philology. But the author 

 has wisely endearoured thus " to connect the btudy of 

 language with the study of other branches of anthro- 

 pology ; for a language is best understood when the 

 habits, customs, institutions, philosophy — the subject- 

 matter of thought embodied in a language are best 

 known. The student of language should be a student of 

 the people who speak the language ; and to this end the 

 book has been prepared, with man\ hints and suggestions 

 relating to other branches of anthropology" (Preface vi.). 

 But besides these matters the chapter contains *hat will 

 be welcomed as a boon by all linguists, a reprint of J. H. 

 Trumbull's masterly paper "On the Best Method of 

 Studying the North American Languages," originally 

 published in the Tnuisaciions oi the .'\merican Philological 

 Association, 1869-70, but strangely neglected by many 

 subsequent writers on the subject. No other treatise 

 perhaps of equal length contains so clear and philosophic 

 an account of the peculiar genius and morphology of 



these ipolysynthetic tongues. A great deal of space is 

 devoted to the question of kinship, the true basis of Indian 

 tribal society, and this intricate subject is fully illustrated 

 by a series of four "kinship charts" or genealogical 

 diagrams, which the original investigator will find of the 

 greatest service in collecting and arranging his materials. 

 The general student will also find them extremel)' useful 

 in comparing the American systems of family relationship 

 with those prevalent especially amongst the Uravidians 

 of the Deccan and the Australian aborigines. Too much 

 importance has perhaps been attached to resemblances of 

 this sort in tracing racial affinities ; but their significance 

 in the history of the evolution of human culture is 

 undeniable. Connubial society develops into kinship 

 society, or the clan, in ivhich all the members are blood 

 relations, whence the tribe and nation. It is remarkable 

 that the connubial, or lowest form, still so prevalent in 

 many parts of the eastern hemisphere, seems to have 

 long disappeared, at least from the northern half of the 

 New World, although some of its customs, especially 

 those associated with kinship, still survive in the more 

 advanced tribal state. This explains the barbaric wealth 

 of family nomenclature with which the Indian languages 

 are still encumbered. In the printed forms, or schedules, 

 of which Chapter III. exclusively consists, the terms of 

 relationship occupy about forty pages, and include 

 hundreds of complicate affinities such as, " my father's 

 elder brother's daughter's daughter's daughter's daughter," 

 "my father's mother's brother's son's son's son's son," 

 " my mother's father's brother's son's daughter's daughter's 

 daughter," "my mother's mother's sister's daughter's 

 son's daughter's daughter," "my mother's elder sister's 

 daughter's daughter' s daughter' s husband.' ' For these, 

 and even more intricate degrees of parentage, many native 

 tongues supply equivalents, which the collectors are 

 accordingly required to discover and insert in the blank 

 columns prepared for the purpose in the schedules. The 

 arrangement of the other matter contained in these 

 schedules seems to be somewhat needlessly involved. At 

 least the advantages are scarcely so obvious as the incon- 

 venience of breaking up the strictly lexical part into 

 upwards of twenty sub-headings, instead of lumping the 

 whole in one general vocabulary arranged alphabetically. 

 Experience has abundandy shown how troublesome is 

 the use of such minutely-classified lists of words even for 

 the compiler. This remark does not of course apply to 

 the hsts of sentences (Schedules 26-9), which appear to 

 have been carefully prepared, and are well calculated 

 to bring out the structure and varied grammatical forms 

 of the Indian languages. A. H. Keane 



LETTERS TO THE EDITOR 



[ The Editor does not hold himself respomible for opinions expressed 

 by his correspondents. Neither can he undertake to return, or 

 to correspond with the writers of, rejected manuscripts. No 

 notice is taken of anonymous communications. 

 The Editor urgently requests correspondents to keep their Utters as 

 short as possible. The pressure on his space is so great that U 

 is impossible otherwise to ensure the appearance even of torn- 

 municatimts containing intereUing and tunel facts.l 



Hot Ice 



The letter of Mr. Perry (Nature, vol. xxiii. p. ■sSS) in 



answer to mine on the subject of Dr. Camelley's experinnHit 



(p 26iV) 1ms remained a long time unanswered, partly because I 



