50 



NATURE 



{May 20, 1880 



testimony to the evanescent character of the Huron 

 language is frankly acknowleged. Many other moot 

 questions are touched with great impartiality, and it is 

 well remarked that divergence of opinion is a healthy 

 sign of life and scientific progress ; for " it is only by the 

 conflict and discussion of theories tliat truth can finally 

 be reached, and the many controversies excited by the 

 science of language show how broadly and deeply the 

 foundations of the science are being laid " (i. 87). 



A statement, however, is made a little further back 

 which will perhaps cause some surprise, as tending to 

 shake these very foundations and call in question con- 

 clusions that seemed almost universally accepted. The 

 theory of evolution, which may be said to underlie all 

 modern thought, and which has already passed almost 

 beyond the pale of discussion, has naturally tended to 

 remove much of the confusion previously associated with 

 the various conflicting opinions entertained regarding the 

 origin of human speech. For if true at all it is evident 

 that this great principle must be of universal application 

 and when applied to language the inference was irresistible 

 that there can be no immutable types of speech, any 

 more than there are immutable animal and vegetable 

 species. Hence the necessary conclusion that all present 

 forms of speech are modifications of previously existing 

 forms, that, however slowh', all are continually shifting^ 

 possibly retrograding under unfavourable conditions, but 

 in the normal state advancing, for the history of evolution 

 is on the whole the history of progress. A careful study 

 of the texture of speech seemed fully to confirm these 

 ii priori deductions, and a general consensus was thus 

 arrived at that there must have been some hypothetical 

 root-state out of which language was slowly evolved, 

 passing successively through lower to higher types, from 

 the isolating to the polysynthetic, agglutinating, incor- 

 porating, inflectional, and analytic orders. 



But in seeming opposition to these views the author 

 holds that it cannot be proved that the prima;val root- 

 languige ever existed, and that "equally unproved is the 

 belief that isolating dialects develop into agglutinative, 

 and agglutinative into inflectional " (p. 75). And at p. 

 131, while admitting the general doctrine of evolution, he 

 seems still to argue for the immutability of linguistic 

 types, though his language is here somewhat deficient in 

 its usual clearness and point. " The Finnic idioms," he 

 writes, " have become so nearly inllectional as to have led 

 a recent scholar to suggest their relationship to our Aryan 

 group ; nex'ertheless they ha\e never cleared the magical 

 frontier between flection and agglutination, hard as it may 

 be to define, since to pass from agglutination to in- 

 flection is to revolutionise the whole system of thought 

 and language and the basis on which it rests, and to break 

 with the past psychological histoiy and tendencies of a 

 speech." 



Here it should be observed that the author may not 

 inconsistently deny the necessary development of agglu- 

 tination into inflection, because he does not regard the 

 latter as a higher type than the former, and because he 

 takes, not the word or root, but the sentence, as the unit 

 and starting-point of all speech. Now the sentence 

 may have been originally cast in an agglutinative 

 form, and if so agglutination would neither imply 

 development in itself nor any necessary further evolu- 



tion in a new direction. This, at least, we take to 

 be the underlying argument, though it appears nowhere 

 explicitly stated in this way. It is stated, however 

 (p. 131), that by taking the sentence as the unit "there 

 is no longer any difficulty in distinguishing between the 

 several families of speech and assigning to each its 

 character and place." 



To all this many will of course reply that to take the 

 sentence as the starting-point is to beg the whole question. 

 It cannot, of course, be denied by the consistent evolu- 

 tionist that there must have been a time when a single 

 articulate utterance supplemented by tone and gesture, did 

 duty for a whole sentence, and in this sense it may be 

 admitted that the sentence is the starting-point of speech. 

 But whether this incipient state can be regarded as con- 

 stituting language, properly so called, is quite another 

 matter, and in any case it could not be predicated of such 

 language that it was either agglutinating or polysynthetic, 

 or even isolating in the sense that Chinese or Annamese 

 is isolating. Here we are, in fact, dealing rather with 

 the germs of the plant than with the plant itself. 



It will further be urged that if "the Finnic idioms 

 have become so nearly inflectional,'' progress from agglu- 

 tination in the direction of inflection is admitted, in which 

 case the fact that " they have never cleared the magic 

 frontier " becomes what the French would call a mere 

 detail, a question of time or other circumstances. The 

 i\Iagyar has already developed an article, and the Dravi- 

 dian tongues possess what look remarkably like true case- 

 endings, while more than one language of the Caucasus, 

 notably Georgian, Chechenz, and Lesghian, have appa- 

 rently passed quite over to the inflecting state. The fact 

 that this transition " revolutionises the whole system of 

 thought and language" will not alarm those evolutionists 

 who necessarily hold that revolution is the law of nature 

 and the order of the universe. Only the great issues are 

 worked out scnsim sine sciisu, and not by violent cata- 

 clysms and fresh creations, as was formerly supposed by 

 unorthodox interpreters of a book which allows of but 

 one creation and one partial cataclysm. Lastly, the 

 critical analysis of agglutination, aad still more of inflec- 

 tion, clearly shows that both are the result of semato- 

 logical and phonetic decay continued over immense 

 periods of time, during which numbers of concrete terms 

 and notional words of all sorts gradually lost their inde- 

 pendence, and thus became transformed to relational 

 particles first loosely tacked on (agglutination), and then 

 completely fused (inflection) with the theme. Thus it is 

 that the passing vagaries of deep thinkers serve but to 

 re-establish on firmer ground the very truths they seem to 

 assail. 



On other questions the work is equally suggestive, and 

 there' are some trenchant remarks at p. 349 of vol. i. 

 which ought definitely to close the doors of the old school 

 of etymologists. " The etymologist must be thoroughly 

 trained in the principles of scientific philology. He must 

 ha\e mastered both phonology and senratology, and he 

 must be well acqiuiinled with more than one of the 

 languages with which he deals. Then and then only can 

 his labours be fruitful ; then and then only will his work 

 be a gain and not a hindrance. False etymologies stand 

 in the way of true ones, and the charlatans who have 

 brought the name of etymology into contempt have 



