2 74 



SCIENCE. 



[Vol. X. No. 252 



languages of British Columbia. All this, it seems to me, argues in 

 favor of the indigenous, American origin of the Eskimo. 



A. F. Chamberlain. 



University College, Toronto, Nov. iz. 



It seems to me that the similarities of sound mentioned in Mr. 

 Chamberlain's letter cannot be admitted as evidence of a connec- 

 tion between the Eskimo and other American languages. The 

 Eskimo words which he classes together are derivatives of entirely 

 different stems, that cannot be traced to a common root. In the 

 first table we recognize the following stems : 7iipta- (' clear weath- 

 er '), nipig- (' to stick '), nipag- (' to vanish '). Under the heading 

 man the words inuk and angitt are classed together, although they 

 have no connection whatever. In comparing languages, compli- 

 ■cated derivatives must not be used, but the words must first be 

 traced to their stems, and the meaning of the stems must be ascer- 

 tained as well as the phonetic laws obtaining in the dialects of the 

 stock, before it is possible to make a satisfactory comparison. 

 Fortuitous coincidences of sound like those given by Chamberlain 

 cannot be admitted as evidence of relationship. F. BOAS. 



New York, Nov. 25. 



Rate of Change in American Languages. 



The letter of Dr. Beauchamp {Science, Nov. 18) opens an inter- 

 esting linguistic question. My own impression is that the rapidity 

 of changes in unwritten, at least American, languages has been 

 overestimated. 



Sagard, in the preface to his ' Dictionnaire de la Langue Hu- 

 ronne' (Paris, 1632), asserted that the Huron was constantly chan- 

 .ging, so that in a generation or two it would seem like a new lan- 

 guage. Two hundred years afterwards, Duponceau took Sagard's 

 very imperfect book, tried it on some intelligent Hurons, and found 

 that " the language had not undergone any essential change " 

 {Memoire siir les Langites de l Ameriqiie dii Nord, pp. 444, 445). 



In 1578 Jean de Lery printed his ' Histoire d'un Voyage faict 

 en la terra du Bresil,' containing long conversations in Tupi. 

 Three hundred years later, Dr. Nogueira republished these conver- 

 sations, with their equivalents in modern Tupi. The differences 

 are surprisingly small, — with proper allowances for dialect and 

 varying phonetics, scarcely more than between Lery's French and 

 the French of to-day (see Nogueira, Apo7ttamentos sobre 

 -Abaheenga on Lingua Geral dos Braszs, Rio de Janeiro, 1876). 



I have recently completed a comparison between the Alagiiilac 

 of Guatemala, which is the most southern dialect known of the 

 Nahuatl, by means of a vocabulary obtained in 1878, with that 

 tongue as spoken in the valley of Mexico in 1550, preserved in the 

 ' Vocabulario ' of Molina. The separation of the two peoples 

 could not have been less than four hundred years ; but the diver- 

 ^gencies are so slight that I could easily have believed the Alagiiilac 

 ■words to have been obtained by a German (my informant was of 

 that nationality) in ancient Tezcuco. 



Dr. Beauchamp, in referring to conflicting orthographies of the 

 same word, points out a real but not the only cause of apparent 

 without actual change in these tongues. He also touches on the 

 confusion liable to occur from the natives forming diverse figura- 

 tive compounds to express objects and ideas new to them. I was 

 •struck with this lately in comparing the expressions in the Lenape 

 for 'faith,' 'regeneration,' 'repentance,' and such theological terms, 

 as introduced, on the one hand, by the Moravian missionaries, and, 

 on the other, independently, by the Anglican Church. They are 

 usually totally dissimilar. 



But a much more curious and important law underlies the appar- 

 ent variability of many American tongues. I refer to that of ' al- 

 ternating consonants ' and ' permutable vowels.' In a number of 

 these languages it is entirely optional with the speaker to articulate 

 any one of three or four consonantal sounds for the same phonetic 

 element. For example : he may at will pronounce the syllable ton 

 either thus, or Ion, nol, rot, etc., alternating the elements /, n, r, t, 

 at will. I have little doubt but that something of the same kind 

 obtained in ancient Accadian, which will explain why the same 

 cuneiform character stands indiscriminately for the sounds kii, tus, 

 j)un, and dur : and the recent researches of Dr. Carl Abel on the 

 phonetic modifications of the ancient Coptic radicals hint strongly 

 ■at the prevalence of this peculiarity in that venerable speech. 



In America, I name as special examples of this the Klamath and 

 the Chapanec. But that these phonetic variations are within fixed 

 limits, and do not involve the integrity of the language, is curiously 

 proved by the last mentioned. Remesal, the eariy ecclesiastical 

 historian of Chiapas, states that the Chapanec was introduced into 

 that department from Nicaragua many generations before the Con- 

 quest; probably it was not later than the year 1300. Now, in 1872, 

 my late friend, Dr. C. H. Berendt, collected in Nicaragua, from a 

 few old Indians, the only survivors of their tribe who spoke its 

 tongue, a number of words and phrases of a dialect called the 

 ' Mangue.' A comparison proves it to have been beyond question 

 a very close relative of the Chapanec, essentially the same in fact, 

 though separated from it for more than five hundred years (see an 

 article on the Mangue by me in the Proceedings of the American 

 Philosophical Society, 1885). 



As in the Turanian tongues, the Turkish, for example, there is a 

 ' vocalic echo,' the leading vowel of the word forcing the others to 

 assimilate to it in sound, so in some American tongues there is a 

 ' consonantal echo,' the presence of one consonantal sound requir- 

 ing more or less changes in the others. The Tupi, the Chapanec, 

 and the Klamath offer examples of the ' consonantal echo,' while 

 a certain degree of the ' vocalic echo ' is observable in the Kiche 

 and Cakchiquel. 



These phonetic laws must be thoroughly understood and al- 

 lowed for, before any one pronounces positively on the rate of 

 change in American languages. Dr. D. G. Brinton. 



Media, Penn., Nov. 23. 



Amnesia. 



The cases cited in Science (Nov. 11, 18, pp. 232, 250) remind me 

 of the following. Some twenty-seven years ago, a neighbor of 

 mine (a young man of twenty-five or under), springing from the 

 vaulting-horse in a gymnasium to catch the trapeze, fell, striking 

 apparently upon his shoulders, and was taken up insensible. Con- 

 sciousness soon returned, perhaps in a fraction of an hour, but 

 there was no recollection of the few hours just previous to the fall. 

 As recovery progressed, however, it was said that his recollections 

 came down closer and closer to the time of the accident ; and that 

 in a week or less he could even remember taking the leap, though 

 not his striking the mattress. 



Whether it be common that the progress of recovery should thus 

 lessen the period covered by the amnesia, might no doubt be 

 learned from such data as many professional athletes could furnish. 

 An athlete once told me how, some years before, he had fallen on 

 his forehead in the circus, and had been taken up for dead. His 

 recovery, I think, had taken several months. He could remember, 

 not indeed the blow, but the sense of powerlessness with which, in 

 mid-air, he had realized that " his balance was lost." But perhaps 

 he did not say whether, a few hours or weeks after the accident, 

 his recollections had come down so far. J. E. Oliver. 



Cornell University, Nov. 18. 



The cases of amnesia mentioned in Science of Nov. 18 recall in 

 my own experience cases which may be of sufficient interest to be 

 recorded. 



When about fifteen years old, I went into a stable to stanchion 

 cows for milking. About an hour afterwards I was found wandering 

 about the yard unconscious, and bleeding profusely from wounds 

 in the face. I have not been able to this day to tell how I was hurt. 

 I have no recollection, beyond going into the stable and fastening 

 a few cows. My hat was found under the cattle's feet. My front 

 teeth were loosened, a hole cut through my lip, and my shoulder in 

 front badly bruised. I was feeling well at the time, and have 

 never fainted, and cannot refer the injury to that cause. The 

 nature of the injury would indicate that it came from the front, 

 and must have appealed to my senses in their normal state. 



From other experiences I have always believed that it is more 

 common to remember the cause of an injury producing temporary 

 unconsciousness than to forget it. I became unconscious once 

 from drowning, but remembered vividly every thing when restored. 

 I was once prostrated by lightning, but remember having seen the 

 flash. 



I think one's remembering the cause of an injury depends largely 



