All conjectures over the origin and relationship of these peoples must await the more trustworthy conclusions which it will be justifiable to draw from the relationship of their languages; only through more exact knowledge of them can we hope to make progress in this broad and obscure field. Several of the North American dialects are already fairly well known to philologists; for a majority of the rest there are, at least, vocabularies which allow a superficial classification according to relationship. Distinguished American philologists, Duponceau, Pickering,[226] Edwin James, Gallatin, and several others have accumulated much material, but there has been no comprehensive work on the subject. This lack has in part been filled by Mr. Albert Gallatin, whose learned work is in the hands of the publishers.[227] Since at the present time the attention of travellers is being directed more and more to this highly interesting subject, these materials are likely to be largely increased in the near future, and to give American scholars the chance of constantly perfecting their work. Some authors, among them McKenney, in his History of the Indian Tribes of North America,[228] do not seem to attach much value to vocabularies collected by travellers. To be sure, these often are handled in a thoughtless and superficial manner; but it must be remembered that in the interior of North America excellent interpreters of the Indian languages may now be found, and that the Indians themselves often fluently speak several languages. Through these agencies interesting contributions to our knowledge of the dialects referred to may always be expected.

In an examination of the tribes of America, the multiplicity of their languages is very striking. These languages are often confined to quite small tribes; their origin is quite correctly explained by Mr. Gallatin.[229] From the interesting work of this scholar it appears, however, that a certain similarity of character belongs to this multitude of tongues—a circumstance that testifies to their common origin as well as to the antiquity of the American population.[230] All this material thus gains value in the eyes of the philologist; and in this connection it is with pleasure to be noticed, that with the extension of these investigations the interest in them in America constantly brings forward diligent co-operators.

Such work would be far easier and much more useful if all the vocabularies were compiled and copied on the same principles. Most of those in existence have been written down by Englishmen, or Anglo-Americans; and as a rule they, like the French, cannot, as Gallatin admits, correctly reproduce the Indian gutturals;[231] yet it is characteristic of all these languages that they abound in gutturals. Another defect in the usual vocabularies, lies in the omission of accents; for these, also, are peculiarly characteristic of the American languages. One of the dialects which I investigated is so difficult to write, that even after the greatest effort I could reproduce but a few words—I refer to that of the Grosventres of the prairies, sometimes called by the English Fall Indians.[232] For this tongue, even the fur trade company had not been able to find an interpreter; and was compelled to carry on all business with those Indians in the Blackfoot language. Most of the other languages of the Missouri Valley can more easily be reproduced by the Germans and Dutch than by other nations; because, as has already been explained, their own speech abounds in gutturals and regularly has hard endings. Consequently the English vocabularies for these tribes are always more or less defective. Pickering felt this very decidedly, and has consequently proposed for his countrymen a notation of tones which is analogous to the German, and which, if it comes into use, will produce most beneficial results. Pickering's scheme requires that the vowels be pronounced in America as follows:

In short, he adopts the German system, completely. In place of the German æ and œ, or ä and ö, he proposes to write in English ae and oe; in place of c to use k, a change that will avoid many misunderstandings; and when the pronunciation is hard, to use z, etc. If this system had been used earlier, a smaller number of errors would have crept into the works on this subject, and into the translations from one language to another. For instance, my vocabularies of the Brazilian peoples have in places been very incorrectly reproduced by French translators.

Another defect in the usual English system, for such vocabularies, arises from the division of all words into syllables, a method which Duponceau very rightly censured; for in this way the correct pronunciation of the words is generally lost. When the word with its accents has been written as a whole, the division into syllables is the next task, in order thus to reach conclusions regarding its inflection and derivation—something quite indispensable for the philologist. A word divided in the English style into its component parts, could be correctly pronounced neither by a German nor by a Frenchman; besides, every language has its own intonation, not common to others. I have, therefore, held it to be the surest way of making the Indian words intelligible to all my readers, to call to my aid the pronunciation of several nations. For instance, on and an, in the Indian languages are spoken, sometimes as in German, sometimes as in French; in such cases I have noted in a parenthesis the system according to which the word or syllable is to be pronounced. Other notes of this character have seemed necessary, and these require the following explanations:—

(1) Sharp c is designated as in French by ç.

(2) Guttur. denotes guttural, i.e., the German or Dutch pronunciation.

(3) aº or oª denotes full a or o, somewhat like aw in English. Here I might have written o͡a; but by using aº and oª, I also indicate that the sound of the lower letter predominates.

(4) d. d. N. (durch die Nase) denotes through the nose: i.e., a nasal sound.