Language and the Study of Language/Lecture IX

by William Dwight Whitney

LECTURE IX.

Uncertainties of genetic classification of languages. "Turanian" family. Dravidian group. North-eastern Asiatic. Monosyllabic tongues: Chinese, Farther Indian, Tibetan, etc. Malay-Polynesian and Melanesian families. Egyptian language and its asserted kindred: Hamitic family. Languages of southern and central Africa. Languages of America; problem of derivation of American races. Isolated tongues: Basque, Caucasian, etc.

In the last lecture, we began a survey of the general dividing lines of human speech, an enumeration and description of the families into which linguistic science has combined the languages thus far brought under her notice. We had time, however, to examine but two of these families, comprehending the tongues of the two great white races which have taken or are taking, after our own, the most conspicuous parts in the history of mankind: they were, on the one hand, the Semitic, a little group of closely related dialects in the south-western corner of Asia, counting as its principal members the Hebrew, Arabic, and Syriac; and, on the other hand, the Scythian, an immense aggregation of greatly varying forms of speech, occupying with its five principal branches—the Ugrian, Samoyedic, Turkish, Mongolian, and Tungusic—a very large, but, in part, a not very valuable, portion of the combined continent of Asia and Europe. We have now to complete our work by passing in cursory review the remaining families. The task may be found, as I cannot help fearing, a somewhat tedious one—consisting, as it must do, to no small extent, in going over a catalogue of unknown or unfamiliar names, belonging to races and tongues that stand far off from our interests; but if its result shall be to give us a comprehensive view of the grand outlines, geographical and structural, of human speech, our hour will not have been spent unprofitably.

It must be borne in mind from the outset that the best classification of human languages now attainable is neither exhaustive, nor equally certain and reliable in all its parts. While nearly the whole field has been explored, it has not been explored everywhere with equal minuteness and care, nor by equally trustworthy investigators. In language, as in geography, there are few extensive regions which need any longer be marked "unknown;" yet there are many of which only the most general features have been determined: and that, perhaps, in part by inference, in part upon information which may turn out incorrect. It may be said in general that, where travellers' reports, or mere vocabularies, have alone been accessible as the ground of classification, the results reached are of superficial character and provisional value. No family of languages can have either its internal or its external relations well established, until its material has been submitted to analysis, the genesis and mode of construction of its forms traced out, and its laws of phonetic change deduced from an examination and comparison of all the accessible phenomena—until, in short, its vital processes are comprehended, in their past history and their present workings. To accomplish this for all existing and recorded human speech will be a slow and laborious task; and, for a long time to come, we must expect that the limits of families will be more or less altered, that languages now separated will come to be classed together, and even that some of those now connected will be sundered. It is not alone true that penetrating study often brings to light resemblances between two languages which escape a superficial examination; it also sometimes shows the illusiveness of others which at first sight appeared to be valid evidences of relationship. In a preliminary comparison, chance coincidences are liable to be overvalued. Moreover, the first tentative groupings are wont to be made by the more sanguine and enterprising class of philologists. The "personal equation," as the astronomers call it, the allowance for difference of temperament, endowment, and skill, has to be applied, certainly not less rigorously, in estimating the observations and deductions of linguistic scholars than those of the labourers in other sciences. There is, on the one hand, the class of facile and anticipative investigators, whose minds are most impressed by apparent resemblances; who delight in construction, in establishing connections, in grouping together extensive classes, in forming grand and striking hypotheses; who are never willing to say "I do not know:" and, on the other hand, there is the class of less ardent and more phlegmatic students, who look beneath superficial resemblances to profounder differences; who call always for more proof; who are ever ready to confess ignorance, and to hold their judgment in suspense; who refuse their assent to engaging theories, allowing it to be wrung from them only by cogent and convincing evidence. Each class has its advantages: the one furnishes the better explorers, the other the sounder critics; the one is the more numerous and the more popular, the other is the safer and the more strictly scientific.

A notable exemplification of this temperamental difference of authorities is furnished us in connection with one of the families of which we have already treated. We saw reason, in the last lecture, to regard with some doubt the genetic relationship claimed to exist between the five great branches of the Scythian family, as being founded too little on actual correspondence of linguistic materials demonstrably derived from a common source, and too much on mere analogies of linguistic structure—analogies, too, which were able to consist with such important differences as separate the jejune dialect of the Manchus from the rich and almost inflective languages of the Finns and Hungarians. We could not pronounce it certain that the family will be able to maintain its integrity in the light of a more thorough and comprehensive investigation. But, on the other hand, we were unable to deny that it may succeed in doing so; and farther, it is altogether possible that recognizable evidences of ultimate connection with the family may be found among other Asiatic tongues, as yet unclassed. Now some linguistic scholars, of no little note and authority, have ventured to give to these possibilities the value of established and unquestionable facts. They have set up an enormous family, which they have styled the "Turanian;" they have allotted to it the agglutinative structure as its distinctive characteristic, and have made it include nearly all known tongues save the Indo-European and Semitic, not in Asia alone, but through the oceanic islands and over the continent of America. Such sweeping and wholesale conglomeration (for we can hardly call it classification), at the present stage of progress of linguistic research, is wholly unscientific, and of no authority or value. It represents only a want of detailed knowledge, and a readiness to give way to loose and unscrupulous theorizing, on the part of its authors, who are, at the very best, anticipators of the result of scientific inquiry—who are even already proved in part its contradictors: for it is long since shown that many of the alleged "Turanian" dialects are hardly less fundamentally different in their structure from the typical languages of the family than is the Greek or the Hebrew. That the inventors of the name Turanian have associated it with such a baseless classification is sufficient reason why it should be strictly rejected from the terminology of linguistic science. Nor has it in virtue of its derivation any peculiar claim to our acceptance. It is borrowed from the legendary history of the Persian or Iranian race, as represented to us chiefly by the Shah-Nameh, or 'Book of Kings,' of Firdusi. There Irej and Tur are two of the three brothers from whom spring the races of mankind; and the tribes of Iran and Turan, their descendants—namely, the native Persians and their neighbours upon the north-east, probably of Turkish kindred—are represented as engaged in incessant warfare upon the frontier of their respective territory. Why we should adopt a term so local in its original application, out of a cycle of legends with which so few of us are familiar, as the name of a race which is claimed to extend from the north-western border of Europe eastward across continent and ocean, widening as it goes, till it spreads along the whole western Atlantic shore, cannot easily be made to appear.

There are especially two groups of Asiatic languages, which have been confidently claimed, and with some show of reason, to belong to the Scythian family. Of these, the first is that occupying the southern portion of the peninsula of India, and commonly called the Tamulian or Dravidian group or family. We have already seen (in the fifth and sixth lectures) that the Sanskrit speaking tribes, of Indo-European race, forced their way into India through the passes on its north-western frontier, almost within the historic period; and that they there took exclusive possession only of the northern portion of the country, including especially the vast plains and valleys of Hindustan proper, with a tract of the sea-coast stretching southward on either hand; dispossessing so far, by reduction to servitude or by expulsion, the more aboriginal inhabitants, but leaving to their former owners the hilly and elevated southern region, the Dekhan, as well as the yet less accessible heights and slopes of the Himalaya chain in the north. Throughout nearly the whole Dekhan, these older races still form the predominant population, and speak and write their own languages. Chief among the latter are the Tamil, occupying the south-eastern extremity of the peninsula, along with most of the island of Ceylon; the Telinga or Telugu, spoken over a yet more extensive region lying north of this; the Canarese, extending from the interior border of the Tamil and Telugu westward almost to the coast; the Malayâlam or Malabar, covering a narrow strip of the south-western coast, from Cape Comorin northwards; and the Tulu, filling a still more restricted area to the north of the Malayâlam. All these are cultivated tongues, and possess written literatures, of greater or less extent and antiquity; that of the Tamil is the most important and the oldest, parts of it appearing to date back as far as to the eighth or ninth century of our era; nothing in Telugu is earlier than the twelfth. The Dravidian races, however, have derived their religion, their polity, and their culture, from the superior race to the north of them, the Hindus; their alphabets are of Hindu descent; their philosophical and scientific terms are borrowed from the rich stores of the Sanskrit; their literary works are in no small part translations or imitations of Sanskrit authors. There are other tribes in the peninsula, of less numbers and importance, wholly uncultivated, and in part of savage manners and mode of life. Some of these—as the Tudas of the Nilagiri hills, the Kotas of the same neighbourhood, and the wild Gonds and Khonds of the hilly country of Gondwana—are proved by their language to be akin with the Dravidian peoples;^[1] others—as the Kols, Suras, and Santals—appear to be of entirely diverse race and speech; relics, perhaps, of a yet more ancient Indian population, which occupied the soil before the incursion of the Dravidians, and was driven out by these, as they, in their turn, by the Indo-Europeans. Once more, outside the borders of India proper, in the neighbouring country of Beluchistan (the ancient Gedrosia), there is found a people, the Brahuîs, whose tongue, though filled with words of Hindu origin, is claimed to exhibit unequivocal traces of a Dravidian basis.

The Dravidian languages are not only, like the Scythian, of a generally agglutinate character, but their style of agglutinative structure is sufficiently accordant with that of the Scythian tongues to permit of their being ranked in the same family, provided that material evidence of the relationship, of a sufficiently distinct and unequivocal character, shall also be discovered. That such has been already found out and set forth, is not to be believed. The investigation has not yet been undertaken by any scholar profoundly versed in the languages of both families, nor has the comparative grammar of the Scythian dialects reached results which can be applied in conducting it and in arriving at a determinate decision. That an outlying branch of the Scythian race once stretched down through western and southern Iran into the Indian peninsula is at present only an attractive and plausible theory, which may yet be established by comparison of languages, when this comparison shall have been made with sufficient knowledge and sufficient caution.

The other group referred to, as having been sometimes claimed to exhibit traces of relationship with the Scythian family, is composed of the languages which occupy the peninsulas and islands of the extreme north-eastern part of the Asiatic continent. Their character and relations constitute a very obscure and difficult problem in linguistic ethnology: whether they make up a group in any other than a geographical sense, whether they are not isolated and independent tongues, is at present exceedingly doubtful. Their linguistic tie, if there be one, is yet to be established.

By far the most conspicuous and important member of the group is the Japanese. It is wholly confined to the islands forming the empire of Japan (and into the northernmost of these, Yesso, it is a recent intrusion; the chief population of the island is Kurilian), and has no representatives or near kindred upon the main-land. So lively attention has been directed to it of late, since the re-opening of the empire to Europeans—its grammars, dictionaries, conversation-books, and the like, are multiplying so rapidly in European languages, and are leading to so much discussion of its linguistic character, that we may hope to see its position ere long definitely established. It has recently been repeatedly and confidently asserted to be "of the Turanian family;" but this is a phrase of so wholly dubious meaning that we cannot tell what it is worth: we shall be obliged to hold our judgments suspended until the general relations of the north-eastern Asiatic languages are better settled. The language is polysyllabic and agglutinative in character, possessing some of the features of construction which also characterize the Scythian tongues. It is of a simple phonetic structure (its syllables being almost always composed of a single consonant with following vowel), and fluent and easy of utterance. Besides the ordinary spoken dialect, there is another, older and more primitive, used as the medium of certain styles of composition: it is called the Yamato. Much, too, of the learned literature of the Japanese is written in Chinese. Their culture and letters come from China, being introduced, it is believed, in the third century of our era; the annals of the empire, however, claim to go back to a much higher antiquity, even to a time some centuries before Christ. It was unfortunate for an inflected tongue like the Japanese to be obliged to resort to China for an alphabet; and although a thoroughly practical and convenient set of characters, of syllabic value, easy to write and to read, was at one time devised, being made out of parts of Chinese ideographs, it is of very restricted use; and the mode of writing generally employed for literary texts is one of the most detestable in the world, and the greatest existing obstacle to the acquirement of the language.

The dialect of the Loo-Choo islands is nearly akin with the Japanese.

The peninsula of Corea, lying in close proximity to the empire of Japan, is occupied by a language between which and the Japanese, though they are not so dissimilar in structure that they might not be members of one family, no material evidences of relationship have been traced and pointed out. The Corean also possesses some literary cultivation, derived from China; but of both language and literature only the scantiest knowledge has reached the West.

Along the coast of Asia north of Corea, and also upon the island of Saghalien or Karafto, and through the Kurile chain of islands, which stretch from Yesso northward to the extremity of the peninsula of Kamchatka, dwells another race, that of the Ainos or Kurilians. They are hairy savages, who live by hunting and fishing, but are distinguished by nobility of bearing and gentleness of manners. Their speech has been sometimes pronounced radically akin with the Japanese, but, apparently, without any sufficient reason. A few of their popular songs have been written down by strangers.

The peninsula of Kamchatka itself belongs to yet another wild race, the Kamchadales; and to the north of these lie the nearly related peoples of the Koriaks and Chukchi, between whom and the American races a connection has been suspected, but not satisfactorily proved. The Namollos, who occupy the very extremity of the continent, next to Behring's straits, are pretty certainly related with the Eskimos of the northern shores of the opposite continent, and thus appear to be emigrants out of America into Asia.

Between the races we have mentioned and the Yakuts of the Lena, that far outlying branch of the Turkish family, finally, live the Yukagiris, another isolated and widely spread people, not proved by their language to be akin with any of their neighbours.

It was the more necessary to glance at the intricate and ill understood linguistic relations of this part of the Asiatic continent, because our eyes naturally turn curiously in that direction, when we inquire whence and how our own American continent obtained the aboriginal population which we have been dispossessing. It is evident that much remains to be done upon the Asiatic side of the straits before the linguistic scholar can be ready for a comparison which shall show with what race of the Old World, if with any, the races of the New are allied in speech.

The south-eastern portion of Asia is occupied by peoples whose tongues form together a single class or family. They fill China and Farther India, and some of the neighbouring parts of the central Asiatic plateau. The distinctive common characteristic of these tongues is that they are monosyllabic. Of all human dialects, they represent most nearly what, as we have already seen reason for concluding, was the primitive stage of the agglutinative and inflective forms of speech; they have never begun that fusion of elements once independently significant into compound forms which has been the principal item in the history of development of all other tongues. The Chinese words, for example, are still to no small extent roots, representing ideas in crude and undefined form, and equally convertible by use into noun, verb, or adverb. Thus, ta contains the radical idea of 'being great,' and may, as a substantive, mean 'greatness;' as an adjective, 'great;' as a verb, either 'to be great,' or 'to make great, to magnify;' as an adverb, 'greatly:' the value which it is to have as actually employed, in any given case, is determined partly by its position in the phrase, and partly by the requirements of the sense, as gathered from the complex of ideas which the sentence presents. We have already had occasion to remark (in the seventh lecture) that somewhat the same thing may be said of many English words; we took love as an instance of one which is now either verb or noun, having lost by phonetic abbreviation the formative elements which once distinguished it as the one and as the other. It is a very customary thing with us, too, to take a word which is properly one part of speech, and convert it into various others without changing its shape: for example, better is primarily an adjective, as in "a better man than I;" but we employ it in connections which make of it an adverb, as in "he loves party better than country;" or a noun, as when we speak of yielding to our betters, or getting the better of a bad habit; or, finally, a verb, as in "they better their condition." Such analogies, however, do not explain the form and the variety of application of the words composing the Chinese and its kindred languages. Of the former possession of formative elements these words show no signs, either phonetic or significant; they have never been made distinct parts of speech in the sense in which ours have been and are so. How different is the state of monosyllabism which precedes inflection from that which follows it in consequence of the wearing off of inflective elements, may be in some measure seen by comparing a Chinese sentence with its English equivalent. The Chinese runs, as nearly as we can represent it, thus: "King speak: Sage! not far thousand mile and come; also will have use gain me realm, hey?" which means, 'the king spoke: O sage! since thou dost not count a thousand miles far to come (that is, hast taken the pains to come hither from a great distance), wilt thou not, too, have brought some thing for the weal of my realm?'^[2]

While all the languages of the region we have described thus agree in type, in morphological character, they show a great and astonishing diversity of material; only scanty correspondences of form and meaning are found in their vocabularies; and hence, the nature and degree of their mutual relationship are still obscure. But the structural accordance is here, evidently, a pretty sure sign of common descent. If monosyllabic tongues were of frequent occurrence among human races, it, for instance, we met with one group of them in China, another in Africa, and another in America, we should have no right to infer that they were all genetically related; for it is, beyond all question, hypothetically possible that different divisions of mankind should be characterized by a kindred inaptitude for linguistic development. When, however, we find the known languages of this type clustered together in one corner of a single continent, we cannot well resist the conviction that they are all dialects of one original tongue, and that their differences, however great these may be, are the result of discordant historic growth.

Infinitely the most important member of the monosyllabic group or family is the Chinese: its history is exceeded in interest by that of very few other known tongues. Its earliest literary records (some of the odes of the Shi-King, 'Book of Songs') claim to go back to nearly two thousand years before Christ, and the annals and traditions of the race reach some centuries farther, so that Chinese antiquity almost exceeds in hoariness both Semitic and Indo-European. China, indeed, in the primitiveness and persistency of its language, its arts, and its polity, is one of the most remarkable and exceptional phenomena which the story of our race presents. It has maintained substantially the same speech and the same institutions, by uninterrupted transmission from generation to generation upon the same soil, all the way down to our own times from a period in the past at which every Indo-European people of which we know aught was but a roving tribe of barbarians. Elsewhere, change has been the dominating principle; in China, permanency. Nor has this permanency been quietism and stagnation. China has had, down even to modern times, no insignificant share of activity and progress, though always within certain limits, and never of a radical and revolutionary character. She has been one of the very few great centres of culture and enlightenment which the world has known; and her culture has been not less original in its beginnings, and almost more independent of foreign aid in its development, than any other. She has been the mother of arts, sciences, and letters, to the races on every side of her; and the world at large she has affected not a little, mainly through the material products of her ingenuity and industry. Repeatedly subjected to foreign domination, she has always vanquished her conquerors, compelling them implicitly to adopt her civilization, and respect and maintain her institutions. That she now at last seems to have become in a measure superannuated and effete, and to be nearing her downfall, under the combined pressure of overcrowded population, a detested foreign yoke and internal rebellion against it, and the disorganizing interference of Western powers, may be true; but it does not become us to regard otherwise than with compassion the final decay of a culture which, taking into account the length of its duration and the number of individuals affected by it, has perhaps spread as much light and made as much happiness as any other that ever existed.

The representative man of China is Confucius, who lived in the sixth century before Christ. He is no religious teacher, but an ethical and political philosopher. In him the wisdom of the olden time, the national apprehension of the meaning and duties of life, found its highest expression, which has been accepted as authoritative by all succeeding ages. He determined how much of the ancient literature should be saved from oblivion: his excerpts from it, historical and poetical, together with his own writings, and the works of his pupils, in which are handed down his own instructions in public and private virtue, form nearly the whole of the Five King and the Four Books, the national classics, the earliest and most revered portion of the national literature. Their continuation and elaboration have engaged no insignificant part of the literary activity of following generations. But, aside from this, almost every department of mental productiveness is represented in China by hosts of works, ancient and modern: in history, in biography, in geography and ethnology, in jurisprudence, in the grammar and lexicography especially of their own tongue, in natural history and science, in art and industry, in the various branches of belles-lettres, as poetry, romance, the drama, the Chinese have produced in abundance what, tried even by our own standard, is worthy of high respect and admiration. No race, certainly, outside the Indo-European and Semitic families, nor many races even of those families, can show a literature of equal value with the Chinese.

Not very much requires to be said in explanation of the structure and history of a language so simple—a language which might be said to have no grammatical structure, which possesses neither inflections nor parts of speech, and which has changed less in four thousand years than most others in four hundred, or than many another in a single century. So restricted, in the first place, is its phonetical system, that its whole vocabulary, in the general cultivated dialect (which has lost the power of uttering final mutes, still preserved and distinctly sounded in some of the popular patois), is composed of only about four hundred and fifty different vocables, combinations of sounds: these, however, are converted into not far from three times that number of distinct words by means of the tones of utterance, which in Chinese, as in some other languages of similarly scanty resources, are pressed into the service of the vocabulary, instead of being left, as with us, to the department of rhetoric and elocution. As a necessary consequence, the several words have a much greater range of signification than in more richly endowed tongues; each seems to unite in itself the offices of many distinct words, the tie of connection between its significations being no longer traceable. External development, the formation of derivative words to bear the variety of derived meanings into which every root tends to branch out, is here almost or quite unknown: internal, significant development has been obliged to do the whole work of linguistic growth. Of course, then, not only the grammatical form, but also the radical significance, is often left to be pointed out by the connection. And here, again, the Chinese finds its nearest parallel, among inflected tongues, in the numerous homonyms (words identical in sound but different in meaning) of our own English; for example, in our three different meet's (meet, mete, and meat), and bear's (bear, verb, bear, noun, and bare, adjective), and found's (found from find, found, 'establish,' and found, 'cast'), and other the like. In the written language, much of this ambiguity is avoided, since each Chinese character represents a word with regard, not to its phonetic form alone, but to its meaning also^[3]—whence comes the strange anomaly that a language composed of but a thousand or two of words is written with an alphabet containing tens of thousands of different signs. The literary style is thus enabled to unite with sufficient intelligibility a wonderful degree of conciseness, to combine brevity and precision to a degree elsewhere unapproached. The spoken language is much more wordy, using, to secure the mutual understanding of speaker and hearer, various devices, which here and there approach very near to agglutination, although they always stop short of it. To no small extent, the Chinese is in practical use a language of groups of monosyllabic roots rather than of isolated monosyllables: a host of conceptions which we signify by single words, it denotes by a collocation of several words: thus, 'virtue' is represented by four cardinal virtues, faith-piety-temperance-justice; 'parent' by father-mother; exceedingly often, two nearly synonymous words are put together to express their common meaning, like way-path, for 'way' (such a collocation being mainly a device for suggesting to the mind the one signification in which two words, each of various meaning, agree with one another); very often, again, a "classifier," or word denoting the class in which a vocable is used, is appended to it, as when we say maple-tree, whale-fish, for maple and whale (many of these classifiers are of very peculiar sense and application); certain words, further, are virtual signs of parts of speech, as those meaning 'get,' 'come,' 'go,' added to verbs; 'place,' making nouns from verbs and adjectives; a relative particle, pointing out the attributive relation; objective particles, indicating an instrumental, locative, dative case; pluralizing words, meaning originally 'number, crowd, heap;' a diminutive sign, the word for 'child;' and so on. There has been here not a little of that attenuation and integration of meaning by which in our own language we have formed so many relational words and phrases; but there is no fusion, no close combination, even, of elements; these are simply placed side by side, without losing their separate individuality. There is no reason assignable why a truly agglutinative stage might not possibly grow out of a condition of things like this; and it is claimed by some that, in certain of the popular dialects (which differ notably from the kwan-hwa, the common dialect of the lettered classes), agglutination, to a limited extent, is actually reached.

While thus the Chinese is, in certain respects of fundamental importance, the most rudimentary and scanty of all known languages, the one least fitted to become a satisfactory means of expression of human thought, it is not without its compensations. The power which the human mind has over its instruments, and independent of their imperfections, is strikingly illustrated by the history of this form of speech, which has successfully answered all the purposes of a cultivated, reflecting, studious, and ingenious people throughout a career of unequalled duration; which has been put to far higher and more varied uses than most of the multitude of highly organized dialects spoken among men—dialects rich in flexibility, adaptiveness, and power of expansion, but poor in the mental poverty and weakness of those who should wield them. In the domain of language, as in some departments of art and industry, no race has been comparable with the Chinese for capacity to accomplish wonderful things with rude and uncouth instruments.

The principal nations of Farther India are the Annamese or Cochin-Chinese, the Siamese, and the Burmese; tribes of inferior numbers, civilization, and importance are the Kwanto, Cambodians, Peguans, Karens, and others. Annamese culture is of Chinese origin; the races of Siam and Burmah emerge from obscurity as they receive knowledge, letters, and religion (Buddhism) together from India. Their languages are, like the Chinese, monosyllabic and isolating; but they are as much inferior to that tongue in distinctness of construction and precision of expression as the people that speak them have shown themselves to be inferior to the inhabitants of China in mental activity and reach. Of indicative words, substitutes for the formative elements of more highly developed languages, they make an extended use. Such auxiliary and limiting words are in Siamese always put before, in Burmese always after, the principal root.

To the same general class of tongues, yet with sundry variations of type, even sometimes appearing to overstep the boundary which divides mere collocation from actual agglutination of elements, are deemed to belong the exceedingly numerous and not less discordant dialects which crowd the mountain valleys on both sides of the great range of the Himalayas, and that part of the plateau of central Asia which lies next north of the range. The linguistic student is lost, as yet, in the infinity of details presented by these dialects, and is unable to classify them satisfactorily. Most of them are known only by partial vocabularies, lists of words gathered by enterprising collectors,^[4] no penetrating investigation and clear exposition of their structure and laws of growth having yet been made. It were useless to detail here the names of the wild tribes to which they belong, or set forth the groupings which have been provisionally established among them. The only one which possesses any historical or literary importance is the Tibetan. Tibet was one of the early conquests of Buddhism, and has long been a chief centre of that religion. It has an immense Buddhist literature, in great part translated from the Sanskrit, and written in a character derived from that in which the Sanskrit is written. Though strictly a monosyllabic language, the Tibetan exhibits some very peculiar and problematical features—in its written but now unpronounced prefixes, and a kind of inflective internal change appearing in many of its words—which are a subject of much controversy among comparative philologists.

With the next great family, the Malay-Polynesian, or Oceanic, we shall not need to delay long. Those who speak its dialects fill nearly all the islands from the coasts of Asia southward and eastward, from Madagascar to the Sandwich group and Easter Island, from New Zealand to Formosa. A few of those which are found nearest to Farther India possess alphabets and scanty literatures, coming chiefly from the introduction among them of religion and culture from India; but the Malay has adopted the Arabic alphabet. Considering how widely they are scattered, there prevails among these languages a notable degree of correspondence of material as well as of structure, and their coherence as a family is unquestionable; but the work of marking out subordinate groups, and determining degrees of relationship, is as yet but partially accomplished for them. Missionaries, American and English, have played and are playing an important part in laying them open to knowledge, as well as in introducing knowledge among those who speak them.

The Polynesian languages, especially those of the eastern division, are of simpler phonetic form than any others spoken by human races: their alphabets contain not more than ten consonants, often as few as seven, and their allowed combinations of sounds are restricted to open syllables, composed of a vowel alone, or of a vowel preceded by a single consonant; of combined consonants, or final consonants, they know nothing. They are polysyllabic, but hardly less destitute of forms than the monosyllabic tongues. Their roots, if we may call them so, or the most primitive elements which our imperfect historical analysis enables us to trace, are more often dissyllabic, but of indeterminate value as parts of speech: they may be employed, without change, as verb, substantive, adjective, or even preposition. All inflection is wanting: gender, case, number, tense, mode, person, have no formal distinctions; pronouns, indicative particles, prepositions, and the like, constitute the whole grammar, making parts of speech and pointing out their relations. Moreover, anything which can properly be styled a verb is possessed by none of these languages; their so-called verbs are really only nouns taken predicatively. Thus, to express 'he has a white jacket on,' the Dayak says literally "he with-jacket with-white," or "he jackety whitey."^[5] As a means of development of signification, the repetition or reduplication of a root is very frequently resorted to; prefixes and suffixes, especially the former, are also applied to the same purpose. Only the personal pronouns have a peculiar kind of variation by number, produced by composition and fusion with the numerals: in this way are often distinguished not only a singular, dual, and plural, but also a tri-al, denoting three: and the numbers other than singular of the first person have a double form, according as the we is meant to include or to exclude the person addressed.

The races to whom belong the dialects we have thus characterized are of a brown colour. But these do not make up the whole population of the Pacific island-world. The groups of little islands lying to the east of New Guinea—the New Hebrides, the Solomon's islands, New Caledonia, and others—are inhabited by a black race, having frizzled or woolly hair, yet showing no other signs of relationship with the natives of Africa. Men of like physical characteristics are found to occupy the greater part of New Guinea, and more or less of the other islands lying westward, as far as the Andaman group, in the Bay of Bengal. They are known by various names, as Negritos, Papuans, Melanesians. Some of their languages have been recently brought by missionary effort to the knowledge of linguistic scholars, and help to prove the race distinct from the Polynesian. In point of material, a wide diversity exists among the dialects of the different tribes; they exhibit almost the extreme of linguistic discordance; each little island has its own idiom, unintelligible to all its neighbours, and sometimes the separate districts of the same islet are unable to communicate together. Yet, so far as they have been examined, distinct traces of a common origin have been found; and in general plan of structure they agree not only among themselves, but also, in a marked degree, with the Polynesian tongues, so that they are perhaps to be regarded as ultimately coinciding with the latter in origin.^[6]

The aboriginal inhabitants of Australia and of parts of the neighbouring islands are by some set down as a distinct race, the Alforas: our knowledge of their speech is not sufficient for us to determine with confidence their linguistic position.

The rank in the scale of languages generally assigned to the ancient Egyptian (with its successor, the modern Coptic), its often alleged connection with the Semitic, and the antiquity and importance of the culture to which it served as instrument, would have justified us in treating it next after the Indo-European and Semitic; but it seemed more convenient to traverse the whole joint continent of Europe and Asia, before crossing into Africa. The chronology of Egyptian history is still a subject of not a little controversy; but it cannot be reasonably doubted that the very earliest written monuments of human thought are found in the valley of the Nile, as well as the most ancient and most gigantic works of human art. There was wisdom in Egypt, accumulated and handed down through a long succession of generations, for Moses, the founder of the Hebrew state, to become learned in; and Herodotus, the "father of history," as we are accustomed to style him, found Egypt, when he visited it, already entered upon its period of dotage and decay. It was a strange country: one narrow line of brilliant green, (but spreading fan-like at its northern extremity), traced by the periodical overflow of a single branchless and sourceless river through the great desert which sweeps from the Atlantic coast to the very border of India; so populous and so fertile as to furnish a surplusage of labour, for the execution of architectural works of a solidity and grandeur elsewhere unknown, and which the absolute dryness of the climate has permitted to come down to us in unequalled preservation. On these monuments, within and without, the record-loving Egyptians depicted and described the events of their national and personal history, the course and occupations of their daily lives, their offerings, prayers, and praises, the scenes of their public worship and of the administration of their state, their expeditions and conquests. Their language has thus stood for ages plainly written before the eyes of the world, inviting readers; but the key to the characters in which it was inscribed, the sacred hieroglyphics, had been lost almost since the beginning of the Christian era; until, in our own century, it has been recovered by the zeal and industry of a few devoted men, among whose names that of Champollion stands foremost. The reconstruction of the ancient Egyptian tongue, though by no means complete, is sufficiently advanced to allow us to see quite clearly its general character. It was but an older form of the modern Coptic. The Coptic has itself gone out of existence within the past three or four centuries, extinguished by the Arabic; but we possess a tolerably abundant Christian Coptic literature, representing two or three slightly different dialects, written in an alphabetic character chiefly adapted from the Greek, and dating back to the early centuries of our era. The differences are comparatively slight between the old Egyptian of the hieroglyphical monuments and the later Coptic, for the exceedingly simple structure of the language has saved it from the active operation of linguistic change. A transitional step, too, between the one and the other is set before us in the series of records, mostly in papyrus rolls, which are called hieratic and demotic, from the characters in which they are written, modified forms of the hieroglyphs, adapted to a more popular use: these records come from the last five or six centuries preceding our era, and represent, doubtless, the popular speech of the period.

A number of other African dialects are claimed to exhibit affinities of material and structure with the language of Egypt. They fall^[7] into three groups: the Ethiopian or Abyssinian, of which the Galla is at present the most important member; the Libyan or Berber, extending over a wide region of northern Africa, from Egypt to the Atlantic ocean; and the Hottentot, embracing the dialects of the degraded tribes of Hottentots and Bushmen at the far southern extremity of the continent: these last have been but recently recognized as showing signs of probable relationship with the rest. The family, as thus made up, is styled the Hamitic (by a name correlative to Semitic and Japhetic): its constitution and relations, however, are still matters of no little difference of opinion among linguistic scholars, and can be fully established only by continued research.

The Egyptian was a language of the utmost simplicity, or even poverty, of grammatical structure. Its roots—which, in their condition as made known to us, are prevailingly, though not uniformly, monosyllabic—are also its words; neither noun nor verb, nor any other part of speech, has a characteristic form, or can be traced back to a simpler radical element, from which it comes by the addition of a formative element. Some roots, as in Chinese, are either verb, substantive, or adjective—thus, ankh, 'live, life, alive,' sekhi, 'write, a writing, writer'—others are only verbs or only nouns. A word used as substantive is generally marked by a prefixed article, which is often closely combined with it, but yet is not a part of it; it has no declension, the objective uses being indicated by prepositions. The personal inflection of the verb is made by means of suffixed pronominal endings, also loosely attached, and capable of being omitted in the third person when a noun is expressed as subject of the verb. Mode and tense are, to a certain limited extent, signified by prefixed auxiliary words. But these pronominal endings, which, when added to the verb, indicate the subject (sometimes also the object), have likewise a possessive value, when appended to nouns: thus, ran-i is either 'I name' or 'my name;' it is literally, doubtless, 'naming-mine,' applied in a substantive or a verbal sense according to the requirements of the particular case: that is to say, there is no essential distinction formally made between a noun and a verb. In the singular number of both articles and pronominal suffixes, as also in the pronouns, there is made a separation of gender, as masculine or feminine. This is a highly important feature in the structure of Hamitic speech, and the one which gives it its best claim to the title of form-language. So far as it goes, it puts the tongues of the family into one grand class along with the Indo-European and the Semitic: these three families alone have made a subjective classification of all objects of knowledge and of thought as masculine and feminine, and given it expression in their speech. But, by its general character, the Egyptian is far enough from being entitled to rank with the Indo-European and Semitic languages, being, rather, but a single step above the Chinese: in many of its constructions it is quite as bald as the latter, and sometimes even less clear and free from ambiguity.

The Egyptian pronouns present some striking analogies with the Semitic, and from this fact has been drawn by many linguistic scholars the confident conclusion that the two families are ultimately related, the Egyptian being a relic of the Semitic as the latter was before its development into the peculiar form which it now wears, and which was described in the last lecture. Considering, however, the exceeding structural difference between them, and the high improbability that any genuine correspondences of so special a character should have survived that thorough working-over which could alone have made Semitic speech out of anything like Egyptian, the conclusion must be pronounced, at the least, a venturesome one. Semitic affinities have been not less confidently, and with perhaps more show of reason, claimed for the Libyan and Abyssinian branches of the so-called Hamitic family. Only continued investigation, and more definite establishment of the criteria of genetic relationship, can determine what part of these alleged correspondences are real, and of force to show community of descent, and what part are fancied, or accidental, or the result of borrowing out of one language into another.

To enter in any detail into the labyrinths of African language and ethnography is not essential to our present purpose, and will not be here undertaken. As a consequence of the extraordinary activity of missionary enterprise and of geographical exploration and discovery in Africa within a few years past, much curiosity and study has been directed towards African dialects; a great mass of material has been collected, and its examination has been carried far enough to give us at least a general idea of the distribution of races in that quarter of the world. A vast deal, however, still remains to be done, before the almost innumerable and rapidly changing dialects of all these wild tribes shall be brought to our knowledge, combined into classes and groups, and understood in their resemblances and differences of material and structure.

Apart from the dialects already mentioned, as belonging to the Hamitic or the Semitic family, the best established and most widely extended group of African languages is that one which fills nearly the whole southern part of the continent, from a few degrees north of the equator to the Cape of Good Hope. It is variously called the Bantu, the Chuana, or the Zingian family; or, by a simple geographical title, the South-African. The material as well as structural coincidences between its numerous members are fully sufficient to prove its unity. Its subdivisions, and the separate dialects composing them, need not here be rehearsed.^[8] None of these dialects has any other culture than that which it has received under missionary auspices in the most recent period. They are all of an agglutinative character, forming words of many syllables, and, in a certain way, they are rich enough in forms, and in the capacity of indicating different shades of meaning and relation. Their most marked peculiarity is their extensive use of pronominal prefixes to the nouns; these are numerous—in some languages, as many as sixteen—and distinguish the number and generic class of the nouns to which they are attached. Thus, in Zulu, we have um-fana, 'boy,' aba-fana, 'boys;' in-komo, 'cow,' izin-komo, 'cows;' ili-zwi, 'word,' ama-zwi, 'words,' and so on.^[9] But farther, these same prefixes, or characteristic parts of them, enter into the formation of the adjectives, the possessive and relative pronouns, and the personal pronouns employed as subject or object of the verbs, agreeing with or referring to the nouns to which they respectively belong: for example, aba-fana b-ami aba-kulu, ba tanda, 'my large boys, they love;' but izin-komo z-ami izin-kulu, zi tanda, 'my large cows, they love.' Thus is produced a kind of alliterative congruence, like the rhyming one often seen in Latin, as vir-o optim-o maxim-o, femin-æ optim-æ maxim-æ. Of inflection by cases the South-African noun has hardly any; the case-relations are indicated by prefixed prepositions. Nor is there a personal inflection of the verbs, except by means of prefixed pronouns. Mode and tense are signified chiefly by auxiliary words, also standing before the main root; but in part by derivative forms of the root, made by suffixes: thus, tandile, 'loved,' from tanda, 'love;' and like suffixes form derivative conjugations of the root, in number and in variety comparable with those which, as was shown in the last lecture, come from the Turkish verb: examples are bonisa, 'show,' bonela, 'see for,' bonana, 'see each other,' bonisana, 'show each other,' bonwa, 'be seen,' etc., etc., from bona, 'see.' Except in the interjectional forms, the vocative and second person imperative, every verb and noun in these languages appears in connected speech clothed with a pronominal prefix; so that a prefix seems as essential a part of one of their words as does a suffix of an Indo-European word, in the older dialects of the family.

A very peculiar feature of the phonetic structure of some of the best-known South-African languages, especially of the Kafir branch (including the Zulu), is the use, as consonants, of the sounds called clicks, made by separating the tongue sharply from the roof of the mouth, with accompanying suction—sounds which we employ only in talking to horses or in amusing babies. As many as four of these clicks form in some dialects a regular part of the consonantal system, each being subject to variation by utterance simultaneously with other sounds, guttural or nasal. It is not a little remarkable that the clicks also abound in the tongues of that isolated branch of the Hamitic family, the Hottentot and Bushman, which is shut in among the South-African dialects: indeed, they are conjectured to be of Hottentot origin, and caught by the other tribes by imitation, since they are found only in those members of the different South-African branches which are neighbours of the Hottentots.

Upon the western coast of the continent, the languages of the family of which we are treating extend as far as into the territory of Sierra Leone; but they are much intermingled at the north with other tongues of a different kindred. A broad band across the continent at its widest part, from Cape Verde on the north nearly to the equator on the south, and eastward to the upper waters of the Nile, is filled with dialects not reckoned as South-African, although possessing a structure in many respects accordant with that which we have just described. Conspicuous among them are the Fulah or Fellatah, the Mandingo, and the tongues of Bornu and Darfur. How far they admit of being grouped together as a single family, and what may be the value of their general structural correspondence with the other great African family, must be left for future researches to determine. One of them, the Vei, has an alphabet of its own, of native invention.

Throughout nearly the whole of northern and central Africa, Arabic influence has for some time past been rapidly spreading, carrying with it a certain degree of civilization, the Mohammedan religion, the Koran, and some knowledge and use of the Arabic language. It is only in this quarter of the world that Semitic faith and speech still continue aggressive.

There remains for consideration, of the recognized great families of human language, only that one which occupies the continent of North and South America. Of this, also, we must renounce all attempt at detailed treatment; it is a theme too vast and complicated to be dealt with otherwise than very summarily within our necessary limits. The conditions of the linguistic problem presented by the American languages are exceedingly perplexing, for the same reason as those presented by the Polynesian and African dialects, and in a yet higher degree. The number, variety, and changeableness of the different tongues is wonderful. Dialectic division is carried to its extreme among them; the isolating and diversifying tendencies have had full course, with little counteraction from the conserving and assimilating forces. The continent seems ever to have been peopled by a congeries of petty tribes, incessantly at warfare, or standing off from one another in jealous and suspicious seclusion. Certain striking exceptions, it is true, are present to the mind of every one. Mexico, Central America, and Peru, at the time of the Spanish discovery and conquest, were the seat of empires possessing an organized system of government, with national creeds and institutions, with modes of writing and styles of architecture, and other appliances of a considerably developed culture, of indigenous origin. Such relics, too, as the great mounds which are scattered so widely through our western country, and the ancient workings upon the veins and ledges of native copper along the southern shore of Lake Superior, show that other large portions of the northern continent had not always been in the same savage condition as that in which our ancestors found them. Yet these were exceptions only, not changing the general rule; and there is reason to believe that, as the civilization of the Mississippi valley had been extinguished by the incursion and conquest of more barbarous tribes, so a similar fate was threatening that of the southern peoples: that, in fact, American culture was on its way to destruction even without European interference, as European culture for a time had seemed to be, during the Dark Ages which attended the downfall of the Roman empire. If the differentiation of American language has been thus unchecked by the influence of culture, it has been also favoured by the influence of the variety of climate and mode of life. While the other great families occupy, for the most part, one region or one zone, the American tribes have been exposed to all the difference of circumstances which can find place between the Arctic and the Antarctic oceans, amid ice-fields, mountains, valleys, on dry table-lands and in reeking river-basins, along shores of every clime. Moreover, these languages have shown themselves to possess a peculiar mobility and changeableness of material. There are groups of kindred tribes whose separation is known to be of not very long standing, but in whose speech the correspondences are almost overwhelmed and hidden from sight by the discordances which have sprung up. In more than one tongue it has been remarked that books of instruction prepared by missionaries have become antiquated and almost unintelligible in three or four generations. Add to all this, that our knowledge of the family begins in the most recent period, less than four hundred years ago; that, though it has been since penetrated and pressed on every side by cultivated nations, the efforts made to collect and preserve information respecting it have been only spasmodic and fragmentary; that it is almost wholly destitute of literature, and even of traditions of any authority and value; and that great numbers of its constituent members have perished, in the wasting away of the tribes by mutual warfare, by pestilence and famine, and by the encroachments of more powerful races—and it will be clearly seen that the comprehensive comparative study of American languages is beset with very great difficulties.

Yet it is the confident opinion of linguistic scholars that a fundamental unity lies at the base of all these infinitely varying forms of speech; that they may be, and probably are, all descended from a single parent language.^[10] For, whatever their differences of material, there is a single type or plan upon which their forms are developed and their constructions made, from the Arctic Ocean to Cape Horn; and one sufficiently peculiar and distinctive to constitute a genuine indication of relationship. This type is called the incorporative or polysynthetic. It tends to the excessive and abnormal agglomeration of distinct significant elements in its words; whereby, on the one hand, cumbrous compounds are formed as the names of objects, and a character of tedious and timewasting polysyllabism is given to the language—see, for example, the three to ten-syllabled numeral and pronominal words of our western Indian tongues; or the Mexican name for 'goat,' kwa-kwauh tentsone, literally 'head-tree (horn)-lip-hair (beard),’ or 'the horned and bearded one'—and, on the other hand, and what is of yet more importance, an unwieldy aggregation, verbal or quasi-verbal, is substituted for the phrase or sentence, with its distinct and balanced members. Thus, the Mexican says "I-flesh-eat," as a single word, compounded of three elements; or if, for emphasis, the object is left to stand separate, it is at least first represented by a pronoun in the verbal compound: as, "I-it-eat, the flesh;" or "I-it-him-give, the bread, my son," for "I give my son the bread."

The incorporative type is not wholly peculiar to the languages of our continent. A trace of it (in the insertion, among the verbal forms, of an objective as well as a subjective pronominal ending) is found even in one of the Ugrian dialects of the Scythian family, the Hungarian; and the Basque, of which we shall presently speak more particularly, exhibits it in a very notable measure. It is found, too, in considerably varying degree and style of development in the different branches of the American family. But its general effect is still such that the linguist is able to claim that the languages to which it belongs are, in virtue of their structure, akin with one another, and distinguished from all other known tongues.

Not only do the subjective and objective pronouns thus enter into the substance of the verb, but also a great variety of modifiers of the verbal action, adverbs, in the form of particles and fragments of words; thus, almost everything which helps to make expression forms a part of verbal conjugation, and the verbal paradigm becomes well-nigh interminable. An extreme instance of excessive synthesis is afforded in the Cherokee word-phrase wi-ni-taw-ti-ge-gi-na-skaw-lung-ta-naw-ne-li-ti-se-sti, 'they will by that time have nearly finished granting [favours] from a distance to thee and me.'^[11]

Other common traits, which help to strengthen our conclusion that these languages are ultimately related, are not wanting. Such are, for example, the habit of combining words by fragments, by one or two representative syllables; the direct conversion of nouns, substantive and adjective, into verbs, and their conjugation as such; peculiarities of generic distinction—many languages dividing animate from inanimate beings (somewhat as we do by the use of who and what), with arbitrary and fanciful details of classification, like those exhibited by the Indo-European languages in their separation of masculine and feminine; the possession of a very peculiar scheme for denoting the degrees of family relationship; and so on.

As regards their material constitution, their assignment of certain sounds to represent certain ideas, our Indian dialects show, as already remarked, a very great discordance. It has been claimed that there are not less than a hundred languages or groups upon the continent, between whose words are discoverable no correspondences which might not be sufficiently explained as the result of accident. Doubtless a more thorough and sharpsighted investigation, a more penetrating linguistic analysis and comparison—though, under existing circumstances, any even distant approximation to the actual beginning may be hopeless—would considerably reduce this number; yet there might still remain as many unconnected groups as are to be found in all Europe and Asia. It is needless to undertake here an enumeration of the divisions of Indian speech: we will but notice a few of the most important groups occupying our own portion of the continent.

In the extreme north, along the whole shore of the Arctic ocean, are the Eskimo dialects, with which is nearly allied the Greenlandish. Below them is spread out, on the west, the great Athapaskan group. On the east, and as far south as the line of Tennessee and North Carolina, stretches the immense region occupied by the numerous dialects of the Algonquin or Delaware stock; within it, however, is enclosed the distinct branch of Iroquois languages. Our south-eastern states were in possession of the Florida group, comprising the Creek, Choctaw, and Cherokee. The great nation of the Sioux or Dakotas gives its name to the branch which occupied the Missouri valley and parts of the lower Mississippi. Another wide-spread sub-family, including the Shoshonee and Comanche, ranged from the shores of Texas north-westward to the borders of California and the territory of the Athapaskas; and the Pacific coast was occupied by a medley of tribes. Mexico and Central America, finally, were the home of a great variety of tongues, that of the cultivated Aztecs, with its kindred, having the widest range.

The linguistic condition of America, and the state of our knowledge respecting it, being such as we have here seen, it is evident how futile must be at present any attempt to prove by the evidence of language the peopling of the continent from Asia, or from any other part of the world outside. We have already noticed that a relationship is asserted to exist between the Eskimo branch of American language and a dialect or two in the extreme north-east of Asia; but the fact that it is a specifically Eskimo relationship is sufficient to prove its worthlessness as a help to the explanation of the origin of American language in general, and to make it probable that the communication there has been from America to Asia, and not the contrary. To enter upon a bare and direct comparison of modern American with modern Asiatic dialects, for the purpose of discovering signs of genetic connection between them, would be a proceeding utterly at variance with all the principles of linguistic science, and could lead to no results possessing any significance or value. One might as well compare together the English, the modern Syriac, and the Hungarian, in order to determine the ultimate relationship of the Indo-European, Semitic, and Scythian families. Sound method (as was pointed out in the sixth lecture) requires that we study each dialect, group, branch, and family by itself, before we venture to examine and pronounce upon its more distant connections. What we have to do at present, then, is simply to learn all that we possibly can of the Indian languages themselves; to settle their internal relations, elicit their laws of growth, reconstruct their older forms, and ascend toward their original condition as far as the material within our reach, and the state in which it is presented, will allow; if our studies shall at length put us in a position to deal with the question of their Asiatic derivation, we will rejoice at it. I do not myself expect that valuable light will ever be shed upon the subject by linguistic evidence: others may be more sanguine; but all must at any rate agree that, as things are, the subject is in no position to be taken up and discussed with profit. The absurd theories which have been advanced and gravely defended by men of learning and acuteness respecting the origin of the Indian races are hardly worth even a passing reference. The culture of the more advanced communities has been irrefragably proved to be derived from Egypt, Phenicia, India, and nearly every other anciently civilized country of the Old World: the whole history of migration of the tribes themselves has been traced in detail over Behring's Straits, through the islands of the Pacific, and across the Atlantic; they have been identified with the Canaanites, whom Joshua and the Israelites exterminated; and, worst of all, with the ten Israelitish tribes deported from their own country by the sovereigns of Mesopotamia! When men sit down with minds crammed with scattering items of historical information, abounding prejudices, and teeming fancies, to the solution of questions respecting whose conditions they know nothing, there is no folly which they are not prepared to commit.

Our national duty and honour are peculiarly concerned in this matter of the study of aboriginal American languages, as the most fertile and important branch of American archæology. Europeans accuse us, with too much reason, of indifference and inefficiency with regard to preserving memorials of the races whom we have dispossessed and are dispossessing, and to promoting a thorough comprehension of their history. Indian scholars, and associations which devote themselves to gathering together and making public linguistic and other archæological materials for construction of the proper ethnology of the continent, are far rarer than they should be among us. Not a literary institution in our country has among its teachers one whose business it is to investigate the languages of our aboriginal populations, and to acquire and diffuse true knowledge respecting them and their history.^[12] So much the more reason have we to be grateful to the few who are endeavouring to make up our deficiencies by self-prompted study, and especially to those self—denying men who, under circumstances of no small difficulty, are or have been devoting themselves to the work of collecting and giving to the world original materials. The Smithsonian Institution has recently taken upon itself the office of encouraging, guiding, and giving effect to the labours of collectors, under special advantages derived from its relation to the Government, with laudable zeal, and with the best promise of valuable results. No department of inquiry, certainly, within the circle of the historical sciences, has a stronger claim upon the attention of such a national institution; and it becomes all Americans to countenance and aid its efforts by every means in their power.

Before closing this cursory and imperfect survey of the varieties of human language, we have to glance at one or two dialects or groups of dialects which have hitherto resisted all attempts at classification. Most noteworthy among these is the Basque, spoken in a little district of the Pyrenees, on both sides of the border between France and Spain, enveloping the angle of the Bay of Biscay, between Bayonne and Balbao. The Basques are well identified as descended from the primitive Iberian population which is supposed to have filled the Spanish peninsula before the intrusion of the Celts: their stubborn and persistent character and the inaccessibility of their mountain retreats have enabled their native idiom successfully to resist the assimilating influences exercised by successive Celtic, Roman, and Gothic conquest and domination. It stands, so far as is yet known, alone among the languages of mankind; kindred has been sought and even claimed for it in every direction, but to no good purpose. It is, then, naturally enough conjectured to be a sole surviving remnant of the speech of an aboriginal race, peopling some part of Europe before the immigration of the Indo-European tribes, perhaps before that of the Scythian; and the possibility that it may be so invests it with an unusual degree of interest. Its structure is exceedingly peculiar, intricate, and difficult of analysis. As we have already had occasion to notice, it possesses much more striking analogies with the aboriginal languages of America than with any others that are known: like them, it is highly polysynthetic, incorporating into its verbal forms a host of pronominal relations which are elsewhere expressed by independent words; like them, also, it compounds words together by representative fragments. But it does not show the same tendency to fuse the whole sentence into a verb; its nouns have an inflection which is much more Scythian than American in type; and there are other differences which distinctly enough discourage the conjecture that it can be historically akin with the tongues of this continent. Some other among the various populations of southern Europe, treated by the ancients as of strange tongue and lineage, and which have now totally disappeared, may possibly have been akin with the Basques: such questions are covered with a darkness which we cannot hope ever to see dispelled.

In Italy are still found the relics of one of these isolated and perished peoples, the Etruscans. They were a race of much higher culture than the Basques, and their neighbourhood to Rome, and their resulting influence, peaceful and warlike, upon her growing polity and developing history, give them a historical importance to which the Iberian race can lay no claim. Inscriptions in their language, written in legible characters, and in some instances of assured meaning, are preserved to our day; yet its linguistic character and connections are an unsolved and probably insoluble problem. Every few years, some one of those philologists whose judgments are easily taken captive by a few superficial correspondences claims to have proved its relationship with some known family, and thus to have determined the ethnological position of the race that spoke it; but his arguments and conclusions are soon set aside as of no more value than others already offered and rejected.

Again, there is found in the mountain-range of the Caucasus a little knot of idioms which have hitherto baffled the efforts of linguistic scholars to connect them with other known forms of speech. Their principal groups are four: the Georgian and the Circassian stretch along the southern and northern shores respectively of the eastern extremity of the Black Sea, and through the mountains nearly to the Caspian; the Lesghian borders the Caspian; and the Mitsjeghian lies between it and the Circassian. The Georgian possesses a peculiar alphabet and a literature; but the whole group, except as it presents a problem for the solution of the linguistic ethnographer, has no special importance.

The Albanian or Skipetar, the modern representative of the ancient Illyrian, has already been spoken of as doubtfully classifiable with the Indo-European languages. If its connection with them shall not finally be made out to the satisfaction of the learned, it, too, will have to be numbered among the isolated and problematical tongues.

One more Asiatic dialect may be worth a moment's notice: the Yenisean, occupying a tract of country along the middle course of the Yenisei, with traces in the mountains about the head waters of that river; it belongs to the feeble and scanty remnant of a people which is lost in the midst of Scythian tribes, and apparently destined to be ere long absorbed by them, but which is proved to be of different race by its wholly discordant language.

The number of such isolated tongues is, of course, liable to be increased as we come to know more thoroughly the linguistic condition of regions of the world which are as yet only partially explored. There is a possibility that many types of speech, once spread over wide domains, may exist at present only in scanty fragments, as well as that some may have disappeared altogether, leaving not even a trace behind.

Notes

1. This is the opinion of Caldwell, from whose excellent Comparative Grammar of the Dravidian Languages (London, 1856) are mainly derived the materials for this account of the family.
2. This example is taken from Schleicher's Languages of Europe in Systematic Review (Bonn, 1850), p. 51.
3. See the twelfth lecture, where this peculiarity of the Chinese mode of writing will be more fully explained.
4. Among these, Rev. N. Brown and Mr. B. H. Hodgson have especially distinguished themselves.
5. Steinthal, Charakteristik etc., page 165.
6. See Von der Gabelentz, Die Melanesischen Sprachen, etc., in vol. viii. (1861) of the Memoirs of the Saxon Society of Sciences.
7. I follow here the classification of Lepsius, given in the second edition of his Standard Alphabet (London and Berlin, 1863), at p. 303.
8. See Lepsius's General Table of Languages, already referred to; and Dr. Bleek's Catalogue of Sir George Grey's Library, at Capetown, 1858.
9. Our examples are taken from Rev. L. Grout's "Zulu-Land" (Philadelphia, 1864), chap. xiv.
10. I make no account here of isolated dialects of an exceptional character, like the Otomi in central Mexico, which is asserted to be a monosyllabic language; nor of others which may exhibit the characteristic features of American speech so faintly, or in such a modified form, as to be hardly recognizable by their structure as American: it remains yet to be determined whether such seeming exceptions do or do not admit of explanation as the result of special historical development. Nor, of course, is the possibility denied that fuller knowledge will bring to light tongues radically and irreconcilably discordant from the general type.
11. A. Gallatin in Archæologia Americana, vol. ii. (Cambridge, 1836), p. 201.
12. This reproach, at least, is about to be removed, by the establishment of a chair of American archæology at Cambridge.