THE ETHNOLOGICAL RESULTS OF DR. LIVINGSTON'S AFRICAN EXPLORATIONS.

It would be too late, even if it were not presumptuous, to offer our readers any general notice of Dr. Livingston's Travels, after the work has been reviewed in almost every English paper. The missionary's character-simple, earnest, robust and determinedhis geriality and observation, give a sustained power and charm to his homely and graphic descriptions of African scenery and life. He reminds us of the older and less literate race of travellers, and the minuteness and loving fidelity of his pictures of animals and plants are almost worthy of Dampier. With little of the roving

voyager's imagination, he has the singular advantage of complete familiarity with the native tribes of the territory he describes, and a mastery of one of the principal languages. The most barbarous people are always found, on close acquaintance, to be possessed of much knowledge of the climate and vegetation, and of the habits of the fauna, of the country occupied by them; and the traveller who enters into their ideas is sure to obtain many valuable and curious facts, amid the mass of puerile and superstitious notions with which their repertory of science is stuffed. Immense as are the additions which Dr. Livingston has made to our knowledge of the interior of southern Africa, the traveller himself is, for the time, the greatest object on the map, and we follow his steps with a constantly increasing admiration of the man. His journies were a prolonged battle not merely with all the ordinary difficulties, hardships and dangers which beset pioneers in barbarous countries, but with the misery and prostration of sickness. We must place him even before those explorers of the N. W. passage who have suffered most, for he voluntarily and singly encountered certain disease and probable death. He is the most indomitable and one of the most sagacious in that long roll of geographical explorers who have gained as much fame for the English race as the colonies we have planted. The light which his labours have thrown on the physical geography of Africa has been universally acknowledged, and Sir R. Murchison, Mr. Macqueen and other scientific men have shown how it affects our previous conceptions of the

What was

character of the interior of the southern peninsula. formerly matter of induction or speculation is now certainty. A bold generalisation of Murchison's has become an observed fact. The water-shed between the great rivers of the east and west coasts-the Zambesi and the Congo-is not a lofty snow-covered mountain chain, but a shallow lake, from the two extremities of which the water flows in opposite directions. Southern Africa is thus an island in the Arabic sense. A similar phenomenon, on a small scale, is found in the south of the Malay peninsula, where the Rio Formosa and the Indau drain the same marsh, so that, in the rainy season, a canoe might be paddled across the Peninsula from sea to sea.

The ethnological results of Dr. Livingston's journies do not appear to have been specially noticed, like the geographical and geological, and we shall therefore draw attention to a few of them. All eyes are now attracted to his return journey up the Zambesi and Lieutenant Burton's attempt to reach Lake Maravi from Zanzibar, and it will be useful to glance back at our knowledge of that branch of the South African tribes among whom their routes will lie. We may first briefly indicate the most recent views that have been taken of African ethnology as a whole.

The Caucasian, Semitic and African languages have a common primary basis, and the two last constitute one family. The African division has two main branches which have been termed the Libyan and the Zimbian. The former is closely connected with Semitic. Its best representatives are Hottentot and Egyptian; but most of the northern and many of the middle languages are mainly Libyan. The Zimbian branch, in its purest condition, comprises all the known languages of Africa south of the Line or within one or two degrees north of it, except the Hottentot. On the N. W. it extends into the middle region along the coast and far over the Niger-Chadda province. In the N. E. a Libyan group, the well known Galla-Saumali, has a strong Zimbian element. In the Nilo-Nigerian zone a sub-formation, which has been termed the Nilotic as it is most strongly marked in the Nubian of the upper Nile, presents some peculiarities of a Scythic character, and its influence is traceable westward to the Mandingo group. From the number of distinct Scythic words in the vocabularies, it is possible that this sub-formation was induced by a real incorporation of an invading Scythic horde from Arabia with Abysinnian tribes, at a period when the African languages had not become flexional. But as the Semito-African and Caucasian alliance, when considered as a whole, shows the same primitive oscillation between a postpositional and a prepositional structure that appears in Himalaic, N. E. Scythoid and American, and as Scythic words are

found abundantly in all the peninsular lines of language that radiate from Upper Asia, it is most probable that, in its basis, the Nilotic sub-formation is simply a very archaic variety of the Semito-African family. Lastly, the western Semitic languages, in all their later stages, have influenced the North African. The Zimbian branch is very broadly distinguished from the Libyo-Semitic by its present form. It is greatly more agglomerative, luxuriant and complex, but it is now almost demonstrated that the Libyo-Semitic groups are not so much imperfectly developed, as impoverished and concreted, forms of the same archaic type. To those who are only familiar with such a family as the IndoEuropean, it is difficult to convey a correct idea of these larger and older families, which, while single in basis, have been more richly and variously developed.* The Semito-African family, while one in roots and ultimate structural tendencies, embraces several subfamilies, the members of each of which have the close and special relationship of the Indo-European languages. The relationship that subsists between the Chinese, the Tibeto-Burman and the Mon Anam tongues presents many points of resemblance. From ages long anterior to history the Chinese have stood apart from the ruder western tribes of the family, who are separated from them by difficult masses of mountains. The improved race has sought out and influenced its more barbarous neighbours, but has been very slightly influenced by them in return. In like manner the Semitic nations have been isolated from the African, but have always played an important part in African history. Much of the Chinese vocabulary, including the numerals, is found in the languages of Tibet and Ultraindia and in the ancient ones of northern India, but generally in older and more varied forms than the present Chinese dialects possess. Precisely similar is the relationship between the Semitic and the African vocabularies. The Semitic numerals, for example, in their existing form, are but a fragmentary and dialectic remnant of an archaic system much of which is better preserved by African tribes, and which can be restored completely and consistently by a comparison of all the dialectic forms of it now extant. For instance little or nothing can be made of 7 in the Semitic sa.ba-ta, se.ba-t &c, (whence our "sabbath" and the "saptu" of our Malay friends) and the Zimbian sa.mba, ta.be &c., except that they have a certain degree of resemblance to each other, which becomes closer when the Semitic masculine suffix is removed. But other African dialects show that the word is compounded of 5 and 2,-e. g. the Ndob sa.mbe is from sa.n 5 and mbe 2, the Yasgua to-mva from

* See p. 92.

nto 5 and mva 2, the Mfut ta be from nta or ta 5 and be 2,-and some of the Zimbian still use the ancient uncompounded and unconcrected names, those of the tribes near the estuary of the Zambesi having such names as m-za.na zi vi.ri, ta.nu na be.li, literally "five and two," ta.nu 5, and be-li &c. 2, retaining the archaic postfixes, while the Mosambique tha.na, za-na like the Chadda dsa-n, tsi-n, se-na &c. illustrate the oscillation between t, d and s, z. When the Hebrew and Arab race first colonised Africa it must have had similar names, and the change from za.na zi vi-ri (ba-ri &c. in other dialects) to sa.ba is the phonologic measure of the chronologic interval between the Kafir and the Arabic forms of speech, and possibly also of the physiological departure of the Kafir man from the Arab or the Arab from the Kafir. In the same manner the Peguan, the Karen and the Mikir numerals preserve the (hinese in older and fuller forms, and carry us back to a period which was ancient when Yu drained the swampy basin of the Yellow River. When we turn to the structure of the various groups that compose the great eastern and western families of the Old World, a range of differentiation is perceived which amazes and perplexes the orthodox ethnologist of the Berlin and London schools. He is called upon to admit an identity in roots, even to the pronouns and native numerals, as absolute as in English and Sanskrit, with a variance in structure greater than that between Turkish and Greek. In this respect the Chino-Himalaic and the Semito-African families present the same kind of facts. Our systems must expand with our knowledge. We cannot force the Siamese, the Tibetan and the Chinese-the Syriac, the Bornui, the Sechuana and the Hottentot,-into moulds of the Indo-European dimensions. But they have a close family agreement nevertheless, and there is no more reason that every large linguistic alliance, should have the same limits of agreement and divergency, than that every fixed star should be a counterpart of the solar system, or that every family of plants or animals should be restricted to precisely the saine range of variation. The truth is that the definition of the word family is inconstant even when applied to the Indo-European brotherhood. It meant one thing when English had not diverged more from the Sanskritic type than Greek has. It means another thing when English has lost most of the ancient flexions. It means one thing to those who admit Bengali into the circle, and another to those who, in doing so, mark it with a bar sinister, or who exclude it altogether. What it may mean at the end of the next thousand years when all the living languages may have changed as much as they have done within the last milleneum, no grammatical prophet can foretell. The truth is that the ethnologist has to deal with genealogies. There

may be a real kinship between groups of tribes, although the consanguinity of every two dates from a different generation. The Indo-Europeans, at various removes backward, may enter into the same family with the Scythians, with the Caucasians, with the Semito-Africans, with the Draviro-Australians, with the Americans, with the Tibetans and Chinese. It is in every cas and at every step a pure question of genealogy. The links cannot be discovered by any less humble and painful method than a patient comparison of facts, but when found, our technical words, which are but the working symbols of the day, must accommodate themselves to our enlarged knowledge by receiving a larger meaning. Dr. Livingston's journies have lain exclusively within the southern portion of the Zimbian area. The coast tribes were more or less known, from the eastern Zumba (Suaheli of the Arabs)— who, in about 1° 30 N., succeed the Saumali, an E. Nilotic tribeto the eastern Mpongwe of the Gabun, whose dialect of pure Zimbian connects the great southern family with the languages of the Guinea coast and the Niger.

The southern division of the Zimbian, consists of three principal groups of tribes,-(1) the so called Kafir of the S. E., known to the Chuanas as Matebele or Makon-kobe, speaking three dialects, the Kosah, the Zulu and the Fingoe, (2) the Chuana,-and (3) the Damara. The Kafir tribes occupy the eastern belt of the continent' nearly up to the Zambesi. Several coast tribes of a more negro character, whom they appear to have preceded, are now subject to them. The coast dialects are imperfectly known, but they are all Zimbian. The eastern Chuana succeed them, the southern tribes under Moshesh being termed Ba-Suto and the northern Ba-Koni. The Chuana tribes do not extend to the southern extremity of the continent like the Kafirs. Their most southern limit is considerably to the north of the Orange River. 'Their northern extension and distribution were unknown prior to Dr. Livingston's researches. The western Chuana are the tribes in and near the Kalahari desert. The Damara, who have recently been well described by Galton, occupy a tract to the N. W. of the desert. They are succeeded by the Western or Kongo-Mpongwe division. The eastern division, ranging from the Zambesi to Barawa (13° N. L.) in the Saumali territory, consists of coast and insular tribes whose dialects are more or less known. They are more closely related to the Kongo than to the southern division. The inland tribes, with the exception of some within 50 to 400 miles of the coast, whose vocabularies have been obtained, are very imperfectly known, no European having explored the interior; but from the affinity between the dialects of the opposite coasts, it is probable that the inner ones are embraced in the same relationship.