Chapter III: Introduction: By Sir Harry Johnston, G.c.m.g., K.c.b

_Sometime H.M. Commissioner, &c., for Northern Zambezia._

Barotseland at the present day is mainly defined as the kingdom of the recently dead Lewanika which lies both to the east and west of the Upper Zambezi, north of Sesheke, the Katima Rapids, the “Caprivi” boundary line of former German South West Africa. The western boundary line of Barotseland is the east bank of the Kwando river up stream to its intersection with the 22nd degree of E. longitude; the northern boundary is the 13th degree of S. latitude; and the eastern limit is a line drawn from the 13th degree of S. latitude to the upper waters of the Kabompo-Lulafuta, and then southward to the Majili river and along the Majili down to its union with the Upper Zambezi near Sesheke.

The master people, still, in this Upper Zambezi state are the A-luyi, who seem to have entered the lands of the Upper Zambezi from the direction of Eastern Angola, if vague tradition be anything to rely on. At any rate the A-luyi became the dominant tribe in this region some three hundred years ago, if not much earlier. Prior to their dominancy the Upper Zambezi regions had been invaded by terrible armies of cannibals, the “Bazimba” of Portuguese East African records and the “Giagas,” “Jaggas” of Portuguese Congo and Angola history. The term “Jagga” seems to have been derived from the title Jaga, which they gave to their chiefs; amongst themselves this tribe or congeries of tribes which played such an amazing part in the history of Central Africa in the latter half of the 16th century was known either as the “Imbángala” of the middle course of the Kwango river or the Va-chibokwe, Va-kioko, Va-chokwe, Bajoko (according to Livingstone), or Ba-jok of South-west Congoland and the Kasai sources. This boiling over of the Va-chokwe—as they are nowadays termed in their original home—caused terrific, population-destroying raids to be made across Northern Angola into Luango and the Congo coast region, and southwards over Barotseland, Central Zambezia, Southern Nyasaland and Moçambique, and northwards up the East Coast to the surroundings of Mombasa. The Va-chokwe were known by several names in the Portuguese records, the Jagga, Ba-zimba, and Bambo (_sing._ “Mumbo.”) “Ba-zimba” seems to have died out as a tribal name, but the Bambo still inhabit the region through which the Lower Shire flows, and the original Va-chokwe, Va-kioko, Va-chibokwe, or Ba-jok are prominent inhabitants in the basin of the extreme upper Kasai, and come into Barotseland or Angola on trading expeditions. They are a singularly independent, quarrelsome, warlike people, not unlikely to give trouble yet to the Belgian controllers of Congoland, to the British peace-keepers in North-West Zambezia, or to the Portuguese in Eastern Angola.

But though Upper Zambezia was traditionally overrun by the Va-chokwe three hundred and fifty years ago, the A-luyi seem soon afterwards to have absorbed the invasive quota of peoples and to have settled down as the rulers of Upper Zambezia till their land was first invaded from the south about one hundred and twenty years ago.

This invasion took place from the direction of Lake Ngami and the invaders seem to have been a section of the Ba-hurutse division of the Bechuana people. They reached the Upper Zambezi—traditionally—at the end of the 18th century, whether as friends or foes, tradition does not say; probably in small numbers and with no racial feud against the A-luyi. Their descendants may be the “Njenji” or “Zinzi” tribe who still speak a dialect of Sechuana and are settled rather high up the Zambezi. They seem however to have long retained the tribal name—Ba-hurutse—which became shortened into Ba-rotse.

These Ba-hurutse colonists of the Upper Zambezi valley apparently preserved some slender connection with the remainder of the great area of Bechuanaland, and had allowed news to reach the Fatherland of their tribe regarding the well-watered region in which they had made a new home. At any rate Sebituane, the son of an erratic woman chief in Basutoland, who in the second decade of the 19th century led her people forth on a mad excursion, gradually found his way to the north-west, and finally at the close of the eighteen-twenties had crossed the Zambezi and brought his followers—now called the Makololo—to the conquest of its upper valley.

By about 1840, he had conquered the A-luyi, whom he and his people called the “Ba-rotse” (Ba-hurutse), by the name of the earlier Bechuana colonists. Livingstone found him a fine-looking, copper-coloured man. “He was far and away the finest Kafir I ever saw,” wrote W. C. Oswell at the time—1851; and many years afterwards (in 1890) he repeated the same thing to me. “Sebituane is a gentleman in life and manner” exclaimed his surprised guests in 1851, when they appreciated to the full his gracious and thoughtful hospitality.

One feels on reading the remembrances of Livingstone and Oswell how intense must have been their regret when a few days after their arrival this great chief fell ill with inflammation of the lungs and passed away. Sebituane’s last days of life were taken up with recounting in a subdued voice (he was suffering from an old wound in the lungs) his wonderful adventures since he had first reached the Zambezi and found his way into Barotseland as a ruler. To the east of his moving horde of Bechuana or Makololo, were the fierce Tebele Zulus under Umsilikazi or Mosilikatse,[1] who likewise strove to conquer for themselves a state in North Zambezia. The pressure of the Amandebele drove Sebituane to the more swampy regions of “Barotseland.”

After Sebituane’s death Livingstone and Oswell left his country, promising to return. His son Sekeletu—inferior to his splendid father in height, physique, and appearance—was placed on the throne and ruled Barotseland for some twelve years. Livingstone had returned to the Upper Zambezi in 1853, determined then to discover the whole secret of its course and of its main tributaries, and to follow up the trail of the Ba-joko traders and the “Mambari” or half-caste Portuguese, who were beginning to renew their trading enterprize with the Upper Zambezi, and ascertain if through Sekeletu’s kingdom they could find their way “á contra costa”—to Portuguese Zambezia and Moçambique.

Livingstone traversed Barotseland (or Uluyi, as it was still more anciently called). Makololo porters and soldiers—some of whom were really A-luyi subjects of Sekeletu—accompanied him to São Paulo de Loanda, and turned back with him from Portuguese civilization to deliver him safely again at Sekeletu’s capital, whence, after many adventures on the Tonga highlands and along the cataract-strewn Zambezi, he reached that river’s delta and the settled towns of the Portuguese.

He returned in 1858 and, accompanied by Charles Livingstone and John Kirk, he once more entered Barotseland and sat with Sekeletu and his subordinate chiefs. But Livingstone had become by then more deeply interested in the problems, geographical and political, of Lake Nyasa and the Shiré river, the ravages of the Yao and Arab slave-traders, the courses of rivers west and north of the Nyasa watershed. Barotseland simmered on; scarcely visited by any European—unless it was some stray hunter who had found his way to the Zambezi. At last, in the year 1878, the pioneers of the great French Protestant Mission—notably Monsieur François Coillard—came into the Lake Ngami basin from Bechuanaland, and thence travelled to Upper Zambezi.

They found this former kingdom of Sebituane and Sekeletu ruled intermittently and disturbedly by a young chief of the old Aluyi dynasty—_Lewanika_. Sekeletu had died in 1864, and a period of struggle then arose in which the Aluyi and allied peoples, with possibly the remnant of the old settlement of the “Barotse” (Ba-hurutse), overcame the fighting caste of the Makololo, killed the males and married the females; and at the end of the struggle—about 1870—-the former dynasty of Aluyi chiefs was re-established.

The French Protestant missionaries (some of whom were Swiss) did much to keep Lewanika and his people in touch with British South Africa, after they were well-established in his country. He was made aware, at the close of the ’eighties, of German ambitions, of Portuguese desires to “protect” his territory—the journeys of Serpa Pinto and Capello and Ivens (1878-1883) had kept him advised of this; and at the same time of the formation of a great Chartered Company which had come to terms with the Matebele Zulus and was extending British political influence over the regions north of the Zambezi.

The “Barotse” people had heard something about the Arabs far back in the 19th century, for Livingstone found an intelligent, enquiring, civil-spoken Arab staying at or near the Barotse capital on the Zambezi in 1855.

Lewanika came to hear of the war with the Arabs in Nyasaland and on Tanganyika into which Germans and English were led, and his sympathies lay with the Europeans. Though Sekeletu had been a slave-trader on rather a considerable scale, Lewanika held such a policy in abhorrence. In 1891 he came to a preliminary understanding with the Chartered Company which has been strengthened by later agreements. In 1902, Lewanika, who had been born and brought up in the heart of Central Africa, travelled _via_ Cape Town to England and was present at the coronation of King Edward VII. He looked a fine and imposing figure in Westminster Abbey, where I—recently returned from Uganda—was presented to him, after the ceremony was over. He had, indeed, as the author of this book says in his introduction, “most charming and courteous manners.” Further, it should be added that in regard to the kingdom of the Barotse, the British South Africa Chartered Company has taken no false step, has incurred no unfavourable criticism, and has received praise for its control of native customs and defence of native rights from the French Protestant and the Church of England missionaries at work in this country and its neighbour states.

* * * * *

The Barotse country, as defined on the sketch map accompanying this book, includes a number of interesting and distinct Bantu languages. It is even possible that a traveller who pushed his explorations to the south-western limits of the kingdom and of the British sphere might find in the valley of the Kwando river a few nomad Bushmen of more normal stature than the stunted Bushmen of Cape Colony, but speaking a tongue of Bushman and not Bantu affinities. The Bushmen were more anciently inhabitants of this land than the Bantu Negroes; but not far away, to the east of Barotseland, we have had a sensational discovery within the last twelve months of a different and peculiar species of man which once inhabited South Central Africa, prior to the penetration thither of the Negro sub-species. This was _Homo rhodesiensis_, the skull, jaw, and limb-bones of which were found in close association with rudely-chipped quartz knives and scrapers, and with the broken bones of antelopes still living in Rhodesia. The place of discovery was some sixty feet below the surface in a cave at the Broken Hill Mine almost in the middle of Northern Rhodesia. The bones found are attributable to at least two personages, the limb-bones and the skull having belonged to a tall man perhaps not less than six feet in height. He possessed tremendously-developed brow ridges, a feature in which Negro man is as a rule more deficient than the European. These, and the large flat face and comparatively small front teeth and other features, suggest an affinity with Neanderthal man (_Homo primigenius_) of Europe. Other points show some resemblance to the black Australian. But the brain-case measured in space only about 1,280 cubic centimetres—a capacity much inferior to the ordinary brain-cases of the Neanderthaloids and Australoids, though, of course, there are occasional examples of Australoid women’s skulls that have a brain capacity of under 1,000 cubic centimetres.

These bony remains of Rhodesian man may have been living creatures as much as forty thousand or as little as ten thousand years ago: there is not enough surrounding evidence to fix the date of them more precisely. All we can say is that the bones of the beasts they hunted and ate are nearly if not quite identical with those of existing species. Probably at the time when this big-browed, gorilla-faced man was alive, the Upper Zambezi, which flows through Barotseland three or four hundred miles westward of Broken Hill Mine, formed a great longitudinal lake in the heart of the Barotse[2] country, the greater part of which still remains very swampy. And when it issued from this expansion it joined the Kwando, the Okavango, and the Zuga to form a vast lake in what is now the North Kalahari Desert. The Guay river was then the Upper Zambezi and the Kafue river joined it as now.

The Kalahari Desert which well nigh ruins the north-western part of South Africa is probably “younger” as a desert even than the Sahara—and the Sahara Desert is a thing of yesterday, possibly much more recent than the existing human species. In the days when Rhodesian man with the huge brow-ridges, long, ape-like face and poorly developed brain ranged across all of Northern Rhodesia that was not under water, he might have passed between the vast Zambezi Lake on the west and the sources of the Guay on the east, and have wandered down into a green, tree-besprinkled South Africa, where he was no doubt killed out by the smaller but far more cunning Bushman or the intelligent Strandloopers who painted pictures on the rock surfaces like their relatives of Europe, the Crô-Magnon men. Long after these antecedent types of invaders had died away came in the vigorous Negroes...? Two thousand years ago? ... who spoke Bantu languages. Whether, between the Strandloopers, Bushmen and Hottentots who invaded the southern portion of South Africa thousands—who yet can say how many?—of years ago and the Bantu Negroes of yesterday (so to speak), other races entered and dwelt in southern and South-Central Africa, we do not know. I have given reasons elsewhere for premising that the flocking south from Equatoria of the Zulus, Bechuana, Karaña, Nyanja, Tonga, Luyi, Hérero, Angola, Luba, Lunda, Yao and Makua tribes and their allies and embranchments was quite a recent episode in the history of Africa, perhaps not more than two thousand years old. We have at present absolutely no knowledge of the incoming of any other type of inhabitant later than the Hottentot and before the Bantu; though there may have been many Negro immigrants belonging to neither group linguistically.

Barotseland, as its recent past becomes revealed to us from the second half of the sixteenth century, seems to have had as a dominant people the _Aluyi_; and as other tribes of importance the _Tonga_ group (Ila-Tonga-Subia) in the south and south-east, the _Luena_ or _Lubale_, _Mbunda_, and _Lujazi_ tribes of northern Barotseland and the adjoining parts of Angola, the _Nkoya-Mbwela_ peoples of eastern Barotseland; and a section of the southern _Luba_ folk. Sporadically there has also occurred a strong invasion of the _Lunda_ people (Ma-bunda) into the northernmost basin of the Zambezi, but it is doubtful whether these immigrants penetrate into the political limits of Barotseland. Similarly a section of another vigorous South Congo people has colonized the Northern Zambezi basin; the Luba, who are known in North-Eastern Barotseland as the _Kahonde_ or _Kaondi_. Then again the enterprising, uppish, cheeky _Va-chokwe_ or “Ba-joko” traders—once the slave-traders of South Congoland and Eastern Angola—circulate through Northern Barotseland and add yet another type of Bantu language to its markets and meetings.

The dominant language at Court and in trade to-day is an almost artificially-manufactured tongue which had come into existence during the last fifty years—_Si-kololo_. This will be found illustrated in its most modern form in the first volume of my _Comparative Study of the Bantu and Semi-Bantu Languages_ (Oxford, 1919). My information has been mainly derived from the _Sikololo Grammar and Vocabulary_ published in London by Dr. Stanley Colyer in 1917, and from a _Sikololo Phrase-book_ produced two years’ earlier by Mr. Stirke, the author of this book, in conjunction with a very noteworthy colleague, Mr. A. W. Thomas, formerly master of the Barotse National School.

_Sikololo_ which has gradually displaced and superceded the Sechuana or Sesuto language of the original Bechwana invaders of a hundred years ago, is a far easier language to speak and pronounce than the difficult and highly peculiar idiom of Central South Africa. To philologists, Sechuana and its later southern specialization, Sesuto, are intensely interesting. Sechuana entered South Africa at as ancient a date as any other Bantu tongue—say sixteen or seventeen hundred years ago, coming thither from South-East Africa. But since then it has become highly specialized. The most specialized form of it was the southernmost, Sesuto, scarcely distinguishable a hundred or more years ago from the Sepedi of the Western Transvaal.

The chief Sebituane, born some hundred and twenty years ago, was a Mosuto, the son of a great crazy woman-chief known as Mantatisi. He carried his Sesuto tongue with him on the amazing journey he made with half his tribe across Bechuanaland to Lake Ngami and the Zambezi. Sechuana or Sesuto was the Court language in Barotseland when Livingstone first reached the Upper Zambezi in 1851. As the years rolled by, and after the Makololo (Basuto) dynasty had given place to a restoration of the Aluyi chieftainship, the original Sesuto spoken by Sebituane and well understood by Livingstone (who could converse in it freely) became greatly changed. It has remained the governing language of Barotseland and has succeeded almost in replacing and killing the remarkable Siluyi of the old rulers; and neither of the beautiful languages of the Tonga-Subia group of Southern and South-Eastern Barotseland has been allowed to take its place.

In Sikololo the pronunciation is much simplified. The rough guttural combination _kx_—of Sechuana-Sesuto gives way to the simple _k_; a hiatus formed by a dropped consonant is filled by _h_; _ts_ is generally changed to _z_; _e_ in the prefixes becomes _i_; _f_ which had become _h_ in Sesuto is generally replaced; the uncouth _tl_, _tlh_ or the aspirated _th_ and _kh_ return to the simple _t_ and _k_; and in general the pronunciation of Sikololo is brought into conformity with the harmonious phonology of the Zambezi languages.

A far more interesting tongue is _Siluyi_, the nearly extinct language of the Aluyi; in its older form, that is to say, as it was spoken round about Lialui (the original governing centre, near the left bank of the Zambezi, in about 15° 15´). Here it was a stately language, using preprefixes, and possessing all the seventeen customary prefixes of the Bantu languages. The second (_Ba_-) and eighth (_Bi_-) of these were in an abbreviated form, because the Luyi tongue had rather a dislike to an initial _b_, though it turned the sixteenth prefix—_Pa_- into _Ba_-, following a practice which recurs a good deal in the Angola tongues.

The principal form in which I have presented this language in the First Volume of my _Comparative Study_ is derived from the studies of Luyi (Siluyi) compiled some twenty to twenty-five years ago by the Rev. E. Jacottet, of the French Protestant Mission. It was soon evident, when the British South Africa Co. officials got to work, that there was at least one separate dialect of Luyi spoken to the East and North-East of the main Zambezi. This was the _Si-kwango_ or _Si-kwangwa_, a tongue in which the preprefixes are dropped. Si-kwangwa seems still to be a living speech among the people in the East-Central part of Barotseland; but the much nobler Luyi of the Zambezi vicinity is dying out, even among the natives of Luyi stock, dying out in favour of mongrel Sikololo.

Allied to Siluyi apparently, is the _Nyengo_ language of the Bampukushu people, a scattered tribe living still (perhaps) in the South-Western portion of Barotseland, on the banks of the Kwando or Linyanti river. There has been no record of this speech since that which was compiled by Livingston about 1853, and which I have reproduced (numbered 82) in my _Comparative Study of the Bantu Languages_.

Still farther south, in the northern part of the basin of Lake Ngami, there were, down to quite recently, the _Ba-yeye_ or _Makoba_ people, whose language also seemed to show kinship with Luyi as well as with the Subia-Tonga group. The imperfect record of this is numbered 81 in my book. It may be extinct as a spoken tongue by now. But its existence where it last lingered on the northern side of Lake Ngami would seem to show that this basin of the shrinking lake (which received a vast but diminishing tribute of water from the Okavango river, and earlier still from the present “Upper Zambezi”) was colonized by the Bantu from the north before it was found and peopled by the Bechuana tribes from the south-east.

So much does Sikwangwa differ from Siluyi nowadays, that it is almost a separate language. If this be so, the Western Zambezi group of my formation would contain at least four languages, _Yeye_, _Nyengo_, _Luyi_ and _Sikwangwa_. I am also informed that Sikwangwa is not the only existing dialect of Luyi, but that two others exist, and are spoken to the North-West and North-East of Lialui and the Zambezi: _Si-koma_ and _Si-kwandi_. These have only been recorded in name, and I have seen no specimens of their vocabularies. It would be interesting to ascertain whether, like Luyi, they retain the use of preprefixes; or whether they have dropped them, as have most of the languages in the western half of the Zambezi basin.

There is a tendency in Siluyi for _b_ and _p_ to take the place of _v_ or _f_ as the initial of the root, although in the prefixes _b_ and _p_ are inclined to drop out: so that _Aba_- becomes _Aa_-; _Ibi_-, _I_; Ubu-, Uu-; and Apa-, Aba-. Thus we have -_pumo_ for “belly,” instead of the familiar -_fumo_; -_bumu_ for “chief,” in place of -_fumu_; -_pumbu_ for “ground” (ordinarily -_vu_) -_buu_ for “hippopotamus” (-_vubu_), -_bula_ for “rain” (-_vula_); and -_pi_ for “war” (-_vita_). The roots for the numerals “six,” “seven,” “eight” and “nine” were in Livingstone’s day, very peculiar and unlike those of any other Bantu tongue; though to-day they are supplemented by paraphrases meaning “five-and-one,” “five-and-two,” etc.

The Rev. Mr. Jacottet has shown how rich Siluyi was in folk-lore and legends. It is a great pity this particularly interesting, expressive Bantu tongue should be allowed to die out through the vogue given to the artificial amalgam of speech known as Sikololo.

_Subia_, spoken in Southern Barotseland, on and near the Zambezi at its bend eastward, is also a language of great interest, said likewise to be dying out under the rivalry of mongrel Sikololo. It was first noticed by Livingstone about 1851, and a grammatical sketch of it was given by Jacottet in 1896. But in the present century, F. V. Worthington and C. F. Molyneux, officials of the Chartered Company, supplied me with a full vocabulary and grammatical information which has been reproduced in my _Comparative Study_, Vols. I and II. Subia is spoken to the south of the Zambezi as far eastward as Pandamatenka; but it belongs to the West Central Zambezia Group, mainly in use on the north side of the great river, and including the Tonga, Ila and Lenje tongues and their dialects. This group is related to the Luba languages of Southern Congoland and to the tongues of the Bangweulu basin and of the Luangwa valley. They are all beautiful-sounding and expressive languages, representing Bantu speech in an attractive form.

The _Tonga_ language is spoken in South-East Barotseland in one or more dialects. The Tonga people are a series of vigorous tribes not very likely to give up their expressive language in favour of the bastard Sikololo or the crabbed Sechuana. They attained some degree of civilization two centuries ago under the teaching of Portuguese Jesuit missionaries. Their attitude towards Livingstone and later travellers, down to the incoming of the British South Africa Company, was hostile. James Chapman, the great explorer of Southern Zambezia and even of Barotseland—in some senses the rival of Livingstone—suffered much from attacks at their hands. So did F. C. Selous. They also warred incessantly against the Zulu invasions under Umsilikazi and Lobengula, and checked the Zuluizing of central Northern Zambezia, which, in default of similar vigorous opposition, took place in the regions east of the Luangwa river.

Similarly related to Tonga and Subia are the _Ila_ (Shukulumbwe) and _Lenje_ tongues, though both lie almost or wholly outside the limits of Barotseland, in the central part of Northern Rhodesia. The Ba-ila tribes, two divisions of which are also known as the _Ba-lumbu_ and _Ba-shala_, are a splendid-looking folk in physical development, who were long recalcitrant to European influence. In fact their healthy plateau country has only been opened up to knowledge during the last twenty years. When they were first seen in the latter part of the nineteenth century they were found to be—the men especially—living in a state of complete nudity, except in the case of the married women. But to make up for the complete neglect of clothing the men devoted particular attention to their head-hair. This was pulled up, greased, lengthened by the insertion of ancestral hair pullings, and fastened to a supple arched, recurved whip, so that it rose a foot to two feet above the occiput. Nowadays I imagine that their complete nudity is modified. The mining surveys, motor roads, and suggestions of railways, together with good and wise government, in and out of the Barotse kingdom, have succeeded in taming the Ba-ila and in drawing them into paid employment; though it is still feared by watchful administrators that they tend to diminish owing to sexual vice which causes the women to become infertile.

Another language of interest and importance is that of the _Nkoya_ and _Mbwela_ tribes living in East Barotseland. It has affinities jointly to the Luyi Group and to that of the North-West Zambezia languages. Preprefixes are not used, but the seventeen prefixes-with-concords seem to be fully represented in primitive forms, except that the seventh prefix (_Ki_-) is represented by _Shi_-, which brings it near to the _Si_- of Luyi. The tenth prefix is lisped as _Thi_-, _Thin_-, &c., a feature reappearing in one or two other tongues of the Group. The eighth prefix is _Bi_- and the locative prefix is _Pa_-. The root for ‘two,’ in the numerals is the East and South Africa form, -_bidi_ or -_biji_ (for -_bili_); not the -_bali_ root that obtains in the other members of the North-West Zambezian Group. Another peculiarity that signalizes Nkoya is the root -_mwa_- for ‘all,’ only to be met with elsewhere in two of the Manyema languages of East Congoland. In Nkoya there is none of that dislike to the consonant _p_, which recurs over and over again in Bantu Africa. But in the other members of the North-West Zambezia Group this obtains, and in these—Luena, Mbunda, Lujazi and Chokwe—the 16th prefix is given as _Ha_- or _A_- instead of _Pa_-. Otherwise they are fairly orthodox, betraying little more “West African” affinities than the use of the -_bali_ root for ‘two’ (instead of -_bili_), which is a characteristic of West Central Africa.

The _Luena_ or _Lubale_ language only enters the north-western part of Barotseland; _Mbunda_ is spoken in separate parts on either side of the Upper Zambezi, also in the north of this state; _Lujazi_ extends from South-East Angola to the western limits of Barotse territory; and Chokwe comes in as a trading language in the extreme north. None of these North-West Zambezia languages use preprefixes.

One other great Bantu Group may be met with in the northern part of Barotseland. Here there are outlying dialects of Luba and Lunda which have penetrated into Barotseland comparatively recently. The Luba tongues of South Congoland are represented by _Kahonde_, a form of South Luba illustrated in my Comparative Study under the designation of 105a. This is spoken on the North-Eastern verge of Barotseland in the basin of the Upper Kabompo river. Southern Lunda, introduced into the northern part of this territory goes under the name of “_Ma-bunda_,” the root of which—_bunda_—is the same as that of “Mbunda,” a quite distinct language of the North-West Zambezia Group allied to Luena, Lujazi, Nkoya, and Chokwe. “Bunda” is evidently a language and tribal name which in the shapes of -_bundu_, -_bundo_, and -_bunda_ haunts Angola and Western Zambezia without implying close relationship between the different forms of speech thus termed. The “Ma-bunda” dialect of Lunda may represent an older form of the latter name, which seems to be a contraction of Lu-unda for Lu-bunda. The language (No. 111) which I have styled “Western Lunda” in my Comparative Study was originally called by its discoverer, the Revd. S. Koelle, “Ru-unda,” suggesting that _Ru_- (= _Lu_-) was the familiar language prefix and -_unda_ or -_bunda_ the root. The loss of an initial _b_ is a frequent occurrence in the Lunda languages.

* * * * *

The distribution of tongues southward from Angola and the Congo Basin over the regions north of the main Zambezi river indicates no causes of hesitancy, no interruptions in the north-to-south progress of the Bantu peoples. But westward of the course of the Luangwa river, as far as the Barotse country extends, or at any rate as far as the main stream of the Zambezi flows, this great river seems to have been a very decided and arresting barrier in the distribution not only of African tribes and languages but of the larger or more remarkable mammals and birds.

With regard to man of course the barrier has not been so effective. Evidently the Aluyi people of long ago crossed the Upper Zambezi and penetrate to the Lower Kwando or Chobe, and across this stream to the vicinity of Lake Ngami, though not much farther west; and in return, immigrants of Bechuana stock entered Barotseland from Lake Ngami in the last century. But there seems to have been a not easily explained halt in human migrations between Western Zambezia and South-West and South Africa. It would be comparatively easy to account for, if one could presume that the present sterile influence of the Kalahari desert prevailed two thousand or even one thousand years ago, and earlier still. But on the contrary from the many indications we can collect and from the legends of the natives in the regions south of the Zambezi, we are inclined to believe that the present arid nature of the lands on the western side of South Africa is quite a modern feature which has gone on rapidly increasing. So rapidly that even I who made a journey in 1882 to the Kunene river from Mossamedes can remember flowing streams (in the dry season) and forests, where the rivers and rivulets are now dry and the woodland is dead.

There seem, in addition, to have taken place in comparatively recent times, changes in the disposition of water supplies, which have diverted much useful moisture from inner South Africa into the Atlantic Ocean. Not such a very long time ago the Kunene river, which brings an abundant supply of fresh water from the great knot of highlands in Southern Angola, bore this water into Lake Etosha in Ovampoland, and beyond that towards the course of the Okavango or the great vanished lake of which Ngami is the surviving vestige: Ngami, which even in Livingstone’s day—the end of the ’forties, some seventy-two years ago—was twice its present size. Into the same basin flowed the mighty Okavango (known much farther north as the Kubango), the Kwando and the upper Zambezi, then the outlet of another vast lake in the Barotse country. Quite possibly these two shallow lakes of huge size sent their overflow of water through the Guay river into the bed of the Zambezi-Kafue, and so out into the Indian Ocean after all; for these waters of Central Africa were completely hemmed in southwards by the plateaux of Damaraland, Bechuanaland and Matebeleland. But in those days the ultimate source of the Zambezi would have lain in the Benguela highlands near the Atlantic Ocean, and not in the Luvale country near to the upper Kasai.

* * * * *

As regards the mammalian fauna and some of the more striking or peculiar of African birds, Barotseland, east of the Upper Zambezi, appears, in common with all the rest of South Central Africa up to the east coast of Lake Nyasa and down to the main course of the Zambezi, to be deficient as compared with Portuguese West Africa, South Africa, and (in a lesser degree) Moçambique. It probably has more in common as a distributional area with the southern part of the Congo basin outside the forest area, and west of the Luapula-Lualaba streams. Down to recent times Barotseland possessed great herds, enormous numbers of a few species of game animals, but not such a great variety of species. Like South Congoland outside the forest zone, it seems never to have had any example of the rhinoceros: indeed, I have no confirmatory evidence of the existence of a rhinoceros or a giraffe, west of the eastern rise of the Nyasa-Tanganyika plateau, or of the waters of Lake Nyasa. The rhinoceros begins to show itself south of Lake Nyasa and east of the Shire river. It recurs in Southern Angola and Damaraland, up to the vicinity of the Western Zambezi, and as far north as the beginning of the lofty plateaux where the Kwanza river rises.

The Kwanza, which divides Angola into two unequal portions, is very noteworthy as a boundary. Its ultimate sources are in the Ngangela, Bihé and Lujazi highlands (plateaus and mountains) which connect the mountain ranges of Benguela on the west with the 5,000-feet-high plateau dividing the Southern Congo basin from that of the Zambezi. This undulating line of heights (rising on several peaks to nearly 8,000 feet) really marks off Southern from Northern Angola; and zoologically Southern Angola extends eastwards almost to the banks of the Upper Zambezi and covers the western limits of Barotseland.

Here you have a remarkable extension of the South African area of zoological distribution. East of this, from the Upper Zambezi to the Shire river and the east coast of Lake Nyasa (with the Zambezi as a southern boundary) there is an interruption in distributional area which affects certain mammals and birds very curiously. For some the interruption is complete. The Ostrich, the true Gazelles, the Oryxes, the Secretary bird, certain types of Vulture, the Striped Hyena, the Chita or Hunting Leopard, the Caracal Lynx, the Aard Wolf (_Proteles_), the Big-eared Fox (_Otocyon_), the small Desert Foxes, the Black-backed Jackal, the _Pedetes_ or Cape Jumping Hare, the Elephant-shrews, the Mountain Zebras, the White Rhinoceros and, in a lesser degree, the Black, are in a measure confined to north-east and eastern Equatorial Africa on the one hand, and to Africa south of the Zambezi and to Southern Angola on the other. Some of the examples cited stretch across to the Bahr-al-ghazal and even Northern Nigeria from Abyssinia and the Eastern Sudan; others reach Senegal and Western Nigeria beyond the forest zone. But all of the examples cited (besides many more less well-known birds and mammals) seem to be absent from North Angola, Northern Zambezia and Nyasaland, and apparently also from Moçambique. In Moçambique, numerous forms may have been exterminated by the white man and the black hunters, who began shooting here in the 17th century. Yet it is difficult to understand that they could thus have eliminated the Ostrich, Springbok, Striped Hyena (represented in South Africa and Angola by the Brown Hyena), Chita, Caracal, Aard Wolf, and _Pedetes_ rodent, or even the Giraffe.

The Giraffe’s distribution conforms somewhat, but not so closely, to similar restrictions. Its least specialized form, perhaps, is the Reticulated Giraffe of Somaliland, whereon the original markings are seen to be white stripes, horizontal and perpendicular, on a red ground. From Somaliland the Giraffe radiates over the eastern Sudan to Northern Nigeria and thence (with gaps) to Senegal. It entirely avoids forested Central and West Africa, though it has a near relation, the Okapi, in the forests of Equatorial Africa east and north of the main Congo river. The distribution of the Giraffe continues south from Uganda and Somaliland to the east of Tanganyika and Nyasa down to the neighbourhood of the Ruvuma river. It has never been reported south of that stream, or until the Lower Zambezi has been _crossed_. A hundred years ago and down to about 1900 it was found in various sub-species and varieties _south_ and _west_ of the Zambezi to Cape Colony and into Southern Angola. But it has never been shown to exist between the Upper Zambezi and the Moçambique coast, north of that river system.

On the other hand the range of the Sable Antelope does not leave out Northern Zambezia and Barotseland. It begins in Eastern Africa, in the latitude and neighbourhood of Mombasa (a fact first suggested in my book on Kilimanjaro, published in 1885), broadens over what was formerly “German East Africa,” passes round the south end of Tanganyika into Southern Congoland and crosses Barotseland into Eastern Angola. Here, in the south-eastern basin of the Kwanza, and possibly in north-western Barotseland as well, it develops its most superb form—_Hippotragus variani_, generally called the Giant Sable—with magnificently developed horns, longer than those of any other type found elsewhere, and a longer, narrower skull. It is possible (from heads I once saw, shot to the west of the main Upper Zambezi and specimens of horns) that the Giant Sable may be found just within the limits of Barotseland.[3]

North of the belt of dense forest between the Liba and Kabompo, on the spongy plains near to the river, Livingstone noticed in the rainy season large numbers of Buffaloes, Elands, Kudus, Roan Antelopes, Gnus, and other game. He even avers that farther north he saw a White Rhinoceros. It was on a Sunday and his encampment was almost surrounded by herds of astonished Mpala antelopes, tsesebes (“bastard hartebeests”), zebras and buffaloes. But although the Black Rhinoceros is undoubtedly found in Southern Angola close up to the Barotse frontier, and the White has been shot immediately south of the Upper Zambezi, no word has come from any other quarter as to the existence of the monstrous White square-lipped species in the direction of the Upper Zambezi and the Congo watershed; so it is possible his eye-sight was deceived.

(From a specimen in the Museum of the New York Zoological Society presented by JOHN JAY PAUL.)]

In general, Barotseland, east of the main Zambezi and of its great north-western affluent the Lungo-e-bungo, seems to possess the mammal and bird fauna of Northern Angola, Southern Congoland, Northern Rhodesia and Nyasaland. Moçambique is richer—up to the Shire river—in possessing the Black (I used to think also the White, and perhaps was not wrong) Rhinoceros, and a few other creatures found also in Eastern Equatorial and in Trans-zambezian Africa. Otherwise Moçambique remains rather an unsolved problem, as to why and how it served as an interruption in the spread of the vertebrate fauna of Pliocene and Pleistocene North Africa down to South Africa, an extension which as regards West-Central and Western Africa may have been barred by the former enormous area of dense forest growth.

* * * * *

In botany, as in vertebrate zoology, Barotseland is part of the region which lies between the Kwanza and Zambezi rivers on the south, and the main-stream of the Congo on the north, from Lake Mweru to the Atlantic Ocean. It has a sufficient rainfall, from 22 inches in the south-west to 42 inches in the north-east, to maintain a fairly luxuriant flora as well as powerful rivers. North of the confluence of the Kabompo and the Liambai or Liba (the Upper Zambezi), there seems to have been seventy years ago a luxuriant belt of forest stretching northwards and occupying the mountainous region between the Liba-Zambezi and the Kabompo or Lulafuta.

The average elevation of the Barotse Valley, south of this confluence, ranges from a little under 3,000 feet at Sesheke to about 3,600 feet at the juncture of the Zambezi and the Kabompo; but the forested country between these rivers must rise in part to at least 4,500 feet. The richness of this forest seems to have made an immense impression on Livingstone in the middle of the ’fifties. Up till then, coming from South Africa through Bechuanaland, he had at most seen scattered trees with occasionally a fine solitary umbrageous specimen, or a palm thicket. But above the confluence of these two rivers he evidently plunged (when he had to land) into a woodland comparable with the splendid forests of the Congo or the Cameroons. His _Missionary Travels_ contains some very good word-painting of the grand growth of the trees, the deep gloom of the forest depths as contrasted with the shadeless glare of Bechuanaland. In spite of the incessantly rainy weather—and in this little-explored region the yearly rainfall must be far in excess of the modest estimate quoted by travellers in the Barotse park-land and plateaux—Livingstone found considerable pleasure in here seeing—for the first time—the _real_ forest display of Central Africa.

North of the Zambezi-Kabompo junction Livingstone noticed the increasing growth of the bamboo on the uplands, as a skirting of the great display of tropical forest which so deeply impressed him as he followed the northward course of the main Zambezi. This—or these—bamboos recur on the elevated plateaux of Eastern Barotseland, and extend over a good deal of the upland on either side of the Upper Kafue, as well as over the five thousand feet crest of the Congo boundary. No doubt they may also be observed within the north-western limits of Barotseland. They are seemingly of the same types as the bamboos of Nyasaland and the elevated portions of Northern Rhodesia south of Tanganyika, and in such case belong to the _Oxytenanthera_ and _Arundinaria_ genera. The only other genus of _arboreal_ bamboo in Africa is _Oreobambos_; but that, I think, is confined in its range to Abyssinia and the Snow mountains of East Equatorial Africa.

The forests of Northern Barotseland have one or more species of wild plantain (_Musa_) and either wild or introduced examples of the oil palm, where the rainfall is over fifty inches per annum. This may have been introduced from Congoland or Northern Angola by the natives, who in the north of the Barotse country are so closely connected with the peoples of the Congo basin; or as in the case of North Nyasaland and Zanzibar Id., it may be a separate species or sub-species of the _Elaïs_ genus.

According to the reports of later travellers, there is much woodland and scrub on the higher land—4,000 to 4,500 feet—on the west side of the Zambezi, south of the Kabompo-Zambezi junction. This woodland is also repeated, less markedly, to the east of the great river where the land likewise rises. The Mankoya country (Eastern Barotseland) has a larger, steadier rainfall than Sesheke and the southern regions; and the Mashukulumbwe table-land, and the valley of the upper Kafue are said to be well and regularly watered countries. The more scattered growth of trees (occasional fine and large trees are common) seems due to human intervention. The natives of the Lukona province, towards Angola, are a wilder, less intelligent race than folk of Tonga kinship on the east, and not so eager to promote tree planting. Bulovale (the region between the Luena and Kabompo rivers) and north of the Kabompo up to the boundary of the Zambezi watershed are well forested regions.

The central valley of the Zambezi—“Borotse proper,”[4] as it might be called, the region which the Barotse (really the Aluyi) have so long dominated—ranges from below the Zambezi-Kabompo confluence on the north to the Gonye Falls on the south, and is about 150 miles long and 60 miles broad. It is the bed—most observers think—of the old lake of the Upper Zambezi, and presents to-day an utterly flat and treeless aspect. Then ensue from Gonye to Katima, some eighty miles of rapids or falls. From Katongo to Mambova, past the now world-famed Sesheke, where Livingstone and Oswell first “discovered” the Zambezi, there are fifty miles of navigable river; then two more rapids at Mambova and Nkalata, and then another navigable stretch of some sixty miles to the vicinity of the Victoria Falls. (These lie well outside the political limits of Barotseland, the boundary striking north along the Majili river, ten miles east of Sesheke.) The plain of the extinct lake is much flooded during the winter rains, when a good deal of central Barotseland is under water. The rainy season seems, however, to be tending more and more towards the spring months (the autumn of South Central Africa).

The geology of Barotseland may be summarized thus: In the north-west, north, east and south-east the surface is a red laterite clay, superimposed on granite which in the higher portions obtrudes itself in mountains and hills. The central valley and most of the south-west has a heavy, white, sandy soil, but with a good deal of alluvium laid down on the sand by river courses or in the bed of the ancient lake. There is a basaltic outbreak in the neighbourhood of the Victoria Falls, but this lies outside Barotseland limits. In the eastern half of the country gold, copper—connected with the copper deposits of Katanga—tin, lead, zinc and iron have been discovered, chiefly in regions between four and five thousand feet in altitude.

* * * * *

In this little book and in the first and second volumes of my _Comparative Studies of the Bantu Languages_ (fed from much the same source) our information concerning Barotseland and its peoples has been greatly implemented. But that fact does not conceal from us that the country, its zoology, botany, peoples, and even its geography (to say nothing of geology) are still far too little known. The portions that lie to the west and south-west of the main Zambezi remain almost _un_known and undescribed, or are only made known to us in works of travel fifty, sixty, seventy years old, which even if accurate in what particulars they give were written at a much lower level of knowledge than exists concerning Africa in general to-day. We know more about Eastern Angola than we do of Western Barotseland. Recent discoveries concerning Angolan Antelopes, Zebras, Rhinoceroses, and other mammalian types make us eager to ascertain how these discoveries affect Barotseland _west_ of the Zambezi. How far do the Bushman tribes extend into Western Barotseland? Are they of normal human stature, like those of South-East Angola? Do they speak languages akin to the Bushman tongues and use clicks? Have they greatly projecting brows like certain Bushmen of the Northern Kalahari, or are they without brow projections at all but very prognathous in the lower part of the face? What is the full tale of the Bantu tribes of Western Barotseland? Are there still unrecorded Bantu languages on the Kwando river? Do the tongues recorded south-west of the Upper Zambezi by Livingstone, seventy-two years ago, still exist?

Does Livingstone’s great forest between the course of the Upper Zambezi and that of the Kabompo still maintain itself, in spite of the reckless spirit of destruction inherent in all uneducated Negro tribes, who since the imposition of peace by the white man have been largely increasing the area of their settlement in Northern Barotseland? If the forest is still there, in whole or in part, why are no botanical reports and collections available? The botany of Barotseland, in common with its zoology, has never been efficiently explored and reported on since the land was first entered. Cattle apparently prosper in the whole country; yet the tsetse fly exists seemingly everywhere. Does it, north of the Zambezi and west of the Luañgwa, convey no germs? The time—it seems to me—has come when the British Empire, from London or from Cape Town, should have this country examined scientifically from west to east and north to south; not with anything but benevolent intentions towards its natives—Aluyi and Batonga, Basubia and Mabunda, Batonga and Baila, Bankoya and Kahonde, Valujazi and Valubale—but with the earnest endeavour to make fully known its resources and defects, its wonders and survivals, and the relation that Western Barotseland bears to the growing menace of South Africa: the spread of the Kalahari Desert, the drying up of once fully habitable land, the cause of ruin in South Angola, in Eastern Damaraland, and the vast, dead region north, east, south, and west of Lake Ngami.

H. H. JOHNSTON.

Photo by Mrs. Marshall]

BAROTSELAND:

EIGHT YEARS AMONG THE BAROTSE.