FOOTNOTES:

[170] For the meaning see Note at the end of the Volume.

[171] Read before the British Association for the Advancement of Science, at Swansea, in 1848.

[172] American Journal of Oriental Literature.

[173] See Rüppell's Reise, &c., that author being the first to give the true affinities of the Koldagi language, i.e. with the Nubian.

B.
KAFFRE ATLANTIDÆ.

The preliminary facts of most importance in the ethnology of the great Kaffre area are two—connected with the language, and from their combined effects giving it the appearance of differing in kind from any other African tongue.

These two peculiarities, which are illustrated from Boyce's Kaffre, and Archbell's Bechuana Grammars, are as follows:—

1. The system of prefixes.—Every Kaffre noun is preceded by an adventitious syllable, apparently destitute of any separate meaning; just as if, in English, we said, instead of—

So far is this principle carried that the words introduced by the missionaries, from our own language, all become thus modified. Hence priest changes to um-priest; pharisee, um-pharisee. I imagine that without this prefix the simple root would be as impossible a form for a Kaffre or Bechuana as a word like ὀρνιθ- (i. e. a root without any concomitant inflection) would be to a Greek. Nevertheless, the Kaffre prefix is no sign of case or number.

In the following words the syllables in italics are the prefixes, wholly independent in origin from the root, and wholly non-radical:—

ENGLISH.KAFFRE.
Personumtu.
Horseihashe.
Chiefinkosi.
Servantisikaka.
Infantusana.
Riverumlambo.
Faceubuso.
Fordakutya.
Peopleabantu.
Wordsamaswe.
Cattleinkomo.
Treesimiti.

2. The euphonic or alliterational concord.—This is a point of Kaffre syntax, and occurs when certain words come together; e. g. in the case of a substantive governing another in the possessive case, or an adjective agreeing with a substantive. In either of these cases the secondary word changes its initial sound into that of the primary one, or into some sound allied to it.

If in English we expressed the relation between the nominative and possessive cases on the same principle that occurs in the Kaffre and Bechuana, we should say instead of—

It may easily be imagined that languages thus characterised, taken along with undoubted points of physical difference, have supplied the grounds for a somewhat broad line of demarcation between the Kaffre and the other Africans. That such a line is natural, is certain; whether it has not been made too broad, is another question.

KAFFRE NATIONS AND TRIBES.

Physical conformation.—Modified Negro.

Language.—Prefixional and alliterational.

Area.—Western, Central(?), and Eastern Africa, from the north of the Equator to the south of the Tropic of Capricorn.

Chief divisions.—1. Western. 2. Southern Kaffres. 3. Eastern Kaffres.

That there is no broad line between the Kaffre and non-Kaffre Africans, on the western side of Africa at least, is shown by the following populations; whereof both the languages, as known by vocabularies, and the physical conformations are intermediate or transitional.

1.
WESTERN KAFFRES.

Beginning with the parts south of the Bimbia area we have between the river of that name and the Portuguese kingdom of Loango—

THE BATANGAS.

Native name.—Bânâka.

Locality.—Sea-coast of Western Africa 3° north latitude, half way between the Camaroons and Gaboon.

Physical appearance.—More Kaffre than Negro; skin more copper-coloured than black; sclerotica clear.

THE PANWES.

Locality.—Eastward to, and more in the interior than the Batangas; from 3° north latitude to 3° south latitude, on the Head-waters of the Gaboon.

THE MPOONGAS.

Locality.—Mouth of the Gaboon.

Then follow the nations of—1. Loango; 2. Congo; 3. Angola; and 4. Benguela; closely allied both in language and appearance, and nations whose place in the Kaffre division has long been recognised.

That there is, however, considerable difference in respect to the physical conformation of the different tribes, is certain; some writers, reducing the native of Portuguese Africa to the Negro, others to the proper Kaffrarian, or South Kaffre, type.

If the difference between these two extremes be rightly estimated by the present writer, the former should prevail along the courses, the latter on the watersheds of the rivers. His information, however, is imperfect upon this point.

2.
SOUTHERN KAFFRES.

Area.—The extra-tropical portion of South Africa, minus the parts south, Walvisch Bay on the west, and the water-system of the Orange River.—Encroaching.

Chief divisions.—1. Amakosas, nearest the Cape. 2. Bechuanas, north of the head-waters of the Orange River. 3. Zulus, north of the Bechuanas, with an undetermined extent inland. Numerous sub-divisions.

Physical conformation.—Cranium, more vaulted and less prognathic than the Negro; hair, tufted, and as such approaching that of the Hottentot; zygomatic development, outwards rather than downwards, so that the cheek-bones become projecting, and the forehead and chin tapering; lips, generally thick, and nasal profile less generally depressed than with the Negro; colour, black, dark brown, clear brown; stature, tall.

Habits.—Pastoral rather than agricultural.

Religion.—Paganism.

Customs.—Circumcision and tattooing.

The Dammaras.—Are the Dammaras Kaffre? This will be noticed in p. [495].

The Kaffres of Lagoa Bay, darker and more Negro-like than the typical Kaffres of Kaffraria, form the transition between the southern Kaffres and the eastern divisions of the tribes of Inhambane, Sofala, and Botonga, and the water-system of the river Yambezi. They are Negro rather than Kaffrarian, their languages being but imperfectly known.

3.
EASTERN KAFFRES.

So are those of Mozambique and Zanzibar; chiefly represented by the Makuas, the Monjous, and the tribes speaking the Suaheli language. A vast accession to our philological data for these parts proves incontestably the Kaffre structure of the languages of the coast from the Cape of Good Hope to nearly 5° north latitude.

But the tribes of the unknown parts of Central Africa, south of the equator, are also, probably, either wholly, or almost wholly, Kaffre. It is this which has induced me to pass sicco pede over the numerous details of the Kaffres of the coast, so as to allow space for a short notice of the newer additions to our knowledge of the inland Kaffres, west and east.

a. West.—The Kazumbi, said to live at such a distance from the coast, as to be obliged to travel three or four moons, before they reach any of the possessions of the Portuguese and to speak a language which resembles, in many words (especially the numerals), the Congo. This is probably the Cazambe of the maps, nearly in the centre of Africa, in 13° south latitude.

The Koniunki.[175]—From some captured Negroes examined by the Rev. T. Arbousset, of the Paris Missionary Society, a few words have been collected of the Koniunki language. They are apparently of the Kaffre class.

The locality of the Koniunkis was also said to be so far in the interior, as for the gang to have been three or four months in reaching the Mozambique coast.

This indicates that they were east of the Kazumbi, whilst the affinity of the language with the Bechuana gives them a southward direction.

The Mazenas, mentioned along with the Koniunkis, as lying between them and the Makuas.[176]

Hence, the Congo, the Kazumbi, Koniunki, and Mazena areas, probably, carry us across the whole continent in (about) 13° south latitude; whilst the likelihood of the southern Koniunki and northern Bechuanas being conterminous, helps to fill up the void spaces north of the parts about Litakú.

b. East.—Parts about Mombaz, Formosa Bay, Lama, Patta, &c.

POCOMO.

Locality.—River Pocomosi (Maro).

Conterminous with the southernmost section of the Gallas.

WANIKA.

Locality.—North and west of Mombaz.

The Mahometanism of the Wanikas, if it exist at all, is of the most imperfect kind. They practise circumcision, it is true; but this is a general African, quite as much as a particular Semitic, rite—"They bury their dead, placing the head to the east; and it is customary, after waiting ten days, to kill a bullock and make a feast, pouring the blood upon the grave." The Wanika man seen by Pickering, "bore the marks of a national designation; consisting of a single notch, filed between the two upper front teeth, with numerous small scars on the breast."

WAKAMBA.

Synonym.—Merremengo.

Locality.—Mixed with, and conterminous with the Wanika.

WATAITA(?).

Locality.—Five days from the coast; conterminous with the Wakamba.

TAVAITI(?).

Locality.—Westward of the Wataita.

Language.—Different from the Chaga and M'Kuafi. Probably akin to the Wanika.

M'SIGUA.

Locality.—Pungany River. Scattered among the Wanika.

M'SAMBARA.

Language.—As known from a vocabulary of Krapf's, closely akin to the Pocomo, Wakamba, Wanika, and M'Sigua.

This last sentence suggests the nature of our reasons for making the tribes just enumerated Kaffre. The dialects of five of them are known by specimens, collected by Krapf, and are very nearly Suaheli. The evidence of the Kaffre origin of the Tavaiti and Wataita is less conclusive.

The M'Kuafi.—Are the M'Kuafi Kaffre? This question will be noticed in p. [501].

It has been suggested that the import of the peculiarities in the structure of the Kaffre languages may have been exaggerated; the effect of such an over-valuation being to isolate the class beyond its proper limit. The following facts are corrective to this view:—

1. The Woloff language is at least one other African tongue, which exhibits the phænomenon of an initial change, a process allied to the euphonic concord.

2. The Celtic tongues of Europe do the same.

3. Apparent instances of prefixed syllables, occur in the Howssa, Yarribean, and probably in other African languages.

Now there are many good reasons for believing that although the effect of such and such-like processes is to give the languages in which they occur a very remarkable external appearance—an appearance which, if we classed tongues and nations on the same principles upon which we class minerals, i.e. irrespective of descent and affiliation, would throw them into solitary and independent groups—they by no means denote the necessity of any inordinately long period for the evolution. All that they do denote is the greater intensity of what may be called the euphonic instinct, combined with a tendency to incorporate elements which, elsewhere, would be kept separate.

A doctrine laid down by Mr. Hales in his Philology to the United States Exploring Expedition, indicating a different classification from the present, deserves notice.

That inquirer considers that the line of affinity runs west and east, rather than north and south; so that the Kaffres of Inhambane, Zanzibar, and Mozambique are more closely allied to those of Loango and Angola than the Kosas, Bechuanas and Zulus of the Cape. The published evidence of the proposition is certainly insufficient.