(Chapter XI, section 10)

North Africa, introduction

North Africa is today an integral part of the Mediterranean world, but it has not always been so. It is land taken over by Mediterraneans, rather than basic Mediterranean country; for this reason it, like Europe, is racially complicated by the survival of Neanderthal-inspired Upper Palaeolithic food-gatherers. This survival is important only in a few places and among small populations, and in this respect North Africa differs greatly from most of Europe. The Mediterranean inroads began here earlier than in Europe, and since North Africa was the highway over which many of the Mesolithic and Neolithic invasions of Europe passed, it is natural that it should have a more thoroughly Mediterranean complexion.

From the beginning of the third millennium onward, northern Africa enjoyed, throughout Egyptian and classical history, the hazy repute of a region peripheral to great centers of culture. From the beginning of the first millennium B.C., the Phoenician colony of Carthage spread eastern Mediterranean civilization into Tunisia; after the fall of Carthage, the Romans extended the enlightened area to include much of Algeria, while the Greeks had already colonized the coast of Cyrenaica. At the time of the Arab invasions, North Africa was fast becoming a backyard of Europe. The advent of Islam brought this process to a violent end, and it did not begin again until after the conquest of Algeria by Napoleon.

Ever since the earliest notices of North Africans on the Egyptian monuments, the native inhabitants of North Africa have spoken Hamitic languages of the closely knit Libyan family. There is very little dialectic difference between them, and it is possible for a Riffian, for example, to speak with an Algerian Kabyle. Similarly, the Berber speech of the natives of Siwa Oasis, on the eastern extremity of the Berber world, is surprisingly like that of the Braber tribes of the Moroccan Middle Atlas, some 3000 miles distant. When contrasted with the complex Cushitic family of Hamitic speech, Berber appears extremely homogeneous, and we are warned by linguistic principles that its spread over the immense Berber area cannot have been too remote in time. It is possible that tones earlier Berber languages have disappeared, and that the present ones owe their distribution to a relatively recent diffusion.

There are, however, remnants of pre-Hamitic speech in various parts of North Africa. The Guanche spoken in the Canary Islands, at the time of the Spanish conquest, early in the fifteenth century, was only party Berber, and contained a large percentage of words of unknown linguistic affiliation.78 In modern Riffian and in other Moroccan Berber dialects, there is still a residue of non-Hamitic words in the local languages. For example, plant names ending in -nt or -nth may be seen in the word iminthi, meaning barley, and in shinti, meaning rye. These words have also been noticed in Indo-European languages of the northern Mediterranean shore, such as Greek and Albanian, and are generally attributed to the so-called Caucasic or Mediterranean linguistic group, which is the B element in Indo-European. It is very likely that agriculture, including the use of these two cereals, was introduced into North Africa by pre-Hamitic peoples.

Although there can be no doubt that Libyan Berber was spoken in the part of North Africa with which the Egyptians were in contact as early as 3000 B.C. and earlier, especially since there is a Libyan element in ancient Egyptian, we cannot assume the same for all of North Africa. It is possible that pre-Hamitic languages were spoken in Morocco and in isolated mountain regions in Algeria and Tunisia until much later, perhaps as late as the time of Christ, since there are strong Riffian traditions of people living in remote valleys who did not speak languages identifiable as tashilhait, or Berber.

According to the Arabian genealogies, all Berbers are descended from two men: Berr ibn Branes and Berr ibn Botr.79 These two Berrs, although possessing the same name, were not related. From them are descended the great families of Berbers such as the Masmuda, Senhaja, and Zenata. Of all these great families the earliest to spread seems to have been the Masmuda or Ghomara branch. This was followed traditionally by the Senhaja, who today include such varied peoples as the Siwans on the borderlands of Egypt, the Tuareg of the Sahara, and the Braber of the Middle Atlas in Morocco. The third great expansion was that of the Zenata, who were known in Roman times in Cyrenaica, but who did not reach Algeria and Morocco until the Middle Ages. In the thirteenth century these Zenata finally invaded Spain, conquering Arabs and earlier Berbers. One may compare the expansions of the Berber families to those of Kelts, Germans, Slavs, etc. in Europe.

Fig. 38. Ancient Libyan.

Redrawn from Bates, O., The Eastern Libyans, Plate 3, p. 120.


Unlike the later writings of mediaeval Arabs, the Egyptian and classical notices of Berbers do not assign to them an orderly descent from a few patrilineal ancestors in a typically Semitic scheme. The Egyptians, throughout their artistic history, took pains to distinguish the Libyans from other peoples by well-defined physical peculiarities. The Libyans are shown as active barbarians, clothed in animal skins, and wearing ostrich plumes in their hair; they are definitely white men, with lighter skins than either Egyptians themselves or Semites. Their faces are usually more sharply cut in profile than those of the Egyptians; the browridges are often prominent, the noses aquiline, the chins pointed, and the beards moderately abundant.

During the Old Empire, the Libyans are depicted as brunets; but in New Empire representations we see a change in the appearance of some of them. One branch, the Tehennu, known to the Egyptians from earlier times, still consists of brunet white men, but another group, the Mashausha, coming from farther west, is definitely blond.80 These two, the new people and the old, joined forces and attacked Egypt from the west. In dress and in other respects, there is nothing to indicate that the Mashausha were not Libyans.

Herodotus, in later times, places the Maxyces in western Libya, and states that they were culturally different from the purely nomadic Libyans to the east. The continuity of the name Mashausha through Maxyces extends to Mazuza, a sub-tribe of Riffians, and to the term Imazighen, by which many of the Berber groups designate themselves, and thamazighth, by which they identify their language.

These Maxyces, or Mashausha, as described by Herodotus, Sallust, and others, seem curiously un-African in some respects. They drive about in chariots, drawn by fiery horses; their garments are covered with gold;
they sacrifice oxen by strangulation, in a central Asiatic manner; the details of their council form of government, as revealed by a study of its modern counterpart, the Ait Arbain, are strangely Altaic.

While it would not be prudent to press this argument too far, it is quite possible that one or more of the invasions of West central Asiatic peoples which reached Palestine during the Bronze Age, or during the time of the earliest use of iron, crossed the Delta into northern Africa and kept moving across a country which offered little feed for cattle and horses, until they reached the Algerian and Moroccan grasslands. Herodotus

specifically states that these people were descendants of Persians. In any case, the horse and chariot entered North Africa from the east; either some Libyans took both from the Egyptians and spread them westward, or a specific people brought them in. The hypothesis of an Asiatic invasion of blond horse-users is not necessary to explain the Mashausha, nor the modern incidence of North African blondism, but, as will be seen later, it agrees perfectly with the present distribution of races in this area.

The history of North Africa during the last five millennia, as dimly outlined by oblique literary and artistic references, and in the absence of adequate archaeology, is not as simple a matter as the early Arab historians, who codified Berber tradition in their own pattern, supposed. It appears to have consisted of a succession of invasionS of Hamitic speaking peoples, mostly nomadic, interspersed with various outsiders, and later of Arabs, into the territory of agriculturalists of Neolithic cultural tradition and of basically European racial character. The Ghomara-Masmuda invasion is one of the earliest which may be salvaged from Berber traditional history, and this was followed by that of the Senhaja, and finally by that of the Zenata. Although the main direction of these expansions seems to have been from east to west, from the Hamitic center to its periphery, this is not true of all of them. The Senhaja, in at least part of their history, moved eastward.

In remote parts of Barbary are still to be found clans and families who cannot trace their ancestry to one of these noble Hamitic lines, or to Arabs, but who admit descent from indigenous heathen or from Christians. These families are called by Marmol “Berbers without name,” and represent the last survival in mountain communities of pre-Hamitic patrilineal family lines, except in those cases in which descent from Romanized Christians of various origins is indicated. Even in the clans named after Hamites or Arabs, the indigenous blood may be strong through continuous female infusion and through adoption.

The Masmuda and Ghomara, who made up the earliest invasion on record, are said to have come from Rio de Oro, as are the Senhaja, according to one tradition. There is, however, a story in both El Bekri and Ibn Khaldun that Ifrikos, the ancestor of the Senhaja, came from the Yemen, not long before the birth of Mohammed. This curious legend is supported in ways unknown to the Arab historians, for cultural traits diffused by some of the Senhaja-speaking peoples include terraced agriculture with irrigation, high earthen tigremts or castles, architecturally similar to those in southern Arabia, textile techniques, textile designs, and pottery forms and decorations all of which are strikingly similar to those in the Yemen.

The Zenata, who appeared in Roman Africa in the third or century A.D. and did not invade northern Morocco and Spain until the twelfth century,81 brought with them the camel, which they passed on to some Middle Atlas Braber tribes, who, separately or in combination with them, developed into the Tuareg. These Zenatan invaders were what Gautier calls les grands nomads chamelliers, the tall, lean, desert people, riding on camels, clothed in blue, and veiled, who trickled along the northern rim of the desert, and who took from Rome the outlying portions of her African empire.

The introduction of the camel changed profoundly the life of the North African plains, although it had little effect on that of the mountains. The wheel disappeared completely; the barbaric Libyans with their bronze and gold vanished from history, and those of them who were not absorbed by the newcomers and who refused to adopt the new economy took to the hills, to found rustic family lines among the mountain farmers. The camels of the newcomers pulled up the grass by the roots, flayed the trunks of all the trees which they could reach, hastened the process of soil erosion, and made the plains of North Africa at last truly African in appearance.

With the introduction of the camel, however, the Sahara became once more suitable for more than a sub-marginal human habitation. At some time during the late Pleistocene or during the periods of post-pluvial climatic change, negroes and negroids had moved up to occupy the oases and mountains of the northern Sahara, and the southern fringe of the Atlas country. Kufra was a negro oasis until the Arabs took it, and the course of the Wed Dra’a is the home of the Haratin, an insufficiently studied group of negroes. With the camel, white men moved down into the Sahara as swiftly riding nomads, enslaving the scattered groups of local negroes, and bringing others up from the Sudan in slave caravans, to cast a negroid tinge across the racial complexion of North Africa, which had hitherto been wholly white man’s country. Most of the slave trading, however, was carried on in Arab times, and indeed, the Arabs arrived in North Africa not long after their most useful animal, the camel.

The Arab invasions of North Africa can be divided into two waves, the first which came directly from Arabia, shortly after the death of the Prophet, and which brought families of aristocratic Arabs from the Hejaz and Yemen. These invaders came mostly without wives, married Berber women, and founded towns and dynasties. Although they converted much of the countryside to Islam, they did not force the Berbers to accept Arabic speech, which was confined, at that time, to the cities. In the eleventh century came the second Arab invasion, which was one of much greater volume and importance. This was the invasion of the Beni Hillal and Beni Soleim, tribes of apostate Bedawin from the Syrian Desert, who had made nuisances of themselves by pillaging caravans. This Hillali element introduced the first numerically important infusion of Arab blood into North Africa. The Beni Hillal and their companions settled first in Cyrenaica; thence some of thcm moved on to the Algerian plateau country, and to the country just south of the Atlas in the Moroccan Sahara, and onward to Rio de Oro. Other bands passed from Algeria through the Taza gateway down the trik es-sultan, to occupy the Moroccan plains along the Atlantic coast, from Safi to Tangier, and inland to Fez and Wezzan.

At present the inhabitants of North Africa are about evenly divided between Arabic and Berber speech, with the former commoner in the east, and the latter in the west. Although the Siwans speak Senhajan, the Cyrenaicans, largely Berber in blood, have been Arabized in language. Aside from the Tuareg, who also speak Senhajan, the next most easterly area of Berber speech lies in southern Tunisia and eastern Tripoli. In Algeria Berber is spoken by two important Berber groups, the Kabyles of the coastal mountains east of Algiers, and the Shawia of the Aures Mountains farther south. Oasis people, such as the Mzabites of Ghardaia, are also Berber speakers, as are the inhabitants of the Tunisian island of Jerba. In Morocco Berbers hold more land than do Arabic speakers; the whole northern strip from east of Melilla nearly to Tetwan, is occupied by Riffians and Ghomarans; the whole Middle Atlas by Senhajan Braber, and the Grand Atlas west of Demnat, by Shluh. In the lowlands east of the Middle Atlas, on the Algerian-Moroccan borderlands, and reaching up into the Riffian territory, are tribes of Zenata.

Throughout North Africa there are tribes and confederations of Arabized Berbers, and also some Berberized Arabs. Language and ethnic origins do not always coincide, and North Africa must be studied as a whole. The present North African peoples, apart from Jews and negroes and European colonists, represent a blend in different proportions between descendants of the old Afalou race, the Mesolithic and Neolithic Mediterraneans, the hypothetical central Asiatic nomads who may or may not have brought in the horse and chariot, the Hamitic-speaking tribesmen whose relationships are east of the Nile and in Ethiopia, and the two waves of Arabs. The regional variation between these elements reflects, in the main, varying proportions of the different components. An exception is seen, however, in the coastal region of Tunisia, where the Carthaginian state had its center, and where there may survive a minor Punic element, and the Islamized descendants of the much more numerous Greek and Italian settlers of the Roman period.


78 Hooton, E. A., The Ancient Inhabitants of the Canary Islands, pp. 16-19.
Abercromby, J., HAS, vol. 1, 1917, pp. 95—129.

79 Coon, C. S., Tribes of the Rif, contains a survey of some of this material. See also Bates, O., The Eastern Libyans.
Bertholon, L., and Chantre, E., Récherches anthropologiques dans la Berberie Orientale.
Fournel, H., Les Berbers.
Gautier, E. F., Les Siècles Obscurs dans l’Histoire du Maghreb; Sahara, the Great Desert.
Gsell, S., Histoire Ancienne de l’Afrique du Nord.
The primary sources for this section are chiefly: Herodotus, Sallust, Procopius, el Bekri, Ibn Khaldun, Marmol, Leo Africanus.

80 Bates, O., The Eastern Libyans, pp. 39—43.
Maspero, G., The Struggle of the Nations, p. 431.

81 As Almohades, or al-Muwahhids.