CHAPTER IX

PLATE 31MOSQUES.PLATE 32MOSQUES.The etymology given by the Arabs to the word “tarki” or “tawarek,” even if not strictly accurate, indicates that the People of the Veil adopted the Faith of Islam long after the other inhabitants of North Africa. When they did so, they appear to have been lukewarm converts and to have retained many practices which the Prophet directed good Moslems to abhor. At Ghat, which was ever under their influence and where numbers of them have always lived, the tradition of their recent conversion may be found in the twoparts of the town, known as the Quarter of Yes and the Quarter of No, from the people who accepted or refused Islam. At so late a period as when the Kel Owi arrived at the end of the seventeenth centuryA.D.the Kel Ferwan whom they drove out of the Iferuan valley in Northern Air were still “heathen,” though we are not told what their religion was. A very early date for the mosques of Air is therefore inherently improbable even if the Kel Geres did found Tefis as the first permanent place of worship for the new Faith. Assuming that the Kel Geres came to Air in the eleventh or twelfth century, the foundation of T’intaghoda mosque some 400 years later is not improbable; and it is not wholly impossible to reconcile such a date with the implications involved in the story of the gift of the mosque of Agades to the second Sultan of Air, who, we believe, reigned half-way through the fifteenth century. I prefer to consider that the mosques as a whole are not very old. Their style of construction demonstrates them to be more recent than the “A type” houses, though admittedly this view might have to be altered in the event of excavations providing additional or contradictory evidence.Apart from the numerous places of prayer marked by a “Qibla” of a few stones laid on the surface of the ground or by a quadrilateral enclosure of small stones, I only came across one site which might have been a pre-Moslem place of worship adapted to the later Faith. In the upper part of the River of Agades, on the south shore below the cliffs, at the entrance of the gulf where the Akaraq valley joins it, there is a square enclosure marked by what looks like the remains of a wall of which only the foundations on the ground level survive. The walls may never at any time have been more than a few inches high; what remains is of stones set in mud cement. At each of the four corners of the square there was a large stone. The four sides, each of some 15 ft. long, were true and square and oriented on the cardinal points. The enclosure was obviously not that of a hut, nor like the ground-plan of any of the houses in Air. In thecentre of the eastern side at a later period two standing stones had been set up. The stones were fossil trees, some other fragments of which were lying loose on the top of the neighbouring cliff. They had obviously been brought by human agency, as curious or interesting stones, from another place at no very remote period.[226]The two standing stones were about 2 ft. 6 in. apart. They were intended to mark the east, but were quite clearly later additions to the place, for they were merely standing, and not built into, the foundation of the enclosure. They were not even symmetrical or exactly in the centre of the side. The enclosure may, I think, be regarded as a pre-Moslem place of worship and not merely as a dwelling-house, because the “Qibla” pillars of an Islamic place of prayer could as readily have been set up elsewhere, had there not been a deliberate design to convert a site from one religious use to another. Its form does not resemble that of any of the usual buildings of Air. In the vicinity was a group of graves, some of which were circular enclosures, while others, obviously more recent in date, were oblong and correctly oriented from the Moslem point of view.The graves and tombs of Air might well form the object of interesting archæological excavation. Many of them display an indubitably non-Moslem appearance. The most common type which continues throughout the period of Tuareg occupation in one form or another is a ring of stones set on edge around a raised area covered with small white pebbles. The grave is too low to be termed a tumulus or mound, it is convex or shaped like an inverted saucer, but the centre rises only a few inches above the surrounding ground. The ring of stones may be roughly circular, oval or elliptical. In the Moslem period the graves are definitely oblong, the major axis being directed north and south, in order that the body may be placed in the grave with the head turned towards the east. The older graves were theround, or elliptical enclosures, the latter with no fixed orientation; the earlier they are the more nearly circular they seem to be. This is especially noticeable in the case of the graves near, and probably contemporary with, the “A type” houses at Tabello. A large central circular grave is often surrounded by smaller oval ones lying in any direction, clustering about a more important burial.The later Moslem graves are smaller, but the practice of covering the surface with white pebbles or chips of quartz continues. The shape becomes narrower, less circular and more inclined to turn into a rectangle. The appearance of head-stones or head and feet stones, which the Arabs call “The Witnesses,” coincides with correct Moslem orientation, but even in modern times it is rare to find any inscription. The few I saw were rough scratchings in Arabic script and sometimes, in T’ifinagh, of some simple name like “Muhammad” or “Ahmed.” I only saw one instance, at Afis, of an inscription of any length; it recorded the interment of a notable sheikh, and was scored with a pointed tool on a potsherd. Neither in the houses nor in the graves of Air is there any evidence of the Tuareg having attempted to cut stone. Even the petroglyphs are hammered and scratched but not chiselled.A great deal has been written about the funerary monuments of North Africa known as the “argem.”[227]They are found in many parts of the Northern Sahara, in the Ahnet mountains and the Adghar n’Ifoghas, and in the Nigerian Sudan, but not in Tuat. They have been reported in the Azger Tassili, at In Azawa on the north road from Air and at several points in Air. Bates reports them in the Gulf of Bomba and in the Nubian cemeteries of Upper Egypt.[228]They are enclosures of piled stones varying in shape from round to square, but generally the former; or they take the form of tumuli containing a cist or tomb. In certain cases the graves are described as surrounded by concentric circlesof stones. The distribution of these “argem” recalls immediately the geographical situation of the Tuareg. It would be easy to assume that their existence was due to this people, were it not for the difficulty that the monuments all appear quite late in date. To quote Gautier[229]: “En résumé la question des monuments rupestres du Sahara, funéraires et religieux, semble élucidée, au moins dans ses grandes lignes. Le problème d’ailleurs, tel qu’il se pose actuellement, et sous réserve de découvertes ultérieures, est remarquablement simple. En autres pays, en particulier dans les provinces voisines d’Algérie et du Soudan, le passé préhistorique se présente sous des aspects multiples. En Algérie les redjems abondent, mais on trouve à côté d’eux des dolmens, quelques sépultures sous roche, pour rien dire des Puniques et Romaines. Au Soudan, comme on peut s’y attendre, en un pays où tant de races sont juxtaposées, le livre de M. Desplagnes énumère des tombeaux de types divers et multiples, poterie, grottes sépulcrales, cases funéraires, tumulus.[230]Rien de pareil au Sahara. On distingue bien des types différents de redjem, les caveaux sous tumulus du nord qui sont peut-être influencés par les dolmens et sépultures romaines, les redjems à soutaches du Tassili des Azguers, les chouchets du Hogar qui semblent nous raconter l’itinéraire et l’expansion des nobles Touaregs actuels. . . . Parmi tant de pierres sahariennes entassées ou agencées par l’homme, on n’en connaît pas une seule qu’on peut soupçonner de l’avoir été par une autre main que Berbère.” But here the difficulty appears, for “ceci nous conduirait à conclure que les Berbères ont habité le Sahara dans toute l’étendue du passé historique et préhistorique si d’autre part tous ces redjems ne paraissaient récents. . . . Les mobiliers funéraires contiennent du fer, et on n’en connaît pas un seul qui soit purement et authentiquement néolithique. Cette énorme lacune est naturellement de nature à nous inspirer la plus grande prudence dans nos conclusions. D’autant plusque, après tout, les monuments similaires algériens, dans l’état actuel de nos connaissances, ne paraissent pas plus anciens.”While the distribution of “argem” seems then to coincide with, and be due to the Tuareg, the “Berbères” to whom Gautier refers arrived in North Africa and spread into the interior before the advent of the metal ages. The last word has certainly not been said regarding the age of these monuments, and in spite of this difficulty of dates I have little hesitation in finding in them evidence of the individuality and racial detachment of the Tuareg stock from that of the other Libyans, who do not seem to have used this funerary apparatus. After all, the late neolithic and early metal ages in inner Libya were hardly separate from one another, and in the south, where we know the Tuareg are only fairly recent arrivals, the lateness of the “argem” is readily understandable. But if we believe them to be due to the Tuareg, the earliest remains in the north must be far older than Gautier supposes.Although certain remains of a presumed funerary or religious nature in Air have been described as “argem,” it has apparently escaped notice that both the pre-Moslem as well as the later graves of the country are all linear descendants of the older and more pretentious monuments. Yet if the term has any significance at all, there has been a tendency perhaps to describe rather too many enclosures as “argem.” Certain examples illustrated by Gautier are probably devoid of any spiritual significance. There are in Air, for instance, especially in the north of the country near Agwau, a number of groups of concentric stone circles, which were simply enclosures round temporary huts or tents. The old hut circles of the T’imia village (Plate 29) show clearly how an isolated example might be assumed to have been a prayer or religious enclosure. Again, the circular heaps of stones at Elazzas resemble the “argem” illustrated by Bates[231]so much that one might be tempted to conclude that theywere such, if it did not happen to be known that they were the raised plinths on which huts used to be constructed. A deduction drawn from the occurrence of the latter might indicate that the origin of the true “argem” was derived from a desire to commemorate in death the only permanent part of a man’s hut dwelling in life. Such an explanation is not only permissible but even probable; it is even possible that in some cases tombs were actually made in the very floor of the hut or side of the pedestal where the deceased had lived.In the lower Turayet valley in Southern Air I passed a number of graves which seemed to suggest an intermediate type between the large prehistoric “rigm” and the later small enclosure of stones covered with white pebbles. The Turayet graves were small circular platforms like the hut foundations at Elazzas, but not more than 10 ft. in diameter with vertical sides a few inches above the ground level and flat tops covered with white stones. The occurrence of these tombs on the Turayet valley, not far from the mouth of the Akaraq valley, where also is perhaps a pre-Moslem place of worship, and the existence of what may prove a pre-Moslem urn burial cemetery at Marandet, all of which places are in the extreme south of Air, are interesting points when it is remembered that the first Tuareg inhabitants of Air came to the country from the south. It may nevertheless be pure coincidence that there seemed to be fewer obviously ancient monuments in Northern Air than in the southern part.The absence of funerary inscriptions is in marked contrast with the profusion of rock writings in Air. Written literature is, however, almost non-existent, but traditional poetry takes its place. The esteem in which poetry is held and the popularity which it enjoys are proof of the intellectual capacity which is present in this people.When it is realised that, alone among the ancient people of North Africa, the Tuareg have kept an individual script, it seems extraordinary that drawing, painting and sculpture should have remained in so primitive a state. Even if we are to admit that the earliest and therefore the best of therock drawings of North Africa are the work of the ancestors of the Tuareg, it is hardly possible to qualify them as more than interesting or curious. Few of them are beautiful. Some of the “Early Period”[232]drawings were executed with precision and care, but even if full allowance is made for the possibility of their having been coloured there are hardly any artistic achievements of merit. They do not bear comparison with the bushman drawings of South Africa, still less with the magnificent cave paintings of the Reindeer Age in Europe. But while some doubt exists regarding the authorship of the early drawings, the later North African pictures can be ascribed to the Tuareg without any fear of controversy. The Tuareg are still engaged in making them, but this modern work is even more crude. The drawings have become conventionalised; the symbols do not necessarily bear any likeness to the objects which they purport to represent.The rock drawings in Air display continuity from bad examples in the style of the early period down to the modern conventionalised glyphs. In most cases both the early and the late work is accompanied by T’ifinagh inscriptions. The earlier drawings represent animals which exist, or used to exist, in Air. The most carefully executed I saw were in the valley leading up from Agaragar to the pass into the Ighazar basin above Faodet. The place was near some watering-point, used by the northern Salt Caravan from Air to Bilma. The pictures were somewhat difficult to see as they had in part been covered by later drawings. The execution was rough, consisting of little more than an outline with a few markings on the bodies of some of the animals. As in the late petroglyphs there was no chiselling or cutting: the lines were made by hammering with a more or less suitable instrument and then by rubbing with a stone and sand. Among the animals thus represented, the giraffe and the ostrich in a wild state survive south of Air. An antelopewith sloping quarters and large lyre-shaped horns, the ox, the camel, the donkey, a horse, a large bird, and the human figure, both male and female, could also be traced. The large antelope I cannot identify for certain, but the large bird is probably the Greater Arab Bustard.In the later work the conventionalised symbols remain fairly constant. The ox is shown as a straight line with four vertical lines representing legs, a clear indication of the hump, and two short horns. The rectangular camel symbol had become so debased that for a long time I was at a loss to interpret it. The representations of the human figure are only curious inasmuch as they emphasise the long robe worn by the Tuareg and sometimes the cross bands over the breast, so typical of the Libyans in the Egyptian paintings. An interesting point in these rudimentary examples of the pictorial art is that even in the early period they portray a similar fauna and habit of life to those of to-day. A faint Egyptian influence may be detected in the human figures. I know of no drawings in Air to compare with the ones found by Barth at Telizzarhen, nor any which appeared to have a religious significance. The most interesting example is certainly that of the ox and cart referred to in the following chapter.The necessity of pictorial expression was evidently less felt than that of poetry, a condition to which nomadism has undoubtedly contributed. Yet even in ornament and draughtsmanship the Tuareg seem once to have reached a higher plane of civilisation in the past than that which they now possess and which their life has led them progressively to abandon.They have little knowledge of history outside their own tribal or group lore with the exception of that modicum of knowledge derived from a superficial study of the Quran. At the same time, men like Ahodu have heard and remembered stories of the past such as those of Kahena, Queen of the Aures, and of her fighting against the Arabs. Their knowledge of local geography is enormous, of the generalform or shape of North Africa small. They know of the Mediterranean and their language has a word for the sea. They have heard of the Nile, of Egypt, of the Niger and of Lake Chad, but they have only very vague inklings of the existence of Arabia or of the whereabouts of Istambul, where the Defender of the Faith lived. They can draw rough maps of local features on the sand and understand perfectly the conception of European maps on a wider scale. When I showed them an atlas with a map of the world and laboriously explained that it was a flat representation of a spherical object, Ahodu and Sidi surprised me by saying that they knew that the world was round, and that if you went in by a hole you would eventually come out on the other side. Duveyrier and others have been surprised at the knowledge of European countries and politics which they have found in the Sahara. The communication of news between distant parts of Africa is highly developed and at times astounding.PLATE 33TIFINAGH ALPHABETIf only on account of their script the Tuareg have deserved more attention in this country than they have received. I have no intention at this juncture of examining either T’ifinagh or Temajegh in detail, as they require study in a volume dedicated to them alone; but, as an ancient non-Arabic script which has survived in Africa, I cannot refrain from a brief description of the former. T’ifinagh is an alphabetic and not a syllabic script, but owing to the abbreviations practised in writing and the absence of all vowels except an A which resembles the Hamza or Alif, it has come to resemble a sort of shorthand. It is usually necessary to know the general meaning of any writing before it can be read. The T’ifinagh alphabet consists of between thirty and forty symbols varying somewhat from place to place. Duveyrier[233]collected an alphabet of twenty-three letters used in the north: Hanoteau,[234]who wrote the best grammar of Temajegh yet published, gives twenty-four letters: Masquerey[235]gives twenty-three letters for the Taitoq dialectand script: Freeman found twenty-five in the Ghadamsi Tuareg dialects. In addition to these letter symbols there are about twelve ligatures of two or sometimes three letters. All these signs are used in Air, but there are also certain additional symbols which may be alternative forms. Of the twenty-three to twenty-five letters in T’ifinagh, some ten only have been derived from the classical Libyan script as exemplified by the bilingual Thugga inscription now in the British Museum. Of these ten letters perhaps five have Punic parallels, while for the thirty known Libyan letters six Phœnician parallels have been found. It has hitherto been assumed[236]that the T’ifinagh alphabet was descended from the Libyan, which, it may be noted, has not yet been found in any inscription proved to be earlier than the fourth centuryB.C.Many theories have been advanced for the origin of the Libyan script, but Halévy is usually accepted as the most reliable authority on the subject. He supposed that the Libyan alphabet was derived from the Phœnician with the addition of certain non-Semitic symbols current nearly all over the Mediterranean. If this were universally admitted as the correct view it would still not be possible to explain why the T’ifinagh alphabet contains so many symbols which are not common to either the Libyan or Punic systems. On evidence which cannot here be examined in detail, it seems easier to believe that the ancestors of the Tuareg brought to Africa, or copied from a people with whom they had been in contact before reaching the Sahara, an alphabet replenished by borrowing certain symbols from a Libyan system partly founded on the Phœnician one. A consideration of this problem, like the one which concerns the Temajegh language itself, must be left to experts to resolve. As much false analogy and loose reasoning have been used on this question as on the subject of the origin of the Libyan races. One thing only seems to me to stand out, namely, that the T’ifinagh alphabet and Temajegh language were not evolved in Africa but came from without, probablyfrom the east or north-east, into the continent, where they developed independently. To postulate an Arabian origin, for instance, for T’ifinagh and Temajegh could not be construed as evidence in support of any theory regarding the origin of the Tuareg themselves. Linguistic evidence is notoriously unreliable from the anthropological point of view, since more often than not it only indicates some cultural contact. The most interesting aspect of the linguistic question is the evidence which it may afford regarding the cultural development of the older Tuareg. In their present stage of development there is no reason for them to have retained, still less for them to have evolved by themselves, any form of script. Their mode of life does not necessitate the use of writing: they are for the most part illiterate or are in process of becoming so. To have had and in so far as they still use T’ifinagh, to have retained an individual script, is to my mind the most powerful evidence in favour of the conclusion to which I have already on several occasions referred, namely, their far higher degree of civilisation in the past.In Air, T’ifinagh is dying out. One tribal group is famous for having retained it in current use more than any other section of the Southern Tuareg. The Ifadeyen men and women still read and write Temajegh correctly if somewhat laboriously. They use it for sending messages to each other or for putting up notices on trees or rocks, saying how one or other of them visited the place. Among most of the other tribes a knowledge of T’ifinagh is confined to the older women and a few men. The younger generation can neither read nor write either in T’ifinagh or in Arabic: the scribes and holy men usually only write in Arabic script. In the olden days all the Tuareg women knew how to write and it was part of their duties to teach the children.The rocks of Air are covered with inscriptions which have neither been recorded nor translated. Owing to the changing linguistic forms of Temajegh and the absence of any very fixed rules for writing it, it is difficult to decipher any butthe modern writings. Words are not separated, vowels are not written, and where one word ends with the same consonant with which the following one begins, a single symbol is usually written for the two.PLATE 34ROCK INSCRIPTIONS IN TIFINAGHT’ifinagh script may be written from left to right or from right to left, or up and down or down and up, or in a spiral or in the boustrophedon manner. The European authors who have written of Temajegh have variously reproduced T’ifinagh running from right to left and from left to right, but the two best authorities, Hanoteau and de Foucauld,[237]have adopted the former direction. It ill becomes me to differ from such learned authorities, but the existence of certain inscriptions in Air leads me to believe that the left to right manner was, there at least, perhaps the most usual system. OnPlate 40is reproduced anArabicinscription written by a Tuareg in Arabic characters running in the wrong direction, namely, from left to right, nor do I think the writer would have made this mistake unless he had been accustomed so to write in the only other script of which he could have had any knowledge, namely, T’ifinagh. The inscription, of course, records the common “La illa ilallah Muhammed rasul Allah.” I came across two or three other instances of the same sort.The T’ifinagh inscriptions in Air, like the pictures with which they are so often associated, belong to all periods. Some of them certainly date back to the first Tuareg invasion.There is a tradition that the Quran was translated into Temajegh and written out in T’ifinagh, a most improper proceeding from the Moslem point of view. But no European has seen this interesting book, which is said to have been destroyed. It may possibly have survived in some place, for Ahodu told me he had once seen a book in Air written in T’ifinagh, though all the documents which I found in the mosques were in Arabic calligraphy. Until a “Corpus” of T’ifinagh inscriptions has been compiled it will be very difficult to make much progress.Such a collection would assist in the study of Temajegh itself, for the language is in a somewhat fluid state, tending to vary dialectically from place to place and period to period. It is one of the languages termed “Berber,” the only connection in which I am prepared to admit the use of this word. By many it is considered the purest of the Berber forms of speech. Although related to such dialects as Siwi and Ghadamsi, and to western forms like Shillugh or the Atlas languages, Temajegh is distinct; it was not derived from them but developed independently, and probably preserved more of the original characteristics.The relationship of the original tongue to the Semitic groups of languages has not yet been defined. The two linguistic families have certain direct analogies, including the formation of words from triliteral verbal roots, verbal inflections, derived verbal formations, the genders of the second and third persons, the pronominal suffixes and the aoristic style of tense. Nevertheless there are also certain very notable differences, like the absence of any trace of more than two genders, the absence of the dual form, and verbs of two or three or four radicals with primary forms in the aorist and imperative only. Berber does not appear to be a Semitic language. But the two are probably derived from a common ancestor.The Air and Ahaggar dialects of Temajegh differ somewhat from each other. They are mutually quite intelligible, and so far as I could judge not more diverse than English and American. Barth stated that, unlike the rest of the Air Tuareg, the Kel Owi spoke the Auraghiye dialect, which is the name often given to the Ahaggar language. The name is, of course, derived from the Auriga or Hawara ethnic group, which, as we shall see, is the name of the parent stock of most of the Ahaggaren tribes. I have it on the best authority, however, of Ahodu, ’Umbellu and Sidi, that the Kel Owi language does not differ materially from the dialect of the rest of Air and am therefore at a loss to be able to explain Barth’s statement.The absence of the Arabic ع (’ain) in Temajegh necessitates its transcription by the letter غ (ghen) which is so characteristic of Berber. In all words, therefore, adopted from the Arabic, and especially in proper names like ’Osman, ’Abdallah, ’Abdeddin, etc., the forms Ghosman, Ghibdillah, Ghabidin are used. The Temajegh letter (yegh) ⵗ orghenis common and so stronglygrasseyéthat it becomes very similar to an R. The difficulty of transcription of the T’ifinagh into European languages is therefore very considerable,[238]for the R and Gh sounds are very confusing. In some T’ifinagh inscriptions the Arabic letter ع is frankly used when Arabic words occur.The great feature of the Temajegh language and of the Tuareg is the diffusion of poetry. It is unfortunately impossible to give any examples in this volume, but the collections made by Duveyrier, Hanoteau, Masquerey, Haardt,[239]and de Foucauld[240]show the natural beauty and simplicity of this art among the People of the Veil. Their prosody is not strict, but nevertheless displays certain formality. Iambic verses of nine, ten and eleven syllables are the most usual forms of scansion, with a regular cæsura and rhymed or assonated terminations. In the matter of rhymes there is considerable freedom: the use of similarly sounding words is allowed. Terminations like “pen,” “mountain” and “waiting” would, for instance, all be permissible as rhymes. Poetry is sung, chanted or recited with or without music. The themes cover the whole field of humanity, from songs of love or thanksgiving to long ballads of war and travel. The Tuareg are in some measure all poets, but the women are most famous among them. They make verses impromptu or recite the traditional poems of their race which are so old that their origin has been forgotten. One hears of women famous throughout the Sahara as the greatest poets of their time.Their way of life is attractive. These famous ladies hold what is called a “diffa,” which is a reception or “salon.” In the evening in front of their fires under an African night they play their one-stringed “amzad” or mandoline and recite their verses. Men from all over the country come to listen or take part. They seem to live and love and think in much the same manner as in Europe those of us do who retain our natural feelings. Only perhaps there are fewergrandes damesin Europe now than in the Sahara.Poetry, music and dancing are all to a great extent branches of a single art in so far as they all depend on rhythm and seek to express the emotions. In Air the syncopated music of the negro has had more influence than in the north, so the “amzad” is less common. Their other instruments are drums, but the lilt of their dance is rather different from that of the south. Their improvised drums are most ingenious. There is the hemispherical calabash floating in a bowl of milk, the note of which varies according to the depth to which the gourd is sunk, and the millet mortar with a wet skin stretched over the mouth by two parallel poles weighed down with large stones lying across their ends. The other various drums of the Southland are also known and used by those who can afford them. The dances of the Tuareg men are done to a quick step on a syncopated beat. The most effective one is a sword dance by a single man running up to the drum and executing a series of rapid steps, with the sword held by both hands at arms’ length above the head. I have never seen any women dancing among the Air Tuareg and it is said not to be their practice. This may be so, for even among the men dancing is relatively uncommon and has probably been borrowed from the south. It seems hardly to be consistent with their grave and dignified demeanour, of which poetry is the more natural counterpart.[216]Jean,op. cit., pp. 82 and 176.[217]Called Assingerma on the Cortier map. Teghazar is the diminutive of Ighazar, and means a small river or torrent.[218]Also spelt Reshwa. Von Bary calls the cone Teginjir, which is inaccurate.[219]Which is also called Tellia, as Barth refers to it.[220]Richardson,op. cit., Vol. II. p. 71.[221]Naturally many more of the “B” houses than of the “A” class still have the roof on them.[222]Cf.Chap. XI.[223]The evidence for these movements is inChap. XI.[224]Jean,op. cit., p. 86.[225]Jean throughout regards the Kel Owi as very ancient inhabitants of Air, but if due allowance is made for (as I think) this error and his traditions are not taken to refer to an earlier period than the one with which this group is associated, they are still valuable, from the comparative point of view.[226]Fossil trees exist in the sandstone hills of Eghalgawen and T’in Wana, a few miles away.[227]Or “rigm” or “rigem” in the singular.[228]Bates,op. cit., App. I.[229]Gautier,op. cit., p. 86.[230]Desplagnes:Le plateau Central Nigérien.[231]Bates,op. cit., App. I., Figs. 90, 93 and 94.[232]According to the classification of Pomel and Flamand. Cf. Frobenius:Hadshra Maktuba, and Flamand,Les Pierres Ecrites.[233]Duveyrier,op. cit.[234]Hanoteau,Grammaire de la Langue Tamachek, Algiers, 1896.[235]Masquerey,Dictionnaire et Grammaire Touaregs(Dialect des Taitoq).[236]As, for instance, by Bates,op. cit., p. 88, following Halévy.[237]De Foucauld,Dictionnaire Touareg-Français, 2 Vols., Alger.[238]Hence the difficulty surrounding the writing of Ghat, or Rat or Rhat. I have used “gh” through this volume, but the French usually use “r.”[239]See especially MM. Haardt and Dubreuil’s account of the Citroën Motor Expedition across the Sahara.[240]In R. Bazin’s life of Père de Foucauld.CHAPTER IXRELIGION AND BELIEFSNominallyat least all the Tuareg of Air are now Moslems with the possible exception of some of the Imghad of the Ikazkazan, who were described to me as Kufara (heathens). Nevertheless, even to-day the Tuareg are not good Moslems, and though, as a general rule, they say their prayers with regularity, they are remiss in such matters as ablutions. These they never perform except with sand or dust, which the Prophet enjoined were only to be resorted to on journeys or where water was scarce.As was explained at the beginning of this volume, the word “Tuareg” is not used by the people themselves. It is used in the first place by the Arabs, in a somewhat derogatory sense. Barth makes no doubt about the etymology of the word Tuareg, or, as he spells it, Tawarek. “. . . if the reader inquires who gave them the other name (i.e.Tuareg), I answer in full confidence, the Arabs; and the reason why they called them so was probably from their having left or abandoned their religion, from the wordترك(as in), ‘tereku dinihum’; for from other evidence which I have collected elsewhere it seems clear that a great part of the Berbers of the desert were once Christians . . . and that they afterwards changed their religion. . . .”[241]The name is written either with a ك or a ق, but according to the learned traveller more often with the former letter. The form “Terga” or “Targa” would, however, if the word is identical radically with “Tuareg,” point to ق being correct in a countrywhere this letter so often becomes a hard ج in the local Arabic. The singular form of “Tuareg” is “Tarki” or “Tarqi,” with both forms of plural,تواركandتاركيون. Duveyrier[242]and nearly all other authorities agree in accepting this etymology, though some have suggested that it meant “The People of the Sand.”[243]Others add, as an alternative explanation for theتركderivation, that it was not so much Christianity from which they fell away but Islam after their conversion, and in support of this their laxity in ritual is quoted. Duveyrier says that they were the “Abandoned of God” on account of the delay in their conversion to Islam and the numerous apostasies which occurred, or else because of their evil and violent habits of life. There is no doubt of the reproach attaching to the word, but the etymology is unsatisfactory. In its original usage it seems to have referred rather to a section of the Muleththemin than to the whole race[244]: if this observation is correct the religious flavour attaching to the word is misleading, and it becomes simply a proper name belonging to a section analogous to that of the Sanhaja and Hawara.The Tuareg of Air observe the usual religious feasts, but their fasting during Ramadhan, which they call Salla Shawal, like their ablutions, is usually excused on the grounds that they are travelling. On the first day of Ramadhan it is customary to visit the graves of ancestors and friends. The feast of Salla Laja or Laya is held on the tenth day of the moon of Zu’lhajja;[245]it is known in Turkey and Egypt as Bairam. On this occasion sheep are slaughtered and the people feast for three days. The feast of Bianu on the 20th of Muharrem is a sort of Saturnalia, and very similar to certain festivities described as occurring in Ashanti. Thefeast lasts for a day and a half and is marked by scenes of joy and happiness, for it commemorates God’s forgiveness of humanity after the Flood. There is much dancing and love-making and laughter, and the old people, the children and the unmarried persons of the villages and camps are sent out of the settlements while the revelries are in progress. The feasts of the Birthday of the Prophet and of the Beginning of the Year are also celebrated. It is customary when a journey is successfully completed to give a sheep to be sacrificed for the poor, and when there is much sickness among men or camels the same habit obtains. When three of our camels had died in rapid succession at Auderas we were urged to make sacrifice, and did so with three sheep.PLATE 35MT. ABATTUL AND VILLAGEI regret that I was never sufficiently fluent in Temajegh to learn much of the superstitions of the Tuareg of Air. Such information can only be obtained after prolonged residence among a people, and superficial conclusions are worse than useless. There is no doubt that underlying all their Islamic practices they hold fundamental beliefs dating from their earlier religious practices, regarding which only very few indications are available. The existence of certain apparently Christian survivals led Duveyrier and other authorities to assert that the Tuareg were Christians before they were converted to Islam, and I am prepared to accept this view in spite of the denials which have been expressed by so eminent a writer as Bates. De Foucauld, I understand, was also doubtful of their having been Christians, for among the earlier beliefs which he found to be retained by the Tuareg of Ahaggar he detected the remains of a polytheistic rather than a monotheistic system. Bates has laboriously collected all the references to religious beliefs among the Eastern Libyans, and any reader interested in the subject cannot do better than refer to his work, for even as far as Air is concerned I can add nothing thereto.[246]There are certain incontrovertible facts which demonstrate the influence, at least, of Christianity among the People ofthe Veil. Much has been written of their use of the cross in ornament, nor can its so frequent occurrence be entirely fortuitous. I am aware that the cross is a simple and effective form of decoration which any primitive people is likely, unless formally prohibited, to have used; but I find it hard to believe that the Tuareg, who, after all, are not so very primitive in their culture, however much of it they may have lost, had no other inducement than a lack of imagination to drag in at every turn this symbol which their religion expressly forbids them to use. Their cross-hilted sword, which has been likened to a Crusader’s, may be a chance example of the use of a design which is as convenient as it is simple, but the tenacity with which they cling to the form, and only to this form, is none the less curious. The cross in T’ifinagh script for the letter “Iet” (T) is doubtless a pure accident occasioned by the rectilinear character of the alphabet. But in that case the absence of the equally convenient diagonal or St. Andrew’s cross is strange. In other instances the appearance of the cross can be even less lightly dismissed. The traditional form of ornamentation on the Tuareg shield is purely and simply the Latin cross rising out of what in design, apparently, is a traditional representation of glory or light, depicted as a radiating mass. Bates argues that the occurrence of a drawing of a shield with a cruciform design thereon upon a rock in Tibesti is an argument against the view which I have adopted, and that the use of this symbol is probably due to a former practice of sun worship which he finds widespread in Libya. But when it is realised how much the Tuareg of Air, to consider only one group, raided in that direction, and how natural it would be for them to commemorate a success by drawing their shield and cross, which they regard as characteristic of themselves, on a rock, his explanation seems rather lame. In the curved top of the iron camel head-piece of Air I am inclined to see another survival of the cross, such as also is probably the square top of their spoons. The pommel of their camel saddle, a design which is always strictly maintained, is anotherconvincing example, especially if the whole equipment is compared with the Tebu sort. In construction the Tuareg and Tebu saddles are very similar, though the cantle of the latter is generally low. The pommel of the Tebu saddle takes the form of a short upright member without any crosspiece or cruciform tendency; it rarely rises much above the level of the rider’s legs. It may be said, on the contrary, that the cross pommel of the Tuareg saddle is the most prominent part of their whole gear. It is of no practical value whatsoever, for the grip of the rider’s legs never reaches as high as the projecting arms of the cross-top, and it is extremely inconvenient for rapid mounting or dismounting in their flowing robes. The cross is also extensively used in ornamenting the leather-work of the saddle, and it plays a considerable part in the traditional metal-work of the more expensive quality.

PLATE 31MOSQUES.

PLATE 31

MOSQUES.

PLATE 32MOSQUES.

PLATE 32

MOSQUES.

The etymology given by the Arabs to the word “tarki” or “tawarek,” even if not strictly accurate, indicates that the People of the Veil adopted the Faith of Islam long after the other inhabitants of North Africa. When they did so, they appear to have been lukewarm converts and to have retained many practices which the Prophet directed good Moslems to abhor. At Ghat, which was ever under their influence and where numbers of them have always lived, the tradition of their recent conversion may be found in the twoparts of the town, known as the Quarter of Yes and the Quarter of No, from the people who accepted or refused Islam. At so late a period as when the Kel Owi arrived at the end of the seventeenth centuryA.D.the Kel Ferwan whom they drove out of the Iferuan valley in Northern Air were still “heathen,” though we are not told what their religion was. A very early date for the mosques of Air is therefore inherently improbable even if the Kel Geres did found Tefis as the first permanent place of worship for the new Faith. Assuming that the Kel Geres came to Air in the eleventh or twelfth century, the foundation of T’intaghoda mosque some 400 years later is not improbable; and it is not wholly impossible to reconcile such a date with the implications involved in the story of the gift of the mosque of Agades to the second Sultan of Air, who, we believe, reigned half-way through the fifteenth century. I prefer to consider that the mosques as a whole are not very old. Their style of construction demonstrates them to be more recent than the “A type” houses, though admittedly this view might have to be altered in the event of excavations providing additional or contradictory evidence.

Apart from the numerous places of prayer marked by a “Qibla” of a few stones laid on the surface of the ground or by a quadrilateral enclosure of small stones, I only came across one site which might have been a pre-Moslem place of worship adapted to the later Faith. In the upper part of the River of Agades, on the south shore below the cliffs, at the entrance of the gulf where the Akaraq valley joins it, there is a square enclosure marked by what looks like the remains of a wall of which only the foundations on the ground level survive. The walls may never at any time have been more than a few inches high; what remains is of stones set in mud cement. At each of the four corners of the square there was a large stone. The four sides, each of some 15 ft. long, were true and square and oriented on the cardinal points. The enclosure was obviously not that of a hut, nor like the ground-plan of any of the houses in Air. In thecentre of the eastern side at a later period two standing stones had been set up. The stones were fossil trees, some other fragments of which were lying loose on the top of the neighbouring cliff. They had obviously been brought by human agency, as curious or interesting stones, from another place at no very remote period.[226]The two standing stones were about 2 ft. 6 in. apart. They were intended to mark the east, but were quite clearly later additions to the place, for they were merely standing, and not built into, the foundation of the enclosure. They were not even symmetrical or exactly in the centre of the side. The enclosure may, I think, be regarded as a pre-Moslem place of worship and not merely as a dwelling-house, because the “Qibla” pillars of an Islamic place of prayer could as readily have been set up elsewhere, had there not been a deliberate design to convert a site from one religious use to another. Its form does not resemble that of any of the usual buildings of Air. In the vicinity was a group of graves, some of which were circular enclosures, while others, obviously more recent in date, were oblong and correctly oriented from the Moslem point of view.

The graves and tombs of Air might well form the object of interesting archæological excavation. Many of them display an indubitably non-Moslem appearance. The most common type which continues throughout the period of Tuareg occupation in one form or another is a ring of stones set on edge around a raised area covered with small white pebbles. The grave is too low to be termed a tumulus or mound, it is convex or shaped like an inverted saucer, but the centre rises only a few inches above the surrounding ground. The ring of stones may be roughly circular, oval or elliptical. In the Moslem period the graves are definitely oblong, the major axis being directed north and south, in order that the body may be placed in the grave with the head turned towards the east. The older graves were theround, or elliptical enclosures, the latter with no fixed orientation; the earlier they are the more nearly circular they seem to be. This is especially noticeable in the case of the graves near, and probably contemporary with, the “A type” houses at Tabello. A large central circular grave is often surrounded by smaller oval ones lying in any direction, clustering about a more important burial.

The later Moslem graves are smaller, but the practice of covering the surface with white pebbles or chips of quartz continues. The shape becomes narrower, less circular and more inclined to turn into a rectangle. The appearance of head-stones or head and feet stones, which the Arabs call “The Witnesses,” coincides with correct Moslem orientation, but even in modern times it is rare to find any inscription. The few I saw were rough scratchings in Arabic script and sometimes, in T’ifinagh, of some simple name like “Muhammad” or “Ahmed.” I only saw one instance, at Afis, of an inscription of any length; it recorded the interment of a notable sheikh, and was scored with a pointed tool on a potsherd. Neither in the houses nor in the graves of Air is there any evidence of the Tuareg having attempted to cut stone. Even the petroglyphs are hammered and scratched but not chiselled.

A great deal has been written about the funerary monuments of North Africa known as the “argem.”[227]They are found in many parts of the Northern Sahara, in the Ahnet mountains and the Adghar n’Ifoghas, and in the Nigerian Sudan, but not in Tuat. They have been reported in the Azger Tassili, at In Azawa on the north road from Air and at several points in Air. Bates reports them in the Gulf of Bomba and in the Nubian cemeteries of Upper Egypt.[228]

They are enclosures of piled stones varying in shape from round to square, but generally the former; or they take the form of tumuli containing a cist or tomb. In certain cases the graves are described as surrounded by concentric circlesof stones. The distribution of these “argem” recalls immediately the geographical situation of the Tuareg. It would be easy to assume that their existence was due to this people, were it not for the difficulty that the monuments all appear quite late in date. To quote Gautier[229]: “En résumé la question des monuments rupestres du Sahara, funéraires et religieux, semble élucidée, au moins dans ses grandes lignes. Le problème d’ailleurs, tel qu’il se pose actuellement, et sous réserve de découvertes ultérieures, est remarquablement simple. En autres pays, en particulier dans les provinces voisines d’Algérie et du Soudan, le passé préhistorique se présente sous des aspects multiples. En Algérie les redjems abondent, mais on trouve à côté d’eux des dolmens, quelques sépultures sous roche, pour rien dire des Puniques et Romaines. Au Soudan, comme on peut s’y attendre, en un pays où tant de races sont juxtaposées, le livre de M. Desplagnes énumère des tombeaux de types divers et multiples, poterie, grottes sépulcrales, cases funéraires, tumulus.[230]Rien de pareil au Sahara. On distingue bien des types différents de redjem, les caveaux sous tumulus du nord qui sont peut-être influencés par les dolmens et sépultures romaines, les redjems à soutaches du Tassili des Azguers, les chouchets du Hogar qui semblent nous raconter l’itinéraire et l’expansion des nobles Touaregs actuels. . . . Parmi tant de pierres sahariennes entassées ou agencées par l’homme, on n’en connaît pas une seule qu’on peut soupçonner de l’avoir été par une autre main que Berbère.” But here the difficulty appears, for “ceci nous conduirait à conclure que les Berbères ont habité le Sahara dans toute l’étendue du passé historique et préhistorique si d’autre part tous ces redjems ne paraissaient récents. . . . Les mobiliers funéraires contiennent du fer, et on n’en connaît pas un seul qui soit purement et authentiquement néolithique. Cette énorme lacune est naturellement de nature à nous inspirer la plus grande prudence dans nos conclusions. D’autant plusque, après tout, les monuments similaires algériens, dans l’état actuel de nos connaissances, ne paraissent pas plus anciens.”

While the distribution of “argem” seems then to coincide with, and be due to the Tuareg, the “Berbères” to whom Gautier refers arrived in North Africa and spread into the interior before the advent of the metal ages. The last word has certainly not been said regarding the age of these monuments, and in spite of this difficulty of dates I have little hesitation in finding in them evidence of the individuality and racial detachment of the Tuareg stock from that of the other Libyans, who do not seem to have used this funerary apparatus. After all, the late neolithic and early metal ages in inner Libya were hardly separate from one another, and in the south, where we know the Tuareg are only fairly recent arrivals, the lateness of the “argem” is readily understandable. But if we believe them to be due to the Tuareg, the earliest remains in the north must be far older than Gautier supposes.

Although certain remains of a presumed funerary or religious nature in Air have been described as “argem,” it has apparently escaped notice that both the pre-Moslem as well as the later graves of the country are all linear descendants of the older and more pretentious monuments. Yet if the term has any significance at all, there has been a tendency perhaps to describe rather too many enclosures as “argem.” Certain examples illustrated by Gautier are probably devoid of any spiritual significance. There are in Air, for instance, especially in the north of the country near Agwau, a number of groups of concentric stone circles, which were simply enclosures round temporary huts or tents. The old hut circles of the T’imia village (Plate 29) show clearly how an isolated example might be assumed to have been a prayer or religious enclosure. Again, the circular heaps of stones at Elazzas resemble the “argem” illustrated by Bates[231]so much that one might be tempted to conclude that theywere such, if it did not happen to be known that they were the raised plinths on which huts used to be constructed. A deduction drawn from the occurrence of the latter might indicate that the origin of the true “argem” was derived from a desire to commemorate in death the only permanent part of a man’s hut dwelling in life. Such an explanation is not only permissible but even probable; it is even possible that in some cases tombs were actually made in the very floor of the hut or side of the pedestal where the deceased had lived.

In the lower Turayet valley in Southern Air I passed a number of graves which seemed to suggest an intermediate type between the large prehistoric “rigm” and the later small enclosure of stones covered with white pebbles. The Turayet graves were small circular platforms like the hut foundations at Elazzas, but not more than 10 ft. in diameter with vertical sides a few inches above the ground level and flat tops covered with white stones. The occurrence of these tombs on the Turayet valley, not far from the mouth of the Akaraq valley, where also is perhaps a pre-Moslem place of worship, and the existence of what may prove a pre-Moslem urn burial cemetery at Marandet, all of which places are in the extreme south of Air, are interesting points when it is remembered that the first Tuareg inhabitants of Air came to the country from the south. It may nevertheless be pure coincidence that there seemed to be fewer obviously ancient monuments in Northern Air than in the southern part.

The absence of funerary inscriptions is in marked contrast with the profusion of rock writings in Air. Written literature is, however, almost non-existent, but traditional poetry takes its place. The esteem in which poetry is held and the popularity which it enjoys are proof of the intellectual capacity which is present in this people.

When it is realised that, alone among the ancient people of North Africa, the Tuareg have kept an individual script, it seems extraordinary that drawing, painting and sculpture should have remained in so primitive a state. Even if we are to admit that the earliest and therefore the best of therock drawings of North Africa are the work of the ancestors of the Tuareg, it is hardly possible to qualify them as more than interesting or curious. Few of them are beautiful. Some of the “Early Period”[232]drawings were executed with precision and care, but even if full allowance is made for the possibility of their having been coloured there are hardly any artistic achievements of merit. They do not bear comparison with the bushman drawings of South Africa, still less with the magnificent cave paintings of the Reindeer Age in Europe. But while some doubt exists regarding the authorship of the early drawings, the later North African pictures can be ascribed to the Tuareg without any fear of controversy. The Tuareg are still engaged in making them, but this modern work is even more crude. The drawings have become conventionalised; the symbols do not necessarily bear any likeness to the objects which they purport to represent.

The rock drawings in Air display continuity from bad examples in the style of the early period down to the modern conventionalised glyphs. In most cases both the early and the late work is accompanied by T’ifinagh inscriptions. The earlier drawings represent animals which exist, or used to exist, in Air. The most carefully executed I saw were in the valley leading up from Agaragar to the pass into the Ighazar basin above Faodet. The place was near some watering-point, used by the northern Salt Caravan from Air to Bilma. The pictures were somewhat difficult to see as they had in part been covered by later drawings. The execution was rough, consisting of little more than an outline with a few markings on the bodies of some of the animals. As in the late petroglyphs there was no chiselling or cutting: the lines were made by hammering with a more or less suitable instrument and then by rubbing with a stone and sand. Among the animals thus represented, the giraffe and the ostrich in a wild state survive south of Air. An antelopewith sloping quarters and large lyre-shaped horns, the ox, the camel, the donkey, a horse, a large bird, and the human figure, both male and female, could also be traced. The large antelope I cannot identify for certain, but the large bird is probably the Greater Arab Bustard.

In the later work the conventionalised symbols remain fairly constant. The ox is shown as a straight line with four vertical lines representing legs, a clear indication of the hump, and two short horns. The rectangular camel symbol had become so debased that for a long time I was at a loss to interpret it. The representations of the human figure are only curious inasmuch as they emphasise the long robe worn by the Tuareg and sometimes the cross bands over the breast, so typical of the Libyans in the Egyptian paintings. An interesting point in these rudimentary examples of the pictorial art is that even in the early period they portray a similar fauna and habit of life to those of to-day. A faint Egyptian influence may be detected in the human figures. I know of no drawings in Air to compare with the ones found by Barth at Telizzarhen, nor any which appeared to have a religious significance. The most interesting example is certainly that of the ox and cart referred to in the following chapter.

The necessity of pictorial expression was evidently less felt than that of poetry, a condition to which nomadism has undoubtedly contributed. Yet even in ornament and draughtsmanship the Tuareg seem once to have reached a higher plane of civilisation in the past than that which they now possess and which their life has led them progressively to abandon.

They have little knowledge of history outside their own tribal or group lore with the exception of that modicum of knowledge derived from a superficial study of the Quran. At the same time, men like Ahodu have heard and remembered stories of the past such as those of Kahena, Queen of the Aures, and of her fighting against the Arabs. Their knowledge of local geography is enormous, of the generalform or shape of North Africa small. They know of the Mediterranean and their language has a word for the sea. They have heard of the Nile, of Egypt, of the Niger and of Lake Chad, but they have only very vague inklings of the existence of Arabia or of the whereabouts of Istambul, where the Defender of the Faith lived. They can draw rough maps of local features on the sand and understand perfectly the conception of European maps on a wider scale. When I showed them an atlas with a map of the world and laboriously explained that it was a flat representation of a spherical object, Ahodu and Sidi surprised me by saying that they knew that the world was round, and that if you went in by a hole you would eventually come out on the other side. Duveyrier and others have been surprised at the knowledge of European countries and politics which they have found in the Sahara. The communication of news between distant parts of Africa is highly developed and at times astounding.

PLATE 33TIFINAGH ALPHABET

PLATE 33

TIFINAGH ALPHABET

If only on account of their script the Tuareg have deserved more attention in this country than they have received. I have no intention at this juncture of examining either T’ifinagh or Temajegh in detail, as they require study in a volume dedicated to them alone; but, as an ancient non-Arabic script which has survived in Africa, I cannot refrain from a brief description of the former. T’ifinagh is an alphabetic and not a syllabic script, but owing to the abbreviations practised in writing and the absence of all vowels except an A which resembles the Hamza or Alif, it has come to resemble a sort of shorthand. It is usually necessary to know the general meaning of any writing before it can be read. The T’ifinagh alphabet consists of between thirty and forty symbols varying somewhat from place to place. Duveyrier[233]collected an alphabet of twenty-three letters used in the north: Hanoteau,[234]who wrote the best grammar of Temajegh yet published, gives twenty-four letters: Masquerey[235]gives twenty-three letters for the Taitoq dialectand script: Freeman found twenty-five in the Ghadamsi Tuareg dialects. In addition to these letter symbols there are about twelve ligatures of two or sometimes three letters. All these signs are used in Air, but there are also certain additional symbols which may be alternative forms. Of the twenty-three to twenty-five letters in T’ifinagh, some ten only have been derived from the classical Libyan script as exemplified by the bilingual Thugga inscription now in the British Museum. Of these ten letters perhaps five have Punic parallels, while for the thirty known Libyan letters six Phœnician parallels have been found. It has hitherto been assumed[236]that the T’ifinagh alphabet was descended from the Libyan, which, it may be noted, has not yet been found in any inscription proved to be earlier than the fourth centuryB.C.Many theories have been advanced for the origin of the Libyan script, but Halévy is usually accepted as the most reliable authority on the subject. He supposed that the Libyan alphabet was derived from the Phœnician with the addition of certain non-Semitic symbols current nearly all over the Mediterranean. If this were universally admitted as the correct view it would still not be possible to explain why the T’ifinagh alphabet contains so many symbols which are not common to either the Libyan or Punic systems. On evidence which cannot here be examined in detail, it seems easier to believe that the ancestors of the Tuareg brought to Africa, or copied from a people with whom they had been in contact before reaching the Sahara, an alphabet replenished by borrowing certain symbols from a Libyan system partly founded on the Phœnician one. A consideration of this problem, like the one which concerns the Temajegh language itself, must be left to experts to resolve. As much false analogy and loose reasoning have been used on this question as on the subject of the origin of the Libyan races. One thing only seems to me to stand out, namely, that the T’ifinagh alphabet and Temajegh language were not evolved in Africa but came from without, probablyfrom the east or north-east, into the continent, where they developed independently. To postulate an Arabian origin, for instance, for T’ifinagh and Temajegh could not be construed as evidence in support of any theory regarding the origin of the Tuareg themselves. Linguistic evidence is notoriously unreliable from the anthropological point of view, since more often than not it only indicates some cultural contact. The most interesting aspect of the linguistic question is the evidence which it may afford regarding the cultural development of the older Tuareg. In their present stage of development there is no reason for them to have retained, still less for them to have evolved by themselves, any form of script. Their mode of life does not necessitate the use of writing: they are for the most part illiterate or are in process of becoming so. To have had and in so far as they still use T’ifinagh, to have retained an individual script, is to my mind the most powerful evidence in favour of the conclusion to which I have already on several occasions referred, namely, their far higher degree of civilisation in the past.

In Air, T’ifinagh is dying out. One tribal group is famous for having retained it in current use more than any other section of the Southern Tuareg. The Ifadeyen men and women still read and write Temajegh correctly if somewhat laboriously. They use it for sending messages to each other or for putting up notices on trees or rocks, saying how one or other of them visited the place. Among most of the other tribes a knowledge of T’ifinagh is confined to the older women and a few men. The younger generation can neither read nor write either in T’ifinagh or in Arabic: the scribes and holy men usually only write in Arabic script. In the olden days all the Tuareg women knew how to write and it was part of their duties to teach the children.

The rocks of Air are covered with inscriptions which have neither been recorded nor translated. Owing to the changing linguistic forms of Temajegh and the absence of any very fixed rules for writing it, it is difficult to decipher any butthe modern writings. Words are not separated, vowels are not written, and where one word ends with the same consonant with which the following one begins, a single symbol is usually written for the two.

PLATE 34ROCK INSCRIPTIONS IN TIFINAGH

PLATE 34

ROCK INSCRIPTIONS IN TIFINAGH

T’ifinagh script may be written from left to right or from right to left, or up and down or down and up, or in a spiral or in the boustrophedon manner. The European authors who have written of Temajegh have variously reproduced T’ifinagh running from right to left and from left to right, but the two best authorities, Hanoteau and de Foucauld,[237]have adopted the former direction. It ill becomes me to differ from such learned authorities, but the existence of certain inscriptions in Air leads me to believe that the left to right manner was, there at least, perhaps the most usual system. OnPlate 40is reproduced anArabicinscription written by a Tuareg in Arabic characters running in the wrong direction, namely, from left to right, nor do I think the writer would have made this mistake unless he had been accustomed so to write in the only other script of which he could have had any knowledge, namely, T’ifinagh. The inscription, of course, records the common “La illa ilallah Muhammed rasul Allah.” I came across two or three other instances of the same sort.

The T’ifinagh inscriptions in Air, like the pictures with which they are so often associated, belong to all periods. Some of them certainly date back to the first Tuareg invasion.

There is a tradition that the Quran was translated into Temajegh and written out in T’ifinagh, a most improper proceeding from the Moslem point of view. But no European has seen this interesting book, which is said to have been destroyed. It may possibly have survived in some place, for Ahodu told me he had once seen a book in Air written in T’ifinagh, though all the documents which I found in the mosques were in Arabic calligraphy. Until a “Corpus” of T’ifinagh inscriptions has been compiled it will be very difficult to make much progress.

Such a collection would assist in the study of Temajegh itself, for the language is in a somewhat fluid state, tending to vary dialectically from place to place and period to period. It is one of the languages termed “Berber,” the only connection in which I am prepared to admit the use of this word. By many it is considered the purest of the Berber forms of speech. Although related to such dialects as Siwi and Ghadamsi, and to western forms like Shillugh or the Atlas languages, Temajegh is distinct; it was not derived from them but developed independently, and probably preserved more of the original characteristics.

The relationship of the original tongue to the Semitic groups of languages has not yet been defined. The two linguistic families have certain direct analogies, including the formation of words from triliteral verbal roots, verbal inflections, derived verbal formations, the genders of the second and third persons, the pronominal suffixes and the aoristic style of tense. Nevertheless there are also certain very notable differences, like the absence of any trace of more than two genders, the absence of the dual form, and verbs of two or three or four radicals with primary forms in the aorist and imperative only. Berber does not appear to be a Semitic language. But the two are probably derived from a common ancestor.

The Air and Ahaggar dialects of Temajegh differ somewhat from each other. They are mutually quite intelligible, and so far as I could judge not more diverse than English and American. Barth stated that, unlike the rest of the Air Tuareg, the Kel Owi spoke the Auraghiye dialect, which is the name often given to the Ahaggar language. The name is, of course, derived from the Auriga or Hawara ethnic group, which, as we shall see, is the name of the parent stock of most of the Ahaggaren tribes. I have it on the best authority, however, of Ahodu, ’Umbellu and Sidi, that the Kel Owi language does not differ materially from the dialect of the rest of Air and am therefore at a loss to be able to explain Barth’s statement.

The absence of the Arabic ع (’ain) in Temajegh necessitates its transcription by the letter غ (ghen) which is so characteristic of Berber. In all words, therefore, adopted from the Arabic, and especially in proper names like ’Osman, ’Abdallah, ’Abdeddin, etc., the forms Ghosman, Ghibdillah, Ghabidin are used. The Temajegh letter (yegh) ⵗ orghenis common and so stronglygrasseyéthat it becomes very similar to an R. The difficulty of transcription of the T’ifinagh into European languages is therefore very considerable,[238]for the R and Gh sounds are very confusing. In some T’ifinagh inscriptions the Arabic letter ع is frankly used when Arabic words occur.

The great feature of the Temajegh language and of the Tuareg is the diffusion of poetry. It is unfortunately impossible to give any examples in this volume, but the collections made by Duveyrier, Hanoteau, Masquerey, Haardt,[239]and de Foucauld[240]show the natural beauty and simplicity of this art among the People of the Veil. Their prosody is not strict, but nevertheless displays certain formality. Iambic verses of nine, ten and eleven syllables are the most usual forms of scansion, with a regular cæsura and rhymed or assonated terminations. In the matter of rhymes there is considerable freedom: the use of similarly sounding words is allowed. Terminations like “pen,” “mountain” and “waiting” would, for instance, all be permissible as rhymes. Poetry is sung, chanted or recited with or without music. The themes cover the whole field of humanity, from songs of love or thanksgiving to long ballads of war and travel. The Tuareg are in some measure all poets, but the women are most famous among them. They make verses impromptu or recite the traditional poems of their race which are so old that their origin has been forgotten. One hears of women famous throughout the Sahara as the greatest poets of their time.

Their way of life is attractive. These famous ladies hold what is called a “diffa,” which is a reception or “salon.” In the evening in front of their fires under an African night they play their one-stringed “amzad” or mandoline and recite their verses. Men from all over the country come to listen or take part. They seem to live and love and think in much the same manner as in Europe those of us do who retain our natural feelings. Only perhaps there are fewergrandes damesin Europe now than in the Sahara.

Poetry, music and dancing are all to a great extent branches of a single art in so far as they all depend on rhythm and seek to express the emotions. In Air the syncopated music of the negro has had more influence than in the north, so the “amzad” is less common. Their other instruments are drums, but the lilt of their dance is rather different from that of the south. Their improvised drums are most ingenious. There is the hemispherical calabash floating in a bowl of milk, the note of which varies according to the depth to which the gourd is sunk, and the millet mortar with a wet skin stretched over the mouth by two parallel poles weighed down with large stones lying across their ends. The other various drums of the Southland are also known and used by those who can afford them. The dances of the Tuareg men are done to a quick step on a syncopated beat. The most effective one is a sword dance by a single man running up to the drum and executing a series of rapid steps, with the sword held by both hands at arms’ length above the head. I have never seen any women dancing among the Air Tuareg and it is said not to be their practice. This may be so, for even among the men dancing is relatively uncommon and has probably been borrowed from the south. It seems hardly to be consistent with their grave and dignified demeanour, of which poetry is the more natural counterpart.

[216]Jean,op. cit., pp. 82 and 176.[217]Called Assingerma on the Cortier map. Teghazar is the diminutive of Ighazar, and means a small river or torrent.[218]Also spelt Reshwa. Von Bary calls the cone Teginjir, which is inaccurate.[219]Which is also called Tellia, as Barth refers to it.[220]Richardson,op. cit., Vol. II. p. 71.[221]Naturally many more of the “B” houses than of the “A” class still have the roof on them.[222]Cf.Chap. XI.[223]The evidence for these movements is inChap. XI.[224]Jean,op. cit., p. 86.[225]Jean throughout regards the Kel Owi as very ancient inhabitants of Air, but if due allowance is made for (as I think) this error and his traditions are not taken to refer to an earlier period than the one with which this group is associated, they are still valuable, from the comparative point of view.[226]Fossil trees exist in the sandstone hills of Eghalgawen and T’in Wana, a few miles away.[227]Or “rigm” or “rigem” in the singular.[228]Bates,op. cit., App. I.[229]Gautier,op. cit., p. 86.[230]Desplagnes:Le plateau Central Nigérien.[231]Bates,op. cit., App. I., Figs. 90, 93 and 94.[232]According to the classification of Pomel and Flamand. Cf. Frobenius:Hadshra Maktuba, and Flamand,Les Pierres Ecrites.[233]Duveyrier,op. cit.[234]Hanoteau,Grammaire de la Langue Tamachek, Algiers, 1896.[235]Masquerey,Dictionnaire et Grammaire Touaregs(Dialect des Taitoq).[236]As, for instance, by Bates,op. cit., p. 88, following Halévy.[237]De Foucauld,Dictionnaire Touareg-Français, 2 Vols., Alger.[238]Hence the difficulty surrounding the writing of Ghat, or Rat or Rhat. I have used “gh” through this volume, but the French usually use “r.”[239]See especially MM. Haardt and Dubreuil’s account of the Citroën Motor Expedition across the Sahara.[240]In R. Bazin’s life of Père de Foucauld.

[216]Jean,op. cit., pp. 82 and 176.

[217]Called Assingerma on the Cortier map. Teghazar is the diminutive of Ighazar, and means a small river or torrent.

[218]Also spelt Reshwa. Von Bary calls the cone Teginjir, which is inaccurate.

[219]Which is also called Tellia, as Barth refers to it.

[220]Richardson,op. cit., Vol. II. p. 71.

[221]Naturally many more of the “B” houses than of the “A” class still have the roof on them.

[222]Cf.Chap. XI.

[223]The evidence for these movements is inChap. XI.

[224]Jean,op. cit., p. 86.

[225]Jean throughout regards the Kel Owi as very ancient inhabitants of Air, but if due allowance is made for (as I think) this error and his traditions are not taken to refer to an earlier period than the one with which this group is associated, they are still valuable, from the comparative point of view.

[226]Fossil trees exist in the sandstone hills of Eghalgawen and T’in Wana, a few miles away.

[227]Or “rigm” or “rigem” in the singular.

[228]Bates,op. cit., App. I.

[229]Gautier,op. cit., p. 86.

[230]Desplagnes:Le plateau Central Nigérien.

[231]Bates,op. cit., App. I., Figs. 90, 93 and 94.

[232]According to the classification of Pomel and Flamand. Cf. Frobenius:Hadshra Maktuba, and Flamand,Les Pierres Ecrites.

[233]Duveyrier,op. cit.

[234]Hanoteau,Grammaire de la Langue Tamachek, Algiers, 1896.

[235]Masquerey,Dictionnaire et Grammaire Touaregs(Dialect des Taitoq).

[236]As, for instance, by Bates,op. cit., p. 88, following Halévy.

[237]De Foucauld,Dictionnaire Touareg-Français, 2 Vols., Alger.

[238]Hence the difficulty surrounding the writing of Ghat, or Rat or Rhat. I have used “gh” through this volume, but the French usually use “r.”

[239]See especially MM. Haardt and Dubreuil’s account of the Citroën Motor Expedition across the Sahara.

[240]In R. Bazin’s life of Père de Foucauld.

RELIGION AND BELIEFS

Nominallyat least all the Tuareg of Air are now Moslems with the possible exception of some of the Imghad of the Ikazkazan, who were described to me as Kufara (heathens). Nevertheless, even to-day the Tuareg are not good Moslems, and though, as a general rule, they say their prayers with regularity, they are remiss in such matters as ablutions. These they never perform except with sand or dust, which the Prophet enjoined were only to be resorted to on journeys or where water was scarce.

As was explained at the beginning of this volume, the word “Tuareg” is not used by the people themselves. It is used in the first place by the Arabs, in a somewhat derogatory sense. Barth makes no doubt about the etymology of the word Tuareg, or, as he spells it, Tawarek. “. . . if the reader inquires who gave them the other name (i.e.Tuareg), I answer in full confidence, the Arabs; and the reason why they called them so was probably from their having left or abandoned their religion, from the wordترك(as in), ‘tereku dinihum’; for from other evidence which I have collected elsewhere it seems clear that a great part of the Berbers of the desert were once Christians . . . and that they afterwards changed their religion. . . .”[241]The name is written either with a ك or a ق, but according to the learned traveller more often with the former letter. The form “Terga” or “Targa” would, however, if the word is identical radically with “Tuareg,” point to ق being correct in a countrywhere this letter so often becomes a hard ج in the local Arabic. The singular form of “Tuareg” is “Tarki” or “Tarqi,” with both forms of plural,تواركandتاركيون. Duveyrier[242]and nearly all other authorities agree in accepting this etymology, though some have suggested that it meant “The People of the Sand.”[243]Others add, as an alternative explanation for theتركderivation, that it was not so much Christianity from which they fell away but Islam after their conversion, and in support of this their laxity in ritual is quoted. Duveyrier says that they were the “Abandoned of God” on account of the delay in their conversion to Islam and the numerous apostasies which occurred, or else because of their evil and violent habits of life. There is no doubt of the reproach attaching to the word, but the etymology is unsatisfactory. In its original usage it seems to have referred rather to a section of the Muleththemin than to the whole race[244]: if this observation is correct the religious flavour attaching to the word is misleading, and it becomes simply a proper name belonging to a section analogous to that of the Sanhaja and Hawara.

The Tuareg of Air observe the usual religious feasts, but their fasting during Ramadhan, which they call Salla Shawal, like their ablutions, is usually excused on the grounds that they are travelling. On the first day of Ramadhan it is customary to visit the graves of ancestors and friends. The feast of Salla Laja or Laya is held on the tenth day of the moon of Zu’lhajja;[245]it is known in Turkey and Egypt as Bairam. On this occasion sheep are slaughtered and the people feast for three days. The feast of Bianu on the 20th of Muharrem is a sort of Saturnalia, and very similar to certain festivities described as occurring in Ashanti. Thefeast lasts for a day and a half and is marked by scenes of joy and happiness, for it commemorates God’s forgiveness of humanity after the Flood. There is much dancing and love-making and laughter, and the old people, the children and the unmarried persons of the villages and camps are sent out of the settlements while the revelries are in progress. The feasts of the Birthday of the Prophet and of the Beginning of the Year are also celebrated. It is customary when a journey is successfully completed to give a sheep to be sacrificed for the poor, and when there is much sickness among men or camels the same habit obtains. When three of our camels had died in rapid succession at Auderas we were urged to make sacrifice, and did so with three sheep.

PLATE 35MT. ABATTUL AND VILLAGE

PLATE 35

MT. ABATTUL AND VILLAGE

I regret that I was never sufficiently fluent in Temajegh to learn much of the superstitions of the Tuareg of Air. Such information can only be obtained after prolonged residence among a people, and superficial conclusions are worse than useless. There is no doubt that underlying all their Islamic practices they hold fundamental beliefs dating from their earlier religious practices, regarding which only very few indications are available. The existence of certain apparently Christian survivals led Duveyrier and other authorities to assert that the Tuareg were Christians before they were converted to Islam, and I am prepared to accept this view in spite of the denials which have been expressed by so eminent a writer as Bates. De Foucauld, I understand, was also doubtful of their having been Christians, for among the earlier beliefs which he found to be retained by the Tuareg of Ahaggar he detected the remains of a polytheistic rather than a monotheistic system. Bates has laboriously collected all the references to religious beliefs among the Eastern Libyans, and any reader interested in the subject cannot do better than refer to his work, for even as far as Air is concerned I can add nothing thereto.[246]

There are certain incontrovertible facts which demonstrate the influence, at least, of Christianity among the People ofthe Veil. Much has been written of their use of the cross in ornament, nor can its so frequent occurrence be entirely fortuitous. I am aware that the cross is a simple and effective form of decoration which any primitive people is likely, unless formally prohibited, to have used; but I find it hard to believe that the Tuareg, who, after all, are not so very primitive in their culture, however much of it they may have lost, had no other inducement than a lack of imagination to drag in at every turn this symbol which their religion expressly forbids them to use. Their cross-hilted sword, which has been likened to a Crusader’s, may be a chance example of the use of a design which is as convenient as it is simple, but the tenacity with which they cling to the form, and only to this form, is none the less curious. The cross in T’ifinagh script for the letter “Iet” (T) is doubtless a pure accident occasioned by the rectilinear character of the alphabet. But in that case the absence of the equally convenient diagonal or St. Andrew’s cross is strange. In other instances the appearance of the cross can be even less lightly dismissed. The traditional form of ornamentation on the Tuareg shield is purely and simply the Latin cross rising out of what in design, apparently, is a traditional representation of glory or light, depicted as a radiating mass. Bates argues that the occurrence of a drawing of a shield with a cruciform design thereon upon a rock in Tibesti is an argument against the view which I have adopted, and that the use of this symbol is probably due to a former practice of sun worship which he finds widespread in Libya. But when it is realised how much the Tuareg of Air, to consider only one group, raided in that direction, and how natural it would be for them to commemorate a success by drawing their shield and cross, which they regard as characteristic of themselves, on a rock, his explanation seems rather lame. In the curved top of the iron camel head-piece of Air I am inclined to see another survival of the cross, such as also is probably the square top of their spoons. The pommel of their camel saddle, a design which is always strictly maintained, is anotherconvincing example, especially if the whole equipment is compared with the Tebu sort. In construction the Tuareg and Tebu saddles are very similar, though the cantle of the latter is generally low. The pommel of the Tebu saddle takes the form of a short upright member without any crosspiece or cruciform tendency; it rarely rises much above the level of the rider’s legs. It may be said, on the contrary, that the cross pommel of the Tuareg saddle is the most prominent part of their whole gear. It is of no practical value whatsoever, for the grip of the rider’s legs never reaches as high as the projecting arms of the cross-top, and it is extremely inconvenient for rapid mounting or dismounting in their flowing robes. The cross is also extensively used in ornamenting the leather-work of the saddle, and it plays a considerable part in the traditional metal-work of the more expensive quality.

Back to Index Next