Navigation – Plan du site
Vestiges et territoire

Local Landscapes and Constructions of World Space: Medieval  Inscriptions, Cognitive Dissonance, and the Course of the Niger

Paulo Fernando de Moraes Farias

Résumés

En Afrique de l’Ouest sahélienne, l’Islam a induit des processus de modification cognitive qui, au cours du temps, contribuèrent à modifier la représentation que l’on se faisait des lieux et de l’histoire. Dans un univers musulman en expansion, elle fut reconstituée en référence aux lieux importants et aux événements fondateurs du monde islamique. Ces changements introduisirent des problèmes de dissonances cognitives dans l’interprétation du passé, les acteurs étant simultanément confrontés à des héritages heuristiques différents et incompatibles. Cet article étudie de tels changements dans la représentation du passé dans la vallée de Tadmăkkăt/Ǝssuk, située au nord du Mali, entre la période médiévale et les années 1980. À compter du XVIIe siècle, suite à des changements dans la nature des structures et des interactions sociales parmi les nomades, certains groupes revendiquèrent de nouveaux statuts empruntés au répertoire islamique. Le rôle commercial tenu par la région au Moyen-Âge fut expurgé des mémoires. Une nouvelle interprétation du passé fut surimposée sur la vallée, dans laquelle même les inscriptions médiévales arabes, pourtant datées, trouvèrent un nouveau sens. Après 1980, la montée du nationalisme Tuāreg allait encore engendrer une interprétation nouvelle du passé de ce site ancien. Une autre forme de dissonance cognitive apparait dans la description du cours du fleuve Niger de Léon l’Africain au xvie siècle et Muḥammad Bello au xixe.

Haut de page

Notes de l’auteur

An earlier version of this paper, under the title “Text as Landscape: Cultural Reappropriations of Medieval Inscriptions in the Seventeenth and Late Twentieth Centuries (Ǝssuk, Mali)”, was published in O. Hulec and M. Mendel (eds.), Threefold Wisdom: Islam, the Arab World and Africa—Papers in Honour of Ivan Hrbek (Prague: Academy of Sciences of the Czech Republic, Oriental Institute, 1993), 53–71. The text and bibliographic references have been updated. We thank O. Hulec and M. Mendel for their kind permission to republish the paper.

Une première version de ce texte a été publiée en 1993, sous le titre « Text as Landscape: Cultural Reappropriations of Medieval Inscriptions in the Seventeenth and Late Twentieth Centuries (Ǝssuk, Mali) », dans l'ouvrage de O. Hulec, M. Mendel, Threefold Wisdom: Islam, the Arab World and Africa—Papers in Honour of Ivan Hrbek, Prague, Academy of Sciences of the Czech Republic, Oriental Institute, 1993, p. 53–71. Le texte et la bibliographie ont été mis à jour. Nous remercions O. Hulec et M. Mendel de nous avoir autorisé à republier cet article.

Texte intégral

1This essay, which focuses on the West African Sahel, investigates the conditions under which history can be forgotten, even by societies containing groups endowed with literacy, a sense of the importance of writing about the past, and knowledge of absolute chronology. We will look at the merging of epigraphic texts into the physical landscape and at what may happen to such texts when resignifications of the landscape take place. We will see how particular ways of relating local space to world space generated new representations of the past and provided foundations for the rights claimed by certain local groups. However, those constructions of space generated interesting problems at other levels of representation (problematic assumptions about the course of the Niger being a case in point).

  • 1  In the transcription of Berber and Arabic words, the following special characters are used in this (...)

2We will examine the Ǝssuk1 valley in northern Mali (figs. 1, 2) in three different historical contexts: the Middle Ages, the seventeenth to mid-twentieth centuries, and the years since 1980.

  • 2  For recent, and very important, archaeological work at Ǝssuk, see S. Nixon, 2009, p. 217-255. The (...)
  • 3  However, on the characteristics and importance of Tifinagh inscriptions, and the relation between (...)

3At Ǝssuk, we find the ruins of a town and a concentration of inscriptions in either Berber (written in Tifinagh characters) or Arabic2. It is the Arabic inscriptions that fall within the scope of the present paper because of their implicit or explicit references to spaces beyond Africa and because many of them include absolute dates3. These Arabic texts are engraved on rock faces and on tombstones distributed over a number of necropolises. They display dates ranging from the early fifth century ah / eleventh century ad to the late eighth century ah / fourteenth century ad.

Figure 1: Ǝssuk / Tadmăkkăt in its regional context

Figure 1: Ǝssuk / Tadmăkkăt in its regional context

P.F. de Moraes Farias, 2003, Maps and Site Plans 1

  • 4  See P. Boilley, 1999, p. 48-49, 57-58.
  • 5  The name “Tuareg” or “Tuāreg” was not created by those to whom it is applied. It derives from the (...)
  • 6  They specialise in Arabic literacy, Islamic rites, educational and medical services, arbitration, (...)

4Ǝssuk is in the Aḍagh of Mali (fig. 1), long referred to in the literature as Aḍagh-n-lfoghas (“Aḍagh of the Ifoghas”), a region that was in close contact with Islamic culture outside West Africa in the Middle Ages. The Ifoghas4 are the Tuāreg5 group who became hegemonic in the area in the nineteenth century, before the beginning of French colonization. Ǝssuk is claimed as their place of origin by the Kăl-Ǝssuk (“those of Ǝssuk”), a number of Tuāreg groups with a specialism in Islamic culture6. Because of this specialism, the Kăl-Ǝssuk are classified as Ălfăqiten (a plural derived from the Arabic al-faqīh, “expert in Islamic law”) and as Inəsləmăn, sing. Anəsləm (a word which means “Muslims”, but which in this context takes the signification of “members of a ‘clerical’ group”). They have a well-known tradition of Arabic writing. In addition, learned men from the Kunta, a Moorish group with a long commercial, Islamic, and scholarly tradition, have been influential in the Aḍagh and have contributed to the construction of modern accounts of the past of the region.

5However, paradoxically, even those who define and legitimise themselves by reference to the medieval Islamic past of Aḍagh have forgotten the medieval history of Islam in Ǝssuk. A mythology of origins narrated by the Kăl-Ǝssuk (and the Kunta) has replaced that history. A large number of the Arabic inscriptions that survive in the Ǝssuk valley are legible and dated. The ability to read these inscriptions persists among Muslim lettrés. Yet this has not prevented the emergence of anachronistic stories about the past of the valley. Rather, this body of stories draws authority from the presence of the epigraphic texts, but brackets away the inscribed dates, and projects the inscriptions back onto an era more remote than these dates. It provides a symbolic idiom for the representation of social relations probably dating from the seventeenth century. In all likelihood, it has been created from that century onwards.

6We will see that since the 1980s a new kind of interest in the past of Ǝssuk has welled up in the Aḍagh. It does not derive from precolonial notions of social classification. Rather, it is in tune with contemporary Tuāreg nationalism.

Text and landscape

7History is unceasingly shed and kept. From what is kept, the past is reconstructed and reinvented, revealed and obscured. At the level of concrete continuity, what is kept are panels of landscape transformed by the interaction of natural processes and human activity, and individual cultural objects—among these, texts fixed in writing and subject to natural erosion and deliberate erasure. Written texts, especially if engraved on stone, often display an awesome durability that allows them to re-emerge after centuries of occultation or neglect. Yet their durability does not ensure that—if read—their content will govern the narratives of the past built upon them.

8Landscapes are themselves inscribed with meaning by stories and images attached to them, and may be read in competing ways. In a weaker metaphorical sense, or in a stronger semiotic one, we say that landscapes accept and solicit reading. Some of these readings may also prove to be of remarkable durability and may be reactivated after centuries. Moreover, even in arid expanses of the Sahara and Sahel, the landscape may include written information: Tifinagh and Arabic graffiti on rock faces, inscribed tombstones, decorative and pious writing on the walls of long-abandoned ruins, and other monumental captions. These are both physical and human geography, and the immediate literary context of the writings on the landscape is the landscape itself—the inherited and reinvented meanings people choose to attach to it. Under certain circumstances, there may be a clash between text and context. One way to solve this problem is to allow the context to assimilate the text, and to force the text into the role of dumb authority for the meanings attached to the non-verbal landscape. This has happened to the medieval inscriptions of Ǝssuk in modern times.

9In the Middle Ages, the Ǝssuk valley was perceived as both a topographical and a religious replica of the Mecca area in Arabia. And, in the modern era, Ǝssuk has been celebrated as the very cradle of Islam in West Africa, hence as a place that somewhat reflects the role played by Mecca in the Arabian Peninsula and the world at large. In the modern narratives, that medieval symbolic replication of space still governs the signification attributed to the Arabic inscriptions present in the Ǝssuk valley. However, it does so in ways that have little to do with what is actually said by the medieval inscriptions engraved on rock faces and tombstones—even though the linguistic competence and literacy required to read them is available. The epigraphic texts survive and are venerated, but as if crossed out, sous rature.

10A genre switch has taken place in the construction of the past. The floating “time of origins” characteristic of oral genres reasserts itself over the time recorded by the inscriptions and imposes its tone on the written-chronicle (ta˒rīkh) genre. Though the modern written accounts maintain an interest in absolute chronology, they turn a blind eye to the dates readily available in the inscriptions. They bring in much earlier dates instead, together with historical figures borrowed from the early history of Islam in Mecca and Medina. These accounts import into Ǝssuk not only a space but also a time that belong elsewhere in the world. The connection they establish between Ǝssuk and Mecca is very different from that proposed by the medieval Arabic epigraphy of Ǝssuk.

  • 7  See P.F. de Moraes Farias, 2003, chapter 7, p. 85.

11The messages embedded on the landscape in Arabic letters are now reduced to physical features of it. They are treated as one constitutive element of the huge non-verbal sign that is the valley itself, with its cliffs and town ruins. The Arabic writing in the inscriptions continues to signify, but does so independently of its verbal contents. Its meaning is now in the stories told about it, but external to it. In the same way, stories are told about other physical features of the valley: the passages through its cliffs are said to have been cut into the rock by Prophet Muḥammad’s son-in-law ˓Alī7.

The medieval period: the town of Tadmăkkăt and the Ǝssuk site

12The epigraphic evidence and urban ruins at Ǝssuk date back to medieval Tadmăkkăt, a town that was one of the most important crossroads of trans-Saharan trade during the period extending from the second half of the ninth century ad to, at least, the end of the first decade of the twelfth century ad. Tadmăkkăt was then frequented by Ibadite and other Muslim traders travelling from North Africa—from Tāhert (until the destruction of this Ibadite town in 296 ah / 909 ad), from Qayrawān (Kairouan) via Wārglān (Ouargla), and from Ṭarābulus (Tripoli) via the Jabal Nafūsa (an Ibadite area) and Ghadāmis (Ghadamès). It was also linked to the sub-Saharan commercial centres medieval Arabic sources called Ghāna (possibly the archaeological site at Kənbi-Sāləḥ, in the south of the Islamic Republic of Mauritania) and Kawkaw—i.e. Găwgăw or Gaawo, in the area of modern Gao and the archaeological sites at Saney (fig. 1).

  • 8  See T. Lewicki, 1981, t. 1, p. 439-444; P.F. de Moraes Farias, 1990a, p. 65-113, especially p. 74, (...)

13After the first decades of the twelfth century AD, the role of Tadmăkkăt as a centre of long-distance trade with North Africa seems to have declined. Yet the town survived, probably on account of its role as a market, and religious centre, catering for the pastoralist economy of the Aḍagh, and because of continuing trade exchanges between this region and the southern Sahara and the Niger valley. In the first half of the fourteenth century ad, it was still known to an external Arabic source as an independent, Muslim, and “Berber” kingdom (though no longer described in connection with trans-Saharan trade). There is also some textual evidence suggesting that the town was still, or had again become, a permanently inhabited place four hundred years later, in the early eighteenth century, but this remains to be confirmed by other written sources and archaeology8.

14For centuries, Tadmăkkăt/Ǝssuk has remained empty of permanent inhabitants, though its well has continued to be regularly visited by pastoralists and their herds.

  • 9 S. Chaker, 1995, p. 146. Other references in P.F. de Moraes Farias, 2003, p. cxliii.

15The medieval Arabic sources do not record the toponym “Ǝssuk” (in Arabic, “Al-Sūq”, pronounced “As-Sūq”). In a work completed in the 1220s ad, Yāqūt (perhaps quoting al-Muhallabī, who wrote in the late tenth century ad) refers to the “kingdom of Tādmāk [or Tādmākka]” and calls its capital Zakram—زكرام—probably a corruption of أكرام (an Arabic transcription of the Berber word aghrəm, which simply means “town, village”). However, in most sources we only find variant Arabic transcriptions (“Tādimakka”, “Tādamakka”, etc.) of the name transcribed in this paper as Tadmăkkăt according to its contemporary pronunciation in the Aḍagh. In Berber, this name means “This one, Mecca”, or “This one [is] Mecca itself” / “Celle-ci c’est la Mecque” (“Màkkàt” is the Tàmašàq form of the name “Mecca”)9. It is a name that equates two distant geographical spaces and identifies the West African town with the cradle of worldwide Islam.

  • 10  Ibnawqal [c. 988 AD], 1938-1939, vol. I, p. 92; Yāqūt [completed 1224-1229 ad], 1866-1870, vol.  (...)

16One reason for this name is that the town was described as being, like Mecca, “situated between two mountains intersected by ravines” (fig. 2). Yet Ibn Ḥawqal (tenth century ad), who first recorded this phrase, and Yāqūt, in this case definitely quoting al-Muhallabī, used these words to describe not Tadmăkkăt, but Awdaghust (a commercial town probably corresponding to the Tagdāwəst archaeological site, in the south of the Islamic Republic of Mauritania). It was later writers—the anonymous author of the Kitāb al-Istibār (c. 1136 AD) and al-Dimashqī (in the early fourteenth century AD)—who brought IbnḤawqal’s “between two mountains” description to bear on Tadmăkkăt. However, the allegedly unique “resemblance to Mecca” of this town had already been asserted by al-Bakrī, in the eleventh century AD, who also stated that the name “Tadmàkkàt” meant “[in the] form of Mecca”10.

Figure 2: Ruins and necropolises at Ǝssuk / Tadmăkkăt

Figure 2: Ruins and necropolises at Ǝssuk / Tadmăkkăt

P.F. de Moraes Farias, 2003, Maps and Site Plans 9

17Tadmăkkăt and Awdaghust played homologous roles in the network of trans-Saharan trade, and there are cliffs east and west of both the Ǝssuk and the Tagdāwəst sites. These two facts help to explain the occasional amalgamation of the descriptions of the two medieval towns, given the vagueness of the notions of sub-Saharan geography held by external Arabic authors. However, other factors were also at work in the superimposition of the image of Mecca on different Sahelian valleys. There was the impulse to link those outposts on the periphery of the Islamic world to the geographical centre of the space of Islamic prayer, in the same way as one did in the very act of turning towards Mecca while praying. It was also the impulse to make local landscapes meaningful in the imported idiom of Islam and to infuse them with certain values.

18One Tadmăkkăt / Ǝssuk inscription, which records no date, but which we assign on palaeographic grounds to the early fifth century ah / eleventh century ad, proclaims in its three last lines:

  • 11  See P.F. De Moraes Farias, 2003, inscription 104, p. 87-88; p. cxli, cxc-cxci, ccxxxi-ccxxxiii.

And there will remain to it [to Tadmăkkăt]
a market [sūq] in conformity to Mecca,
and the book [the Qur’ān] will remain.11

19This epigraphic text does not compare local and Meccan topography. Rather, it is the commercial and religious roles of Tadmăkkăt / Ǝssuk that are directly modelled on Mecca.

20Such comparisons and identifications, generated at first by Muslim outsiders living in partibus infidelium and keen to impose a familiar frame of reference on an alien space, later had a role to play within indigenous, West African Muslim culture. They became part of a wide process of geographical, genealogical, and historical transferences.

Spatial metaphors, holy geography, and the course of the Niger: the grounding of an idiom for the description of local social structure and worldwide history (seventeenth–twentieth centuries)

  • 12  See G.-R. de Gironcourt, 1911, p. 198-206.

21The region (fig. 1) extending from the Bentiya sites, which correspond to the medieval and early modern Songhay town of Kukiya (or Kutyia), to the Tadmăkkăt / Ǝssuk site has a modern history of being treated as a symbolic arena. It has been depicted as a scene of competition between opposite forces transferred, over time and space, from the sacred history sited in the Middle East. The Muslim cemeteries at Bentiya are traditionally associated with the alleged presence in the area of Prophet Muḥammad’s Companions (al-Ṣaḥāba in Arabic, and Eshahaba or Saab in Songhay).12 However, this alleged presence contrasts sharply with the image of an earlier Kukiya that was a centre of “paganism” not only in West Africa, but also in connection with the ever-reiterated duel between prophets and anti-prophets, which is at the heart of the Qur’ān and which is seen as extending back to before Muḥammad’s lifetime. The clash between Mūsā (the Biblical Moses) and the sorcerers of Pharaoh (Qur’ān VII:103–121) is reflected in a passage in one of the seventeenth-century Timbuktu chronicles:

  • 13 ʿAbd al-Ramān al-Sa˓dī, 1964, p. 4 [text]; p. 6–7 [trans.]; J.O. Hunwick, 1999, p. 6.

The town of Kukiya [is] very ancient. It is on the river shore in the territory of Songhay. It [already] existed in the time of Pharaoh, and it is said that it was from there that he gathered the sorcerers for the trial of strength with him to whom God spoke [Mūsā]—peace be upon him.13

  • 14 Al-Idrīsī [1154 ad], 1969, p. 10, 14, 19, 45 [text]; p. 11-12, 17, 23, 53 [trans.].

22The missing link between this passage and the Qur’ān is the oral, and written, lore attaching a reputation for sorcery to certain cities and assigning the geographical origins of Pharaoh’s sorcerers to some “ancient city” or other in the present Nile valley, or much further to the west. Particularly important in this context are the writings of al-Idrīsī (twelfth century ad). This author stated that the Nile had two branches, one running through Egypt and the other flowing through West Africa. In connection with West Africa, he refers to a riverine city, Kūgha (a town name often confused with that of Kukyia / Kutyia by medieval Arabic authors). His text says that Kūgha was famous for its sorceresses and linked by routes to Ghāna and Kawkaw—i.e. Găwgăw, Gaawo, or Gao (fig. 1). Other al-Idrīsī passages, which deal with Nubia and Egypt, mention the Nilotic town of Būṣīr, from which sorcerers were said to have come to Pharaoh, and then go on to state that another Nilotic town, Anṣinā, was the home of the sorcerers Pharaoh mobilised against Prophet Mūsā. The text also mentions another Nubian city, Kusha (a name easily confused with that of Kukyia / Kutyia as pronounced in Songhay)14. Thanks to the idea that the Niger and the Nile were branches of the same river, and the practice of equating different town names, it became possible for later authors like Al-Sa˓dī to identify Kukyia / Kutyia in West Africa as the place of origin of Pharaoh’s “sorcerers”.

  • 15  See M. Delafosse, 1913, p. 5-6.

23However, it was also possible to represent in other ways the assumed link between Pharaonic religion and West African traditional religions. West African stories narrated at Nioro, in Soninke country, and published by Delafosse state that after Pharaoh drowned in the sea the inhabitants of his kingdom fled to West Africa, bringing their religion with them. These stories do not refer to the Nile / the Niger15.

24Barth noted down an echo of these various alternative tales at Burem (fig. 1), on 10 June 1854, and proposed an historisante interpretation of it:

  • 16  See H. Barth, 1965, III, p. 464.

There is a remarkable tradition that a Pharaoh once came from Egypt to this spot, and returned. This story would at least imply an early intercourse with Egypt, and should not, I think, be viewed incredulously.16

25The postulated connection with Pharaonic “paganism” was a widespread ideological assumption among West African Muslims. It was not dependent on only one geographical conception. However, it went particularly well together with the notions that the Niger flowed out of, or into, the Nile—two variants of the mental map of African hydrography inherited by Muslim cultures from the geography of antiquity.

  • 17  See H. Barth, 1965, III, p. 491.
  • 18  For fascinating discussions of this map and its context, see C. Lefebvre, 2008, vol. 1, p. 106-110 (...)

26This mental map generated problems of cognitive dissonance as it clashed with empirical information. Yet, it survived such clashes. One of Barth’s Tuāreg friends suggested that an expedition from England should be sent up the Niger to initiate trade with the region. It is not clear whether it was locally believed that the ships would be coming from the Mediterranean or the Atlantic17. However, there is no such doubt in the case of Muḥammad Bello, caliph of Sokoto from 1817 to 1837. He provided Clapperton with a map produced by learned men at his court in which the course of the Niger is represented (fig. 3)18. An Arabic caption states:

  • 19  This Arabic word can mean either “sea” or “a large river”. Here it clearly refers to a river.

هذا  بَحرُ  کوارَ وهو الذي  يصل  الى  مصر المسمى  بالنيل  
This is the baḥr [river19] Kwāra [the Hausa name of the Niger], which reaches Miṣr [Egypt] and is called the Nile.

Figure 3: Map showing the course of the Niger given to Clapperton at Muḥammad Bello’s Court

Figure 3: Map showing the course of the Niger given to Clapperton at Muḥammad Bello’s Court

E.W. Bovill, 1966, Mission to the Niger, IV, The Bornu Mission, 1822-1825, Published for the Hakluyt Society, Cambridge, Cambridge University Press, Second Series, vol. CXXX, Inserted between pages 698 and 699

  • 20  Cf. D. Denham, H. Clapperton, W. Oudney, 1826, vol. 2, p. 302, 304-305, 309-310, 313-314, 331, and (...)

27On the map, after flowing between Borgu on the west and Nupe on the east, the Niger turns east on its way to the Nile valley. Nevertheless, in conversations with his visitor, Bello drew on the sand the course of the Niger / Kwāra and said that the river ended in the Atlantic. The caliph also expressed the wish that the king of England should send ships up the river for regular contact20. The Niger had been a link with the Nile and the Islamic East, but was about to turn round towards contacts with Western Europe via the Atlantic.

  • 21  See J. Cooper, 2007; L. Festinger, 1957.

28Precisely what intracultural negotiations took place, within Sokoto scholarship, between its two incompatible descriptions of the course of the Niger is an intriguing question. It falls beyond the scope of the present paper, but needs to be tackled by the anthropology and psychology of cognition. According to cognitive dissonance theory, when people hold two cognitions inconsistent with each other they find themselves in a state of uncomfortable tension. Hence, they are driven to take steps to reduce the dissonance and resolve the inconsistency21. Did this apply to the discrepant ideas about the Niger held in the Sokoto court? If it did, what avenues did Sokoto scholars follow in order to reconcile their cognitions?

  • 22  See the perceptive analysis of Leo’s text by F.-X. Fauvelle-Aymar, B. Hirsch, 2009, p. 83-102, esp (...)

29Similar questions may be asked of Leo Africanus and his text (1526). Like al-Idrīsī, but unlike the authors of the Sokoto-produced map we have discussed, Leo believed the Niger to be a branch of the Nile running east to west. At the same time, he claimed direct empirical knowledge of the course of the Niger, as somebody who had navigated it from Timbuktu towards the west. However, his knowledge, supposedly corroborated by personal experience (not simply based on book learning), led him to deny that the Niger flows west to east22. Does this mean that Leo never actually went to Timbuktu and the other parts of West Africa he says he visited? Or should we also consider the possibility that he did visit those places, but enjoyed a higher degree of tolerance to cognitive dissonance than we are now prepared to grant him?

  • 23 M. Last, 1980, p. 162-178.

30Whichever the case, it is clear that idioms borrowed from the classical Muslim heritage of learning for describing African reality were double-edged mental tools. They obfuscated certain features of that reality. Nevertheless, they also allowed the construction of serviceable representations of other features of it, through the use of historical metaphor and analogical geography—as discussed by Last in connection with the Kano Chronicle23 and as in the case of the name “Tadmăkkăt”. The descriptions of the course of the Niger went beyond metaphor and analogy, and actually identified that river with the Nile.

  • 24  See P.F. De Moraes Farias, 1990b, p. 109-147.

31All these forms of representation have been widely used by West Africans, and outsiders like Leo Africanus, when resorting to Muslim (and later Christian) idioms of explanation. Another example of this are the various kinds of assumptions fused into West African genealogies that stretch back to Arabia and into, say, the different accounts of the origins of the Yoruba put forward by Muḥammad Bello, the Aròkin of Òyò, Reverend Samuel Johnson, and al-Ḥājj Ādam al-Ilūrī24.

32The modern precolonial narratives of the origin of Ǝssuk / Tadmăkkăt deploy similar resources. They continue to circulate, and it is to their study that we must now return.

The resignification of the landscape and epigraphy of Ǝssuk in the wake of social change in the seventeenth-century Aḍagh

  • 25  See E. Hodgkin, 1987, p. 431-436.
  • 26  On the Iwəlləmmədăn, see Ghubăyd ăgg-Ălăwjeli, 1975; Altinine agg-Arias, 1974; A. Richer, 1924; F. (...)

33A new balance of political power, and new forms of social stratification and division of labour, emerged among the West African Tuāreg from the seventeenth century ad onwards. In the Aḍagh (fig. 1), the seventeenth century was a time of upheaval. A time for loyalty switches, redefinition of social roles, reshaping of group identities, and adoption of new group appellations. A new confederation of tribes emerged in the area—the Iwəlləmmədăn—and eventually forced another umbrella group—the Kăl-Tadmăkkăt (“those of Tadmăkkăt”)—out of the Aḍagh and towards the region around Timbuktu. Disputes about the rules of transmission of power appear to have occurred, and the equilibrium between available pastures and the size of the herds was apparently broken25. The displacement of the Kăl-Tadmăkkăt was followed by the emigration and split of the Iwəlləmmədăn confederation itself26. Some Kăl-Ǝssuk groups attached themselves to sections of the Iwəlləmmədăn and followed them out of the Aḍagh. Other Kăl-Ǝssuk emigrated from the Aḍagh independently.

  • 27 On the social organisation and pastoralist economy of the Tuāreg, see J. Nicolaisen, I. Nicolaisen, (...)

34In the form in which they emerged from these upheavals, Tuāreg societies represent themselves on different classificatory grids. On one of these, they are divided into tewšiten (a term often translated as “tribes”), which in principle are descent groups derived from a common ancestor and which are often grouped together into “confederations” or “drum groups”. But, on another classificatory grid, they are organized into categories of status and specialism, which are designated by terms such as Imušagh (or, according to the language branch concerned, Imašəghăn / Imajəghăn), usually translated in the literature as “noblemen, noble warrior groups”); Iməghad / Imghad (“tributary warrior groups”); Ălfăqiten or Inəsləmăn (members of “clerical” groups, marabouts, sing. Anəsləm); Inăăn (artisans working on metal and wood); Ighăwelăn (descendants of freed slaves who lived in farming communities of their own); and Eklăn (descendants of former slaves living with, and working for, other status categories)27.

  • 28  See P. Bonte's introduction to S. Bernus et alii., 1986 and in the same book P. Bonte, 1986, p. 23 (...)
  • 29  J.S. [B.] Lecocq, 2002, p. 16-23.

35In practice, while some tewsitenconsist of people who all belong to the same social stratum, others are defined territorially rather than genealogically and bring together different status categories—but with the ancestor of reference being claimed only by the highest status group28. Also, the status terms listed above do not uniformly apply across the geographical space inhabited by Tuāreg societies. Besides, the received translations of these terms are often misleading. As cogently argued by Baz Lecoq29, a new analysis of Tuāreg social stratification is needed.

  • 30  See ʿAbd al-Ramān al-Sa˓dī, 1964, 320 [text], 483 [trans.] and H.T. Norris, 1975, p. 109.
  • 31 H. Barth, 1965, vol. III, p. 761.

36But it is clear that the Kăl-Ǝssuk have processed their memories of Ǝssuk in the light of a perceived polarity between “clerical groups” and “warrior groups”. Their choice of name is very significant. It is ultimately to Tadmăkkăt, and to its Mecca-like role in the Sahel, that they trace back their distinctive corporate identity. Yet they do not share the name of the Kăl-Tadmăkkăt warrior groups who were forced to move out of the Aḍagh and into the Timbuktu region in the seventeenth century30. Instead, the Kăl-Ǝssuk adopted a name of Arabic origin. In the Tămašăq language, except for the word ewet or ewuit, which was recorded by Barth in the Timbuktu area31, but which is not current elsewhere, the words used for “market” are loan words: hebu, borrowed from Songhay, and əssuk, from the Arabic. Literally, “Kăl-Ǝssuk” means “those of the market town”.

  • 32  See Sīdī Mawlāy Muḥammad al-Hādī, Naṣīḥat al-˒umma, MS 94 of the collection formerly kept in the A (...)

37Yet the Kăl-Ǝssuk identification with Ǝssuk is not with a Muslim trade ethos, but with an ethical drama. This pseudo-historical dramatic episode represents two things. One is the early establishment of Islam in the Tuāreg lands. The other is the post-seventeenth century structural relationship between the Inəsləmăn “clerical” groups and the Imušagh “warrior” groups, as depicted from the point of view of Kăl-Ǝssuk ideology. The medieval role of Tadmăkkăt as a market for trans-Saharan trade has ceased to be historically pertinent to the Kăl-Ǝssuk and hence has been forgotten by them. However, their nineteenth-century literature states that Tadmăkkăt and Ǝssuk are the same town32. The names “Kăl-Ǝssuk” and “Kăl-Tadmăkkăt” both refer to that town, but these two names came to reflect the contradistinction perceived by the clerical groups as central to Tuāreg society. This notion is also central to the anachronistic reinterpretation of the history of Tadmăkkăt / Ǝssuk by the Kăl-Ǝssuk and the Kunta.

38The ethical drama contrasts all the “clerical” groups not only with the Kăl-Tadmăkkăt, but also with the Iwəlləmmədăn and all other “warrior” groups. It provides the Kăl-Ǝssuk, and their cultural specialism, with a high social position in the land by right of conquest.

39It is a myth that places itself in the first century ah / seventh century ad, and leaves out the period to which the Ǝssuk epigraphic evidence dates back (fifth century ah / eleventh century ad–eighth century ah / fourteenth century ad).

  • 33  See P.F. de Moraes Farias, 2003, inscription 106, p. 89.

40The earliest dated Ǝssuk inscription, incised on a rock face33, proclaims

هذا سنة أربعة وٲربعة مئة  
This is the year four hundred and four [404 AH / 1013–14 ad].

41By contrast, a Kăl-Ǝssuk ta˒rīkh (probably written in the nineteenth century) dates the supposed Muslim military conquest of Ǝssuk as follows:

  • 34  The quotations are from a four-page anonymous and untitled text, two copies of which are kept in t (...)

عقبة المستجاب في عام احدي و ستين من الهجرة ......
...... و فتح  ال
ڷه  السوق  على  يده
˓Uqba al-Mustajāb in the year sixty-one of the Hijra [61 ah / 680–81 ad]
………….and Allāh conquered Al-Sūq [Ǝssuk] by his hand.34

42The ta˒rīkh starts from the coming of the Arab Muslim conqueror ˓Uqba to North Africa in 61 ah / 680–81 ad. It then brings him to wage jihād at Ǝssuk. It goes on to say that a large number of al-aḥāba (Prophet Muḥammad’s companions) accompanied ˓Uqba to Ǝssuk, and that many of them are buried there. Those who belonged to the Muhājīrūn (Prophet Muḥammad’s Meccan companions who followed him to Medina) are buried on one side, and those who were Anṣār (Medinan companions of the Prophet) are buried on the other. Thus, the text puts forward a would-be explanation for the distribution of the necropolises around Ǝssuk (fig. 2). It also claims that the names of those al-aḥāba may still be read on the tombstones. One assumes that many pious and Arabic-reading visitors to the Ǝssuk necropolises, having failed to find such names (and dates) there, nevertheless would feel able to confirm their alleged presence—as Sokoto scholars were able to confirm imaginary maps of the course of the Niger, which clashed with their empirical knowledge.

  • 35  See J.M. Abun-Nasr, 1971, p. 68-69.

43Of course, the myth transfers to Ǝssuk well-known episodes of the conquest of North Africa by Muslim armies in the seventh century AD, and puts a new gloss on them35. The historical conflict between ˓Uqba and his Berber adversary Kusayla is moved across the Sahara from the Maghrib to the Aḍagh. Ǝssuk is represented as a mighty “pagan” city, which was nevertheless taken over by the superior might of Islam through holy war. ˓Uqba arrives as conqueror but his violence is righteous, and his might proves superior to “pagan” might. Clearly, no right resting only on “pagan”-like might (as in the case of the “warrior” groups) could possibly exceed the right to possess, and to belong, inherited by the Inəsləmăn “clerical” groups from ˓Uqba al-Mustajāb’s conquest.

44This myth exists in many oral and written versions. In it, Ǝssuk is a paradigmatic meeting place, in which Muslims from the heartland of Islam will settle and intermarry with newly converted locals. It is a metaphor for the conversion of the Sahel to Islam or, rather, for the creation of Muslim communities in the area. In the “time of origins” of the myth, opposite ways of life meet and clash, and at one level Islam definitively conquers its contrary. However, some of the original difference perversely re-emerges in disguise, and will renew evil and lead to murder and destruction. The initial conflict is transformed at one level into a true fusion of Arabs and Berbers under the aegis of Islam. This is presented as the origin of Kăl-Ǝssuk identity. But, at another level, that conflict is just hidden behind a façade of nominal, hypocritical conversion, liable to relapse into unholy violence at any moment. This is an implicit depiction of the ethos of the Tuāreg “warrior” groups, as seen from the point of view of Kăl-Ǝssuk ideology.

  • 36  See H.T. Norris, 1975, p. 22, 25; P.F. de Moraes Farias, 2003, p. cxliv.

45Like the Kăl-Ǝssuk themselves, ˓Uqba was not to stay forever at Ǝssuk. However, he stayed there long enough to build mosques and establish Islam. Kusayla, the defeated Berber ruler of “pagan” Ǝssuk, also stayed there pretending to have become a Muslim. (In some versions, a villainous and violent role somewhat similar to Kusayla’s is played by the Songhay conqueror Sonyi Ali, whose ambiguous attitude towards Islam is recorded in Sahelian traditions, and who is described in such versions as responsible for the decadence of Ǝssuk and dispersion of its inhabitants.36)

  • 37  See H.T. Norris, 1975, p. 24-25; P.F. de Moraes Farias, 2003, p. cxxxvi-cxxxvii.

46According to the Kăl-Ǝssuk ta˒rīkh we have cited, eventually ˓Uqba and most of his army leave Ǝssuk on their way back to Arabia, never to return. Kusayla attaches himself to ˓Uqba’s army but later kills him—not on a North African battlefield as narrated by the medieval historical sources, but at prayer in Walāta (Iwalātǝn), during the celebration of the Muslim festival of ˓Īd al-Kabīr (the location of this episode in Walāta reflects the linkage between these Kăl-Ǝssuk stories and the Kunta narratives37). ˓Uqba dies while engaging in a peaceful activity emblematic of Islam and the Inəsləmăn. Significantly, it is not a warrior’s heroic death like that of his historical namesake killed in North Africa in the 680s ad, but it allows him to remain the conqueror of Ǝssuk unvanquished in the battlefield.

  • 38 H.T. Norris, 1975, p. 17, 25.
  • 39  See H.T. Norris, 1986, p. 126-127, 185, 192, 220, who comments on this text and translates it. See (...)

47As Norris has perceptively noted, in his martyrdom, and in the role of water diviner also attributed to him, ˓Uqba incarnates two central aspects of the ideal image of the Anəsləm38. His example can be contrasted with the traditional code of behaviour of the “warrior” groups, which was geared neither to the peaceful practice of Islam nor to holy war, but to raiding and armed self-defence. Nevertheless, the stories represent ˓Uqba as a victorious military conqueror, and this was important to the Inəsləmăn in the context of their relations with other groups within modern precolonial Tuāreg society. The point is perhaps made clearer in a late nineteenth-century text by Shaykh Muḥammad Iknan wa-n Fadasen al-Sūqī al-Gunahānī, a Kăl-Ǝssuk scholar. It is a long polemical letter, written to refute claims that it was legal to raid the Kăl-Ǝssuk and seize their possessions. It defends the rights of the Kăl-Ǝssuk, by arguing that the land was theirs before the rise to power of the Iwəlləmmədăn39.

48From the seventeenth century to the onset of the colonial period and beyond, the myths about the coming of ˓Uqba to the Aḍagh performed important social functions, in a land where inter-group violence was endemic, but where, at the same time, the warrior and the cleric were called to balance reciprocal obligations or keep their imbalance within tolerable limits. Yet that body of mythology worked against the preservation of accurate memories of the history of Ǝssuk / Tadmăkkăt, and against the proper reading of the epigraphic information at the site. It went further than the old medieval analogies between the physical landscapes of Mecca and Ǝssuk, and between the historical characteristics of the two cities. Companions of the Prophet, transferred from the Middle East and North Africa to the Sahel, were directly superimposed on the Ǝssuk landscape, and a chronology borrowed from the first century ah / seventh century ad was brought to bear on the local inscriptions. This infused the landscape with new meaning, but this meaningfulness devoured history.

The years since 1980

49In 1987, after an evening of conversation about the origins of Islam in Ǝssuk, a group of Idnan pastoralists of the Aḍagh, otherwise not much given to scholarly pursuits, dug a trench in the ruins of that town. They discovered human bones and some stone slabs bearing Arabic inscriptions, which they could not read. The stones were then carefully stored, in the house they had built in the valley to be the permanent seat of their (not government-recognised) cooperative. In a nutshell, the unofficial excavation and cooperative represented a new kind of interest in Ǝssuk among the populations of the region. This new interest combined economic and other motivations.

50It partly resulted from environmental degradation and increased difficulty in obtaining water supplies. Few permanent wells were within the reach of the pastoralists exploring the pastures of the middle segment of the Telămse (Tilemsi) valley (fig. 1). Two of those wells were at Ǝssuk. This had encouraged those Idnan to establish a permanent presence there.

51Houses were built, and stones taken from the old ruins were used to lay the foundations of their banco walls. Other areas were marked out for building and vegetable plots were established on the bed of the Ărojj wa-n Ǝssuk (fig. 2). A tragic blow to these initial efforts came on 13–15 September 1987, when a flash flood destroyed several houses and killed two people and many animals. Yet the cooperative building and one or two houses survived the onslaught of the waters, as did the pastoralists’ interest in the valley as a building site.

52However, among young intellectuals in Kidal, the capital of the Aḍagh, interest in the eventual reurbanisation of Ǝssuk had been growing for some time before 1987, in the context of discussions about the need for a general revival of Kăl-Tămašăq culture and the Islamic dimension of its heritage. The severe droughts of 1968–73 and 1983–85 had been rightly seen as a threat to the very survival of the way of life of the Kăl-Tămašăq. The local intelligentsia had become acutely aware of the need to protect and renew that way of life, and more vocal in the expression of their grievances against the Malian central government. Hence environmental, political, and religious considerations gravitated towards one another to outline a “nationalist” platform, which held appeal for both pastoralists and intellectuals.

  • 40  On these uprisings see P. Boilley, 1999; G. Klute, 1999, p. 455-472; B. Lecocq, 2010.

53It was in this context that the idea of a “return” to Ǝssuk took shape. Many of those favouring it had a “western” education and had learned about the history of Tadmăkkăt / Ǝssuk at school. But they were also steeped since childhood in the legends about the valley, its ruins, and its (largely unread) inscriptions. Both sources of information came together to direct their attention to Ǝssuk, rather than to other archaeological sites in the region. Theirs was an interest fostered no longer by the dynamics of precolonial categories in Kăl-Tămašăq society, but by modern cultural nationalism. This interest predated the armed uprisings that took place between 1990 and 1996, and again between 2006 and 200940.

  • 41 Dawlat is a loanword from Arabic meaning “state”, “dynasty”, “power”, “empire”.
  • 42  See P.F. de Moraes Farias, 2003, inscriptions 154a.1, 154a.2, 154b.1, 154b.2, p. 130-133; p. cxl-c (...)

54In the years following the end of the first of these uprisings, an association for the investigation of the history of Ǝssuk and Dawlat-n-Tadmăkkăt (“the Tadmăkkăt state”41) was formed in Kidal. In fact, in the medieval days when Tadmăkkăt / Ǝssuk was a great commercial centre, unusual political institutions—not yet fully understood by historians—were apparently developed there42. Further research on the originality of those institutions is to be welcomed. But, for some, the emphasis is rather on recovering from the past the image of an ancestral, “classic” state for the Tuāreg of today—a people without a national state of their own. From this point of view, a new resignification of the Ǝssuk landscape is now in progress.

Haut de page

Bibliographie

ʿAbd al-Ramān al-Sa˓dī, 1964 [reprint of the 1898-1900 publication], Ta’rīkh al-Sudān, O. Houdas with E. Benoist (ed. and trans.), Paris, Adrien-Maisonneuve.

Abun-Nasr, J.M., 1971, A History of the Maghrib, Cambridge, Cambridge University Press.

Al-Bakrī, 1965, Al-Masālik wa’l-Mamālik, M.G. De Slane (partial ed. and trans.), Description de l'Afrique septentrionale par Abou Obeïd el-Bekri, Paris, Adrien-Maisonneuve.

Al-Dimashqī, 1866 [reprinted Leipzig 1923], Nukhbat al-dahr, M.A.F. Mehren (ed.), St. Petersburg.

Al-Idrīsī, 1866 [reprinted Amsterdam, Oriental Press, 1969], Description de l'Afrique et de l'Espagne, Dozy, R.P.A., De Goeje, M.J. (partial ed. and trans.), Leiden, E.J. Brill.

Altinine agg-Arias (ed. and trans.), 1974, Traditions historiques des lwellimidan par Digga ăg-Khamməd Ekhya [2nd ed., Niamey, Centre régional de documentation pour la tradition orale, Centre nigérien de recherche en sciences humaines].

Al-ʿUmarī, 1963, Masālik al-abṣār, extracts inalā al-Dīn al-Munajjid, Mamlakat Mālī ‘inda l-jughrāfyiyīn al-Muslimīn (ed. and trans.), Beirut, vol. 1.

Anonymous, 1958, Kitāb al-Istibṣār fī Aʿjāib al-Amṣār, Saʿd Zaghlūl ʿAbd al-amīd (ed.), Alexandrie, Imprimerie de l'université d'Alexandrie.

Barth, H. 1965, Travels and Discoveries in North and Central Africa, London, Cass, 3 vol.

Bernus, E., 1981, Les Touaregs nigériens, Paris, ORSTOM.

Bernus, S., et alii. (eds.), 1986, Le fils et le neveu, Cambridge, Cambridge University Press; Paris, Maison des sciences de l’homme.

Boilley, P., 1999., Les Touaregs Kel Adagh, Paris, Karthala.

Bonte, P., 1986, “La tawshet est-elle un groupe de filiation?”, in S. Bernus et alii. (eds.), Le fils et le neveu, Cambridge: Cambridge University Press and Paris, Maison des sciences de l’homme, p. 237-275.

Casajus, D., 1987, La tente dans la solitude, Cambridge, Cambridge University Press; Paris, Maison des sciences de l'homme.

Chaker, S., 1995, Linguistique berbère, Paris, Louvain, Peeters.

Claudot-Hawad, H., 2001, Éperonner le monde’: Nomadisme, cosmos et politique chez les Touaregs, Aix-en-Provence, Édisud.

Claudot-Hawad, H., 2006, “Marginale, l’étude des marges?”, in H. Claudot-Hawad (ed.), Berbères ou arabes?, Paris, Éditions Non Lieu.

Cooper, J., 2007, Cognitive Dissonance: Fifty Years of a Classic Theory, London/Los Angeles, Sage Publications.

Delafosse, 1913, M., “Traditions historiques et légendaires du Soudan occidental, traduites d'un manuscrit arabe inédit”, Bulletin du Comité de l'Afrique française. Renseignements coloniaux et documents, 8, p. 293-306; 9, p. 325-329; 10, p. 355-368.

Denham, D., Clapperton, H., Oudney, W., 1826 [2nd ed.], The Narrative of Travels and Discoveries in Northern and Central Africa in the years 1822, 1823, and 1824, London, John Murray, 2 vol.

Fauvelle-Aymar, F.-X., Hirsch, B., 2009, “Le ‘pays des Noirs’ selon Léon l’Africain: géographie mentale et logiques cartographiques”, in F. Pouillon et alii. (eds), Léon l’Africain, Paris, Karthala & IISMM.

Festinger, L., 1957, A Theory of Cognitive Dissonance, Evanston, Row, Peterson.

Fisher, H.J., 1990, “The Vanquished Voice: Shaykh Muhammad's Defence: A Case Study from the Western Sahara”, in D. Henige, T.C. McCaskie (eds.), West African Economic and Social History. Studies in Memory of Marion Johnson, Madison, University of Wisconsin, African Studies Program, p. 47-61.

Gast, M., 1968, Alimentation des populations de l'Ahaggar - étude ethnographique, Paris, Arts et Métiers Graphiques.

Ghubăyd ăgg-ălăwjeli, 1975, Histoire des Kəl-Dənnəg, Copenhagen, Akademisk Forlag.

Gironcourt, G.-R. de, 1911, ‘Les inscriptions de la nécropole de Bentia’, Comptes rendus de l'Académie des inscriptions et belles-lettres, Paris, Académie des inscriptions et belles-lettres.

Hodgkin, E., 1987, Social and Political Relations on the Niger Bend in the Seventeenth Century, Ph. D. thesis, University of Birmingham, Centre of West African Studies.

Hopkins, J.F.P., Levtzion, N., 1981, Corpus of Early Arabic Sources for West African History, Cambridge, Cambridge University Press.

Hunwick, J.O., 1985, Sharī˓a in Songhay, Oxford, New York, Oxford University Press for the British Academy.

Hunwick, J.O., 1999, Timbuktu and the Songhay Empire, Leiden, Brill.

Ibnawqal, 1938-1939, Kitāb ūrat al-ar, J.H. Kramers (ed.), Ibn Ḥawqal, Muḥammad ibn ‘Alī al-Naṣībī Abū al-Qāsim. Kitāb Ṣūrat al-arḍ, Leiden, E.J. Brill, 2 vol.

Klute, G., 1999, “Vom Krieg zum Frieden im Norden von Mali”, in H.-P. Hahn, G. Spittler (eds.), Afrika und die Globalisierung, Münster, Hamburg; London, LIT Verlag, p. 455-472.

Last, M., 1980, “Historical Metaphors in the Kano Chronicle”, History in Africa, 7, p. 162-178.

Lecocq, J.S. [B.], 2002, That Desert is Our Country: Tuareg Rebellions and Competing Nationalisms in Contemporary Mali (1946-1996), Ph. D. thesis, University of Amsterdam.

Lecocq, B., 2010, Disputed Desert: Decolonisation, Competing Nationalisms and Tuareg Rebellions in Northern Mali, Leiden, Brill.

Lefebvre, C., 2008, Territoires et frontières. Du Soudan central à la République du Niger, 1800-1964, thèse de doctorat en histoire, université de Paris 1 Panthéon Sorbonne.

Lefebvre, C., Surun, I., 2008, “Explorations et transferts de savoir: deux cartes produites par des Africains au début du xixe siècle”, Mappemonde, 92. URL: http://mappemonde.mgm.fr/num20/articles/art08405.html

Lewicki, L., 1981, “Les origines et l'islamisation de la ville de Tādmakka d'après les sources arabes”, in Le sol, la parole et l'écrit. Mélanges en hommage à Raymond Mauny, Paris, Société française d'histoire d'outre-mer, p. 439-444, 2 vol.

Marty, A., 1975, Histoire de l'Azawagh nigérien de 1899 à 1911, Paris, Mémoire de l'École de hautes études en sciences sociales.

Marty, A., 1985, Crise rurale en milieu nord-Sahélien et recherche coopérative. L'expérience des régions de Gao et Tombouctou, Mali, 1975-1982, thèse de doctorat d'État, Tours, Université François Rabelais.

Moraes Farias, P.F. de, 1985, “Models of the World and Categorical Models: The ‘Enslavable Barbarian’ as a Mobile Label”, in J.R. Willis (ed.), Slaves and Slavery in Muslim Africa, London, Cass, vol. 1, p. 27-46.

Moraes Farias, P.F. de, 1990a, “The Oldest Extant Writing of West Africa: Medieval Epigraphs from Ǝssuk, Saney, and Egef-n-Tăwăqqast (Mali)”, Journal des africanistes, 60(2), p. 65-113.

Moraes Farias, P.F. de, 1990b, “‘Yorùbá Origins’ Revisited by Muslims”, in P.F. de Moraes Farias, K. Barber (eds.), Self-Assertion and Brokerage, Birmingham, University of Birmingham, Centre of West African Studies, p. 109-147.

Moraes Farias, P.F. de, 2003, Arabic Medieval Inscriptions from the Republic of Mali: Epigraphy, Chronicles and Songhay-Tuāreg History, Oxford, Oxford University Press for The British Academy.

Moraes Farias, P.F. de, 2006, “Touareg et Songhay: histoires croisées, historiographies scindées”, in H. Claudot-Hawad (ed.), Berbères ou arabes?, Paris, Éditions Non Lieu, p. 225-262.

Nicolaisen, J., 1963, Ecology and Culture of the Pastoral Tuareg, Copenhagen, The National Museum.

Nicolaisen, J., Nicolaisen, I., 1997, The Pastoral Tuareg, Copenhagen, Rhodos International Science and Art Publishers, London, Thames and Hudson, 2 vol.

Nicolas, F., 1950, Tamesna, Paris, Imprimerie nationale.

Nixon, S., 2009, “Excavating Essouk-Tadmakka (Mali): New Archaeological Investigations of Early Islamic Trans-Saharan Trade”, Azania. Archaeological Research in Africa, 44(2), p. 217-255.

Norris, H.T., 1975, The Tuaregs, Warminster, Aris and Phillips.

Norris, H.T., 1982, The Berbers in Arabic Literature, London, New York, Longman, Beirut, Librairie du Liban.

Norris, H.T., 1986, The Arab Conquest of the Western Sahara, London, Longman, Beirut, Librairie du Liban.

Oxby, C., 1978, Sexual Division and Slavery in a Twareg Community, Ph. D. thesis, University of London, SOAS.

Prasse, K.-G., 1972-1974, Manuel de Grammaire touarègue (tahaggart), Copenhagen, Akademisk Forlag, 3 vol.

Rasmussen, S.J., 1992, “Ritual Specialists, Ambiguity and Power in Tuareg Society”, Man, n. s., 27(1), p. 105-128.

Richer, A., 1924, Les Oulliminden, Paris, Larose.

Swift, J., 1979, The Economics of Traditional Nomadic Pastoralism: the Tuareg of the Adrar n Iforas (Mali), Ph.D. thesis, University of Sussex.

Yāqūt, 1866-1873, Mu‘jam al-buldān, F. Wüstenfeld (ed.), Jacut's geographisches Wörterbuch, aus den Handschriften zu Berlin, St. Petersburg, Paris, London und Oxford, 6 vol., Leipzig, Brockhaus.

Haut de page

Notes

1  In the transcription of Berber and Arabic words, the following special characters are used in this paper: Ă, ă = a very brief “a”; Ǝ, ǝ = a neutral vowel pronounced approximately like the “a” in the English word “China”, the “o” in the English word “police”, or the first “e” in the French word “premier”; š = “sh” as in the English word “ship”, or “ch” as in the French word “chat”. Emphatic consonants have a dot underneath (e.g. ḍ, ḥ, ṛ, ṣ). The digraph “gh” represents a phoneme pronounced approximately like the French r grasseyé. The sign ˒ represents the Arabic hamza. The sign ˓ transliterates the Arabic letter ghayn (ex: ‘Alī, al-˓Umarī, ˓Uqba). A macron is placed over long vowels (e.g. ā). The “u” is pronounced as the “ou” in the French word “lourd”.

2  For recent, and very important, archaeological work at Ǝssuk, see S. Nixon, 2009, p. 217-255. The excavated sequence dates from the mid-first millennium ad to c. 1400.

3  However, on the characteristics and importance of Tifinagh inscriptions, and the relation between them and the Arabic inscriptions, see P.F. de Moraes Farias, 2003, chapter 7, p. 85-89, 92-95.

4  See P. Boilley, 1999, p. 48-49, 57-58.

5  The name “Tuareg” or “Tuāreg” was not created by those to whom it is applied. It derives from the Saharan Arabic form Tǝwāṛǝg—see K.-G. Prasse, 1972-1974, vol. 1, p. 10. Tămašăq is the Berber language spoken by the Tuāreg of the Aḍagh and of parts of the Niger valley. Hence, they are“Kăl-Tămašăq”, a name meaning “those of [the] Tămašăq [language]”. In other regions, in which variants of the same Berber language are spoken, one finds comparable names like “Kǝl-Tǝmajǝq”, “Kǝl-Tamahaq”, etc. In the present paper, the name “Tuāreg” covers all these Berber-speaking populations. However, on the sensitive issues concerning the use of this name, see (among others) P. Boilley, 1999, p. 10, note 6, p. 625; H. Claudot-Hawad, 2001, p. 6; H. Claudot-Hawad, 2006, p. 220-222; and also P.F. de Moraes Farias, 2006, p. 229, note 2. The Tuāreg of the Aḍagh are also known as “Kăl-Aḍagh” (“those of the Aḍagh”).

6  They specialise in Arabic literacy, Islamic rites, educational and medical services, arbitration, praying for rain, and supplying amulets.

7  See P.F. de Moraes Farias, 2003, chapter 7, p. 85.

8  See T. Lewicki, 1981, t. 1, p. 439-444; P.F. de Moraes Farias, 1990a, p. 65-113, especially p. 74, 77-79. For the fourteenth century ad, see al-˓Umarī [c. 1338], 1963; and J.F.P. Hopkins, N. Levtzion, 1981, p. 274. On the eighteenth-century evidence, see H.T. Norris, 1975, p. 110-12, quoting the Fatḥ al-Shakūr [1799-1800 ad] of al-Bartīlī (also known as al-Barritaylī); and P.F. de Moraes Farias, 2003, p. cxliii.

9 S. Chaker, 1995, p. 146. Other references in P.F. de Moraes Farias, 2003, p. cxliii.

10  Ibnawqal [c. 988 AD], 1938-1939, vol. I, p. 92; Yāqūt [completed 1224-1229 ad], 1866-1870, vol. I, p. 400; vol. II, p. 938; Anonymous, 1958, p. 223; Al-Dimashqī, 1923, p. 239; Al-Bakrī [1068 ad], 1965, p. 181 [text], p. 339 [trans.]. See also J.F.P. Hopkins, N. Levtzion, 1981, p. 46, 85, 149, 168, 210; and J.O. Hunwick, 1985, p. 6, n. 2.

11  See P.F. De Moraes Farias, 2003, inscription 104, p. 87-88; p. cxli, cxc-cxci, ccxxxi-ccxxxiii.

12  See G.-R. de Gironcourt, 1911, p. 198-206.

13 ʿAbd al-Ramān al-Sa˓dī, 1964, p. 4 [text]; p. 6–7 [trans.]; J.O. Hunwick, 1999, p. 6.

14 Al-Idrīsī [1154 ad], 1969, p. 10, 14, 19, 45 [text]; p. 11-12, 17, 23, 53 [trans.].

15  See M. Delafosse, 1913, p. 5-6.

16  See H. Barth, 1965, III, p. 464.

17  See H. Barth, 1965, III, p. 491.

18  For fascinating discussions of this map and its context, see C. Lefebvre, 2008, vol. 1, p. 106-110, vol. 2, Annexes, p. 78-82 and C. Lefebvre, I. Surun, 2008, p. 17-21.

19  This Arabic word can mean either “sea” or “a large river”. Here it clearly refers to a river.

20  Cf. D. Denham, H. Clapperton, W. Oudney, 1826, vol. 2, p. 302, 304-305, 309-310, 313-314, 331, and the map facing p. 330. See also, on the amalgamation of the Niger with the Nile, P.F. De Moraes Farias, 1985, vol. 1, p. 27-46.

21  See J. Cooper, 2007; L. Festinger, 1957.

22  See the perceptive analysis of Leo’s text by F.-X. Fauvelle-Aymar, B. Hirsch, 2009, p. 83-102, especially p. 85, 89, 95-96.

23 M. Last, 1980, p. 162-178.

24  See P.F. De Moraes Farias, 1990b, p. 109-147.

25  See E. Hodgkin, 1987, p. 431-436.

26  On the Iwəlləmmədăn, see Ghubăyd ăgg-Ălăwjeli, 1975; Altinine agg-Arias, 1974; A. Richer, 1924; F. Nicolas, 1950.

27 On the social organisation and pastoralist economy of the Tuāreg, see J. Nicolaisen, I. Nicolaisen, 1997; J. Nicolaisen, 1963; S.J. Rasmussen, 1992, p. 105-128; E. Bernus, 1981; S. Bernus et alii., 1986; D. Casajus, 1987; A. Marty, 1985; A. Marty, 1975; J. Swift, 1979; C. Oxby, 1978; M. Gast, 1968.

28  See P. Bonte's introduction to S. Bernus et alii., 1986 and in the same book P. Bonte, 1986, p. 237-275.

29  J.S. [B.] Lecocq, 2002, p. 16-23.

30  See ʿAbd al-Ramān al-Sa˓dī, 1964, 320 [text], 483 [trans.] and H.T. Norris, 1975, p. 109.

31 H. Barth, 1965, vol. III, p. 761.

32  See Sīdī Mawlāy Muḥammad al-Hādī, Naṣīḥat al-˒umma, MS 94 of the collection formerly kept in the Assemblée Nationale, Niamey, but now transferred to the Institut de Recherches en Sciences Humaines in Niamey; quoted by H.T. Norris, 1975, p. 22.

33  See P.F. de Moraes Farias, 2003, inscription 106, p. 89.

34  The quotations are from a four-page anonymous and untitled text, two copies of which are kept in the Institut de France (Fonds De Gironcourt MS VI-2410-133). It was collected by G.-R. de Gironcourt in 1912 from a Kăl-Ǝssuk chief whose name he (and other French authors) recorded as “Mohammed Ouguinatt” at the Fombalgo pool west of the Niger, approximately 50 kilometres north-west of Bentyia (fig. 1). H.T. Norris, 1975, p. 22-23 provides a translation of this text. ˓Uqba al-Mustajāb means “˓Uqba whose prayers are answered”; see H.T. Norris, 1982, p. 55-56, 239.

35  See J.M. Abun-Nasr, 1971, p. 68-69.

36  See H.T. Norris, 1975, p. 22, 25; P.F. de Moraes Farias, 2003, p. cxliv.

37  See H.T. Norris, 1975, p. 24-25; P.F. de Moraes Farias, 2003, p. cxxxvi-cxxxvii.

38 H.T. Norris, 1975, p. 17, 25.

39  See H.T. Norris, 1986, p. 126-127, 185, 192, 220, who comments on this text and translates it. See also the excellent discussion of the same text by H.J. Fisher, 1990, p. 47-61.

40  On these uprisings see P. Boilley, 1999; G. Klute, 1999, p. 455-472; B. Lecocq, 2010.

41 Dawlat is a loanword from Arabic meaning “state”, “dynasty”, “power”, “empire”.

42  See P.F. de Moraes Farias, 2003, inscriptions 154a.1, 154a.2, 154b.1, 154b.2, p. 130-133; p. cxl-cxli, cxlv-cxlvi, ccxvi.

Haut de page

Table des illustrations

Titre Figure 1: Ǝssuk / Tadmăkkăt in its regional context
Crédits P.F. de Moraes Farias, 2003, Maps and Site Plans 1
URL http://afriques.revues.org/docannexe/image/896/img-1.png
Fichier image/png, 255k
Titre Figure 2: Ruins and necropolises at Ǝssuk / Tadmăkkăt
Crédits P.F. de Moraes Farias, 2003, Maps and Site Plans 9
URL http://afriques.revues.org/docannexe/image/896/img-2.png
Fichier image/png, 326k
Titre Figure 3: Map showing the course of the Niger given to Clapperton at Muḥammad Bello’s Court
Crédits E.W. Bovill, 1966, Mission to the Niger, IV, The Bornu Mission, 1822-1825, Published for the Hakluyt Society, Cambridge, Cambridge University Press, Second Series, vol. CXXX, Inserted between pages 698 and 699
URL http://afriques.revues.org/docannexe/image/896/img-3.png
Fichier image/png, 458k
Haut de page

Pour citer cet article

Référence électronique

Paulo Fernando de Moraes Farias, « Local Landscapes and Constructions of World Space: Medieval  Inscriptions, Cognitive Dissonance, and the Course of the Niger », Afriques [En ligne], 02 | 2010, mis en ligne le 25 février 2011, consulté le 23 octobre 2017. URL : http://afriques.revues.org/896 ; DOI : 10.4000/afriques.896

Haut de page

Auteur

Paulo Fernando de Moraes Farias

Honorary Senior Research Fellow, Centre of West African Studies, University of Birmingham, United Kingdom

Articles du même auteur

Haut de page
  • Logo Institut des mondes africains
  • Logo DOAJ – Directory of Open Access Journals
  • Les cahiers de Revues.org