Navigation – Plan du site

AccueilThématiques04Déconstruire les modèlesArguing sovereignty in Songhay

Déconstruire les modèles

Arguing sovereignty in Songhay

Plaider la souveraineté en pays songhaï
Bruce S. Hall


La littérature récente en archéologie, histoire et anthropologie à propos du développement de la complexité sociale et politique en Afrique lance un défi à nos modèles anciens de la formation de l’État, et notamment des empires médiévaux sahéliens, comme celui du Songhaï. Comme nous le savons, il y de multiples voies vers la complexité qui ne mènent pas forcément à la formation d’un État ; et il y avait une distribution hétérarchique du pouvoir dans plusieurs formations politiques africaines. Néanmoins, l’historiographie des États précoloniaux en Afrique de l'Ouest sahélienne et celle du rôle de l’islam dans ces formations politiques restent attachées à un modèle de l’État tiré des géographies et chroniques arabes. On y insiste encore sur le pouvoir militaire et sur un rapport ambivalent entre islam et des formes autochtones d’autorité. Dans cet article, nous proposons une réinterprétation de l’exercice de la souveraineté et de sa rhétorique dans l’empire songhaï qui met en avant les modalités par lesquelles les dirigeants pouvaient prétendre à une autorité islamique ou la mettre en cause. Les dirigeants songhaïs revendiquaient une autorité religieuse qui allait au-delà du pouvoir de coercition. Plutôt qu’un rapport ambivalent entre le domaine religieux musulman et le pouvoir séculier, l’autorité islamique constituait le principal fondement de la prétention par les souverains songhaïs à un pouvoir étendu.

Haut de page

Texte intégral


  • 1  This is how his name is written in the Taʾrīkh al-fattāsh (M. Kâti, 1964, p. 100). He is also know (...)
  • 2  There are earlier examples of this strategy in the case of the Almoravids and the kingdom of Takru (...)

1In 1493, a usurper, whom we know as Askia al-ḥājj Muhammad Ture, captured the reins of power in imperial Songhay by defeating an army led by Sunni Baru,1 the rightful heir to the throne of the great warrior king Sunni Ali Beer, who had died the year before under suspicious circumstances. After deposing Sunni Baru, Askia Muhammad established a new dynasty of ‘Askias’ who would rule imperial Songhay with different degrees of skill and success until the state’s destruction at the hands of an invading Moroccan army in 1591. In both the seventeenth-century Arabic chronicles of Songhay history written in Timbuktu, and in much of the modern historiography of the Songhay Empire, Askia Muhammad is represented as a novel political figure because of the extent to which he sought to legitimize his rule on explicitly Islamic credentials.2 Askia Muhammad became a patron of Muslim scholars and holy men in commercial towns such as Timbuktu and he undertook the pilgrimage to Mecca in 1496-7 soon after establishing himself in power. Upon his return to West Africa, he endeavored to put his rule on a sound Islamic footing by winning the approval of a prominent North African scholar named Muḥammad b. ʿAbd al-Karīm al-Maghīlī (d. 1504), who visited the Songhay capital of Gao in 1498. The written text which records Askia Muhammad’s questions and al-Maghīlī’s answers offers us a unique window onto medieval Sahelian statecraft. Above all, it reveals an argument made by Askia Muhammad that his regime would respect the letter of Islamic law. This adherence to the law was contrasted with the behavior of his predecessor Sunni Ali Beer who, despite his claim to be a Muslim, was described as both unjust in his treatment of Muslims and insincere in his profession of Islam.

  • 3  E. Tyan, 2011.
  • 4  J. Hunwick, 1985, p. 81.
  • 5  J. Hunwick, 1996, p. 176.

2In this article, I will offer a reading of Askia Muhammad’s questions to al-Maghīlī which highlights the arguments that the Songhay ruler made about sovereignty and his own authority. Askia Muhammad claimed to be a just Muslim ruler and as such, sought recognition by Muslims residing far outside the geographic core of the Songhay state. By asserting this authority, Askia Muhammad invoked a complex of Islamic political concepts about the threat of disorder and the necessity of submission or allegiance (Ar. bayʿa) required of all Muslims to a lawful amīr.3 In Mālikī jurisprudence, allegiance can theoretically be withheld from a ruler only when he acts so improperly that he can only be defined as an unbeliever. Al-Maghīlī reported that there was a ḥadīth in which the Prophet Muhammad had said: “Whoever withholds his hand from [the bonds of] obedience shall meet God on the Day of Judgement having no argument in his defence; and whoever dies with no oath of homage (bayʿa) upon his neck dies a jāhilī death.”4 So when Askia Muhammad denigrated his predecessor Sunni Ali Beer as a non-believer, he was making a necessary argument if his authority was to be recognized as legitimate. What this implies is that Askia Muhammad sought support not just from the Muslim scholars in Timbuktu and elsewhere who had opposed Sunni Ali Beer, what John Hunwick calls the ‘religious estate’,5 but also from a far wider group of people who had backed Sunni Ali Beer on what were probably very similar terms of necessary allegiance to a just Muslim ruler. I will argue that when Askia Muhammad’s arguments are understood in this way, we are forced to rethink the nature of power and authority in the imperial Songhay state, and to accord a greater importance to the symbolic authority of Islam in the projection of its power, under the ‘Sunni’ dynasty as much as that of the Askias who followed.

3One of the main arguments made by Askia Muhammad in his questions to al-Maghīlī is that his predecessor Sunni Ali Beer was unworthy of the authority he claimed, precisely because he acted in ways that made his Muslim identity suspect:

  • 6  J. Hunwick, 1985, p. 69.

Sunni Ali’s father was the sultan of its people and his mother was from the land of Far, and they are unbelieving people who worship idols of trees and stones; they make sacrifices to them and ask their needs at them. If good befalls them they claim that it is those idols who gave it to them and if it does not befall them they believe that those idols withheld it from them. Thus they do not undertake military expeditions until they have consulted them and when they arrive back from a journey they go to them and stay beside them. These idols have custodians who serve them and act as intermediaries between the people and them. Among these people are soothsayers and magicians to whom they likewise have recourse.6

  • 7 Ibid., p. 69-70.

4By such arguments Askia Muhammad claimed that Sunni Ali was stamped by these non-Muslim practices and beliefs, a fact which explained the injustice of his rule: “Sunni Ali, from childhood to manhood used to frequent them a great deal to the extent that he grew up among them and became stamped with their pattern of polytheism and with their customs.”7

  • 8 Ibid., p. 70.

5While Askia Muhammad admits that Sunni Ali Beer claimed to be a Muslim, he argues that his actions pointed to his origins among non-believers: “[T]hough he had grown up from childhood to manhood among his maternal uncles and had become stamped with their pattern of behavior, it was his habit to make a lip profession of the two shahadas and other similar words of the Muslims without knowing their significance.”8 Despite his claims to being a Muslim, he prevented those at his court from carrying out the five daily prayers or fasting during Ramadan. Askia Muhammad connected Sunni Ali Beer’s impiety with the injustice of his rule:

  • 9 Ibid., p. 71.

Another practice of his was making lawful the [shedding of the] blood and [seizure of the property] of Muslims. He put to death scholars, jurists and pietists, women, infants and others, mutilating some by ablation of the genitals or by cutting off the nose and hands. He seized property, enslaved [free] women and sold free men to an extent that cannot be measured. His wickedness in the land through these [acts] and the like is well known; nothing like it has ever been heard of in Islam.9

6These claims about Sunni Ali Beer’s insincerity and ignorance as a Muslim were not important just because they demonstrated that Sunni Ali Beer relied on non-Muslim forms of political legitimacy (even if he did). More than this, they were meant to demonstrate that allegiance to Sunni Ali Beer had been misplaced because he had not been a just Muslim ruler. Askia Muhammad was therefore right to overthrow the Sunni dynasty.

  • 10  See for example, J. Hunwick, 1966, p. 296-317; S.M. Cissoko, 1996; N. Levtzion, 1978, p. 338; L. K (...)
  • 11  For this distinction, see N. Levtzion, 1973, p. 184.

7In the older scholarship on medieval statecraft in Sahelian West Africa, the ‘magician king’ Sunni Ali Beer and the ‘Muslim pilgrim’ Askia al-ḥājj Muhammad represented ideal types in the long struggle imagined between Islam and non-Muslim ‘traditional’ religious and political traditions. The advent of Askia Muhammad’s rule was seen as marking a shift in the role of Islam in the political history of the region’s major states, from a religion of commerce that coexisted with more local religious practices at the court of important states, to an explicit Islamic idiom of statecraft.10 Nehemia Levtzion described the contrast and conflict between the purportedly ‘syncretic’ Islam of Sunni Ali Beer and the more ‘pious’ Askia Muhammad, as setting the pattern of the subsequent history of Sahelian Africa. Muslim reformers, especially those of the jihadist movements launched in the eighteenth and nineteenth centuries by Fulbe scholars across West Africa, consistently made the claim that Sahelian Muslim rulers were in fact, by their failure to respect Islamic norms, non-Muslims.11

  • 12  P.F. de Moraes Farias, 2003, p. xlvii-xlix, lxix.
  • 13 J. Hunwick, 1999, p. 91.
  • 14 Ibid., p. 91.
  • 15 Ibid., p. 103.

8The problem with much of the scholarship on Songhay, and other pre-colonial Sahelian states, is that it tends to both accept uncritically the story of Songhay statecraft largely as it was reported in the two principal seventeenth-century chronicles, the ‘Taʾrīkh al-sūdān’ and the ‘Taʾrīkh al-fattāsh’.12 The Arabic chronicles of Songhay history tell us that Sunni Ali Beer, after coming to power in 1464, launched a series of successful military campaigns which extended the territory controlled by the Songhay state westward from its core territory centered at Gao and Kukiya in the eastern Niger Bend. When he captured Timbuktu in 1468, many of the merchants and Muslim scholars of the town fled and took refuge in Walata, 430 kilometers to the west. Some of those who remained were killed, and as a consequence, Sunni Ali Beer is represented in the texts written in Timbuktu as a “great oppressor and notorious evil-doer.”13 We are told that he mocked Islam and that his true loyalty was to non-Muslim traditional religious rites. “He tyrannized the scholars and holy men, killing them, insulting them, and humiliating them.”14 Once Askia Muhammad took power, he sought to redress the mistreatment of the Muslim scholars whom he made into his supporters and advisors. As the ‘Taʾrīkh al-sūdān’ puts it, “the most felicitous and well-guided one became the ruler on that day, and was amīr al-muʾminīn and khalīfa of the Muslims […]. Through him, God Most High alleviated the Muslims’ distress, and eased their tribulation.”15

  • 16  P.F. de Moraes Farias, 2003, p. lxxiii.
  • 17  This language was elevated even further in the nineteenth-century forgery of parts of the Ta’rīkh (...)
  • 18  C. Blum and H. Fisher, 1993; J. Hunwick, 1996; P.F. de Moraes Farias, 2003; J. Boulègue 2007, p. 5 (...)

9The seventeenth-century chronicles were written to celebrate the history of Songhay as a continuously existing, coherent state over centuries, which reached its zenith in the reign of the ‘pious’ Askia Muhammad.16 The chronicles elevate the language of the just Muslim ruler – raised in status to ‘khalīfa’, or political successor to the Prophet Muhammad17 – in their description of Askia Muhammad’s authority, suggesting that it was a completely new basis of political power, independent of the tradition of Songhay statecraft inherited from Sunni Ali Beer. Instead of interpreting these claims as new, or bound up with the particularity of Askia Muhammad’s efforts to win legitimacy for his rule, I argue that Muslim religious authority was an essential feature of medieval Sahelian statecraft more generally. The idea that Sunni Ali Beer marks the apogee of a form of kingship bound up with local religious or ritual specialists, hostile to Muslims and distinct from Islam, is a construction of the seventeenth-century Arabic chronicles written in Timbuktu. It is not consistent with the interplay of power and authority which was central to the state tradition in the Sahel, and which more recent scholarship has begun to highlight.18

Rethinking state formation and heterarchy

10Over the last several decades, there has been significant rethinking of some of the major paradigms of state formation in the fields of history, anthropology, and archaeology of Africa. In Susan McIntosh’s recent effort to synthesize this literature, and to give Africa a bigger place in the theoretical models of the development of social and political complexity more generally, she has written that:

  • 19 S.K. McIntosh, 1999a, p. 4.

[Africa] provides a rich corpus of material relevant to an understanding of societies in which central authority, often of a ritual nature, is paired with a power structure that is diffuse, segmentary, and heterarchical, as well as societies in which considerable complexity is achieved through horizontal differentiation and consensus-based decision-making. The distribution of power among several corporate entities (e.g. lineages, secret societies, cults, age grades) can be regarded as a strategy that has successfully resisted in a variety of ways the consolidation of power by individuals.19

  • 20 S.K. McIntosh, 1999a, p. 8; A. Stahl, 1999, p. 35-55; A. Stahl, 2001, p. 12-15.

11McIntosh and others have argued that evolutionary models of social and political development fail to account for the ‘pathways to complexity’ encountered in Africa. In order to avoid the teleology which assumes a path to ever greater centralized hierarchy and control, we are encouraged to think about ways in which complexity was achieved without concomitant political centralization or social hierarchy.20

  • 21  M. Fortes, E.E. Evans-Pritchard, 1940.
  • 22  A. Southall, 1970, p. 230.
  • 23 Ibid., p. 231.
  • 24  A. Southall, 1988, p. 52.

12In one of the earliest iterations of political anthropology in Africa, the oft-cited African Political Systems edited by M. Fortes and E. E. Evans-Pritchard and published in 1940, a fundamental distinction was made between acephelous societies whose political structure is organized into competing lineage segments, and states which possess a centralized administrative organization.21 When Aiden Southall published his classic study of the Alur in 1956, he revealed the existence of political systems that fit neither the typology of acephelous societies nor centralized states. Using the example of the Alur, located north of Lake Albert straddling the border of modern Uganda and the Democratic Republic of Congo, Southall wrote that “a great deal of Alur expansion and domination over other people appears to have taken place almost without the use of force, and it is the voluntary element in the submission of subject groups which remains an interesting problem.”22 He concluded that “[t]o the extent that other groups submitted to incorporation into the Alur system without the application of force to induce them to do so, this must be attributed to sociological and moral rather than material factors.”23 In effect, peripheral groups attached themselves to the Alur state because it offered them access to valuable forms of ritual or religious capital, which were associated with shared ideas of sacral kingship and rainmaking. Southall coined the term ‘segmentary state’ to describe this kind of polity “in which the spheres of ritual suzerainty and political sovereignty do not coincide. The former extends widely towards a flexible, changing periphery. The latter is confined to the central, core domain.”24

13Fortes and Evans-Pritchard had highlighted the composite nature of many African states, and the political necessity of balancing central authority with regional autonomy:

  • 25  M. Fortes, E.E. Evans-Pritchard, 1940, p. 12.

Local chiefs represent the central authority in relation to their districts, but they also represent the people under them in relation to the central authority. Councilors and ritual functionaries represent the community’s interest in the preservation of law and custom and in the observance of the ritual measures deemed necessary for its well-being. The voice of such functionaries and delegates is effective in the conduct of government on account of the general principle that power and authority are distributed. The king’s power and authority are composite. Their various components are lodged in different offices. Without the cooperation of those who hold these offices it is extremely difficult, if not impossible, for the king to obtain his revenue, assert his judicial and legislative supremacy, or retain his secular and ritual prestige.25

  • 26  J.H. Elliot, 1992, p. 56; H.G. Koenigsberger, 1989, p. 135-153; P.S. Stern, 2008, p. 257-261.
  • 27  D.T. Niane, 1995, p. 75-77.

14We might borrow language from the literature on early-modern Europe to say that medieval Sahelian states can be usefully described as ‘composite monarchies’. In this kind of state, the center is relatively weak, relying extensively on other nodes of authority, usually granting extensive autonomy to regional elites and employing a generous system of patronage.26 In the famous Sundiata epic of the founding of the Mali Empire, there is a scene at the climax of the narrative in which the twelve leaders of the chiefdoms (‘kafus’) united under Sundiata’s military command symbolically put their spears into the ground in front of him to announce their submission. This symbolic investiture of Sundiata as king is then followed by the return of the spears to the leaders, which reveals that imperial Mali will be ruled as a composite monarchy of these constituent chiefdoms.27

  • 28  M. Fortes, E.E. Evans-Pritchard, 1940, p. 16.
  • 29  N. Kodesh, 2010, p. 7; D. Shoenbrun, 1999, p. 139-141.
  • 30  I. Kopytoff, 1999, p. 94-95; S.K. McIntosh, 1999a, p. 2.

15The older literature on African statecraft often focused on the sacral nature of political authority. Fortes and Evans-Pritchard argued that in many African states, rulers exercised a largely symbolic authority over their subjects: “An African ruler is not to his people merely a person who can enforce his will on them. He is the axis of their political relations, the symbol of their unity and exclusiveness, and the embodiment of their essential values […]. His credentials are mystical and are derived from antiquity.”28 Southall’s ‘segmentary state’ built on this idea, except that it distinguished between sacral authority and effective power. More recent literature has focused on the heterarchy of power in the development of political complexity in Africa, highlighting the multiple sources of authority that extend well beyond rulers. Very often, political rulers or military specialists vie with people who possess types of knowledge that allow them to claim an ability to affect the relationship between the natural, social, and supernatural domains.29 Complex political communities are constituted by the dynamic interplay of their different constituencies, but not in a linear direction towards more hierarchical and centralized political forms. It is perhaps just as likely that state systems based on kingship will dissolve into systems of regional chiefship or village headship, as the other way around. The political system in any society is the product of the configuration and struggles of different power holders to define how they want to live, within the constraints that they find themselves.30

  • 31  T. Earle, 1997, p. 13.
  • 32  J. Goody, 1971, p. 72.

16If many African polities were characterized more by extensive ritual authority than coercive power, we need to also account for the materiality of that authority. There is a debate in the literature on state formation about the extent to which even ritual authority is based in material relations. Timothy Earle, for example, argues that “although there are many routes to social complexity and many blind alleys, the creation of politically expansive and centralized institutions requires that the available sources of power be rooted in economic control.”31 The problem in much of Africa, according to Jack Goody, was that it was not possible for states to base their power on the extraction of staple finance from subject populations because of the relatively low agricultural yields in much of the continent. Levels of agricultural production did not depend on state protection or organization (i.e. irrigation), and broadly, there was little difference in the agricultural output of producers living under state authority, and those living outside of it. For Goody, the power of many African states was primarily predatory, based on what he called the ‘means of destruction’, rather than on rents or taxes from economically productive subjects.32

  • 33  Others include C. Raynaut et al., 1997, p. 63; H. Fisher, 1973, p. 366; R. Mauny, 1975, p. 505-506
  • 34  R.M. Netting, 1972, p. 233; quoted in S.K. McIntosh, 1999a, p. 14.

17At a general level, the argument of Goody, among others, assumes that Sahelian states based themselves on coercion and an ability to extract economically valuable resources from subject populations and neighboring peoples subject to their raids.33 From the Arabic written record of local chronicles, and from oral epics, it is clear that military power was important to Sudanic statecraft. However, these models overestimate the extent to which power was exclusively the result of coercion, and undervalue or minimize the importance of the religious or ritual authority exercised by these states. What is missing in these models of statecraft based on coercion is the way in which “common symbols, shared cosmology, and the overarching unity of fears and hopes made visible in ritual” could be mobilized to constitute a wider “moral community” that transcended locality or lineage to constitute a state.34

  • 35  P. de Maret, 1999, p. 161.
  • 36 Quoted in P. de Maret, 1999, p. 161.

18The example of the Luba, in south-central Africa, is illustrative of how ideas about statecraft have changed in the literature on some parts of Africa where a pre-colonial state tradition was once important. The older models of a centralized Luba state have now been replaced by a conception of royal Luba power that was more of an ideological construct than an empirical reality. According to Pierre de Maret, Luba lacked the capacity to maintain a standing army and as such, the power of the Luba kingdom was more symbolic than military: “It had thus to rely on ideology and to share its prestige with the heads of largely autonomous regional communities which sought legitimacy and identity through the manipulation of various myths, emblems, and a rich array of regalia.”35 According to Mary Nooter Roberts, “Luba power rested not with a single dynastic line of kings or with a single ‘center’, but with a constellation of chieftaincies, officeholders, societies, and sodalities that validated claims to power in relation to what we suggest to be a largely mythical center. The paradox is explicit: there was no real center, yet belief in one allowed cultural integration of an entire region.”36

19This model of extensive symbolic authority has been used by David Edwards to describe the ancient Nilotic state of Meroe. In that region:

  • 37  D. Edwards, 1998, p. 190.

[s]ubsistence resources had a generally low productive capacity with low surpluses. Some direct royal control may have been exercised over them through taxation/tribute backed by the ultimate sanction of force, and probably also by the ritual power of the monarchy. In common with many later African kingdoms, however, in the absence of elaborate administrative organizations, more direct controls were probably limited to a relatively small core area.37

  • 38 Ibid., p. 191.

20According to Edwards, elites outside of the core area of the state depended on central power for forms of wealth that could not be produced locally: “Royal power is likely to have been far more closely linked to, and dependent on, the control of procurement of non-utilitarian resources, especially through trade... [T]here is every reason to see [imports] as ‘prestige-goods’, important for defining royal and elite status and for the public display of power and prestige.”38

  • 39  T. Insoll, 2003, p. 212-213.
  • 40  This is the argument of J. Spaulding, 2007, p. 81-86.
  • 41  D. Edwards, 1998, p. 188-189.
  • 42  D. Wengrow, 2006.
  • 43  S. Nixon, 2009, p. 251.

21The ability of the medieval states of the Sahel to control – or benefit from – long-distance trade has long been recognized as an important feature of their emergence and success, although current archaeological interpretation suggests that state formation preceded the development of regular trans-Saharan trade.39 Edwards argues that in Meroe, most long-distance trade in prestige items took the form of ‘embassy trade’ between states, rather than by market mechanisms, which only developed much later.40 If this is true, the power of the Meroetic state outside of its core area rested on its ability to distribute imported prestige goods to regional elites, who in turn depended on their connection with Meroe to mark their status in their home regions. Archeological evidence of imported prestige items such as jewelry, metal goods, lamps, faience, and wine and oil residues have been found in the orbit of Meroe, suggesting that redistribution to regional elites was an important mechanism for the projection of Meroe’s symbolic authority.41 David Wengrow has made the argument that a very similar dynamic of monopolization of long-distance trade – or wealth finance – by regional elites in pre-dynastic Upper Egypt was essential to the process of state formation there.42 Recent archeological work at Saharan trade sites indicates that high-value trade goods were sometimes sent south across the Sahara to West Africa and presumably distributed amongst elites. For example, Sam Nixon found a Qingbai Chinese porcelain pot and remnants of silk textiles at Essouk in northern Mali, an important staging site on the Saharan trade to the Niger Bend.43

  • 44  P.F. de Moraes Farias, 2003, p. 3-9.
  • 45  R. McIntosh, 1993; cited in S.K. McIntosh, 1999b, p. 76.
  • 46  S.K. McIntosh, 1999b, p. 76.

22An important shift in statecraft in the Sahel occurred when the region was articulated into international Muslim trade networks. Established relationships with Muslim traders provided African rulers with access to wealth finance from parts of the world otherwise inaccessible. Early evidence of this in the Niger Bend is the marble used in tombstones that were imported from Muslim Spain to Gao in the twelfth-century.44 In her analysis of the development of social complexity at Jenne-Jeno on the Middle Niger, Susan McIntosh argues that in this especially productive ecological niche characterized by high population density, occupational and craft specialization, and long-distance trade, there was no concomitant development of political centralization before the introduction of Islam. The settlement pattern of the area at its height of prosperity between 800-1000 CE was not that of a single town, but a series of clusters in close proximity to each other. To Rod McIntosh, this suggests the desire of heterogeneous groups of people to be near the center, but not of it; to function as part of an urban entity without being subsumed by it.45 Susan McIntosh argues that the “maintenance of settlement boundaries seems to reflect a degree of autonomy and resistance to the center, the spatial manifestations of which disappear with the great settlement reorganization that took place between AD 1200 and AD 1400, coincident with the conversion of local leaders to Islam and the expansion of the Empire of Mali. In this period, the settlement pattern took on the more familiar characteristics of a city-state.”46 How precisely conversion of leaders to Islam caused a reconfiguration of Sahelian statecraft and settlement is the next question to be addressed.

Islam and state in the Sahel

  • 47  J. Hunwick, 1996. p. 180.
  • 48  N. Levtzion, 1973, p. 186.
  • 49  N. Levtzion, 2000, p. 63.
  • 50  T. Insoll, 1996, p. 45-48.

23One of the crucial issues in the historiography of the Songhay Empire is the extent of the autonomy and authority exercised by what John Hunwick calls the Muslim ‘religious estate’.47 How much independence did the Muslim religious estate have from those who controlled the state apparatus? This question – posed more broadly – has been at the center of much of the historiography of the medieval state tradition in Sahelian West Africa. Nehemia Levtzion, for example, argued that Muslim traders were welcomed by many medieval polities because of the importance of trans-Saharan trade. The kings of Ancient Ghana “let the Muslims practice their religion without interference.” Both the Muslims and the king “preferred to live apart. The Muslims were interested in retaining an autonomous Muslim community, whereas the king tried to restrict Islamic foreign influence over his subjects.”48 According to Levtzion, an especially clear illustration of this relationship lies in the spatial distance that purportedly separated Muslim merchants from the royal court in many medieval capitals.49 However, archeological evidence suggests that the dual settlements described in the medieval Arabic geographical sources were not as distinct as Levtzion suggested. According to Timothy Insoll’s work at Gao for example, the Muslim town of Gao-Saney was not founded by Muslims, nor was it exclusively Muslim at the time it was described in Arabic sources.50

  • 51  N. Levtzion, 1973, p. 188.

24In Levtzion’s typology of Islamization, spatial separation between Muslims and non-Muslims is followed by a phase in which “Islam was transmitted by the Muslim traders who had hailed from the Sahel to the local people […] when closer relations developed between the hospitable chiefs and their Muslim guests.”51 Yet even in situations in which Islamization had become more widespread, Levtzion characterized the relationship between Islam and state as a straddling act:

  • 52 Ibid., p. 190.

Islamized kings were placed in a delicate position between an influential Muslim minority (who lived close to the center, monopolized the trade and had extensive outside relations) and the majority of their pagan subjects. They were thus obliged to hold a middle position between Islam and the traditional religion; they were neither real Muslims nor complete pagans. From this middle position some chiefs or dynasties might turn towards the true Islam, while others might fall back to regain closer relations with the traditional religion.52

  • 53  C. Harrison, 1988; D. Robinson, 2000.

25Levtzion’s notion of ‘true Islam’ is anachronistic and reminds us of an older tradition based in colonial scholarship which denigrated the form of Islam practiced in sub-Saharan Africa as ‘Islam noir’, distinct from purportedly more authentic versions practiced in places such as the Middle East.53 It also fails to account for the fact that many people from the Sahel were considered full Muslims by their contemporaries elsewhere in the Muslim world. Certainly at the height of Mali’s imperial power in the fourteenth century, Mali’s kings were recognized as Muslims, going on the pilgrimage to Mecca and supporting Muslim scholarship in their realm. However, Levtzion argued that:

  • 54  N. Levtzion, 1973, p. 198.

a closer review of the contemporary evidence reveals the survival of pre-Islamic customs which sustained elements of the traditional religion. The outcome was not syncretism, nor the molding of Islamic and traditional elements, but rather a dualism, in which the two systems existed side by side. These two cultural systems did not exist in abstraction, but were represented by social groups within the empire. Whereas Islam gained ground in the urban centers, in the trading community and among the ruling estate, it had little impact on the rural communities, which remained closely attached to the traditional religion.54

  • 55  M. Abitbol, 1979, p. 55; J. Boulègue, 2007, p. 520-521.

26The distinction between Islamic religious authority associated with long-distance trade and a more ‘traditional’ authority based on locality and lineage has been a common way of understanding statecraft in the Sahel. The challenge of any sovereign was to transcend these poles of authority, a possibility which seems blocked in Levtzion’s model of dualism.55

  • 56  N. McHugh, 1994; R.S. O’Fahey, 1980.

27John Hunwick has attempted to frame the problem of contested authority in a different way, focusing on a distinction between what he calls secular power and (Muslim) religious authority. Using the case of the Songhay Empire, Hunwick presents a typology which draws on the literature describing the privileges that were granted to holy men in Nilotic Sudan:56

  • 57  J. Hunwick, 1996, p. 184.

A clearer picture of relations between the ruling estate and the religious estate in Songhay may be gained through the accounts of gifts of such things as slaves, land, cattle, cloth and cash, as well as exemption from state obligations and taxes. A further sign of the respect which at least some of the askiyas had for the religious estate is to be seen in the power of intercession that some of the holy men could exercise with the rulers, and the recognition by rulers of the quality of hurma that holy men and their houses, as well as mosques, might possess.57

  • 58  However, he does not find evidence for this (Ibid., p. 189-190).

28Hunwick argues that even under Sunni Ali Beer, the Muslim religious estate in the Songhay Empire received many gifts and grants of privileges from the state, presumably in exchange for religious services offered to rulers in prayers, production of amulets, etc.58 He emphasizes the autonomy of the religious estate – especially in Timbuktu – from state power in the Songhay Empire under the Askia dynasty.

  • 59  M. Gomez, 1990, p. 5.

29Michael Gomez has argued that Hunwick and others have overestimated Timbuktu’s autonomy because they have relied on the writings of scholars from Timbuktu as their main sources for imperial Songhay history. According to Gomez, the Songhay state had a much firmer political control over Timbuktu and the scholars and merchants who lived there than most historians allow. Likewise, Gomez rejects the argument made by Hunwick that the Muslim religious estate wielded significant influence on the royal court at Gao.59 For Gomez, the Songhay state maintained a highly organized administration and military control over the western Niger Bend, including Timbuktu, because it relied on the revenues generated by agriculture and trade in this region.

  • 60  P.F. de Moraes Farias, 2003, p. clxx-clxxvi; D. Lange, 1991, p. 255-256; D. Lange, 1994, p. 299-30 (...)

30The problem with these depictions of relations between Muslims and state officials is that they assume that Islam was a distinct social and political force that was fundamentally foreign to the cultures and societies of the Sahel. The implication of this approach is that when Askia Muhammad made his explicitly Islamic appeal to justify his rule after deposing Sunni Ali Beer’s rightful heir, he invoked a cultural form that was foreign, even if it had many local adherents. P.F. de Moraes Farias’s work on the epigraphic tradition in the Niger Bend, which dates back significantly earlier than the locally-produced written tradition, tells a different story about the relationship between Islam and local politico-religious culture. The Timbuktu chronicles were seventeenth-century political interventions meant to offer a usable past of continuous Songhay political organization and dynastic coherence to the rulers of the Songhay successor state established by the Moroccan invasion in 1591. But the chronicles made such a fundamental distinction between the evil, not-fully-Muslim, Sunni Ali Beer and the good Muslim Askia Muhammad that they appear to describe a shift in the importance of Islam for statecraft in the Sahel. The archaeological and epigraphic evidence makes clear that there were many Muslims at early Songhay sites such as Bentia / Kukiyia, which the ‘Taʾrīkh al-sūdān’ identifies as a very ancient Songhay site, possibly even a former Songhay capital before it was moved to Gao, although this is a subject of much historiographical debate.60

31The presence of Muslims in political contexts that the chronicles assign exclusively to local non-Muslim dynamic traditions such as Kukiya suggests an important corrective to the way that we understand the role of Islam in the medieval Sahel. Instead of thinking of Islam and local non-Muslim religious practice as necessarily distinct and opposed, it is better to conceive of multiple and overlapping spheres of religious and political authority that have long existed together. As Moraes Farias argues:

  • 61 Ibid., p. lxxxvi.

[…] We must give short shrift to the still widespread historical stereotype that Niger-Bend Islamic culture was essentially a betrayal, and an absolute rejection, of truly (by definition ‘pagan’) Soŋoy culture. This stereotype, nurtured by a romantic opposition of ‘the alien’ to ‘the authentic’, must be replaced by the acknowledgement that different forms of Niger-Bend cultural praxis have constituted themselves in history by drawing on cultural repertoires of various origins. Of course, each of these forms of praxis has placed different (often vastly different) emphasis on each of the repertoires they combined or borrowed from. But it would be mistaken for us to forget that cultures are echo chambers resonant with other cultures.61

  • 62 Ibid., p. cxxi.

32In the case of the early Songhay political formation, we must understand that not only were there significant numbers of Muslims present at any given time, but that these people were an important component of the wider society in terms of the specialized knowledge and access to certain prestige trade goods. As such, Islam held a kind of authority, even to many people who were non-Muslims. As Moraes Farias suggests, “[i]t makes little sense to postulate that the emphasis on symbols from the external world was simply a façade, good for impressing foreigners but meaningless to most insiders. On the contrary, the capacity to wield those symbols, and to manipulate access to foreign commodities and foreign information, had become a significant resource in internal politics.”62 The remainder of this article will focus on one way in which Songhay rulers argued for Islamic authority over disparate peoples, even if they were unable to enforce their claims with the exercise of coercive power.

Arguing over Authority

  • 63  J. Hunwick, 1985, p. 79.
  • 64 Ibid., p. 81.

33In one of his questions to al-Maghīlī, Askia Muhammad is reported to have asked the following: “[S]ome of the Muslims to our east and west, having heard of me, asked if they could enter into allegiance to me. Should I grant them this or should I confine myself to ruling [only] the land which God caused us to inherit from Sunni Ali?”63 In this question, the Arabic term ‘bayʿa’, for allegiance, is invoked. What Askia Muhammad was asking was whether he could claim an authority as a just Muslim ruler that went beyond the scope of his effective power. This is an argument for precisely the kind of symbolic authority and voluntary adherence that we have been discussing. Al-Maghīlī responded by explaining that there are three types of lands: an ‘ungoverned land’ (‘bilād al-sāʾiba’) whose people have no amīr, a land whose people have a lawful Muslim ruler, and a land in which the purportedly Muslim amīr imposes unlawful taxes and commits oppressive acts more generally. Askia Muhammad is told to encourage Muslims residing in the ‘ungoverned land’ to enter into allegiance to him: “Allow such people to pay homage to you and enter into obedience to you. If they refuse that, then compel them to do so with whatever means are at your disposal, for it is not lawful for a group of Muslims to be without a ruler.”64

  • 65  E. Burke III, 1980.
  • 66  H. Touati, 1996, p. 102.
  • 67  I use the form ‘sāʾiba’ because that is how the term appears in the West African context. The spel (...)
  • 68  H. Touati, 1996, p. 114-115; H. Touati, 1993, p. 98.

34The idea of an ‘ungoverned land’ is important in the theoretical framework of Islamic statecraft because it represents the absence of Islamic order. The term ‘bilād al-sāʾiba’ has long been used in North Africa to refer to those areas beyond the control of states, often in mountainous or desert areas. In Morocco especially, the distinction between the lands controlled by the power of the central state (the ‘makhzan’), and those lands where anarchy and disobedience prevail (the ‘sība’) was an important vulgate used by the French colonial regime to understand and administer Moroccan society.65 However, it cannot be simply dismissed for having become part of the colonial lexicon for Muslim societies.66 Houari Touati has argued that in the Mālikī legal literature of early modern North Africa and the Sahara, there are three principal meanings associated with the ‘bilād al-sāʾiba’,67 all of which imply an opposition to order as it is defined by Islam. One sense is to follow customary legal practice rather than Islamic law (ʿurf versus fiqh). It can also be the equivalent of the idea of ignorance before the advent to Islam (jāhiliyya versus Islam). The third usage is for contexts in which there is no centralized political authority over a region at all, but instead what we might think of as ‘tribal’ councils (jamāʿa versus sulṭān).68 It is the last meaning that is familiar to students of North African history, and it is the sense of the term used in Askia Muhammad’s question.

35The most obvious solution to the problem of ‘sāʾiba’ was the imposition of lawful Muslim rule. But there is also a very important abstract, or symbolic, sense for this bundle of ideas. Where people live under local systems of custom and rule, they define themselves as outside of Islamic political order. However, Muslims living in such contexts should still do everything in their power to make an allegiance to a just Muslim ruler, even if the reach of his power does not touch them. As such, the symbolic authority of the just ruler extends much further than his temporal power. The ultimate objective is the spread of Islamic order everywhere, thereby eliminating the ‘bilād al-sāʾiba’ altogether.

  • 69  J. Hunwick, 1985, p. 79-80.
  • 70 Ibid., p. 130-131.

36Following the larger question about the allegiance of far-away Muslims to Askia Muhammad was a related query about another Sahelian Muslim state whose ruler is described as unjust, and “who oppresses people and wherever he finds goods belonging to the Muslims seizes them […]. He intercepts caravans coming into his territory, forces them to halt, and searches their loads, values what is in them and takes from them what he claims to be zakāt. But he does not give those entitled to zakāt anything [from it].”69 Despite these acts, the ruler claimed to be a Muslim and argued that his actions were sanctioned by Islamic law. Askia Muhammad’s question is whether this ruler should be considered an oppressor (ẓālim) or an unbeliever (kāfir), and what actions could be taken against him? The issue of interest for our purposes is not the actions recommended against this ruler, but the fact that Askia Muhammad claimed an authority over the Muslims residing in this purportedly Muslim state. It is the same argument about allegiance that we have seen above, and it turns on whether the neighboring state should be considered to be in a condition of ‘sāʾiba’, outside of Islamic political order. Al-Maghīli responded that Askia Muhammad had an obligation to intervene in this situation, and to bring justice and his authority to the Muslims living there.70

  • 71  Hunwick argues that he did intervene on several occasions (Ibid., p. 27).
  • 72 Ibid., p. 26.
  • 73  P.F. de Moraes Farias, 2003, p. 215.

37The importance of al-Maghīlī’s replies rests less on the authorization given to Askia Muhammad to intervene militarily in the affairs of neighboring Muslim states,71 than on his claims to a wider symbolic authority as a just Muslim ruler. Hunwick has argued that at the time that Askia Muhammad took power in what was a poly-ethnic Songhay empire, “the authority of Islam appears to have been that which had the widest appeal among the diverse elements.”72 It seems clear that there was a constituency for Askia Muhammad’s claims to be a just Muslim ruler who deserved widespread allegiance among Sahelian Muslims living beyond the reach of Songhay coercive power. What this suggests is that the characterization of Sunni Ali Beer as paying ‘lip-service’ to Islam should not be taken at face value. Sunni Ali Beer was a Muslim. Even if we cannot know the extent or precise character of the Islamic practice of those among whom Sunni Ali grew up – they too were almost certainly Muslims. To be a Muslim did not mean a rejection of local culture, but it did add another pole of authority for people in the Sahel. The Muslim-inflected tradition of statecraft that emerged in the Sahel was oriented to the wider Muslim world outside of West Africa, but also implicated in processes by which Sahelian space and moral community were redefined to account for the presence of Muslims.73


  • 74  J. Hunwick, 1999, p. xli
  • 75  A. Haour, 2007, p. 50.
  • 76  P.F. de Moraes Farias, 2003, p. lxxiii.

38According to John Hunwick, by the end of Askia Muhammad’s reign in 1529, Songhay was a “loosely knit empire […], which controlled the Middle Niger very firmly, and somewhat more tenuously a large swathe of territory stretching from the river Senegal in the west to the Aïr massif in the east, from the borders of Borgu in the south to the salt pans of Taghāza in the north.” In most of the tributary territories, former rulers continued to exercise political control and “simply paid tribute to the ruler at Gao. In the core territories some local rulers were left in political control, while a parallel Songhay official looked after Songhay interests and ensured loyalty.”74 Hunwick and others have described a variety of bureaucratic offices in the empire which suggests a fairly intrusive state apparatus, and an ability to project coercive Songhay power by military force. Anne Haour has remarked that imperial Songhay, as it is described in this historiography, is an “improbably large” polity.75 It is difficult to know how much credence we should put in Songhay’s ability to project power over great distances because our sources for these geographical characterizations are seventeenth-century chronicles written to celebrate the history of Songhay as a continuously existing, coherent state over centuries.76 What I have proposed in this article is that if we measure the scale and breadth of Songhay in terms of its claims to authority as a Muslim state, it was at least as large as Hunwick described it above. I have offered a reading of Askia Muhammad’s arguments about the extent of his authority as primarily religious and symbolic in nature. The implication of this argument is that we cannot interpret Sunni Ali Beer’s Islamic identity as incidental; but instead must see that power in Songhay was contested within a shared Islamic framework of legitimate authority.

  • 77  U.dan Fodio, 1978, chap. 2; J. Hunwick, 1985, p. 129.
  • 78  A.A. Batran, 2001, p. 134.

39It is important to emphasize that although Askia Muhammad made a claim about his own authority, we do not know how well his argument was received by those for whom it was intended. However, it is very clear that the text which recorded Askia Muhammad’s questions to al-Maghīlī became very influential over time for other political leaders seeking to exercise their own authority. The question of allegiance (bayʿa) and the ‘bilād al-sāʾiba’ are important in the writings of Usman dan Fodio at the beginning of the nineteenth century in Hausaland.77 In the context of his jihad against the Hausa states, Usman dan Fodio argued that any Muslim living under rulers deemed by him not to be Muslims would be treated as enemies, just as any other non-Muslims in war. The same issue appears in the legal literature in nineteenth-century Mauritania, except in this case from the perspective of the Muslims living in areas controlled by rulers who claim the right to allegiance because of their authority as just Muslim rulers. For some merchants especially, it was preferable in practice not to swear allegiance to a just Muslim ruler because in that way payment of zakāt, a required tithe on all Muslims, could be avoided. In an undated legal opinion written by al-Shaykh Sīdi al-Mukhtār al-Ṣaghīr al-Kuntī (d. 1847),78 the issue of allegiance of Muslim merchants from Mauritania to the Fulbe state led by Amadou Lobbo on the Middle Niger is addressed. The questioner asked that Sīdi al-Mukhtār al-Ṣaghīr explain:

  • 79  B.S. Hall, 2011, p. 99.

the state (ḥāl) of the Fulbe in general, and specifically in the case of Amadu Lobbo and his people, and all the rest of his amīrs […]. Is it required of everyone to swear allegiance to them if they are not already under allegiance to another power? […]. Or, are [the Fulbe] like all of the rest of the vanquished (al-mutaghallibīn) who are to be dealt with, as the poet said: “Their house remains their house, and their land remains their land?” In this case, one does not pay zakāt to them voluntarily […].79

40In the question, the letter-writer uses the term ‘bilād al-sāʾiba’ to characterize the Middle Niger region controlled by Amadou Lobbo’s state. In Sīdi al-Mukhtār al-Ṣaghīr’s response, he argued that the questioner should swear allegiance to Amadou Lobbo, whose claims to be a just Muslim ruler should be accepted as valid.

41These nineteenth-century cases highlight something very important about imperial Songhay under Askia Muhammad. His claims to authority as a just Muslim ruler were interventions that required consent by many of those whom he sought to bring under his control. Just as the Songhay state under Askia Muhammad engaged in military conflicts with neighboring polities, it also sought to project a religious authority. We must assume that like its military power, Songhay’s religious authority was highly contested by others who had reason to make similar claims. What needs to be understood is that these kinds of symbolic or religious arguments had become a crucial dimension of statecraft in the Sahel. They would remain so until European colonial conquest at the end of the nineteenth century.

Haut de page


Abitbol, M., 1979, Tombouctou et les Arma : De la conquête du Soudan nigérien en 1591 à l’hégémonie de l’Empire Peul du Macina en 1833, Paris, Maisonneuve & Larose.

Batran, A.A., 2001, The Qadiryya Brotherhood in West Africa and the Western Sahara : The Life and Times of Shaykh al-Mukhtar al-Kunti, 1729-1811, Rabat, Université Mohammed V-Souissi.

Blum, C., Fisher H., 1993, “Love for Three Oranges, or, The Askiya’s Dilemma: The Askiya, al-Maghili and Timbuktu, c. 1500 A.D.”, Journal of African History, 34 (1), p. 65-91.

Boulègue, J., 2007, “L’affichage du religieux dans les conflits politiques en Sénégambie du xve au xviiie siècle”, in C. Deslauriers et D. Juhé-Beaulaton (éds), Afrique, terre d’histoire, au cœur de la recherche avec Jean-Pierre Chrétien, Paris, Karthala.

Burke III, E., 1980, “The Sociology of Islam: The French Tradition”, in M. Kerr (éd.), Islamic Studies: A Tradition and Its Problems, Malibu, Undena Publications, p. 73-88.

Cissoko, S.M., 1996, Tombouctou et l’Empire Songhay: Épanouissement du Soudan nigérien aux xve-xvie siècles, Paris, L’Harmattan.

dan Fodio, U., 1978, Bayān wujūb al-hijra ʿalā ʾl-ʿibād, F.H. El Masri (éd. et trad.), Khartoum, Khartoum University Press.

Earle, T., 1997, How Chiefs Come to Power: The Political Economy in Prehistory, Stanford, Stanford University Press.

Edwards, D., 1998, “Meroe and the Sudanic Kingdoms”, Journal of African History, 39 (2), p. 175-193.

Elliot, J.H., 1992, “A Europe of Composite Monarchies”, Past and Present, 137, p. 48-71.

Fisher, H., 1973, “‘He Swalloweth the Ground with Fierceness and Rage’”: The Horse in the Central Sudan II. Its Use”, Journal of African History, 14 (3), p. 355-379.

Fortes, M., Evans-Pritchard, E.E., 1940, “Introduction”, in M. Fortes et E.E. Evans-Pritchard (éds), African Political Systems, Londres, Oxford University Press.

Gomez, M., 1990, “Timbuktu under Imperial Songhay: A Reconsideration of Autonomy”, Journal of African History, 31 (1), p. 5-24.

Goody, J., 1971, Technology, Tradition and the State in Africa, Cambridge, Cambridge University Press.

Hall, B.S., 2011, A History of Race in Muslim West Africa, 1600-1960, New York, Cambridge University Press.

Harrison, C., 1988, France and Islam in West Africa, 1860-1960, Cambridge, Cambridge University Press.

Haour, A., 2007, Rulers, Warriors, Traders, Clerics: The Central Sahel and the North Sea, 800-1500, Oxford, Oxford University Press.

Hunwick, J., 1966, “Religion and State in the Songhay Empire, 1464-1591”, in I.M. Lewis (éd.), Islam in Tropical Africa, Londres, International African Institute, p. 296-317.

Hunwick, J., 1985, Sharīʻa in Songhay: the replies of al-Maghīlī to the questions of Askia al-Ḥājj Muḥammad, Oxford, Oxford University Press.

Hunwick, J., 1994, “Gao and the Almoravids Revisited: Ethnicity, Political Change and the Limits of Interpretation”, Journal of African History, 35 (2), p. 251-273.

Hunwick, J., 1996, “Secular Power and Religious Authority in Muslim Society: The Case of Songhay”, Journal of African History, 37 (2), p. 175-194.

Hunwick, J., 1999, Timbuktu and the Songhay Empire: Al-Sa`dī’s Taʾrīkh al-sūdān Down to 1613 and Other Contemporary Documents, Leiden, E.J. Brill.

Insoll, T., 1996, Islam, archaeology, and history. Gao Region (Mali) ca. AD 900-1250, Cambridge Monographs in African Archaeology 39, Oxford, Tempus Reparatum.

Insoll, T., 2003, The Archaeology of Islam in Sub-Saharan Africa, Cambridge, Cambridge University Press.

Kaba, L., 1984, “The Pen, the Sword, and the Crown: Islam and Revolution in Songhay Reconsidered, 1464-1493”, Journal of African History, 25 (3), p. 241-256.

Kâti, Mahmoud Kâti ben El-Hâdj El-Motaouakkel, 1964, Tarikh El-Fettach ou Chronique du chercheur pour servir à l’histoire des villes, des armées et des principaux personnages du Tekrour, O. Houdas et M. Delafosse (éds et trans.), Paris, A. Maisonneuve.

Kodesh, N., 2010, Beyond the Royal Gaze: Clanship and Public Healing in Buganda, Charlottesville, University of Virginia Press.

Koenigsberger, H.G., 1989, “Composite States, Representative Institutions, and the American Revolution”, Historical Research, 62, p. 135-153.

Kopytoff, I., 1999, “Permutations in patrimonialism and populism: the Aghem chiefdoms of Western Cameroon”, in S.K. McIntosh (éd.), Beyond Chiefdoms: Pathways to Complexity in Africa, Cambridge, Cambridge University Press, p. 88-96.

Lange, D., 1991, “Les Rois de Gao-Sané et les Almoravides”, Journal of African History, 32 (2), p. 251-275.

Lange, D., 1994, “From Mande to Songhay: Towards a Political and Ethnic History of Medieval Gao”, Journal of African History, 35 (2), p. 275-301.

Levtzion, N., 1971, “A Seventeenth-Century Chronicle by Ibn al-Mukhtār: A Critical Study of Taʾrīkh al-Fattāsh”, Bulletin of the School of Oriental and African Studies, 34 (3), p. 571-593.

Levtzion, N., 1973, Ancient Ghana and Mali, Londres, Methuen.

Levtzion, N., 1978, “Islam in West African Politics: Accommodation and Tension between the ‘ulamā’ and the Political Authorities”, Cahiers d’études africaines, 18 (71), p. 333-345.

Levtzion, N., 2000, “Islam in the Bilad al-Sudan to 1800”, in N. Levtzion et R. Pouwels (éds), History of Islam in Africa, Athènes, Ohio University Press, p. 63-91.

Maret, P. de, “The power of symbols and the symbols of power through time: probing the Luba past”, in S.K. McIntosh (éd.), Beyond Chiefdoms: Pathways to Complexity in Africa, Cambridge, Cambridge University Press, p. 151-165.

Mauny, R., 1975, Tableau géographique de l’Ouest Africain au Moyen âge d’après les sources écrites, la tradition et l’archéologie, Amsterdam, Swets & Zeitlinger.

McHugh, N., 1994, Holymen of the Blue Nile: The Making of an Arab-Islamic Community in the Nilotic Sudan, 1500-1850, Evanston, Northwestern University Press.

McIntosh, S.K., 1999a, “Pathways to complexity: an African perspective”, in S.K.McIntosh (éd.), Beyond Chiefdoms: Pathways to Complexity in Africa, Cambridge, Cambridge University Press, p. 1-30.

McIntosh, S.K., 1999b, “Modeling political organization in large-scale settlement clusters: a case study from the Inland Niger Delta”, in S.K.McIntosh (éd.), Beyond Chiefdoms: Pathways to Complexity in Africa, Cambridge, Cambridge University Press, p. 66-79.

McIntosh, R., 1993, “The pulse theory: Genesis and accommodation of specialization in the Middle Niger”, Journal of African History, 34 (2), p. 181-220.

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

Netting, R. McC., 1972, “Sacred Power and Centralization: Aspects of Political Adaption in Africa”, in B.Spooner (éd.), Population Growth: Anthropological Implications, Cambridge, MIT Press, p. 219-244.

Niane, D.T., 1995, Sundiata: An Epic of Old Mali, London, Longman.

Nixon, S., 2009, “Excavating Essouk-Tadmakka (Mali): new archaeological investigations of early Islamic trans-Saharan trade”, Azania, 44 (2), p. 217-255.

O’Fahey, R.S., 1980, State and Society in Dar Far, Cambridge, Cambridge University Press.

Raynaut, C. et al., 1997, Societies and Nature in the Sahel, New York, Routledge.

Robinson, D., 2000, Paths of Accommodation: Muslim Societies and French Colonial Authorities in Senegal and Mauritania, 1880-1920, Athènes, Ohio University Press.

Saad, E., 1983, Social History of Timbuktu: The Role of Muslim Scholars and Notables, 1400-1900, Cambridge, Cambridge University Press.

Shoenbrun, D., 1999, “The (in)visible roots of Bunyoro-Kitara and Buganda in the Lakes region: AD 800-1300”, in S.K. McIntosh (éd.), Beyond Chiefdoms: Pathways to Complexity in Africa, Cambridge, Cambridge University Press, p. 136-150.

Southall, A., 1970, Alur Society: A Study in Processes and Types of Domination, Nairobi, Oxford University Press.

Southall, A., 1988, “The Segmentary State in Africa and Asia”, Comparative Studies in Society and History, 30 (1), p. 463-491.

Spaulding, J., 2007, The Heroic Age in Sinnār, Trenton, Red Sea Press.

Stahl, A., 1999, “Perceiving Variability in Time and Space: Evolutionary Mapping of African Societies”, in S.K. McIntosh (éd.), Beyond Chiefdoms: Pathways to Complexity in Africa, Cambridge, Cambridge University Press, p. 35-55.

Stahl, A., 2001, Making History in Banda: Anthropological Visions of Africa’s Past, Cambridge, Cambridge University Press.

Stern, P.S., 2008, “‘A Polity of Civil & Military Power’: Political Thought and the Late Seventeenth‐Century Foundations of the East India Company‐State”, Journal of British Studies, 47 (2), p. 253-283.

Touati, H., 1993, “La loi et l’Écriture. Fiqh, ʿurf et société au Maghreb d’après les Ajwiba d’Ibn Nāṣir (m. 1085/1674)”, Annales Islamologiques, 27, p. 93-108.

Touati, H., 1996, “Le prince et la bête. Enquête sur une métaphore pastorale”, Studia Islamica, 83, p. 101-119.

Tyan, E., 2011, “Bayʿa”, in Encyclopaedia of Islam, 2e édition, Brill Online.

Wengrow, D., 2006, The Archaeology of Early Egypt: Social Transformations in North-East Africa, 10,000 to 2650 BC, Cambridge, Cambridge University Press.

Haut de page


1  This is how his name is written in the Taʾrīkh al-fattāsh (M. Kâti, 1964, p. 100). He is also known as Abū Bakr Dao in the Taʾrīkh al-sūdān (J. Hunwick, 1999, p. 102).

2  There are earlier examples of this strategy in the case of the Almoravids and the kingdom of Takrur (N. Levtzion, 1973, p. 184).

3  E. Tyan, 2011.

4  J. Hunwick, 1985, p. 81.

5  J. Hunwick, 1996, p. 176.

6  J. Hunwick, 1985, p. 69.

7 Ibid., p. 69-70.

8 Ibid., p. 70.

9 Ibid., p. 71.

10  See for example, J. Hunwick, 1966, p. 296-317; S.M. Cissoko, 1996; N. Levtzion, 1978, p. 338; L. Kaba, 1984, p. 242; E. Saad, 1983, p. 42.

11  For this distinction, see N. Levtzion, 1973, p. 184.

12  P.F. de Moraes Farias, 2003, p. xlvii-xlix, lxix.

13 J. Hunwick, 1999, p. 91.

14 Ibid., p. 91.

15 Ibid., p. 103.

16  P.F. de Moraes Farias, 2003, p. lxxiii.

17  This language was elevated even further in the nineteenth-century forgery of parts of the Ta’rīkh al-fattāsh (N. Levtzion, 1971).

18  C. Blum and H. Fisher, 1993; J. Hunwick, 1996; P.F. de Moraes Farias, 2003; J. Boulègue 2007, p. 520-521.

19 S.K. McIntosh, 1999a, p. 4.

20 S.K. McIntosh, 1999a, p. 8; A. Stahl, 1999, p. 35-55; A. Stahl, 2001, p. 12-15.

21  M. Fortes, E.E. Evans-Pritchard, 1940.

22  A. Southall, 1970, p. 230.

23 Ibid., p. 231.

24  A. Southall, 1988, p. 52.

25  M. Fortes, E.E. Evans-Pritchard, 1940, p. 12.

26  J.H. Elliot, 1992, p. 56; H.G. Koenigsberger, 1989, p. 135-153; P.S. Stern, 2008, p. 257-261.

27  D.T. Niane, 1995, p. 75-77.

28  M. Fortes, E.E. Evans-Pritchard, 1940, p. 16.

29  N. Kodesh, 2010, p. 7; D. Shoenbrun, 1999, p. 139-141.

30  I. Kopytoff, 1999, p. 94-95; S.K. McIntosh, 1999a, p. 2.

31  T. Earle, 1997, p. 13.

32  J. Goody, 1971, p. 72.

33  Others include C. Raynaut et al., 1997, p. 63; H. Fisher, 1973, p. 366; R. Mauny, 1975, p. 505-506.

34  R.M. Netting, 1972, p. 233; quoted in S.K. McIntosh, 1999a, p. 14.

35  P. de Maret, 1999, p. 161.

36 Quoted in P. de Maret, 1999, p. 161.

37  D. Edwards, 1998, p. 190.

38 Ibid., p. 191.

39  T. Insoll, 2003, p. 212-213.

40  This is the argument of J. Spaulding, 2007, p. 81-86.

41  D. Edwards, 1998, p. 188-189.

42  D. Wengrow, 2006.

43  S. Nixon, 2009, p. 251.

44  P.F. de Moraes Farias, 2003, p. 3-9.

45  R. McIntosh, 1993; cited in S.K. McIntosh, 1999b, p. 76.

46  S.K. McIntosh, 1999b, p. 76.

47  J. Hunwick, 1996. p. 180.

48  N. Levtzion, 1973, p. 186.

49  N. Levtzion, 2000, p. 63.

50  T. Insoll, 1996, p. 45-48.

51  N. Levtzion, 1973, p. 188.

52 Ibid., p. 190.

53  C. Harrison, 1988; D. Robinson, 2000.

54  N. Levtzion, 1973, p. 198.

55  M. Abitbol, 1979, p. 55; J. Boulègue, 2007, p. 520-521.

56  N. McHugh, 1994; R.S. O’Fahey, 1980.

57  J. Hunwick, 1996, p. 184.

58  However, he does not find evidence for this (Ibid., p. 189-190).

59  M. Gomez, 1990, p. 5.

60  P.F. de Moraes Farias, 2003, p. clxx-clxxvi; D. Lange, 1991, p. 255-256; D. Lange, 1994, p. 299-300; J. Hunwick, 1994, p. 257.

61 Ibid., p. lxxxvi.

62 Ibid., p. cxxi.

63  J. Hunwick, 1985, p. 79.

64 Ibid., p. 81.

65  E. Burke III, 1980.

66  H. Touati, 1996, p. 102.

67  I use the form ‘sāʾiba’ because that is how the term appears in the West African context. The spelling ‘sība’ is a Moroccan dialectical variation.

68  H. Touati, 1996, p. 114-115; H. Touati, 1993, p. 98.

69  J. Hunwick, 1985, p. 79-80.

70 Ibid., p. 130-131.

71  Hunwick argues that he did intervene on several occasions (Ibid., p. 27).

72 Ibid., p. 26.

73  P.F. de Moraes Farias, 2003, p. 215.

74  J. Hunwick, 1999, p. xli

75  A. Haour, 2007, p. 50.

76  P.F. de Moraes Farias, 2003, p. lxxiii.

77  U.dan Fodio, 1978, chap. 2; J. Hunwick, 1985, p. 129.

78  A.A. Batran, 2001, p. 134.

79  B.S. Hall, 2011, p. 99.

Haut de page

Pour citer cet article

Référence électronique

Bruce S. Hall, « Arguing sovereignty in Songhay »Afriques [En ligne], 04 | 2013, mis en ligne le 29 mai 2013, consulté le 05 décembre 2023. URL : ; DOI :

Haut de page


Bruce S. Hall

Assistant Professor, Department of History, Duke University

Haut de page

Droits d’auteur


Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search