In range, content and aim, the epic-sized volume edited by Martin Sterry and David Mattingly displays a kinship with the proliferating ‘companions’ that now festoon our bookshelves. In effect, it is a handbook of the history, culture and economy of the ancient Sahara. The collection is the third in a projected four-volume series of the Trans-SAHARA project whose aim is to demonstrate much earlier and more extensive systems of Saharan connectivity in antiquity. The sixteen separate studies cover all geographic zones from the Nile to the Atlantic, from the Maghrib in the north to the Sudan in the south, from late prehistoric to early Islamic developments. As such, it extends and develops the picture in the detailed survey of the Sahara in antiquity offered by Mario Liverani at the turn of the millennium (Studi Storici 47 (2006), 1003–57). The project's aim is to establish that the Sahara was more intensely populated, developed and connected in the protohistoric times than it has been subsequently—although the reader must be cautioned that in the contributors’ periodisation of history the ‘protohistoric’ extends as late as 650 C.E. (4). Covering the vast sand and rock sea of the Sahara, the first half of the volume concentrates on the early history of the oasis centres. The latter half offers a series of special case studies.
Given the range and detail of the data covered—principally but not only archaeological—it is very useful that the editors pause at the end of the section perhaps most remote from the readers of this Journal (that on the Saharan oases) to offer a discussion of the general consequences of the complex of evidence for the overall argument (ch. 8), a happy tactic that they repeat for the volume as a whole (ch. 19). Furthermore, where literary evidence and the usual stratigraphical markers are so absent, the authors laudably publish extensive tables of revised 14C dates (twenty pages of them for the southern Saharan oases alone: 279–97) that provide a fix on the temporal span of the settlements concerned. For the most part, the argument focuses on the urban functions of desert oases. To what extent the development of the Saharan oases can be described as ‘urban’ is something that the reader will have to judge. The maximum extent that any oasis reached in the challenging environment of the desert was about ten hectares, admitted by the authors to be a rather small hamlet in Mediterranean terms; most were much smaller in size. The largest Saharan oases would not even make it into the smallest of Roman town sizes in north Africa on the list proffered by Andrew Wilson (table 10.1). Even so, the difficulty is less one of the absolute size of individual oasis communities than it is of the enormous distances isolating them and the lack of anything describable as an urban network of which they were part. Modest evidence is offered to confirm the pre-Roman emergence of the oasis settlements; the real period of their efflorescence, however, is Roman—and it is the interactions with a big imperial state that seems to be the critical element in their history. The oasis settlements grew in size, wealth and complexity in their dialogic interaction with large states of the Mediterranean and the Nile River valley. The end of this process in Mediterranean antiquity was the imbrication of the oases with the large ‘tribal confederations’ of the desert's northern edge, whose emergence seems to be part of the same dynamic. As the writers emphasise, the Saharan ‘polities’ cannot be territorial states in ‘a conventional sense’ (703). Rather, it seems that these ‘polities’ (whatever else they were) consisted of networks of communication across truly vast expanses of desert, networks in which they served as nodes in extraordinarily thin filaments of contact. As the editors say, in this situation the idea of territorial states is probably misplaced. It was, rather, the control of networks that was important (703)—where family and kin groups, religious associations and other personal connections seem to have been more important. The extent to which these communities were ‘states’ (a list is provided: fig. 19.1) in any sense that the term is normally used is seriously open to question.
The core exemplary case of a Saharan polity is that of the Garamantes (53–111, 696–700, 711–15), in many ways the firmest documented of all the desert communities. Their settlements appear like super-thin strings of pearls, literally hundreds of tiny sites strung in three separate bands over hundreds of kilometres along wadi depressions in the desert more than a thousand kilometres south of the Mediterranean. Typologically, they remind one of the linear dir, an ecology supporting ribbon-like strands of settlements along the Saharan periphery of the Maghrib (e.g. J. Despois, L'Afrique du Nord, 3rd edn (1964), 106–10). The Garamantes are important for the argument here since they are frequently used as a standard of measurement by which the potential of other oasis communities is assessed. How far their case can be extended to the rest of the Sahara, especially when their very exceptional size and complexity (715) is considered is questionable. Like many other Saharan communities of the time, theirs was a dimorphic community that had both sedentary and pastoral nomadic elements. It also seems to be one of the rare cases—the large oases of the arid lands west of the Nile, like those of Dakhla, would probably be another—where the balance of power enabled the oasis dwellers to engage with mobile pastoralists on more equal terms (ch. 3). In most oasis environments in the Sahara in historical times, where most oasis communities were smaller and more isolated, the balance of power—bluntly, of violent force—favoured armed pastoralists who not infrequently reduced the sedentarist oasis dwellers to various types of servitude and dependence. In the particular case of the oases in the western desert of Egypt and those of the desert hinterland of Tripolitania, it hardly seems accidental that the most profound urbanisation (i.e. the extension of intensified settlements) took place at the time of Roman imperial hegemony, as is made clear in Anna Boozer's study of the Egyptian oases and by Mattingly and Sterry's own study of the Garamantian oases of the Fazzan (chs 3 and 4). During the long peace enforced by Roman rule, these settlements were able to leverage the power of a nearby imperial state to their advantage in ways that the more remote and far more isolated oasis communities in the central and western Sahara would not have been able to do.
Rather than the particulars—many of which are new and useful—perhaps the most telling benefit of the volume its recentring of our assumed perspective, a Mediterranean one, with a different focal point from which the familiar histories of the Nile valley or the Maghrib can be seen in a different light. Although the new evidence does shed light on many aspects of the history of Sahara and its immediate environs in antiquity, the rationale behind the decision to focus on ‘state formation’ escapes me. There is so very little evidence of anything like states—even on the very loose ‘non-European’ definitions of state that are proffered (27–30)—that it seems that another point of focus, such as the nature of a fluid connectivity, would have been more useful. I must say that I am as open as anyone to new ideas of what might constitute a state, but the models here still seem to be ones strongly influenced by modern western theorists. The evidence that is cumulated and presented in vivid detail paints a long-term picture of a different Saharan environment in which huge distances, mobility and the linking of isolated nodes of more intensive sedentary settlement produced a species of pattern of compression and expansion which might have benefitted from a more systematic application of Horden and Purcell's ideas on the Mediterranean. If powerful religious associations (ch. 18) and trading families can be important connecting entities, then why not found an analysis on these more appropriate organising entities? One also has to wonder why an African thinker like Ibn Khaldūn, who offered different patterns of state formation based on kin and community solidarity and cohesion, ‘asabiyya, might not have received more consideration.
There remains a contrarian idea that the early history of this region might have been operating under its own strong peculiarities, as pointed out by the reviewer some time ago (MHR 18 (2003), 93–125). Perhaps a cautionary reading for readers of this Journal, even if the essay covers neither their period nor their subject matter, is the contribution by Judith Scheele (ch. 18). It is a bracing antidote to the sometimes exaggerated claims, surfeit of high theory and the truly exiguous data bearing on the problem. One might consider, as one example, ch. 11 (‘Numidian State Formation in the Tunisian High Tell’), where the chasm between the many pages devoted to theory and the actual evidence is particularly alarming. Scheele's study of what has been documented of power relations in the Sahara in more recent times, where she has reasonably good evidence to sustain the conclusions—better, in any event, than most of the archaeologists of antiquity—suggests serious cautions. The growth of centres, emergence of hierarchical social orders and development of trade contacts and institutions like slavery do not necessarily imply the parallel development of other structures—such as a state, for example. She coyly suggests that her cautions might not be true of other times, like those of the Garamantes, but it seems more likely that she has right on her side. In short, it is frustrating to understand why the editors and most of the contributors are so heavily wedded to the concept of ‘the state’ and allied ideas when so much of their evidence seems to point in different directions, more flexible ones that would better fit the long-term history of the Sahara. The emergence of dimorphic polities on the peripheries of the Syrian Desert (e.g. M. Sommer, MedAnt 9.2 (2006), 428–9, referring to Michael Rowton's seminal work), for example, presents well-documented ways in which pastoral nomadic groups and agricultural populations related to each other. They suggest useful parallels for the means by which more complex communities might have emerged out of combinations of the mobile and the settled. As is noted (ch. 6, 250–1), the working of such implicated forces along desert peripheries powered the formation of dynastic powers in the Islamic Maghrib. Whereas this does not seem to have happened under the high Roman empire, the first signs of the phenomenon by late Roman antiquity is a development of some significance, as Mattingly's studies of the Laguatan ‘tribal confederation’ have shown.
Despite the apparent size of this collection, better communication between the volumes in the series is a desideratum. In all of the speculation about trade contacts, possibly in slaves, possibly in textiles and other items, between the oases and the more fertile agrarian lands to the north and the south, it is striking not to find any reference to the Tariff of Zaraï (CIL 8.4508 + 18643; see P. Trousset, AntAfr 38/39 (2002), 355–73), most probably because it had been mentioned in an earlier volume in the series (D. Mattingly et al. eds, Trade in the Ancient Sahara (2017), 259–84: see the review by R. B. Hitchner, Libyan Studies 49 (2018), 200–01). In this list of portoria or transit dues on goods most probably coming out of the Sahara, right at the top of the list are mancipia, slaves, and a little further down a whole number of textiles and woven cloths (S. Guédon, AntAfr 50 (2014), 111–23). Elsewhere in the tariff are listed agricultural produce like dates and nuts, and minerals like salt and alum, also probably from Saharan sources. No guesswork is needed. Similar papyrological records (not, alas, covered by Boozer in ch. 4), mainly ‘customs receipts’ for items transported from oasis environments into the Nile valley, confirm much the same range of produce coming out of the desert (e.g. P.Oxy. 12.1439, 14.1652b; P.Grenf. 2.50c; BGU 3.697, 867; P.Amh. 2.117; P.Tebt. 2.461: imports of figs, alum, etc.).
The collected volume edited by John Scheid and Michel Zink, presentations at a journée d’études on north Africa held in 2018, is as compact as the Sterry–Mattingly is expansive. As Jehan Desanges indicates in his preface, the study of the indigenous ethnic groups of north Africa has made substantial advances since he first provided a systematic catalogue of the known tribes in 1962. Although small in scale, these studies do add substance to that knowledge. Lofti Naddari's study, undergirded by new epigraphical discoveries, clarifies the identity and structure of the Musunii Regiani, an ethnic group whose lands lay south and east of Roman Sufetula (Sbeïtla), and outlines their gradual development of Roman-type institutions. There follows a pair of contributions, by Michel Christol and Christine Hamdoune, on ethnic groups in Mauretania Caesariensis which, as both authors emphasise, was a very different environment from that of the proconsular province to the east. Christol presents new epigraphic evidence on an important Roman administrator of indigenous gentes, an official who often functioned as a high-ranking subordinate of the provincial governor. Hamdoune, by contrast, is more concerned with the self-presentation of the tribal elites themselves—how they used Latin writing in public but preferred to exploit indigenous themes in iconographic representations of themselves and their activities. She points out the importance of fortified redoubts on their fundi or highland estates as a core of their identity. Both contributors emphasise the role of military service and the focus of the Roman tribal administrators on recruitment of manpower. Konrad Vössing's study of the representation of African ethnic groups by Procopius and Corippus argues that, despite their manifest generic differences, both writers had a strong interest in presenting Maures and other African ‘barbarians’ as undifferentiated ‘hostile masses’. The argument has some force where Procopius is concerned, but seems much less convincing for the detailed ethnographic picture found in the Byzantine epic poet. The attempt by Ahmed M'charek to trace continuity in the identity of the Misiciri into the post-Arab conquest period, based mainly on linguistic similitudes, seems more open to question. The argument for continuity is not as compelling, say, as the one that he made for Maghrawa in central Tunisia (MEFRA 100 (1988), 731–60). Finally, a detailed survey by René Rebuffat on Tripolitania provides a synoptic overview of the various ethnic groupings over the whole of Graeco-Roman antiquity.
Taken together, the two volumes provide a global overview of the Roman and extra-Roman faces of the northern half of continental Africa. Whatever judgements one might make on the success (or not) of their inroads into the problem of state formation or tribal identities, they are worthy of investigation by those in the field. The Mattingly–Sterry project, in particular, offers not only a necessary supplement to Mediterranean-centred perspectives, but also a vivid demonstration of how much our knowledge of the Saharan environments of antiquity has been transformed by the acquisition of very large numbers of new archaeological data.
