The peoples of North Africa have long captured the interest of colonial writers, both ancient and more recent. Modern scholarship on Roman Africa, however, descends primarily from the post-colonial turn exemplified by the work of Abdallah Laroui and Marcel Bénabou in the 1970s.Footnote ¹ The interest in the history of North Africa in itself, rather than solely as part of a wider European empire, has brought with it a new, but hazy, understanding of North Africa’s place as a “frontier” of intense economic and cultural engagement.Footnote ² In recent decades, however, the increased pace of archaeological research has brought with it an influx of data about the people and places outside urban centers.Footnote ³ David Mattingly (M.) has taken this new data as an opportunity to explicitly unravel the colonial contexts that dominated descriptions of North Africa under both the Roman Empire and French colonial rule. The result is a comprehensive reimagining of North African peoples and landscapes.

M.’s intervention begins with the recognition that many of the histories written about North Africa have been “firmly built on a colonialist discourse” (279). Top-down colonial biases have even affected the name of the discipline: “Africa in the Roman Empire” is preferred over “Roman Africa” to avoid the baggage of imperial terminology. The impact of colonialism on the available evidence for North Africa looms large. Textual evidence, while easily accessible, has often been folded into Oriental discourses that “other” marginalized peoples. Archaeological work has focused on urban, monumental, and “high culture” sites, which are often poorly recorded. In contrast, we are still missing critical evidence for military sites and rural centers. This colonial archaeology has consistently overemphasized foreign influences on Africa’s development. As a result, M. identifies the most significant difficulty in the study of North Africa as a lack of evidence for North Africans themselves – not Greek, Punic, or Roman interlopers – and for the contribution of pre-colonial communities to North Africa’s heritage.

A major theme of M.’s analysis is the failure of the Romanization paradigm to describe North Africa. In M.’s view, Romanization and its related theoriesFootnote ⁴ cling too closely to colonial epistemologies: cultural influences are one-way, binary distinctions are made between nomadic and settled, as well as “civilized” and “barbarian,” and indigenous agency is ignored (53). Instead, M. draws upon the fast-growing body of post-colonial literature of the last two decades, influenced by Edward Said’s Orientalism and focusing on Roman imperialism.Footnote ⁵ The difficulty of this constellation of approaches is, as Mattingly notes, rooted in the tensions between agency and structure: how can one answer “Can the subaltern speak?”Footnote ⁶ without minimizing the intentionally destructive effects of imperialism.Footnote ⁷

Striking this difficult balance is the crux of the analysis by M. that follows: the weight of past bias in the scholarship pushes him to defend himself at the start from the inevitable charge that he is “anti-Roman” in searching for a post-colonial model that highlights local identity, tradition, and agency in an imperial mix (57). M. continues this work through a focus on “discrepant identities,” an approach that he has long championed,Footnote ⁸ which privileges the intentional ideological construction of communities over strict ethnographic labels. M.’s formulation of discrepant identity here is differentiated from its previous versions in its scrutiny of unequal power structures between different populations in Roman Africa.Footnote ⁹

Phoenician colonies in Africa, first brought into the limelight of Anglophone scholarship by María Eugenia Aubet’s The Phoenicians and the West, come first in M.’s analysis.Footnote ¹⁰ The study of the Phoenicians has benefitted greatly from the rise of rural archaeology in the 1990s to study the fragmentary hinterland of Punic colonization.Footnote ¹¹ This work has all but dispelled the myth that Punic colonization was less impactful or disruptive than later Greek colonies.Footnote ¹² The concomitant question of a coherent Punic “identity” has also been the focus of scholarly attention. Modern social constructionist approaches have taken the material culture to identify multiple “punicities,” which responded to local conditions.Footnote ¹³ M. shares this constructive focus on shared culture and material traditions, eschewing Greco-Roman,Footnote ¹⁴ and modern popular,Footnote ¹⁵ conceptions of a “Punic” identity. M. argues that Punic settlers shared similar burial customs and a Semitic religious system, which included human sacrifice in the cult of Baal and Tinnit (83). Their urban structures, however, were diverse: many cities along the Tunisian and Libyan coasts, such as Kerkouane, exhibit locally driven urban growth, and they are quite irregular in size and shape (80–101). These conclusions lead M. to argue that “a unified Libyphoenician ethnic identity is a misleading construct” (102). M.’s argument therefore builds and expands upon earlier scholarship that connects the difficulty of distinguishing specific Libyphoenician identity to its colonial context.Footnote ¹⁶

M.’s concentration on the North African Iron Age (NAIA)Footnote ¹⁷ establishes a critical framework within which colonial ventures in Africa disrupted a connected web of societies with shared social and technological characteristics. M.’s project is particularly interested in emphasizing the size and sophistication of these indigenous peoples’ settlements. Of particular importance is the continuity of settlement at Althiburos, which dates to the North African Iron Age (133–39),Footnote ¹⁸ and the density of Numidian sites across the well-populated Maghreb (table 4.2). In addition, M. sees the development of formerly Numidian sites into Phoenician and Roman communities along the coast of North Africa, for instance, as just part of a long-running NAIA trend that also extends deeper into the heartland (150). This is also seen among the Mauri, inhabiting the western half of Morrocco, who have left layers of pre-Roman development at the sites of Volubilis Footnote ¹⁹ and Rirha.Footnote ²⁰ M. identifies the North African Iron Age as an era of shared socio-political trends, a compelling narrative thanks to the comprehensive view of the (in places, still scantyFootnote ²¹ ) archaeological evidence to date.

M.’s sub-regional focus travels from west to east, crossing the Numidians (126–67), Gaetulians (169–77), and Garamantians (177–89) respectively in Chapters 4 and 5. The latter two are especially tricky to pinpoint since both Roman sources and archaeological investigation have demonstrated that their semi-sedentary societies straddled the Mediterranean and Saharan worlds (189–209). The role of the Garamantes in cross-regional trade as a proxy for the importance of Roman African territories has been the subject of fierce debate, although previous data have largely been limited to chance finds and models of archaeologically invisible goods, such as enslaved persons.Footnote ²² M. adds a regional survey of palaeobotanical data to these arguments, clearly demonstrating that the Garamantians cultivated a substantial set of agricultural products (a “crop package”) prior to Phoenician investment (187). M’s synthesis of oasis communities from the Syrtes to the central Sahara region and southern Morocco makes clear that the extent and complexity of NAIA Gaetulian communities was far greater than previously recognized.Footnote ²³ Moreover, the constellation-like clustering of dense hillforts scattered throughout pre-desert biomes shows that transhumant and settled communities were not fundamentally opposed (209). This last point is overdue in bringing the conversation on North Africa into line with the long-standing anthropological consensus that (semi-)nomadic and settled communities were part of the same systems.Footnote ²⁴ Overall, the array of evidence makes M.’s case for an expansive, complex, and evolving North African Iron Age compelling. These chapters are undoubtedly the highlight of the book.

Chapters 6 and 7 offer a critical review of the size and culture of the Roman military community in North Africa. The Roman army has long been seen as distinct from the communities in which it was embedded, especially in linguistic and religious terms: the promotion of Latin,Footnote ²⁵ in particular, as well as traditional customs,Footnote ²⁶ created a unique culture for Roman soldiers.Footnote ²⁷ The extent of the “forcing effect” of the army’s culture on the broader population is debatable,Footnote ²⁸ and this creates a clear difficulty for critiques of the traditional Romanization paradigm.Footnote ²⁹

M.’s work seeks to redefine both the scale and the impact of the Roman army at a local level. Purposefully eschewing the term “frontier,” which is fundamentally intertwined with imperial policy,Footnote ³⁰ M. instead uses “security zone” (246) to describe the Roman army’s operations as an army of occupation. This security zone is populated through an exhaustive survey of Roman military sites across North Africa (244–49), which will undoubtedly provide a useful foundation for future research.

Having established the size and embeddedness of the Roman army in the North African landscape in Chapter 6, M. then turns to the identities of this group in Chapter 7. M. aligns himself with the tradition of scholarship that identifies the Roman army as a distinct feature of a non- (or even anti-) Roman landscape.Footnote ³¹ Evidence from Numidian Lambaesis and Tripolitanian Gholaia underpins prominent case studies for the religious and iconographic diversity of Roman garrisons compared to their local region (310–14, table 7.3). This compilation work aims to argue for a Roman military community that, while conspicuous, held itself “aloof” from the indigenous populations around it (318). In this way, M. affirms the army’s importance to the story of an imperial Roman Africa, while also arguing for a limit on the army’s impact on North African identities writ large.

Chapters 8 and 9 question the foundations of Roman “urban success.” Cities have dominated both scholarly and popular conceptions of North Africa as evidence for the wealth and power of the Roman colonial project.Footnote ³² Determining how many Romans inhabited these cities, however, is a trickier proposition. M. follows the population estimates of Andrew Wilson to argue for the over-representation of Libyphoenician and coastal towns among the ranks of Roman cities.Footnote ³³ Despite the level of uncertainty in these numbers, they do well to expose general trends in the inconsistent connections between legal status, territory size, and urban accoutrements in North African cities (347–55). M. constructs a typology of various “urban trajectories” in the region (355–77), intending to make the evidence from these cities comprehensible again. Chapter 9 turns to the epigraphic data to explore the identity of these community types, which demonstrate substantial multilingualism in employing Latin, Greek, Punic, and Libyan languages in naming and inscriptional practices (386–405). M.’s other focus is on burial types: although the evidence is very poorly documented, regional burial monuments such as the cupula point a way forward to understanding diachronic change without relying on linguistic markers.Footnote ³⁴ Given the relatively erratic nature of the evidence outside of major cities (415), most of M.’s remarks are suggestive rather than conclusive. Future research will surely elucidate these patterns further.

Chapters 10 and 11 tackle the beleaguered history of rural sites in North Africa. As colonial focus on large, coastal Roman centers pervades the discipline, the inclusion of rural landscape analysis is still a rarity.Footnote ³⁵ Despite extensive attention over the last 40 years,Footnote ³⁶ M. laments the unevenness of work estimating the population of rural North Africa or quantifying economic and social changes through ceramic assemblages (443–47). The potential for more sites to be found is clear: M.’s discussion of the varied circumstances of centuriation efforts across North Africa and Carthage is particularly effective in explaining the intensification of agricultural production in the Roman Period (473–74). His analysis of the linguistic and religious culture of North Africa’s rural majority reveals similar patterns to those of city life: socially contingent language practices predominate, often in Latin or Punic (or some combination), but almost equally often with Libyan personal names (485–90). The persistence of Saturn cult iconographyFootnote ³⁷ provides further evidence for a religious tradition that continued to thrive across North Africa after the North African Iron Age (497–504). Each of these trends independently confirms M.’s overarching thesis that while the Roman rural boom relied on the forcible reshaping of rural communities’ structures, the survival of settlement traditions, linguistic ties, and religious activity demonstrates the resilience of North African peoples under an imposed imperial structure. The strength of M.’s theory of discrepant identities is clear in its ability to explain the continuity of the many overlapping and seemingly contradictory methods of self-fashioning in this diverse landscape.

Chapter 12 reflects quickly on the major industries of North Africa under Roman rule. A multi-scalar approach briefly touches upon different resources (grain, olive oil, marine animals, stone, slaves, pottery) and their trade across the region. This overview feels out of place, as it lacks the emphasis on identity construction that is characteristic of the previous chapters. Moreover, the space devoted to this subject is too limited to achieve a widespread synthesis of North Africa’s remarkable economic growth.Footnote ³⁸ M. cites, but does not connect, recent arguments about the importance of the Saharan slave trade,Footnote ³⁹ and other potential bodies of evidence, such as the development of export facilities at ports (550–59).Footnote ⁴⁰ This chapter should serve as a springboard for future work that draws the remains of industrial production facilities in Roman North Africa into a common economic narrative.Footnote ⁴¹

Between Sahara and Sea is exceptional for its intensive approach to the archaeology of North Africa from the NAIA to the Late Antique period. M. brings together a vast quantity of archaeological material, much of it from projects he led personally, in an encyclopedic fashion. This is a very long book, and there are places where the reader might appreciate a less extensive discussion, particularly in chapters 6 and 12, where the rapidly expanding connections cannot be covered in requisite depth. Nevertheless, taking local communities and local identities seriously, and thereby escaping the Romans’ ethnographizing gaze, has paid great dividends here.Footnote ⁴²

This book can and should be the stepping-stone to a surge of archaeological and historical interest in North Africa. M. highlights the glaring gaps in fieldwork, especially the excavation of rural sites beyond the coastal band, that obscure the history of African peoples before the arrival of Punic, Greek, and Roman colonists. There is certainly a need for more research within border zones, and increasingly outside them as well. Future explorations might also move into phenomenological or environmental readings of Roman frontier zones as areas where an individual experiences Roman power (in terms of what crops are grown, how roads are managed, or what is traded) long before, and possibly much more vividly than, one experiences actual Romans. This work may further flesh out the experience of discrepant identities in their specific African context and add comparative material to investigations of imperial power and identity in other Roman provincial contexts. M.’s contribution here marks a turning point in the narrative of Africa in the Roman Empire and will undoubtedly be an essential work for years to come.