Navigation – Plan du site

AccueilThématiques16Remapping the Senegambian megalit...

Remapping the Senegambian megaliths: Beyond the culture area paradigm

Une nouvelle cartographie des mégalithes de la Sénégambie : au-delà du paradigme des aires culturelles
Cameron Gokee, Adama Harouna Athie, Ibrahima Thiaw et Assane Ndiaye

Résumés

Entre 1969 et 1982, Victor Martin et Charles Becker ont mené une série de missions de reconnaissance pour documenter des milliers de sites dans le cadre de leur Inventaire des Sites Protohistoriques de la Sénégambie. Cet ensemble de données est depuis devenu une référence fondamentale de l’archéologie de la Sénégambie centrale, où les cartes de répartition des sites continuent de définir une « zone mégalithique » et de fournir un contexte aux fouilles menées sur des sites de référence bien connus. Nous soutenons cependant que les données spatiales et quantitatives de l’Inventaire peuvent contribuer davantage à appréhender les sites mégalithiques comme des lieux inscrits dans des paysages sociaux complexes. Nous commençons par examiner comment l’Inventaire est né d’un paradigme de zone culturelle et comment il le perpétue à travers des stratégies de cartographie basées sur la sélection et la simplification. Nous plaidons ensuite pour un paradigme plus relationnel en archéologie sénégambienne en développant une cartographie critique des sites mégalithiques récemment numérisés pour MAEASaM (Mapping Africa’s Endangered Archaeological Sites and Monuments). Nous utilisons ici les résultats de relevés récents et des outils géostatistiques pour évaluer et visualiser les incertitudes de l’Inventaire, tout en nous efforçant de mieux illustrer les monuments mégalithiques comme des expressions matérielles d’interactions sociales à différentes échelles. Ce faisant, nous soulignons comment de nouvelles approches des données « legacy » de l’Inventaire et d’archives spatiales similaires pourraient contribuer à la décolonisation de l’archéologie africaine.

Haut de page

Notes de la rédaction

You will find all the tables mentioned in the text in the ‘Attachments’ section of this article. There is also a clickable link whenever the table is mentioned in the article.

Texte intégral

We would like to express our gratitude to the many scholars, both past and present, whose foundational research on Senegambian megaliths has made this discussion possible. We are particularly indebted to Victor Martin and Charles Becker; our critical engagement with the Inventaire is a testament to the importance of their work. We also thank MAEASaM for supporting the digitization and survey efforts involved in this study, and editors Léa Roth, Chloé Capel, and Gérard Chouin for the opportunity to contribute a not-quite-cold case to this special issue of Afriques. Lastly, we are grateful to two anonymous reviewers for their constructive feedback. Any remaining errors are entirely our own.

Introduction: A not-quite-cold case

  • 1 V. Martin, C. Becker, 1984; recently published as C. Becker, V. Martin, 2021.
  • 2 H. Bocoum, L. Laporte, 2021.

1Between 1969 and 1982, the French scholars Victor Martin and Charles Becker undertook a series of reconnaissance missions across Senegal and The Gambia to document settlement mounds, shell middens, earthen tumuli, and megalithic monuments dating to the 16th century or earlier. The resulting Inventaire des Sites Protohistoriques de la Sénégambie1 (hereafter Inventaire) has since become a mainstay in regional archaeology, where the distributions of these site types give shape to four protohistoric “culture areas” (Fig. 1). In the Megalith Zone, for example, the monumental sites mapped by Martin and Becker have been and continue to be an important frame of reference for field missions and heritage management efforts, including the inscription of the Stone Circles of Senegambia as a UNESCO World Heritage Site in 2006.2

  • 3 F.G. Richard, 2009; I. Thiaw, 2012; A.W. Southhall, 1970.
  • 4 A.F.C. Holl, H. Bocoum, 2017.
  • 5 L. Laporte et al., 2017, 2024.

2The Inventaire is therefore not a true “cold case” gathering dust on the rickety shelves of some colonial archive or national museum. Nevertheless, we argue that archaeologists have overlooked the potential for this dataset to challenge the very same culture area paradigm—rooted in colonial-era logics of tribal, ethnic, or national boundaries—that shaped its conception and realization in the first place.3 Even where recent projects move to interpret regional patterns in megalithic monument types from the Inventaire, their research agendas remain focused on the extraction of data related to architecture and burial, commemoration, and ritual practices at well-known type-sites, such as Sine-Ngayène4 and Wanar,5 long implicated in defining the Megalith Zone as a culture area. A reluctance to seriously consider spatial and quantitative data in the Inventaire could, of course, reflect real concerns about biases arising from the survey methods employed by Martin and Becker. If, however, we can work around these issues, this massive corpus of “legacy” data could help to further unpack old archaeological assumptions and address new anthropological questions about past socio-spatial relations in central Senegambia.

3To explore this possibility, we first retrace the history of research on Senegambian megaliths to demonstrate how the Inventaire emerged from, and now sustains, a culture area paradigm—particularly through mapping strategies of selection and simplification. We then advocate for a more relational paradigm by developing a critical cartography of megalithic sites recently digitized for Mapping Africa’s Endangered Archaeological Sites and Monuments (MAEASaM). Specifically, we use recent survey data and geostatistical tools to identify (and mitigate) potential biases in the Inventaire as a step towards alternative mapping strategies that can better illustrate the uneven relations of tradition and authority materialized in the megalithic monuments of central Senegambia. In so doing, we emphasize how fresh approaches to legacy data from the Inventaire and similar spatial archives could aid with efforts to decolonize African archaeology as we look beyond the culture area paradigm and an over-emphasis on excavation.

Figure. 1. Map of Senegambian protohistoric culture areas and monumental sites in the Senegambian Megalith Zone

Figure. 1. Map of Senegambian protohistoric culture areas and monumental sites in the Senegambian Megalith Zone

Excavated sites mentioned in the text: (1) Wanar; (2) Mbolop-Tobé; (3) Sine-Ngayène, Ngayène II, and Santhiou-Ngayène; (4) Wassu; (5) Kodiam; (6) Tiékène-Boussoura; and (7) Saré-Diouldé.

Illustration by C. Gokee (2025), redrawn from V. Martin, C. Becker, 1974b, p. 404.

Mapping the archaeology of central Senegambia

  • 6 H. Bocoum et al. 2002; C. Descamps, F. Descamps, 2010; J. Maley, R. Vernet, 2013; S. Canós-Donnay, (...)

4Senegambia generally refers to a region encompassing the Senegal and Gambia river basins, as well as areas drained by the Casamance and Cacheu rivers to the south (Fig. 1). Today these rivers and their tributaries help to sustain the modern nations of Senegal and The Gambia, but they have been shaping human geography in this corner of West Africa for thousands of years6—from the arrival of Pleistocene foragers to the late Holocene spread of diverse agro-pastoral communities. From AD 500 onwards, some of these waterways began to attract dense concentrations of human settlement, precipitating the development of complex societies of varying size, scale, and political organization, with urban centers, sociopolitical stratification, long-distance trade, and religious orientations, including conversion to Christianity and Islam. After AD 1500, the rivers of Senegambia became openings to an Atlantic world where the trade in slaves would contribute to the rise of predatory states and the unraveling of local social fabrics, making the region vulnerable to the establishment of French and British colonial boundaries.

5This historiography of Senegambia emerges, in large part, from more than a century of amateur and professional archaeology at sites ranging from prehistoric (Stone Age) quarries and encampments to protohistoric (Iron Age) settlements, shell middens, earthen tumuli, and megalithic monuments—these latter spread over 38,000 km2 of tropical savanna between the Gambia and Saalum river basins. In what follows, we summarize three generations of research in this region to show how interpretations of megalithic funerary practices, and associated social identities and institutions, have evolved in relation to broader trends in archaeological method and theory across francophone West Africa. At the same time, we consider how the persistence of certain mapping practices, particularly those based on the Inventaire, have contributed, whether explicitly or implicitly, to the intellectual inertia of a culture area paradigm in the archaeology of Senegambia, and particularly its central region.

Colonial era

  • 7 J.W. Maxwell, 1898.
  • 8 E.g. E.T. Hamy, 1904; C. Duchemin, 1905; J.L. Todd, G.B. Wolbach, 1911; H. Parker, 1923; P. Jouenne(...)
  • 9 R. Mauny, 1961, p. 164-172.
  • 10 P. Ozanne, 1965; P.O. Beale, 1965.

6Antiquarian interest in Senegambian megaliths began in 1891 when the French colonial administrator Captain Pineau dug around some laterite monoliths at the site of Samé, where he evidently found nothing of historical value.7 Over the next few decades, however, other French and British amateurs routinely discovered human remains and grave goods during their expeditions to document and excavate megalithic monuments across central Senegambia.8 Following these early missions, French archaeologist Raymond Mauny first directed professional excavations in 1956 at the sites of Diam-Diam and Sine-Ngayène,9 while British archaeologists undertook similar fieldwork in the 1960s at Wassu and neighboring sites in The Gambia.10 By the end of this era, antiquarians and archaeologists had excavated some 50 monuments (and dozens of human burials) from 31 different sites concentrated between the lower Gambia and Saalum rivers.

  • 11 P. Jouenne, 1930.
  • 12 H. Parker, 1923.

7Given their methods and prejudices, most of these scholars were content to present their discoveries as evidence for the funerary function of megalith circles, though some offered further interpretations touching on cultural processes. Pierre Jouenne, for example, read the arrangements of megaliths, including frontal stones topped by disc-shaped buttons, as evidence for a solar cult,11 while Henry Parker argued that these monuments were the work of Carthaginian craftsmen from North Africa.12 Implicit in these theories was the idea that the megalith builders were a single population occupying a discrete territory defined by the distribution of megalith sites (Fig. 2), but lacking direct historical ties to Sereer, Mandinka, Wolof, and other contemporary ethnic groups in this same region.

Figure 2. Map of the Senegambian Megalith Zone captioned “Sketch map of the Gambia Valley: The distribution of Stone Circles is Shaded.”

Figure 2. Map of the Senegambian Megalith Zone captioned “Sketch map of the Gambia Valley: The distribution of Stone Circles is Shaded.”

H. Palmer, 1939, Fig. 1.

Culture historical era

  • 13 de Barros, 1990.
  • 14 V. Martin, C. Becker, 1984.
  • 15 Mbanar in Wolof; podom in Sereer.
  • 16 V. Martin, C. Becker, 1974b; Becker et al., 1987.

8Between the 1960s and 1980s, the cultural identities of the megalith builders and other past populations emerged as key inquiries in Senegambian archaeology. For a new generation of archaeologists informed by the culture area paradigm prevalent in francophone West Africa,13 resolving these questions depended on the systematic recording, collection, classification, and dating of archaeological remains viewed as normative cultural traits. At the forefront of this agenda, the French scholars Victor Martin and Charles Becker initiated a census of Senegambian sites from 1969 to 1973, which they continued to expand and revise over the next decade. This project culminated in the Inventaire des Sites Protohistoriques de la Sénégambie14—a record of several thousand sites including qualitative and quantitative information about their surface features and artifacts (see discussion below). Using these data, Martin and Becker then mapped the distributions and densities of shell middens, tumuli,15 megaliths, and habitation and iron-smelting sites to define four protohistoric culture areas (Fig. 1) and model historical demographics within them.16

  • 17 G. Thilmans, C. Descamps, 1974, 1975; Thilmans et al., 1980.

9These culture areas quickly became a frame of reference for describing and explaining the spatio-temporal patterns observed at excavated sites across the region. In the Megalith Zone, for instance, Guy Thilmans and Cyr Descamps focused their research on several type-sites variably associated with the ruins of earthen tumuli, stone tumuli, stone circles, and megalith circles.17 By extracting grave goods, human remains, and radiocarbon samples from these latter three monument classes, these scholars were able to establish an approximate chronology for megalith building (200 BC to AD 1600) and further subdivide the Megalith Zone into three regional facies: a Western facies (exemplified by Sine-Ngayène) with more abundant earthen tumuli and megalith circles and burials associated with pottery, iron weapons, and copper bracelets; an Eastern facies (exemplified by Saré-Diouldé) with more abundant stone tumuli associated with numerous burials, but few grave goods; and a Central facies (exemplified by Tiékène-Boussoura) characterized by the full range of monuments, as well as burials with frequent dental modifications, but few weapons, ornaments, or pottery. Drawing on a schematic map of these patterns, Thilmans and Descamps concluded that the megalithic funerary tradition had originated in the Central facies before diffusing first to the west and later to the east.

  • 18 J. Joire, G. Duchemin, 1955; Descamps et al., 1978.
  • 19 G. Thilmans, C. Descamps, 2006.
  • 20 Unless otherwise noted, radiocarbon dates (two-sigma standard deviation) in the text have been cali (...)
  • 21 G. Thilmans et al., 1980.

10At the same time, the overlapping presence of earthen tumuli in the Western facies of the Megalith Zone raised questions about affinities with the Tumulus Zone, a culture area defined by sites with massive burial mounds measuring 10–80 m in diameter. Earlier excavations of tumuli at Rao-Nguiguélah and Rao-Massar (1086-1405 calAD),18 along with those by Thilmans and Descamps at Ndalane (1050-1153 calAD),19 had revealed depositional histories with one or more primary burials, sometimes in chambers holding a wooden bed overlaid by a thatched hut, and graves goods such as iron spears, copper bracelets, and gold, glass, and carnelian beads.20 This pattern differed from the multi-burial complexes found in the Megalith Zone, where many, but not all, earthen tumuli also had frontal monoliths.21 So how did these latter monuments fit into the spread of a megalithic funerary tradition across Senegambia?

  • 22 A. Gallay et al., 1982; A. Gallay, 2006.

11To answer this question, Alain Gallay carefully plotted data from the Inventaire to identify Mbolop-Tobé as a type-site within the overlapping ranges of megalithic monuments and earthen tumuli.22 Here the excavation of one megalith circle focused on the recovery of pottery vessels around the frontal stones, while the excavation of an earthen tumulus revealed a complex depositional history beginning with the burial of two individuals and one dog and concluding with a shallower, more recent interment. Gallay ultimately came to read this primary burial as a socio-political elite interred with his wife and/or other subservient individuals (morts d’accompagnement), similar to the complex mortuary rituals inferred from multiple inhumations in megalith circles and stone tumuli. After rejecting anomalous 6th millennium calBC radiocarbon dates for this monument (see Table 1), Gallay further argued that earthen tumuli post-date megalith circles in the Western facies, citing their peripheral position at most sites.

  • 23 e.g. M.A. Klein, 1968; C.A. Quinn, 1972; R.-P.H. Gravrand, 1983.
  • 24 F.G. Richard, 2015.
  • 25 Lomb in Sereer.
  • 26 C. Becker, V. Martin, 1982.

12Interpreting megalith circles and other funerary monuments in central Senegambia also required archaeologists of this era to engage with ethnohistorical evidence and intercultural interactions, either in the form of migration or conquest, resulting in the acculturation of diverse social practices, including burial customs.23 According to oral traditions, the original inhabitants of this region were Sosse, a Mande-speaking population organized into small-scale communities with leadership tied to ancestral land rights. Then, from the early to mid-2nd millennium, Sereer groups migrated from Takrur in the north and Mandinka elites (Guelwaar) arrived from Gaabu in the south to establish the kingdoms of Siin and Saluum, though these narratives may have been embellished to support more recent identity politics across the region.24 Complementing these traditions, written accounts from the 16th century onwards describe the burial of Mandinka and Sereer elites in smaller low mounds25 associated with prestige and staple goods and, in some cases, their wives and slaves—echoing the mortuary rites inferred from excavations into megalithic monuments.26 For better or worse, this ethnohistorical context could not categorically resolve the identity (ethnicity) of the megalith builders, but it did further cement the culture area paradigm as a foundation for new questions about their funerary practices and social processes.

Current era

  • 27 S.K. McIntosh, R.J. McIntosh, 1993.
  • 28 A.E. Lawson, 2003.
  • 29 A.F.C. Holl, H. Bocoum, 2017.

13Building on this foundation, the archaeology of central Senegambia has developed in two directions since the 1990s. First, several projects have sought to contextualize megalithic monuments within their regional landscapes by documenting nearby settlements, iron furnaces, and other features or locales excluded from the Inventaire. Using this dataset and a map of the Megalith and Tumulus Zones to orient their research, Susan and Rod McIntosh, for example, led a systematic survey of the Bao Bolon drainage to identify shallow habitation sites attesting to a long history of occupation by small agro-pastoral communities.27 Subsequent fieldwork by Amy Lawson similarly found small settlements clustered around megalithic sites in The Gambia, which she interpreted as a landscape of shifting cultivators (serial sedentism) whose monumental cemeteries staked out their ancestral ties to the land.28 This reading gains support from more recent surveys in the Bao Bolon area by Augustin Holl and Hamady Bocoum, who argue that megalithic cemeteries would have defined territorial access to cultivable land and other resources for populations dispersed along seasonal watercourses.29

  • 30 A.F.C. Holl, H. Bocoum, 2006, 2017.

14A second direction has moved to more accurately describe past funerary practices through the detailed excavation and dating of different monument architecture types. Holl and Bocoum, for example, aimed to compare depositional histories within and between the sites of Sine-Ngayène, Ngayène II, and Santhiou-Ngayène.30 Excavations of a double circle and ceremonial space at Sine-Ngayène exposed four phases of monument construction and use circa AD 700–1450, while excavations at two earthen tumuli uncovered elite burials with grave goods similar to those from Ndalane and Rao further north. Contrary to earlier interpretations by Thilmans and colleagues, Holl and Bocoum determined that most megalith circles have multiple secondary burials, while some stone circles and stone tumuli have few or no human remains. From this, they proposed that lineage-based societies in the region gradually shifted from emphasizing individual identity (primary burials) to collective identity (secondary and transitional burials) between 1360 BC and AD 1600 as larger groups came together to pool resources and control territory. Regrettably, the poor documentation of excavations and disastrous curation of human remains from this project precludes a more detailed interpretation of the associated megalithic funerary rites associated with these burials.

  • 31 L. Laporte et al., 2018, 2024.
  • 32 L. Laporte et al., 2022.
  • 33 L. Laporte et al., 2017.

15In this regard, ongoing research by a Franco-Senegalese team under the direction of Luc Laporte has proven more instructive.31 Since 2005, their excavations of megalith circles at Wanar and an earthen tumulus at Soto have revealed a three-step process for burying and commemorating the dead. The first step involved digging a burial pit, sometimes extended laterally into a small chamber, for the placement of human remains. The second step was the monumentalization of this burial place with a tumulus or a circle of laterite standing stones (from nearby quarries) interspersed with dry stone or organic material to become the façade for a “house of the dead.” The third step was the commemoration of the monument, and underlying inhumations, with subsequent burials and/or the placement of ceramic vessels around the frontal stones. Crucially, this team has modeled radiocarbon dates from Wanar and previously excavated sites to demonstrate that these steps, and their associated monument types, do not follow a strict evolutionary sequence, although the megalith circle construction clusters around AD 1000–1300.32 They also make careful use of ethnographic and archaeological evidence, including site distributions from the Inventaire, to situate megalithic funerary practices within their wider culture-historical context.33

  • 34 For a critical discussion, see Laporte et al., 2024, p. 1331-1336.

16Collectively, these recent projects, despite their different methodologies and interpretations, reveal the archaeological landscape of central Senegambia to be quite complex. Spatio-temporal and formal variability in the megalithic cemeteries does not map perfectly onto past funerary practices and social identities.34 One can nevertheless observe some very general trends in the forms and functions of monument “types” originally defined by Thilmans and Descamps (Table 1):

  • 35 A. Gallay, 2006; A.F.C. Holl, H. Bocoum, 2006.
  • 36 J. Joire, G. Duchemin, 1955; S. Magnavita, I. Thiaw, 2015; G. Thilmans, C. Descamps, 2006.
  • 37 C. Becker, V. Martin, 1982.

17(1) Earthen tumuli (Fig. 3a) range from larger (>20 m diameter) to smaller (<10 m diameter) mounds of sand often, but not always, with peripheral ditches and one or more frontal monolith stones several meters to the east. As noted above, excavations at three tumuli in the Bao Bolon area have found primary burials for one or two adults whose elite status can be inferred from accompanying inhumations and/or grave goods such as iron spears and copper ornaments.35 This pattern has parallels in funerary mounds further north,36 as well as ethnographic accounts of traditional Sereer burials for village chiefs and other high-status men.37

  • 38 A. Delvoye et al., 2016.
  • 39 e.g. L. Laporte et al., 2017, p. 97; D. Thiam, 2021.
  • 40 L. Laporte et al., 2017, 2018.

18(2) Megalith circles (Fig. 3b) are laterite monoliths arranged in single rings (4–5 m diameter) or concentric double rings (7–9 m diameter) with one or more frontal monoliths or V-shaped stones. Some of these monuments, such as those at Tiékène Boussoura, mark primary burials of one to several individuals with iron spearheads and metal and stone jewelry, though most contain complex secondary burials with intermingled remains and grave goods from up to 50 individuals (sometimes in association with primary burials). The associated frontal stones were undoubtedly places for commemoration offerings to these ancestors, as attested by deposits of diverse, and often numerous, pottery vessels.38 Although megalith circles are unique to central Senegambia, archaeological examples of laterite monoliths have been identified elsewhere across West Africa,39 and architectural analogies for interpreting these monuments as “houses of the dead” can be found among the Bassari, Coniagui, and other societies in southern Senegambia and Guinea.40

  • 41 L. Laporte et al., 2024, p. 1291-1292.
  • 42 A.F.C. Holl, H. Bocoum, 2017.
  • 43 L. Laporte et al., 2017, 2018.

19(3) Stone circles (Fig. 3c) have one or more courses of laterite blocks organized into large (5–10 m diameter) and small (3–5 m diameter) rings, frequently with one to four frontal stones to the east. Recognizing that shallow excavations at Ngayène II and Santhiou-Ngayène likely missed deeper human remains,41 the large stone circles appear to mark primary burials for a few individuals with iron spearheads and bodily adornments. According to Holl and Bocoum, the number of secondary burials with few grave goods in the smaller stone circles, and in the upper levels of the large ones, suggests that these monuments were spaces for symbolic offerings and/or ritual decomposition of bodies, perhaps prior to their placement in megalith circles.42 These monuments likewise have parallels with tombs made today by the Bassari and neighboring societies.43

  • 44 L. Laporte et al., 2024, p. 1258-1267.
  • 45 L. Laporte et al., 2024, p. 55.
  • 46 e.g. C. Dupuy et al., 2006; K. Sanogo, N. Coulibaly, 2013.

20(4) Stone tumuli (Fig. 3d) are mounds covered by laterite blocks (6–10 m diameter), often with one or more frontal monolith stones to the east. Beneath this superficial similarity, however, one can distinguish at least three different modes of construction.44 For some stone tumuli, the laterite blocks simply cap an earthen mound, while for others, including those excavated at Saré-Diouldé and Tiékène-Boussoura, they overlie a broad stone circle. For still others, the blocks appear to represent wall collapse above circular foundations standing several courses high, in which case their interpretation as mortuary monuments, rather than as domestic hut foundations, requires verification.45 While we await excavations to resolve these issues of form and function, it is sufficient to note that stone tumuli in central Senegambia may contain both primary and secondary burials, similar to those in the upper Senegal–Niger region,46 consistent with their use in complex funerary sequences emphasizing both individual and collective identities.

Figure 3. Typology of monuments in central Senegambia

Figure 3. Typology of monuments in central Senegambia

(a) earthen tumuli (SEN 963); (b) megalith circle (SEN 558); (c) stone circle (SEN 640); and (d) stone tumuli (SEN 962).

Photos by A. Athie (2024).

  • 47 A.F.C. Holl, H. Bocoum, 2017; L. Laporte et al., 2024.

21For contemporary archaeologists in central Senegambia, recognizing variability within and between these monument types—whether from primary funerary practices or secondary processes of reuse and ruination—represents a critical departure from assumptions of uniformity and (ethnic) continuity inherent in the culture area paradigm. Recent investigations at Sine-Ngayène and Wanar, for example, leverage this variability to interpret past mortuary traditions and socio-political institutions in Bao Bolon region.47 At the same time, these and other projects often retain elements of the culture area paradigm, most notably through the continued use of terms akin to “Megalith Zone” or “Western facies.” Furthermore, an enduring focus on type-sites encourages the extrapolation of local artifact/feature typologies and chronologies to megalithic cemeteries across the region, thereby impeding the analysis of variability required to study social processes and interactions at this wider scale.

From culture areas to social interactions

  • 48 G.M. Feinman, J.E. Neitzel, 2020.
  • 49 For discussions of these trends in Senegambian archaeology, see F.G. Richard, 2009; I. Thiaw, 2012.

22The persistence of the culture area paradigm is not unique to the archaeology of central Senegambia. As discussed by Feinman and Neitzel,48 this framework emerged in the mid-19th century to classify and interpret archaeological traits in terms of social evolution, ethno-national heritage, and, most notoriously, biological race.49 By the early-20th century, these traits came to define archaeological cultures/traditions as analogous to contemporary ones, with historical change primarily attributed to diffusion and migration. Despite theoretical shifts to view past cultures as adaptive systems or relational structures, the culture area paradigm continues to constrain research in two ways. First, an emphasis on type-sites and elites—such as those entombed in megalithic monuments—can obscure broader patterns of variability in status, production, exchange, and consumption in past societies. Second, reifications of culture areas usually force complex socio-political and economic networks into bounded units, thereby impeding our ability to study how social affiliations form and dissolve through interactions at multiple scales.

  • 50 E. Bollwerk, 2015.
  • 51 For examples, see A. Delvoye, 2022, Fig. 2; A. Gallay, 2010, Fig. 1; A.F.C. Holl, H. Bocoum, 2017, (...)

23According to Elizabeth Bollwerk,50 archaeological cultures/traditions owe much of their intellectual inertia to mapping practices of selection and simplification. In central Senegambia, for example, amateur scholars initially defined the Megalith Zone as distinct by selecting certain qualitative traits (megaliths) for emphasis, then simplifying their geographic distributions into a discrete region on thematic maps (e.g. Fig. 2). The Inventaire lent an air of objectivity to this approach by providing quantitative data to delimit the Megalith Zone and neighboring culture areas (e.g. Fig. 1). Even recent studies, which explicitly draw connections to other West African mortuary traditions, continue to orient research around distribution maps of monumental sites and/or feature types where the visual concentration of dots in central Senegambia can implicitly reinforce the illusion of a coherent, spatially bounded cultural entity at this regional scale.51

  • 52 J.B. Harley, 2001; D. Wood, 2010; for archaeological applications of these perspectives, see papers (...)
  • 53 A.M. MacEachren et al., 2005.
  • 54 e.g. C. Knappett, 2011.
  • 55 E. Bollwerk, 2015.

24To dispel this illusion, we need to acknowledge that maps are not objective depictions of the archaeological record (or past realities); they are subjective constructs tied to specific research agendas and power relations.52 Thematic maps depicting sites with static dots and culture areas with solid lines are simply an archaeological custom. So, if we want to study social interactions materialized by artifacts, features, and sites at multiple scales, then we need to better visualize the uneven and often uncertain relations among them.53 One strategy now gaining traction in archaeology relies on network visualization to illustrate these relations with unweighted or weighted links.54 Another recent strategy, outlined by Bollwerk, makes use of map symbology, including scale, transparency, and crispness, to convey the “fuzziness” of boundaries in the archaeological record arising from both the complexity of social interactions in the past and the limitations of our methodologies in the present.55

25Finding inspiration in this latter approach, we propose that data from the Inventaire can help us move beyond representations of central Senegambia as a coherent Megalith Zone, which may seem ironic given their ongoing role in the reification of this culture area. Nevertheless, we argue that GIS-based interpolation and visualization of site locations—along with qualitative and quantitative data often overlooked in the Inventaire—can contribute to a relational perspective on megalithic cemeteries as places emerging from social interactions at multiple scales.

Revisiting the Inventaire des Sites Protohistoriques de la Sénégambie

26Culture area maps work through the selection and simplification of archaeological data to depict seemingly objective boundaries. Alternative maps therefore need to acknowledge variability and uncertainty in these data, which can arise from diverse social practices in the past as well as post-depositional processes and the survey methods and technologies recording them in the present. So how do these various sources of uncertainty impact our ability to visualize the megalithic landscape of central Senegambia using data from the Inventaire?

Reconnaissance (1969–1982)

  • 56 V. Martin, C. Becker, 1972, 1974a, 1981, 1984.
  • 57 V. Martin, C. Becker, 1984, p. 35-40.

27To answer this question, it is important to understand that Martin and Becker continually expanded and revised the Inventaire over more than a decade,56 before producing a final report and map of protohistoric sites for the Atlas National du Sénégal. In central Senegambia, they oriented successive field missions around a hydrographic network divided into 66 sectors on national basemaps with a 1:200,000 scale (Fig. 4). Unfortunately, the Inventaire gives little explanation of their methods beyond the sequential numbering of sites within each sector and naming them in relation to nearby villages,57 from which we can surmise the use of guides familiar with local megaliths. For each site, Martin and Becker focused on recording the number of different monument types: earthen tumuli, megalith circles, isolated monoliths, and tombelles (a category combining stone circles and stone tumuli). They also noted the spatial arrangement of laterite monoliths in the megalith circles and frontal alignments, the number of unique features such as V-shaped frontal stones and double megalith circles, and the presence of carved cupules, buttons, or disks on the monoliths.

28For the final Inventaire, Martin and Becker listed 1988 megalithic sites spread over an elliptical area covering 38,000 km2 and stretching more than 350 km from the town of Kaolack in the west to Goudiry in the east. These encompass 16,000 monuments and 28,288 laterite monoliths categorized as: 2849 earthen tumuli without frontal stone, 668 earthen tumuli with frontal stone(s), 1053 megalith circles, 6849 tombelles without frontal stone, 2182 tombelles with frontal stone(s), and 3300 isolated monoliths or monolith groups. Before mapping these data, however, we must carefully consider the impact of terminological issues and spatial inaccuracies resulting from the methods and technologies available more than 40 years ago.

Figure 4. Map of sites recorded in Inventaire for the Megalith Zone of central Senegambia

Figure 4. Map of sites recorded in Inventaire for the Megalith Zone of central Senegambia

V. Martin, C. Becker, 1984, carte 4a-b.

Reappraisal (1993–2024)

  • 58 G. Malaperdas et al., 2023.

29Our approach to these uncertainties began by integrating central Senegambian sites from the Inventaire into a GIS database following established guidelines for the digitization of historic maps.58 First, we scanned (300 dpi) the six regional maps (1:200,000 scale) included with the Inventaire, then applied tools available in ArcGIS Pro 3.3 (ESRI) to georeference these images using 6–8 permanent control points, such as roads and political boundaries, visible on a Hybrid Imagery basemap. Importantly, the total root mean square error (RMSE) for these control points had a range of 41.4–49.3 m across the six maps, giving us a measure of the spatial accuracy for transformations of the originals at this regional scale. Next, we digitized the 1987 site locations as vector (point) features in our GIS database and assigned them unique reference numbers linked to site-specific monument counts and descriptions in the Inventaire.

Figure 5. Map of central Senegambia showing megalithic sites recorded in the Inventaire and areas re-surveyed by later projects

Figure 5. Map of central Senegambia showing megalithic sites recorded in the Inventaire and areas re-surveyed by later projects

Illustration by C. Gokee (2025).

  • 59 S.K. McIntosh, R.J. McIntosh, 1993.

30Next, we compared these site locations with those from four recent surveys (Fig. 5) to assess their positional accuracy. The first attempt to critically assess the Inventaire was a 1988–1989 survey across western Senegal directed by McIntosh and McIntosh.59 Focusing on questions about the geographic context of monumental features and a supposed lack of settlement areas in the Megalith and Tumulus Zones, this project aimed to explore functional, chronological, and spatial relations among sites surveyed in a 1-km radius around the megalithic complexes of Sine-Ngayène, Mbolop-Tobé, Koupar, Djiguimar, and Kaymor. Despite the rarity of surface ceramics, the McIntoshes did find several habitation and iron-smelting locales with superficial deposits on the terraces above the seasonal Bao Bolon drainage and its tributaries (Table 2). They also noted damage to several sites documented by Martin and Becker (e.g. 15-115, 15-140) where farmers had broken up and moved megaliths to slow soil erosion. This suggests that the Inventaire could, depending on the scale of site destruction over the past five decades, be a better document of megalithic monuments in central Senegambia than the contemporary archaeological record.

  • 60 A.E. Lawson, 2003.

31From 1998 to 2000, Lawson initiated a similar project to provide archaeological context for megalithic sites further south in The Gambia.60 A full-coverage survey in two areas centered on Wassu (each approximately 25 km2) made use of 20–50 m transects to identify more than 250 sites, including settlements, fortifications (tata), iron-smelting workshops, European trading posts, and megalithic monuments (Table 2). Although Lawson recorded the latter as individual stone circles (n=153) and earthen tumuli (n=48), a visual inspection of her figures 9.21–9.22 suggests that these clustered into more than 70 sites—nearly double the number recorded in the Inventaire for the same area. This difference no doubt reflects the challenges faced by Martin and Becker, as francophone scholars based in Senegal, for conducting surveys in The Gambia, an anglophone country then lacking an institutional framework for archaeological research. At the same time, the greater number of monuments in the Inventaire hints at their long-term attrition due to land management practices similar to those observed by McIntosh and McIntosh.

  • 61 A.F.C. Holl, H. Bocoum, 2017.

32Building on these surveys and earlier excavations, Holl and Bocoum returned to the western Megalith Zone in 2002 to launch a regional study of mortuary ritual and social organization in megalith-building societies.61 Over the next nine years, they undertook large-scale excavations at Sine-Ngayène and two neighboring sites, as well as a systematic survey along 15 km of the Petit Bao Bolon drainage. Holl and Bocoum made use of handheld GPS units, now more widely available in archaeology, to record the coordinates for dozens of habitation sites, grinding areas, and megalith quarries, in addition to 34 of the 39 monumental sites catalogued by Martin and Becker in this same study area (Table 2). At first glance, this seems to provide support for the loss of megalithic sites after decades of farming, but the greater number of monumental features found by Holl and Bocoum suggests that plowing and erosion could make some features, such as stone circles/tumuli, more visible over time. Martin and Becker also recorded fewer earthen tumuli along the Petit Bao Bolon, but this more likely reflects a tendency by Holl and Bocoum to classify isolated monoliths as the frontal stones for these monuments—an issue we further consider below.

  • 62 e.g. E. Huysecom, 1987.

33Finally, one of us (Athie) recently led a team from the Institut fondamental d’Afrique noire (IFAN) to resurvey megalithic sites for MAEASaM. This program encourages the use of legacy data and remote sensing to identify, map, document, monitor, and ultimately protect archaeological sites under threat in many African countries. In Senegal, this has involved digitizing the Inventaire and other legacy datasets62 to extract spatial, chronological, and cultural information for 15,292 sites and, using visual surveys of GoogleEarth imagery, to identify another 13,770 possible sites, including those with shell middens, earthen and stone tumuli, and megalith circles. Focusing on three different areas (Fig. 5), the IFAN team set out in 2024 to document 46 megalithic sites and discover nearby settlement and iron furnaces, drawn from a sample of 646 sites in the Inventaire and another 79 located in GoogleEarth. Field verification at each site involved flying UAVs to obtain aerial photos of the monuments and recording their geographic coordinates with hand-held GPS units (+/- 5 m precision).

34These coordinates, along with those for another 31 sites matched from the Petit Bao Bolon survey, provide a basis for evaluating the site locations georeferenced from maps in the Inventaire. Although coordinates derived from GoogleEarth were fairly precise, those estimated from georeferencing had average positional accuracies of 425–551 m with a RMSE of 174–274 m. This means that we need to be very cautious using spatial data from the Inventaire to measure site-specific associations with geophysical features, such as watercourses and soil types, which can vary over several hundred meters. We can, however, make effective use of these site locations for regional analyses, such as clustering and interpolation, where positional (in)accuracies are relatively small when projected against the 38,000 km2 area of the SMZ.

35Comparing monument tallies between the Inventaire and subsequent surveys (Table 2), we find that the degree and direction of their differences are rather ambiguous. On the one hand, Martin and Becker identified fewer monumental sites than Lawson in The Gambia and fewer monuments than Holl and Bocoum along the Petit Bao Bolon, which confirms that their reconnaissance did not achieve complete coverage of the SMZ archaeological record. On the other hand, Martin and Becker observed more sites than Holl and Bocoum and more features in areas later surveyed by Lawson, which suggests that the Inventaire is, at least in some cases, a more complete record of monumental sites adversely impacted by farming over the past few decades. As we discovered during the IFAN resurvey, however, the accuracy of monument counts could vary from one site to another: on some, there were fewer monuments; and on others, there were more monuments. This is likely because past monument counts were not always exhaustive, and agricultural activities are continuing to convert the monuments, particularly the earthen tumuli, into arable land.

  • 63 See C. Descamps, 1980.

36Other uncertainties in the Inventaire arise from the classification scheme employed by Martin and Becker.63 As summarized above (Table 1), there are four main monument types in central Senegambia—earthen tumuli, megalith circles, stone circles, and stone tumuli—as well as several hybrid forms and size variations, each associated with different burial and commemoration practices. The Inventaire, however, does not distinguish between large and small earthen tumuli, nor between stone circles and tumuli (grouped together as “tombelle”). Nevertheless, excavations of these monument types suggest that earthen tumuli mark primary burials, while large (>5–6 m diameter) stone circles and tumuli mostly contain primary burials and small (<5–6 m diameter) ones mostly contain secondary or transitional burials. In other words, size seems to have been a more significant distinction among stone circles/tumuli than their construction techniques. By glossing over these differences, the Inventaire hinders our ability to chart the dynamics of these specific funerary practices, but the co-presence of stone circles/tumuli at many sites suggests that we can carefully consider them together as features of a shared mortuary system at a regional scale.

37More difficult to interpret are the isolated monoliths (or groups of monoliths) recorded at 1147 sites (58%) in the Inventaire. Although these appear to demarcate open ceremonial spaces at some sites, most are likely the frontal stones for earthen tumuli or stone circles/tumuli that have been buried or destroyed. Because we cannot discriminate between these possibilities, we choose to maintain this type as a placeholder for tumulus-style funerary practices. This issue, together with those outlined above, emphasizes the need to account for uncertainties with site locations and monument types/counts in the Inventaire as we move from culture area maps to more relational views of megalithic cemeteries as places within an uneven social landscape.

Remapping the megaliths

  • 64 J.W. Crampton, 2010.
  • 65 N. Lambert, C. Zanin, 2016.

38Our approach to remapping the Senegambian megaliths takes inspiration from recent trends in critical cartography.64 By recognizing subjectivity in map-making, this approach encourages us to visualize uncertainty in archaeological data, even as we read these data as evidence for the social production of space. Because site distribution maps often fail to capture such nuance, we experiment with map symbology—including color, texture, shape, and size65—to represent ambiguity and complexity in site positions/attributes from the Inventaire. Specifically, we employ graduated symbol and multivariate mapping tools in ArcGIS Pro 3.3 (ESRI) to explore how mortuary monuments emerged from social interactions at multiple scales within and beyond the so-called Megalith Zone.

Graduated symbols

  • 66 G. Thilmans et al., 1980.
  • 67 L. Laporte et al., 2017.
  • 68 V. Martin, C. Becker, 1974b.
  • 69 A.F.C. Holl, H. Bocoum, 2017, p. 96.

39As illustrated in Figure 6, we can begin to represent variability among Senegambian megalith sites by using graduated symbology to scale them based on monument quantities from the Inventaire. Although this map does not convey uncertainty from depositional processes and survey biases, the side-by-side comparisons of these data do help to visualize regional patterns previously observed by Martin and Becker and subsequent archaeologists. Spatial clustering in the graduated symbols, for example, nicely conveys preferential site placement along the Gambia and Saluum rivers and their seasonal tributaries. Meanwhile, the variable distributions of these symbols loosely fit the three facies defined by Thilmans and Descamps: a Western facies with the highest densities of earthen tumuli and megalith circles; a Central facies with the greatest diversity in monument types and architectural variations such as double circles and V-stones; and an Eastern facies with the highest densities of stone tumuli and few megalith circles.66 Figure 6 also emphasizes regional trends previously charted in GIS by Laporte and his team: monoliths, megalith circles, and platforms for complex funerary rites are prevalent in ethnographic populations and archaeological sites to the south and east, whereas earthen and stone tumuli, largely for primary inhumations, are more often found in historical populations and archaeological sites to the north and east.67 Finally, this map works to depict a spatial hierarchy of megalithic site sizes, ranging from 1 to 181 monuments, referenced in arguments about past population densities68 and/or the territorial organization of “autonomous self-sustaining small mixed farming communities.”69

Figure 6. Spatial distributions of the monument types and architectural features recorded in the Inventaire

Figure 6. Spatial distributions of the monument types and architectural features recorded in the Inventaire

Illustration by C. Gokee (2025).

40The graduated symbols in Figure 6 also help to question the Megalith Zone as a coherent culture area. Fuzziness in the boundaries between regional facies, for example, cannot be simply explained away as sampling bias in the quest to link contemporary cultural groups to the archaeological sites of their ancestors. Rather than downplay the appearance of megalith circles in the Eastern facies or stone circles/tumuli in the Western facies, we should ask how these minor trends attest to the messy, uneven ways in which people acted to reproduce and/or transform social relations from generation to generation. In this regard, differing monument counts, despite the uncertainties outlined above, hint at the unique history of each site within a multi-scalar mortuary landscape. As summarized in Figure 7, a vast majority of megalithic cemeteries (n=1214) have five or fewer monuments, presumably marking local, short-term events focused on the burial and commemoration of one or a few people. Meanwhile, a small minority of sites (n=12) have more than 80 monuments, perhaps reflecting their long-term use over centuries, the seasonal or permanent aggregation of regional populations, and/or the social authority and territorial claims of local elites. Whatever the case may be, a site-size map invites us to think about megalithic cemeteries as places implicated in the relational production of social identities and institutions across central Senegambia prior to the 16th century.

Figure 7. Counts for (a) mortuary monuments and (b) laterite monoliths at sites recorded in the Inventaire

Figure 7. Counts for (a) mortuary monuments and (b) laterite monoliths at sites recorded in the Inventaire

Illustration by C. Gokee (2025).

Multivariate models

41Our current understanding of megalithic place-making comes mainly from the excavation of type-sites, particularly Sine-Ngayène and Wanar in the Bao Bolon drainage. As outlined above, the diverse funerary rites associated with earthen tumuli, megalith circles, and stone circles/tumuli at these sites can be understood as materializing a complex field of social relations, including age, gender, lineage, status, and so forth. The emplacement of these relations is not, however, unique to large cemeteries in one corner of the Megalith Zone. From a regional perspective, the overlapping, uneven distributions of monument types (Fig. 6) hint at the ongoing assemblage of burial and commemoration practices through social interactions at multiple scales. A lack of chronological control prevents us from charting these interactions through time, but we can use Inventaire data to draw maps exploring the interplay between site-specific and regional variability in megalithic cemeteries, as well as the uncertainty in these data.

  • 70 J. Conolly, M. Lake, 2006, p. 97-101.
  • 71 L. Laporte et al., 2017; D. Thiam, 2021.

42Using an Empirical Bayesian Kriging tool, we first transformed monument type distributions into surface models with 2-km-x-2-km resolution (Fig. 8). Briefly, this tool applies Bayesian statistics to spatial covariance in data values between nearby points to interpolate values, and their associated uncertainties, for grid units across a study area.70 Kriging is particularly useful for archaeology because it does not require a uniform distribution of data (monument counts), and it weights clusters of points (sites) together to account for potential sampling biases. At first glance, the resulting models simply reaffirm well-known regional trends (Fig. 6), but the “fuzziness” of their color saturation does more, we argue, to illustrate unevenness in monument distributions and uncertainty around positional and attribute data from the Inventaire. Monument estimates extending to the edges of the survey area also encourage us to question the boundaries of a Megalith Zone by searching for sites missed by Martin and Becker and by visualizing relations to earthen tumuli further to the northwest and laterite monoliths to the south.71

Figure 8. Surface models (2 km x 2 km) interpolated from monument type counts recorded in the Inventaire

Figure 8. Surface models (2 km x 2 km) interpolated from monument type counts recorded in the Inventaire

Illustration by C. Gokee (2025).

  • 72 R.D. Drennan, 2009, p. 299-307.

43Next, we looked for correlations across these trends using Principal Components Analysis (PCA), a statistical technique to reduce the dimensionality of large datasets by defining a sequence of principal components (PC), each of which represents the direction (eigenvector) and amount (eigenvalue) of variance between all pairs of variables.72 By showing the relative contributions (loading) of these variables, this method helps to identify and visualize clusters in archaeological data. The results of our analysis (Table 3) found that PC1, with positive loading from stone circles/tumuli and negative loading from earthen tumuli, accounts for a majority (75.5%) of the variance in our interpolated monument counts. This is followed by PC2 (19.3%) with positive loading from stone circles/tumuli and earthen tumuli, and PC3 (3.4%) with positive loading from isolated monoliths and, to a lesser extent, earthen tumuli, megalith circles, and stone circles/tumuli. The positive loading of this latter category in all four components is almost certainly due to its conflation of two or more different monument types (see above), while the weak loading of megalith circles is consistent with their clustering at a regional scale. Future fieldwork could help to resolve these interpretations, but for the moment this exploratory PCA invites us to consider how site-specific associations in funerary monuments shaped, and were shaped by, complex social interactions across the wider landscape.

Figure 9. Composite RGB image for the first three principal components from the PCA for surface models of monument type counts

Figure 9. Composite RGB image for the first three principal components from the PCA for surface models of monument type counts

Illustration by C. Gokee (2025).

44To visualize this relational perspective, the thematic map in Figure 9 displays our first three components together as RGB color bands. Concentrations of blue (PC1) and green (PC2) are consistent with relatively discrete traditions of burial beneath stone circles/tumuli in the southeast and earthen tumuli in the northwest, while smaller, more dispersed concentrations of red (PC3) appear to represent funerary practices involving megalith circles in the west and isolated monoliths in the east. We can observe similar patterns in Figure 10 where the RGB color bands simply display our surface models for megalith circles, earthen tumuli, and stone circles/tumuli. Both maps effectively use color hue and value to depict the unevenness of these concentrations and the fuzziness of their boundaries, thereby offering an alternative to the conceptual confines of the Megalith Zone. In this way, they open the door for future research on megalithic cemeteries as places enmeshed in social relations at multiple scales.

Figure 10. Composite RGB image for surface models of monument type counts

Figure 10. Composite RGB image for surface models of monument type counts

Illustration by C. Gokee (2025).

Discussion: Site inventories for critical cartography in African archaeology

  • 73 V. Martin, C. Becker, 1984, p. 4.

[I]l serait heureux que le présent travail serve de point de départ à des travaux archéologiques plus complets dans tous les secteurs de la Sénégambie protohistorique, à cause de l’importance et de la richesse des vestiges qu’il annonce et inventorie. De même, il devrait être un outil pour faire comprendre la richesse et la diversité des vestiges historiques et donc pour promouvoir une action efficace en vue de la conservation, de la protection et de l’étude d’un patrimoine inestimable73

45Martin and Becker envisioned the Inventaire as a starting point, not a destination, for archaeological research and heritage management in Senegambia. In this regard, their project has proven quite successful over the past few decades, particularly between the Gambia and Saluum rivers where monumental site distributions have contributed to the definition of a Megalith Zone and its inscription as UNESCO World Heritage. For archaeologists, these data have also helped orient surveys to situate megaliths in their local landscapes and excavations to reconstruct the diverse array of funerary practices within them. Alongside these advances, however, site-distribution maps drawn from the Inventaire have worked, both explicitly and implicitly, to perpetuate colonial-era interests in classification and boundary-making inherited from a culture area paradigm.

46A turn to critical cartography offers one possible path away from this paradigm and its long epistemological shadow in Senegambian archaeology. By questioning the objectivity of maps, this approach invites us to better visualize uncertainty and variability in the archaeological record as impartial evidence for social relations (rather than boundaries) emerging from long-term interactions among people, places, and things. Our return to the not-quite-cold case of spatial and quantitative data for megalithic sites in the Inventaire therefore began by assessing uncertainties around the methods employed by Martin and Becker several decades ago. Comparisons with more recent surveys, including fieldwork in support of MAESaM, found that site coordinates georeferenced from Inventaire maps are usually within several hundred meters of those obtained from hand-held GPS units or satellite imagery. And although the Inventaire may undercount megalithic sites and monuments in some areas, particularly along the Gambia River, these numbers may be more accurate than those from recent surveys in areas where farming has toppled monoliths or erased earthen and stone tumuli.

  • 74 Compare interpretations in A.F.C. Holl, H. Bocoum, 2017; L. Laporte et al., 2024.

47These uncertainties are bothersome, but they do not preclude us from using the Inventaire to map central Senegambian megaliths in ways that align with a more relational paradigm for archaeology. We can, for example, employ graduated symbols (Figs. 5–6) to emphasize spatial variability in megalithic monument types and their associated funerary practices. We can also employ interpolation tools in GIS (Fig. 8) to smooth over sampling biases in the Inventaire and visualize uneven relations within and between monument type distributions too often shoehorned into the Megalith Zone and its Western, Central, and Eastern facies (Figs. 9–10). Such mapping strategies encourage us to read the geographies of earthen tumuli, megalith circles, and stone circles/tumuli as complex histories of interaction between people, and their ancestors, at multiple scales. For example, recent interpretations of local diversity in monuments and funerary practices at the excavated type-sites of Sine-Ngayène and Wanar have focused on the equally diverse social identities of people entombed within them (e.g. age, gender, lineage, status), as well as the interplay between these identities and the production of collective and/or individual social authority.74 By emphasizing regional variability in megalithic sites/types, our maps can help link these interpretations to historical accounts of Sereer migration from the north, Mandinka (Guelwaar) conquest from the south, and trade along the Gambia River—hypotheses to be tested through the future analysis, not simply visualization, of geospatial data.

  • 75 See discussions in E. Bollwerk, 2015; G.M. Feinman, J.E. Neitzel, 2020.
  • 76 D.C. Conrad, 2008; see also U. Dieckmann, 2021.
  • 77 E.g. C. Knappett, 2011; for a recent application of network analysis in African archaeology, see N. (...)
  • 78 N. Gupta, R. Devilliers, 2017.

48In other words, our re-mapping of these data is only a first step towards excising the culture area paradigm from Senegambian archaeology.75 A second step could involve greater engagement with the histories of natural and social landscapes maintained by local descendent communities, similar to work by David Conrad using topographical references in Mande oral epics to relate contemporary political concerns and cultural identities to past events and processes in the Upper Niger region.76 A third step could include network analyses to depict and quantify relations between sites in abstract socio-material space,77 which would also provide another workaround to uncertainty in coordinates georeferenced from the Inventaire. Yet one more step, and the goal of digitizing the Inventaire for MAEASaM, would harness the power of geovisualization tools in web-based GIS to enable our interactive exploration of relations among monumental sites at multiple scales.78

  • 79 P.R. Schmidt, 2009; J. Lydon, U.Z. Rizvi, 2010.
  • 80 Y. Hamilakis, 2025; U.Z. Rizvi, 2025.
  • 81 A. Lorde, 1984.
  • 82 S. Venkatesan et al., 2024.

49Importantly, our call to revisit the Inventaire resonates with a broader movement to decolonize archaeological practice in and beyond Africa.79 Much of this work seeks to break with colonial systems of thoughts and methodologies that linger in modern research practices. Contemporary cartographies of Senegambian megalithism, for example, remain bound up with a culture area paradigm and its colonial articulations with geography, race, and ethnicity. Building on a decolonial perspective of refusal and delinking,80 we can challenge this conceptual framework through critical cartography using legacy data, geospatial technologies, ground-truthing methodologies built on more sincere collaborations between archaeologists and local communities. Drawing the Inventaire into this project runs the risk of perpetuating imperial imaginaries and their modus operandi—like trying to use the master’s tools to dismantle the master’s house81—but our stance is informed by a decolonial agenda in which it matters who uses the tools and how and why they use them.82 Neither the house nor the tools can be attributed exclusively to the master, insofar as the building took place on expropriated land using expropriated labor. Moreover, it is not the tools, but the intelligence behind them, that define human creations. As a multiracial, multiethnic, multinational, and multigenerational team, we are aware that archaeological methods and theories stand on cracking colonial foundations in dire need of replacement. By applying new mapping strategies to old data from the Inventaire, we propose to further demolish the culture area paradigm in Senegambian archaeology and clear the way for relational interpretations and curatorial possibilities that blend more coherently with local worldviews.

  • 83 Y. Hamilakis, 2025.
  • 84 S. Canós-Donnay, I. Thiaw, 2026; I. Thiaw, 2020.

50Past challenges to this decolonizing project include a dearth of oral traditions related to Senegambian megalithism and a lack of collaboration between archaeologists and local communities, who are still largely made invisible in the process of knowledge production as simple informants or laborers. Unfortunately, foundational research, including the Inventaire, has made these challenges seem insurmountable, but meaningful transformations are possible if we actively pursue more sincere partnerships to learn with, instead of about, people living today in central Senegambia. This might entail novel approaches to research and preservation of archaeological sites, historical landscapes, and museum collections, including human remains, envisioned through a prism of care and repair.83 One such prism is the Senegambian concept of teranga, a social value of hospitality with deep roots in regional and global entanglements that can act as a catharsis for living together in ways that positively structure social space and promote resilience.84 Moving forward, the articulation of critical cartography with local aesthetics of teranga would offer more semantically and historically appropriate views of archaeological landscapes across Senegambia than the continuing use of culture areas.

  • 85 S.M. Spivey-Faulkner, 2021.
  • 86 R. Allen, B. Ford, 2019.
  • 87 L. Laporte et al., 2024, p. 1292.
  • 88 C. Gokee, I. Thiaw, 2018.
  • 89 A. Delvoye, 2022.

51This aligns with a growing trend for archaeologists to move away from excavation, with its parallels to colonial modes of resource extraction, and toward the use of non-destructive methods, such as satellite or geophysical remote sensing,85 and new analyses of legacy collections, such as museum collections and unpublished reports.86 In central Senegambia, amateur and professional archaeologists have already exhumed the remains of more than 600 people,87 often with little concern for repatriating these bones or engaging with descendent populations.88 This is not to deny the value of future excavations, but rather to insist on using fieldwork to create real community partnerships and enable more effective use of legacy collections. An example of the former strategy comes from archaeological research one of us (Thiaw) undertook on Gorée Island, where the discovery of six historical graves became a basis for difficult conversations about race, ethnicity, and class in the past. An example of the latter strategy comes from work by Adrien Delvoye using the ceramic sequence from Wanar to reassess the classification and chronology of pottery sherds from other megalithic sites in the IFAN collections.89 We hope these efforts provide a model for future research to enhance the interpretive value of site data in the Inventaire by excavating a more geographically and architecturally diverse sample of monuments classified by Martin and Becker.

  • 90 R. Guitat, 1970; E. Huysecom, 1987.

52In the meantime, we encourage the creative use of new geospatial methods and technologies to reanimate other “legacy” spatial datasets in West Africa. Beyond central Senegambia, one could examine the relational production of tradition and authority through the thousands of large earthen tumuli, shell middens, settlement mounds, and iron-smelting sites also catalogued in the Inventaire. Or one could make similar use of Stone Age site inventories, and associated artifact collections, available for Senegal and neighboring countries to study the social landscapes of earlier foraging communities.90 Our remapping of megalithic site data from the Inventaire proves that decades-old inventories can do more than define culture areas and point the way to promising excavation sites—they can help us pose new questions about social interactions unfolding at local and regional scales.

Haut de page

Bibliographie

Allen, R., Ford, B. (dir.), 2019, New life for archaeological collections, Lincoln, University of Nebraska Press.

Beale, P.O., 1965, The Anglo-Gambian Stone Circles Expedition, 1964/65, report presented to the President of the Gambia, Bathurst.

Becker, C., Diouf, M., Mbodj, M., 1987, “Les sources démographiques de l’histoire de la Sénégambie”, Annales de démographie historique, p. 15-31.

Becker, C., Martin, V., 1982, “Rites de sépulture préislamique au Sénégal et vestiges protohistoriques”, Archives suisses d’anthropologie générale, Genève, 46, p. 261-293.

Becker, C., Martin, V., 2021, “Répertoire des sites mégalithiques du Sénégal et de Gambie”, in H. Bocoum, L. Laporte (dir.), Paysages mégalithiques du Sénégal et de la Gambie : un patrimoine mondial, Dakar, Histoire Générale du Sénégal Éditions, p. 207-274.

Bocoum, H., Camara, A., Diop, A., Diop, B., Lame, M., Thiam, M. (dir), 2002, Éléments d’archéologie ouest-africaine : V. Sénégal, Dakar, Éditions Sépia.

Bocoum, H., Laporte, L. (dir.), 2021, Paysages mégalithiques du Sénégal et de la Gambie : un patrimoine mondial, Dakar, Histoire Générale du Sénégal Éditions.

Bollwerk, E., 2015, “From cultural complexes to complex social topography: A history of spatial approaches to Native cultural landscapes in the Middle Atlantic”, Bulletin of the History of Archaeology, 25, p. 1-14. DOI: 10.5334/bha.256

Bronk Ramsey, C., 2009, “Bayesian analysis of radiocarbon dates”, Radiocarbon, 51, p. 337-360.

Canós-Donnay, S., 2024, “Archaeology of Senegambia”, in Oxford research encyclopedia of African history, Oxford, Oxford University Press. DOI: 10.1093/acrefore/9780190277734.013.1376

Canós-Donnay, S., Thiaw, I., 2026, “Archaeology and community engagement in the Senegambia: From colonial mindsets to creative Interactions”, Journal of Community Archaeology & Heritage, 11, p. 74-87. DOI: 10.1080/20518196.2025.2592002

Conolly, J., Lake, M., 2006, Geographical information systems in archaeology, Cambridge, Cambridge University Press.

Conrad, D.C., 2008, “From the banan tree of Kouroussa: Mapping the landscape in Mande traditional history”, Canadian Journal of African Studies, 42, p. 384-408.

Crampton, J.W., 2010, Mapping: A critical introduction to cartography and GIS, Chichester, Wiley-Blackwell.

de Barros, P., 1990, “Changing paradigms, goals and methods in the archaeology of Francophone West Africa”, in P. Robertshaw (dir.), A history of African archaeology, London, James Currey, p. 155-172.

Delvoye, A., 2022, “Les traditions céramiques du mégalithisme du Sénégal et de Gambie : une nouvelle périodisation pour l’Afrique de l’Ouest aux 1er et 11e millénaires de notre ère”, Journal of African Archaeology, 20, p. 230-258.

Delvoye, A., Laporte, L., Bocoum, H., Bernard, R., Cros, J.-P., Djouad, S., Dartois, V., Lejay, M., Athié, A., Cissé, F., Quesnel, L., 2016, “L’art et la manière : Approche technologique des céramiques de dépôt dans le mégalithisme sénégambien – Le cas de la nécropole de Wanar (Sénégal)”, Journal of African Archaeology, 14, p. 115-134.

Descamps, C., 1980, “Compte-rendu : Sites protohistoriques de la Sénégambie in Atlas National du Sénégal (Martin & Becker, 1977)”, Journal des africanistes, 50, p. 161-163.

Descamps, C., Descamps, F., 2010, La préhistoire au Sénégal. Recueil de documents, Dakar, Nouvelles Éditions Africaines du Sénégal.

Descamps, C., Hebrard, L., Sarr, C., Thilmans, G., Diop, O., 1978, “Âges radiométriques du quaternaire d’Afrique au nord de l’équateur”, Bulletin de l’association sénégalaise pour l’étude du Quaternaire, 52-53, p. 11-177.

Dieckmann, U. (dir.), 2021, Mapping the unmappable: Cartographic explorations with Indigenous peoples in Africa, New York, Columbia University Press.

Drennan, R.D., 2009 (2nd ed.), Statistics for archaeologists: A commonsense approach, New York, Springer.

Duchemin, C., 1905, “Les mégalithes de Ia Gambie”, L’Anthropologie, 16, p. 633-638.

Dupuy, C., Saliège, J.-F., Tavier, M., 2006, “Trois inhumations sous tumulus pierrier à Dialaka (Ouest malien, bassin supérieur du Sénégal”, in C. Descamps, A. Camara (dir.), Senegalia : Étude sur le patrimoine ouest-africain, Paris, Éditions Sépia, p. 239-249.

Feinman, G.M., Neitzel J.E., 2020, “Excising culture history from contemporary archaeology”, Journal of Anthropological Archaeology, 60, 101230.

Gallay, A., 2006, “Le mégalithisme sénégambian: une approche logistique”? in C. Descamps, A. Camara (dir.), Senegalia : Étude sur le patrimoine ouest-africain, Paris, Éditions Sépia, p. 205-223.

Gallay, A., 2010, “Seriation Chronologique de la céramique mégalithique sénégambienne (Sénégal, Gambie), 700 cal BC – 1700 cal AD”, Journal of African Archaeology, 8, p. 99-129.

Gallay, A., Pignat, G., Curdy, P., 1982, “Mbolop Tobé (Santhiou Kohel, Sénégal) : contribution à la connaissance du mégalithisme sénégambien”, Archives suisses d’anthropologie générale, Genève, 46, p. 217-259.

Gestrich, N., Schardt, J.-M.P., 2024, “Pottery décor as networks on the Middle Niger, Mali”, Azania: Archaeological Research in Africa, 59, p. 396-432.

Gillings, M., Hacıgüzeller, P., Lock, M. (dir.), 2018, Re-mapping archaeology: Critical perspectives, alternative mappings, London, Routledge.

Gokee, C., Thiaw, I., 2018, “Human remains and archaeologies of identity in Senegal”, in B. O’Donnabhain, M.C. Lozada (dir.), Archaeological human remains: Legacies of imperialism, communism, and colonialism, New York, Springer, p. 141-153.

Gravrand, R.-P.H., 1983, La civilisation sereer : “Cosaan,” les origines, Dakar, Les Nouvelles Éditions Africaines.

Guitat, R., 1970, “Carte et répertoire des sites néolithiques du Sénégal”, Bulletin de l’Institut fondamental d’Afrique noire, Série B, 32, p. 1125-1134.

Gupta, N., Devillers, R., 2017, “Geographic visualization in archaeology”, Journal of Archaeological Method and Theory, 24, p. 852-885.

Hamilakis, Y., 2025, “A decolonial archaeology of refusal, care and repair”, Antiquity, p. 1–12. DOI: 10.15184/aqy.2025.10252

Hamy, E.T., 1904, “Quelques observations sur les tumulus de la vallée de la Gambie présentées à l’occasion d’une exploration récente de M. le Capitaine Duchemin”, Académie des Inscriptions et Belles-Lettres - Compte-rendu des séances, p. 560-569.

Harley, J.B., 2001, The new nature of maps: Essays in the history of cartography, Baltimore, Johns Hopkins University Press.

Holl, A.F.C., Bocoum, H., 2006, “Variabilité des pratiques funéraires dans le mégalithisme sénégambien : le cas de Sine Ngayène”, in C. Descamps, A. Camara (dir.), Senegalia : Étude sur le patrimoine ouest-africain, Dakar, Éditions Sépia, p. 224-234.

Holl, A.F.C., Bocoum, H., 2017, Megaliths, cultural landscape and the production of ancestors, Saarbrücken, Éditions Universitaires Européennes.

Huysecom, E., 1987, Die archäologische Forschung in Westafrika, Munich, C.H. Beck.

Joire, J., Duchemin, G., 1955, “Découvertes archéologiques dans la région de Rao (Bas-Sénégal)”, Bulletin de l’Institut français d’Afrique noire, Série B, 17, p. 249-333.

Jouenne, P., 1930, “Les monuments mégalithiques du Sénégal : Les roches gravées et leur interprétation cultuelle”, Bulletin du Comité d’études historiques et scientifiques de l’Afrique occidentale française, p. 309-399.

Klein, M.A., 1968, Islam and imperialism in Senegal: Sine-Saloum, 1847-1914, Stanford, Stanford University Press.

Knappett, C., 2011, An archaeology of interaction: Network perspectives on material culture and society, Oxford, Oxford University Press.

Lambert, N., Zanin, C., 2016, Practical handbook of thematic cartography: Principles, methods and applications, Boca Raton, CRC Press.

Laporte, L., Bocoum, H., Delvoye, A., Sanogo, K., Polet, J., Ceesay, B., Cros, J.-P., Athié, A., Djouad, S., Ndiaye, M., Armbruster, B., Ballouche, A., Eichhorn, B., Garnier, A., Lespez, L., Mathé, V., Robion-Brunner, C., 2017, “Les mégalithes du Sénégal et de la Gambie dans leur contexte régional”, Afrique : Archéologie & Arts, 13, p. 93-119.

Laporte, L., Cros, J.-P., Bocoum, H., Teixeira, M., Delvoye, A., Ndiaye, M., Djouad, S., Quesnel, L., 2018, “Mégalithes du Sénégal et pratiques sépulcrales : l’hypothèse de rites funéraires différés”, Journal des africanistes, 88, p. 116-147.

Laporte, L., Lefèvre, J.-C., Ndiaye, M., Oberlin, C., Delvoye, A., Stern, M., Faye, A.K., Ballouche, A., Cros, J.-P., Djouad, S., Quesnel, L., 2022, “Deux millénaires ou deux siècles de constructions mégalithiques au Sénégal et en Gambie ? Nouvelles dates pour Wanar et révision des contextes archéologiques”, Afrique : Archéologie & Arts, 18, p. 53-88.

Laporte, L., Ndiaye, M., Delvoye, A., Cros, J.-P., Ballouche, A., Lamotte, P., Djouad, S., Quesnel, L. (dir), 2024, La nécropole aux amants pétrifiés : ruines mégalithiques de Wanar (région de Kaffrine, Sénégal), Oxford, Archaeopress.

Lawson, A.E., 2003, Megaliths and Mande states: Sociopolitical change in the Gambia Valley over the past two millennia, PhD dissertation, Ann Arbor, University of Michigan.

Lorde, A., 1984, “The master’s tools will never dismantle the master’s house”, in Sister outsider: Essays and speeches, New York, Crossing Press, p. 110–113.

Lydon, J., Rizvi, U.Z. (dir.), 2010, Handbook of Postcolonial Archaeology, New York, Routledge. DOI: https://doi.org/10.4324/9781315427690

MacEachren, A.M., Robinson, A., Hopper, S., Gardner, S., Murray, R., Gahegan, M., Hetzler, E., 2005, “Visualizing geospatial information uncertainty: What we know and what we need to know”, Cartography and Geographic Information Science, 32, p. 139-160.

Magnavita, S., Thiaw I., 2015, “Nouvelles recherches archéologiques dans la zone des tumuli du Sénégal”, Nyame Akuma, 82, p. 3-10.

Malaperdas, G., Grethe, R., Zacharias, N., 2023, “Depicting the past: The value of old maps and topographic diagrams in cultural heritage through GIS”, Journal of Archaeological Science: Reports, 52, 104276. DOI: 10.1016/j.jasrep.2023.104276

Maley, J., Vernet, R., 2013, “Peuples et évolution climatique en Afrique nord-tropicale, de la fin du Néolithique à l’aube de l’époque moderne”, Afriques, 4. DOI : 10.4000/afriques.1209

Martin, V., Becker, C., 1972, “Notes sur les sites protohistoriques de la Sénégambie”, Bulletin de l’Association sénégalaise d’étude du Quaternaire africain, 35-36, p. 23-48.

Martin, V., Becker, C., 1974a, Répertoire des sites protohistoriques du Sénégal et de la Gambie, Kaolack, CNRS.

Martin, V., Becker, C., 1974b, “Vestiges protohistoriques et occupation humaine au Sénégal”, Annales de démographie historique, p. 403-429.

Martin, V., Becker, C., 1981, Atlas de protohistoire sénégambienne, Kaolack, CNRS.

Martin, V., Becker, C., 1984, Inventaire des sites protohistoriques de la Sénégambie, Kaolack, CNRS.

Mauny, R., 1961, Tableau géographique de l’Ouest Africain au Moyen Age, Dakar, Institut français d’Afrique noire.

Maxwell, J.W., 1898, “Stone-circles in Gambia”, Geographical Journal, 12, p. 522-527.

McIntosh, S.K., McIntosh, R.J., 1993, “Field survey in the tumulus zone of Senegal”, African Archaeological Review, 11, p. 73-107.

Ozanne, P., 1965, “The Anglo-Gambian Stone Circles Expedition”, Research Review, 1, p. 32-36.

Palmer, H.R., 1939, “Stone circles in the Gambia Valley”, Journal of the Royal Anthropological Institute, 69, p. 273-283.

Parker, H., 1923, “Stone circles in the Gambia”, Journal of the Royal Anthropological Institute, 53, p. 173-228.

Quinn, C.A., 1972, Mandingo kingdoms of the Senegambia, Evanston, Northwestern University Press.

Reimer, P., Austin, W., Bard, E., Bayliss, A., Blackwell, P., Bronk Ramsey, C., Butzin, M., Cheng, H., Edwards, R., Friedrich, M., Grootes, P., Guilderson, T., Hajdas, I., Heaton, T., Hogg, A., Hughen, K., Kromer, B., Manning, S., Muscheler, R., Palmer, J., Pearson, C., van der Plicht, J., Reimer, R., Richards, D., Scott, E., Southon, J., Turney, C., Wacker, L., Adolphi, F., Büntgen, U., Capano, M., Fahrni, S., Fogtmann-Schulz, A., Friedrich, R., Köhler, P., Kudsk, S., Miyake, F., Olsen, J., Reinig, F., Sakamoto, M., Sookdeo, A., Talamo, S., 2020, “The IntCal20 Northern Hemisphere radiocarbon age calibration curve (0–55 cal kBP)”, Radiocarbon, 62, p. 725-757.

Richard, F.G., 2009, “Historical and dialectical perspectives on the archaeology of complexity in the Siin-Saalum (Senegal): Back to the future?”, African Archaeological Review, 26, p. 75-135.

Richard, F.G., 2015, “‘The very embodiment of the Black peasant?’ Archaeology, history, and the making of the Seereer of Siin (Senegal)”, in F.G. Richard, K.C. MacDonald (dir.), Ethnic ambiguity and the African past: Materiality, history, and the shaping of cultural identities, Walnut Creek, Left Coast Press, p. 87-118.

Rizvi, U.Z., 2025, “The form of a shadow: Decolonial practice, refusal, and feminist killjoys”, in A. Nativ, G. Lucas (dir.), Shadow archaeologies: In the shadow of antiquity or for other modes of archaeological worldmaking, London, Routledge, p. 11-22.

Saint-Seine, J., 1939, Lettre à Théodore Monod, Toubacouta, 12-IV-1939 (Document XV-5), Archives de l’Institut français d’Afrique noire, Sénégal.

Sanogo, K., Coulibaly, N., 2013, “La problématique des ‘tumulus pierriers’ au Mali”, Afrique : Archéologie & Arts, 9, p. 101-112. DOI : 10.4000/aaa.330

Schmidt, P.R. (dir.), 2009, Postcolonial archaeologies in Africa, Santa Fe, School for Advanced Research Press.

Southhall, A.W., 1970, "The illusion of tribe", Journal of Asian and African Studies, 5, p. 28-50.

Spivey-Faulkner, S.M., 2021, “Juggling sand: Ethics, identity, and archaeological geophysics in the Mississippian world”, Journal of Archaeological Science: Reports, 36, 102882.

Thiam, D., 2021, “Observations sur le mégalithisme bissau-guinéen : recherches préliminaires”, Nyame Akuma, 95, p. 19-24.

Thiaw, I., 2012, “Colonization and the development of archaeology in Senegal”, in S.J. van der Linde, M.H. van den Dries, N. Schlanger, C.G. Slappendel (dir.), European archaeology abroad: Global settings, comparative perspectives. Leiden, Sidestone Press, p. 349-374.

Thiaw, I., 2020, “Archaeology of two pandemics and teranga asethetic”, African Archaeological Review, 37, p. 475-479.

Thilmans, G., Descamps, C., 1974, “Le site mégalithique de Tiékène-Boussoura (Sénégal) : fouilles de 1973-1974”, Bulletin de l’Institut fondamental d’Afrique noire, Série B, 36, p. 447-496.

Thilmans, G., Descamps, C., 1975, “Le site mégalithique de Tiékène-Boussoura (Sénégal) : fouilles de 1974-1975”, Bulletin de l’Institut fondamental d’Afrique noire, Série B, 37, p. 259-306.

Thilmans, G., Descamps, C., 2006, “Fouille d’un tumulus à Ndalane (région de Kaolack, Sénégal)”, in C. Descamps, A. Camara (dir.), Senegalia: Étude sur le patrimoine ouest-africain, Dakar, Éditions Sépia, p. 235-238.

Thilmans, G., Descamps, C., Khayat, B., 1980, Protohistoire du Sénégal, Tome I : Les sites mégalithiques, Dakar, Institut fondamental d’Afrique noire.

Todd, J.L., Wolbach, G.B., 1911, “Stone circles in the Gambia”, Man, 11, p. 161-164.

Venkatesan, S., Ntarangwi, M., Mills, D., Gillespie, K., Davé, N., Backhaus, V., 2024, “A decolonial anthropology: You can dismantle the master’s house with the master’s tools”, Critique of Anthropology, 44, p. 99-140.

Wood, D., 2010, Rethinking the power of maps, New York, The Guilford Press.

Haut de page

Notes

1 V. Martin, C. Becker, 1984; recently published as C. Becker, V. Martin, 2021.

2 H. Bocoum, L. Laporte, 2021.

3 F.G. Richard, 2009; I. Thiaw, 2012; A.W. Southhall, 1970.

4 A.F.C. Holl, H. Bocoum, 2017.

5 L. Laporte et al., 2017, 2024.

6 H. Bocoum et al. 2002; C. Descamps, F. Descamps, 2010; J. Maley, R. Vernet, 2013; S. Canós-Donnay, 2024.

7 J.W. Maxwell, 1898.

8 E.g. E.T. Hamy, 1904; C. Duchemin, 1905; J.L. Todd, G.B. Wolbach, 1911; H. Parker, 1923; P. Jouenne, 1930; H.R. Palmer, 1939; J. Saint-Seine, 1939.

9 R. Mauny, 1961, p. 164-172.

10 P. Ozanne, 1965; P.O. Beale, 1965.

11 P. Jouenne, 1930.

12 H. Parker, 1923.

13 de Barros, 1990.

14 V. Martin, C. Becker, 1984.

15 Mbanar in Wolof; podom in Sereer.

16 V. Martin, C. Becker, 1974b; Becker et al., 1987.

17 G. Thilmans, C. Descamps, 1974, 1975; Thilmans et al., 1980.

18 J. Joire, G. Duchemin, 1955; Descamps et al., 1978.

19 G. Thilmans, C. Descamps, 2006.

20 Unless otherwise noted, radiocarbon dates (two-sigma standard deviation) in the text have been calibrated using OxCal 4.4 with the IntCal20 curve (Bronk Ramsey 2009; Reimer et al. 2020).

21 G. Thilmans et al., 1980.

22 A. Gallay et al., 1982; A. Gallay, 2006.

23 e.g. M.A. Klein, 1968; C.A. Quinn, 1972; R.-P.H. Gravrand, 1983.

24 F.G. Richard, 2015.

25 Lomb in Sereer.

26 C. Becker, V. Martin, 1982.

27 S.K. McIntosh, R.J. McIntosh, 1993.

28 A.E. Lawson, 2003.

29 A.F.C. Holl, H. Bocoum, 2017.

30 A.F.C. Holl, H. Bocoum, 2006, 2017.

31 L. Laporte et al., 2018, 2024.

32 L. Laporte et al., 2022.

33 L. Laporte et al., 2017.

34 For a critical discussion, see Laporte et al., 2024, p. 1331-1336.

35 A. Gallay, 2006; A.F.C. Holl, H. Bocoum, 2006.

36 J. Joire, G. Duchemin, 1955; S. Magnavita, I. Thiaw, 2015; G. Thilmans, C. Descamps, 2006.

37 C. Becker, V. Martin, 1982.

38 A. Delvoye et al., 2016.

39 e.g. L. Laporte et al., 2017, p. 97; D. Thiam, 2021.

40 L. Laporte et al., 2017, 2018.

41 L. Laporte et al., 2024, p. 1291-1292.

42 A.F.C. Holl, H. Bocoum, 2017.

43 L. Laporte et al., 2017, 2018.

44 L. Laporte et al., 2024, p. 1258-1267.

45 L. Laporte et al., 2024, p. 55.

46 e.g. C. Dupuy et al., 2006; K. Sanogo, N. Coulibaly, 2013.

47 A.F.C. Holl, H. Bocoum, 2017; L. Laporte et al., 2024.

48 G.M. Feinman, J.E. Neitzel, 2020.

49 For discussions of these trends in Senegambian archaeology, see F.G. Richard, 2009; I. Thiaw, 2012.

50 E. Bollwerk, 2015.

51 For examples, see A. Delvoye, 2022, Fig. 2; A. Gallay, 2010, Fig. 1; A.F.C. Holl, H. Bocoum, 2017, Fig. 3.2; L. Laporte et al., 2024, I.4: Figs. 1-4.

52 J.B. Harley, 2001; D. Wood, 2010; for archaeological applications of these perspectives, see papers in M. Gillings et al., 2018.

53 A.M. MacEachren et al., 2005.

54 e.g. C. Knappett, 2011.

55 E. Bollwerk, 2015.

56 V. Martin, C. Becker, 1972, 1974a, 1981, 1984.

57 V. Martin, C. Becker, 1984, p. 35-40.

58 G. Malaperdas et al., 2023.

59 S.K. McIntosh, R.J. McIntosh, 1993.

60 A.E. Lawson, 2003.

61 A.F.C. Holl, H. Bocoum, 2017.

62 e.g. E. Huysecom, 1987.

63 See C. Descamps, 1980.

64 J.W. Crampton, 2010.

65 N. Lambert, C. Zanin, 2016.

66 G. Thilmans et al., 1980.

67 L. Laporte et al., 2017.

68 V. Martin, C. Becker, 1974b.

69 A.F.C. Holl, H. Bocoum, 2017, p. 96.

70 J. Conolly, M. Lake, 2006, p. 97-101.

71 L. Laporte et al., 2017; D. Thiam, 2021.

72 R.D. Drennan, 2009, p. 299-307.

73 V. Martin, C. Becker, 1984, p. 4.

74 Compare interpretations in A.F.C. Holl, H. Bocoum, 2017; L. Laporte et al., 2024.

75 See discussions in E. Bollwerk, 2015; G.M. Feinman, J.E. Neitzel, 2020.

76 D.C. Conrad, 2008; see also U. Dieckmann, 2021.

77 E.g. C. Knappett, 2011; for a recent application of network analysis in African archaeology, see N. Gestrich, J.-M.P. Schardt, 2024.

78 N. Gupta, R. Devilliers, 2017.

79 P.R. Schmidt, 2009; J. Lydon, U.Z. Rizvi, 2010.

80 Y. Hamilakis, 2025; U.Z. Rizvi, 2025.

81 A. Lorde, 1984.

82 S. Venkatesan et al., 2024.

83 Y. Hamilakis, 2025.

84 S. Canós-Donnay, I. Thiaw, 2026; I. Thiaw, 2020.

85 S.M. Spivey-Faulkner, 2021.

86 R. Allen, B. Ford, 2019.

87 L. Laporte et al., 2024, p. 1292.

88 C. Gokee, I. Thiaw, 2018.

89 A. Delvoye, 2022.

90 R. Guitat, 1970; E. Huysecom, 1987.

Haut de page

Table des illustrations

Titre Figure. 1. Map of Senegambian protohistoric culture areas and monumental sites in the Senegambian Megalith Zone
Légende Excavated sites mentioned in the text: (1) Wanar; (2) Mbolop-Tobé; (3) Sine-Ngayène, Ngayène II, and Santhiou-Ngayène; (4) Wassu; (5) Kodiam; (6) Tiékène-Boussoura; and (7) Saré-Diouldé.
Crédits Illustration by C. Gokee (2025), redrawn from V. Martin, C. Becker, 1974b, p. 404.
URL http://journals.openedition.org/afriques/docannexe/image/5785/img-1.jpg
Fichier image/jpeg, 593k
Titre Figure 2. Map of the Senegambian Megalith Zone captioned “Sketch map of the Gambia Valley: The distribution of Stone Circles is Shaded.”
Crédits H. Palmer, 1939, Fig. 1.
URL http://journals.openedition.org/afriques/docannexe/image/5785/img-2.jpg
Fichier image/jpeg, 320k
Titre Figure 3. Typology of monuments in central Senegambia
Légende (a) earthen tumuli (SEN 963); (b) megalith circle (SEN 558); (c) stone circle (SEN 640); and (d) stone tumuli (SEN 962).
Crédits Photos by A. Athie (2024).
URL http://journals.openedition.org/afriques/docannexe/image/5785/img-3.jpg
Fichier image/jpeg, 779k
Titre Figure 4. Map of sites recorded in Inventaire for the Megalith Zone of central Senegambia
Crédits V. Martin, C. Becker, 1984, carte 4a-b.
URL http://journals.openedition.org/afriques/docannexe/image/5785/img-4.jpg
Fichier image/jpeg, 185k
Titre Figure 5. Map of central Senegambia showing megalithic sites recorded in the Inventaire and areas re-surveyed by later projects
Crédits Illustration by C. Gokee (2025).
URL http://journals.openedition.org/afriques/docannexe/image/5785/img-5.jpg
Fichier image/jpeg, 632k
Titre Figure 6. Spatial distributions of the monument types and architectural features recorded in the Inventaire
Crédits Illustration by C. Gokee (2025).
URL http://journals.openedition.org/afriques/docannexe/image/5785/img-6.jpg
Fichier image/jpeg, 981k
Titre Figure 7. Counts for (a) mortuary monuments and (b) laterite monoliths at sites recorded in the Inventaire
Crédits Illustration by C. Gokee (2025).
URL http://journals.openedition.org/afriques/docannexe/image/5785/img-7.jpg
Fichier image/jpeg, 79k
Titre Figure 8. Surface models (2 km x 2 km) interpolated from monument type counts recorded in the Inventaire
Crédits Illustration by C. Gokee (2025).
URL http://journals.openedition.org/afriques/docannexe/image/5785/img-8.jpg
Fichier image/jpeg, 474k
Titre Figure 9. Composite RGB image for the first three principal components from the PCA for surface models of monument type counts
Crédits Illustration by C. Gokee (2025).
URL http://journals.openedition.org/afriques/docannexe/image/5785/img-9.jpg
Fichier image/jpeg, 477k
Titre Figure 10. Composite RGB image for surface models of monument type counts
Crédits Illustration by C. Gokee (2025).
URL http://journals.openedition.org/afriques/docannexe/image/5785/img-10.jpg
Fichier image/jpeg, 508k
Haut de page

Pour citer cet article

Référence électronique

Cameron Gokee, Adama Harouna Athie, Ibrahima Thiaw et Assane Ndiaye, « Remapping the Senegambian megaliths: Beyond the culture area paradigm »Afriques [En ligne], 16 | 2025, mis en ligne le 27 mars 2026, consulté le 14 mai 2026. URL : http://journals.openedition.org/afriques/5785 ; DOI : https://doi.org/10.4000/161vx

Haut de page

Auteurs

Cameron Gokee

Department of Anthropology, Appalachian State University, USA

Adama Harouna Athie

Institut fondamental d’Afrique noire, URICA/Laboratoire d’archéologie, Université Cheikh Anta Diop, Sénégal

Ibrahima Thiaw

Institut fondamental d’Afrique noire, URICA/Laboratoire d’archéologie, Université Cheikh Anta Diop, Sénégal

Assane Ndiaye

Institut fondamental d’Afrique noire, URICA/Laboratoire d’archéologie, Université Cheikh Anta Diop, Sénégal

Haut de page

Droits d’auteur

CC-BY-NC-ND-4.0

Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont susceptibles d’être soumis à des autorisations d’usage spécifiques.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search