Bryn Mawr Classical Review

Bryn Mawr Classical Review 2012.05.30

Joann Freed, Bringing Carthage Home: the Excavations of Nathan Davis, 1856-1859. University of British Columbia studies in the ancient world, 2.   Oxford; Oakville, CT:  Oxbow Books for the Department of Classical, Near Eastern and Religious Studies, University of British Columbia, 2011.  Pp. 264.  ISBN 9781842179925.  $96.00.  

Reviewed by Matthew M. McCarty, Yale University (

In 1856, responding to a petition from a little-known Anglican missionary, the British Foreign Office decided to fund the first major, long-term archaeological work at Carthage. The missionary, Nathan Davis, sought to uncover Punic artifacts to populate the halls of the British Museum. Instead, his excavations at Carthage, Gammarth, and Utica produced almost exclusively Roman and late antique material, including an important series of mosaic floors that he lifted and sent to London. Although he published a popular account of his work, Davis’ excavations have largely been effaced from the historiographic and topographic record of Carthage; Freed’s monograph represents the first attempt to systematically reconstruct Davis’ excavations, offering a lively narrative account of Davis’ work. Although the book is as much an attempt to rehabilitate Davis’ (often deservedly) poor reputation as it is an account of his excavations, overall Freed offers not only a rich and carefully researched description of Davis’ campaigns in their social, political and personal contexts, but also important new data for better understanding the topography of Roman and late antique Carthage. Thanks to Freed’s monograph, it is no longer possible to deny the significance of Davis’ contributions or to despair wholly at the impossibility of siting his finds within the ancient city.

Davis has been largely written out of the historiography of research at Carthage, due at least partially to the two conflicts that frame Freed’s account: between national traditions and between amateur and professional archaeology. Davis’ plundering has been compared unfavorably to the more “scientific” approaches of his contemporaries, and given the later predominance of French scholarship on Carthage, his early excavations are generally either ignored or quickly dismissed. In Clementine Gutron’s magisterial account of nineteenth- and twentieth-century archaeology in Tunisia (not cited by Freed), for example, only once does Davis emerge into the body of her text from brief mentions in the footnotes: as playing host to Flaubert in spring of 1858.1

As for the historiographic aspects of Freed’s monograph, the author resists delving into postcolonial theory, focusing instead upon detailed description of Davis’ circumstances and his concrete archaeological contributions: the results of his excavations and his technique for lifting mosaics. What is clear from Freed’s account is that the history of European archaeology in North Africa cannot be told solely—or even primarily—as a series of succeeding or competing -isms (imperialism, colonialism, Orientalism, nationalism, postcolonialism), but instead must be recounted in terms of individual historical agents, their experiences, their social milieux and networks, and particular circumstances (including overcrowding at the British Museum). While competition between the British and French for influence in Tunisia may have partially driven the Foreign Office to invest directly in the excavation, equally important were personal friendships (Davis’ ties to Bey Mohammed and his court, which secured permission for his work), acquaintances (the European community in Tunisia), and rivalries (especially with Charles-Ernest Beulé, professor of archaeology at the Bibliothèque Impériale), and Freed’s narrative rightly brings these factors to the fore. Given the confused (and sometimes contradictory) locations of trenches recorded by Davis and visitors to his excavations, the Herculean effort of attempting to place Daviss’ digs in a systematic way has rarely been attempted (save in a limited manner by Paul Gauckler, to whose errors Freed devotes a short appendix). Freed locates Davis’s sites primarily by overlaying Daviss’ published map with a plan of the Roman street grid (fig. 4.12) and cross- checking the result against Daviss’ descriptions, Daviss’ hitherto unpublished sketch-plans, the surveyed points on Falbe’s 1833 map of Carthage, and features the locations of which are assured by subsequent work. Freed’s proposals for locating Daviss’ sites in relation to the Roman street grid are appended at the end of this review to supplement Freed’s table on p. 82, which offers a concordance of Davis’ different numbering systems and Falbe’s map points, and her fig. 14.2, which focuses on the locations of the mosaics.

Despite Freed’s careful method, though, there remain the cartographic problems inherent in working from points that were set by eye rather than by survey. While it may be attractive to seek near-exact placement of Daviss’ finds, Freed is overly optimistic in her ability to accurately site Daviss’ digs. While she acknowledges that points on Daviss’ map can frequently be between 30-50m off (p. 39: hardly an insignificant distance!), this does little to shake her confidence in mapping his sites, resulting in a precision beyond what is possible from Davis’ records. Take, for example, Davis’ Site 35, his final excavation (admittedly, as with all of the work he conducted in 1859, among the least well and most confusingly recorded), where Davis found the mosaic of Nereids and Tritons: Freed’s overlay and final map (figs. 4.12 and 14.2) place the site near Decumani 5-6S and Cardo 1W (mistakenly called 1E in Freed’s text, p. 195), a location she accepts based on its proximity to the large hydraulic complex known as the Turris Aquaria, for which the marine themes of the mosaic seem suited. Freed even suggests the mosaic could have paved a forecourt to the complex. Davis’ written account, however, explicitly places the site between two features surveyed onto Bordy’s 1897 French army map of Carthage: the village of Douar ech-Chott and Carthage’s circular harbor. The mosaic must thus be sited east of the village in a field adjacent to the cemetery northeast of the village. When the street grid is overlaid onto Bordy’s map, the Cardo Maximus cuts through the center of the village, and thus, if we trust Davis’ written description over his imprecise map, the mosaic must have been found E of the Cardo Maximus. This also moves the mosaic several blocks away from the Turris Aquaria (Cardo 2W), and makes an association between the two monuments highly unlikely. Here, as with several other points, Freed is far too optimistic about how precisely it is possible to locate Davis’ finds and associate them with known archaeological features in Roman Carthage. Caution is thus necessary before fully accepting her hypothesized placements.

Indeed, meaningful re-contextualization of Davis’ finds is only possible when coupled with later excavations in the same areas, and it is here that Freed’s account is most productive: in particular, regarding a large residential complex furnished with the Mosaic of Months and Seasons (Site 8, Davis’ first mosaic find), and the Punic votive stelae from Davis’ so-called “Punic mine” (Site 15). For the former, using a 30m-offset based on Davis’ mis-plotted nearby Circular Monument, Freed suggests that the mosaic comes from immediately adjacent to the triclinium of the Maison des Chevaux, excavated in the 1960s. Given the size of this house and its association with a large columned hall to the south (the so-called Baths of Gargilius), Freed suggests that the entire complex may have served as the proconsul’s palace. Again, given the uncertainties tied to Davis’ eyeballed cartography, expecting a uniform offset for the points he plotted may be overly optimistic. Nevertheless, placing the mosaic conclusively in this area does attest the continued importance and wealth of the neighborhood on the southwest slope of the Odeon Hill in the fourth century. For the stelae from Site 15, Freed draws attention to the fact that their location and recorded context—built into a later wall (probably, as she rightly suggests, late antique)—closely match the context of thousands of stelae found by Sainte-Marie in 1874 In fact, Freed suggests that Sainte-Marie’s trench E may have been on the same site as Davis’ “Punic Mine.” At any rate, it is clear that Davis and Sainte-Marie mined the stelae from the same structure, and my own examination demonstrates that the dimensions and decoration of Davis’ stelae match Sainte-Marie’s almost exactly,2 confirming that they form a single group. The fate of the stelae between the abandonment of the tophet after Carthage’s destruction in 146 BCE and the late-antique reuse remains uncertain.

In spite of my few qualms regarding the interpretive limitations that Freed sometimes oversteps, her monograph represents a valuable exercise in archival archaeology, a strong complement to Gutron’s Franco-centric account of archaeology in Tunisia, an important contribution to the topography of Roman Carthage, and a solid model for future studies seeking to re-contextualize a range of materials from the Maghreb.

Site 1: three vaulted chambers (cisterns or supporting platform?): Dec. 4S, Card. Max.
Site 2: Dec. 2S, Card. Max.- 2E
Site 3: Dec. 5N, Card. 16E
Site 4: latrine of Antonine Baths: Dec. 3-4N, Card. 17-18E
Site 5: slope of Bordj-Djedid: Dec. 6-7N, Card. 16-18E
Site 6: Hill of Juno (further precision impossible)
Site 7: Dec. 5N, Card. Maximus
Site 8: Mosaic of Months and Seasons: Dec. 3-4N, Card. Max.-1W
Site 9: Circular Harbor: Dec. 4S, Card. 13-14E
Site 10: precision impossible, given very different locations on sketch-maps
Site 11: Dec. 6N, Cards. 16-20E
Site 12: just inside W wall of Antonine Baths
Site 13: Vandal Victory Mosaic: Dec. 8N, Card. 21E
Site 14: Mosaic of Baskets of Fish & Fruit: Dec. 2N, Card. 6E
Site 15: “Punic Mine”: Dec. 1-2N, Card. 5E
Site 16: “Carthaginian House”: Dec. 5-6N, Card. 2E
Sites 17-20: precision impossible and largely not discussed by Davis
Site 21: House E of Theater: Dec. 4-5N, Card. 7-8E
Site 22: House with Mosaic of the Gazelles: Dec. 3-4N, Card. 2-3E
Site 23: Circular Monument: Dec. 3-4N, Card. 2-3E
Site 24: Fallen column, perhaps re-excavated by Senay: Dec 3-4N, Card. Max.-1E
Site 25: Dec. 1S, Card. Max.
Sites 26-27: tombs along road between La Malga and Sidi Bou Said
Sites 28-29: villas at Gammarth
Sites 30-31: catacombs at Gammarth
Site 32: Byzantine mausoleum at Gammarth
Utica site A: between “Citadel” and House of the Cascade
Utica site B: house near forum
Site 33: Mosaic of Sirens: Card. 2W, Dec. 2N
Site 34: Vandal Hunt Mosaic: Dec. 3-4N, Card. 6W
Site 35: Mosaic of Nereids and Tritons: Dec. 5-6S, Card. 1W


1.   C. Gutron, L’archéologie en Tunisie (XIXe-XXe siècles). Jeux généalogiques sur l’Antiquité (Paris 2010), 201.
2.   Controlling for state of preservation of the British Museum stelae (BM) published by C. Mendleson (Catalogue of the Punic Stelae in the British Museum. London 2003) and using H. Bénichou-Safar’s data (“Les stèles «dites de Sainte-Marie» à Carthage,” in H. Devijver and E. Lipinski, eds, Punic Wars, 353-64. Leuven 1989) for the Sainte-Marie stelae (SM), the iconography is matched nearly perfectly (by no means true of all strata and sectors of the Salammbô tophet): the forearm-and-hand motif appears on 22% of BM and 25% of SM; the caduceus on 20% of BM and 22% of SM; the crescent-and-disc on 8% of SM and 6% of BM; and the “sign of Tanit” on 26% of BM and 33% of SM. The average dimensions differ slightly, though both fall within the standard deviation: 12.8cm width for BM and 13.7cm for SM; 7.8cm thickness for BM and 6.7cm for SM.
3.   G. Lapeyre, “Fouilles récentes à Carthage,” CRAI 1935: 82-4.

Comment on this review in the BMCR blog
Read Latest
Index for 2012
Change Greek Display
Books Available for Review
Bryn Mawr Classical Commentaries

BMCR, Bryn Mawr College, 101 N. Merion Ave., Bryn Mawr, PA 19010
HTML generated at 19:33:58, Wednesday, 16 May 2012