This neatly presented book has a clear argument: that the material evidence of eight building complexes employed by other scholars to study slave sales in the Graeco-Roman world has, in fact, little to do with them. It understands itself as a response to work carried out by a number of Italian archaeologists—Braconi, Coarelli, Cocco, Fentress (whom I count as an honorary Italian)—with a focus on a sub-set of papers published in 2005 in the Journal of Roman Archaeology by Braconi, Coarelli and Fentress under the title ‘Selling people: five papers on Roman slave-traders and the buildings they used’.1 In what she calls ‘a critical reassessment’, Monika Trümper reviews the evidence of the Agora des Italiens at Delos, the Building of Eumachia in Pompeii, the Crypta Balbi in Rome, the Tempio Rotondo in Ostia, the Basilica in Herculaneum, the Chalcidicum in Lepcis Magna, the Prytaneion in Magnesia on the Maeander, and the Serapeum in Ephesus. For each of these, Trümper offers an identification other than that of a slave market, championing a porticus in all cases. The negative conclusion on the buildings’ use as slave markets, based on a negation of the concept of a slave market as a building type, is aided by a comparison of the ancient evidence with an idea of what a slave market is or, rather, was. That idea, evidently, could not be got from the ancient evidence, and Trümper therefore makes use of what she calls ‘a cross-cultural perspective’: a chapter of twenty pages in which she summarises knowledge on five slave markets known from more recent times (Istanbul, Cairo, Marrakesh, Havana, and places in the United States of America).
Few would disagree, I think, with Trümper’s contention that the evidence adduced by the scholars named for their identification of slave markets is not watertight—least of all some of the scholars subject to Trümper’s pen. Elizabeth Fentress, for instance, states in her introduction to the papers on ‘Selling people’ that ‘we cannot prove any of it’ (180); but given that her goal is ‘to re-open for discussion the subject of Roman slave markets’ because ‘it is high time we started looking [for slave markets]’, this seems actually unproblematic. The gap between the criticism launched by Trümper and the aims and arguments of those she takes to task is already brought out on page 1 of her study: there, we read that ‘the identification of ancient monuments as slave markets depends upon the assumption that buildings were designed specifically for use as slave markets’. This definition is produced by Trümper, and it is not shared by all of those she criticises. Moreover, we are not told why only a clearly defined building type could be linked archaeologically with slave sales. Trümper is not unaware of the gap: in a footnote on page 33, she cites Fentress’s statement that ‘the fact that the building could function as a market for slaves hardly implies that this was its only function’, and she states more generally in the text that ‘a certain multifunctionality or flexibility in use is conceded to most potential slave markets that would have been used for other purposes when no slaves were being sold’ (33). She contends however that ‘it is nevertheless assumed that the buildings were foremost constructed to serve as slave markets’ (33-34). There is no footnote to corroborate her contention, whilst the title of the papers she aims at speaks merely of slave traders ‘and the buildings they used’. And if one actually reads those papers, a range of approaches and assumptions comes to light: only Filippo Coarelli speaks clearly of ‘un tipo architettonico’ (212) in his discussion of the Agora des Italiens; yet, his aim is ‘continuare la discussione’ (197), not to provide the final word. I would myself not believe for a moment that the construction of this building can be rendered intelligible through reference to slave uprisings in the 130s, not least because I do not regard the Sicilian Insurrection of the 130s as a slave rebellion, but particularly because I think the identification of architectural structures on the basis of historical frameworks unsatisfactory. Yet, I have no general problem with the idea that the building could have been used for slave sales for at least some of the reasons put forward by Coarelli—whatever formal building type we wish to see in it. Concerning the contributions by Braconi and Fentress, the hunt for a specialised building is in any case soon over: Braconi states in his short paper on the chalcidicum at Lepcis Magna that ‘mi pare possibile candidarlo anche alla funziona di mercato di schiavi’, not least because ‘la grande piazza porticata poteva costituire un ottimo luogo di raccolta di merce (anche) umana’ (219). Anche! Similarly, Fentress speaks in her discussion of the Eumachia building at Pompeii of ‘the building’s varied focus’ and ‘its multiple rôle’ (229) and says that ‘the peculiar features of the Eumachia building are nowhere equalled’ (230): no building type here. And whilst Fentress implies that building consent may have been granted specifically with slave sales in mind, at no point does she suggest that this was the building’s only or primary function: in fact, she goes to town to labour the multi-functionality of this and other buildings under review, and she emphasises that these may largely have been used for activities other than slave sales. Thus, whilst she is keen to understand more clearly the ‘ specificity of the chalcidicum ’ (222), she speaks quite loosely of its ‘use for auction-blocks’ (232); and ‘a need for clearly-defined spaces’ for slave sales (232-233) is explored in combination with other uses. The building of Eumachia is for Fentress the result of the ‘construction of a building that combined a commercial and an imperial function’ (229): purpose-built, yes, albeit for a range of functions. Between Coarelli, Braconi and Fentress, there exist, then, important nuances in their study of buildings that might have served as slave markets, and not a single view of what a slave market was or how we might identify one in the archaeological record.
But the notion of the purpose-built slave market also sits uncomfortably in Trümper’s own exploration: for having defined a slave market thus, she labels the half a dozen or so modern structures that provide her cross-cultural perspective as slave markets – although according to her own assessment the structures at Cairo, Marrakesh and New York were not built for this purpose. She concludes that ‘purpose-built slave markets were overall rare’ (15). Then, she argues, ‘taking the lowest common denominator of “modern” slave markets into account, one would expect ancient equivalents to be fully integrated, closed complexes for a fairly regular, year round trade’ (19), listing thereafter the necessary structures, e.g. courtyards, or showrooms. Having made little attempt to justify her selection of the modern comparanda that produced this lowest common denominator (Istanbul yes, New York no), Trümper also does not explain why we should think slave trading to have attracted similar ‘material’ responses in societies that are vastly different otherwise. But without addressing these kinds of questions it is not clear what kind of comparative insights the modern data could bring and what relevance the evidence for a modern society having ‘constructed clearly identifiable purpose-built slave markets on a regular basis’ (1) (or not, as the case may be) should have for the ancient world. The posited structural analogy between modern and ancient thus finds a justification in building the target that Trümper aims to dismantle: for whenever the ancient buildings lack the postulated criteria, they cannot be identified as slave markets. In the context of a critique that thrives on the exposure of differences, it is perhaps odd to conclude, then, that different places ventured different ‘material’ solutions for slave sales (83-84).
In her final chapter, Trümper sums up her critique: the lack of a comprehensive discussion of all remains, selective citation of published work, a lack of attention to the buildings’ history, and disregard for due contextualisation, local and cross-regional (75). One presumes that an example of good practice is constituted by Trümper’s monograph on the ‘Agora des Italiens’ of 2008, based on her 2004 Habilitationsschrift, without which much of Graeco-Roman Slave Markets is not intelligible.2 In contrast, the Italian phalanx has failed to identify ‘the building type of a Graeco-Roman purpose-built slave market with specific characteristics’ (83). Yes, perhaps. But the reason is that they were not necessarily looking for a Bautypus, but that they primarily tried to answer questions, each in their own way. In doing so, they have not provided what Trümper calls ‘truly conclusive evidence’ (83); then again, they were fully aware that they ‘cannot prove any of it’. Being archaeologists, they may moreover feel perplexed by the suggestion that such evidence would ideally be found in the form of dedicatory inscriptions, i.e. in a text (83). All the same, some of Trümper’s reservations are anything but conclusive: that, for instance, slave selling would not have been possible at the chalcidicum in Herculaneum because it ‘would have seriously impeded traffic’ (61) defies just about everything we know about the workings of markets; and the silence over the hypothetical nature of ‘the passageway in the rear wall of the propylon ’ (or chalcidicum ?) at Delos which is central to her identification of the building as a porticus (43; cf. Fig 11 and 14, and 2008: 21-33 and 51-52) is unhappy in a work that demands of others—and promises in turn—‘hard proof’ (40); and what ‘a concentration camp-like slave market’ (44) is, I do not know.
It is not just that Graeco-Roman Slave Markets has no positive argument of its own that makes this slim book on an important aspect of Roman slavery a fairly disappointing read. The entirely negative take on the attempt of others to address a significant gap in our knowledge of Roman slavery is not stimulating. That one or other of Trümper’s Italian targets may have operated with a more flexible definition of a building type than Trümper, failing to demonstrate sufficient Gründlichkeit in their discussions of potential slave markets in what are, after all, fairly short contributions, is well possible. But it is certainly impossible that we may arrive at better interpretations unless ‘we started looking’. Trümper, on the other hand, has got us no closer to understanding where and how the Romans traded slaves. Instead, we have been provided with a list of holes in the arguments of others, and a ‘serious’ list at that: one cannot escape the feeling that the odd dose of Italian fantasia is being met with a good helping of German Ernst. The focus on architectural arguments might moreover have been made plain through a sub-title, e.g. A typological study. As it stands, Graeco-Roman Slave Markets gives the impression of contributing to a historical debate—but the question of the building type is actually marginal for the exploration of ‘where and how the many slaves that lived and worked in the private and public spheres were traded’ (1).
So, there is enough here that would make for a decent review article, foregrounding for discussion questions of method and goal, ideally combined with suggestions as to where to go next. But no one would believe all and every flight of fancy that scholarship is able to produce in the first place, and to pack one’s frustration between two hard covers is not an obvious solution. Perhaps, such doubt as that behind Graeco-Roman Slave Markets is in any case better expressed over a glass of Torgiano in the course of a nice pranzo at the Caffè di Perugia than on 84 much pricier pages of text. And with a bit of flexibility of mind, one might even conceive of less rigid conceptions of building types and their uses: perhaps, the Romans could think of a porticus also as a slave market—without the need for architectural separation of functions?
1. JRA 18 (2005), 180-240. The papers sprang from a 2001 conference at the British School at Rome, and include also contributions by John Bodel on slave traders, and Giuseppe Pucci on the iconography of slave sales.
2. M. Trümper, Die ‘Agora des Italiens’ in Delos. Baugeschichte, Architektur, Ausstattung und Funktion einer späthellenistischen Porticus-Anlage (2008).