All Souls College

The monuments with which this paper is concerned are a group of votive stelai from Tunisia and Algeria dedicated to Baal / Saturn. This paper analyses their formal composition and examines the impact of Rome on the iconography of the cult. Although aspects of the stelai change, with the replacement of neo-Punic inscriptions of the pre-Roman Libyco-Punic phases by Latin ones in the Roman period, and the introduction of Roman style clothes for the worshipping figures who were formerly represented naked, in some parts of Numidia the same elements of ritual equipment and votive offerings persist throughout. So too does the basic composition and the simplistic, naïve style of representation. The cult appealed predominantly to the lower strata of society, and the stelai suggest that at this social level Romanization was limited, affecting clothes and language much more than cult systems and belief. Elsewhere, in the more Romanised coloniae of Numidia and the highly urbanised regions of Africa Proconsularis, there is a much greater degree of cultural borrowing from Roman art and religious practice.

This volume contains seventeen papers by leading Roman historians and archaeologists, discussing approaches to and methods of analysing the performance of the economy of the Mediterranean world under Roman imperial rule in the period c. 100 BC to AD 350. An introductory chapter defines the constituents and characteristics of economic integration, growth and decline in the Roman economy, and the ways in which quantifiable and proxy data can be used to measure them. This is followed by analyses of approaches in specific economic sectors: demography, urbanisation and settlement patterns, the agrarian economy, patterns of trade and commerce, metal supply and coinage. The volume offers a comprehensive survey of the opportunities for advancing understanding of the economic and technological development of the Mediterranean world under Roman rule by combining old and new evidence from archaeological and documentary sources.

The South Etruria survey results indicated a much denser concentration of settlement in the southern part of the survey area, closer to Rome. It seems clear that the proximity of Rome encouraged settlement here, and it is frequently assumed that farms and villas in this area were engaged in growing fruit and vegetables to serve the markets of Rome (Potter, Changing Landscape; Morley, Metropolis and Hinterland). However likely this is, until now it has remained largely an assumption, supported in part by literary sources which relate more to the south and east of Rome (Tibur etc.) than to the Tiber Valley to the north; and by theoretical models. There is however a mass of data in the Tiber Valley survey archives and in published material (Forma Italiae etc.) which can be used to ground this analysis in the archaeology. Irrigation and drainage schemes demonstrate investment in land improvement for some form of agriculture, and one class of structures provides a clue as to the kinds of products grown. A number of rural sites possess large cisterns (200-700 m3) far exceeding the water-supply needs of the farm or villa, and in many cases these are not associated with villa baths. Such cisterns are most plausibly interpreted as being for irrigation: a study carried out on similar hydraulic features to the east of Rome (Thomas and Wilson, PBSR 1994), using software to estimate irrigation needs for different crops in various kinds of soils, suggested that the capacity of such cisterns was insufficient for irrigating significant areas of wheat or vines, but would have suited the irrigation needs of vegetable gardens or fruit orchards. The scant epigraphic evidence seems also to support the use of cisterns of this type for irrigated horticulture. We have here a suggestive indicator to support the hypothesis of settlement in the immediate vicinity of Rome being geared to growing perishable crops which could not be transported too far to market, and indications of substantial investment in hydraulic infrastructure to support this irrigated horticulture.

I risultati del South Etruria survey delineano una maggiore concentrazione di insediamenti nel settore meridionale, quello più vicino a Roma, dell’area ricognita. Sembra chiaro che la vicinanza a Roma favorì lo sviluppo insediativo, ed è stato spesso sottolineato che le fattorie e le ville in quest’area erano impegnate a rifornire di frutta e ortaggi i mercati di Roma (Potter, Changing Landscape; Morley, Metropolis and Hinterland). Per quanto verosimile, questo assunto è rimasto fino ad ora un’ipotesi, supportata in parte - oltre che da modelli teorici - dalle fonti letterarie che però riguardano essenzialmente le aree a sud e a est di Roma (Tibur etc.) e non la valle del Tevere. Esiste comunque una massa di dati negli archivi del Tiber valley survey e nelle pubblicazioni (Forma Italiae etc.) che può essere utilizzata per supportare a livello archeologico tale analisi. Gli impianti di irrigazione e di drenaggio indicano investimenti volti a migliorare le potenzialità agricole del territorio, ed una tipologia di strutture sembra costituire lo specchio del tipo di sfruttamento del suolo. Un certo numero di siti rurali possiede grandi cisterne (200-700 m3) assolutamente sproporzionate per i bisogni di una fattoria o di una villa, ed in molti casi esse non sono associate ad impianti termali. Tali cisterne dovevano dunque essere utilizzate per l’irrigazione: uno studio di questo tipo su simili apprestamenti idraulici ad est di Roma (Thomas and Wilson, PBSR 1994), condotto utilizzando stime di calcolo per il fabbisogno d’acqua dei vari tipi di coltivazioni e di suoli, suggerisce che la capacità di queste cisterne era insufficiente per irrigare vaste coltivazioni di grano o vigneti, ma avrebbe potuto supplire ai bisogni di orti e frutteti. La scarsa documentazione epigrafica sembra anch’essa confermare l’uso di questo tipo di cisterne per l’orticoltura. Abbiamo quindi un importante indicatore a sostegno dell’ipotesi che le tenute agricole vicine a Roma erano destinate alla coltivazione di prodotti altamente deperibili che non potevano essere trasportati da troppo lontano, ed indicazioni di consistenti investimenti verso le infrastrutture idrauliche destinate a supportare tali forme di orticoltura.

This article presents a previously unpublished Roman travertine relief showing scenes of breadmaking, currently in the restaurant Romolo in Trastevere in Rome. It presumably came originally from a tomb monument, possibly in the vicinity, and might be dated on grounds of material and style anywhere between the very late Republic and the Flavian period. From left to right it shows two men delivering of sacks of grain, a man loading grain into an animal-driven mill, three men keading dough by hand, three more shaping loaves, and one putting loaves into the oven. The article discusses parallels in other reliefs of bakery scenes, and highlights the importance of this one for the evidence that it provides for the extent of division of labour in a fairly large-scale bakery, in which the breadmaking process is divided into stages each carried out by different groups of people.

Excavations conducted by the Society for Libyan Studies since 1999 at the Greek city of Euesperides (Benghazi, Libya), directed by Paul Bennett, Ahmed Buzaian and Andrew Wilson, have been investigating evidence for purple dye production from Murex trunculus shellfish. Numerous spreads of crushed shell waste have been recorded, and one of these has been excavated. Two phases of dye production activity were revealed, both dating to the first half of the third century B.C. Dye preparation had taken place over large mud brick hearth surfaces in the courtyard of a former house. There was little trace of permanent infrastructure apart from the hearth surfaces, and the archaeologically diagnostic features of this production are large deposits of crushed Murex shell and ash. It is likely that the dyestuff was heated in portable vessels (probably metal) mounted on stands or tripods over the hearths; the crushed shell was dumped among the embers of the fires afterwards. No built vats or tanks were found.
Studies on the shell species and breakage patterns by Estíbaliz Tébar Megías provide insights into the methods used for collecting and preparing the Murex shells, and forms the basis for a preliminary quantification of production. It is likely that the manufacture of purple dye at Euesperides was linked also to wool production; and that wool or textiles were among the goods exported in return for the large amount of imported trade goods found at the site.

This paper discusses the problems of trying to reconstruct ancient populations from surface survey data. It builds on recently suggested methods for other areas of Italy and applies them to the Biferno Valley to try and model plausible population numbers for the region in the Roman period. Important points concerning the feasibility and trustworthiness of such an exercise include factors of survey design and methodology, and particularly survey recovery rates. Possible ranges of population estimates are discussed for the Early and Late Roman periods with reference to different types of sites.

"This paper examines a group of workshops revealed by clearance excavations in the 1920s-1940s at the Roman city of Sabratha in Libya. The excavations of the domestic and commercial quarters around the forum were never published, but among the remains still visible at the site are at least 18 workshops characterised by groups of 2, 3 or 4 vats whose size and arrangement closely recalls those at known fish-salting factories along the Tunisian coast. It is proposed that the workshops at Sabratha are also for salting fish, or for garum production, or both. The workshops have surprisingly little space for preparation or other activities, and in this respect are closer to the small fish-salting workshops recently discovered at Pompeii than to the large factories of the Straits of Gibraltar.
This paper presents the evidence for the workshops, and discusses the reasons for their interpretation as fish-salting establishments, and their possible date. Some reflections are offered on the possible capacities of the workshops, and the nature of their products. The wider significance of the Sabratha fish-salting establishments is that, like the larger workshops of Baelo Claudia, Pompeii, Tyritake and Chersonesos, they provide further evidence for urban-based production of salted fish products."

This paper proposes a new model for the diffusion of foggara-based irrigation across the Sahara in ancient and medieval times. Recent fieldwork by the Fazzan Project has established that the foggaras of the Wadi al-Ajal are of Garamantian origin (last centuries BC / early centuries AD), and appear to have been used perhaps until the early middle ages (ninth to eleventh centuries), but probably not beyond this. Abandonment of some of the foggaras may even have begun as early as the fourth century AD. It is argued that foggara irrigation technology was introduced from Egypt in the second half of the first millennium BC, and enabled the development of a Garamantian agricultural society in Fazzan, which controlled trans-Saharan trade. From Fazzan, foggaras spread north to the fringes of the Garamantian world, to southern Tunisia and the southern Aurès in the Roman period.
An apparent collapse in north to south trans-Saharan trade in the late Roman period (fourth/fifth centuries AD), linked to the decline of the Tripolitanian coastal cities, weakened the Garamantian state, not least by affecting supplies of slaves used in foggara construction and maintenance. Coupled with a declining water table, problems of labour and maintenance gradually led to the abandonment of most foggaras, and a shift to smaller-scale agriculture supported by wells. Meanwhile, however, development of oasis zones in the west central Sahara (especially the Touat and Gourara) facilitated the subsequent development of new north to south trade routes through what is now the Algerian Sahara. These oases are today the zones of most highly developed foggara use anywhere outside Iran; local tradition puts foggara use here as early as the eleventh century AD, and there are grounds for thinking it may go back to the seventh century if not earlier. There are strong similarities, in construction and nomenclature, between the foggaras of Fazzan and those of the west central Sahara, and it is most likely that the foggaras of the Touat and Gourara were introduced from Fazzan. From these oases the foggara subsequently spread to the Tidikelt, Tafilelt and Figuig. Fazzan played a pivotal role in the spread of foggara irrigation across the Sahara; not only did the technology spread along trans-Saharan trade routes, but it enabled the development of oases as trading centres. As such, the history of the foggara in the Sahara is inseparable from the history of trans-Saharan trade.

Cette communication presente les resultats des fouilles menées par la Society for Libyan Studies à la cité grecque d’ Euesperides (Benghazi, Libye) entre 1999-2003. Les fouilles ont découvertes une partie des défenses de la cité, avec des maisons construites à interieur de la muraille, et aussi un bâtiment de rang social élevé, avec une mosaïque hellénistique faite en technique mixte des tesserae et des cailloux. Cet bâtiment a remplacé une maison avec un andron dont le sol était en mosaïque de cailloux. La datation du destruction de la maison et sa remplacement par le bâtiment avec les mosaïques en technique mixte est assez précise, grace à la découverte des monnaies qui donnent un terminus post quem de 261 avant J.-C., et un terminus ante quem de 250 avant J.-C. est donné par l’abandon du site après le mort du roi Magas. Le reconstruction des maisons et bâtiments si proche à l’abandon du site suggère que la citè a été abandonée non par fur et mesure parceque le port était en train de s’ensabler, main tout à fait soudain, probablement comme transfert de la population par la reine Berenice, à un nouveau site qui s’appellait Berenice.

La quantification des trouvailles céramiques montre que Euesperides se trouvait dans un cadre des routes commerciales qui traversaient la Méditerranée. La 90% pourcent des céramiques fines etaient importées, dont une groupe importante de céramique commune Punique. Les amphores pour vin, huiles et poissons salées viennent de toutes régions circum-Mediterranéens, des Etroits de Gibraltar à l’Égéen du nord. Parmi les exportations de Euesperides étaient probablement le silphium, et des textiles ou de laine tintés en pourpre, comme indique la fouille d’un atelier de la 3ème siècle avant J.-C. qui produisait la teinture de pourpre du Murex, et aussi l’identification des grands dépotoirs des Murex cassés des ateliers divers.

Relatively little attention has been paid to the archaeology of textile productionin North Africa. The purpose of this paper is to review the evidence for the production and dyeing of textiles, to assess how we may distinguish between the remains of fulling establishments and dyeworks, and between different kinds of dyeing; and to see what can be deduced about the scale and organisation of production. It must be stressed that such a picture will necessarily remain very sketchy until more sites have been excavated with a view to answering these sorts of questions. Because minor finds such as spindlewhorls, loomweights and needles from North African sites are rarely and poorly published, I shall focus on the evidence for physical plant and infrastructure, which is necessarily related primarily to the stages of fulling and dyeing. This is essentially an archaeology of soaking and trampling; both fulling and dyeing require vats and tubs. Fulling involves trampling cloth in tubs in a detergent solution to remove the lanolin from the fibres, followed by rinsing in vats, and then drying, carding, bleaching and pressing. The vats and tubs have no need to be heated. Dyeing, however, may well involve heating the solution, and may not have involved treading or trampling. Sites discussed in this paper include Timgad, Thuburbo Maius, Tiddis, Cuicul (Djemila) Hippo Regius, Meninx, and Berenice (Benghazi).