BMCR 2006.04.07

Le littoral de la Tunisie: Étude géoarchéologique et historique. Études d’Antiquités africaines

, Le littoral de la Tunisie : étude géoarchéologique et historique. Etudes d'Antiquités africaines,. Paris: CNRS éditions, 2004. 308 pages : illustrations, maps ; 28 cm.. ISBN 2271062608 €49.00.

How did people utilize coastal environments in antiquity? What evidence do we have for changes to the Mediterranean shoreline in the last 3,000 years? Did human activities or the climate cause greater modification to the central North African coastline in this time? A survey team combining geologists and archaeologists set out to answer these questions through an examination of the entire 1,300-km coastline of Tunisia beginning in 1987. Hédi Slim, Pol Trousset, Roland Paskoff, and Ameur Oueslati (hereafter the authors) recognized archaeological remains provided essential evidence for dating coastal changes, and a study of archaeological markers over such a long stretch of coastline had never been undertaken. They developed a balanced geoarchaeological approach which permitted the project to build a database related to both the coastal environment and human activities. Le littoral de la Tunisie is the final publication of this project, which will be familiar to specialists in this field from the dozen or so preliminary reports already available.1

This book consists of four sections: an introduction; a characterization of the southern, central, and northern coastal regions of Tunisia; a catalogue of the sites recorded; and broad conclusions about coastal geomorphology and human activities. The brief introduction mentions the main objectives and methodology of the survey and credits the members who participated during 10 years of fieldwork. It is followed by a discussion of the Tunisian coastline, broken into three regions. The analysis of each region contains a short description of its geomorphology, archaeological remains, and relevant ancient textual sources. The catalogue comprises a geomorphological and archaeological description of 210 sites beginning at the Libyan border and proceeding north and west to the Algerian border. It includes 49 sites involved in fishing or fish-processing in Tunisia, more than double the number previously known along the entire North African coast. Photographs, plans, sketches with architectural and/or geological features are frequently included and are of high quality. Surface pottery is mentioned briefly,2 and bibliographic information is included for published sites.

I have treated the first three sections of the book briefly in order to concentrate on the important conclusions in the final section, which have implications for those interested in issues of shoreline environments, sea-level fluctuations, Holocene geomorphology, climatic vs. anthropogenic change, quarrying, fishing and fish-products, trade, and the Roman economy. Here the authors have argued for a much greater role for the fishing activities in the economic growth of North Africa than previously recognized. They have also documented substantial changes throughout the entire shoreline of the country and determined (rather alarmingly) that 75 percent of the coast is currently retreating, while 20 percent remains stable and only 5 percent is advancing.

The main contribution of the archaeological members of the project is the discovery of evidence for fish-processing industries. Prior to the survey of the Tunisian coastline, ‘factory’ sites with more than 10 tanks had been identified in the country at just 2 locations, Neapolis and Sullecthum. Le littoral de la Tunisie presents evidence for 12 ‘factories’ and 32 sites possessing a smaller number of tanks. One reason for the large increase in numbers is that tanks used to produce garum and salsamenta look similar to cisterns commonly found in baths, houses, and rural sites throughout Tunisia. The authors sketch out the distinctive criteria they use to identify fish-processing tanks: they were constructed in the ground, they did not have vaults, their exterior angles were reinforced to withstand the weight of the fish and salt within, and their interior walls were waterproofed and joined with quarter-round moldings. In addition, the authors claim that fish-processing sites share three common characteristics: a coastal location, a proximity to salt flats, and the availability of water. These guidelines should prove very useful for other researchers.

Other interesting results emerge from the discussion. Survey techniques proved very good at identifying fishing-related activities (an additional four sites contained fishponds and seven had deposits of murex trunculus shells for producing purple dye) but fishing per se was hard to identify from surface observation and only a very small number of bronze fishhooks or terracotta fishing weights were found. Ancient authors and the well-known mosaics with fishing scenes in the Bardo and Sousse Museums suggest fishing techniques probably practiced in antiquity (lines, casting nets, harpoons, and ‘lobster pots’), but ethnographic studies of modern Tunisian fishermen offer an even wider array of techniques not indicated by ancient sources, such as fixed nets, palm-frond traps, and barriers. Since the Greek and Roman authors who write about fishing rarely mentioned North Africa, confirmation for the use of these methods remains elusive.

Quarrying is the other major economic activity the authors consider. They discuss the exploitation of the two main types of sandstone found along the coast, the Cap Blanc and Rejiche formations. The former, on the northern coast and in the Gulf of Tunis, supplied building material for Carthage and Utica. The latter exists in the Sahel region and was used widely in the port towns of Horrea Caelia, Hadrumetum, Leptiminus, and Sullecthum. It was even transported from the coast 35 km inland to Thysdrus (El Jem) to build the famous amphitheater, the third-largest in the Roman world. The project’s investigations reveal that 18 of the 24 quarries investigated along the Tunisian coastline were partially or completely submerged. They thus provide good indications that the sea-level has risen in most areas during the last 3,000 years somewhere between 0.3 and 0.5 m, even if, the period of activity at the quarry could only be dated broadly.

Coastal issues are paramount in a country like Tunisia where the majority of the population, today and in the past, has settled near major ports. The authors conclude that there is no single explanation for the advance, loss, or stability along the shore mentioned above, but there are three main factors involved: sea-level fluctuation; shoreline displacement; and an “erosive crisis” at the end of antiquity. Visitors to both Sullecthum and Mnaka notice waves lapping at ancient walls, for example, but in the former case it is due to a general glacio-eustatic rise in sea-level since antiquity of 0.5 m and in the latter to the severe local wave action which has eroded the beach. At Carthage the situation is more complex. During antiquity the coastline was advancing due to alluvial material deposited in the Gulf of Tunis from the Medjerda and Miliana rivers. Thus, all material from the earliest phase of the city (8th century BCE) has been found more than 100 m behind the modern shoreline. Later Punic and Roman remains lie closer to the shore. Yet, after the 16th century CE when the tombolo connecting Tunis and Carthage had formed, the situation reversed, and today the remains of a Roman insula built in the 2nd century CE between Kardo XVIII and Kardo XIX lie under water. The placement of dams along the two rivers in the mid-20th century CE has prevented sediments from entering the Gulf of Tunis and exacerbated the problem of rising sea levels at Carthage and elsewhere in the Gulf. In the one major instance where the coast has advanced into the sea since antiquity, the Medjerda river delta, the former port town of Utica is now situated 12 km inland, and the former island containing Castra Corneliana (Kalaat el-Andaluss) has been joined to the mainland.

The primary question that emerges is: what role have humans played in these coastal changes? The authors argue that an increase in the quantity as well as the intensity and seasonality of rainfall was responsible for the destruction of slopes by runoff. This is the “erosive crisis” they place at the end of antiquity, in which readers will recognize shades of Claudio Vita-Finzi’s “Younger Fill”, which has itself stimulated much debate.3 The authors do not exclude the possibility that human actions caused some of the crisis too, but they point out that significant changes took place throughout the 1300-km study area and appeared in both densely and sparsely settled areas. Thus, they emphasize the role of a climatic shift in producing a regime of intense and heavy rainfall which brought severe erosion to dunes, shorelines, and terraces. This is a cautious but judicious approach to a problem with potentially far-reaching consequences, especially in this era of global warming. Will it stand up to scrutiny as others examine the results? I am not an expert who can judge the geological conclusions of the project, but while it seems to me that climate change may have caused widespread changes at the end of antiquity, the project downplays the effects of more localized human activities, both past and present. The development of agriculture in the fertile Tunisian Tell certainly increased sediments transported down the Medjerda and Miliana rivers, producing as much as 450 sq. km of new land near Utica as well as large deposits of sediment in the Gulf of Tunis. It seems logical to conclude that human activities, even if restricted in extent, have had major consequences along the Tunisian coastline.

For now, this book stands as one of the best treatments of a large section of the Mediterranean coastline. In North African archaeology, it represents one of the few times that archaeologists and geomorphologists have collaborated; the effectiveness of the partnership leads one to wonder why more such studies are not underway.4 It should be influential in stimulating further research, which ought to proceed using more intensive methodologies to attack smaller geographical areas and more focused questions.5 From a broader perspective, the authors have clearly shown that fish-products deserve consideration with grain, olive oil, and African Red Slip ware as the primary exports from Africa Proconsularis, and that this province now must be regarded as a major source of garum and salsamenta alongside Spain and Morocco. They do not engage with wider debates about trade or growth in the Roman economy, however, although this would have been welcome. It would not be surprising to see results from Le littoral de la Tunisie appearing in these discussions soon.


1. For previously published overviews of the project, see M. Bonifay, A. Oueslati, R. Paskoff, H. Slim, and P. Trousset, “Programme tuniso-français d’étude du littoral de la Tunisie: bilan des travaux 1987-1990,” Bulletin de l’Institut National du Patrimoine 1 (1990) [1992]: 95-116; R. Paskoff, H. Slim, and P. Trousset, “Le littoral de la Tunisie dans l’antiquité: cinq ans de recherches géoarchéologiques,” Comptes Rendus à l’Academie des Inscriptions et Belles-Lettres (1991): 515-46. For references to other preliminary reports, see Le littoral de la Tunisie, pages 253-254, 294-297.

2. A chronological summary of the pottery results for each site is included at pages 223-226. Le littoral de la Tunisie does not include a full discussion of the pottery from these sites which has been published separately as M. Bonifay, C. Capelli, T. Martin, M. Picon, and L. Vallauri, “Le littoral de la Tunisie: étude géoarchéologique et historique (1987-1997). La céramique,” Antiquités africaines 38-39 (2002-2003) [2005]: 125-202. Specialists will undoubtedly wish to consult this article at the same time as they read Le littoral de la Tunisie. It is unfortunate that the pottery was not published with the rest of the survey results. As I have learned while preparing this review, it may prove difficult to acquire the publication quickly if your institution does not subscribe to Antiquités africaines.

3. C. Vita-Finzi. The Mediterranean valleys: geological changes in historical times. (Cambridge 1969).

4. The other notable example of such collaboration is G. W. W. Barker, D. D. Gilbertson, G. D. B. Jones, and D. J. Mattingly, eds. Farming the desert: the UNESCO Libyan Valleys Archaeological Survey. Volume 1. Synthesis; Volume 2 Gazetteer and Pottery (Paris and London 1996).

5. The methodology of the survey is covered very briefly in the book, and it is not really possible to determine how the authors located sites. Survey techniques employing closely-spaced fieldwalkers would likely yield more sites, and also smaller prehistoric sites, which were not the focus of the authors in this project, however. What is needed above all is excavation of fish-processing sites, as the authors have undertaken at Neapolis. Their preliminary report is: L. Slim, M. Bonifay, and P. Trousset, “L’usine de salaison de Neapolis (Nabeul), premiers résultants des fouilles 1995-1998,” Africa 17 (1999) 153-197.