Version classiqueVersion mobile

Fish & Ships

 | 
Emmanuel Botte
, 
Victoria Leitch

Seconde partie. Études archéologiques sur l'Afrique

Ancient marine resource exploitation in the coastal strip of Cyrenaica: Some evidence of fish-related industry in Cyrenaica?

Mohamed A Hesein

Texte intégral

Introduction

  • 1 Wilson 2004; Buzaian 2009. Oil and wine presses in Cyrenaica are currently being surveyed and stu (...)
  • 2 Wilson et al. 1999; Lloyd 1977/1979.
  • 3 Wilson, Megías 2008.
  • 4 The data demonstrates in this article is a preliminary examination of some archaeological remains (...)

1The economic and productive activities of Roman Cyrenaica have received little attention in scholarship. Most of the information about manufactures and production is very restricted. Olive and wine presses, well known from other parts of North Africa, have not yet been fully documented or studied in this region1. The most important work has been carried out at the ancient sites of Euesperides and Berenice, both of which lie beneath the modern city of Benghazi2, revealing considerable information regarding the economic and trade activities of this part of Cyrenaica. However, almost nothing has hitherto been known about the level of coastal exploitation in the region and its economic role in Cyrenaica. Although some archaeological missions are working at a few coastal sites in this region, many questions regarding the economic and manufacturing roles of these sites have not been answered. Information about fish-related manufacturing and production is still very obscure. Until now only rare studies have examined the scale of exploitation of the marine resources in the region3 or the extent to which the Cyrenaican coastal inhabitants used the sea as a source of income. A survey carried out by the author in 2010 and 2012 has revealed new evidence of abundant archaeological remains scattered along the shore of Cyrenaica4, indicating intensive exploitation of marine resources. The purpose of this paper is to analyse the evidence to assess the likelihood of fish-related production activities in the area.

Fig.1. The distribution of sites within the study area
(© Mohamed A Hesein).

Fig.1. The distribution of sites within the study area(© Mohamed A Hesein).

1. Area studied

2The field survey documented six sites that were probably associated with the exploitation of marine resources. The sites are distributed within an area measuring 50 km west-east (Fig. 1) lying along the coastal strip of Al-Jabel Al-Akhdar (Green Mountain). Starting from the ancient site of Kainopolis (El-Agla), located about 35 km east of the ancient site of Ptolemais (Tolmeitha), to the area of Noat1, about 15 km to the west of Apollonia (Susa).

2. The archaeological remains

2.1. Cape of Phycus (Al-Mamloh)

3Al-Mamloh is an archaeological site that lies about 7 km east of Phycus. Here, a number of archaeological remains are preserved in an area covering more than 1 km2, including remains of a fortification, a villa and an industrial complex. This latter structure lies 35 m from the sea and covers an area of 2000 m2 (Fig. 2).

Fig. 2. The possible fish-processing site at cape of Phycus (Al-Mamloh)
(© Mohamed A Hesein).

Fig. 2. The possible fish-processing site at cape of Phycus (Al-Mamloh)(© Mohamed A Hesein).

4Large-scale excavation is needed to understand the function of the site, however, from the ground we can assume that the site comprised a rectangular structure surrounded by external walls. There is a well located 9 m in front of this structure which is still in use today.

5The building consists of a set of tanks of different shapes and sizes, mainly rectangular, built of stone and lined with opus signinum (Fig. 2a). By tracing the foundations the following features can be identified:

6Eight tanks (numbered from 1 to 8), which seem to vary in size and shape. Most are rectangular apart from tank 6 which has curved corners. The dimensions of the tanks are 2.80 × 2.40 m, 1.50 × 2.50 m, 1.60 × 2.40 m, 1.60 × 1.50 m, 1.60 × 1.40 m, 1.50 × 2.50 m, 1.20 x 1.30 m, 1.20 × 1.13 m and 1.00 × 0.70 m. Depths are uncertain.

Fig. 2a. Detail of waterproofing in Vat 3, facing west (© Mohamed A Hesein).

Fig. 2a. Detail of waterproofing in Vat 3, facing west (© Mohamed A Hesein).

7Three basins can be only partly traced (9, 10 and 11). Although their shape and size cannot be accurately determined, it is probable that they were the same shape as the basins next to them. For instance, basin 9 is more likely to be similar in shape to basins 1 and 2.

2.2. Phycus (Zawiet El-Hamama)

8Phycus lies 25 km to the north-west of Cyrene and about 30 km west of Apollonia. The archaeological remains that appear to be related to fish production are located in two different areas (sites 1 and 2) within the ancient site of Phycus (Figs. 3, 3a and 3b).

Fig. 3. Possible fish-related productions at the ancient site of Phycus
(Zawiet El-Hamama) (© Mohamed A Hesein).

Fig. 3. Possible fish-related productions at the ancient site of Phycus(Zawiet El-Hamama) (© Mohamed A Hesein).

Fig. 3a. Distribution of tanks in site 1 at Phycus
(Zawiet El-Hamama) (© Mohamed A Hesein).

Fig. 3a. Distribution of tanks in site 1 at Phycus(Zawiet El-Hamama) (© Mohamed A Hesein).

Fig. 3b. Initial plan for the site 2 (tanks?) at Phycus (Zawiet El-Hamama) (© Mohamed A Hesein).

Fig. 3b. Initial plan for the site 2 (tanks?) at Phycus (Zawiet El-Hamama) (© Mohamed A Hesein).

2.2.1. Site 1

9The site lies on the eastern headland of the promontory. This zone contains more than 30 tanks or vats (Fig. 3a) distributed as seven groups in an area of 2000 m2.

Group one

  • 5 It is remarkable that the most of the vats recorded at site 1 at Phycus and sites 1 and 2 at Al-H (...)

101 - A row of five vats cut into the rock vertically (Numbered from 1 to 5). Vat 2 (Fig. 3c) measures 1 m2, while vats 3 and 5 are rectangular measuring 0.70 × 0.60 m and 0.90 × 0.95 m respectively. Vats 1 and 4 are bottle-shaped 5, their opening diameters are 1 m and 1.4 m. The depth was not identified for this group because of the accumulation of soil that covered them. All vats are lined with waterproof plaster (opus signinum).

Fig. 3c. Vat 2 within group 1, facing west
(Photo: Mohamed A Hesein).

Fig. 3c. Vat 2 within group 1, facing west(Photo: Mohamed A Hesein).

112 - A row of ten vats (numbered from 6 to 15) cut into the rock. All the vats are lined up three by three. The vats vary in size and shape, six having a round mouth (vats 6, 7, 8, 10, 11, 13, 14 and 15) measuring 0.80, 1, 0.75, 1.15, 0.75, 1.25, 1 and 1.5 m in diameter respectively and two are oval (vats 9 and 12) measuring 1.4 and 1.25 m. All the vats are lined with opus signinum. The first line of three vats (12, 13, 14 and 15) was built half a metre lower than the ground level of the others in the group.

Group two

12This is a set of three vats (16, 17 and 18). Vats 16 and 17 are rectangular, measuring 1.20 × 0.95 m and 4.36 × 3.87 m respectively. Vat 18 is bottle-shaped with an opening diameter of 1.20 m. As with Group One, the three vats are lined with opus signinum.

Group three

13This set of 6 vats (19, 20, 21, 22, 23 and 24) (Fig. 3d). is located 3.70 m north of Group Two. Vats 19, 20, 21 and 22 are bottle-shaped, with outlet diameters of 1.50, 1.40 and 1.20 m respectively. Vats 22 and 23 were inter-connected measuring (2.0 × 3.0 m) and were faced with opus signinum. The eastern side of vats 21 and 22 facing the sea is completely destroyed by erosion, which allows us to estimate its depth as more than 1.5 m.

Fig. 3d. Side of group 3, looking south (© Mohamed A Hesein).

Fig. 3d. Side of group 3, looking south (© Mohamed A Hesein).

Group Four

14To the north of Group Three is a set of three bottle-shaped vats (25 (Fig. 3e), 26 and 27). Vat 25 has been eroded by the sea, leaving its southern side visible in section. It has a circular opening measuring 0.85 m in diameter. It is about 2 m in depth with its cylindrical shaft lined with opus signinum. Vats 26 and 27 are located to the north of the previous one and are bottle-shaped, measuring 1.70 and 1.20 m respectively. Their depths have not been identified as they are full of soil. Both are lined with opus signinum.

Fig. 3e. An imagined vat which is similar to the shape of vat 25 in Phycus, the total capacity of this vat is about 5.25 m3. The lid shown in this image is just imaginary, and its shape does not rely upon archaeological evidence (© Mohamed A Hesein).

Fig. 3e. An imagined vat which is similar to the shape of vat 25 in Phycus, the total capacity of this vat is about 5.25 m3. The lid shown in this image is just imaginary, and its shape does not rely upon archaeological evidence (© Mohamed A Hesein).

Group five

15This group is a set of three vats (28, 29 and 30) of varying shape and size. Vats 28 and 29 are bottle-shaped with outlet diameters of 1.23 m and 0.89 m respectively. Vat 30 is rectangular measuring 1.30 × 1.24 m. There is a rectangular cut in rock located between vat 28, 29 and 30 of unclear purpose. The depth of all these cannot be determined because they are full of soil, as for nearly all the vats at this site.

Group Six

16This group is located to the northwest of Group Four and contains four vats (31, 32, 33 and 34). These are the largest bottle-shaped vats (Fig. 3f) at the site with opening diameters of 2.25 m, 2.48 m, 2.8 m and 3.1 m respectively, again with undetermined depths. All four vats are lined with opus signinum.

Fig. 3f. Vats 32, 33 and 34 within group 6, facing south-west (© Mohamed A Hesein).

Fig. 3f. Vats 32, 33 and 34 within group 6, facing south-west (© Mohamed A Hesein).

Group seven

17A set of four vats (35, 36, 37 and 38) of varying size and shape, located between 4 and 7 m apart. Vats 35, 36 and 37 are bottle-shaped with opening diameters of 1.2, 1 and 0.76 m respectively and are lined with opus signinum. The remaining vat (38) has a rectangular opening leading to a large rectangular tank measuring about 3.0 × 2.0 m with a depth of at least 2 m. The tank is lined with opus signinum. This tank most probably served as a water cistern.

18There are two channels cut into the rock running from the sea into the site. One runs north to south and is about 2 m wide and about 1 m deep. The other runs from the southeast to the northwest and is about 2 m wide and 0.20 m deep.

2.2.2. Site 2

19This site is located to the west of site 1. It is difficult at the moment to identify the whole plan of this site as it is covered by sand. Some of the visible remains indicate a complex building lying beneath the sand. The general impression is that it possessed a set of tanks or basins (?) of different sizes and shapes (mostly rectangular) built of stone and lined with opus signinum (Fig. 3b).

2.3. Aptouchou (Al-Haniya)

20A number of vats were found along the shore of the ancient site of Haniya, 13 km west of Phycus. The vats can be divided into two groups according to their location (Fig. 4).

Fig. 4. Distribution of vats within site 1 and 2 at ancient Aptouchou (Al-Haniya) (© Mohamed A Hesein).

Fig. 4. Distribution of vats within site 1 and 2 at ancient Aptouchou (Al-Haniya) (© Mohamed A Hesein).

2.3.1. Site 1

21A set of vats (numbered from 1 to 8) lies on the rocky area overlooking the sea. All the vats are cut into the rock and are circular with neck diameters of 0.90, 0.80, 0.90, 0.80, 0.80, 0.70, 0.70, and 0.80 m respectively with unidentified depths (Fig. 4a).

Fig. 4a. Vat 3 within site 1 at Aptouchou (Al-Hanya), facing south-east (© Mohamed A Hesein).

Fig. 4a. Vat 3 within site 1 at Aptouchou (Al-Hanya), facing south-east (© Mohamed A Hesein).

2.3.2. Site 2

22This group occupies a small rocky island adjoining the shore, located 200 m to the west of site 1. The depth of the water around the island varies. For instance, on the shore side the water does not exceed 0.50 m in depth, whereas on the other sides, the water is 1.50 m to 2.50 m in depth. The island and site 1 serve to form a bay.

23Site 2 consists of 12 circular vats in a line (Fig. 4b), cut into the rock. These vats vary in size with their diameters being: 1.10, 0.85, 1.10, 1.150, 1.60, 2.80, 1.70, 0.90, 2.20, 1.70, 2.30 and 2.40 m. In both areas there are small slots cut into the rock 0.2 m in diameter. This might suggest that the vats were covered, and that the slots supported wooden columns holding some sort of roof. All of the vats in both areas are lined with opus signinum.

Fig. 4b. Vats 4, 5 and 6 within site 2 at Aptouchou (Al-Hanya), facing south (© Mohamed A Hesein).

Fig. 4b. Vats 4, 5 and 6 within site 2 at Aptouchou (Al-Hanya), facing south (© Mohamed A Hesein).

2.4. Kainopolis (Al-Agla)

24Al-Agla lies 30 km north-west of Phycus and 35 km east of the ancient site of Ptolemais. As with other sites that have been found along this coast, there are a number of archaeological remains scattered around the site which might belong to fish-related production (Fig. 5).

Fig. 5. Google Earth imagery illustrating the distribution of remains at Kainopolis (Al-Agla).

Fig. 5. Google Earth imagery illustrating the distribution of remains at Kainopolis (Al-Agla).

Fig. 5a. Vats by the shore in Al-Agla, facing north (© Mohamed A Hesein).

Fig. 5a. Vats by the shore in Al-Agla, facing north (© Mohamed A Hesein).

25A set of twelve vats lies in shallow water (numbered from 1 to 12), installed in four rows (Fig. 5a). Each row consists of three vats of 2 m diameter. Because they are covered by water and sand their depths cannot be identified and it is difficult to assess whether they are lined with opus signinum.

  • 6 Lloyd 1977; Buzaian 2000.

26The second set of vats (numbers 13 and 14) is located 10 m to the south of the first set and consists of two circular vats built of small stones mixed with opus signinum and small ceramic fragments (Fig. 5b). The upper parts of the vats are eroded away. The bases have 2 m diameters. The remaining height of both vats does not exceed 0.20 m. They seem to be similar in size to vats recorded at the ancient sites of Berenice and Taucheira6.

Fig. 5b. Vat 13, facing west (© Mohamed A Hesein).

Fig. 5b. Vat 13, facing west (© Mohamed A Hesein).

Fig. 5c. Vat 15, facing east (© Mohamed A Hesein).

Fig. 5c. Vat 15, facing east (© Mohamed A Hesein).

27The third set of vats comprises two vats (numbered 15 and 16) built of stones and lined with a thick layer of opus signinum to the south-west of vats 13 and 14. It is difficult at the moment to reconstruct the whole structure without excavation (Fig. 5c). However, on the ground it seems that they are rectangular, similar to one of the examples to the west.

28A vat (number 17) built of stone and lined with a thick layer of opus signinum is located to the south-west of vat 16. Unlike the previous examples, this one takes a large circular shape with a diameter of 4 m.

29It is worth mentioning that these vats or basins exist with other archaeological features within the building complex. Large-scale investigation and excavation is needed to get a more detailed picture of these sites.

3. Discussion

  • 7 Trakadas 2004a; Curtis 1983; 1991.

30In evaluating the significance of the ancient fish-processing industry in Cyrenaica, two points should be considered. First, we must note the references made by ancient writers to the high demand for marine products in the Roman period, especially the growing demand for fish sauces such as garum and liquamen.7 Secondly, we need to assess if there was a growth in this industry over time at coastal sites with the construction of new sites or the development of pre-existing ones. Certainly the exploitation of marine products seems to have increased during the Roman period in this area.

3.1. Evidence from ancient texts

  • 8 Scylacis Caryandensis 108; Strabo Geography 20; Ptolemy Geography IV; Stadiasums Magni Maris; Syn (...)

31The ancient texts mention more than 25 sites along the coast of Cyrenaica8, of which my survey only covers a small number. Some pre-existing ports flourished and developed into independent cities. For example, Apollonia, the port of Cyrene, was established as a small harbour to serve the city, but grew considerably and became an independent city. The same scenario happened at Barce and its port (Ptolemais). These examples show that access to the sea went beyond the need for an easy means of transportation and indicate a growing interest in the active exploitation of marine resources. Additionally, Cyrenaica is not an isolated example, but displays a similar path of development noted at many sites around the Mediterranean.

32Synesius (letter 148) describes villagers who live far south of Cyrenaica. He stated there were no urban agglomerations, roads or commercial areas in that region. Two important statements can be extracted. First, there was production and marketing of salt in the coastal region of Cyrenaica. As he said in the letter «the people here do not take to the sea, even for the purpose of getting their salt». Secondly, Cyrenaican people, at least who live near to the sea, were aware of the importance of the sea as a source of food, as he said “the people here refuse to believe that the sea too is able to nourish mankind”.

3.2. The evidence from the archaeology – fish or other productions?

33The intensive installations along the shore of Cyrenaica offer possible confirmation for the large-scale exploitation of marine resources. The presence of a large variety of vats on, or a few metres from the sea requires us to consider the exact purpose of their construction. However, if we admit theoretically that the main purpose of these vats was to keep and preserve fresh water, then why would a site such as Phycus reserve an area for more than 35 vats cut into the rock? There are four wells (two of which are still in use today) a few hundred metres west of the promontory where the vats were cut. Why did they need these huge quantities of water? I think it is inaccurate to adopt the interpretation of fresh water storage.

34Hypothetically, it is more practical to build one large cistern for keeping water instead of building 6, 9 or 12 aligned vats containing a low volume of water. Additionally, the deliberate arrangements of the vats indicate that the purpose was for an elaborate process and different types of products.

  • 9 Kenrick 2013.

35In addition, the water cistern that found in Cyrenaica had a different geometric arrangement. In a personal communication, Ahmed Emrage confirms that in his recent field survey in Cyrenaica, three types of water cistern have been identified. The first type has a narrow opening (Fig. 6a) leading to a huge water collecting square chamber; the depth of this type is up to 6 m. The second is a vaulted rectangular cistern containing one or two chambers (Fig. 6b), more than two in some cases, divided by an arched wall. This type of water cistern has been found in different areas in Cyrenaica such as Ptolemais.9 The third is a huge rectangular shape, with a roof sometimes supported by a column.

Fig. 6a. Water cistern, type 1, facing north (from Emrage forthcoming).

Fig. 6a. Water cistern, type 1, facing north (from Emrage forthcoming).

Fig. 6b. Example of vaulted cistern that found in Cyrenaica, facing north (from Emrage forthcoming).

Fig. 6b. Example of vaulted cistern that found in Cyrenaica, facing north (from Emrage forthcoming).

36Presumably, some of the larger vats I have identified were for the collection and preservation of water. Such quantities suggest that there was an industrial process being carried out nearby which required a huge amount of water.

  • 10 Wilson 2006, p. 527.
  • 11 Wilson 2006; Wilson, Tébar Megías 2008.
  • 12 Wilson 2003; Tébar Megías, Wilson 2008.

37Another production process that can be rejected is purple dye manufacture. The possibility of such production is not compatible with rock-cut vats. As Andrew Wilson states «the production of purple dye required heating the shellfish, which could not be done in sunken vats».10 He confirms that the new evidence that comes from purple dye production sites at Meninx (Jerba) and Euesperides (Benghazi) suggests that this production did not need vats11. Furthermore, the absence of remains of murex shells or burning remains at the sites described above supports this view. Indeed, these vats do not seem to have been intended for textile dyeing, leather tanning or rinsing. Such tanks would require the vat to be built above ground level with their bases at floor level12.

  • 13 Buzaian 2012.

38Nevertheless, the absence of permanent elements such as oil presses around these installations excludes the possibility that these vats were for keeping olive oil products. Additionally, presses were mainly built in the hinterland of the main cities and not on the coast. This has also been confirmed by the recent archaeological field survey by Ahmed Buzaian13.

  • 14 Wilson 2001.
  • 15 In terms of location, it should differentiate between vats located inside the circuit walls of ci (...)
  • 16 Flemming 1972; Lloyd 1977; Buzaian 2000; Wilson 2001.

39There remain two possible productions which might have carried out in these vats: wine fermentation and fish-related production. Wilson14 refers to a number of bottled-shaped vats which have a similar structure to those presented above. These vats have been recorded inside the circuit wall of some coastal cities, namely Ptolemais, Taucheira, Berenice15 and another set of five vats was located in the rocky area by the shore of Apollonia16. Wilson argues that such vats are more likely to be associated with wine production using a treading floor.

  • 17 Wilson 2001, p. 32.

40However, Wilson does not seem to demonstrate any concrete evidence for the existence of treading floors associated with those vats, except the one at Taucheira which he seems to be in some doubt about. He states that «since no olive pressing structures have been found nearby (such as orthostats, press beds or olive mills), I am inclined to believe that they were more likely to have been used for the fermentation of wine. This suggestion receives some support from the presence a little to the north, though not in strict demonstrable association with the vats, of a feature which I would identify as a wine treading floor»17.

  • 18 Curtis 1991.
  • 19 Bernal et al. 2009.
  • 20 Based on personal observation of these vats.
  • 21 Flemming 1972.

41Although I have some doubt about the presence of fish-related production within cities, as this would produce a strong and unpleasant smell during the process18, garum workshops have been found inside the ancient city of Pompeii19. Indeed, there is no sign of an associated treading floor near the vats mentioned by Wilson at Apollonia20. Adding to that the occurrence of submerged vivaria21 100 m to the north-east of these vats, it seems very likely that these vats could have been associated with fish-related production.

  • 22 Buzaian 2000; Wilson 2001.

42Looking at another example from Apollonia, work carried out by a French mission revealed a house containing six vats (with depths of 3.00 m) sunk into the ground, similar to those found at Taucheira22. Samples from the vats have been taken for chemical analysis; the first results did not show any evidence of fish processing in these vats. Subsequently, in personal communication, the French chemist who did the analysis confirmed to me that, after comparing the results from the Apollonia vats with those from other fish production sites around the Mediterranean, fish sauce might be present in the samples. If the new chemical evidence confirms that fish products were in these vats, it strongly suggests that such vats inside the city walls could be involved, to some degree, in fish production.

  • 23 Balagrae is an ancient city locating inland about 15 km to the west of Cyrene (see Fig. 1). Buzai (...)

43However, square or rectangular area with drainage which might have connected to these vats does not seem to exist in my survey area. This excludes the hypothesis of an association with wine-treading vats such as the one found at the ancient site of Balagrae23.

44Also, it is a mistake to assume that wine was pressed inland and then brought in animal skins to the shore for storage in these vats while awaiting shipping. It is impractical on several grounds, not least the multiple transhipments of wine or olive oil, between pressing and shipping. Add to that, cutting one big vat in the ground instead of six is more beneficial for storing a single type of substance (as explained above for water). Additionally, it makes more sense to fill wine or olive oil directly from transport skins into amphorae and then to store them in a warehouse while waiting for shipping, marketing or local consumption.

  • 24 Yorke, Davidson forthcoming.

45The existence of these vats a few metres from the sea, and especially on the promontories and small islets seems to suggest the possibility that they could be used for fish-related products. Similar examples have been noted by the shore in Ptolemais. Many vats cut into the rock in the promontory have been recorded by Yorke and Davidson and have been initially interpreted as a fish factory24. Supporting this view, are the large quantities of fish-related production sites that have been found along the shores of Mediterranean. Some of these industries were cut into the rocky areas.

  • 25 Ameur 2005.
  • 26 Id.
  • 27 Yorke, Davidson 1969.
  • 28 Trakadas 2004b.

46Identical forms were found in different places along the coast of Tunisia. At the ancient site of Gummi (Mahdia), for example, 153 vats cut into the rock were found. They come in four main shapes: 64 circular (almost oval), 39 rectangular, 5 cylindrical and 1 square25. This set of vats lie on a peninsula which overlooks the passage for seasonal fish migration. Such geometrical structures and shapes were found at other sites, such as Thaenae, Sullecthum, Sidi Daoud and El-Mamoura in Tunisia and have been interpreted as cetariae for the fish industry26. In Algeria, according to Yorke and Davidson, a number of circular fish vats cut into the rock were found in Tipasa and other rectangular fish tanks cut into the rock were found at the ancient site of Cherchel27. In southern Spain and Morocco, as a third example, fish production tanks of different sizes and shapes were found cut into the rock (Fig. 7)28.

Fig. 7. Cetariae cut into the rock at Praia de Angeiras (Trakadas 2004b, p. 59).

Fig. 7. Cetariae cut into the rock at Praia de Angeiras (Trakadas 2004b, p. 59).
  • 29 Id.

47Some of these features seem to have the same character as those found on the coast of Cyrenaica. Several fish vats (cetariae) were found along the coast between Santa Pola and Punta de L’Arenal cut into the rock in the promontory areas29. Circular vats, for example, sunk into the ground were also found at the ancient site of Baelo. Furthermore, in Ceuta, there are large circular lined holes cut into the ground located within a fish installation.

  • 30 Id.

48It is worth mentioning that these types of features found in different fish production sites such as Calpe and Punta de L’Arenal have been interpreted as vats for producing fish sauces30.

  • 31 Davaras 1974.
  • 32 Slim et al. 2004.
  • 33 Flemming 1972.
  • 34 For further discussion of vivaria, see my forthcoming thesis.

49Elsewhere in the Mediterranean, on Crete, which faces the coast of Cyrenaica, rows of vats varying in size and cut into the rock were found at Chersonesos and Siteia. The vats served as vivaria for storing live fish during the migratory seasons, to be sold later in the markets rather than as breeding basins31. Similarly, several rock-cut tanks connected to the sea by channels were found in Tunisia, and were used for breeding or holding fish for salting32. These types of tank connected to the sea by channels were identified also in Apollonia33 and other sites in Cyrenaica34.

  • 35 Slim et al. 2007; Curtis 1983.
  • 36 Manilius 5.656-81; Columella 12.55.4.
  • 37 Bernal et al. 2009; Wilson 2006.
  • 38 Trakadas 2004b.

50Producing fish sauces and other fish products does not require a specific shape or size of container35. Manilius and Columella state that salted fish were placed in succession in large cetariae or ceramic containers36. Supporting this idea, the dolia that have been found in the garum workshop at Pompeii suggest (Fig. 8) that they were used as containers for making garum37. The circular shape is highly suggestive of an installation for making garum sauce, as such structures facilitate stirring, which is part of the process38.

Fig. 8. Dolia that have been found in the garum workshop at Pompeii (© Mohamed A Hesein).

Fig. 8. Dolia that have been found in the garum workshop at Pompeii (© Mohamed A Hesein).

51The second set of examples of what seem to be fish tanks comprise rectangular tanks set next to each other. These tanks were found in different locations in the surveyed area such as Phycus, Cape of Phycus and Kainopolis. It is difficult to trace the whole structures as most of them are covered by sand or disturbed by human activities. Large-scale excavation is needed to obtain a comprehensive picture of the remains, their depth and their exact function. Generally, the structures are probably sets of tanks of different shapes and sizes mainly rectangular. They were built of brick and the interior walls were faced with opus signinum.

  • 39 Wilson 1999.
  • 40 Slim et al. 2004; Ben-Lazreg et al. 1995.

52However, my preliminary interpretation, related to the size, shape and location a few metres from the sea, is that they were fish salting tanks. These tanks seem to have the same structural format to examples found in Sabratha and interpreted as salted fish tanks by Andrew Wilson39 and those found along the coast of Tunisia40. The square and rectangular tank shapes are probably the most common types for salted fish cetariae that have been found around the Mediterranean basin.

Conclusion

53The rectangular and square vats scattered along the shore of Cyrenaica such as those at Phycus and Cape of Phycus might be attributed, with some confidence, to the production of salted fish. However, the neighbouring bottle-shaped vats should be interpreted with caution at the present time. Wine fermentation is one possibility, but garum production should not be discounted.

54Large-scale excavation and chemical analyses are required to prove this idea conclusively and to provide evidence for what was carried out at these manufacturing sites. However, my interpretation is based on the comparison of vats of similar shapes and sizes along the coast of North Africa, and the location of these vats close to the sea. It seems likely that the function of these sites could be fish-related processing, with garum production being carried out in the bottle-shaped vats, and salted fish processing in the rectangular or square tanks.

I would like to express my greatest gratitude to Dr. Victoria Leitch for her continuous support and encouragement. I am grateful to Dr. Nick M. Ray and Dr. Matthew Hobson for their feedback and advice on the early draft of this paper. I am most grateful to Ahmed Emrage and David Davidson for sharing with me their research before publication. A special thanks to Professor David Mattingly, my PhD supervisor, for his useful and fruitful suggestions and guidance. Finally, I wish to thank my friend Ahmed El-Hayan for motivating and helping me during my field survey. Any errors or mistakes in the article are mine

Bibliographie

Ameur 2005 : Ameur (Y.) (2005) - Recherches sur les installations halieutiques à Gummi à l’époque antique. In : Hassen M. (ed.), Byzacium antique et Sahel médiéval : Urbanisme et Occupation du Sol. Colloque sur l’urbanisme au Sahel organisé à Monastir (20-21Juillet 2001), Université de Tunis I, 2005, p. 37-80.

Ben Lazreg et al. 1995 : Ben Lazreg (N.), Bonifay (M.), Drine (A.), Trousset (P.) - Production et commercialisation des salsamenta de l’Afrique ancienne. In : Trousset (P.) (ed.), Productions et exportations africaines: actualités archéologiques en Afrique du Nord antique et médiévale. VIe colloque international sur l’histoire et l’archéologie de l’Afrique du Nord (Pau, octobre 1993 - 118e congrès), Paris, 1995, p. 103-142.

Bernal et al. 2009 : Bernal (D.), Cottica (A.), Zaccaria (A.) - El garum de Pompeya y Herculano (2008-2012). Sintesis de la primera campana del proyecto hispano-italiano, Informes y Trabajos 3, 2009.

Bouzian 2012 : Bouzian (A.) - Oil presses in Cyrenaica [lecture], annual meeting of Society of Libyan Studies, 4 December 2012, London.

Buzaian 2000 : Buzaian (A.) - Excavations at Tocra (1985-1992), Libyan Studies 31, 2000, p. 59-102.

Buzaian, Bentaher 2002 : Buzaian (A.), Bentaher (F.) - Preliminary report on two seasons of excavations at Balagrae (al-Beida), Libyan Studies, 33, 2002, p. 125-132.

Buzaian, Bentaher 2006 : Buzaian (A.), Bentaher, (F.) - Excavations at Balagrae (al-Beida) 2001-2003. In : Fabbricotti (E.), Menozzi (O.), Cirenaica : studi, scavi e scoperte. Atti del X Convegno di Archeologia cirenaica (Chieti 24-26 novembre 2003), BAR International Series, Oxford, 2006, p. 153-162.

Curtis 1983 : Curtis (R.) - In defense of garum, Classical Journal, 78 (3), 1983, p. 232-240.

Curtis 1991 : Curtis (R.) - Garum and Salsamenta: Production and Commerce, Studies in Ancient Medicine, Leiden, 1991.

Davaras 1974 : Davaras (C.) - Rock-Cut Fish Tanks in Eastern Crete, ABSA, 69, 1974, p. 87-93.

Emrage forthcoming : Emrage (A.) - Roman forts and fortified farms in Cyrenaica, PhD thesis, Leicester, University of Leicester.

Flemming 1972 : Flemming (N.) - Cities in the Sea, London, 1972.

Hesein forthcoming : Hesein (M.) - Bridging the Eastern and Western Mediterranean: The Roman Harbour Sites on the Coast of Cyrenaica, Libya, PhD thesis, Leicester, University of Leicester.

Jones, Little 1971 : Jones (B.), Little (H.) - Coastal Settlement in Cyrenaica, Journal of Roman Studies, 61, 1971, p. 64-79.

Lloyd 1977 : Lloyd (J.) (ed.) - Excavations at Sidi Khrebish, Benghazi (Berenice), I, Tripoli, Supplements to Libya Antiqua, 1977.

Lloyd 1979 : Lloyd (J.) (ed.) - Excavations at Sidi Khrebish, Benghazi (Berenice), II, Tripoli, Supplements to Libya Antiqua, 1979.

Polish Archaeological Mission in Ptolemais, Libya 2011 : Preliminary Reports [Online] Available from < http://ptolemais.pl/pdf/Ptolemais_Preliminary_Report_2007-2009.pdf > [30 September 2012].

Slim et al. 2004 : Slim (H.), Trousset (P.), Paskoff (R.), Oueslati (A.) - Le littoral de la Tunisie. Étude géoarchéologique et historique, Études d’Antiquités africaines, Paris, 2004.

Slim et al. 2007 : Slim (L.), Bonifay (M.), Piton (J.), Sternberg (M.) - An example of fish salteries in Africa Proconsularis: the officinae of Neapolis (Nabeul, Tunisia). In : Lagóstena (L.), Bernal (D.), Arévalo (A.) (eds), Cetariae 2005. Salsas y Salazones de Pescado en Occidente durante la Antigüedad. Actas del Congreso Internacional (Cádiz, 7-9 de noviembre de 2005), BAR International series, 1686, 2007, p. 21-44.

Tébar Megias, Wilson 2008 : Tébar Megias (E.), Wilson (A.) - Classical and Hellenistic textile production at Euesperides (Benghazi, Libya): preliminary results. In : Alfaro Giner (C.), Karali (L.) (eds.), Vestidos, textiles y tintes (Purpureae Vestes II): Estudios sobre la Produccion de Bienes de Consumo en la Antigüedad, Universidad de Valencia, Servicio de Publicaciones, Valencia, 2008, p. 49-59.

Trakadas 2004a : Trakadas (A.) - The archaeological evidence for fish processing in the Western Mediterranean. In : Bekker-Nielsen (T.) (ed), Ancient Fishing and Fish-Processing in the Black Sea Region, Aarhus, 2004, p. 47-82.

Trakadas 2004b : Trakadas (A.) - Ancient marine resource exploitation in the Western Mediterranean: the contribution of the fish-salting industry of Mauretania Tingitana (Morocco), Institute of Classical Archaeology, Aarhus University, Denmark [Online] Available from <http://hmapcoml.org/projects/m&b/Documents/Trakadas_HMAP-Med.pdf > [15 December 2012].

Wilson 1999 : Wilson (A.) - Commerce and industry in Roman Sabratha, Libyan Studies, 30, 1999, p. 29-52.

Wilson 2001 : Wilson (A.) - Urban economies of late antique Cyrenaica. In : Kingsley (S.), Decker (M.) (ed), Economy and exchange in the East Mediterranean during late Antiquity, Oxbow Books, Oxford, 2001, p. 28-43.

Wilson 2003 : Wilson (A.) - The archaeology of the Roman fullonica, Journal of Roman Archaeology, 16, 2003, p. 442-466.

Wilson 2004 : Wilson (A.) - Cyrenaica and the late antique economy, Ancient West and East, 3.1, 2004, p. 143-154.

Wilson 2006a : Wilson (A.) - Fishy business: Roman exploitation of marine resources, Journal of Roman Archaeology, 19.2, 2006, p. 525-537.

Wilson 2006b : Wilson (A.) - New light on a Greek city: archaeology and history at Euesperides. In : Fabbricotti (E.), Menozzi (O.) (eds.), Cirenaica : studi, scavi e scoperte. Atti del X Convegno di Archeologia cirenaica (Chieti 24-26 novembre 2003), BAR International Series, Oxford, 2006, p. 141-152.

Wilson, Tébar Megías 2008 : Wilson (A.), Tébar Megías (E.) - Purple dye production at Hellenistic Euesperides (Benghazi, Libya). In : Napoli (J.) (ed.), Ressources et activités maritimes des Peuples de l’Antiquité, Actes du Colloque international de Boulogne-sur-Mer (12-14 mai 2005), Boulogne-sur-Mer, 2008, p. 231-238.

Wilson et al. 1999 : Wilson (A.), Bennett (P.), Buzaian (A.), Ebbinghaus (S.), Hamilton (K.), Kattenberg (A.), Zimi (E.) - Urbanism and Economy at Euesperides (Benghazi): preliminary report on the1999 season, Libyan Studies, 30, 1999, p. 147-168.

Yorke, Davidson 1969 : Yorke (R.), Davidson (D.) - Roman Harbours of Algeria, Underwater Association Report, 1969.

Yorke, Davidson 1973 : Yorke (R.), Davidson (D.) - Tolmeita & Tocra. Roman Harbours of Cyrenaica, Underwater Association Minisymposium, 1973.

Yorke, Davidson forthcoming : Yorke (R.), Davidson (D.) - The Hellenistic Harbour of Ptolemais.

Notes

1 Wilson 2004; Buzaian 2009. Oil and wine presses in Cyrenaica are currently being surveyed and studied by Ahmed Buzaian.

2 Wilson et al. 1999; Lloyd 1977/1979.

3 Wilson, Megías 2008.

4 The data demonstrates in this article is a preliminary examination of some archaeological remains which seem to be more likely related to fish production. For more discussion and analytical studies of these remains, see my forthcoming PhD thesis (Hesein forthcoming).

5 It is remarkable that the most of the vats recorded at site 1 at Phycus and sites 1 and 2 at Al-Hanya seem to be bottle-shaped, widening out below a narrow neck. In any case, the diameter that has been provided is the measurement of the opening mouth.

6 Lloyd 1977; Buzaian 2000.

7 Trakadas 2004a; Curtis 1983; 1991.

8 Scylacis Caryandensis 108; Strabo Geography 20; Ptolemy Geography IV; Stadiasums Magni Maris; Synesius 170, 171.

9 Kenrick 2013.

10 Wilson 2006, p. 527.

11 Wilson 2006; Wilson, Tébar Megías 2008.

12 Wilson 2003; Tébar Megías, Wilson 2008.

13 Buzaian 2012.

14 Wilson 2001.

15 In terms of location, it should differentiate between vats located inside the circuit walls of cities (e.g. those mentioned by Wilson), and those along the shore (e.g. those identified by the author in the surveyed area, and those by the shore of Ptolemais mentioned by Yorke and Davidson).

16 Flemming 1972; Lloyd 1977; Buzaian 2000; Wilson 2001.

17 Wilson 2001, p. 32.

18 Curtis 1991.

19 Bernal et al. 2009.

20 Based on personal observation of these vats.

21 Flemming 1972.

22 Buzaian 2000; Wilson 2001.

23 Balagrae is an ancient city locating inland about 15 km to the west of Cyrene (see Fig. 1). Buzaian, Bentaher 2002/2006.

24 Yorke, Davidson forthcoming.

25 Ameur 2005.

26 Id.

27 Yorke, Davidson 1969.

28 Trakadas 2004b.

29 Id.

30 Id.

31 Davaras 1974.

32 Slim et al. 2004.

33 Flemming 1972.

34 For further discussion of vivaria, see my forthcoming thesis.

35 Slim et al. 2007; Curtis 1983.

36 Manilius 5.656-81; Columella 12.55.4.

37 Bernal et al. 2009; Wilson 2006.

38 Trakadas 2004b.

39 Wilson 1999.

40 Slim et al. 2004; Ben-Lazreg et al. 1995.

Table des illustrations

Titre Fig.1. The distribution of sites within the study area(© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-1.jpg
Fichier image/jpeg, 276k
Titre Fig. 2. The possible fish-processing site at cape of Phycus (Al-Mamloh)(© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-2.jpg
Fichier image/jpeg, 104k
Titre Fig. 2a. Detail of waterproofing in Vat 3, facing west (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-3.jpg
Fichier image/jpeg, 404k
Titre Fig. 3. Possible fish-related productions at the ancient site of Phycus(Zawiet El-Hamama) (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-4.jpg
Fichier image/jpeg, 80k
Titre Fig. 3a. Distribution of tanks in site 1 at Phycus(Zawiet El-Hamama) (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-5.jpg
Fichier image/jpeg, 112k
Titre Fig. 3b. Initial plan for the site 2 (tanks?) at Phycus (Zawiet El-Hamama) (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-6.jpg
Fichier image/jpeg, 92k
Titre Fig. 3c. Vat 2 within group 1, facing west(Photo: Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-7.jpg
Fichier image/jpeg, 716k
Titre Fig. 3d. Side of group 3, looking south (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-8.jpg
Fichier image/jpeg, 1,1M
Titre Fig. 3e. An imagined vat which is similar to the shape of vat 25 in Phycus, the total capacity of this vat is about 5.25 m3. The lid shown in this image is just imaginary, and its shape does not rely upon archaeological evidence (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-9.jpg
Fichier image/jpeg, 64k
Titre Fig. 3f. Vats 32, 33 and 34 within group 6, facing south-west (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-10.jpg
Fichier image/jpeg, 836k
Titre Fig. 4. Distribution of vats within site 1 and 2 at ancient Aptouchou (Al-Haniya) (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-11.jpg
Fichier image/jpeg, 88k
Titre Fig. 4a. Vat 3 within site 1 at Aptouchou (Al-Hanya), facing south-east (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-12.jpg
Fichier image/jpeg, 160k
Titre Fig. 4b. Vats 4, 5 and 6 within site 2 at Aptouchou (Al-Hanya), facing south (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-13.jpg
Fichier image/jpeg, 864k
Titre Fig. 5. Google Earth imagery illustrating the distribution of remains at Kainopolis (Al-Agla).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-14.jpg
Fichier image/jpeg, 432k
Titre Fig. 5a. Vats by the shore in Al-Agla, facing north (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-15.jpg
Fichier image/jpeg, 868k
Titre Fig. 5b. Vat 13, facing west (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-16.jpg
Fichier image/jpeg, 1,1M
Titre Fig. 5c. Vat 15, facing east (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-17.jpg
Fichier image/jpeg, 1,3M
Titre Fig. 6a. Water cistern, type 1, facing north (from Emrage forthcoming).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-18.jpg
Fichier image/jpeg, 216k
Titre Fig. 6b. Example of vaulted cistern that found in Cyrenaica, facing north (from Emrage forthcoming).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-19.jpg
Fichier image/jpeg, 212k
Titre Fig. 7. Cetariae cut into the rock at Praia de Angeiras (Trakadas 2004b, p. 59).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-20.jpg
Fichier image/jpeg, 444k
Titre Fig. 8. Dolia that have been found in the garum workshop at Pompeii (© Mohamed A Hesein).
URL http://books.openedition.org/pccj/docannexe/image/1689/img-21.jpg
Fichier image/jpeg, 138k

Auteur

Department of Archaeology and ancient History. University of Leicester. UK. Department of Archaeology. University of Omar Al-Mukhtar, Al-Bayda, Libya. medhesein@hotmail.com

Le texte et les autres éléments (illustrations, fichiers annexes importés) sont sous Licence OpenEdition Books, sauf mention contraire.

Lire

Open access

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search