The Pyla-Koutsopetria Archaeological Project: Cyprus, the Near East, and trade in Late Antiquity

[slide] [slide] For the past four years, the Pyla-Koutsopetria Archaeological Project has been conducting an ongoing survey of a substantial Late Roman coastal site, Koutsopetria, located between the Dhekelia cantonment and the outskirts of modern Larnaka. [slide] To date, almost 64 hectares of this mid-sized settlement have been covered by a pedestrian survey, which aims to identify the extent and nature of the site, and to place the survey results within the framework of both regional and wider Mediterranean contexts.[1] From early analyses of ceramics recovered from survey, this paper will suggest that Koutsopetria, a community located in the middle ground between city and country, enjoyed a prominent position that was at least as dependent on local exchange networks as it was on larger extended markets. 

Evidence at the site indicates that Koutsopetria was a centre of human habitation since the Late Bronze Age, during which time the heights of Kokkinokremmos were fortified.[2] [slide] Geomorphological work carried out by Dr. Jay Noller for the Cyprus Geological Survey Department in 2003 and 2004 identified a probable ancient harbour of about thirty hectares in the south of the survey area.[3] [slide] [slide] Artefacts retrieved from survey indicate steady and increasing occupation between the Archaic and Early Roman periods. [slide] The dominance of Late Roman material recovered, however – over 80% of the assemblage – reflects the vitality of the site during this period. In addition to an abundance of roofing tile and amphorae, which together comprise one half of the Late Roman material, Cypriot, African and Phocaean red-slipped fine wares were also present, [slide] with the majority of forms recovered dating from the sixth to seventh centuries. Survey revealed a dramatic decline in artefact density following the Late Roman period, suggesting that the seventh century marked the apogee of activity at the site.

In general terms, the picture from Koutsopetria is consistent with what is known about the social and economic situation in the Late Antique Near East. The steady economic decline of the mid-third century, shared by most of the Eastern provinces, began a reversal in the fifth century and some areas witnessed a dramatic increase in population and prosperity. This economic resurgence is revealed in the archaeological record by, among others, an increase in the quantity and often quality of public and private buildings, as well as wide distributions of amphorae and fine wares.[4] In Cyprus, a number of field surveys and excavations, such as those at Kalavasos-Kopetra, Palaipaphos, Petrera and Koutsopetria have produced evidence of this Late Antique social and economic revival.[5]  For example, the analysis of table wares shows that habitation at Kopetra increased dramatically in the sixth and early seventh centuries.[6] Ceramic data from the Canadian Palaipaphos Survey Project supports a prosperous phase between approximately 400 and the first half of the sixth century, and work at Petrera has offered a similar conclusion.[7] New churches were constructed in this period, too, at Kopetra and Petrera.[8] [slide] At Koutsopetria, the existence of a Late Roman building identified by its excavator as an Early Christian Basilica, as well as the widespread surface evidence of walls, the presence of Roman or Late Roman hydraulic systems, and the discovery in the summer of 2006 of Late Roman sub-floor packing, suggests that building activity was prominent at the site. [slide] Additionally, in 2004 an olive press weight and crusher stone not incompatible with a Late Roman date were recovered.[9] The regular occurrence of significant fragments of what is apparently Proconnesian marble, imported from its production site on the Sea of Marmara, provides further indication of the material prosperity of the site in the Late Roman period. [slide] Further afield, the situation is similar. Tchalenko’s survey of the Belus massif in Syria, south of Antioch, updated in recent years by selective excavation and subjected to a re-analysis by Georges Tate, still demonstrates the remarkable wealth of the region.[10] Tate argued that the population of the area increased by a factor of four in some places, and that the increase in the number of epitaphs of Syrians overseas reflects the resulting burden on local production and the availability of living space.[11] [slide]

The numerous, well-constructed dwellings and the intricately detailed churches and bathhouses offer conspicuous testament to the new wealth of the region.[12] Further indications of the economic vitality of Late Antique life in the Levant are visible in the wide distribution of LR1 amphorae, significant numbers of which were manufactured near Antioch and probably served, among other tasks, to export the surplus olive oil and wine produced in the region. During the same period, the overseas prominence of Gaza wine amphorae, connected in particular to the prosperous group of Negev villages – Mampsis, Elusa, Oboda, and others – reflects, as Tate noted, that the picture in Syria was not unique.[13]

[slide] Further south, excavations at Aqaba, at the tip of the Red Sea, complement this picture. Closely linked to the Negev, Aqaba enjoyed a prosperous existence as a centre for regional and extra-regional exchange and pilgrimage traffic well into the seventh century.[14] The Roman Aqaba Project recorded a significant increase in the quantity and diversity of amphorae passing through the port, with examples including Gazan and Egyptian amphorae as well as Dressel class 44, which Parker linked to olive oil from northern Syria.[15] As in Syria, it is probable that despite the clear indications of material prosperity, the Negev villages lacked a sophisticated economic system oriented towards the export of their goods. While the relative importance of both local and long-distance trade is hotly contested and the subject of an ongoing debate, the local movement of Syrian olive oil and Gazan wine for transshipment via regional centres such as Aqaba or Antioch certainly supports the substance and prominence of local trade.[16]

While Koutsopetria fits comfortably within the general social and economic trends of Late Antiquity, we can also use the sizeable assemblage of ceramic evidence to assess more specifically the site’s role in Late Antique Mediterranean exchange networks. It remains clear that despite certain limitations, pottery is immensely useful both as an ‘index’ of exchange and an instrument by which we might interpret the archaeological record.[17] As mentioned earlier, the ceramic corpus at Koutsopetria includes the three main Late Roman finewares – Cypriot, African and Phocaean red slips, as well as a significant quantity of LR1 transport amphorae. Given the fact that finewares rarely, if ever, traveled alone, the presence of non-Cypriot wares reflects the reciprocal trade or exchange of goods from the site, and we may surmise that the finewares recovered at Koutsopetria reflect the passage of other, archaeologically invisible cargoes, originating in areas with wider connections to the Aegean and the West.[18] Small quantities of North African amphorae when considered with ARS present at Koutsopetria provide further indications of the broad economic links to the site.

[slide] The presence of these artifacts together is consistent with other Cypriot sites such as Kopetra, Palaipaphos and Petrera, where the proportions of CRS and PRS, in particular, are broadly comparable.[19] CRS is dominant, accounting for over 50% of the assemblage at all four locations, while PRS and ARS take second and third place, respectively, and Egyptian Red Slip, ERS, a distant fourth.[20] Within this data there are more striking similarities between these three sites. CRS form 9 was very common at both Koutsopetria and Kopetra,[21] and PRS form 10 and its derivatives were also prevalent at both of these sites and, in addition, further west in the CPSP survey area.[22] Furthermore, ARS 105, dating to the final phase of the ware in the early to mid-seventh century, was the most common identified ARS form at both Koutsopetria and Kopetra.[23] The recurring incidence of these very late forms underscores the contemporaneity of Kopetra and Koutsopetria in particular. One significant difference, however, is the much higher incidence of ARS in general at Koutsopetria.[24] This suggests that Koutsopetria was perhaps a local centre for redistribution and re-exchange of the ware, especially given the site’s greater exposure to the coast and apparently prominent harbour relative to Kopetra and Petrera. It is also possible that, as postulated for the incidences of only a few forms of ARS and PRS at Anemurium, dominance of a single form of ARS at Koutsopetria may reflect the lengthy distance from the source, with only the most popular form reaching Near Eastern ports. It is also possible, however, that the high percentage of ARS 105 was a result of a consistent and reliable supply.[25] Overall, however, the ceramic data suggests that the similarities between Kopetra, an inland agricultural village, Petrera, a coastal site, and Koutsopetria suggest that they were all part of the same broad socio-economic trend in Late Antiquity and, on a more specific level, were members of the same regional exchange network which connected these essentially non-urban locations.[26]

What can be inferred, however, about Koutsopetria’s greater place within Mediterranean networks? First, to date, only slight indications of Egyptian amphorae or certain examples of ERS have been recovered from the survey. Egyptian material – especially fineware – is frequently rare in significant quantity outside Egypt, with its scarcity in Cyprus sometimes linked to restrictions on the carriage of excess cargo on the annona,[27] but the data from Koutsopetria stands in contrast to that from Kopetra and Petrera, and is also highlighted against the broad evidence for Egyptian material from Dreamer’s Bay, Paphos and Amathous.[28] The presence of Egyptian finewares and amphorae at these more westerly sites reflects the fact that the main sea routes to Egypt were found far to the west of Koutsopetria. [slide] Paphos, in particular, enjoyed a long-standing role as a ‘regional hub’ for the south and southwest of the island and could claim historic links to Egypt. The small amounts of Egyptian ceramics at Kopetra and Petrera, central sites close to the south coast, probably demonstrate links with the larger trade routes to Egypt through nearby metropolitan centres. Reciprocal indications of the busy sea trade between Cyprus and Egypt are found in the prominence of both genuine and imitation Cypriot material, finewares and amphorae, in northern Egypt at places such as Alexandria and Marina el-Alamein.[29]

Eastern Cyprus, on the other hand, enjoyed far closer links with the northern Levant than Egypt, Israel and Cilicia and this is reflected in the archaeological record. In the Early Roman period, western Cypriot pinched-handled amphorae and Cypriot sigillata are rare in Syria.[30] [slide] As Hayes has demonstrated with data from five sites – Resafa, near the Euphrates in northern Syria, Antioch near the coast, Hama, to the south, and Jalame and Ein Boqeq in Palestine – CRS, staging a Late Antique comeback, enjoyed a wider southerly distribution and constitutes only a small proportion of finds in northern Syria, where PRS continued to dominate the market.[31] East of Resafa, too, excavations at the fortress city of Halabiyya-Zenobia on the Euphrates recovered no CRS, perhaps reflecting the practical limitations of onward transport into the interior.[32] Elsewhere in Syria, excavations at Déhès in the Belus massif region reinforces the overall picture of PRS dominance, with only a few examples of Cypriot wares recovered. The majority of the pottery was locally and regionally produced.[33] At Hama, which Hayes argued occupied ‘a middle position’ in the differing distribution patterns between north and south – reflected here in this table – CRS, like its predecessor Cypriot sigillata, is found in greater quantity.[34] Outside of the scope of this data, we might mention that CRS is also present at Aqaba,[35] and in some quantity at Caesarea, where the dominance of CRS form 1 led Tomber and Magness to suggest that the harbour city represented the ‘entry for its import into Palestine’.[36] Striking amongst the distribution of CRS in the southern Levant is its prevalence in a sample of 568 finewares from Sumaqa, in Palestine, whose prosperity was roughly contemporary with the towns and villages of the Belus massif. In contrast to the situation at Déhès, however, the Cypriot imports at Sumaqa dominated with an impressive 59%. Kingsley argued that this represented cultural sophistication and prosperity resulting from the wine trade, but it may also reflect the availability of different types of goods in different markets, as well as long-established trade routes.[37]

Excavations at the major eastern Cypriot religious and civic centre of Salamis-Constantina have linked the city with moderate Egyptian trade,[38] but the balance of evidence suggests that eastern Cyprus continued, in Late Antiquity, to be economically oriented towards the northern Levant. If a west Cypriot source for CRS is to be accepted,[39] its wide distribution in Egypt and the southern Levant can be explained by its passage from western ports, either direct or in the empty boats of the annona returning to Egypt from where they were subsequently transshipped. Certainly a number of authors have argued for a Cypriot stop on the annona routes to both Rome and Constantinople,[40] and, north west of Paphos, Cape Drepanon’s ornate Proconnesian marble churches and late flourishing have been connected to its apparent role as a stopping-point of the annona.[41]

The dearth of Egyptian material, as well as the apparent lack of Gazan or Palestinian wares to date, are not the only clues to an eastern economic orientation for Koutsopetria. Of the Early Roman artefacts at the site, Eastern Sigillata A, made in the Antioch region, was abundant. We may also note once more the large proportion of LR1 amphorae at Koutsopetria, whose point of origin is not yet determined.  [slide] Cypriot sources for LR1 are well known and one probable production site at Zygi-Petrini, near Kopetra, is not far.[42] Conversely, other sub-types were produced in northern Syria and, as mentioned earlier, were likely used to transport the surplus agricultural produce from the region. The LR1 at Koutsopetria may reflect the redistribution, either internally or externally, of Cypriot or Syrian goods. LR1 were used to carry oil as well as wine and it is tempting to connect the olive press from the site with local olive oil production. Seven sub-categories of LR1 have been found at Koutsopetria, and more classification work is required to determine their origins. For now, therefore, this hypothesis must remain speculative.

[slide] Finally, the pale yellow roof tiles present at Koutsopetria were probably manufactured near Salamis.[43] Indeed, the site’s relative proximity to this city – where substantial Late Roman urban redevelopment included the construction of churches and a basilica complex which received pilgrims, some arriving from Syria – underscores the likelihood of an eastern trade connection linked to local interests, with goods arriving alongside pilgrimage traffic for redistribution throughout the island through centres such as Koutsopetria.[44]

At the same time, it is important to recognize that Eastern and Western economic orientations – one towards the Levant, the other towards the Aegean and the West – were not necessarily mutually exclusive. While therefore we may argue that Koutsopetria enjoyed a closer relationship with the East of the island, the fineware forms found at the site underscore Koutsopetria’s overlap with centres of Western and Aegean trade elsewhere in Cyprus, highlighting the fact that the site’s prosperity was as much a part of local and regional successes elsewhere as any attempts it may have made to exploit the long-distance exchange markets; and the most common forms of ARS, CRS and PRS found throughout Syria and Israel-Palestine are quite often the same as those favoured at Kopetra, Petrera, and Koutsopetria, reinforcing the interconnectedness of the Eastern Mediterranean in Late Antiquity.[45]

A mid-sized settlement which was neither rural village, like Kopetra, nor urban centre, like Kition or Salamis, Koutsopetria stands out from the other Cypriot sites discussed in this paper. Yet it also shares distinct similarities with some of them, Kopetra in particular. The diverse ceramic corpus recovered from both sites underscores their participation in vibrant and larger exchange networks which extended well beyond the shores of Cyprus. Still, it seems probable that, like Kopetra, which was described as ‘economically and culturally subordinate to the large coastal cities of the island’,[46] and ‘a second-tier collection point for local products and a redistribution point for imported goods’,[47] Koutsopetria was not a significant nexus of direct long-distance exchange, but was instead heavily reliant on local exchange networks which both supported and were a result of the new prosperity in Late Antiquity. On an established road to Salamis, and possessing a substantial religious centre, Koutsopetria was perhaps a convenient stopping point for travelers and a useful midway point for inland villages to acquire and exchange goods arriving at the substantial harbour without the need to travel to Kition.[48] Goods produced in the Koutsopetria hinterland, destined for export from between the economic catchments of Kition and Salamis, could also have favoured the site as a point of transshipment. In wider Mediterranean networks, which linked urban and rural areas in the distribution of goods and ceramics, we may surmise therefore that Koutsopetria was fully connected to the wider Roman world via a ‘decentralised’ system or a ‘link’ within a larger ‘economic chain’ – a member of a local economy within a wider series of local, as well as larger, economies.[49] In addition, the gradual softening of the traditional urban-rural divide, where cities were ‘parasitic centres of consumption’, and the new emphasis instead on the important interactions between the two, offers some clues as to the potential role of a mid-size settlement like Koutsopetria in the redistribution of material which passed through there.[50] In this respect, the inferred relationship between Koutsopetria and local urban centres such as Kition or Salamis may find parallels in what has been suggested for the role of urban sites such as Antioch and Aqaba for the reciprocal distribution and export of local products.[51] Further work remains to be completed at the site, and it is our hope that its eventual publication will add in a small way to the ongoing study of a crucial period in the history of both Cyprus and the Near East.[slide]

Greg Fisher

Keble College

Acknowledgments

My thanks are due to William Caraher, R. Scott Moore, David Pettegrew, Michael Fronda, and Cheri Williams for reading drafts of this paper and offering their suggestions. The support of the Social Sciences and Humanities Research Council of Canada, McGill University, The Craven Fund, Keble College, The Keble Association, and the Canadian Centennial Scholarship Fund is gratefully acknowledged. 

Figures

Figure 1

Relative fineware proportions in Syria, Jordan and Palestine, drawn North to South, for sites discussed by Hayes (2001). Numbers indicate certain finds, either sherds, bases, or other diagnostics.

Bibliography