Archaeological Evidence from the Island of Susac, Croatia: Preliminary Report
Philippe Della Casa & Bryon Bass
Dept. of Prehistory, University of Zurich

Abstract

This paper briefly details the Neolithic sites recently documented on the distant Croatian island of Susac. The island had seen no systematic archaeological survey before the 1997 research expedition. Lithic artifacts such as blades, scrapers, glossed sickle blades and points suggest exploitation and production of local (Adriatic) flint resources. Depleted cores and abundant debitage indicate at on-site flint reduction and tool preparation. Attributes of the Impressed (Impresso) and Middle Neolithic Painted pottery assemblages from Susac are described. Susac provides an important link in the island chain that spans the east-west axis of the Adriatic Basin and coupled with other Neolithic insular sites, expands the concept of early trans-Adriatic travel and blue water resource exploitations.

1. Introduction

The 1997, 1998 and 2000 Susac Expeditions were conducted by the multinational members of the Korcula Archaeological Research Group KARG (Bass et al. 1997; Bass 1998; Radic et al. 1998; 2000). Logistical elements such as fresh water, food re-supply, and transportation limited the seasons on Susac to general surface survey and test excavations on identified archaeological sites. While systematic archaeological survey had not been conducted on Susac prior to the 1997 expedition, a Roman-Palaeochristian-Medieval structural complex near the Sv. Mihovil church had been visited on previous occasions (1980; 1986 and Rismondo 1990). However, these investigations were limited to that locality.

2. The island of Susac

Scientific and general literary references pertaining to Susac are limited. The island was not specifically mentioned by either Greek or Roman historical geographers. Susac is mentioned in early historic references, but only as a periphery island of minimal interest. For example, Susac was featured on Medieval sailing charts and apparently functioned as a useful navigational aid for sailors en route to other maritime destinations (Kozlicic 1997). The waters near this obscure island were often disputed as commercial Venetian fishing fleets and local islanders competed for access to the rich fishing grounds (Novak 1952). Most later historical descriptions of the Susac tend to be extremely brief (e.g. Burton 1879) and provide nothing more than general and often vague geographic details.

Location
Susac lies off the central Dalmatian Coast of Croatia and is situated within the southern island group along with Korcula, Lastovo, and Mjlet (figure 1). The "line-of-sight" connections to the nearby islands, as noted elsewhere in the Adriactic (e.g. Forenbaher et al. 1994), also apply to Susac. On most days, the islands of Vis, Hvar, Korcula, and Lastovo are visible to the north and west. The remote island of Palagruza is usually visible to the south west and with the right atmospheric conditions, the faint coastline of Italy's Promontorio del Gargano can be seen on the far western horizon. Limited sections of the southern peaks of the Dinaric Alps are just visible on the Croatian mainland, although the near-shore islands obscure most of the view. Visibility categories recently applied to islands in the Mediterranean Basin can be assigned to Susac (Patton 1996). Following the model set out by Patton, Susac would best qualify as a "B" category island (i.e. can be reached without sailing out of the sight of land) (see Bass 1998).

Present inhabitants
Operational personnel occupy the lighthouse situated on the southern promontory, and a sheep and goat herder maintains a farmstead near the Sv. Mihovil church. One footpath roughly extends across the island, but the main thoroughfare is between the lighthouse and the farmstead. Susac rarely has more than five residents at any given time, although this can fluctuate as family members visit or lighthouse caretakers rotate their duties. The extensive agricultural terracing found on the larger islands and mainland is absent, although a few drystone walls for historical land divisions and terraces for drying fishing nets are present (Novak 1952).

Geology, relief, vegetation
Susac's landmass covers roughly 4.6 sq. km, with a width varying between 0.3 and 1.8 km. The highest elevation is Susac vrh, situated at 239 meters above sea level. Tectonically, the island is in the south Adriatic unit of a palaeo-orogenic complex, now comprised of subsiding ridges. The northern section of the island is an Upper Jurassic formation, while the southern promontory, with its connecting lowland saddle, is Lower Cretaceous. The soils vary in quality and type, and there is abundant supply of limestone breccia. A freshwater lokva, or seasonal pond, is found above of the drainage leading to the bay at Dol. The submarine topography rapidly drops off from the island's coastline, with the greatest near-shore depths off the northern coast. In certain locations, the depths plunge beyond the 90-meter mark less than one hundred meters from the island's current shoreline.
Strictly for descriptive purposes, Susac can be divided into three topographic areas: the northern hills, the lowland saddle, and the southern promontory (figure 2). The northern hills have a fairly gradual relief. Dense vegetation covers nearly all of the hilltops and surrounding slopes and land survey in this area proved to be quite difficult. The lowland saddle stretches from the Sv. Mihovil church to the southern slopes of the bay at Uvala Duga. Most of the eastern shoreline, from Dol to Uvala Duga, has a gradual relief. The shore zone in some areas is cobble and gravel, but most of the inter-tidal zones are comprised of jagged, wave-cut limestone. This is in direct contrast with the western shoreline of the lowland saddle, which consists mainly of sheer cliffs and steeply sloped escarpment. The peak on the southern promontory rises to ca. 80 masl. The island's lighthouse, built in the nineteenth century during the period of Austro-Hungarian control, is situated at the top of the promontory. The cliffs surrounding the promontory are sheer and plunge directly into the sea.
The vegetation on Susac consists primarily of pistacia (Pistacia lentiscus), juniper (Juniperus phoenicia), thistle (Carthamus sp.), asparagus (Asparagus acutifolius), and rosemary (Rosmarinus officinalis), but more than twenty five floral species have been collected by the KARG for identification. Wild olive trees and the makija scrub brush also thrive on the central and northern sections of the island. Terrestrial faunas include numerous species of birds, lizards, mice, and rats. Freshwater for human consumption does not exist, so the inhabitants collect rainwater in cisterns cut out of the limestone.

3. Neolithic finds (1997, 1998 and 2000 Susac Expedition)

Several sites with Neolithic finds were recorded during the 1997 survey, and two of them investigated during the 1998 and 2000 seasons (Figure 2). The following discussion serves only as a preliminary presentation of the material. Further research and publications concerning these materials are forthcoming.

SU027
Although this site was documented during the 1997 season, the Early and Middle Neolithic element did not come to light until the test excavations of the 1998 season. The site is situated on a small rise near the modern farmstead and consists of an extremely dense artifact scatter that extends beyond a few hundred square meters. At the center of the site, the surface scatter contains a maximum of 40 flint finds (debitage, tool fragments) and at least 140 fragments of prehistoric pottery per sq m. Numerous complete tools were found, including one polished axe (so called "green stone"). Eight obsidian fragments were also sampled on the surface of the site and examined for provenience by Robert Tykot (University of South Florida).
Dry stone walls run around the central part of the site creating an artificial terracing towards the sea side and some sort of delimitation or fortification on the inland side (figure 3). As ascertained by the 2000 test excavations, the oldest of these stone walls belong to the Neolithic period, while later structures might be of Roman or later date.
Surface artifacts contain diagnostic elements from the Early Neolithic, Middle Neolithic, Copper Age, Bronze Age, and local Iron Age. The mixed nature of the surface finds suggests that the site has deflated. Test excavations during 1998 support this theory, as the deposit with similar multi-period artifacts proved to be extremely shallow and poorly stratified in unprotected zones of the site. However, the sub-surface tests in 2000 led to the discovery of reasonably stratified deposits under the tumble of an ancient wall structure, containing a mixture of terrestrial and marine shells, animal bones and diagnostic artifacts (Trichromatic Painted pottery, flint and obsidian tools) that relate to Middle Neolithic contexts in Korcula (Vela Spilja; see Cecuk and Radic 1995), in Italy and Greece, as well as some Early Neolithic Impressed Ware sherds (figure 4). This primary midden deposit (or stratified secondary deposit) is dated to around 4800 Cal BC by radiocarbon (5895±65 BP).
Animal bones from the Middle Neolithic deposit comprise both domesticated species (sheep and/or goat) and sea mammals such as otter and seal along with sea birds and fish. There is a broad variety of mollusc remains (Patella coerulea, Monodonta, Cepaea nemoralis, Theba pisana).

SU002
Contrary to the situation at SU027, the integrity of the Early Neolithic evidence at this site appears reasonably secure. The pottery assemblage relates to the earliest phases of the so-called "Adriatic variant" of the Early Neolithic impressed pottery cultures, correlating to a radiocarbon date of around 5800 Cal BC (6925±65 BP) from one of the test units. Finds from other periods (i.e. later prehistoric, Greek, Roman, or Medieval) were not detected in the immediate vicinity.
The site is positioned in a watershed that slopes down to the western end of Uvala Duga (figure 5). The eastern limits of the site are ca. 50 meters from the high tide mark. Limestone breccia is scattered throughout the area and recent pedologic transport does not appear to be dynamic. Goat grazing seems to have had an impact on some of the artifacts, as indicated by freshly broken sherds. During the 1997 season, a grid was laid across the surface scatter and individual artifact positions were plotted. Pottery that did not appear to be directly threatened by trampling was left in context.
During the 1998 and 2000 seasons, test units were excavated to investigate the subsurface integrity of the site. Only one unit contained evidence of a primary deposit, and it appears that a proportion of the site is in an erosional deposition within the gully. Charcoal, marine shell and land snail, along with an abundance of Impresso sherds, debitage, and tool fragments, were also sampled in the excavations.
The wares present at SU002 (figure 6) all fall into the "A" phase of the chronology developed for East Adriatic, Early Neolithic impressed pottery by Mueller (1988 and 1991; 1994). Numerous motifs have been assigned to the A1, or earlier phase. Fingernail and pinched impressions, arranged in either wide or narrow horizontal patterns, are common. Impressions from the wavy edges of Cardium shells (so-called Cardial Impressed Wares) and the smooth arched edges of Pectunculus shells (Glycimeris glycimeris), usually arranged in single and double rows covering most of the exterior surface of the vessel, are found. Triangular, circular, and semicircular stamped impressions, arranged in rough rows and applied with the tips of lithic, wood, and bone implements, are frequent. In many examples, especially the earliest wares, the impressions have been applied to the surface of the vessel in a random manner.
Classification of the sherds by fabric generally corresponds with the three categories noted at the Early Neolithic site at Tinj-Podlivade (Chapman and Mueller 1990; Chapman et al. 1996). Numerous base and rim fragments were sampled, including thin-walled sherds and rims with lip impressions. A few monochrome sherds are also present, but these form a small minority of the SU002 assemblage. The fragments are from recognized Dalmatian Early Neolithic forms such as dishes, bowls, globular vessels, straight-sided vessels, and miniature vessels (Mueller 1991 and Chapman et al. 1996). A few lug fragments are found in the assemblage.
The lithics from SU002 form an interesting and broad Early Neolithic "tool kit" (figure 7). Retouched primary and secondary flakes are present. As noted at other Early Neolithic sites in Dalmatia, the assemblage does have a strong tendency towards blades. Examples occur with and without retouch, and some of the final products appear as blade-based end scrapers. Two sickle blades and two small polished river pebbles were also collected at the site. Finally, bilaterally retouched or damaged points, possibly used for opening molluscs (somewhat reminiscent of the Sipontian assemblage from Coppa Nevigata [Puglisi 1955]), are present.

4. Preliminary Discussion

Depleted cores found at SU002 and a large flint nodule from SU027 raise the issue regarding sources for the lithic materials. Although source material has yet to be located on Susac, it is interesting to note some lithic samples from both SU002 and SU027 resemble the general physical descriptions of the cherts from a nodular and tabular flint source recently confirmed on Palagruza (figure 8). Furthermore, the presence of an Early Neolithic site on Palagruza (Forenbaher et al. 1994; Kaiser and Kirigin 1994; Kirigin 1995; Forenbaher and Kaiser 1997; Kaiser and Forenbaher 1999) in conjunction with the source materials (cf. Marchesetti 1876; Burton 1879) indicates a possible supply for the Susac materials.
However, and even more interesting when investigating long distance exchange networks, a Gargano (and thus Italian) provenance of a larger part of the lithic material including the nodule from SU027 has recently been suggested by petrographical micro-facies analysis (Jehanne Affolter, University of Neuchatel). Furthermore, Robert Tykot's analysis of obsidian fragments from SU027 indicates a Liparian origin of all the pieces sampled so far (figure 9).
The presence of sickle blades in a remote insular context is intriguing. Glossed sickle blades are known from the Early Neolithic site at Prato Don Michele (Tine's Phase I, Mueller's Impresso A) on the Tremiti Islands (see Whitehouse 1987). However, the Tremiti site is described as a settlement and at this stage, there is little evidence to support a similar occupation on Susac. Therefore, if Susac was visited for the purpose of harvesting wild flora and fauna, then an alternative picture of maritime exploitations may be drawn.
The situation obviously changed in the Middle Neolithic, with domesticated animals being introduced to structured settlements on the island, while there is still ample evidence for the exploitation of maritime resources.
The pottery from the Early Neolithic sites is confined to the "A" phase of the Early Neolithic chronology. Generally, Dalmatian sites with corresponding pottery indicate a generous ca. 5500-6500 Cal BC time range. However, sites closer to Susac, such as Vela Spilja on Korcula or Gudnja pecina on the Peljesac Peninsula, fall within the 6000-6500 Cal BC range (Chapman 1988; Chapman and Mueller 1990, for uncalibrated dates and available laboratory numbers). This implies that sites closer to the "island bridge" (figure 10) linking the opposing Adriatic shores exhibit earlier dates. Obviously this requires further substantiation.
The concept of an "island bridge" is not new to the literature, but the evidence from Susac, in conjunction with Vis and Palagruza, now confirms deep-water ventures into the Adriatic during the Early and Middle Neolithic. It is now possible to incorporate evidence from the "island bridge" into a broader archaeological and historical framework (Kaiser and Forenbaher 1999; Bass 1998).


Acknowledgements

The KARG maintains a research station at the Centar za kulturu-Arheoloski muzej, in the town of Vela Luka on the island of Korcula. The authors would like to thank the Opcina Vela Luka for their invaluable and continued support. They are very grateful to the Kommission zur Foerderung der wissenschaftlichen Forschung an der Universitaet Zuerich for the financial aid to the 2000 Susac season. The team members of the 1997, 1998 and 2000 Susac Expedition were: Prof. Dinko Radic, Prof. Branko Kirigin, Dr. Anka Radic, Dr. Zoe Bass, Tea Katunaric, Fredy von Wyl, Anton Bacic, Eric Rupley, Georgia Mortzou, Luke Adams, Helena Tomas, Marina Perica, Mario Farcic, Ante Mirosevic, Emanuela Jochum, Adrian Huber and Irene Ebneter.


Bibliographical References

Bass, B. 1998.
Early Neolithic offshore accounts: remote islands, maritime exploitations, and the trans-Adriatic cultural network. Journal of Mediterranean Archaeology 11 (2): 37-62.

Bass, B., Kirigin, B. and Radic, D. 1997.
Susac '97. Obavijesti Hrvatskog arheoloskog drustva 29/3 (Zagreb), 70-72.

Burton, R. F. 1879.
A Visit to Lissa and Pelagosa. Journal of the Royal Geographic Society 49, 151-190.

Cecuk, B. and Radic, D. 1995.
Vela Spilja: Pretpovijest otoka Korcule. Katalog Izlozbe. Dubrovnik: Knezev Dvor.

Chapman, J. 1988.
Ceramic Production and Social Differentiation: The Dalmatian Neolithic and the Western Mediterranean. Journal of Mediterranean Archaeology 1 (2): 3-25.

Chapman, J. and Mueller, J. 1990.
Early Farmers in the Mediterranean Basin: the Dalmatian evidence. Antiquity 64: 127-134.

Chapman, J.C., Shiel R., and Batovic, S. 1996.
The Changing Face of Dalmatia. Leicester University Press and The Society of Antiquaries, London.

Fiskovic. I. 1980.
O ranokrsçanskim spomenicima naronitanskog podrucja. Izdanja Hrvatskog arheoloskog drustva 5 (Split), 213-256.

Fiskovic, I. 1986.
Prilog proucavanju porijekla predromanicke skulpture na juznom Jadranu. Starohrvatska prosvjeta ser. 3/15 (Split), 133-163.

Forenbaher, S. and Kaiser, T. 1997.
Palagruza, jadranski moreplovci i njihova kamena industrija na prijelazu iz bakrenog u broncano doba. Opuscula ArchÊologica 21 (Zagreb), 15-28.

Forenbaher, S., Gaffney, V., Hayes, J., Kaiser, T., Kirigin, B., Leach, P. and Vujnovic, N. 1994.
Hvar, Vis, Palagruza: A Preliminary Report on the Adriatic Island Project. Vjesnik za arheologiju i historiju dalmatinsku 86 (Split), 13-52.

Kaiser, T. and Forenbaher, S. 1999.
Adriatic Sailors and Stone Knappers: Palagruza in the Third Millennium BC. Anitquity 73, 313-324.

Kaiser, T. and Kirigin, B. 1994.
Palagruza, arheolosko srce Jadrana. Arheo 16 (Ljubljana), 65-71.

Kozlicic, M. 1997.
Plovidba Jadran 14.-16. stoljeca u putnickim izvjescima. Radovi-Razdio povijesnih znanosti 35/22 (1995/1996-Zadar), 257-279.

Marchesetti, C. de. 1876.
Descrizione dell'isola di Pelagosa. Bolletino della Societ‡ Adriatica di scienze naturali 2 (Trieste), 283-306.

Mueller, J. 1988.
Cultural Definition of the Early Neolithic and its Interaction in the Eastern Adriatic. Berytus 36, 101-125.

Mueller, J. 1991.
Die ostadrische Impresso-Kultur: Zeitliche Gliederung und kulturelle Einbindung. Germania 69/2, 311-358.

Mueller, J. 1994.
Das ostadrische Fruehneolithikum-Die Impresso-Kultur und die Neolithisierung des Adriaraumes. Praehistorische Archaeologie in Suedosteuropa-Band 9. Berlin: Wissenschaftsverlag Volker Spiess.

Novak, G. 1952.
Dokumenti za povijest ribarstva na istocnoj obali Jadranskoga mora - (1) Otoci Vis i Hvar. Zagreb: Jugoslavenske Akademije Znanosti i Umjetnosti.

Patton, M. 1996.
Islands in Time-Island Sociogeography and Mediterranean Prehistory. London: Routledge.

Puglisi, S. 1955.
Industria microlitica nei livelli a ceramica impressa di Coppa Nevigata. Revista di scienze preistoriche 10, 19-37.

Radic, D., Bass, B. and Della Casa, P. 1998
Arheoloska istrazivanja na otoku Suscu: sezona 1998. Obavijesti Hrvatsko Arheolosko Drustva 30 (3): 55-59.

Radic, D., Bass, B. and Della Casa, P. 2000
Arheoloska istrazivanja na otoku Suscu: sezona 2000. Obavijesti Hrvatsko Arheolosko Drustva 32 (3): 59-62.

Rismondo, T. 1990.
Arheoloski ostaci s otoka Susca. Obavijesti Hrvatskog arheoloskog drustva 22/3 (Zagreb), 37-39.

Whitehouse, R. 1987.
The First Farmers in the Adriatic and their Position in the Neolithic of the Mediterranean. Premieres Communautes Paysannes en Mediterranee occidentale. Colloque International du C.N.R.S., Montpellier (1983). 357-365. Paris (1987).