Island dynamics and Minoan expansion in the Aegean: the K ythera Island Project

In recent years archaeologists have become increasingly interested in the investigation of island societies. At a global level, discoveries in the Pacific, Caribbean, Mediterranean and elsewhere have greatly improved our understanding of the antiquity and dynamics of island life. Now archaeologists at the Institute, together with other colleagues, have embarked on a long-term interdisciplinary study of the island of Kythera in the Aegean.


I
sland archaeology in the Mediterra nean benefits from many unique advantages, not least the remarkable wealth of detailed archaeological data accumulated over the past cen tury or so, and the potential for integration with rich historical data over more than two-and-a-half millennia.Another partic ular challenge of Mediterranean archaeol ogy is to explore the range of insular life in a sea of islands that display vast diversity in geographical and cultural configuration and in the relationships between islands and the continental fringe. 1 The Kythera Island Project (KJP) aims to shed light on patterns of insular life in the Mediterra nean as exemplified by the past and present of the Aegean island of Kythera.

An island in the stream
The Greek island of Kythera lies in the southwestern corner of the Aegean Sea, between the rugged capes of the southern Peloponnese and the much larger island of Crete (Fig. 1).Its position midway between these two geographically and culturally distinct regions makes Kythera, and its much smaller neighbour Antikythera, an ideal location in which to study the role of an island as a stepping stone, but also a filter, in movements of people, animals, plants, objects and ideas.Kythera has indeed acted as such since an undeter mined time far back in the Ice Age, when the full-size ancestors of Crete's first bizarre and sadly now-extinct fauna of pygmy hip popotami, elephant and deer crossed fr om the Peloponnese via the island and its inter vening seaways.Much more recently, in 1941, the island's coasts served as a refuge for survivors from Royal Navy vessels sunk while patrolling the straits during the Battle of Crete, and even today ferry-links tie Kythera into the economies of western Crete and the southern Peloponnese.
In between these chronological poles, Kythera has witnessed the earnings and goings of a remarkable variety of people, cultures, states and empires, including pio neer settlers near the end of the Neolithic, later arrivals from Minoan Crete, possibly Phoenicians who (at least according to later tradition) introduced the cult of the eastern sex-goddess Aphrodite; rival soldiers from  Sparta and Athens, and the successive imperial emissaries of Rome, Byzantium, Venice and, for a few decades after the Napoleonic wars , Britain.Throughout this period a parallel and no less vital story has been the creation, maintenance and refor mation of local island identities, by people of long or recent residence on the island, and among Kytherans living abroad.

C:rKYTHERA
In 1998, for these and other similar reasons, the island was chosen as the focus of the KIP, an interdisciplinary programme of fi eld-and laboratory-based research that aims to explore the long-term dynamics of the island's cultural and natural history. 2 The KIP is based at the Institute in London, but the team also includes researchers from the Fitch Laboratory ofthe British School at Athens, from the universities of Sheffield, Oxford, Cambridge, and from Baylor Uni versity (Texas).The team's expertise covers archaeology, geoarchaeology, the scientific analysis of pottery, clays and metalworking debris, geographical information systems (GIS), vegetational analysis, historical geog raphy and social anthropology.This cov erage is soon to be extended by the addition of an island biogeographer, whose task will be to analyze patterns of fauna!coloniza tion and transference between the island and adjacent areas.

Discovering the Minoans ofK ythera
One of the most challenging questions addressed by the KIP is the pattern of Kythera's Bronze Age history.In this respect, our fi eldwork builds upon the foundations laid in the 1960s by the archaeologists Nicolas Coldstream (later Y ates Professor at UCL) and George Huxley, who excavated the island's prin cipal coastal entrep6t at the site of Kastri (Fig. 2). 3 This excavation was notable for uncovering startling evidence of an impor tant colony of Minoans (people from Bronze Age Crete) that underlay the later Classical, Roman and early Byzantine levels.This colony thrived in particular during the first half of the second millen nium BC (the Middle and early Late Bronze Age), a period that saw the Minoan palace states of Crete, centred at places such as Knossos and Phaistos (Fig. 1), expand their power across much of the southern Aegean, creating networks of trade and perhaps political dominance that secured the inflow of metals and other materials to the palatial centres.
The settlement at Kastri was contempo rary with other island sites that display marked Minoan influence, the most fa mous of which is Akrotiri (Fig. 1) on Thera (also known as Santorini) -which was preserved for posterity in a Pompeii-like condition by an eruption of the island's volcano.Intrigu ingly, on Kythera the lowest stratified level at Kastri, sitting on bedrock, revealed even earlier pottery of Cretan type, dating to the Early Bronze or pre-palatial period (c.2500-2000 BC), and broadly co-existent with nearby material of local type, whose "minoanized" -an admittedly unmelliflu ous term, yet one that has the virtue of strict neutrality concerning the ethnic and cul tural issues involved-began early, and they involved complex interactions with indig enous people already living on the island.
A further breakthrough in our under standing of minoanization on K ythera was made early in the 1990s by the leading Greek archaeologist Yiannis Sakellarakis, who led the excavation of a spectacular peak-top religious sanctuary of distinctive Minoan type on the summit of the mountain of Agios Georgios (St George), which dom inates Kastri, its coastal plain and much of inland Kythera (Fig. 2). 4 On Crete, such peak sanctuaries are considered to have played a key role in the integration, via shared ritual practices, of communities scattered across the territory controlled vis ually from the peak.The recent discovery at Agios Georgios therefore raised the possi bility of a Minoan-period sacred landscape extending across at least part of the island.More generally, it underlined how little was known about the process of minoanization on Kythera beyond specific sites -in short about minoanization as a phenomenon in the island's wider cultural landscapes.

Archaeological fieldwork 1998-99 5
The centrepiece of the KIP's archaeological fieldwork is a 4-year intensive field survey of the landscapes of central Kythera, cov ering all periods from earliest prehistory to the present.This was considered essential if isolated windows on the past -such as sites of Classical, Roman and Byzantine date -were to be placed within a broader regional context.Moreover, the experience of the past quarter century of research has established that such fi eld surveys are an essential component of any serious attempt to analyze the long-term history of Aegean societies and landscapes.Accord ingly, in the summers of 1998 and 1999, a group of 20-30 students began surveying the thorny sun-blasted terrain ofthe island by fieldwalking (Fig. 3).The group con sisted of postgraduates and undergradu ates from the Institute, as well as from other universities in the UK and abroad.Con scious of the need to build bridges between a large foreign archaeological team and local people, we place special emphasis on the participation of many Greek students, most of whom have themselves been trained at UCL.So far, slightly under 30km 2 of Kythera has been investigated in this way.The survey teams are made up of between four and seven people apiece, walkers in each team being spaced at 15 m intervals and traversing small areas known as tracts.6 Information on densities of archaeological material, together with a range of other ancient and modern data, are recorded for each tract.By 1999 some 3300 tracts had beenfi eldwalked (Fig. 4), creating a mosaic of data that facilitates the identification of archaeological sites.These sites are sub sequently investigated in more detail.usually by gridding and the controlled col lection of material.So far 86 archaeological sites have been encountered, the vast majority of which were unknown prior to the present survey.The climax of the sur vey to date has been surface investigation of Kastri itself by tract walking, gridding, kite-borne photography (Fig. 5) by Tim Cunningham of Louvain University, and detailed documentation of the site's rock cut cemeteries.The initial results suggest that Kastri was considerably larger than hitherto suspected, with an area of c. 9-10 ha, at least two thirds of which appears to have been occupied in the heyday of the Minoan colony, during the first half of the second millennium BC.
Several other aspects of our work are also well advanced.Sampling for thin section analysis under the microscope of pottery types found in each occupation phase at Kastri, and of the island's diverse clay sources, has been undertaken at the Fitch Laboratory.This is shedding new light on the technological transition from indigenous (probably originally Pelopon nesian) to Minoan pottery-making tradi tions at Kastri, on wider patterns of pottery making on Kythera, and on off-island trade.In the summer of 1999, a promising Kastri, Agios Georgios and other known Figure 3 Surveying a Ky theran landscape by fi eldwalking, summer 1999.start was made by our geoarchaeological team from Sheffield University, who began to identify indicators of sea-level change, as well as signs of soil movement probably triggered by human action and the earliest traces of terracing to conserve soils and reduce erosion.In addition, we have begun an analysis of the island's metal sources and of the evidence for ancient metalwork ing found during survey.Finally, mention must be made of the geographical information system used for the storage and analysis of data.It is designed and run by James Conolly and is used not only on Kythera but also as an MA training package in London.It allows sophisticated modelling of islanders' activ ities in three-dimensional space, the explo ration of intervisibility between the Minoan peak sanctuary and contemporary settle ments, and, more generally, it plays a key role in the continuing integration and inter rogation of the project's expanding dataset.

Preliminary results and prospects
After two field seasons, the KIP has already substantially expanded our knowledge of Kythera's early history.The island's first major horizon of settlement dates to the Early Bronze Age (c.3200-2000 BC), al though several clues indicate the presence of settlements by the last phases of the pre ceding Neolithic.The material culture associated with this horizon confirms that the first Kytherans came from the nearby Peloponnese.The first Minoans arrived at Kastri by c. 2500 BC, and interaction be tween them and indigenous communities in t; .heisland's interior is hinted at by finds of a few sherds of Minoan-type pottery at settlements that preserve a local!Pelopon nesian material culture.However, so far no further candidates for Minoan settlements of this date have been encountered.
The transition to the second millen nium BC, when the first Minoan palace states emerged in Crete, also saw the expansion of settlement at Kastri, but ini tially a complete emptying ofthe rest ofthe landscape.Within some three centuries, however, when Minoan palatial society reached its most elaborate form in the Neopalatial phase, the landscape of Kythera suddenly began to be re-settled with many small farmsteads, whose mate rial culture is clearly derived from that seen at Kastri and on Crete itself (Fig. 6).Full interpretation of this hitherto unrec ognized phenomenon must await more complete study of the material, but it seems likely to signal a further, and prob ably quite differently configured, horizon of colonization.The later second millen nium BC (including the Mycenaean period) saw another sharp reduction of settlement that lasted in its most extreme form until the Classical period, when the population ofKythera again expanded.These cycles of settlement expansion and collapse have continued fr om the aftermath of the Clas sical period right up to the present day.
The extreme oscillations in Kythera's fortunes bear witness to the plenitude of the island's history.Our research over the next few years will focus on the fuller doc umentation and explanation ofthese oscil lations, their integration with wider changes in the outside world, and their implications for shifting island identities and environments.This will involve con tinued fieldwork but also investigation of archival, anthropological, ecological and biogeographical data, in order to achieve the first holistic analysis of the dynamics of insularity on an Aegean island.

•Figure 1
Figure 1 The southwestern Aegean, showing th e location of th e island of Ky thera and the Bronze Age sites mentioned in the text.

Figure 2
Figure2View of the site of Ka stri, located on the promontory in the fo reground, with the peak sanctuary of Agios Georgios on the summit of the mountain in th e distance.ultimatestylistic origins lie to the north, on the Peloponnesian mainland.In exactly what sense this pre-palatial Minoan pottery makes Kastri the oldest Minoan colony in the Aegean remains open to debate.But clearly, the demographic and social proc esses by which Kastri and Kythera becan1e "minoanized" -an admittedly unmelliflu ous term, yet one that has the virtue of strict neutrality concerning the ethnic and cul tural issues involved-began early, and they involved complex interactions with indig enous people already living on the island.A further breakthrough in our under standing of minoanization on K ythera was made early in the 1990s by the leading Greek archaeologist Yiannis Sakellarakis, who led the excavation of a spectacular peak-top religious sanctuary of distinctive Minoan type on the summit of the mountain of Agios Georgios (St George), which dom inates Kastri, its coastal plain and much of inland Kythera (Fig.2). 4 On Crete, such peak sanctuaries are considered to have played a key role in the integration, via shared ritual practices, of communities scattered across the territory controlled vis ually from the peak.The recent discovery at Agios Georgios therefore raised the possi bility of a Minoan-period sacred landscape extending across at least part of the island.More generally, it underlined how little was known about the process of minoanization on Kythera beyond specific sites -in short about minoanization as a phenomenon in the island's wider cultural landscapes.

Figure 4
Figure 4 Map of Ky thera, generated by GIS, showing the extent of the survey area and of the tracts (solid black) within it that were fi eldwalked in 1998-99.

Figure 6 Figure 5
Figure 6 Minoan Neopalatial painted sherds from a surfa ce site in central Kyth era.