Introduction

The purpose of this paper is to discuss the processes of settlement and abandonment during the medieval period in the Nemea Valley, located in the northeast part of the Peloponnesos, in southern Greece. The Nemea Valley Archaeological Project (NVAP) exemplifies the contribution of landscape archaeology to the understanding of the Greek countryside. In the last twenty five years, regional survey projects have produced a rich surface record of prehistoric as well as historic times. However, interest in the archaeology of the “later” historic periods, especially the medieval and post-medieval, was slow to develop. The early survey projects focused on the material remains of more remote time-periods. For example, the Minnesota-Messenia Expedition (MME), a pioneering project that introduced the methods of archaeological survey to Greece, placed emphasis on the Bronze Age. MME served as a model for subsequent research. Although MME expressed interest in “later” periods, these were approached through the study of historical records rather than through material documentation (Topping 1972). Many practitioners at the time would suggest that the medieval and post-medieval pasts lay beyond the domain of archaeology. The existence of documents became a hindrance that delayed the development of material approaches to the study of the recent Greek past. How to handle documents and material remains has been the subject of intense debate in the field of historical archaeology since its inception (Leone 1988; Leone and Potter 1988; Little 1992; Paynter 2000; Schuyler 1978, 1988). Such debates have taken place only recently in the field of Greek archaeology (Kardulias 1994a, b; Small 1995). From its beginnings, Greek archaeology stressed the primacy of texts over material remains; its purpose was to provide the material evidence of the truth revealed in texts. Its subservient relationship to philology significantly influenced the direction of fieldwork (Athanassopoulos and Wandsnider 2004, pp. 1–2). Ivor Noël Hume’s (1964) well-known characterization of historical archaeology as a “handmaiden to history” is also applicable here.

Interest in the material record of the recent past has grown substantially in the last two decades. In contrast to the earlier projects of the 1950s–70s, the second generation of surveys, from the 1980s onwards, took an active interest in the archaeological documentation of the medieval and post-medieval countryside. Currently, most survey projects incorporate the medieval, post-medieval, and the early modern pasts in their research design and publications. Some of these projects include the Boeotia Archaeological Survey (Bintliff 1996, 2000; Bintliff and Snodgrass 1985; Vroom 2000, 2003), the Laconia Survey (Cavanagh et al. 1996, 2002), the Pylos Archaeological Project (Davis 1998; Davis et al. 1997), the Asea Valley Survey (Forsen and Forsen 2003), the Eastern Korinthia Archaeological Survey (Tartaron et al. 2006), the Australian Paliochora Archaeological Survey, and the Sydney Cyprus Survey Project (Given et al. 1999; Given and Knapp 2003). Several other recent projects in Greece and the Eastern Mediterranean are discussed by Susan Alcock and John Cherry (2004, pp. 244–248). These surveys have produced substantial bodies of medieval to early modern material remains and underlined the need for better chronological control. They have also encouraged the development of methods that integrate archival information with survey data. Thus, the archaeological surface record of the medieval and post-medieval times offers new opportunities and presents new challenges. There is a need to create a dialogue between the archaeological remains and the existing textual record, to develop context-specific narratives informed by historical approaches. Earlier presentations of the medieval and later components of archaeological surveys are quite limited in that regard. They treated the historical and archaeological records as two separate entities. It was political history that provided the textual frame while the archaeological data were expected to “fill in” the gaps of the historical record. Events such as the Slavic migration in the seventh century, or the coming of the “Franks” and the Fourth Crusade (1204) were viewed as key factors that presumably led to the decline of settlement and abandonment of the countryside. Increasingly, however, it has become clear that the archaeological remains are silent about the dramatic events of history. Thus, the Fourth Crusade did not lead to immediate shifts that can be observed in the material record of important regional centers such as Corinth. In fact, a significant change can be discerned several decades later, when Corinthian ceramic production declined and Italian imports predominated (Sanders 2000, 2003; Williams and Zervos 1994, 1995).

Archaeological surveys have revealed a much more complex picture of localized episodes of abandonment and resettlement. Their narratives emphasize processes such as the organization of agricultural production, interregional trade as well as political, and demographic factors. Shifts in the density and pattern of settlement reflect economic expansion and contraction, different agricultural strategies, as well as population fluctuations. Dispersed patterns have been associated with dense population, access to commercial markets, and intensive agricultural activity (Jameson et al. 1994). In contrast, economic contraction is linked with nucleation and subsistence agriculture (Runnels and Van Andel 1987). Furthermore, these shifts may reflect different patterns of land tenure. A dispersed pattern has been associated with small holdings, while a nucleated pattern may suggest that an elite minority had control of the land (Alcock 1993; Cherry et al. 1991; Davis 1991; Halstead 1987). Overall, the prevailing models of interpretation place emphasis on economic factors. Context specific narratives that integrate the archaeological remains and the existing textual record remain underdeveloped.

One of the promising frameworks for integrating different data sources, the textual and the archaeological, is the Annales approach, which recognizes the interrelation of three scales of time: the long term (longue durée), the medium term (conjoncture) and the short term (l’histoire événementielle). These three different time scales operate contemporaneously (Braudel 1972). In archaeology the Annales perspective has been applied to different cultural sequences (Bintliff 1991a; Iannone 2002; Knapp 1992; Staniforth 2003). The most attractive aspect of the Annales for archaeology is its de-emphasis of the historical event and the attention it pays to long-term historical processes. It is these qualities that make it particularly suitable for linking regional archaeological data with textual evidence. Survey data, overall, inform us about the long-term scale, the process of change usually seen in the countryside over thousands of years. The longue durée is a history of humans and their environment, “a history in which all change is slow, a history of constant repetition, ever-recurring cycles” (Braudel 1972, p. 20). The medium term corresponds to cycles of expansion and contraction operating in the landscape at different periodicities. Peregrine Horden and Nicholas Purcell (2000, p. 37) suggest that “conjonctures involve wars and economic systems; secular trends, in short, in every aspect of social history.” According to these authors, the cycles of the medium term last for anything up to fifty years. In contrast, John Bintliff (1991b, p. 7) suggests that the conjonctures operate over several generations or centuries. This longer span of time proposed by Bintliff is more useful for archaeology. Textual material belongs to the short-term, the time scale of a generation or so and supplies a political narrative, linked to historical events. It is “the history of particular persons and events, the study of traditional political and diplomatic narrative, the realm of individual consciousness in all its blindness” (Horden and Purcell 2000, p. 37).

The goal of this paper is to contribute to our understanding of the processes of settlement and abandonment of the medieval countryside by combining archaeological with textual information. Of interest here is the Annales temporal scale of the medium term, which corresponds to the social and economic cycles recorded in the historical and the archaeological records. The discussion will now turn to the Nemea region during the medieval period, from the eleventh to the fifteenth centuries. Medieval is preferred here as a chronological term, because it is broad and allows comparison of developments in Greece to other parts of the Mediterranean and Europe. The term “Byzantine,” which derives from the name of the Eastern Roman Empire, Byzantium, is more frequently used in the literature. Conventionally, the long history of Byzantium (324–1453) is subdivided into early (fourth to mid-seventh centuries), middle (mid-seventh to 1204) and late (1204–1453). Thus, the chronological focus here is the latter part of the middle and the late periods. In this discussion the term Byzantine is used to indicate political affinity rather than chronology. Next, the archaeological evidence for land use and habitation in Nemea will be examined, along with social, political, and economic processes which shaped the medieval countryside.

Nemea Valley: The Medieval Archaeological Evidence

The Nemea region has been the focus of archaeological research for several decades. Nemea is best known for the Sanctuary of Zeus, a Panhellenic Classical site, where the biennial Nemean Games took place. Like the Olympic Games, the Nemean Games included athletic competitions. Excavations in the Sanctuary area began in the 1970s and have revealed many structures including the Classical stadium, where the games were performed (Miller 1990).

The Nemea Valley Archaeological Project, undertaken from 1985–1990, consisted of the excavation of a prehistoric site, Tsoungiza, along with an intensive regional survey in the surrounding area. The goals of NVAP were to document human activity and habitation in the Nemea Valley and to study them in their ecological, economic, social, and historical contexts (Wright et al. 1990). The survey employed a diachronic perspective and sophisticated methodologies that recorded “sites,” as well as “off-site” material. NVAP “sites” refer to field units that produced a significant concentration of archaeological material. The density threshold used for the recognition of sites typically ranges between thirty to fifty sherds per one hundred square meters (Cherry et al. 1991, p. 46). In recent years the term “site” has fallen out of fashion and has been replaced by “findspot” or “place of special interest” (POSI). NVAP predates this development; the analyzed archaeological material derives from site as well as off-site collections.

In the Nemea area there is a proliferation of sites dating to the twelfth and thirteenth centuries (Fig. 1). They are dated on the basis of diagnostic ceramic wares, which include various styles of sgraffito, slip painted, green and brown painted, late sgraffito, glaze painted, matt-painted, and a small number of Italian and other imports. Also very common are plain glazed and cooking wares. The majority of these sites are located near the arable land, on the lower slopes of the hills surrounding the Nemea valley, and in smaller valleys in the southern part of the survey area. The Classical city of Phlious, in the northwest part of the survey, has a substantial medieval component, concentrated in the southwestern corner of the site.

Fig. 1
figure 1

Map of the Nemea Valley Survey with Medieval sites

Full size image

Two large sites (nos. 600 and 700) stand out, consisting of dense scatters of ceramics spread over several fields. Site 600 located near the Sanctuary of Zeus and the Classical stadium, covers an area of approximately 34 ha. Site 704, to the south, has an estimated size of 47 ha. Smaller sites vary in size; they usually cover less 1 ha. (e.g., site 602 = 0.7 ha; site 510 = 0.6 ha.). Sites 600 and 704 are contemporary with the majority of small sites dispersed throughout the studied area. Because of their size, they may be described best as the remnants of hamlets, established near good agricultural land. Both are located in the same general area, site 600 to the east of the Sanctuary of Zeus and site 704 on the Tretos Pass, just outside the Nemea valley.

At site 600 the highest densities of medieval material occur around a spring known locally as the “medieval” or “Turkish” fountain (Fig. 2). In addition to the survey material, the excavations at the Sanctuary have produced plentiful evidence of farming activities dating to the twelfth and thirteenth centuries, in the form of farming plots and an irrigation ditch (Miller 1975, pp. 155–157). Heavy concentrations of ceramics and coins were also found during the excavations of the north end of the Stadium (Miller 1975, pp. 162, 169, 1976, p. 194).

Fig. 2
figure 2

Map of Site 600 and vicinity

Full size image

The majority of the small sites scattered throughout the survey area probably represent isolated farms or field houses established near fertile lands. A specialized site, a pottery workshop (site 510) located in the southern part of the survey area, provides clear evidence of local ceramic production. Collections from site 510 include fragments of clay rods and thinner, curved hooks indicating the presence of one or more pottery kilns (Wright et al. 1990, fig. 27, c–h). The rods were inserted in the kiln walls to create shelves, and the thinner cylindrical pieces were used as hooks to suspend small items during firing (Naumann 1971; Papanikola-Bakirtzis et al. 1992, p. 32). Similar kiln debris is known from Corinth (Morgan 1942, pp. 21–22), and also Serres, in northern Greece (Papanikola-Bakirtzis 1999, pp. 222–223). Overall, the density of ceramic materials and settlement in the survey area indicate that the cultivated area was expanding during the twelfth and thirteenth centuries. These remnants of medieval habitation must reflect the intense level of agricultural activity in the Nemea region during this time.

Existing documentation supports this interpretation. Several historians have suggested that the eleventh, and in particular, the twelfth centuries were a time of unprecedented economic and demographic growth that encompassed the urban centers and the countryside of medieval Greece (Harvey 1989; Hendy 1985; Laiou 2002a; Lefort 2002). Archival records at the regional level are available only for some regions of Greece. The well preserved monastic archives of Mount Athos provide detailed information for parts of northern Greece. Unfortunately, the records of the central government, which were compiled for all Byzantine territories at frequent intervals, have not survived. Successive destructions of Constantinople, the Byzantine capital, are responsible for the disappearance of the centralized administrative records. An exception is the cadaster of Thebes, an official copy of a cadastral and fiscal register dated to the second half of the eleventh or the first quarter of the twelfth century (Lemerle 1979, p. 193; Svoronos 1959). This tax-register provides a glimpse at patterns of land use in central Greece. It reveals a complex pattern of land ownership with scattered, small land holdings and dense settlement. In most Byzantine territories, land increasingly came under the control of powerful landowners beginning in the tenth century. This process was kept in check temporarily by legislation that the strong rulers of the Macedonian dynasty introduced during the tenth and early eleventh centuries. Legislation was a response to the increasing economic and military strength of a powerful group of aristocrats who were seen as a threat to the central administration. However, the effort to maintain a more equitable distribution of land was not successful. Subsequent conflict between the civil and the military aristocracy weakened the state and contributed to significant losses of Byzantine territories—e.g., loss of Asia Minor to the Seljuk Turks in 1071, Norman expansion in southern Italy and the Balkan provinces. With the accession of the Komnenoi to the throne in 1081, the military aristocratic families emerged victorious and came to control the state apparatus. Alexios I Komnenos radically transformed the political structure and culture of Byzantium by relying on the imperial family and its allies for managing the state (Neville 2004, pp. 31–32). From that point onward, the growth of landed estates accelerated and transformed most peasants into dependent farmers. In some regions of Byzantium, entire villages and estates became property of privileged ecclesiastical or lay landowners, especially in the northern regions of Macedonia and Thrace. In the past, this feudalization process was viewed as the source of Byzantine economic and social decline (Ostrogorsky 1954). More recent scholarship has reversed this picture and suggests that the re-organization of agricultural production, represented by the growth of the large estate, was a stimulus that contributed to economic and demographic growth beginning in the eleventh century (Harvey 1989; Lefort 2002). However, the consolidation of land under powerful landowners may not have been uniform in all territories. For example, in the region covered by the cadaster of Thebes, the accumulation of property seems to have occurred in a more piecemeal way, resulting into a less consolidated pattern of land ownership. Overall, the view derived from the Theban tax-register shows increasing demand for land and intensification in agricultural production (Harvey 1982–83).

Agricultural production based on small to medium size properties, similar to the Theban example, fits well with the Nemea evidence. The pattern of residential dispersion in Nemea suggests that labor intensive agricultural strategies were employed, including the cultivation of gardens and probably vineyards. Furthermore, the significant amount of glazed pottery recovered from fields in the Nemea area provides direct evidence for the participation of rural households in trade networks. It indicates growth in the disposable income of the peasants, who were able to purchase expensive glazed tablewares. Economic growth in Nemea might be associated with commercial activities in the region’s major urban center, Corinth. Increased availability of trade opportunities may have contributed to the opening up of new fields and an increase in the area of land under cultivation in rural areas like Nemea.

Trade opportunities were, in turn, tied to political developments. Of interest here is the relation of Byzantium to Western Europe beginning in the eleventh century. Overall, Western Europe was entering a period of expansion, evident in the growing power of the Italian maritime cities of Venice, Genoa, and Pisa. In 1082, the Venetians were granted exemptions from Byzantine trade taxes as a reward for Emperor Alexios I Komnenos’ assistance against the Normans. Venetian commercial expansion grew steadily in Byzantine territories. Similar exemptions and privileges were soon granted to Genoa and Pisa. The Italian traders became very active during the twelfth century, and their commercial activities led to an increased volume of exchange that included luxury goods, such as silk, as well as foodstuffs and agricultural products (Laiou 2002c). The city of Corinth was one of the main Peloponnesian centers involved in interregional trade. For example, the Venetians exported large quantities of olive oil from Corinth (Harvey 1989, p. 147). Increased demand for produce coincided with the intensification of agricultural production. Over time, trade dominated by the Italian city-states brought the Eastern Mediterranean region in close contact with Western European markets.

The interaction between East and West took a different form in the late eleventh century due to the Crusades. The Western Crusaders succeeded in establishing small independent polities in the East, based on the feudal model imposed on lands re-gained from the Muslims. Increasingly, however, the relation between the West and Byzantium became antagonistic. The turning point was the Fourth Crusade, which, under Venetian leadership, resulted in the capture of Constantinople in 1204. The Latin capture of the city and the breakdown of the centralized Byzantine state led to political fragmentation. Several small political units were created and replaced Byzantium. The main ones were the Latin Empire of Constantinople, the Principality of Achaia (Peloponnesos), several Venetian colonies established in coastal areas (Coron, Modon, Crete, Euboea, some Aegean Islands, the Ionian Islands), the Despotate of Epiros, the Empire of Nicaea, and the Empire of Trebizond (Laiou 2002b, p. 24). Constantinople was recaptured by the Greek ruler of Nicaea in 1261 and the Byzantine state was revived. In the Peloponnesos (known as the Morea during the late medieval period) several fortresses were recaptured and parts of the peninsula returned to Greek control. The diminished Byzantine state, however, was not able to function as a unifying force. It was immersed in dynastic quarrels and civil wars, which contributed to its continuing decline in the fourteenth and fifteenth centuries (Reinert 2002, pp. 263–270). In the mid-fourteenth century Byzantine territories in the Peloponnesos were brought together and formed the Despotate of the Morea, a successful polity that lasted until 1460, when it was conquered by the Ottoman Turks (Runciman 1980).

Thus, the political and social structure of medieval Greece became extremely fragmented and decentralized after the Latin conquest of 1204; prolonged conflict, civil wars, and the plague seem to have taken a heavy toll on the rural population. There are numerous narrative and documentary sources that point to demographic decline and impoverishment in the countryside (Longnon and Topping 1969; Panagiotopoulos 1985). Starting in 1347, the Black Death spread throughout Byzantium and Western Europe and led to a severe demographic crisis. The effects of the plague in the Morea can be deduced from documents of the mid-fourteenth century concerning fiefs belonging to Italian feudal lords, the Acciaiuoli. The documents mention a significant number of deserted staseis (households) in villages located in the southern and western parts of the peninsula (Longnon and Topping 1969, pp. 67–68, 266–267). In addition, Venetian sources discuss the devastating effects of the plague in their colonies of Modon and Coron (Panagiotopoulos 1985, pp. 61–68).

A drastic change took place in the Nemea region during the late thirteenth century when archaeological evidence of settlement and agricultural activity becomes scarce. A complex of sites on a precipitous hill, identified in historical sources as Polyphengi, dominates the region (Fig. 3). The steep hill controlled the passes to the south and the west, routes of significance in the medieval period. On the eastern side of the hill, an impressive post-medieval monastery, Panagia tou Vrachou, still stands against the cliff. The summit is fortified by a wall and a square tower (Fig. 4). A double-arched cistern is located near the tower, at the edge of the cliff. The ruins of a village spread over a sloping plateau below the summit. On the south side of the hill, at the level of the settlement, there is a fortified cave-chapel with a partially preserved wall painting depicting the Presentation in the Temple (Figs. 5 and 6). This may be the site of the earliest church built on Polyphengi, which is first mentioned in sources dating to 1402 (Kordosis 1981, p. 368). Diagnostic pottery from the settlement and the lower slopes of Polyphengi includes Italian imports (protomajolica, Italian painted sgraffito, majolica) ranging in date from the late thirteenth to the fifteenth centuries (Fig. 7).

Fig. 3
figure 3

View of mountain Polyphengi

Full size image
Fig. 4
figure 4

View of tower on summit of Mt. Polyphengi

Full size image
Fig. 5
figure 5

View of Mt. Polyphengi and fortified cave-chapel

Full size image
Fig. 6
figure 6

Preserved wall painting in rock shelter depicting the presentation at the Temple

Full size image
Fig. 7
figure 7

Pottery from settlement on Mt. Polyphengi

Full size image

The earliest reference to the castle/mountain occurs in a Western source, the French version of the Chronicle of the Morea, under the name Polifant. In several other historical sources, it is mentioned under a number of different names: Polifant, Agios Georgios, San Giorgio, Sancto Giorgio de Polisengo, S. Zorzi Tropico (Kordosis 1981, pp. 179, 183). Clearly, its location on the main route connecting the western and central Morea to Corinth was the main reason for the growth of Polyphengi in the Late Medieval period. The castle was mostly under Western control and served as the center of a small area for the collection of taxes owed to the Castellany of Corinth. Polyphengi is also mentioned in Venetian documents of the second half of the fifteenth century. It is likely that the fort and town were forcefully taken by the Ottomans, who then transferred inhabitants from the Nemea area and Arcadia, in central Peloponnesos, to the vicinity of Constantinople. Chalkokondyles, the Byzantine historian who describes the Ottoman conquest of the Peloponnesos during 1457–1460, does not mention Polyphengi by name in his account. However, his description of the conquest of a town located on a precipitous mountain most likely refers to it. The topography, the route followed by the Ottoman army, and other features seem to fit the specifics of the historical account (Kordosis 1981, pp. 180–183).

The pattern of settlement in the late thirteenth to the fifteenth centuries provides a stark contrast to habitation during the twelfth to thirteenth centuries in the Nemea region. As mentioned before, the Late Medieval period was a time of political fragmentation, continuous competition and conflict. Thus, defense and security played an important role in the choice and location of settlement. Many settlements developed around pre-existing or newly established kastra (castles) that provided relative security from frequent incursions. There are numerous examples in the Peloponnesos, the best known site being Mystras, which became the refuge of the Byzantine aristocracy and capital of the Despotate of the Morea. Late Byzantine kastra-villages share a number of common characteristics. They are built on inaccessible hilltops with commanding views, around a “keep,” a small fortified area, which contains a tower and often a cistern. The village extends below the keep with free standing houses built along parallel contour lines. The settlements vary in size, ranging from fifteen to 300 houses (Kourelis 2002, pp. 55–56; Sigalos 2003, pp. 211–217). In the district of Corinthia, where Nemea belongs, there are several known kastra-settlements; a few have been investigated by archaeologists, notably the sites of Agios Vasilios (Kardulias 1997), and Mount Tsalika (Gregory 1996). A late medieval fortified settlement has been excavated at Panakton, in central Greece. The village was inhabited for less than a century, from the middle of the fourteenth to the early fifteenth century. It consisted of at least thirty houses built on a series of terraces covering an area of approximately a hectare. The community was built around a towered stronghold (Gerstel et al. 2003). The site of Polyphengi is comparable to Panakton in terms of layout and size.

Thus, the conditions of the Late Medieval period led to the abandonment of farmsteads and the concentration of the population in nucleated, fortified settlements, a process similar to incastellamento (the growth of fortified sites on hilltops) that took place in Italy during the tenth to eleventh centuries. Besides offering security, hilltop sites served other functions, they were important for the collection of taxes and reflect increased control of local populations; they played a role in the creation and reproduction of relations of power. John Moreland (1992) views the creation of castellos as an effort to reinforce territorial and political control through the consolidation of lands and the aggregation of people. In the Peloponessos, the nucleation of settlement was a response to conditions of insecurity, conflict, and warfare, which increased in the late thirteenth century and intensified in the fourteenth and fifteenth centuries. These fortified settlements served as central points in the efforts of competing powers to establish territorial control.

The settlement trends outlined here are manifestations of different political, social, and economic processes. The abundant evidence for dispersed habitation in the Nemea area in the twelfth to thirteenth centuries reflects the intense level of agricultural activity during a period of demographic growth and expansion of trade. The drastic change seen in the late thirteenth century, when nucleation of settlement became the norm, is the result of insecurity, conflict, and the extreme fragmentation of the social and political structure of the Peloponnesos after the Latin conquest. These trends fit well into the Braudelian medium-term scale, the conjuncture; they correspond to cycles of expansion and contraction, settlement and abandonment operating in the landscape. Thus, the archaeological data provided by regional survey projects, including NVAP, have added a material dimension to historical narratives and opened new paths to the study of the medieval and post-medieval Greek countryside.