WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION

Between sea and mountain: the archaeology of a 20th century “small world” in the upland basins of the southeastern Korinthia
Abstract Between the villages of Sophiko and Korphos in the southeastern Korinthia are a number of geographically well-defined and fertile upland basins or poljes, each one accompanied in modern times by a cluster of farmsteads and used for agriculture and pastoral activities. The heavily forested slopes adjacent to these basins were systematically exploited for resin production, a flourishing industry in the wider region especially after World War II, which is now in serious decline. Although physically isolated from major urban centers, the microecologies represented by these settlements played a vital role in the 20th century in the subsistence of its local population, which originated primarily in the nearby mountainous village of Sophiko. Placing these isolated, yet deeply interconnected places into their regional context provides another key case-study for the contingent character of the Greek countryside in the 19th and 20th century. Between 2001 and 2009, the authors investigated these basins, with a primary focus on the largest, known locally as Lakka Skoutara, through two archaeological projects: the Eastern Korinthia Archaeological Survey (2001-2003) and the Saronic Harbors Archaeological Research Project (2008-09). The former studied Lakka Skoutara as part of its emphasis on the archaeology of the modern period (19th-20th centuries), while the latter conducted archaeological investigations in several of these basins as part of a larger regional survey of the Saronic coastline. Lakka Skoutara presented a remarkably robust assemblage of material including domestic and religious architecture, agricultural installations, and ceramics scatters. This material reflects the dynamism of changing land use patterns in the Greek rural landscape as well as the formation processes and life cycles of use, reuse, and abandonment connected to domestic residence. By combining archaeological survey with oral information obtained from local residents, we were able to reconstruct part of the landscape history of this small, low-density rural settlement and its relationship to the wider world. This micro-level analysis of the site complements the broader perspectives offered by regional level data collection, oral history, and comparative studies from elsewhere in Greece. Fieldwork at Lakka Skoutara and its neighboring poljes foreground the historical processes affecting the archaeology of the countryside over the last two millennia by 1

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION complementing the traditional practices of intensive pedestrian archaeology with methods that document the dynamic archaeological environments visible in contemporary countryside recorded over the much narrower horizon of a decade of field work.

2

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION I. Introduction The Lakka Skoutara basin is located some xx km from the coast in the Southestern Corinthia. It is one of a number of basins that stretch in a north-south string from the village of Sophiko to the Saronic coast near the monastery of the Panayia of Steiri. The basins, locally called lakkas, are surrounded by steep slopes which are wooded today, but were clearly terraced in earlier times. Lakka Skoutara is bounded to the north by a ridge marked Rachi Zarakas on the Greek Army Mapping Service 1:5000 maps and topped by a geodetic marker of 517.02. The height of Rouxthi forms the western boundary of the Lakka Skoutara. A deep ravine called Zastani Rema runs from the northwest to the southeast and provides access to the coast to the south of the village of Korphos. The eastern boundary of the Lakka is composed of the steep hill of Prixea. The modern route to Lakka Skoutara comes from a dirt road accessible from the main Corinth-Epidauros highway. The road passes through a broad and relatively shallow lakka called Vathy before curling its way over the western side of Rachi Zarakas and descending into the Lakka Skoutara basin. The route of this modern road probably passes at a lower elevation than its premodern predecessor which seems to have kept slightly higher on the slope of the ridge and allowed for a more gentle route over the western part of Rachi Zarakas and more gradual descent into the basin itself. The road then continues east through three more lakkas before emerging to the west of the Byzantine church of the Panayia of Steiri. History of Work in the Region Early travelers rarely passed through the southeastern Corinthia, and even less frequently through the hilly region between Sophiko and the coast. In fact, only the relatively sizeable villages of Sophiko and Korphos appear the early maps of the area prepared by the Expédition scientifique de Morée. The census conducted by the same expédition, however, records a number of places inhabited by only a few families in the Corinthia suggesting that the surveyors were aware that small places on the scale of Lakka Skoutara did exist. While it is possible that one of the small places recorded in the French census was, in fact, the small settlement at Lakka Skoutara, there is no obvious link between the places recorded in their list and Lakka Skoutara. It may be that the area escaped the careful eye of 19th century geographers and topographers because it was likely only accessible via a monopati. By the late 19th century, and presumably earlier, the main cart path from Sophiko to Korphos passed to the south of the string of upland basins. A. Miliarakis description of the ancient and modern political geography of the area 3

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION published in 1886 recorded the cart road from Korphos to Sophiko as running through the Larisi basin and by the Byzantine church of the Taxiarchis (Miliarakis 1886, p. 142; Dixon 2000, p. 69). Even the 1893 discovery of the “Find of Sophiko” a group of almost 1000 late 3rd century B.C. coins in Korphos harbor by sponge fisherman did little to attract significant attention to the area. In fact, the chance discovery of the Find of Sophiko at the harbor of Korphos reflects the style of investigation of Miliarakis and most early travelers to the area who typically adhered to major transportation that linked economic and population centers. At a superficial level the most visible infrastructure of the early Modern Greek economy exerted a definitive influence on the parts of the countryside that revealed evidence for ancient activities. Even during the 20th century, the investigation of the Sophiko area was largely confined to areas accessible by roads or clearly visible from population centers. No travelers visited the rugged area between Sophiko and Korphos. Despite these limitations, the archaeologists, many of whom were based at the site of Corinth itself on the Isthmus, did provide a basic context for the region of the southeastern Korinthia into which we could place the site of Lakka Skoutara. Fowler and Gebauer, as well as other early commentators, noted the remains and artifacts present at the site of Ay. Paraskevi to the northwest of the Rachi Zarakas ridge (Fowler, Corinth I, pp. 99-102; Gebauer 1939, p. 270). In the 1970s, J. Wiseman produced a plan of the ClassicalHellenistic fortifications at the site, and explored the surrounding area albeit relatively superficially in his survey of the Corinthian countryside. More recently, the peripatetic Y. Pikoulas and Y. Peppas have spent some time in the area and identified a few of the ancient and conspicuous Medieval monuments in the region including a Classical or Hellenistic tower at a site on the Early Modern cart road called Are Mbartze. In 2000, Michael Dixon wrote his dissertation on the results of an extensive survey focused on the area around the harbor of Korphos and its territory to the southwest, and argued that the boundary between the Corinthia and the Epidauria ran through this region during Classical and Hellenistic times. Dixon’s survey concentrated primarily on the Hellenistic and Classical period and documented a series of fortified sites that he identified with Classical or Hellenistic efforts to defend a prosperous local population. Over the past few years, the Saronic Harbors Archaeological Research Project under Thomas Tartaron and Daniel Pullen have worked north of Korphos along to coast with a primary emphasis on Bronze Age remains at Kalamianos. The site that they have documented represents a major Late Bronze Age center in the area which appears to have had a significant presence in the surrounding countryside. In fact, the work of SHARP will likely propose Late Bronze Age dates for some of the features that Dixon documented in the countryside. 4

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION While there is now considerable evidence for Classical to Hellenistic and Bronze Age activities in the area, thus far the area has escaped from the discussion of the Late Roman settlement explosion that has become such a common feature in discussions of Greek countryside. While traces of a Roman and Late Roman presence exist in the area including a fragment of a floor mosaic, reused architectural fragments in the various Byzantine churches, and a Roman period inscription built into the Panagia at Steiri, there is little in the way of a sustained Roman imprint on the local landscape. Our work at Lakka Skoutara will go some way to fill this gap. In contrast to the absence of evidence for Roman and Late Roman activity, the remarkable collection of Byzantine monuments surrounding Sophiko village has generated considerable scholarly attention. As early as 1930s, A. Orlandos conducted a thorough survey of the Byzantine and Post-Byzantine churches in the neighborhood of Sophiko including the Byzantine church at Steiri, the monastery of the Theotokos, the church of the Taxiarchis, and several well-preserved Post-Byzantine churches both along the road linking Korphos to Sophiko and in the village of Sophiko itself (Orlandos 1935). He published the results of this survey in the first volume of his AMBE. In 1975, T. Gristopoulos updated Orlandos’ work, but did not expand the number of monuments (Gristopolos 1975). Recently M. Kappas and Y. Fousteris has suggested that two of the post-Byzantine churches Ay. Antonios at Tourla and its neighboring church of the Hypapanti may, in fact, be of Byzantine date. The notable concentration Byzantine and Post-Byzantine monuments in the vicinity of the village of Sophiko led Timothy Gregory to conduct an early intensive a survey of the basin to the east of the village and the documentation of the fortifications on Mt. Tsalikas by Timothy Gregory. The longstanding, if superficial, scholarly interest in Southeastern Corinthia is sufficient to establish that this area saw consistent occupation from the prehistoric period through the Byzantine period with a gap in the recorded evidence coming during the Roman period. Most of the researches, however, have concentrated along the major route leading from the Corinthia to the Epidauria which provided access to the area for its ancient and modern travelers and scholars alike. As important for the local residents of this region, however was presence of a good harbor at Korphos which provided access to the Saronic community and ensured that the local inhabitants were never far removed from the networks that linked together the main economic and political centers of the Greek world. In the Early Modern period, the village of Sophiko Local history work…

5

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION This study benefits from a range of archaeological studies of the modern landscapes of the northeastern Peloponnese. In the Corinthia itself, S. Sutton’s work has examined the emergence of modern villages in the context of a broader diachronic study of settlement patterns in the Nemea Valley and with particular sensitivity to processes that led to the development and transformation of rural settlement, including kinship patterns, external economic activities, and the administrative power of the Greek state.1 Effie Athanassopoulou has explored similar trends the pattern of Medieval settlement in the Nemea Valley. Extensive research in the Southern Argolid and the Methana peninsula has also provide many points of comparison for settlement and economy in the southeastern Corinthia.2 The district of Sophiko and Korphos was historically as closely tied to the Argolid and the Saronic regions as the Corinthia, and, under Venetian rule, the district was even linked administratively to the communities in the district of Porto Porro in the Argolid. Beyond providing points of comparison for our work, these studies have employed a range of approaches that inform our own study of the modern countryside. These include, for example, efforts to understand ancient and medieval settlement systems from the patterns, behaviors, and signatures of settlement and land use in the modern era,3 but also, more generally, the integration of the modern period within diachronic studies of landscapes.4 Our work also benefits from the recent emphasis on understanding the close ties between the modern Greek countryside and the increasingly globalized economy of the 19th and early 20th century. By focusing on the responses of the rural world to economic opportunities and challenges in recent centuries, scholars have undermined the persistent views of Greek peasants as backward denizens of an unchanging, autochthonous Greek countryside,5 while diachronic studies have shown that inhabitants of even relatively remote or marginal Greek countrysides participated in a global Greek and Mediterranean economy. The realization that the Greek peasant occupied the rural landscape with a range of strategies returned attention to the diverse number of rural structures and settlement types in the Greek countryside. As early as the 19th century scholars had begun to document the villages, hamlets, komidia and kalyvia present throughout the countryside. Kalyvia, loosely structured agglomerations of seasonal or occasional residences, have long been understood as the

1 2

Sutton 1994. Jameson, Runnels, and van Andel 1994; Mee and Forbes 1997. 3 E.g., Murry and Chang 1981; Murray and Kardulias 1986. 4 Jameson, Runnels, and van Andel 1994; Forbes 1997; Koukoulis 1997. 5 Sutton 1994.

6

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION forerunner to villages in Greece.6 Scholars had often read this process as being stimulated by the end of Ottoman power and the resulting shifts in population in the first half of the 19th century. This shift in how scholars regard the nature of rural, post-independence Greece, complements the ongoing revisions of settlement patterns in the Peloponnesus during the later Ottoman period. The traditional arguments posed that Christian settlements were restricted to mountainous, upland areas in the Peloponnesus during the Ottoman period because the elite, Ottoman çiftliks occupied the more fertile lowland areas. According to this argument, the population of the upland areas remained small until Greek independence when it expanded into the recently-vacated Ottoman lands. This provides a foundation for the idea that the Greek peasantry remained in almost suspended animation during the Ottoman period only to be revived with the emergence of an independent Greek state. Wagstaff and others, however, have challenged this argument on two grounds. First, they have shown that based on known settlement elevations there is very little evidence to suggest that the population of the Ottoman Morea abandoned bottom lands to found settlement in the relative safety of less economically viable upland areas. Moreover, they cite Peter Topping’s important studies of the Venetian census for the 18th century Morea to argue that the population of the Ottoman Morea appears to have gradually increased from the 18th to the 20th century probably owing to the relatively stability of Ottoman political control over the territory and the continued arrival of immigrants, particularly Albanians, from other Ottoman lands. The tumultuous decades following Greek independence did effect settlement in the countryside. Ibrahim Pasha’s army inflicted damage on the Corinthian countryside as the area saw regular military activities during the struggle for independence.7 In fact, the destruction of the war of independence prompted the American philhellene and philanthropist Samuel Gridley Howe to found the refugee settlement of Washingtonia on the Isthmus near the present day village of Hexamilia. While it seems likely that the vicinity of Sophiko was spared the activity of Turkish troops, the area did not avoid its share of internecine conflicts which typified this period. The Notaras family, based in Trikala, raided the village of Sophiko and burned both its vineyards and surrounding forest lands.8 Despite these destructions and the dreadful picture put forward by travelers through the area during this time, the destructions of the early 19th century may have had little significant impact on the rural population. A resilient and flexible rural

6 7

Beuermann 1954, pp. 226-238; Sivignon 1981, Wagstaff 1982, pp. 20-24 Wagstaff 1978, p. 303 8 McGrew 1985, p. 3.

7

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION population appear to have reoccupied destroyed villages and returned quickly to cultivate various crops in the aftermath of episodes of destruction.9 The general themes in the recent study of the modern Greek countryside reflect a growing awareness of a resilient and flexible peasantry who engaged in a wide range of strategies to extract their livelihood from an environment shaped as much by local resources as regional and global economic and politic realities. The willingness to abandon the timelessness of the rural Greek peasant requires a concomitant change in how archaeologists have viewed the physical remains of Greek rural life. The most recent efforts to articulate an archaeology of the modern Greek countryside, H. Forbes’ detailed study of the Methana peninsula, followed ethnographic lines by combining a careful reading of local kinship structures and the Saronic economy to shed light on the varied meanings embedded in rural landscapes. His approach, however, was less a sustained treatment of the actual material evidence present in the countryside than an ethnographic study of familial relations that sometimes related to physical landscape. Our work at Lakka Skoutara foregrounds the landscape as the contingent arena for diachronic historical and archaeological processes of settlement. It should, then, offer a significant contribution to scholarship on the Greek countryside by highlighting regional and local economic trends evident in documentary studies, oral interviews, and the short-term and long-term archaeological patterns. To understand the dynamic nature of both the archaeological and the historical landscape, we have applied a diverse range of approaches. Central to our documentation of the site of Lakka Skoutara was an intensive diachronic pedestrian survey, a 10-year longitudinal study of the domestic structures in the basin, and historical and ethnographic research using both documentary sources and interviews with individuals who lived and work in the lakka. Each method brings a different scale of analysis to our research ranging from the relatively short term study of the houses to the exceedingly long term perspectives offered by diachronic survey. This multidimensional approach can potentially integrate varied evidence to demonstrate the contingent character of this Greek countryside and the meanings to its inhabitants. II. The Archaeological Survey A. Methodology The site of Lakka Skoutara was initially documented in 2001 as part of an extensive survey of the area between the harbor village of Korphos and the village of Sophiko. The goal of
9

Wagstaff 1978, p. 305-306.

8

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION this extensive survey was to discover the pre-modern route between the two settlements and the work determined that one potential route passed through the site of Lakka Skoutara. In a series of Extensive Discovery Units (6258-6533, 6537-6552), the extensive survey identified a significant scatter of ancient and modern ceramics as well as the presence of a dozen abandoned houses, agricultural installations, and a 20th century church building associated with a modern route through the rugged interior of the Saronic coastline. By the end of the 2001, this little valley suggested a fascinating past, with ceramics dating from prehistory to the modern era and 20th century houses that were a formation process wonderland. The following year, the Eastern Korinthia Archaeological Survey revisited the area to conduct a proper intensive survey and a proper documentation of the modern houses in the Lakka (see below).10 Since there were limited resources and time available for the survey, the team decided to sample three transects across the landscape. These transects would followed basic geomorphological divisions in the basin by capturing some part of the slopes of the Lakka, the alluvial fans that produced rocky soil throughout the northern section of the basin, and the more-stable and less rocky red soils that marked the basin floor. This method was consistent with the geomorphological division of units practiced throughout the EKAS survey area and allowed us to control for the significance and in a coarse way, the chronology of various erosional processes. The units were also positioned to capture areas immediately surrounding six of the houses which represented various states of abandonment. This sampling method produced 92 units in three groups with an average unit size of 2335 sq meters and a total area of 2.1 ha. [Although we played around with more intensive kinds of mapping, e.g., chasing densities] The EKAS team walked each unit at 10 meter spacing with each fieldwalker counting every artifact that appeared 1 m to each side of their swath. This procedure sampled 20% of the area of each unit for density. The variation of artifacts present in each unit was sampled according to the chronotype system in which field walkers collected one example of each unique type of artifact.11 The ceramics team analyzed these artifacts in the field and the results were keyed into an Access database which was linked to a GIS database. The units surveyed at Lakka Skoutara produced around 2200 artifacts per ha (walked), which is considerably higher than the approximately 1500 sherds per ha produced by the units in the main survey transect on the Isthmus, but still below the 3000 sherds per ha often considered to be the benchmark for site density in the Eastern Mediterranean. There were, however, 25
10 11

See Tartaron et al. 2006 for a full overview of the methods. For the Chronotype system in the context of the EKAS project, see Tartaron et al. 2006; Caraher, Nakassis, and Pettegrew 2006.

9

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION units covering approximately hectare which exceed the 3000 artifacts per hectare standard for site density. The field team designated one area of the site as a Localized Cultural Anomaly (LOCA) owing to the high quantity of Final Neolithic material concentrated at the conjunction of six moderate-density units with varying qualities of surface visibility. In these areas, the team conducted a more intensive form of artifact collection. The teams selected a 20m x 20 m square in each DU of the LOCA and performed a total ChronoType collection in that square. One example of each type of artifact was collected from each of the four squares to produce a complete sample of every type of ceramic present. The squares were located relatively close to one another, since the FN-EH material was not randomly distributed throughout each DU, but rather clustered together. Consequently, the LOCA collection units were concentrated in an effort to capture the area with the highest artifact concentration. GPS coordinates were taken at the SW corner of each LOCA collection square providing a fixed point from which to map the units. The 2.1 ha surveyed represented slightly less than 1% of 245 ha of arable land available to a settlement. This figure includes all contiguous land of less than an 18 degrees slope which represents the steepest slope under cultivation in the basin albeit with the help of terrace walls. Terrace walls are common on the northern side of the basin and generally begin around 8 degrees of slope, although some terrace walls, perhaps better described as check dams appear at lower grades.

B. Distributional Data The standard and local collection survey produced 926 artifacts in 625 batches. The periods represented in this assemblage of artifacts produced in these survey units represented over 6000 years of human occupation from the Final Neolithic period to the modern day. Examining the assemblage produced by our standard DU chronotype sample showed that 33 periods appeared in the survey. Since the chronotype system enables the ceramicist to place an artifact in a wide range of categories from the most broad (Ceramic Age) to very specific and narrow periods (Early Medieval). This means that artifacts collected from the field are categorized into sometimes overlapping categories. For example, a query for all possible Medieval pottery should includes pottery that is both certainly Medieval and material that could only be dated to a broad range of time which would include the Medieval period. Aoristic 10

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION analysis provides one way to show the chronological distribution of artifacts in the basin that takes into account the different degrees of precision in our dating of the artifacts. This kind of representative analysis assumes that an artifact has an equal chance of appearing during any year across its entire span of possible dates. So if an artifact is dated to the Late Roman period with a date from between 400 and 700 A.D., the artifact has a 1/300 chance of appear in each year. While it is important to emphasize that this is simply a model for the chronological distribution of ceramics, it is a useful way to represent the relative quantity of material datable to each century. As chart x. shows, there is activity at the site for nearly the entire historical period with a sharp increase in activity in the most recent century. For the material collected from the 20 x 20 meter squares designated as a LOCA, the distribution of material looks markedly different. The graph shows significant spikes in the Final Neolithic and Early Bronze Age as well as a substantial spike in the Roman to Late Roman period with some activities continuing through the Medieval period. In the Early Modern period, after around 1800 AD, artifact densities trail off in sharp contrast to the artifact assemblages produced through more traditional collection methods. The small areas subjected to LOCA/chronotype collection were over 100 meter from any of the modern houses which tended to produce most significant quantities of Early Modern and Modern material.

11

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION

The spatial distribution of material across the basin likewise produced broad patterns that allow us to make some general observations regarding the settlement patterns at Lakka Skoutara. Since it was not a total coverage survey with a goal of sampling the entire basin, it is important to understand that our analysis only captures about 1% of land available for use throughout the basin’s history. Moreover, our survey was limited by the densely forested slopes on the southern and eastern side of the basin, fenced areas, and slopes too steep to accommodate easily our survey procedure. <<<<discussion of prehistoric material here?>>>> The earliest historic period of settlement in Lakka Skoutara appears to be the Archaic period. Three pieces of fine and semifine ware present a slight trace of material datable to the Archaic period. Most material from before the Roman conquest of Greece, however, can only be dated broadly to such categories as the Archaic-Hellenistic period or the Classical-Hellenistic periods. The presence of fine ware and kitchen ware in the assemblage suggests that the area served in some domestic function, but the small number of sherds present in the assemblage, however, recommends against any definite arguments for chronology or specific function. The sherds were concentrated primarily in the central area of the basin, immediately south of the modern church dedicated to Ay. Katerini, where soil depth is good and there is less cobbly soil 12

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION washing down from the hill slopes. Units 7001 and 7002 had particularly high visibility and especially diverse assemblages. A few artifacts appeared in unit 7058. This unit had a lower visibility (40%), but a very diverse assemblage of material. It is notable that the higher intensity LOCA collection conducted in this unit did not produce any material dating from the Archaic to Hellenistic period. By the Roman period the basin appears to have used more intensively and extensively. There are particular concentrations of material both in the same units where Archaic – Hellenistic artifacts appeared, but also further east. This suggests that there may have been continuity in the basic settlement structure of Lakka Skoutara and that an existing settlement expanded during the Roman and Late Roman periods. This pattern would appear to coincide with the more general expansion of settlement noted by surveys across Greece. While little pottery can be conclusively dated to the Early Roman period except for a few pieces of Eastern Sigillata A and Koan type amphora handles, there is a substantial quantity of Roman and Late Roman kitchen ware, coarse ware, and amphora sherds across the entire extent of the basin. Some fragments of roof tile may also date to the Roman period hinting at the possibility of substantial architecture in the area. It is notable that we found relatively few Late Roman fineware sherds, but an abundance of spirally groove and wheel ridged artifacts. This may suggest that the area that we sampled were not utilized for domestic activities during the Later Roman period, but for more commercial purposes or for more limited forms of seasonal habitation which would not produce the robust deposits of fineware. Echoing the substantial amount of Medieval activity documented in the vicinity of the village of Sophiko, activity continued in Lakka Skoutara into the Medieval period. There is an abundance of material from the Medieval period clustered around a number of fields in the east central area of the basin and in the far southwestern edge. The artifact datable to the Medieval period range from storage and utility wares to kitchen ware and fine and semifine table wares. This diverse assemblage would seem to represent settlement in the area. It is tempting to associate the material at the southwestern corner of the Lakka with the presence of a well. While it is impossible to know whether this well was in use during the Medieval period, it may well be that its position at a join in the fractured bedrock of the basin provided access to ground water in an earlier period as well. Unfortunately, our sampling technique did not allow us to cover the entire basin floor so it is impossible to determine whether Medieval activities extended across the entire basin. That being said, it could be that during the Medieval period, settlement retreated from the deepest soils in the central part of the basin and moved to the east and slightly up the 13

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION slope. This contrasts the Roman and earlier Greek period activities in the area that were focused on the lowest points of the basin. It may be that the Medieval settlement positioned itself to maximize the opportunity to cultivate the less rocky soils in the center of the basin by leaving them free from settlement. In contrast to the Medieval period, the modern material which dates from the 19th and 20th centuries extends outward from the remains of several of contemporary houses which stand near the center of Lakka Skoutara. In addition several Early Modern to Modern houses stand along the terraced northern slope of the basin and like their counterparts in the lower fields, produced halos of material. While ceramic roof tiles account for a substantial part of the modern assemblage from the basin, the area also produced the complete range of material expected in a domestic assemblage. This includes fine and table wares types including Canakale and Didymoteichon wares, kitchen and cooking wares, and storage vessels. The increase in material from the Medieval to Early Modern and Modern period might well suggest an increased in activity in the area during this time. At the same time, we might expect a significant modern overburden extending across the survey area as the modern houses slowly collapse and scatter their archaeological signatures across the surface.

III. A Study of Houses in Time and Space A. Methods for Documenting Houses At the end of the 2001 EKAS field season, a small team returned to Lakka Skoutara to begin recording the numerous abandoned houses, their architecture and associated features, and archaeological assemblages. Our aim was to document the houses and their environment in a way that would allow reconstructing the cultural formation processes—construction phases, habitation and discard, abandonment, and post-abandonment uses of the house—that continue to transform the houses. Aware of broader scholarly discussions about the interpretation of rural sites in ancient landscape,12 ethnoarchaeological and modern survey approaches in Greece,13 and the dynamic nature of the Greek village,14 we intended our study to contribute to an

12 13

Binlitff and Snodgrass 1988; Osborne 1992; Alcock, Cherry, and Davis 1994; Pettegrew 2001. Murray and Change 1981; Murray and Kardulias 1986; Whitelaw 1991; Murray and Kardulias 2000. 14 Sutton 1994.

14

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION understanding of the character of settlement, the nature of abandonment, and archaeological signatures and meanings of habitation in the context of diachronic landscapes.15 We documented several houses in a preliminary way in 2001 which allowed us to refine our methods fully the following year. Our methodology, as it developed, consisted of three distinct components: 1) recording through detailed description and photography the houses, their architecture, and associated materials, 2) collecting information on artifact densities around the modern structures through surface survey, and 3) conducting oral interviews with the house owners and inhabitants of Sophiko. In the previous section, we discussed the character of artifact densities, and we will elaborate on the oral interviews below, so here we can focus on detailed descriptions of the houses and their assemblages between 2001/2002 and 2009. However, because the oral interviews relate so directly to the interpretation of the houses (and vice versa), we will integrate in this section relevant information gained from the interviews with Mr. Perras, Mr. Zographos, and others. Our recording procedures included basic descriptive fields (e.g., “Artifactual Material”) as well as interpretive assessments (e.g., “Function & Land Use”). To facilitate the process of description, we assigned numbers to the houses that we later associated with individual home owners through oral interviews. We noted the location of each house, its size and dimensions, orientation, and associated features; the artifacts present inside and outside (within 15 meters) of the house in terms of their types, quantities, and conditions; the different phases of habitation, construction styles, and building functions; and the current condition of buildings and area, including ground cover and visibility. Table 1 below shows the table we had developed by the 2002 field season for collecting data. In addition to such descriptions, we also photographed the interior and exterior of the houses over nearly a decade, providing data that allows us to image the houses through time.16

15

Studies of Medieval and post-Medieval vernacular architecture and housing in Greece are numerous and represent a diversity of theoretical and methodological approaches ranging from typological descriptions to analysis of settlement pattern hierarchies and social or economic behavior. For discussion and examples, see Megas 1951; Wagstaff 1965; Philippides, editor, 1983-1990, with multiple volumes for different regions of Greece including the Corinthia (Chrysafi-Zografou 1987); Clarke 1995, 2000; Sutton 1995; Cooper 2002; and Sigalos 2004. Philippides 1983, 33-49, Sigalos 2004, 21-49, provide useful reviews of the literature for Greek vernacular architecture. On the multiplicity of meanings in modern houses, see Forbes 2007, 309-314. 16 In addition to the photos in this article, we have also made images available online.

15

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION

Table 1. Form used to collect data for houses Following our initial study of the houses in 2001 and 2002, we returned to the area for several days in three subsequent seasons (2004, 2006, and 2009) to record the cultural processes and land use that are altering the houses, their functions, and assemblages. As the houses vary in their current function, condition, and position in the valley, we were unable to record every house during every season. Moreover, several houses that are maintained and occupied seasonally (#s 1, 7, 8, 9, and 15) were inaccessible and not documented systematically, while 16

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION another house (#12) seems to have disappeared sometime after our visit in 2002, perhaps due to the widening of the main gravel road running through the valley; these six houses (#s 1, 7-9, 12, and 15) will rarely be included in the following tables and discussion. Several other houses (#s 11, 16, and 17) are high on the northern slopes of the lakka basin, survive only in their wall foundations, and are overgrown with weeds; these we recorded only in the 2002 and 2009 seasons. For the following discussion, the reader is encouraged to visit the online website that accompanies this publication to view photos of individual houses through time. A summary of the houses is listed below in Table 2.

17

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION Table 2. Summary of Houses

B. Agriculture and Land Use Most of the houses at Lakka Skoutara represent seasonal farmsteads oriented to agricultural enterprises in the surrounding valley. Throughout Greece and the Aegean, farmsteads vary in size and plan according to regions, but there are nonetheless consistent patterns including, for example, house types, floor plans, architectural styles, and domestic functions, among others.17 As the vernacular architecture will be discussed in the following section, we can focus our attention here on the associated landscape features that linked houses to agricultural enterprises at different points in time (Table 3). From oral interviews we know that the inhabitants of the valley supported themselves in the 20th century with subsistence agriculture, especially the cultivation of wheat, olives, and vegetables. However, they also exploited the valley for specialized products like charcoal, pine wood, and resin which were carried down to the nearby harbor at Korphos for export, while pottery and tiles from Aegina could be imported through the same harbor. Agricultural features and installations such as gardens and external bake ovens (phournoi) that have today fallen out of use appear regularly enough at and around these houses to materialize the economic activities of the inhabitants of the lakka at a former point in time. The principal crop cultivated in the valley in recent times is the olive, and even today the valley sees a flurry of activity during the late fall as residents from Sophiko return for the harvest. Aerial photos show that olives comprise nearly all the land of the valley under cultivation (Fig.), a fact confirmed by a pedestrian survey which documented this fruit in 80.4% of the survey units (n=74 of 92). The groves are not neatly divided, however, and Mr. Perras explained to us the complexities of cultivation and ownership patterns during the valley’s heyday: individuals and families commonly owned specific trees in other people’s fields, and, although they could make no claim over the field, did have access to individual trees.18 Even the church owns many olive trees throughout the valley that were donated by individuals as a pledge
17

Murray and Kardulias 1986, 31-32, for example, note the importance of associated gardens and location on flat arable land in the S. Argolid; Whitelaw 1991, 412-419, suggests that the most consistent characteristics are interior and exterior whitewash, the use of plaster on interior walls, at least two rooms, and associated porches, interior fireplace, and external oven. 18 This pattern of land ownership is not uncommon in Greece, and is at least in part a result of dowry and inheritance practices. Cf. Forbes 2007, 164-165, 168-169.

18

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION (τάµα) and support the costs of maintain the church.19 Trees owned by individuals or the church are often marked with painted signs or initials to indicate ownership.

Table 3. Features Associated with the Houses We do have a sense of the longevity of olive cultivation in the valley from both the archaeological study and oral testimony. In our pedestrian survey of a small sample of the valley, we measured the smallest and largest olive tree in each of the fields examined. We noted many very young olive trees, with circumferences 0.04-0.10 m, planted in recent years and often surrounded by make-shift fences to keep out the grazers. But most of the groves recorded are evidently more mature, as the average minimum circumference for all 92 survey units was 0.44 m and the average maximum circumference was 1.85 m. These figures are consistent with Mr. Perras’ comment that the trees are generally from 50 to 80 years old. Some of the groves must
19

Cf. Forbes 2007, 347-348, who notes that Aleppo pine trees in Methana served the same end.

19

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION be much older, in fact, for we measured a number of trees with double trunks and base circumferences of 3-6 m; the three largest trees noted had circumferences of 6.75 m, 6.80 m, and 8.55 m. These larger trees must be hundreds of years old and consequently point to the history of olive cultivation at the lakka. Interestingly, there is actually very little physical evidence for the cultivation of olives in the valley beside the trees themselves, small scattered piles of branches pruned from the trees during the winter, and ashy circles produced by burning the trimmings.20 The fieldwalls that circumvent some of the groves point to ownership or cultivation patterns and the ladders and drinking containers found in several houses are probably connected with olive harvesting activities today. The processing of the fruit, however, occurred elsewhere in recent times. Mr. Perras informed us that olives could not be processed in the valley for lack of water and were carried to Sophiko where they were processed through horse-powered machines, and, later, modern machines. The only installation in the valley related to olive production is an enormous trapetum mortarium located 100 (?) m east of the church just below the modern gravel road. This crusher base is 1.85 m wide and at least 0.38 m high and has a form comparable to presses elsewhere in the Peloponnese typically dated to the Roman-Late Roman period.21 Our informants did not remember this press ever used for any purposes and called it “ancient,” dating to the period before the modern settlement. Indeed, the top of the mortarium is now quite damaged and, in its current location adjacent to a gravel road, there would be not enough room to operate the press effectively. The other major crop cultivated in the valley in modern times has left an equally engimantic temporal signature. On the one hand, no one cultivates grain in the valley today and the ubiquity of olive groves might easily make one forget that this pattern is a recent one. But we know from our informants that many of the olive groves were planted in the last century and replaced fields cultivated in grain. The terracing on the northern slopes of the valley, in fact, provide good evidence for grain cultivation for the surviving field stone retaining walls would be unnecessary for the olive trees that are now contained with them; unfortunately, we have no basis for dating the terrace construction in the valley.22 The same can be said for the field walls

20 21

Cf. Murray and Kardulias 1986, 27. Kardulias and Runnels 1995; Foxhall 1997, 262-264, and Figs. A.1 and A.3. The use or reuse of these kinds of presses may stretch into the Medieval period. Foxhall notes examples of reuse in later periods in which the central column was cut out of the mortaria. 22 Terraces are notoriously difficult to date and scholars have suggested periods ranging from the Bronze Age to the 19th century. For discussion, see Whitelaw 2001, 405-410.

20

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION through the valley that may have functioned initially to keep the goats out of grain fields.23 That four of the houses have associated external bake ovens (cf. Table 3) introduces additional material evidence for cereal cultivation and bread making, even if we should not conclude that the ovens were used only to bake bread. The most obvious evidence for grain processing, however, are six threshing floors in the lakka, all constructed in a similar manner with flat cobble-sized stones interlocked to create circular areas ranging from 13.90 to 18.20 m (average diameter of 16.30 m).24 Only one of the floors (#3) is slightly different, raised a half meter above the ground on a low platform; Mr. Perras called this one “ancient” and noted that even his grandfather could not recall it ever being used.25 All the floors, however, lack central holes which indicate the use of doukeni sledges with flint or obsidian blades rather than an animal tethered to the central pole.26 The alonia are located generally located some distance from the houses, a pattern that might at first suggest that the modern inhabitants chose to construct threshing floors in areas away from the houses and open to the wind. In fact, our informants tell us that it is the other way around: the threshing floors were already present when their forefathers settled the area and in turn influenced the location of houses. There is some good material support for this in House #4, one of the oldest houses in the valley, which was built directly over part of the aloni and therefore indicates a threshing floor predating the house and out of use when the house was inhabited; the family who lived in this house used a different threshing floor not far to the south of the house. In general, the alonia must represent a phase of land use in the lakka dating back at least to the early 20th century since young to mature olive trees dominate the valley today, and were, as we noted above, planted over half a century ago. Most of the alonia, then, were out of use in the post-WW II era,27 and some had fallen into disuse by the early 20th century. Olives and grain do not require a significant supply of water, which is severely lacking in the valley. Mr. Perras explained that water was not trapped in the area due to geological factors

23 24

Whitelaw 1991, 410. This range is slightly smaller than the 2 threshing floors documented by Murray and Kardulias 1986, 26, which showed diameters of 20 m, but much larger than those documented on Kea, with diameters between 6 and 8 m (Whitelaw 1991, 424) 25 Mr. Perras commented on how the people who lived there before his people were smart and understood that a raised platform for a threshing floor would help air the wheat before threshing; his own people used threshing floors at ground level and the wheat took longer to dry and therefore became stale. 26 One of the home owners, Mr. Perras, gave us detailed descriptions of how the inhabitants of the lakka used the “ντουγένι” for threshing. This is the same doukani used in other parts of Greece like the Argolid and the Cyclades. The animal pulls a rectangular wooden board with rows of blades (metal or stone) that separate the grain from the chaff. See Murray and Kardulias 1986, 26. 27 Threshing floors elsewhere stay in use as late as the early 1970s: Murray and Kardulias 1986, 26; Clarke 2000, 121.

21

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION but washed away to sea, and he could point to only a single well in the southwestern part of the lakka. The inhabitants relied on cisterns built typically close to their houses and filled by the winter rains. Five of the houses (n=5) have cisterns that would provide easy access to water and the large concave, collection surfaces that funneled rainwater into the smaller cistern (also called aloni) range in size from 6.00 to 9.60 m (average diameter = 8.20 m). Although originally constructed of sizable fieldstones, they were all refurbished with concrete tops at some point since the 1950s. Besides olive and grain, the only other kind of fruit tree observed today were a few interspersed almond, apricot, and pear trees. Pine trees, however, are common in the valley and cover the surrounding hills, and resin processing was a significant industry into modern times. The surrounding pine forests were not owned by the community but by individuals,28 who tapped the resin in metal collectors and processed closer to home. Basins for processing resin are found inside two of the houses (#s 10 & 14) and an additional house (# 2) has a basin immediately outside. These three surviving basins, however, are a pale physical reflection of a once vital industry. More visible material evidence for resin processing comes in the form of the ubiquitous resin collectors found on and around pine trees in the region and also the paths on the wooded hillsides that were created long ago to access the trees. In recent decades the resin industry has plummeted sharply due to a decline in the product’s value beyond the region. The pine forests are returning to their natural state and the only evidence of this once significant activity is the scatter of rusty collectors on the hillside and the paths that lead to them.29 While these traditional forms of land use have declined sharply in recent decades, the valley remains open for herding goats, evident in the reuse of certain houses as animal keeps (see below), the make-shift wooden hedges erected around recently planted olive trees, and the paths that weave through the surrounding pine slopes and are maintained by the goats alone. There is also a sizable animal fold immediately to the south of House #3 which has made use of the house’s deep cistern since its abandonment some decades ago; in 2009, the goats wandered freely inside grazing the weeds that had grown among the ruins. This may be a recent pattern of land use, but, interestingly, our informants tell us that Yiorgos Mertikas, the great-grandfather of Mr. Perras and Mr. Zographou and one of the first settlers in the valley, obtained titles of land
28

In most areas of Greece, uncultivated land like pine forest is typically owned and cultivated by the state or the community; trees in the Sophiko region are individually owned. Cf. Forbes 2007, 165. 29 One of our informants, Mr. Zographos, told us that he himself owned many pine trees that he continues to exploit albeit less intensively than in the past. Greek legislation prohibits the destruction of pine trees but the trees no longer have significant value for their resin. The older inhabitants of Sophiko receive a pension for their years of hard work in resin production.

22

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION ownership from Ottoman authorities to raise animals. Smaller animals, like chickens, were kept in enclosures near the houses, while draught animals were kept in associated yards or within the house itself. Indeed, at once house an aged donkey resides whose primary purpose today seems to be to keep the owner of the house, Mr. Zographou, company during his daily visits to his house. C. The Houses: Architecture and Building Phases Most of the houses at Lakka Skoutara are single-storey “long house” types common to the Peloponnese and southern central Greece in the Early Modern Era (See Table 2).30 They are typically 9-12 meters long and 4-6 m wide with total area between 50 and 70 meters (mean: 52.26 m; median 58.80 m), and constructed with fieldstone walls, mud mortar, and tiled roofs; four houses (#s 5, 9, 13, and 16) have noticeably smaller dimensions and two of these (9 and 13) are recent constructions built with storage, rather than residence, in mind.31 The houses are usually oriented roughly north-south,32 with windows and doors on the long east-west walls; the doors are almost always on the east façade. Enclosed courtyards are found at a number of houses immediately outside the main doorway, and are often associated with external installations like cisterns, gardens, chicken coops, and bake ovens.33 We note that the sizes of all these buildings (including the storehouses) are comparable to houses documented elsewhere in central Greece and the Peloponnese.34 The floor plans of the house reflect an agricultural mainland style with interior space arranged linearly into one or two rooms (Table 3).35 Several of the houses (#s 6, 11, 16, and 17) collapsed long ago and survive only in low foundation walls; overgrowth of vegetation makes it impossible to reconstruct floor plans but we can surmise either one or two rooms. The five documented houses that remain standing, however, all suggest comparable plans, with the long north-south dimension of the house divided into northern and southern rooms by an interior
30

See Sigalos 2004, 51-70, for a regional-based overview of the forms and styles of Byzantine-Early Modern architecture. Sigalos draws particular attention to the long house type with broad façade, which is predominant in the Greek mainland and especially the Peloponnese during the Ottoman and Early Modern periods (57, 61-63, 169176). 31 Cf. Murray and Kardulias 1986, 31. 32 Four houses (4, 8, 9, 13) have an east-west orientation but we note that two of these are storehouses. 33 See Sigalos 2004, 61-62, for the courtyard as an integral component of the house. 34 Murray and Kardulias 1986, Table 1, pp. 28-29, shows similar figures of 9-20 sq m for storehouses and 50-93 sq m for houses. Cooper 2002 , 37, suggests typical dimensions of 10-12 x 6-8 m for 19th and early 20th century houses in the nomoi of Achaia and Elea, which is slightly wider than our 4-6 m range. Clarke 2000, 112-113, suggests 10 x 6 m houses are common for late 19th-early 20th century houses in a village in Methana. And Sigalos 2004, 88-109, notes dimensions for Late Ottoman and Early Modern longhouses in towns and villages in Boiotia generally in the range of 11-14 x 5-8 m, although houses are occasionally much longer. 35 Sigalos 2004, 59.

23

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION plastered partition wall or slight rise in elevation.36 The elevated room (usually the northern one) is smaller and contains a fire place, windows and niches on the east and / or west walls,37 and furniture such as beds, benches, and tables; it comprised the main living and sleeping space for the home owners. The larger room typically features a simple earth floor, the house door, and an additional window and was used for, among other things, an interior work space, storage, and the housing of animals;38 this much is evident in the agricultural implements (e.g., ladder), straw on the floor, and resin processing basins visible in some of the houses. The houses have low pitched roofs constructed of long beams, a lattice of intertwined branches, and tiles. The most common type of tile covering the typical house in the valley is the buff and (red) brown Lakonian tile, although other types of machined-produced tile (e.g., the glossy red “Marseilles”-type) are also occasionally found. Most of the houses have now lost their full set of tiles, but where they remain suggests that 2,000 tiles is not uncommon (e.g., House #7); the longest house (#10) makes use of about 2,700 tiles. Although the houses tend to be roofed with the same type of tile, our survey and study demonstrated different tile types at several of the houses that point to successive roofing episodes. The architecture itself show building phases that can be dated by construction styles and evident refurbishment.39 On the one hand, it is relatively easy to differentiate Early Modern (pre-1950) from Modern construction episodes in the houses since the former makes use of a traditional vernacular style of construction common to longhouse construction—coursed field stone walls about 0.50-0.75 m thick, chinked with small stones and tiles, filled with mud mortar, plastered, and whitewashed—while the latter makes use of construction materials like cinder blocks and concrete reinforcements that have been in use in the region only since the 1960s. Cinder blocks used to reinforce pediments (House #2, 5) and long walls (House #3, 5), or as the principal building material altogether (#9 & 13), indicate distinct modification episodes of recent decades; the same is true of brick and concrete capping used in conjunction with older building materials. On the other hand, it is less obvious that the incorporation of fieldstones into a cinder block house would represent an intentional effort to create continuity with the original house of the 1920s, as Mr. Perras informed us about his own reconstructed home, or that houses preserved today entirely in fieldstone represent refurbishments using traditional styles in recent times.
36

The partition walls are constructed of vertical branches or thin canes, covered with mud, plastered and whitewashed. The elevation difference between rooms is created by a ledge of plastered stone or cement. House #14 is the only house where the floor (cement) is the same level throughout. Cf. Sigalos 2004, 62. House 37 The niches are often used for icons: Sigalos 2004, 89. 38 Cf. Sigalos 2004, 103. 39 See Foster 2002, 130.

24

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION These living realities are sometimes nearly undetectable in the archaeological material itself but come to life in interviews with informants.

25

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION Table 4. Plan and Construction Style of Houses

The same complexities are evident in examining the relationship between adjacent house remains that could suggest building migration, additions, or phases in accordance with changing household needs and the necessity of occasional reconstruction. It is obvious that House #13, the cinder block storehouse, was constructed in recent decades and succeeds as the next phase the remains of a long house whose fieldstone walls are still visible today. Or similarly, one can note that House #2 reuses the eastern wall of a former house lying immediately to the west which survives only in low foundation walls and a sunken depression in the earth (1-2 m deep) that must have been caused by the collapse of the bedrock beneath the former house; the current house clearly is a newer version of its neighbor and recycles the fieldstones into the new building. In other cases, however, the exact relationship of associated structures or the multiplicity of enclosed spaces is less clear. At House #4, for instance, low foundation walls of two rooms to the west of the house could represent, variously, earlier phases of the structure, later extensions to accommodate new members, associated buildings and enclosures, or, as Mr. Perras explained, the divided living space of two brothers who were unable to get along. The oral interviews provide significantly more detail about the human experiences of the changing household that take us back ultimately into the 19th century. According to one informant, House #3 dates to the 1920s but it was maintained and refurbished in the same place for nearly a century—even if there is nothing obvious in the architecture that identifies its earlier from later habitation.40 Mr. Zographos’ house (#7) is situated within 20 m of the foundations of the house (#8) of his grandfather, Ioannis Mertikas, who died in 1947 at 103 years old; his own house marks a more recent construction of the family that has outlived that of his grandfather’s home. On the other hand, Mr. Perras’ house (#5), which is also a relatively recent refurbishment (early 1980s) in cinder block, lies immediately over his father’s house built in the 1920s and intentionally incorporates the former structure’s foundations and field stone walls to emphasize continuity. Interestingly, though, the low foundations of nearby House #6 represent not a house preceding his father’s (#5) but one that his father built during the German occupation in the early 1940s to accommodate the new needs of the household who were then living in the valley year
40

Coulton and Foster’s enormous catalogue of village houses for the nomoi of Achaia and Elea (2002) give numerous examples of houses dated by oral testimony or datestones to the 19th century, a pattern that is not uncommon elsewhere (Whitelaw 1991, 417). The longevity of houses in the same location is understandable if we accept Forbes’ observation for Methana (2007, 229-230) that inhabitants thought of a rebuilt house as the same house as the one it replaced.

26

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION round. The seemingly old foundations of House #6 are actually much more recent in time than the foundations now incorporated into a cinder block structure. These kinds of intricacies in building lifecycles must be common to all the houses in the valley, even if we have neither the archaeological clues nor the oral stories to decode them.41 Our informants referred to older houses in several places that are totally invisible today, the ephemeral building material presumably incorporated into later structures and features.42 The oldest inhabitants of the lakka remember seeing some of these houses as children built of mudbrick walls and even mudroofs. Mudbrick construction is common to vernacular architecture of the Peloponnese generally,43 but is nearly absent at Lakka Skoutara, and the informants’ memory in this respect adds a vital clue to the longevity of this semi-village in the area. The different construction styles that can be observed at the houses, then, manifest the multiple phases of building, repair, and refurbishment that are always present albeit not always obvious at houses surviving only in foundation or those vanished altogether. In fact, as the following sections will show, the houses of the lakka are constantly being transformed even within the dynamic landscape of a valley that can often seem abandoned. D. Domestic Lifecycles If the agricultural installations and houses themselves reflect points and episodes of habitation through time, the equipment and artifacts at the houses represent the varied processes of habitation, functional shift, reuses, and abandonment in a landscape tied to the broader global forces transforming the northeastern Peloponnese in the 20th century.44 Hence, while habitation was typically seasonal in the valley, with land owners residing permanently in Sophiko and visiting their land during peak agricultural months, we have also learned of times when people inhabited the valley semi-permanently, as, for example, during the turbulent 1940s when World War II and the subsequent Civil War made life in Sophiko unbearable. The abandoned landscape that seems to characterize the valley today is itself a product of the decline of small41

Although the construction of adjacent houses often occurs to accommodate new members (daughters-in-laws and grand children) of the extended families, building function can be quite complex. See Clarke’s example (2000, 119, 123) of a large family in Methana purchasing a house in the 1920s immediately adjacent to their own and using it for, successively, storage, temporary village school, and the residence of the family of son and daughter-in-law, and eventually, grandparents. Cf. Sigalos 2004, 62, for the use of adjacent houses as residences for married children and stables. 42 Mr. Zographos pointed out a place near his grandfather’s house (#8) where he remembered his his greatgrandfather’s kalyvi (hut) being with its dirt roof. Mr. Perras noted that House #10, belonging to Anthanasios Kalimanis, was the ancestor to a house just to the north (now gone) owned by Athanasios’ grandfather Yiorgios. Such vanished houses are often incorporated into later constructions: cf. Clarke 2000, 116-117. 43 Foster 2002, 139. 44 On the multi-functional character of houses, see Clarke 2000; Murray and Kardulias 1986.

27

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION scale agriculture across Greece since the 1960s. The last family residing permanently in the valley had moved out by the early 1980s, and the remaining standing houses have been occupied only for the shortest duration during the fall olive harvest. Yet, Mr. Zographos still drives out to his country house nearly every day from his permanent residence in neighboring Sophiko. Such contingent forms of settlement and land use have left material correlates in the landscape that defy facile definitions of “habitation” and “abandonment.” Consider the associated artifact assemblages as they reflect on the functions of the buildings.45 We would expect that these farmsteads should produce a range of artifacts that point to domestic function, including at least the basic furnishings common to early modern seasonal houses.46 Obviously “domestic” assemblages are exceptional, however, and most of the houses are missing equipment like beds, tables, chairs, and kitchen utensils (Table 5).47 The houses that survive only in their foundations (#s 4, 6, 11, 16, and 17) have only associated light tile scatters and occasional artifacts inside and outside the structure, while most of the assemblages at other houses indicate non-domestic functions (see below). Only two houses (#s 5 and 14) showed a variety of habitation material including furniture, containers, clothing, tiles, and various assorted metal and plastic artifacts. At Mr. Perras’ house (#5), the material was scattered in the collapse all about the floor of the house, but in House #14 containers and glass bottles were still present on wooden shelves.48 Both houses fell into disuse but had not been depleted of the household goods, perhaps because the home owners were unable to visit in their older age, or because the children inheriting the properties saw no point to continue their parents’ investment. Mr. Perras himself was 80 years old in 2001 when he showed us around the valley and, although he had not visited his house in some 10 years, became quite upset when he saw it in ruins.49 Most of the houses produce the sorts of assemblages that we would expect from abandonments in which the objects and equipment were recycled elsewhere. Half of the houses recorded (#s 4, 6, 8, 11, 16, and 17) survive only in their foundations and have left (at most) very few tiles and sherds. Since these were abandoned long ago during the period when the lakka was used regularly, it is probable that the materials were carried off and reused elsewhere before or during abandonment. Domestic objects and equipment were essentially stripped from Houses 2
45 46

See Murray and Chang 1981 and Murray and Kardulias 1986, Table 3, for a discussion of artifacts and function. Village houses and farms in early modern times typically had basic utilitarian equipment like a bed, chairs, tables, utensils for cooking, eating, and drinking, wine barrels, olive oil containers. See Clarke 2000, 110-113, 117, 124. 47 Admittedly, this is in part a result of our sample. We were unable to access and documented several of the houses that are still in use (1, 7, 9, and 15). For a discussion of the functional characteristics (including ambiguities) of artifacts and equipment at modern settlements and herder’s sites, see Murray and Chang 1981; Murray and Kardulias 1986. 48 No objects were found in the wooden niches in the northern wall or the shelves on the east walls. 49 On the emotive power of abandoned lands, cf. Forbes 2007, p 326-327.

28

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION and 3 before their conversion into animal pens, while some houses (e.g., House #10) still standing and in use today no longer have the most basic household equipment including, for instance, storage vessels, plates, and utensils. Assemblages at several of the houses reflect specific shifts in building function following occupation. The domestic assemblages of Houses 2 and 3 were so depleted during and after abandonment that there is nothing inside the houses that specifically suggests habitation. The burlap bags, wool, medicine bottles, glass and plastic containers, and manure, among others, reflect first the conversion of these buildings into animal pens and, now that the roofs have collapsed, open areas for animal grazing.50 The small cinder block house (#13), on the other hand, replaced a longhouse two decades ago and was clearly built with storage in mind. There we recorded not only some provisional discard within the structure (construction material like bricks, cinder blocks, stacked tiles and wood) but also equipment useful during the fall olive harvest like a ladder, plastic chairs, and a burlap bag.

50

In 2009, we observed goats stationed in an animal pen 50 meters to the south walking openly among the ruins of House #3. The reuse of houses as animal pens is not uncommon in the Greek countryside: cf. Forbes 2007, 231233.

29

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION

Table 5. Artifacts recorded inside the houses Few of the artifacts found outside the houses contribute to positive assessments about habitation despite the fact that the courtyard and surrounding fields would have been principal 30

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION arenas for domestic activities (Table 5).51 The presence of tractor tires, resin collectors, shotgun shells, and branch piles around several buildings points to the relatively recent use of the use of the land for, respectively, plowing, resin processing, hunting, and olive cultivation. The light scatter of ceramic, metal, plastic, and glass containers found at some houses (#s 2, 4, 6, 13) point to storage and/or consumption of food and liquids (e.g., sardine cans, the Nescafe frappe shaker) but could well reflect seasonal visits to the valley during the olive cultivation, or behaviors completely unrelated to the use of the house.52 The same can be said for the sole of the shoe, the comb, sock, and small mirror—their presence is too random and limited to suggest anything other than human use of the area.53 Construction material, however, is a common type of artifact and at least confirms the presence of buildings in the form of slumped and collapsed walls and roofs (#s 2 and 3), stacked tiles or wood (# 13), and light scatters of tile and brick (#s 4, 10, 11, 13, 14, and 16). This much is obvious already from the surviving fieldstone walls, but these eventually will fade into the land and the construction debris will become more important in defining the former habitation. Overall, the amount of artifacts noted in and around most of the houses is rather low (cf. Tables 5 and 6). Less than half (5 of 11) of the carefully documented houses contained assemblages inside that were substantial; field stones and a few tile fragments were the principal signature of the other six buildings. Outside the houses, artifact scatters were normally small, with only occasional moderate-density clusters of potsherds and trash (#s 4 and 13). As we noted earlier, our area of pedestrian survey incorporated several houses and showed that tile especially was an important signature of these buildings, while light scatters of table wares, kitchen wares, and storage vessels were observed in the fields around the houses. These observations at least allow us to conclude that while artifact clusters (including especially tile) may sometimes constitute signatures of habitation, that lower-density scatters are often all that is left of former houses in the modern countryside.

51 52

Sigalos 2004, 61-63. The laundry detergent and plastic water bottle around house #2 was observed in 2009 and was discarded from the gravel road above and not the house. Some of the plastic water bottles have presumably fallen from olive trees where they had been recycled as tree ornaments. 53 Murray and Chang 1981, Fig. 3, and Murray and Kardulias 1986, 33, note clothing, shoe soles, and socks associated with animal folds.

31

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION

Table 6. Artifacts recorded outside the houses Among the most interesting data we collected in our study was a series of observations about the state of the houses over an 8 year period between 2001/2002 and 2009 (see Table 5). It is true that in many of the houses, we noted no discernible changes, but the alterations that we did observe were not insignificant. As one example, we obtained two quite different snapshots of the objects inside House #10 between 2002 and 2009. During the first year, we noted the large southern room covered with hay, and observed an overturned table, plow, small metal 32

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION basin, ladder, wood pile, and several small objects (scissors, metal, leather, key), suggesting use for storage and animal pen; in 2009, the northern and southern rooms were significantly cleaner and neater (the hay had disappeared) and still included the table and key, but otherwise had a quite different assemblage: a wood saw, an empty barrel, broom, and a stack of 219 tiles.54 Even abandoned houses in the lakka show signs of artifact movement. We documented the stripping of hundreds of whole tiles from the roof beams of House #2 between 2001 and 2004,55 an event symbolizing the recycling behaviors that are common at the houses.56 Outside the houses, we observed small changes between 2002 and 2009 that indicate the houses remain centers of active land use. At two of the maintained houses (10 and 13), tiles, a small mirror, and socks disappeared between 2002 and 2009 while a barrel and water trough appeared. A Nescafe frappe shaker and plastic water bottle at House 16 were newly discarded probably during the October harvest while the laundry detergent container above House 2 suggests random discard from the road above. And abandoned and ruined structures were evidently good places to pile pruned olive branches inside and around (see Table 6). Finally, we observed in this brief span of time the rapid deterioration of the walls and roofs of the abandoned houses themselves (see images). Over this eight year period, we observed the progressive collapse of the tile roofs of several abandoned houses (#s 2, 3, 5, and 14) resulting in the wooden beams and hundreds of tiles falling inside and partially outside the structures. At two of the structures (#3 and 14), the houses floors were no longer visible by 2009, and in fact, were largely inaccessible, with roof beams and debris blocking entry; the household items left in the house were buried beneath the bulk of the building itself. The loss of the roof has typically entailed also the rapid deterioration of the walls as exposure to the elements has eroded the mud mortar and fieldstone walls have fallen out. *********************************************************** In the preceding sections, we have highlighted how rural agricultural houses and their associated artifacts, features, and environments in this small world reflect the contingency of
54

Both years suggest that the house was being used mainly for storage of agricultural and domestic implements, but the particular artifacts being stored were quite different. Cf. a similar list of agricultural implements documented at “storehouses” in the southern Argolid: Murray and Kardulias 1986, 31. 55 The event is driven home by a story told by Mt. Perras about two brothers who owned together an old house and had an irresolvable dispute, which resulted ultimately in one brother leaving but not before stripping his half of the roof tiles! The story highlights not only the curate behavior common to rural activities at Lakka Skoutara, but the relational and personal dimensions behind the observed archaeological patterns. Such kinship disputes that involved property ownership was relatively common in 20th century Greek villages. Cf. Forbes 2007, 164, 168, 232-234. 56 The storage of building supplies (brick, tile, cinder blocks) as provisional discard at several houses.

33

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION habitation and abandonment over periods of time ranging from decision-making moments to the patterns of centuries, and in tune with the broader region and world. Ultimately, these dynamic processes confound our definitions and categories. Given the tendency for seasonal occupation in the valley, can we say that Lakka Skoutara was ever fully inhabited in the Early Modern era other than during the war years of the 1940s? On the other hand, has it ever really been abandoned? The seasonal return of the inhabitants of Sophiko for the olive harvest, at least, shows how the abandonment of habitation need not mean the departure from the land. We will return to these questions in the final discussion, but before we do this, we can turn to a more direct discussion of the memory and meaning in this landscape reflected in the oral interviews.

IV. The Oral Interviews V. Toward an Archaeological History of the Rural Landscape

34

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION VI. References Alcock, S.E., J.F. Cherry and J.L. Davis. 1994. “Intensive survey, agricultural practice, and the classical landscape of Greece.” In I. Morris (ed.), Classical Greece: Ancient Histories and Modern Archaeologies, Cambridge, pp. 137-70. Bintliff, J. and A. Snodgrass. 1988. “Off-site pottery distributions: a regional and interregional perspective.” In Current Anthropology 29, pp. 506-513. Caraher, W. R., and L. Diacopoulos. 2004. “Less than a Village: Patterns of Rural Settlement in the Landscape of the Southeastern Korinthia” (Paper, San Francisco). Caraher, W. R., D. Nakassis, and D. K. Pettegrew. 2006. “Siteless Survey and Intensive Data Collection in an Artifact-Rich Environment: Case Studies from the Eastern Corinthia, Greece,” JMA 19.1, pp. 7-43. Chrysafi-Zografou, Metaxoula. 1987. Greek Traditional Architecture. Corinthia. Athens. Clarke, M.H. 2000. “Changing House and Population Size on Methana, 1880-1996: Anomaly or Pattern?” in Sutton 2000, Stanford, pp. 107-124. --- 1995. “From Shelters to Villas: Changing House and Settlement Form on Methana, 18801987,” In Pavlides and Sutton 1995, pp. 511-36. --- ______“Continuity and change in the family estate on Methana, Greece, 1900-1995,” in Ethnographika ??, pp. ??? Cooper, F. (ed.). 2002. Houses of the Morea: Vernacular Architecture of the Northwest Peloponnesos (1205-1955), Athens. ---2002. “The Morea Project.” In Cooper 2002, Athens, pp. 16-21. Coulton, M.B. and H.B. Foster, 2002. “Catalogue of Village Architecture,” in Cooper 2002, Athens, pp. 146-413. Creighton, O.H. and J.R. Segui. 1998. “The ethnoarchaeology of abandonment and post-abandonment behaviour in pastoral sites: evidence from Famorca, Alacant Province, Spain.” Journal of Mediterranean Archaeology 11, pp. 31-52. Diacopoulos, L. 2004. “????” in Effie F. Athanassopoulos and LuAnn Wandsnider (eds.), Mediterranean Archaeological Landscapes: Current Issues, Philadelphia. Dixon, M. 2000. “Disputed Territories: Interstate Arbitrations in the Northeastern Peloponnese, ca. 250– 150 b.c.” (diss. Ohio State Univ.).

35

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION Forbes, H. 1996. “The uses of the uncultivated landscape in Modern Greece: a pointer to the value of the wilderness in antiquity?”, in G. Shipley and J. Salmon (eds.), Human Landscapes in Classical Antiquity: Environment and Culture. London. Forbes, H. 2007. Meaning and identity in a Greek landscape: an archaeological ethnography. Cambridge. ---1997. “Turkish and Modern Methana,” in Mee and Forbes, Liverpool, pp. 101-117. Foster, H.B. 2002. “Village Architecture of the Morea,” in Cooper 2002, Athens, pp. 130-145. Fotiadis, M. 1995. “Modernity and the Past-Still-Present: Politics of Time in the Birth of Regional Archaeological Projects in Greece,” AJA 99, pp. 59-78. Foxhall, L. 1997. “Ancient Farmsteads, Other Agricultural Sites and Equipment.” In Mee and Forbes, Liverpool, pp. 257-68. Friedl, E. 1962. Vasilika: A Village in Modern Greece. New York. Halstead, P. 1987. “Traditional and Ancient Rural Economies in Mediterranean Europe: Plus ca Change?”, JHS 107, pp. 77-87. Koukoulis, T. 1997. “Catalogue of Churches,” in Mee and Forbes, Liverpool, pp. 211-256. Mee, C. and H. Forbes (eds.). 1997. A Rough and Rocky Place: the Landscape and Settlement History of the Methana Peninsula, Greece, Liverpool. Megas, Giorgos. 1951. The Greek House: its evolution and its relation to the houses of the other Balkan peoples. Athens. Murray, P. 1980. “Discard Location: the Ethnographic Data,” AA 45, pp. 490-502. Murray, P. and C. Chang. 1981. “An Ethnoarchaeological Study of a Contemporary Herder’s Site,” in JFA 8, pp. 372-381. Murray, P. M., and P. N. Kardulias. 1986. “A Modern-Site Survey in the Southern Argolid, Greece,” JFA 13, 21-41. ---2000. “The Present as Past: An Ethnoarchaeological Study of Modern Sites in the Pikrodhafni Valley,” in Sutton 2000, Stanford, pp. 141-168. Osborne, R. 1992 “‘Is it a farm?’ The definition of agricultural sites and settlements in ancient Greece.” In B. Wells (ed.), Agriculture in Ancient Greece, Stockholm, 21-28. Pavlides, E., and S.B. Sutton. 1995. Constructed Meaning: Form and Process in Greek Architecture. Minneapolis. Pettegrew, D.K. 2001. “Chasing the Classical Farmstead: Assessing the Formation and Signature of Rural Settlement in Greek Landscape Archaeology,” JMA 14(2), pp. 189−209. Philippides, D., ed. 1983-. Greek Traditional Architecture. Vol 1. Athens. 36

WORKING PAPER DO NOT CITE WITHOUT AUTHORS’ PERMISSION Sigalos, E. 2004. Housing in Medieval and Post-Medieval Greece. BAR International Series 1291. Oxford. Sutton, S.B. (ed.). 2000. Contingent Countryside: Settlement, Economy, and Land Use in the S. Argolid Since 1700. Stanford. --- 1995. “Crumbling walls and bare foundations: the process of housing in Greece,” in Pavlides and Sutton 1995, pp. 319-44. --- 1994. “Settlement Patterns, Settlement Perceptions: Rethinking the Greek village,” in Kardulias, P.N., ed., Beyond the Site: Regional Studies in the Aegean Area. Lanham, Maryland, pp. 313-335. Tartaron, T. F., T. E. Gregory, D. J. Pullen, J. S. Noller, R. M. Rothaus, J. L. Rife, L. Diacopoulos, R. L. Schon, W. R. Caraher, D.K. Pettegrew, and D. Nakassis. 2006. “The Eastern Korinthia Archaeological Survey: Integrated Methods for a Dynamic Landscape,” Hesperia 75.4, pp. 435-505. Tzortzopoulou-Gregory, L. 2008. “Cemeteries in the Countryside: An Archaeological Investigation of the Modern Mortuary Landscape in the Eastern Corinthia and Northern Kythera.” In W.R. Caraher, L.J. Hall, and R.S. Moore, eds., Archaeology and History in Roman, Medieval and Post-Medieval Greece: Studies on Method and Meaning in Honor of Timothy E. Gregory, Aldershot. Wagstaff, J.M. 1982. The Development of Rural Settlement: A Study of the Helos Plain in Southern Greece. Amersham, England. --- 1965. “Traditional Houses in Modern Greece.” In Geography 50.9, pp. 58-64. Wagstaff, J., and S. Augustson. 1982. “Traditional Land Use,” in Renfrew, A.C., and J. Wagstaff, eds., An Island Polity. Cambridge, pp. 106-33. Whitelaw, T.M. 1991. “The Ethnoarchaeology of Recent Rural Settlement and Land Use in Northwest Keos,” in J.F. Cherry, Jack L. Davis, and E. Mantzourani, eds., Landscape Archaeology as Long-Term History: Northern Keos in the Cycladic Islands. Monumenta Archaeologica 16. Los Angeles, pp. 403-454.

37