The Karystian Kampos Survey Project: Methods and Preliminary Results

THE KARYSTIAN KAMPOS SURVEY PROJECT: METHODS AND PRELIMINARY RESULTS

Tankosić, Ž.1 and Chidiroglou, M.2

1Indiana University, Department of Anthropology, Student Building 130 Bloomington, IN 47405, USA 2 Hellenic Ministry of Culture, 20 Bouboulinas Str., 10682 Athens, Greece, ([email protected])

Received: 30/11/2009 Accepted: 16/02/2010 Corresponding author: [email protected]

ABSTRACT The Karystian Plain (the Kampos) is part of the Karystia that had not been archaeo‐ logically explored in a systematic way despite its obvious economic importance for the inhabitants of southern Euboea. In the course of two seasons of fieldwork the Kampos Survey was able to cover approximately forty percent of the designated survey area. In the process we located 36 previously unknown archaeological findspots ranging in date from the end of the Neolithic to the Late Byzantine times. In this paper we present our preliminary results. Our data suggest that the Kampos was as important to the ancient Karystian as it is to the area’s inhabitants today.

KEYWORDS: Karystian Plain, prehistory, Classical and Roman periods, obsidian, sur‐ face survey 12 TANKOSIC, Ž. & CHIDIROGLOU, M.

INTRODUCTION tential of the Kampos as well as its geo‐ In this paper we present preliminary graphical location between Karystos and results of the two seasons of intensive ar‐ the sheltered deep water port at Marmari chaeological field survey of the Karystian were added incentives to the research pro‐ Plain or Kampos. The project is organized ject. The Karystia, although somewhat by the Southern Euboea Exploration Pro‐ marginal in modern times, had been an ject (SEEP) with the permission from the important region in the past. Previous Hellenic Ministry of Culture. The project work in the area by both the Greek Ar‐ is funded by the Institute for Aegean Pre‐ chaeological Service represented by the th history (INSTAP) and by the personal 11 Ephorate of Prehistoric and Classical funds of the late Professor Malcolm Wal‐ Antiquities and by SEEP provided a size‐ lace. The fieldwork part of the project took able body of data to support this assertion. place during the summers of 2006 and Although the eastern part of the Kampos 2007. Currently, the detailed study of the had been explored in the course of previ‐ material and other data collected in the ous work, large gaps in our knowledge of field is under way by an international the area, potentially very important for team of scholars. reconstructing the Karystian past, still re‐ The Kampos is located northwest of the mained. It was our goal to fill in those Bay of Karystos (fig 1) and stretches gaps. roughly east‐west, starting at the head of The survey target area in the plain cov‐ the bay, where it is the broadest and ers approximately 260 ha (fig 2), including where the modern town of Karystos is lo‐ the lower foothills of the Karpaston range cated, and tapering somewhat towards to the north and of the Paximadhi Penin‐ Marmari on the western coast. sula to the south.

Figure 2 Topographic map of the Kampos with the boundaries of the survey area. North is up. The boundary of the survey area in the east coincides with the westernmost ex‐ tent of Dr. Donald Keller’s dissertation Figure 1 The Karystian Kampos viewed from survey (Keller, 1985). Our aim was to the east. South Euboean Gulf and east coast of learn as much as possible about the use of Attica in the background. the Kampos in the past and to locate the On the north it is bordered by the Kar‐ focal points of that use in the form of paston range and on the south by the Bay findspots. Since the Kampos is under of Karystos and the northern foothills of rapid modern development, obtaining the Paximadhi Peninsula. It is the largest information on the exact location of ar‐ agriculturally viable piece of land in the chaeological sites in the area will help in area and is also fairly well watered, espe‐ their protection. An important goal of the cially in the spring. This agricultural po‐ Kampos survey was to compare in the THE KARYSTIAN KAMPOS SURVEY PROJECT: METHODS AND PRELIMINARY RESULTS 13 same setting some of the methods used previously in the field by SEEP in order to gauge their reliability and effectiveness. In addition, we wanted to examine whether some of the routes that exist today in the Kampos have been used as thoroughfares in the more distant past. Finally, we wanted to examine the effects of alluvia‐ tion on the visibility of ancient remains in Figure 3 Survey coverage. Dark gray – the 2006 the plain. transects. Light gray – squares surveyed in 2007. Courtesy of Google Earth.

METHODS “Stratified sampling” in this particular case meant that the squares to be surveyed In 2006 the survey methodology con‐ were not chosen randomly but with re‐ sisted of two to four people walking in spect to specific predetermined guide‐ transects determined mainly by natural or lines; e.g., to include all of the different man‐made features of the terrain (fig 3). geomorphological features present in the This method intentionally corresponded survey area as well as taking into account to the “route survey” approach used by some of the results of the 2006 season. The SEEP east of the Bay of Karystos on the 100 x 100 m squares were further divided Bouros‐Kastri Peninsula. Our transects into ten 10 x 100 m transects, each sur‐ followed paved and unpaved roads, foot‐ veyed by a team member. Total collection paths, gullies, ravines, and current or sea‐ was carried out on the surface of all the sonal water flows; in short, all features of areas designated as findspots, while the the terrain that could serve as thorough‐ thin material scatter between the findspots fares. Additionally, two long arbitrary was recorded (in case of non‐diagnostic transects that roughly coincided with the material) or recorded and collected (diag‐ edges of the survey area were surveyed on nostic sherds, obsidian, etc.). the north and south foothills bordering We were able to locate 20 findspots in the Kampos. The team members were also 2006 and 16 in 2007. Judging only by the allowed to depart from the transects and number of discovered findspots it would investigate any areas that seemed promis‐ be possible to conclude that the extensive ing as potential locations of findspots. method employed in 2006 was more pro‐ Only a small representative sample of ma‐ ductive. However, the degree of efficiency terial was collected during this phase of notwithstanding, the route‐survey method the survey and only from locations that was not able to address other aspects of were designated as findspots. the use of the Kampos—namely the offsite The 2007 survey was executed using the distribution of archaeological materials. intensive stratified sampling method. The Moreover, many of the findspots discov‐ survey team in the 2007 season consisted of ered in 2007 are located not far from the seven to ten people in the field at any time. 2006 survey transects. This shows the in‐ The entire survey area was divided into adequacy of the extensive survey when 100 x 100 m squares, which were then sur‐ trying to locate all past remains in a given veyed independently or in clusters using area and especially the ones located at un‐ the stratified sampling approach (fig 3). expected locations. Therefore, the main 14 TANKOSIC, Ž. & CHIDIROGLOU, M.

`difference between the methods used in “settlements” as they are traditionally de‐ 2006 and 2007 field seasons is not in the fined based on surface finds—we did not quantity but in the quality of data that find any architectural remains and pottery was acquired. is present only in the form of a few very small fragments. Most of the prehistoric RESULTS OF THE SURVEY findspots are characterized by obsidian scatters, some of which are among the We were able to locate and record 36 largest found in Greece thus far (William new archaeological findspots in the course Parkinson pers. comm.). Good illustra‐ of the survey. As we suspected, the plain tions of this are findspots 06N16, 07S28, was not a forgotten area in the past but and 07N35 (fig 4). was the location of lively activity in both prehistoric and historic periods. The prehistoric finds were the most surprising. At the beginning of the project one of our working hypotheses was that prehistoric remains would be largely cov‐ ered by alluvium. Consequently, we only expected to find evidence of the prehis‐ Figure 4 Locations of findspots mentioned in toric use of the plain in areas located on the text. Triangles mark prehistoric and circles higher grounds and on other locations Classical/Hellenistic/Roman findspots. North is up. away from the heaviest effects of alluvia‐ tion. Contrary to our expectations, we The findspot 06N16 is located on a have found at least 15 findspots that are ridge north of the modern road to Mar‐ either purely prehistoric or have a prehis‐ mari and above the small Ay. Photeini toric component. They are located on both chapel. The site consists of a large number schist outcrops and on the flat alluvial soil of obsidian flakes, tools, and core frag‐ of the plain. Moreover, some findspots ments (ca. 400 pieces) and small pottery located on sloped terrain (e.g., findspot fragments. The evidence suggests blade 06N16) did not show evidence for defla‐ and flake production on the site (William tion and displacement of materials, which Parkinson pers. comm.). Three tanged would have been the case if they were af‐ points also came from this findspot. fected by alluvial or colluvial shifts of the Findspot 07S28 is located in the western topsoil. This suggests that either the major part of the Kampos, around a rocky out‐ alluvial events that participated in the crop at the edge of a modern farmstead. A formation of the Karystian Kampos origi‐ total of 287 obsidian pieces were collected nated earlier than our earliest prehistoric from the area of about 100 x 50 m. The finds (dated to the Final Neolithic or at significance of this site is that almost the least the EBA) or that alluviation in the entire lithic reduction sequence is repre‐ plain was topical—i.e., it affected different sented among the finds. Findspot 07N35 is parts of the plain in different ways. Final located on the eastern side of the survey answers to these questions can only be area and in the immediate vicinity of the provided by geological analyses. What is excavated EBA site of Ay. Georgios. The puzzling is that we did not find any pre‐ findspot consists of a large obsidian scat‐ historic findspots that could be termed ter of about 2500 fragments stretching THE KARYSTIAN KAMPOS SURVEY PROJECT: METHODS AND PRELIMINARY RESULTS 15 over an area of at least 150 x 100 m. We periods (fig 4) with parallels in Attica and also surveyed parts of this findspot in most of the Cycladic islands. more detail using a 2 x 2 m grid. Our goal Findspot 06N14 is located at the north was to check for any patterning in the ma‐ central edge of the Kampos, on a rocky terial distribution that would help us outcrop. Three sets of steps cut on two identify the specific activities taking place different levels in the natural rock lead to at this location. The 07N35 assemblage a platform at the summit of the outcrop consists of tools made on pressure flakes, and to several niches. Parts of a dry‐stone many of which can be identified as wall of N‐S direction are visible to the notches and borers, but no blade cores west of the outcrop. Black glazed sherds, were found. The material is chronologi‐ belonging to skyphoi and a kantharos and cally consistent with the LN/EB period plain ware sherds, together with millstone (William Parkinson pers. comm.). On the fragments and a black glazed rooftile were basis of current data this findspot could be found at this site. These finds, together tentatively interpreted as a specialized with the form and location of the pits, the site, a place where particular chipped niche, the rock cut steps leading to it, and stone tools were produced or modified to the possible perivolos wall suggest that the be used for specific, yet unclear, activities. remains belong to a small Classical rural There is little evidence for blade produc‐ sanctuary. Another probable sanctuary tion at the site, which suggests that the with a rock cut niche on a stone outcrop blades were produced elsewhere. This facing east with surface finds of Roman seems even more probable given that the times was identified at findspot 06N19. nearby EBII site of Ay. Georgios, exca‐ An additional rectangular cut was found vated by the 11th Ephorate, produced evi‐ on the top of the outcrop. Plain and dence for the full reduction sequence ac‐ combed Roman sherds were observed at companied by a wealth of lithic material this site. (Sapouna‐Sakellaraki 1992 and personal Remains of a Byzantine chapel (06N11) observations). were found not far from the northeastern We located a limited number of Classi‐ border of the survey area. It is possible cal and Hellenistic findspots in the north‐ that some of the Cape Mnima sandstone western and western region of the Karys‐ blocs used in its construction belong to an tian plain during the 2006 and 2007 survey earlier edifice—perhaps a Roman tower. A seasons. The large number of Roman poorly preserved circular stone construc‐ findspots, as well as the large amount of tion (06N12) preserved at a height of ca. Roman material scatter, in comparison to 0.50 m and located in the immediate vicin‐ the distribution range of the Classical to ity of 06N11 could represent the remains Hellenistic findspots, can hypothetically of a round Classical tower (cf. for Attica reflect varying degrees of intense Roman e.g. Suto 1993, 3, table 1); however, at this farming versus pastoral activities in the point there is no sufficient evidence to Kampos, from the 5th century B.C. to the support that. The other, perhaps more 2nd century A.D. Due to the visible archi‐ likely, possibility is that these remains tecture and their prominent locations, represent an old threshing floor. sanctuaries were the most clearly visible Many findspots consisted of scatters of remnants of the Classical and Hellenistic Early to Late Roman plain and combed ware pottery of different intensity includ‐ 16 TANKOSIC, Ž. & CHIDIROGLOU, M. ing pithos sherds, amphora and jug strap tested in the Kampos during the EBA and handles, rim and base parts, other utilitar‐ Roman periods, a fact that indicates inter‐ ian vessel fragments, and column frag‐ est in the resources of the plain during ments and tiles. Roman constructions or these times. Intense olive or wine cultiva‐ building remains were visible at several tion could be the reason during the Ro‐ locations, such as findspot 06S07, where man period. Due to the lack of palynologi‐ parts of a water pipe were found appar‐ cal studies, this hypothesis remains a pos‐ ently in situ. A number of cist graves, sible but unsubstantiated way of explain‐ probably dating to the Roman period, ing the density of Roman sites. At the were found at findspot 06S08 (Chidi‐ moment, it is more difficult to interpret roglou 1997, 405 and 1998, 363‐366). the interest that prehistoric populations At other findspots Roman surface ma‐ had with the Kampos, but it is likely that terial was mixed with Byzantine, as at it was also based on its significant agricul‐ findspots 06S05 and 06S06 and at the site tural potential. of the modern church of Eisodia Theoto‐ The scarce Classical to Hellenistic finds kou, where traces of a Byzantine chapel can be variously explained. One should (06N13), together with six blocks of Cape keep in mind the factor of chance in locat‐ Minima sandstone, two fragmentary cipol‐ ing sites in alluvial soils or areas with long lino columns and a small white marble centuries of cultivation and of building Ionic capital were also found. Blocks and material recycling. Moreover, the perti‐ parts of stone olive or wine presses were nent material evidence has not been stud‐ found reused in various modern rural in‐ ied in full and some reinterpretations are stallations in the Kampos, serving as a possible. On the other hand, one could reminder of the perennial recycling of an‐ hardly expect the Karystians, who suf‐ cient building materials. A good example fered during the Persian wars and were of this practice is findspot 06N02 where a then for years penalized by the Athenians Roman sandstone block serves as a door‐ (Herodotus, 6.99. 2, 8.112. 2, 8.121.1, 9.105; step to the modern installation and a Thucydides 1.98.3, 4.42.1, 4.43.3, 7.57.4, schist counterbalance block of an ancient 8.69.3; also see Wallace 1972, 171‐254), to olive press (cf. Brun 2004, 98, 102) has expand their cultivation in the Kampos in been incorporated into the (relatively) these difficult years. modern threshing floor nearby. To conclude: despite our relative suc‐ cesses more work remains to be done in DISCUSSION AND CONCLUSIONS the Kampos, not only in the study of data collected during the most recent survey, Our results allow us to formulate sev‐ but also in the form of further archaeo‐ eral tentative conclusions. The area of the logical exploration of the areas not yet Karystian Kampos, probably due to its surveyed and in the form of additional economic importance, has been inhabited detailed research at particular findspots. from Prehistoric to modern times. It seems that the densest human presence is at‐

ACKNOWLEDGMENTS We would like to thank all the people involved with the organization and execution of the Kampos Survey. We are grateful to INSTAP, Indiana University’s Schrader Fund, THE KARYSTIAN KAMPOS SURVEY PROJECT: METHODS AND PRELIMINARY RESULTS 17 and Malcolm Wallace for generously providing funds for this project. We are particu‐ larly grateful to Dr. Donald Keller for his involvement in the survey and for providing valuable comments on an earlier draft of this paper.

ABBREVIATIONS ArchDelt Archaeological Report (Αρχαιολογικόν Δελτίον)

KODAI Kodai. Journal of ancient history

REFERENCES Brun, J.P. (2004) Archéologie du vin et de l’huile de la Préhistoire à l’époque hellénistique, Paris. Chidiroglou, M. (1997) Karystos, Xanemo Palaeohora (Κάρυστος, Ξάνεμο Παλαιοχώ‐ ρας , οικόπεδο Ιω. Γκιλιπάθη) (in Greek). ArchDelt 52 (Chronika), 404‐405. Chidiroglou, M (1998) Karystos, Xanemo (Κάρυστος, Ξάνεμο, οικόπεδο Ιω. Γκιλιπάθη) & Οικόπεδο Γυμνασίου –Τεχνικού‐Επαγγελματικού Λυκείου) (in Greek). Ar‐ chDelt 53 (Chronika), 363‐366. Keller, D.R. (1985) Archaeological Survey in Southern Euboea, Greece: A Reconstruction of Human Activity from Neolithic Times through the Byzantine Period. Ph.D. Thesis, Indiana University. Sapouna‐Sakellaraki, E. (1992) building plot of public electrical company, at Aghios Georgios Kampos (Οικόπεδο Δ.Ε.Η. εκτός σχεδίου πόλεως στη θέση Άγιος Γεώργιος Κάμπου)(in Greek). ArchDelt 47 (Chronika), 177‐178. Suto, Y. (1993) Isolated Farms in Classical Attica, KODAI 4, 1‐19. Wallace, M.B. (1972) The history of Karystos from the sixth to the fourth centuries B.C. Ph.D. Thesis, University of Toronto.

Editions of ancient texts:

How, W.W. and Wells, J. (1928) A Commentary on Herodotus, Oxford. Jones, H.S. and Powell, J.E. (1954‐1955) Thucydidis. Historiae, Oxford.