HESPERIA 76 (2OO7) THE BUSY COUNTRYSIDE Pages 743-784 OF LATE ROMAN CORINTH

Interpreting Ceramic Data Produced by Regional

Archaeological Surveys

abstract

Using data generated by the Eastern Korinthia Archaeological Survey, the author examines the evidence for the frequently attested "explosion" of Late Roman settlement in the Corinthia, assessing the degree towhich the differ ential visibility of pottery from the Early and Late Roman periods affects our over perception of change time. Calibration of ceramic data to compensate for amore differences in visibility demonstrates continuous pattern of exchange, use on habitation, and land the Isthmus during the Roman era. The author also excavated and surface from other compares assemblages regional proj and new of the ceramic evidence ects, suggests ways interpreting produced by archaeological surveys.

In conventional narratives of the Late Roman period, Corinth, like the run entire province of Achaia, shares in the of afflictions known to histo to A.D.: rians and chroniclers of the 3rd 6th century earthquakes, plagues, barbarian invasions, abandoned lands, and oppressive taxation, among other disruptive forces.1 Over the last 20 years, however, scholars have questioned,

1.1 am to Robert as well as from audience and grateful the codirectors of Schon, reading critiquing of drafts of this to at the annual comments the Eastern Korinthia Archaeological response presentations manuscript; their have of the Insti Survey (EKAS), Timothy Gregory and meetings Archaeological greatly improved the argument and Daniel to the tute in Pullen, and field director, of America Philadelphia (2004) style of the paper. Thomas for and San the was Tartaron, encouraging Diego (2007), American My research supported by me to and the EKAS School of Classical Studies at Athens from analyze present grants the Ohio State University as tea Roman data, first part of my disser talk series (February 2006), and the (Ruth Higgins Award for Summer tation on the Late Roman Corinthia in Archaeological Survey Meetings orga Research the Department of History, now in nized the Canadian and Finnish Isthmia Fellow (Pettegrew 2006), and this by History Department The here in am study. analysis presented Institutes Athens (2006). I ship, College of Humanities G. Mi makes use of Microsoft Access to Amelia and especially grateful Brown, chael Riley International Academic ArcGIS data structures created William by Caraher, Timothy Gregory, Fund, and Office of International Edu Richard Rothaus and Lee Anderson. Anthony Kaldellis, Dimitri Nakassis, cation Dissertation Research Grant), In this I have Nathan and Kathleen at preparing article, Rosenstein, the Jacob Hirsch Fellowship the as well as to benefited from conversations with Slane, Tracey Cullen, Mark American School of Classical Studies Yannis Lolos, Melissa Moore Morison, Landon, and the two anonymous (2005-2006), and Conference Partici Clare reviewers for their Funds from Pickersgill, Guy Sanders, and Hesperia thorough pation Messiah College.

? The American School of Classical Studies at Athens 744 DAVID K. PETTEGREW

a diminished, and contextualized these terrors in light of growing corpus of archaeological research indicating that the social and economic life of the was Late Roman province anything but depressed. The principal impetus a for the revision has been series of archaeological survey projects that have an produced evidence for explosion of settlement in the countrysides of Greece during the Late Roman period. If the territories of Late Roman Achaia appear to have been thriving, how then could the province have been in a state of general decline? consensus on The current the ancient countrysides of Greece is that a the proliferation of habitation indicates Late Roman revival in the social a and economic life of the province of Achaia after depression earlier in the Roman period.2 Beginning in the 4th century A.D., according to this view, the province experienced significant agricultural intensification and economic prosperity, tied perhaps to population growth, the production or an of olive oil for export, imperial policy of promoting smallholding are farmers. Whatever the cause, Late Roman remains highly visible in a a the Greek countryside, fact that should indicate healthy, not depressed, one economy. A similar pattern of proliferating settlement has led scholar a to speak of the "busy countryside" of Late Roman Cyprus,3 description that is also fitting for Greece and other regions of the Aegean. Despite the widespread recognition of this pattern, however, there has a been little scholarship dealing with number of key interpretive problems, especially the "source criticism" of survey pottery. Archaeologists have long more recognized that the material culture of the later Roman period is never to measure visible than that of other periods, but have attempted our the degree to which such differential visibility affects interpretation of change between the earlier and later Roman periods. How busy, in reality, was the countryside of late antiquity compared with that of the preceding and following periods? In this study I address the issue of change in the Roman countryside on a based critical analysis of the ceramic data collected by the Eastern a Korinthia Archaeological Survey (EKAS). I show how the perception of Late Roman "settlement explosion" in the Corinthia and other regions of Greece is significantly affected by the differential visibility of the Early and turn a Late Roman periods, which in is product of the high visibility of Late Roman pottery, the nature of archaeological survey sampling regimes, and the well-developed distribution networks of late antiquity. While my analysis of a a the material reveals phenomenon different from that suggested by simple, literal reading of the evidence, in the end it reinforces rather than detracts a a eastern from picture of vibrant Late Antique economy inGreece and the Corinthia, and suggests that the structures of Corinthian trade and settlement that in an earlier Roman continued into the 6th and 7th developed period 2. For historical discussion general the broad cultural transformations of the era. recent seeWard-Perkins centuries, despite and syntheses, a I begin with broader exposition of the pattern and problems of 2000a, p. 321; Banaji 2001, pp. 16-17, in the Kosso the Late Antique countryside, and of the way which Greece and 214; Shipley 2002, pp. 329-331; ceramic 2003, pp. 31-52; Chavarria and Lewit Corinthia in particular fit into that pattern. Next, I present the 18-19. data for the Roman from EKAS and the material 2004, pp. period analyze along 3. Rautman 2000. Marcus Raut as as ceramic with the Late Roman data from other surveys well of Late Roman published man's "busy countryside from excavated contexts. I then discuss the of as the for assemblages implications Cyprus" served inspiration our the title of the article. this analysis for understanding of the Roman countryside and suggest present THE COUNTRYSIDE OF LATE ROMAN CORINTH BUSY 745

different ways of calibrating the data to compensate for differential vis some state ibility. Finally, I draw historical conclusions about the of the Corinthia in late antiquity.

THE LATE ANTIQUE COUNTRYSIDES OF GREECE

The last two decades of scholarship have transformed and greatly expanded our understanding of the history of the Late Roman countryside in the eastern Mediterranean. Older visions of abandoned lands, autarkic estates, to exploited coloniy and general economic decline have given way depictions of healthy and prosperous territories with prolific medium-sized farms, strong village centers, and well-connected economies. This vibrant rural world lasted to the end of the 6th century, and there is now evidence sug gesting that, inmany regions of the eastern empire, the prosperity continued into the 7th and 8th centuries as well.4 re The basis for the recent revision lies above all in the widespread gional archaeological research occurring in both the eastern and western a provinces, coupled with better understanding of the ceramic chronologies an for the period.5 Greece and the Aegean have assumed important role eastern in the Roman world generally, ranking alongside Israel, Cyprus, as and Syria countries in which the rural landscape has received careful a archaeological investigation. This is product of the frequent archaeologi rescue cal work conducted in Greece, including both ongoing excavations numerous a and regional survey projects (Fig. 1), which have documented countryside filled with Late Antique sites.6 A remarkably consistent regional pattern of proliferating Late Roman across a settlement Greece and the Aegean has fueled revisionist views of strong Late Antique economy7 Table 1 lists intensive regional surveys that on have produced data settlement patterns between the Late Hellenistic and Early Byzantine periods, and indicates the ways in which these proj ects have defined the subphases of the Roman period.8 From central and

r?a 18-19. 4. See, generally, Ward-Perkins and Lewit 2004, pp. nagh 2000, p. 106; Shipley 2002, 2000a; Hirschfeld 2001; Banaji 2001; 8. AEP: Jameson, Runnels, and van pp. 268-273, 326-336. PRAP: Alcock Bowden, Lavan, and Machado 2004. Andel 1994, pp. 255-256,400-404, et al. 2005, pp. 152 (table 1), 164 (ta For older views, see, e.g., Rostovtzeff table A.l. Methana Survey: Bowden bles 3,4), 167 (tables 5, 6), 179-188, 194-204. Patras 1926;Jonesl964,p.812. and Gill 1997a, 1997b.NVAP: Wright Survey: Petropoulos 5. For settlement patterns in the west, et al. 1990, pp. 616-617; Alcock 1993, and Rizakis 1994, pp. 198-207. Boeotia see Lewit 1991. On the importance of pp. 41,43-44; Kosso 2003, pp. 31-52, Survey: Bintliff and Snodgrass 1985, as a source on data to her survey for understanding the based made available by 1988a; Bintliff 1991. Stanford Skourta Late rural see the directors. Berbati-Limnes Plain French Antique economy, Ward project Survey: 1990, pp. 35-36; Munn and Munn 1990. Perkins 2000a, pp. 315-317; Chavarria Archaeological Survey: Forsell 1996, Oropos Survey Lewit 336-337. Asea For 60 and 2004, pp. 4-6. pp. Valley Survey: Project: Cosmopoulos 2001, pp. a 6. For synthesis incorporating the s?n, Fors?n, and Lavento 1996; Kari 64, 78-79. SEEP: Kosso 2003, pp. 31 rescue see 52. Keos results of excavations, Avra vieri 2003; Fors?n and Karivieri 2003, Northern Survey: Cherry, mea and Mantzourani 327 1997. For general discussion of pp. 307-312; Fors?n and Fors?n 2003, Davis, 1991, pp. see Other regional surveys, Alcock 1993, p. 334.Megalopolis Field Survey: Lloyd, 347,481. archaeological surveys and mentioned in this but pp. 33-49; Shipley 2002, pp. 329-331; Owens, Roy 1985; Roy, Owens, study omitted from 1 await Kosso 2003, pp. 31-52. and Lloyd 1988; Roy, Lloyd, and Table further analysis and fuller 7.Ward-Perkins 2000a, p. 321; Owens 1989; Lloyd 1991. Laconia publication of the Roman and Late Roman Banaji 2001, pp. 16-17,214; Chavar Survey: Lawson 1996; Mee and Cava material. 746 DAVID K. PETTEGREW

Figure 1.Map of Greece and the southern Greece to the the Late the locations of Aegean islands, Antique period appears Aegean, showing to a time and a mentioned in the be of settlement expansion recovery, with only few excep regional surveys as text: eastern Ne tions. Moreover, the data presented inTable 1 indicate, the Late Roman (1) Corinthia; (2) a mea Berbati-Limnes area; explosion of settlement is emphasized by dearth of sites immediately valley; (3) (4) southern (5)Methana; before and after: the period of material abundance is sharply defined by Argolid; (6) Boiotia; (7) Skourta plain; periods of material absence. southern recent (8) Oropos; (9) Euboia; These sharp contrasts have been central to discussions of the (10) northern Keos; (11) Patras; province of Achaia in the Early Roman and Early Byzantine periods. At (12) Asea (13) one valley; Megalopolis; end of the spectrum, the absence of Early Roman material emphasizes (14) Pylos; (15) Laconia; and the of the Late Roman. Susan Alcock, for argues that strength example, (16) Kythera Roman imperialism dramatically restructured the Late Hellenistic-Early new Roman landscape, leading to entirely patterns of land distribution and a nucleated settlement before reversal in the later Roman period led again a to dispersed settlement pattern.9 At the spectrum's other end, scholars to have linked Late Antique Achaia broader discussions about the end of a new the Roman world and the creation of Byzantine society: the absence contrasts of settlement in Early Medieval Greece sharply with the ubiqui even a tous settlement of the preceding period, and may be product of the latter's demographic health, if overpopulation and the overtaxing of the an soil in late antiquity led to ensuing violent reversal.10 The most common explanation of the Late Antique pattern (Table 2) is a an that it represents recovery and expansion of settlement and agriculture, a more presumably indicating healthier economy, intensive agricultural 9. Alcock 1993. contrast to a 10. 1994. practices, widening markets, and/or population growth, in Gregory TABLE 1. INTENSIVE SURVEYS AND PERCEIVED PATTERNS OF ACTIVITY

Roman Project Chronological Definitions Late Hellenistic Early Middle Roman Late Roman Early Byzantine

Recovery and b.c. Early Roman (50 expansion, Rapid decline a.D. 200); Middle Roman at Argolid Exploration Decline Low level Low level gradually from late 6th Project (AEP) (a.d. 200-400); Late Roman first, peaking century (a.d. 400-650) from late 4th to 6th century

= Roman Early Roman Contraction by (100 B.c.-A.D. 100) and Low level, but Low level, but Gradual increase late 6th century, some con Methana Survey Middle Roman (a.d. 100 Decline increase from increase from through 4th and sites 300); Late Roman (a.d. 300 Late Hellenistic Early Roman 5th centuries tinuing into 7th 700) century

b.c Nemea Valley Early Roman (30 a.d. Archaeological 250); Late Roman Decline Low level Slight increase Marked decline Project (NVAP) (a.d. 250-650)

Roman (30 b.c Berbati-Limnes Early Decline from a.d. 150); Middle Roman Archaeological Classical-Helle Low level Slight increase Increase (a.D. 150-300); Late Roman Survey nistic to Roman (a.D. 300-700)

b.c Early Roman (1st century later 2nd a.D.); century Increase from Middle Roman (later 2nd cen cen Asea Valley Survey Low level later 1st Low level Increase Dark Age tury-early 4th century a.D.); tury A.D. Late Roman (4th-6th cen tury A.D.)

Early Roman (to 3rd century Megalopolis Field a.D.); Late Roman (from Decline Low level Increase Survey 3rd century a.d.)

cen Early Roman (lst-3rd Decline, but = tury A.D.); Middle Roman Continued from Lower level Laconia Survey High level higher than Dark Age (3rd-4th A.D.); Late Roman abandonment? century Late Roman Early Roman (5th-7th century a.d.)

Pylos Regional Early Roman (31 b.c a.d. Archaeological 400); Late Roman High level High level High level Decline Project (PRAP) (a.d. 400-700)

Patras Survey Decline Increase Decline Marked decline

Significant Late increase from Hellenistic-Early Decline in 7th b.c-a.d. Boeotia Survey Roman (200 300); Decline Low level 4th century, and Late century Roman (a.d. 300-650) especially 5th and 6th century

Early Roman (lst-3rd cen Stanford Skourta Decline after mid tury a.D.); Late Roman Decline Low level Increase Plain Survey 6th century (4th-6th century a.d.)

Early Roman (a.d. 1-200); Oropos Survey Middle Roman (a.d. 200 Low level, but Decline Low level Expansion Project 400); Late Roman slight increase (a.d. 400-700)

Roman (100 b.c Southern Euboea Early a.d. 200); Middle Roman Dramatic Exploration Project Decline Low level Marked decline (a.d. 200-400); Late Roman increase (SEEP) (a.d. 400-600)

Northern Keos Roman (a.d. 1-300); Early Decline Low level Significant Survey Late Roman (a.d. 300-700) 748 DAVID K. PETTEGREW

TABLE 2. INTERPRETATIONS OF ROMAN PATTERN BY SURVEY PROJECTS

Project Interpretation

Economic in the recovery Late Roman period. The fragmentation of the Roman in new new empire resulted development of regional markets and trading networks (AEP) Argolid Exploration Project from the 4th a.d. to late century The Argolid's connection markets in olive oil stimulated settlement intensification and population growth.

estates Initial depopulation and predominance of larger in the Early Roman Methana Survey period, followed by intensification of agriculture in the Late Roman period. A.D. Prosperous and flourishing in 5th and 6th centuries

Early Roman pattern indicative of nucleated settlement. Return of popu Berbati-Limnes perhaps Archaeological Survey to in lation and prosperity the valley late antiquity.

to Early Roman (later lst-2nd century a.d.) shift from hamlets and villages rural Asea Middle Roman Roman Valley Survey villas; decline; Late flourishing with rural villas. Greater a.d. prosperity and possibly higher population until later 6th-early 7th century

Decline of rural in the Roman as a result of Field economy Early period, perhaps Megalopolis Survey redistributed wealth economic in the and population; recovery Late Roman period.

to Diverse settlement trends according survey area, quality of land, and proximity to in to Sparta. General reduction settlement from Hellenistic Roman. Early in most Laconia Survey Roman peak followed by decline and probable abandonment of survey area a.d. to from 4th century No archaeological evidence for sites dating the 7th-9th century a.d.

and the Increased levels of dispersed settlement artifacts throughout Roman in sites some period, with differentiation types of and preference for coastal Pylos Regional Archaeological Project locations, settlement area. Pattern more (PRAP) although patterns vary by survey suggests land but in the amount intensive use, changes of identified pottery could also be access to wares. explained by changing levels of imported

economic recession economic Early Roman followed by Late Roman revival.

Boeotia Survey Late antiquity prosperous, with expanding population, agriculture, economy, and settlement.

in Stanford Skourta Plain Survey Prosperity late antiquity.

return More human activity in the Late Roman countryside, indicating the of Oropos Survey Project small-scale agriculture and greater overall prosperity.

in not clear. Causes of significant increase Late Roman activity entirely Possibly a of in the Late Hellenistic result of the restructuring territory and depopulation Northern Keos Survey more extensive cultivation in the Roman before period followed by Early period, a return to intensive cultivation in the Late Roman period.

a inhabited and before 11. see previously sparsely countryside large-scale regional For the general pattern, abandonment in the 7th Kosso has even that the century.11 Cynthia argued Gregory 1985,1994; Bintliff and van Andel ubiquity of rural sites in this period indicates that the imperial government Snodgrass 1985, p. 148; tax and Runnels 1987, pp. 102-104,109, encouraged economic development in the region by granting breaks Bintliff" and for intensified cultivation.12 113-117; Snodgrass 1988a; Bintliff 1991; Kardulias, Gregory, and these attempts at historical there Despite important interpretation, Sawmiller 1995, pp. 3-5,16-17. For has not been much discussion of the the boom problems presented by sources for Table 2, see n. 8, above. and-bust pattern of Roman settlement in Greece or of other wrinkles 12. Kosso 2003. THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 749

in the general tapestry of the Late Antique period.13 A few surveys, for have shown no or even a in Late Ro example, great upturn, downturn, man as settlement.14 There is frequent regional variation well: settlement some areas expansion begins in the 2nd to 4th century in (Megalopolis, the Berbati valley, Methana, Oropos, northern Keos), but in the later 4th or 5th century in others (Boiotia, the southern Argolid); the expansion is some dramatic and explosive in regions (Boiotia, the southern Argolid, northern Keos), but gradual and slight in others (Oropos, Methana, the Berbati and Nemea valleys); and the degree of rehabitation of earlier sites, or the presence of earlier material at Late Roman sites, also varies from one region to another. Most critically, however, archaeologists have not often applied the principles of "source criticism" to the analysis and interpreta tion of their survey data.

Source Criticism, Differential Visibility, and the Problem of Pottery Studies

source a In the parlance of survey archaeologists, criticism is way of under standing the biases of survey data by questioning the relationship between amounts the types and of collected artifacts (the sample) and the original a assemblage of artifacts (the total population).15 The process involves closer same we a critical examination of the data, in the way that might question literary source, and it recognizes that contextual analysis and interpretation source of the must precede any attempt to construct a historical narrative or draw conclusions.

Mediterranean archaeologists have long recognized many of the factors that create and distort the composition and appearance of artifact scatters, our an and thereby affect understanding of the evidence. There is active and to vibrant scholarship devoted the ways inwhich varying surface visibility, geomorphological processes, cultural formation processes, taphonomy, plow ing and smearing, manuring, bioturbation, and other rural activities influ ence the of artifacts in surface contexts.16 an inter recognition However, pretive scholarship that deals with the pottery itself, and attempts to assess the visibility, diagnosticity, and representativeness of ceramics within and between chronological periods, is less well developed. Although surveyors

13. Brief but discussions area important survey (Shipley 2002, pp. 268-273, 15. Rutter 1983; Alcock 1993, in reason to appear Alcock 1993, pp. 49-53; 326-336). There is good be pp. 49-53; Mille? 1985,1991a, 1991b, Bintliff and Sanders 2000a, pp. 6-7; lieve, however, that this underrepresen 2000a, 2000b; Caraher, Nakassis, and 2004. to tation may be due the dearth of Pettegrew 2006, pp. 21-26. 14. The for 16. Pylos survey, example, imported fine wares, poor preservation E.g., Wilkinson 1982; Ammer has evidence of of ana man produced consistently rims, and the lack of imported 1985,2004; Bintliff and Snodgrass levels of settlement between the see van high phoras: Lawson 1996, pp. Ill, 122 1988b; Jameson, Runnels, and and Late Hellenistic Roman periods 123;Mee and Cavanagh 2000, p. 106; Andel 1994, pp. 228-246; Alcock, et al. Two For and Davis et al. (Alcock 2005, pp. 179-188). Shipley 2002, p. 270. the Late Cherry, 1994; Zangger a prominent examples of Late Roman Roman downturn in Patras, see Petro 1997; Bintliff, Howard, and Snodgrass downturn are Laconia and Patras. In poulos and Rizakis 1994, p. 201. The 1999; Barker et al. 2000; Fentress 2000; the Laconia Survey the Late Roman cause is unclear, but it may be related to Bintliff 2000b; Terrenato 2000; Pette is the period poorly represented generally, identification problems discussed grew 2001; Van de Velde 2001. except in the southeastern sector of the below. 75o DAVID K. PETTEGREW have recognized that the relative visibility and invisibility of pottery from our different periods might distort picture of the transition between peri ods, there have been, until recently, few efforts to understand the problem or correct for it.17 can We describe the problem of "differential visibility" in the following to scatters on way. Survey archaeologists typically assign dates artifact the a can basis of relatively small group of artifacts that be assigned relatively precise chronological values (e.g., African Red Slip form 50, Late Roman to 2 amphora). I refer these important diagnostic artifacts, which repre a are to sent given period by virtue of the fact that they easy recognize as and therefore frequently identified, "type fossils."18 The number of type a fossils available for use by survey project is dependent first and foremost on our it is the general state of knowledge of Mediterranean pottery, but also tied closely to ceramic studies of particular regions. A period's vis ease can on ibility is determined by the with which it be recognized the a basis of its type fossils. A greater number of recognizable types permits a more confident assignment of chronological value and greater level of visibility for the period in question, while fewer types reduce diagnostic confidence and erode visibility. This iswhy the increasing study of locally wares one most in produced is of the significant developments regional a archaeological survey today: it introduces wider range of type fossils, a to thereby increasing period's visibility.19 Visibility is also tied directly a a project's specialized knowledge: ceramicist who has studied Classical to cooking fabrics, for example, may be able produce higher-resolution use chronological information for that period, while the of fieldwalkers to who have been specifically trained recognize obsidian bladelets may lead a to significant improvement in the visibility of prehistory. Scholarship dealing with the differential visibility of surface finds has had the greatest impact in the interpretation of prehistoric landscapes. In amuch-debated case, John Bintliff, Phil Howard, and Anthony Snodgrass Neolithic and have argued that the poorly fired, friable pottery of the not in and for that Bronze Ages has simply survived well the soil matrix, a or remains reason few potsherds obsidian bladelets might be all that of to a many prehistoric sites. In addition postulating hypothetical vanished also that pottery population reduced by taphonomic processes, they argue fieldwalkers trained to recognize pottery tend to overlook obsidian blades must in the field, and they conclude that the number of sites be calibrated to an accurate of a from low-density scatters in order generate map region's prehistoric settlements.20 a Similarly, there is growing recognition that regional survey projects more may be overlooking Medieval sites because the relevant pottery is

is see the work of Melissa Moore Mori observed the Boeotia 17. The problem acknowledged assemblages by on For recent studies of local in, e.g., Rutter 1983. For discussion son (Moore 2000,2001) Hellenis Survey. to a tic-Late of utilitarian wares from urban see and attempts promote system of Antique groups contexts, wares Clare Slane Slane and Sanders calibration, seeMille? 1985,1991a, from southern Epirus; 2000,2003; 1991b, 2000a, 2000b; Bintliff, Howard, Pickersgill s study (Pickersgill and 2005. and Snodgrass 1999. Roberts 2003) of Roman fine and 20. Bintliff, Howard, and Snodgrass coarse wares and with discussion and debate in 18. Caraher, Nakassis, and Pette from Sparta; Joanita 1999, Barker et al. 2000. See also Bintliff grew 2006, p. 22. Vroom's employment (1998,2003, 19. See Patterson 2000; Vroom 2004) of the concept of "horizontal 2000c; Davis 2004. case for surface 2003,2004. For specific studies, stratigraphy" post-Roman THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 751

case friable and less diagnostic.21 In the of Italy, for example, scholars have concluded that the reduced visibility of the early Middle Ages in the a archaeological record is probably product of weaker material signatures use (caused by, for example, poorly fired pottery, the of nonceramic storage a an containers, and generally lighter material culture) rather than indicator an of absence of population.22 For excavated sites in Britain, John Schofield even a a has argued that handful of Early Medieval potsherds may represent contrast vanished settlement, in to the Roman period, when robust artifact are scatters common.23 In Greece, Guy Sanders has suggested that the practice of glazing vessels during the 12th and 13th centuries has produced a more visible material signature that overshadows the less conspicuous remains from the earlier centuries of the Byzantine period and thus distorts our understanding of population size and distribution.24 source The application of criticism to the study of archaeological survey data from Roman Greece, by contrast, is not especially well established, although there is great need of it due to the significant disparity in visibility between the earlier and later halves of the period.25 While the Early and are Late Roman periods both highly visible due to widely distributed type as fossils, such the rims and bases of well-studied African Red Slip forms, more the Late Roman period is significantly visible because of the greater number of type fossils derived from common utilitarian vessels such as can or amphoras, which be recognized by surface treatment other attributes. a In the field, ridging, combing, and grooving signal diagnostic sherd to the fieldwalker, distinguishing such fragments from plain, undecorated sherds; most one on in surveys the is picked up, the other remains the ground. same During analysis, the surface treatment also bolsters the confidence a of the ceramicist in attaching specifically Late Roman date to the sherd, to a as rather than assigning it less precise chronological grouping, such Roman or Ancient.

In the following pages I examine the data from the Eastern Korin as as thia Archaeological Survey (EKAS), well that from other regional to projects, in order demonstrate the degree to which visibility issues af our fect interpretation of settlement patterns. I argue that the degree of at difference in period visibility the level of collection, typing, and analysis can some cases so in be great that failure to adjust for it would lead one to draw distorted historical conclusions from the data. In the case of the we Corinthia, should interpret the abundance of Late Roman ceramics not as a evidence for sudden explosion of settlement, but as a reflection of a continuing phase of investment in Corinthian territory extending into the Late Roman period.26

21. For a recent overview, see Chris Bowden and Gill 1997b, p. 77;Mee terns of Roman settlement them tie 2004. and Forbes Bintliff selves: see 1997b, p. 39; 2000a, Pettegrew, forthcoming a, 22. Ward-Perkins 324 Mee and in 2000a, pp. pp. 6-7; Cavanagh 2000, and prep. For previous discussions 327. p. 106; Shipley 2002, p. 270; Berlin of Roman-Late Roman settlement 23. Schofield [1989] 2000. andHeath inAlcock et al. 2005, and land use in the Corinthia, see 24. Sanders 2003, pp. 394-395. pp. 194-204. Wiseman 1979, pp. 444-446; Gregory 25. Such have sometimes problems 26. In the present study I focus spe 1985; Engels 1990, p. 24; Romano been noted scholars with the on by dealing cifically Roman and Late Roman 1993,2003; Rothaus 1994; Kardulias, Roman ceramics and their effect on the inter and period: see, e.g., Cherry, Davis, Gregory, Sawmiller 1995; Kardu andMantzourani 1991, p. 331; Lloyd pretation of settlement and land-use lias 2005. 1991, p. 188;Alcock 1993, pp. 49-53; patterns. Elsewhere I address the pat 752 DAVID K. PETTEGREW

ROMAN POTTERY IN THE EASTERN CORINTHIA

was The EKAS project carried out from 1997 to 2003 in the area between Ancient Corinth and Isthmia, directly east of the villages of Hexamilia and Xylokeriza, and transecting the course of the ancient road between Isthmia and Corinth. The survey methods of EKAS were distributional, across surveyors walking transects small tracts (Discovery Units), collect a raw ing from their swaths both count of artifacts (pottery, tile, lithics, a count and other) and total of chronotypes (unique pottery types). As the methods and scope of the survey have been fully published elsewhere,27 to here I wish only emphasize briefly three points about the collection strategy. sets First, the data generated by the survey allow the analyst to quantify count per unit both the total of artifacts of different classes (e.g., pottery, tile, as as lithics), well the total sample of chronotypes (e.g., Late Roman African Red Slip forms 104-106).28 The chronotype system samples the diversity of a artifact types encountered in each survey unit, thereby allowing systematic assessment of the distribution of specific kinds of cultural material across a area. survey Since chronotypes have both functional (e.g., fine ware) and chronological (e.g., Late Roman) values, it follows that chronotype data as as also provide quantifiable information about the functional well the use chronological character of particular units. In the present study I the term "total count" to refer to the sample of chronotypes rather than to the seen course total number of chronotypes in the of the survey29 Second, because the chronotype system discourages the collection an common of duplicate artifacts, it has inherent bias against especially a types of artifacts that appear repeatedly in surveyor's swath. The sample as ware of frequently appearing artifact types, such Combed and Spirally to Grooved body sherds, is likely underrepresent the total number of arti facts seen, while the sample of artifact types that appear in low to moderate such as fine will more the total number amounts, wares, closely approximate encountered. The chronotype system thus allows for the kind of analysis common ar conducted in this study precisely because of its bias against are tifact types. If anything, the "source problems" discussed in here likely to be even more severe than because the indicated, sample underrepresents common the number of diagnostic body sherds and other especially Late Roman artifact types.30

in would the four 27.Tartaronetal.2006. 29. For example, four fieldwalk turn, analyze only ers 28. For full discussion of the walking four different swaths in collected sherds. its a unit each note three 30. For a more detailed discussion chronotype system, including guid survey might its in Late Roman combed of the and the in ing principles, implementation body sherds, potential problems one volved in the the field, the nature of the data it pro but they would collect only per quantification using its its swath. while the total number see Na duces, potential for analysis, and Hence, chronotype system, Caraher, see of combed sherds encountered and 10 drawbacks, Meyer 2003; Meyer and body kassis, Pettegrew 2006, pp. Gregory 2003; Gregory 2004; Caraher, in the unit would be 12, the count of 13; Tartaron et al. 2006, pp. 457 11 this for the unit would be 465. Nakassis, and Pettegrew 2006, pp. chronotype 13; Tartaron et al. 2006, pp. 457-465. only four, and the EKAS ceramicists, THE BUSY OF LATE ROMAN COUNTRYSIDE CORINTH 753

a Finally, permit restriction requiring the analysis of artifacts in the field discouraged the kind of thorough scrutiny and reexamination that is a museum or some typical of analysis in laboratory: kinds of pottery (e.g., were not were various classes of local wares) always precisely typed but instead included within broader chronotype groupings (e.g., Roman fine some ware).31 This may explain why local pottery types discussed in recent studies of Corinthian pottery appear to be absent in the following analysis.32 In spite of these limitations, the EKAS survey produced an abundance of ceramic finds relevant to the questions raised in this study.33

The Busy Countryside

On the surface, the Late Roman period in the eastern Corinthia appears we to have been very busy (Fig. 2).34 If look simply at the number of finds, more regardless of their spatial distribution, there is far pottery from the or Late Roman period than from either the preceding following periods (Table 3). Late Roman pottery forms 4.5% of the total number of analyzed = narrow finds (n 38,337), making it the best represented of the periods in the EKAS data.35 The Early Roman period, by contrast, produced one less than percent of the total number of artifacts recovered through normal Discovery Unit survey, and the total count for the Early Medieval = was a one period (n 17) bare fraction of percent of the total number of finds. The picture presented by the various broad periods that overlap the b.c Roman period is similar: only in the broad Roman period (31 a.d. 700) does the number of finds compare with that of the narrow Late Roman sample. Late Roman material is also found in more Discovery Units than material of any other period (Fig. 2), appearing in 43% of all survey = units (n 1,336); by contrast, Early Roman pottery and Early Medieval

et ceramic data from see 31. Tartaron al. 2006, pp. 446 quantities of imported pottery. That EKAS, Caraher, one most exten 448,466-467; Caraher, Nakassis, and Corinth itself is of the Nakassis, and Pettegrew 2006, pp. 21 12-13. In-field cities in Tartaron et al. Pettegrew 2006, pp. sively excavated Roman the 26; 2006, pp. 463-465; in eastern com a. artifact processing, of course, did Mediterranean also allows Pettegrew, forthcoming mean volve detailed notes, digital photo parison of the Roman pottery recorded 35. By "narrow periods" I and artifact and less in graphs, illustrations, by EKAS with that recovered in the those than 500 years length; can to center. an b.c this data sometimes be used excavations of the urban For they include Early Roman (31 a.d. subtype chronotyped artifacts. overview of the Corinth excavations, 250), Late Roman (a.d. 250-700), see 32. E.g., many of those discussed in Corinth XX; for the Roman pottery, and Early Medieval (a.d. 700-1200). see are Slane and Sanders 2005. Slane 2000,2003; Slane and San "Broad periods" greater than 500 eastern an 33. The Corinthia is ders 2005. years in length; they include Hellenis for b.c-a.d. appropriate region addressing such 34. The following statistics quan tic-Early Roman (323 250), It was well suited for b.c-a.d. Roman questions. agri tify artifacts collected through typical Roman (31 700), culture in b.c-a.d. and consequently inhabited and pedestrian survey methods from Early Medieval (31 1200), farmed and it tensive exclude Roman-Medieval b.c-a.d. throughout antiquity, lay Discovery Units; they (31 1800), within the of the most in b.c-a.d. territory well all artifacts recovered nonsystematic and Roman-Modern (31 2000). connected commercial of Roman not city ways (e.g., grab samples) and from Table 3 does include the broadest which acted as a central other kinds of units that the Roman Greece, trading survey (e.g., exper periods overlap period, hub between east and west. We should imental, extensive). For other discus such as Ancient, Post-Prehistoric, or to therefore expect find significant sions of the Roman-Late Roman Ancient-Medieval. 754 DAVID K. PETTEGREW

Isthmia

Artifacts m /10,000 sq Kenchreai 1-100

101-200

2p1^30Q

300+

llbrnet?rs

Figure 2. Density per hectare of Late Roman pottery in the main EKAS area survey

TABLE 3. ARTIFACTS RECORDED BY EKAS IN NARROW AND BROAD PERIODS

Period Count ofArtifacts % Overall Artifacts Count ofUnits % Overall Units Artifacts per Unit

Narrow Period

Early Roman 331 0.86 193 14.4 1.7 Late Roman 1,707 4.45 577 43.2 3.0 Early Medieval 17 0.04 14 1.0 1.2

Broad Period

Hellenistic-Early Roman 19 0.05 18 1.3 1.1 Roman 2,210 5.76 600 44.9 3.7

Roman-Early Medieval 3 0.01 3 0.2 1.0 Roman-Medieval 108 0.28 45 3.4 2.4 Roman-Modern 7 0.02 5 0.4 1.4 THE OF LATE ROMAN CORINTH BUSY COUNTRYSIDE 755

Isthmia

m Artifacts /10,000 sq Kenchreai 1-100

101-200

^300+

Figure 3. Density per hectare of Roman in the main occur 14.4% and 1.0% of all units Early pottery pottery in, respectively, only (Figs. 3, are on EKAS survey area 4).36Moreover, in units with Late Roman pottery there average 3.0 a as Late Roman sherds per unit, density twice great as that of Early Ro man and Early Medieval material in corresponding units, and comparable to only the broader Roman period (3.7 artifacts per unit). Indeed, much a of the eastern Corinthia is covered by nearly continuous carpet of Late Antique artifacts of fluctuating but high density. an Taken at face value, this pattern would seem to support interpreta or tion of settlement expansion, population explosion, intensive agriculture use and land in the final phase of the Roman period. As the discussion that follows demonstrates, however, we should not take the data set at face value, we to but must instead subject it to closer contextual analysis if wish under use over stand patterns of exchange, ceramic deposition, and land time.

36. 2-6 the for an area of m2 if the Figures display density predict 10,000 3,000 m2), but calculating density by of were the same as that of a more of artifacts each period per unit, artifact density hectare provides larger and calculated hectare the number size was per (i.e., the given unit). Unit typically easily comprehensible value than the one a meter. of artifacts of the period that would much smaller than hectare (ca. 2,000 number of artifacts per square 75^ DAVID K. PETTEGREW

isthmia

m Artifacts /10,000 sq Kenchreai 1-100:

? ^Kilometers

4. Figure Density per hectare of Source Criticism of the EKAS Data Early Medieval pottery in the main EKAS area 4 most common survey Table lists the 12 Late Roman chronotypes in the EKAS ware ware data and shows the dominance of Spirally Grooved and Combed count in the total (RBHS) of Late Roman wares.37 These two chronotypes alone form the majority (62.8%) of Late Roman pottery and a substantial count portion (2.8%) of the total of artifacts analyzed by EKAS.38 use as a The of "spiral grooving" and "combing" basis for Late Roman chronotypes derives from the terminology and chronologies established by Henry Robinson in his study of Roman pottery from the excavations in terms the Athenian Agora.39 The appear frequently in the archaeological are common literature because the surface treatments they denote inGreece s and the Aegean and because Robinson work, which is still the chief author coarse-ware ity for Roman-period chronologies in the region, linked the

37. The RBHS 38. ware abbreviation ("Rims, Wheel-Ridged also forms therefore been grouped by EKAS and a substantial of the with the broader Bases, Handles, Body Sherds") portion overall Roman period rather refers to the total count of counts. the feature than the more artifacts; Although of wheel specific Late Roman RBH ("Rims, Bases, and Handles") ridging is often linked to the Late period. denotes those Roman it not uncommon 39. only diagnostic frag period, is in Agora V, p. 6. the 1st ments, excluding body sherds. and 2nd centuries and has THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 757

TABLE 4. MOST COMMON LATE ROMAN CHRONO TYPES IN EKAS

Total Count ) Chronotype ofLR Chronotypes

Spirally Grooved ware 702 41.1 Combed ware 371 21.7 Amphora, Late Roman 2 108 6.3 Kitchen ware, Late Roman 96 5.6 82 4.8 Amphora, Palestinian Phocaean ware 68 4.0

Medium coarse ware, Late Roman 57 3.3 Phocaean ware 3 46 2.7 Roman 1 23 1.4 Amphora, Late Late Roman 22 1.3 Amphora, Red Slip, Late Roman 18 1.1 Amphora, Aegean Red 1 16 0.9

Total 1,609 94.2

surface treatments to the Late Roman period. In Athens, Robinson dated to spiral grooving the 4th through 6th centuries A.D., especially the 5th a.d. and 6th centuries, and combing to the 6th and 7th centuries Recent on commerce work Roman has shown the frequency of spiral grooving and on eastern combing amphoras and transport vessels of the Mediterranean occurs on generally, although such surface treatment also other shapes and as as forms, well in other periods.40 or In the Late Roman period generally, body sherds with combed grooved surface treatments derive from Late Roman 2 amphoras and from as as Palestinian and Gaza-type amphoras, well from other Late Roman utilitarian vessels, both open (basins) and closed (pitchers and jugs), of types to a.d. most sources dating from the 4th the early 7th century The likely are for such sherds among the EKAS material specifically Late Roman 2 amphoras,41 Palestinian and other Late Roman amphoras, open and closed medium coarse-ware vessels, and basins, all of which were identified with some frequency from rims, bases, and handles (Table 5).42 Other surveys have linked combed and grooved decoration to later Byzantine amphora types, but rims and handles from Byzantine amphoras are scant in our

of am ware are a 40. Common Late Roman (LR) be confused with those Gaza characterized by specific and the handles with Nieder fabric and surface and con types include, e.g., Carthage and phoras treatment, = Benghazi LR Amphora type 2, and bieber 77 EKAS chronotype Aegean sist almost exclusively of body sherds. Red The of bases and Palestinian and Gaza-type amphoras. Amphora (K. Slane, pers. comm.). recovery rims, toes, The surface treatment alone and as well as closer Grooving, ridging, and combing also obviously handles, analysis occur on does not indicate a Late of the fabric and surface treatment of vessels of later periods, specifically coarse but must be considered can allow to a especially Byzantine and plain Antique date, body sherds, assignment wares: at on in more as e.g., Sara?hane, Late conjunction with clay, color, fabric, specific chronotype, such Late as form. 2 The 1 Roman-Byzantine amphoras, well and Roman amphora. Late Roman as 61 41. is not a source other vessels (Hayes 1992, pp. These amphoras have horizontal amphora likely for in on on their and were or 79); and Sparta, Byzantine plain grooving shoulders spirally grooved combed sherds, in the account for wares, cooking wares, and amphoras probably produced Aegean. but may other ridged body in (Sanders 1993, pp. 268-283). In Cor Cf. Munn 1985; Karagiorgou 2001. sherds found frequently the survey occurs on of 42. There is some in the area and dated to the broad Roman inth, grooving amphoras overlap Frankish date with han in 4 high-arched chronotypes listed Tables and 5. period. dles; the body sherds of this type could Spirally Grooved ware and Combed 758 DAVID K. PETTEGREW

TABLE 5. MOST COMMON LATE ROMAN AMPHORA AND COARSE-WARE TYPES IN EKAS (RBH ONLY)

Chronotype RBH Count

Amphora, Late Roman 2 107 Medium coarse ware, Late Roman 43

Amphora, Palestinian 22 1 Amphora, Late Roman 22 Amphora, Aegean Red 1 14 Late Roman Amphora, 13 Basin, Late Roman 11

Total 232

are an source survey area and consequently unlikely for many of the combed and grooved sherds. coarse wares In the EKAS data, amphora sherds and medium with surface grooving and combing significantly alter the proportional makeup of Late Roman functional classes (Table 6), forming 83% of analyzed Late wares. treat Roman By contrast, Early Roman amphoras lack these surface are ments, and the functional classes of the Early Roman period found in more wares coarse wares equal proportions: fine 38.3%, 36%, and kitchen are wares 24.8%. Although the Early and Late Roman periods represented vs. vs. by similar numbers of fine-ware (127 165) and kitchen-ware (82 96) sherds, Late Roman coarse-ware sherds grossly outnumber those from vs. eastern the Early Roman period (1,417 119). In the Corinthia, then, are more a utilitarian vessel fragments important indicators of Late Roman an presence than of Early Roman presence, while for the Early Roman wares wares are more period, fine and kitchen proportionally important period signatures. to The explanation for such proportional differences appears be meth a our to odological, product of differing abilities recognize amphoras and coarse wares two Late Roman coarse-ware sherds from the periods. body are recognizable by both fieldwalkers and ceramicists because of their dis are tinct surface treatment, while Early Roman body sherds not. Indeed, one as coarse ware or only of the 119 fragments identified Early Roman a = amphora in the EKAS data is body sherd, while 83.5% (n 1,183) of are Late Roman coarse-ware and amphora fragments body sherds. While wares as the Early Roman presence is known almost entirely from fine such

TABLE 6. ROMAN ARTIFACTS IN EKAS BY FUNCTIONAL CLASS

Functional Class ER Count %ER Pottery LR Count %LR Pottery Roman Count % Roman Pottery

Coarse ware and 119 36.0 83.0 59.1 amphora 1,417 1,305 Fine ware 127 38.3 165 9.7 438 19.8 Kitchen ware 82 24.8 96 5.6 294 13.3 0.9 6 0.4 34 1.5 Lamp 3 Other 23 1.3 139 6.3

Total 331 100.0 1,707 100.0 2,210 100.0

= = ER Early Roman; LR Late Roman THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 759

to a Eastern Sigillata, the rims and handles of amphoras, and, lesser extent, kitchen-ware sherds, the Late Roman period has the advantage of highly recognizable medium-coarse body sherds derived from utilitarian vessels. occur Since utilitarian vessels much more frequently in the countryside than do fine wares, and since body sherds greatly outnumber rims, bases, and no more handles, it is surprise that the Late Roman period is far visible in the field than the Early Roman period.43 on Although Late Roman material may appear thick the ground in the eastern Corinthia relative to that of the preceding and following periods, statistical analysis suggests that the discrepancy is caused mainly by the significant differential visibility of the periods. The Late Ro man a "explosion" in the EKAS data is product of the ubiquity of easily identifiable body sherds, while the visibility of the Early Roman period on depends pottery less often encountered in surface survey: fine wares, rims, bases, and handles. Scholars have noted the problem of differential enormous visibility before, but the figures presented above indicate the degree of difference in visibility. In the discussion that follows I show how the problem of differential visibility has affected other survey projects in Greece and the eastern Mediterranean, and suggest ways to correct and calibrate for it.

ASSESSING OTHER LATE ANTIQUE COUNTRYSIDES

Many of the intensive archaeological surveys carried out within the last a manner generation have published their finds in complete enough to permit critical analysis of the data and evaluation of the changes between not periods. Other projects, while yet completely published, have presented one some enough of the data to allow to form impression of the evidence for various periods. Although any critical review of the data from these projects will be incomplete, the information available indicates that the common problem of differential ceramic visibility for the Roman period is to Greek surveys generally. In the following discussion, I reexamine the ceramic data from several survey projects that have presented their results in a numerical form that can to same be subjected the kind of analysis applied above to the EKAS to measure to data. In doing so, I attempt the degree which regional surveys generally have been affected by the bias of highly diagnostic pottery. Since each of these surveys sampled the original population of artifacts in a dif ferent way, I also highlight the ways inwhich archaeological and historical are interpretations influenced by these sampling regimes.

43. Surface treatments such as would more presumably become still and do not combing grooving always visible if precise identification of the extend over the whole of an were body undecorated sherds possible. This or but are sometimes not amphora vessel, conclusion does greatly undermine restricted to shoulders and necks. It the argument presented here, however, even coarse wares follows that though the Late since Late Roman are still Roman is more period already highly visible, substantially visible than those of it is nevertheless underrepresented and the Early Roman period. DAVID K. PETTEGREW 760

TABLE 7. ROMAN POTTERY COUNTS IN THE METHANA SURVEY, EKAS, AND NORTHERN KEOS SURVEY

EKAS Keos MethanaSurvey Northern Survey Period Count % of Total Roman Count % of Total Roman Count % of Total Roman

Early Roman 315 28.2 331 16.2 32 13.9 Late Roman 801 71.8 1,707 83.8 199 86.1

Total 1,116 100.0 2,038 100.0 231 100.0

The Methana Survey

a The Methana Survey has provided full publication of its finds, completely and systematically annotated.44 The survey divided the Roman period into b.c-a.d. three subperiods: Early Roman (100 100), Middle Roman (a.d. 100-300), and Late Roman (a.d. 300-700). For the sake of comparison with the EKAS data, however, I have combined the Early Roman and a b.c-a.d. Middle Roman periods into broader Early Roman period (100 300).45 The artifact-sampling strategy for theMethana Survey favored the collection of feature sherds.46

A comparison of the figures for Early and Late Roman pottery in both to EKAS and theMethana Survey (Table 7) shows the degree which Late Roman pottery dominates in both bodies of data.47 In theMethana Survey, a Early Roman represents larger proportion of the total Roman count than a are it does in EKAS, but this is only matter of degree: the proportions roughly comparable in both surveys. Most interesting is the degree of correspondence between the RBHS two proportions for the periods (Tables 8-10). At Methana, ordinary body wares com sherds constitute 71.4% of that signal the Late Roman period, a pared with 71.2% for EKAS. In both surveys body sherds constitute lower at as percentage of the total sample of Early Roman pottery; Methana, in a more EKAS, rims, bases, and handles play important role in signaling

EKAS b.c. 44. Gill, Mee, andTaylor 1997. The defined by (1st century example of every unique chronotype) are on a 7th The term vs. selective Both figures presented here based through century a.D.). grab sampling. as Roman" refers to the EKAS influence historical count of the pottery published in the "Early sampling systems between the 1st b.c. in similar catalogue of finds. The analysis does period century interpretation ways, although not to and the 3rd and "Late the include pottery assigned bridg century a.D., chronotype system underrepre Roman" refers to the EKAS sents common ing periods (e.g., Hellenistic-Early period especially chronotypes, from the middle of the 3rd such as sherds. Roman, Late Classical-Early Roman, century spirally grooved body the Because the constitute Middle Roman-Late Roman), but it through 7th century. Hence, body sherds probably was tenta a of the Late Roman does include pottery that "Middle Roman" period of the larger percentage Methana falls before a.d. of the Corinthia than the tively dated (e.g., "possiblyLR"). Survey 300, pottery figure can 45. Bowden and Gill 1997a, pp. 84 it be subsumed under the slightly given inTable 9 (71.2%) suggests. Roman" defined with the numbers for 90; 1997b, p. 77. The period denoted by broader "Early period When compared term "Roman" in the Me EKAS b.c-a.d. the between the simple by (31 250). Methana, greater disparity to 46. Mee and Forbes 35. the numbers of and Late Roman thana Survey corresponds generally 1997b, p. Early 47. The data sets for in data results the "EarlyRoman" period in EKAS. To chronological sherds the EKAS likely EKAS and Methana were from the of Late avoid confusion, throughout the follow Survey pro greater frequency use term duced two different Roman in the EKAS ing discussion I the "Roman" by sampling sys amphoras to to as tems: the refer the entire Roman period chronotype system (a single territory. THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 761

TABLE 8. EARLY ROMAN POTTERY (BY EXTANT FRAGMENT) IN THE METHANA SURVEY AND EKAS

Methana Survey EKAS Portion Count % Total Count Total%

Rim76 24.1 49 14.8 Base51 16.2 33 10.0 Handle 74 23.5 111 33.5 Body sherd 114 36.2 136 41.1 Other 0 0.02 0.6

Total 315 100.0 331 100.0

TABLE 9. LATE ROMAN POTTERY (BY EXTANT FRAGMENT) IN THE METHANA SURVEY AND EKAS

Methana Survey EKAS Portion Count % Total Count % Total

Rim130 16.2 322 18.9 Base24 3.0 37 2.2 Handle 75 9.4 110 6.4 Body sherd 572 71.4 1,215 71.2 Other0 0.0 23 1.3

Total801 100.0 1,707 100.0

TABLE 10. ROMAN POTTERY IN THE METHANA SURVEY AND EKAS (RBH AND BODY SHERDS)

RBH % Body Sherd % Total %

Early Roman/Methana 201 63.8 114 36.2 315 100.0 Late Roman/Methana 229 28.6 572 71.4 801 100.0 Early Roman/EKAS 195 58.9 136 41.1 331 100.0 Late Roman/EKAS 492 28.8 1,215 71.2 1,707 100.0

the Early Roman period than they do the Late Roman period. The very different RBH:S ratios in the Early and Late Roman periods are imme diately apparent in both projects (Table 10). The explanation for the prominent Late Roman presence in the as seems to Methana Survey, in EKAS, be the high visibility of Late Ro man and a combed grooved body sherds, which constitute large percentage = of the total count of Late Roman pottery: 43.1% (n 345), and 12.4% 48. Bowden and Gill 87 = 1997a, pp. (n 99), respectively.48 Removing such body sherds from the Late Roman 88. The 345 Combed ware sherds listed mix count more deflates the total of Late Roman artifacts inMethana by in the artifact are said to catalogue than 50%. These sherds also have a tremendous effect on the number of represent or other closed amphoras sites identified the as an are to by survey. all sherds forms, and linked by the authors Dismissing body identify Berenice Late Roman 1 and Late ing category, regardless of fabric or surface treatment, would eliminate Roman 2 26% of the Late Roman sites to amphoras. (from 58 43), but only 5.5% of the Early 762 DAVID K. PETTEGREW

Roman sites (from 36 to 34).49With this calibration the number of Late a Roman sites, instead of representing 61% increase from the Early Ro man a period, indicates much gentler increase of 26.5%. Such is the ef on fect of diagnostic Late Roman body sherds Methana artifact and site populations.50

The Northern Keos Survey

The Northern Keos Survey employed grab samples of potentially diagnostic artifacts to characterize the chronological character of their survey area.51 The investigators recorded 31 sites with some kind of Roman pottery, either Early Roman (lst-3rd century a.D.), Late Roman (4th-early 7th or century a.D.), Roman (lst-early 7th century a.d.).52 Nine of these 31 sites could be dated specifically (but not exclusively) to the Early Roman a period and 26 had specifically Late Roman phase; hence, Late Roman a sites outnumbered Early Roman by factor of 2.9. The total count of Late was Roman pottery approximately six times that of Early Roman pottery (Table 7). ware at at Fine appears two-thirds of all sites which either Early Ro man or Late Roman is represented. The relative proportion of body sherds at sites of each period differs, however: Early Roman body sherds were no was reported at only four of nine Early Roman sites, and site dated to on the Early Roman period the basis of body sherds alone.53 By contrast, or 22 of 26 Late Roman sites (84.6%) produced combed, grooved, ridged a body sherds datable to the Late Roman period. At approximately third = of the sites (n 8), the Late Roman component appears to have been on identified only the basis of body sherds, usually with combed, grooved, or case ridged surface treatment. As in the of the Methana Survey and EKAS, eliminating body sherds from the counts diminishes the number to of Late Roman sites significantly (from 26 18), and reduces the increase in site numbers between the Early and Late Roman periods from nearly 300% to 200%. This still represents an increase, but a substantially smaller one than before.

49. The Late Roman sites that zourani 1991, p. 481, for chronology, would lose "site" status are MS4, MS11, and pp. 327-347, for a discussion of MS12, MS15, MS55B, MS102, ceramic deposition in the countryside MS104, MS108, MS109, MSI 13, in the Greek and Roman periods. The are MS116, MS124, MS205, MS214, and figures presented here derived from MS220. Ten of these sites yielded only the counts of artifacts listed in Sutton or et al. 1991. include sites where combed grooved Late Roman body They fewer than three artifacts of a sherds. The Early Roman sites that given would be eliminated areMS60 and date were found, but do not include MS213. off-site finds. As in the case of the 50. This differential diagnosticity is, Methana data, I have also omitted pot in fact, highlighted by the investigators tery dated to broader bridging periods as a reason for caution: see Bowden and such as "Classical-Late Roman."

Gill 1997a, p. 84. 53. Early Roman body sherds et 28-30. included fine-ware one 51. Cherry al. 1991, pp. mainly sigillata; sherd was noted. 52. See Cherry, Davis, and Mant ridged body THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 763

The Oropos Survey Project

In the Oropos Survey Project field teams collected only diagnostic artifacts, dividing the Roman period between Early Roman (lst-2nd century A.D.), Middle Roman (3rd-4th century A.D.), and Late Roman (5th-7th century were or a.d.).54 There 30 "certain" "possible" findspots that could be dated some to part of the Roman period and 5 "tentative" findspots. Of these 30 an a likely sites, 9 had Early Roman phase, 14 had Middle Roman phase, a and 21 had Late Roman phase. were on The nine Early Roman findspots identified almost entirely one the basis of rims, bases, and handles; only site yielded plain Early were Roman body sherds that considered diagnostic.55 By contrast, the were on Middle and Late Roman periods identified mainly the basis of treatments. body sherds and surface Although approximately half of the Middle Roman findspots yielded rims, bases, and handles, wheel-ridged were (mainly body) sherds the predominant Middle Roman artifact both in overall quantity and in frequency at sites. Similarly, although about half of the 21 sites produced Late Roman rims, bases, and handles, combed to a.d. were body sherds dating the 5th-7th century the predominant arti fact type found, and the basis for a confident attribution of Late Roman we remove as an date. Again, if body sherds identifying period index, the numbers of Early, Middle, and Late Roman sites change from 9:14:21 to are 8:7:10. Both the Middle Roman and Late Roman upturns severely a deflated by this calibration, and the result is very different picture of the settlement pattern of the Roman period.

The Pyla-Koutsopetria Archaeological Project

a The Pyla-Koutsopetria Archaeological Project is recently completed, a gridded site survey of 40-hectare harbor town outside Larnaca in south eastern one a recent Cyprus.56 It is of series of projects in Cyprus that have produced well-documented Late Roman assemblages of sizable rural Late a Roman sites.57 Using the chronotype system, the project sampled total of about 8,500 pieces of pottery from survey units with predominantly are Late Roman activity.58 The largest group of Late Roman chronotypes tiles (44% of all Late Roman material), which can be dated on the basis of good stratigraphie excavations at the site and other Late Roman sites in are Cyprus. If tiles excluded, amphoras, coarse wares, and medium coarse

54. Cosmopoulos 2001, pp. 60-64, 56. Caraher et al. 2005; forthcom man date accounts for over 40% of the 84-122. On the strat I thank and fellow entire over artifact-sampling ing. my colleagues analyzed assemblage and see egy, Cosmopoulos 2001, pp. 26-31. codirectors of the project, R. S.Moore 80% of the total number of artifacts Because the Middle Roman and William for that can be dated to a narrow period Caraher, encouraging period at both the and me to discuss the ceramic data. one Oropos overlaps Early (i.e., lasting fewer than 500 years); the Late Roman in 57. Other include the sites most of the can periods EKAS, projects remaining pottery only it cannot be subsumed within either of Maroni Petrera et al. to (Manning be dated broadly theAncient His but must be discussed and toric b.c-a.d. period separately. 2002) Kopetra (Rautman 2003), period (750 749). Other 55. find and the Cosmopoulos 2001, p. 113, Sydney Cyprus Survey Project specific chronological periods, includ 91/22. One other site was dated on are spot (Given and Knapp 2003). ing Early Roman, represented in the basis of an Roman Early lamp. 58. Pottery of specifically Late Ro very small amounts. 764 DAVID K. PETTEGREW wares constitute the majority of the Late Roman material (85%), while fine wares or wares (11%) and kitchen cooking (4%) make up the remainder. Moreover, as in the case of the EKAS data, the majority of medium coarse are on and amphora sherds body sherds (68%), identified the basis of spiral grooving and combing; rims (6%), bases (2%), and handles (24%) together a count for only third of the total sherds of this class. By contrast, body a sherds represent minority (31%) of the total count of fine wares dated are to the Late Roman period, while rims (54%) predominant. Although was at the Early Roman period poorly represented the site, the majority are ware to coarse of sherds (61%) fine and only 12% belong the class of ware and amphoras.

Other Surveys

The examples discussed above could be multiplied by the addition of other extensive and intensive surveys in Greece, although rarely have the finds been recorded in sufficient detail to allow statistical analysis of wares, so are more functional classes, and parts of vessels, the results necessarily a impressionistic. The Boeotia Survey, for example, employed collection a strategy that selected potentially diagnostic artifacts.59 In sample of 454 to Late Antique Early Byzantine sherds from 30 sites in the region, Joa wares nita Vroom noted that fine constituted only 6% of this material; the overwhelming majority consisted of amphoras (29%), especially Late Ro man was 2, and beehive fragments (62% of all finds).60 The Boeotia Survey so fortunate to recognize many Late Antique sites, since almost all of the even were finds (perhaps the Late Roman 2 amphoras) locally produced. were a Indeed, if it not for the identification of single type of pottery, the Late Roman beehive fragments, the remains of this period would appear on much thinner the ground. This conclusion is consistent with the results of many other topo in graphic and intensive surveys in Greece and the Mediterranean which or a Late Roman component has been identified only mainly on the basis or treatment on of combed, ridged, grooved surface body sherds. The an Argolid Project, for example, divided the Roman period into Early (50 a b.c-a.d. 200), aMiddle (a.d. 200-400), and Late phase (a.d. 400-650), but the catalogue of sites suggests that Early and Middle Roman pottery wares occur was rarely identified with much confidence, while Late Roman numerous coarse as frequently, with amphora sherds and domestic wares, as well red-slipped fine wares, and the occasional coin, lamp, roof tile, and a an cooking vessel.61 Whether the apparent disparity is product of actual or a difference in ceramic abundance merely difference in ceramic visibility can a only be determined by closer examination of the data. treat 59. Vroom 308. Altogether, these analyses indicate that the distinctive surface 2004, p. a 60. Vroom 2004, pp. 308-324, esp. ments of some types of Late Roman pottery provide higher degree of table 2B. turn to in and con p. 311, diagnosticity, which in leads greater confidence dating 61. Jameson, Runnels, and van tributes to the in the field. the period's higher visibility Indeed, tendency Andel 1994, pp. 415-538. of to sherds reinforces survey projects sample only potentially diagnostic 62. Cf. Caraher, Nakassis, and corrects A 2006. rather than for these biases toward easily recognized types.62 Pettegrew THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 765

sampling method that favors body sherds with distinctive surface treatments and decoration is likely to exaggerate the differences between periods, while as an removing body sherds identifying class significantly diminishes the some not evidence for the Late Roman period in regions. This is to argue a that the pattern of settlement expansion in the Late Roman period is a product of survey method alone, but rather that differential visibility is reason in primary for the relative abundance of Late Antique material both artifact and site catalogues.

QUANTITATIVE COMPARISONS WITH EXCAVATED SITES

Before discussing the implications of this rereading of the ceramic evidence to for the interpretation of Roman settlement patterns, it will be useful compare typical ceramic surface assemblages for the Early and Late Roman periods with corresponding ceramic assemblages from excavated contexts.63 a con The introduction of quantitative data from range of excavated Roman texts provides interesting points of comparison between material identified on the basis of stratigraphie context and deposits (excavation assemblages) on and material typically identified the basis of recognized and sampled a a type fossils (survey assemblages). Such comparison demonstrates that sum typical Early Roman surface assemblage (i.e., the of identified type an fossils ofthat date) differs significantly from expected total population of Early Roman artifacts, while typical Late Roman surface assemblages, are coarse more which characterized by larger percentages of wares, closely resemble recovered from excavation. assemblages an The quantification of pottery has been important component of excavation and survey projects for three decades, especially in the western assem Mediterranean.64 Although the practice of quantifying excavated a more blages is recent development in the eastern Mediterranean, it is an common becoming increasingly way of presenting Roman and Medi eval ceramic finds.65 The counting of artifacts, of course, has long played a r?le in the presentation of survey data, but the quantification of the types of artifacts found in intensive survey and the interpretation of sur are more vey data and settlement history in light of such counts recent

developments.66

63.1 appreciate Kathleen S lane's include Sanders 1987,2003; Papado on insightful critique and suggestions poulos 1991; Slane 2000,2003. For recent of the following discussion. examples the practice applied to Roman and Medieval see 64. On the value of quantification, pottery, see Rautman et al. Riley 1976, pp. 125-131; 1979, 2000,2003; Manning pp. 97-111; Slane 2003. Important 2002; Gerstel et al. 2003. western 66. See and Perkins early studies in the Mediter Fentress 1989; ranean includeHayes 1976; Riley Poulter 1998, pp. 464-475,503-511; Rautman et 1976,1979; Fulford and Peacock 1984, 2000,2003; Manning al. pp. 253-262,273-275. 2002; Vroom 2004; Caraher, Nakassis, 65. Quantified studies inGreece and Pettegrew 2006. 766 DAVID K. PETTEGREW

While these studies indicate that excavated assemblages vary greatly in the relative proportions of functional classes, other similarities between surface and excavation assemblages assemblages encourage comparison.67 con In the following discussion I consider three examples of excavated texts of Roman date for which the pottery has been completely collected, an an recorded, counted, and catalogued: industrial complex in Corinth, a urban house in Carthage, and rural Early Christian basilica in Cyprus. a My aim is not to establish standard of proportions, since the amount and at proportions of pottery of each functional class vary each site, depending largely upon the nature of the context. Rather, my goal is to demonstrate to the degree which typical Early and Late Roman surface assemblages dif fer from the range of corresponding excavated assemblages in terms of the criteria discussed above.68

Corinth: Industrial Buildings East of the Theater

For comparison with the data from EKAS, the most immediately relevant s recov work is Kathleen Slane quantitative study of the ceramic material ered from four buildings, probably used for industrial purposes, excavated in the 1980s east of the theater at Corinth.69 The excavations generated nearly 12 tons of Roman pottery, of which Slane has studied 127,370 pieces. The material dates principally between the 1st and 4th centuries, with the greatest amount belonging to the Early Roman period (late 2nd-early a at 3rd century);70 there is break in the depositional sequence the begin ning of the Late Roman period (early 4th to mid-5th century), after which the ceramic sequence continues to the 7th century. Slane's study tabulates the relative percentages of functional classes over time, with an eye toward delineating shifts in imports and local production. Amphoras constitute 47% of the overall pottery by count, with the highest percentages in the 1st and 2nd centuries and in the 5th century. Fine wares show the same to general pattern, with the highest percentages in the late 1st early 2nd a century (10%-12%) and again in the 5th century (12%-14%), and low wares point in the later 2nd and 3rd centuries (5%-7%). Cooking and plain

on the 67. The following discussion 68.1 have tabulated functional the differences between Early assumes similarities coarse and Late Roman I that between the classes (e.g., fine wares, wares, periods, attempt, and kitchen and extant as far as to relate these ex composition of excavated deposits wares) parts possible, are to on count rather than cavated to the same surface assemblages be expected. (RBHS) based assemblages count com There may, of course, be differences weight, since facilitates periods. found in town 69. Slane comm. between assemblages parison with counted artifacts from 2000,2003, pers. is no reason The first of these studies is based main and countryside, but there regional surveys. Since the classes coarse on a of material to think that such differences would defined by EKAS (fineware, ly the fine wares, body a In both coarse that is more sensitive to and make comparison impossible. ware, medium ware, kitchen imports, are not to does not calculate for and urban and rural contexts, for example, ware) exactly equivalent amphoras we ratios of excavation coarse wares. The discussion should expect that the rims, those used by the projects, present to will on two on for the bases, and handles body sherds I base my conclusions only relies especially Slane 2003; not An criteria: the ratio of fine wares see 333 and differ greatly. ideal comparison broad percentages by RBHS, p. to of excavated and surface assemblages other wares, and the ratio of RBH figs. 19.11,19.12. contexts same to the 70. K. Slane would compare from the body sherds. Because preceding (pers. comm.). is not region, but this always possible. discussion of the survey data focused COUNTRYSIDE OF LATE ROMAN CORINTH THE BUSY 767

the most: constitute on a wares, by contrast, vary together they average little more than 40% of the pottery, but in the 3rd century that number increases to as much as 60%, while the numbers of amphoras and fine wares decline.71

Carthage: Roman Peristyle House

a In domestic deposits from Late Roman peristyle house excavated by the some University of Michigan in 1975, amphoras constituted 50%-60% of finds dated to the 1st century a.D., aswell as those dated to the 5th and 6th cen amounts turies, with significantly lower in the 2nd and 3rd centuries. ware African Red Slip comprised 8%-10% of the Roman-period ceramic material after the 2nd century72 strat The published counts and weights of pottery types in each of 13 igraphie units show differing relative frequencies for the parts of vessels.73 Body sherds typically make up 80%-95% of each deposit, rims 4%-10%, wares bases l%-4%, and handles l%-3%. For fine specifically, rims (20% a 40%) and bases (generally, 9%-25%) constitute greater proportion of the a as total count, and body sherds much lower percentage (as low 45%, but for most groups, 50%-65%).74 Handles, whether fine or coarse, form a consistently low percentage of the overall assemblage.

Maroni Petrera, Cyprus: Early Christian Basilica

an Sturt Manning and his colleagues investigated Early Christian basilica excava atMaroni Petrera in Cyprus between 1990 and 1997.75 Limited tions there produced 4,202 potsherds (85.5 kg), although most of these were two were found in the plow zone; only closed deposits revealed, the to n = first dating the Early Roman period (early 2nd century a.D.; 559), the second to the Late Roman period (late 4th to early 5th century a.D.; = n 128). are There considerable differences between the two deposits, but in both, amphoras have the highest proportional representation of the ceramic groups: 33% of the Early Roman, 66% of the Late Roman. Fine wares constitute a = as small percentage of the total count: 6.3% (n 35) of the Early Roman = wares are semblage and 12.5% (n 16) of the Late Roman. Cooking much 71. ware more n = are On average, cooking makes important in the Early Roman group (17%; 95) than they in the 17% of the overall and n = = up assemblage, Late Roman (0.8%; 1). Pithoi constitute 5% (n 28) of the Early Roman ware 25%. do not fluctuate = plain Lamps count and 11% (n 14) of the Late Roman. when appear, as above 2%-3% across the entire Lamps, they period. are they do in the Early Roman deposit, proportionally insignificant. Regard 72. Hayes 1976, pp. 84,114. are less of functional class, sherds make the bulk of the 73. The following percentages body up pottery by count based on the figures presented in Riley (Early Roman 87.3%; Late Roman 79.7%); of the remaining RBH tables 1-15. are more common 1976, pp. 132-156, fragments, rims than bases and handles. In both deposits 74. Riley 1976, tables 3a, 5a, 5b, 7a, are the amphora fragments predominantly body sherds, which form 94.1% 8a, 11a, lib, 12a, 13a, 13b. and 89.4%, of the total counts for and Late Roman am et 2002. respectively, Early 75. Manning al. For the see a fine wares, however, sherds make 57% of the pottery, pp. 41-57, with break phoras. Among body up only on total in both the and Late Roman RBH down of counts pp. 44-47, tables count; Early periods, fragments a 6.1,6.2. constitute greater proportion of fine wares than of coarse wares. 768 DAVID K. PETTEGREW

Comparison of Excavated and Surface Assemblages

Any excavated assemblage is bound to reflect the context from which the we pottery derived, and should consequently expect great variety in the one functional and morphological makeup of assemblages from site to are another. Moreover, the relative proportions of RBHS closely related to functional classes: amphoras, cooking wares, plain wares, and fine wares fragment differently, producing different sherd sizes and different numbers of rims, bases, handles, and body sherds. Consequently, these proportions will also vary, reflecting the functional character of the assemblage and the context that produced it.76 In the excavated assemblages discussed above, two are however, consistent patterns stand out, which comparable to the seen patterns in the data from EKAS and other regional survey projects. First, at none of the three sites do fine wares account for more than 15% of the overall ceramic population, and they typically range between 6% case and 10%. This remains the in spite of shifts in the relative proportions of functional classes between periods. Although there exist archaeological wares a contexts in which fine do represent far greater proportion of the a total Roman ceramic population,77 the data from broad range of contexts ware are indicate that proportions of fine usually less than 20%, and often much lower.78

Second, the evidence for the relative proportions of RBHS at Carthage andMaroni Petrera suggests that despite significant variation, body sherds constitute the great majority of pottery (80%-95%) counted in both Early a and Late Roman deposits, RBH forming consistent minority (5%-20%). more wares a The proportionally significant fine become in deposit, the more RBH proportions approach those of body sherds, but at most sites most and in assemblages, body sherds form the vast majority of finds.79

76. In the case studies discussed 78. other can be of Two examples steady minority the finds, usually to case 6%?10% of the occasion above, for example, amphoras ranged added the studies presented assemblage, from 33% to 66% or more of the ce above. Excavations conducted in the ally lower or higher (Fulford and wares ramic population, cooking 1% 1960s by Dumbarton Oaks and the Peacock 1984, pp. 253-254,273-275). and wares at 79. I two 30%, plain 20%-50%; Archaeological Museum of Istanbul Again, offer additional local vs. varied at imported proportions the church of St. Polyeuktos Sara examples. Excavations between 1976 between sites. sherds formed a in some 1978 in the and isthmus Body ?hane Istanbul generated and lower city of the to of Torone of much lower proportion (ca. 50%) 350,000 400,000 sherds dating from produced 5,241 pieces to to overall population of fine-ware deposits the Late Roman Early Modern pottery attributed six types of Late than did for ware and am In Roman Of these they plain periods (a.d. 400-1900). both Late amphora. sherds, a of the = were phoras, product relatively Antique and Byzantine deposits, fine 87.3% (n 4,577) body fragments, size fine-ware wares = were smaller of the original form about 10% of all finds; only while only 12.7% (n 664) RBH. vessels. in later as these numbers are based Byzantine periods, amphoras Moreover, on wares on that 77. E.g., Limyra the Lycian become less important, do fine only sherds could be assigned where fine wares make some wares come to a to a there was a coast, up and kitchen represent specific amphora type; 29% of the The Late Roman material. greater proportion of the total (Hayes larger group of 5,598 body sherds that a at 3 high percentage may be product of 1992, pp. 3,53). Excavations Car probably represented Late Roman not be the location of the excavation, adjacent thage by the British Academy between amphoras but could designated to to these monumental buildings and shops, 1975 and 1978, mainly of Late Antique that class with certainty, and or of the role as an see were excluded from the If we city's episcopal and Byzantine domestic buildings, pro analysis. and site. Cf. Vroom ceramic were were to add these sherds to the pilgrimage 2004, duced data that quantified group, ware a 93.9% of pp. 291-294,307-308. by weight. Fine constituted body sherds would represent THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 769

two can These patterns be compared with the assemblages of Roman, Early Roman, and Late Roman date produced by regional survey projects. The EKAS data for the Late Roman period (Table 6) yield proportions are for the functional classes that closest to those of assemblages produced excavation: an of coarse by overwhelming predominance wares, ampho ras, and kitchen wares, with much smaller amounts of fine wares.80 In on the Early Roman and the broad Roman periods, the other hand, fine wares are proportionally overrepresented (38.3% and 19.8%, respectively), coarse wares are a while the percentages of and amphoras much lower, fact coarse wares were as that suggests that many Roman and Early Roman to as or signed broader chronological groupings such Roman-Medieval are Ancient. In other surveys, these the kinds of pottery that might be ignored altogether. nor In neither EKAS theMethana Survey do the RBHS ratios for the an Early and Late Roman periods (Tables 8-10) correspond well to that of excavated assemblage with roughly 80%-90% body sherds and 10%-20% more RBH, but in both surveys the Late Roman percentages closely ap proximate the expected proportions. In the Late Roman period, body sherds constitute 71.4% of the total Late Roman count atMethana and 71.2%

in EKAS, whereas the percentages for Early Roman body sherds in both at are we surveys (36.2% Methana; 41.1% in EKAS) much lower than might on our expect based knowledge of excavated assemblages. The Early and Late Roman surface samples, in other words, differ significantly in their to same similarity fully excavated assemblages of the periods.

DISCUSSION: UNDERSTANDING ROMAN SURFACE ASSEMBLAGES

Throughout the preceding analysis I have argued that the Early Roman and Late Roman periods have significantly different degrees of visibility in two are typical regional surveys, and that the periods therefore very unevenly most represented in archaeological surface samples. The type fossils of the are wares earlier period generally fine and RBH sherds that in excavated contexts normally constitute only 5%-20% of the overall assemblage;81 the type fossils of the later Roman period also include fine wares and RBH coarse-ware as sherds, but they include body sherds well, which often a form significant proportion of excavated ceramic assemblages. To draw archaeological and historical conclusions from the number of type fossils

the total count, and RBH would fall to functional class labeled "Other" in the 81. In some cases fine-ware sherds 6.1% totals for the broader Roman an even only (Papadopoulos 1991, p. 82). period may constitute smaller fraction. ex a At Pyrgouthi in the Berbati valley, (seeTable 6) ismainly due to ceramic In survey of the Late Roman forti cavation of a Late tiles were not a more Antique farmstead roof that assigned fication and Early Christian basilica at at by the Swedish Institute Athens specific date. Forty-eight nonceramic Louloudies, south of Thessaloniki, a total of of ware to produced 8,500 sherds, artifacts (glass, architectural fragments, Late Roman fine amounted less which 11.8% = were of amount (n 1,000) RBH ground stone, tesserae, plaster, etc.) than 1% by weight the total were (Hjohlman, Penttinen, andWells 2005, also included in this count, but of pottery collected (M. Beckmann in these not p. 234). do greatly affect the overall Poulter 1998, pp. 503-511). 80. The greater significance of the proportions of the sample. 77? DAVID K. PETTEGREW

alone, without correcting for these differences in visibility and sampling to biases, is compare apples and oranges, exaggerating the evidence for a the later period relative to the earlier. This simple fact has variety of implications for the praxis of regional archaeological survey generally, the on as interpretation of changing settlement patterns the basis of surface our semblages, and understanding of regional diversity in the Roman and Late Roman landscapes of Greece.

Implications for Archaeological Survey Method

con The conclusions presented here have two major implications for those a ducting archaeological surface survey. First, survey projects have respon sibility to explain how their sampling strategies have produced their ceramic are data sets and how their historical interpretations derived from that data. a murmur Despite growing inMediterranean survey circles against uncrit some an ical quantification,82 counting pottery is in respects indispensable prerequisite for archaeological interpretation. Although quantitative studies a a crew add degree of intensity that may slow down survey whose aim is move to efficiently through the countryside looking for sites, it is essential a to to understand how ceramic sample relates the original ceramic popula tion, and thereby affects historical interpretation.83 Regional surveys must amanner publish their ceramic finds in complete and transparent enough to permit a reader to follow the entire process that leads to the interpreta tion of settlement patterns, and in a way that allows for and encourages reanalysis of the kind presented above. on Second, the recent emphasis in ceramic publications fabric analysis wares one and local is of the most valuable developments in regional survey, for it significantly increases the number of type fossils available for certain periods.84 The local ceramic typologies produced byjoanita Vroom in her analysis of post-Roman ceramic assemblages for the Boeotia Survey, for to example, have made it possible fill out the settlement history of the to region.85 Melissa Moore Morison has examined Hellenistic Late An ware tique utilitarian pottery from southern Epirus, establishing regional on groups for survey pottery based p?trographie analysis and dated with a are to knowledge of locally excavated pottery; these groups then used measure changing levels of imports and local wares.86 Clare PickersgilTs coarse wares ongoing study of locally produced Roman and fine from Sparta to should provide similar insights and help clarify the confusing picture of the Roman period presented by the data from the Laconia Survey.87 Such are studies of local ceramics indispensable for the interpretation of survey are or data, especially in regions where imports few lacking.

in 82. See, e.g., Fentress 2000. 85. Vroom 1998,2003,2004. Pickersgill and Roberts 2003, which 10 83. Caraher, Nakassis, and Pette 86. M. G. Moore 2000,2001. the authors analyze pottery groups from the Roman stoa and theater at grew 2006. 87. PickersgilTs doctoral dissertation examines Roman coarse and conclude that 84. See, generally,Moody 1985;Hag (in progress) Sparta (pp. 593-597) gis andMook 1993; Armstrong and and fine wares from excavated contexts locally produced pottery, rather than Hatcher et al. Kiri at and new fulfilled local demand 1997; Moody 2003; Sparta produces typological imported wares, atzi 2003; Alcock et al. 2005, pp. 194 sequences for local wares. Some of the throughout the Roman period. are in 204; Broodbank and Kiriatzi 2007. results of this study reflected THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 771

TABLE 11. FACTOR INCREASE BETWEEN EARLY AND LATE ROMAN PERIODS IN THE METHANA SURVEY AND EKAS (BY RBHS VS. RBH)

Roman Late Roman Factor Increase Roman Late Roman Factor Increase Early Early RBHS RBHS byRBHS RBH RBH byRBH 315 Methana 801 2.5 201 2291.1 331EKAS 1,707 5.2 195 4922.5

Measuring Change in Ceramic Deposition, Land Use, and Habitation

uneven Given the problems of differential visibility and representation we use discussed above, how should survey data to interpret historical no we change in the landscape? There is simple computation by which can correct for Early Roman underrepresentation, and any attempt at as calibration must be tentative insofar it rests upon unstable premises to move a a and attempts from known sample to putative total ceramic population that must always remain unknown. Nevertheless, attempts at are to a analysis and calibration, however imperfect, preferable simple and uncritical comparison of the sample populations of both periods, and they can to a lead better understanding the Roman data and the settlement

patterns they represent. to The simplest way minimize the effects of differential visibility with out to two on calibration is compare the periods the basis of similar artifact types. Rather than simply comparing the total number of sherds for each one can period, eliminate body sherds from the counts altogether, since are so these unevenly represented. The results of the application of this to are method theMethana Survey and EKAS data displayed inTable 11, inwhich the factor increase between Early and Late Roman based on total counts can to on be compared the increase based only RBH. With this adjustment the Late Roman presence is much reduced: the explosion of a Late Roman settlement becomes gentler upturn in the eastern Corinthia a and pattern of stability inMethana. Figures 5 and 6 display the adjusted area. pattern for the EKAS Contrasted with Figures 2 and 3, which show the density of RBHS (i.e., the total count) for each period, Figures 5 and 6, are on which based the density of RBH without the body sherds, show a less explosive increase in the Late Roman period, although a material upturn is still evident.88 A related way of measuring relative differences between periods is to restrict the comparison of ceramic remains to functional classes that are to less susceptible radical differences in relative visibility and identification case (cf.Table 12, below). In the of the EKAS data, comparison of the fine a wares, kitchen wares, and lamps shows variable degree of increase between the Early and Late Roman periods, but nothing like the change that is sug 88. The density scale in Figs. 5 and we gested by comparison of the total counts for each period. If also exclude 6 has been to better the adjusted display coarse wares body sherds from the medium and amphoras and instead rely reduced amounts of pottery resulting on RBH the coarse-ware from counts, number of Late Roman the removal of the body sherds solely fragments and to of from to a much closer to the Roman count permit easier comparison the drops 1,417 235, figure Early relative densities of the two periods. of 118. Measured in this way, the factor increase in ceramic material from 772 DAVID K. PETTEGREW

Isthmia

Artifacts m /10,000 sq Kenchreai 1-25 If 2^-100

A'-jt0t~17?;\ * - . V>'v

fe^^'for Ikjroeters

5. hectare of to on Figure Density per the Early the Late Roman period ranges from 1.2 to 2.1, depending Roman rims, bases, and han which class one takes as the standard of measurement. Early dles in the main EKAS area more survey Yet another, subtle, method of comparison is to analyze differ ences in the numbers of rims, handles, and bases in each period (Tables 8, 9). In the eastern Corinthia the number of bases (33 ER vs. 37 LR) and vs. constant handles (111 ER 110 LR) remains between periods, but the rims increase significantly (49 ER vs. 322 LR). InMethana the number of rims almost doubles (76 ER vs. 130 LR), but handles remain constant (74 vs. vs. ER 75 LR) and bases decrease by 50% (51 ER 24 LR). Comparing the periods in this way shows that, at least in these two surveys, an increase in rims is chiefly responsible for the overall increase in the count of RBH an for the Late Roman period. The disadvantage of such approach is that to the results will vary according the shifting ratios of functional classes with different RBHS ratios: amphoras, for example, produce far fewer rims 89. On the other and bases than fine-ware hand, amphora plates.89 rims tend to be better than If our to remained constant between preserved ability identify pottery periods, fine-ware rims due to their thickness, we also be able to reclaim some of the dated to the broader as might pottery and amphoras also have handles, Roman more do not. period by using the ER:LR ratio of the precisely dated sherds plates LATE THE BUSY COUNTRYSIDE OF ROMAN CORINTH 773

Isthmia

Artifacts m /10,000 sq Kenchreai 1-25

26-100 ' % 101-176 ^

Kilometers

Figure 6. Density per hectare of Late in each of the functional classes. For example, EKAS identified 127 Early Roman rims, bases, and handles in Roman fine wares and 165 Late Roman fine wares, a ratio of 43.5% to the main EKAS survey area 56.5%. The number of fine-ware sherds dated to the broad Roman period same we are is 438; using the ratio, would estimate that 191 of these Early Roman and 247 Late Roman. The addition of these numbers to those of counts the precisely dated sherds would produce adjusted total fine-ware of 318 and 412 for the two periods. The adjusted counts for this and other are functional classes presented inTable 12.90Although these adjustments affect the total counts, they do not appreciably change the overall ratio of to Late Roman.91 90. These calculations are based on Early most not and Late the for the Roman in however, that surveys do figures period Accepting, identify Early coarse ware Roman at we can a of calibration factors Table 6. The figures for and pottery equal rates, apply variety are RBH amphoras calculated using to correct for the difference. Ratios derived from ceramic assemblages a only. produced by excavation (see above) provide useful starting point. As 91. the Using unadjusted figures, an RBH:S ratio of we can use the known value of = suming expected 10:90, Late Roman (n 502) outnumbers Roman RBH = in the EKAS to estimate an = a of Early (n 195) sample expected Early Roman (n 330) by factor sherd of such a calibration factor would increase 1.52; the adjusted factor increase is body population 1,755; to 1.49. the total Early Roman ceramic population 1,950, approximately six 774 DAVID K. PETTEGREW

TABLE 12. FACTOR INCREASE IN EKAS FOR FUNCTIONAL CLASSES BETWEEN EARLY AND LATE ROMAN PERIODS BASED ON ADJUSTED COUNTS

Adjusted Adjusted Factor Functional Class Roman Late Roman Roman Ratio Roman Late Increase Early Early Roman

Coarse ware and amphora (RBH) 118 235 399 33.4% :66.6% 5012.0251 127Fine ware 165 438 43.5% :56.5% 4123181.3 Kitchen ware82 96 294 46.1% :53.9% 2542181.2 3 Lamp6 34 33.3% :66.7% 29 2.114

Total 330 801 5021,196 1.5

= times the number of ER sherds (n 331; Table 6, above) identified by EKAS. The use of a more moderate RBH:S ratio of 20:80 reduces the

estimate to 780 body sherds, which would increase the total Early Roman same ceramic population to 975. Performing the calculations for Late = Roman RBH (n 492) would bump up the LR ceramic population to 4,920 or (using the 10:90 ratio) 2,460 (using the 20:80 ratio). This calibration significantly changes the relative ceramic representation of the two periods, producing Late Roman totals only 2.5 times greater than the Early Ro man totals. as we can Similarly, taking 10% the expected percentage of fine wares, an estimate from the 127 Early Roman fine-ware sherds expected popula coarse tion of 1,143 and cooking-ware sherds, bringing the potential total to population of Early Roman material 1,270. These figures suggest that = the EKAS total count for the Early Roman period (n 331) underrepre sents a at the original population by factor of least three and possibly much more. same to count Applying the calibration the Late Roman fine-ware = an coarse (n 165) produces estimate of 1,485 and cooking-ware sherds a a to and potential total population of 1,650, figure very close the actual = count of identified LR pottery (n 1,707). This calibration, then, reduces the degree of difference between the Early and Late Roman periods: the a adjusted LR population outnumbers the adjusted ER population by factor of only 1.3. The analyses presented above, in which the factor of increase ranges between 1.2 and 2.5, demonstrate that the Early and Late Roman ceramic are more populations in EKAS in fact equal than they initially appear. are in Moreover, because the two periods of unequal length?281 years case b.c-a.d. at the of the Early Roman period (31 250), and least 450 in the case of the Late Roman (a.d. 250-700)?modest increases in the on total amount of pottery might be expected. Based the proportion of we to 281 to 450, might expect the Late Roman ceramic population be 1.6 a constant rate times greater than the Early Roman, assuming of ceramic across deposition the entire Roman period. This number falls well within the calibrated range of 1.2-2.5. an Calibration and correction for differential visibility is important step in the process of interpreting change in the Roman countryside. Survey to archaeologists must carefully examine their data in order determine to whether such problems of differential visibility apply their regions, and THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 775

serve if so,which calibration factors will them best. This is true not only for those whose primary interest lies in the Early and Late Roman periods, but for those studying other "boom-and-bust" cycles in the ancient landscape as well. Might the expansion of settlement in the Late Classical period, for a example, also be at least partly consequence of similar source issues?

Reading the Late Roman Landscape of Greece

us to Finally, this study encourages consider different paradigms for in terpreting the overall pattern of activity in Greece and the Aegean during the Roman period. Previous interpretations have related the explosion of to Late Roman material the proliferation of Late Roman sites and, by to as inference, historical and economic factors such population growth, an increase in intensive agriculture, changes in imperial land-use policy, and general economic prosperity. Levels of Late Roman abundance may, a s however, also speak to complex issues like the extent of region connec tions to broader networks of exchange and distribution over time. The a critical question iswhether lack of Early and/or Late Roman pottery in some a regions indicates genuine absence of rural settlement, or whether a it simply reflects poor sample for the period, caused by limited regional access to more a on easily identifiable imported pottery and dependence local wares that remain unidentified.92 to Archaeologists have tended prefer the former explanation, but there some to is evidence suggest that the latter explanation is often the correct one, for the pattern of dramatic increase in Late Roman settlement ismost access to striking in regions with the readiest coastal sites and exchange systems (Fig. l).The island of Keos, the southern Argolid, and theMethana are to sea peninsula all situated close the and positioned along major trade are and distribution routes, and all regions where surveys have produced abundant evidence for Late Roman expansion.93 By contrast, the pattern of as Roman settlement in inland regions such the Nemea valley, the Berbati area valley, Megalopolis, the Asea valley, and the Laconia survey is less seem an clear.94 Even Boiotia, which might to be exception because it is

92. Cf. the and criti more most seen as a con thoughtful produced significantly than inland periods should be cal discussion Andrea Berlin and n. by regions (see 94, below). The evidence sequence of the isolation of the region Sebastian Heath of similar from east-west rather than a of interpretive Pylos suggests that sign depopulation, and in the data collected the networks and on wonders whether the problems by exchange dependence greater frequency Alcock et al. wares of wares Pylos survey: 2005, imported changed significantly Middle Byzantine in the val 194-204. to a pp. from the Hellenistic the Late Roman ley might simply be result of greater 93. For see the comments of levels local The Keos, periods, with higher of imports production. Berbati valley Sutton on to a.d. a area (1991, p. 253) the island's from the 4th the 6th century survey, with study approximately well Berlin and Heath have the same size as that of counted "outward-looking economy," suggested that EKAS, connected with of over time in vs. a few of Italian plenty imports. fluctuations imported only pieces sigillata and in the southern local fine ware and have a total Similarly, Argolid, amphoras of only 58 fragments of Red Slip van Andel and Runnels 1 lo ware a (1987, pp. greatly affected the number of rural (Forsell 1996, pp. 330-331), bare return ll 7) have linked the of prosper settlements identified for different fraction of the EKAS total.At Megalo in the Late Roman to the et the of settlement ity period periods: Alcock al. 2005, pp. 194 polis picture recovery to the sea. The 204. in the seems region's proximity Pylos Late Roman period clear, while fewer et but the is not survey area, producing 94. Sutton (inWright al. 1990, growth explosive (Roy, ceramics than the imported Keos, pp. 657-659) suggests that the paucity Lloyd, and Owens 1989, pp. 149-150); southern and still of in the Nemea Argolid, Methana, imports valley during Lloyd (1991, p. 188) has noted the 776 DAVID K. PETTEGREW

not geographically or commercially advantaged, yet still conforms to the typical pattern of Late Roman settlement expansion, turns out to prove the rule, for not only was the region weak in imports, but the ceramic evidence on for the Late Roman period is based almost entirely locally produced imitations of Late Roman amphora forms and beehives.95 An argument that relates ceramic abundance to distribution networks would also explain the Roman ceramic data from EKAS.96 Although EKAS a area surveyed intensively much smaller overall (ca. 4 km2) than many other regional surveys, the amount of fine-ware and imported amphoras recorded for both the Early and Late Roman periods exceeds that of many areas a other regions. Because of the survey location at Mediterranean crossroads, and its proximity to the port of Kenchreai, the sample includes were many ceramic types that widely distributed in the Roman period, with a s resulting increase in the period visibility (Table 13).97 That the lst-2nd and 5th-6th centuries appear to shine most brightly in this analysis does mean saw not necessarily that those centuries the highest level of settle were ment; itmay simply indicate that these the periods of greatest regional connection to broader networks of exchange.98 on To settle such questions, further research is needed the relationship between patterns of exchange, period visibility, and exurban settlement. a There may be relationship between the density of rural settlement and

would be much less clear. distribution is divided between difficulty of recognizing diagnostic Ro period systems man 96. Cf. those who favor state-driven material because imported table Pettegrew, forthcoming b, explana wares in for a fuller discussion of the tions Wickham reached the countryside only interpreta (Whittaker 1983; In the tion of the boom-and-bust in the small quantities. Asea valley the cycle 1988,1998; Abadie-Reynal 1989; a the "abandon Fulford Durliat and those investigators found flourishing Late Corinthia, including 1996; 1998) a ment" of the 3rd-4th centuries. who favor based on mar Roman landscape, but also noted explanations a 97. The dates to ket or demand in paucity of imports and dependence assigned pottery (overview Kingsley on wares in 13 are from and Decker see also the other local during the Early and types Table derived 2001b; 1972 recent work on articles in the same Late Roman periods (Karivieri 2003, Hayes and from volume, especially pp. 275-276,288; Fors?n and Karivieri the Roman pottery from Corinth. Slane Kingsley 2001 andWard-Perkins 2003, p. 307). In Laconia, the investi has shown (CorinthXVIII.2, pp. 47-54; 2001). On widespread wealth and a that Eastern in late see gators have suggested that lack of 2000; 2003, pp. 330-331) purchasing power antiquity, ware and A is uncommon in both urban Blake imported red-slipped Sigillata 1978, pp. 436-440; Banaji 1996; account for our and contexts after the middle and amphoras might poor sanctuary Maguire 2001; Banaji 2001, the Roman of the 1st that Eastern with and understanding of Late century a.D.; pp. 60-65,218-221, response the area: see n. B can date to the 1st cen discussion in Kehoe 2003. There is also period in survey 14, Sigillata early but occurs in 2nd a middle course: seeWard-Perkins above, and cf. Pickersgill and Roberts tury a.D., especially and can even be 168. See 2003 for the importance of local century deposits pushed 2000b, pp. 369-381; 2001, p. at into the 3rd and that Wickham for the ceramic production Sparta. century; ?andarli 2005, pp. 693-824, see ware is used as as the late 1st cen role of aristocratic demand in 95. For the paucity of imports, early generat 106-107. In Vroom's but has its in the 2nd to However we ex Hayes 2000, pp. tury, heyday ing regional exchange. ceramic breakdown of 19 sites with Late Ro 3rd century. Some of the fine-ware and plain the patterns, imported man sherds (2004, pp. 308-324), fine red-slipped sherds noted inTable 13 commodities such as fine wares and wares ware local wares. Al were distrib (locally produced Askra presumably represent amphoras certainly widely show that came uted in the Mediterranean from the 5 th especially) constitute 6% of all Late though the data imports from both the western and eastern Med to 7th cf. Blake 436 Roman finds; amphoras (probably century: 1978, pp. of Late the latter was a much more Wickham locally produced imitations iterranean, 440; 1988, p. 190; 1998; coarse is not Roman 2) 29%; wares 3%; important source; this surprising, Reynolds 1995;Ward-Perkins 2000b; and 62%. that the area lies immedi and locally produced beehives given survey Kingsley 2001, pp. 55-58; Kingsley north and west of Kenchreai. Decker Rautman Without the amphoras and beehives, ately 2001b, pp. 11-13; on the evidence for settlement in this 98. The scholarship Late Roman 2003. THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 777

TABLE 13. COUNTS OF ROMAN FINE WARES (BY EXTANT PART) IN EKAS

Date Rim Base Handle Total Count Chronotype Body Sherd

Roman

1st B.c-late 7th a.d. Roman fine ware 29 14 92 138 1st B.c-late 7th a.d. Roman Red Slip 68 39 110 217 1st B.c-late 7th a.d. Roman semifine ware 13 3 8 26 2nd-7th a.d. ARS 12 4 41 57 Subtotal 122 60 251 438

Early Roman

1st B.c-3rd a.d. ER semifine ware 0 0 0 2 1st B.c.-3rd a.d. ER Red Slip 3 1 7 11 1st B.c-3rd a.d. Eastern Sigillata 8 2 16 26 1st B.c-lst A.D. Arretine 1 0 0 1 1st B.c-lst A.D. Eastern Sigillata A 3 5 10 18 lst-3rd a.d. Eastern Sigillata B 2 2 12 16 a.d. 1st century Eastern Sigillata Bl 1 0 6 7 lst-3rd a.d. Eastern Sigillata B2 10 5 29 44 a.d. ware lst-3rd ?andark 0 0 2 2 Subtotal 28 15 82 127

Late Roman

3rd-4th a.d. ARS 50 1 0 0 0 1 3rd-late 7th a.d. LR fine ware 2 0 0 0 2 3rd-late 7th a.d. LR Red Slip 9 5 0 4 18 Late 4th-late 7th a.d. Egyptian Red Slip 0 0 0 2 2 Late 4th-late 7th a.d. Cypriot Red Slip 0 1 0 2 3 4th-7th a.d. Phocaean ware 39 19 0 10 68 5th-6th a.d. Phocaean ware 3 44 0 0 2 46 5th-6thA.D. ARS 99 9 0 0 0 9 5th-6th a.d. ARS 104-106 0 0 0 a.d. ware Late 6th-early 7th Phocaean 10 0 0 0 Subtotal 120 25 0 20 165

Total 270 100 353 730

= = = ER Early Roman; LR Late Roman; ARS African Red Slip

as some the development of systems of exchange, scholars have argued, we not assume or but need that the relationship is necessarily linear even even wares more regular.99 It is conceivable that imported fine might be or abundant in the countryside during periods of lower overall settlement population.100 A deeper understanding of ceramic chronologies, supply, and us to distribution will certainly help understand historical change in Ro man landscapes.

99. The economic made difficult to evaluate. An obvious test argument uted through commercial networks. For the van would be with Late Roman another the for southern Argolid by Andel comparison argument linking develop settlement in an inland ment of Late to and Runnels (1987, pp. 113-117), patterns region Antique settlements eco in but in such the of see relating settlement, population, Greece, regions pat the growth markets, Sarris 2004. nomic access to tern growth, and Mediter is often less visible because of the 100. See, e.g., Sanders 2004, ranean is but markets, fascinating paucity of diagnostic artifacts distrib pp. 165-166. 778 DAVID K. PETTEGREW

CONCLUSION: THE LATE ROMAN CORINTHIA

The eastern Corinthia, lying between Corinth, its eastern harbor, Kenchreai, and the site of Isthmia, was one of the busiest territories in Roman Greece, as is evident from the abundance and distribution of artifacts throughout the area.1011 have argued that the shifts in ceramic abundance between the are a Early and Late Roman periods product of artifact-sampling systems common to regional surveys and the differential visibility of the two pe are riods in the field, and that they also connected to broader patterns of supraregional exchange. Calibration of the survey data to correct for the an difference in visibility reduces the evidence for explosion of Late Roman settlement and reveals amore constant and continuous pattern of ceramic as a deposition (and, by inference, of habitation well), rather than dramatic cycle of boom and bust. Still, the quantity of imported amphoras and fine ware testifies to the large number of houses, farmsteads, villas, shrines, and agricultural installations that existed in the shadow of Corinth during the as as on Late Roman period, well to the vitality of systems of exchange the Isthmus into the 7th century.102 This generally positive depiction of Late Roman Corinth has, of course, also been recently confirmed by scholarship on the urban center.103

Other historical factors may also have contributed to the ubiquity of Late Antique pottery in the eastern Corinthia. What effect, for example, did the construction of the Hexamilion fortress and the establishment of a on re garrison in the early 5th century have the local economy and the gional distribution of pottery?104 Did the flurry of Early Christian church an as construction provide economic stimulus well?105 However such factors are may have influenced the ceramic record, they further examples of ways era. in which the Isthmus remained "busy" in the Late Roman as on to an Just recent scholarship Corinth itself has begun revise so overly pessimistic picture of the Late Antique city, the history of activ a ity and settlement in the territory east of Corinth needs to be read in more positive light.106 Every indication suggests that this busy countryside to continued function and flourish throughout the entire Roman period, in spite of the disasters of plague, earthquake, and invasions. If the Corinthia as a suffered from the 3rd-century crisis that affected the empire whole,

lated the local on role one to consider for the 101. Although I have focused here economy: the interesting on source of the the ceramic analysis, I have army, government, and the Corinthia. in the elsewhere discussed the patterns of church in creating localized economic 105. On churches Corin on in see Sanders 2005. Roman and Late Roman settlement hotspots the Late Roman empire, thia, Caraher 2003; the Isthmus as revealed by the EKAS see Fulford 1996, pp. 158-162; Kings For consideration of the same issue see Bowden 151 data: Pettegrew 2006, forthcoming a, ley and Decker 2001b, pp. 5-11; Dunn elsewhere, 2003, pp. in 2004. 154. forthcoming b, prep. Karagiorgou (2001) has sug 102. On the of the that the wide distribution 106. For revisions to the histo importance P?lo gested pat in east-west trade terns of Late Roman 2 in of town and see ponn?se generally amphoras riography country, routes this see Abadie the and the Balkans should and during period, Aegean Gregory 1985; Kardulias, Gregory, be understood in terms of Sawmiller Sanders Reynal 1989, p. 56. military 1995; 1999,2004, 103. Slane 2000; Slane and Sanders involvement in these areas, since these 2005; Rothaus 2000; R. S.Moore 2000; 2005. vessels contained olive oil used for the Robinson 2001; Caraher 2003; Slane of border Such 104. For the impact of the Hexamil provisioning troops. and Sanders 2005; Brown 2005, forth see of not Kardulias ion fortress, Kardulias 2005. Recent amphoras are, course, restricted coming a, forthcoming b; that the mili to border but the of scholarship has suggested areas, possibility 2005; Pettegrew 2006, forthcoming a, a connection is nevertheless an Rife et al. 2007. tarization of society could have stimu such forthcoming b; THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 779

permanent discontinuity in rural settlement and building activity was not a the result. If late-4th-century earthquake made Corinth tremble, there is no on evidence that its long-term effects the countryside were crippling. If or the Heruls the Visigoths ransacked the region, the countryside recovered. at a Even in the wake of the 6th-century plague and time of alleged Slavic was some invasion, imported pottery being used and deposited at of the on major rural sites the Isthmus. Only in the 7th and 8th centuries do the lights of the Corinthian crossroads dim and go out. This localization of the regional economy marks the decisive end of the role of the Isthmus in the life of the ancient city.

REFERENCES

C. 1989. "Les am Abadie-Reynal, Armstrong, P., and H. Hatcher. 1997. phores protobyzantines d'Argos "Byzantine and Allied Pottery, (IV-VI si?cles)," in Recherches sur la Phase 2: Past Works on Materials Actes du c?ramique byzantine. colloque Analysis and Future Prospects," dAth?nes inMaterials organis? par VEcolefran?aise Analysis of Byzantine et FUniversit? de II ed. H. Strasbourg Pottery, Maguire, Washing de recherches sur (Centre VEurope ton, D.C., pp. 1-8. centrale et A. 1997. Le du sud-orientale), Ath?nes, Avramea, P?loponn?se 8-10 avril 1987 IVe au VIIIe si?cle: et (BCH Suppl 18), Changement ed. V. D?roche and J.-M. Spieser, persistances (Byzantina Sorbonensia Paris, pp. 47-56. 15), Paris. V = H. S. the Agora Robinson, Pottery of Banaji, J. 1996. "The Circulation of Roman Period Princeton as an (Agora V), Gold Index of Prosperity in 1959. the Central and Eastern Mediter S. E. 1993. Graecia ranean in in Alcock, Capta: Late Antiquity," Coin Roman The Landscapes of Greece, Finds and Coin Use in the Roman Cambridge. World. The Thirteenth Oxford Sympo S. A. M. A. B. Har sium on and His Alcock, E., Berlin, Coinage Monetary S. N. and rison, Heath, Spencer, tory, 25.-27.3.1993, ed. C. E. King D. L. Stone. 2005. "Pylos Regional and D. G. Wigg, Berlin, pp. 41-53. Part VII: -. 2001. in Late Archaeological Project, Agrarian Change Historical Geometric and Aristo Messenia, Antiquity: Gold, Labour, Late through Roman," Hesperia 74, cratic Dominance, Oxford. 147-209. and A. 1991. pp. Barker, G., J. Lloyd, eds. S. F. and L. Roman Alcock, E., J. Cherry, J. Landscapes: Archaeological Davis. 1994. "Intensive in theMediterranean Survey, Survey Region and the Clas of Agricultural Practice, (Archaeological Monographs the sical Landscape of Greece," in Clas British School at Rome 2), London. sical Greece: Ancient Histories and C. Barker, G., Mee, W. Cavanagh, Modern ed. I. M. Archaeologies, Morris, R. Schon, S. Thompson, J. Bint 137-170. P. Cambridge, pp. liff, Howard, and A. Snodgrass. A. 1985. "Plow-Zone to Ammerman, J. 2000. "Responses The Hidden in Experiments Calabria, Italy," Landscape of Prehistoric Greece,' JFA 12, pp. 33-40. by J. L. Bintliff, P.Howard, and -. 2004. "Farewell to the Garden A. M. Snodgrass (JAM 12.2, Decem of Eden: Survey Archaeology after ber 1999); JMA13, pp. 100-123. the Loss of m Innocence," Archaeo Bintliff, J. 1991. "The Roman Coun Field in Past in logical Survey Cyprus: tryside Central Greece: Observa Future Potentials. History, Proceedings tions and Theories from the Boeotia a Held of Conference by the Archaeo Survey (1978-1987)," in Barker and logicalResearch Unit of theUniversity Lloyd 1991, pp. 122-132. 1-2 ed. -. 2000a. on of Cyprus, December, 2000, "Beyond Dots the M. 177-182. Iacovou, London, pp. Map: Future Directions for Surface K. PETTEGREW 780 DAVID

Artefact in in The Revisited: Survey Greece," Technology, Demography, Future ace in and in the of Surf Artefact Survey Landscape Prepalatial ed. M. and 241-274. Europe, J. Bintliff, Kuna, Aegean," AJA 111, pp. N. 3-20. A. R. for Venclov?, Sheffield, pp. Brown, 2005. "Waiting the -. 2000b. "The of Barbarian Concepts Invasions: Reconciling in 'Site' and 'Off-Site' Archaeology Archaeological and Historical inNon at Surface Artefact Survey," Evidence for Barbarians Corinth Destructive to in Late Techniques Applied Antiquity" (paper, Boston Landscape Archaeology (The Archae 2005). of -. a. or ology Mediterranean Landscapes Forthcoming "Banditry ed. M. F. 4), Pasquinucci and Tr? Catastrophe? History, Archaeology, ment, Oxford, pp. 200-215. and Barbarian Invasions of Roman -. 2000c. The in Frontiers in "Deconstructing Greece," Shifting Sense of Place'? Settlement Late 6: Barbar Sys Antiquity Romans, Field and the Historic and the the tems, Survey, ians, Transformation of A Roman Record: Case-Study from Central World, ed. R. Mathisen and Greece," PPS 66, pp. 123-149. D. Shanzer. -. to Bintliff, J. L., P. Howard, and A. M. Forthcoming b. "Lost the 1999. Snodgrass. "The Hidden Countryside? Abandonment and Landscape of Prehistoric Greece," Continuity in Late Antique Corinth JMA12, pp. 139-168. and Thessaloniki," International Historical Bintliff, J., and A. Snodgrass. 1985. Journal of Archaeology. W. R. "The Cambridge/Bradford Boeo Caraher, 2003. "Church, Society, tian Expedition: The First Four and the Sacred in Early Christian Years/'/i^ 12, pp. 123-161. Greece" (diss. Ohio State Univ.). -. 1988a. "The End of the Caraher, W. R., R. S. Moore, J. S. Nol from K. Roman Countryside: A View ler, and D. Pettegrew. 2005. "The the in First Millennium East," Vyhi-Koutsopetria Archaeological Western in the First Papers: Europe Project: First Preliminary Report Millennium a.D. (BAR-IS 401), (2003-2004 Seasons)," RDAC 2005, ed. H. S. R.Jones, Bloemers, Dyson, pp. 245-268. -. and M. Biddle, Oxford, pp. 175-217. Forthcoming. "The Vyhi-Kout -. Dis 1988b. "Off-Site Pottery sopetria Archaeological Project: tributions: A Regional and Inter Second Preliminary Report (2005 regional Perspective," CurrAnthr 29, 2006 Seasons)," RDAC pp. 506-513. Caraher, W. R., D. Nakassis, and D. K. 1978. Blake, H. "Medieval Pottery: Pettegrew. 2006. "Siteless Survey Technical Innovation or Economic and Intensive Data Collection in an in in Italian Ar Artifact-Rich Change?" Papers Environment: Case 1: Lancaster Seminar. from Eastern chaeology The Studies the Corinthia, Recent Research in Prehistoric, Greece; JMA 19, pp. 7-43. and Medieval and T. 2004. Classical, Archaeology, Chavarria, A., Lewit. on Pt. 2 (BAR Suppl. 41), ed.H. McK. "Archaeological Research the Blake, T. W. Potter, and D. B. Late Antique Countryside: A Bib Whitehouse, Oxford, pp. 435-473. liographic Essay," in Bowden, La Bowden, H., and D. Gill. 1997a. "Late van, and Machado 2004, pp. 3-51. Roman Methana," inMee and Cherry, J. F, J. L. Davis, and E. Man Forbes 84-91. eds. 1991. 1997a, pp. tzourani, Landscape -. in as 1997b. "Roman Methana," Archaeology Long-Term History: Mee and Forbes 1997a, pp. 77-83. Northern Keos in the Cycladic Islands 2003. Vetus: The Bowden, W. Epirus (Monumenta Archaeologica 16), a Late Prov Los Archaeology of Antique Angeles. ince, London. Cherry, J. F, J. L. Davis, E. Man Bowden, W., L. Lavan, and C. Ma tzourani, and T. M. Whitelaw. 1991. in chado, eds. 2004. Recent Research "The Survey Methods," Cherry, on Late and Mantzourani the Antique Countryside Davis, 1991, (LateAntique Archaeology 2), pp. 13-35. N. of Leiden. Christie, 2004. "Landscapes Broodbank, C, and E. Kiriatzi. 2007. Change in Late Antiquity and First "The 'Minoans' of Kythera the Early Middle Ages: Themes, THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 781

1993. Directions, and Problems," in Land Fors?n, J., B. Fors?n, and M. Lavento. Haggis, D., and M. Mook. "The Kavousi Coarse Wares: A Bronze scapes of Change: Rural Evolutions 1996. "TheAsea Valley Survey: in Late and the Mid 1994 in Antiquity Early A Preliminary Report of the Age Chronology for Survey the dle N. Ages, ed. Christie, Aldershot, Season," OpAth 21, pp. 73-97. Mirabello Area, East Crete," AJA pp. 1-37. Francovich, R., H. Patterson, and 97, pp. 265-293. Corinth = Corinth: Results Excavations G. eds. 2000. 1972. Late Roman of Barker, Extracting Hayes, J. Pottery, Conducted American London. by the School of Meaningfrom PloughsoilAssemblages Classical Studies at -. Athens (The Archaeology of Mediterranean 1976. "Pottery: Stratified = in Excava XVIII.2 K.W. Slane, The Landscapes 5), Oxford. Groups and Typology," Demeter andKore: E. B. 1990. in tions at Conducted Sanctuary of The French, "Archaeology Carthage 1975, by Roman and Princeton AR 2-82. the ed. H. Pottery Lamps, Greece, 1989-90," 36, pp. University ofMichigan, J. 1996. 1990. Fulford, M. G. "Economic Hot Humphrey, Tunis, pp. 47-123. = -. XX C.K. Williams II and spots and Provincial Backwaters: 1992. Excavations at Sara?hane N. Cen the in Bookidis, eds. Corinth, the Modelling Late Roman Econ Istanbul 2\ The Pottery, Princeton. -. tenary: 1896-1996, Princeton 2003. omy," in Coin Finds and Coin Use 2000. "The Current State of B. Cosmopoulos, M. 2001. The Rural in the Roman World. The Thirteenth Roman Ceramic Studies inMedi Ancient Greek on and terranean or History of City-States: Oxford Symposium Coinage Survey, Handling The from in Oropos Survey Project (BAR-IS Monetary History, 25.-27.3.1993, Pottery Surveys," Franco D. 1001), Oxford. ed. C. E. King and G. Wigg, vich, Patterson, and Barker 2000, L. 2004. "Are the Davis, J. Landscapes Berlin, pp. 153-177. pp. 105-109. of Greek Prehistory Hidden? A Fulford, M. G., and D. P. S. Peacock, Hirschfeld, Y. 2001. "Habitat," in Inter in eds. 1984. Excavations at Late on the Comparative Approach," Side-by Carthage: preting Antiquity: Essays Side Postclassical G. Survey: Comparative Regional The BritishMission 1.2: The Avenue World, ed. W. Bow Studies in theMediterranean du Pr?sident Habib P. O. Area, Bourguiba, ersock, Brown, and Grabar, ed. S. E. Alcock and F. Salammb?: The and J. Cherry, Pottery Other Cambridge, Mass., pp. 258-272. 22-35. Ceramic the Shef Oxford, pp. Objects fom Site, Hjohlman, J., A. Penttinen, and Dunn, A. 2004. "Continuity and field. B.Wells. 2005. Pyrgouthi:A Rural in Change the Macedonian Coun Gerstel, S. E. J., M. Munn, H. E. Site in theBerbati Valleyfrom the from Gallienus to E. A. H. Iron to Late tryside Justinian," Grossman, Barnes, Rohn, Early Age Antiquity: in and Machado and M. Kiel. 2003. "A Late Medi Excavations Swedish Bowden, Lavan, by the Institute 2004, pp. 535-586. eval Settlement at Panakton," Hes atAthens 1995 and 1997 (SkrAth4?, 1998. "Les conditions Durliat, J. du peria 72, pp. 147-234. 52), Stockholm. commerce au VIe in si?cle," The Gill, D., C. Mee, and G. Taylor. 1997. Jameson, M. H., C. N. Runnels, and Sixth Distribu "Artefact inMee van A Century: Production, Catalogue," and T. H. Andel. 1994. Greek and ed. R. tion, Demand, Hodges Forbes 1997a, pp. 282-343. Countryside:The SouthernArgolid and W. 89-117. and A. B. 2003. The to the Present Bowden, Leiden, pp. Given, M., Knapp. from Prehistory Day, Engels, D. 1990. Roman Corinth: An Sydney Cyprus Survey Project: Social Stanford. Alternative Model the Ancient to H. for Approaches Regional Archaeological Jones, A. M. 1964. The Later Roman 284-602: A City, Chicago. Survey (Monumenta archaeologica Empire, Social, Economic, E. 2000. "What Are We Los and Fentress, 21), Angeles. Administrative Survey, Norman, For?" in E. Counting Francovich, Gregory, T. 1985. "An Early Byzan Okla. and Barker tine at near Patterson, 2000, Complex Akra Sophia Karagiorgou, 0.2001. "LR2: A Con 44-52. 411 on pp. Corinth," Hesperia 54, pp. tainer for the Military Annona Fentress, E., and P. Perkins. 1989. 428. the Danubian Border?" in Kingsley African -. 1994. "Counting Redslip Ware," "Archaeology and Theo and Decker 2001a, pp. 129-166. on AfrRom 5, pp. 205-214. retical Considerations the Tran Kardulias, P. N. 2005. From Classical to R. 1996. "The Roman sition to Forseil, Period," from Antiquity theMiddle Byzantine: Social Evolution in Late in The Berbati-Limnes in the in Be and the Fortress at Archaeolog Ages Aegean Area," Antiquity Isthmia, 1988-1990 Site: icalSurvey, (SkrAth 4?, yond the Regional Studies in the Greece (BAR-IS 1412), Oxford. ed. B. 285 ed. P. N. P. E. and 44), Wells, Jonsered, pp. Aegean Area, Kardulias, Kardulias, N.,T. Gregory, 343. 137-159. Sawmiller. 1995. "Bronze and Lanham, Md., pp. J. Age Fors?n, B., and A. Karivieri. 2003. "The -. 2004. "Less is Better: The Late Antique Exploitation of an Roman-Early Modern Periods: Quality of Ceramic Evidence from Islet in the Saronic Gulf, Greece," in Fors?n and Conclusions," Fors?n Archaeological Survey and Practical JFA 22, pp. 3-21. 307-331. for Low in A. 2003. in 2003, pp. Proposals Impact Survey Karivieri, "Roman Pottery," and B. Fors?n. 2003. aMediterranean Fors?n, J., The Context," inMedi Fors?n and Fors?n 2003, pp. 275 Asea An Arcadian terranean 289. Valley Survey: Archaeological Landscapes: Mountain Current ed. E. Valleyfrom thePalaeolithic Issues, Athanassopou Kehoe, D. P. 2003. "Aristocratic Domi Period untilModern Times los and L. Philadel nance in Late Roman (SkrAth4?, Wandsnider, the Agrar Stockholm. 15-36. 51), phia, pp. ian Economy and the Question 782 DAVID K. PETTEGREW

of Economie M. Growth,"/iM 16, J. Pontig, and E. C. Ribeiro. pp. 711-721. 2002. The Late Roman Church at S. 2001. "The Economic Maroni Petrera: and Kingsley, Survey Salvage Impact of the PalestinianWine Excavations 1990-1997, and Trade in Late in Other Traces Roman Remains in Antiquity," Kings of and Decker 44-68. the Lower Maroni ley 2001a, pp. Valley, Cyprus, Kingsley, S., and M. Decker, eds. Nicosia. 2001a. and in and Economy Exchange Mee, C, W. Cavanagh. 2000. "The East Mediterranean the during Late Hidden Landscape of Prehistoric a Antiquity. Proceedings of Conference Greece: A View from Laconia and at Somerville College, Oxford, 29th Methana, "/AM 13, pp. 102-107. May, 1999, Oxford. Mee, C, and H. A. Forbes, eds. 1997a. -. 2001b. "New New A and Place: Land Rome, Rough Rocky The Theories on Ex and Settlement Inter-Regional scape History of the to change: An Introduction the East Methana Peninsula, Greece, Liver Mediterranean in Late Economy pool. in -. 1997b. Antiquity," Kingsley andDecker "Survey Methodology," 2001a, pp. 1-27. inMee and Forbes 1997a, pp. 33-41. E. 2003. and N. 2003. Kiriatzi, "Sherds, Fabrics, Meyer, "Pottery Strategy and Sources: the in Given and Clay Reconstructing Chronotypes," Knapp Ceramic Landscapes of Prehistoric 2003, pp. 14-16. inMETRON: and T. E. Kythera," Measuring Meyer, N., Gregory. 2003. the Bronze Aegean Age. Proceedings of "Pottery Collection, Pottery Analy theNinth International Con Aegean sis, and GIS Mapping," inGiven New Yale ference, Haven, University, and Knapp 2003, pp. 48-52. 18-21 April 2002 (Aegaeum24), Millett, M. 1985. "Field Survey Cali ed. K. P. Foster and R. Laffineur, bration: A Contribution," in Archae Li?ge, pp. 123-130. ologyfrom thePloughsoil: Studies in C. K. 2003. The the Collection and Kosso, Archaeology of Interpretation of Public in Late Roman Greece Field C. Policy Survey Data, ed. Hasel (BAR-IS 1126), Oxford. grove, M. Millett, and I. Smith, 1996. Lawson, J. "The Roman Pottery," Sheffield, pp. 31-37. in The Laconia -. 1991a. or Survey: Continuity "Pottery: Population and in a Greek Rural Land Tarra Change Supply Patterns? The Ager 2: Data conensis in scape Archaeological (BSA Approach," Barker and Suppl. 27), ed.W. Cavanagh, Lloyd 1991, pp. 18-26. R. -. J. Crouwel, W. V. Catling, and 1991b. "Roman Towns and G. Shipley, London, pp. 111-123. their Territories: An Archaeological T. 1991. in and Lewit, AgriculturalProduction Perspective," City Country in Roman a.d. in the Economy, 200?400 the Ancient World, ed. J. Rich and (BAR-IS 568), Oxford. A. Wallace-Hadrill, London, 1991. the 169-189. Lloyd, J. A. "Farming High pp. -. lands: Samnium and Arcadia in the 2000a. "The Comparison Hellenistic and Roman of Surface and Early Impe Stratified Artefact in in rial Periods," Barker and Lloyd Assemblages," Non-Destructive 1991, pp. 180-193. TechniquesApplied toLandscape E. Lloyd, J. A., J. Owens, and J. Roy. Archaeology (The Archaeology of 1985. "TheMegalopolis Survey Mediterranean Landscapes 4), in of ed. Arcadia: Problems Strategy M. Pasquinucci and F. Tr?ment, in Field and Tactics," Archaeological Oxford, pp. 216-222. in Britain and -. 2000b. Survey Abroad, "Dating, Quantifying, ed. S. and F. H. Macready Thomp andUtilizing Pottery Assemblages from Surface in son, London, pp. 217-224. Survey," Franco in Maguire, H. 2001. "The Good Life," vich, Patterson, and Barker 2000, Late 53-59. Interpreting Antiquity: Essays pp. on the Postclassical World, ed. G. W. Moody, J.A. 1985. "TheDevelop P. and ment of a Bronze Bowersock, Brown, O. Gra Age Coarse Ware bar, Cambridge, Mass., pp. 238-257. Chronology for the Khania Region Manning, S.W., A. Manning, R. Tom ofWest Crete," TUAS 10, pp. 51 ber, D. A. Sewell, S. J.Monks, 65. THE BUSY COUNTRYSIDE OF LATE ROMAN CORINTH 783

L. inHonor E. Flavia in Moody, J., H. Robinson, J. Francis, ing of Timothy Gregory, Augusta Corinthiensis; and L. Nixon. 2003. "Ceramic Fab ed.W. R. Caraher, L. J. Hall, and Corinth XX, pp. 279-301. ric R. S. London. M. 1926. Social and Analysis and Survey Archaeol Moore, Rostovtzeff, -. b. Roman ogy: The Sphakia Survey,"BSA 98, Forthcoming "Regional Economic History of the and the Boom-and-Bust Oxford. pp. 37-105. Survey Empire, the M. 1994. Moore, M. G. 2000. "'Surveying Epi Countryside: Rereading Archae Rothaus, R. "Urban Space, rote and for and Villas in Pottery: Ceramics, Cuisine, ological Evidence Episodic Agricultural Space, in the Late Roman in Structures Social History in Southern Epirus, Abandonment Roman Late Corinth," b.c -a.d. International rurales et soci?t?s P. N. Greece, 300 500" (diss. Corinthia," Journal of antiques, ed. Boston Univ.). Historical Archaeology. Doukellis and L. G. Mendoni, -. -. In 2001. "Roman and Late An prep. "Surveying the Isth Paris, pp. 391-396. of mus: Patterns of Settlement in the -. 2000. First tique Pottery Southern Epirus: Corinth, the City Roman-Late Roman Corinthia." Greece: An Urban Late Some Results of the Nikopolis of History of in Foundation and and P. Roberts. 2003. Cult and Survey Project," Pickersgill, C, Antique Religion (Religions Destruction: and North "New on Roman in the Graeco-Roman World Nikopolis Light Sparta: 139), western from the Greece, ed. J. Isager, Athens. Roman Pottery Sparta Leiden. in BSA Moore, R. S. 2000. "Trade the East Theatre and Stoa," 98, Roy, J., J. A. Lloyd, and E. J. Owens. ern A.D.: 1989. under the Ro Mediterranean, 100-700 pp. 549-597. "Megalopolis 1998. at man in Renais The Ceramic Evidence" (diss.Ohio Poulter, A. G. "Field Survey Empire," The Greek BSA 463-511. sance in the Roman State Univ.). Louloudies," 93, pp. Empire (BICS L. 2000. "The Coun Munn, M. H., and M. Z. Munn. Rautman, M. Busy Suppl. 55), ed. S.Walker and Roman 1990. "On the Frontiers of Attica tryside of Late Cyprus," A. Cameron, London, pp. 146 and Boiotia: The Results of the RDAC 2000, pp. 317-331. 150. -. Late and Stanford Skourta Plain Project," 2003.ACypriot Village of Roy, J., E. J. Owens, J. A. Lloyd. in in the in the in Essays Topography, History, Antiquity: Kalavasos-Kopetra 1988. "Tribe and Polis the Chora and Culture Vasilikos R.I. at in of Boiotia (Teiresias Valley, Portsmouth, Megalopolis: Changes Settle P. 1995. Trade in the Western ment in to Suppl. 3), ed. A. Schachter, Mon Reynolds, Pattern Relation Synoe treal, pp. 33-40. Mediterranean, A.D. 400-700: The cism," npaKxix? rov XII Aiedvov? Munn, M. L. Z. 1985. "A Late Ro CeramicEvidence (BAR-IS 604), Ivve?piov KXao?Kri? ApxawXoyia?, man Kiln Site in theHermionid, Oxford. AOijva, 4-10 ZexTSju?piov 1983, Greece," AJA 89, pp. 342-343 Rife, J. L., M. M. Morison, A. Barbet, vol. 4, Athens, pp. 179-182. R. K. D. H. and B. 1983. on (abstract). Dunn, Ubelaker, Rutter, J. "Some Thoughts 1991. F. Monier. 2007. of Papadopoulos, J. K. "Roman "Life and Death the Analysis Ceramic Data Gen at a in Roman Amphorae from the Excavations Port Greece: The erated by Site Surveys," in Archaeo 67 Kenchreai 2002 in theMediterranean ztTorone;ArchEph 1989, pp. Cemetery Project, logical Survey 103. 143-181. 2006," Hesperia 76, pp. Area (BAR-IS 155), ed.D. R. Keller H. 2000. "The Current State D. 137 Patterson, Riley, J.A. 1976. "The Carthage Sys and W. Rupp, Oxford, pp. of EarlyMedieval andMedieval tem for the Quantification of Pot 142. Ceramic Studies inMediterranean in Excavations at tery," Carthage Sanders, G. D. R. 1987. "An Assem in Conducted the of Frankish at Cor Survey," Francovich, Patterson, 1975, by University blage Pottery and Barker 110-120. ed. H. 2000, pp. ofMichigan, J. Humphrey, inth," Hesperia 56, pp. 159-195. D. -. at Petropoulos, M., and A. Rizakis. Tunis, pp. 125-156. 1993. "Excavations Sparta: 1994. -. "Settlement Patterns and 1979. "Coarse Pottery," in The Roman Stoa, 1988-91: Pre in the at Sidi Landscape Coastal Area of Excavations Khrebish Ben liminary Report, Part 1:Medieval Patras: Preliminary Report," JRA 7, ghazi (Berenice) II, ed. J. A. Lloyd, Pottery," BSA 88, pp. 251-293. -. pp. 183-207. Tripoli, pp. 91-467. 1999. "A Late Roman Bath at D. K. 2001. the A. 2001. and Pettegrew, "Chasing Robinson, B. "Fountains Corinth: Excavations in the Panayia Classical Farmstead: the the Culture ofWater at Roman Assessing Field, 1995-1996," Hesperia 68, Formation and of Rural Univ. Signature Corinth" (diss. of Pennsyl pp. 441-480. Settlement in Greek -. 2003. Landscape vania). "Recent Developments b.c. Archaeology,"/M4 14, pp. 189-209. Romano, D. G. 1993. "Post-146 in the Chronology of Byzantine -. on 2006. "Corinth the Isth Land Use in Corinth and Planning Corinth," in Corinth XX, pp. 385 mus: an 44 Studies of the End of of the Roman Colony of b.c.," in 399. Ancient Ohio in Roman -. 2004. in Landscape" (diss. The Corinthia the Period "Problems Interpret State Univ.). (JRA Suppl. 8), ed.T. E. Gregory, ing Rural and Urban Settlement in -. a. "The End 9-30. a.d. Forthcoming of Ann Arbor, pp. Southern Greece, 365-700," Ancient Corinth? Views from the -. 2003. Centu in Rural Evo "City Planning, Landscapes of Change: in and His Landscape," Archaeology riation, and Land Division in Ro lutions inLate Antiquity and the inMedieval and Post-Medieval man Corinth: Colonia Laus Iulia Middle N. tory Early Ages, ed. Christie, on Greece: Studies Method and Mean Corinthiensis and Colonia Iulia Aldershot, pp. 163-193. 784 DAVID K. PETTEGREW

-. and Mantzourani. 1991. -. Production 2005. "Archaeological Evi Davis, E. 2000b. "Specialized "Gazetteer and in Late dence for Early Christianity and the of Archaeological Sites," Exchange," Antiquity: in and and a.d. 425?600 End of Hellenic Religion in Cor Cherry, Davis, Mantzourani Empire Successors, in Urban in Roman 69-162. ed. A. inth," Religion 1991, pp. (CAHXLV), Cameron, T. T. E. D. B. and M. Corinth: Interdisciplinary Approaches, Tartaron, F, Gregory, J. Ward-Perkins, Whitby, ed. D. N. Schowalter and S. J. Pullen, J. S. Noller, R. M. Rothaus, Cambridge, pp. 346-391. L. R. L. -. 2001. Friesen, Cambridge, Mass., J. Rife, L. Diacopoulos, "Specialisation, Trade, pp. 419-442. Schon, W. R. Caraher, D. K. Pette and Prosperity: An Overview of and D. Nakassis. 2006. "The of the Late Sarris, P. 2004. "Rehabilitating the grew, the Economy Antique in Great Estate: Aristocratic Prop Eastern Korinthia Archaeological Eastern Mediterranean," Kings a erty and Economic Growth in the Survey: Integrated Methods for ley and Decker 2001a, pp. 167 Late in 178. Antique East," Bowden, Dynamic Landscape," Hesperia 75, Lavan, and Machado 2004, pp. 435-505. Whittaker, C. R. 1983. "Late Roman pp. 55-71. Terrenato, N. 2000. "The Visibility of Trade and Traders," in Trade in the ed. Schofield, A. J. [1989] 2000. "Under Sites and the Interpretation of Field Ancient Economy, P. Garnsey, Results: Towards an K. and C. R. standing Early Medieval Pottery Survey Analysis Hopkins, Whittaker, in 163-180. Distributions: Cautionary Tales and of Incomplete Distributions," London, pp. and Barker C. 1988. Sherlock Their Implications for Further Francovich, Patterson, Wickham, "Marx, 460 60-71. and Late Roman Com Research," Antiquity 63, pp. 2000, pp. Holmes, in van and C. Runnels. 183-193. 470, repr. Landscapes from Antiq Andel, TH., merce,"//^ 78, pp. S. 1987. the A Rural -. 1998. "Overview: uity, ed. Stoddart, Cambridge, Beyond Acropolis: Production, in pp. 109-119. Greek Past, Stanford. Distribution and Demand," The Distribu Shipley, G. 2002. "The Survey Area in Van de Velde, P. 2001. "An Extensive Sixth Century: Production, to the Hellenistic and Roman Periods," Alternative Intensive Survey: tion, and Demand, ed. R. Hodges in Laconia in the Riu Mannu and W. 279 The Survey: Continuity Point Sampling Bowden, Leiden, pp. and in a Greek Rural Land 292. Change Survey Project, Sardinia,"/M/f 14, 1: and 24-52. -. 2005. the scape Methodology Interpreta pp. Framing Early tion ed.W. Cava 1998. "Medieval and Post Middle and theMediter (BSA Suppl. 26), Vroom, J. Ages: Europe V. Medieval from a Site in Oxford. nagh, J. Crouwel, R. W. Catling, Pottery ranean, 400-800, 257 Boeotia: A Case T. 1982. "The Definition and G. Shipley, London, pp. Study Example Wilkinson, J. in Means 337. of Post-Classical Archaeology of Ancient Manured Zones by Slane, K. W. 2000. "East-West Trade Greece," BSA 93, pp. 513-546. of Extensive S herd-Sampling Tech -. Ceramics 223-233. in Fine Wares and Commodities: 2003. After Antiquity: niques,"/^ 9, pp. and in the 7th 1979. "Corinth and Rome The View from Corinth," Rei Cre Society Aegean from the Wiseman, J. to A.c.: A I: 228 b.c-a.d. tariae Romanae FautorumActa 36, the 20th Century Case Study 267,"^V7WII.7.1, Central Leiden. 438-548. pp. 299-312. from Boeotia, Greece, pp. -. -. "Late F. L. 2003. "Corinth's Roman Pot 2004. Antique Pottery, Wright, J.C, J. Cherry, J. Davis, and Trade in the East E. S. B. and tery: Quantification and Meaning," Settlement, Mantzourani, Sutton, ern 1990. "The Nemea in Corinth XX, pp. 321-335. Mediterranean: A Preliminary R. F. Sutton Jr. of Ceramics from A Slane, K. W., and G. D. R. Sanders. Comparison Valley Archaeological Project: and in 2005. "Late Roman Horizons Limyra (Lycia) Boeotia," Preliminary Report," Hesperia 59, at and Machado 579-659. Established Corinth," Hesperia Bowden, Lavan, pp. M. E. B. 74, pp. 243-297. 2004, pp. 281-331. Zangger, E., Timpson, S. F. Knauss. Sutton, R. F., Jr. 1991. "Ceramic Evi Ward-Perkins, B. 2000a. "Land, La Yazvenko, Kuhnke, and J. Archae dence for Settlement and Land Use bour, and Settlement," in Late An 1997. "The Pylos Regional to Hellenistic and a.D. 425 Part II: in the Geometric tiquity: Empire Successors, ological Project, Landscape in and 600 A. Evolution and Site Periods," Cherry, Davis, (CAHX1V), ed. Cameron, Preservation," 549-641. Mantzourani 1991, pp. 245-263. B. Ward-Perkins, and M. Whitby, Hesperia 66, pp. 315-345. Sutton, R. F., Jr., J. F Cherry, J. L. Cambridge, pp.

David K. Pettegrew Messiah College department of history

P.O. BOX 3051 ONE COLLEGE AVENUE

GRANTHAM, PENNSYLVANIA 1J02J

[email protected]