caraher pettegrew james vayiaoffprint 2010

8/8/2019 Caraher Pettegrew James VayiaOffprint 2010

http://slidepdf.com/reader/full/caraher-pettegrew-james-vayiaoffprint-2010 1/33

dining in t he san ctua ry of de me ter and kore 1

Volum e 7 920 1 0

Copyright © The American School o Classical Studies at Athens, originally published in Hesperia 79 (2010), pp. 385–415. This oprint is supplied orpersonal, non-commercial use only. The defnitive electronic version o thearticle can be ound at <http://dx.doi.org/10.2972/hesp.79.3.385>.

Hesperia

The Jour nal of the Am eric an School

of Classical Studies at Athens



hesperia

Tracey Cullen,Editor

Editorial Advisory Board

Carla M. Antonaccio, Duke University

Angelos Chaniotis, Oxford University Jack L. Davis, American School of Classical Studies at Athens

A. A. Donohue, Bryn Mawr College Jan Driessen, Université Catholique de Louvain

Marian H. Feldman, University of California, BerkeleyGloria Ferrari Pinney, Harvard University

Sherry C. Fox, American School of Classical Studies at Athens Thomas W. Gallant, University of California, San DiegoSharon E. J. Gerstel, University of California, Los Angeles

Guy M. Hedreen, Williams College Carol C. Mattusch, George Mason University

Alexander Mazarakis Ainian, University of Thessaly at VolosLisa C. Nevett, University of Michigan Josiah Ober, Stanford University

John K. Papadopoulos, University of California, Los Angeles Jeremy B. Rutter, Dartmouth College

A. J. S. Spaworth, Newcastle UniversityMonika Trümper, University of North Carolina at Chapel Hill

Hesperia is published quarterly by the American School o Classical Studies atAthens. Founded in 1932 to publish the work o the American School, the jour-nal now welcomes submissions rom all scholars working in the felds o Greek archaeology, art, epigraphy, history, materials science, ethnography, and literature,

rom earliest prehistoric times onward. Hesperia is a reereed journal, indexed in Abstracts in Anthropology, L’Année philologique, Art Index, Arts and HumanitiesCitation Index, Avery Index to Architectural Periodicals, Current Contents, IBZ:

Internationale Bibliographie der geistes- und sozialwissenschaftlichen Zeitschriften-literatur, Numismatic Literature, Periodicals Contents Index, Russian Academy of Sciences Bibliographies, and TOCS-IN . The journal is also a member o CrossRe.

The American School o Classical Studies at Athens is a research and teachinginstitution dedicated to the advanced study o the archaeology, art, history,philosophy, language, and literature o Greece and the Greek world. Establishedin 1881 by a consortium o nine American universities, the School now servesgraduate students and scholars rom more than 180 afliated colleges and uni-

versities, acting as a base or research and study in Greece. As part o its mission,the School directs ongoing excavations in the Athenian Agora and at Corinthand sponsors all other American-led excavations and surveys on Greek soil. Itis the ofcial link between American archaeologists and classicists and the Ar-chaeological Service o the Greek Ministry o Culture and, as such, is dedicatedto the wise management o cultural resources and to the dissemination o knowl-edge o the classical world. Inquiries about programs or membership in theSchool should be sent to the American School o Classical Studies at Athens,6–8 Charlton Street, Princeton, New Jersey 08540-5232.



w. r . carahe r, d . k . pet tegre w, and s . james 386

o the territory near Sophiko and its harbor, Korphos.2 There has beencomparatively less work on the rugged coastline stretching rom the

Isthmus to the bay o Frangolimano (Fig. 1). The neglect o this area isunderstandable. Most scholars have argued that the main ancient roadthrough the southeastern Corinthia bypassed this region, proceeding southto the Epidauria and Argolid along the inland route through the modern

villages o Galataki and Rhyto beore emerging in the ertile plain nearSophiko (Fig. 1).3 Even today the area near the village o Katakali is ruggedcountry, relatively undeveloped, and lacking in paved roads. The diculty in accessing this area, compared to the well-trod Isthmus, has contributedto the relative neglect o its topography and antiquities.

In 2001 and 2003, members o the Eastern Korinthia Archaeological Sur- vey (EKAS) conducted eldwork in this region, concentrating their activitiesnear Lychnari Bay and the peninsula o Vayia (Fig. 2).4 Although the main goal

o this work was to document an Early Bronze Age site on the Vayia penin-sula,5 EKAS also conducted an extensive survey o the surrounding territory.

The survey revealed three signicant, undocumented sites with preservedarchitecture o Late Classical to Hellenistic date, which we have called Lych-nari Tower, Ano Vayia, and Kato Vayia. With the encouragement o thedirectors o EKAS, Timothy Gregory and Daniel Pullen, and a study permitprovided by the Greek Ministry o Culture, we conducted a short study sea-son in 2008 to complete the documentation o the remains at these sites.

Fge 1. Ma he heaecha, hg mde adae e, ad maj e.

W. R. Caraher

2. For the Isthmus, see Tartaronet al. 2006; Caraher, Nakassis, andPettegrew 2006. For the area aroundKorphos and Sophiko, see Dixon 2000;

Tartaron et al., orthcoming.3. Fowler 1932, pp. 99–101; Wise-

man 1978, p. 127; Dixon 2000, pp. 61–62.

4. See Tartaron et al. 2006 or anoverview o the EKAS aims andmethods.

5. Tartaron, Pullen, and Noller2006.



tow er s a nd forti f i c ati ons at vay i a 387

In this article, we seek to place these three newly discovered ruralinstallations into their local context, with the principal aim o providingspecic inormation on the Classical and Hellenistic landscape o theeastern Corinthia. A careul reading o the local topography, the ceramicartiacts associated with the sites, and other ortied sites in the Corinthiaprovides evidence or the military unction o these buildings. Moreover, the

sites near Lychnari Bay reinorce this stretch o the Corinthian countrysideas a productive and strategically important coastal environment and a sig-nicant corridor or regional communication. In addition to highlightingthe signicance o this micro-region or the Corinthian polis, we also aimto contribute in a small way to the broader discussion o the unction o rural towers and associated installations in antiquity.

topoGrAp Hy o F tHE VAyiA A rEA

The Vayia region occupies a key place within communication and travel networks between Corinth’s proximate chora on the Isthmus and its more

distant southern coastal territory. The rocky spine o Mt. Oneion ormsthe dramatic southern boundary to the fat plain o the Corinthian Isthmus(Fig. 3), stretching rom the imposing rock o Acrocorinth to the harbortown o Kenchreai in the east. To move south by land along the easterncoast o the Corinthia, avoiding both the city o Corinth and the ortica-tions near Kenchreai, would have required crossing Mt. Oneion throughthe Maritsa pass, ortied during the Late Classical period.6 Once south o the mountain ridge, there were several routes through the rugged country

Fge 2. Ma Lha ba adhe Vaa ela. W. R. Caraher

6. Stroud 1971a; Caraher andGregory 2006.




o the southeastern Corinthia that provided access to cultivable valleysand unortied settlements, roads west into the Argolid and south into theEpidauria, and several natural embayments.

The bay o Lychnari is one o the best natural inlets on the jaggedcoast o the eastern Corinthia (Fig. 4; c. Fig. 2). While no evidence orancient harbor works has been ound there, its sheltered aspect and fatbeach would have been well suited or ancient ships. 7 The peninsulaknown as Vayia shields the small bay rom the east, and the rocky hilltopo Lychnari protects the bay below rom the western wind. Lychnari Bay opens inland onto a broad valley bounded to the north and east by thecoastal ridge and to the south by the abrupt mountains o the southeasternCorinthian interior. The valley bottom (Fig. 5) provides relatively easy passage rom the vicinity o Lychnari Bay through the nearby village o Katakali northwestward to the low hills south o Oneion, the villages o Kato Almyri, Loutro Elenis, and Galataki, and the ancient settlement o Solygeia (Fig. 1). Continuing north and passing to the east o the low hillo Stanotopi, the countryside opens onto the Isthmus o Corinth and theharbor town o Kenchreai.

Immediately to the east o Lychnari Bay, a small, pebbly beach sits atthe mouth o the Vayia River, a seasonal torrent that cuts deeply throughthe coastal ridge as it descends rom the mountains o the central Corin-

thia (Fig. 6; c. Fig. 2). Walking inland rom this beach, it is easy to reachLychnari Bay by ollowing a corridor south o Ano Vayia.8 Turning to theeast (Fig. 6), an ascent up the steep but not unmanageable bank o the VayiaRiver aords access to a high pass (Fig. 7) that runs below and to the southo the coastal height o Kaki Rachi/Babouri. This pass leads eastward tothe bay o Frangolimano, rom which a traveler can proceed inland, pastthe ortied Classical site o Ayia Paraskevi and onward toward the valley o Sophiko and the Epidauria beyond.9

7. Wiseman 1978, p. 132.8. This hill, like the coastal penin-

sula and the small bay, is locally reerredto as Vayia.

9. Wiseman 1978, pp. 127–128;Dixon 2000, pp. 68–70.

Fge 3. M. oe, veed mhe h (he ihm), hg heMaa a. Photo D. K. Pettegrew




Fge 4. Lha ba ad he Vaa ela, veed m he e (Lha e). Photo K. R. Pettegrew

Fge 5. Lad e oe ad

he ihm, veed m he ea (A Vaa). Photo W. R. Caraher

This route to the southeast rom Lychnari Bay is suggested by morethan the topography alone: there are stretches o a narrow built pathascending the eastern side o the Vayia River valley toward a high valley immediately to the south o the hill o Kaki Rachi (see Fig. 7). Today this high valley is thoroughly terraced, and olive trees continue to be cul-tivated. A cinderblock eld house shares the valley with two abandonedlong houses in the advanced stages o collapse. Proceeding east throughthe pass, the path continues along its northern side where it cuts into the




Fge 6. Ma hg he a ha m Vaa ba ad Fag-lma ba. W. R. Caraher

Fge 7. pa ad Kak rah/ba ad Faglma ba,

veed m he e. Photo W. R.Caraher




10. In recent times, resin collectorsused this path, and many o the pinetrees show scars rom this activity. Theremains o a roughly built stone basinor collecting resin indicate that thispath was used in the early 20th century,i not beore.

11. Peppas 1990, pp. 239–241; 1993,p. 136, plan 2. The modest enceinteconsists o drystone walls 1–1.5 m

pine-covered, southern/inland slopes o the coastal ridge. The course o the path itsel rarely exceeds a gentle slope and gradually descends towardthe large bay at Frangolimano, keeping largely to the lower slopes o thecoastal ridgeline.10 Despite the densely wooded and rugged appearance o the countryside, the pass is easy to traverse, and one can walk rom Vayia toFrangolimano in a little over an hour. It is worth noting that Peppas has

identied a small rubble ortication o medieval date on the southernslope o Kaki Rachi, which would have been well suited to block move-ment along this route.11

Ano VAyiA

The most extensive remains in the region o Lychnari and Vayia stand atopthe hill (156 masl) that we have called Ano Vayia to distinguish it rom thesite o Kato Vayia below. The site consists o a rectangular complex, whichis oriented north–south and constructed o rough polygonal masonry, and a

circular tower (Fig. 8). The most imposing eature o the rectangular com-plex is its western wall (Fig. 9), which is over a meter wide and consists o two aces with a rough cobble core. The wall is preserved in three coursesand stands to a height o 1.20 m. The largest stones in this ace exceed ameter in length and show signs o having been worked to t snugly withtheir neighbors; the inner ace o the wall is largely obscured by the tumbleo the building, but it was apparently built o smaller stones. In several placesalong the course o the wall, it is clear that the builders cut back bedrock to orm a solid base or the building and, in some cases, even incorporatedbedrock outcrops into the lower courses o the walls themselves.

This style o rough polygonal construction is common to rural struc-tures in the Corinthia. We nd similar masonry at Kephalari station,

the site o Are Bartze, the towers at the Hill o the Windmills, and inthe substantial walls at the site o Ayia Paraskevi.12 We can contrast thistype o wall construction with the technique used at the square Stanotopitower on the eastern end o the Oneion ridge, where squared blocks arearranged in more or less regular courses.13 The rough quarry-aced blockso Stanotopi have more in common with the careul ashlar constructionused in towers in Attica, the Megarid, and the Aegean islands, and rep-resent a more rened technique than that seen at Ano Vayia.14

At the western wall’s midway point, there is a break o slightly over2.0 m where the bedrock was clearly trimmed back to create an entranceto an east–west corridor between the northern and southern parts o the

wide. Peppas’s proposal o a medievaldate or the walls, however, must betreated with some caution, as it appearsto be based almost exclusively on thedrystone construction style. Elsewherein the Corinthia, it is possible to datesimilar drystone walls to the Classical–Hellenistic period on the basis o asso-ciated ceramic evidence; we are notsuggesting that the walls documented

by Peppas are necessarily ancient indate, but merely noting that possibility.

We were unable to locate Peppas’sortication in 2008.

12. Wiseman 1978, pp. 128–129;Dixon 2000, pp. 64–65; Dixon 2005.

13. Stroud 1971a, pp. 129–133.14. See examples in Morris and

Papadopoulos 2005, pp. 157–180; Young 1956a.




building (Fig. 8). The structure north o the corridor is in the orm o anirregular rectangle with its north wall running 8.0 m northeastward at anobtuse angle rom the main western wall to take advantage o a naturalbedrock terrace and several substantial bedrock outcrops. The northern

wall o the northern structure averages 0.75 m in width and is slightly narrower than the western wall. At a point 4.20 m along its length, thenorth wall is joined by another north–south wall running roughly paral-

lel to the western wall o the complex, orming the eastern side o thestructure. The only clear evidence or a south wall to this structure is apoorly preserved and simple partition wall (7.60 m rom the north wall)that could not have borne signicant weight. At the southern end o theeast wall, a narrower (0.52 m) and rougher wall runs east, toward thetower, or slightly over 5.60 m. Unlike other walls at the site, this wall isconstructed o roughly stacked eldstones. The inormal construction style

Fge 8. pla he claal–Helle fa a AVaa. W. R. Caraher and D. K. Pettegrew




and the absence o larger stones allow the possibility that it represents alater phase in the use o the site.15

In contrast to the northern structure, the southern structure appears tobe more regular in plan. Its southern wall orms a neat right angle with themain north–south wall o the building. While its eastern wall is obscuredin tumble and vegetation, enough o its course is visible to indicate that it

was substantial and well dened. Only the northern wall o the southernstructure maniests the same shoddy construction technique seen in the

partition wall and eastern wall o the northern structure. Its width variesrom less than 0.60 m to 1.0 m. The narrower, western parts o the walldo not preserve any rubble core, which could indicate a later phase o rebuilding or modication.

The corridor separating the southern and northern structures runseastward to the oundation o a round tower with careully coursed stonespreserved with roughly cut, curved proles on their outer aces (Figs. 8,10).16 From the visible remains, it appears that only the lowest and outercourse o the tower remains in situ, suggesting a structure with a diametero 6.20 m. On the tower’s western side, a second course o stones may bepreserved, but generally the upper courses o the tower have scattered downthe steep northern and eastern slopes o the Ano Vayia hill. The round

tower is clearly a component o the rest o the compound, but the exactarchitectural relationship is unclear. The poor quality o the constructiono the eastern wall projecting rom the north structure makes it dicultto determine whether this wall should be understood as belonging to thesame phase as the tower and north–south compound, or as a later wallthat may have served a purpose entirely unrelated to the main phase o construction on the site.

15. Young (1956b, pp. 124–126)describes a similar wall at the Cli

Tower near Sounion.16. The masonry o the tower’s

oundation is similar to the roughpolygonal style o the building to the

west, but it is more careully coursed.

Fge 9. se he ee all a A Vaa, veed m he h-

e. Photo K. R. Pettegrew




Intensive Survey

In conjunction with the initial mapping o the architectural remains at thesite in 2003, members o the EKAS project conducted a small-scale inten-sive pedestrian survey o the Ano Vayia hill (Fig. 11). The goal o thissurvey was to sample material rom the hill, to determine the extent o the site, and to produce a data set comparable with that collected rom themain EKAS transect on the Isthmus. The last goal required that we con-

duct our survey o the hill using basically the same technique that EKASemployed elsewhere in the survey area. As we have analyzed many o theadvantages and limitations o this method elsewhere,17 we will includehere only a summary o the methods employed and ocus instead on theresults o this survey.

The most signicant obstacle to conducting survey around the site wasthe dense vegetation covering the entire hill. Pine trees with low branches,

Fge 10. the e a A Vaa, veed m he hea.Photo W. R. Caraher

17. Caraher, Nakassis, and Pette-grew 2006; Tartaron et al. 2006.




in particular, made it impractical, i not impossible, to survey the entire hill-side, so we decided to ocus our eorts on three transects descending the

slopes to the north, west, and south sides o the hill; the eastern slope wastoo steep to survey. Consistent with our procedure elsewhere, the survey onAno Vayia involved eldwalkers at 10-m intervals examining the surace1.0 m to either side o their swath through the unit. The walkers countedevery artiact and collected artiacts in accordance with the principles o thechronotype sampling method. This method dictated that the eldwalkershould collect one example o each unique type o artiact. In practice, thismeans that a walker could collect one example o a rim, handle, base, andbody sherd o each abric or surace treatment present in their 2-m-wideswath through the unit. This method ensured that we would produce atleast one example o each type o artiact present in the unit, and it alsoprovided an inormal indicator o the requency or density o particular

types o artiacts present in a unit because the walkers could collect as many as ve examples o a single vessel type.18

To make our sample a bit more robust than the typical EKAS survey transect, we surveyed units that were slightly smaller (1,300 m2) than thetypical EKAS unit (median size: 2,100 m2). Smaller units also suited thegeological and topographical complexity o the environment. We shouldnote that we did not systematically survey the buildings on the site, but

Fge 11. Ma hg he de aa ad A Vaa.Contour interval 4 m. W. R. Caraher

18. For discussions o the chrono-type system within the context o EKAS, see Tartaron et al. 2006,pp. 457–465; Caraher, Nakassis, andPettegrew 2006, pp. 11–13; Pettegrew 2007, p. 752.




instead collected grab samples o diagnostic artiacts visible amid the tumble without disturbing the basic arrangement o the allen stones.

It is perhaps unsurprising that the densest concentration o materialoccurred around the architecture at the top o the hill and that artiactdensities declined dramatically urther down the slope. The units imme-diately adjacent to the collapsed buildings showed artiact densities o nearly

2,000 artiacts per hectare; this number is comparable to the generally highartiact density documented by EKAS across the busy Corinthian Isth-mus.19 Relatively poor surace visibility and hillslope erosion may partly account or the declining artiact densities on the slopes, but there was noevidence or ancient or modern construction on the slopes aside rom severalmodern terraces. Despite the diculties encountered in this environmentand the relatively coarse resolution o our survey, it is clear that the materialin the immediate vicinity o the collapsed buildings represents a distinctand localized phenomenon in the landscape.

Artifacts and Distributional Data

Systematic survey o units around Ano Vayia using the chronotype systemproduced a total assemblage o 90 artiacts, which consisted largely o pot-tery and tile (96%); three obsidian bladelets and a piece o medieval–modernglass were the only nonceramic artiacts noted in the survey. Approximately 75.6% (n = 68) o the artiacts date specically to the Classical–Hellenisticperiod; these artiacts consist mainly o ragments o coarse utilitarian ves-sels, storage jars, and pithoi (Fig. 12; see also Figs. 13, 14, below). Classical–Hellenistic coarse and medium-coarse pottery accounts or 31.1% o thetotal survey assemblage (n = 28: 2 rims, 5 handles, 21 body sherds); someo these sherds belong to amphoras such as Corinthian A and B. Pithosragments (n = 26) constitute 28.9% o the total artiact count and 38.2%

o Classical–Hellenistic artiacts.Other ceramic classes o Classical–Hellenistic date are present in small

numbers: kitchenware (n = 6) and roo tiles (n = 6) each make up 6.7% o the total artiact count and 8.8% o the Classical–Hellenistic material,

while two sherds (2.9% o Classical–Hellenistic) were identied as seminetableware. Overall, the Classical–Hellenistic assemblage is predominantly coarse material that originated rom storage vessels and various utilitarianshapes, with small quantities o kitchenware, tiles, and ne ware.

Besides Classical–Hellenistic pottery, the survey also recorded a smallpercentage (4.4%) o sherds dating to either the Archaic–Classical or thebroader Archaic–Hellenistic period,20 which represent either an earlierphase at the site or, more probably, were let by the same inhabitants who

deposited the Classical–Hellenistic sherds discussed above.21 In addition,although it is impossible to know with certainty, several pieces (n = 7) o medium-coarse ware may also derive rom the same occupational phase;the material is clearly ancient, but otherwise undiagnostic. The survey units also produced a small number (n = 6) o medieval and modern arti-acts that accounted or 6.7% o all artiacts analyzed rom the systematicsurvey.

19. See Caraher, Nakassis, andPettegrew 2006.

20. From the systematic survey,these include an Archaic–Classicalamphora, Corinthian A amphora,pithos ragment, and medium-coarse

ware.21. Since these sherds could date to

either the Archaic or Classical period,it is impossible to narrow down the

identication one way or another.As discussed in the artiact cataloguebelow, however, none o the materialcollected looks specically Archaic, andsince the principal signature at the siteis Classical–Hellenistic, it is reasonableto iner that these sherds date to theClassical period.




In order to supplement the systematic collection o artiacts romsurvey units on the hillside around the structure, we collected grab sampleso diagnostic artiacts visible in the architectural collapse itsel and in theimmediate vicinity o the buildings. This second sample was biased towardeature sherds and diagnostic artiacts, with the aim o gathering as muchinormation as possible about the unction and date o the buildings. The as-semblage o 61 artiacts collected through grab sampling largely mirrors thatcollected rom the systematic survey units. The majority o material datesto the Classical–Hellenistic span (65.6%, n = 40), with meager amounts o Archaic–Classical sherds (14.8%, n = 9), a ew ancient coarse-ware sherdsand tiles (13.1%, n = 8), and a single ragment o a hopper mill (probably o

Late Classical–Hellenistic date; see below). Unlike substantial rural domes-tic structures o Classical–Hellenistic date elsewhere, Ano Vayia producedonly a small number (4.9%, n = 3) o later artiacts in grab samples.22

The unctional characteristics o the Classical–Hellenistic ceramicartiacts (n = 40) generally refect the Classical–Hellenistic survey assem-blage: a large number o pithos sherds (32.5%, n = 13) and medium-coarsesherds (22.5%, n = 9), and very small amounts o kitchenware (5.0%, n = 2);no tablewares in ne or semine abrics were collected in the grab sample.

The grab samples conrm the picture o an overall assemblage consistingmainly o storage or utility wares such as pithoi and transport amphoras.

The major dierence between the Classical–Hellenistic assemblagesproduced by the systematic chronotype collection and the grab sample col-

lection is that roo tiles constitute the most requent artiact class (40%,n = 16) o the latter but only 8.8% o the ormer. The relatively larger num-ber o Classical–Hellenistic tiles relates not to sampling method but to theconcentration o tiles in and around the collapsed building itsel. The tilescollected as grabs (29.5% o total grabs) include 16 plain and painted Laco-nian tiles o Classical–Hellenistic date and two unslipped Corinthian tiles,datable by abric to the Archaic–Hellenistic period but clearly associated

Fge 12. tal aemlage lead ae ae m A Vaa.Photo K. R. Pettegrew

22. The Archaic–Classical pottery collected as grab samples includes oneCorinthian A amphora handle andeight pithos sherds. Later post-Classi-cal material includes a Roman amphorasherd, a Late Roman stewpot ragment,and a Late Medieval coarse-ware sherd.




with the building. While 18 tile ragments are not very many or a build-ing o this size, their presence at least demonstrates that the structure wasrooed. The dearth o roo tiles in and around the structure refects eitherthe stripping o tiles rom the building during or ater abandonment, 23

or that they still lie buried beneath the rubble debris. The appearance o Laconian and Corinthian tiles together suggests that there might have

been multiple phases o construction or that the building was erected lessin accordance with aesthetics and more in line with practical concerns.

Interpretation of Artifacts and Architecture

The relatively inormal construction style o the structures on Ano Vayiasuggests that these buildings represented a less substantial investment inthe landscape than one might expect or a place o long-term occupationin use or a generation or more. The assemblage collected rom the struc-ture at Ano Vayia, as well as rom the surrounding survey units, comple-ments this interpretation in several ways. First, the ceramic material dates

primarily to a single period (Classical–Hellenistic), and the relatively ew pieces o Archaic–Hellenistic pottery can probably be associated with thesame episode o occupation. Evidence or later use o the site in the LateRoman, Early Medieval, Late Medieval, and Early Modern eras is scantand suggests occasional visits to the area, not episodes o reurbishmentand reoccupation. In sum, the architecture at Ano Vayia represents a singleperiod o occupation sometime in the Classical–Hellenistic era that let adiscrete concentration o material in the coastal landscape.

Second, the artiact assemblage associated with the building is pri-marily utilitarian in nature. The predominance o ragments o pithoi andamphoras suggests that storage was a priority at the site, perhaps in orderto compensate or the apparent lack o cisterns at the top o the hill. The

presence o cooking wares and a small hopper mill ragment indicatesactivities related to ood preparation. Fine wares are ew, represented only in the survey sample by two body sherds o semine abric. The picture wehave rom ceramic artiacts, then, is one o low-intensity occupation thatlet comparatively homogeneous debris in the landscape.

The ceramic artiacts suggest that Ano Vayia was a habitation sitealong the Corinthian coast, but that the occupation was neither intensivenor o long duration. Ano Vayia did not produce the kind o basic ceramicassemblage that has come to be associated with rural armsteads in theGreek world, with their recognizable quantity and diversity o amphoras,kitchenwares, and ne wares.24 Nor is the ceramic signature at this site

23. Pettegrew 2001, pp. 196–202.24. See Pettegrew 2001 or a sum-

mary o scholarly discussion aboutarmsteads, problems o denition, andthe typical rural assemblage, and Pette-grew 2002, or the argument that arti-act scatters representing “armsteads”minimally produce a varied “package”o artiact types such as kitchenware,

amphoras, and ne ware. Identiyingan ancient rural building as a arm-stead will, o course, always be prob-lematic; the character o artiact as-semblages says a great deal more aboutthe intensity and investment o landuse and occupation over time. For ad-ditional discussion o the importanceo ne ware and kitchenware in de-

ning rural armsteads, see White-law 1998, and Bintli, Howard, andSnodgrass 2007, pp. 39–42, where theauthors observe that Archaic–Classicalarmsteads in Kea and Boiotia haveassemblages with over 50% ne ware,and also high proportions o kitchen-

wares.




consistent with typical domestic contexts in the Classical period, whichshow substantial quantities o ne-ware vessels and drinking cups in bothurban and rural settings.25 The paucity o ne ware and kitchenware at AnoVayia suggests instead limited rates o ceramic discard as well as a meagersystemic assemblage o pots and household wares.26 In short, the ceramicmaterial makes Ano Vayia look much more like an occasionally occupied

rural building than a typical amily arm.Archaeological investigations in the broader Corinthia can provide

useul points o comparison. The typical surace signature o Classical–Hellenistic habitation documented by EKAS, which ocused on the well-inhabited Isthmus, consisted o abundant pithoi, amphoras, storage vessels,kitchenwares, and painted roo tiles; ne wares were especially numerous.

The more homogeneous assemblage at Ano Vayia nds local parallels withassemblages rom some o the small orts and deensive outposts o Classi-cal–Hellenistic date in Corinthian territory. Wiseman’s survey o the ort o Mt. Lysi in the Geranian pass, or instance, yielded only coarse-ware sherdsand painted Laconian tiles,27 and the watchtower at Ayios Sostis in the

southern Corinthia produced mainly large coarse-ware vessels and paintedLaconian tiles, with only small amounts o black-glazed ware.28 Closer tothe Vayia region, the systematic documentation o the Classical–Hellenisticortications in the Maritsa pass on Mt. Oneion recorded an assemblagecomparable to the assemblages rom Ano Vayia: predominantly tile, abundantamphoras (including Corinthian A and B), requent coarse ware, inrequentkitchenware, and negligible ne ware.29 While a ew Corinthian orticationsites have produced the diverse array o pottery that one would expect roman ancient habitation,30 this kind o site generally lacks abundant ne wares.

There is no single ceramic signature that can distinguish a Corinthian“ortied garrison” rom a Corinthian “armstead,” and we can expect thereto be overlap in the types o signatures let by such buildings. Nonethe-

less, the ceramic assemblages at sites identied as garrisons, guard posts,

25. The archaeological signature o brothels, houses, and industrial spaces

was one o the themes in the recentcolloquium “Houses o Ill Repute: TheArchaeology o Brothels, Houses, and

Taverns in the Greek World” at the110th Annual Meeting o the Archaeo-logical Institute o America in Philadel-phia (2009). In that colloquium, Kath-leen Lynch discussed a Late ArchaicAthenian household assemblage, com-paring it with the rural Dema and Varihouses as well as with urban houses atOlynthos (Lynch 2009). One conclu-sion o that paper, and in act, the en-tire session, was that ne wares, andspecically drinking cups, are propor-tionally dominant in Archaic andClassical domestic assemblages. Seealso Jones, Sackett, and Graham 1962;

Jones et al. 1973; and Foxhall 2001.26. One could argue that the site

was stripped o its occupational as-semblage during or ater abandon-ment, but occupation did, in act, leavea signature in the orm o numerouscoarse-ware and pithos ragments inand around the structure. Since ne

ware and cooking ware, moreover, aremuch more likely than coarse wares toragment and generate sherds, we

would have expected to nd some

sherds rom these wares at the build-ing i it had been occupied or a signi-cant length o time. The rarity o ne

ware and kitchenware most likely rep-resents sporadic occupation and notintensive domestic habitation.

27. Wiseman 1978, pp. 20–22.28. Wiseman 1978, pp. 114–116.

29. Caraher and Gregory 2006,pp. 340–345.

30. At the Late Classical tower andenclosure at Stanotopi on Mt. Oneion,or example, Stroud (1971a, pp. 130–131) recorded black-glazed ne-ware

vessels (oinochoe, skyphoi, kantharoi),Corinthian A and B amphoras, Corin-thian tiles, and painted Laconian roo tiles, although this diversity can be ex-plained by longer use or more inten-sive occupation. In his examination o aLate Classical ortied garrison stationat Kephalari Station, located on a passleading rom Ayios Vasilios in the Co-rinthia to Mycenae, Wiseman (1978,pp. 118–120) recorded numerous sur-ace sherds, including coarse wares andseveral types o ne wares (e.g., Corin-thian, Argive, and black-glazed Attic).



w. r . carahe r, d . k . pet tegre w, and s . james400

and orts do tend to be simpler and more uniorm than those ound inspecically domestic contexts (with their diverse array o ne ware). Or, toput it another way, Corinthian ortications typically produce less diverseand robust ceramic signatures because they represent less intensive andshorter-term episodes o habitation in the landscape than is typically thecase or armsteads. In light o these observations, we can associate the

assemblage rom Ano Vayia more closely with the assemblages o Clas-sical and Hellenistic military buildings and orts in the Corinthia thathave a predominantly coarse-ware signature and provide evidence orshort-term occupation and use. As we argue below, however, this inter-pretation o Ano Vayia as a kind o garrison building nds support romadditional orms o evidence, such as the nature o other sites in the area,Corinthian topography, and the general context o 5th–4th-century b.c. Corinthian history.

Catalogue

The original study o the Ano Vayia ceramics in 2001 and 2003 by mem-bers o EKAS placed most o the material within the broad periods o Archaic–Classical or Archaic–Hellenistic. In 2008, a restudy o the pot-tery was undertaken by Sarah James, resulting in the ollowing catalogue.

The items included in the catalogue are intended to represent the overallassemblage in terms o chronology and range o types using the availabledata. The catalogue is based on a small sample o artiacts brought back to the Isthmia Excavation House. Many other artiacts that were analyzedaccording to chronotype procedures (see above) were let in the eld, andor these we retained the date and identication assigned at the time o the original analysis; they were not part o this restudy.

As noted above, the assemblage is marked by the presence o a large

pithos (with our nonjoining pieces) and several amphoras. The pithos(4) rim prole and abric have tentatively been identied as dating to theHellenistic period. A predominance o Corinthian B amphora sherds andhandles supports a date rom the 5th to 3rd century b.c. Additional chrono-logical indicators are the numerous roo tiles, which are the typical Classical–Hellenistic type (6), and the ragment o an andesite hopper mill slab (8).Hopper mills are most common in the 4th–3rd century b.c. in the Corinthiaand Argolid, which suggests a Hellenistic date or this piece.31 In the virtualabsence o ne ware, it is dicult to date this material more precisely thanto within a couple o centuries.

Note that all catalogue measurements are in meters.

1 Cooking pot Fig. 137615-510. P.L. 0.054, est. Diam. (rim) 0.120.Medium-coarse yellowish red (5YR 4/6) clay with requent small–large

silver sparkling inclusions, requent large white angular inclusions, and ew large voids.

Ancient

2 Laconian cover tile

7612-4. Max. pres. dim. 0.059.

31. Kardulias and Runnels 1995,pp. 110, 121. More generally, hop-per mills were in use throughout theAegean rom the late 5th century tothe 1st century b.c., but they were mostpopular in the Early Hellenistic period;see, e.g., Pulak et al. 1987, pp. 41–42,no. HW 38, gs. 10, 11.




Medium-coarse yellow-red clay (5YR 5/6) with ew small to medium-sized

black and white inclusions and rare sparkling inclusions. Worn black paint onexterior and edge.Archaic–Classical

3 Corinthian A amphora Figs. 13, 14

7615-507. Th. (body) 0.010.Coarse reddish yellow (7.5YR 7/6) clay with a dark gray core. Rare medium

to very large angular gray and brown inclusions and rare very large voids.C. example in Koehler 1978, pl. 14, no. 12.Mid-6th century b.c.

4 Pithos Figs. 13, 14

7615-501. Est. Diam. (rim) 0.50, Th. (body) 0.022.

Coarse, very pale brown clay (10YR 7/4) with common large to very largeangular red inclusions and many voids in its pocked surace. Squared rim.

Hellenistic?

5 Table amphora

7615-517. Est. Diam. (rim) 0.20, max. pres. dim. 0.034 x 0.051.Medium-coarse, pale yellow (2.5Y 7/4) abric with ew small to medium

black and red inclusions.Classical–Hellenistic

Fge 13. pe m A Vaa:kg 1, cha A am-ha 3, ad h 4. Scale ca. 1:4.Drawings K. R. Pettegrew, S. Van Horn,and W. R. Caraher

Fge 14. Aa m A Vaa.clke m e le: h 4,he mll agme 8, chaa le 7, cha A amha 3.Photo A. Porter

1

3

4




6 Painted Laconian tile

7614-502. P.L. 0.107, p.W. 0.074, Th. (edge) 0.021, Th. (body) 0.017.Coarse reddish yellow (7.5YR 8/6) abric, pocked and eroded. Frequent me-

dium to large red and white inclusions. Worn black paint. Slightly thickened edge.Classical–Hellenistic

7 Corinthian pan tile Fig. 147605-1. P.L. 0.158, p.W. 0.094, Th. 0.025.Coarse, pale brown abric (10YR 7/4), cracked and pocked. Frequent very large

black and gray inclusions with rare, angular, red and white inclusions.Archaic–Hellenistic.

8 Hopper mill Fig. 14

7612-502. P.L. 0.07, Th. 0.03.Lower slab ragment with diagonally incised lines. Andesite rom Nisyros.C. Pulak et al. 1987, pp. 41–42, no. HW 38, gs. 10, 11.Probably 4th–3rd century b.c.

LycHnAri towEr

The second site documented by EKAS lies on the hill o Lychnari, im-mediately to the west o the bay with the same name (Figs. 6, 15–17). Onits eastern side, some 20 m to the southeast o a geodetic marker, are theremains o another round tower. Like the ortications at Ano Vayia, thetower has coursed, rough polygonal masonry that includes stones o mas-sive size. The walls are very well preserved, with an outer ace o largerstones and an inner ace o smaller, but still substantial, stones and with acobble core between the aces. The outer ace is traceable or two-thirdso the circuit and measures over 8.0 m in diameter, with walls over a meter

in width. The inner ace stands to a greater height than the outer ace andgives the remains the appearance o a wedding cake’s stepped construction.

There is no reason, however, to think that this refects the original designo the tower, as the top courses o the outer ace are not nished and inseveral places reach the same height as the inner ace.

Unortunately, the preserved walls have a maximum height o only 1.5 m (Fig. 17), so little can be said regarding the original elevation o the tower. Young’s inormal estimate o heights or these towers, however,suggests that their height could be 2–2.5 times their diameter.32 I this iseven a rough indicator, the tower may have stood to over 15 m in height.Some indication o the original height o the tower might come rom thelow mound o material assembled around the base o the nearby geodetic

marker. The marker stands ca. 2 m high on an articial mound o earthand large stones. Among these stones are numerous blocks o pink andgray conglomerate. It seems probable that these stones were piled aroundthe articial mound or the geodetic marker both to prevent erosion and toelevate the marker above the level o the ruined tower. I these cut blocksoriginally came rom the tumble o the nearby tower, they would suggestthat the tower stood to a considerable height. 32. Young 1956b, pp. 134–135.

Fge 15 (opposite, top). the Lh-a e, veed m he h.Photo K. R. Pettegrew

Fge 16 (opposite, bottom). pla he claal–Helle e a Lha. W. R. Caraher and D. K.Pettegrew




Another eature o this tower is worth noting. The tumble that llsthe central part o the tower orms a small rounded depression (1.0 x 0.60x 0.30 m). Similar eatures have been noted in other rubble structures inthe Corinthia. A possible parallel to this depression is ound in the cairnsdocumented by Dixon near the harbor o Korphos some 10 km to thesouth o Lychnari.33 Dixon suggested that these eatures could be hollowsthat served as bases or herms or stelai marking the border between theCorinthia and the Epidauria during the Hellenistic period.34 Across thebay o Lychnari, however, EKAS teams noted similar depressions in thecairns on the ridge o Vayia, which have been dated to the Early Bronze

Age on the basis o material embedded in the cairns and the measuremento rillenkarren on the stones.35 The cairns at both Korphos and Vayia vary in diameter rom 5 to 10 m, as do the central depressions. Consideringthe dierences in size, unction, and date among the cairns, or now wecan only regard this common eature on a case-by-case basis; at Lychnari,the depression in the center almost certainly relates to a postdepositionalprocess that caused stones to be removed ater the tower’s collapse.

The tower at Lychnari can be dated to the Classical–Hellenistic periodon the basis o pottery embedded in the building’s tumble (Fig. 18) andscattered around the general area. The assemblage, which due to permitrestrictions we could not document in detail, included ragments o pithoi, amphoras, Corinthian tiles, and painted tiles.36 This chronologi-

cal range would be consistent with the rough polygonal masonry style,and we can note or comparanda a similar tower documented by Lolosat the site o Tsakouthi in Sikyonia.37 The Tsakouthi tower is 8.30 m indiameter, with a double wall approximately 1.30 m in width. Like thetower at Lychnari, the Tsakouthi tower was built in a rustic constructionstyle, with ew drated edges and a rough combination o polygonal andtrapezoidal blocks. Lolos dates this tower on the basis o its constructionstyle and the artiacts in the area to ater the 5th century b.c.38

Fge 17. se he e all a Lha, veed m he h-

e. Photo W. R. Caraher

33. Dixon 2000, pp. 87–89.34. Dixon 2000, p. 89.35. Tartaron, Pullen, and Noller

2006, p. 151.36. We observed only one later

ceramic ragment in the vicinity o thetower, a Broneer type XXVIIA lamp(late 1st–early 2nd century a.d.); seeCorinth IV.2, pp. 90–102.

37. Lolos 1998, pp. 233–234.38. Lolos 1998, p. 234.




KAto VAyiA

The nal group o architectural remains that is likely to date to the Clas-sical–Hellenistic period is located at Kato Vayia on the Vayia peninsula,

which projects northwestward into the Saronic Gul and shelters theeastern side o the harbor o Lychnari (Figs. 4, 19, 20). The remains onthe peninsula are so poorly preserved that it is not possible to determinetheir complete plan. They exist amid a scatter o ceramic material thatis very similar to the utilitarian and coarse material ound around AnoVayia and the tower at Lychnari. Moreover, the rubble construction styleis similar to that o the ortications documented at both Stanotopi andon the heights o Mt. Oneion.

The most clearly dened eatures at Kato Vayia are a series o longrubble walls and extensive piles o tumble. The best-preserved wall runsor nearly 40 m rom southeast to northwest, curving slightly to ollow thenatural contours o the ridge and bounding the western side o the levelarea along the top o the Vayia peninsula. This western wall is constructedo unworked, local gray limestone stacked in irregular courses to orm twoaces approximately a meter apart, with cobble ll between the aces.

The stretch o tumble eventually emerges as another clearly dened wall extending or close to 50 m, east to west, across the northern side o the ridge (see Fig. 20). Like the western wall, this wall ollows the contourso the ridge and runs immediately above the steep slope that orms thenorthern side o the ridge. Unlike the western wall, however, this wall is

more careully articulated, showing clear right-angle turns that suggestbuildings or rooms amid long stretches o tumble. The most well-denedstructure occurs about midway along the northern wall, where a small, ir-regularly shaped building projects 3–3.5 m to the south. The walls o thissmall structure are 0.75 m wide and o the same construction style as the

walls elsewhere on the ridge. About 4 m to the east o this building, theeast–west wall takes a sharp turn toward the north and, ater another 4 m,returns again to the east. It is possible that the small building in conjunction

Fge 18. tal aemlage lead ae ae m Lha.Photo K. R. Pettegrew




with this abrupt zigzag in the wall’s course could represent an entrance to anenclosure on the top o the ridge, although its western and southern sidesare poorly preserved and its eastern wall is represented only by an extensivescatter o tumble.39 Three walls o similar construction style mark out theremains o another small building measuring approximately 3.0 x 7.0 m insize, which abutted the northern side o the southern wall o the enclosure.

While it is impossible to oer a denitive interpretation o this com-plex o walls at Kato Vayia, the uniormity o the ceramics associated withthe structures and the extensive system o rubble walls again suggest aortication o the Classical–Hellenistic period. The closest analogy in theCorinthia or this kind o inormal construction are the walls on Stanotopiand Oneion, which are similarly constructed o rubble masonry and situatedatop strategically signicant heights.40

Fge 19. tal e le all a Ka Vaa, veed m he e. Photo W. R. Caraher

39. The dense vegetation and carpeto pine needles obscure the course o the eastern wall. The only place wherethe remains are clearly articulated is onthe southern side o the ridge.

40. Caraher and Gregory 2006;Stroud 1971a.




Discussion: Function, topoGrApHy, AnD History

Over the last several decades, regional programs o archaeological researchhave populated the Greek countryside with Classical and Hellenistic arm-steads, buildings, monuments, and places associated with the ephemeralactivities o rural lie. Among the most debated types o sites are rural tow-ers, which recent scholarship has associated with guard stations and com-munication beacons, ortied arms, and outposts or intensive agriculturalactivities such as slave-driven mining endeavors.41 In these assessments,outlined in the recent sweeping study by Morris and Papadopoulos,42 Corinthian sites have remained conspicuously absent. This refects thepaucity o towers in the Corinthian countryside as well as the dicultnature o the written sources or the region’s economy, settlement structure,and military organization. Despite the absence o scholarly discussion on

Fge 20. pla he claal–Helle fa a KaVaa. W. R. Caraher and D. K. Pettegrew

41. The works o Young and Fracchiaestablished the basic criteria or as-sessing the unction o tower sites inthe Greek countryside. Young ocusedhis work on towers in southeasternAttica and the island o Siphnos, andFracchia extended his arguments to her

study o towers in the Argolid: Young1956a, 1956b; Fracchia 1985. See alsoOsborne 1986, 1992.

42. Morris and Papadopoulos 2005.In addition to synthesizing and com-piling research on a wide variety o Greek towers and rural installations

(pp. 157–167), the authors proposethat many rural towers served to houseslaves who worked with high-laborand high-value crops such as vines, orin mines such as those in southeasternAttica.




rural installations in the Corinthia, it is nevertheless clear rom epigraphicsources and also rom a growing body o data produced by archaeologicalsurveys that Corinthians did live in the countryside, oten at some distancerom the urban core o Corinth.43

The ortications near Vayia and Lychnari contribute to our under-standing o the Corinthian countryside as well as to the broader topic o

ortied rural structures in Greece. A careul examination o the remainsat Vayia and around Lychnari Bay oers several reasons or concludingthat these sites unctioned to guard this agriculturally rich and strategi-cally signicant stretch o the Corinthian coastland. These arguments arebased upon the associated artiact assemblage, architectural design o thebuildings, and their place within the local topography.

As we have already seen, the artiact assemblage rom Ano Vayiaindicates low-intensity habitation in the countryside. The Lychnari towerand the rubble ortication at Kato Vayia also produced ceramic objectsconsistent with this assessment. The sites may have accommodated someshort-lived domestic activities, but these were not sustained or permanent.

Most critically, the sites near Vayia produced ceramic assemblages thatbear little resemblance to those rom long-term rural habitations identiedelsewhere in the Corinthia or to those rom typical Classical–Hellenisticarmsteads in Greece. On the other hand, the artiacts rom these build-ings do have immediate comparisons with material rom Corinthian sitesidentied as ortications and towers. For this reason, along with the reasonsdiscussed below, we avor interpreting the sites at Vayia as buildings orsmall garrisons around Lychnari Bay.

The location o these towers in the local topography provides ad-ditional reason to conclude that they were principally used or protectionand deense rather than domestic and agricultural enterprises. At thehighest point in their landscapes, the Lychnari and Ano Vayia towers

(Fig. 2) were not positioned to acilitate the economic exploitation o thelocal landscape. The towers are some distance removed rom tillable landand cannot thereore be easily understood as part o intensive agriculturalinvestment (e.g., viticulture) or the kind o intensive cultivation necessary to support rural industries such as mining—as scholars have posited ortowers in other parts o the Greek world.44 Indeed, throughout the easternCorinthia, the EKAS project demonstrated that Classical–Hellenistic ruralhabitation tends to be concentrated on the most agriculturally productiveland (e.g., the plain o the Isthmus) rather than on more marginal lands(e.g., the lower slopes o Oneion).45 In this respect, it is interesting to notethe lack o substantial Roman reuse o the sites o Lychnari and Ano Vayia,

43. For arguments based largely onepigraphical and literary evidence, seeStroud 1968; Salmon 1984, pp. 413–419; Stanton 1986; Dixon 2000,pp. 291–293. For recent discussions o Classical settlement in the eastern Co-rinthia, see Caraher, Nakassis, and Pet-tegrew 2006, pp. 14–21; Tartaron et al.2006, pp. 494–513; Caraher et al. 2009.

Isthmus, see Caraher, Nakassis, andPettegrew 2006, pp. 14–21; Tartaronet al. 2006, pp. 494–513. See alsoCaraher et al. 2009, or discussion o the distributional data in the valley o Lakka Skoutara, near Sophiko, whereClassical–Hellenistic material is com-mon in the valley itsel, but stops below the slopes.

44. Generally, see Morris and Papa-dopoulos 2005. Moreover, a numbero towers in the Argolid and southernCorinthia do not occupy advantageousor superior topographic positions; see,e.g., Lord 1938, 1939; Lord, Frantz,and Roebuck 1941; Hjohlman, Pent-tinen, and Wells 2005.

45. For Classical settlement on the




which could indicate that later inhabitants viewed the location o thesebuildings as too marginal or agricultural exploitation.46 The locations o theLychnari and Ano Vayia structures in no way exclude typical agriculturalactivities—as the presence o the hopper mill ragment suggests—but thesebuildings were clearly not positioned to maximize such activity.

The visibility o these towers also complicates their identication as

ortied homesteads or amily arms. While their positions on top o theLychnari and Ano Vayia ridges aorded commanding views o the land-scape, they also increased their visibility and vulnerability rom the principalland and maritime routes that they overlooked (see below). I the unc-tion o these towers was solely or local landowners to protect their ownhuman or material property, there are more obvious locations that wouldhave provided better views toward the land with less exposure to dangerrom the sea. We can contrast the towers near Lychnari with the pyramidaltowers in the Argolid, which, as Fracchia observed, lacked the kind o com-manding view appropriate or a watchtower. Notably, the towers in theArgolid also diered in the presence o dense scatters o ceramics and

agricultural processing equipment consistent with rural habitation.

47

The location o the towers and buildings at Vayia, however, does makesense i they unctioned as military installations guarding agricultural land,transportation corridors, and coastal zones. The Lychnari tower is locatedat the top o the Lychnari hill and seems to be positioned to overlook thebay and the northern coast o the Corinthia (Figs. 4, 5), while the AnoVayia tower overlooks the pass rom Frangolimano as well as the VayiaRiver valley (Figs. 6, 7). Indeed, both towers were clearly intervisible(Figs. 2, 6) and presumably were placed to work together to monitoractivities in the area o Lychnari and Vayia. The tower at Ano Vayiaoverlooked movement through the pass rom Frangolimano as well asagricultural land to the south, but the height o the coastal ridge o Kaki

Rachi compromised its view o the northern coast o the Corinthia andthe Saronic Gul islands. The tower on Lychnari, in contrast, did not oera clear view o the pass but provided a good view o the northern coast o the Corinthia, including most o the Saronic Gul and islands. Together,the ortications could have unctioned to hinder, block, or prevent enemy passage through two o the best natural harbors o the eastern Corinthia,Lychnari and Frangolimano.

In this respect, we see two plausible purposes served by the structuresat Lychnari Bay. The rst is that the towers unctioned within a broadersystem o Corinthian deense aimed at preventing systematic incursioninto the Corinthian chora by guarding or even blocking signicant land andmaritime transportation routes. Small garrison units stationed at Lychnari

46. (Late) Roman reuse o ruralClassical–Hellenistic buildings andtowers is very common in Greecegenerally (see Hjohlman, Penttinen,and Wells 2005; Pettegrew 2006) andin the Corinthia specically, where themajority o Classical–Hellenistic sites(75% o EKAS units [n = 561 o 750]

and over 90% o larger sites) show evi-dence or Roman-period reuse. Thereuse o sites is not only economically sensible, as it continues the materialinvestment in buildings on the land,but it maintains and renews social tiesto places over centuries.

47. Fracchia 1985, p. 688.




or Ano Vayia would, o course, have been no match or a ull-scale invadingarmy, but they could at least have communicated with orces positionedcloser to the Isthmus (at Stanotopi, Oneion, and Kenchreai, and evenAcrocorinth) and thereby provided an early warning signal o imminentdanger to land closer to Corinth.48 Since these other sites lacked a clear

view o the bays along the southern coast o the Corinthia, the towers at

Lychnari Bay could have provided a rst watch. While it may be hard toimagine an invading army choosing this route or an attack on Corinth, it

would have been oolish to leave the pass unguarded, or a orce landingat Lychnari or Frangolimano and moving west into Corinthian territory

would have been completely hidden by the coastal heights and out o view o Corinthian positions near Oneion and the Isthmus.

The role o towers in guarding passes and establishing regional military communication networks is well documented in nearby regions o the Greek mainland.49 Ober and Munn have both shown how rural towers in Atticabelonged to networks o routes, towers, and ortied sites that unctionedtogether or local deense in the Late Classical world.50 Recently, Lolos

and Marchand have demonstrated the close link between towers and roads,arguing that city-states used towers to control trac through the country-side.51 As noted above, Lolos documented a tower at Tsakouthi in Sikyoniathat is similar in size and construction technique to the round tower on theheight o Lychnari; he argued that it overlooked a signicant roadway link-ing the Sikyonian plain to the region around Stymphalos.52 In these contexts,rural towers unctioned mainly as signal stations across the countryside,connecting military orces, rural communities, and polis centers separatedby long distances and rocky terrain. The impressive views aorded theLychnari and Ano Vayia towers must have extended the infuence o any orce stationed in the ortications on the Vayia peninsula.

The ortications o Vayia and Lychnari also nd good parallels in Co-

rinthian (and their allies’) eorts to guard or block vulnerable passes inthe mountainous regions o Corinth. Wiseman, Smith, and others haveassociated a network o towers with the road network that passes romthe southern Megarid into the Corinthia via either the Kaki Skala or overpasses through Mt. Geraneia.53 Further south, the large and complex orti-ed site o Ayia Paraskevi, near the modern village o Sophiko, overlooks aertile plain and several major lines o communication and travel throughthe southeastern Corinthia.54 While this site could represent a ortiedoutpost or a village o the Corinthian interior, its position also suggestsa military unction not unlike that o the “border orts” along the Attic-Boiotian rontier.55 Similarly, the impressive array o rubble ortications

48. Wiseman 1978, p. 58.49. Although the military interpre-tation has been critiqued and has otenbeen replaced by agricultural explana-tions, it still works well or certainregions. For summary o recent work,see Morris and Papadopoulos 2005,pp. 157–167.

50. Ober 1985; Munn 1993.51. See Lolos 1998, pp. 242–244;

Marchand 2009, pp. 130–137.52. Lolos 1998, pp. 233–234.53. Wiseman 1978, p. 17; Smith

2008, pp. 24–25. For Geraneia, see Wiseman 1978, pp. 24–26; Smith 2008,pp. 31–32.

54. See n. 2, above.55. There is evidence that the or-

tication at Ayia Paraskevi even hadperipheral ortication in the area,

which would have allowed it to monitorits neighborhood more eectively; seeDixon 2005. For the “border orts” o Attica and Boiotia, see Ober 1983,1985, 1987a, 1987b; Cooper 1986,2000; Camp 1991; and Munn 1993.For Corinthian tribes and trittyes, seeStroud 1968; Salmon 1984, pp. 413–419; and Stanton 1986.




along the ridge o Oneion must represent eorts to control passage acrossthe eastern ridge o the mountain and through the rugged interior o theCorinthia—even i those walls should represent a temporary occupation by a oreign power.56 In this respect, it is worth noting the requency o Clas-sical–Hellenistic rural towers and buildings o military unction at higherelevations in the eastern Corinthia. Such a consistent pattern highlights

the close connection o the region with travel corridors and the interest inguarding its territory. Whether this impetus or ortiying the southeastCorinthia was essentially centralized (rom the state o Corinth itsel) ordecentralized (depending on groups o local landowners) is beyond thethreshold o our evidence.

A second plausible purpose served by the ortications in the Vayiaregion may have been to house small garrisons stationed to provide immedi-ate and eective response to small-scale raids and banditry on agriculturallands west o the bay. In this scenario, soldiers stationed in ortied campssuch as that at Kato Vayia could have orestalled opportunistic raidingepisodes that would damage local agricultural endeavors and domestic

acilities such as arms and storage acilities. Protection o this sort wouldhave been particularly important during certain seasons, such as the latespring grain harvest and the late all olive harvest.57 We can also expectthat such ortications could have provided protection both or and romorces engaged in naval activities in the Saronic Gul (see below).

Use o the site at Kato Vayia to house garrisons makes sense in view o the inormal yet substantial rubble walls ound there. The site nds parallels

with the so-called ortied camps in Attica documented by McCredie, 58 such as the hastily ortied positions at Koroni.59 Both Koroni and Vayiastand on a coastal peninsula with easy access to a protected natural embay-ment, and in construction, both share a casemate style o architecture withsmall rooms constructed in drystone masonry stacked against a ortication

wall.60

While the precise unction o the substantial ortied site o Koroniis disputed,61 the small size and inormal construction style o the Vayia ort

would t McCredie’s criteria or ortied camps. A orce stationed at KatoVayia would have been able to respond quickly to small orces attemptingto come ashore in Lychnari Bay or the Vayia River delta or pass throughthe valley rom the east.

The imprecise dates or the ceramic material at Vayia and the paucity o literary sources that deal directly with Corinthian aairs during the Classicaland Hellenistic periods make it dicult to do more than speculate on theoccasion or the construction o the ortications.62 On the one hand, the

56. Stroud 1971a; Caraher and

Gregory 2006.57. Morris and Papadopoulos 2005,pp. 158–163.

58. McCredie 1966.59. Vanderpool, McCredie, and

Steinberg 1962.60. The absence o evidence or

trimming stones to t in the wall may indicate hasty construction at bothKoroni and Vayia, comparable to the

ortications on both Oneion and

Stanotopi. This pattern could suggestshort-term response to some emergency or strained military or political circum-stances. The evidence or later repairor modication is even more inormalthan that or the building’s originalconstruction, suggesting that the ad hoc nature o the structure carried throughits entire lie span. The shoddy con-struction o the apparent repairs and

ancillary walls seems more appropri-

ate i we consider this building tohave been occupied episodically ratherthan in a systematic or consistentashion.

61. See Lauter-Bue 1989; Caraherand Gregory 2006, p. 347.

62. See comments in Stroud 1971a,pp. 139–145; 1971b; Caraher andGregory 2006, pp. 345–347.




coarse archaeological dating means that the ortications discussed in thisarticle can only be broadly dated to the 5th to 3rd centuries b.c., and may have been used and developed episodically over these three centuries. It ispossible, then, that more than one o the scenarios proposed here apply tothe site as it was used over time. On the other hand, the general dearth o textual evidence or exurban activities provides the greatest challenge or

any scholar intent on advancing an argument about the unction o towersin the Greek countryside. Textual evidence can provide suitable backgroundcontexts—contemporary political or military conditions—against whichthe archaeological remains stand out.

Despite the lack o specicity in the textual sources, we do have a body o ragmentary literary evidence that provides potential context or the in-ormal rural ortications documented around Lychnari Bay. Throughoutthe Classical and Hellenistic periods, there were numerous opportunitiesand reasons to erect ortications along the Saronic Gul. In the 5th century,these sites protected the Corinthian coast against opportunistic raids by sea,such as the Athenian sack o Solygeia in 425 b.c., which demonstrated the

vulnerable state o Corinthian territory south o the Isthmus.

63

Thucydidesalso describes a clash in 412 b.c. between Athenian and Peloponnesianorces in the Saronic Gul at the deserted harbor o Speiraion, north o the Corinthia-Epidauria rontier.64 In that account, Athenian naval orcesstationed on a Saronic island ollowed a Peloponnesian feet o three dozenships, attacking them in the harbor and on the beach. The Peloponnesiansdespaired o guarding their ships in such a deserted place, and even consid-ered burning them, but eventually resolved to pull their entire feet high ontodry land and station their troops nearby. Corinthians and other neighboringinhabitants arrived the ollowing day to reinorce them, and the Pelopon-nesian feet shortly thereater made a sally and escaped to Kenchreai. Whilethere would be problems in identiying Speiraion with Vayia,65 this account

suggests another possible scenario or the origin o ortied garrisons alongthe southern Corinthian coast, and shows the way in which neighboringtroops (πρόσχωροι) were useul or deending even relatively isolated harbors.

By the 4th century, Corinth was ully engaged in the tumultuous poli-tics o internecine warare, and the movement o troops through Corin-thian territory demonstrated both the vulnerability o the city’s chora andthe need to ortiy specic corridors through the territory. In 370/69, orexample, Theban troops under Epaminondas passed easily through theeastern part o the Corinthia.66 In 366, the Argive general Peisias movedtroops rom Argos to the heights o Oneion through the rolling hills northo Solygeia, showing a viable route rom (and into) the Peloponnese thatbypassed the traditional concentration o Corinthian ortications around

the polis center.67 Such examples highlight how the regular movement o troops through the Corinthian chora in the early 4th century precipitatedan increased eort to maintain guards, ortications, and orces in thecountryside. We know, or example, that as early as 366 b.c. the Atheniansplaced garrisons throughout the Corinthian countryside, which were soonreplaced by Corinthian orces when political relations between the twostates soured.68 We can easily imagine that these orces could have con-structed rubble ortications such as those near Vayia.

63. For Solygeia, see Thuc. 4.42–45; Fowler 1932, pp. 97–99; Stroud1971b; 1994, pp. 269–280; Wiseman1978, pp. 56–58.

64. Thuc. 8.10–15, 20. See Wise-man 1978, pp. 136–140; Salmon1984, pp. 5–7, 336–338; Stroud 1994,pp. 297–299; Dixon 2000, pp. 77–78.

65. Wiseman (1978, p. 140) andSalmon (1984, pp. 5–7) have identiedSpeiraion with the harbor o Korphos,urther to the south. Dixon (2000,p. 78) has suggested that the neighbor-hood orces might have been stationedat the nearby tower o Ara Bartze.

66. Xen. Hell. 6.5.50–52. See Stroud1971a, p. 139; Caraher and Gregory 2006, p. 346.

67. Xen. Hell. 7.1.42. See Stroud1971a, pp. 140–141.

68. Xen. Hell. 7.4.4–5. See Stroud1971a, p. 140.




There were also occasions or ortication throughout the early Hel-lenistic period. Stroud asserts that Philip o Macedon probably ortiedkey parts o the Corinthia, even i we have no idea about the speciclocations.69 At the end o the 4th century, Alexander, son o Polyperchon,seems to have ortied points in the eastern Corinthia to prevent Kassanderrom taking the city.70 Diodorus Siculus reports that there were two orts

near the port town o Kenchreai; Stroud and others have associated these with the inormal rubble ortications on Mt. Oneion.71 The orticationsmentioned in this survey reveal the challenges acing any eort to link theragmentary textual sources or the history o Corinth’s territory to specicsites during the later 4th and 3rd centuries. At the same time, these sourcesindicate that there were more episodes o ortiying the Corinthia than wecan securely identiy in the archaeological record. It is reasonable to believethat some o these eorts to ortiy the Corinthia involved garrisons whoconstructed inormal outposts to guard the countryside and the coastalinlets to the territory.

The textual evidence or placing garrisons and orts in the countryside

provides more than one plausible context or the construction o ortica-tions around Lychnari Bay. As we suggested above, these structures, whichshare with other Corinthian towers a similar place in local topography,design, and construction style, represent eorts to saeguard the Corin-thian chora, the garrisons stationed on the coast, and the major routes intothe region. I their rough construction and irregular design indicate a lesssignicant investment in the countryside than the rural towers typically studied by scholars, these eatures are nevertheless consistent with theepisodic character o rural Corinthian ortications in general.

Historical evidence or Corinthian ortication during the Classical–Hellenistic period has tended to ocus on eorts to block military orcesrom moving through the Isthmus.72 Generally, the various states that

sought to ortiy the Isthmus were not concerned with deending Corin-thian territory and interests per se. In contrast to these better-known orti-cations, the towers and sites at Vayia did little to protect the Peloponnesegenerally, or armies moving southward could easily bypass them by sea andland in the vicinity o Epidauros, where there was a more convenient pointo access to the Peloponnesian centers at Argos and urther south. 73 It islogical to read the sites at Vayia instead as installations o the Corinthianstate or local Corinthian citizen landowners, whose aim in ortiying wasto prevent episodic but destructive local raids and to guard against incur-sions deep into Corinthian territory. The ease with which an army couldpass north rom the bay at Lychnari or even Frangolimano and ravage theCorinthian chora, or continue north to Oneion and the Isthmus, made

the ortication o this coastline a crucial component o any strategy toprotect Corinthian territory, and turned these inormal ortications intoa signicant eature o Corinth’s landscape.

69. Stroud 1971a, p. 142.70. Diod. Sic. 19.63.4. See Stroud

1971a, p. 142.71. Stroud 1971a, p. 143; Caraher

and Gregory 2006, pp. 346–347.72. Wiseman 1963.73. Diod. Sic. 19.54.3.




rEFErEncEs

Bintli, J., P. Howard, and A. Snod-grass. 2007. Testing the Hinterland:The Work o the Boeotia Survey (1989– 1991) in the Southern Approaches tothe City o Thespiai (McDonald In-stitute or Archaeological Research,Monographs), Cambridge.

Camp, J. McK., II. 1991. “Notes onthe Towers and Borders o ClassicalBoiotia,” AJA 95, pp. 193–202.

Caraher, W. R., and T. E. Gregory.2006. “Fortications o MountOneion, Corinthia,” Hesperia 75,pp. 327–356.

Caraher, W. R., T. E. Gregory, D. K.Pettegrew, and L. Tzortzopoulou-Gregory. 2009. “Between Sea andMountain: The Archaeology o a

20th-Century ‘Small World’ in theUpland Basins o the SoutheasternKorinthia” (paper, Modern Greek Studies Association, Symposium 21,Vancouver 2009).

Caraher, W. R., D. Nakassis, and D. K.Pettegrew. 2006. “Siteless Survey and Intensive Data Collection in anArtiact-Rich Environment: CaseStudies rom the Eastern Corinthia,Greece,” JMA 19, pp. 7–43.

Cooper, F. A. 1986. “Epaminondasand Greek Fortications,” AJA 90,p. 195 (abstract).

———. 2000. “The Fortications o Epaminondas and the Rise o theGreek City,” in City Walls: The Urban Enceinte in Global Perspec-tive (Studies in Comparative Early Modern History), ed. J. Tracy,Cambridge.

Corinth = Corinth: Results o ExcavationsConducted by the American School o Classical Studies at Athens, Cam-bridge, Mass.

I.1 = H. N. Fowler and R. Still- well, Introduction, Topography, Archi-tecture , 1932

IV.2 = O. Broneer, TerracottaLamps, 1930

Dixon, M. 2000. “Disputed Territo-ries: Interstate Arbitrations in theNortheastern Peloponnese, ca. 250–150 b.c.” (diss. Ohio State Univ.).

———. 2005. “Epigraphy and Topo-graphical Survey: The Case o theCorinthian-Epidaurian Border inthe Hellenistic Period,” in Ancient Greece at the Turn o the Millennium:

Recent Work and Future Perspectives.Proceedings o the Athens Symposium,18–20 May 2001, ed. N. M. Kennelland J. E. Tomlinson (Publicationso the Canadian ArchaeologicalInstitute at Athens 4), Athens,pp. 137–145.

Fowler, H. N. 1932. “Corinth and theCorinthia,” in Corinth I.1, pp. 18–114.

Foxhall, L. 2001. “Colouring in theCountryside: Response to David K.Pettegrew, ‘Chasing the ClassicalFarmstead,’ JMA 14.2 (December2001),” JMA 14, pp. 216–222.

Fracchia, H. M. 1985. “The Pelopon-nesian Pyramids Reconsidered,” AJA 89, pp. 683–689.

Hjohlman, J., A. Penttinen, andB. Wells. 2005. Pyrgouthi: A Rural Site in the Berbati Valley rom the Early Iron Age to Late Antiquity: Excavations by the Swedish Institute at Athens, 1995 and 1997 (SkrAth 4°,52), Stockholm.

Jones, J. E., A. J. Graham, L. H. Sack-ett, and M. I. Geroulanos. 1973.“An Attic Country House below the Cave o Pan at Vari,” BSA 68,pp. 355–452.

Jones, J. E., L. H. Sackett, and A. J.Graham. 1962. “The Dema House

in Attica,” BSA 57, pp. 75–114.Kardulias, P. N., and C. N. Runnels.1995. “The Lithic Artiacts: FlakedStone and Other Nonfaked Lith-ics,” in Artiact and Assemblage: The Finds rom a Regional Survey o the Southern Argolid, Greece, ed. C. N.Runnels, D. J. Pullen, and S. Lang-don, Stanord, pp. 74–139.

Koehler, C. G. 1978. “Corinthian Aand B Transport Amphoras” (diss.Princeton Univ.).

Lauter-Bue, H. 1989. “Die Festungau Koroni und die Bucht von Porto

Raphti: Ein Beitrag zur GeschichteAthens im 3 Jh. v. Chr.,” in Attische Festungen: Beiträge zum Festungs-wesen und zur Siedlungsstruktur vom 5. bis zum 3. Jh. v. Chr. (AttischeForschungen 3, MarbWPr 1988),ed. H. Lauter, H. Lauter-Bue, andH. Lohmann, Marburg, pp. 67–102.

Lolos, Y. A. 1998. “Studies in the Topography o Sikyonia” (diss. Univ.o Caliornia, Berkeley).

Lord, L. E. 1938. “The ‘Pyramids’ o Argolis,” Hesperia 7, pp. 481–527.

———. 1939. “Watchtowers and For-tresses in Argolis,” AJA 43, pp. 78–84.

Lord, L. E., M. A. Frantz, and C. Roe-buck. 1941. “Blockhouses in theArgolid,” Hesperia 10, pp. 93–112.

Lynch, K. 2009. “Pottery rom a LateArchaic House by the Numbers,” Archaeological Institute o America,110th Annual Meeting Abstracts 32,Philadelphia, pp. 163–164.

Marchand, J. C. 2009. “Kleonai, theCorinth–Argos Road, and the ‘Axiso History,’” Hesperia 78, pp. 107–163.

McCredie, J. R. 1966. Fortifed Military

Camps in Attica (Hesperia Suppl. 11),Princeton.Morris, S. P., and J. K. Papadopoulos.

2005. “Greek Towers and Slaves:An Archaeology o Exploitation,” AJA 109, pp. 155–226.

Munn, M. H. 1993. The Deense o Attica: The Dema Wall and the Boio-tian War o 378–375 B.C., Berkeley.

Ober, J. 1983. “Two Ancient Watch-towers above Aigosthena in theNorthern Megarid,” AJA 87,pp. 387–392.

———. 1985. Fortress Attica: Deense

o the Athenian Land Frontier, 404– 322 B.C. ( Mnemosyme Suppl. 84),Leiden.

———. 1987a. “Early Artillery Towers:Messenia, Boiotia, Attica, Megarid,” AJA 91, pp. 569–604.

———. 1987b. “Pottery and Miscella-neous Artiacts rom Fortied Sitesin Northern and Western Attica,”Hesperia 56, pp. 197–227.

Osborne, R. 1986. “Island Towers: TheCase o Thasos,” BSA 81, pp. 167–178.

———. 1992. “‘Is It a Farm?’ The

Denition o Agricultural Sitesand Settlements in Ancient Greece,”in Agriculture in Ancient Greece.Proceedings o the Seventh Interna-tional Symposium at the Swedish Institute at Athens, 16–17 May,1990, ed. B. Wells (SkrAth 4˚, 42),Stockholm, pp. 21–28.

Peppas, Y. 1990. Μεσαιωνικές σελίδες

της Αργολίδος, Αρκαδίας, Κοριν -

θίας, Αττικής , Athens.




Smith, P. J. 2008. The Archaeology and Epigraphy o Hellenistic and Roman Megaris, Greece (BAR-IS 1762),Oxord.

Stanton, G. R. 1986. “The Territorial Tribes o Korinth and Phleious,”ClAnt 5, pp. 139–153.

Stroud, R. S. 1968. “Tribal Boundary Markers rom Corinth,” CSCA 1,pp. 233–242.

———. 1971a. “An Ancient Fort onMount Oneion,” Hesperia 40,pp. 127–145.

———. 1971b. “Thucydides and theBattle o Solygeia,” CSCA 4,pp. 227–247.

———. 1994. “Thucydides and Cor-inth,” Chiron 24, pp. 267–302.

Tartaron, T. F., T. E. Gregory, D. J.Pullen, J. S. Noller, R. M. Rothaus,

J. L. Rie, L. Tzortzopoulou-

Gregory, R. Schon, W. R. Caraher,D. K. Pettegrew, and D. Nakassis.2006. “The Eastern KorinthiaArchaeological Survey: IntegratedMethods or a Dynamic Land-scape,” Hesperia 75, pp. 453–523.

Tartaron, T. F., D. J. Pullen, R. K.Dunn, L. Tzortzopoulou-Gregory,A. Dill, and J. I. Boyce. Forth-coming. “The Saronic HarborsArchaeological Research Project(SHARP): Investigations at Myce-naean Kalamianos, 2007–2009,”Hesperia 80.

———. 1993. Μεσαιωνικές σελίδες

της Κορινθίας , Athens.Pettegrew, D. K. 2001. “Chasing the

Classical Farmstead: Assessing theFormation and Signature o RuralSettlement in Greek LandscapeArchaeology,” JMA 14, pp. 189−

209.———. 2002. “Counting and Color-

ing Classical Farms: A Response toOsborne, Foxhall, and Bintli et al.,” JMA 15, pp. 267–273.

———. 2006. “Reurbishing the Farms: The Reuse o the Rural Landscapein Late Roman Greece,” in CommonGround: Archaeology, Art, Science,and Humanities. The Proceedings o the 16th International Congress o Classical Archaeology, Boston, August 23–26, 2003, ed. C. C. Mattusch,A. A. Donohue, and A. Brauer,

Oxord, pp. 33–35.———. 2007. “The Busy Countryside

o Late Roman Corinth: Interpret-ing Ceramic Data Produced by Regional Archaeological Surveys,”Hesperia 76, pp. 743–784.

Pulak, C., R. F. Townsend, C. G. Koeh-ler, and M. B. Wallace. 1987. “TheHellenistic Shipwreck at Serçe Li-manı, Turkey: Preliminary Report,” AJA 91, pp. 31–57.

Salmon, J. B. 1984. Wealthy Corinth: A History o the City to 338 B.C., Oxord.

Tartaron, T. F., D. J. Pullen, and J. S.Noller. 2006. “Rillenkarren at Vayia:Geomorphology and a New Classo Early Bronze Age FortiedSettlement in Southern Greece,” Antiquity 80, pp. 145–160.

Vanderpool, E., J. R. McCredie, and

A. Steinberg. 1962. “Koroni: APtolemaic Camp on the East Coasto Attica,” Hesperia 31, pp. 26–61.

Whitelaw, T. 1998. “Colonisation andCompetition in the Polis o Keres-sos: The Development o Settle-ment in North-West Keos romthe Archaic to the Late RomanPeriods,” in Kea-Kythnos: Historyand Archaeology. Proceedings o an International Symposium, Kea-Kythnos, 22–25 June 1994 (Melete-mata 27), ed. L. G. Mendoni andA. J. Mazarakis Ainian, Paris,

pp. 227–257. Wiseman, J. R. 1963. “A Trans-Isthmian

Fortication Wall: Notes on Helle-nistic Military Operations in theCorinthia,” Hesperia 32, pp. 248–275.

———. 1978. The Land o the Ancient Corinthians (SIMA 50), Göteborg.

Young, J. H. 1956a. “Ancient Towerson the Island o Siphnos,” AJA 60,pp. 51–55.

———. 1956b. “Studies in South At-tica: Country Estates at Sounion,”Hesperia 25, pp. 122–146.

William R. Caraher

University of North Dakota

department of history

276 centennial drive

merrifield hall, stop 8096

grand forks, north dakota 58201

wi l l [email protected]

David K. PettegrewMessiah College

department of history

one college avenue

box 3051

grantham, pennsylvania 17027

[email protected]

Sarah James

University of Texas at Austin

department of classics

1 universit y station c3400

austin, texas 78712 –0308

sarah . james@mai l .utexas . edu

caraher pettegrew james vayiaoffprint 2010

Documents