Ancient Greek Houses and Households: Chronological, Regional, and Social Diversity 9780812204438

Citation preview

Ancient Greek Houses and Households

This page intentionally left blank

Ancient Greek Houses and Households Chronological, Regional, and Social Diversity

E D I T E D B Y B R A D L E Y A . A U LT AND LISA C. NEVETT

University of Pennsylvania Press Philadelphia

Copyright © 2005 University of Pennsylvania Press All rights reserved Printed in the United States of America on acid-free paper 10

9

8

7

6

5

4

3

2

1

Published by University of Pennsylvania Press Philadelphia, Pennsylvania 19104-4011 Library of Congress Cataloging-in-Publication Data Ancient Greek houses and households : chronological, regional, and social diversity / edited by Bradley A. Ault and Lisa C. Nevett. p. cm. ISBN 0-8122-3875-3 (cloth : alk. paper) Includes bibliographical references and index. 1. Dwellings—Greece—History—To 1500. 2. Housing—Greece—History— To 1500. I. Ault, Bradley A. (Bradley Allen), 1961–. II. Nevett, Lisa C. DF99 .A53 2005 307.3′0938 22—dc22 2005042095

Dedicated to the memory of Michael “Mike” H. Jameson 1924–2004

This page intentionally left blank

Contents

List of Illustrations 1. Introduction Lisa C. Nevett

ix

1

2. Structural Change in Archaic Greek Housing Franziska Lang 12 3. Security, Synoikismos, and Koinon as Determinants for Troad Housing in Classical and Hellenistic Times William Aylward 36 4. Household Industry in Greece and Anatolia Nicholas Cahill 54 5. Living and Working Around the Athenian Agora: A Preliminary Case Study of Three Houses Barbara Tsakirgis 67 6. Between Urban and Rural: House-Form and Social Relations in Attic Villages and Deme Centers Lisa C. Nevett 83 7. Houses at Leukas in Acarnania: A Case Study in Ancient Household Organization Manuel Fiedler 99 8. Modest Housing in Late Hellenistic Delos Monika Trümper 119

viii

Contents

9. Housing the Poor and Homeless in Ancient Greece Bradley A. Ault 140 10. Summing Up: Whither the Archaeology of the Greek Household? Bradley A. Ault and Lisa C. Nevett 160 Glossary

177

List of Contributors Index

179

181

Acknowledgments

190

Illustrations

Maps 1.1 2.1 2.2 3.1

Greece and the Aegean, sites in text 2 Greece and the Aegean, Archaic sites 15 Distribution of apsidal and oval houses 17 The Troad 38

Figures 2.1. 2.2. 2.3. 2.4. 3.1. 3.2. 3.3. 4.1. 4.2. 4.3. 5.1. 5.2. 5.3. 5.4. 6.1. 6.2. 6.3. 6.4.

Plans of Archaic houses mentioned in the text 16 Zagora, Andros, houses 21 Kastanas, houses 23 Access and room arrangement, Geometric and Archaic houses 25 Troad houses, schematic plans 40 Ilion, schematic plan of Hellenistic house walls, south slope 41 Ilion, schematic plan of Hellenistic house walls, Lower City 43 Olynthus, distribution of household industry 56 Olynthus, House A 6. 57 Sardis, Archaic Lydian houses 61 Athens, houses south of South Stoa I 68 Athens, House of Simon 71 Athens, House of Mikion and Menon 72 Athens, Houses C and D 73 Thorikos, courtyard houses 86 Thorikos, Velatouri Hill, Industrial Quarter 87 Ano Voula, remains of houses 91 Ano Voula, Kalampoka plot 92

x

List of Illustrations

6.5. Rhamnous, phrourion 94 7.1. Leukas, House AII.2 102 7.2. Leukas, House BII.3 102 7.3. Athens, Classical house on Areopagus 103 7.4. Corinth, Classical house, “Terracotta Factory” 103 7.5. Kassope, House 5 104 7.6. Leukas, House AII.5 105 7.7. Leukas, House AII.6 105 7.8a. Leukas, pottery, Houses AII.5, AII.6 107 7.8b. Leukas, pottery, room o′/p′ 107 7.8c. Leukas, pottery, room q 107 7.8d. Leukas, pottery, room h 107 7.9. Leukas, Houses AII.5, AII.6, pottery by function and shape 108 7.10. Leukas, House AII.6 window pillar 109 7.11. Leukas, House AII.6 with room identiWcations 114 8.1.a. Quartier de l’Inopos: complexes α–γ 121 8.1.b. Quartier du stade: Îlot II, complexes ζ–θ 121 8.1.c. Quartier du théâtre: normal house III Z 121 8.1.d. Quartier du stade: Îlot I, complexes ε–ε′′′ 121 8.2a. Quartier du théâtre: complexes III, H, Y; 27/29. 124 8.2b. Quartier du théâtre: complexes IV c a–d/6; IV c 7–13 124 8.3. Quartier du théâtre: distribution of complexes 131 8.4. Quartier de Skardhana: distribution of complexes 132 8.5. Quartier du stade: distribution of complexes 133 9.1. Delos, Maison du Lac; Ano Siphai, small house 143 9.2. Olynthus, House A, House C 145 9.3. Athens, Kerameikos, Building Z 148 9.4. Olympia, Leonidaion; Epidauros, Katagogion 151 9.5. Corinth, South Stoa; Nemea, Xenon 153 9.6. Kassope, Katagogion 154

Table 8.1.

Delos, mixed complexes and mixed/taberna complexes

127

Chapter 1

Introduction Lisa C. Nevett

The aim of this volume is to bring together a series of case-studies in which the archaeological evidence for housing is used to address a variety of questions about Greek households. Our focus is on settlements in Greece itself and Asia Minor during the Archaic, Classical, and Hellenistic periods (Map 1.1), and the chapters included here form part of a wider body of research in which new ways of approaching this material are being explored, both to provide fresh angles on familiar questions and to open up new areas for investigation. Also implicit is a testing of different ways of thinking about and interpreting domestic architecture and its associated Wnds, and an exploration of the range of problems to which such data can be applied. The use of domestic architecture and assemblages as a source in this way is a relatively recent phenomenon: in the past, the small scale and simple construction of most Archaic and Classical Greek houses meant that they received only limited attention in comparison with the contemporary public architecture. During the late nineteenth and early twentieth centuries there was a tendency to combine discussion of the layout of archaeologically recovered housing with what little could be gleaned about architecture and spatial organization from textual sources. Thus, while the newly-discovered remains of Late Bronze Age palatial structures were used to illuminate the aristocratic dwelling of the “Homeric world,” and the Hellenistic houses being excavated on Delos provided a quasi-evolutionary link between Greek and Roman domestic life, the lack of any Classical houses with comparable levels of preservation meant that dwellings of this date were necessarily understood principally through textual evidence.1 The twentieth century brought a progressive enrichment of the archaeological database, including a range of evidence for houses of the Wfth and fourth centuries. Detailed studies of individual sites began early on to concentrate on increasingly detailed descriptions of the architectural, and subsequently also the artifactual, evidence.2 Key sites include Olynthus

Map 1.1. Greece and the Aegean with sites mentioned in text (see also Maps 2.1 and 3.1).

Introduction

3

in northern Greece, where an absence of public architecture led to the excavation of large numbers of houses and detailed recording of the architecture and many of the associated small Wnds, with the aim of providing evidence for daily life in the Classical period. Work at Zagora on Andros has offered an insight into the organization of domestic and public space during the eighth century, at the time when the Classical polis was starting to emerge. Further extensive excavation on Delos has also been important because of the excellent preservation of the domestic architecture and the complementary information about the surrounding public buildings, which have enabled detailed analysis of decorative schemas and of the urban environment as a whole.3 While textual sources have not been entirely absent from these discussions,4 the main emphasis has been on assembling and ordering detailed information relating to the archaeological remains of the houses. In some instances comparisons have been drawn between structures from different sites in order to highlight similarities and contrasts in architecture and layout.5 One of the main features of this work has been the use of a typology based on the architecture surrounding the central open courtyard of houses at different sites excavated during the early twentieth century. In most cases, the entrances to the individual rooms lead off here and are sheltered by a covered portico. This sometimes takes the form of a recessed prostas, opening only into a limited number of rooms. In other examples, a longer colonnade projects into the court forming a pastas, which is sometimes extended around all four sides of the court to form a peristyle. In a fourth type, more recently deWned, some of the rooms are linked via a large, covered hearth-room or Herdraum, rather than through the open court. This typology functions as a descriptive tool for discussing the basic layout of houses, although it need not necessarily always imply fundamental differences in the ways in which space was originally used in these houses.6 Alongside this largely descriptive work, there has recently been an increased emphasis on interpretation. This has often continued to rely heavily on models drawn from textual sources in an effort to understand how different spaces were used.7 The most comprehensive and best-known example is W. Hoepfner and E.-L. Schwandner’s volume Haus und Stadt im klassischen Griechenland (1994; a second, revised edition of the original, published in 1986) in which detailed exploration of the organization of household space at a number of different sites is juxtaposed with consideration of the public architecture and layout at those sites. Reconstructions of individual buildings and of city plans as a whole are all geared to supporting a general thesis that the ideals of equality seen in some philosophical texts were put into practice in the classical polis in allotting public and private space. Importantly, the work is one of the

4

Lisa C. Nevett

Wrst to make the vital link between architecture and the broader cultural context in which it was constructed.8 This link is vital because it implies that the material record can be used as a source through which to explore speciWc social problems, as well as simply offering a picture of “daily life,” which was the most ambitious aim of most previous studies. Recognition of the potential of the archaeological evidence for Greek housing has opened the way for a new generation of work which has sought to move beyond the descriptive approach, using this material as a relatively independent form of evidence. As a result we now have a clearer picture of households and their relationship with the broader cultural context. During the Dark Age and Archaic periods there was an increasing architectural distinction between different types of activities, and recognizably different living, civic, and cult spaces emerged.9 Apsidal houses, which had been common in some areas, disappeared. At the same time there was an increase in house size and a radical change in spatial organization: by the late Wfth and fourth centuries the earlier two- or three-room structures with sequentially arranged rooms had been transformed. Many now offered as many as eight to ten different spaces, reached individually via a central open court.10 These houses normally possess only a single street entrance, and the central court acted as a distributive space, channeling communication between the different rooms. When large numbers of houses are compared, a similar picture of spatial organization emerges across Greece and Asia Minor, and also in the Greek cities of Sicily and southern Italy. This arrangement has been termed the “single-entrance, courtyard house.”11 The majority of rooms in such houses have yielded little in the way of specialized architecture and contained a range of different types of Wnds, suggesting a relatively Xexible pattern of usage which may have varied through the day or between seasons.12 It also seems that storage of goods and produce, together with small-scale workshops for craft production, formed a major component of many houses (see Cahill, Chap. 4; Tsakirgis, Chap. 5), and that a range of strategies were adopted for supporting the household, with the emphasis varying between storage and craftproduction in a manner which contradicts any assumption that all households would have aimed at self-sufWciency.13 A few rooms do stand out as having been designed with speciWc functions in mind, because of the architectural features they contain (although not every house has an identiWable example of each type, and many spaces may have been used for more than a single purpose). DeWnable elements regularly found at numerous sites include food preparation facilities with traces of ash and coarseware vessels, and an adjacent bathing facility furnished with water-proof plaster and sometimes a terracotta bath-tub (for an example, see house AII.5–6 at Leukas, discussed by Fiedler in Chap. 7). Perhaps the most

Introduction

5

widely discussed complex of features occurs in a room convincingly identiWed as an andron, which may be equipped with facilities that include a decorated Xoor and/or walls, a decorated anteroom, a marked border for couches, and an off-centre doorway to allow for positioning of couches around the walls. The evidence for couches suggests that one function of such rooms may have been as a setting for symposia. The presence of the andron—a separate male space—is one of the most dominant images of the Classical Athenian household projected by the few relevant textual sources. A parallel female space—the gunaikon—is also frequently mentioned. Modern scholars have usually interpreted these references to mean that Greek houses were divided spatially into separate male and female quarters. Nevertheless, very few houses exhibit the kind of binary structure which might support such a hypothesis, so that it seems likely that in most cases a more complex distinction is being drawn between the main area of the house, used by members of the family, and the andron, where male guests may have been secluded.14 The provision of a decorated andron is part of a wider change which saw the house increasingly playing a symbolic role, designed to project an image of the wealth and status of the householder.15 It has been argued that the changing form and role of the house is closely linked with changes taking place in the broader social and political spheres, including the demise of the democratic polis.16 In each case, these conclusions have emerged from the identiWcation of patterns in the organization of domestic space. In contrast with previous research, the approach taken is normally inductive, rather than relying on textual evidence to frame hypotheses which are then tested using the archaeology. Variability from site to site in the degree of preservation and the extent to which different components of the material record (particularly the Wnds) have been recorded and published, means that analytical methods have been tailored to suit the material available in each instance. At the same time, the speciWc issues addressed in each case require emphasis on differing aspects of the evidence, in order to extract the most relevant patterns in the data.17 There has, nevertheless, been a growing consensus on which aspects of the material record are likely to be important, with increasing attention paid to recording and analyzing Wnds and their distribution, and to contextualizing architectural and decorative elements. In addition, statistical analysis is becoming an increasingly frequent tool in order to identify patterns in larger samples of data. Following the example set by modern architectural analysis, attention has also focused on patterns of communication between rooms.18 The degree to which textual sources can and should be incorporated is still somewhat controversial. In rare instances texts do help relatively directly to illuminate particular points related to the archaeology: at

6

Lisa C. Nevett

Delos, collation of epigraphic evidence has revealed details about the customary make-up of farms on the island ( J. H. Kent 1948) and the local use of architectural vocabulary provides insights into the use of architectural space on the island (Hellmann 1992). For the majority of sites, however, there is little or no information of this sort, so that although they are vital in helping to establish the general cultural context, texts play no direct role as evidence for domestic spatial, or indeed social, behavior. Many recent studies of Greek domestic space therefore start from a common basic assumption and draw on a range of shared approaches to the evidence. What has been lacking, however, is any wide-ranging and systematic way of establishing the broader social and cultural signiWcance of many of the patterns identiWed in the data—a crucial step, since a single pattern might have a variety of viable alternative explanations. Research taking place on households in other archaeological contexts and also on households in some modern cultures, has aimed to bridge this gap by looking for systematic relationships between spatial organization and social behavior, and a few very general cross-cultural associations have been identiWed.19 These are, however, of limited utility in the ancient Greek context, where the resolution of the connections established to date cannot match the complexity of the issues under investigation, because of the relative sophistication of our understanding of Greek culture, both through archaeology and through textual and iconographic sources. For this reason, the precise nature of the relationship between domestic space and society, and also to some extent the methods by which this can be studied, currently need to be explored in a speciWcally Greek context. The present volume is intended both to contribute to this process and to broaden the range of issues tackled. A major aim is to seek a better understanding of the extent of, and reasons for, variability in domestic organization. Relatively little has been done to look for patterning on a regional, rather than site-based level, and the main focus of such work has inevitably been on constructing normative explanatory models. While these are a useful Wrst step in trying to understand Greek households, attention also needs to be paid to highlighting and exploring factors involved in variation between households at different sites, in different areas, at different periods and/or belonging to different social groups. Our emphasis here is therefore on exploring groups of households which are beyond the scope of the normative models recently put forward (Hoepfner and Schwandner 1994; Nevett 1999). We look at patterns of spatial organization in houses lying outside the central Classical period (Lang, Chap. 2; Trümper, Chap. 8); at communities relatively distant from the areas on which most current models are based (Cahill, Chap. 4; Fiedler, Chap. 7; Aylward, Chap. 3); and at social groups which

Introduction

7

are atypical of those most commonly studied (Tsakirgis, Chap. 5; Trümper, Chap. 8; Nevett, Chap. 6; Ault, Chap. 9). In the course of the volume we begin to examine a range of intersecting issues which have not previously been studied extensively using this form of evidence, including the roles played by marginal social or ethnic groups (Cahill, Chap. 4; Tsakirgis, Chap. 5; Trümper, Chap. 8; Nevett, Chap. 6; Ault, Chap. 9) and the relationship between households and their wider social, cultural and political contexts (Aylward, Chap. 3; Nevett, Chap. 6; Trümper, Chap. 8; Cahill, Chap. 4). By juxtaposing a variety of evidence, questions and approaches in a single volume in this way, we aim to explore the range and versatility of the archaeological remains of households as a source, while at the same time contributing to a broader understanding of the relationship between social behavior and domestic organization in the Greek world. Our goal is to shed some light on variability in households and to push forward the boundaries of what can be done using this type of material, providing a collection of papers which are of value both individually and as a group. Dates are B.C. throughout unless otherwise speciWed. Spellings of ancient place names are those in most common usage. In editing the chapters by Lang, Fiedler, and Trümper, certain expressions translated directly from the German have been retained in order to remain consistent with the authors’ other published work and with the scholarly tradition in which those authors are working. Notes 1. For example, Rider 1916. This phenomenon is explored further in Nevett in press. 2. Examples of this approach are Paris 1884; Couvé 1895; Robinson 1929– 1952; Holland 1944. This remains one of the most dominant and important forms of scholarship on Greek houses: for example Grandjean 1988; Makaronas and Giouri 1989; Ducrey et al. 1993; Reber 1998; Siebert 2001. 3. Olynthus: Robinson 1929–1952; Zagora: Cambitoglou et al. 1971; Cambitoglou et al. 1988; Delos: principally the Exploration Archéologique de Délos Series, especially Chamonard 1923, 1924, 1933; Deonna 1938; Bruneau et al. 1970; Siebert 2001. 4. For instance, Robinson’s publication of the material from Olynthus included sections entitled “Testimonia Selecta ad Domum Graecam Pertinentia” and “Reference List of Some Greek Words Concerned with the House” (Robinson 1946, 399–452, 453–471). 5. For example Graham 1972; Jones 1975; Krause 1977. 6. The prostas was originally identiWed at Priene (Wiegand 1904, 285–287). The term pastas was Wrst applied in the context of Olynthus, initially to a large room (Robinson 1930, 46), and subsequently to the portico (Robinson et al. 1938, 143–144, with Wg. 5)—the meaning which it has retained in later work on

8

Lisa C. Nevett

Greek houses. Use of the word peristyle became prominent in the context of Delos (Chamonard 1922, esp. 125). The Herdraum was deWned based on the housing Ammotopos and Kassope, both in Epirus (Dakaris 1989, 38–41, 43–47; Hoepfner and Schwandner 1994, 146–150). For further discussion of the deWnition and utility of these categories, see Aylward, Chap. 3; Nevett 1999, 22–25, 103; Nevett in press. 7. For example Reber 1988; Pesando 1987, 1989; Karadedos 1990. 8. Another example is Susan Walker’s brief article on women in Greek households (Walker 1983). 9. On space in general, see Lang 1996; on the separation between private and cult spheres, see Mazarakis Ainian 1997. 10. For further detail on these changes, see Lang, Chap. 2; Lang 1996, 104– 108; Nevett 2004; and Nevett in prep. chap. 2. 11. Nevett 1995, 1999, 103. 12. Nevett 1999, 37–38; Cahill 2002, 150–153; Ault in press, chap. 3. Compare Foxhall 2000, 491. 13. Cahill 2002, esp. chap. 6; Ault in press, esp. chap. 4. At Olynthus Cahill also identiWes a spatial zoning within the city of households which seem to have engaged in similar strategies (Cahill 2002, 224–225). 14. Interpretations of speciWc houses as composed of separate male and female spaces include Walker 1983 and Karadedos 1990. Doubts about the workability of the binary model are expressed by Jameson 1990a,b. For reinterpretation of the andron-gunaikon distinction, see Raeder 1988, Nevett 1995, 1999, esp. 155. 15. Kreeb 1988; Kiderlen 1995; Trümper 1998; Nevett 1999; Westgate 1997– 98; Walter-Karydi 1998. On other aspects of the use of the house as a symbol of wealth, see also Nevett, Chap. 6. 16. Walter-Karydi 1998. 17. Thus, even where the same body of data is being analyzed, different methods can be adopted (for example the analyses of the material from Olynthus by Nevett and Cahill which explore different aspects of those households: compare Nevett 1999, chap. 4, which focuses on drawing out similarities between structures, and Cahill 2002, passim, where the emphasis is on identifying and interpreting variation). 18. This is an aspect of spatial organization stressed by Hillier and Hanson (1984, esp. 143–151) and used in Nevett 1999. Compare also Lang, Chap. 2. 19. For example S. Kent 1990a,b; Rapoport 1990. Literature Cited Ault, B. A. In press. The Houses. The Organization and Use of Domestic Space. Excavations at Ancient Halieis, 2. Bloomington: Indiana University Press. Bruneau, P., C. Vatin, U. T. Bezerra de Meneses, G. Donny, E. Lévy, A. Bovon, G. Siebert, V. R. Grace, M. Savatiannou-Pétropoulakou, E. Lyding Will, and T. Hackens. 1970. L’îlot de la maison des comédiens. Exploration Archéologique de Délos 27. Paris: École Française d’Athènes. Cahill, N. D. 2002. Household and City Organization at Olynthus. New Haven, Conn.: Yale University Press. Cambitoglou, A., A. Birchall, J. J. Coulton, and J. R. Green. 1988. Zagora 2: Excavation of a Geometric Town on the Island of Andros. Excavation Season 1969; Study Season 1969–1970. Athens: Archaologikis Hetairias.

Introduction

9

Cambitoglou, A., J. J. Coulton, J. Birmingham, and J. R. Green. 1971. Zagora 1: Excavation of a Geometric Town on the Island of Andros. Excavation Season 1967; Study Season 1968–9. Sydney: Sydney University Press. Chamonard, J. 1922–1924. Le quartier du théatre. Exploration Archéologique de Délos 8.1–3. Paris: École Française d’Athènes. ———. 1924. Le quartier du théatre, 3: construction et technique. Exploration Archéologique de Délos 8. Paris: École Française d’Athènes. ———. 1933. Les mosaïques de la maison des masques. Exploration Archéologique de Délos 14. Paris: École Française d’Athènes. Couvé, L. 1895. “Fouilles à Délos.” Bulletin de Correspondence Hellenique 19, 460– 516. Dakaris, S. I. 1989. Κασσπη: Νετερ ες ανασκαϕ ς (1977–83). Ioannina: Ioannina University. Deonna, W. 1938. Le mobilier délien. Exploration Archéologique de Délos 18. Paris: École Française d’Athènes. Ducrey, P., I. Metzger, and K. Reber. 1993. Eretria: Fouilles et recherches 8: le quartier de la Maison aux Mosaïques. Lausanne: École Suisse d’Archéologie en Grèce. Foxhall, L. 2000. “The Running Sands of Time: Archaeology and the ShortTerm.” World Archaeology 31, 484–498. Graham, J. W. 1972, “Notes on Houses and Housing-Districts at Abdera and Himera.” American Journal of Archaeology 76, 295–301. Grandjean, Y. 1988. Études Thasiennes 12: recherches sur l’habitat Thasien à l’epoque grecque. Paris: École Française d’Athènes. Hellmann, M.-C. 1992. Recherches sur le vocabulaire de l’architecture grecque d’après les inscriptions de Délos. Paris: Boccard. Hillier, B. and J. Hanson. 1984. The Social Logic of Space. Cambridge: Cambridge University Press. Hoepfner, W. and E.-L. Schwandner. 1994. Haus und Stadt im klassischen Griechenland. Wohnen in der klassischen Polis 1. Rev. ed. Munich: Deutscher Kunstverlag. Holland, L. B. 1944. “Colophon,” Hesperia 13, 91–171. Jameson, M. H. 1990a. “Private Space and the Greek City.” In The Greek City from Homer to Alexander, ed. O. Murray and S. Price, 171–195. Oxford: Clarendon. ———. 1990b. “Domestic Space in the Greek City-State.” In Kent 1990c, 92–113. Jones, J. E. 1975. “Town and Country Houses of Attica in Classical Times.” In Thorikos and the Laurion in Archaic and Classical Times, ed. H. Mussche, P. Spitaels, and F. Goemaere-De Poerck, 63–140. (Miscellanea Graeca 1). Ghent: Belgian Archaeological Mission in Greece Karadedos, G. 1990. “Υστεροκλασικ σπ τι στη Μαρωνε α Θρκης.” Egnatia 2, 265–297. Kent, J. H. 1948. “The Temple Estates of Delos, Rheneia and Mykonos.” Hesperia 17, 243–338. Kent, S. 1990a. “Activity Areas and Architecture: An Interdisciplinary View of the Relationship Between Use of Space and Domestic Built Environments.” In Kent 1990c, 1–8. ———. 1990b. “A Cross-Cultural Study of Segmentation, Architecture, and the Use of Space.” In Kent 1990c, 127–152. ———, ed. 1990c. Domestic Architecture and the Use of Space. Cambridge: Cambridge University Press. Kiderlen, M. 1995. Megale Oikia: Untersuchungen zur Entwicklung aufwendiger griechischer Stadthausarchitecktur von der Früarchaik bis ins 3 Jh.v.Chr. Hürth: Verlag Martin Lange.

10

Lisa C. Nevett

Krause, C. 1977. “Grundformen des Griechischen Pastashauses.” Archäologische Anzeiger 1977, 164–179. Kreeb, M. 1988. Untersuchungen zur Figürlichen Ausstattung Delischer Privathäuser. Chicago: Ares. Lang, F. 1996. Archaische Siedlungen in Griechenland: Struktur und Entwicklung. Berlin: Akademie Verlag. Makaronas, C. and E. Giouri. 1989. Οι οικες αρ παγς Ελνης και ∆ιονσον της Πλλας. Athens: Athens Archaeological Society. Mazarakis Ainian, A. 1997. From Rulers’ Dwellings to Temples: Architecture, Religion and Society in Early Iron Age Greece (1100–700 B.C.). Jonsered: Paul Åströms Forlag. Nevett, L. C. 1995. “Gender Relations in the Classical Greek Household: The Archaeological Evidence.” Annual of the British School at Athens 90, 363–381. ———. 1999. House and Society in the Ancient Greek World. Cambridge: Cambridge University Press. ———. 2004. “Domestic Space as a Means of Exploring Social Change: Household Organisation and the Formation of the Classical Greek Polis.” In Wohnformen und Lebenswelten, ed. A. Hoffmann and M. Droste. 11–20. Frankfurt: Peter Lang Verlag. ———. In press. “Greek Houses as a Source of Evidence for Social Relations.” In Building Communities: House, Settlement and Society in the Aegean, ed. R. Westgate, N. Fisher, and J. Whitley. Athens: British School of Archaeology. Studies of the British School of Archaeology at Athens. ———. In preparation. Domestic Space and Social Organisation in Classical Antiquity. Cambridge: Cambridge University Press. Paris, P. 1884. “Fouilles de Delos: maisons du second siècle av. J.-C.” Bulletin de Correspondence Héllenique 8, 473–496. Pesando, F. 1987. Oikos e Ktesis: la casa greca in eta classica. Perugia: Quasar. ———. 1989. La casa dei greci. Milan: Longanesi. Raeder, J. 1988. “Vitruv, de architectura VI 7 (aediWcia Graecorum) und die hellenistische Wohnhaus und Palastarchitektur.” Gymnasium 95, 316–668. Rapoport, A. 1990. “Systems of Activities and Systems of Settings.” In Kent 1990c, 9–20. Reber, K. 1988. “AediWcia Graecorum: zu Vitruvs Beschreibung des grieschischen Hauses,” Archäologischer Anzeiger 1988, 653–666. ———. 1998. Die klassischen und hellenistischen Wohnhäuser im Westquartier. Eretria. Ausgrabungen und Forschungen, 10. Lausanne: Payot. Rider, B. C. 1916. The Greek House: Its History and Development from the Neolithic Period to the Hellenistic Age. Cambridge: Cambridge University Press. Robinson, D. M. 1929–1952. Excavations at Olynthus. 14 vols. Baltimore: Johns Hopkins University Press. ———. 1930. Excavations at Olynthus. Vol. 2. Architecture and Sculpture: Houses and Other Buildings. Baltimore, Johns Hopkins University Press. ———. 1946. Excavations at Olynthus. Vol. 12. Domestic and Public Architecture. Baltimore, Johns Hopkins University Press. Robinson, D. M., and J. W. Graham. 1938. Excavations at Olynthus. Vol. 8. The Hellenic House. Baltimore: Johns Hopkins University Press. Siebert, G. 2001. L’îlot des bijoux, l’îlot des bronzes, la maison des sceaux. Exploration Archéologique de Delos 38. Paris: Éditions E. de Boccard. Trümper, M. 1998. Wohnen in Delos: Eine baugeschichtliche Untersuchung zum Wandel der Wohnkultur in hellenistischer Zeit. Rahden: Verlag Marie Leidorf GmbH.

Introduction

11

Walker, S. 1983. “Women and Housing in Classical Greece.” In Images of Women in Classical Antiquity, ed. A. Cameron and A. Khurt, 81–91. London: Croom Helm. Walter-Karydi, E. 1998. The Greek House: The Rise of Noble Houses in Late Classical Times. Athens: Athens Archaeological Society. Westgate, R. 1997–98. “Greek Mosaics and Their Architectural and Social Context.” Bulletin of the Institute of Classical Studies 42, 93–115. Wiegand, T. and H. Schraeder, 1904. Priene: Ergebnisse der Ausgrabungen und Untersuchungen in den Jahren 1895–1898. Berlin: Georg Reimer.

Chapter 2

Structural Change in Archaic Greek Housing Franziska Lang

A household consists of a group of people sharing a common place of residence, who by virtue of their joint behavior, function as a social and economic unit.1 With the exception of widows, the unmarried, or cohabiting brothers and sisters, the basic element of the household in ancient Greece was the family, consisting of parents and children together. An exceptional form of household was one that comprised a group, unconnected by kinship, who shared the same architectural and structural dwelling complex (Ault, Chap. 9). In Archaic Greece, a household could have been composed of one or more families, sometimes including nonfamily members such as slaves (Laslett 1972, 25–38; Allison 1999b, 4). The size of a household and the internal division of a house depend, among other things, on the following factors: the occupants—whether they comprise a nuclear or an extended family; the rules of marriage—whether patrilocal, matrilocal, or ambilocal; the source of income—whether based on agriculture, transhumance or pastoralism; and the engagement of males in military service or politics (Pfälzner 2001, 27–34). These observations also show that a household has different dimensions: the architectural dimension of the house (as a co-residence), the economic activity of the household, and the family as a social unit. As is well known, in order to identify households archaeologically we have to rely on the architecture of houses and the Wnds within them. This is, however, not unproblematic. First, the ground plan of excavated houses is commonly used for the classiWcation of houses within a settlement. This is Wne if the settlement has only one occupation phase, but if it was inhabited over some generations, the size of the household, the composition of the family, the function of the house, and changing activities over time can produce changes to the ground plan. The later building phases of a house could differ completely, in function as well as in physical appearance, from the earlier phases. This fact must be considered during classiWcation, otherwise the assumption is that no change of

Structural Change in Archaic Greek Housing

13

any kind took place from the original foundation until the abandonment of the house. A second problem is that few settlements are excavated in their entirety, so that the ground plans of a few excavated houses that are not necessarily representative of the whole variety of existing house-types are taken, pars pro toto, as characteristic of the whole settlement, and its interpretation may even be based on a single house. These factors are especially critical where the aim is to reconstruct the social hierarchy of a settlement according to the ground-plan of the houses.2 This problem particularly affects the pre-Classical periods in Greece, when a standardized settlement plan is not yet a common feature. A third difWculty is that social structure in Greece is highly variable. For example, the household might be identical with a house as an architectural structure, but on the other hand the family might live in more than one house. In addition, the pattern of living differs regionally so that the transposition of results from one region to another should not be undertaken without detailed consideration of the potential for variability. Finally, a detailed analysis requires detailed excavation reports, but typically only the architectural features of domestic structures are reported in detail, without adding complementary information about the Wnds and exact Wnd locations. Fortunately, in recent years this attitude has changed and more studies present a detailed analysis of rooms and their inventories, so that critical examination and a comparative social analysis of the archaeology of Greek settlements are becoming increasingly feasible. The intense interest in ancient housing—nowadays very fashionable— has focused mainly on the domestic architecture of the Early Iron Age or the Classical and Hellenistic periods (Drerup 1969; Fagerström 1988; Hoepfner and Schwandner 1994).3 For a long time, the period in between, the Archaic, was not a particular topic of interest.4 Excavators digging the Classical period stopped before reaching the Archaic phases, or excavators interested in the pre-Archaic periods destroyed the Archaic strata often without sufWcient documentation. In a few settlements some houses were documented but most of their remains are quite sparse. Finally, when Archaic houses were excavated, only their architecture was typically published, without the Wnds. This means that interpretations relating to Archaic houses are, in most cases, based exclusively on architectural characteristics, and mostly wall foundations at that. Because of this fact, the approach to Archaic buildings is generally two-dimensional (concentrating on structural remains) and not three-dimensional (which would include consideration, for example, of their decoration, wall-painting, or windows), integrating artifacts and activity-areas for living, consumption, production, distribution, and so on (compare the approaches taken by Margueron 1983, 16–20; Ashmore and Wilk 1988, 4; and Allison 1999b, 4).

14

Franziska Lang

Even if this introduction sounds sceptical, and some points are indeed problematic and need more consideration, there are good reasons to be optimistic. The material is extensive enough to enable trends in Archaic housing to be described (as indicated in Map 2.1, showing sites mentioned in this chapter). Ultimately this period is particularly interesting because we are able to appreciate the fundamental changes which occurred in the Greek way of life from the Early Iron Age to Classical times. Indeed, the well-attested sociopolitical and artistic developments of the Archaic period are generally and to some degree reXected in domestic architecture as well.

Typology and Organization of Archaic Houses One means of interpreting housing is through a typology of ground plans and architectural design. Houses are classiWed in accordance with their ground-plan and size. For Early Iron Age houses such a typology was developed by Drerup (1969). A fundamental difference exists between apsidal or oval and rectangular ground plans. Rectangular houses are subdivided into further types, such as antae or square houses. In addition, Drerup distinguished a “Langhaus” from a “Breithaus” based on the position of the entrance in the Early Iron Age. In the former, the entrance is on the short side, in the latter the entrance is on the long side of the house. These house-types are common in the Early Iron Age (Fagerström 1988; Mazarakis Ainian 1997b). From the tenth century B.C. onward houses with oval or apsidal shape spread throughout the whole of Greece with the exception of Crete. Examples of such houses are known at Assiros, Athens, Eretria, and Smyrna (Drerup 1969; Mazarakis Ainian 1997b, 43–113). At the same time, different types of rectangular houses can be studied, for example at Emporio on the island of Chios, founded in the eighth century (Boardman 1967). Sprawling down the slope of a hill, these houses were erected in isolation. The settlement reveals one-room or two-room houses and antae houses of various sizes (Fig. 2.1a–b). The houses of the Early Iron Age had a simple ground-plan and usually one room, or two rooms lying one behind the other (Fig. 2.1a–d), although in Crete there is some variation, as at Vrokastro (Hayden 1983; Lang 1996, 78–83). An obvious change in house-types can be recognized in Greece from the seventh century onward. Some of the former house-types disappear entirely (for example, the square “Breithaus”) and there is a remarkable decline of the apsidal house-type (Map. 2.2). That this process was not accidental can be seen at Eretria (Fig. 2.1d) and Miletus, where apsidal houses are remodeled, or directly overbuilt by rectangular-shaped ones (Lang 1996, 85; Morris 1998, 14). This seems to suggest that from the

Map 2.1. Greece and the Aegean with Archaic sites mentioned in Chap. 2.

Fig. 2.1. a–b. Emporio, Chios; c. Thorikos, Attica; d. Eretria, Euboea; e. Aigina; f. Limenas, Thasos; g. Dreros, Crete; h. Koukounaries, Paros; i. Onythe, Crete; j. Vroulia, Rhodes; k. Kopanaki, Messenia.

Structural Change in Archaic Greek Housing

17

Late Geometric period there was a planned replacement of apsidal and oval houses. In the northern Greek settlement at Assiros, however, the contrary development took place. A pre-existing, large, rectangular housecomplex was replaced by two large, apsidal houses in the Late Geometric period. The shape of the antae or apsidal houses survived in the Archaic period in other parts of Greece, but was now used for sacred purposes, namely, as temples (Lang 1996, 73). Apart from this, new house-types appeared and exhibit a general pattern ranging from one- or two-room houses to multiple-room houses (Fig. 2.1). The ground-plans of the latter are compact and have a common access-area connecting the rooms. A further fundamental change in the Archaic period affected the arrangement of rooms. Since the Early Iron Age, the rooms had been situated one behind the other. In Archaic times they were placed next to each other paratactically, thereby creating a radial arrangement (Fig. 2.1e–k). In front of the back rooms at least one additional room was constructed. These new ground plans can be

Map 2.2. Map showing the distribution of apsidal and oval houses in the lateGeometric and Archaic periods.

18

Franziska Lang

categorized according to their anterooms. Dwellings in Aegina (Fig. 2.1e), Thasos and Corinth had a three-room scheme: one room in front of two paratactically arranged rooms. A pastas -like anteroom is known in Corinth, Onythe (Fig. 2.1i) and Xobourgo. Alternatively, the rear rooms were preceded by a court, as at Vroulia on the island of Rhodes (Fig. 2.1j) or the houses on the Kalabaktepe at Miletus, where these houses appear in a row. Examples of compactly arranged rooms can be seen in the approximately square houses with a couple of rooms from Dreros on Crete (Fig. 2.1g; Drerup 1969) or Koukounaries on Paros (Fig. 2.1h; Schilardi, 1987, 228–231; Kiderlen 1995, 26–27). A similar compact design is found in multi-room houses at Kopanaki in Messenia (Fig. 2.1k) and Tsikalario on the island of Naxos. These new house-types reXect changes in society to be discussed below. Another signiWcant change occurred in Archaic times with the separation of private and ofWcial buildings. A different political and social organization required speciWc types of buildings, whereupon new types of architecture arose. This started with special buildings for the gods, with the earliest temples being erected in the late eighth century. It ended with different buildings for each political authority, a pattern best documented in Athens. There is no space here to expand on this topic but it is a further indication of the differences between Early Iron Age and Archaic society. In addition to house-types themselves, the arrangement of houses within settlements informs us about the intention of the inhabitants in structuring space, and about the composition of the community. To this end one should distinguish between single-phase and multi-phase sites. In single-phase settlements the original founding-plan and the intention associated with it is still recognizable, whereas in multi-period sites changes in house-structure over time are quite likely (as noted above). Generally, changes can affect settlements partially or as a whole and can be expressed in the remodelling of former houses and/or erecting new houses. Various factors are responsible for this. Changes in individual houses are often connected with the sociobiological cycle of marriage, reproduction, death, the changing of generations, and so on. Signs of this may to some extent be visible in the house. An increase in house size could be the result of a birth, or a reduction by death (Pfälzner 2001, 34–35). In Greece, patrilocal residence was predominant. In architectural terms this meant that when a son married he continued to live in the house of his parents, perhaps adding onto it. This was only possible in settlements with detached houses. In settlements with an agglomeration of houses, such extensions could not be made, so either remodelling subdivided the interior of the house or the son built a new residence elsewhere in the settlement. It is important to

Structural Change in Archaic Greek Housing

19

note that in the latter case the numbers of houses within a settlement would increase, without there being an actual growth in population. Changes in the overall settlement could result from a variety of factors: such modiWcations were accepted by the whole community; the leading group exerted pressure; external powers put pressure on the community; or economic factors forced modiWcation. The particular explanations which are valid in each case must be studied through detailed analysis of the archaeological facts. The arrangement of houses within a settlement indicates the relationship between the settlers and the extent of formalization of the inhabited space. In general, the settlements in the Early Iron Age are of two types, with houses erected at a distance from one another, as at Koukounaries (Schilardi 1978, 195–210; 1979, 236–248), or with houses forming an agglomeration, as at Ag. Andreas on the island of Siphnos (Philippaki 1978, 192–194). In both instances, the arrangement was haphazard without any regular network of streets. During the Geometric period at Zagora on the island Andros, large one-room houses and some two-room houses were agglomerated (Cambitoglou et al. 1988; Fig. 2.2). In the second half of the eighth century, houses were remodeled by subdividing the interior and adding new rooms. The one-room houses of the earlier phase (Fig. 2.2a) were replaced with multi-room dwellings (Fig. 2.2b). This modiWcation might have been necessary due to the expansion of families, and stands as an example of substantial structural change in a settlement leading to the increased standardization of housing and settlement structure. An example of a settlement plan standardized from the time of its conception is Vroulia, on the island of Rhodes, founded at the earliest in the seventh century (Kinch 1914; Fig. 2.1j). The houses form a row, with two or more rooms behind a court. The rooms lie adjacent to one another, in a paratactic arrangement. In this settlement no subsequent modiWcation occurred, so the visible plan mirrors the original conception of the founders: planned rows of houses.

Internal Organization of Houses Up to this point I have considered the relationship between the house and its environment, the surroundings of the outside world. I now turn to the house itself (cf. Jameson 1990). Besides the ground-plan, the furnishing and arrangement of rooms are indicators of the functional and social conceptions of dwellings. I do not, however, intend to reconstruct social structure in detail, since in most settlements only a few houses have been excavated, the Wnds are often not published in detail, and other sources of data are lacking. My intention is rather to show which

20

Franziska Lang

parameters can give us an idea about structural changes in Archaic Greek housing, and to describe some trends which need to be explored in further studies. There are many dimensions of the house which relate to the social, economic, technological, sociopsychological, symbolic, functional, and representative spheres. The Social Sphere The “social space” (Parker Pearson and Richards 1997) in a house can be deWned with reference to the number of rooms, the subdivision of the house, and the position of doorways that indicate the nature of internal communication. The number of rooms affects internal differentiation in terms of function, arrangement and segregation. In one-room houses multi-functional utilization is dominant, whereas in multi-room houses the possibility of a mono-functional use for each room within the house increases. In the Early Iron Age one- and two-room houses predominate (Mazarakis Ainian 1997b, tables I–VIII). The arrangement of rooms is sequential; because one room is situated behind another, the rear room was only accessible via the front room. In the late eighth century, the Wrst examples of houses with a more complex pattern of spatial organization appeared (for example, at Zagora, Fig. 2.2.b). The houses had more than two rooms but a linear arrangement was still dominant, not to be replaced by a radial room arrangement until the Archaic period. Now, the rooms were placed alongside one another in a paratactic manner. Furthermore, the houses included a courtyard, as seen at Zagora, Miletus, Limenas in Thasos; or a corridor, for example, at Athens and Dreros (Lang 1996, 94–97). As mentioned above, multi-period sites give us the chance to study changes in the architecture and activity-areas of houses as a result of functional, social, and other shifts over time. These changes—distinguishable through the different phases of the house—were a result of the developmental cycle of the household (Pfälzner 2001, 34). This developmental cycle was, in turn, inXuenced by both the family cycle (its sociobiological development) and the economic cycle (the development and extent of household activities). Changes can be recognized at multi-period sites like Zagora (Fig. 2.2. a; Cambitoglou et al. 1988). Here, the older houses were surrounded by open spaces. Within the houses were benches which might have been used for sitting, sleeping or as storage/work space. All the tasks of daily life were carried out in the one room of the house, which was multi-functional in nature. There were no divisions between a living-room (in this sense of oikos) and storage-room. If there were a need to subdivide areas, this would have been determined by social convention; for instance, an area

Fig. 2.2. Zagora, Andros. a. Houses of the Wrst half of the eighth century; b. houses of the second half of the eighth century.

22

Franziska Lang

for men separated from that for women, for older people from that for younger people, and so on. Privacy, in the modern sense, could not have existed within such a house. Curtains or some similarly ephemeral material could have accomplished little more than to restrict visibility. Naturally, when weather permitted, work would have taken place outside the house. Similarly, family life could also have played out in the public sphere, since many family activities could have taken place either inside or outside the house. The entire settlement could be viewed as the setting for daily life. Even if the houses were encircled by a fence, transparency was still there. In the second half of the eighth century, Zagora witnessed an essential change that affected the overall character of the settlement. Each house was given a courtyard, a small, a middle-sized, and a large room, which was the former central room of the one-room house (Fig. 2.2b). A long wall divided the new settlement into a western area with houses in which the rooms were arranged in linear gradation and an eastern area where the rooms had a more radial arrangement (Lang 1996, 104–105). This alteration might indicate changes in the internal structure of the community and family. The reproduction of the same house-type across the settlement seems to show that it was carried out with the consent of all the residents of Zagora, or that the inhabitants were compelled by a leading group or external power.5 If common consent existed, then the degree of supposed social stratiWcation was not high,6 unless hierarchical relationships were expressed in a symbolic way difWcult to verify by archaeological methods, such as through differentiation by costume. Although there is no site other than Zagora where a similar transformation can be proven across a whole settlement, there are examples like Miletus where a comparable development from few-room houses to multipleroom houses can be seen. Looking over the evidence for Greece, one can identify various regional patterns. In particular, a completely different picture emerges in Macedonia. At Thessaloniki Toumba and Assiros complex houses with many rooms existed from the eleventh until the ninth centuries, which were remodelled in subsequent periods (see below). A house-complex built in this way was erected in Macedonian Kastanas in the Late Geometric period (Fig. 2.3a; Hänsel 1989, 232–254). This house-complex, with at least twelve rooms together with an adjacent street, has been completely excavated. Four rooms are connected internally by doorways. The excavated hearth and implements indicate different functions for each room and make it likely that the four rooms form one unit. Single rooms in this house-complex were not connected internally to the rest of the house; their entrances opened out to the street. In some of these single rooms hearths were found, possibly indicating that these were

Fig.2.3. Kastanas. a. Geometric houses; b. Archaic houses.

24

Franziska Lang

individual house-units. The one-room unit with hearth suggests a multifunctional area, while the four-room unit seems to be divided into different functional sections. This larger structure could be explained as the complex of an extended family living in separate units of the house or as one house complex shared by different families. In the Archaic period at Kastanas some changes can be identiWed. The large house-complex was reorganized. The four-room unit was altered to a two-room unit, while few of the former one-room units continued to exist. The other units disappeared in favour of an open space. In the later sixth century, another fundamental change can be observed (Fig. 2.3b). The former multiple-room house-complex was overbuilt with two-room detached houses (Hänsel 1989, 304–315). The rooms now were used multifunctionally. The new houses follow almost the same orientation as the former house-complex. A comparable shift occurred at other sites like Assiros (Wardle 1987, 315–320). Presumably, this remodeling was the result of signiWcant changes in social and economic structures in this region (Lang 1996, 105–106). Another instance of the modiWcation of houses can be seen at Miletus, but here the remodeling did not seriously affect the structure of the ground plan as it did at Zagora and Kastanas (Lang 1996, 208–212; Senff 2000). Such examples obviously reXect responses to new requirements within their respective communities. They had notable consequences for the internal organization of the house, and as such for the household and the way of life in the community. Internal communication within the house is determined by the number and arrangement of the rooms, the structure of access-ways and room functions. These parameters were a part of the innovations. The position of doorways within a house is indicative of the structure of internal communication. The way in which this internal communication changed decisively from the Early Iron Age to the Archaic period is best illustrated schematically. In Fig. 2.4, row A, entrances (crossed circle), rooms (empty circle), transitional areas (Wlled circle), and rooms with exactly two points of access (double circle) are each represented by different symbols. Lines indicate accessibility between rooms. Rooms with only one or two entrances are differentiated from transitional rooms. It is important to note that these access diagrams do not consider function, Wnds, or shape of the rooms in the house (Brown 1990, 94–99). Therefore different ground plans can be shown to have the same access pattern, as at Zagora, Kastanas and Vroulia. To consider the crucial factor of the internal systematics of dwellings, a second set of diagrams showing the size of rooms and their position in the house is included. Fig. 2.4, row R, employs the same conventions as Fig. 2.4, row A, adding smallest rooms (dotted circle), largest rooms (empty square), and largest rooms serving as transitional areas (Wlled square).

Fig. 2.4. Scheme of access (A) and room arrangement (R) in Geometric (italic) and Archaic houses: crossed circle = entrances; empty circle = rooms; Wlled circle = transitional areas; double circle = rooms with exactly two points of access; dotted circle = smallest rooms; empty square = largest rooms; Wlled square = largest rooms serving as transitional areas.

26

Franziska Lang

These diagrammatical representations of access patterns and room arrangements for examples of Early Iron Age (italicized ) and Archaic houses show that organizational differences are present. While in the Early Iron Age most of the rooms in houses were arranged in a linear series, Archaic houses had a more radial structure and possessed rooms adjacent to one other which each had separate access. As the houses still had only a single entrance, the new pattern of organization required a different access-system. A new type of room was therefore incorporated which functioned as a transitional space from which people could have access to all other parts of the building. This was a precondition for restricted or communal access, and supported the separation of rooms facilitating the distinction of function and of status within the household. This transitional space was usually a hall or a courtyard. Three different types of room now existed: the room with one door, a “walk-through” room with two doors, and a transitional space with at least three doors. These types dominated in domestic architecture for the subsequent periods, whereas the sequential arrangement of rooms, which was quite usual in the Early Iron Age, was maintained in later periods only in the sacred architecture of temples. Finally, room arrangement and accessibility are indicators of room function and the level of privacy. Degree of privacy can be measured in the number of rooms which were not walk-through rooms. Since in Early Iron Age houses rooms were arranged sequentially, walk-through rooms had always existed. Rooms in this sequence had an identical level of potential privacy, except for the back room, which logically offered the most seclusion. Rooms with separate access became usual after the Early Iron Age. It has been presumed that the function of the rooms with lower accessibility is more speciWc than that of the circulation space (Bernbeck 1997, 196). As already mentioned, with this system of access a greater diversity of habitation was possible inside the house: differentiation by age, gender, status, function, and so forth, could all be accomplished more easily if necessary.7 The consequences of the new room arrangement and access patterns were fundamental, and these principles did not change during the following centuries. The sizes of the dwellings and of each room are of interest too. Generally speaking, Early Iron Age houses were smaller than Archaic ones. This is a result of the greater number of rooms in Archaic houses. Many Early Iron Age houses contained one large room of considerable size, sometimes with a small anteroom, as at Emporio or Thorikos (Figs. 2.1 a–c). In houses with more than one room, the sizes of the rooms were generally not equal: usually one room was larger. From the late eighth century onward, there was a tendency towards conformity with one larger room and a few smaller, more-or-less equal-sized rooms. Often the common

Structural Change in Archaic Greek Housing

27

transitional space (a courtyard or hall) was the largest “room” in the dwelling, as is shown by the Wlled squares in Fig. 2.4 (Lang 1996, 87–103). Another aspect of social structure is gender ( Jameson 1990, 104). There is no solid evidence for the gender-speciWc division of space within Archaic houses such as we assume for the Classical period.8 And, while there are many scenes of male banqueting on painted pottery from the Archaic period, there is no evidence that such activities were already taking place in private houses, where there simply was no space for them. Perhaps, the gatherings depicted were held in some more communal structure, such as a hestiatorion. The interpretation of rooms in Archaic houses as andrones (Hoepfner 1999, passim) seems to be a transposition of the Classical situation. The Economic Sphere Economic aspects of a household can be inferred from the architecture, furnishings, tools, the identiWcation of activity areas, and various kinds of remains like slag, macro-botanical residues, bones, and so on. House construction is also an indicator of economic conditions, through the building materials and techniques used: the larger rooms of Archaic times demanded larger wooden beams, and stone foundations were used instead of the rubble masonry of Early Iron Age houses, and tiles instead of thatch, reeds, or stone slabs, and so on (Fagerström 1988, 106–123). In a subsistence-farming economy, installations and rooms with speciWc functions, such as stables or storage, were necessary. In the Early Iron Age pithoi or similar vessels placed on benches were quite common. In the Archaic period the number of houses with rooms containing benches and pithoi declined in favour of speciWc architectural storage features such as cellars (seen, for example, in examples of subterranean rooms at Corinth). In Vroulia and Onythe holes in the walls were excavated which were Wlled with sherds, bones, and other residues (Lang 1996, 116). The Technological Sphere Craftsmanship in a segmented society, like that of the Early Iron Age, is connected to what Sahlins referred to as the “domestic mode of production” (Sahlins 1974). In many Early Iron Age settlements indications of metal-working have been found in house-groups (as, for example, at Oropos: Mazarakis Ainian 1997a, Wg. 2). Archaic period furnaces and kilns have been excavated in public spaces like the agora (Athens) or temples (Olympia) and in houses (e.g., at Corinth) in different areas (Lang 1996, 166–167, 170).9 By this time the mode of production was no longer limited to the domestic domain. The new requirements of the rising

28

Franziska Lang

poleis for public building projects demanded a new form of collective production, as is demonstrated by new architectural features. Increasing familiarity with new materials also allowed for the development of useful tools and devices (e.g., clamps and dowels). Many of the new production techniques can only have been mastered by specialists, who were dependent upon close collaboration with others for the realization of a project. This required professionals, and the self-sufWcient subsistence economy began to be supplemented by the economy of product exchange, as demonstrated, for example, with the creation of the agora. Prestige objects such as bathtubs have been found in some Archaic houses, probably demonstrating the wealth or the status of the inhabitants (Lang 1996, 128–136). Craftsmanship and technological developments inXuenced architectural design. New skills determined to what extent building materials could be worked. New types of construction led to technological innovations. In Archaic times techniques of quarrying and dressing stones improved radically. This affected above all the stone walls of the houses: whereas in the Early Iron Age “the stone socle . . . is not to be regarded as a foundation . . . of a wall, but to keep the moisture of the ground off the mud brick wall-foot” (Fagerström 1988, 99), in the Archaic period a new masonry technique appeared having two important effects on house construction. First, new construction techniques utilizing headers and stretchers in ashlar masonry or well-dressed polygonal stones allowed the thicker wall that is the precondition for a second story. Second, mud-brick construction was reinforced by a transverse timber framework laid within the wall itself. This framework and the thicker walls could carry more weight, allowing more substantial beams to be used for the roof, so that supporting posts were no longer necessary. Therefore, in the seventh century the use of uprights started to decline. They had disappeared almost entirely by the sixth century (Fagerström 1988, 122–124; Lang 1996, 108– 111), with the exception of northern Greece, where walls were still erected using the technique of wattle and daub. Another signiWcant invention was the clay roof tile. The earliest examples of tiles in the post-Mycenaean period seem to have appeared on temples at the end of the eighth century (Schwandner 1990). Tile roofs have less weight than stone slabs and can span a greater distance without supports. A newly specialized mode of production allowed for the serial output of tiles and their standardization in size and shape. The Sociopsychological Sphere The sociopsychological dimensions of houses can be studied via the patterns of communication they enabled: Archaic houses were more compact, complex, and allowed different kinds of communication for the

Structural Change in Archaic Greek Housing

29

household, both within the domestic sphere and with the outer world. In the one- or two-room houses of the Early Iron Age one entered directly into the family sphere, whereas in Archaic houses the visitor Wrst enters a “neutral” area, a transitional space (generally a court or corridor). The inhabitants of houses without a courtyard would, upon leaving the house, immediately perform some of their domestic tasks in the outside world where a stranger could see them work: eye contact with neighbors could be made and unexpected encounters were possible. Houses closed to the outside world, such as Archaic courtyard houses, did not allow this kind of accidental contact. Communication could only occur upon intentional entry into the house. The courtyard of such houses could be used for any sort of outdoor work. As a result the residents were more isolated from their neighbours and from the outside world, and the possibilities of contact with that world were more controlled. One aspect of nonverbal communication is visibility between rooms. This can be understood from the way in which “pathways” through the house are organized: radial Archaic houses have a non-axial alignment of doorways and lack intervening spaces from which rooms branch off (Figs. 2.1e–k and 2.4). In contrast, the sequential arrangement of rooms in houses of the Early Iron Age is determined by a design with a clear visual connection between the rooms and axially aligned entrances (Figs. 2.1a–d and 6). A further aspect of non-verbal communication is the hierarchy of rooms expressed by different furnishings and decoration. A raised threshold may imply a psychological restriction on who could enter the house or the room. In addition, barriers like low walls or columns could also restrict visibility through the house. The Symbolic Sphere The symbolic dimension is a further aspect of domestic architecture. Symbols operate as a medium to convey information about, for example, the status of the household and its residents, ideas or beliefs, ideological or ethnic afWliations, a separation of private spheres within the house for individual residents, and so on. The door—like the city gate—is a point of transition from an outside world to an inside world and is highly symbolic (Carsten and Hugh-Jones 1995, 40–42). It mediates between the world of the collective (the male world) and the world of the family (the female world).10 Thresholds can symbolize the border between different spheres and indicate representative and spiritual rooms (Parker Pearson and Richards 1997, 25). It may be that the mere existence of a threshold, its construction, and the material used, indicates further different symbolic spheres within the house. Sills are less common in the Early Iron Age houses than in the later Archaic ones.

30

Franziska Lang

As the number of rooms in a house increases, so too does the probability that individual rooms served restricted functions. This evolution can be seen from the Early Iron Age to the Archaic period. The Early Iron Age one- or two-room houses had multi-functional rooms, whereas the Archaic multi-room houses offered more possibilities for mono-functional rooms. Functional divisions are most clearly indicated by changes in the architecture, material culture, and related features. The Representative Sphere In some houses conspicuous rooms have been excavated: in Corinth, where a couple of houses have rooms below ground level, and in Emporio, Vroulia, and Tsikalario, rooms without doorways were found. It seems likely that these rooms were used for storage (Lang 1996, 139). A further architectural hint as to the function of a room is the nature of its Xoor surface: for example, waterproof Xooring is presumably used for bathrooms or stables as at Smyrna or Corinth (Lang 1996, 168, 239). Installations like hearths or storage pits or circular stone Xoors give clear evidence of room function (Lang 1996, 114–117, 138–140). Whereas the construction and usage of hearths remained constant, the number of rooms containing storage pits declined in the Archaic period. Distinctive Wnds such as bathtubs (as at Vroulia and Smyrna) and fragments of pithoi (for example, at Onythe and Zagora), cooking ware or loomweights are also indicative of speciWc functions. Instances where several of these objects have been found together are identiWable as multifunctional spaces. This is often seen in courts or oikoi (main living rooms). These were the preferred locations for domestic activities such as cooking, eating, and talking, and normally they were not restricted in access, visibility, or by domestic rules. Often these rooms had multiple points of access. Single-access rooms ensure a subdivision of the house by function (be it working, eating, sleeping), and social structure (where status, age, or gender restrictions may have been enforced). The function of rooms and houses can also be understood in terms of their spatial context. At Dreros and Koukounaries multiple-room dwellings with similar ground plans (Fig. 2.1.g, h) are situated beside temples. In both dwellings distinctive Wnds were made: at Koukounaries a cache of seals (for which this dwelling was named “house of the seals”), at Dreros steatite vases. It may be that both buildings had some speciWc purpose connected with a cult or served as ofWcial buildings within the settlements (Lang 1996, 183, 188). Certain nonessential Wttings and furnishings were used for the decoration of the house, and the self-portrayal and representation of household members. These could be expressed either in sophisticated architectural decoration or by speciWc objects such as statues and terracottas. Innovations

Structural Change in Archaic Greek Housing

31

in the architecture of Archaic temples were adopted in the architecture of public and private buildings (Schwandner 1990). For example, in Thasian Limenas or Elian Babes the adoption of architectural features like anteWxes or tiles, Wrst used in sacred architecture, emphasized the representative character of the house (Lang 1996, 112). Ornamented fragments of furniture-legs demonstrate the changing customs associated with common meals during the Archaic period. These objects belong to a new item of furniture, the couch. Adopted from the Near East, the practice of lying on klinai instead of sitting became common during the Archaic period and is also well-documented in Archaic vase painting (Fehr 1971; Murray 1990).

Conclusions From the end of the Early Iron Age onward a greater differentiation of settlements and houses, with new settlement- and house-types, can be identiWed. In the Archaic period the range of ground plans common to the Early Iron Age developed in such a way that simple features became more complex. The multiple-room house emerged, which followed the same ground plan within a single settlement and showed a greater desire for standardization. The number of rooms within houses increased, creating the preconditions required for a re-arrangement of the household. The multi-functional, one-room house was replaced by a house with several mono-functional rooms. The increased number of rooms in this new arrangement may be indicative of a change in family organization. These transformations were regionally different. In northern Greece settlement-plans and house-types developed in the opposite direction from those in southern Greece: the settlement structure was altered from a complex to a simpler layout. In southern Greece settlement density increased; while in the northern region the settlements with agglomerated houses were replaced by detached houses. In houses in the north, the functional complexity generated by many rooms was given up in favour of a few, multi-functional, rooms. The changes in northern Greece may indicate a change in socioeconomic structure accompanying a return to large, few-room houses. In this contribution I have tried to offer some thoughts on the archaeology of Archaic houses through a detailed analysis of their ground-plans. However, physical parameters are only one side of the coin. The other side is the fact that changes within the household need not necessarily have had any effect on the architecture. Hence it is important to consider the archaeological formation processes of household deposits in such an analysis. In order to do this, more reliable information must be gained: the inventory of the houses must be considered as an important source of information for the social structure within a house. Methodical activity-zone

32

Franziska Lang

analysis would help to provide some hints of alterations to communal life within the house. Such analysis is also necessary in order to aid in identifying the social status of its inhabitants. Similarly, recurrent use of the same house-type does not necessarily signify social equality. As already mentioned, hierarchies could have been expressed in very different ways. Even if such evidence is not yet available, the progressive transformation of a ground plan over time is nevertheless a very interesting indication of a structural shift within the family and in its social affairs (Lang 2002, 274–299). Changes in internal family structure can be recognized through the modiWcation of a multi-functional one- or two-room house to create a multiple-room house, the introduction of an integral transitional area that allowed functions to be distributed over different areas of the house, and by providing separate access to individual rooms. This was the beginning of a new and more individualistic approach to life, since the residents within the house had more rooms, each of which could be reached separately. At the same time, the family’s relationship to the outer world was altered by enclosing the house with high walls so that accidental contact with passers-by was no longer possible. The house and its inhabitants, as the smallest social unit of a society, formed a microcosm that was simultaneously networked with its neighbors and the whole settlement as a macrocosm (Lang, in press). There is a strong interdependence between the two, so that the analysis of houses offers us only one side of a complex phenomenon with various dynamics. In the Archaic period, the buildings acquired different characters expressed in ground-plans which were speciWc to their functions. Buildings that were not for private purposes, but for political-administrative and sacred functions were constructed (for example, tholoi and temples respectively). The differentiation of private and public architecture had begun. Ground plans soon became standardized making it possible for anyone—even a modern archaeologist—to identify the intended function of a structure simply by looking at it. Thus a building became a symbol, and this was the beginning of the codiWcation of Greek architecture. The changes in domestic architecture are a further clue with respect to the new, fundamentally different structures of family and society in Archaic Greece that are also reXected in well-known examples of sculpture and vase-paintings. But that is the topic of another paper. Notes 1. I would like to express my thanks to K. Haswell, B. Ault, and L. Nevett for their proofreading. 2. Cf. Kiderlen (1995) and Mazarakis Ainian (1997b), where classiWcations are often based on a single ground plan.

Structural Change in Archaic Greek Housing

33

3. Notable exceptions are the more general older studies such as Rider (1916) or Oelmann (1927). The Wrst detailed analysis of the architectural features of ancient houses came with the excavation of Olynthus, begun in 1929 by D. M. Robinson. 4. 4. A Wrst attempt to study this material was made by Weickert (1929). His focus on sacred buildings was due to the lack of excavated Archaic houses. See also the more general comments in Jeffery 1976, Snodgrass 1980, and Morris 1998. 5. It should be noted that in spite of this observation, historical data about Zagora are completely lacking. 6. Rooms 19 and 22/23 (Figs. 2.2a.b) are often seen as comprising a ruler’s dwelling (e.g., Mazarakis Ainian 1997b, 171–175). This interpretation is based on the size of the rooms, the benches with pithoi, the hearth, and the orientation towards the temple, although the latter was built in the sixth century, long after abandonment of Zagora at the end of the eighth century. Such a reading does not entirely convince me: there are other rooms of the same size with benches and fragments of pithoi and other rooms facing to the south. A more detailed consideration of the matter is necessary, for which there is unfortunately no space here. 7. To treat access relationships as social relationships constitutes a simpliWcation of a complex social system (which also integrates factors such as policy and economy, ideology and religion, and climate and resources; Samson 1990a, 10–11; Brown 1990, 103). However, patterns of access need to be explained: the fact is that a fundamental change of spatial and access arrangements occurred from the Early Iron Age to the Archaic period. 8. Although recent studies have shown that gender division within Classical houses was not quite as restrictive as was once believed (Nevett 1995; Goldberg 1999). 9. For Athens, see Kavvadias and Kawerau 1906, 120; Brann 1962, 110–111; and Baziotopoulou-Valavani 1994, 45–54. For Corinth, see Williams and Fisher 1971, 7; 1979, 125–128. 10. However, in recent years new studies have questioned the legitimacy of these time-honored beliefs (Leach 1999, 194–196). Literature Cited Allison, P. M., ed. 1999a. The Archaeology of Household Activities. London: Routledge. ———. 1999b. “Introduction.” In Allison 1999a, 1–18. Ashmore, W., and R. R. Wilk. 1988. “Household and Community in the Mesoamerican Past.” In Household and Community in the Mesoamerican Past, ed. R. R. Wilk and W. Ashmore, 1–27. Albuquerque: University of New Mexico Press. Baziotopoulou-Valavani, E. 1994. “Ανασκαϕς σε αθηναικ κεραµικ εργαστρια των αρχαικν και κλασικν χρνων.” In The Archaeology of Athens and Attica Under the Democracy, ed. W. D. E. Coulson, O. Palagia, T. L. Shear, Jr., H. A. Shapiro, and F. J. Frost, 45–54. Oxford: Oxbow. Bernbeck, R. 1997. Theorien der Archäologie. Tübingen: A. Francke-Verlag. Brann, E. T. H. 1962. Late Geometric and Protoattic Pottery. Athenian Agora 8. Princeton, N.J.: American School of Classical Studies at Athens. Boardman, J. 1967. Excavations in Chios 1952–1955: Greek Emporio. Annual of the British School of Archaeology at Athens, Supplement 6. Athens: British School of Archaeology at Athens.

34

Franziska Lang

Brown, F. E. 1990. “Comment on Chapman: Some Cautionary Notes on the Application of Spatial Measures to Prehistoric Settlements.” In Samson 1990b, 93–109. Cambitoglou, A., A. Birchall, J. J. Coulton, and J. R.Green. 1988. Zagora 2: Excavation of a Geometric Town on the Island of Andros. Excavation Season 1969; Study Season 1969–1970. Athens: Archaologikis Hetairias. Carsten, J., and S. Hugh-Jones, eds., 1995. About the House: Lévi-Strauss and Beyond. Cambridge: Cambridge University Press. Drerup, H. 1969, Griechische Baukunst in geometrischer Zeit. Göttingen: Vandenhoeck and Ruprecht. Fagerström, K. 1988. Greek Iron Age Architecture. Developments Through Changing Times. Göteborg: Paul Aströms Förlag. Fehr, B. 1971. Orientalische und griechische Gelage. Bonn: Bouvier. Goldberg, M. Y. 1999. “Spatial and Behavioural Negotiation in Classical Athenian City Houses.” In Allison 1999a, 142–161. Hänsel, B. 1989. Kastanas: Die Grabung und der Baubefund. Berlin: Spiess. Hayden, B. 1983. “New Plans of the Early Iron Age Settlement of Vrokastro.” Hesperia 52, 367–387. Hoepfner, W. 1999. “Archaik.” In Geschichte des Wohnens, vol. 1, 500 v.Chr.–500 n.Chr. Vorgeschichte—Frühgeschichte—Antike, ed. W. Hoepfner, 129–199. Stuttgart: Deutsche Verlags-Anstalt. Hoepfner, W. and E.-L. Schwandner, 1994. Haus und Stadt im klassischen Griechenland. 2nd ed. rev. Munich: Deutscher Kunstverlag. Jameson, M. H. 1990. “Domestic Space in the Greek City-State.” In Domestic Architecture and the Use of Space. An Interdisciplinary Cross-Cultural Study, ed. S. Kent, 92–113. Cambridge: Cambridge University Press. Jeffery, L.H. 1976. Archaic Greece: The City-States c. 700–500 B.C. London: Methuen. Kavvadias, P. and G. Kawerau. 1906. Die Ausgrabungen der Akropolis vom Jahre 1885 bis zum Jahre 1890. Athens: Estia. Kiderlen, M. 1995. Megale Oikia: Untersuchungen zur Entwicklung aufwendiger griechischer Stadthausarchitektur von der Früharchaik bis ins 3.Jh.v.Chr. Hürth: M. Lange. Kinch, K. F. 1914. Fouilles de Vroulia. Berlin: Reimer. Lang, F. 1996. Archaische Siedlungen in Griechenland. Struktur und Entwicklung. Berlin: Akademie Verlag. ———. 2002. Klassische Archäologie: Eine Einführung in Methode, Theorie und Praxis. Tübingen: A. Francke Verlag. Lang, F. In press. “House—Community—Settlement: The New Concept of Living in Archaic Greece.” In Building Communities: House, Settlement and Society in the Aegean, ed. R. Westgate, N. Fisher, and J. Whitley. Studies of the British School of Archaeology at Athens. Athens: British School of Archaeology. Laslett, P. 1972. “Introduction: The History of the Family.” Household and Family in Past Time, ed. P. Laslett. 1–89. Cambridge: Cambridge University Press. Leach, E. 1999. “Discussion: Comments from a Classicist.” In Allison 1999a, 190–197. Margueron, J.-C. 1983. “Notes d’archéologie et d’architecture orientales.” Syria 60, 1–24. Mazarakis Ainian, A. 1997a. “Ανασκ ϕη Σκ λας Ωρωπο.” Praktika for 1997, 47–77. ———. 1997b. From Rulers’ Dwellings to Temples: Architecture, Religion, and Society in Early Iron Age Greece (1100–700 B.C.). Jonsered: Aström.

Structural Change in Archaic Greek Housing

35

Morris, I. 1998. “Archaeology and Archaic Greek History.” In Archaic Greece: New Approaches and New Evidence, ed. N. Fisher and H. van Wees, 1–91. London: Duckworth. Murray, O., ed. 1990. Sympotica: A Symposium on the Symposion. Oxford: Clarendon. Nevett, L. 1995. “Gender Relations in the Classical Greek Household: The Archaeological Evidence.” Annual of the British School of Archaeology at Athens 90, 363–381. Oelmann, F. 1927. Haus und Hof im Altertum: Untersuchungen zur Geschichte des antiken Wohnbaus. Berlin: de Gruyter. Parker Pearson, M. and C. Richards, 1997. “Ordering the World: Perceptions of Architecture, Space and Time.” In Architecture and Order. Approaches to Social Space, ed. M. Parker Pearson and C. Richards, 1–37. London: Routledge. Pfälzner, P. 2001. Haus und Haushalt. Wohnformen des dritten Jt. v.Chr. in Nordmesopotamien. Mainz: P. von Zabern. Philippaki, B. 1978. “Ανασκαϕ ακροπολως Αγ. Ανδρας, Σϕνου,” Praktika for 1978, 192–194. Rider, B. C. 1916. The Greek House: Its History and Development from the Neolithic Period to the Hellenistic Age. Cambridge: Cambridge University Press. Sahlins, M. 1974. Stone Age Economics. London: Tavistock. Samson, R. 1990a. “Introduction.” In Samson 1990b, 1–18. ———. ed. 1990b. The Social Archaeology of Houses. Edinburgh: Edinburgh University Press. Schilardi, D. 1978. “Ανασκαϕ Π ρου.” Praktika for 1978, 195–210. ———. 1979. “Ανασκαϕ Π ρου.” Praktika for 1979, 236–248. ———. 1987. “Ανασκαϕ Π ρου.” Praktika for 1987, 228–231. Schwandner, E.-L. 1990. “Überlegungen zur technischen Struktur und Formentwicklung archaischer Dachterrakotten.” Hesperia 59, 291–300. Senff, R. 2000. “Die archaische Wohnbebauung am Kalabaktepe in Milet.” In Die Ägäis und das westliche Mittelmeer: Beziehungen und Wechselwirkungen vom 8. bis 5.Jh.c.Chr, ed. F. Krinzinger, 29–37. Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Snodgrass, A. 1980. Archaic Greece. Berkeley: University of California Press. Wardle, K. A. 1987. “Excavations at Assiros Toumba.” Annual of the British School of Archaeology at Athens 82, 315–335. Weickert, C. 1929. Typen der archaischen Architektur in Griechenland und Kleinasien. Augsburg: Filser. Williams C. K., II. 1979. “Corinth 1978: Forum Southwest.” Hesperia 48, 105–144. Williams C. K., II and J. E. Fisher. 1971. “Corinth 1970: Forum Area.” Hesperia 40, 1–51.

Chapter 3

Security, Synoikismos, and Koinon as Determinants for Troad Housing in Classical and Hellenistic Times William Aylward

The Troad is an important source for houses in the Greek world by virtue of its location on the Dardanelles, along the only maritime passage between the Black Sea and the Mediterranean. Military operations of Xerxes and Alexander the Great focused attention on the Dardanelles as a border between Europe and Asia, and the strategic importance of the Troad was not lost upon its inhabitants. Cities and territory in the Troad changed hands on numerous occasions in Classical and Hellenistic times, and the political and military instability of the region inXuenced the character of domestic life there. This chapter focuses on certain regional phenomena that shaped the character of domestic architecture in the Troad, namely, security offered by city walls, synoikismos, and intercity status relationships, especially when deWned by a koinon.1 Also addressed here is the close geographical position of the Troad to regions known for two distinctive domestic architectural types, conventionally identiWed as pastas and prostas houses. Olynthus and Delos, known for the pastas -type house, lie to the west and south of the Troad, respectively. Colophon and Priene, known for the prostas -type house, lie to the south of the Troad on the Ionian coast, and Thracian Abdera, known for the same house type, lies to the northwest.2 House types in the Troad are therefore a potentially informative source for modern systems of classiWcation for ancient Greek architectural types, and, perhaps more importantly, for assessing regional diversity in ancient domestic architecture that deWes conventional classiWcation. There is great inconsistency in the quality of evidence for Greek housing in the Troad, and interpretations must be distilled from a broad mix of archaeological and literary sources. The past Wfteen years of systematic archaeological research at Ilion have made this site a focal point for the investigation of domestic architecture in the region. There is great

Security, Synoikismos, and Koinon

37

promise for comparanda from other Troad cities, especially Assos and Alexandreia Troas, but, in advance of further archaeological reports for these sites, historical sources continue to provide the primary basis for interpretation. This account of Greek housing in the Troad comprises two parts. The Wrst is a presentation of the evidence for domestic architecture in the Troad, arranged chronologically, and with an unavoidable focus on Ilion. Following this, discussion turns to the social context for houses in the Troad, with a view to understanding connections between domestic architecture and the region’s political, demographic, and geographic framework.

The Physical Evidence Map 3.1 shows the principal settlements in the Troad from Archaic through Hellenistic times.3 Aeolian Greek occupation appears to begin in the eighth and seventh centuries at Kebren in the interior and at Ilion near the northwest coast. At Ilion, Geometric pottery suggests settlement as early as the eighth century B.C., but this is not conWrmed by any known domestic structures.4 By Archaic and Classical times, settlements on the coast and in the interior appear to have formed a network of cities from the Dardanelles to the Gulf of Edremit. In the course of the sixth and early Wfth centuries, parts of the Troad were in contact with Mytilene, Athens, and Persia, and these connections almost certainly inXuenced the character of domestic architecture, although there is no speciWc evidence from the Troad to demonstrate this.5 New foundations and political mergers of neighboring communities (synoikismoi ) of the Hellenistic age altered the demography of the Troad, transformed civic identity, and involved extensive appropriation of interior lands by coastal cities with new economic clout and religious consequence. Builders in the Troad in the Greek period used materials readily at hand. This clear concern for economy is consistent with practice at Assos and Neandreia, where buildings make almost exclusive use of local granite for foundations and walling. At Ilion, where such sturdy material as granite was not available locally, builders often cut into the soft marly bedrock on the plateau between the Scamander and Simois Rivers to create platforms for building. Foundations and walls were then built from unworked Weldstones, most often limestone, bonded with earthen mortar. An abundance of terracotta roof tiles at Ilion shows that sturdy walls would have been necessary to support heavy roofs. The luxury of tiled roofs appears to have been offset by restricted clear space between walls and supports. The earliest known domestic architecture in the Troad comes from the Classical period at Neandreia, about 30 km. south of Ilion (Fig. 3.1a;

38

William Aylward

Fatmann-Rey 1996, 15–42). Remains are sparse, and they consist of a single course of stone foundation, with some stones cut to size and others unworked. Nor are there remains to assist in the reconstruction of the walling or rooWng of the house. The topography here is rather steep, but there is no evidence for terracing. The plan indicates that the house had been oriented to face southeast, perhaps to shelter it from prevalent northwest winds along the west coast of the Troad. The arrangement of the rooms suggests that the architectural plan is of a pastas-type house, and Fatmann-Rey compares the plan to variants of this type known from Eretria, Olynthus, and Maroneia (1996, 25–30). The pastas-type house is outside Neandreia’s Aeolian heritage and architectural association with the Ionian cultural milieu, where the prostas-type predominates. This

Map 3.1. The Troad with sites mentioned in Chapter 3.

Security, Synoikismos, and Koinon

39

apparent inconsistency with conventional deWnitions for house types is actually a Wtting forebear to the diversity of domestic architecture encountered in the Troad in later periods, which includes variants of both pastas- and prostas-type houses. Just after 310, Antigonos Monophthalmos founded the eponymous city of Antigoneia, later renamed Alexandreia Troas by Lysimachos, on the west coast of the Troad, and he settled it by synoikismos (Strabo 13.1.33, 47, 52; compare Cohen 1995, 145–146). Recent geophysical survey at the site has brought to light an extensive grid of insulae arranged within the Lysimachean city wall.6 The insulae detected in the geophysical survey ostensibly date to the Roman period, but they probably derive their orientation from an earlier Hellenistic city plan, as was the case at Ilion (for which, see below). Behind the cavea of the city’s theater, the surveyors have identiWed what appears to be a house of the pastas type, and they have compared its architectural plan to houses at Olynthus (Fig. 3.1b; Papenberg and Schrader 1999, 75–76). At Ilion, there are traces of two houses on the south slope of the acropolis, and these appear to date prior to the construction of a new temenos portico for the Sanctuary of Athena Ilias (Fig. 3.2). Remains of these houses were poorly preserved, and they only give a partial indication of plan and orientation.7 The construction of the temenos portico has now been linked to the construction of a new city wall for Ilion in the third quarter of the third century.8 Prior to this time, inhabitants could have relied only on the remnants of the prehistoric citadel wall for protection, and this explains the location of the fourth-century houses in close proximity to the acropolis and the circuit of the prehistoric fortiWcations. Construction of the new temenos portico would have forced the inhabitants of these houses to relocate, but the contemporary construction of a new city wall around the Lower City would have provided ample secure alternatives for housing. On the southern edge of the plateau later enclosed by Ilion’s Hellenistic city wall there are also traces of houses contemporary with those on the south slope of the acropolis, but perhaps not inhabited for longer than a single generation, from about 260–240 (Fig. 3.3, Phase H1; Aylward 1999, 161–164). Remains consisted of a northern and southern group of walls, separated by about 5.00 m, each consisting of fragmentary foundations, remnants of tamped earthen Xoors, and terracotta roof tiles.9 Foundation walls were between 0.52 and 0.60 m wide, and composed entirely of uncut Weldstones, none larger than 0.50 m on any side. Some of these walls had been founded in rock-cut trenches, presumably as a means to compensate for the gradient of the plateau. The northern group had at least two small contiguous rooms, one about 1.50 m by 1.10 m and the other about 2.00 m wide, but of unknown length. Remains of

Fig. 3.1.a. Neandria, schematic plan of the so-called pastas house; b. Alexandria Troas, schematic plan of the so-called pastas house; c. Ilion, schematic plan of the Portico House; d. Ilion, schematic plan of the Gateway House; e. Ilion, schematic plan of the Quarry House.

Fig. 3.2. Ilion, schematic plan of Hellenistic house walls on the south slope of the acropolis.

42

William Aylward

the southern group did not give a clear indication of the size or number of rooms. The meager architecture of the houses suggests transient or short-lived habitation, perhaps connected to the cultivation of unprotected Welds or the quarrying of local bedrock for building on the citadel or city fortiWcations. Construction is dated by a bronze coin of Alexandreia Troas, struck no earlier than 261, discovered in the masonry of one of the house walls (Troy coin C473; compare Aylward 1999, 160). A later phase of more permanent housing superimposed on these remains shows that these buildings were abandoned soon before or during construction of the Hellenistic city wall in the third quarter of the third century. Three discrete houses have been identiWed among the more permanent remains of housing in the Lower City that follow upon the completion of Ilion’s new fortiWcations: the Portico House, the Gateway House, and the Quarry House (Figs. 3.1c–e, 3.3, Phase H2; Aylward 1999, 164– 172).10 Datable Wnds from these houses show that occupation may have continued here as late as the third quarter of the second century and so the absolute dates for this phase of domestic architecture at Ilion are best given as ca. 225–130. In terms of building materials, houses in this phase are quite similar to those of the previous phase, and materials salvaged from the earlier houses, especially prefabricated materials such as roof tiles, must have played a signiWcant role in their construction. Builders in this phase were more invasive in their removal of bedrock for foundation platforms, and this is consistent with the evidence for larger houses with courtyards and facilities for water management, such as wells, cisterns, and drains. The architectural plan of the Portico House is only partially preserved, and it is not clear if it resembled a house of the prostas or pastas type (Figs. 3.1c, 3.3). Some foundations consisted of rubble masonry and others roughly dressed limestone blocks with earthen mortar for joining. Signs of mud brick walling gave an indication of the material used for the superstructure, and terracotta roof tiles likewise for the roof. A room about 3.30 m² appears to have formed the northeast corner of the house, and this was connected to a porch through a doorway on the southwest wall. The porch may have opened onto an enclosed courtyard, where there was evidence for a storage pit and a rough stone pavement. The porch was just under 2.00 m deep, and signs of vertical supports show that it was roofed. The masonry and Wll of a well located inside the porch suggest that it was in use during the occupation of the house ( Jablonka 1995, 50– 53).11 Walls abutting the exterior of this room may have been later additions to the building. The Gateway House is so called because it was built inside a gateway in the rock-cut fortiWcation system that had surrounded the Lower City

Fig. 3.3. Ilion, schematic plan of Hellenistic house walls from two phases in the Lower City.

44

William Aylward

of the prehistoric settlement (Figs. 3.1d, 3.3). Parts of the rock-cut ditch may have remained exposed in Hellenistic times, and these perhaps functioned as an open cistern. House remains consisted of three separate rooms that may have at one time belonged to the same dwelling—a north room and a south room, separated by a narrow gap, and a further room to the west. The north and south rooms were each 4.00 m wide, and of unknown length. For the foundations, some preserved up to 0.70 m high, the builders used roughly-worked limestone for wall facing, and softer limestone cut from bedrock for the rubble cores of walls, which were on average about 0.50–0.60 m wide. These materials were bonded with earth with a high ceramic content derived from earlier occupation of the area. Traces of mud brick gave an indication of the upper walling, and the plumb faces of some walls suggest that some interior spaces may have been plastered. The north room of the Gateway House had a doorway in the east wall, and it would have looked out over the open section of the prehistoric rock-cut fortiWcation, which by this time may have functioned as an open cistern. This arrangement, in combination with several pits with Hellenistic Wlls from the same area, attests to a focus of domestic activity on the east side of the house. Therefore, the builders seem to have taken into account the region’s prevailing southeasterly winds when orienting their houses. Wind damage to homes may have been reduced by pitching roofs down to the south and east, along the slope of the plateau and away from persistent winds from the northwest. The south room of the Gateway House had a Xoor remnant of tamped earth, two rock-cut pits for storage, and a simple hearth of irregular oval shape on a platform of burnt clay, 1.60 by 0.60 m in size, in the northeast corner.12 The hearth could have supported a small open Wre or a brazier, with smoke vented through the wall or ceiling, although no traces of ventilation Wxtures survived. The gap between the north and south rooms appears to have supported a water channel, perhaps for draining run-off from rooftops or efXuent from the house. The west room was built over a section of the prehistoric rock-cut fortiWcation that had been Wlled in for domestic construction. A fragmentary stone pavement between the north and west rooms could have been a courtyard or a street, but there is no good indication as to how this space might have communicated with either room. If the pavement is the remainder of an enclosed courtyard, then this may have been a central living space within a multiroom house. If it belonged to a street, then the west room would have been part of an adjacent house or city block. The Quarry House is so named because it was built over a back-Wlled quarry that had provided building material for public works on the citadel and for Ilion’s Hellenistic city wall (Figs. 3.1e, 3.3).13 Like other

Security, Synoikismos, and Koinon

45

contemporary houses in this area, the Quarry House was deliberately sited near a source of water, in this case a well, which had probably serviced the quarry at one time. Architectural features in the Quarry House consisted of a well, a water channel built from roof tiles on the west side of the house, and two walls that deWne a north and south room, the former with a tamped earthen Xoor, and the latter with a fragmentary stone pavement. A destruction layer of burnt mud brick and roof tiles was found above Xoor level in the north room. In general, these features were similar in orientation, building materials, and methods of construction to the Portico and Gateway Houses. The builders appear to have added additional stone curbing to the upper edge of the rock-cut shaft of the well in order to bring its mouth up to Xoor level inside the house. The gutter on the west side of the house probably collected runoff from the eaves. The fragmentary stone pavement in the south room suggests provisions for the diversion of rainwater across a possible enclosed courtyard, but the gutter outside the house suggests this area may have been roofed. Occupation of these houses appears to have continued at least into the second quarter of the second century, and in the case of the Quarry House, habitation appears to have been brought to an abrupt end by Wre (Berlin 1999, 144–145). The next phase of domestic architecture in this area corresponds to the Augustan revival of Ilion at the end of the Wrst century and the installation of a new orthogonal plan of insulae across the Lower City.

Context and Interpretation I now propose to investigate city walls, synoikismos, and the concentration of power in a regional koinon as components of the social context for housing in the Troad. Discussion of how the Troad’s houses reXect the region’s position close to localities known for pastas- and prostas -type houses follows. The inhabitants of Neandreia in the Classical period were clearly drawn to the city’s remote acropolis for purposes of safety.14 Likewise, in the Wfth through early third centuries at Ilion, security concerns undoubtedly lay behind house construction in the shadow of the sanctuary of Athena Ilias and the prehistoric citadel wall, Ilion’s only means of defense prior to the construction of the city’s Hellenistic fortiWcations. During this period, Ilion endured the army of Xerxes (480), paid tribute to Athens (425–424), and succumbed to Charidemos of Oreos (ca. 360) and renegade Gallic mercenaries (278).15 Toward the end of this period there are remains of far more ephemeral domestic buildings on the plateau at the southern edge of the unfortiWed Lower City (Fig. 3.3, Phase H1), where housing probably never consisted of more than a

46

William Aylward

loose agglomeration of temporary buildings and farmland.16 This situation brings to mind Pecírka’s model of homestead farms in Classical and Hellenistic Greece (Pecírka 1973, 114–115)—loosely arranged, but within deWnable boundaries. At Ilion, the religious and Wnancial institution of the Sanctuary of Athena Ilias provided the common ground by which the loose topographical arrangement of the community was connected. After Alexander the Great crossed the Hellespont in 334, he made Ilion a polis, liberated the city from taxation, made dedications to Athena Ilias, and ordered the construction of new public buildings.17 His successor consolidated resources and population in the Troad into the hands of only a few cities at the expense and livelihood of the region’s smaller towns. Antigonos Monophthalmos founded the eponymous city of Antigoneia (later renamed Alexandreia Troas by Lysimachos) in 310, and he settled it with a synoikismos involving the cities of Kebren, Skepsis, Larisa, Kolonai, Hamaxitos, and Neandreia.18 Antigonos also put Ilion’s sanctuary of Athena Ilias at the center of a new Troad koinon. In exchange for tax beneWts at Ilion, the new institution was enriched by private benefactions of Malousios of Gargara for the construction of a theater and possibly a new temple of Athena.19 Epigraphic and numismatic evidence shows that the koinon prospered under the leadership of the Sanctuary of Athena Ilias, often at the expense of other Troad cities (Robert 1966, 18–25, 34–46). For example, member cities paid taxes to the koinon and tribute to Athena Ilias at festivals, and the Sanctuary generated additional revenue from interest loans and rent on temple land.20 Lysimachos maintained a focus on Ilion and Alexandreia Troas. The latter he fortiWed with a new city wall. If the Lysimachean city wall at Alexandreia Troas was as long as Strabo says it was, about forty stades (ca. 7.4 km), the city could have easily supported a residential district of about 400 ha and a potential capacity of up to 100,000 people.21 It is unlikely that the synoikismos enacted by Antigonos for Antigoneia (later Alexandreia Troas) brought 100,000 residents to that city, but the event could easily have been the largest coordinated movement of people the Troad had ever seen. Lysimachos may have further concentrated the population of the Troad at Ilion with synoikismos.22 This would have greatly increased the number of Troad residents with Wnancial responsibilities to the Sanctuary of Athena Ilias, and it would have transformed Ilion into a competitor with Alexandreia Troas for human and capital resources, as well as for commercial activity.23 Ilion’s synoikismos may have involved Achilleion, Birytis, Gentinos, Glykeia, Kenchreai, Skamandreia, Sigeion, Thymbra, “the Village of the Ilians,” and other villages of the Simois Valley.24 This list, combined with sources for cities with Wnancial obligations to the Troad koinon, suggests that at least twenty Troad cities, including

Security, Synoikismos, and Koinon

47

independent communities such as Abydos, Assos, and even Alexandreia Troas, were to varying degrees dependent upon the Sanctuary of Athena Ilias in the Hellenistic period (Magie 1950, 66, 869–870; Aylward 1999, 174, 182, notes 86, 89). The Hellenistic city wall built around Ilion in the third quarter of the third century was about 3,600 m long, and the total area enclosed within it was about 65 ha. The struggle between Antiochos Hierax and Attalos I for control of Asia Minor (ca. 241–228) appears to have set the stage for the new fortiWcation (Aylward 1999, 175). This conXict would have threatened not only unprotected housing, but also the independence of the entire city. The archaeological remains for housing in the Lower City before and after the construction of the new fortiWcations reveal an obvious distinction in permanence (Fig. 3.3), and this shows rather clearly that settlement continued to be motivated by protection in the Hellenistic period. A fair estimate for a comfortable population in this newly enclosed space is about 14,000.25 There is epigraphic evidence for the enrollment of new citizens at Ilion in the following generation, and the inWlling of quarries to create land for housing, as demonstrated by the archaeological record of the Quarry House, suggests a high demand for land in the Lower City at this time (Pomeroy 1997, 205–206; Frisch 1975, no. 64). I now turn to evidence in the Troad for variation in house plans of the pastas and prostas type and to the possible social determinants behind them.26 The evidence supports the notion that the frequency with which new houses are brought to light by Weldwork is inversely proportional to the integrity of conventional classiWcations for house types. For example, it is now well known that the prostas-type, a hallmark of domestic architecture in Ionia, was also used in the West. Likewise, it comes as no surprise that variants of the pastas-type, so well known from Olynthus and Eretria, also occur in the Troad at Alexandreia Troas and Neandreia. Indeed, there was a certain universality about both types (Tsakirgis 1989, 278–279). At Ilion, the Portico House is, at Wrst glance, consistent with prostastype houses of Ionia (Fig. 3.1c). The plan featured a covered porch that divided a court from a room or suite of rooms behind. But there are a number of uncertainties in the plan, including the porch’s western termination, the number of supports, and the number of rooms behind it. Moreover, the location of the extant square room behind a porch with a southern orientation is a recognized feature of andrones in both pastasand prostas-type houses.27 In the absence of a complete plan, it is difWcult to classify the Portico House as a Wxed architectural type. Given the date of this house near the end of the third century, over half a century after a substantial reorganization of territory and communities in the Troad, the lack of a clear manifestation of house-types is not surprising, and perhaps to be expected.

48

William Aylward

More complete house plans are found elsewhere in the Troad. As presented above, these include two variants of the pastas-type house in the Greek East. Alexandreia Troas has produced evidence for a pastas-type house above the city’s theater (Fig. 3.1b), and a variant of the same type is preserved in a near complete set of building foundations near the acropolis of Neandreia (Fig. 3.1a). Important here is the participation of Neandreia in the synoikismos for Alexandreia Troas as early as the third century. Also important is that, while some cities forced to relocate to Alexandreia Troas were later released from the merger and refounded, as in the case of Skepsis, Neandreia was never resettled (Strabo 13.1.33, 47). It is conceivable that the Ionian city of Neandreia, where the pastastype house has been identiWed in the Classical-period city, was the source for the pastas-type housing now known at Alexandreia Troas. The occurrence of two pastas-type houses in neighboring cities outside the conventional region identiWed for this type merits further investigation. Here it should be noted that the variant of the prostas-type house suggested for Ilion’s Portico House follows upon the same era of synoikismos in the Troad. Whereas the prostas-type elements of the Portico House lend themselves to classiWcation as indigenous architectural traits, the example of transmission of a house type via synoikismos from Neandreia to Alexandreia Troas suggests that connections between demographic shifts and domestic architecture should not be ruled out for Ilion. The available evidence suggests that houses in the Troad are not readily categorized within conventional geographic boundaries for prostasand pastas-type houses as deWned by systems of classiWcation popular in the 1960s and 1970s, when a focus on architectural form obscured the importance of social context.28 This is not to say that ancient builders in the Troad were not trained in established methods and techniques. Nor that local determinants, such as material, climate, and topography, did not bring consistency to domestic architecture. It is to say, rather, that the appearance of houses was at times probably less a consequence of any one region, condition, or cultural group, and more a reXection of social context, and, as shown by cases presented here, demographic phenomena. Synoikismos, because of its transformative effect on the identity, economy, military security, and health of a community, provides a meaningful lens through which to view the diversity of house plans found in Troad cities. This notion is consistent with the loss of civic identity experienced by synoikized communities, as in the case of Lebedos, and with the diversiWcation of cult activity and architecture among synoikized communities in western Greece, such as Dyme and Patras.29 Extreme consequences of demographic shift on domestic architecture are perhaps best illustrated by the anoikism for Olynthus, where rapid implementation of military goals produced secure, uniform housing, and by the Periklean

Security, Synoikismos, and Koinon

49

synoikism at Athens, where overcrowding hastened plague among displaced refugees barred from permanent residency and basic housing.30 Demography, of course, is only one of many determinants for the appearance of houses. For example, wealth and affordability have been identiWed as leading factors behind the absence of distinguishable types of houses at Pergamon in Hellenistic and Roman times (Wulf-Rheidt 1999, 309–314), and gender-based separation of domestic space has been proposed as a determinant for the Greek oikos (Nevett 1999, 156–175; 1994). The sum of all variables that went into the making of houses far exceeds the few that are detectable in the archaeological record. Even for those variables for which there is physical evidence, predictable relationships cannot be assumed for material remains and any social signiWcance such remains might impart (Nevett 1999, 156). Accordingly, houses like those in the Troad that are products of such unpredictable human processes might be best left unclassiWed. The foregoing discussion of the social consequences of Troad housing is distilled somewhat in a remark by the geographical commentator on Homer, Demetrius of Skepsis (ca. 205–130), reported by Strabo (13.1.27), that Ilion was so neglected that the rooftops were not even tiled. The Portico House, the Quarry House, and the Gateway House (Figs. 3.1c–e, 3.3) are among the types of houses that Demetrios would have seen at Ilion during his alleged visit there ca. 190 B.C. These houses at the southern edge of the Lower City had tiled roofs, and the same can be expected for more robust housing one would have encountered closer to the city center.31 Demetrios argued against Ilion’s claim, also reported by Strabo (13.1.25), that it was the site of Homeric Troy, but Skepsis had rivals in the Troad, and Demetrios’s view has not been free from allegations of bias.32 Demetrios’ ancestors at Skepsis had been synoikized at Alexandreia Troas under Antigonos, despite the king’s pledge of freedom and autonomy, but they were later restored to Skepsis by Lysimachos. Skepsis was well known for its intellectual culture, primarily because of its location near Assos, but the autonomy it had gained from Lysimachos was lost at the Peace of Apamea in 188, when Rome made Skepsis subject to Pergamon as punishment for siding with Antiochos III.33 And so the Skepsis of Demetrios’s day was eclipsed by free cities of the Troad like Ilion and Alexandreia Troas with special connections to Rome. Moreover, Demetrios would have known Ilion as a newly walled city and the center of the koinon to which Skepsis probably paid dues. The ceramics, amphoras, and roof tiles recovered from house deposits at Ilion that date to the lifetime of Demetrios bear out a level of comfort and opportunity commensurate with Ilion’s proximity to the Sanctuary of Athena Ilias, the Troad koinon, and Rome.34 Reasons for the disparity between Demetrios’s view of Ilion as a neglected city without tiled roofs and the archaeological

50

William Aylward

materials showing otherwise are perhaps best understood in the context of competition between Troad cities tainted by the previous century’s dramatic shifts in territory, resources, and population. The characterization of Ilion presented by Demetrios, hardly a dispassionate observer, not only tells us about the value the ancients put on stability and comfort in housing, but it also reXects how city walls, the transformative power of synoikismos, and the authority of a koinon probably shaped regional perceptions about domestic architecture. Notes 1. For permission to conduct research at Troy, I thank Manfred Korfmann and C. Brian Rose. I am grateful to the editors and anonymous readers for helpful comments on this chapter. 2. For the distribution of pastas- and prostas -type houses, see, e.g., Cahill 2002, 75–77; Nevett 1999, 170–171; Aylward 1999, 172; Hoepfner and Schwandner 1994, 185, 214–216 and Wgs. 1761–77. The distinction is more relevant for building and appearance than for function and meaning; cf. Cahill 2002, 78; Nevett 1999, 171. 3. For an overview of this topic, see Cook 1973, 360–368. 4. Rose 1999, 61–63; Cook 1973, 101, 360–361; Blegen 1958, 147, 249–250. 5. For the Troad in the Archaic and Classical periods, see Tenger 1999, 125–147; Cook 1973, 178–180, 264, 351, 363–364. 6. Papenberg and Schrader 1999, 37–83. I thank Kutalmis¸ Görkay for calling this to my attention. 7. In quadrats E/F8/9: Rose 1999, 41–46; 1998, 96; Blegen 1958, 248, 249, 291–293, Wgs. 61–66, 336. 8. For the dating of the temenos portico and city wall, see Tekkök 2000, 85–96; Rose 1997, 93–101. 9. For the tiles, see Hasaki 1999, 225–236. 10. The houses come from quadrats w28, C29, and y28/29, and they were all built over and around a defensive ditch that enclosed the Bronze Age Lower City as early as the Troy VI period. 11. In 1994 the well of the Portico House was almost completely excavated (removing 17 m. of Wll) and assigned a founding date contemporary with the construction of the house walls, i.e., ca. 225 B.C. 12. Cf. the storage jars and oval hearth in a Hellenistic house at Draphi in Attica ( Jones, Sackett, and Graham 1962, 102, n. 29). 13. For these building projects, see Rose 1997, 98; 1995, 99; Jablonka 1995, 54. 14. The city may have been a safe haven for Troad peoples who feared attacks from the Peloponnesian Xeet (cf. Thucydides 3.31, 3.33) 15. Xerxes: Herodotus 7.43. Athenian tribute list: Meritt et al. 1939, 291, 492– 493. Charidemos: Demosthenes, Against Aristocrates 23.154–155. Gauls: Strabo 13.1.27; cf. Rose 1997, 93, 98; Leaf 1923, 142, 147. 16. Quarrying on the bedrock plateau for buildings on the acropolis may also have inhibited residential development in the Lower City at this time (cf. Rose, 2003, 46; 1998, 85–86). 17. Diodorus Siculus 17.17–18; Plutarch, Life of Alexander 15.7–9; Arrian 1.12.1; Strabo 13.1.26; cf. Cohen 1995, 155–156; Leaf 1923, 143–144; Frisch 1975, nos. 25, 121–123.

Security, Synoikismos, and Koinon

51

18. Strabo 13.1.33, 47, 52 (cf. Ricl 1997, 4–8, 15–16); Cohen 1995, 145–146; Cook 1973, 364; Magie 1950, 66, 869–870; Leaf 1923, 235–236. 19. Rose 2003, 62–63; Cohen 1995, 153–154; Verkinderen 1987, 247–269; Frisch 1975, xii–xiii, nos. 1, 10; Magie 1950, 66; Leaf 1923, 141–143. 20. Verkinderen 1987, 258–259; Frisch 1975, no. 18 (ca. 300 B.C.), no. 10 (77 B.C.); Magie 1950, 66, 140–142, 239, 869–870, 1016–1019, 1119–1120. 21. Strabo 13.1.26, on Ilion, must have confused Ilion for Alexandreia Troas (cf. Rose 1997, 93–98; Cohen 1995, 153–154; Leaf 1923, 141–143, 236). 22. Strabo (13.1.26) is the only source for this event, and so its historicity must be approached with caution, especially since the Lower City of Ilion was not fortiWed until later in the century. But the historical record for Alexandreia Troas indicates that the synoikismos came Wrst and the city wall followed. 23. Cf. also the land grants of Antiochos I which attached 8,000 plethra (ca. 720 ha) of arable land in the Troad to the city of Ilion in the third century (Welles 1934, 60–71, nos. 10–13; Cook 1973, 365–367). 24. Rhoiteion and Gergis were added by Rome in 188 (Livy 38.39.10). 25. Aylward and Wallrodt 2003, 107; Aylward 1999, 173. This estimate, as well as the one made above for Alexandreia Troas, is based on an average density of 250 persons per ha (cf. Jameson, Runnels, and van Andel 1994, 549–554, with Tables B.2, B.3; Ault 1994, 42–43). 26. Cf. Tsakirgis 1996, 777; Nevett 1995, 94; Ault 1994, 226–231. The analysis presented here has an obvious focus on architectural types and appearance. Functionality, spatial organization, privacy, gender, and segregation are among the important components of architectural analysis that lie beyond the scope of this chapter. 27. Nevett 1995, Wgs. 6.1–6.20; Hoepfner and Schwandner 1994, 318–320, Wg. 303; Jones, Sackett, and Graham 1962, 105, with note 40. 28. Most notably, Drerup 1967; but cf. also, Krause 1977, Graham 1966. 29. For Lebedos, see Ager 1998; Welles 1934, 15–32, nos. 3–4. For northwestern Greece, see Houby-Nielsen 2001, 262–268. Some degree of independence could be maintained by synoikized cities, as in the case of Rhodes; cf. Gabrielsen 2000; Demand 1990, 89–94. 30. Thucydides 2.17, 52; Cahill 2002, 35–44; Demand 1990, 74–83, 87, 95–97. 31. The temple of Athena Ilias on the acropolis had marble roof tiles at this time (Rose 2003, 46), too, and other public buildings around the acropolis had terracotta tiles, some stamped “ΙΛΙ” for Ilion (Rose 1993, 101, 103 Wg. 7). 32. Kagan 1984, 22–23 (cf. Miller 1994, 269–270; Pfeiffer 1968, 249–251). For Demetrios, see Diogenes Laertius 5.84; Strabo 13.1.27, 55 (cf. Leaf 1923, xxvii–xxviii). For Skepsis, see Strabo 13.1.33, 52–54 (cf. Cohen 1995, 24, 145–147; Cook 1973, 345–347; Leaf 1923, 169–171, 268–275). 33. Ma 1999, 89, 163, note 2000, 283; Kagan 1984, 22–23. 34. Berlin 1999, 147–151; Hasaki 1999, 225–236; Lawall 1999, 216–217. Literature Cited Ager, S. L. 1998. “Civic Identity in the Hellenistic World: The Case of Lebedos.” Greek, Roman, and Byzantine Studies 39, 5–21. Ault, B. A. 1994. “Classical Houses and Households: An Architectural and Artifactual Case Study from Halieis, Greece.” Ph.D. dissertation, Indiana University. Aylward, W. 1999. “Studies in Hellenistic Ilion: The Houses in the Lower City.” Studia Troica 9, 159–186.

52

William Aylward

Aylward, W. and J. Wallrodt. 2003. “The Other Walls of Troia: A Revised Trace for Ilion’s Hellenistic FortiWcations.” Studia Troica 13, 89–112. Berlin, A. M. 1999. “Studies in Hellenistic Ilion: The Lower City. StratiWed Assemblages and Chronology.” Studia Troica 9, 73–158. Blegen, C. W. et al. 1958. Troy IV. Settlements VIIa, VIIb, and VIII. Princeton, N.J.: Princeton University Press. Cahill, N. 2002. Household and City Organization at Olynthus. New Haven, Conn.: Yale University Press. Cohen, G. M. 1995. The Hellenistic Settlements in Europe, the Islands, and Asia Minor. Berkeley: University of California Press. Cook, J. M. 1973. The Troad: An Archaeological and Topographical Study. Oxford: Clarendon Press. Demand, N. 1990. Urban Relocation in Archaic and Classical Greece: Flight and Consolidation. Norman: University of Oklahoma Press. Drerup, H. 1967. “Prostashaus und Pastashaus.” Marburger Winckelmanns Programm, 6–17. Fatmann-Rey, G. 1996. “Versuch der Rekonstruktion eines Wohnhauses von Neandria.” Asia Minor Studien 22, 15–42. Frisch, P. 1975. Die Inschriften von Ilion. Bonn: Habelt. Gabrielsen, V. 2000. “The Synoikized Polis of Rhodes.” In Polis and Politics: Studies in Ancient Greek History, ed. P. Flensted-Jensen et al. 177–205. Copenhagen: Museum Tusculanum Press. Graham, J. W. 1966. “Origins and Interrelations of the Greek and the Roman House.” Phoenix 20, 3–31. Hasaki, E. 1999. “Studies in Hellenistic Ilion: A Note on the Rooftiles in the Lower City.” Studia Troica 9, 225–236. Hoepfner, W., and E.-L. Schwandner. 1994. Haus und Stadt im klassischen Griechenland. Rev. ed. Wohnen in der klassischen Polis 1. Munich: Deutscher Kunstverlag. Houby-Nielsen, S. 2001. “Sacred Landscapes of Aetolia and Achaia: Synoecism Processes and Non-Urban Sanctuaries.” In Foundation and Destruction: Nikopolis and Northwestern Greece, ed. J. Isager, 257–271. Athens: Danish Institute at Athens. Jablonka, P. 1995. “Ausgrabungen südlich der Unterstadt von Troia im Bereich de Troia VI-Verteidigungsgrabens. Grabungsbericht 1994.” Studia Troica 5, 39–79. Jameson, M. H., C. N. Runnels, and Tj. Van Andel 1994. A Greek Countryside: The Southern Argolid from Prehistory to the Present Day. Stanford, Calif.: Stanford University Press. Jones, J. E., L. H. Sackett, and A. J. Graham. 1962. “The Dema House in Attica.” Annual of the British School at Athens 57, 75–114. Kagan, J. H. 1984. “Hellenistic Coinage at Scepsis After Its Refoundation in the Third Century B.C.” American Numismatic Society Museum Notes 29, 11–24. Krause, C. 1977. “Grundformen des Griechischen Pastashauses.” Archäologische Anzeiger, 164–179. Lawall, M. L. 1999. “Studies in Hellenistic Ilion: The Lower City. The Transport Amphoras.” Studia Troica 9, 187–224. Leaf, W. 1923. Strabo on the Troad. Cambridge: Cambridge University Press. Ma, J. 1999. Antiochus III and the Cities of Western Asia Minor. Oxford: Oxford University Press. Magie, D. 1950. Roman Rule in Asia Minor. Princeton, N.J.: Princeton University Press. Meritt, B. D. et al. 1939. The Athenian Tribute Lists. Vol 1. Cambridge, Mass.: Harvard University Press.

Security, Synoikismos, and Koinon

53

Miller, Stella G. 1994. “Architectural Terracottas from Ilion.” In Proceedings of the International Conference on Greek Architectural Terracottas of the Classical and Hellenistic Periods, December 12–15, 1991. Hesperia, Supplement 27, ed. N. Winter, 269–273. Princeton, N.J.: American School of Classical Studies. Nevett, L. C. 1994. “Separation or Seclusion? Towards an Archaeological Approach to Investigating Women in the Greek Household in the Fifth to Third Centuries BC.” In Architecture and Order: Approaches to Social Space, ed. M. Parker Pearson and C. Richards, 98–112. London: Routledge. ———. 1995. “The Organisation of Space in Classical and Hellenistic Houses from Mainland Greece and the Western Colonies.” In Time, Tradition and Society in Greek Archaeology, ed. N. Spencer, 89–108. London: Routledge. ———. 1999. House and Society in the Ancient Greek World. Cambridge: Cambridge University Press. Papenberg, I. and P. Schrader. 1999. “Geophysikalische Untersuchungen im Stadtgebiet von Alexandria Troas,” Asia Minor Studien 33, 37–83. Pecírka, J. 1973. “Homestead Farms in Classical and Hellenistic Hellas.” In Problèmes de la terre en Grèce ancienne, ed. M. I. Finley, 113–147. Paris: Mouton. Pfeiffer, R. 1968. History of Classical Scholarship from the Beginnings to the End of the Hellenistic Age. Oxford: Clarendon Press. Pomeroy, S. B. 1997. Families in Classical and Hellenistic Greece: Representations and Realities. Oxford: Clarendon Press. Ricl, M. 1997. The Inscriptions of Alexandria Troas. Inschriften griechischer Städte aus Kleinasien 53. Bonn: Rudolph Habelt. Robert, L. 1966. Monnaies antiques en Troade. Hautes études numismatiques 1. Geneva: Droz. Rose, C. B. 1993. “The 1992 Post-Bronze-Age Excavations at Troia.” Studia Troica 3, 97–116. ———. 1995. “The 1994 Post-Bronze-Age Excavations at Troia.” Studia Troica 5, 81–105. ———. 1997. “The 1996 Post-Bronze-Age Excavations at Troia.” Studia Troica 7, 73–110. ———. 1998. “The 1997 Post-Bronze-Age Excavations at Troia.” Studia Troica 8, 71–113. ———. 1999. “The 1998 Post-Bronze-Age Excavations at Troia.” Studia Troica 9, 35–71. ———. 2003. “The Temple of Athena at Ilion.” Studia Troica 13, 27–88. Tekkök, B. 2000. “The City Wall at Ilion,” Studia Troica 10, 85–96. Tenger, B. 1999. “Zur Geographie und Geschichte der Troas.” Asia Minor Studien 33, 103–180. Tsakirgis, B. 1989. “The Universality of the Prostas House.” Paper presented at the Ninetieth Annual Meeting of the Archaeological Institute of America, Baltimore. Abstract in American Journal of Archaeology 93 (1989), 278–279. ———. 1996. “Houses and Households.” American Journal of Archaeology 100, 777–781. Verkinderen, F. 1987. “The Honorary Decree for Malousios of Gargara and the koinon of Athena Ilias,” Tyche 2, 247–269. Welles, C. B. 1934. Royal Correspondence in the Hellenistic Period. New Haven, Conn.: Yale University Press. Wulf-Rheidt, U. 1998. “The Hellenistic and Roman Houses of Pergamon.” In Pergamon: Citadel of the Gods: Archaeological Record, Literary Description, and Religious Development, ed. H. Koester, 299–330. Harvard Theological Studies, 46. Harrisburg, Pa.: Trinity Press International.

Chapter 4

Household Industry in Greece and Anatolia Nicholas Cahill

A common trope in ancient Greek literature is that craftsmen and traders were socially marginalized: that they were not allowed to participate in the political or social life of the city as fully as those who depended on agriculture as their primary source of livelihood. At Thebes, according to Aristotle, there was a law that no one who had not kept away from the agora for the last ten years might be admitted to ofWce (Politics 1277b; compare 1328b–1329b). Euripides’ mother was slandered for being a vegetable seller, and there are innumerable other examples of the low status of industrial workers. Granted, this view is primarily espoused by the land-owning aristocracy, Xenophon, Plato, Aristotle and other such writers, whose opinions form the bulk of surviving ancient Greek literature. But it has informed our understanding of Greek society to an extent perhaps not generally appreciated. Archaeologists have looked for industrial areas in Greek cities at the outskirts of the city, for instance, in the “potters’ quarters”— expecting craftsmen to be geographically as well as socially marginalized; or in the densely occupied “industrial districts” of cities such as Athens (compare Tsakirgis, Chap. 5).

Olynthus and the Greek World One of the most completely excavated Greek poleis, Olynthus, offers a rather different perspective on Greek household industry. On the borders with Macedonia, Olynthus is itself arguably near the limits of the Greek world, but is in many ways the best site for exploring the Greek household. Its violent destruction by Philip II in 348 B.C. left a wealth of artifacts on the Wnal destruction Xoors of its houses, and its relatively careful excavation and complete publication give us an unparalleled insight into the economic strategies pursued by different households, including household industry (Cahill 2002).

Household Industry in Greece and Anatolia

55

By household industry I mean the production or processing of goods for sale or consumption outside the household. Most households produced or processed their own food, cloth, and other staples: industries of a sort, but aimed at maintaining the self-sufWciency of the household, not at creating closer economic ties with other households in the polis. Other types of production, by contrast, were meant primarily for consumption outside the household, and tied the household in to wider economic networks, either by barter or exchange on the market. Some of these industries can be distinguished by their nature. Sculptors, manufacturers of weapons, potters, and coroplasts would have had only limited use for the objects they produced. They must have exchanged their products for money or other goods, and such households can be distinguished fairly easily by the non-domestic artifacts they leave behind. Other households produced normal household products such as cloth or food, but on a larger scale than the household required. The surplus again must have been traded or sold. In the archaeological record, such households can be distinguished by the unusual quantities of ordinary household artifacts they possessed, rather than speciWcally “industrial” artifacts such as molds or unWnished sculptures. The identiWcation of household industry in these cases is correspondingly more difWcult, especially as we may have difWculty deWning “usual” quantities. A relatively large number of households seem to have engaged in such domestic industry at Olynthus. Figure 4.1 shows the distribution of household trade and industry at the site. At least a quarter of the roughly 100 houses excavated at the site preserve traces of some sort of industry, and had the houses been excavated more carefully (they were dug over the course of four Weld seasons between 1928 and 1938), we would almost certainly have evidence for industry in more of them; the plan shows only a minimal interpretation. Processing agricultural goods was the most common household industry at Olynthus. About Wfteen houses were equipped with facilities for crushing olives, pressing grapes, grinding grain and the like—facilities with more capacity than would be needed to supply the daily requirements of the household. Such equipment as olive crushers and pressing Xoors represented a sizable investment, and would have been used only at certain seasons (Amouretti 1992, 1994). During that time, however, they probably could have processed much more than the produce of a typical olive orchard or vineyard, and were probably rented out and served other households, in return for a portion of the oil or wine, or for cash. Aristotle relates how Thales predicted a bumper crop of olives, optioned all the presses for that year, and so made a fortune (Politics 1259a).

Fig. 4.1. Olynthus, distribution of household industry.

Household Industry in Greece and Anatolia

57

One house involved with agricultural processing, House A 6, had an olive crusher or orbis in the court (Fig. 4.2). A sloping cement platform adjoining the court led to a sunken area where some facility had been robbed or eroded out. This might represent a pressing Xoor with a collecting basin at one end. In the same courtyard were found 12 upper grindstones of various types. This household seems to have engaged not only in the production of olive oil, but also in grinding grain to meal or Xour on a large scale (Cahill 2002, 241–244). The economic success of the household is suggested by the fact that the house was expanded beyond its original allotment, taking over part of the adjoining house to the south. Other houses have different types of installations. House A xi 10, for instance, had a special room with foundations for a press and two pithoi, and a second, cement-Xoored workroom to its north, with a large drain

Fig. 4.2. Olynthus, House A 6.

58

Nicholas Cahill

to the street (Graham 1953). A number of the houses with such processing equipment also had large rooms, occupying up to one-third of the total Wrst-Xoor area of the house. One such room, in house A 4, contained 15 pithos lids, suggesting that it was used as a storeroom for crops awaiting processing or for the Wnished oil or wine. Similar rooms in other houses may have served similar purposes. Weaving was another important household industry. Again, many houses at Olynthus preserved groups of loomweights belonging to one or two looms, enough to keep the household supplied with cloth. But two houses contained far larger quantities of loomweights. In one house, A v 9, the four rooms adjoining the courtyard each contained about 20 loomweights—enough for one loom. Four looms, taking up much of the welllit space of the house, are probably more than the household would need for its own use; these were very probably used to weave cloth for use outside the household. A nearby house, A viii 7/9, contained even more loomweights: 297 in all, in two hoards, probably not set up on looms at the time the city was destroyed, but awaiting use. A typical Greek loom seems to have used between 20 and 40 loomweights; 297 weights therefore represents the equipment for between 7 and 15 looms. Like house A 6, this was a largerthan-normal house, two houses which had been combined into a single residence by removing the wall which separated their pastades. A few other houses had large numbers of loomweights, house A iv 9, for example, contained 133 weights, probably belonging to two or three looms. Although it is unclear here, as elsewhere, whether the households were weaving cloth for internal or external use, the facts that this house adjoined the agora and had three shops opening onto Avenue B, all of which contained unusually large numbers of coins, suggests that the household was engaged in retail trade as well as large-scale textile manufacture. Two masons or sculptors lived in the row of houses along the west side of the city, to judge from the unWnished stone objects in their houses. One of these (house A 5) specialized in stelai, altars and louters; the other (house A 10) in architectural blocks such as capitals. Three or so houses seem to have engaged in baking and cooking on a larger scale than normal. One (A viii 8) may have been a baker; two others had kitchens and shops for the production of food, more than a single household could probably eat (houses D v 6 and ESH 4). Manufacture is rather less well represented among the excavated houses. At least one coroplast is located near the agora (house B i 5). In a room of this house were found 13 Wgurine molds. There was no trace of a kiln at this or other Olynthian houses, however. A manufacturer of slingbullets lived near the south end of the city, in what seems to be a relatively poor area (house C -x 5).

Household Industry in Greece and Anatolia

59

There are many more aspects of speciWc household industries at Olynthus that cannot be addressed here. Many houses have basins, workrooms, installations, foundations, and other facilities and equipment which do not seem to be part of the normal domestic assemblages, and at least some of these were probably used in a form of household industry. But a number of generalizations may be offered about the place of this “banausic” activity. First of all, it is by no means uncommon. Evidence for the production of goods for consumption outside the household is found in a large number of houses; and we must take into account that these early excavations undoubtedly missed a great deal of evidence. Moreover, it is found in houses whose architecture and contents suggest that the owners were relatively well-to-do, not economically or socially marginalized. Second, industry is not geographically marginal, conWned to the outskirts of Olynthus or to industrial areas. Rather, the houses which engaged in such production were located in the center of town, on the North Hill; and were well distributed throughout the hill. There is, on the other hand, a curious absence of evidence for household industry in the later, supposedly more upper-class “Villa Section” to the east. I suspect that this is the result of different economic strategies in these two regions of the city (Cahill 2002, 281–288). Third, domestic industry is not marginalized within the household, but is closely integrated with other household activities. Many houses, such as A 6 and A iv 9, had rooms which opened directly onto the streets, and are therefore usually interpreted as shops or workshops. If these households had wanted to, they could have segregated industrial activities from the rest of the house, relegating them to these separate rooms. Instead, production is generally located in the heart of the household: in the courtyard and in rooms adjoining the court. The shops seem to have served primarily as mediating spaces between domestic production and public sale. This is not always the case; but the degree of integration of domestic industry into the rest of the household is striking. Finally, there is the question of who actually worked the equipment found in the archaeological record, the loomweights, grindstones, olive presses and the like. Most historians and archaeologists estimate that households in ancient Greece ranged in size from four to six people, plus household slaves—a nuclear family (Gallant 1991). But some houses at Olynthus had equipment requiring a signiWcantly larger number of workers to operate it. We have houses with enough loomweights weights for 4 to 15 looms, implying a substantial number of women to keep all those looms occupied. The twelve upper grindstones in house A 6 suggest that a dozen or so workers were involved in grinding grain for Xour or meal. Closely tied as it was to the daily domestic areas of the household, industry

60

Nicholas Cahill

clearly had an important effect on the organization and makeup of the household. There is no direct evidence for who these workers were, whether they were members of an extended family, hired labor, or slaves. This is a key question, however. Under some circumstances, members of the extended, free family could be put to work. One is reminded of the passages in Xenophon’s Memorabilia, in which Aristarchus explains to Socrates how his female relatives have been forced to move in with him, swelling his household to 14 members plus slaves (whom he does not enumerate). Aristarchus claims that he cannot provide for them all, but Socrates asks why they should starve, since other, equally large households are making money through household industry: by making groats, baking bread, weaving clothes and the like—exactly the sorts of industries most commonly attested at Olynthus. Aristarchus answers that his household is composed of free women, theirs of slaves. To this Socrates simply suggests that he put them to work doing jobs which they already, as well trained Greek women, know how to do: weaving, cooking, baking and the like. Aristarchus, taking the conventional position, protests that these are not suitable activities for well-bred women; but Socrates (and economic necessity) prevails (Xenophon, Memorabilia 2.7). Despite Socrates’ practical but unorthodox viewpoint, the expectation here and elsewhere is that household industry was performed by slaves, and only under exceptional circumstances were members of the free, extended family expected to participate in proWt-making endeavors. This is, of course, a social norm rather than a historical reality. Ideally, we might be able to distinguish slave from free occupants of a household by looking at, for instance, the eating and drinking utensils, cooking areas and the like. But Olynthus was not carefully excavated enough to make this possible; and in general, the material culture of slaves has not been easily recognized in the archaeological record (Morris 1998). Nonetheless, given the widespread expectation that such work would be done by slaves, I suspect that these households owned and employed relatively large numbers of them.

Household Industry at Sardis A comparison of this situation to that of non-Greek houses puts these patterns in a wider perspective. At Sardis, the capital city of the Lydians in western Anatolia, recent excavations have uncovered parts of unusually well-preserved houses dating to the Archaic period (Fig. 4.3). The houses are located just inside the fortiWcation wall of the Lydian city at its western edge. Built in the late seventh or early sixth century, they were destroyed, together with the fortiWcation, when Cyrus the Great captured

Household Industry in Greece and Anatolia

61

Sardis about 547. The houses were buried under destruction debris from their own walls and from the nearby fortiWcation, and left more or less untouched afterwards. They are therefore similar to the houses at Olynthus in being short-lived and violently destroyed, leaving a huge number of artifacts on the destruction Xoors which document the use of space in the Wnal phase of the house’s life. On the Xoors of one partlyexcavated house, for instance, were found some 224 complete vases, 4 sets

Fig. 4.3. Sardis, Archaic Lydian houses from the western edge of the city.

62

Nicholas Cahill

of grindstones, 113 loomweights, as well as a variety of objects of iron, bronze, bone, ivory and other materials, some of the most complete household assemblages known. The deep burial of the houses and the wellpreserved Roman remains above them limited the area that could be excavated; indeed, only parts of three houses have been uncovered. But they show interesting evidence of household industry, and this raises questions about the relationship between industry and household organization (Cahill 2000). In the larger area excavated, shown in Figure 4.3, two different houses can be distinguished. One consists of an open courtyard (Areas 4 and 6), with two or more rooms opening onto it: a kitchen (Area 3) and a workshop for glassworking (Area 5). The second house (Area 1) was added to the south of this house in the Wrst half of the sixth century B.C. A small sondage about 50 m. south exposed part of another house; the impression is that this whole region of the city was densely occupied. Unlike the other rooms excavated so far, which were left undisturbed, the northernmost room (Area 5) was dug out shortly after the destruction, apparently to recover its valuable contents. In the backWll of this room were found about 3.5 kg of opaque red glass cullet and 0.5 kg of transparent yellowish cullet. Other Wnds from the room included one complete melon bead made of segments of different colored glass, fragments of another similar bead, a small saw, a bronze bowl or crucible, and other tools. In the two southern corners of the room were mudbrick and stone benches with blackened top surfaces, apparently used in the working of glass. This room, then, was probably used as a workroom for making glass objects. One unusual feature of this room is its relatively elaborate decoration. The walls of other rooms of this house, like most houses here and elsewhere in the Archaic period, were either left as plain mudbrick, or were plastered over with mud plaster. The walls of this room alone were plastered in a Wnely polished plaster with a great deal of straw temper and a sparkling, golden-colored micaceous wash. This covered not only the south and west walls, but also the working benches, whose edges were also reinforced with wooden boards, now burned away. Cullet is raw glass awaiting melting or molding into Wnished objects. This particular type of opaque blood-red cuprite glass is known primarily from Near Eastern sites, from Nimrud, Persepolis, Hasanlu, and is described in a Middle Babylonian glass-making text (Brill and Cahill 1988). It is not known from Greece at this early period. It is unclear whether the cullet was manufactured here from raw materials, or was manufactured elsewhere and then softened and worked in this workshop. The manufacture of glass from silica and other raw ingredients would require a furnace, which was not located within the excavated

Household Industry in Greece and Anatolia

63

area, although it could have been in another part of the house. This absence might suggest that the cullet was manufactured elsewhere and only worked into Wnished articles here. On the other hand, among the Wnds from the room were three fragments of coral, which are also mentioned in the Akkadian glass-making texts as one of the raw ingredients of glass (Oppenheim et al. 1970). Other raw ingredients, such as plant ash and sand or pebbles, would not have been recognizable in the burned and churned-up debris of the room. The composition of this cullet also differs from other contemporary samples, in having little or no lead in it; this might point to a local origin, at Sardis if not in this house, rather than being imported from afar. Evidence of other household industry was found in the courtyard of the house. A reused, neckless hydria was found shattered on the ground; scattered around the fragments were many small lumps of slag-like material. This substance is currently under analysis; tentatively it seems to be the vitriWed leavings of some kind of iron working or smelting operation. Why this worthless material would be carefully collected in the courtyard rather than being discarded is unknown, but it seems to document the involvement of this household with another type of industry. The details of glass manufacture in this relatively early and wellpreserved workshop are important and interesting, but not the primary subject of this chapter. Instead I want to consider the relationship between industry and other domestic activities in this house. The other excavated spaces of the house were not primarily industrial, but contained the usual range of domestic artifacts and installations. One room was a kitchen, with a bench for grinding grain, a hearth for cooking and baking, and a large number of vessels for cooking, eating and drinking—including 7 cooking pots, 5 cooking pot stands, and a pile of 23 stemmed dishes. The partly excavated courtyard contained another hearth, with more cooking vessels on it and nearby; 3 pithoi; 113 loomweights, 2 more sets of grindstones, and large quantities of pottery and other Wnds. The quantity of vessels for cooking, eating and drinking in this house is quite astonishing. The small space—only about 80 m² of the domestic portion of this house, less than half the area of a typical Olynthian house—contained cooking equipment including at least 17 cooking pots, 10 cooking stands, breadtrays, and a spit. Some of these were set up on hearths and associated with food, others stored on shelves or set up at the base of the wall, ready to be used. Eating and drinking vessels were also remarkably numerous: 31 stemmed dishes, 30 skyphoi, 16 oinochoai, 17 lydia, 10 lamps, and 78 other vessels. This number of vessels for cooking, eating and drinking, and the quantity of other equipment, such as the 4 sets of grindstones and 105 loomweights—enough for more than one loom—raises the issue of household

64

Nicholas Cahill

size. This incompletely excavated house had pottery to feed two dozen and more people simultaneously, multiple cooking areas with many cooking vessels of different sizes and types, and grindstones and loomweights to employ a number of women. The assemblage contrasts with the (also partially excavated) house to the south. Here, a much smaller space was dug, but it offered evidence for only domestic activities, in quantities much more consistent with a nuclear household. It contained, for instance, 8 large stemmed dishes rather than 30, 4 cooking pots rather than 17, 1 hearth, 7 skyphoi instead of 31, and 9 oinochoai (some of which were quite small) rather than 19. Obviously this is a much smaller area than that of the northern house; but there are other signiWcant differences. For instance, activities which in the northern house are spatially separated take place in a single space here, suggesting a smaller area and household size. Moreover, while the proportions of vessel types are similar in the two houses, the absolute quantities are different. The enormous assemblage of objects in the northern house for preparing food, cooking, eating, drinking and other household activities points to a much larger household size than expected, sufWcient for three or more nuclear families. I would propose that this is connected to the presence in this house of domestic industry. Some have questioned whether this building was actually a house, or was primarily another sort of structure—a glass workshop with attached living quarters, for instance. We would want to see the structure completely excavated, of course, but ultimately this turns into a matter of terminology rather than substance. In my view, a building which shelters a full range of domestic activities, such as food storage, preparation, cooking and eating, and weaving, is best characterized as a house, even if other activities go on in it as well. And as we have seen, domestic industry plays an important role in the houses of other cultures, such as the Greeks. We would not deny that the houses at Olynthus are houses, simply because they contained equipment and spaces used for nondomestic production. The questions really are, what kinds of activities characterize this building—both domestic and industrial—and how does this affect the nature and makeup of the household? If we assume that the household here was larger than the normal nuclear family, were the people who lived here related by birth, by occupation, by some hierarchical relationship—master and slaves—or what? What brought this household together? This sort of question is among the most difWcult to get at from the archaeological record alone, particularly when we have only part of a house to deal with. But one of the most striking Wnds from the houses was a pile of 23 identical stemmed dishes on the Xoor in the northeast corner

Household Industry in Greece and Anatolia

65

of the kitchen. Most of these, 18 of the 23, bear marks, incised after the dish was Wred, some of which are repeated on more than one dish. For instance, two dishes bear a cross in the center of the plate and a pi-shaped sign on the underside. Three different dishes bear a snake-like symbol, and two dishes bear a labrys-like symbol—one twice. All these are incised in different “handwritings,” showing that they were done by different people. It is not clear just what these marks represent. Other stemmed dishes from Sardis bear incised grafWti; indeed grafWti seem most common on this shape. But this collection of dishes and marks is not easy to parallel or explain. I suggest that the marks identify the owners of the dishes, and that shared marks identify shared social ties among the owners—for instance, ties of blood. We could have a group of related families inhabiting this house, sharing tasks of food preparation and eating and drinking. On the other hand, there is no evidence in the pottery and other Wnds for any social hierarchy such as between slaves and free inhabitants: the dishes and cups are all virtually identical. The situation is therefore perhaps quite different from that of the Greek household industry. In conclusion, we must be careful to think away our industrial and post-industrial expectations and perspectives when studying the ancient household and economy. In the modern world we expect the workplace to be separate from the home, we expect people to commute to work and leave their business there. This was clearly not the case in antiquity, where home and workshop were often one and the same place. But the impression given by so many ancient—and modern—writers, that industry and those who practiced it, were socially and physically marginalized ought to be reconsidered. Granted, in Greece, much of the household industry may have been done by slaves rather than free members of the household. But the industry itself is not relegated to marginal areas of the household or of the city. It is front and center. The social situation in other cultures such as Lydia, on the other hand, was quite different from that in Greece; and household industry may have taken place in a very different environment, and affected the household in a very different manner. Literature Cited Amouretti, M.-C. 1992. “Oléiculture et viticulture dans la Grèce antique.” In Agriculture in Ancient Greece: Proceedings of the Seventh International Symposium at the Swedish Institute at Athens, 16–17 May, 1990, ed. B. Wells, 77–86. Stockholm: Swedish Institute at Athens (Svenska instituet i Athen 4o.42). Amouretti, M.-C. and J.-P. Brun, eds. 1994. La production du vin et de l’huile en Méditerranée. BCH Supplement 26. Paris: Diffusion de Boccard. Brill, R. H. and N. D. Cahill. 1988. “A Red Opaque Glass from Sardis and Some Thoughts on Red Opaques in General.” Journal of Glass Studies 30, 16–27.

66

Nicholas Cahill

Cahill, N. D. 2000. “Lydian Houses, Domestic Assemblages, and Household Size.” In Across the Anatolian Plateau: Readings in the Archaeology of Ancient Turkey. Annual of the American Schools of Oriental Research 57, 173–185. ———. 2002. Household and City Organization at Olynthus. New Haven, Conn.: Yale University Press. ———. In press. “Lydian Houses, Domestic Assemblages and Household Size.” Near Eastern Archaeology. Gallant, T. W. 1991. Risk and Survival in Ancient Greece. Stanford, Calif.: Stanford University Press. Graham, J. W. 1953. “Olynthiaka 2: An Industrial Establishment on the North Hill.” Hesperia 22, 196–207. Morris, I. 1998. “Remaining Invisible: The Archaeology of the Excluded in Classical Athens.” In Women and Slaves in Greco-Roman Culture: Differential Equations, ed. S. R. Joshel and S. Murnaghan, 193–220. London: Routledge. Oppenheim, A. L., R. H. Brill, D. Barag, and A. von Saldern. 1970. Glass and Glassmaking in Ancient Mesopotamia. Corning, N.Y.: Corning Museum of Glass Press.

Chapter 5

Living and Working Around the Athenian Agora: A Preliminary Case Study of Three Houses Barbara Tsakirgis

Since 1931, archaeologists working under the auspices of the American School of Classical Studies in Athens have uncovered major portions of the Classical Athenian Agora. In the process of revealing its public buildings, the excavators have exposed the remains of residential and industrial districts on the periphery of the public space. These areas, located largely on the slopes of the Kolonos Agoraios and the Areopagus, and in the valley between the Areopagus and the Pnyx, were clearly densely populated from the Wfth century onward, and the houses located there offer a view of Wfth-century private life that balances the better-known public monuments (Thompson and Wycherley 1972, 173–185). While many of the houses were simply places of residence, some served a dual purpose as places of living and working. As part of a larger project to describe and analyze the domestic remains around the Athenian Agora, I here examine these habitation areas and enter speciWcally into three houses so that I might consider the evidence for living and working on the margins of the civic space of the Agora. The American excavators have uncovered about a dozen houses belonging to the Classical period. Exploration around the public center indicates that while there is some housing in the Late Archaic period to the north, west, and south of the area later occupied by the Classical Agora, habitation becomes concentrated in these quarters after the Wrst quarter of the Wfth century, probably after the clean-up of central Athens following the Persian invasion and the rebuilding of the public space. Evidence of Late Archaic houses includes the incompletely preserved walls below the later Eleusinion (Miles 1998, 12–16), one house under the Classical block on the north slope of the Areopagus (Fig. 5.1), and the partially extant remains of a house under the Roman temple next to the Stoa Poikile (Shear 1997, 512–514); the most eloquent feature of this last house is

Fig. 5.1. Athens, block of houses south of South Stoa I (courtesy of the Agora Excavations, American School of Classical Studies at Athens).

Living and Working Around the Athenian Agora

69

its well, Wlled after the Persian invasion of the city (Lynch 1999). Literary evidence locates Classical houses to the north, west, and south of the public area, all areas where such structures have in fact been found. The Classical houses thus far excavated accord with Demosthenes’ statement (3.25–26; cited in full by Ault, Chap. 9) that the Athenian house, even that of a wealthy and inXuential man of the Wfth century, was not an extensive or impressive place. First, a word about the typical Athenian Classical house plans. Unlike the houses at Olynthus, or Piraeus, or other sites where residential districts were planned all at one time, the Athenian houses vary considerably in size and plan.1 The Classical Athenians did not take advantage of the post-Persian clean-up to rebuild in regular blocks with uniform house lots, and thus only one block of Classical houses, south of South Stoa I, is known in Athens (Fig. 5.1; Thompson and Wycherley 1972, 177–179); within this block located on the north slope of the Areopagus there is no uniformity in the dimensions of the houses. A roughly centrally placed courtyard is a standard feature of each house; this court is usually entered directly from the street, rather than from a vestibule. A few of the courts have traces of posts which probably supported a simple shed roof to form a basic portico. There is no good evidence for a complete peristyle in any Classical house around the Agora.2 While the houses in the Classical block are generally rectangular or square in their footprint, other Athenian houses of the Classical period, including several we will look at closely, are not. The irregularity of these other houses is due to the lack of a street grid in Athens, a feature well known in contemporaneous and slightly later residential areas in Piraeus and Olynthus, as noted above. Variable also is the number of rooms and the range of types encountered. While the half dozen or so houses in the Classical block lack a vestibule, some others in Classical Athens (e.g., Houses C and D in the Industrial district; Fig. 5.4, below) do possess this feature which could serve both as entrance to the house and buffer between the inhabitants in the house and the outside world. All of the houses examined here, and in fact all but two in this area of central Athens, lack a recognizable andron, the setting for the symposium.3 Whether the absence of this distinct architectural setting speaks against the symposium having taken place in these houses, or whether we should see the event of the symposium even in houses where there is no formal andron, is a point I will return to later. Literary evidence tells us that houses could be used for industrial as well as for domestic activity. Demosthenes’ inheritance included a house with slaves who labored in the attached workshop (27.19.26; 28.12). While some workshops must have been large in size, for example, that which accommodated the 120 slaves fashioning shields (Lysias 12.19), most

70

Barbara Tsakirgis

factories of the Classical period were fairly small (Finley 1974, 137). Thus the large cobbler’s workshop mentioned by Aeschines in Against Timarchus (97) might well have Wt into one of the houses to be examined here; the workshop had probably nine or ten slaves. The three houses considered below are all to the southwest of the Classical Agora. Two streets exit the public space at its southwest corner and run toward the Pnyx; the House of Simon (Fig. 5.2) is bordered on its east and northwest by these two streets. The house is located at the very edge of the open area of the Agora; one of the horos stones of the civic center abuts the acute northeast angle of the building. The House of Mikion and Menon (Fig. 5.3) is just down the street from Simon’s place, on the easterly of the two streets. The third house to be surveyed was built originally as two houses in the so-called Industrial District (Fig. 5.4), further to the southwest from Mikion’s and Menon’s house, in the valley between the Areopagus and the Pnyx. This general area to the southwest of the Classical Agora seems, from the evidence of these houses and other buildings, to have been given over to the residence and workspace of craftsmen. The incompletely excavated remains of the House of Simon belong to a lot irregular in shape (Fig. 5.2); the house was not completely excavated due to the overlying Hellenistic Middle Stoa and the Roman period civic ofWces (D. Thompson 1960; Jones 1975). What was revealed of the house, constructed after the post-Persian clean-up of Athens, is an open central court, Xanked on the east by two rooms and on the west by one. The court’s northwest wall is the street wall, and there is no excavated trace of the street entrance into the house. One earlier well underlies the court and a second is contemporaneous with the use of the house into the fourth century. When Dorothy Thompson excavated the house in the 1950’s, she found the beaten earth Xoors of both the court and surrounding rooms carpeted with short iron nails. Averaging 0.015 m in length, of the hundreds recovered, many have corroded and fused together into large masses in the Agora Excavations’ storage tins. In and among these nails, Thompson also recovered numerous bone eyelets (with an average diameter of 0.0155–0.025 m), which she interpreted along with the nails as working material for the cobbler’s trade. Thompson combined these humble remains with a cup base inscribed “belonging to Simon” and proposed that the cobbler resident in this centrally located house was none other than Simon, the natural philosopher and friend of Socrates, who, according to Diogenes Laertius (2.13.122), was accustomed to visit Simon in his shop and discuss philosophy. This is not the place to argue for or against Thompson’s proposal of the cobbler’s identity, although I should note that, while the nails and eyelets were recovered inside the house, the cup base was found outside

Living and Working Around the Athenian Agora

71

in the street; Thompson suggested that its Wndspot outside the house may have indicated its use as a doorknocker and street sign. What is relevant for the present discussion is that the working materials were recovered all over the excavated parts of the house. Whether once stored in now decomposed cloth or leather sacks, or sitting on wooden shelves, the nails and eyelets suggest by their distribution that the cobbler (Simon or

Fig. 5.2. Athens, House of Simon (courtesy of the Agora Excavations, American School of Classical Studies at Athens).

Fig. 5.3. Athens, House of Mikion and Menon (courtesy of the Agora Excavations, American School of Classical Studies at Athens).

Fig. 5.4. Athens, Houses C and D, Wfth and fourth centuries (courtesy of the Agora Excavations, American School of Classical Studies at Athens).

74

Barbara Tsakirgis

otherwise) used both the courtyard and the roofed rooms as working space. This arrangement is logical enough, especially if we factor in time of day and season into the interpretation of the house. The courtyard and its available well water could have been the setting for work on a warm, sunny day, and the sheltered rooms opening off this central unroofed space could have served the same purpose in inclement weather. Also found in the house, and not part of Thompson’s original discussion, were two fragments of obsidian blades or bladelets (0.019 and 0.021 m in length respectively), possibly used for cutting leather, and a bone needle, possibly for stitching the leather. The obsidian blades could have been used in a knife like that held by a cobbler at work on a black Wgured pelike by the Eucharides Painter in the Ashmolean Museum (Beazley 1956, 396.21; illustrated in Boardman 1974, 143, Wg. 229).4 In 1932, Dorothy Burr excavated the southern limits of what was later to be called the House of Mikion and Menon (Fig. 5.3), but the area was so disturbed that the exploration was abandoned after one season. Thompson eventually published the contents of the so-called Demeter Cistern that was later recognized as one of two in the house’s courtyard (D. Thompson 1954). In the late 1960’s excavators returned to the area and revealed a many-roomed house which clearly served as both residence and workshop from just after the post-Persian clean-up down into the third century (Shear 1969, 383–394; Miller 1974; Rotroff 1997, 24– 25). Like the so-called House of Simon, the House of Mikion and Menon is arranged around an irregular but roughly central court which has one wall bordering on the street. As can be seen from the plan, the excavators restored a possible entrance from the street into the court, for want of a better or certain entrance. This court is occupied by two cisterns, one of which was dug through by a well in the latest phase of the house. The cisterns contain a mix of both household and industrial debris, including marble chips, fragments of a marble basin reused for sculpting practice, unWnished sculpture, sculptor’s tools, various pots and pans (especially kantharoi), terracotta Wgurines, and foodstuff (olive pits, grape seeds, and perhaps apricot pits.) In this house, the Xoors were littered with marble chips, dust, and unWnished sculpture, largely found in the rooms which border the street on the east and south. No iron chisels, punches, or points were found,5 but among the sculptor’s debris were the inscribed bone stylus from which is derived the name of the earliest resident of the house (MIKION EPOI[ESEN]) as well as numerous lead strips, a stone pounder and a faceted piece of pumice. These last two items were found with one of the lead strips in 1932 and were recently uncovered in the storage containers from that season’s excavations. The lead strip and six others like it, which are approximately square in section and average 0.10–0.15 m. long, have

Living and Working Around the Athenian Agora

75

several parallels from the Agora, two of which were also found in association with marble and poros chips.6 Similar strips have been found at Aegina (Furtwangler 1906, 424), Nemea (Zimmer 1990, 56), Isthmia (Rostoker and Gebhard 1980, 350; 360), and from the Workshop of Pheidias at Olympia (Schiering 1991, 164–167). Hansgeorg Bankel identiWed the Aeginetan examples as pencils used for making preliminary drawings on marble (Bankel 1984), and the Wndspot of the Athenian lead strips, in and amongst the debris of stoneworking, suggest that they were for the same purpose. Bankel demonstrated the line the lead strips would leave when drawn across a piece of stone. A photograph taken under a microscope of one of the examples from the Agora also provides evidence for their use. The pointed end of the pencil is compressed, a condition produced when drawing on stone takes place. The stone pounder, smoothed for comfortable holding on one end, is fractured on the other, where the stone repeatedly struck something else, perhaps the end of a chisel. The faceted piece of pumice represents the Wnal phase of the marble working process; it is a polisher, rubbed into planes by the repeated smoothing of the sculpted marble. These tools were found with the marble chips in the rooms on the southern side of the house; little marble and no tools were recovered in the incompletely excavated northern section of the house. From the very disturbed areas of the western rooms come a few pieces of wall plaster, painted red and black; since virtually no walls were found in this area of the house, it is impossible to say if their Wndspots prove them to have been originally on walls in this part of the house. The incompletely excavated rooms to the north of the courtyard and its two cisterns seem to be laid out around a second courtyard, perhaps the focus for domestic life in the building. Parallels for stoneworking and sculpting establishments are surprisingly difWcult to uncover in a world accustomed to seeing sculpted and inscribed stone. A sculptor’s workshop in southern Italy at Baiae is best known for its plaster casts (Zimmer 1990). While the Workshop of Pheidias at Olympia produced parallels for the lead pencils, the other materials (such as fragments of glass and the large scale terracotta moulds used for forming it) found there and the location of the workshop itself are very unlike the Wnds and location of the House of Mikion and Menon (Schiering 1991). A slightly better, although much later parallel is the fourth-century sculptor’s workshop at Aphrodisias, where numerous unWnished large-scale sculptures were found (Rockwell 1991). The two rooms of the studio at Aphrodisias are thought to have been both workrooms and studio for the display of the statues to customers, an arrangement similar to that which I propose for the courtyard of the House of Mikion and Menon.

76

Barbara Tsakirgis

Rodney Young excavated further south the so-called Industrial District in the 1930s and 1940s (Young 1951); there he found marble chips in numerous layers throughout the area, and accordingly named the street which crosses it the Street of the Marble Workers. Houses C and D (Fig. 5.4), which lie on the east of the Street of the Marble Workers, were built, like the House of Simon and the House of Mikion and Menon, after the post-Persian clean-up of Athens. The two Wfth-century Houses C and D have plans similar to one another, with an entrance vestibule from the street, a central court, and the larger rooms placed on the north side of the house.7 The two houses were combined into one in the fourth century, probably after the construction of the Great Drain, and at that time the northern House D was given over to industrial activity while the southern House C seems to have been used for more exclusively domestic purposes. In the fourth-century level of the courtyard of House D there was a hearth and associated slag of both iron and bronze, all of which suggest an industrial use of the house court in the second phase of the building. Since no casting pit of the sort seen elsewhere around the Agora was found here, the house was probably used as a smithy for reshaping reWned or already cast metal rather than as a foundry.8 A curse tablet found in the Wll of the house contains imprecations against bronze-workers, and has been used to support the identiWcation of the fourth-century space as a bronze smithy. A recent republication of the tablet has cast doubt on the signiWcance of its Wndspot (it should be in a well or underground in order to be close to the chthonic deities who would oversee the curse), and the authors suggest that the tablet was unknowingly dug up elsewhere, mixed into earth to create the mudbrick walls, and deposited only secondarily into the house when the walls collapsed (Curbera and Jordan 1998). If they are correct, the curse tablet should be removed from association with the house’s occupant. The metal slag and the hearth support the location of metalworking activities here, even without the conWrmation of the tablet. That the southern half of the house, the original House C, was used in the fourth century as the living quarters can be seen most clearly from an assemblage of items found in one room on the eastern side of the court. Twenty-two pyramidal loomweights, a spindle whorl, and portions of a brazier of the broad Classical type speak clearly of women’s work (Amyx 1958, 229–230). This is the largest single collection of loomweights recovered from one place in an Athenian house; unfortunately their distribution at the time of recovery, in a heap or fallen in a line from the loom, is not known. The brazier, the spindle whorl, the loomweights, and presumably a warp-weighted loom could have been used here, or in the neighboring courtyard when weather permitted. With water from the courtyard well and access to a number of other rooms around the unroofed

Living and Working Around the Athenian Agora

77

space, the original southern house provided ample accommodation for the domestic activities of the people living here. How are the three houses considered here alike? Both the House of Mikion and Menon and the combined House C/D seem to indicate a logical desire to separate the living and the working spaces; there is not enough preserved of the House of Simon to be able to tell whether a similar separation existed there. The living area of House C/D is clearly located in the southern portion, the original House C, while the incompletely excavated rooms northwest of the court in the House of Mikion and Menon seemed to have served the same purpose. Courtyard space is essential to all three houses, but admittedly it is practically universal in Classical houses; none of the courts has a colonnade, but this lack is also common for the period. In the House of Simon and the House of Mikion and Menon, the court borders on the street and in the House of Mikion and Menon it may possibly have doubled as the entrance to the house. This openness and easy access may have provided not only welllit working space for the cobbler and sculptor alike, but also a large and convenient “showroom” for customers. Note that the later house/workshop C/D has a vestibule in each half of the building and thus a buffer between both industrial and domestic interior and the street. Why this buffer was necessary in the smithy is a mystery; perhaps it exists because of the noise and heat of the smithing work or is simply a survival from the earlier plan. The House of Simon again offers only an incomplete picture, as it could have had a vestibule in the unexplored portion of the house. Absent from all three houses, at least in their excavated state, is the square room with raised border and off-center doorway that we have learned to call the andron, the men’s room used for dining and drinking. Does the absence of a formal room for the andron mean that such an event, integral to men’s bonding and entertainment and the very social structure of the polis, did not take place in these houses? I suspect not, since the symposium is a transient activity practiced by men, not a room, and three, Wve, or more couches could be arranged temporarily in one room of these houses when needed, and moved to a more convenient spot when not.9 The modest Late Archaic and Classical house located west of the Stoa Poikile has no andron, yet Kathleen Lynch in her careful study of its well deposit has shown that drinking in both informal groups and formal arrangements occurred there (Lynch 1999). The question is, of course, still open because of the incomplete excavation of two of the houses and because of the possibility that any or all of the three were occupied by metics, whom we might not expect to host or regularly participate in the symposion. Metics were known to be regularly employed in the banausic trades carried on in these three houses and so could have

78

Barbara Tsakirgis

been the residents here. For example, Cephalus, the owner of the shield factory mentioned above, was a metic. While the Athenian workshop houses are on the margins of the Agora, they are in the heart of densely populated residential areas. Each house has not only a street-side location on a major thoroughfare leading from the Agora to important destinations within and without the city, for example, the Pnyx and the Piraeus, but each is also located at a crossroads, a position which allows ready access by potential clientele as well as the possible opportunity to advertise wares and services to many passersby. Aeschines (1.124) says that shops in the Athens of his day are located along streets, and Isocrates (Areopagiticus 15; Against Callimachus 9) mentions that shops were places to sit and talk. Another feature of Athenian shops mentioned by Aeschines (1.124) is the mutability of the space. In noting the street-side location of shops, Aeschines also says that a shop is known and called by the occupation of the owner or resident. Thus any single house can be a doctor’s ofWce, a laundry, or a smithy, depending on the profession of the man who works currently there. This mutability of space can be seen particularly in House C/D, and is especially possible when there are few or no industrial installations necessary for a trade. Thus the domestic use of the original House D can easily be replaced by the smithy, with the simple addition of a hearth for heating the metal to sufWcient softness for working. At numerous other sites around Greece there are houses with workrooms and fairly heavy industry located in the heart of residential areas. Five examples, dated from the Wfth through the Wrst centuries, illustrate this mix of industrial and domestic establishments. The plan of House A vii 8 at Olynthus was altered sometime after the original construction; the original andron and anteroom on the eastern side of the house was converted into a workroom (Hoepfner and Schwandner 1994, 112). Access to this new installation was from the house’s original court, itself entered only from the street door. Similar conversion of living space into industrial is seen in the third century at Kassope in House 5, where a potter’s kiln was built into part of the house once occupied by the hearth-room, a characteristic space in northwestern Greek houses (Hoepfner and Schwandner 1994, 157; Zimmer 1999, 565). As in the converted house at Olynthus, the area containing the kiln is entered from the original courtyard of the house. Similarly, oil presses are an integral part of the houses on the Rachi at Isthmia (Anderson-Stojanovic 1996). To the House of the OfWcial at Morgantina in central Sicily, in its Wrst-century phase, a potter’s workshop was added; the workshop, including three kilns, occupied the northern court and rooms of the very sizeable two-court house (Tsakirgis 1984, 223–224; Cuomo di Caprio

Living and Working Around the Athenian Agora

79

1992, 16–20). At least one of four shops attached to the House of the Mended Pithos, also at Morgantina, contained metal waste similar to that found in the smaller northern court of House C/D in the industrial district in Athens and was likely also used as a smithy (Tsakirgis 1984, 170). These Wve examples reveal both similarities to and differences from the house-workshops in Athens and suggest that there was no single model followed for such establishments in the Greek world. At Olynthus, Kassope, the Rachi, and in the House of the OfWcial at Morgantina, there was very little separation of the areas used for industrial activity from those used for domestic life. While that separation was likely the ideal, as suggested by the House of Mikion and Menon and House C/D in Athens, space and available Wnancial resources probably did not always allow the homeowner to create distinct areas for living and working. In the House of the Mended Pithos, the two areas are so clearly distinct, with no through passage from one to the other, that the possibility exists that the homeowner was not the smith working in his shop, but perhaps merely the landlord of the smith. One clear similarity does remain between these Wve parallels outside Athens and the houses near the Athenian Agora; the separation of residential and industrial activities in a greater setting, that is, city blocks, was not the norm in Classical Greece. A homeowner could Wnd himself living next to a marbleworker or a smith or a dyeworker, a violation of the zoning laws current in North America and many other industrialized nations today.10 Before concluding, I would like to note that not all workshops were combined with houses in classical Athens. In a study of the Attic orators and the archaeological evidence, M. Bettalli explored whether workshops or ergasteria were locations speciWcally and exclusively given over to public shops and work spaces (Bettalli 1985). Certainly the buildings explored here served as both workshops and houses, and, as I have suggested, salesrooms, but there are some structures in the vicinity of the Agora which were not obviously places of residence, with the possible exception of a shop slave or two. Several multi-room complexes, including a Classical example on the north side of the Athenian Agora being studied by Thomas Milbank, seem to have been on the order of the synoikia mentioned by Aeschines (1.124; see also Ault, Chap. 9), a multiroomed building rented by several different men to serve an industrial function. The Wnds are almost purely industrial in nature and the buildings lack any of the architectural characteristics of a house, for example, a vestibule, or a centrally located courtyard. That such structures existed in Classical Athens along with the house/workshop speaks of a certain Xexibility of the Athenians in their assignment of function to space.

80

Barbara Tsakirgis

Notes 1. For the urban plan and houses at Olynthus and Piraeus, see Hoepfner and Schwandner 1994, 68–113, 22–42. 2. The central of three houses in section Omega on the north slope of the Areopagus is restored with a complete peristyle. The reader should examine both the actual state plan and the restored plans as published in Shear (1973, 146–156). The state plan shows that there is no physical evidence for a complete peristyle; there is merely space for it in the center of the very incompletely preserved house. 3. Those with an andron are the central of three Classical houses under a large Late Roman house on the north slope of the Areopagus (Shear 1973, 146–156) and the so-called House of the Greek Mosaic in the valley below the Pnyx (H. Thompson 1966; Thompson and Wycherley 1972, 180–182; and Graham 1974, 47, where the house is called the House of the Wheel Mosaic). 4. The obsidian bladelets have been lost, so it has not been possible to examine their surfaces for traces of the abrasions that might have resulted from leatherworking. 5. That iron tools were used by sculptors is known from epigraphical evidence, e.g., Kirchner 1927, no. 1673 (iron tools used at Eleusis; old ones sold as scrap for the new); Dürrbach 1912, no. 161, l.107 (iron tools for stone workers at Delos.) Iron tools, two point chisels and a Xat chisel, were recovered from a later (third to fourth century A.D.) sculptor’s workshop at Aphrodisias (Rockwell 1991, 139.) 6. One was found in the working level of the Stoa of Zeus. The other was recovered in section BE. 7. This type can now be called, after Nevett 1999, the “single-entrance, courtyard house.” 8. For evidence of metal-working found around the Athenian Agora, see Mattusch 1977. 9. This opinion is argued more fully and more forcefully by Lynch in press. 10. This mixing of residential quarters and industrial activity is explored in Schwandner (1988), as well as by Cahill (Chap. 4) and Trümper (Chap. 8). Literature Cited Amyx, D. 1958. “The Attic Stelai: Vases and Other Containers.” Hesperia 27, 163–307. Anderson-Stojanovic, V. R. 1996. “The University of Chicago Excavations in the Rachi Settlement at Isthmia.” Hesperia 65, 57–98. Bankel, H. 1984. “Griechische Bleistifte.” Archäologische Anzeiger 1984, 409–411. Beazley, J. D. 1956. Attic Black-Figure Vase Painters. Oxford: Clarendon. Bettalli, M. 1985. “Case, botteghe, ergasteria: note sui luoghi di produzione e di vendita nell’Atene classica.” Opus 4, 29–42. Boardman, J. 1974. Athenian Black Figure Vases. London: Thames and Hudson. Cuomo di Caprio, N. 1992. Fornaci e OfWcine da Vasaio tardo-ellenistiche. Morgantina Studies. 3, Princeton, N.J.: Princeton University Press. Curbera, J. B. and D. R. Jordan. 1998. “A Curse Tablet from the ‘Industrial District’ near the Athenian Agora.” Hesperia 67, 215–218. Dürrbach, Felix. 1912. Inscriptiones Graecae. Vol. 11, Pt. 2. Inscriptiones Deli. Berlin: George Reimer.

Living and Working Around the Athenian Agora

81

Finley, M. I. 1974. The Ancient Economy. Berkeley: University of California Press. Furtwangler, A. 1906. Aegina: Das Heiligtum der Aphaia. Munich: Franz in Komm. Graham, J. W. 1974. “Houses of Classical Athens.” Phoenix 28, 45–54. Hoepfner, W., and E.-L. Schwandner. 1994. Haus und Stadt im klassischen Griechenland. Rev. ed. Wohnen in der klassischen Polis 1. Munich: Deutscher Kunstverlag. Jones, J. E. 1975. “Town and Country Houses of Attica in Classical Times.” In Thorikos and Laurion in Archaic and Classical Times, ed. H. Mussche, 63–140. Miscellanea Graeca 1. Ghent: Comité des Fouilles Belges en Grèce. Kirchner, Iohannes 1927. Inscriptiones Graecae. Vol. 2, Pt. 2, Inscriptiones Atticae. Berlin: de Gruyter. Lynch, K. 1999. “Pottery from a Late Archaic Athenian House in Context.” Ph.D. dissertation, University of Virginia. Lynch, K. In press. “More Thoughts on the Space of the Symposium.” In Building Communities: House, Settlement and Society in the Aegean, ed. R. Westgate, N. Fisher, and J. Whitley. Studies of the British School of Archaeology at Athens. Athens: British School of Archaeology. Mattusch, C. C. 1977. “Bronze- and Ironworking in the Area of the Athenian Agora.” Hesperia 46, 340–379. Miles, M. M. 1998. The City Eleusinion. Athenian Agora. 31, Princeton, N.J.: American School of Classical Studies. Miller, S. G. 1974. “Menon’s Cistern.” Hesperia 43, 194–245. Nevett, L. C. 1999. House and Society in the Ancient Greek World. Cambridge: Cambridge University Press. Rockwell, P. 1991. “UnWnished Statuary Associated with a Sculptor’s Studio.” In Aphrodisias Papers 2, ed. R. R. R. Smith and K. Erim, 127–143 Journal of Roman Archaeology Supplementary Series no. 2. Rostoker, W. and E. R. Gebhard. 1980. “The Sanctuary of Poseidon at Isthmia: Techniques of Metal Manufacture.” Hesperia 49, 347–363. Rotroff, S. I. 1997. The Athenian Agora. Vol. 29, Athenian and Imported Wheelmade Tableware and Related Materials. Princeton, N.J.: American School of Classical Studies. Schiering, W. 1991. Die Werkstatt des Pheidias in Olympia, zweiter Teil: Werkstattfunde. Berlin: Walter de Gruyter. Schwandner, E. L. 1988. “Handwerkerviertel in Gründungsstädten des 5. und 4. Jahrhunderts.” In Πρακτικ του ΧΙΙ ∆ιεθνος Συνεδρου Κλασικς Αρχαιολογας, Αθνα, 4–10 Σεπτεµβρου,1983, 4, 183–187. Athens: Hypourgeion Politismou kai Epistemon. Shear, T. L., Jr. 1969. “The Athenian Agora: Excavations of 1968.” Hesperia 38, 382–417. ———. 1973. “The Athenian Agora: Excavations of 1971.” Hesperia 42, 121–179. ———. 1997. “The Athenian Agora: Excavations of 1989–1993.” Hesperia 66, 495–548. Thompson, D. B., 1954. “Three Centuries of Hellenistic Terracottas, I.” Hesperia 23, 72–107. ———. 1960. “The House of Simon the Shoemaker.” Archaeology 13, 234–240. Thompson, H. A. 1966. “Activity in the Athenian Agora 1960–1965.” Hesperia 35, 37–54. Thompson, H. A., and R. E. Wycherley. 1972. The Athenian Agora. Vol. 14, The Agora of Athens. Princeton, N.J.: The American School of Classical Studies. Tsakirgis, B. 1984. “The Domestic Architecture of Morgantina in the Hellenistic and Roman Periods.” Ph.d. Dissertation, Princeton University. Young, R. S. 1951. “An Industrial District of Ancient Athens.” Hesperia 20, 135–288.

82

Barbara Tsakirgis

Zimmer, G. 1990. Griechische Bronzegusswerkstatten. Mainz am Rhein: von Zabern. ———. 1999. “Handwerkliche Arbeit im Umfeld des Wohnens.” In Geschichte des Wohnens, vol. 1, 500 v. Chr.–500 n.Chr. Vorgeschichte—Frühgeschichte—Antike, ed. W. Hoepfner, “Kapital III: Die Epoch der Griechen. 5: Einzelprobleme,” 561–75. Stuttgart: Deutsche Verlags-Anstalt (Wüstenrot Stiftung, Deutscher Eigenheimverein e. V., Ludwigsburg).

Chapter 6

Between Urban and Rural: House-Form and Social Relations in Attic Villages and Deme Centers Lisa C. Nevett

Studies of the size and distribution of the population of Athens and Attica have suggested that for most of the Wfth and fourth centuries more than two thirds of Athenian citizens are likely to have been resident outside the asty, in dispersed farmsteads and, more frequently, in the rural and coastal villages which formed the core of many of the extra-urban Attic demes (for example Rosivach 1993).1 While the city of Athens must have played a major role in the political, economic and social lives of some inhabitants of these demes, epigraphic evidence has shown that for others the main forum for social and political activity is likely to have been the deme center (Osborne 1985, esp. 88–92; Stanton 1994, 217). This material also suggests that there were signiWcant differences between those who chose to focus their activities on the deme, and those who came to prominence on the wider social and political stage of the asty, with the latter group tending to consist of those who were able and willing to spend signiWcant periods of time away from their property in the deme (Whitehead 1986, 317–326). There are, then, likely to be contrasts both in the social experience and possibly also the economic resources of those citizens who were active at the level of the polis as a whole, as opposed to those whose lives were focused more locally. The nature of the sources means that our text-based models for Athenian social relations are likely to be unavoidably dominated by the norms of those citizens who spent time in the city, while others whose lives were mostly restricted to the extra-urban demes are probably underrepresented. I begin to redress this balance by exploring aspects of the social relationships of households located in the Attic demes from an archaeological perspective. I argue that the limited amount of surviving evidence for housing in deme centers indicates that patterns of social interaction may have been somewhat different in these communities from those of the city itself.

84

Lisa C. Nevett

Excavation both at Athens and at other large settlements has revealed high numbers of houses following a relatively standardized pattern of organization, the single-entrance, courtyard house.2 The only street door in such houses generally opens onto an open central court, which in turn provides access to various rooms, their entrances normally shaded by a portico (a prostas, pastas, or peristyle). Various aspects of the organization of space suggest conclusions about patterns of relationships between the individuals who once occupied such houses. The interior provides a closed environment to which access is controlled by limiting contact with the outside world to a single street door. Many houses possess a specialized dining room or andron which could effectively have been used to isolate visitors from the remainder of the household, using a closable door and sometimes also an anteroom. These arrangements suggest that there was a desire to control interaction between household members and the wider community, which corresponds in many respects with what can be deduced about Athenian households from textual sources. In the archaeological record, this pattern of organization is found widely, in urban houses from a variety of areas of the Greek world. As excavation increases the number of houses known, it is becoming apparent that alongside the houses which follow this model, there are also a limited number of others which do not. Many of the exceptions are very small in size, and therefore lack the space for such arrangements (for example, at Athens and Olynthus: Nevett 1999, 156–158). There are also a few larger structures where such measures could apparently have been put in place, but where this was not done. The existence of such houses raises the possibility that their occupants did not closely conform to the social ideals of the oikos, as articulated in the textual and epigraphic sources, and that there may instead have been alternative patterns of daily activity and social interaction. The houses found in the smaller communities of Attica form a coherent group of such structures, offering an opportunity to investigate social behavior and to compare relationships within these households with those of the city of Athens itself.

The Evidence Thorikos Traces of small rural settlements have been found widely distributed through the Attic countryside, but examples of complete houses to compare with the Athenian sample are not, at present, numerous. The most extensive evidence from a single deme center comes from Thorikos, in southeast Attica (principally Mussche et al 1964–1990), where the excavated material includes residential complexes of Wfth- and fourth-century

Between Urban and Rural

85

date (see Figs. 6.1, 6.2). At least some of the houses here seem to have served an industrial and/or a commercial function, in addition to being residential, and this combination parallels observations which have been made in the context of a number of larger, urban sites, including Athens, Olynthus, and Delos (see the chapters by Cahill, Tsakirgis, and Trümper in this volume). At Thorikos and in a variety of outlying structures from smaller, village settlements in the surrounding area, the domestic quarters are often combined with facilities for processing silver ore, which was mined in the area. The important role played by mining and processing raises the possibility that Thorikos may not be a typical example of an Attic deme: it seems likely that as a consequence of the need for a substantial workforce for ore extraction and processing, the population included signiWcant numbers of slaves. If so, their presence may have had an effect on the material culture of the area (compare Morris 1998, 197–211), or at the very least might have resulted in the provision of large amounts of accommodation which lacked the amenities expected by citizen or metic families. This possibility can be explored by comparing the evidence from Thorikos with the limited amount of material available from other Attic deme sites. The houses at Thorikos retain the basic layout seen in the single entrance, courtyard house described above.3 Nevertheless, closer comparison reveals several fundamental contrasts with Athens and other large settlements in the way in which space was organized, and I would suggest that these can be interpreted as indicating underlying differences in patterns of social interaction both between individuals and between households. At Thorikos there is sometimes more than one entrance from the street (for example Figs. 6.1d, f, g). This arrangement would have increased access to the interior of the house by enabling members of the household to come and go relatively easily. There would thus have been relatively limited potential for controlling contact between certain parts of the house and the exterior, in comparison with the single-entrance pattern of organization. Once inside, although the courtyard was a standard element at Thorikos, it often seems to have played a somewhat different, less dominant, role than was the case in the single-entrance, courtyard house, with many of the rooms entered sequentially or as suites, rather than individually from the court or portico (for example Figs. 6.1b, d, f, g). This arrangement would have impeded any attempt to monitor activity in different parts of the house and movement between different rooms, from a single location. A Wnal aspect of the spatial syntax of the Thorikos houses which deserves comment is the fact that there seems to have been less variability in room size than was the case in Athenian houses. In particular, mean

Fig. 6.1. Thorikos, courtyard houses.

a.

Fig. 6.2. Thorikos. a. the southwest slope of Velatouri Hill; b. the Industrial Quarter.

88

Lisa C. Nevett

room size tends to be larger, which suggests that less attempt was made to create intimate and private spaces. This might imply a lack of the kinds of activities for which such rooms were used at Athens, or perhaps more simply, that there was less specialization and more Xexibility at Thorikos in the way in which different spaces were used. Two aspects of the architecture of the Thorikos houses are also potentially important: one of these is the extent of the decoration employed. In comparison with, for example, wealthy Roman dwellings, Greek houses of this period are relatively sparsely decorated. Nevertheless, at Athens and elsewhere, by the beginning of the fourth century BC the larger houses did sometimes feature decorative elements including mosaic Xoors, painted plaster walls and columned courtyards (Walter-Karydi 1998 passim; 1996). Decoration was habitually conWned to the court and the andron (a single room, sometimes with a dedicated vestibule), the probable location for the symposium. At Thorikos peristyle courts are unknown. Mosaic Xoors are also absent even during the fourth century, by which date they were appearing regularly elsewhere. The excavators identify an andron at the site in one house (Fig. 6.1f), based on the presence of red painted wall-plaster (in a room to the northwest of the western court: Mussche et al. 1965, 31 and 33). Nevertheless, additional architectural characteristics found in rooms of this type elsewhere, namely a single off-center doorway and a raised border on the Xoor, are lacking. There is also no decorative mosaic Xoor, and the room has two entrances, which as far as we know would be exceptional for a specialized dining room at this date. Andrones are not completely absent from houses in the region: for instance a relatively standard example of fourth century date was found incorporated into a large compound at nearby Souriza, which comprised both residential quarters and ore processing facilities (Tsaimou 1979 passim; Conophagus 1980, 375–389; Jones 1982, 179). Still, the apparent absence of andrones from Thorikos itself is striking, and, I suggest below, potentially signiWcant in social terms. Evidence for changes made to various properties through time makes it possible to look at the choices made about how to adapt built form and syntax to changing functional and social requirements. Some households sought to increase the amount of roofed accommodation available to them. In the insula 3 tower compound (Fig. 6.1g) the unroofed space was reduced by extending existing rooms out into the courtyard (Spitaels 1978, 57–59). Elsewhere, more dramatic alterations were undertaken which involved the annexation of adjacent land or property, as was done in the western part of insula 3, during the fourth century (Mussche 1990, 57). There is some evidence that the extra space created in this way was used, not only for strictly domestic purposes, but also to provide storage

Between Urban and Rural

89

or workshop space: in washery 1, for example (Fig. 6.1b), facilities for ore-processing were added in the later Wfth century to an existing (earlier Wfth century) residential structure (Mussche, 1998, 50). In some cases existing buildings were also subdivided: for example the western half of insula 3 was split into two separate houses during the Wfth century and continued to be occupied and adapted into the fourth (Mussche 1990, 61). Such changes may have been made in order to suit a succession of different households, or to accommodate an individual oikos during different phases of its life-cycle. One other notable aspect of the domestic architecture at Thorikos is the presence of the well-built circular tower in insula 3 (Figs. 6.1g, 6.2b) which is associated with an early phase of occupation during the early Wfth century (Mussche 1967, 63). The excavators tentatively suggest that this may originally have been part of a country estate which was later incorporated within the boundaries of the settlement (Spitaels 1978, 109). It is the case that such structures have most frequently been found as part of isolated farm complexes, rather than within larger nucleated settlements (see, for example, Lohmann 1989, 64–65 with note 160). Nevertheless, no real evidence is offered in support of the hypothesis that this was the origin of the Thorikos tower, and given that it is surrounded by other buildings it seems equally possible that what we have is a different kind of structure, a tower house inside a nucleated settlement. Indeed, there are two, or possibly three, other round towers at Thorikos, two of which lie on the fringes of the industrial quarter and may therefore also have been incorporated within the built-up area (see Fig. 6.2a).4 Various functions have been suggested for these towers, including a link with the mines (for example, Morris 1998, 209–210). Nevertheless, towers are not limited to mining areas, they are found more widely, both in Attica and in other regions (see, for example, Lohmann 1989, 64–65). It therefore seems likely that they could also have a more general use, and their possible signiWcance is discussed further below. In sum, the spatial syntax of the houses at Thorikos suggests that they were relatively open to interaction with the wider community and that activity in the house itself would have been less easy to restrict than was the case with many of the houses in Athens. The individual household spaces seem likely to have been used relatively Xexibly. Architecturally, the houses show relatively little sign of elaboration, and although alterations were made to some of the Wfth-century structures during the fourth century, these apparently did not include the addition of decorated andrones or colonnaded courts. The most striking architectural features are, instead, the circular towers incorporated into insula 3 and into other structures at the site.

90

Lisa C. Nevett

Ano Voula The more fragmentary material of the late Wfth to mid-fourth centuries date from another deme center, Ano Voula, ancient Halai Aixonides, reveals characteristics which are in many ways similar to those of the Thorikos houses. Unlike Thorikos, Halai Aixonides seems not to have played any specialized economic role, but instead probably relied largely on agriculture to support its population, and is therefore likely to be an example of “what a deme unconstrained by special factors looks like” (Osborne 1985, 26). It thus seems justiWed to take the patterns of spatial organization at the site as an example of more widespread patterns seen in other Attic deme centers. The house reported in the most detail, the rectangular structure on the Papacharalambos plot, was occupied from the late Wfth to mid-fourth centuries, and was progressively extended into the adjacent street (Fig. 6.3d). The interior follows the canonical layout, with rooms opening off a central courtyard (Andreou 1989, 61–62, 1994, 197–201). Like many of the Thorikos houses, however, in at least one phase it had two separate outside doors, at front and rear, and would therefore have been more accessible from the outside world than the single entrance courtyard house. The most extensive evidence of housing at Ano Voula is provided by the irregular insulae on the Kalampoka plot (Travlos 1988, 475; Andreou 1994, passim; Hoepfner 1999, 251–256). The plot includes several complete or partial blocks of buildings which are divided by streets of varying widths and orientations (Fig. 6.4). Outlines of the structures which make up the blocks are sometimes fragmentary and the chronological development of the buildings in the area is unclear. Nevertheless, among the remains a variety of public and private buildings can be identiWed. These include the foundations of three different stone-built circular bases for towers which would have risen to several storeys (Travlos 1988, 467).5 The best-preserved tower is surrounded by a cluster of buildings in the northwestern sector, where both the exterior walls and the interior partitions of an insula are to a large extent preserved. The positions of the doorways are not always clear and the boundaries between the individual properties are therefore difWcult to establish securely. It is possible that, as at Thorikos, plot size changed through time. (The presence of what is apparently a blind alley to the northwest of the tower, running southeast between this block and its neighbour, suggests the possibility that alterations may have taken place: Fig. 6.4) Various alternative interpretations of the use of space in this area can be suggested: the block may have consisted of two courtyard houses, each with an extensive cluster of ancilliary buildings which were perhaps used for production or storage. Hoepfner has argued that these structures

Fig. 6.3. Ano Voula, remains of houses.

92

Lisa C. Nevett

were actually separate properties, making a total of four houses in all (Hoepfner 1999, 253–256). The pastas-like area lying between the open courts and inner rooms may have made the principal quarters—to the northwest of the court in each case—quite dark, if it supported a wall of full-height. Nevertheless, where the exact position of doorways can be reconstructed, in two out of three cases they seem to have been aligned, which would have enabled the maximum amount of light to penetrate. (This is in contrast with many houses in larger urban centers, where doorways seem deliberately to have been off-set, which would have restricted daylight entering the interior, although it would have meant that sight-lines

Fig. 6.4. Ano Voula, the Kalampoka plot.

Between Urban and Rural

93

were blocked and the privacy of individual rooms was thereby increased: Nevett 1999, 73.) No signs of interior decoration are reported, and although Hoepfner reconstructs specialized andrones in three of his four houses (Hoepfner 1999, 253–256), there is no architectural or artefactual evidence cited to support this idea, which is presumably based on an assumption that such rooms must have been present in most houses. The patterns found in the dwellings on these plots are supplemented by more fragmentary information from fully or partially excavated structures elsewhere in the area of the ancient settlement at Ano Voula.6 Even taken together, this evidence is not straightforward to interpret, but there are nevertheless a variety of parallels here for some of the unusual features of the spatial organization at Thorikos, identiWed above. A complex pattern of arrangement of rooms into suites is visible both on the Kalampoka plot and elsewhere, and there is at least one instance where access to the house seems to have been relatively open since there are sometimes several street entrances (Fig. 6.3d). There is also an apparent absence of decorative features such as colonnaded courtyards, mosaic Xoors or andrones, even though the houses are mainly of fourth-century date and are of a reasonable size. The circular towers of the Kalampoka plot parallel the towers identiWed at Thorikos, suggesting that such structures could be an integral element of domestic buildings in the deme centers, even where mining was not a feature of the local economy. Rhamnous Our current state of knowledge makes it difWcult to compare Thorikos and Ano Voula with any other deme center. There are, nevertheless, some hints that similar features may have been more widespread: at Rhamnous an insula which included a circular tower like those at Thorikos and Ano Voula has been found just within the southeast gate of the outer circuit wall of the fortress (Fig. 6.5; Staïs 1891, 14; Pouilloux 1954, 68; Petrakos 1976, 39–41; Petrakos 1999, 83). It is possible that the tower may have been part of a courtyard house, although the role of this part of the site has been debated and the chronology and architectural details are unclear. (Petrakos dates the tower to the Wfth century, although he interprets this and the surrounding buildings as being defensive in character, linked with the nearby south gate of the citadel: Petrakos 1999, 83).

Interpretation and Conclusions: Social Life in the Attic Demes The number of residential buildings known from Attic deme settlements is obviously limited at present, and it is particularly difWcult to establish to what extent the houses currently known are likely to be representative

Fig. 6.5. Rhamnous. a. the phrourion; b. southeast area of the phrourion, detail.

Between Urban and Rural

95

of the majority of deme houses during this period. Our picture will become clearer as more evidence comes to light. Nevertheless, at present I would suggest that despite the specialized role of Thorikos as a mining center, there are similarities between the houses there and at Ano Voula, and that as a group these deme houses contrast signiWcantly with those of larger communities like Athens. This may well imply that there were corresponding differences in the social lives of households in the two types of settlement: if the measures taken at Athens and elsewhere to control access to and movement around the house are indicative of concern over the integrity of the household, then the relative lack of evidence for such control in these deme houses may indicate either a corresponding lack of such concerns or a different method of demonstrating adherence to these ideals. Here it may be relevant to consider the relative size of Athens in comparison with the other Attic settlements: Finley estimates the male citizen population of the city at around 15,000 in the Wfth century (Osborne 1985, 64), whereas Osborne estimates the male citizen population of a deme center at only around 150 individuals (Osborne 1985, 45). Even allowing a huge margin for error in each of these calculations, the magnitude of the likely difference in population size must have had implications for the patterns of social relations in the two types of community. In a small settlement there is likely to have been closer personal contact between individuals, with fewer strangers, and different households more aware of the activities of others. It seems likely that in the deme centers the balance between formal and informal methods of social control was slightly different from that struck in the city itself, with the watchful eye of neighbours and friends being more pervasive and a greater role played by the kinds of peer pressure and gossip, which have already been shown to have played an important role at Athens generally (see, for example, Cohen 1991, 97; Hunter 1994, 96–119; Lewis 1996, 9–13). The absence of an andron—even in the larger properties—suggests that in the deme s the house itself may not generally have served as the location for the kind of formalized entertainment of visitors which seems to have taken place in the andrones of city houses. Such occasions may have been the preserve of those whose social network encompassed the city itself, and perhaps did not extend to the majority of the inhabitants of the deme s, whose social lives centered upon their fellow demesmen. If guests were entertained in these deme houses, it seems that they must have been received in a relatively plain, undecorated space. Perhaps the inhabitants of a smaller community were more closely acquainted, so that there was less of an incentive to try to make an impression on a visitor by creating ostentatious surroundings. The absence of a decorated andron does not, however, necessarily indicate that these deme houses did not serve as status symbols in any respect.

96

Lisa C. Nevett

I would suggest that the towers may be signiWcant here: a massive, multistorey, stone-built tower would have been a very visible feature in the streetscape of one of these settlements. Such a structure may have played a functional role as a secure storage place for agricultural produce or for other goods and possessions (as in the much-cited passage of Demosthenes: 47–56). I would also suggest, however, that it may have had an additional symbolic function, demonstrating that its owner had the wealth to spend on constructing a massive circular building, which although relatively tall and highly visible among the rectilinear forms of the surrounding streets, may not have made for economical use of space. In addition, if—as some of the textual sources suggest—towers were used for storage of produce or goods, possession of one may also have been an explicit statement of (or at least, claim to) personal wealth in terms of resources. Such a situation would parallel the use of towers in residential contexts in other, more recent societies. For example, in Mediaeval Chechnya “a glance at a . . . complex reveals whether the man who lived in it was wealthy, what his trade was, and, sometimes, what his social status was” (“Medieval Stone Towers in Chechnya”). In sum, although current evidence is limited, the organization of space in houses from Attic deme sites seems to support the impression that there were subtle differences between the social lives of households in these communities and those of households in Athens itself. Given the differences in scale between the population of the city and those of the individual extra-urban deme centers, this is perhaps not surprising, but it does offer an indication—however preliminary—of the way in which the broader geographical and social context of an individual household could potentially have had an inXuence on the nature of social relationships, both between individuals and between households. Notes 1. I would like to thank Bradley Ault and John Ellis Jones for their help with references for this chapter, and Todd Whitelaw for preparing preliminary drafts of the Wgures. 2. What follows is a summary of more detailed arguments presented brieXy in Nevett 1995 and in full in Nevett 1999. 3. My discussion here is based on the houses in the so-called industrial quarter, together with house 1 immediately to the west of the theatre (Mussche et al. 1965; Donnay 1967; Mussche et al. 1968; Mussche 1971; Mussche et al. 1990; Spitaels 1978; Mussche 1998). 4. A list of towers at Thorikos is provided by Spitaels (Spitaels 1978, 108–109), and updated by Mussche (1998, 57–58). I have not included the square tower (number 3) here since its shape and position suggest that it is likely to have had a different function from that of the circular ones. 5. A number of other circular towers are also reported from building plots

Between Urban and Rural

97

elsewhere in the area of the ancient settlement (Alexandri 1973–74a; Alexandri 1973–74b; Tsaravopoulos 1990, 74). 6. These include two houses on the Kaloumenos plot in Odos Vithinias (Tsivarakos 1969), a house on the Nenka plot in Odos Dragatsaniou (Liankouras 1973–74), and portions of several structures from the Logothetis Plot in Odos Delphon (Kyriazopoulo 1985): see Figs. 6.3 a–c, e. Literature Cited Alexandri, O. 1973–74a. “Ανω Βολα: δς Αννυµος κα ∆αιδλου.” Archaiologikon Deltion 29, 158. ———. 1973–74b. “Ανω Βολα: δς Βασ. Γεωργου κα Βουλιαγµνης.” Archaiologikon Deltion 29, 60–63. Andreou, I. 1989. “Οδ!ς Πατριρχ!υ Γρηγορου Ε’.” Archaiologikon Deltion 44, 61–62. ———. 1994. “Ο δ'µος των Αιξονδων Αλν,” in Coulson, Palagia et al. 1994, 191–209. Cohen, D. 1991. Law, Sexuality and Society. Cambridge: Cambridge University Press. Conophagus, C. 1980. Le Laurium antique et la technique grecque de la production de l’argent. Athens: Ekdotike Hellados. Coulson, W. D. E., O. Palagia, T. L. Shear, Jr., H. A. Shapiro, and F. J. Frost, eds. 1994. The Archaeology of Athens and Attica Under the Democracy. Oxford: Oxbow. Donnay, G. 1967. Thorikos 2: Le quartier industriel. Brussels: Comité des Fouilles Belges en Grèce. Hoepfner, W. 1999. “Athen und Attika.” In Geschichte des Wohnens, vol. 1, 500 v.Chr.–500 n.Chr. Vorgeschichte—Frühgeschichte—Antike, ed. W. Hoepfner, 223– 260. Stuttgart: Deutsche Verlags-Anstalt (Wüstenrot Stiftung, Deutscher Eigenheimverein e. V., Ludwigsburg). Hunter, V. 1994. Policing Athens: Social Control in the Attic Lawsuits, 420–320 BC. Princeton, N.J.: Princeton University Press. Jones, J. E. 1982. “The Laurion Silver Mines: A Review of Recent Researches and Results.” Greece and Rome 29, 169–183. Kyriazopoulo, A. 1985. “Βολα—Ανω Βολα.” Archaiologikon Deltion 40, 54–62. Lewis, S. 1996. News and Society in the Greek Polis. London: Duckworth. Liankouras, A. 1973–74. “Οδς ∆ραγατσανου.” Archaiologikon Deltion 29, 60–61. Lohmann, H. 1989. “Das Kastro von H. Giorgos (Ereneia).” Marburger Winckelmannprogramm 1989, 33–66. “Mediaeval Stone Towers in Chechnya.” Electronic document, http://www. chechnyafree.ru/index.php?lng=eng§ion=carchitectureeng&row=2, accessed 3rd July, 2004. Morris, I. 1998. “Remaining Invisible: The Archaeology of the Excluded in Classical Athens.” In Women and Slaves in Greco-Roman Culture, ed. S. Murnaghan and S. R. Joshel, 203–220. London: Routledge. Mussche, H. F. 1967. “Le quartier industriel.” In Thorikos 3: 1965, ed. J. Bingen, H. F. Mussche, J. Servais, J. de Geyter, T. Hackens, P. Spitaels, and A. Gautier, 57–71. Brussels: Comité des Fouilles Belges en Grèce. ———. 1971. Thorikos 5: Le quartier industriel. Brussels: Comité des Fouilles Belges en Grèce. ———. 1990. “Insula 3.” In Thorikos 9: 1977/1982, ed. H. F. Mussche, J. Bingen, J. E. Jones, and M. Waelkens 12–62. Ghent: Comité des Fouilles Belges en Grèce.

98

Lisa C. Nevett

———. 1998. Thorikos: A Mining Town in Ancient Attika. Ghent: Comité des Fouilles Belges en Grèce. Mussche, H. F., J. Bingen, J. Servais, R. Paepe, T. Hackens. 1968. Thorikos 1963: rapport préliminaire sur la première campagne de fouilles. Brussels: Comité des Fouilles Belges en Grèce. Mussche, H. F., J. Bingen, J. E. Jones, M. Waelkens. 1990. Thorikos 9: 1977/82. Ghent: Comité des Fouilles Belges en Grèce. Mussche, H. F., J. Servais, J. Bingen, and T. Hackens. 1965. “Thorikos 1963: rapport préliminaire sur la première campagne de fouilles,” L’Antiquité Classique 34, 5–46. Mussche, H. F. et al., 1964–1990. Thorikos. Gent: Comité des Fouilles Belges en Grèce. Nevett, L. C. 1995. “Gender Relations in the Classical Greek Household: The Archaeological Evidence.” Annual of the British School at Athens 90, 363–381. ———. 1999. House and Society in the Ancient Greek World. Cambridge: Cambridge University Press. Osborne, R. 1985. Demos: The Discovery of Classical Attika. Cambridge: Cambridge University Press. Petrakos, B. 1976. “ Α ) νασκαφ+ Ραµνου-ντος.” Praktika 1976, 5–50. ———. 1999. Ραµνους Ι. Τοπογρ αφα. Athens: Archaeological Society. Pouilloux, J. 1954. La Forteresse de Rhamnonte. Paris: Boccard. Rosivach, V. J. 1993. “The Distribution of Population in Attica.” Greek, Roman and Byzantine Studies 34, 391–407. Spitaels, P. 1978. “Insula 3. Tower Compound 1.” In Thorikos 1970/1971, 7. Rapport préliminaire sur les septième et huitième campagnes de fouilles. P. Spitaels, J. Bingen, A. Uyttendaele, F. Blondé, K. van Gelder, A. Cheliotis, and A. Helsen, 39– 110. Brussels: Comité des Fouilles Belges en Grèce. Staïs, B. 1891. “Εν Ραµνου-ντι.” Praktika 1891, 13–18. Stanton, G. R. 1994. “ The Rural Demes and Athenian Politics.” In Coulson, Palagia etal. 1994, 217–224. Travlos, J. 1988. Bildlexikon zur Topographie des Antiken Attika. Tübingen: Wasmuth. Tsaimou, I. 1979. “ .Ο ανδρνας” του- ‹‹πλυτ'ριου Σµου›› στ+ Σορηςα τ+ς Λαυρεοτικ'ς.” Athens Annals of Archaeology 12, 15–23. Tsaravopoulos, A. 1990. “Ανω Βολα.” Archaiologikon Deltion 45, 74. Tsivarakos, E. 1969. “Ανασκαφ+ ε4ς 5Ανω Βολαν.” Archaiologikon Deltion 24, 89–90. Walter-Karydi, E. 1996. “Die Nobilitierung des griechischen Wohnhauses in der Spätklassischen Zeit.” In Basileia: Die Paläste der hellenistischen Könige, ed. W. Hoepfner and G. Brands, 57–61. Mainz: von Zabern. Walter-Karydi, E. 1998. The Greek House: The Rise of Noble Houses in Late Classical Times. Athens: Athens Archaeological Society. Whitehead, D. 1986. The Demes of Attica. Princeton, N.J.: Princeton University Press.

Chapter 7

Houses at Leukas in Acarnania: A Case Study in Ancient Household Organization Manuel Fiedler

Already in the late nineteenth and again in the Wrst half of twentieth century, Th. Wiegand and H. Schrader, followed by D. M. Robinson, recognized the importance of artifact distributions for analyzing domestic space in ancient Greek houses.1 They published ground-plans of houses from Priene and Olynthus, marking Wnd-spots of small Wnds and pottery (Wiegand and Schrader 1904, 325, with Wg. 365; Robinson 1946, pl. 136). Nevertheless, until now very few Greek houses have been published together with their Wnds material.2 And architectural features alone generally do not allow us to clarify the use of every room in a private house.3 As a rule, whether small Wnds and pottery can actually contribute to the analysis of a household’s organization depends primarily upon depositional and post-depositional processes (Schiffer 1987; LaMotta and Schiffer 1999; Sommer 1991). While an ad hoc abandonment in the case of catastrophic circumstances, as, for example, at Pompeii,4 would ideally leave behind relatively clear and complete assemblages of material reXecting household activities, far more difWcult is the interpretation of “normal” abandonment scenarios, when inhabitants quit their homes in an organized and systematically conducted fashion, leaving behind only things of no further value (i.e., garbage).5 An instance where a city was abandoned under “normal” circumstances can be found at Leukas. Here, several private houses have been excavated in the past few years (Dousougli 1993; Fiedler 1999, 2003). The Hellenistic House AII.6, with its well-preserved Wnds, shows how certain developments in Greek society were reXected in its ground plan and through household organization.

The Polis of Leukas Leukas was founded as a Corinthian colony in the seventh century B.C. The settlement was situated on the Acarnanian-Leucadian strait between

100

Manuel Fiedler

the island and the Acarnanian shore, and so lay directly on the essential trade route between Greece and the Adriatic Sea. The size of the town (covering ca. 100 ha), references in the primary sources, and its impressive remains, for example, the longest stone bridge known from ancient Greece (Negris 1904; Fiedler 2003, 22–25) and an impressive harbor mole with extensive defense posts nearby (Dörpfeld, Goessler et al. 1927, 269– 271; Murray 1982, 266, 1988), illustrate that Leukas must have been an important city in northwestern Greece. For several decades during the third and second centuries Leukas gained political signiWcance as the capital of the Acarnanian league, “caput Acarnaniae” as Livy (33.17.1) informs us. After 31 B.C., when the battle of Actium took place northeast of the island, Octavian founded Nikopolis ad Actio as a monument of his victory. Henceforth, speaking of the Greek cities around the Ambracian Gulf, Strabo says, “most of which, or rather all, have become dependencies of Nikopolis” (10.2.2; trans. H. L. Jones, Loeb ed., 1961). But the archaeological evidence indicates that several of these settlements were forsaken altogether, as were Kassope and Ambrakia (Hoepfner and Schwandner 1994, 117; Schwandner 2001, 112; Karatzeni 1999, 243– 244). It appears that the inhabitants of Leukas were also required to leave their homes in order to colonize the newly founded “city of victory” (Fiedler 1999, 2003; Morris 2001, 287).

Street System and Houses Recent excavations have shown that Leukas was laid out on an orthogonal grid, probably already in the sixth century B.C. The insulae were divided into approximately square house plots. These plot sizes inXuenced the ground-plans of private houses over the subsequent centuries. The plan of the Late Archaic or Early Classical House AII.2 (Fig. 7.1) was designed in three parallel, north-south oriented strips in which two or three rooms of different sizes were positioned around a deep courtyard (k/m). Similar arrangements can be found in other Classical and Hellenistic houses at the site, and are seen in the Hellenistic House BII.3 (Fig. 7.2), with its central courtyard (G1) and adjacent rooms that were also located in three parallel, north-south oriented strips. Obviously, those ground-plans which were predetermined by the Archaic house-plots were developed in Late Archaic or Early Classical times and inXuenced the house’s design subsequently into the Classical and Hellenistic periods. Contrary to the Classical “Typenhäuser” (as deWned by Hoepfner and Schwandner 1994), it is not the ad-hoc repetition of ground-plans that was carried out at Leukas, but rather the houses’ layout followed local traditions in domestic architecture as well as the different requirements of the house-owners or occupants.

Houses at Leukas in Acarnania

101

Several houses in Athens and in Corinth show similar ground-plans. A Classical house on the northeastern slopes of the Areopagus (Fig. 7.3) consisted of several rooms which were arranged around a deep courtyard (Shear 1973, 150–151). In the so-called “Terracotta-Factory” of Corinth, a private house with a coroplast’s workshop of the Wfth century (Fig. 7.4), the rooms surrounded a central courtyard (Williams 1981, 418–420). These examples demonstrate that this general house-type was distributed over wide areas of ancient Greece, and was not locally or regionally restricted. Elsewhere in northwestern Greece similar houses are known as well. For example, at Kallipolis (Aetolia) houses with similar ground-plans, dating to the fourth and third centuries B.C., have been excavated (Themelis 1999). The late Classical Herdraumhäuser in Kassope show a related layout (Hoepfner and Schwandner 1994, 146– 152), but with the signiWcant difference that one central room, the oikos, dominated the houses (Fig. 7.5). Functions of rooms are generally difWcult to specify in each of these houses. Courtyards, oikoi, or andrones are often recognized by their characteristic positions or sizes, or by built features contained in them. In Leukas it seems that in the Late Archaic/Early Classical House AII.2 (Fig. 7.1) one of the larger rooms with adjacent chambers could have been used as the oikos (i/j or o), and the room p near the street could have functioned as an andron. The andron of the Hellenistic House BII.3 (Fig. 7.2) can be identiWed by its characteristic plaster Xoor with raised edges for klinai (room G2). But often such identiWcations remain hypothetical if, apart from the architectural evidence, information about the Wnds is lacking.

The Hellenistic House AII.6 at Leukas In another Hellenistic house, located close to the middle of the ancient town, a great deal of artifactual evidence was recovered. The analysis of architecture and built features, as well as study of the distribution of different kinds of artifacts, clariWes how rooms were used and provides information about the household’s structure.6 Ground-plan The house was built upon stone foundations that supported mud-brick walls. It consists of an initial core built at the end of the third or the beginning of the second century (House AII.5: Fig. 7.6), and an extension that was added from the neighboring property probably in the Wrst century (see House AII.6, Fig. 7.7). The building was bounded by a street (stenopos) to the north and an alleyway (ambitus), which served largely for drainage, to the south.

Fig. 7.1. Leukas, House AII.2.

Fig. 7.2. Leukas, House BII.3.

Fig. 7.3. Athens, Classical house on the northeast slopes of the Areopagus (courtesy of the Agora Excavations, American School of Classical Studies at Athens).

Fig. 7.4. Corinth, Classical house, also known as the “Terracotta Factory” (courtesy of the Corinth Excavations, American School of Classical Studies at Athens).

104

Manuel Fiedler

In the earlier Hellenistic phase (Fig. 7.6) the building had an approximately square ground-plan of 13 by 14 m. (covering 182 m²). Its division into three parallel strips, so characteristic for other houses in Leukas, can still be discerned here, but the irregularity of room sizes makes it less in evidence. The house consisted of seven rooms (on the groundXoor) that were oriented west and east of a courtyard. This courtyard was divided into a forecourt (m′) and a rear court (i /′ k′). A street entrance must have existed onto the forecourt (m′) but remains unexcavated. Next to the forecourt m′ lay the largest rooms of the house, o′/p′ in the eastern, and h′ in the western third of the house plot. Both rooms were equipped with large, double-winged doors. In the south wall of m′ a small doorway gave access to a second courtyard, i /′ k′. By passing through this southern courtyard all the other rooms could be reached: one room (n′) to the east, a group of three small chambers ( j ′, f, and g ′) to the west, and a separate small bathroom with bathtub (l ′). South of the house, running parallel to the alleyway, a small drain ( f *) was unearthed which was probably used as a latrine. The house was later enlarged (House AII.6, Fig. 7.7) so that it now measured 15.80 by 14 m (covering 220 m²): three rooms of similar sizes (s, r, and q) were added along the east. Two of them (s and r) were annexes to the large room o′/p ′, and room q could be reached from both rooms r and n′. It is unknown whether a second storey existed, but the ground-plan does show a distinct separation between the different parts of the house. While the rear court i /′ k′ functioned to provide access to the rooms in the south of the house, the forecourt gave access to both of the large rooms in the north, o′/p′ and h′. Between the front and rear halves of the house access was gained only through the small doorway between m′ and i /′ k′, or through the more remote passage from o′/p′, via

Fig. 7.5. Kassope, House 5.

Fig. 7.6. Leukas, House AII.5.

Fig. 7.7. Leukas, House AII.6.

106

Manuel Fiedler

the small chambers r, q, n′ to the courtyard i /′ k′. From this it is clear that the front and rear portions of the house were deliberately separated from one another. In order to explain this in terms of the household’s structure the functions of the individual rooms must now be analyzed. Site Formation and Artifact Deposition During excavation of the house some 600 small Wnds and a large quantity of pottery were documented. The minimum number of pottery vessels (MNV) represented amounts to about 1055. This Wgure was derived largely by counting rim sherds.7 Most of the Wnds and pottery are preserved in fragmentary condition. Very few whole objects survived. Analysis of the stratigraphy suggests that most of the pottery sherds and other Wnds, from both phases of the house, were integrated into earthen Xoor surfacings that were constituted of multiple layers of mud. When vessels and other objects were broken they must have been removed from the house, but some “lost” fragments remained behind and eventually became part of the house Xoors.8 The more completely preserved objects must have been abandoned at the same time as the houses themselves, under non-violent circumstances, at the end of the Wrst century. Considering the processes of artifact deposition and the circumstances of the building’s abandonment, it should be noted that Wnd-spots are not necessarily identical with the location of an object’s utilization.9 But statistical analyses of the artifact distribution by room and comparison of artifact densities across the house, together with consideration of Wxed furnishings and the Wnish of rooms, their size and position, can provide information regarding room functions and the activities that took place within them.

Finds and Functions of Rooms The pottery can be categorized into three main groups: the Wne wares, the plain wares and the cooking wares (Figs. 7.8, 7.9). Vessel shape, in addition to classiWcation by wares, correlates to function. The Wne wares include vessels that were used for serving and consuming food (such as plates and bowls) as well as for drinking (beakers, in particular). Figure 7.9 shows the breakdown of the varieties of vessels present. Plain ware pottery was used for storing and preparing food (for example pithoi, amphorae, lekanai/bowls), and cooking pots and pans (chytrai, lopades) were produced in special heat-resisting cooking wares. Approximately 40 percent of the pottery found in the house is of the Wne ware variety, 37 percent is plain ware, and 23 percent is cooking ware (Fig. 7.8a).

Houses at Leukas in Acarnania

107

Analysis of Rooms Room o′/p′, the largest in the house, was equipped with a wide doorway (to be entered from m′) and a sizable window. This window is attested by a window-pillar, or mullion (Fig. 7.10). The Wnd-spots of its three preserved fragments indicate that the window opened onto the street. The walls of the room were decorated with colored plaster (mostly red, but also yellowish-green and bluish-gray). Its Xoor, as in the case of most other rooms, was made of beaten earth. A hearth was built above the Xoor in the southern part of the room.

a.

Leukas house AII.5&6 pottery, all rooms (MNV 1055)

Fine Wares 40%

b.

pottery oikos o'/p' (MNV 335)

Cooking wares 24%

Cooking wares 23% Fine Wares 51%

Plain wares 25%

Plain wares 37%

c.

pottery kitchen q (MNV 162)

Fine Wares 20%

d.

pottery andron h' (MNV 33) Cooking wares 6%

Cooking wares 37% Fine Wares 55%

Plain wares 39%

Plain wares 43%

Fig. 7.8a. Leukas, percentages of Wne, plain and cooking ware pottery, Houses AII.5 and AII.6, based on MNV counts. Fig. 7.8b. Leukas, percentages of Wne, plain, and cooking ware pottery, Houses AII.5 and AII.6, room o′/p′. Fig. 7.8c. Leukas, percentages of Wne, plain, and cooking ware pottery, Houses AII.5 and AII.6, room q. Fig. 7.8d. Leukas, percentages of Wne, plain, and cooking ware pottery, House AII.5 and AII.6, room h.

House AII.5&6: 1055 vessels (Minimum number of vessels – MNV) Fine wares MNV % Plain Wares MNV Drinking: Storing: Beakers 147 13,9 Amphorae 46 Eating: Amphorae’slids lids 28 Dishes 175 16,6 Flagons/Bottles 126 Serving: Jars 15 Plates 15 1,4 Pithoi 6 Large Bowls 2 0,2 Preparing: Small bowls 37 3,5 Mortaria 3 Flagons 14 1,3 Bowls 135 Other forms: Other forms: Pyxides 7 0,7 Stands 16 Lekythoi 6 0,6 Louteria 2 Unguentaria 2 0,2 Unguentaria 6 Miniature vessels 8 0,8 Miniature vessels 4 others 6 0,6 others 4 > 419 39,7 > 391

%

0,3 12,8

Cooking-Wares Cooking: Pots Pans Lids Other forms: Stands Cooking bowls Cooking flagons Pompeian red plates

1,5 0,2 0,6 0,4 0,4 37,6

>

4,4 2,7 11,9 1,4 0,6

MNV

%

110 96 23

10,4 9,1 2,2

5 7 3 1

0,5 0,7 0,3 0,1

245

23,2

Fig. 7.9. Leukas, Houses AII.5 and AII.6 with percentages by function and shape, based on MNV counts.

Houses at Leukas in Acarnania

109

The wide variety of artifacts recovered there indicates very different activities taking place in room o′/p′. The preparation of food is attested by completely preserved millstones (for grinding Xour), a Wnger-pestle and two mortaria. Fragments of some 80 cooking vessels—and in addition a completely preserved bronze cooking pot—were also found in o′/p′. This is the largest concentration of cooking vessels in the house and accounts for 24 percent of all the pottery (MNV 335) recovered from the room (Fig. 7.8b).

Fig. 7.10. Leukas, window pillar from House AII.6, room o′/p′ (H. 0.558 m).

110

Manuel Fiedler

Various storage vessels like amphorae, jars, or plain ware pots indicate that food and drink were probably stored in room o′/p′ as well. The percentage of Wne wares from room o′/p′, at approximately 51 percent of the overall ceramic assemblage, is higher than in other rooms in the house. This should mean that the beakers, plates and other table vessels were used here in the context of dining. The production of textiles is attested by a signiWcant number of loomweights, spindle whorls, and bronze needles. Other objects like Wshhooks, iron chisels, an axe, and Xint artifacts were found here, so that some tools which were used outside the home were obviously stored in room o′/p′. In conclusion, the largest room in the house was equipped with a hearth, had walls of painted plaster, was accessed through a wide doorway, and bore a large window facing onto the street. It was probably used as a dining room, and the household prepared and stored food, produced textiles, and stored a variety of tools here. It can therefore be recognized as the oikos, a room central to the social life of the household. Room s was a small chamber next to the oikos o′/p′, from where it could be entered by a small doorway. It was the only room in the house with a water resistant mortar Xoor, and its walls were protected by water resistant colorless plaster. A small drain led to a main canal outside the house. Although room s yielded very little in the way of artifactual material (only 15 MNV), it probably served as a bathroom. Its bathtub or washbasin (louterion) must have been removed during the house’s abandonment. Room r was the other chamber adjacent to and accessible from room o′/p′. It also provided a route through to the next rooms to the south. Although a quantity of pottery and small Wnds, including lamps, was found here (accounting for 112 MNV), no particular function for room r can be speciWed. Most probably, items like pottery and tools were stored here. Room q was the kitchen. A hearth was built in a corner and its earthen Xoor was mixed with ashes. Flint chips recovered here (as well as in the oikos) could well have been residual by-products of setting the hearth Wres. Pottery was recovered here with a high frequency, accounting for some 162 MNV. While the percentage of cooking vessels is the highest of the house (at approximately 37 percent of the overall assemblage from the room), conWrming that room q was used as a kitchen, Wne wares were relatively rare, comprising only 20 percent of the room’s assemblage. (See Fig. 7.8c, in comparison to the assemblages from the oikos room o′/p′, Fig. 7.8b). Also frequent were plain ware bowls that could have been used for preparing food. A great many plain ware jars, Xasks, and amphorae were found in the kitchen, room q, where they functioned as containers for food and liquids. Since a pithos was found in room n′, this space appears to have been designated as a storeroom. It was positioned toward the back of the house

Houses at Leukas in Acarnania

111

to be entered from the kitchen (room q) and through a small door from the most remote corner of courtyard i /′ k′. The percentages of plain and cooking ware MNV’s are high (from a total of 132 MNV, 46 percent and approximately 22 percent respectively; with numerous jars, amphorae, and bowls represented). Some of the vessels are quite well preserved. Courtyard i /′ k′, in the southern half of the house, had an irregular ground plan. The southeastern area was wider than the northwestern, perhaps to accommodate a stairway (no traces of which are actually preserved) to an upper Xoor. In the northeastern corner existed what can only be described as a kind of stone bench. A well was located opposite, in a niche of the northwestern corner of i /′ k′. This was a perfect location, in the middle of the house, surrounded by shading walls that protected the well-water against the sun and algae growth. Relatively little pottery was found in i /′ k′, a scenario comparable to all other spaces in the western half of the house, testifying that activities utilizing pottery took place largely in rooms of the building’s eastern half. In i /′ k′ only some 76 MNV can be accounted for altogether. Fragments of plain ware jars, bowls and amphorae attest that the courtyard was utilized for everyday tasks. Cooking wares occurred at the very low rate of 15 MNV, while Wne ware assumed a normal percentage, dominated by beakers. Other table vessels were few in number. Obviously, the courtyard was normally not used for cooking activities or dining. That textile production took place in the courtyard is shown by the spindle whorls that were found here in the highest concentration of the whole house, as well as by bronze needles and loomweights that were recovered. A few small fragments of millstones were also scattered in this area. Some of the activities that took place in the oikos could therefore be transferred to courtyard i /′ k′, too, perhaps during more agreeable weather or seasons. Room l ′′ was a small annex which was used as a bathroom in the earlier Hellenistic house (AII.5) and was altered by blocking up the doorway and removing the bathtub. A new entrance was made next to the well. The water-resistant plaster Xoor remained, but a basin connected to the main drain was installed now in the southern half of the room. Obviously, this chamber was adapted for tasks in which fresh water from the well was utilized, and which could be efWciently drained away. Room l ′′ could therefore have functioned as a kind of washroom, perhaps for clothes and/or dishes. No special function could be determined for the chamber j ′ which gave access to two rooms behind, rooms f and g ′′. Few sherds were found here and these account for only about 40 MNV, Wne ware vessels being dominate (at approximately 54 percent), while cooking and plain ware vessels were relatively few. The small chamber f in the southwestern corner of the house led to

112

Manuel Fiedler

the next room, g ′′, and probably to the latrine f * behind this room. The nature of the pottery recovered in room f is different from the adjacent rooms. Again, few sherds were found, providing a MNV of approximately 70. These were dominated by plain ware vessels like jars, amphorae and other more generic pots (making up ca. 59 percent of the assemblage). And while the percentage of cooking vessels is very low (approximately 11 percent), Wne ware is present in a relatively normal proportion (some 30 percent). All things considered, room f seems to have been used primarily for storage. Behind the house-wall, south of room f in area f*, a small drain was built which Xows off to the parallel main canal (ambitus). This annex f* may have served as a latrine, to be entered from room f. Room g ′′ is rectangular in shape and approximately 10m² in area. It was reached by passing through the two anterooms j ′ and f. Surrounded by adjacent rooms it was clearly a dark space that was closed off from natural light and fresh air. Nevertheless, room g′′ was an important part of the house as is seen by its Wnish and the artifact assemblage recovered there. It was one of only three rooms with painted plaster walls, from which red, yellowish, black, and speckled grayish plaster-fragments are preserved. Several fragments of bronze tools (a spatula, a stylus, a medical instrument known as a kauterion) as well as of bronze vessels were found in this room. The percentage of Wne ware pottery is high (accounting for approximately 50 percent of a small assemblage of only some 45 MNV), but beakers and dishes are unusually rare in contrast with other table vessels such as small bowls and jars that dominate. The percentage of plain ware vessels is also high (approximately 44 percent), but cooking vessels are infrequent (representing only some 3 MNV, or approximately 7 percent, of the assemblage). It seems that valuable vessels were kept in this room. In general, the function of this space is difWcult to assess on the basis of the Wnds. It could have been used as a kind of “living-room” considering the painted plaster walls, but because of its remote position, its small size and darkness, it is unlikely to have been functioned as an oikos or andron. The valuable Wnds (compared to those recovered from other rooms) remind one of the description of an Athenian household by Xenophon (Oikonomikos 9.3), who mentions that valuable goods should be securely stored in the house’s principal sleeping-room (thalamos). A function as a kind of “living room” or thalamos should be considered. Room h′ was positioned near the street entrance to the house. It was Wnished with red and white painted plaster walls and an off-center, doublewinged door equipped with a stone threshold. Since the room’s function was altered after abandonment of the rest of the building,10 its original purpose cannot be speciWed on the basis of the artifacts recovered there, although the ceramic assemblage does not show any strong evidence for

Houses at Leukas in Acarnania

113

cooking or food preparation (see Fig. 7.8d; where the relatively large plain ware assemblage is based on numerous amphorae and jar fragments from later the post-abandonment phase). But its position next to the street as well as the details of its Wnish are enough to prove that room h ′ was laid out as an andron. Area m′ represents the space onto which the entrance to the house from the street must once have opened (located along its unexcavated northern portion). It can therefore be identiWed as a kind of an entry vestibule.11 The wide doorways, which seem to be designed as such for display purposes, led to the two adjacent rooms (o′/p′ and h′) could indicate that m′ was not roofed but designed as a forecourt to i /′ k′. As in the neighboring room h′, no Wnds connected with this phase are preserved. Only an iron spade, a garden tool which must normally have been used outside the house, was found in the southeastern corner, next to the entrance to the oikos room o′/p′.

Conclusion Inhabited from end of the third or the beginning of the second century to the late Wrst century, this house consisted Wrst of seven rooms on the ground Xoor arranged around two courtyards (House AII.5), then subsequently of ten rooms (House AII.6). Three of these rooms, rooms o′/p′, h′, and g′′, were Wnished with painted plaster walls (Fig. 7.11). All rooms with the exception of the small chambers s and l ′′ bore earthen Xoors. The plan shows a separation of rooms and courtyards between the front and the back parts of the house (Fig. 7.11). The largest room was positioned in the front area. In addition to its painted plastered walls it was equipped with a wide door and an ornamented window. Inside the room was the only Wreplace of the house (apart from that in the kitchen). Therefore, room o′/p′ must have been the main living room or oikos, the center of the household’s social life. The pottery and other artifacts recovered hint at a wide variety of activities taking place here including the preparation, cooking and consumption of food. Textiles were produced and goods were stored here as well. Next to the oikos several smaller rooms with particular functions were arranged: the bathroom (s), and, accessible through chamber r (where a variety of different kinds of pottery was found), the kitchen (q) and a store-room (n′). Such a combination, a large oikos with adjacent small chambers, was very common in Greek houses since Classical times. It is one of the characteristic features of Classical houses at Eretria (Reber 1998, 154–165; the so-called “Dreiraumgruppe” in the tradition of Archaic “Dreiraumhäuser”). The arrangement is also well attested in Piraeus (Hoepfner and Schwandner 1994, 31–42), at Olynthus (Hoepfner and Schwandner

114

Manuel Fiedler

1994, 82–86, 100–102; Cahill 2002, 153–161, where it is termed the “kitchen-complex”), as well as in the late Classical “Herdraumhäuser” of Kassope (Hoepfner and Schwandner 1994, 146–150). That this group of rooms was still common in the Hellenistic period is in evidence in the houses of New Halos (Haagsma 2003), in the “Herdraumhäuser” of Orraon (Hoepfner et al. 1999, 395–410), at Komboti (Hoepfner and Schwandner 1994, 320, with Wg. 305), and in houses of Kassope’s later phases (Hoepfner and Schwandner 1994, 157, with Wg. 151) which were abandoned at the same time as Leukas’ House AII.6. At New Halos, Orraon, and Komboti, the functions of the smaller chambers are well attested by the archaeological evidence: they were used as bathrooms, weaving-rooms, store-rooms, or kitchens.12 When the Hellenistic house in Leukas was enlarged, in phase AII.6, rooms serving exactly these same functions were added to the oikos o′/p′: the bathrooms s, and the kitchen q, which gave access now to the store-room n′. Obviously, with the enlargement from phase AII.5 to AII.6 the inhabitants recreated a very traditional “core” long manifest in Greek houses.

Fig. 7.11. Leukas, House AII.6 with room identiWcations.

Houses at Leukas in Acarnania

115

Opposite the oikos, the andron (h′) lay in the front part of the house. It was decorated with painted plaster walls and equipped with a wide, double-winged door. Positioned next to the entrance, the andron was isolated from all other rooms. Since Classical times this was the preferred location for such a room. Here were conducted the symposia which included guests from outside the household.13 The small room g′′, entered through two antechambers ( j ′ and f ), was the third of the house’s rooms with painted plaster walls. Additionally, small Wnds and pottery recovered here were of higher quality than in most of the other rooms. In contrast to the oikos, this room was located in the most remote part of the house, obviously as a kind of private “livingroom,” perhaps a thalamos. The courtyard was separated into a forecourt (m′) and a rear court (i /′ k′). The doorway from m′ to i /′ k′ was narrow in comparison with the entrances from m’ to the oikos (o′/p′) and andron (h′). Obviously, visitors who entered the forecourt m′ were directed to the andron with its wide door, or to the oikos, but were kept away from the more remote areas of the house. The rear court and its surrounding rooms were screened off from the forecourt with a dividing wall. In the back of the house access to rooms was strictly controlled. The door from the rear court to the store-room (n′) in which the household provisions were kept, lay in the furthest part of the courtyard, while the latrine ( f *) and room g ′′ were accessed only via small anterooms. The andron (h′) and the oikos (o′/p′), where most of the social activities of the household took place, were designed as relatively open, semi-public spaces, whereas the more private aspects of domestic life was sheltered from the outside world and conWned to the rear portion of the house. This division into semi-public and more strictly private spheres already existed in Late Classical houses, for example, at Eretria (Reber 1998) and Maroneia (Lavas and Karadedos 1991), but with the difference that the andron (or, in the case of Eretria, multiple andrones) was positioned in the representative portion of the house, while the oikos was located in its more private reaches. It is remarkable in House AII.6 at Leukas that the oikos was located just inside the entry from the street. Its wide door was situated within the forecourt (and not the rear court, which would have been equally possible) and the large window faced onto the street (and not onto one of the courtyards). This should mean that the activities which took place in this room were to some extent open to the outside world. With its andron and oikos, House AII.6 at Leukas demonstrates the persistence of traditional elements of Greek private architecture which were already vanishing in other parts of the Hellenistic world, as on Delos (Trümper 1998; Hoepfner 1999, 514–522) or at Morgantina (Tsakirgis 1984, 387–392). At these sites, new representative elements like oeci maiores

116

Manuel Fiedler

and exedrae were introduced. Representative life (including symposia) as well as private life took place in these new rooms which dominated the houses that possessed them. This continuity of tradition in domestic architecture is found also in the late houses at Kassope, Orraon, and Komboti, all of which were abandoned at the same time as House AII.6 at Leukas, and in the earlier Hellenistic houses of New Halos. But although House AII.6 utilizes certain traditional elements, social change is evident, especially in the opening up of main areas of the house to the outside world.

Notes 1. I am very grateful to Dr. A. Dousougli and Dr. K. Zachos (12th Ephorate of Prehistoric and Classical Antiquities, Ioannina) for their generous permission to study the site, and to the German Archaeological Institute for Wnancial support. I would also like to thank B. Ault and L. Nevett for their invitation to participate in the colloquium that led to my contribution to this volume. For a more detailed analysis of the houses at Leukas, see Fiedler 2003. 2. Ault and Nevett 1999; Ault 1994, 10–17; Ault in press, chap. 1. Haagsma 2003 also takes account of artifact distribution in interpreting room function in the houses at Nea Halos, but the site itself was subject to only shallow deposition as well as extensive disturbance of its stratigraphy by modern plowing. 3. See Hoepfner and Schwandner 1994 and Hoepfner 1999, which frequently employ hypothetical attributions to room functions. 4. But see Allison 1992 and Siggis 1999, both of whom show that even in Pompeii Wnds’ distribution does not fully represent the ancient reality of daily life. 5. Therefore, Wnds from settlements which were abandoned under nonviolent circumstances are preserved largely by “negative selection” (see Eggers 1986, 266–268). 6. For a more detailed analysis and further consideration of methodological issues, see Fiedler 2003. 7. The total number of pottery sherds recovered from the house is, of course, much higher. 8. Fiedler 2003, 50–53. For similar evidence from Halieis, see Ault 1994, 74–78; Ault and Nevett 1999, 49–50; and Ault in press, chap. 1. And in Roman houses, see Thüry 2001, 28–30. 9. See Schiffer 1987, Sommer 1991, and LaMotta and Schiffer 1999, for methodological problems. 10. Room h′ continued to be used as a storeroom after the abandonment of the rest of the house. For details, see Fiedler 2003. 11. No other entry to the house was revealed in the course of excavation. While the northern extent of room h′ also lay along the unexcavated portion of the house, it cannot contain an external doorway because its doorway with area m′ was intended to be closed from inside of room h′, as observable from cuttings in the stone threshold here. Thus, there is no other possible location for an external entry than along the northern wall of area m′. 12. Bathroom: Orraon (House 1), Komboti. Weaving-room: Orraon (House 1). Storeroom: Orraon (House 1), Komboti, New Halos (House of Agathon, House of the Snake, House of the Ptolemaic Coins; more probably a storeroom

Houses at Leukas in Acarnania

117

than an andron, considering the kind of bench there, which is unusual for andrones). Kitchen: Komboti, New Halos (House of the Coroplast). For Orraon, see Hoepfner et al. 1999, 395–410; for New Halos, Haagsma 2003; for Komboti, Hoepfner and Schwandner 1994, 320; and W. Hoepfner and E.-L. Schwandner, pers. comm. 13. For andrones at Olynthus, see Hoepfner and Schwandner 1994, 98–99, and Cahill 2002, 180–190, with pl. 1; at Halieis, Ault 1994, 234–237, and Ault, in press, chap. 3; at Kassope, Hoepfner and Schwandner 1994, 146–152; and at Orraon, Hoepfner et al. 1999, 404–405. Literature Cited Allison, P. M. 1992. “The Distribution of Pompeian House Contents and its SigniWcance.” Ph.D. dissertation: University of Sydney. ———. ed., 1999. The Archaeology of Household Activities. London: Routledge. Ault, B. A. 1994. “Classical Houses and Households: An Architectural and Artifactual Case Study from Halieis, Greece.” Ph.D. dissertation, Indiana University. ———. In press. The Houses: The Organization and Use of Domestic Space. Excavations at Ancient Halieis 2. Bloomington: Indiana University Press. Ault, B. A. and L. C. Nevett. 1999. “Digging Houses: Archaeologies of Classical Greek and Hellenistic Domestic Assemblages.” In Allison 1999, 43–56. Cahill, N. 2002. Household and City Organization at Olynthus. New Haven, Conn.: Yale University Press. Dörpfeld, W., P. Goessler, et al. 1927. Alt-Ithaka. Ein Beitrag zur Homer-Frage. Studien und Ausgrabungen auf der Insel Leukas-Ithaka. Munich: GräfelWng. Dousougli, A. 1993. “Νοµ
s Λεuxδαs.” Archaiologikon Deltion 48, 293–300. Eggers, H. J., 1986. Einführung in die Vorgeschichte. 3rd ed. München-Zürich: Piper. Fiedler, M. 1999. “Leukas. Wohn- und Alltagskultur in einer nordwestgriechischen Stadt.” In Hoepfner, 1999, 412–426. ———. 2003. “Antike Häuser in Leukas: Wohnhausarchitektur und Fundmaterial aus einer nordwestgriechischen Stadt des 6. bis 1. Jh.v.Chr.” Ph.D. dissertation, Freie Universität Berlin. Haagsma, M. J. 2003. “The Houses in New Halos.” In Housing in New Halos: A Hellenistic Town in Thessaly, ed. H. R. Reinders and W. Prummel, 37–79. Lisse: A. A. Balkema. Hoepfner, W., ed. 1999. Geschichte des Wohnens. Band I: 500 v.Chr. – 500 n.Chr. Vorgeschichte—Frühgeschichte—Antike. Stuttgart: Deutsche Verlags-Anstalt (Wüstenrot Stiftung, Deutscher Eigenheimverein e. V., Ludwigsburg). Hoepfner, W., S. Dakaris, K. Gravani, and E.-L. Schwandner. 1999. “Orraon: Eine geplante Kleinstadt in Epirus.” In Hoepfner 1999, 384–411. Hoepfner, W., and E.-L. Schwandner. 1994. Haus und Stadt im klassischen Griechenland. Rev. ed. Wohnen in der klassischen Polis 1. Munich: Deutscher Kunstverlag. Karatzeni, V., 1999. “Ambracia During the Roman Era.” In L’Illyrie méridionale et l’Épire dans l’Antiquité III, ed. P. Cabanes, Actes du IIIe colloque international de Chantille, 1996. 241–247. Paris: de Boccard LaMotta V. M. and M. B. Schiffer. 1999. “Formation Processes of House Floor Assemblages.” In Allison 1999, 20–29. Lavas, G. and G. Karadedos. 1991. “Mauerwerk, Bodenbeläge und Anstrichtechnik eines spätklassischen Hauses in Maroneia, Thrazien.” In Bautechnike der Antike: Diskussionen zur Archäologischen Bauforschung, vol. 5, ed. A. Hoffmann, et al. 140–147. Mainz: P. von Zabern.

118

Manuel Fiedler

Morris, S. P. 2001. “The Towers of Ancient Leukas.” Hesperia 70, 285–347. Murray, W. M. 1982. “The Coastal Sites of Western Akarnania: A TopographicalHistorical Survey.” Ph.D. dissertation, University of Pennsylvania. Murray, W. M. 1988. “The Ancient Harbour Mole at Leukas, Greece.” In Archaeology of Coastal Changes. Proceedings of the First International Symposium, “Cities on the Sea: Past and Present,” Haifa, Israel, September 22–29, 1986, ed. A. Raban, 101–116. BAR International Series 404. Oxford: British Archaeological Reports. Negris, P. 1904. “ Vestiges antiques submerges.” Athener Mitteilungen 29, 354–360. Nevett, L. C. 1999. House and Society in the Ancient Greek World. Cambridge: Cambridge University Press. Reber, K. 1998. Die klassischen und hellenistischen Wohnhäuser im Westquartier: Eretria. Ausgrabungen und Forschungen 10. Lausanne: Editions Payot. Robinson, D. M. 1946. Excavations at Olynthus, vol. 12. Domestic and Public Architecture. Baltimore: Johns Hopkins University Press. Schiffer, M. B. 1987. Formation Processes of the Archaeological Record. Albuquerque: University of New Mexico Press. Schwandner, E.-L. 2001. “Kassope, the City in Whose Territory Nikopolis Was Founded.” In Foundation and Destruction: Nikopolis and Northwestern Greece, ed. J. Isager, 109–115. Athens: Danish Institute at Athens. Shear, T. L. 1973. “The Athenian Agora: Excavations of 1971.” Hesperia 42, 121– 179. Siggis, B. 1999. “Vita cognita; Die Ausstattung pompejanischer Wohnhäuser mit Gefäßen und Geräten—untersucht an ausgewählten Beispielen.” Ph.D. dissertation, Universität Köln. Sommer, U. 1991. Zur Entstehung archäologischer Fundvergesellschaftungen: Versuch einer archäologischen Taphonomie. Studien zur Siedlungsarchäologie 1. Bonn: Habelt Universitätsforschungen zur Prähistorischen Archäologie, Bd. 6. Themelis, P. 1999. “Ausgrabungen in Kallipolis.” In Hoepfner 1999, 427–440. Thüry, G. E. 2001. Müll und Marmorsäulen: Siedlungshygiene in der römischen Antike. Mainz: P. von Zabern. Trümper, M. 1998. Wohnen in Delos: Eine baugeschichtliche Untersuchung zum Wandel der Wohnkultur in hellenistischer Zeit. Internationale Archäologie 46. Rahden: Verlag Marie Leidorf GmbH. Tsakirgis, B. 1984. “The Domestic Architecture of Morgantina in the Hellenistic and Roman Periods.” Ph.D. dissertation, Princeton University. Wiegand, Th. and H. Schrader. 1904. Priene: Ergebnisse der Ausgrabungen und Untersuchungen in den Jahren, 1895–1898. Berlin: G. Reimer. Williams, C. K., II. 1981. “The City of Corinth and Its Domestic Religion.” Hesperia 50, 408–421.

Chapter 8

Modest Housing in Late Hellenistic Delos Monika Trümper

The domestic architecture on the Cycladic island of Delos is often cited as an example of the increased luxury in Late Hellenistic housing compared with its Classical predecessors (Nevett 1999, 164–166). Indeed, evidence found in some of the houses supports the thesis that lifestyles became more lavish in the Hellenistic period. Characteristic of these houses are elaborate architectural features including peristyles, statuary, and rich decoration of walls and Xoors with painted frescoes and tessellated mosaic pavements. However, an extensive overview of all living units revealed in the French excavation since 1873 offers a more differentiated picture: the spectrum of living units ranging from simple lodgings above shops to extended lavish peristyle-houses is much broader in the cosmopolitan trade port of Delos than in “Hippodamian” Classical cities with their standardized plots and houses. Within this spectrum the majority of the Delian living units are not only smaller in size but also less well equipped than the average Classical house in cities like Olynthus, Priene, Piraeus, and Miletus. This can best be demonstrated by comparing Classical houses with the prevailing house-type in Delos which comprises a central courtyard with service room(s), a vestibule opening off to the front, and a group of two or three “living” rooms to the rear (the Delian “normal-house”; Trümper 1998, 107–108). In these examples, which have an average ground Xoor area of 120 m2, no space was reserved for an andron or another room for the reception of outsiders as in Classical houses. The fact that Delian domestic architecture offers quite a number of living units ranging even below the small “normalhouse” is little known because these have rarely attracted archaeologists’ attention. Whereas the different house types, even modest examples, have been thoroughly studied (Chamonard 1922/24; Trümper 1998), a vast number of unidentiWable, unappealing structures (multifunctional complexes with shops, workshops, magazines, humble dwellings, etc.) remain unpublished or largely unexcavated, probably because they failed to yield rich Wnds or reWned architecture.1

120

Monika Trümper

This chapter does not claim to offer a complete picture of modest housing in Delos; rather, it intends simply to call attention to the existence of humble dwellings and to provide some initial remarks on their interesting features. These preliminary results are based on my own research in Delos and have proWted much from references to other sites and studies. Among the latter the Vesuvian cities have to be mentioned Wrst because their well-preserved domestic architecture includes many modest dwellings that are the focus of several recent studies (WallaceHadrill 1994, 91–117; Pirson 1999) In what follows, the “types” of modest living units in Delos will be presented brieXy before discussion of their architectural and urban context. My concluding observations will address the socio-historical aspects of these buildings, above all whether the different types of modest housing in Delos can be associated with speciWc social strata, and assess whether the social structure of Late Hellenistic Delos is reXected in its domestic architecture.

The Range and Character of Modest Living Units The kinds of structures that could serve as temporary or even permanent lodgings at the bottom level of the vast range of residential space vary, depending on cultural habits and social norms. Since these norms are difWcult to grasp, especially for the lowest social strata, the identiWcation of habitable structures apart from clearly deWned houses is difWcult and open to speculation. Epigraphy, literary sources, and legal texts help to a certain extent by providing some information on habitable units. Thus, it is known that Roman tabernae not only functioned as shops, workshops, or taverns, but also offered simple accommodation, for example, on the pergula or mezzanine level (Pirson 1999, 19–20).2 For Delos, inscriptions offer detailed information on the rented property of the sanctuary of Apollo: between 314 and 167 the temple let separate workshops, integrated into buildings or forming independent complexes. Some of these workshops possessed upper storeys that could either be rented together with the ground Xoor rooms or else let separately from them. Whereas the inscriptions in some cases deliberately use different terms for shops, workshops, and living units, thus suggesting a clear functional differentiation between working and living, other rented units seem to have been conceived as multifunctional, serving both needs (Hellmann 1992, 48–51, 99–100, 138–140, 423–426). These various sources, together with the extensive archaeological record from Pompeii, provide a Wrm basis for the identiWcation of modest living space in Delos. As at Pompeii, the taberna Wgures as one of the key elements. It is certainly not surprising to Wnd commercial buildings, and among

Modest Housing in Late Hellenistic Delos

121

them tabernae, in a Xourishing trade port, but it is remarkable that the extraordinary Wgure of over 500 tabernae in Delos has, up to now, never been examined exhaustively (Chamonard 1922/24, 207–215; cf. Trümper 2004). The Delian tabernae can be identiWed by separate, often quite narrow, entrances from the street. As in the Vesuvian cities, the taberna as an architectural type was multifunctional, covering a wide range of uses from shop to workshop, a production center for food, a storeroom or simple tavern, all the way through to humble accommodation. In theory, every taberna, even a single-room one, might have served as a living and sleeping place for the tabernarius, the more so if it comprised a mezzanine and/or an upstairs room. Since only a few walls of tabernae (especially in the Quartier du théâtre) have been preserved to a sufWcient height to show holes for beams, it can only be assumed (but not proved) that upper Xoors are part of the standard design of these complexes. To F. Pirson the unrestricted residential function of Pompeian tabernae

Wall 'Late' wall (TH III Z)

II A

θ' θ

B ζ

α nA

α'/β'

Rue

du s

Marble

tade

Gneiss

Sa

rap

β

Window

η η'

ieio

γ

Blocked door

δ

Sta

Opus tessellatum

de

a.Quartier de l'Inopos: Mixed complex α−γ a:

Terracotta Sewer

b.Quartier du stade: Îlot II, mixed complex ζ-θ b: g

d

IE

6 f

e

Z b

c

ε'

a

ε

ε'' ε'''

4

IC

0

Rue 5

c: c.Quartier du théâtre: Normal-house III Z with row of 5 tabernae

Rue

Stad

du s

tade

e

0

d: d.Quartier du stade: Îlot I, mixed complex ε−ε'''

Fig. 8.1. Delos. a. Quartier de l’Inopos: mixed-complex α–γ; b. Quartier du stade: Ílot II, mixed-complex ζ–θ; c. Quartier du théâtre: normal house III Z with row of 5 tabernae ; d. Quartier du stade: Ílot I, mixed-complex ε–ε′′′.

10 m

122

Monika Trümper

only seems likely in those cases which included at least one back- or sideroom in addition to the shop itself (Pirson 1999, 53–55). According to these criteria not even a quarter of all Delian tabernae could have served as humble dwellings. However, some of the single-room tabernae were certainly conceived as living units as proven by special, double thresholds: these consist of one doorway leading into the ground-Xoor room and a smaller one which gave access to a separate staircase up to the upper storey (Trümper 1998, Wgs. 87–88). Theoretically, these upstairs apartments could have extended over neighboring rooms or houses, thus forming much more extensive dwellings than indicated by the corresponding ground-Xoor size. But this assumption is, at least in some cases, contradicted by the archaeological record. In the Quartier du théâtre a row of Wve tabernae opening onto the much-frequented Rue 5 was built together with the normal-house III Z (Fig. 8.1c).3 Four of these shops—and probably all of them—were provided with double thresholds. Since two tabernae (2, 4) back onto the courtyard of the house their upper Xoors cannot have been larger than the ground Xoor quarters below, and therefore any possible extension to the northeast was excluded. It is hard to imagine that the elaborate double thresholds were intended merely for storerooms in the upper Xoor. In this particular case, therefore, the reconstruction of single-room upstairs apartments seems to be the only plausible alternative. The simple wooden ladders or steep staircases that have to be reconstructed within the majority of doublestorey tabernae (solid stone bases or staircases being scarce in tabernae), do not allow any conclusions concerning the character and use of the mezzanine or upper Xoor. Regardless of accessibility, the accommodation in or above shops can be taken as the simplest form of living unit recognizable in the archaeological record of Delos. Although the necessary water-supply and even facilities for waste water disposal can be found on the ground and upper Xoors in some tabernae, most inhabitants of these humble dwellings must have contented themselves with the few public wells and cisterns hitherto identiWed, or must have made an agreement with some better-equipped neighbor. The group of units which offers a slightly higher standard of living, and which lies between the tabernae with one or two rooms and modest normal-houses, is quite heterogeneous, and the boundaries between the categories are far from clear-cut. The main feature common to this intermediate group is a strong commercial orientation, indicated by the obligatory incorporation of tabernae, which were integrated into the layout of the units or even connected with them through doors. In contrast with the simple tabernae mentioned above, examples of this group also include other elements such as a courtyard, water supply, and rooms in addition to the shops that could have been used as living rooms.4

Modest Housing in Late Hellenistic Delos

123

Although in their standard layout these complexes share some similarities with houses, they lack their characteristic form, structure, and suites of rooms. Some examples may illustrate the wide range of variation within this group: • Complex ζ–θ in Insula II of the Quartier du stade (104 m2; Fig. 8.1b) consists of a courtyard (θ) with large cistern, a room (θ′) opening off the courtyard, and two shops (η, ζ), one (η) with a staircase entered separately from the Rue du stade (Plassart 1916, 232–234). Originally the extent of this complex was slightly smaller in the south, although it had the same number of rooms. The courtyard was, in this phase, connected with one of the tabernae (η) and opened with a door onto the plot to the east, which was still un-built (Trümper 1998, 222–223, Wg. 24). As a result of several transformations the complex was, in its last phase, divided into separate units: a unit opening to the south with a courtyard and a room (θ, θ′), two interconnected shops (η, ζ), and an independent upstairs apartment. This apartment had not formed part of the original plan and extended over the shops and the Rue du stade, with columns supporting the projecting section. Despite the clear division into three independent units, two of which offered more or less humble accommodation, the courtyard-complex might still have provided the necessary water supply for the other two. • Complex ε–ε′′′ in the same quarter (64 m2; Fig. 8.1d) comprises a suite of three interconnected rooms, of which only the northern room (ε) is accessible from the Rue du stade (Plassart 1916, 174–175). This room, which includes a small latrine (ε′), might have served as a shop, or as the presence of a drain suggests, as a courtyard or a simple covered entrance-room. The stone steps between rooms ε′′ and ε′′′ form the solid base of a staircase continued with wooden stairs, which led to upstairs rooms.5 The original complex was smaller, facing a public street with a sewer to the north. It had two or three rooms with an entrance door (later blocked) in room ε′′ which probably served as a taberna. The unusual position of the staircase, just to the right of this Wrst entrance door, might be due to the fact that it initially served an independent upper apartment. This earlier, mixed-complex, with restricted space for living, was enlarged at the expense of the northern street and was transformed into a unit probably used exclusively for living (depending upon the function attributed to room ε). The remodeled unit was fairly well equipped, with a latrine established over the public sewer, at least two living rooms on the ground Xoor, and further rooms in the upper storey. The latter rooms must now have

124

Monika Trümper

belonged to the corresponding ground Xoor, since the staircase is situated between the ground-Xoor living rooms. In both phases water must have been drawn from public wells or cisterns (not yet found in this quarter) or from the facilities of neighbors. • Such arrangements for water collection were not necessary in complex III H in the Quartier du théâtre (155 m2; Fig. 8.2a) which includes a courtyard (b) with large cistern, a suite comprising a large room (c) and an annex-room (d), two shops (21, 25) and a vestibule (a) leading to one of the main shopping streets in Delos, the Rue du théâtre (Chamonard 1922/24, 39). The different transformations are difWcult to interpret: Originally, there were probably only a large regular room c/d and taberna 25, which extended further west. When the annex-room d was created, taberna 25 was reduced and later its connecting door to room d was blocked (Trümper 2002, 156, n. 47). The unusual size of the cistern seems to exceed the needs of this rather

théâtre

Wall Blocked door Marble

A

Gneiss

G

Marble chips

F Opus tessellatum

I

Sewer

E

N a:a.Quartier du théâtre: Mixed complexes III H, Y Mixed/taberna-complex 27/29 Rue supérieu

re du théâtr

e

Rue 4

0

C 2

b: b.Quartier du théâtre: Mixed complex IV C a-d/6 Mixed/taberna-complex IV C 7-13

Fig. 8.2. Delos, Quartier du théâtre. a. mixed-complexes III, H, Y, mixed taberna-complex 27/29; b. mixed-complex IV c a–d/6; taberna-complex IV c 7–13.

10 m

Modest Housing in Late Hellenistic Delos

•

•

•

•

125

modest complex and suggests that it was used as a common water supply for several units nearby. Complex III Y in the same quarter (185 m2; Fig. 8.2a) is fairly well equipped, with two shops (31, 33), a vestibule on the Rue du théâtre (Ruelle β), two suites each comprising a larger room with an annex (b, c and d, e), and a small central room (a) which might have served as a courtyard with drainage, or at least as a light source (Chamonard 1922/24, 50). Of the two shops opening off the central space (a), one was subsequently connected with a mixed-complex to the south (which included TH III 27/29). The unusual plan of this unit might be due to its insertion into an open plot between two existing buildings, but its peculiar bipartition could also have been aimed, right from the beginning, at creating two separate units, each with a shop and a group of living rooms (33 b, c – 31 d, e). In addition, both would have proWted from some common space (Ruelle β, “courtyard” a). Complex IV C in the Quartier du théâtre (59 m2; Fig. 8.2b) has an unusual shape and features due to several complicated transformations (Chamonard 1922/24, 52). It still comprises a courtyard (a) with cistern, a living room (d) of astonishingly small depth compared to its width, a small taberna (b) with drainage (latrine?), and a taberna (6) on its rear side.6 Originally taberna 7 formed part of this unit, and room d was divided into two rooms, so that the complex must have had a fairly regular shape. Nevertheless it cannot have had more than three, or at most four, small rooms and was certainly already surrounded by tabernae in the north and east (numbers 6, 15, 13, 11). The complex adjacent to the former—TH IV C 7, 11, 13, 15 (47 m2; Fig. 8.2b)—possesses a suite of three interconnected tabernae (7, 11, 13) and one independent taberna (15) (Chamonard 1922/24, 52). Because of its composition it has, at Wrst sight, to be excluded from this intermediate-size, mixed-use category. But taberna 11 is paved with rough slabs of gneiss and equipped with drainage, an altar, and a staircase placed just to the left of the south entrance door. This might have functioned as common, arterial space or a light source (open or only partly roofed) for the other two tabernae and for an independent upstairs apartment above tabernae 13 and 15. The original plan provided a row of three fairly regular shops (11, 13, 15) to which taberna 7 was added in a later phase at the expense of its western neighbor (TH IV C a–d, see above). Simultaneously, taberna 11 might have been remodelled in order to provide some modest residential facilities. A similar development can be reconstructed in the case of complex III 27/29 (121 m2; Fig. 8.2a) in the same quarter (Chamonard 1906, 587– 588). Both the open space between shops 27a and 29 and the existing houses (TH III H and I) were successively built and remodeled. The room

126

Monika Trümper

behind taberna 27a (27b) contained a mill and a well, and opens onto the paved Ruelle α, which is equipped with a sewer. The denomination Ruelle is misleading because this is actually an ordinary vestibule between shops 27a and 29, which broadens between rooms 27b and 31b to form a narrow courtyard. It ends in a corridor leading to room 31c, which had been created at the expense of the famous Maison de Cléopâtre (TH III I) and had originally been divided into two rooms of unequal size. It is uncertain whether these two rooms were used for accommodation, and if so, by whom. They may have served the tenants of interconnected rooms 27a/b or those of rooms 31a/b (since room 31b possesses a door to the end-corridor of Ruelle α). Alternatively they may have served as independent shops or as storerooms for both complexes (III 27a/b and III Y). This complex clearly demonstrates the problems and limits of interpretation. • Complex α–γ (98 m2; Fig. 8.1a) occupies the southeastern edge of the Insula of the Maison à une seule colonne, in the Quartier de l’Inopos. It consists of two tabernae (α, β) of irregular shape, opening off one of the main streets of this quarter, and a rectangular backroom (γ) of unusual size, which has not been fully excavated. Since the partition wall of the shops does not run through the whole depth of the rooms to the south wall of room γ, a common space (α′/β′) precedes the back room, which is consequently accessible from both tabernae. It remains an open question what function is to be attributed to this peculiar back room, since it belongs to an early phase of the insula, while the Maison à une seule colonne and the tabernae (α, β, δ) were added subsequently (Trümper 1998, 244–246, with Wg. 61). Back-room γ might have been used as a common storeroom or production center, as a common living space for two parties, or else as a living room for one party who managed two shops. Complete excavation of this enigmatic room might solve these questions, but until then, the complex has to be rated as a marginal candidate that could also be counted among the group of simple tabernae with one or two rooms. This short overview of eight complexes could be completed by describing the Wve further examples included in Table 1. This table lists the most important features of the complexes but cannot provide detailed information on their history, which in most cases is quite complicated. Within this heterogeneous type, two groups of unequal size can roughly be discerned. Eleven examples, by far the majority of the thirteen complexes, comprise both shops and separate living rooms, most of which were also provided with a courtyard. Here these are called “mixed-complexes,” alluding to their double function as commercial or productive working space, and as living space.7 In the remaining two units, an additional

Table 8.1. Delos, mixed-complexes and mixed/taberna -complexes

Building

Size including tabernae in m2

Size Number Number of excluding of tabernae tabernae tabernae connected in m2 in total with complex

“Type” of complex

Number of rooms (without Courtshops) yard

Ilot des bronzes V

52

—

1?

1

mixed complex

2

x

Quartier du stade ε-complex

64

54

1?

1?

mixed complex

2–3

x?

Quartier du stade ζ–θ complex

104

48

2

mixed complex

1

x

Îlot de la maison à une seule colonne α–γ complex

98

TH II C o-t TH III H

60

2

2

111

47

1–2

?

155

130

2

TH III U

150

106

1

TH III X

175

103

TH III Y (incl. 31b)

185

90 without 31b)

TH III V

208

mixed complex?

1

Water supply

Latrine

x x

Staircase

Upper floor separately accessible

x?

x?

x? x

x (2)

x

Date

after 167 B.C.

Plassart 1916, 165s

after 167 B.C.

Plassart 1916, 232ss

after 167 B.C.

unpublished; plan: Chamonard 1922/24, pl. XXVIII

after 167 B.C.?

mixed complex

2–3

x?

mixed complex

2

x

x

1

mixed complex

2–3

x

x

2

1

mixed complex

2–3

x

Chamonard 1922/24, 49

?

2

2

mixed complex

4

x?

Chamonard 1922/24, 50

?

102

4 at least

?

mixed complex

2–3?

x

Chamonard 1922/24, 49

?

x x

x?

x

TH IV C a-d/6

59

36

2?

1?

mixed complex

1–2

x

x

TH III 27/29

121

84

2

1

mixed/taberna complex

2?

x?

x

TH IV C 7-13

47

—

2–3

2?

mixed/taberna complex

x?

x?

Publication

Siebert 2001, 76s

x?

x

x

Chamonard 1922/24, 32

?

Chamonard 1922/24, 39

?

Chamonard 1922/24, 48s

?

Chamonard 1922/24, 52

?

Chamonard 1906, 587s

?

Chamonard 1922/24, 53

?

128

Monika Trümper

living room cannot be identiWed on the ground Xoor, but both include elements which distinguish them from ordinary tabernae, for example common arterial spaces (“courtyards” or light sources), a water supply, altars, or access to independent upper-storey apartments. Their categorization as “mixed/taberna-complexes” reXects their ambiguous character and their strong similarities with either mixed-complexes, or with pure taberna-complexes.8 In both groups signs of luxury such as lavish decoration of walls and Xoors, or the presence of statuary, are very scanty. Given the small number of living rooms in these complexes it is not possible to discuss methods for exploring the differential use of living space. Even if some examples do include characteristic room-suites with a large room (“oecus maior”) and an annex-room, offering some potential for differentiation, most rooms must have been multifunctional and had to meet all the needs of their inhabitants. Even if the tabernae in these complexes are excluded, the ground Xoor area is in many cases much greater than might have been expected. Indeed, the sizes range from 36 m2 to 130 m2 (see Table 1). Thus, some mixed-complexes offer considerably more space for living than the normalhouses with standard features, the smallest of which cover an area of only 65 m2 (Trümper 1998, 166–169). The group of mixed- and mixed/taberna-complexes listed in Table 1 could be considerably enlarged by enumerating all the buildings and insulae which probably comprised similar complexes, but which have not been sufWciently excavated or published and thus cannot be assessed fully.9 Consequently, any conclusions and statistics must be preliminary. It can only be surmised that these small units played a much more important role than has been assigned to them in the literature. Many modest living units—mixed-complexes as well as single tabernae— were equipped with upper storeys that either supplemented the ground Xoor or were used independently. The separate use of upstairs rooms or apartments is conWrmed by the evidence of staircases which either had their own entrance from the street or were situated right next to or just inside the entrance door of a shop, thus being easily accessible without intruding into the ground-Xoor unit (Trümper 1998, 94–98).10 Although the upper Xoors of mixed-complexes or even single tabernae offered accommodation equally modest to that on the ground Xoor, independent upper apartments cannot, as a rule, be included with these more modest dwellings. On the contrary, according to the Wnds, most upper storeys of houses were decorated much more lavishly than the rooms below. Therefore, their status must have been equal if not superior to the ground Xoor rooms, even if they lacked their own water supply (Trümper 1998, 90–106).

Modest Housing in Late Hellenistic Delos

129

Architectural and Urban Context of Modest Living Units So far, only the architectural data (size, layout, and installations) of various modest living units have been presented. But since none of them form a freestanding, autonomous block, their immediate architectural context, together with the overall urban context, have to be discussed. With which types of buildings were they grouped or combined? What kind of relationship existed between them and their neighbors? Did they, for example, function as satellite units of large rich houses as is often observed in Pompeii (Wallace-Hadrill 1994; Pirson 1999)? Were they concentrated in certain quarters or streets? Apart from the obvious location of tabernae on busy shopping streets, public places, and the harbor-front, it is highly revealing to study their immediate surroundings. Of the 507 tabernae hitherto identiWed in Delos, the contexts of 24 percent cannot be evaluated because they were not sufWciently recovered, 21 percent are combined with mixed-complexes and pure taberna-complexes, 21 percent with public and sacred buildings (stoai, agorai, sanctuaries), only 12 percent with houses, 9 percent with clubhouses (such as the Établissement des Poseidoniastes de Bérytos), 8 percent with large commercial establishments (such as the Magasin des colonnes and the Magasin à la baignoire), and only 5 percent are entirely independent. If the tabernae with uncertain contexts are excluded, of the remaining 387 examples, 27.5 percent belong to mixed-complexes and pure taberna-complexes, 27.5 percent to public and sacred buildings, 15.5 percent to houses, 12 percent to clubhouses, 11 percent to large commercial establishments, and 6.5 percent to no building at all (Trümper 2004, 134, graphic 2). Among those buildings which were domestic in nature (that is, houses and mixed complexes), tabernae are predominantly linked to the lower levels of housing. Whereas the existence of tabernae is taken as characteristic of mixed-complexes, not more than a third of all houses include shops. Among the tabernae integrated into domestic buildings, 53 percent belong to houses of the lowest category including, above all, normalhouses. A further 25 percent are assigned to those of the middle category, and only 22 percent are associated with elite housing, mainly comprising extensive, lavish peristyle-houses. Even if these numbers convey interesting information concerning the distribution of Delian tabernae, an overall evaluation of their signiWcance is only possible when comparing them to the archaeological records of other sites. The best reference for such comparative studies is, as usual, provided by the well-preserved and well-documented Vesuvian cities. Detailed statistics for rented tabernae demonstrate that in Pompeii only 21 percent of the tabernae were integrated into public buildings and

130

Monika Trümper

10 percent into pure taberna-complexes, but by far the majority, 69 percent, were linked to houses.11 Among the tabernae assigned to domestic units, 56 percent co-occur with the highest status domus structures, 11 percent with those of middle status, and a mere 2 percent with dwellings of the lowest category (Pirson 1999, 137–141). The comparison between the location of Pompeian and Delian tabernae is instructive on many points. In Pompeii tabernae were predominantly integrated into big houses, often grouped in long impressive lines along the front on both sides of the main entrance; they Wgured simultaneously as a source and symbol of wealth, and as a sign of social status and power; consequently, they had an entirely positive connotation. By way of contrast, the owners of large, rich Delian houses were obviously neither interested in nor dependent upon the income of tabernae, nor were they compelled or attracted to establish them as a symbol of social status. On the contrary, tabernae were clearly linked to lower status housing and thus most probably to the lower social strata. Series of tabernae form a typical feature of normalhouses where they do not Xank the front entrance, since the houses themselves usually open onto quiet side streets and are only of modest width. Instead, the tabernae lie along the sides of houses adjoining major streets (Figs. 8.3–8.5). The characteristic distribution of Delian tabernae can only partly be explained by the location of domestic units within the town. Within all quarters, houses of different sizes are consistently interspersed with mixedcomplexes and tabernae. There is no differentiation into “rich” and “poor,” into noble or lower class quarters. Furthermore, in all quarters large houses on important roads show blank façades, either planned right from the beginning or as result of subsequent integration of shops. Thus, the reasons for the preferred combination of tabernae with lower level accommodation are, at least partly, to be sought in speciWc economic and social conditions differing considerably from the Pompeian context. Because of their integrated tabernae, mixed-complexes predominantly Xank busy streets and public places. These complexes have been established in all quarters of Delos, above all in those with a high commercial proWle like the Quartier du théâtre and the Quartier de Skardhana. They were not laid out as part of large-scale building programs comprising clear-cut blocks with several units, but result from single measures performed by individual owners. Therefore it is not surprising that mixedcomplexes occupy plots of small size and/or unfavorable shape that were not appropriate for standard house-types. While some of these plots are clearly to be interpreted as undesirable leftover terrain, others have a surprisingly regular shape and must have been deWned as part of an ofWcial process of allotment. Accordingly, irregular complexes like TH III Y and X (Figs. 8.2a, 8.3) were inserted between existing buildings, exploiting

Modest Housing in Late Hellenistic Delos

131

all space available. In contrast, unit TH III H (Figs. 8.2a, 8.3) was obviously erected on a regular plot, together with a row of shops on the Rue du théâtre, and the adjoining habitation III G. All of these were added to existing houses in the south and the west and clearly deWned by the course of the Rue du théâtre. The multiple-complex TH IV C (Figs. 8.2b, 8.3) occupies the eastern end of the smallest insula in the Quartier du théâtre. Narrowing considerably from west to east, its eastern portion formed an ordinary plot, but could hardly have been Wlled with a normal courtyard-house. The peculiar complex TH III U (Fig. 8.3) goes back to an early phase in the development of the Quartier du théâtre and was erected as a freestanding independent block with its entrances to the eastern Rue 6, which originally could have served as one of the main streets. The complex ζ–θ in the Quartier du stade (Figs. 8.1b, 8.5) was built at the same time as house II A to the north, but as an individual unit, its trapezoidal plot being determined by the course of the adjacent streets. The nearby complex ε–ε′′′ (Figs. 8.1d, 8.5) with its regular row of rooms was also planned together with its neighbor (house I E);

Maison aux stucs

'Hôtellerie'

113

Quartier de l'Aphrodision

Rue 4

Rue 2

Îlot II

Îlot VI Rue

érieure Rue sup

âtre

du thé

Îlot IV

du théâtre

C C

Rue

4

Rue 2

supérieure

118

Îlot VIII

Rue

Rue 3

120

Îlot XIII

du théâtre

du théâtre

Ru

e1

2

Ru

ed

u th

éâ tre

U

27/29

121

X

V

G

A

Rue 2

e6

Ru

Îlot I Rue

45

Îlot X

117 Quartier du théâtre

H

Y

e5 Ru

16 12 8

122

e?

,Q

ua

α

Pl ag

e?

6 34

2

122

,G

r èv

e1 1

e? Ru

i?,

Ru

Z

10 Ru

e6

Îlot III

Ru e6

Îlot XIIb

0

117 etc.

β

122

Îlot XIIa

γ 4 1

Îlot XI

GD-Number taberna House Mixed complex Magasin/commercial building (122) Meeting place/building of association or group (113)

Fig. 8.3. Delos, Quartier du théâtre: distribution and context of tabernae and mixed-complexes.

132

Monika Trümper

however, due to the sloping terrain it is situated on a considerably higher level than the adjacent house and was not connected with it. This list of the individual developmental sequences of mixed-complexes could be continued, adding more variations to what is already a fairly broad spectrum. But it should by now have become clear that the modest mixed-complexes cannot generally be interpreted as leftover plots, which form the last steps in the development of urban space. On the contrary, in some of the insulae visible today it is clear that they were deliberately chosen as a viable alternative among the different house-types, and they might therefore provide evidence of a correspondingly differentiated society.

Social Strata and the Organization of Households in Modest Living Units Wealthy individuals are certainly not to be expected as tenants in the modest living units on Delos. However, it is debatable which social strata and what kinds of households were accommodated in these humble 72 68

67

64 b

64

67 b

62

62 b

56

52

63

57 b 55 b P

61 57

54

J

59 A 59 B b

55

M

52 etc.

59 B 58 59 C 59 C b

GD-Number taberna House Mixed complex

Meeting place/building of association or group 'Public' Building

59 D

Fig. 8.4. Delos, Quartier de Skardhana: distribution and context of tabernae and mixed-complexes.

Modest Housing in Late Hellenistic Delos

133

dwellings, and on what terms. Was it really members of the lowest social orders who inhabited the tabernae and mixed-complexes, and what sorts of individuals belonged to these groups? What social spectrum and structure can be reconstructed for Delian society? The Wrst step is to distinguish clearly between the society of independent Delos (314–167) and the cosmopolitan population of the free port after 167, because the structure and composition of the two differed considerably (Vial 1984; Roussel 1916/1987). In order to fully interpret the modest living units they have to be dated within one of these periods. This is a difWcult, and in many cases even impossible, task since reliable data are mostly missing. While all the examples outside the Quartier du théâtre can with more or less certainty be assigned to the period after 167, the chronology of this particular quarter is far from certain. Even if the following reXections are limited to the period of the free port in order to simplify matters, no clear picture emerges: despite the excellent archaeological record, sufWciently detailed information on Delian society is lacking and many points remain controversial. For example, it is still disputed how long the Athenian cleruchy, established after

80

76 etc.

GD-Number taberna House Mixed complex

Îlot II

Meeting place/building of association or group Public (agonistic) building

79

Îlot I

ζ−θ complex

78

ε−ε''' complex 77 76

Fig. 8.5. Delos, Quartier du stade: distribution and context of tabernae and mixed-complexes.

134

Monika Trümper

167, really functioned and performed. In the last quarter of the second century it might have been replaced by a multinational assembly, which reXected the social structure of the cosmopolitan population including Athenians, Romans, Italians, and other foreigners. But the different sources provide no details about the size and importance of the different ethnic groups, the social structure, and the organization of society. Although families of some importance and the formation of local polisélites of an old type are still attested epigraphically, it is far from certain that the family was the only dominant focus of the social structure in the milieu of the trade port. In addition, voluntary associations composed according to ethnicity or profession played an increasingly important role. Against this background any attempt to identify the social level of the owners of speciWc house-types is condemned to failure. Indeed, recent efforts to ascribe certain houses to individuals known by name and, above all, to clear social strata, have not been successful (Rauh 1993, 213–233; Boussac and Moretti 1995; Trümper 1998, 208–210, 212–214, 218–220, 234–241). N. K. Rauh (1993) has suggested that the servile strata consisting of slaves and liberti predominated in the trade port’s permanent population; they would have inhabited houses of all categories, either as owners or as housekeepers for a wealthy patronus residing only temporarily in Delos. If this were the case, an association of slaves could have Wgured as owners and residents of a lavish, high-status peristyle-house, the Maison de l’Hermès. Following Rauh’s argument, who, then, should be imagined as the tenants of all of the modest living units? Is it conceivable that there was a broad enough range within the one and only servile stratum to match the wide spectrum of living units, from pergulae to peristyle-houses? Since no clear answers can be given to these questions, one would rather, for the time being, stick to the simple assumption that modest living-units were constructed and inhabited by the less welloff, while the large, lavish peristyle-houses were in the possession of rich owners—regardless of their ethnicity, origin, and social status. It is equally problematic to choose the opposite approach, trying to infer from the architecture how a complex was actually inhabited or used. However, a careful analysis of the architecture with reference to better-documented examples might yield some instructive conclusions. For the Vesuvian cities recent studies have aimed at correlating the social structure of Roman society with the archaeological record, above all employing insulae with a wide range of units differing in character and size, such as the Insula Arriana Polliana (VI 6). Wallace-Hadrill has suggested that such insulae, as “big houses,” offered a “potential Xexibility of use, for a large extended family, for an ostentatious slave household, for a workforce (slave and free), or for letting” (Wallace-Hadrill 1994, 110).

Modest Housing in Late Hellenistic Delos

135

Whereas the patronus who Wgured as owner of the whole insula resided in the large central atrium-peristyle house with members of his family, the surrounding units (tabernae, small dwellings, upstairs apartments) could either have been given to some of his dependents (slaves, liberti, etc.) or let to strangers. In examining and interpreting the archaeological data, Pirson (1999, 165–173) claimed to distinguish strictly between those units integrated into the main house by connecting doors and those that were entirely separate and autonomous. The latter could have been rented to an outsider, while the integrated units were presumably assigned to a member of the extended family, or to some other dependent of the patronus. These assumptions are based on a correlation between particular social groups and different types of residential units, which, for Roman society, is fairly well supported by the literary sources and epigraphy. In Delos, similar principles of common ownership and the deliberate integration of a diverse population characterized by clear hierarchical distinctions might be inferred from the architecture by recognizing instances of the simultaneous construction (and/or remodeling) of several units which belong to visibly different types but are nevertheless connected by doors. Extensive building programs encompassing several complexes have been identiWed in Delos, but have rarely produced such combinations of varied structures. This can be demonstrated by the following examples. • The most widespread modular design consists of a block with two normal-houses of fairly equal size; this could alternatively be combined with a row of shops adjacent to the long side of one of the houses (Trümper 1998, Wgs. 45, 47, 50). It is possible to identify instances of both fully integrated and completely autonomous tabernae, for example, TH III A (Fig. 8.3) and TH III Z (Figs. 8.1c and 8.3) respectively. Whereas the latter could have been rented to any person, the integrated examples, following Pirson (1999, 165–173), are more likely to have been managed by the tenants of the house to which they belonged. Assuming that a “traditional” nuclear family inhabited this small normal-house, it is still debatable whether slaves were present in such a household (cf. Nevett 1999, 13), let alone served as possible tabernarii. The owner of the house might equally well have used the tabernae himself to earn a living. Thus, working and living in this household would have been organized in more or less the same way as in the mixed-complexes, distinctions being conWned to differing standards of living. • In one instance, a form of row-house insula (Îlot des bronzes) with six normal-houses was built in two phases. Since these houses are of fairly

136

Monika Trümper

equal size and present a relatively standardized layout they were presumably designed for households of the same social stratum (Trümper 1998, 180–190, Wgs. 7–9; Siebert 2001, 55–84, plan III—with incorrect reconstruction of the history of this insula; Trümper 2001). The independent mixed-complex V, in the southwest corner of this block, did not form part of the original layout but results from a subsequent reorganization of the associated terrace that originally comprised one normalhouse. By extending this terrace to the north its owner gained enough space to establish a new normal-house and the mixed-complex, which he might either have let proWtably or sold. • One single example, the Îlot de la maison des masques, fulWlls the condition set out above and was completed in one phase with a variety of different living units which are densely interwoven. Four single tabernae, a medium peristyle-house, and two small, well-equipped, but irregular courtyard-houses, are grouped like satellites around a lavish peristyle-house in the center (Trümper 1998, 248–253, Wgs. 37–38). The rooms of the latter are arranged in such a favorable manner around the peristyle-courtyard that unattractive ill-lit corner rooms do not exist. While the resulting shape is completely irregular, this is not perceptible from the inside, and hence has no negative effect on the peristyle-house. It does, however, on the two small ones. Since remodeling took place across the entire insula, probably simultaneously, it is tempting to ascribe such modiWcations to the initiative of a single individual who owned the entire insula. This individual could have occupied the large peristyle-house, allotting the other units to some of his dependents or else letting them proWtably to strangers. The second alternative is conceivable because all the ground Xoor dwellings as well as two additional upstairs apartments, unknown in their extent, were completely independent; however, no unit possessed its own well or cistern but had to use a common water supply consisting of a large cistern in the southwest corner of the insula. Whereas the inhabitants of the two peristyle-houses had direct access to this cistern, the tenants of the smaller units could only approach it by a separate corridor in the southeast corner of the insula, and thus, had considerably further to carry their supply. To sum up, in general the “big house” certainly did not predominate in Delos, but instead a pattern of multiple independent units of differing size is characteristic of the city’s domestic architecture and residential quarters. Like their Pompeian counterparts, the owners of high-status houses in Delos might still have disposed of an impressive entourage of slaves; equally, they will probably have engaged in the most proWtable business on Delos, the wholesale trade of luxury goods and slaves. Their

Modest Housing in Late Hellenistic Delos

137

social status and Wnancial power were certainly reXected in their large and lavish houses. However, the domestic architecture suggests an absence of social pressure for the upper classes to visibly surround themselves and their residences with varied groups of dependents. The mixed-complexes that have been the focus of this chapter were generally constructed separately. An independent party of modest Wnancial means might very well have bought a plot and then built and used such a complex. Since the concept of mixed-complexes includes a combination of modest living facilities with commercial space, it is likely that both were used by the same individuals. This conclusion seems inevitable for examples where living rooms and shops are interconnected by doors. Even independent shops might have been managed by the inhabitants of an associated house, rather than let to strangers. Because of the integration of tabernae with residential structures, locations on busy streets or areas were preferred. Although the corresponding plots were often of an unfavorable shape, and therefore not appropriate for standard housetypes of regularized plan, with the exception of some obviously marginal plots, the majority of mixed-complexes are astonishingly regular and even quite large. These complexes are to be found in all quarters of Delos where they mix indiscriminately with living units of all types. As in Pompeii, tenants of tabernae, and thus of mixed-complexes as well, will not have belonged to the upper class. At the same time they will probably not have constituted the bottom level of Delian society either. Inhabitants of mixed-complexes were certainly better off than those who lived in or above a shop with only one or two rooms on the ground Xoor and/or upstairs. Even these living units, modest but still independent, might not have accommodated the poorest (slaves), who lived instead within the conWnes of large houses and households. Since such a structure for Delian society can only be assumed, but not proven, tabernae and mixed-complexes represent the lowest form of living that can so far be detected within the archaeological record. Thus, since residing in such a unit meant living at the lower end of a wide range of habitations, it is highly probable that this also coincided with living on a lower rung of Delian society. Notes I would like to thank B. Ault and L. Nevett for giving me the opportunity to publish my studies on modest housing in Delos here, although I did not participate in the original colloquium that spawned this volume. Furthermore, I am greatly indebted to the director of the École Française d’Athènes, R. Etienne, and the Epimeletos in Delos, P. Chatzidaki; without their liberality and support the necessary research in Delos could never have been effected. 1. Even the most recent excavations in the Quartier de Skardhana have

138

Monika Trümper

concentrated on houses and left unWnished several tabernae and mixedcomplexes (Bruneau et al. 1970, 9–10; Siebert 2001, 103–105). (For “mixed complexes,” see below.) 2. To simplify matters, the term “shop” in this paper is used as a synonym for taberna, although the modern word does not imply the broad spectrum of functions of the Latin term. 3. For the units of the Quartier du théâtre the following abbreviations are used: TH III Z = Quartier du théâtre, Insula III, house Z, or TH III 27/29: Quartier du théâtre, Insula III, complex with entrances 27/29. 4. Living rooms are deWned here in a broad sense as those accommodating all household activities (including those which are usually pursued in service rooms or even courtyards), as opposed to the commercial functions of tabernae. 5. According to Plassart (1916, 175), the steps formed only a short staircase connecting the two rooms on different levels, but the position of the staircase, its solid construction and the existence of another connecting door between those rooms exclude such an interpretation. 6. The Xoors of room d and taberna 6 are situated 1.70 m below the level of the courtyard TH IV C a, which is not enough for a real cellar or basement. Room d was thus probably completely Wlled with earth up to the level of the courtyard; it could not have been extended over taberna 6 unless the latter had only been used as a storeroom with a low ceiling. 7. Working and living were not strictly separated and pursued in clearly segregated spaces as in the postindustrial western world; thus, “work” refers here to commercial activities and covers, for example, the production and sale of goods, or the retail of food and drink. Production exceeding the requirements of a household and testifying to commercial interests can only rarely be identiWed in Delian houses. Evidence for such activities seems to be mostly from a “late” phase of use (Roman) and equally late remodeling. Nevertheless, commercial activities could well have taken place in houses, but must have been limited to the negotiation or conclusion of contracts; these have not left traces except in one instance, the Maison des sceaux, where some 16,000 seals from a private business archive have been found (Siebert 2001, 85–98; cf. Trümper in press). 8. For the taberna-complexes which form autonomous freestanding blocks (insulae) and consist only or mainly of tabernae, see Trümper 2004, 135–136. 9. Utilizing the system followed in Bruneau and Ducat (1983), where GD # is short for Guide de Délos Monument #; for example, GD 57: the eastern extension of the Établissement des Poseidoniastes which has several mixed/tabernacomplexes; GD 59B: an insula north of the Îlot de la maison des comédiens which has separate taberna- and mixed-complexes or simple courtyard-houses; GD 59C: an insula north of the Îlot des bronzes with tabernae and probable mixed-complex(es); GD 122: complex ε with one normal-house, one simple courtyard-house, several mixed- or taberna-complexes; GD 85: Quartier du théâtre Îlot XIV with many tabernae, and two simple courtyard-houses or mixedcomplexes (Trümper 2002, 189, Wg. 10; Trümper 2004, 152–154, Table 1). 10. Staircases to separate upper Xoor units could also be completely external, that is, constructed against the façades of houses; but none of these can be found with tabernae or mixed-complexes (Trümper 1998, 92– 94). External staircases in Pompeii form separate corridors incorporated into insulae, which are inserted between independent units (Pirson 1999, 210–41). This kind of external staircase is rare in Delos; so far only one example is known, in the Quartier du théâtre Insula XIV (Trümper 1998, 92, note 462, pl. 90.3). 11. Of the 577 Pompeian tabernae (Gassner 1986, 83), 270 examples (more than half) could have been rented (Pirson 1999, 137). Detailed statistics are only

Modest Housing in Late Hellenistic Delos

139

provided by Pirson (1999, 137–141), while Gassner (1986, 74) simply states that the majority of tabernae were situated in or among houses.

Literature Cited Boussac, M.-Fr. and J.-Ch. Moretti. 1995. Review of Rauh 1993. Topoi 5(2), 561–572. Bruneau, Ph. et al. 1970. L’îlot de la maison des comédiens. Exploration archéologique de Délos 27. Paris: Éditions E. de Boccard. Bruneau, Ph., and J. Ducat. 1983. Guide de Délos. Paris: Éditions E. de Boccard. Chamonard, J. 1906. “Fouilles de Délos: Fouilles dans le Quartier du théâtre.” Bulletin de Correspondance Hellénique 30, 485–631. Chamonard, J. 1922/24. Le Quartier du théâtre: étude sur l’habitation délienne à l’époque hellénistique. Exploration archéologique de Délos 8. Paris: Éditions E. de Boccard. Gassner, V. 1986. Die KauXäden in Pompeii. Vienna: Verband der wissenschaftlichen Gesellschaften Österreichs. Hellmann, M.-Chr. 1992. Recherches sur le vocabulaire de l’architecture grecque d’après les inscriptions de Délos. Bibliothèque des écoles Françaises d’Athènes et de Rome 278. Paris: Éditions E. de Boccard. Nevett, L. 1999. House and Society in the Ancient Greek World. Cambridge: Cambridge University Press. Pirson, F. 1999. Mietwohnungen in Pompeji und Herkulaneum. Untersuchungen zur Architektur, zum Wohnen und zur Sozial- und Wirtschaftsgeschichte der Vesuvstädte. Studien zur antiken Stadt 5. Munich: Verlag Dr. Friedrich Pfeil. Plassart, A. 1916. “Fouilles de Délos: Quartier d’habitations privées à l’est du stade.” Bulletin de Correspondance Hellénique 40, 145–256. Rauh, N. 1993. The Sacred Bonds of Commerce: Religion, Economy, and Trade Society at Hellenistic Roman Delos. Amsterdam: Gieben. Roussel, P. 1916/1987. Délos, colonie athénienne. Paris: Éditions E. de Boccard. Siebert, G. 2001. L’îlot des bijoux, l’îlot des bronzes, la maison des sceaux. Exploration archéologique de Délos 38. Paris: Éditions E. de Boccard. Trümper, M. 1998. Wohnen in Delos: Eine baugeschichtliche Untersuchung zum Wandel der Wohnkultur in hellenistischer Zeit. Internationale Archäologie 46. Rahden: Verlag Marie Leidorf GmbH. ———. 2001. Review of Siebert 2001. Topoi 11 (2), 793–808. ———. 2002. “Das Quartier du théâtre in Delos. Planung, Entwicklung und Parzellierung eines ‘gewachsenen’ Stadtviertels hellenistischer Zeit.” Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung 117, 133–202. ———. 2004. “Wohnen und Arbeiten im hellenistischen Handelshafen Delos. Kontexte und Verteilung der tabernae.” In Wohnformen und Lebenswelten im interkulturellen Vergleich, ed. M. Droste and A. Hoffmann 125–159. Frankfurt: Peter Lang Verlag. ———. In press. “Die Maison des sceaux in Delos—Ein ‘versiegelter’ Fundkomplex? Untersuchungen zur Aussagekraft und Interpretation der Funde eines durch Brand zerstörten hellenistischen Wohnhauses.” Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung. Vial, C. 1984. Délos indépendante. Bulletin de Correspondance Hellénique Supplement 10. Paris: Éditions E. de Boccard. Wallace-Hadrill, A. 1994. Houses and Society in Pompeii and Herculaneum. Princeton, N.J.: Princeton University Press.

Chapter 9

Housing the Poor and the Homeless in Ancient Greece Bradley A. Ault

“The nude Cynic fears no Wre for his tub (dolia); if broken, he will make himself a new house tomorrow, or keep it repaired with clamps of lead. When Alexander beheld in that tub its mighty occupant, he felt how much happier was the man who had no desires than he who claimed for himself the entire world.” —Juvenal, Satire 14.308–314; trans. G. G. Ramsey, Loeb ed., 1957 Diogenes: “You will take no thought for marriage or children or native land: all that will be sheer nonsense to you, and you will leave the house of your fathers and make your home in a tomb or a deserted tower or even a jar (πθον).” . . . “Leading this life you will say that you are happier than the Great King.” —Lucian, Philosophies for Sale 9; trans. A. M. Harmon, Loeb ed., 1929

In these two passages from the ancient sources, the father of Cynic philosophy is alternately admired and derided for making his home in a large storage jar. Since it is Diogenes’ own deliberate choice to reject the creature comforts of materialism, Juvenal Wnds him noble. Lucian, on the other hand, in a tone more satirical than Juvenal himself, Wnds it ridiculous: giving up the trappings of normal domestic life on the pretence that it will bring true fulWllment is pure nonsense. Nevertheless, Lucian’s Diogenes does indicate three sorts of places where the homeless might be found dwelling, in tomb structures, abandoned buildings, or pots. In what follows I would like to consider the archaeological evidence, or apparent lack thereof, for lower class housing, as well as to explore alternative domestic and living arrangements. I will underscore what has been perceived as our limited ability to identify the poor and

Housing the Poor and Homeless

141

the homeless in the archaeological record, while at the same time emphasizing that these individuals were present throughout society and played signiWcant roles in its overall landscape. Finally, I hope to clarify the semantics of “homelessness” in ancient Greece and will suggest a range of possible archaeological examples that I take to illustrate its varying states, from evidence for the dwellings of prostitutes to those needing temporary lodgings while traveling.

The Poor First off, who were the poor, the pene-stai, in ancient Greece? Ultimately they must be distinguished from one class of the homeless, the destitute, or ptochoí (for whom, see below). This distinction is stressed by both Liddell and Scott’s Greek-English Lexicon (πνης) and the Oxford Classical Dictionary (penia). The ranks of the poor comprised the “dependent” poor, slaves in particular, many of whom would have been foreign born. They also included the “independent” poor, the lowest class of citizens ( Jameson 1994, 61). In Athens, these were the thetes, whose importance as rowers in the Athenian navy is well attested.1 Resident aliens, too, might belong to the independent but impoverished masses (cf. Thür 1989). Given the legendary simplicity of even the great men of ancient Athens, it might be that we are unable to deWnitively identify the dwellings of the poor on architectural grounds at all. Recall Pseudo-Dicaearchus (1.1), discussing Athens: Most of the houses are mean, the nice ones few. A stranger would doubt, on seeing it Wrst, if this were really the renowned city of the Athenians. (F. Jacoby, ed., Die Fragmente der griechischer Historiker 2, 254; trans. Robinson 1946, 417, no. 69)

And Demosthenes (Olynthiac 3.25) states that: Out of the wealth of the state they set up for our delight so many fair buildings and things of beauty, temples and offerings to the gods, that we who came after must despair of ever surpassing them; yet in private they were so modest, so careful to obey the spirit of the constitution, that the houses of their famous men, of Aristides or of Miltiades, as any of you can see that knows them, are not a whit more splendid than those of their neighbors. (Trans. J. H. Vince, Loeb ed., 1930)

The Un-Free And so we should not really be surprised either that a study by Ian Morris, appropriately entitled “Remaining Invisible,” was essentially unable to distinguish the mining slaves of Thorikos on the basis of either architectural or ceramic grounds that might have associated them with their attested Thracian and Anatolian homelands (Morris 1998). At the same

142

Bradley A. Ault

time, however, the buildings in the Industrial Quarter at Thorikos are riddled with small cubicles that could as well have housed slaves as they did mine owners and their families or overseers left in charge (Fig. 6.1). It is worth noting the likelihood that due to these and other social circumstances the dwellings here deviate markedly from the expected syntax of Greek domestic architecture (a point made above by Nevett, Chap. 6). A similar situation holds for Delos, where not only were thousands of slaves bought and sold on a regular basis (Strabo 14.5.2), but slaves are also believed to have served as bailiffs for prominent families who frequently left their businesses and properties otherwise unattended (cf. Rauh 1993). The city was Wlled with human chattel whether or not we are able to locate it with certainty in space. Notable from Delos are grafWti from the small, claustrophobic room C in the Maison du Lac (Fig. 9.1a) one of which reads, “Behold the land of Antioch with its abundant Wgs and water. Oh savior Maeander, you come to our aid and give us water” (my translation; cf. Severyns 1927). Generally interpreted as having been scrawled by a Carian slave, and quite a literate one at that, the room may have served as his or her quarters.2 F. Hugh Thompson’s study, The Archaeology of Greek and Roman Slavery (Thompson 2003), takes into consideration slaves’ residences by drawing together much of the available published evidence. But even in these instances it is circumstantially derived from (small) room size, (marginal or dark) location, or (lack of) appointment. Treating agricultural and mining slaves separately, Thompson offers the (long-standing) opinion that the towers associated with farmsteads were used to house slaves (Thompson 2003, 56–59, 65), although Nevett (Chap. 6) raises another possibility. And at Thorikos, too, he notes the identiWcation of possible slave quarters associated with ore-washeries in the Laurion, citing examples at Megala Peuka, Thorikos (Washery 1. Fig. 6.1b), in the Soureza valley, and at Agrileza (Thompson 2003, 149–150, with references). Possible instances at Delos are presented as well, again drawing on previously published examples, including the Masion du Lac (Thompson 2003, 61–64, with references).

Size Doesn’t Matter Trümper’s study (Chap. 8) convincingly identiWes the existence of a great many small, simple, multipurpose, commercially oriented residences at late Hellenistic Delos. Many of these tabernae were occupied or even owned by representatives of the lower social spectrum, as well as leased to them. However, since many individuals also owned more than one residence, either as alternate dwellings or as rental properties, house size should not be the ultimate criterion by which poor dwellings are identiWed,

Fig. 9.1. a. Delos, Maison du Lac; b. Ano Siphai, Boeotia, small house (?).

144

Bradley A. Ault

however tempting it may be to do so (as Nicholas Cahill reminds me, pers. com., November 2000). The example of a simple, diminutive threeroom structure dating to the Wfth or fourth century, and built obliquely to the light defensive wall at Ano Siphai in Boeotia (Fig. 9.1b), has been taken to show not only the persistence of forerunners dating back to the Middle Bronze Age “megaron” houses, but perhaps also a dwelling of the working poor.3 Whether it is a house or something else, however (e.g., a hestiatorion or dining hall), remains open to speculation.

Synoikiai Also, the primary sources speak not infrequently of synoikiai.4 These are multiple occupancy structures, the identiWcation of which ranges from apartments, to boarding houses, to hotels or shops. As in the case of small dwellings, we are again tempted to associate synoikiai with a poorer or at least transient population. Although examples of such buildings as distinct from the katagogia discussed below are not readily identiWable, Tsakirgis (Chap. 5) cites the recent excavation and identiWcation of a complex on the north side of the Athenian Agora that is posited to have functioned primarily as rental ergasteria or workshops. Two standard-sized house plots at Olynthus show subdivision into two and as many as four dwellings respectively (Houses A viii 10 and C –x 7; Fig. 9.2a, b).5 If we interpret them as having borne a second storey, there is the potential here for doubling the number of rooms (and therefore dwellings as well) from the ground Xoor. And since it has a bearing on what is to follow, it should be noted that the designation of synoikiai could also be applied to brothels (Kraynak 1984, 20, citing Isaeus 5.19–21). In the end, it seems likely that given the relatively infrequent archaeological investigation of private buildings, as well as our inability to identify synoikiai (and hence to misidentify them as single occupancy dwellings), we are looking past examples which lay right before us.

The Homeless The greatest sources of homelessness in human history have been natural and man-made disasters. And while droughts and Xoods, earthquakes and hurricanes are unavoidable, the same cannot be said for war and its effects upon civilian populations.6 Accordingly, the recurrent nature of these calamities assures an ongoing stream of homeless populations at any one time. But if we can scarcely identify the poor in places where they are known to have resided in great numbers, how should we go about deWning, let alone identifying, those who are typically conceived of as the homeless or destitute?

Fig. 9.2. Olynthus. a. House A viii 10; b. House C –x 7.

146

Bradley A. Ault

During the Peloponnesian War, when Athens swelled with those who had Xed the countryside for the security of the city’s walls, Thucydides (2.52) tells us that the refugees lived in “suffocating shacks” (καλ βαις πνιγηραι ς). In the same setting, Plutarch (Life of Pericles 34) refers to “shacks and suffocating tents” (οκµασι µικροι ς και σκηνµασι πνιγηροι ς). It comes as no surprise that the combined effect of Athens in summer, the overcrowding of the city, and these substandard housing arrangements only fueled the plague that swept through the populace in 430. It may be that only under the most exceptional of circumstances can we attempt to Wx the ptochoí in space. Their association with beggary, for example, suggests that traces of habitation would largely be limited to residence in tents (skenai) or other impermanent structures and squatter activity.7 And dependent as they were upon the good will of others, they will not often have strayed too far from urban centers ( Jameson 1994, esp. 62–63). Archaeologically, we are unlikely to identify the remains of the temporary shelters that will have been used by the homeless unless we speciWcally set out to look for them.8 These individuals would always adapt themselves to any available niche within the topography of the ancient polis, from public spaces to alleyways, and intramural and as well as extramural zones.9 In the past as today, they will have turned up in whatever setting could accommodate them. While pre-historians and those working in the classical archaeology of the Roman provinces, or in the historical archaeology of other periods, are attuned to the detection of post-holes and more ephemeral traces of structures, as well as the diligent recovery and interpretation of total artifact assemblages, the settlement archaeology of ancient Greece lags behind in these areas. That said, instances of squatting, typically conWned to settlement abandonment contexts, have been identiWed at numerous sites, including Halieis (House D; third century), Olynthus (especially in the area of the Northwest Quarter; fourth century), and Thorikos (the tower compound in Insula 3; Roman).10 The very real but archaeologically inconspicuous nature of impoverishment is conjured up in the following passage from Aristophanes’ Wealth (534–546), where Chremylus is defending his newfound wealth against the goddess Poverty: What good can you ever provide, but blisters from the bath-house stove, and starving little children, and a rabble of old women? I can’t tell you the number of lice and mosquitoes and Xeas, so many there are, that buzz around our heads and irritate us, waking us up and saying “You’re hungry, so up you get.” We have to wear rags, not a decent cloak; instead of a bed, a rush-strewn pallet full of bugs, which wake us up if we’re asleep. A rotten rush mat instead of a carpet; instead of a pilllow, a dirty great stone by the head.

Housing the Poor and Homeless

147

To eat, instead of loaves of bread, shoots of mallow; instead of barley cake, weedy radish leaves; instead of a chair, the top of a broken wine jar, instead of a kneading trough, a rib of storage cask, also smashed; (trans. Fisher 1998, 94–95)

So much for my pessimism. A more optimistic approach to dealing with the problem is to acknowledge that we are confronted by numerous possible instances of the poor and homeless in the archaeological record already at hand. Simply put, I take “homelessness” foremost to mean those deprived of the conceptual rather than necessarily the physical aspects of the oikos. In what follows I would urge that we consider the state of homelessness in its wider semantic Weld. So, for example, and in keeping with the martial theme which with this section began, soldiers on campaign, sleeping in tents, are among the varieties of the homeless. As the passage from Lucian cited at the outset of this chapter plainly states, the Greeks understood the oikos, the house and household, its property and the domestic economy as it were, to include not only one’s possessions, but one’s kinfolk. Admittedly, wife and children were considered property. Aristotle even points out that “the poor, not having any slaves, must employ both their women and children as servants” (Politics 1322b–1323a; trans. B. Jowett, Cambridge, 1988). But they also transcend mere chattel. Similarly, an oiketes, or domestic slave, could equally be considered a member of the household ( Jameson 1990, 104). Family, as much as property, was seen as essential to and encompassed by the concept of the oikos. A sense of being, of belonging, and legitimacy was conferred upon one who was part of an established oikos. If one lacked any of these, they really did occupy a marginal if not a liminal place within Greek society.11 The state of being homeless, then, could be a temporary or long-term one. It could be transient or permanent. It could range from extreme unpleasantness (as for Chremylus), to a way of life, to constituting a happy state (if we are to believe Diogenes). Homelessness, as the Weld archaeologist or itinerant scholar knows well, is not necessarily a bad thing. It may carry with it longing for hearth and home, but often (as for Odysseus) it is also a time of liberation and adventure.

Porneia A unique category of the homeless in ancient Greece, as I am now arguing we deWne them, was occupied by prostitutes. Socially marginalized, they lacked the legitimizing trappings of family life and therefore stood outside the realm of the normal oikos. Hetairai and pornai, the higher and lower class varieties of prostitutes respectively, were nevertheless very much a feature of the polis and certainly played important roles in

Fig. 9.3. Athens, Kerameikos, Building Z. a. Phase 1; b. Phase 2; c. Phase 3.

Housing the Poor and Homeless

149

it. Frequently their places of work and residence were combined, since they will not usually have been admitted to typical family life. (The case of the famed courtesan Neaira provides a notable exception; Demosthenes 59.) In Athens, and likely in other cities as well, at least some of the porneia or brothels are actually thought to have come under state control or had taxes levied on them.12 Eva Keuls has stated that “We have no knowledge of the conditions prevailing in the brothels of Corinth and Athens, but there is no reason to assume that they were any more commodious than the dark and stinking holes in which Roman whores practiced their trade” (Keuls 1985, 156). However, recent excavations by the German Institute just within Athens’ Sacred Gate have revealed a structure that has for good reasons been identiWed as a porneion. This is the so-called Building Z (Fig. 9.3).13 And it does not resemble a “dark and stinking” hole. It might even be considered “commodious.” Now the Kerameikos was renowned for its brothels (e.g., Alexis 206). Indeed, the very name of the neighborhood was a by-word for them.14 Building Z occupies a 600 m² plot built up against the city wall just south of the Sacred Gate. Its three principal phases span the third quarter of the Wfth to the end of the fourth century. In its earliest conWguration, Building Z1 (Fig. 9.3a), at least Wfteen rooms clustered around two small courtyards. Although the excavators identify it as a private house, I would note that it is quite unlike any other such domestic structure we know from the later Wfth century.15 Moreover, there is evidence that at least three of the rooms around the central courtyard served as banqueting facilities: Room C was preceded by a mosaic anteroom; Rooms O and P, like Room C, and others along the western limits of the house, had red plastered walls, and yielded signiWcant amounts of Wne ware pottery associated with the consumption of food and drink.16 Building Z1 was destroyed by an earthquake between 427 and 420, which, we are told, buried the structure along with a great many possessions of the building’s inhabitants. Late in the 420s, Building Z2 (Fig. 9.3b) was erected over the site of its predecessor. The two courtyards were combined into one and the range of rooms around it were standardized in size. This structure was probably destroyed in 404, along with the razing of the city wall at the end of the Peloponnesian War. In this instance very little of the artifact assemblage associated with the building was preserved.17 The excavators again identify a private residence here, perhaps under the same ownership as Building Z1. After lying in ruined abandon for several decades, in the second half of the fourth century Building Z3 was established (Fig. 9.3c).18 The number of cell-like rooms around the southern and western perimeters of

150

Bradley A. Ault

the building was increased, as was its water supply. In addition to a well, three enormous interconnected cisterns were fed from rainwater collected off of the surrounding roof. But it is the artifact assemblages from the structure, “almost completely preserved in situ,” the excavators tell us (Knigge 1991, 93), that is so suggestive of the lives of those who inhabited Building Z3. The distribution of large numbers of loomweights suggests that each room was provided with its own loom. Hundreds of Wne ware vessels associated with the consumption of food and drink were also recovered. And Wnally, many Wgurines, amulets, and other items of jewelry are suggestive of associations with Aphrodite as well as a foreign origin for the building’s inhabitants.19 It is at this point that the excavators quite willingly identify Building Z3 as a tavern and brothel. The textile manufacturing that is quite clearly attested from the structure as well points to the cottage industry taking place there at the hands of its occupants when they were not otherwise engaged.20 Building Z3 was destroyed, perhaps by an earthquake, around the end of the fourth century. I am quite willing to identify Building Z in all three of its Wfth- and fourth-century incarnations as a porneion, although we may equally be able to detect in its evolution the development from a katagogion or hostelry to a brothel. I believe that all the evidence points to it. Neither the architecture nor the artifacts recovered there suggest a simple domestic structure. With Building Z, then, we are truly in the presence of housing for the homeless.21

On the Road In Aristophanes’ Frogs, Dionysus is about to undertake a journey to Hades. Before embarking, he stops in at the house of Heracles (actually dressed as Heracles!) to ask advice: But I have come, wearing this costume to mimic you so you might tell me the friends who entertained you when you went to get Cerberus. Tell me, too, the harbors, bakeries, brothels (πορνε ι α), resting places, forks in the road, springs, roads, cities, dwellings, and the hostels (πανδοκευτρας) with whom the fewest bugs were to be found. (109–114; trans. Kraynak 1984, 207–208)

So just as today’s vacationers, business travelers, and conference goers are temporarily without a home, in antiquity, those traveling on everything from diplomatic missions to religious pilgrimages might have availed themselves of lodgings.22 Appearing only in the Wfth century as a specialized architectural type, hostels have been identiWed at sanctuaries as well as in urban settings. It is Thucydides, referring to an example at Plataea, who Wrst employs the term katagogion (3.68). The association of katagogia is frequently with that

Fig. 9.4. a. Leonidaion, Olympia; b. Epidauros, Katagogion.

152

Bradley A. Ault

of housing pilgrims at cult sites, where they could also overlap with attendees at athletic events. The most famous preserved example is that of the Leonidaion at Olympia (Fig. 9.4a; Kraynak 1984, 49–65).23 Erected in the 320s by Leonidas of Naxos, it measures 74.8 by 81.08 m, was surrounded by an Ionic colonnade and had a central court with a Doric peristyle. Although heavily remodeled in the Roman period, its original fourth-century phase may have had as many as 50 rooms, those on the west side being considerably more spacious. Between 100 and 300 people could have used the building (it never rose beyond a single storey in height), and these are likely to have been ofWcials of some sort, perhaps the theoroi sent out by various poleis represented at the sanctuary.24 An even more impressive hostel was built at Epidauros in the late fourth to early third century (Fig. 9.4b; Kraynak 1984, 63–73). Measuring 76.3 m. on each side, this is the largest documented Greek katagogion. There are 70 rooms of varying size that could have accommodated guests, and these are clustered around four intercommunicating courtyards with Doric peristyles.25 Although evidence for a second storey is inconclusive, estimates are that the building could hold between 140 and 420 lodgers. The need for such a sizable hotel at a busy place like Epidauros is obvious, for in addition to its ongoing cult functions, athletic and theatral events would have drawn visitors.26 The alternative design to these squarish courtyard-centered hostels consists of rectangular structures, often resembling stoas. Notable examples of this type were erected in the late 4th century along the southern side of a racetrack in what would later become the forum of Roman Corinth, and within the sanctuary of Zeus at Nemea. The so-called South Stoa at Corinth measures 165 m in length by 25 m deep (Fig. 9.5a; Broneer 1954; Kraynak 1984, 108–117). Fronted by 71 Doric columns, there was an internal row of 34 in the Ionic order. Behind this deep porch, 35 suites of rooms opened, each of which was equipped with a drawshaft connected to a fresh water conduit fed by the Peirene fountain. There is evidence for a run of second storey rooms as well, reached by stairwells at each end of the building. It appears that the ground Xoor served to house restaurants and taverns, while the upstairs chambers provided short-term occupancy. Roughly contemporary with the South Stoa at Corinth, is the xenon from Nemea (Fig. 9.5b; Birge, Kraynak, and Miller 1992, 99–187). Measuring 85.89 m. in length and with a depth of 19.78 m, the building was originally laid out with 14 rooms. Although it lacked the colonnaded porch that deWned the South Stoa at Corinth, it is restored with second storey chambers along its northern extent. This seems to have been a highly specialized hostel. Erected in conjunction with a bath complex to the west, it appears to have been devoted to housing athletes participating in the Nemean Games.

Fig. 9.5. a. Corinth, South Stoa; b. Nemea, Xenon.

Fig. 9.6. Kassope, Katagogion. a. Phase 2; b. Phase 1.

Housing the Poor and Homeless

155

A katagogion with an urban and more strictly commercial orientation exists at Kassope in Epirus (Fig. 9.6a; Kraynak 1984, 73–79; Hoepfner and Schwandner 1994, 127–132). Located behind a stoa on the northwestern side of the agora there, and dating to the last quarter of the third century, it measures 30 by 32.6 m. 17 rooms opened onto a peristyle court with octagonal columns. The building is reconstructed with a second storey as well. This is, to my knowledge, the only example of such a structure integrated into an orthogonal city plan, where it occupies an area equivalent to four house parcels. Equally interesting is the predecessor building at Kassope (Fig. 9.6b), dating from the time of the city’s foundation and layout in the midfourth century, which took up only two house plots and resembled on a smaller scale the hall-like structures we have encountered at Corinth and Nemea. While the original identiWcation of the building as a katagogion has been challenged, its function having been alternatively interpreted as a market hall, there seems to be no reason why it cannot have been put to a number of different uses simultaneously or successively.27 The katagogion, then, and other types of alternative residences, from synoikiai to porneia, are multifunctional structures, designed to meet a variety of needs while at the same time providing the most basic one of shelter. In this regard, they are like the house itself. We can see across the many contexts of habitation for the homeless, as broadly deWned here, that all positions in the social hierarchy are represented. From humble tents or more ad hoc structures to Wne hotels, housing for the homeless mirrors the social spectrum. We can never hope to locate the barrel of Diogenes. Nor, I might add, should we have any desire to. But we can be sure that homelessness of one sort or another was as constant a feature of ancient Greece as it is in the early twenty-Wrst century. Notes 1. The relationship between the peasant class and the egalitarian tendencies of many poleis is explored in Wood 1988. 2. Interesting in light of what follows about prostitutes as one of the varieties of the homeless, Rauh interprets the Maison du Lac as a high-class brothel (Rauh 1993, 206–215). For a general presentation of the Maison du Lac, see Trümper 1998, 213–214, with 212, Wg. 17. 3. Hoepfner and Schwandner 1994, 322, Wg. 307. The structure has an area of ca. 92 m². By way of comparison, the smallest house identiWed in the lower town at Halieis, House A, covers ca. 133 m² (Ault, in press). 4. Cf. the discussion of rental property in Hennig 1999; as well as the references to synoikia collected in Frier 1977, 32, with note 31; Kraynak 1984, 20; and Nevett 1999, 158. 5. Hoepfner and Schwandner 1994, 110–112, with Wg. 88; Robinson 1946,

156

Bradley A. Ault

54–57, with pls. 44–47 (A viii 10), and 264–269, with pls. 229–231 (C –x 7; which is the same as Hoepfner and Schwandner’s E –x 7). And whereas Cahill does not agree with the reconstruction or interpretation of Hoepfner and Schwandner (pers. com., November 2000), he has identiWed what he believes to be an external stairway in House A iv 9 (Cahill 2002, 109), which could be taken as leading to a separate dwelling on the second Xoor. I am not certain but that this is not merely a variant arrangement of the recessed prothyron construction of the doorway. 6. Cf. Driessen’s study of “crisis architecture” or “warchitecture” (Driessen 1995, with many useful references), although here the concern is largely with identifying changes in the plan and use of monumental structures in times of general social upheaval or collapse. And although not dealing with homeless populations per se, Nowicki’s consideration of Dark Age refugee settlements in ancient Crete (Nowicki 1999) is another example of how those working in earlier periods of Greek archaeology are recognizing and exploring varying social conditions. 7. For a discussion of tents used by religious pilgrims, with many references, see Dillon 1997, 209–211; and Goldstein 1978, 8–100. 8. On this see, Cribb 1991, especially chap. 6: “Nomad Architecture and Domestic Space,” 84–112. 9. Scobie (1986) collects a good number of useful comparanda from the Roman world. 10. Halieis (Ault and Nevett 1999, 48, with 49, Wg. 4.2, and 50, table 4.2; Ault, in press); Olynthus (Cahill 2002, 49–61); Thorikos (Spitaels, et al. 1978, 109). 11. Cf. the study by Connor (1985) on kataskaphê, the physical destruction of the house. Along with the denial of proper burial to oneself or one’s relations, the conWscation of property, imposition of exile, curses and Wnes (one or more of which always accompanied it), kataskaphê was one of the most injurious punishments levied in ancient Greece and Wnds parallels in other societies as well. 12. On state control of brothels see Athenaeus 13.569d–f; on their taxation see, Graham 1998, 37 (citing Boeckh 1886, 404–405), and 39, where he notes, “We may be hesitant to believe that Solon established prostitutes at Athens (Athenaeus 13.569d–f), but the state’s interest in prostitution is clear.” Similarly, Rosivach (1995, 2–3) believes that Athenaeus’ claim that Solon established publicly owned brothels in Athens originated in a commonplace joke, probably from Attic Old Comedy, based on the cult epithet of Aphrodite Pandemos. I am grateful to L. C. Nevett for bringing this latter reference to my attention. The stele from the harbor at Thasos (ca. 470–460) is convincingly argued by Graham (1998) to contain provisions on the regulation of prostitution. 13. While the Wnal publication of the building remains forthcoming, preliminary discussions as well as illustrations can be found in Davidson 1997, 85–90; Kiderlen 1995, 39–42; Knigge 1991, 88–94; and Lind 1988. Judith Binder, I am told (Susan G. Cole, pers. comm., November 2000), has a theory that Building Z served a very different function: that it was the residence of the girls who were selected to weave the peplos of Athena. This is interesting in light of the structure’s location just within the Sacred Gate, as well as the quantities of loomweights recovered therein (a matter considered in a different light below). 14. Cf. the other sources collected in Lind 1988. City gates and the roadways leading out of town were a frequent haunt of prostitutes, as were the cemeteries that often lined them. The Via Appia of ancient as well as contemporary Rome provides a striking parallel case in point. 15. It is Buildings Z1 and Z2 that Kiderlen identiWes as examples of megale oikia (1995, 39–42). The existence of such elaborate houses is, to my mind, not convincingly proven prior to the fourth century (cf. Ault 1998).

Housing the Poor and Homeless

157

16. Knigge 1991, 90, with 31, Wg. 27 (a Panathenaic Amphora dated ca. 450–425), and 91, Wg. 81. 17. A votive deposit of miniature vessels and terracotta Wgurines was recovered from Building Z2, but its precise location is not speciWed (Knigge 1991, 92, with Wg. 83). 18. Knigge notes that a crudely cut inscription from the general area, and dating to the years of its abandonment, pertains to the auctioning off of an available parcel for 450 drachmas (Knigge 1991, 92). 19. Knigge 1991, 93, with Wg. 85, 94, with Wg. 86 (a silver amulet dated 380–370, depicting Aphrodite as the evening star); cf. Davidson 1997, 86. 20. It is Davidson who makes the point most forcefully (Davidson 1997, 86– 90). Cf. the pun relating a prostitute’s work of weaving and sex in the concluding portion from the passage of Strabo cited in note 21 below. 21. It should be noted that Building Z stands as the most convincingly identiWed example of a porneion in the ancient Greek world. Famed brothels are attested, to be sure, particularly those associated with sanctuaries of Aphrodite (cf. Strabo 8.6.20, on the example at Acrocorinth, reputed to have 1000 “priestesses”). A complex near Markopoulo, in the Attic deme of Myrrhinous, identiWed as dedicated to the worship of Aphrodite, complete with what may be facilities for temple prostitutes, has only recently been announced and was discovered in the context of laying out the equestrian events complex for the 2004 Olympics. 22. A fundamental source for this section is the 1984 Ph.D. dissertation by L. Kraynak, “Hostelries of Ancient Greece” (Kraynak 1984). This includes extensive consideration of the primary source material pertinent to these structures including their terminology and what we know about conditions in them, as well as their basic architectural features. Kraynak goes on to explore 13 examples in some detail before offering her conclusions. A briefer, more synthetic treatment may be found in D’Arrigo 1996. 23. Hoepfner has recently re-identiWed the building in its initial phase as functioning not as a katagogion but as a hestiatorion, or banquet hall, based on the long tradition of such facilities in sanctuaries as well as its similarity to Macedonian palace architecture (particularly the examples at Aigai and Pella: Hoepfner 1996, 36–40). 24. Although there is certainly no reason why in the off-seasons the Leonidaion or other katagogia attendant to major sanctuaries would not have been available for private individuals to stay in. 25. The inXuence of domestic architecture upon the two western external doorways to their respective courtyards has been noted by Kraynak (1984, 65), who argues that they show a marked resemblance to the recessed prothyra used for the principal doorways in houses at a number of sites (including Olynthus and Halieis; see Ault, in press). 26. Of course, only a fraction of those visiting sanctuaries like Epidauros at any one time will have been housed in such buildings. Most pilgrims will have stayed in tents or simply out in the open; cf. the references cited above in note 6. 27. Hoepfner and Schwandner (1994, 129–130) question the interpretation of the original excavator, S. Dakaris (Dakaris 1989, 32–38), on the basis of Wnds attested from the building, which include numerous weights and a marble measuring table. Accordingly, they see in the building type a forerunner of the Roman macellum. Kraynak herself notes the Xexibility of the terminology for hostelries vis-à-vis their functions as attested in the literary sources (Kraynak 1984, 9–21).

158

Bradley A. Ault

Literature Cited Ault, B. A. In press. The Houses. The Organization and Use of Domestic Space. Excavations at Ancient Halieis 2. Bloomington: Indiana University Press. Ault, B. A. 1998. Review of Kiderlen 1995. Journal of Hellenic Studies 118, 244–245. Ault, B. A. and L. C. Nevett. 1999. “Digging Houses: Archaeologies of Classical Greek and Hellenistic Domestic Assemblages.” In The Archaeology of Household Activities, ed. P. Allison, 43–56. London: Routledge. Birge, D. E., L. H. Kraynak, and S. G. Miller. 1992. Excavations at Nemea, Vol. 1, Topographical and Architectural Studies: The Sacred Square, the Xenon, and the Bath. Berkeley: University of California Press. Boeckh, A. 1886. Die Staatshaushaltung der Athener. 3rd ed. Vol. 1 Berlin: Reimer. Broneer, O. 1954. The South Stoa and Its Roman Successors. Corinth 1.4. Princeton, N.J.: American School of Classical Studies. Cahill, N. 2002. Household and City Organization at Olynthus. New Haven, Conn.: Yale University Press. Connor, W. R. 1985. “The Razing of the House in Greek Society.” Transactions of the American Philological Association 115, 79–102. Cribb, R. 1991. Nomads in Archaeology. Cambridge: Cambridge University Press. D’Arrigo, M. 1996. “Il katagogion: un ediWcio tra il pubblico e il privato.” In Ricerche sulla casa in Magna Grecia e in Sicilia, ed. F. D’Andria and K. Mannino, 89–107. Lecce: Congedo. Dakaris, S. I. 1989. Κασσ πη: Νε τερ ες ανασκαϕ ς (1977–83). Ioannina: Ioannina University. Davidson, J. 1997. Courtesans and Fishcakes: The Consuming Passions of Classical Athens. New York: St. Martin’s Press. Dillon, M. 1997. Pilgrims and Pilgrimage in Ancient Greece. London: Routledge. Driessen, J. 1995. “‘Crisis Architecture’? Some Observations on Architectural Adaptations as Immediate Responses to Changing Socio-Cultural Conditions.” Topoi 5, 63–88. Fisher, N. 1998. “Rich and Poor.” In Cambridge Illustrated History of Greece, ed. P. Cartledge, 76–99. Cambridge: Cambridge University Press. Frier, B. W. 1977. “The Rental Market in Early Imperial Rome.” Journal of Roman Studies 67, 27–37. Goldstein, M. 1978. “The Setting of the Ritual Meal in Greek Sanctuaries: 600–300 B.C.” Ph.D. dissertation, University of California, Berkeley. Graham, A. J. 1998. “The Woman at the Window: Observations on the ‘Stele from the Harbour’ of Thasos.” Journal of Hellenic Studies 118, 22–40. Hennig, D. 1999. “Vermietung.” In Geschichte des Wohnens, vol. 1: 500 v.Chr.–500 n.Chr. Vorgeschichte—Frühgeschichte—Antike, ed. W. Hoepfner, 596–600. Stuttgart: Deutsche Verlags-Anstalt (Wüstenrot Stiftung, Deutscher Eigenheimverein e. V., Ludwigsburg). Hoepfner, W. 1996. “Zum Typus der Basileia und der Königlichen Andrones.” In Basileia. Die Paläste der hellenistische Könige, ed. W. Hoepfner and G. Brands, 1–43. Mainz: P. von Zabern. Hoepfner, W. and E.-L. Schwandner. 1994. Haus und Stadt im klassischen Griechenland. Rev. ed. Wohnen in der klassischen Polis 1. Munich: Deutscher Kunstverlag. Jameson, M. H., 1990. “Domestic Space in the Greek City-State.” In Domestic Architecture and the Use of Space: An Interdisciplinary Cross-Cultural Study, ed. S. G. Kent, 92–113. Cambridge: Cambridge University Press. ———. 1994. “Class in the Ancient Greek Countryside.” In Structures rurales et sociétés antiques: Actes du Colloque de Corfou (14–16 Mai 1992), ed. P. N. Doukellis

Housing the Poor and Homeless

159

and L. G. Mendoni, 55–63. Annales littéraires de l’Université de Besancon 508. Paris: Les Belles Lettres. Keuls, E. C. 1985. The Reign of the Phallus: Sexual Politics in Ancient Athens. New York: Harper and Row. Kiderlen, M. 1995. Megale Oikia: Untersuchungen zur Entwicklung aufwendiger griechischer Stadthausarchitektur von der Früharchaik bis ins 3 Jh.v.Chr. Hürth: Verlag Martin Lange. Knigge, U. 1991. The Athenian Kerameikos: History—Monuments—Excavations. Athens: Deutsches Archäologisches Institut. Kraynak, L. H. 1984. “Hostelries of Ancient Greece.” Ph.D. dissertation, University of California, Berkeley. Lind, H. 1988. “Ein Hetärenhaus am Heiligen Tor? Der Athener Bau Z und die bei Isaios (6, 20 f.) erwähnet Synoikia Euktemons.” Museum Helveticum 45, 158–169. Morris, I. 1998. “Remaining Invisible: The Archaeology of the Excluded in Classical Athens.” In Women and Slaves in Greco-Roman Culture: Differential Equations, ed. S. R. Joshel and S. Murnaghan, 193–220. London: Routledge. Nevett, L. C. 1999. House and Society in the Ancient Greek World. Cambridge: Cambridge University Press. Nowicki, K. 1999. “Economy of Refugees: Life in the Cretan Mountains at the Turn of the Bronze and Iron Ages.” In From Minoan Farmers to Roman Traders: Sidelights on the Economy of Ancient Crete, ed. A. Chaniotis, 145–171. Heidelberger althistorische Beitrage und epigraphische Studien, 29. Stuttgart: F. Steiner. Rauh, N. K. 1993. The Sacred Bonds of Commerce: Religion, Economy, and Trade Society at Hellenistic Roman Delos, 166–87 B.C. Amsterdam: Gieben. Robinson, D. M. 1946. Excavations at Olynthus vol. 12, Domestic and Public Architecture. Baltimore: Johns Hopkins University Press. Rosivach, V. J. 1995. “Solon’s Brothels.” Liverpool Classical Monthly 20, 2–3. Scobie, A. 1986. “Slums, Sanitation and Mortality in the Roman World.” Klio 68, 399–433. Severyns, A. 1927. “Deux ‘grafWti’ de Délos.” Bulletin de Correspondence Hellénique 51, 234–243. P. Spitaels, J. Bingen, A. Uyttendaele, F. Blondé, K. van Gelder, A. Cheliotis, and A. Helsen, 1978. Thorikos 1970/1971, 7. Rapport préliminaire sur les septième et huitième campagnes de Fouilles. Ghent: Comité des fouilles Belges en Grèce. Thompson, F. H. 2003. The Archaeology of Greek and Roman Slavery. Reports of the Research Committee of the Society of Antiquaries of London 66. London: Duckworth. Thür, G. 1989. “Wo wohnen die Metöken?” In Demokratie und Architektur: Der hippodamische Städtebau und die Entstehung die Demokratie (Konstanzer Symposion vom 17. bis 19. Juli 1987), ed. W. Schuller, W. Hoepfner, and E.-L. Schwandner, eds., 117–121. Wohnen in der klassischen Polis 2. Munich: Deutscher Kunstverlag. Trümper, M. 1998. Wohnen in Delos: Eine baugeschichtliche Untersuchung zum Wandel der Wohnkultur in hellenistischer Zeit. Internationale Archäologie 46. Rahden: Verlag Marie Leidorf GmbH. Wood, E. M. 1988. Peasant-Citizen and Slave: Foundations of Athenian Democracy. London: Verso.

Chapter 10

Summing Up: Whither the Archaeology of the Greek Household? Bradley A. Ault and Lisa C. Nevett

The foregoing nine chapters adopt a wide variety of approaches to the study of Greek houses and households, and are uniWed by their archaeological perspective. They encompass a range of chronological and geographical contexts, as well as social, cultural, and economic ones. Far from underscoring any sort of homogeneity (whether in terms of ground plans, typologies, or domestic organization as a whole), the overarching themes have been change, diversity, and adaptability. Houses are seen as “meaningful architecture” (Locock 1994), constituting a unique form of “social space” (Parker Pearson and Richards 1994). They therefore merit consideration as an independent and highly signiWcant form of evidence, as each contribution to this work demonstrates. The buildings themselves, and in some cases their artifact assemblages, have been read as primary sources, as “texts” for documenting both the minutiae of lives within them and how these lives articulated with wider social and economic spheres. In what follows we would like to offer some thoughts about what we take to be major themes introduced and points pursued throughout this work.

Texts Versus Archaeology? In her introduction to the volume, Nevett provides a brief summary of past study of Greek houses in classical archaeology. Indeed, it is brief because the examples are relatively few, at least in terms of sites where such remains were the primary focus. ConWrming the traditional (misguided) adage that “archaeology is the handmaid of history,” she outlines how in the nineteenth and twentieth centuries ancient houses were seen as illustrative of the literary sources, which were, in true circular fashion, used to substantiate and elucidate the conclusions drawn from archaeology.

Summing Up

161

This raises the textual conundrum: for while the study of Classical Greece has at its disposal an embarrassment of riches in terms of primary, written, source material, matters domestic Wgure there largely as background for other issues rather than as a major focus. Moreover, the literary sources are severely biased, not only by their Athenocentrism, but also by their upper-class, male, and moralizing perspective. The one ancient source that does explicitly deal with “The Greek House” is the manual De Architectura written by the Roman architect Marcus Vitruvius Pollio in the Wrst century B.C. (Vitruvius 6.7). But what could Vitruvius actually know about housing in Greece in the Archaic and Classical periods? Nothing, of course, although some of his observations are, in fact, valid ones, since he surely had access, if only secondhand, to Greek houses of the Hellenistic period in Southern Italy and Sicily (compare Graham 1966, 16; and Tsakirgis 1989). So perhaps, employed judiciously, he can assume a place as a useful source, if a rather late and geographically removed one. Nevertheless, this is not the way he has always been used. For example, prostas versus pastas porches are not distinguished by Vitruvius, they are merely so designated. (“This space is named ‘prostas’ by some, and ‘pastas’ by others”; 6.7.1, trans. Rowland 1999). It is only considerably more recent scholarship that has attempted to assign the term to a speciWc “type” of porch meeting speciWc criteria. In an effort to bridge the gap, Aylward, in Chapter 3 here, emphasizes the Xuidity of this designation for Greek houses in the Troad. Elsewhere, Ault has suggested using the neutral term “transverse hall” to refer to the porch generally opening off of the north side of the courtyard and giving access to the primary living rooms of the house (Ault 2000, 488–89; Ault in press “a,” chap. 3). Porches can be seen in other regions and in various examples included in this volume, from eighth-century Zagora to Wfth-century Attica and beyond. The emphasis throughout this volume is on the positive role archaeology can play in its own right, furnishing material capable of interpretation beyond the strictures imposed by literary evidence. Archaeology, as employed here, is liberated from servitude to history. As Nevett states in the introduction, conclusions drawn “have emerged from the identiWcation of patterns in the organization of domestic space” (Nevett, Chap. 1). This is demonstrated through examples of changes in house plans over time and through space, as well as through variability in the use of household space. While Nevett enumerates the various types of rooms that have been identiWed in houses, she reinforces the multifunctional nature of most. Although “the most widely discussed feature is that convincingly identiWed as an andron” (Nevett, Chap. 1), andrones are conspicuous in this volume by their general absence from the houses under consideration (a

162

Bradley A. Ault and Lisa C. Nevett

matter taken up by both Nevett and Tsakirgis in their respective contributions, and returned to again below). Meanwhile, in recent years archaeology has helped us to rethink the most enduring myth of the Greek house and its organization, namely that domestic space was strictly gendered, divided into women’s and men’s quarters (respectively the gynaikonitis and andronitis). This model was Wrst refuted by Jameson (1990a, b), followed by Nevett (1994, 1995, 1999); both argued against assertions made by, for example, Walker 1983 and Keuls 1985. Many houses were clearly not segregated along inXexible female and male lines, but rather, the use of space depended on the nature of interpersonal and status relationships, and it was largely the scheduling of activities which helped to limit unwanted contact between respectable women and non-kin males. Even this may be an extreme view inapplicable to smaller communities, as Nevett suggests in her consideration of the Attic demes (Chap. 6). At the same time, dual courtyard houses such as those at Eretria, and the Herdraum houses of Epirus (perhaps also including related houses in Acarnania such as those at Leukas, discussed by Fiedler in Chapter 7), do show a physical ordering which might indicate more rigorously enforced social separation in a few communities (Karadedos 1990, Reber 1988). Archaeology is also capable of shedding light on other aspects of the symbolic dimensions of houses. Elements of personal display and self-promotion, formerly recognized primarily in the andron, can also be discerned in the elaboration of house façades (Nevett in press), the incorporation of elements of public architecture into the vocabulary of the domestic, in the rise of domestic luxus generally (Walter-Karydi 1998), and perhaps even in the inclusion of towers as part of the house compound (Nevett, Chap. 6). The analytical tools employed by the authors contributing to this study include the recognition of signiWcant patterning in ground plans and plotting of the distribution of Wnds. Important features include decoration and architectural elaboration, intercommunication between rooms and the locations of Wnds and architectural elements. Given the strengths of the archaeological record, the natural conclusion to be drawn here is that, except in unusual instances, the primary value of texts lies in helping to establish speciWc cultural contexts and in raising issues for investigation. Admittedly, for this, they are vital. Having disavowed a dependence upon literary sources as a sole means of establishing “the precise nature of the relationship between domestic space and society” (Nevett, Chap. 1), Nevett also questions the effectiveness of one-size-Wts-all anthropological models, and of the kind of abstract theory employed in other contexts to bridge the gap between the data and our full understanding of it. Law-like propositions and statements about the nature of Greek households will only be possible when they are formulated with reference

Summing Up

163

to the Greek world itself. And in spite of the obvious utility of middle range theory, such models must take into consideration regionally-based analyses, where possible, to complement site-based ones (much as Aylward and Nevett have begun to do here for the Troad and Attica respectively). Ultimately, throughout the volume as a whole the authors stress the need to go beyond the normative model of Greek houses and households, in order to explore their wide chronological, geographical, and social diversity. These themes, by which the volume is organized, are cross-cut by a number of topics which recur in different papers, and help to round out our picture of the nature of Greek oikoi and of their connection with a whole range of aspects of social and economic life.

Oikoi DeWned Before considering wider issues, it is appropriate to begin by thinking in more detail about the evidence itself and how it might be interpreted. Lang’s contribution on Archaic housing in Chapter 2 provides a good frame of reference. She offers an analytical approach which is potentially relevant to the study of houses of all periods, and goes some way towards developing a context-speciWc theoretical model for using the archaeological evidence of housing to explore the organization of ancient society. The issues raised in Lang’s contribution recur in others’ and speak to many of the concerns of household studies in general (for example, spatial differentiation based on gender and activity areas; signiWcance of reception areas; the functional signiWcance of typological distinctions). It is especially worthwhile to see the application of archaeological data to these matters in Greek contexts, and the good results that it can yield. Opening with a broad-based deWnition of the household, Lang continues to explore its implications throughout her text. Considerations such as house or household size, economic base, and social standing within and obligations to the community help to explain the variability of the examples she introduces and discusses. Of course, the very nature of Archaic Greek housing, and the character of the other sources that survive to document the period, are different from the situation for the Wfth century on. Surviving remains are often more ephemeral, their ground plans are less architecturally complex, and there is less literary and iconographic evidence against which to situate the archaeological material. Nevertheless, by studying the surviving evidence for variability in the material remains themselves, it is possible to begin to appreciate some of the differences between the households once occupying those spaces. A basic aspect of this is how ground plans and typologies reXect variant and changing household organization. Between the tenth and sixth

164

Bradley A. Ault and Lisa C. Nevett

centuries radical changes in the architectural form of houses accompany a divergence between private and ofWcial buildings, with the latter proliferating beyond the most obvious type, the temple (compare Mazarakis-Ainian 1997). A parallel source for revealing the social and domestic organization of the community is the overall settlement plan, its layout as well as its history, in as much as these can be known. Lang identiWes a number of cultural spheres that can be discerned by studying domestic space: the social, economic, technological, sociopsychological, symbolic, functional, and “representative” or display. A site like Zagora is ideal for studying settlement-wide changes in these areas over time (Lang, Chap. 2, Fig. 2.2), but they can also be observed elsewhere, albeit on a more limited scale. At the same time, the pattern of increasing segmentation of domestic space in communities like Zagora and Miletus is at odds with the situation occurring in Macedonia (e.g., at Kastanas: Lang, Chap. 2, Fig. 2.3), where multi-room structures were replaced with one- or two-room houses. While the precise reasons behind such shifts in organization cannot be determined, they must have had a social and economic basis; therefore, the mere fact of their recognition remains important and provides a basis for further study. As well as concrete and material manifestations of houses, Lang also shows how it is possible to address their more symbolic aspects, which depend to a greater extent on inference (for example, the suggestion that doorways mark transitions). The important symbolism of house and home is highlighted from a very different perspective in the contribution of Ault (Chap. 9), which explores the concept of the oikos by looking at those who were either temporarily or permanently without one of their own. These range from the poor, enslaved, or otherwise disenfranchised, to the transient and travelers of any variety, and they show how great a range of non-traditional and alternative types of residence or dwelling there in fact were. By examining the physical as well as conceptual varieties of homelessness, the centrality of the notion of the physical oikos is also emphasized. While all of these approaches yield valuable results, a variety of the caveats are brought out by a number of contributors: for example, it is a difWcult but important task to distinguish short-lived from long-lived features in any given house (houses and their distribution of functions are likely to evolve over time, as the household lifecycle waxes and wanes: compare Gallant 1991, 11–33). Similarly, the incomplete nature of excavation means that archaeologists rely on only a sample of any given site to speak for the whole. Regional differences (of both intra- and interregional varieties) affect the formulation of generalizations as well, as does the “highly variable” social structure that certainly existed across the Greek world. Finally, the widespread lack of artifact analysis, particularly

Summing Up

165

the study of assemblages and their spatial distribution, greatly hampers interpretation of domestic social life in houses. Fiedler’s presentation of the material from Leukas (Chap. 7) goes a long way to providing a case-study for several of the foregoing points. The houses there feature a distinctive layout over the course of the Classical and Hellenistic periods. Divided into zones Xanking the courtyard, the area of the house nearest the street entry is the most public. It is here that the most prominent living rooms of the house are located, including the andron as well as the oikos (e.g., Fiedler, Chap. 7, Fig. 7.11). Space becomes increasingly private as one penetrates the interior. Such a shift from front to back, from public and open to private and enclosed, is observable in houses elsewhere. But it is especially dramatic at Leukas, and is coupled with a series of architectural controls, such as the use of screen walls and consecutively entered suites of rooms. Indeed, the houses at Leukas are not unlike the Herdraum houses of Epirus in this way. House AII.5/AII.6 at Leukas also demonstrates how a careful analysis of the Wnds, including their quantiWcation and distribution, can aid in the analysis of spatial function. The minimum number of vessels of 1055 from the house (dating from the third or second to the Wrst century) is on the order of the few similar studies carried out on household inventories at other Greek sites (Ault and Nevett 1999). The importance of being able to deWne the oikos precisely in terms of the range of activities taking place there is underscored when it comes to looking at the way in which the domestic realm encompasses other activities. This is clear in the context of Cahill’s discussion of Lydian Sardis. The remains of the sixth-century glass-workshop there (Cahill, Chap. 4, Wg. 4.3) are remarkable not only for what survived of its contents, but also for its built installations which served as working surfaces, as well as its elaborate Wnish with polished plaster containing gleaming micaceous inclusions. Sardis presents a unique example of an Archaic Lydian city within the Greek sphere where high-end luxury craft production and raw materials extraction have been thoroughly documented, ranging as it does from glass to gold (Ramage and Craddock 2000). Within the same structure as the glass workshop a kitchen and part of a court both yielded up domestic installations and abundant assemblages. Cahill estimates that there is enough “pottery to feed two dozen and more people simultaneously,” “three or more nuclear families” (Cahill, Chap. 4). But is the building a “house”? This is a question of semantics, of course. And taking the broad interpretation adopted by this volume, the answer is “yes.” Given the full range of domestic activities represented, there can be little question that we are looking at a house. Parallel scenarios from the Greek world involving craft production in a “domestic” context are widespread, in fact, wherever there are houses that have

166

Bradley A. Ault and Lisa C. Nevett

been explored. Athens (Tsakirgis, Chap. 5), Delos (Trümper, Chap. 8), Halos (Reinders 1988; Reinders and Prumell, 2003), Halieis (Ault 1999), Olynthus (Cahill, Chap. 4), and the Rachi settlement at Isthmia (AndersonStojanovic 1996, 1997a) all come to mind. The problematic interpretation of the so-called “Citadel House Area” of Hellenistic Mycenae is instructive here. Originally seen as a series of buildings comprising “dye works” (Bowkett 1995), reconsideration of the evidence convincingly reidentiWes the remains as houses with oil or wine pressing installations (Anderson-Stojanovic 1997b).

Oikoi in the Economy: Variant Strategies Cahill’s discussion turns on one of the most prominent themes running through the volume, namely the economy: how did the economic role of the house contribute to shaping its built form and social function? What was the role of individual oikoi in the wider economic sphere? And to what extent did the economic status and occupation of an individual householder determine the way in which his oikos presented and conducted itself socially. The juxtaposition of household industry at Sardis with that at Olynthus provides valuable insight into the overall nature of the domestic economy. An important aspect is the prevalence and prominence of domestic industrial activity in the urban landscape as seen through the archaeology. This stands in contradistinction to literary sources that hold such pursuits in low esteem and contrasts with traditional scholarship which has sought to locate such activity in workers’ ghettos on the outskirts of town or in tightly clustered neighborhoods. Here, the idea of the self-sufWcient household producing everything required to meet its own needs is replaced by one involved in market exchange as a consumer household. Olynthus stands as the representative example of a Greek city with a wide range of “industrial” activities observable among its households, from sculptors to sling bullet makers (even though this latter “trade” should remind us of the siege to which the city was subject by Philip of Macedon in 348, and how warfare affects the normal conduct of life; compare Ault, Chap. 9). House A 6 at Olynthus (Cahill, Chap. 4, Fig. 4.2) furnishes an example of a range of activities taking place in the domestic setting. Here, there is evidence for the pressing of oil and grinding of grain on a large scale. Especially notable is the presence of the head of a female statue recovered from the courtyard, which was also crowded with pressing equipment and parts of twelve grindstones, as well as a Wnished andron opening off the “pastas” located there. Clearly, the economic livelihood

Summing Up

167

of the household existed alongside displays of status and must have made these possible by underwriting their cost. In addition to a demonstrable connection between evidence for manufacture and the presence of shop space appended to houses, there are half a dozen instances of what must surely have been very large storerooms within the insulae of the North Hill (Cahill 2002, 247–248). What remains in question, however, is whether these storerooms and their contents (likely to have been agricultural in nature) were the private property of a single household, or were more widely utilized. Such facilities may shed light on social organization and the domestic economies of Olynthus. For example, if such storerooms were communal rather than private, might they have been accessible to the sort of kin-based, regionally afWliated, or profession oriented neighborhoods that Cahill has proposed elsewhere (Cahill 2002, 221; 1992; 1991, 90)? Recalling the anoikismos (a “moving inland”) of 432 that brought together the population which settled the North Hill, there were no doubt many extended family groupings who came from the surrounding countryside and whose bonds would have remained strong in the setting of their new home. Compared with the widespread evidence for household industry on the North Hill at Olynthus, the Villa Section, which lay below in the plain to the east, appears to have been lacking. This seems unlikely to have been the result of greater wealth possessed by the inhabitants of the Villa Section. Indeed, sales inscriptions recovered from the site indicate that these properties were worth less than those on the North Hill (Nevett 2000; Cahill 2002, 276–281, 293–299). Rather, the difference appears to be the result of variant economic strategies employed in the two districts. Overall, one has the impression of the complete interpenetration of industrial and domestic realms at Olynthus. This is paralleled and supported by evidence from elsewhere, as at Athens (Tsakirgis, Chap. 5) and Halieis (Ault 1999, in press “a,” “b”). The question remains, however, of who performed the tasks of domestic industry. Cahill concludes that slave labor is likely, and on a large scale in some instances. It is also worth mentioning the work of Scheidel (1995, 1996) at this juncture, where extensive evidence is presented for the large-scale contribution of women to the ancient, especially agricultural, workforce (something which Nevett’s observations in Chapter 6 might also support). Similarly, Cahill’s own point about the possibility of extended groupings of kin-based or other afWliated groups residing in close proximity to one another in the city (already mentioned above), offers another possible concentrated source of labor. Cahill also suggests that there was a distinction between Greek domestic industrial establishments founded on slave labor, and their Lydian

168

Bradley A. Ault and Lisa C. Nevett

counterparts. On the basis of grafWti on 18 of 23 stemmed dishes found stacked in the kitchen of the glass workshop at Sardis, he suggests that the Lydian workshops were more independent, more well-off, and less hierarchically structured. This is an interesting scenario, even if other possibilities present themselves. What is successfully accomplished here is the effort to dispel the notion of industrial pursuits as being socially marginalized in Greek and Lydian society, whoever was overseeing it. Tsakirgis (Chap. 5) also takes up the matter of domestic industry, this time in Athens. She argues that the layout and features of Athenian houses differ from what are now recognized as “typical” elements of Classical Greek houses elsewhere, noting how the unplanned sprawl of the city, in so far as we know it, prevented the layout of regular blocks of insular housing. This led to the irregular shape of many houses, as well as to their variable size, some appearing to have been conceived from the outset as quite small. Similar constraints imposed by the urban plan are encountered at Delos as well (Trümper, Chap. 8). Tsakirgis also comments that Athenian houses near the Agora lack andrones, which have been documented in only two examples. And while there are no examples of the more “canonical” forms of vestibule, such as those recessed entryways often referred to as being of “prothyron” type (as in House A 6 at Olynthus: Cahill, Chap. 4, Fig. 4.2), variant arrangements are present (Tsakirgis, Chap. 5, Figs. 5.1 and 5.4). Another recurring feature is the kopron, a waste disposal pit, now documented in or just outside a number of Athenian houses (Ault 2000, esp. 555–556). In addition, the most recurrent element is a central courtyard. On balance, then, Athenian houses are not so very unlike contemporary houses elsewhere in Greece. Of the three Athenian houses presented in some detail, two appear to show a clear separation between work and living areas, as in the case of the glass workshop at Sardis. The general lack of private andrones noted in these contexts and in other houses near the Agora could perhaps be a result not only of our limited knowledge of Classical housing in the city center, but also of the increased availability of public dining rooms there, as exempliWed by those in the South Stoa (Travlos 1971, 334–337). Tsakirgis also makes the important point (taken up by Nevett as well in her treatment of housing in the Attic demes in Chapter 6) that this does not necessarily indicate a lack of symposia, which could easily have been more ad hoc affairs. But the question remains, in either case, does this reXect the status of the houses’ inhabitants, and in particular, could these have been properties occupied by Athens’ class of resident aliens, the metics (compare Thür 1989)? Paralleling Cahill’s observations about the examples at Olynthus, Tsakirgis emphasizes the fact that these workshop houses are located in the heart of the city in residential areas bordering on public space. They

Summing Up

169

are prominently situated along major roadways and at crossroads providing maximum exposure to potential clientele. Finally, there is the important observation that while “workshop houses” and household industry are prevalent at Athens and elsewhere, more specialized and exclusively workshop spaces are documented in both literary sources and archaeological examples. The evidence from the Agora offers a window into the lives of craftsman inhabitants of the urban core of Athens and graphically illustrates the extent to which dwelling and productive activities were intertwined at the domestic level in the Greek world (as is the case in pre-Industrial societies generally). A further perspective on the relationship between the house, household, and trade is offered by Trümper’s contribution (Chap. 8), which directs our attention away from Athens and the Classical period—a traditional focus of research. Instead, Trümper deals with the evidence from Delos. Eschewing the high status housing for which the island is so justiWably well known, she turns her attention to some of its small and simple dwellings that have been largely overlooked. Not unlike those at Athens and Olynthus considered by Cahill and Tsakirgis, these examples combine living and working spaces. Also, and especially important, Delos serves as a bridge between the Greek and Roman worlds. This is demonstrated by the fact that the best parallels for the types of less ostentatious buildings encountered at Delos are to be found in Pompeii and Herculaneum. Trümper shows how architecture can be employed as an index of socio-economic status. The hierarchy of building types she proposes for Delos is a means to developing a better understanding of the economic livelihood of the island. An important observation is that more than 500 multi-purpose tabernae have been documented at the site, and that such small structures far outnumber the familiar, more expansive, private dwellings. The types of modest dwelling range from tabernae to the simplest of “normal” houses. Her employment of a very close and systematic reading of the architectural remains demonstrates not only the great variety, but also the multiple phases, of such structures. In addition, concern is taken to emphasize the importance of second story rooms, often reached by separate external stairways accessible only from the street. Trümper explores how such structures were integrated into the urban landscape of Delos, considering their distribution and location with regard to neighboring streets, buildings and other topographical features. Her conclusions provide a dramatic demonstration of the effectiveness of household archaeology in addressing wider social and economic issues. Typically associated with low-status to modest housing or other mixed-use complexes, these residences differ from examples at Pompeii, over 50 percent of which were attached to high status housing. Thus, the tabernae

170

Bradley A. Ault and Lisa C. Nevett

of Delos seem to be operating in a very different social and economic milieu from their Campanian counterparts. The examples at Pompeii are typically integrated into house façades and hence are generally assumed to be rented spaces under the ownership, even if not under the direct proprietorship, of the householder. In Delos, however, the tabernae are not afWliated with larger dwellings (architecturally, at least), and so are apparently the result of middle- to lower-class entrepreneurship. At the same time, the independence of the tabernae, mixed-complexes, and the like on Delos suggests that the livelihood they represent was seen as a viable alternative to, and perhaps readily accepted alongside, the more “prestigious” mercantile sectors of the city’s population. Ultimately, like their counterparts at Athens and Olynthus, the Delian residences are seen as centers of business, whether they are grand peristyle houses of wealthy traders or the considerably simpler dwellings of humble merchants occupying “service” roles in the local economy.

Homo Politicus—Homo Domesticus The papers just discussed demonstrate the manner in which various aspects of the economic activity of households were intimately connected with their spatial organization and social relationships. But, as other contributions make clear, there are further, extra-domestic, factors which played a major role in shaping Greek households, and which can therefore be explored using the evidence of housing. In Chapter 3, Aylward takes up the evidence for Greek housing in the Troad, and relates aspects of its physical organization to the local political situation. Over the years the region has yielded remains of various domestic structures, however vestigial and suggestive. The information from Ilion, published in greater detail elsewhere (Aylward 1999), is supplemented with more recent Wndings from Neandria (Fatmann-Rey 1996) and Alexandria Troas (Papenberg and Schrader 1999). At these latter sites excavation and geophysical survey have recovered the plans of two houses, both of which are argued to be of pastas type. Aylward’s aim is not to present the material in its own right, but to see how it squares with conventional typological classiWcations of Greek houses. Another aspect of his study is the proposition that domestic architecture in the Troad reXects the political fortunes of the region during the Classical and Hellenistic periods. If he is correct, this is an important instance of the response of material culture to speciWc historical events (be it synoecism or Gallic invasion), and again demonstrates the fact that classical archaeology is unusual in having at its disposal important contextual data sets in the form of written historical sources, albeit they are

Summing Up

171

potentially biased. It also supports one of Lang’s initial points about the archaeological potential of housing as a measure or index of social change, a point revisited in subsequent chapters as well. Important for considering the typology of houses in the Troad is their location between regions traditionally associated with the presence of pastas and prostas houses (on mainland Greece and along the Ionian coast, respectively). Aylward’s ultimate conclusion, that such distinctions may not always be so clear-cut on the ground, with some houses perhaps defying classiWcation, provides yet one more conWrmation of the recent dismissal of the prostas/pastas distinction in the ancient context (for example, Nevett 1999, 81). In the Troad the organization of the domestic sphere may be related to widespread political conditions, and to the integration of individual poleis into a larger koine. But political structures, which inXuence the relationship between individual communities, can also be shown to have had an effect at a different scale, between groups within a single polis. In Chapter 6 Nevett highlights some of the ways in which such relationships may work by looking at housing in several of the 140 odd deme centers that made up this extraordinarily large polis. What we seem to see on the basis of the limited available evidence, is that certain conventions of domestic architecture encountered elsewhere, as in the “single entrance courtyard house” type, are subject to alteration in smaller, more rural settlements. This suggests that different social conventions were operating both within the household and in its relationship with the surrounding community. In spite of the limited survival and exploration of domestic structures in the demes, evidence from the two best-known examples, Thorikos and Ano Voula, suggests that they possess characteristics very different from those of Athenian houses. In particular, the organization of domestic space seems less concerned with controlling access to and movement within houses. The smaller size of the demes may account for this, where a different system of social controls was in place. Similarly, the lack of andrones may imply that the role of the symposium, and the function it served, varied from the Athenian institution or was accomplished through alternative means. That this translates into formal symposia as urban phenomena, with less formal occasions being conducted in the demes, may be a preliminary conclusion here. In the absence of andrones and other typical status markers, Nevett suggests that the towers seen recurrently at deme sites may have fulWlled a similar role, in addition to their more utilitarian aspects. In sum, a whole set of differences in organization and conduct may be observable by looking carefully at the evidence from sub(-)urban domestic settings in Attica.

172

Bradley A. Ault and Lisa C. Nevett

Conclusion In the course of this discussion we have tried to draw together a few of the themes running through the papers in this volume, by highlighting aspects of the individual papers which move beyond the data itself, and which contribute to wider debates. Our aim has not been to be comprehensive, but merely to suggest that the evidence for housing can be invoked in order to address a broad range of issues. We also want to suggest that, taken together, this volume represents more than simply the sum of its parts: although the geographical, chronological and social range discussed here is relatively broad, and a number of different approaches are employed, the juxtaposition of the individual case-studies suggests a variety of wider conclusions. One obvious point is simply the diversity of the physical, social, and economic groupings falling into the category of oikos. At the same time, however, the variability of the oikos makes it a valuable tool for investigating a whole range of aspects of the ancient Greek world which are understood imperfectly, if at all, from surviving textual evidence. Thus it becomes possible to study the domestic economy and its integration within the larger economic sphere, or to look at the effects of large-scale political circumstances on individual households and small communities. It is also feasible to compare communities in different areas of the Greek world and to get a better sense of the amount of change taking place in them over time. These discussions show that it is possible to widen the focus from simply the appearance of domestic architecture and the functioning of domestic social relations, which have tended to dominate previous work on Greek housing. Finally, these papers should not be seen as exhaustive treatments by any means. They are, rather, suggestive—opening up possible avenues for future research while at the same time showing how many challenges still remain. At the level of the data itself, it will be clear from the studies of Cahill, Fiedler, and Tsakirgis that the more detailed the information available from individual sites and structures, the more we can ultimately say about patterns of activity taking place there, and the better we can come to grips with social groupings and economic processes. It is thus vital to focus Weldwork not only on extracting information about the architecture of individual houses, but also about the objects found in those houses, for it is only this which will enable us to reconstruct patterns of activity in detail. At the same time, the approach taken by Lang suggests the possibility that careful work at the theoretical level can improve our analytical and interpretative framework, so that we can make better use of the information already at our disposal. The juxtaposition of material from different areas of the Greek world (for example,

Summing Up

173

in the contributions by Aylward, Cahill, and Fiedler), and the comparison between larger and smaller communities, and households of different status in areas that are better-known (as in the examples presented by Nevett and Trümper), shows that studying variability can help to move us beyond the stereotypical images which persist in our understanding of Greek households. Finally, the chapter by Ault demonstrates that, however valuable the archaeological evidence, prudent use of texts alongside this material can still yield valuable insights. Indeed, the real excitement, as well as the challenge, of working in the Greek context lies in our ability to use these different perspectives and sources in order to arrive at an almost uniquely detailed understanding of Greek society.

Literature Cited Anderson-Stojanovic, V. 1996. “Excavations in the Rachi Settlement at Isthmia, 1989.” Hesperia 65, 57–90. ———. 1997a. “Dye Works or Olive Press: A Reconsideration of Installations in the Rachi Settlement at Isthmia,” paper presented at the 99th Annual Meeting of the Archaeological Institute of America, Chicago. Abstract in American Journal of Archaeology 102 (1998): 369. ———. 1997b. Review of L. C. Bowkett (1995). Journal of Hellenic Studies 117, 244–245. Ault, B. A. 1999. “Koprones and Oil Presses at Halieis: Interactions of Town and Country and the Integration of Domestic and Regional Economies.” Hesperia 68, 549–573. Ault, B. A. 2000. “Living in the Classical Polis: The Greek House as Microcosm.” In The Organization of Space in Antiquity, ed. S. G. Cole. Classical World ( Journal of the Classical Association of the Atlantic States. Special Issue) 93, 483–496. Ault, B. A. In press “a.” The Houses: The Organization and Use of Domestic Space. Excavations at Ancient Halieis, 2. Bloomington: Indiana University Press. Ault, B. A. In press “b.” “Oikos and Oikonomia: Greek Houses, Households and the Domestic Economy.” In Building Communities: House, Settlement and Society in the Aegean, ed. R. Westgate, N. Fisher, and J. Whitley, Studies of the British School of Archaeology at Athens. Athens: British School of Archaeology. Ault, B. A. and L. C. Nevett. 1999. “Digging Houses: Archaeologies of Classical Greek and Hellenistic Domestic Assemblages,” in The Archaeology of Household Activities, ed. P. Allison, 43–56. London: Routledge. Aylward, W. 1999. “Studies in Hellenistic Ilion: The Houses in the Lower City,” Studia Troica 9, 159–186. Bowkett, L. C. 1995. Well-Built Mycenae, 36. The Hellenistic Dye-Works. Oxford: Oxbow. Cahill, N. D. 1991. “Olynthus: Social and Spatial Planning in a Greek City.” Ph.D. dissertation, University of California, Berkeley. ———. 1992. “Korkyra Melaina and the Distribution of Land in Greek Colonies,” paper presented at the 94th Annual Meeting of the Archaeological Institute of America, New Orleans. Abstract in American Journal of Archaeology 97 (1993) 345–346.

174

Bradley A. Ault and Lisa C. Nevett

———. 2002. Household and City Organization at Olynthus. New Haven, Conn.: Yale University Press. Fatmann-Rey, G. 1996. “Versuch der Rekonstruktion eines Wohnhauses von Neandria.” Asia Minor Studien 22, 15–42. Gallant, T. W. 1991. Risk and Survival in Ancient Greece. Reconstructing the Rural Domestic Economy. Stanford, Calif.: Stanford University Press. Graham, J. W. 1966. “Origins and Interrelations of the Greek House and the Roman House.” Phoenix 20, 3–31. Jameson, M. H. 1990a. “Domestic Space in the Greek City-State.” In Domestic Architecture and the Use of Space: An Interdisciplinary Cross-Cultural Study, ed. S. Kent 92–113. Cambridge: Cambridge University Press. ———. 1990b. “Private Space and the Greek City.” In The Greek City: From Homer to Alexander, ed. O. Murray and S. Price, 171–195. Oxford: Oxford University Press. Karadedos, G. 1990. “Υστεροκλασικ σπ τι στη Μαρωνε α Θρκης.” Egnatia 2, 265–297. Keuls, E. C. 1985. The Reign of the Phallus. Sexual Politics in Ancient Athens. New York: Harper and Row. Locock, M. ed. 1994. Meaningful Architecture: Social Interpretations of Buildings. Worldwide Archaeology Series, 9. Avebury: Aldershot. Mazarakis-Ainian, A. 1997. From Rulers’ Dwellings to Temples. Architecture, Religion, and Society in Early Iron Age Greece (1100–700 B.C.). Jonsered: Aström. Nevett, L. C. 1994. “Separation or Seclusion? Towards an Archaeological Approach to Investigating Women in the Greek Household in the Fifth to Third Centuries B.C.” In Parker Pearson and Richards 1994, 98–112. ———. 1995. “Gender Relations in the Classical Greek Household.” Annual of the British School at Athens 90, 363–381. ———. 1999. House and Society in the Ancient Greek World. Cambridge: Cambridge University Press. ———. 2000. “A Real Estate ‘Market’ in Classical Greece? The Example of Town Housing.” Annual of the British School at Athens 95, 329–343. ———. In press. “Domestic Façades: A ‘Feature’ of the Landscape of Greek Poleis?” In Inside the City in the Greek World: Studies of Urbanism from the Bronze Age to the Hellenistic Period, ed. S. Owens and L. Preston. Oxford: Oxbow. Papenberg, I. and P. Schrader. 1999. “Geophysikalische Untersuchungen im Stadtgebiet von Alexandria Troas.” Asia Minor Studien 33, 37–83. Parker Pearson, M. and C. Richards, eds. 1994. Architecture and Order: Approaches to Social Space. New York: Routledge. Ramage, A. and P. Craddock. 2000. King Croesus’ Gold: Excavations at Sardis and the History of Gold ReWning. Cambridge, Mass.: Harvard University Press, 2000. Reber, K. 1988. “AediWcia graecorum: Zu Vitruvs Beschreibung des griechischen Hauses.” Archäologischer Anzeiger 1988, 653–666. Reinders, H. R. 1988. New Halos: A Hellenistic Town in Thessalia, Greece. Utrecht: HES Publishers. Reinders, H. R. and W. Prummel, eds. 2003. Housing in New Halos. A Hellenistic Town in Thessaly. Lisse: A. A. Balkema. Scheidel, W. 1995. “The Most Silent Women of Greece and Rome: Rural Labour and Women’s Life in the Ancient World (I).” Greece and Rome 42, 202–217. ———. 1996. “The Most Silent Women of Greece and Rome: Rural Labour and Women’s Life in the Ancient World (II).” Greece and Rome 43, 1–10. Thür, G. 1989. “Wo wohnen die Metöken?” In Demokratie und Architektur: Der hippodamische Städtebau und die Entstehung die Demokratie (Konstanzer Symposion

Summing Up

175

vom 17. bis 19. Juli 1987), ed. W. Schuller, W. Hoepfner, and E.-L. Schwandner, eds., 117–121. Wohnen in der klassischen Polis 2. Munich: Deutscher Kunstverlag. Travlos, J. 1971. Pictorial Dictionary of Ancient Athens. London: Thames and Hudson. Tsakirgis, B. 1989. “The Universality of the Prostas House.” Paper presented at the 90th Annual Meeting of the Archaeological Institute of America, Baltimore. Abstract in American Journal of Archaeology 93 (1989), 278–279. Vitruvius. 1999. Ten Books on Architecture. Trans. I. D. Rowland, commentary and illustrations by T. N. Howe. Cambridge: Cambridge University Press. Walker, S. 1983. “Women and Housing in Classical Greece.” In Images of Women in Antiquity, ed. A. Cameron and A. Kuhrt, 81–91. London: Croom Helm. Walter-Karydi, H. 1998. The Greek House: The Rise of Noble Houses in Late Classical Times. Athens: Archaeological Society at Athens.

This page intentionally left blank

Glossary

Ambitus : an alleyway, often laid longitudinally through a housing block (insula). It may be associated with above ground drainage as much as offering a thoroughfare. Anoikismos : the process whereby the populations of several settlements relocated from coastal areas, moving inland, to form a single, larger nucleated center. Antae : extensions of the lateral walls of a building to form a shallow porch, sometimes thickened at their terminations to form pillars. Andron/andronitis : name applied in an archaeological context to a square room with an off-center doorway, presumably allowing for the placement of couches. In the textual sources the andron is named as the location of the symposium and is traditionally interpreted as a male-associated space. Exedra : a square or rectangular recessed space opening off the courtyard, not closed off by a wall or possessing a separate doorway. Gunaikon/gunaikonitis : term used in the textual sources for the domestic area of the house and interpreted as referring to a female space. The extent to which the archaeological evidence supports this literal interpretation has been debated. Herdraum/hearth-room: large, interior space with a central hearth which dominates the layout and therefore gives a distinctive form to houses at a few sites in northern Greece, including Kassope and Ammotopos. Hestiatorion : communal banqueting hall, often encountered in the setting of a sanctuary. Insula : a block of houses, varying in size and number, generally bounded by streets on all four sides. Katagogion : a hotel or hostel. Kline : banqueting couch. Koinon : league, federation, commonwealth, or state, as distinct from the polis.

178

Glossary

Kopron : stone-lined pit found in archaeological context s in the courtyards of houses and used for disposal of household wa ste. Labrys : a double-axe. Oecus maior : literally a “great hall,” a multi-purpose room used for living, reception, and dining, usually elaborated with plastered and painted walls, as well as expensive Xoor treatments, such as mosaic. Oikos : term used in the textual sources for a house and its associated property, but the primary living-room of houses may also be so designated. Pastas : portico in an interior court, standing in front of entrances to individual rooms. One of a variety of different types of portico which have been used as a basis for deWning distinctive house-types (the others being prostas and peristyle). Peristyle: colonnade running around four sides of internal courtyard of a house, and used to deWne the peristyle house type. Prostas : portico in internal court of a house sheltering the entrance to the major rooms (rather than running along the whole side of the court). Prothyron : a doorway, in particular, the external doorway to the house, often recessed into the exterior wall. Representational or representative space: used to designate a space interpreted as demonstrating the status of the house-owner to visitors, through architectural embellishment. Symposium : drinking party known from textual and iconographic sources and assumed to have been a male occasion, although also featuring female slaves and/or entertainers. The layout of excavated andrones seems to have been designed to accommodate such occasions (although of course they may also have been used for other activities). Stenopos : a street. Synoikia : a multiple residence or space for shops; apartment and or workshop complex. Synoikismos : the process whereby the populations of several settlements relocated to form a single, larger nucleated center. Taberna : a small shop, workshop, or tavern, generally included along the frontage of a (Roman) building. Thalamos : a room located within the core of the house , generally associated with sleeping or storage.

Contributors

Bradley A. Ault is Associate Professor of Classics at the University at Buffalo. William Aylward is Assistant Professor of Classics at the University of Wisconsin, Madison. Nicholas Cahill is Associate Professor of Art History at the University of Wisconsin, Madison. Manuel Fiedler received his Ph.D. from the Free University of Berlin. Franziska Lang is on the faculty of the Seminar for Classical Archaeology of Humboldt University, Berlin. Lisa C. Nevett is Assistant Professor of Classical Studies and the History of Art at the University of Michigan, Ann Arbor. Monika Trümper is Assistant Professor in the Institute of Archaeology, University of Heidelberg. Barbara Tsakirgis is Associate Professor of Classics and Art History at Vanderbilt University.

This page intentionally left blank

Index

Italicized page numbers indicate Wgures and tables. Abdera: location of, 2 ; prostas-type houses at, 36 Abydos: location of, 38 ; relocation and, 48 Achilleion, relocation and, 46 Aegina: ground-plans at, 16, 18; lead strips at, 75; location of, 15 Aeschines, 70, 78, 79 Ag. Andreas: arrangement of houses at, 19 agricultural processing, as household industry, 55, 56. See also Attic deme centers Agrileza, slave quarters at, 142 Alexander the Great, 36, 46 Alexandreia Troas: city wall of, 46, 51n22; coin of, 42; founding of, 39, 46; location of, 38 ; pastas-type houses at, 39, 40, 47, 48, 170; synoikismos of, 46, 48, 49 altar, 125 ambitus (canal or alleyway), 112 Ambrakia, abandonment of, 100 American School of Classical Studies, 67 amulets, 150 andron : absence of (summary), 161–62; absent in Athenian Classical houses, 69, 77–78, 168; absent in Attic deme centers, 88, 93, 95, 171; decoration of, 86, 96; features of, 5, 101; at Ilion, 47; interpretation of room as, 27; at Leukas, 101, 102, 107, 113, 115–16, 165; at Olynthus, 166 anoikismos, 48, 167 Ano Siphai (Boeotia): location of, 2 ; poor dwellings at, 143–44

Ano Voula: description of, 90, 92–93, 171; ground-plans at, 91; Kalampoka plot at, 90, 92, 93; location of, 2 ; Papacharalambos plot at, 90, 91; Thorikos compared with, 93, 95 Antigoneia (Troad). See Alexandreia Troas Antigonos Monophthalmos, 39, 46 Antiochos III, 49 Antiochos Hierax, 47 Aphrodisias: location of, 2 ; workshop at, 75, 80n5 Aphrodite, 150, 157n21 archaeology: household identiWcation in, 12–13; of Roman provinces versus ancient Greece, 146; textual evidence juxtaposed to, 1, 3–7, 160–63, 173. See also architecture; methodology Archaic houses: approach to, 12–14; distribution of apsidal, 17 ; economic aspects of, 27; representative aspects of, 30–31; room arrangement in, 24, 25, 26, 29; size of, 26–27; social space of, 20–27; sociopsychological aspects of, 28–29; spatial organization of, 4, 19–31; summary of, 31–32, 163–64; symbolic aspects of, 29–30; technological aspects of, 27–28; typology and organization of, 14–19 Archaic period: documentation of, 13–14; sites in, 2, 15 ; spatial organization in, 4 architecture: approach to, 1, 3–7; codiWcation of, 32; divergence between private and ofWcial, 18, 164; private adaptations from public, 31;

182

Index

socioeconomic status reXected in, 169–70; vocabulary of, 6 Aristarchus, 60 Aristophanes, 146, 150 Aristotle, 54, 55, 147 artifacts/Wnds: as critical source, 31–32, 172; depositional and post-depositional processes and, 99, 106, 116n2; function of rooms indicated by, 30, 106, 107, 108, 109, 110–11, 112, 126; hierarchy of rooms indicated by, 29; lack of analysis of, 164–65; “negative selection” of, 116n5. See also speciWc objects Asia Minor: map of, 2 ; struggle over, 47 Assiros: house types at, 14, 17; location of, 15 ; room arrangement at, 22, 24 Assos: building materials at, 37; location of, 38 ; in synoikismos, 48 asty, deme centers compared with, 83–84 Athenaeus, 156n12 Athenian Classical houses: Agora, 67–82; excavation of, 67, 69; implications of, 168–69; industrial and domestic activities in, 70–79, 85, 166, 167; irregularity of, 69; mutability of space in, 78; room arrangement in, 101, 103. See also Athens/Piraeus Athenocentrism, 161 Athens/Piraeus: brothels of, 147–50, 156n12; Building Z at, 148, 149–50, 156n13, 156n15; cleruchy of, 133–34; Great Drain in, 76; House C/D (Industrial District) at, 70, 73, 76, 77, 78; House of Mikion and Menon at, 70, 72, 74, 77; House of Simon at, 70–71, 77; House of the Greek Mosaic at, 80n3; house types at, 14; Industrial District of, 70, 73, 76, 77, 78; lack of street grid in, 69; location of, 2 ; poor of, 141; population of, 95; post-Persian clean-up of, 67, 69; public dining rooms in, 168; residential and workshop areas in, 78; room arrangement at, 20; social relations in, 83; synoikism at, 49; unplanned sprawl of, 168; war refugees in, 146. See also Athenian Classical houses athletic events, 150, 152 Attalos I, 47 Attic deme centers: approach to, 83–84; description of housing in, 84–90; ground-plans of houses in, 86, 87, 91,

92, 94 ; social relations in, 84, 95–96; spatial organization of houses in, 90–93, 171 Babes: architectural adaptations at, 31; location of, 15 back-room of taberna, 126 Baiae (Italy), workshop at, 75 Bankel, H., 75 bathrooms, 104, 110, 111, 114 bathtubs, 28, 30, 104 beads, 62 Bettalli, M., 79 Binder, J., 156n13 Birytis, relocation and, 46 bone objects, 70–71, 74 bronze objects: food preparation, 109, 112; household industry, 62; smithy, 76; textile production, 110, 111 brothels (porneia), 147–50, 155 Cephalus, 78 Charidemos of Oreos, 45 Chechnya, stone towers in, 96 circulation patterns: approach to, 5; doorways as indicators of, 24, 25, 26; multifunctional rooms and, 162; sociopsychological sphere and, 28–29 cisterns. See wells and cisterns city walls: of Alexandreia Troas, 46, 51n22; housing shaped by, 50; of Ilion, 39, 41, 42, 43 ; security of, 36 Classical houses: features of, 113–16; Late Hellenistic compared with, 119; at Leukas, 100, 102 ; room arrangement in, 101, 103. See also Athenian Classical houses Classical sites: gendered domestic space at, 5, 27; homestead farms at, 46; literary sources on, 161; map of, 2 cleruchy, of Athens, 133–34 cobbler’s workshop (Athens), 70–71, 74 coins, 42 Colophon: location of, 2 ; prostas-type houses at, 36 Connor, W. R., 156n11 cooking wares: household industry and, 56, 58, 63–64; room function indicated by, 30, 106, 107, 108, 109, 110–11, 112 coral, 63 Corinth: brothels at, 149; ground-plans at, 18; hostel at, 152, 153 ; location of, 2 ;

Index room arrangement in, 101, 103 ; room functions in, 30; South Stoa at, 152, 153 ; Terracotta Factory at, 101, 103 coroplasty, 56, 58, 101 courtyards: of Athenian Classical houses, 69, 70, 74, 75, 76, 77, 101; of Attic deme centers, 85, 90, 92–93; of brothels, 148 ; of Delian “normal-house,” 119; dual, 162; of House AII.5/6 (Leukas), 104, 105, 106, 111, 114, 115; household industry in, 58, 62, 63, 74; as multifunctional, 30; room arrangement and, 20, 22; of tabernae-complexes, 122, 123, 124–25; as widespread, 168 craftsmen and traders: assumptions about, 54, 166; identity of, 59–60. See also household industry; shops and workshops Crete, refugee settlement in, 156n6 crisis architecture (concept), 156n6 cullet, 62–63 curse tablet, 76 Cynic philosophy, 140 Cyrus II (Cyrus the Great), 60 Dakaris, S., 157n27 Dark Age, spatial organization in buildings in, 4 Davidson, J., 157n20 decoration and decorative elements: of andron, 86, 96; of Attic deme centers, 88, 93; of brothels, 149; context of, 5; of glass workshop, 62; hierarchy of rooms indicated by, 29; of House AII.6 (Leukas), 107, 110, 113, 115; representative sphere and, 30–31; of tabernae-complexes, 128 Delian modest housing units: approach to, 119–20; architectural and urban context of, 129–32; implications of, 169–70; range and character of, 120–28; social strata represented in, 132–37 Delos: architectural changes at, 115–16; as bridge between Greek and Roman worlds, 169; epigraphic evidence at, 6; excavation of, 3, 137–38n1; ground-plans of, 121 ; Îlot de la Maison des masques at, 136; industrial and domestic activities in, 85, 138n7, 166; location of, 2 ; Maison à une seule colonne at, 126, 127 ; Maison de Cléopâtre at, 126; Maison de l’Hermès at, 134; Maison

183

des sceaux at, 138n7; Maison du Lac at, 142, 143, 155n2; “normal-house” of, 119, 135–36; pastas-type houses at, 36; Quartier de l’Inopos at, 121, 126; Quartier de Skardhana at, 130, 132, 137–38n1; Quartier du Stade at, 121, 123–24, 127, 133 ; Sanctuary of Apollo at, 120; slaves of, 142, 143 ; social strata of, 132–36; tabernae in, 121, 142. See also Delian modest housing units Delos, Quartier du théâtre: external staircase in, 138n10; ground-plans of, 121, 124 ; tabernae and mixed-complexes in, 122, 124–26, 130–31 Demetrius of Skepsis, 49, 50 demographic changes: housing types as reXective of, 48–49; wartime, 144, 146, 155. See also synoikismos Demosthenes, 69, 141 developmental cycle, house as reXective of, 20 Dicaearchus (Pseudo-), 141 Diogenes, 140, 155 Diogenes Laertius, 70 domestic mode of production (concept), 27–28. See also household industry doorways, external: of House AII.5/6 (Leukas), 104, 105, 106, 112, 114, 116n11; of katagogion, 157n25; multiple, of houses in Attic deme centers, 85, 86, 90, 92–93, 95; role in household/ community interactions, 84, 85; symbolic aspect of, 29; of tabernae and mixed-complexes, 122, 123, 135 doorways, internal: aligned, in Attic deme centers, 92–93; circulation patterns indicated by, 24, 25, 26; of House AII.5/6 (Leukas), 113, 115; of tabernae, 122 drainage, 76, 123, 125. See also water channels Dreros (Crete): cache of seals at, 30; ground-plans at, 16, 18; location of, 15; room arrangement at, 20 Drerup, H., 14 Driessen, J., 156n6 Dyme, cult activity and architecture at, 48 Early Iron Age houses: arrangement of, 19; metalworking in, 27; room arrangement and functions in, 20, 29, 30; size of, 26–27; typology of, 14

184

Index

eating and drinking vessels: at House AII.6 (Leukas), 106, 107, 108 ; household industry at Sardis and, 61, 63–65 economy: house as reXective of, 20, 27, 166–70; of product exchange, 28, 166 Emporio: ground-plans at, 16 ; house types at, 14; location of, 15 ; room functions at, 30; size of rooms at, 26 Epidauros: katagogion at, 151, 152; location of, 2 Epirus: Herdraum house of, 162, 165 equality, spatial reXection of, 3–4 Eretria: ground-plans at, 16 ; house types at, 14, 113, 115, 162; location of, 2, 15 ; oikos at, 113 ergasteria. See shops and workshops exedra, appearance of, 116 eyelets, bone, 70–71 family: anoikismos and, 48, 167; as essential to oikos, 147 family cycle: in Attic deme centers, 89; house as reXective of, 18–19, 20, 32; variable patterns of, 13 Fatmann-Rey, G., 38 Wgurines, 150, 156n17 Wnds. See artifacts/Wnds Wne wares, 106, 107, 108, 110, 112, 149–50 Finley, M. I., 95 Wre, evidence of, 45 Xoor surfaces: agricultural processing and, 57–58; of Attic deme centers, 88; of Gateway House (Ilion), 44; of House AII.6 (Leukas), 106, 107, 110, 111, 113; mosaic, 88; of Quarry House (Ilion), 45; room function indicated by, 30; of tabernae-complexes, 125 Gassner, V., 138–39n11 gender, spatial organization and, 5, 27, 49, 162 Gentinos, relocation and, 46 Geometric houses: arrangement of, 19; distribution of apsidal (or oval), 17 ; room access diagrams for, 24, 25, 26 German Institute, 149 glassworking, 62–63, 165, 168 Glykeia, relocation and, 46 grafWti, 142, 168 Graham, A. J., 156n12 grindstones/millstones: household

industry and, 57, 59, 61, 62, 63, 166; room function indicated by, 109, 111, 126 gunaikon, features of, 5 Haagsma, M. J., 116n2 Halai Aixonides. See Ano Voula Halieis: industrial and domestic activities in, 166, 167; location of, 2 ; smallest house in, 155n3; squatting in, 146 Hamaxitos: location of, 38 ; relocation and, 46 hearths: of Athenian Classical houses, 76; of Gateway House (Ilion), 44; of House AII.6 (Leukas), 107, 110; household industry and, 61, 63; room function indicated by, 30. See also Herdraum (hearth-room) Hellenistic Greece: homestead farms in, 46; map of, 2 Hellenistic houses: Classical compared with, 119; at Ilion, 39, 41, 42, 43 ; at Leukas, 100, 102. See also Delian modest housing units Herculaneum, Delos compared with, 169 Herdraum (hearth-room): deWnition of, 7–8n6; description of, 3; Leukas house compared with, 165; social space in, 162 Hoepfner, W.: on Ano Voula houses, 90, 92, 93; interpretive work of, 3–4; on katagogion, 157n27; on Leonidaion, 157n23 homelessness and homeless population: identifying dwellings of, 140, 144, 147; prostitutes as, 147–50; semantics of, 141; sources of, 144; travelers as, 150–55; varieties of, 147, 164 Homer, 49 household: abandonment scenarios of, 99; composition of, 12, 59; of Delian modest living units, 132–36, 169–70; economic role of, 20, 27, 166–70; identiWcation of, 12–13; living and working space separated in, 77; settlements as interdependent with, 32; size of, 63–64; support strategies of, 4–5; variability of, 6–7, 163–64. See also family; oikos ; spatial organization household industry: assumptions about,

Index 54, 59, 65; in Athenian Classical houses, 70–79, 85, 166, 167; in brothels, 150, 156n13; deWnition of, 55; domestic activities integrated with, 59, 62, 63, 69–70, 78–79, 85, 165–68; in Olynthus, 55–60, 78, 79, 85, 166–67; in Sardis, 60–65 houses: apsidal (or oval), 4, 14, 17; arrangement of, 19; deWnition of, 64; destruction of, 156n11; dismissal of prostas/pastas distinction for, 171; public buildings separated from, 18; rectangular, 14, 15 ; short- vs. long-lived features of, 164–65; single-entrance, courtyard type of, 80n7, 84; size of, 142–44; as social space, 160; symbolic dimensions of, 5, 29–30, 162, 164. See also Archaic houses; Athenian Classical houses; Attic deme centers; Classical houses; Delian modest housing units; Early Iron Age houses; Geometric houses; Hellenistic houses hydria, 63 Ilion: archaeological research at, 36–37; competition of, 49–50; Gateway House at, 40, 42, 44, 49; implications of, 170–71; location of, 38 ; pastas-type houses at, 40 ; physical evidence at, 37–45; as polis, 46; Portico House at, 40, 42, 47, 48, 49, 50n11; Quarry House at, 40, 42, 44–45, 49; Sanctuary of Athena Ilias at, 39, 41, 46–47, 51n31; settlement of, 37; social context of, 45–50; synoikismos of, 46–47 industrial district/quarter: of Athens, 70, 73, 76, 77, 78; of Thorikos, 87, 142. See also household industry iron objects, 70–71, 80n5, 110, 113 ironworking, 63 Isocrates, 78 Isthmia: industrial and domestic activities in, 166; lead strips at, 75; location of, 2 ; oil presses at, 78 Jameson, M., 162 jewelry, 150 Juvenal, 140 Kallipolis: Classical houses at, 101; location of, 2

185

Kassope: abandonment of, 100; Classical house at, 101, 104, 114; continuity of traditional architecture at, 116; household industry at, 78, 79; katagogion at, 152, 154, 155; location of, 2 Kastanas: location of, 15; multiroom structures at, 164; room arrangement of houses at, 22, 23, 24 katagogion (hostel): description of, 150, 152, 155; ground-plans of, 151, 153, 154 ; synoikiai distinguished from, 144 kataskaphê, 156n11 Kebren: location of, 38 ; relocation and, 46; settlement of, 37 Kenchreai, relocation and, 46 Keuls, E. C., 149, 162 Kiderlen, M., 156n15 kilns, 78 kitchens, 107, 110, 114. See also cooking wares klinai (couches), 5, 31, 101 Knigge, U., 157n18 koinon, housing shaped by, 50, 171; Sanctuary of Athena Ilias as center of, 46; security concerns and, 36 Kolonai: location of, 38; relocation and, 46 Komboti: continuity of traditional architecture at, 116; location of, 2 ; room functions at, 114 Kopanaki: ground-plans at, 16, 18; location of, 15 kopron (waste disposal pit), 168 Koukounaries: arrangement of houses at, 19; ground-plans at, 16, 18; location of, 15 ; steatite vases at, 30 Kraynak, L. H., 157nn22, 25, 27 Larisa: location of, 38 ; relocation and, 46 Late Geometric period: room arrangement in, 22, 23, 24 latrines, 104, 112, 115, 123 lead strips, 74–75 leatherworking, 74 Lebedos, loss of civic identity at, 48 Leonidas of Naxos, 152 Leukas: abandonment of, 100, 106, 116; history of, 99–100; location of, 2 ; street system and houses of, 100–101 Leukas, House AII.5/6: abandonment of, 99; analysis of, 107–13; ground-plan of, 101, 104, 105, 106, 114 ; room functions in, 106; summary of, 113–16, 165

186

Index

liberti, living situation of, 134–35 Limenas/Thasos: architectural adaptations at, 31; ground-plans at, 16, 18; location of, 15 ; room arrangement at, 20 Livy, on Leukas, 100 loomweights: in brothels, 150, 156n13; household industry and, 58, 59, 62, 63, 76; room function indicated by, 30, 110, 111 Lucian, 140, 147 Lydia, household industry in, 65 Lynch, K., 77 Lysimachos, 39, 46, 49 macellum (Roman), 157n27 Malousios of Gargara, 46 marble chips and dust, 74–76 market exchange, 28, 166 Markopoulo, Aphrodite worship at, 157n21 Maroneia: house types at, 115; location of, 2 masonry/sculpting: as household industry, 56, 58; workshop for, 74–76, 80n5. See also stone quarrying and dressing Megala Peuka, slave quarters at, 142 megale oikia, 156n15 “megaron” houses, 144 metalworking, 27, 76 methodology: activity-zone analysis, 31–32; anthropological models and, 162–63. See also archaeology; architecture; artifacts/Wnds; textual evidence metics, 77–78, 168 Middle Bronze Age, 143 middle range theory, 163 Milbank, T., 79 Miletus: house types at, 14, 18; location of, 2, 15 ; number of rooms at, 22; room arrangement at, 20, 24, 164 mill, 126. See also grindstones/millstones mining and ore processing, 85, 89, 90, 141–42 Morgantina: architectural changes at, 115–16; household industry at, 78–79; House of the Mended Pithos at, 79 Morris, I., 141 mud-brick construction, 28 Mycenae: household industry at, 166; location of, 2 nails, 70–71

Neandreia: Classical remains at, 37–39; location of, 38 ; pastas-type house at, 40, 47, 48, 170; relocation and, 46, 48; security concerns of, 45 needles, 74, 110, 111 Nemea: katagogion at, 152, 153 ; lead strips at, 75 New Halos (Nea Halos): continuity of traditional architecture at, 116; location of, 2 ; room functions at, 114 Nikopolis, founding of, 100 Nowicki, K., 156n6 obsidian bladelets, 74, 80n4 Octavian, 100 oecus maior, appearance of, 115–16 oikos, deWnition of, 147, 163–66; determinants of, 49, 113–14; of House AII.5/6 (Leukas), 107, 110, 113–15, 165; at Kassope, 101, 104; as multifunctional, 30, 138n4; of tabernae-complexes, 125; variability of, 172–73. See also household olive oil processing, 57, 78, 166 Olympia: lead strips at, 75; Leonidaion at, 150, 151, 152, 157n23; location of, 2 ; lodgings at, 150, 151, 152 Olynthus: anoikism for, 48; destruction of, 54; excavation of, 1, 3, 33n3; industrial and domestic activities in, 55–60, 78, 79, 85, 166–67; location of, 2 ; oikos at, 113–14; pastas-type houses at, 7–8n6, 36, 166; spatial zoning in, 8n13; squatting in, 146; storeroom in, 167; subdivided dwelling in, 144, 145 Onythe: ground-plans at, 16, 18; location of, 15 Orraon: continuity of traditional architecture at, 116; location of, 2 ; room functions at, 114 Osborne, R., 95 pastas-type houses: cultural milieu of, 38; at Delos, 36; description of, 3; at Olynthus, 7–8n6, 36, 166; in Troad, 39, 40, 47, 48, 170; use of term, 7–8n6, 161, 171 Patras, cult activity and architecture at, 48 patrilocal residence, 18 patronus role, 135 Peace of Apamea, 49 Pecírka, J., 46 Peloponnesian War, 1, 146

Index Pergamon, housing types at, 49 peristyle-type houses: absence of, 88; at Delos, 136; description of, 3; hostel as, 150, 152, 155; lack of evidence for, 69, 80n2; use of term, 7–8n6 Petrakos, B., 93 Pheidias, 75 Philip II of Macedon, 54, 166 Piraeus, oikos at, 113 Pirson, F., 121–22, 135, 138–39n11 pithoi fragments, 30, 57, 63, 110–11 plague, 146 plain wares, 106, 107, 108, 110–11, 112 Plassart, A., 138n5 Plutarch, 146 polis, 5, 46, 50, 171 Pompeii: domestic assemblages of, 99; external staircases in, 138n10; Insula Arriana Polliana at, 134; tabernaecomplexes in, 120–22, 129–30, 137, 138–39n11, 169–70 poor, 140–44. See also homelessness and homeless population porches, 42, 161, 168. See also pastas-type houses; peristyle-type houses; prostas-type houses porneia (brothels), 147–50, 155 pottery: room function indicated by, 106–13. See also cooking wares; eating and drinking vessels; Wne wares Priene: location of, 2 ; prostas-type houses at, 7–8n6, 36 privacy, 26, 92–93, 95. See also circulation patterns prostas-type houses: at Abdera, 36; cultural milieu of, 38; description of, 3; in Troad, 47–48; use of term, 7–8n6, 161, 171 prostitutes, 147–50, 156n14, 157nn20–21. See also porneia (brothels) prothyron-type porch, 156n5, 168 public spaces and buildings: architectural adaptations from, 31, 162; as dining rooms, 168; private buildings separated from, 18, 164; production in, 27–28; tabernae-complexes and, 129–32 Rauh, N. K., 134, 155n2 regional patterns: of housing changes, 31, 164–65; importance of, 163; of internal domestic spatial organization, 22, 24 representative sphere: house as reXective of, 30–31

187

Rhamnous: circular tower at, 93, 94 ; location of, 2 Robinson, D. M., 33n3, 99 Rome: archaeology of, 146; prostitutes of, 156n14 rooms: access diagrams for, 24, 25, 26; arrangement of, 17–18, 20–27, 164; as multifunctional, 30, 138n4, 161–62; number of, 29, 30; size of, 26–27, 85, 88; visibility between, 29 Rosivach, V. J., 156n12 rural settlements. See Attic deme centers Sahlins, M., 27 sanctuaries and cult sites, 150, 152, 157n24, 157n26. See also temples Sardis: household industry at, 61–65, 165–66, 168; houses destroyed in, 60; location of, 2 Scheidel, W., 167 Schrader, H., 99 Schwandner, E.-L., 3–4, 157n27 second storey. See upper storey security concerns, 36–37, 45–47. See also city walls self-sufWciency, assumption of, 4, 166 settlements: abandonment of, 99, 100, 106, 116; arrangement of houses within, 18–19; change in, over time, 172–73; household industry locations in, 54, 59, 65; households as interdependent with, 32; plan and history of, 164; as setting for daily life, 22; single-phase and multiphase types of, 18. See also houses; rooms shops and workshops: as buffer to domestic space, 77; identiWcation of, 144; as separate from domestic space, 79; size of, 69–70; spatial organization and, 4; status of, 166; street-side locations of, 58, 59, 78, 168–69. See also household industry; taberna complexes Sigeion: location of, 38 ; relocation and, 46 silver mining and processing, 85, 89, 90, 141–42 Simon, 70 Skamandreia, relocation and, 46 skenai (tents), 144, 146, 155, 157n26 Skepsis: location of, 38 ; refounding of, 48; relocation and, 46, 49 slaves: dwellings of, 141–42, 143; household industry and, 59–60, 65, 69, 167; as household members, 147; living

188

Index

situation of Delian, 134–35; in mining and ore processing, 85, 141–42 slingbullets, 58 smithy, 76, 79 Smyrna: house types at, 14; location of, 15 ; room functions in, 30 social structure: alternative patterns of, 84, 85, 96; houses as reXective of, 20–27; reconstruction of, 13; tabernae-complexes and, 129–30, 132–37; variable patterns of, 13. See also spatial organization sociopsychological sphere: house as reXective of, 28–29 Socrates, 60, 70 Solon, 156n12 Soureza valley, slave quarters at, 142 spatial organization of houses: activities distinguished in, 4; context-speciWc model of, 163–64; controlling access in, 85, 88–89, 171; gender and, 5, 27, 49, 162; patterns of, 5–6; regional patterns of, 22, 24; socioeconomic status reXected in, 169–70. See also circulation patterns; privacy; social structure spatial zoning, 8n13 spindle whorls, 76, 110, 111 squatter activity, 146 staircases: internal or external, 138n10; of tabernae-complexes, 123, 124, 125, 128, 138n5 statue, female, 166 stemmed dishes, 61, 64, 65 stone quarrying and dressing: debris of, 74–75; iron tools for, 80n5; technical developments in, 28; in Troad, 37. See also masonry/sculpting storage of goods and produce: in Attic deme centers, 88, 96; changes in, 27; of House AII.6 (Leukas), 110, 114; household industry and, 58; pits for, 30, 42; spatial organization and, 4, 167 Strabo: on Alexandreia Troas, 46, 51n22; on Ilion, 49; on Nikopolis, 100; on weaving and sex, 157nn20–21 streets: at Leukas, 100–101; shops on, 58, 59, 78, 168–69; tabernae-complexes on, 129–32, 137 symbolic sphere: circular tower and, 96; house as reXective of, 5, 29–30,

162, 164; tabernae-complexes and, 129–30 symposia, 5, 69, 77–78. See also andron synoikia, 79, 144, 155. See also porneia (brothels) synoikismos, 36, 37, 48–49, 50 taberna complexes: architectural and urban context of, 129–32; range and character of, 120–28, 142; social strata represented in, 132–37; summary of, 169–70; use of term, 138n2 technological sphere: house as reXective of, 27–28 temples, 18, 30, 31. See also sanctuaries and cult sites tents (skenai), 144, 146, 155, 157n26 textual evidence: archaeological juxtaposed to, 1, 3–7, 160–63, 173; contradictions of, 166; as resource, 162–63, 170–71 thalamos, 112, 115 Thasos: stele from, 156n12. See also Limenas/Thasos Thessaloniki Toumba: room arrangement at, 22 thetes, 141 Thompson, D. B., 70–71, 74 Thompson, F. H., 142 Thorikos: Ano Voula compared with, 93, 95; description of, 84–85, 88–89; ground-plans at, 16, 86 ; Industrial Quarter of, 87, 142; location of, 2 ; mining slaves of, 85, 141–42; size of rooms at, 26; spatial organization at, 85, 88–89, 171; squatting in, 146 thresholds, 29, 122 Thucydides, 144, 150 Thymbra, relocation and, 46 tile roofs: standardized tiles for, 28; in Troad, 37, 42, 49, 51n31 towers: at Ano Voula, 90, 91, 92, 93; function of, 96, 142, 162, 171; at Rhamnous, 93, 94 ; at Thorikos, 86, 87, 89 trade port (Delos), 133–34, 136–37 transitional space, 26, 27, 29. See also courtyards transverse space, use of term, 161 travelers: lodgings for, 151, 153, 154 ; varieties of, 150, 152, 155

Index Troad (Dardanelles): competition among cities of, 49–50; map of, 38 ; military and political context of, 36–37, 45–47, 170–71. See also Ilion Troy. See Ilion Tsikalario: ground-plans at, 18; location of, 15 ; room functions in, 30 un-free. See slaves upper storey: development of, 28; of katagogion, 152, 155; of subdivided house, 144; of tabernae-complexes, 120, 121, 122, 123, 128 variability: of domestic organization, 6–7, 163–64; of family cycle, 13; of porches, 161; studying patterns of, 172–73 vases, steatite, 30 vestibules: of Athenian Classical houses, 69, 76, 77, 168; of House AII.5/6 (Leukas), 113; of tabernae-complexes (Delos), 124, 125, 126 Vitruvius (Marcus Vitruvius Pollio), 161 voluntary associations, 134 Vroulia: arrangement of houses at, 19; ground-plans at, 16 ; location of, 15 ; room access pattern at, 24; room functions in, 30 Walker, S., 162 Wallace-Hadrill, A., 134

189

warchitecture (concept), 156n6 water channels, 44, 45 weaving, 56, 58. See also loomweights Weickert, C., 33n4 Weigand, Th., 99 wells and cisterns: of Athenian Classical houses, 69, 70, 74, 75; of brothels, 150; of Gateway House (Ilion), 44; of House AII.5/6 (Leukas), 111; of Portico House (Ilion), 42, 50n11; of Quarry House (Ilion), 45; of tabernaecomplexes (Delos), 124–25, 126 window and window pillar, 107, 109, 110, 113 women in agricultural workforce, 167 work, use of term, 138n7 Xenophon, 60, 112 Xerxes, 36, 45 Xobourgo: ground-plans at, 18; location of, 15 Young, R., 76 Zagora: arrangement of houses at, 19; courtyards at, 20, 22; excavation at, 3; ground-plans of, 21 ; location of, 15 ; room access pattern at, 24; room arrangement at, 20, 164

Acknowledgments

This volume results from a colloquium organized for the 102nd General Meeting of the Archaeological Institute of America, held in San Diego, California, on January 5th, 2001. Originally entitled Households at the Margins of Greek Society, all of the papers presented in San Diego are included here in more fully developed form. In addition, two further contributions were solicited, those of William Aylward and Monika Trümper, which serve to broaden the already wide ranging scope of the project. Recognizing that variability rather than marginality was a more appropriate overarching theme, we revised the title of the book to Ancient Greek Houses and Households: Chronological, Geographical, and Social Diversity. Especial thanks are due to the following individuals and organizations. James. R. McCaw worked extensively on many of the drawings appearing here (especially those in Chapters 2, 6, and 9, and Maps 1.1 and 2.1). Margie Towery was responsible for compiling the index. The Department of Classical Studies at the Universiy of Michigan, Ann Arbor, generously provided funds for the indexing. To the Samuel Kress Foundation, for making travel funds available for Manuel Fiedler and Franziska Lang, and the Research Committee of the Faculty of Arts at the Open University for Lisa Nevett, to attend the initial conference in San Diego. To the Archaeological Institute of America for accepting and hosting our colloquium. At the University at Buffalo, Stephen L. Dyson, Park Professor of Classics, encouraged and aided our road to publication in a variety of ways, and the Dean’s Fund, of the College of Arts and Sciences, helped to provide subvention for the cost of illustrations. At the University of Pennsylvania Press, Walda Metcalf originally approached us with the offer of publication; and Eric Halpern, Director of the Press, and Alison Anderson, Managing Editor, have shown us unXagging support in seeing it through. Finally, we are grateful to two anonymous reviewers of the manuscript for their helpful suggestions. Finally, our collective debt to Mike Jameson, who, sadly, passed away on August 18, 2004, is acknowledged through the dedication of this work to him and his memory.