Household and City Organization at Olynthus [Illustrated] 0300084951, 9780300084955, 9780300133004

Citation preview

Household and City Organization at Olynthus

Household and City Organization at Olynthus nicholas

Yale University Press

cahill

New Haven and London

Published with assistance from the foundation established in memory of Philip Hamilton McMillan of the Class of 1894, Yale College. Copyright © 2002 by Yale University. All rights reserved. This book may not be reproduced, in whole or in part, including illustrations, in any form (beyond that copying permitted by Sections 107 and 108 of the U.S. Copyright Law and except by reviewers for the public press), without written permission from the publishers. Designed by Charles Ellertson Set in Quadraat and Wilson types by Tseng Information Systems, Inc. Printed in the United States of America by Edwards Brothers, Inc. Library of Congress Cataloging-in-Publication Data Cahill, Nicholas. Household and city organization at Olynthus / Nicholas Cahill. p. cm. Includes bibliographical references and index. isbn 0-300-08495-1 (alk. paper) 1. Olynthus (Extinct city)—Buildings, structures, etc. 2. City planning—Greece—Olynthus (Extinct city)

3. Dwellings—

Greece—Olynthus (Extinct city) df261.o53 c3

2001

938'.1—dc21

2001026769

A catalogue record for this book is available from the British Library. The paper in this book meets the guidelines for permanence and durability of the Committee on Production Guidelines for Book Longevity of the Council on Library Resources. 10 9 8 7 6 5 4 3 2 1

Contents

Preface vii Acknowledgments xi chapter 1 Greek City Planning in Theory and Practice 1 chapter 2 History and Archaeology at Olynthus 23 chapter 3 The Houses Described 74 chapter 4 The Houses Organized 148 chapter 5 The Organization of Blocks 194 chapter 6 The Economies of Olynthus 223 appendix 1

Cluster Analysis of Room Areas, Five-Cluster Solution 289

appendix 2

Sales Inscriptions from Olynthus 293 Notes 301 Bibliography 343 Illustration Credits 370 General Index 371 Index of Houses and Buildings, Blocks, Trenches, and Streets 377 Index of Artifacts 381

Preface

πόλις ἐν σκοπέλῳ κατὰ κόσμον οἰκεῦσα σμικρὴ κρέσσων Νίνου ἀφραινούσης A polis on a barren rock, small, but settled in an orderly fashion, is greater than senseless Nineveh. (Phokylides of Miletus, 6th c. b.c., fr. 4)

The concept of order was central to the creation of a Greek city. A wellordered state would endure; a poorly ordered one would fall into stasis and disintegrate. This order encompassed both the social organization of the state—its laws, government, tribal structure, and other aspects—and its physical structure; and its physical organization was closely linked to its social order. Planning a city, then, was not simply a matter of finding a suitable site, laying out blocks, establishing the trace of the city walls, deciding on the sites of the agora, the temples, and the other public buildings. It was the manifestation of an ideal, a model of the community and of the world translated into physical form. It was the realization of an abstract view of civic space. To Aristotle, as to any Greek, ‘‘every polis is composed of oikoi.’’ 1 The oikos, both in its social meaning as ‘‘household’’ and its physical manifestation as ‘‘house,’’ was the basic building block of the polis, in both its social meaning as ‘‘community’’ and its physical meaning as ‘‘city’’ (or ‘‘city-state’’). The study of the Greek city begins, therefore, with the study of the Greek house. This book considers some of the relationships between house and city, between household and community, as they were worked out in practice at Olynthus in northern Greece (fig. 1). This polis was occupied for a short period of time, for eighty-four years at the most. It was then violently destroyed, leaving tens of thousands of artifacts on the final floors of its houses, and for the most part never reoccupied. A large part of the city was excavated between 1928 and 1938 by David M. Robinson, who published his findings in fourteen massive volumes.2 vii

[To view this image, refer to the print version of this title.]

figure 1. Map of Greece Its unique history, extensive excavation (which uncovered more than a hundred houses), and full publication make Olynthus the best-documented site for the study of household and urban organization in Classical Greece. Only at Olynthus can we study the remains of a planned city occupied for less than three generations, and so relatively unmodified by later rebuilding, and consider not only the architecture of houses but their contents as well, with well-preserved assemblages on the final destruction floors. We can investigate not only how the houses and city were planned and built, but how space was actually used; we can reconstruct the intended organization of civic and domestic space, and how that organization was realized in practice. We have unique evidence for the layout and use of domestic space; for the occupations and aspirations of the households; for the domestic and urban economies and how they articulate with one another. We can viii

Preface

consider not only the typical house, but the range of variation among contemporary houses and their contents: variation which is related to differences in origin, status, family ties, occupation, economic strategies, and the like. We can analyze neighborhood and regional planning in the city, consider its house blocks as not only physical units of civic organization but social units as well, and evaluate larger regional patterns in the city. We can compare the ideologies of Greek household organization with how houses were actually constructed, examining what sorts of spaces were built and how they were intended to be used, and then how those spaces were actually used. In short, the archaeology of Olynthus offers a fuller and richer picture of Greek domestic and civic life than almost any other Greek site.

Preface

ix

Acknowledgments

One of the great pleasures of this study has been working with so many distinguished and helpful scholars, who have shown just what a cooperative venture archaeology is. My first thanks go to the members of the Olynthus team who recalled to me their experiences and memories of working at the site. Paul Clement, Walter Graham, William A. MacDonald, George Mylonas, John Travlos, and Gladys Weinberg answered my innumerable questions about the excavation, stratigraphy, artifacts, and other matters, and brought to life an excavation which I would otherwise know only from terse records and black-and-white photographs. To them I am indebted not only for their help with Olynthus, but for sharing their recollections of a life in archaeology half a century ago and giving me a deeper appreciation for the history of my chosen field. Lucy Turnbull, then-director of the museum at the University of Mississippi, kindly allowed me to study and copy the fieldbooks, plans, photographs, and other unpublished records stored in Robinson’s archives. Without the records she so generously provided, much of this study would have been impossible. Julia Vokotopoulou kindly discussed her own excavations at the site and allowed me to work with some of the finds from Olynthus in the Thessalonica Museum. Many scholars have contributed to this book, from its early stages as a PhD dissertation through the present publication. J. K. Anderson, Bradley Ault, John Camp, Wolfram Hoepfner, Michael Jameson, John Kroll, Rob Loomis, Carol Mattusch, Stephen Miller, Ian Morris, Naomi Norman, Mark Rose, Susan Rotroff, Curtis Runnels, David Stronach, Ronald Stroud, Ruth Tringham, Barbara Tsakirgis, and Charles Williams II offered advice, comments, and criticisms at different stages, and I have learned and profited greatly from their expertise. In developing the Internet site for the project, Ross Scaife, Neel Smith, Anne Mahoney, and Robert Chavez were of invaluable assistance and inspiration. Naturally, errors remain mine alone. Crawford H. Greenewalt Jr. is owed a special debt of thanks for his continual support and comments, and for setting such high standards of scholarly integrity, erudition, and for his open-minded approach to archaeological questions. xi

The staff of Yale University Press, particularly Harry Haskell, and Eliza Childs have been continually helpful in rounding off the many rough edges and correcting the numerous inconsistencies and infelicities of the text. The inconsistent spellings of Greek names and terms is such a familiar issue to most readers that it hardly requires explanation or apology. In general, I have tried to use the more familiar Latin spellings of more familiar names—hence ‘‘Olynthus’’ rather than ‘‘Olynthos’’—but retained the Greek versions of less familiar words. Except where otherwise noted, all texts and translations from Greek and Latin are from the Loeb Classical Library.

xii

Acknowledgments

Household and City Organization at Olynthus

chap ter one

Greek City Planning in Theory and Practice

ἔνθεν ἀναστήσας ἄγε Ναυσίθοος θεοειδής, εἷσεν δὲ Σχερίῃ ἑκὰς ἀνδρῶν ἀλφηστάων, ἀμφὶ δὲ τεῖχος ἔλασσε πόλει καὶ ἐδείματο οἴκους καὶ νηοὺς ποίησε θεῶν καὶ ἐδάσσατ’ ἀρούρας. From here godlike Nausithoös had removed [the Phaiakians] and led a migration, and settled in Scheria, far away from men who eat bread, and driven a wall about the city, and built the houses, and made the temples of the gods, and allotted the holdings. (Homer, Odyssey 6.7–10, trans. Lattimore)

In this earliest reference to Greek colonization, the basic elements of founding a new polis are already in place: the uninhabited land, the construction of fortifications and the temples of the gods, the division and allotment of agricultural land, and the building of houses, presumably on lots assigned to the colonists like the farmland. In its essence, the process remained basically the same for a thousand years. The goal of this chapter is not to provide a history of Greek city planning.1 Rather, it will consider a few literary accounts and historical cities which develop issues concerning the relations between polis and household, issues which relate to the understanding of the archaeological remains at Olynthus. These issues include the composite nature of the citizen body in new cities and the consequent need to unify a diverse population, the correspondence between physical organization and social structure, and the importance of the distribution of land as a mechanism for achieving unity and order. A variety of situations could lead to the creation of a new Greek city. Colonization and clerouchies far from the homeland were frequent and familiar phe1

nomena. Villages, small towns, and cities were joined in synoikisms (literally ‘‘a dwelling together’’) to form a single larger city. Cities were destroyed and rebuilt, or moved and refounded on a new site, or expanded onto a new terrain. According to one estimate, by the fourth century b.c. one-third of all Greeks who lived in cities, lived in cities which had been newly founded since the Geometric period.2 This continuous founding and rebuilding of cities gave the Greeks ceaseless opportunities to develop theories about the organization of the ideal state, and left traces of those theories in philosophical writings, in the accounts of historians, in inscriptions, and in the remains of the cities themselves. The processes of colonization, synoikism, and refoundation, repeated over the centuries, were important motivating factors in the development of the polis as a characteristically Greek institution. The problems of organizing a new settlement—of drafting laws, designing a city, building its walls, dividing the land, distributing temples to the gods and houses to the citizens—had to be worked out over and over, and the process of solving these problems gave the planners and the inhabitants of these new communities fresh insights into the nature of human society. By the sixth century b.c. and probably earlier, theorists and oikists had begun to design ideal states based on principles of cosmology and natural history, and to apply those ideas to the foundation of actual cities. They began to recognize explicitly and study the dual aspect of the polis, as concrete city and as social community, and to redefine the relation between those twin aspects to create a society whose social and spatial organizations were correspondent and reflected higher moral and philosophical ideals. In a general way, such redefinition and restructuring occurs in any city foundation. But these processes are particularly important in highly planned and organized communities like the Greek colonies where new laws, new constitutions, new systems of land tenure, and the like were drawn up at the outset, and the city and citizen body laid out and organized taking future expansion into consideration. We know little about the earlier civic theorists. We know some names, but none of their written works survive save Plato’s and Aristotle’s. Aristotle’s Politics, the most complete surviving discussion of ideal states, mentions a few notables: Hippodamus of Miletus, Phaleas of Chalcedon, and Plato; and these are all rather late in the development of Greek urban planning. We can draw some conclusions from extant city plans and from surviving laws and constitutions, but while this evidence suggests that the planners of these early communities were concerned with many of the same problems as later theorists, and seem to have come to sophisticated conclusions about the planning and organization of towns, it does not for the most part help us to reconstruct their thoughts. Their theories were nevertheless familiar to a contemporary audience: Aristophanes could draw a laugh from his Athenian audience by parodying these thinkers in the Birds.3 The reforms of early lawmakers, such as Cleisthenes in Athens and Aletes in Corinth, also demonstrate their concern that the physical organization of the polis reflect its social 2

Greek City Planning

organization (and vice versa) and help us identify different means of achieving such correspondence.

h i p p oda m u s of mi l e t u s The most famous Greek urban theorist, and the earliest of whom we have any real knowledge, was Hippodamus of Miletus. If he wrote treatises they are long gone, and we know of his personality and thoughts primarily from Aristotle’s brief description in the Politics.4 Although (wrongly) famous as the inventor of the gridplanned city and discussed today mainly as an architect and city planner, Aristotle tells us that Hippodamus ‘‘wished to be a man of learning in natural science generally, [and] was the first man not engaged in politics who attempted to speak on the subject of the best form of constitution’’ (Politics 1267b). Hesychius and Photius, both late lexicographers, describe Hippodamus as a meteorologos, a natural philosopher. Hippodamus thus seems to have come to city planning from the theoretical, rather than the practical side of things; he was concerned not solely with the physical layout of cities but in the ordering of an ideal society, and he designed his ideal city to accommodate such a community.5 But he was also involved in the planning of a number of historical cities: the Piraeus is securely attributed to his hand, and Thurii and Rhodes are also associated with him.6 Thus although we know few details of his utopia, we can fill out some of his thoughts, at least about its physical appearance, from its realization at those sites. Aristotle tells us that Hippodamus ‘‘discovered the division of poleis’’ (τὴν τῶν πόλεων διαίρεσιν εὗρε). This could refer to the physical planning of the city—not the invention of the grid plan, which was already ancient when Hippodamus was born, but some other aspect of the city’s organization—as well as to the division of the polis as a community of citizens. It very probably refers to both, and to the correspondence between physical and social planning.7 Hippodamus organized his ideal state in a tripartite system. The polis, of 10,000 citizens, was divided into three sections based on occupation: one section of artisans, one farmers, and the third soldiers. Likewise the land was to be divided into three parts, religious, public, and private sections; the laws were organized into three classes, wanton assault, damage, and homicide; and the magistrates were to attend to three subjects, public matters, matters relating to aliens, and matters relating to orphans. Such an attention to numerology is sometimes attributed to either Hippodamus’s background in Ionian natural science or to Pythagorean influence, but it is encountered in other political and architectural works, for instance in the Laws of Plato, and could be seen as characteristic of general Greek ideas about city and social planning rather than of Ionian thought in particular.8 All citizens of Hippodamus’s state were not landholders: the private section of land was owned by the farming class, while the public section of land was deGreek City Planning

3

voted to producing food for the soldiers, and the artisans presumably lived off their work.9 But although the majority of the population did not own land, all citizens were equally enfranchised: Aristotle says that the demos was composed of all three ‘‘sections’’ ( μέρη) of citizens, and formed the assembly which elected the city magistrates.10 The aim of Hippodamus here is to create a state in which, although both the citizens and the civic space are divided into specialized sections, these sections are uniform, parallel, and equivalent to one another. Thus although his state is organized along different lines from such previous states as Cleisthenes’ Athens, it is still not explicitly hierarchical in its social organization nor in its conception of space. Aristotle further tells us that he ‘‘cut up’’ (κατέτεμεν) the Piraeus, and refers to cities planned ‘‘according to the newer and according to the Hippodamian manner’’ (Pol. 1330b). Hippodamus was a practicing planner, and aspects of his thought can be understood through how it was carried into practice. Hesychius and Photius provide glosses on Ιπποδάμου νέμησις, ‘‘the nemesis of Hippodamus,’’ explaining it as ‘‘Hippodamus, son of Euryboön and a meteorologos, divided (or distributed) the Piraeus for the Athenians.’’ 11 These nemeses thus seem to be related to the ‘‘division of cities,’’ at least of the Piraeus. A group of boundary stones from the Piraeus documents these nemeses.12 The inscriptions are dated by letter forms to the fifth century, and some at least probably belong with Hippodamus’s replanning. Of particular interest are three boundary stones delimiting the nemeses of the city (ἄστυ) and of the Mounichia hill. These inscriptions read: [ἄ]ι τ[ ε῀σ]δε τ῀ες hοδ῀ο τ῀ειδε hε Μονιχίας ἐστι νέμησις, ‘‘Up to this road here is the nemesis of the Mounichia,’’ and two stones read ἄι τ῀ες hοδ῀ο τ῀εσδε τὸ ἄστυ τ῀ειδε νενέμεται, ‘‘up to this road here the City has been ‘nemesized.’ ’’ 13 The word νέμησις is derived from the verb νέμω, a rather general verb meaning ‘‘to divide, apportion, distribute,’’ and is occasionally used, by itself and in compounds, in contexts of urban planning for the division of land or citizens into smaller sections, for distribution of land to citizens, and the like.14 McCredie suggests that in this context νέμησις had a technical meaning, and translates it as ‘‘‘plan,’ ‘layout’ or even ‘grid,’ rather than simply ‘occupation’ or some such.’’ 15 The boundary stones themselves directly attest only one nemesis, the Mounichia, a hill in the eastern part of the city. This stone was found in situ just northwest of the hill. The other two stones, reading ἄι τ῀ες hοδ῀ο τ῀εσδε τὸ ἄστυ τ῀ειδε νενέμεται, were not found in situ.16 McCredie translates the inscriptions ‘‘Here, up to this street, the City has been planned,’’ suggesting that the ἄστυ, as distinct from the Mounichia and other regions, formed one nemesis.17 But the different phrasing of the two inscriptions might suggest a different meaning, and the second text might just as easily be translated ‘‘Here, up to this street, the City has been divided into nemeses.’’ The City would then not be a single nemesis but a series of nemeses, of which the Mounichia is one. Unfortunately the actual plan of the Piraeus is rather poorly known, since the 4

Greek City Planning

Sullan destruction left the city in ruins and the modern port has destroyed or buried what was left. The little that is known, primarily from nineteenth-century observations, suggests that the hills of Mounichia and Akte were laid out without a true grid plan. Since the Mounichia formed one nemesis it is possible that the Akte formed another. The flatter central part of the Piraeus, in contrast, was laid out in a regular grid, with four main streets oriented northeast-southwest, and a number of main streets (five?) oriented northwest-southeast. These seem to define a series of large parcels of land, about 250 × 275 m, which were subdivided by smaller streets into house blocks.18 This type of hierarchical divisive planning, with wide streets defining ‘‘major rectangles’’ which are then subdivided into blocks, is a method quite different from that of cities like Olynthus, where streets of equal width divide the city into blocks, without larger arteries or clearly divided sectors. Hippodamus’s most significant contribution to city planning, then, is probably this special method of division of land and territory. Although Aristotle describes only three broad functional categories of land in Hippodamus’s ideal state, his ‘‘division of cities’’ seems to be more complex, flexible, and generally applicable than a simple division of land by function. As McCredie points out, it is this aspect of Hippodamus’s planning, rather than any innovations in orthogonal street patterns, which established his position as the father of Greek city planning. At least in Aristotle’s account of Hippodamus’s thought, we find established most of the technical terminology used later in planning cities. Hippodamus’s discovery of the ‘‘division of cities’’ and his ‘‘distribution’’ of the city into nemeses set the groundwork for later planners, both of utopias, such as Plato’s Magnesia and Aristotle’s city in the Politics, and of real cities.

plato’s laws Of Plato’s three utopian works, the Republic, the Statesman and the Laws, the last is his fullest account of the design and establishment of a new city. Published in 346 b.c., a year after Plato’s death and two years after the destruction of Olynthus, the Laws is couched in terms of a plan to establish a colony—Magnesia—near the south coast of Crete. Although replete with details about the location of the city, the origins of its citizens, and so forth, and although Plato claims that this is not going to be an ideal utopia but will take into account the imperfections of human nature, the Laws is not meant to be a blueprint for an actual city, any more than the dialogue is intended as a record of a real conversation.19 Nonetheless, Plato’s prescription for the foundation of Magnesia seems to agree in many respects with what we know or would expect of the planning of real Greek cities, and just as many of his laws are based on actual Greek codes, his account of the foundation of his Cretan ‘‘almost-utopia’’ may reflect, in some respects, actual Greek practice.20 Greek City Planning

5

Plato’s city is to be established in the middle of the countryside, far enough from the sea to be safe from the pernicious influences of ports and commerce.21 It is surrounded by farmland sufficiently fertile to make the state self-sufficient, and by mountains and forests for other resources. Moreover, it will have no neighbors, for the land has been deserted for many ages; so the problems of relations with the outside world will be minimized. The setting is staged as an ideal environment for the nurture of civic virtue and for the creation of a polis based on principles of what is best for men, rather than what is necessary or expedient in the political climate of a real Greek city. A Composite Community The population of Magnesia is to be mixed, from all over Crete and the Peloponnese: a sort of joint colonization and synoikismos led by settlers from Knossos.22 These were the most common methods of assembling a citizen body: from groups emigrating from their homeland as colonists to found or join another city, or by whole communities moving and joining together to form a larger polis. Plato discusses the advantages and disadvantages of such a composite community, made up of citizens from different origins, different γένη: It would not be equally easy for States to conduct settlements in other cases as in those when, like a swarm of bees, a single clan goes out from a single country and settles, as a friend coming from friends, being either squeezed out by lack of room or forced by some other such pressing need. . . . All such cases [of a single group emigrating] are in one way easier to manage, as regards settling and legislation, but in another way harder. In the case where the race is one, with the same language and laws, this unity makes for friendliness, since it shares also in sacred rites and all matters of religion; but such a body does not easily tolerate laws or polities which differ from those of the homeland. . . . On the other hand, the clan that is formed by fusion of various elements would perhaps be more ready to submit to new laws, but to cause it to share in one spirit and pant (as they say) in unison like a team of horses would be a lengthy task and most difficult. (Laws 708C–D)23 Another advantage of having diverse sources of colonists, says Plato, is that one can recruit only the best citizens and so avoid the common afflictions of crime, stasis, and other ‘‘diseases of the polis.’’ For Plato admits that laws alone are unable to cure the strife, the demands for redistribution of land, the cancellation of debts that were so feared by the landowners of Greek cities. The remedy for this dissension must lie with the landowners themselves, who must be willing to redistribute their own property and cancel debts. ‘‘For when a state is obliged to settle such strife by law, it can neither leave vested interests unaltered nor yet can it in any wise alter them, and no way is left save what one might term that of ‘pious 6

Greek City Planning

aspiration’ and cautious change, little by little, extended over a long period’’ (Laws 735C–737B). Inequalities and Civil Strife This concern about civic strife between citizens was a decisive consideration in the design of a new community, for Plato and for the planners of actual cities. In both the Republic and in the Laws, Plato attributes such strife to economic inequalities among citizens, a view shared by some other Greek urban theorists: ‘‘It is, as we assert, necessary in a state which is to avoid that greatest of plagues, which is better termed disruption [διάστασις] than dissension [στάσις], that none of its citizens should be in a condition of either painful poverty or wealth, since both these conditions produce both these results; consequently the lawgiver must now declare a limit for both these conditions’’ (Laws 744D). Theories regulating the amount of citizens’ wealth were proposed by other lawmakers. ‘‘The question of property, they say, is universally the cause of party strife,’’ writes Aristotle, who then describes the constitution of one of these theorists, Phaleas of Chalcedon (Pol. 1266a-b). This man was the first to introduce the idea of equalizing the property of citizens, by edict in a newly founded city or by regulating dowries in cities that already existed. In other states, too, ‘‘some of the laws that were enacted . . . in early times prohibited the ownership of more than a certain amount of land.’’ Aristotle criticizes Phaleas’s single-minded focus on inequality of property, saying that ‘‘equality of property among the citizens is certainly one of the factors that contribute to the avoidance of party faction; it is not however a particularly important one.’’ But in his long discussion of the causes of stasis, Aristotle himself writes, ‘‘For stasis is everywhere due to inequality . . . for generally the motive for stasis is the desire for equality.’’ 24 Aristotle’s ‘‘equality’’ is not simply economic equality, but social and political as well, and ‘‘equality’’ itself may be defined in a number of ways; but economic inequality is certainly among the motives. And from the eighth century on, planned colonies seem to have attempted to equalize the amount of land distributed to each citizen and so maintain a balance between poverty and wealth. The lawmaker may avoid the evils brought about by economic inequalities, writes Plato, through proper distribution when the community is laid out: ‘‘But as for those to whom—as to us now—God has given a new state to found, and one free as yet from internal feuds,—that those founders should excite enmity against themselves because of the distribution of land and houses [διὰ τὴν διανομήν τῆς γῆς τε καὶ οἰκήσεων] would be a piece of folly combined with depravity of which no man could be capable’’ (Laws 737B). But how to do this? ‘‘What then would be the plan of a right distribution? First, we must fix at the right total the number of citizens; next, we must agree about the distribution of them, into how many sections [ μέρη] and each of what size, Greek City Planning

7

they are to be divided; and among these sections we must distribute, as equally as we can, both the land and the houses’’ (737C). In his more ‘‘realistic’’ Laws, Plato does not attempt to absolutely equalize the wealth of all the citizens as he does in the Republic. Instead he institutes four property classes, depending on the amount of wealth each colonist brings to the new city; a citizen may however move from one class to another if he becomes richer or poorer. Through the principle of ‘‘proportionate inequality’’ (τῷ ἀνίσῳ συμμέτρῳ) citizens are given offices and honors according to their wealth and class; and this, he argues, should reduce friction among the citizens and lead to greater friendship. The property classes also play a role in the election of the council, in taxes and contributions, and in disbursements.25 The difference in wealth between the richest and poorest citizens is to be strictly limited, however. The lower limit is set at the citizen’s allotment, which is to be inalienable; while the upper limit is three, or four, or five times this amount.26 If a citizen accumulates more than this amount, the surplus is given to the city and ‘‘to the gods who have the state in their keeping.’’ Citizens are not allowed to accumulate gold, silver, or other precious metals, nor are they allowed to keep foreign currency; the state will use its own currency which has only token value.27 The Division of Citizens and Polis The number of households in Magnesia is fixed at 5,040, a number which attains an almost mystical significance in the Laws.28 In keeping with his aim of the equitable and harmonious division and distribution of land and citizens, Plato chooses this number because it is divisible by all the whole numbers from one to twelve, excepting eleven; in all it is divisible by fifty-nine divisors. In choosing this number, Plato’s aim is to allow the citizen body to be divided evenly into the various subdivisions a polis needs to function and, through this process of rational and harmonious subdivision, to achieve (or at least foster) the harmony among the citizens which will allow the state to avoid stasis.29 That very self-consistency we must now do our best to consider in conjunction with the proposed division of the state into twelve parts, inquiring in what conspicuous way the twelve parts, which in their turn admit of being divided into very many ways,—these and their immediate subdivisions, and those which spring from them, until we get down to the 5,040 individual citizens—and such divisions will give you your phratries, your demes, and your villages, and besides these, your military divisions and levies, yes, and your money-values and your measures, whether of solids, liquids, or weights —how all these, I say, are so to be fixed by law as to harmonize with and to fit in with each other (746D, text and translation England).30 8

Greek City Planning

Plato thus institutes three different principles of the subdivision of the citizen body, all of which, he believes, will enhance civic harmony: a diverse population from various states, which may be more receptive to a new law code; a wellthought-out system of subdividing the population into smaller groups, tribes, phratries, demes, and the like; and a system of property classes. These all crosscut each other, so that for instance the tribes are all of equal wealth with, presumably, members of the four property classes distributed equally among them. On the other hand, Plato rejects Hippodamus’s division of citizens by occupation. All citizens own land or at least belong to land-owning families, and all participate in the military defense of the territory, while crafts and other such activities are expressly forbidden to citizens and even the slaves of citizens, and are restricted to metics. The citizen body is thus homogeneous, not specialized, and the tribes all equivalent and interchangeable. This homogeneity is further extended to the spatial organization of the state, which thus reflects the social structure of the polis. For understanding the physical planning of this semi-utopia, and the relation between its physical and social organization, the most explicit section of the Laws is in Book 5, where Plato prescribes how the city and countryside will be divided, the allotments distributed, and the population organized (Laws 745B–747E). Although the description is of a purely hypothetical state, Plato uses a terminology which seems to reflect actual practice, and his procedure may, in some respects, follow that of genuine city planners. Although this is one of the fullest preserved descriptions of the founding of a state, his description of the physical planning of the polis is brief and sketchy: the physical realization of his system is of less interest than the design of that system and the ‘‘moral planning’’ of the citizens through proper laws. This attitude is prevalent, not only among philosophic writings but among Greek historians and in documentary sources concerning the foundations of actual cities. The lawgiver must set aside a fortified acropolis for the city gods—Hestia, Zeus, and Athena. This area is kept independent of the land which is to be divided, assigned to one purpose or another, and allotted to the citizens; it remains a separate entity within the polis but unconnected with its physical and social divisions. Like the citizen body, the city and countryside (chora) are divided up on a number of levels: After this, he [the lawmaker] must divide twelve sections, first setting aside a sacred precinct for Hestia, Zeus and Athena, calling this the acropolis and enclosing it with a ring-wall; starting from which [i.e., the acropolis] he must cut the city itself and all the country into twelve sections. It is necessary that the twelve sections be equal, by making those of good land small, and those of inferior land larger. He must divide [the land into] 5,040 kleroi, and further cut each of these in two, and join two pieces into a composite allotment, each containing a near plot and a distant plot—joining the plot Greek City Planning

9

nearest the city with that nearest the border, and the second nearest with the second furthest, and all the rest likewise. And with these double-allotments he must use the same device which we just now spoke of, about the poor land and good, making them equal through the greatness or smallness of the distribution. He must also distribute the citizens into twelve sections, making the twelve sections as equal as possible with respect to the other property [i.e., other than the allotted land], after a census of all [the citizens] has been made. And finally he must allocate twelve sections to the twelve gods as kleroi, and name the allotted section after each god and dedicate it to him, and name the tribe itself [after the god]. And then he must cut up the twelve sections of the city in the same manner as he distributed the other country, and he should distribute to each [landholder] two houses, one near the center and the other near the borders. And thus the settlement will be completed.31 The polis is first divided into twelve ‘‘sections’’ ( μέρη), city and country independently, starting from the acropolis.32 The countryside sections are made equivalent in worth, rather than equal in size, by making those of fertile land smaller and those of poor land larger. The urban core (ἄστυ) is likewise divided into twelve sections. The citizen body, too, is to be divided into twelve sections (also μέρη), again making each section equivalent in wealth. These ‘‘sections’’ of the city, the territory, and the citizen body are obviously parallel, and although in this rather elliptical passage Plato never explicitly states that one section of citizens will inhabit one section of city and territory, that seems to be his intention. Sections are appointed to each of the twelve gods as allotted holdings, kleroi, and these would seem to refer to the sections of territory; but then each is consecrated to the god and the tribe named after him; this would imply a section of citizens. The logical interpretation is that the tribe (both a φυλή and a μέρος) of citizens named after each god will inhabit the section (also a μέρος) of territory which is that god’s allotted kleros.33 Likewise, craftsmen (δημιουργοί, who are not citizens but nevertheless are necessary to the functioning of the state) are to be divided into thirteen ‘‘sections,’’ one of which will be settled in the city, and the others distributed through the country sections.34 Plato thus describes a perfect correspondence between the social and physical organization of the polis. Citizens and territory are divided into sections, which are identical in wealth, fertility, and structure. These tribal sections form semiindependent units within the state. Each section of the countryside has its own internal administration: a central village with temples, marketplace, artisans for self-sufficiency, its own fortress and defense organization, and its own courts.35 The sections are also religious units, each with its own god or daemon or hero assigned to it, with sacred land, sanctuaries, and urban and rural festivals.36 On a lower level, the agricultural territory is divided into 5,040 kleroi corre10

Greek City Planning

sponding to the 5,040 households, and each kleros divided in two: these halves are then joined, one near to the city with one far from the city, to form one composite kleros. Like the sections, these kleroi are to be made equivalent in worth rather than equal in size. The city, too, is to be divided ‘‘in the same manner’’ as the countryside into 5,040 kleroi. Each citizen thus has three parcels: one of the 5,040 city kleroi, and one of the composite countryside kleroi, which is composed of two halflots. Each citizen also receives two dwellings, ‘‘one near the center and one near the outskirts.’’ By this Plato probably means a city house and a country house, as was usual in Classical Greece, although in theory he could instead mean one house on the half-kleros near the city, the other in the half-kleros far away. Making sections of land and kleroi equivalent in worth or produce rather than equal in size was practiced in some historical Greek states. Two bronze tablets from Heraclea in Lucania record the resurveying and rental of sacred properties of Dionysus and Athena Polias, and although Dionysus has more than three times as much land as Athena—roughly 330–350 hectares as opposed to 93–98 hectares—the amount of arable land is much closer: Dionysus has 109–117 ha, Athena 83–88. This may have resulted from a deliberate attempt to equalize the holdings of the two deities by giving Dionysus a larger area of less fertile land.37 In a related manner, Cleisthenes equalized the representation in the ten tribes of the three different regions of Attica through a proportional reallotment of the demes depending on their population, an analogous process. And at Sparta, the πολιτικὴ χώρα, land belonging to the citizens rather than the περίοικοι, was divided into lots which were equal in yield rather than in extent.38 Plato is a philosopher, not an architect or an oikist, and so does not describe in detail how the sections and kleroi are to be organized in the city and the countryside: how the sections are to be laid out, how they are distinguished from one another, whether there are to be boundaries between them, how city blocks are to be designed, houses constructed, and so forth. He does lay down some of the building codes for public buildings, particularly about city walls. Like so many of his contemporaries, Plato believed that ideally ‘‘walls should be made of bronze and iron, rather than earth’’; but like any pragmatist of the fourth century, he realized that a real city would need walls.39 He therefore prescribes that if a city must have walls, then ‘‘the building of the private houses must be arranged from the start in such a way that the whole city may form a single wall; all the houses must have good walls, built regularly and in a similar style, facing the roads so that the whole city will have the form of a single house, which will render its appearance not unpleasing, besides being far and away the best plan for ensuring safety and ease of defense’’ (779B). This scheme is actually used at Olynthus, where the fortification wall formed the back walls of the houses in Rows A and A', and probably along the east side of the city as well. Plato’s Magnesia is thus a complex and highly organized community, its social Greek City Planning

11

organization mirrored in its physical layout. The twelve tribes which form basic divisions of the citizens are reflected in the twelve sections of the city and of the countryside, and in the religious festivals and other civic institutions. The 5,040 households are distributed into 5,040 kleroi, whose tripartite nature also manifests the various concerns of the citizens: as members of an urban community, as agricultural smallholders, as defenders of the territory. Both the ‘‘sections’’ and the kleroi are geographically mixed, each containing segments of city, country, and borderland, thus integrating the different regions of the state and spanning local concerns, just as Cleisthenes tried to span local concerns in his reorganization of Attica. The division of territory into villages, demes, and the like may further reflect social ties; while other types of land division, such as functional separation of public and private spaces, will reflect more general views of the relations between man, the polis, and the gods. We will see many of these equivalences repeated in other theoretical works, such as Aristotle’s Politics, and manifested in actual cities.

a r i s t o tl e ’ s p o l i t i c s Whereas Plato’s Laws develops a complex, detailed system for establishing a new society, the Politics of Aristotle is more analytic, and his thoughts about constitutions, social orders, and the design of states tend to be descriptive rather than prescriptive and utopian. As an analysis of ideal and existing states, of the causes of revolution and dissension in historic states, of the relation between oikos and polis, the Politics is probably our most important source for understanding Greek urbanism. Here, however, I only wish to consider Aristotle’s rather short section of prescriptions for the founding of the ideal state in Book 7. Aristotle makes only general recommendations about the size of the population and territory of his state. They should be neither too large nor too small; for a state that is too small cannot be self-sufficient, while one that is too large cannot be governed. ‘‘Ten people would not make a city, and with a hundred thousand it is a city no longer.’’ 40 He does not give figures as Plato or Hippodamus do, but only rules of thumb: the population should be ‘‘the greatest surveyable number required for achieving a life of self-sufficiency,’’ and the territory ‘‘should be large enough to enable its inhabitants to live a life of leisure which combines liberality with temperance.’’ 41 Aristotle bases the organization of his ideal state on his analysis of the elements necessary to its existence: food, handicrafts, arms, property, public worship, and a system of deliberation and jurisdiction. From this he defines the occupations necessary to each state: farmers, craftsmen, soldiers, a propertied class, priests, and judges. These occupations lead to the question of the division of labor

12

Greek City Planning

within the state; for the distribution of labor, says Aristotle, is the main cause of the differences between constitutions.42 Some of these occupations may be relegated to noncitizens, however, if they are considered unsuitable for the elements which are true ‘‘parts’’ of the state.43 His system is unusual, though, in relegating not only industry, trade, and commerce to noncitizens, but also farming, on the grounds that farming prevents the leisure ‘‘necessary both for growth in goodness and for the pursuit of political activities.’’ 44 Property, he argues, should be owned by citizens, to allow them that leisure; property ownership therefore will not be a separate class or occupation.45 This leaves the military, judicial, and religious functions as the proper occupations for a citizen, and Aristotle distributes these to the young, mature, and elder citizens respectively, so that each citizen performs all three functions but in different phases of his life.46 Thus Aristotle does not construct a class system based on property, as does Plato, but a system based on occupation. In addition to dividing the citizens into occupations, he divides them into smaller groups which share ‘‘common tables’’ (συσσίτια). The organization of these groups is not specified in detail, but in part at least they have a military function since the messes of some of the younger citizens will be in guard posts distributed along the walls. Similar systems of defense based on social units are prescribed by Aeneas Tacticus, and were found in Smyrna, Stratonikeia, and elsewhere.47 He does not mention civic groups such as tribes, demes, or other such divisions, but he must have taken these divisions for granted as parts of any state, which need not be specified or described in detail. Aristotle is much more conscious than Plato of the relations, often hostile, between neighboring states, and many of his prescriptions about the siting, layout, and division of states are determined by military considerations. He is also more concerned with the friendly relations between states: with the market, imports and exports, and with diplomatic relations. His state does not exist in a vacuum with no neighbors, no foreign currency or other contacts with the outside world, as does Plato’s Magnesia; it exists within the contemporary realities of Greece at the beginning of the Hellenistic period, and these realities constrain the design of his state so that its physical layout cannot perfectly reflect his social ideals. Aristotle’s description of the siting and organization of a city is justly famous, although confusing in a number of points.48 The question of the proximity of the sea to his ideal state is a concern to Aristotle as it is to Plato. Unlike Plato, however, he recognizes the advantages of commerce and naval power to a real-world polis and so recommends that the state have a port, but that it be far enough away to allow the state to regulate and legislate the dangers of strangers and foreign contacts to the moral development of the citizens.49 The specific siting of a city, says Aristotle, depends on its constitution: ‘‘an acropolis is suitable for oligarchy and monarchy, a level plain suits the character of a democracy; neither suits an aristoc-

Greek City Planning

13

racy, for which a number of different strong places is preferable’’ (Pol. 1330b). He again draws parallels between the social organization of a state and its physical situation. Like Plato’s Magnesia, Aristotle’s city is placed in the center of its territory for ease of communication and transportation. The siting should be determined by four considerations: health, political life, military security, and water. It should be on sloping ground, preferably facing east or towards east breezes or, failing that, facing south.50 It should be difficult of access for the enemy but easy for the inhabitants to escape from. It should have plentiful fresh water and healthy breezes; and if all the water is not equally pure, drinking water should be kept separate from other sources. Like other planners, Aristotle divides the territory of his ideal state into sections ( μέρη). He first divides the land into two sections, public and private.51 Each of these sections he then divides in half: the public section into land used to support religious functions and land used to support the ‘‘common tables,’’ and the private section into a district nearer to the city and a district nearer to the borders of the territory. Each citizen will then receive one plot in each section. Like Plato, he wishes to avoid civil strife over the ownership of land near the city and also wants all citizens to ‘‘share in both districts’’ so that all have a common interest in defending the frontier. These sections are thus more similar to Hippodamus’s tripartite functional division than to Plato’s semi-independent districts, just as Aristotle’s division of citizens into groups based on occupation is more similar to Hippodamus’s than to Plato’s tribes. Aristotle is one of the few authors to make specific comments about the design and layout of a city, its street pattern, blocks, and houses. In an important but somewhat obscure passage, he recommends that

Ἡ δὲ τῶν ἰδίων οἰκήσεων διάθεσις ἡδίων μὲν νομίζεται καὶ ησιμωτέρα πρὸς τὰς ἄλλας πράξεις ἂν εὔτομος ᾖ καὶ κατὰ τὸν νεώτερον καὶ τὸν Ἱπποδάμειον τρόπον, πρὸς τὰς πολεμικὰς ἀσφαλείας τοὐναντίον ὡς εἶχον κατὰ τὸν ἀρχαῖον όνον· δυσείσοδος γὰρ ἐκείνη τοῖς ξενικοῖς καὶ δυσεξερεύνητος τοῖς ἐπιτιθεμένοις. The arrangement of private dwellings is considered to be more pleasant and more convenient for other purposes if it is regularly planned [εὔτομος], both according to the newer and according to the Hippodamian manner; but for security in war [the arrangement is more useful if it is planned in] the opposite [manner], as it used to be in ancient times. For that [arrangement] is difficult for foreign troops to enter and find their way about in when attacking. (Pol. 1330b, adapted from the Loeb translation) Aristotle is clearly contrasting regularly with irregularly planned cities, and seems to suggest that the ‘‘Hippodamian manner’’ is a special instance of the 14

Greek City Planning

‘‘newer,’’ being both εὔτομος, ‘‘regularly planned,’’ but also having other characteristics, such as division into larger regions (nemeses?).52 The meaning and implications of the term εὔτομος (literally ‘‘well cut,’’ but in this context generally translated ‘‘well divided, regular,’’ or ‘‘regularly planned’’) are not completely certain. Kondis takes εὔτομος as applying only to the residential portion of the city.53 To him, it implies ‘‘cut by frequent streets’’—and hence, through its easy communication, a ‘‘well-cut arrangement’’ (εὔτομος διάθεσις) would be convenient for other purposes but vulnerable and easy to penetrate in wartime. Although in this passage Aristotle is only discussing the residential district of the city, I suspect that the term ought to apply to the plan of the whole city. Τέμνω, κατατέμνω and related terms meaning ‘‘to cut’’ are frequently used to describe the process of laying out an entire city and its surrounding territory. We have already seen how Hippodamus himself was said to have ‘‘cut up’’ (κατέτεμεν) the Piraeus, and how Plato ‘‘cuts’’ (τέμνειν) the city and the countryside into twelve pieces. These processes are applied to whole cities and their territory, not simply to residential areas. Εὔτομος then should be a feature of the plan of the entire city; Aristotle notices it particularly in the context of the residential quarter because the geometric layout of blocks, the regularity of the streets which divide them, and the relation between streets, avenues, alleys, and other such dividing features of the city plan would have been most obvious there. Public areas and buildings, on the other hand, would be characterized by more monumental architecture, distinctive siting, and other features besides their layout in the city grid. A ‘‘well-cut arrangement of private houses’’ might imply a specific geometric pattern to Aristotle, a particular relation between the length and width of blocks (or the rhythm of streets and cross streets), or between wider avenues and narrower streets (πλατεῖαι and στενωποί ); but what that geometric pattern would be is difficult to say. Aristotle only describes it as ‘‘pleasing and convenient’’ but says that its disadvantage is that it is easy for an enemy to penetrate and explore; and those could be true of almost any type of grid plan. Aristotle’s solution to this dilemma is that

διὸ δεῖ τούτων ἀμφοτέρων μετέχειν (ἐνδέχεται γὰρ ἄν τις οὕτω κατασκευάζῃ καθάπερ ἐν τοῖς γεωργοῖς ἃς καλοῦσί τινες τῶν ἀμπέλων συστάδας) καὶ τὴν μὲν ὅλην μὴ ποιεῖν πόλιν εὔτομον, κατὰ μέρη δὲ καὶ τόπους· οὕτω γὰρ καὶ πρὸς ἀσφάλειαν καὶ κόσμον ἕξει καλῶς. Hence it is well to combine the advantages of both plans (for this is possible if the houses are laid out in the way which among the farmers some people call ‘on the slant’ in the case of vines), and not to lay out the whole city in straight streets, but only certain parts and districts, for in this way it will combine security with beauty. (Pol. 1330b) Greek City Planning

15

No really satisfactory explanation for Aristotle’s simile to a vineyard has been offered. A σύστας, literally ‘‘a closely planted vineyard,’’ has generally been interpreted as the Roman quincunx, in which grapevines are planted in the spaces between rows, thus: , in contrast to vines planted in rows (κατὰ στοῖχον). In a long discussion of this passage, however, Kondis emphasized that the ancient sources do not specify a particular geometric arrangement of vines, but only that the vines are planted very close together. He therefore suggests that Aristotle is describing an arrangement of dwellings which minimizes the number of streets, packing the houses closer together and reducing the number of access routes into the heart of the city. The arrangement of the Piraeus, Rhodes, and Thurii, for instance, with a fairly small number of wide avenues ( plateiai) dissected by narrower streets (stenopoi) which would allow access to the houses but not easy passage through the city to foreign troops would satisfy this interpretation.54 But if the main feature of Aristotle’s arrangement of houses was that it was closely packed, there are certainly less ambiguous ways to express that idea. His simile ought to be to the arrangement of the vines rather than to their proximity to one another. In modern vineyards, one of the most striking features of the arrangement of vines is that as you move through the field, a variety of different alignments appears from different points: from one vantage point you look straight down a row, from another vantage point only a few feet away you can see through the vines at an angle, and new alignments and paths seem to appear at every moment. Aristotle may envision an arrangement which is geometrically regular but more complex than a simple grid: one which includes diagonal streets, for instance; this would make the arrangement confusing to foreign troops. Another possible explanation might be derived from the way a closer planting of grapevines is achieved by using the quincunx pattern. A quincunx may be considered as a grid of grapevines whose interstices have been planted with more vines, thus achieving the ‘‘dense planting’’ described by ancient authors: . Those extra vines essentially block the paths between the vines, thus meeting Aristotle’s demand that the arrangement be ‘‘difficult for foreign troops to enter and find their way about in when attacking.’’ Translated into a system of blocks rather than vines, his arrangement might look like this: . There would still be straight access in one direction, but in the other direction the offset blocks would interrupt any direct route from the outside of the city to its heart. By applying this system ‘‘in sections and regions,’’ a city planner could ori16

Greek City Planning

0

100 m

figure 2. Plan of Goritsa ent the offsets so that most of the major accesses from the gates were blocked. The system would thus be ‘‘well-cut’’ and secure at the same time. No examples of these exact schemes of city planning are known, and they remain theoretical constructions, whether Aristotle’s or merely my own. But peculiarities of the layouts of a few cities may be associated with this passage of the Politics. The city of Goritsa in Thessaly was laid out in the last quarter of the fourth century—just after the Politics was written. Although Goritsa could have been laid out on a very regular grid oriented northeast-southwest, it was in fact oriented almost due north-south, forcing a more irregular layout of blocks (fig. 2). The city is divided into a number of regions with different layouts, some with the blocks Greek City Planning

17

oriented north-south, some oriented east-west; and at the boundaries of these regions the blocks are offset from one another so that only a few streets run from the gates into the center of the city. The city thus might be seen as laid out both as a closely planted vineyard and in sections and regions, with offsets between the sections.55 Likewise, an offset in the grid of Stymphalus has been interpreted as planning ‘‘in sections and regions.’’ 56 It is tempting to attribute irregularities in the grid plan in the Villa Section of Olynthus to such considerations as well (see chapter 2). Within the ‘‘well-cut arrangement’’ of the city, Aristotle specifies the location of a number of institutions (Pol. 1331a-b). Certain temples, and the common tables of the officials, are together given a separate location in the city, distinguished by natural prominence and fortified (or at least naturally defensible) from the neighboring sections of the city. Below this is an area reserved exclusively for the full citizens. Aristotle compares it to the Thessalian ‘‘free agora’’ and stipulates that it should be free of buying and selling, reserved for leisure activities. The commercial agora, on the other hand, is to be set apart, outside the rest of the city, in a central spot convenient to both port and countryside. This will also be the civic center, with law courts, administrative buildings such as the offices of the market officials and city officials, and other matters of business. Other temples must be set apart, their locations governed by the requirements of their separate cults. The locations of these institutions are thus determined primarily by topography and by natural qualities of the land, just as the location of the city itself is determined by the topography suitable for its constitution. Aristotle is again striving for a complete functional division of space: he does not divide the city into parallel sections like Plato’s μέρη, but into regions which are both topographically and functionally separated. These same principles govern the division of the territory of the city. There are to be temples to gods and heroes, administrative centers, guard posts for defense and other features of the city. Aristotle spends little time describing these, however, but ends his discussion of the organization of his city: ‘‘But to linger at this point over the detailed statement and discussion of questions of this kind is a waste of time. The difficulty with such things is not so much in the matter of theory but in that of practice; to lay down principles is a work of aspiration, but their realization is the task of fortune. Hence we will relinquish for the present the further consideration of matters of this sort’’ (Pol. 1331b).

i de a l a n d p r a c t i ce But while an ideal constitution and organization were philosophically desirable, they were not possible in practice, a fact even Plato recognized in his Laws. This discrepancy between theory and practice introduced tensions into the Greek po18

Greek City Planning

lis. The initial conditions set by the oikist, or founder of a city, were held sacred and not lightly changed; yet change was constant in cities. Land was bought and sold; divided among multiple offspring; or united through marriage, purchase, or rental. The principle of equal allotments of land, which was applied in the earliest colonies, was counterbalanced by inequalities in status and wealth among the citizen body; and as Aristotle pointed out, this could lead to stasis. Greek historical and epigraphic sources are often quite taciturn about the structure and makeup of actual citizen bodies, however, and we know even less about the relation between the organization of citizens and the layout of the city. Jones’s study of public organization in Greece emphasizes how disparate our surviving sources are and how frequently we are forced to reconstruct hypothetical systems of organization from a few preserved tribe names or casual and often ambiguous observations.57 Many Greek cities had composite populations, with citizens drawn from different cities or regions of Greece. Olynthus, whose population was a mix of natives, Bottiaeans, Chalcidic settlers who settled on the South Hill, new immigrants arriving in the anoikismos of 432 b.c., and later arrivals during the fourth century, is only one example. The difficulties of unifying a diverse population, which Plato alludes to in his description of Magnesia, would have been common to any such composite city: Aristotle notes that ‘‘hence most of the states that have hitherto admitted joint settlers or additional settlers have split into factions.’’ A famous example of such a composite city dissolving into factional strife is the city of Thurii, founded on the site of Sybaris in southern Italy. The neighboring city of Croton had wiped out the majority of the Sybarites in 511 b.c. and laid the city to waste. After a lapse of fifty-eight years, according to Diodorus, the city was refounded as a synoikism with a group of Thessalians; but they too were driven out shortly afterwards by the Crotoniates. The refugees sent ambassadors to Greece and invited settlers from the Spartans and the Athenians. The Athenians sent a group of colonists in ten ships led by Lampon and Xenocritus, and also announced the formation of the colony to the Peloponnesian cities, inviting whoever wished to join. Thurii was thus from the first a very heterogeneous community, formed from the original Sybarites, Athenians, and Peloponnesians. It was also a community with an all-star cast: among the many famous thinkers who emigrated were Hippodamus himself (who is thought to have planned the city, although no ancient source actually says he did), Herodotus, Thucydides the son of Melesias, Lysias, and other notables.58 Tensions among citizens of such a mixed community led to stasis, however: For a short time only did the Thurians live together in peace, and then they fell into serious civil strife, not without reason. The former Sybarites, it appears, were assigning the most important offices to themselves and the lower ones to the citizens who had been enrolled later; their wives they also Greek City Planning

19

thought should enjoy precedence among the citizenesses in the offering of sacrifices to the gods, and the wives of the later citizens should take second place to them; furthermore, the land lying near the city they were portioning out in allotments among themselves, and the more distant land to the newcomers. And when a division arose for the causes we have mentioned, the citizens who had been added to the rolls after the others, being more numerous and more powerful, put to death practically all of the original Sybarites and took upon themselves the colonization of the city. Since the countryside was extensive and rich, they sent for colonists in large numbers from Greece, and to these they assigned parts of the city and gave them equal shares of the land. Those who continued to live in the city quickly came to possess great wealth, and concluding friendship with the Crotoniates they administered their state in admirable fashion. Establishing a democratic form of government, they divided the citizens into ten tribes, to each of which they assigned a name based on the nationality of those who constituted it: three tribes composed of peoples gathered from the Peloponnesus they named the Arcadian, the Achaean, and the Eleian; the same number, gathered from related peoples living outside the Peloponnesus, they named the Boeotian, Amphictyonian, and Dorian; and the remaining four, constituted from other peoples, the Ionian, the Athenian, the Euboean, and the Islander. They also chose for their lawgiver the best man among such of their citizens as were admired for their learning, this being Charondas. (Diod. Sic. 12.11) Unequal distribution of land and privileges led to the dissolution of the state; and the remedy was to distribute equal shares of land, establish a rational division of the citizen body—in this case based on origin—and a well-ordered system of laws, which Diodorus describes at some length. Diodorus also describes the layout of the city, which he attributes to the Athenians led by Lampon and Xenocritus:

νομίσαντες εἶναι τοῦτον τὸν τόπον τὸν δηλούμενον ὑπὸ τοῦ θεοῦ περιέβαλον τεῖχος, καὶ κτίσαντες πόλιν ὠνόμασαν ἀπὸ τῆς κρήνης Θούριον. τὴν δὲ πόλιν διελόμενοι κατὰ μὲν μῆκος εἰς τέτταρας πλατείας, ὧν καλοῦσι τὴν μὲν μίαν Ἡράκλειαν, τὴν δὲ Ἀφροδισίαν, τὴν δὲ Ὀλυμπιάδα, τὴν δὲ Διονυσιάδα, κατὰ δὲ τὸ πλάτος διεῖλον εἰς τρεῖς πλατείας, ὧν ἡ μὲν ὠνομάσθη Ἡρῴα, ἡ δὲ Θουρία, ἡ δὲ Θουρῖνα. τῶν ὑπὸ δὲ τούτων [τῶν] στενωπῶν πεπληρωμένων,a τὰς οἰκίας b ἡ πόλις ἐφαίνετο καλῶς κατεσκευάσθαι. ὑπὸ δὲ erased in P, S; Vogel: τούτων δὲ τῶν στενωπῶν πεπληρωμένων; Casevitz: ὑπὸ δὲ τούτων τῶν στενωπῶν πεπληρωμένων b codd., accepted by Castagnoli, Kondis. Wesseling, Casevitz, Vallet etc.: ταῖς οἰκίαις a Kondis;

Believing this to be the place which the god had pointed out [in an oracle], they threw a wall around it, and founding a city there they named it Thurium 20

Greek City Planning

[To view this image, refer to the print version of this title.]

figure 3. Plan of Thurii. After D. Mertens, in D. Mertens and E. Greco, ‘‘Greek Architecture in the West,’’ in The Greek World: Art and Civilization in Magna Graecia and Sicily (New York, Rizzoli, 1996) 259

after the spring. They divided the city lengthwise with four avenues, the first of which they named Heracleia, the second Aphrodisia, the third Olympias, and the fourth Dionysias, and crosswise they divided it with three avenues, of which the first was named Heroa, the second Thuria, and the last Thurina. And when the rectangles enclosed by these avenues were filled out with streets, the city appeared well laid out in its residential districts. (Diod. Sic. 12.10, adapted from the Loeb translation) Diodorus’s account distinguishes between πλατεῖαι and στενωποί (wide avenues and narrow streets), creating a two-tiered street system. An initial division by means of four lengthwise πλατεῖαι and three πλατεῖαι across the width of the city was followed by a subdivision of these large areas by means of στενωποί.59 This has been confirmed in the limited excavation of this site. The high water table, deep burial and the later occupation layers of the Roman city of Copiae have hindered Greek City Planning

21

our understanding of the Greek remains, but the results have suggested that the city was divided by streets ca. 13 m wide into major rectangles 1,000 Greek feet (295 m) wide by perhaps 1,200 feet long. These major rectangles were then subdivided by smaller streets, ca. 3 m wide, forming blocks with a ratio of 1:2 (fig. 3). The use of round proportions and measures suggests that the major rectangles were a primary unit of the layout of the city, as Diodorus’s account suggests. The results are not secure enough, either in chronology or physical extent, to confirm the general process of layout described by Diodorus, but they are certainly not inconsistent with his account and give us some hope that we may understand this city better. ‘‘City planning’’ encompasses more than simply deciding where the streets, agora, and temples will be. A city is built for its citizens to inhabit, and in some ways the most crucial and yet least understood part of the process is determining how the community will interact with its physical environment, and how the physical environment should be tailored to fit the community. This seems to have been a primary concern of early Greek planners and theorists. We know so much less about the organization of real poleis, however, that it is difficult to document such correspondences between social and physical layouts. On one hand there is evidence at some cities for deliberate, public acts arranging tribes and other public units of organization locally, so that each tribe occupied a certain area of the city. These examples are somewhat rare, but show an interest in that correspondence between social and geographic divisions of the city. Other processes lead to a more ‘‘natural’’ ordering of the citizen body: the distribution of land, for instance, may cause citizens of like interests or affiliations to choose plots of land in the same region of the city (see ‘‘The Distribution of Land at Korkyra Melaina’’ in chapter 5). This could lead to a fairly well-defined system of zoning which reflects private concerns and ties, rather than the public organization of the city. Such a system of zoning can be distinguished at Olynthus.

22

Greek City Planning

chapter two

History and Archaeology at Olynthus

Olynthus lies between the westernmost and central fingers of the Chalcidic peninsula in northern Greece, about 2.5 km inland from the sea (figs. 1, 4). The country immediately surrounding the city is rolling fields, well drained and plentifully supplied with water. To the north, the Polygyros hills rise to some 1,000 m. The city was built on two flat-topped hills rising 30–40 m above the surrounding plain (figs. 5, 6). The original settlement was on the smaller and more steepsided South Hill. The North Hill was later laid out as a planned settlement with a strict grid plan (see below). In the later fifth and fourth centuries, the city expanded onto the plain to the east, in an area referred to as the ‘‘Villa Section.’’ A narrow ridge extending southwards from the southeast corner of the North Hill is known as the East Spur Hill (ESH). The urban geography of Olynthus reflects the major historical periods of the city, and we can compare housing and urban organization in different phases of the city’s life. The site offers many advantages. The fields around the city are today lush with green alfalfa, wheat, and olives, and in antiquity the soil was considered particularly fertile.1 The region is rich in timber for shipbuilding, and in antiquity grew figs (for which the city is named), grapes, and olives, as well as grain, beans, and fruit. Horses and livestock grazed in the hills. The Sandanus River (modern Resetnikia) flows by the foot of the city and supplied a convenient source of water (although water was also piped into the city from the hills to the north). Today the river is swollen in winter and quite low in summer, but in antiquity it seems to have been greater: a bridge seems to have been built over the river near the South Hill (fig. 6, Sec. L). ‘‘The country itself possesses ship-timber and has revenues from many ports and many trading-places, and likewise an abundant population on account of the abundance of food’’ (Xen. Hell. 5.2.16). 23

4,560,000 m 4,540,000 m 4,520,000 m

My gdo n

Pella

Amphipolis

ia Argilos

4,500,000 m

La Koro ke ne ia

Lake Bolb

e

Apollonia Stagira

i k e t t o B

4,480,000 m

Dikaia Methone

Strepsa

Aineia

15

2

m

30

5

610 m

Polichne? (Smixi)

m

Akanthos

4,460,000 m

Stolos

Olynthus

Assera Piloros

Mecyberna

Ak te

Singos Sermylia

Dion

Pa l

Aphytis

e

4,420,000 m

Torone

Scione

4,400,000 m

Mende

Sarte

n ia ho

len

S it

4,440,000 m

Potidaea/ Cassandreia

110,000 m (UTM zone 35) 130,000 m

150,000 m

170,000 m

190,000 m

210,000 m

230,000 m

250,000 m

270,000 m

figure 4. Map of the Chalcidice The urban history of Olynthus can be divided into fairly distinct periods. An early phase began with the first Greek habitation of the site and lasted until the Peloponnesian War. The rebellion of Olynthus and other Chalcidic communities from the Athenian Empire in 432 b.c. led to an anoikismos or ‘‘moving inland,’’ in which the populations of some neighboring cities moved to Olynthus to form a larger and more defensible city. In the later fifth and fourth century, the Chalcidic League, with Olynthus as its capital, grew in strength and population, becoming the predominant power in this part of Greece. In the fourth century, the League came into conflict with the rising power of the Macedonians, until Philip II de-

24

History and Archaeology at Olynthus

[To view this image, refer to the print version of this title.]

figure 5. View of the North Hill, Olynthus creed that there wasn’t enough room for them both. Outmaneuvered by Philip and betrayed by its own commanders, Olynthus was captured and its inhabitants sold into slavery in 348 b.c. The city was essentially abandoned, its houses in ruins. A few settlers returned to the site, either as squatters or as a garrison; but they had little impact either historically or archaeologically. The history of Olynthus as an independent polis ends, then, with its destruction in 348, only eighty-four years after the anoikismos. These events, particularly the anoikismos and the destruction, dominate our interpretation of the archaeological remains at Olynthus and tend to encourage a view of the site which may be oversimplified. Following a brief outline of the layout of the city, in this chapter I will discuss a few problems which are relevant to the planning and urban history of Olynthus in the later fifth and fourth centuries. The early remains are therefore of less interest than the Classical, while the questions surrounding Philip’s destruction of Olynthus are of great interest for understanding the houses and domestic assemblages. And the problem of the reoccupation of the site, which has been raised and discussed with some vehemence in recent years, must be resolved if we are to try to use the published data with any confidence.

History and Archaeology at Olynthus

25

Ro wA

´

OLYNTHUS

Ave. C

Ave. A

St. x

Row A

Ave. D

North Hill North Cemetery

100 m

Ave. B

0

ESH 6

St. v ESH 1-3 / Tr. 13

East

Agora

H. of the Wash Basin

Spu

Riverside Cemetery

Tr. 7

Ave. F

St. i

n

Ave. G

l

r Hil

Reset

ik

ia River

H. of Many Colors Villa CC Sec. N/ Tr. 10 “Civic Center”

V. of Good Fortune

H. of the Tiled Prothyron H. of the Twin Erotes

V. of the Bronzes

Fortification wall?

c. G

Se

Sec. L bridge?

H. of the Comedian

S. Villa Se

Sec. J/ K

c. F

Villa Section

Sec. E

fountainhouse

Tr. 4

South Hill Sec. Q/ Tr. 5-6

Sec. O (C -x 5-7)

theater??

Byzantine church; Neolithic settlement (Tr. 1-3)

figure 6. Plan of Olynthus

a b r i e f t o u r of the c i t y The architecture and city plan of Olynthus are familiar from Robinson’s publications and numerous later discussions. An extensive description is therefore not necessary here, but a short account will help set the stage for the more detailed descriptions and analyses of the houses (fig. 6).2 The South Hill The South Hill was settled in a somewhat irregular fashion, with clusters of rooms and, probably, shops facing out onto a simple network of streets. Two streets ran roughly north-south along the east and west sides of the hill, separated from the brow of the hill by a row or two of rooms. Two cross-streets running roughly eastwest were excavated, and there were probably others. The plan is therefore somewhat similar to that of the North Hill, although less regular.3 A public area was built in the north part of the hill in the fifth century, and other public buildings were constructed at the northern tip of the hill. The North Hill: Blocks and Streets The North Hill was laid out in an orthogonal plan which, with some irregularities, extended over the entire hill (fig. 7). Houses were mostly built in blocks of ten, composed of two rows of five houses separated by a narrow alley. On the east side of the city, however, the blocks were shortened to allow streets to follow the topography of the hill. The streets are oriented almost due north-south and eastwest. North-south arteries were labeled ‘‘avenues’’ by the excavators, east-west arteries ‘‘streets.’’ Avenues were labeled A, B, C, D, E, and F from west to east; streets were labeled in roman numerals i, ii, iii, up to xiii, beginning at the south end of the North Hill, with the streets south of these labeled −i, −ii, etc. Blocks were identified by the intersection of street and avenue at their southwest corner: hence block A v is bounded on the west by Avenue A and on the south by Street v. Within the blocks, houses were numbered from the northwest: house A v 1 is at the northwest corner of block A v; A v 2 is at the southwest corner of A v; A v 3 is the second house from the west on the north half of A v, and so forth. Row A and the East Spur Hill Along the west side of the city a single row of houses lies between Avenue A and the city wall, in a scheme recommended by Plato.4 This was referred to as Row A, and the excavated houses numbered from A −5 at the north (adjoining the gate in Avenue A) to A 13 at the south. North of the gate, another row of houses continues to the tip of the hill, referred to as Row A'. The northernmost six houses History and Archaeology at Olynthus

27

Blocks A iv - B ix and Row A 1

3

5

7

9

0

50 m

A -1

house numbering scheme 4

6

8

10

A1

2

A4

A3

A2

Block A viii

Block A vii

A5

B vii 1

A6

Block B vi

A8

A7

Block A vi

A9

Block A v

Avenue B Block A iv

Avenue A

A 12

A 11

A 10

Bv1

A 13

Fortification Wall

B vii 2

figure 7. Blocks A iv–B ix and Row A

were partly explored; these houses were probably reoccupied after the destruction of the city (below, ‘‘Olynthus after 348 b.c.’’). This row of houses seems to continue around the eastern side of the hill, where, however, it was more irregularly planned and the circuit wall was not definitely uncovered. Houses ESH 1 to ESH 4 seem to belong to this scheme, and traces of house walls continue the row farther to the south. Fortifications The fortification wall of the city was preserved mostly in foundations, about 0.8 m wide. It was traced primarily along the west side of the North Hill, where the wall forms the back walls of the houses of Row A.5 A gate at the north end of Avenue A was one of the main entrances into the city; other gates were presumably located to the south, probably between the North and South Hills and elsewhere. The circuit wall continues around the hill to the north, forming the back wall of Row A'. At the tip of the North Hill, it turns to the east in a rounded bastion to follow the contour of the hill. Possible traces of a fortification wall were discovered along the brow of the East Spur Hill, and the arrangement of houses and buildings along the east side of the North Hill suggests that the east brow of the hill formed the original limit of the city, with the circuit wall ringed inside by a row of houses analogous to Rows A and A'.6 The wall was not traced farther south than the East Spur Hill, and the southeastern limit of the city was not determined. Unpublished plans show a stretch of what may be the fortification wall and tower exposed in a trial trench west of the Villa of the Bronzes (fig. 8).7 If this is the fortification, it may account for irregularities in the city grid in this region. The House of the Comedian probably lay within the city wall, therefore; it seems to be one of the earlier houses on the site. The fortification wall must have been built when the city was expanded in 432 b.c. When the city expanded into the Villa Section, however, the eastern fortification may have been torn down, leaving few traces for Robinson to excavate. On the South Hill, a section of fortification wall was uncovered at the north end of the hill, just west of Section N.8 The date of the wall was not determined; it may have been part of the enclosure that ringed the North Hill or an earlier circuit that protected only the South Hill before the anoikismos. The South Hill must have been fortified in both periods, however: Artabazus’s siege of Olynthus in 479 b.c. shows that it was fortified at least that early (below, ‘‘The Early Period at Olynthus’’). The Villa Section On the plain to the east of the North and South Hills was a district which Robinson dubbed the Villa Section. The houses in this area were not labeled according to the scheme used on the North Hill but were named after some distinguishing feature History and Archaeology at Olynthus

29

in their plan or contents: the Villa of Good Fortune after the mosaic inscription in one of its rooms; the House of Many Colors after its painted decoration, and so forth (figs. 6, 8). The layout of the Villa Section was not completely clarified. A number of houses were excavated between Avenues F and G, and trial trenches explored the city grid in the vicinity; but wide areas were not opened up as they were on the North Hill. The grid in this part of the city is oriented 2–3° nearer to magnetic north than the grid on the North Hill. In general, the blocks seem to consist, like those on the North Hill, of ten houses in two rows of five. However, unlike on the North Hill, open spaces were left between some of the houses. In some cases, these empty spaces probably resulted from irregular land distribution, creating house plots which were larger than normal (such as the Villa of Good Fortune or the South Villa) and leaving adjacent small plots empty, or perhaps used as gardens associated with adjacent houses (below, chapter 6, ‘‘gardens’’). In other cases, house plots of normal size may have simply been left undeveloped. The existence of these undeveloped areas and larger-than-normal house plots suggests that the process of land division and distribution was different in this part of the city. Whereas on the North Hill each block or row of houses was built as a single unified structure, the house plots in the Villa Section were allotted less uniformly, perhaps over a longer stretch of time, and the houses built one by one rather than as blocks (chapter 5). Regular street intersections were followed along Avenue F for almost half a kilometer north of house ESH 4. However, in the southeast part of the city, between the House of the Comedian and the Villa of Good Fortune and Villa of the Bronzes, the layout of houses was somewhat irregular. The east-west streets to the east of Avenue F are offset about 5.6 m to the south of their course to the west. The house fronts along Avenue F do not line up, and the course of the avenue here is unclear. The reason for this irregularity may be the trace of the earlier city wall here, as described above. This area would therefore mark the intersection of the two city grids. The full extent of the Villa Section was not determined. It must have been fortified, but no trace of the later circuit was found. House walls were found as far as the Byzantine church of St. Nicholas, some 455 m east of the East Spur Hill, but Robinson suggests that this area was outside the regular grid, as the intervening valley flooded in most years.9 Graham, however, suggests that the city might have extended this far east, bringing its total dimensions to about 1000 × 900 m.10 The South Hill has an area of some 7 hectares. The North and East Spur Hills together measure some 28 ha, including the valley between the two hills. The extent of the Villa Section was not determined: roughly 16 ha can be documented, but the section may continue considerably farther. The built-up urban area documented by the excavations therefore totals about 51 ha, but the Villa Section could

30

History and Archaeology at Olynthus

The Villa Section

Ave. F

0

Excavation trench

Ave. G

ESH 5

ESH 4 House of Many Colors

Villa CC

Street -ii

House of the Comedian

Villa of Good Fortune

House of the Tiled Prothyron

House of the Twin Erotes

f city

Line o

Street -iii

32 BC

wall, 4 ?

Villa of the Bronzes

Street -iii

South Villa

Ave. F

Street -iv

figure 8. Reconstructed Plan of the Villa Section

50 m

be somewhat larger. Of those more than 51 ha, about 4.5 ha was actually excavated. This is a fairly small proportion of the whole town, but is more than has been excavated at almost any other Greek urban site. Public Buildings Two public areas were excavated on the South Hill. At the north end of the hill was a poorly preserved area with two ‘‘arsenals’’ or stoas, and fragments of other large buildings.11 This area was heavily robbed after the destruction, leaving little preserved remains. To the south of this area was a space Robinson called the ‘‘Civic Center,’’ including a large building—probably an assembly hall—built largely of reused ashlar blocks. Its construction dates to the fifth century b.c., probably to the period between the Persian capture in 479 and the anoikismos of 432 b.c. 12 The southern end of this public building was later built over by houses, indicating that the building went out of use before the city was destroyed, perhaps replaced by the public buildings on the North Hill.13 Another public area was located towards the south end of the North Hill. This consisted of an open plaza, one block wide (about 85 m) and perhaps 137 m long, apparently free of buildings.14 Three public buildings surrounded the plaza on its north side and northeast corner. On the north was a poorly preserved stoa-like building taking up the southern part of block A iv—probably a stoa facing onto the square. At the northeast corner was another stoa-like building with a central colonnade, A iv 10 (the so-called bouleuterion). To the south of this was a fountainhouse, A iii 9. This square was probably the agora of the city, although Robinson and Graham argued that it was an ‘‘open area for military maneuvers,’’ and Hoepfner and Schwandner claim that it was a sanctuary. There is, however, no evidence for any building in the open square, although it was explored with a trial trench over most of its length. I prefer, therefore, to identify this as the agora of the city (below, chapter 6, ‘‘The Agora’’).15 A hollow at the south end of the South Hill was tentatively identified in the first year of excavation as the theater of Olynthus. Trial trenches dug in 1931 at the bottom of the hollow, however, failed to locate any remains of the orchestra.16 Sanctuaries Leaving aside Hoepfner and Schwandner’s unsupported identification of a sanctuary at the south end of the North Hill, no sanctuaries have been found in the city. The main sanctuary of the city may have been outside the city walls. The treaty between Philip II and the Chalcidians decreed that a copy was to be set up ‘‘in the sanctuary of [Artemis at Olynthus]’’ (ἀναθεῖν ἐς τὸ ἱε[ ρ]ὸν . τῆ[ς Ἀρτέμιδος ἐν Ὀλύνθοι]). This inscription was found ‘‘in a field slightly south and to the west 32

History and Archaeology at Olynthus

of the northern half of the village of the village [of Myriophyto], that is . . . some three-quarters of a mile to the west of . . . Olynthus.’’ 17 Limited excavation there uncovers no traces of building, however, and it remains uncertain whether this marks the site of an extramural sanctuary or whether the inscription was transported here later. A dump of terracottas found on the South Hill just east of the Civic Center might be refuse from an early sanctuary. Reused blocks built into the public building in the Civic Center and fragments of painted stucco moldings and other decoration may come from this archaic sanctuary.18 There is plentiful evidence of household cult, and some domestic buildings in which ritual artifacts are unusually common, such as the House of the Tiled Prothyron (below, chapter 3, ‘‘The House of the Tiled Prothyron’’). Water Supply Although the Sandanus River flowed just below the city, other arrangements were made for drinking and washing water. Two fountain houses were discovered in excavations. One, whose southwestern corner was preserved, was set near the entrance road at the southeast edge of the South Hill. This ashlar building was fed by a pressure pipe, which probably led to springs in the Polygyros mountains some 8 km away. Black-figured sherds found on the floor of the building suggest a date in sixth century b.c., a date accepted by Crouch and others, who places this among the earliest pressure systems in Greece.19 A second fountain house was built in the northeast corner of the agora (building A iii 9). It was fed by a large terracotta pressure pipe, laid in a tunnel three to six meters below ground. The pipe continued across the northern part of the agora, turning a corner at the northwest corner of the square, and continued northward beneath Avenue A, leaving the city through the gate at the northern end of the avenue.20 This too was probably fed from the hills north of the city. The fountainhouse was drained through another pipe leading off to the west. Some 631 amphora toes were found in and around the building. Cemeteries Three cemeteries were partly explored by Robinson’s team. Some 560 graves were excavated at the main one, known as the Riverside Cemetery, which was located just to the west of the city (fig. 6). To the north, the North Cemetery produced only some 30 graves, more widely scattered than at the Riverside Cemetery. The third burial ground was the East Cemetery, located on a small knoll some 700 m east of the tip of the South Hill. In addition, a Macedonian-type chamber tomb was excavated some 2 km west of the city. Robinson believed that these three graveyards were the only ones used by the Olynthians.21 History and Archaeology at Olynthus

33

historical overview The Early Period at Olynthus Olynthus was already ancient in 432 b.c. In the Neolithic period there was a small settlement on the South Hill. No Bronze Age remains are preserved on the site, but nearby Agios Mamas was inhabited during that era.22 For our purposes, however, occupation really begins in the seventh century, when a town was founded on the South Hill. The date of this earliest occupation is not securely fixed. Imports of Corinthian and Attic pottery seem to begin in the sixth century, but Mylonas dates local, ‘‘pre-Persian’’ pottery to as early as 700, while suggesting that occupation may have begun even earlier.23 The history of the town before the Persian wars is little known. At some point it was taken, according to Herodotus, by the Bottiaeans, a local tribe which had been driven here by the Macedonians.24 Together with the other towns in the area, Olynthus supplied troops and ships to Xerxes in 480. But during the Persian retreat from Greece in 479 b.c., Artabazus besieged and captured the city, suspecting that it would revolt from the king. ‘‘Having besieged and taken Olynthus, he brought these men [the Bottiaeans] to a lake and cut their throats, and delivered their city over to the charge of Critobolus of Torone and the Chalcidian people; and thus the Chalcidians gained possession of Olynthus.’’ 25 Artabazus did not kill all the Bottiaeans living at Olynthus, however, and they were an important element of the citizen body later on, minting their own coinage in parallel with the Chalcidic.26 After 479, then, there will have been an immigration into Olynthus of new settlers from nearby Chalcidic communities, somewhat akin to the influx at the beginning of the Peloponnesian War; and the city, already ethnically mixed, became even more diverse, including Chalcidians, Bottiaeans, and probably remnants of the original, pre-Bottiaean population. The archaeological remains from this period are primarily located on the South Hill. A destruction level found in a number of spots has been attributed to Artabazus’s capture of the city.27 The early levels on the South Hill were not excavated extensively or carefully enough to reconstruct the plan of the settlement, but the few early walls which can be identified in the fieldbooks seem to follow the later plan in a general way. The plan of the South Hill may well have been similar in the early and later phases. More than sixty deep bell-shaped pits were cut into bedrock around the Civic Center on the South Hill, and then filled in some time in the earlier fifth century.28 These are mostly located in the streets along the east and west sides of the hill, and in the public area near the center of the hill. This suggests that they were public rather than private constructions. They were identified by Robinson as either granaries or cisterns. The bedrock here is too porous to make them effective cisterns, and most of them were not plastered or waterproofed. If they were used as cisterns, therefore, they would have been quite inefficient, although perhaps usable in a siege. Their location in open streets and the plaza makes it unlikely 34

History and Archaeology at Olynthus

that they were used as granaries under normal circumstances, since they would quickly have become damp and the grain moldy, not to mention the hazard they would have posed (and still pose) to pedestrians. They may have been dug during the emergency preparations for Artabazus’s siege in an attempt to provision the city with food and water, and filled in shortly after the capture of the city in 479 b.c. There are no signs that the North Hill was occupied prior to its layout as a gridplanned city. A few isolated coins and sherds found in this area are the only objects which predate the anoikismos of 432 b.c., and these probably attest the great length of time silver coins could stay in circulation rather than early occupation of this hill. Until 432 b.c., Olynthus was a relatively small city. An original member of the Delian League, she paid only two talents of tribute from 454 to 432 b.c., much less than some of the neighboring Chalcidian cities. By comparison, Mende usually paid eight, Torone six to twelve, Scione six to fifteen, and Sermylia, only about 13 km from Olynthus, paid three to six talents during those years.29 Before the anoikismos, therefore, Olynthus was by no means the largest city in the region, and its choice as the stronghold of the region during the Peloponnesian War and the new capital of the Chalcidic League must have been due to other factors: its central location, distance from the sea, and the availability of a naturally defensible area on which to build. The Anoikismos and the Planning of the North Hill Receiving from the Lacedaemonian government a promise to invade Attica, if the Athenians should attack Potidaea, the Potidaeans, thus favoured by the moment, at last entered into league with the Chalcidians and Bottiaeans, and revolted. And Perdiccas induced the Chalcidians to abandon and demolish their towns on the seaboard, and settling inland at Olynthus, to make that one city a strong place: meanwhile to those who followed his advice he gave a part of his territory in Mygdonia round Lake Bolbe as a place of abode while the war against the Athenians should last. They accordingly demolished their towns, removed inland, and prepared for war. (Thuc. 1.58) In August or September of 432 b.c., on the eve of the Peloponnesian War, Perdiccas, the king of the Macedonians, persuaded a number of Chalcidic cities to move inland and form a single fortified city at Olynthus.30 This anoikismos, or ‘‘moving inland,’’ was the impetus behind the expansion of the old settlement on the South Hill onto the broad, flat-topped hill to the north, and the laying out of a new grid-planned section of the city there. It is the most critical single event in the urban history of Olynthus. Thucydides does not mention just which Chalcidic cities took part in the reHistory and Archaeology at Olynthus

35

bellion and anoikismos. The revolt was not universal: many of the larger cities, such as Mende, Scione, Aphytis and Torone, remained loyal to Athens, continued to pay tribute to Athens, and were captured by the Chalcidians only later in the war. Argilus, Assera, Gale, Dicaea, Mecyberna, Phegetus, Singus, Scabala, Scapsa, Sermylia, Spartolus, Stolus, and Strepsa, however, are all absent from the Athenian Tribute Lists of 432 b.c., as is Olynthus itself, and other cities stopped paying in subsequent years. The region which joined in the revolt thus probably included the northern parts of the peninsulas of Pallene and Sithonia (Potidaia and its territory, Olynthus, Mecyberna, Singus, Gale) and probably much of the mainland Chalcidic peninsula (fig. 4).31 If all these cities rebelled from Athens and stopped paying tribute, it does not follow, of course, that their populations moved to Olynthus. There is, unfortunately, little direct evidence for the movement of specific populations to the fortified center. Strabo implies that the majority of them joined the move at this time, but this is certainly mistaken because a number of cities held back from the revolt and did not leave their homes.32 The inhabitants of three cities almost certainly moved inland to Olynthus: Mecyberna, Singus, and Gale. These three towns were assessed by the Athenians in 425 and 421 b.c. with the trivial contribution of 10 drachmas, although before the revolt they had paid from thirty to more than two thousand times this amount. The assessments may have been lowered in part because these cities were forced to contribute to Athenian garrisons stationed in them, and to reward the citizens who remained for their loyalty to Athens; but the insignificant sums must also reflect the inability of these cities to pay more because they had been all but depopulated in the revolt.33 The locations of these three poleis, on the coast in the area which rebelled from the Athenians, is perfectly consistent with this interpretation. It should be noted, however, that they were not completely depopulated— a remnant of loyal pro-Athenian citizens continued to pay tribute—nor were the towns completely torn down, as Perdiccas had tried to persuade the Chalcidians to do.34 The Peace of Nicias, concluded between Sparta and Athens in 421 b.c., includes the stipulation that ‘‘the Mecybernaeans, Sanaeans and Singaeans shall inhabit their own cities, as do the Olynthians and Acanthians.’’ 35 It seems that the inhabitants of those cities who had moved to Olynthus were now being required to return to their original homes; this stipulation actually is one reason for interpreting the tiny assessments of Mecyberna and Singus as reflecting the depopulation of those towns. The dioikismos or dispersal of population demanded by the peace was never carried through, however. Thucydides immediately follows his account of the terms of the peace with the Chalcidians’ refusal to honor the agreement to which they had not been a party. But the anoikismos involved more than just these three communities. Assera and Pilorus, both coastal towns lying within the probable area of the revolt, stopped 36

History and Archaeology at Olynthus

paying tribute in 432 and may have moved to Olynthus. Sermylia too may have joined the move inland somewhat later. And other populations were involved as well. In 430 b.c., for instance, the survivors of the siege of Potidaia fled to ‘‘the Chalcidice and wherever they could find a place to live,’’ and many of them probably settled in Olynthus.36 And in 423, when Scione and Mende revolted from Athens, Brasidas sent their women and children to Olynthus for safekeeping while they awaited the Athenian attack, and some of these might have become permanent residents.37 The country dwellers in the chora of Olynthus must also have fled into the defensible city during Athenian campaigns, just as the Attic villagers flocked into Athens during these years.38 Perdiccas’s offer of land around Lake Bolbe, some 50 km away, may have been made to avoid redistributing the territory around Olynthus to these new settlers, who otherwise would be landless. The Growing Population of Olynthus Allowing even a minimal reckoning of the states which participated in the anoikismos, then, a great influx of new inhabitants must have streamed into Olynthus in 432 b.c. and the following years. Since the raison d’être of the move was to form a defensible city, this population must have been housed in a walled area rather than scattered through the countryside. This influx must have caused a sudden housing crisis at Olynthus, and this circumstance is usually thought responsible for the expansion of the original town. The population increase at Olynthus, however, is difficult to estimate from historical sources. A number of scholars have tried to deduce the population and size of ancient cities from the amount of tribute they paid to the Athenian Empire. Although the relation between tribute and population was probably not simple or direct, these lists may give us a broad idea of the relative sizes and resources of the communities involved. Before the revolt, Olynthus consistently paid 2 talents (12,000 drachmas) of tribute yearly to the Athenian Empire. The cities of Mecyberna, Singus, and Gale, whose populations moved to Olynthus, paid varying amounts of tribute in the years prior to the anoikismos. Their combined tribute averaged about 2.7–3.2 talents per year, about half again as much as the tribute of Olynthus itself.39 We can probably conclude that the combined populations of those three cities alone exceeded that of Olynthus in 432. Of the other cities which may have joined in the anoikismos, Sermylia paid 4.5 talents just before the revolt, while Pilorus and Assera totaled about 0.6 talents. The refugees from Potidaia were probably quite numerous, as were the women and children from Scione and Mende (Potidaia had paid 6 talents of tribute, increased to 15 talents about 436 b.c.). To judge purely from the tribute paid by the cities which revolted from Athens, we might estimate that the number of immigrants in the years following the revolt was at least three times as large as the original population of Olynthus itself, and perhaps considerably more.40 History and Archaeology at Olynthus

37

Deriving absolute figures for the populations represented by these tribute payments, however, is frustrating and ultimately inconclusive. Estimates based on tribute for the population of Olynthus before the anoikismos range widely, from about 1,500 inhabitants to some 6,400 or more, and for the cities of Mecyberna, Singus, and Gale, from perhaps 2,250 to 9,600 inhabitants.41 West, on the other hand, relying on Beloch’s population estimates and on the numbers of troops in Chalcidic forces during the Peloponnesian War, calculated that one talent represents between 180 and 200 citizens, or a free population of 625 to 700—less than a quarter of Ruschenbusch’s estimate. He figures that before the Peloponnesian War, Olynthus had ‘‘a citizen body of four or five hundred. . . . This would make it a little village of about two thousand inhabitants, slaves and metics included.’’ Such widely differing figures inspire no great confidence in these methods. There is no evidence for settlement outside the South Hill at Olynthus before 432 b.c., and although some proportion of the citizens may have lived in the countryside, we may assume that the majority lived in this fortified settlement.42 A density of 150 people per hectare of built-up area is commonly accepted as a minimum figure.43 This would suggest a population of only 1,050 people on the South Hill, which has an area of only some seven hectares. Even if the hill was much more densely settled, it would be next to impossible to fit 6,400 people, the figure suggested by Ruschenbusch, in this limited space. Although these attempts to quantify the increase in population are frustratingly inexact, we can at least be assured that the impact of the anoikismos on the old town of Olynthus must have been overwhelming. The new immigrants will have greatly outnumbered the original inhabitants, tripling the population or more. The thousands of new fugitives will have needed housing, food, water, agricultural land to support themselves, and they will have placed new burdens on such city facilities as market space and the like. The political status and future of these immigrants will also have caused other sorts of questions: were they to remain in Olynthus only for the duration of the war and then return to their original cities, or did they intend to remain in the new, expanded city? Did they remain citizens of their original homelands or become citizens of the expanded Olynthus? Were they to be supported exclusively by the land in Mygdonia lent by Perdiccas, or did the anoikismos also entail a distribution of new agricultural land around Olynthus as well? In 432 b.c., then, Olynthus, already a diverse community of Chalcidians, Bottiaeans, and other peoples, became even more composite, with all the social problems of these states described by Plato and others. The Planning of the North Hill The anoikismos of 432 b.c. was the impetus for laying out a new region of the city, the North Hill, with a modern grid plan.44 There is no evidence that the northern part of the site was occupied before the reign of Perdiccas II. The only finds from earlier than his reign are a very few coins, found mostly in hoards whose 38

History and Archaeology at Olynthus

closing dates to much later or in the debris from Philip’s destruction, and a few black-figured sherds, again all from later contexts. Because many of the houses were excavated to bedrock, this negative evidence carries real weight. There are, however, forty-four coins of Perdiccas II from Olynthus, thirty-seven of which come from the North Hill. Although these coins probably circulated for some time before they were lost or deposited in hoards, and their distribution therefore does not reflect the extent of occupation in Perdiccas’s reign, the difference between the distribution of his coins throughout the North Hill and that of the earlier coins and vases, almost exclusively on the South Hill, strongly supports the conclusion that the North Hill was planned during the reign of this monarch. Absolute dates for most of the houses are harder to come by. Early floors were not distinguished in excavation, and although some objects can be assigned to strata which predate the final destruction floors, these may not necessarily belong with the earliest floors of the houses. Three deposits, however, offer two termini post quem and one terminus ante quem for the construction of houses on the North Hill, and these offer some insights into the sequence and absolute chronology of building here. A silver coin of the Thracian tribe of Trieres was found below the west wall of house A 11, that is, the west city wall, which must have been built immediately following the anoikismos. The very close resemblance between the Apollo heads on these coins and those on Chalcidic coins of Groups B through F led Clement to date the coins of Trieres to the last third of the fifth century, providing a terminus post quem for the fortification wall and this house no earlier than the anoikismos of 432—exactly what we would expect.45 Another deposit predating the construction of houses was found in a test trench along Street vii, under house D vii 4. This burned stratum ran about 0.4 m under the west wall of the house, and contained lamps, terracottas, and other finds. The latest lamps belong to Howland’s Type 24A, traditionally dated from late in the third quarter of the fifth century through the fourth quarter of the century.46 The terracottas are less easy to date, but none of the four figurines were apparently made from the same molds as figurines from destruction floor levels. The deposit probably reflects transitory occupation of some kind, perhaps while the new inhabitants were waiting or actually constructing their houses, and provides another terminus post quem in the last third of the fifth century for the houses in this area. The only secure terminus ante quem for construction on the North Hill is a hoard of silver coins found in house B v 1, whose closing ought to postdate the building of that house.47 This hoard contained three tetrobols of Perdiccas II, an early Athenian owl, twelve tetrobols of Acanthus, and three early Chalcidic tetrobols of Groups B and C. Clement dates the closing of the hoard to about 421 b.c. on the basis of the Chalcidic coins, although a later date has also been suggested.48 This house, and therefore this block and presumably the surroundHistory and Archaeology at Olynthus

39

ing blocks as well, ought therefore to have been built within a decade or so of the anoikismos. The Sequence of Construction on the North Hill Although the North Hill was probably planned about 432 b.c., Olynthus, like Rome, was not built in a day. As Graham and Robinson observed, the houses of the North Hill were built over a period of time, starting from the south end of the hill and working north. The houses of Row A on the west edge of the hill, for example, were constructed so that the north wall of each house was used as the south wall of the house to its north. It is unclear, though, whether they were built one after another over a period of time or were built in one campaign but without bonding the walls. The two rows of houses in block A vii were apparently built in different stages, separated by some length of time. Not only are the two rows very different in design and construction, but two earlier, irregular buildings were incorporated into the north row of houses (fig. 7). These buildings respect the southern row of houses and are separated from it by the usual drainage alley; they must have been built after the southern row. The houses of the north row were subsequently built around these buildings, leaving parts of one intact and destroying other parts.49 For some period of time, then, only the southern half of this block stood, and the space that was to become the northern half was unbuilt and available for these sheds or storage buildings. The length of time represented by this sequence is more difficult to determine. If house B v 1 was built by ca. 421, however, the process probably went fairly quickly—within a couple of decades at the most. The admittedly scant archaeological evidence, then, suggests that the blocks of houses on the North Hill were not begun before the revolt and anoikismos of 432 b.c., but that many were built within a fairly short period after that time. This is perfectly consistent with the historical evidence for a large influx of people into the city during these years. The Extent and Population of Olynthus in the Anoikismos Because defense was a primary concern, only the most easily fortified area was probably laid out and walled at this time. This probably included the whole North Hill and the East Spur Hill, the wall crossing from Avenue F. The less easily defended, low-lying areas were probably not included within the defenses: the Villa Section, down on the plain east of the city, and perhaps the valley between the North Hill and the East Spur Hill. The traces of defense wall along the east side of the city and the arrangement of houses along the brow of the North and East Spur Hills suggest that this was the original edge of the planned region. This area probably would have enclosed roughly 500 houses.50 How many people does this represent? The most careful analysis of the ancient Greek household, by Gallant, suggests that households varied in size cyclically, ranging from 2 to 5 coresident kin, generally numbering 4 to 5 adults, adolescents, and chil40

History and Archaeology at Olynthus

dren.51 Some households may have been larger, of course; and slaves are omitted from his calculations. This would suggest a new population of some 2,000–2,500 people, plus slaves. Other estimates of household size tend to be larger. Robinson and Graham estimate 6 to 8 free inhabitants but reconstruct a smaller settlement on the North Hill in 432 b.c., arriving at about the same number of people.52 Hoepfner and Schwandner postulate a household of 10 including slaves, for a total population after the anoikismos of about 6,000, increasing to 7,000 during the Peloponnesian War.53 Raeder estimates 10 to 12 people per house at Priene, where houses are rather smaller than those at Olynthus.54 All these estimates, however, are rather less than even the minimum number of people who Ruschenbusch and Zahrnt would estimate moved to Olynthus in the anoikismos (Ruschenbusch: 7,600–9,000; Zahrnt: 4,050–6,880). We would have to pack 15 to 18 people into each house to accommodate Ruschenbusch’s estimated population of Mecyberna, Singus, and Gale on the North Hill; and more if we include the refugees from Potidaia, Mende, and Scione and the other Chalcidic towns. Perhaps only a small proportion of the populations of Mecyberna, Singus, and Gale moved to Olynthus during the revolt, only a few hundred families rather than the thousands which one would estimate from the tribute payments, leaving the majority in their old, vulnerable homes. It is also possible that a larger area of Olynthus was planned and built up, although that would make the new city considerably harder to defend. But I suspect rather that it is misleading to estimate ancient populations using the methods employed by Zahrnt and Ruschenbusch. Although the Tribute Lists probably give us a useful impression of the relative sizes and strengths of different cities, the translation of quotas into absolute populations is problematic, and their estimates accord neither with the archaeological evidence nor with the numbers of troops mentioned by Thucydides. It is worth noting, however, that West’s estimate, made in 1918 before the application of modern demographic techniques, results in a minimum immigration of roughly 490–640 citizens, which, since some houses will probably have included more than one adult male (fathers with grown sons, for instance), is compatible with the archaeological evidence. Political Changes The political changes which accompanied this anoikismos have been long debated. Was the Chalcidic League, the most important political entity here in the fourth century, formed at this moment, or was it formed later, as Xenophon and Demosthenes seem to imply? Did the Chalcidians who left their own cities and moved inland to Olynthus become citizens of a newly created (or previously existing?) state or league of the Chalcidians? Or did they remain citizens of their original homes, becoming ‘‘Singians living in Olynthus’’ or the like? Or did they become History and Archaeology at Olynthus

41

citizens with full civic rights of the polis of Olynthus (rather than of a Chalcidic federal state)? 55 In the fourth century the official name of the political unit centered at Olynthus, documented in treaties and coins, was οἱ Χαλκιδεῖς or τὸ κοινὸν τῶν Χαλκιδέων, ‘‘the Chalcidians’’ or ‘‘the association of the Chalcidians.’’ 56 The clearest account of the workings of this state comes from Xenophon’s Hellenica. The year is 382 b.c., and the cities of Acanthus and Apollonia, threatened by the rapid growth of the Chalcidians, have sent ambassadors to Sparta, asking them to send troops to the Chalcidice to quell the growing power. Cleigenes, the Acanthian ambassador, tells the Spartans that ‘‘to be sure, almost all of you know that Olynthus is the largest city on the coast of Thrace. These Olynthians, in the first place, attached to themselves some of the cities with the provision that all should live under the same laws and be fellow-citizens, and then they took over some of the larger cities also.’’ He goes on to warn them that ‘‘such of the cities as share in the citizenship [of the Chalcidic League] unwillingly, these, I say, will quickly fall away if they see any opposing force presenting itself; if, however, they once become closely connected by reciprocal rights of intermarriage and of property, which have already been voted, and find that it is profitable to be on the side of the conqueror . . . then, it may be, this confederacy will no longer be so easy to break up’’ (Xen. Hell. 5.2.11–20). By contrast, the Acanthians wish to remain autonomous and use their own ancestral laws (Xen. Hell. 5.2.14).57 Xenophon lists a number of principles which governed the Chalcidians in the 380s. The cities are to share common laws and common citizenship and have rights of intermarriage, and citizens can hold property in any Chalcidic city. The date of the creation of the League remains disputed. Robinson, Clement, West and other American scholars argued that it was formed in 432 b.c. during the anoikismos, and this is perhaps still the orthodox opinion. Hampl, however, believed that the league already existed in 479 b.c.; Gaebler opted for the period 425–421 b.c.; whereas Gude and more recent scholars such as Zahrnt and Demand have preferred a later date, in the fourth century.58 Our first indisputable references to the League begin in the fourth century. In about 393 b.c., the League concluded a fifty-year peace with Amyntas III; this treaty may have been renewed or reworked somewhat later, in the 380s. In the copy of the treaty which was found at Olynthus in the nineteenth century, Amyntas is negotiating with ‘‘the Chalcidians,’’ rather than the Olynthians. Both sides are prohibited from forming alliances with Amphipolis, the Bottiaeans, Acanthus, and Mende, thus defining rough boundaries for the region controlled by the Chalcidic League. The preserved portion describes regulations for the import and export of timber and other materials, and customs dues which must be paid on certain products.59 Other fourth-century sources further document the Chalcidian League: Xeno42

History and Archaeology at Olynthus

phon’s account of the Spartan expedition to Olynthus; treaties between the Chalcidians and other states and kingdoms including the Athenians and Philip; and other state documents preserved on stone and in literary accounts.60 Before the fourth century, however, we have few relevant epigraphic documents. The Athenian Tribute Lists document the conditions Athens would want to impose on the Chalcidians rather than their actual situation, and it is easy to read more into the various ways the states are listed than the documents probably warrant. Literary sources are somewhat ambiguous. Thucydides does not explicitly describe the formation of a new state in the anoikismos; the ‘‘single strong polis’’ the Chalcidic communities formed when they moved inland to Olynthus might not necessarily imply a singular political state, but simply a physical stronghold, analogous to the use of πόλις to mean ‘‘acropolis.’’ Throughout his account of the war, Thucydides usually refers to ‘‘the Chalcidians’’ rather than to Olynthians, Mecybernians, and so forth, even though the official policy in Athens was to refer to cities alone rather than groups or organizations.61 It is thus possible that Thucydides’ terminology reflects the existence already during the Peloponnesian War of either a unitary polis or a federal league.62 Some incidents in his account suggest that the Chalcidians were organized into a formal state this early. They seem to act as a body throughout the war, rather than as individual cities. When the Athenians attacked Spartolus in 429 b.c., for instance, the Bottiaeans sent to Olynthus for help, and in response, Chalcidian (rather than Olynthian) hoplites and other troops were sent. When Brasidas and the Chalcidians had captured Torone and were besieging the fortress of Lecythus, they demanded that the Athenian troops evacuate the peninsula ‘‘since it is Chalcidian’’ (ὡς οὔσης Χαλκιδέων; Thuc. 4.110–15). Their demand can be understood as a claim for control even of Chalcidic states like Torone which had not joined in the rebellion, and thus that they considered themselves a confederacy of all Chalcidian states throughout the peninsula. But Thucydides’ usage of the term Χαλκιδεῖς is ambiguous and does not necessarily imply that there was a formal state of that name. If the troops stationed at Olynthus were still citizens of their original poleis but living in Olynthus for the duration of the war, then ‘‘Chalcidians’’ would be an accurate and convenient way of describing them. The Chalcidian coinage is often cited as evidence that the League was founded at the beginning of the war, rather than later. Robinson and Clement dated the earliest issues of the silver coinage inscribed ΧΑΛΚΙΔΕΩΝ to about 432 b.c., in part because they are similar in style to coins of Perdiccas II and show similar attrition rates in hoards to those coins.63 Although others have pointed out weaknesses in their argument, no better interpretation or more convincing dating has emerged.64 But a coinage bearing the legend ‘‘of the Chalcidians’’ does not necessarily imply that they were minted by a state or league; the legend could simply indicate that they were intended to be used throughout the Chalcidice, or that they were minted by the Chalcidians who were then living at Olynthus but not formally History and Archaeology at Olynthus

43

organized. While perfectly compatible with an early date for a League, therefore, the coinage and Thucydides’ accounts do not prove its existence at this time. A parallel which favors a date for the formation of a Chalcidic League at the beginning of the Peloponnesian War is provided by their neighbors, the Bottiaeans. These people had formed their own federal league, with a common boule and probably common strategoi in addition to local magistrates in each city, by about 422 b.c. Were it not for an inscription found in Athens, recording a treaty between the Athenians and this League, we would not know of its existence. But the fact that the Bottiaeans had organized themselves into such a federal state early in the Peloponnesian War makes it that much more plausible that the Chalcidians could have done so as well, as Thucydides’ language seems to suggest.65 The move inland therefore was probably accompanied by a political reorganization of the Chalcidians, forming a federal league of Chalcidic cities, perhaps loosely organized, but recognizing both a central authority and a degree of autonomy of the member states. How does this affect our understanding of the political affiliation of the new immigrants to Olynthus? Membership in a common league would ensure certain civic rights to all citizens and might have had provisions even this early for the new immigrants to maintain citizenship of their old cities while nevertheless enjoying civic rights while resident at Olynthus. These sorts of assurances would have been desirable before people would want to invest in constructing new houses at Olynthus and committing themselves, at least to some degree, to that city as their home.66 In addition, the formation of such a league would allow the Chalcidians to abandon their homes and move to Olynthus without officially dissolving their own poleis—a politically expedient move which would allow people to move back to their original homes at the end of the war, should they want to do so, while in the meantime maintaining their status at Olynthus. Again,Thucydides’ account of the revolt and the anoikismos, of the coastal Chalcidians tearing down their towns and moving en masse to Olynthus, is probably somewhat exaggerated: the small towns did not disappear but continued to exist independently (although in some cases on a greatly reduced scale).67 The Growth of the City in the Fourth Century Together with the economic and military power of the Chalcidians, the population of Olynthus grew rapidly during the fourth century, and the city expanded to accommodate the immigrants. In his description of the growth of Olynthus, Cleigenes, the Acanthian ambassador, tells the Spartans that the Olynthians have ‘‘not less than eight hundred hoplites and far more than that number of peltasts; while as for horsemen, if we also become united with them, they will have more than one thousand.’’ Later on, Xenophon describes an Olynthian cavalry of 600 horsemen.68 Demosthenes, speaking of the same period, attributes to the Olyn44

History and Archaeology at Olynthus

thians 400 cavalry and a total force of 5,000 men. Only thirty years later, he says, when Philip captured the city, their total force numbered more than 10,000, with a thousand cavalry.69 Unfortunately, neither Xenophon nor Demosthenes distinguishes between the forces of the city of Olynthus and those of the whole Chalcidic League.70 But nonetheless, the population of the Chalcidic capital city must have grown together with the League, and we should interpret the expansion of the planned section of Olynthus in this context. The Villa Section is probably a later expansion of the city, to accommodate the increasing population. The house blocks of this region do not line up with those on the North Hill, and the orientation of the streets in this region is about 2–3° different from that of the earlier grid. The grids do not connect: streets on the North Hill apparently did not continue over the brow of the hill to join those of the Villa Section, but stopped at their original dead ends leaving communication between the two parts of the city somewhat restricted.71 The residential district on the plain, then, was probably laid out in a separate, later campaign. Some of these houses can be dated. Silver hoards 6 and 7, from house ESH 4, just outside the eastern city wall, show that this house was built before ca. 379 b.c.72 This allows a rough terminus post quem for the expansion beyond the original limits of the city, probably beginning shortly after the close of the Peloponnesian War when Chalcidic power began its ascent. This suburb was only partly explored, and it could not be determined whether it was enclosed by a defense wall or not. The many artifacts found in these houses, however, including such weapons as the intact shield from the Villa of the Bronzes, probably imply that this part of the city was enclosed and defended. Had it been left unfortified, and the houses abandoned during the siege, the contents would probably have been removed either by their owners or by Philip’s army in the final days of the city’s life. Philip’s Destruction of Olynthus Olynthus was captured and sacked by Philip of Macedon in 348 b.c. This destruction is responsible for the unusually fine preservation of the houses and their contents. The great value to archaeologists of a violently destroyed site like Olynthus, of course, is that many artifacts are found in destruction debris on the floors of the houses, rather than in dumps or fills as in most other types of sites, so that they are still in something close to a primary context.73 The details of the attack, siege, and capture can be pieced together, albeit imperfectly, from the account of Diodorus, from speeches of Demosthenes, and from a few other sources.74 The alleged cause of the war was that Olynthus had given refuge to Philip’s two half-brothers, Arridaeus and Menelaus; but relations were already strained. Philip first invaded the Chalcidice in summer 349, captured at least one town by siege and destroyed it, and forced others to capitulate and History and Archaeology at Olynthus

45

defect from the League.75 In late summer, however, the Chalcidians concluded an alliance with Athens, and Athenian reinforcements arrived. In early spring, 348, a second Athenian army responded to the Chalcidians’ entreaties and invaded the territories Philip had captured the year before, but Philip soon followed by again invading the Chalcidice. Two battles ensued, and Philip emerged the victor in both. City after city, betrayed by pro-Macedonian factions, opened their gates to him, including Torone, one of the largest of the Chalcidic towns; Apollonia; Sane; and Mecyberna, harbor of Olynthus itself. As he approached, the Olynthians offered peace. But it was too late: Philip’s answer was that there was not room for the both of them; either they would have to leave Olynthus, or he Macedonia, for all time.76 And so ‘‘he confined them [the Olynthians] to the defense of their walls; then in the continuous assaults that he made he lost many of his men in encounters at the walls; but finally bribed the chief officials of the Olynthians, Euthycrates and Lasthenes, and captured Olynthus through their treachery. After plundering it and enslaving the inhabitants he sold both men and property as booty. By doing so he procured large sums for prosecuting the war and intimidated the other cities that were opposed to him’’ (Diod. Sic. 16.53). Philip made Olynthus an example, demonstrating to future opponents the price of resistance; Euthycrates and Lasthenes became infamous for their treason; and Olynthus in turn became a proverbial example of how no city, no matter how rich and powerful, can survive if its leading citizens are corrupted. It may have been the treachery of the hipparch Lasthenes that led to the capture, probably shortly before the siege, of 500 cavalry together with their weapons: a record number, says Demosthenes.77 Philip’s outrageous treatment of the captured city, and later of its enslaved inhabitants, likewise were cited to illustrate the barbaric cruelty of the Macedonian king.78 Demosthenes claims that he destroyed this and neighboring cities ‘‘so ruthlessly that a traveler would find it hard to say whether they had ever been inhabited.’’ 79 This is certainly an exaggeration, but one an Athenian audience would at least find plausible. Archaeological evidence of the siege and fighting at the walls and in the city was widespread. Slingbullets and arrowheads were found throughout the site (fig. 9). That many of these were spent in the final battles is shown by the inscriptions on some of them: eleven or twelve slingbullets and six arrowheads bear the name of Philip himself; others are inscribed with the names of Hipponicus, one of Philip’s generals, and Cleobolus, an unidentified general, probably one of Philip’s.80 Slingbullets were concentrated particularly on the east side of the South Hill, in Sections F and G, where more than eighty-five missiles were found in a relatively small area. This was perhaps the site of one of Diodorus’s ‘‘continuous assaults.’’ On the North Hill, the houses along Avenue B contained many more slingbullets than houses elsewhere, suggesting more intense fighting here figure 9. Distribution of Slingbullets and Arrowheads 46

History and Archaeology at Olynthus

Distribution of Slingbullets and Arrowheads slingbullet arrowhead 0

50

100 m

5 n8 s tha head ore w : m arro & G ts & c. F lle Se ngbu sli

than in the rest of the city. Interestingly, relatively few slingbullets were found in the streets and avenues themselves; the majority came from inside the houses, although it is hard to imagine that a sling would be a very useful weapon within the confines of a private home. In part this is because only houses were extensively excavated, leaving the streets undug; but even where streets were excavated, relatively few missiles were found. It may be that the defenders took to the flat areas of the house roofs and were pelted there, or that missiles in the streets were collected during or after the fighting. No casualties of war were found in the excavations. Probably the site was cleared and the bodies buried sometime after the destruction. Evidence of violent destruction and abandonment was also common at Olynthus. A number of houses showed evidence of intense burning.81 Many contained large numbers of objects in situ on the floors, including intact or complete vases, terracotta figurines and other breakable objects, large collections of loomweights, and such relatively valuable objects as metal vessels, grindstones, and pithoi, objects which at other, peacefully abandoned sites would have been carried off. Abandonment and Looting Some at least of the inhabitants of Olynthus seem to have left the city before the siege in 348 b.c. The prices of houses plummeted in the years just before Philip’s attack, apparently as foresighted inhabitants, recognizing the impending threat posed by the Macedonians, sold their property and presumably moved to safer locations.82 Some of the houses in the city were probably abandoned before the destruction and their assemblages depleted; the buildings were perhaps then used for dumping refuse, introducing objects into their assemblages. Those who remained at Olynthus will have lived for some time under siege or the threat of siege, and this undoubtedly affected the contents of their houses. They may have brought belongings and perhaps relatives, slaves, and livestock in to the city from the surrounding countryside, crowding the city as Athens was crowded during the Peloponnesian War. They may have made extra provisions for food storage; existing supplies of food may have been depleted. After its capture, the city was extensively looted. It is therefore no surprise that valuable objects, such as metal vessels, silver and gold coins, and the like, were found very infrequently in the houses. Certain types of assemblages are depleted, for example, equipment for symposia—with interesting archaeological consequences (below, chapter 4, ‘‘The Andron’’). The soldiers may have plundered stored foodstuffs, too, to supply their continued march: of the many pithoi excavated at the site, few or none had any significant contents preserved.83 Less precious but still usable objects were also salvaged and reused, by the survivors or by neighboring communities. Large stone grinding implements, olive crushers, and other worked stone implements were frequently salvaged. No complete sets of olive crushers were found, for instance; broken stones and perma48

History and Archaeology at Olynthus

nent installations like crushing floors are all that remains of what was probably a more extensive industry (below, chapter 6). Bathtubs, pithoi, and other features were also removed, sometimes leaving gaps in the floor where they had been set. Worked blocks were robbed from public buildings and elsewhere, leaving such telltale traces as Macedonian royal coins which clearly postdate the destruction. Moreover, some houses and areas may have been plundered more thoroughly than others. Philip’s soldiers may have concentrated on looting the richest houses, and later diggers too knew that some areas were more productive than others, with the result that the richer houses, which we would expect to contain larger, more diverse and more luxurious assemblages, could end up containing many fewer objects. Another difficult question is the extent to which the furnishings and contents of the houses were scattered from their original locations when the houses were destroyed. With more careful excavation, we might have been able to distinguish which objects had fallen from a height, which were resting on the floor, and which had been hurled to the spot where they were found; but this kind of information is mostly lacking. In some cases, detailed information recorded in the fieldbooks allows us to reconstruct the distribution of finds through a house, but the records are not always this complete. Some objects were also deliberately hidden to prevent their being looted. Four of the eight hoards of silver coins discovered in the excavations date to this period, as does probably the fine bronze brazier from house A xi 10, and a few other objects. We must be aware of the effects of such processes on the distribution of artifacts, even if we can only rarely control for them. Olynthus after 348 b.c. Although Demosthenes describes how Philip destroyed Olynthus and other cities ‘‘so ruthlessly that a traveler would find it hard to say whether they had ever been inhabited,’’ his description of the utter desolation is exaggerated.84 The city and the Chalcidic League were destroyed, but occupation did not cease. The territory of Olynthus was taken as royal land. A recently published inscription from Cassandreia records a grant of land by Lysimachus to a Macedonian, Limnaios son of Harpalos, including land ‘‘in the Olynthia’’ (ἐν τῆι Ὀλυνθίαι). The 360 plethra of land (about 36 ha) lay at Trapezous, probably the low hill about a kilometer southeast of the city. The inscription probably dates to 285/4 b.c., and shows that by this time the territory of the city had become royal property and much of it dispersed by Philip and his successors to Macedonians and their loyal followers.85 In 316 b.c., Cassander ‘‘founded on Pallene a city called Cassandreia, after his own name, uniting with it as one city the cities of the peninsula, Potidaia, and a considerable number of the neighboring towns. He also settled in this city those History and Archaeology at Olynthus

49

of the Olynthians who survived, not a few in number.’’ 86 Diodorus does not say, however, where these Olynthians were living, whether on the ancient city site, in a nearby settlement, or scattered over a wider area. But the name of Olynthus survived. An inscription of the second century a.d. found nearby also mentions the name of the city; the name and a community of some sort seems to have survived, probably as a small village.87 A funerary inscription found in a field north of the North Hill lists the names Leonidas, Derdas ‘‘the Macedonians,’’ presumably inhabitants of some settlement in the area, perhaps on the hill itself. The gravestone probably dates to the second half of the fourth century, between the destruction of Olynthus and the founding of Cassandreia.88 Historical and epigraphic sources tell us little about whether the city itself was reoccupied. The massive destruction level, overwhelming number of coins which predate 348 b.c., and Demosthenes’ description of the utter desolation of the site have generally convinced most scholars that Olynthus was for all practical purposes deserted. This is a critical issue, as artifacts found there have served a fundamental role in establishing the chronology of Greek archaeology in the fourth century. Sparkes and Talcott, for example, in their authoritative publication of Attic black-glazed and plain pottery from the Athenian Agora, cite comparanda from Olynthus 108 times; and this publication is the basis in turn for dating innumerable other objects and sites.89 Indeed, one of Robinson’s objectives when he began to excavate Olynthus was ‘‘to unearth objects of the period just before the destruction of August, 348 b.c.’’ 90 He made much of the fixed historical cutoff for the site provided by Philip’s destruction of Olynthus. Because they were found at Olynthus, for example, Robinson argued that certain types of artifacts which had previously been considered Hellenistic were in fact made earlier, in the Classical period. ‘‘We can now also date before 348 b.c. many works of art which previously would have been put in the Hellenistic age.’’ And, ‘‘Even the Hellenistic types of caricature and realism, such as satyrs and negroes, were begun in Hellenic times at Olynthus and I have found no reason for changing my belief that all our terra-cottas date before August, 348 b.c. No evidence of later occupation of the hills has yet come to light. Those scholars who are still unbelievers would do well to study carefully the volumes on Coins and Vases. If the Olynthian hills had been inhabited in Hellenistic times, we ought to find Alexander coins and Hellenistic vases.’’ 91 After the season of 1934, however, the excavators were forced to alter their position slightly. A number of coins of Alexander the Great, Cassander, and ‘‘Anonymous Bronzes’’ of the Macedonian royal mint, dating to the reigns of Alexander and his successors, were found at the site, particularly in the ‘‘Northwest Quarter’’—Row A' and blocks A xi–A xiii. Robinson and his staff concluded that ‘‘the Northwest Quarter was abandoned about the time of the reign of Cassander and the rest of the excavated area about 348.’’ 92 By 1935, some reviewers already doubted that Robinson’s confidence in the ter50

History and Archaeology at Olynthus

minus post quem non of 348 b.c. was completely justified. W. S. Ferguson, in his review of Gude’s History of Olynthus, notes that ‘‘between a quarter and a third of the datable Olynthians known by name lived after 348 b.c. . . . [Cassander] may have combed the Greek world for them [the Olynthians he resettled at Cassandreia], but it is more likely he found them collected somewhere. They need not have gathered at Olynthus, though that was the most natural place for them to have congregated; but if they had lived on the old site for the intervening thirty years the archaeological materials hardly admit of such close dating as to disprove their presence there. . . . It seems to me probable that an Olynthus was reconstituted after 348 b.c.’’ 93 But the Olynthus team stuck to their guns. In Olynthus 8 Robinson and Graham argued that ‘‘the view that the city was thoroughly destroyed and ceased to be inhabited in the year 348 b.c. has also gained more and more general acceptance. . . . Out of a total of nearly seventy reviewers of the Olynthus volumes dealing with the classical city, only two have definitively rejected this terminus post quem non.’’ 94 Responding to another review of the Olynthus series, Clement argued that because 3,528 of the excavated coins date to before 348 b.c., and only 60 coins date to between 348 and 316, there is a 98.3 percent chance that any coin from Olynthus dates from before 348 b.c.95 Such arguments from sheer quantity, whether of scholarly opinion or of ancient artifacts, carried a great deal of weight, and the cutoff date of 348 b.c. for the site was generally accepted. Further doubts about this terminus post quem non have been raised in an attempt to solve problems in the chronology of early Hellenistic pottery. Excavations at Koroni, a small site on the east coast of Attica, and other sites have shown that pottery which had been traditionally dated to the last quarter of the fourth or early third century must date to about a generation later, to about 270–260 b.c., leaving a thirty-year gap in the chronology of early Hellenistic ceramics.96 A number of scholars look to Olynthus for a solution to this discrepancy. James Dengate proposed to fill this gap by arguing that Olynthus was not abandoned in 348 b.c., but was extensively occupied until 316 when Cassander founded Cassandreia. Thus some of the material from Olynthus could be downdated from 348 to 316 b.c., filling conveniently the problematic gap in the pottery chronology.97 Like Robinson and Clement, Dengate turns to statistical arguments about the total proportion of coins found at the site, comparing the numbers of coins dating before and after the destruction. But he finds many more late coins than did Clement: in addition to ‘‘154 coins from Olynthus dating after 348 b.c. down to Cassander’s reign,’’ he argues that we should consider the 126 coins which Clement dated to before and after 348 as dating after 348, because ‘‘most Greek cities in the Aegean area did not begin regular issues of bronze coinage until after ca. 350 b.c.’’ 98 He thus counts 280 coins dating between 348 and 316, as opposed to the 60 listed by Clement. By contrast, Dengate continues, there are 11 coins dating from 316 to the time of Augustus, 18 from the Roman empire, and 13 History and Archaeology at Olynthus

51

Byzantine coins, 4 Venetian, 1 Frankish, and 3 Turkish. ‘‘Thus in the 2,300 years that have passed since Cassander removed the population of Olynthus only 50 coins were recorded from the excavation, while human activity for the 32 years between Philip’s destruction and the founding of Cassandreia is evidenced by 280 coins. . . . This is an average of 10 coins for every year while after 316 b.c. it is only about 1 coin for every 50 years. I am willing to accept that shepherds dropped the post-Cassander coins.’’ 99 The large number of coins dating between 348 and 316 b.c., Dengate concludes, is enough to prove that Olynthus was extensively reoccupied after its destruction by Philip. The reason for the mix-up is that ‘‘the excavators perhaps, because the 316 b.c. abandonment did not leave such a wealth of material as the 348 destruction, confused the two and attributed it all to 348 b.c. It is convenient to my argument that the 32 years difference between the two layers probably mixed in the publications of the Olynthus material is exactly the same span of time (about a generation) noted in the difference between the coins and the pottery from Koroni.’’ He therefore concludes that ‘‘not only the dated coins, but also pottery, lamps, and other objects from Olynthus can date after 348, [and this] allows us to fill the 30-year gap between the dates of the pottery and the coins found at Koroni.’’ 100 The absolute date, whether 348 or 316 b.c., does not particularly matter for the purposes of this study, for understanding the uses of space and the layouts of houses and of city. But if there were really more than one stratigraphic level in most houses, and if the excavators did not distinguish earlier and later phases but mixed the destruction level of 348 b.c. with the debris abandoned in 316, then the assemblages would be unsuitable for the kinds of analyses I want to do in this study. Archaeological Evidence from Olynthus In evaluating Dengate’s arguments, we may limit the period in question to the years between Philip’s destruction in 348 and the foundation of Cassandreia in 316 b.c. Although later objects were found at Olynthus, including later Hellenistic, Roman, Byzantine, and later coins and artifacts, all of these came from surface and upper strata, none from contexts significant to the occupation of the houses. Ferguson quite correctly points out that many Olynthians are attested in historical and epigraphic sources, not only in the period 348–316 but down into the third, second, and succeeding centuries; but none of these sources specify that they are living at Olynthus after 316 b.c. Indeed, most of them mention Olynthians living (or dying) somewhere else: soldiers in the employ of Hellenistic monarchs, a sculptor working in Rhodes, immigrants buried in Athens and elsewhere.101 According to the archaeological and historical evidence, then, by the end of the fourth century, Olynthus lay abandoned and buried. In considering the archaeological evidence for occupation between 348 and 52

History and Archaeology at Olynthus

316 b.c., though, we must be clear about what we are examining and where it was found. Both the Olynthus team and its critics make some of the same mistakes in method. For instance, both sides tend to analyze all the coins found at the site as one grand statistic. They then calculate percentages of this total from different periods to derive historical conclusions. But the finds from Olynthus do not constitute a single homogeneous mass of material: they consist of many individual assemblages from house floors, graves, upper fills, hoards, pits, chance finds, and other contexts, assemblages which must be treated separately and in detail. Moreover, Dengate and others combine coins which can be absolutely dated, for instance coins bearing the name of Alexander or Cassander, with coins dated by style or numismatic opinion. The distinction is critical in a case like this where we are not simply trying to determine a date for the site but establishing a chronology for all of Greek archaeology. We need to be very sure that we are basing that chronology on the firmest evidence and not simply establishing a circular argument. Archimedes boasted, ‘‘Give me a place to stand, and I will move the earth!’’ and we are in a similar quandary: we need a place to stand, something unquestionably fixed in the shifting universe of early Hellenistic chronology. Of all the 109 or so different mints and royal issues represented at Olynthus, only 4 types can be absolutely dated to the years between the fall of Olynthus and the foundation of Cassandreia: the coins of Alexander the Great; Cassander; ‘‘anonymous issues of the royal mint,’’ minted by Alexander, Cassander, and his successors; and one posthumous issue of Philip II.102 These are the only surely datable artifacts, and with these we must begin our investigation of the city after its destruction. Fifty-five such securely datable coins were found at Olynthus (table 1). These 55 coins show that Olynthus was reoccupied after its destruction, and that Demosthenes was exaggerating when he described the lifeless desolation of the city. But the distribution of the coins is striking, as Robinson and Graham noted (fig. 10).103 The Northwest Quarter accounts for 44 coins, or four-fifths of the securely late coins found at Olynthus. Nearly every house along Row A' contained late coins, and in significant quantities, up to 18 in a single house. And these quantities are even more remarkable because the houses were not completely excavated, but only trenched along the walls. Had whole houses been excavated instead of narrow trenches, how many more late coins would have been recovered? Moreover, early Hellenistic Macedonian issues constitute a large proportion of the coins found in the houses of the Northwest Quarter—up to more than 50 percent in the houses with significant numbers of coins, averaging 29 percent overall. This is an enormous contrast to the coin assemblages of houses in the rest of the city, which contained at most one late coin each. Houses in the Northwest Quarter, then, were very likely reoccupied, as Robinson and others concluded. What about the rest of the city? Two areas at Olynthus History and Archaeology at Olynthus

53

table 1: Macedonian Royal Coins, 348–316 b.c. Findspot

Number of Coins

Notes

Northwest Quarter A xi 9 A xi 10 A xii 9 A' 7 A' 8 A' 9 A' 10 A' 11 Total

4 4 1 2 1 18 8 6 44

Rest of North Hill A iii 9 A iv 10 Av5 A vii 9 Tr. 7 ESH 4 Total

1 2 1 1 1 1 7

depth not recorded depth not recorded found in dirt basket, depth unknown depth not recorded surface find depth not recorded

South Hill Sec. N Tr. 10 Total

2 1 3

0.6 m below surface depth not recorded

Other Treaty inscription findspot total

one in topsoil

1 55

produced light concentrations of late Macedonian coins: the public buildings at the east end of block A iv, which contained 3 late coins, and the public buildings at the north end of the South Hill (dug as Trench 10 and Section N), which also contained 3 late coins, 2 of them at least at a fairly low level.104 All these buildings were extensively robbed, as they were among the few buildings at the site built of cut stone, and it is most likely that the late coins were dropped here by stonerobbers, perhaps the remnant of population living in the Northwest Quarter. There is no architectural or stratigraphic evidence that any of these buildings were rebuilt after the destruction; the strata were quite mixed up, probably by the stone-robbing. figure 10. Distribution of Early Hellenistic Macedonian Royal Coins 54

History and Archaeology at Olynthus

Of the houses in the rest of the city, only three on the North Hill and one on the East Spur Hill produced late coins. One coin was found on the modern surface in Trench 7, and another was found in the dirt basket while excavating house A v 5; the stratigraphic contexts of the other two are unknown. None of these coins, however, can be securely associated with the use, destruction, or reoccupation of these houses. The total number of late coins, then, is not nearly as important as their distribution, concentrated in one region of the city, and found only in upper fills or in disturbed or unknown contexts elsewhere. Less Datable Coins When Dengate and others claim that late coins are distributed throughout the site, they include both datable Macedonian royal coins and less securely datable issues, including coins minted both before and after the destruction, coins whose chronology depends on style and historical probability rather than absolute dates, and coin types which have not been securely attributed. Clement identified 166 coins from Olynthus which could have been minted either before or after 348, and 94 coins which he calls ‘‘questionable’’ in date. The numismatic arguments for dating each of these mints are beyond the scope of this study. But one can make arguments from the distribution of these coins at Olynthus itself. If, on one hand, Dengate and Rose are correct and some or all of these problematic issues date after 348 b.c., then they should have been in circulation together with the Macedonian royal coins, and they should show similar patterns of distribution, assuming no other factors are in play. If, on the other hand, these coins are distributed differently at Olynthus from the Macedonian royal issues, then we must explain this discrepancy. Either the coins really postdate the destruction and the areas where they were found were reoccupied, but for some reason the people living there did not use, or did not lose, Macedonian royal coins as often as the people living in the Northwest Quarter did; or the coins are not securely datable and should be dated by their context rather than vice versa. In fact, very few of the issues in question are restricted to the Northwest Quarter.105 Notable among those which are restricted to this region are coins of Megara, Philippi, and other independent mints. Many of the problematic coins found here, however, belong to a series bearing the same types as coins of Macedonian kings but lacking a legend identifying the king. Clement described these coins as belonging to an ‘‘Uncertain King,’’ and divided them into four types.106 Types 3 and 4 were found only in the Northwest Quarter and at the findspot of the Philip inscription, a site which also produced a coin of Alexander the Great. Of Type 1, the specimens with the head facing right are concentrated in the Northwest Quarter, with a few examples elsewhere, while those with the head facing left are mostly found outside the Northwest Quarter. We are probably justified, therefore, in considering Types 3 and 4, and probably most of the right-facing coins of 56

History and Archaeology at Olynthus

Type 1, as belonging to Alexander or his successors and postdating the destruction. None of these coins carry the chronological weight of the early Hellenistic royal issues, but they should probably be considered among those issues. If we consider all these coins, the coins of Megara, Philippi, Histiaea, Elaeus, the two Unattributed issues, and the late Uncertain Kings as later than 348 b.c., it raises the total of late coins to 83, of which 63 come from the Northwest Quarter. Another major difference between the coin assemblages in the Northwest Quarter and those in the rest of the city is the proportion of coins of the kings of Macedonia. Coins of Philip and his successors (including Anonymous Bronzes and Uncertain Kings) make up almost half the coins in the Northwest Quarter; in the rest of the city, coins of kings from Perdiccas to Philip make up only about 6.9 percent of the total coins.107 After Philip’s conquest of the Chalcidice, I would suggest, the pattern of monetary circulation changed drastically. Now part of the Macedonian kingdom, the proportion of royal issues in use around Olynthus probably rose dramatically. Chalcidic coins ceased to be minted, of course (although old ones may have stayed in circulation for a considerable time) while other autonomous coinages, even if some of them continued to be minted, were probably not as widely circulated.108 The distribution patterns of late coins at Olynthus demand an explanation; we cannot lump all the coins together into a single statistic and lose sight of important details. If a large part of Olynthus had been reoccupied, rather than just the Northwest Quarter, then there is no obvious explanation for why people living in the Northwest Quarter would use primarily early Hellenistic Macedonian royal coins, while those living in other reoccupied areas would use exclusively autonomous Greek coinages. The most economical explanation is that securely late coins are concentrated in the Northwest Quarter because only that area was reoccupied, and they are all but absent from the rest of the site because that area was not inhabited after Philip’s destruction. We can therefore do much better than calculate the simple statistical chances that a given object dates to before or after 348, based on the overall quantities of early and late coins. The restricted distribution of late coins allows us to treat the reoccupied area of the city separately. Vases and Lamps Rotroff ’s study of late Classical and early Hellenistic pottery generally supports the traditional chronology. She contrasts the bulk of the pottery from Olynthus to pottery from Building Z in the Kerameikos, which was violently destroyed between ca. 318 and 307 b.c.—just at the time when Dengate and others would see the final abandonment of Olynthus—and finds that ‘‘it is scarcely possible that a substantial amount of the Olynthos material dates as late as 316 b.c., for, if it did, the two groups [Olynthus and Building Z] would be closely similar—and they are not.’’ 109 History and Archaeology at Olynthus

57

table 2: Cups of Late Shapes Catalogue

Findspot

Kantharoi Olynthus 5, no. 728 Olynthus 13, no. 513A Olynthus 5, no. 532

Tr. 4 nr. Grave 377 Tr. 4

Cup-kantharoi Olynthus 13, no. 500 Olynthus 13, no. 501

D iii 4 B vi 9

Skyphoi Olynthus 13, no. 583

Ave. B & A iii 9

Room

Depth

bc j pit

.9 m = floor .65 m

.5 m

Together with Brian Sparkes, however, Rotroff finds a number of vases which seem later than the bulk of the pottery from Olynthus. Six kantharoi and other cups are particularly noticeable for their slender proportions, plain rims, and other apparently late features (see table 2).110 Two pieces which Sparkes and Rotroff isolate as particularly distinctive are both problematic. The tall one-handled kantharos, Olynthus 5, no. 728, one of the most distinctive cups from Olynthus, was found in Trench 4 in 1928, but it was mistakenly republished in Olynthus 13 as coming from house C −x 5.111 The excavations in 1928 were casual, and the cup’s stratigraphy is impossible to reconstruct. It is possible, therefore, that it comes from an upper level, although no other obvious signs of reoccupation were noted here; its context is at any rate uncertain. The second cup, a kantharos with a tall plain rim, was found near Grave 377, and was published as Olynthus 13, no. 513A; this same cup however was illustrated mistakenly as no. 510A.112 Found outside a grave, this vase cannot be associated with the destruction level at Olynthus. It seems likely that the households which reoccupied the site after Philip’s destruction continued to bury their dead in the traditional cemeteries, thus introducing late vases. There are no other objects which can be associated with this cup. These vases, from outside a grave and from an unknown context in a trench on the South Hill, may well be later than 348; because they do not necessarily come from occupation contexts they do not affect the issues of the reoccupation of the houses. Some of the other pots that Rotroff considers suspicious, however, come from contexts which show no other evidence of reoccupation. The cup-kantharos with molded rim, Olynthus 13, no. 500, was found in a trial trench on the East Spur Hill, which exposed parts of two rooms of house D iii 4. These rooms also contained a red-figured hydria, a late work of the Painter of Olynthos 5.156.113 Other vases 58

History and Archaeology at Olynthus

by this painter are found in houses at Olynthus which show no other signs of reoccupation, and this vase almost certainly dates before 348 b.c. The rooms also contained a complete oinochoe, an almost intact stamnos pyxis, a lopas, an intact bowl with inturned rim, a black-glazed plate, a large, complete terracotta female head and a more fragmentary female figurine, twelve loomweights, a lead lamp (?), and a variety of other artifacts, whose quantity, condition, and similarity to other objects certainly from the destruction level make it very probable that they predate Philip’s capture of the city. Although the notes do not record which objects came from which room, they were all found at the same level and very probably all belong together. Likewise, the cup-kantharos from B vi 9 comes from a pit below the latest floor, together with a red-figured lekanis lid, the name-piece of the Painter of Salonica 38.290, and a number of other artifacts.114 This pit must predate the latest occupation of the house, since the wall between rooms j and k was built over the pit. Both pit and wall might conceivably date to after Philip’s destruction, but the complete absence of late Macedonian coins from not only this house but this whole region of the city argues against this. It is more likely that this cup predates the destruction; moreover it probably predates it by some span of time since it had been discarded in the pit and a wall built over the pit by the time of the destruction. These cups, if they did postdate the destruction, would be anomalous among the otherwise unremarkable domestic assemblages. One would expect them to be accompanied by late coins, other vases of unusual shape, or chronologically diagnostic artifacts. Instead, it would be preferable to reexamine the criteria used to date them late, or to simply consider them anomalously tall and narrow in shape, but without chronological implications. The Northwest Quarter It is surprising that none of these typologically late vases come from the Northwest Quarter. The houses in Row A' produced very few artifacts other than coins, and those are mostly small, undistinguished finds. This is probably due to both the incomplete excavation of the houses and the fact that these houses were abandoned peacefully rather than violently destroyed.The destruction debris was probably cleaned out by the later inhabitants, who took most of their belongings with them to Cassandreia. But significant assemblages of objects were found in houses in blocks A xi, A xii, and A xiii. In addition to four early Hellenistic coins, house A xi 10, for instance, contained more than 121 artifacts, including a deposit of vases, plastic vases, and other artifacts in the pastas of the house, as well as a cluster of 57 loomweights in the kitchen. Are these assemblages debris from the destruction of 348 and if so, can we distinguish a later occupation level? Or were these artifacts abandoned here at the founding of Cassandreia in 316 b.c.? If so, they should be diagnostically later than the material from the rest of the city. Or do the assemHistory and Archaeology at Olynthus

59

blages contain both destruction debris from 348 and later material abandoned by the settlers of Cassandreia, mixed by excavators who did not recognize two distinct levels? If Dengate is correct and there was an extensive reoccupation of Olynthus which was not recognized by Robinson’s team, here, if anywhere, is where we should look. Most of the vases, terracottas, and other finds from these houses are very similar to artifacts from the main part of the city. Rotroff comments that the vases ‘‘do not differ from the bulk of pottery from the site.’’ 115 The terracottas offer an even clearer situation. Of nineteen figurines found in houses of blocks A xi, A xii, and A xiii, eight were made from the same molds as figurines known from the part of the city which was not reoccupied. The molds might have survived the destruction and been reused to make these figurines, but it seems much more likely that the figurines themselves predate Philip’s attack.116 Most of these artifacts therefore seem to predate Philip’s destruction. These houses were buried very shallowly, only about 30 cm deep, and their stratigraphy is often unclear. There is no explicit record in the fieldbooks of two distinct occupation levels, but none of the late coins seem to have been found in direct association with the assemblages of vases and lamps, but come from higher fills or unrecorded contexts. While one can sometimes tentatively reconstruct the levels and assemblages of the 348 destruction floors in these houses, there are virtually no artifacts which can be associated with the late coins. The fact that the destruction debris was not cleared out of the houses suggests that these houses were not repaired and reoccupied, but were allowed to lie abandoned and were used as dumping grounds by the late inhabitants. This of course could have introduced other types of late artifacts in addition to coins. None, however, can be identified on stratigraphic or typological grounds. Even the Northwest Quarter, then, was probably only partly reoccupied. The resettlers probably rebuilt and lived in the houses in Row A', as those houses produced large numbers of late coins.These houses produced very few other artifacts, however. The floors were probably cleared of pre-348 debris after the destruction, and when the survivors were moved to Cassandreia, they took their intact belongings with them, dropping only a few coins. Moreover, the houses were only partly excavated, and sherd material was not saved. The houses in blocks A xi–A xiii, in contrast, were not cleared out and reoccupied and produced relatively fewer late coins, and so they seem to show a different sort of use, perhaps a combination of looting and dumping. To sum up, we cannot treat the artifacts published in the Olynthus volumes as a single group which can be uncritically dated to before 348 b.c. The presence of a significant number of early Hellenistic coins proves that there was some activity at Olynthus until 316 b.c., and the scattered later coins and other finds document sporadic human presence there even after the foundation of Cassan-

60

History and Archaeology at Olynthus

dreia. We must therefore evaluate the specific context of artifacts from Olynthus when using them to establish dates for Greek pottery, lamps, terracottas, or other objects. The distribution of early Hellenistic Macedonian royal coins is quite conclusive: the datable late coins are concentrated heavily in the Northwest Quarter. This area was reoccupied, but the rest of the city shows no evidence of reoccupation as such. The survivors salvaged stone and other reusable materials and artifacts from the ruins of the city, losing a few artifacts in the process, buried their dead, and dumped waste near their houses, but they did not reoccupy houses outside the Northwest Quarter. We can therefore treat the destruction levels as relatively clean of later material and consider the destruction assemblages as more or less contemporaneous groups. It also means that we can continue to use objects from the destruction level as a fixed point for the chronology of fourth-century Greece. We must remember, though, that not all objects from Olynthus come from the destruction level. Objects from the Northwest Quarter may date between 348 and 316 b.c. (although many date earlier), whereas those from upper fills, graves, and other less secure contexts cannot be dated by their context with such assurance, and cannot carry the chronological weight of the destruction assemblages. Robinson’s Excavation Olynthus holds a special place in the history of Classical archaeology. The director of the excavation, David M. Robinson, was professor of archaeology, epigraphy, and literature at Johns Hopkins University for many years, training generations of American classicists. The excavation staff lists read like a Who’s Who of American and Greek archaeologists. Among the many distinguished scholars who got their early training there are Walter Graham, John Travlos, George Mylonas, Saul and Gladys Weinberg, Paul Clement, Lawrence Angel, William A. McDonald; the list goes on and on. The excavation of the city was rapid by modern standards. In 1928, Robinson employed a team of more than 200 workmen; in his last season, Robinson estimated that they were removing about sixty tons of earth a day. In four seasons of excavation between 1928 and 1938, Robinson’s team excavated more than five hectares of the city and a portion of nearby Mecyberna. They uncovered more than one hundred houses, public buildings, streets, trial trenches, and more than six hundred graves—an achievement unrivaled at any other site in the Greek world. Such a pace of work would have been disastrous at most sites, and certainly Olynthus was not excavated with the care with which it would be today. But two things saved the site and make its excavation a unique achievement in the history of Classical archaeology: the relatively simple stratigraphy, which meant that the loss of exact stratigraphic context was generally not as great as it would have

History and Archaeology at Olynthus

61

been at most sites, and the careful recording system set up by J. Walter Graham, which ensured that sufficient information was recorded to make it possible to reconstruct the assemblages and, to some extent, the stratigraphy. The workmen excavated by removing passes of earth over a large area, up to one or two houses in extent at a time. This made close stratigraphic control impossible and made it difficult to distinguish pits or other intrusions. Luckily, the stratigraphy of the North Hill and Villa Section was generally quite simple.117 A layer of topsoil, which was fairly sterile, usually rested directly on the final destruction level. A dense stratum of collapsed rooftiles was encountered in most rooms, and most of the artifacts were found beneath this. In some rooms the rooftiles were concentrated in one corner where they had slid when the roof collapsed. Cement, cobble, and mosaic floors were obviously easy to distinguish in excavation. Earth floors were sometimes noted specifically by their texture, and artifacts were often described as resting directly on the earth floor. Artifacts found directly on the floors should belong with the final assemblages of the house. In other cases, however, earth floors were not specifically noted, and the stratigraphic positions of artifacts were simply recorded as absolute depths below surface. It is not always clear in these cases whether an artifact was on the floor or in some other context. Some objects whose exact contexts were not recorded very probably belonged with the final floor deposit. Large, complete, breakable objects such as vases are not likely to have been trampled into the earth floors or washed in with upper fills. But sherds, fragmentary artifacts, and small and less breakable objects, such as coins and loomweights, could have been trampled into the earth floor, washed in with later deposits above the floor, or brought in with earlier fills. Whether objects resting on the final floors were in a primary use context is harder to say. The objects may have been stored here for use elsewhere; the room may have been used to store refuse rather than as a working or living space; or the house could have been abandoned before the destruction of the city and used as a dumping ground for refuse from other houses (see below, ‘‘Domestic Assemblages and the Pompeii Premise’’). Some houses were excavated to well beneath floor level, and these fills produced a few objects as well. But since the North Hill and Villa Section were occupied for a relatively short period of time, there were fewer early floors and other stratigraphic complexities than at longer-lived sites. Recent excavations at the site have noted such early floors.118 The South Hill, in contrast, was stratigraphically much more complex, with a series of earlier levels predating the fourth-century phase. It is difficult to sort out the stratigraphy of many areas, and to distinguish objects coming from the mid-fourth-century destruction level from those found in earlier fills. Moreover, excavation here uncovered only a few complete or recognizable houses. For these reasons I have generally omitted the South Hill from the discussion of the houses,

62

History and Archaeology at Olynthus

although it would be very interesting to compare the organization of the older and newer quarters of Olynthus. In the first year of excavation, 1928, only the trench in which an object was found was recorded; more specific information was only sporadically noted, for instance in the case of large stone objects and coins. Because each trench encompassed a number of houses, most of the finds from this season cannot be assigned to a particular house or room. During the second season in 1931, however, J. Walter Graham instituted a new recording system, which documented the house, room, depth below surface, and sometimes the one-meter square where each artifact was found. More specific contexts were often noted as well: whether an object came from an upper fill, or from the layer of rooftiles found in most rooms, from the floor level itself, or from such other contexts as pits or burned deposits. These records in the fieldbooks and publications make it possible to reconstruct the room assemblages in many houses. Not all objects were recorded, saved, or published. The fourteen volumes of the Olynthus series published the vast majority of the whole vases, lamps, coins, terracotta figurines, recognizable metal objects, and other artifacts from the site. However, loomweights, grindstones, storage amphoras, coarse vases, pithoi, nails, bits of wire or metal, unidentifiable lumps of lead, and other household objects were found in great quantities but not systematically included in the final publications. They were, however, recorded in the fieldbooks, and their distribution can be reconstructed in part from these records. Coarse and unpainted pottery vessels suffered the most. Field notes occasionally mention that there were ‘‘many coarse vases and sherds’’ in a room, but few were collected or mended. Indeed, to collect, sort, and mend the quantities of coarse pottery the excavations must have produced would have been the work of squadrons of conservators for many lifetimes.119 We therefore lack much information about cooking pots, coarse storage vessels, and the like. The incomplete collection and recording of artifacts introduced systematic biases into the records. For instance, the publications and field notes record a much greater number of small vases than large.120 This probably does not reflect the actual numbers of vases in use in ancient houses because small pots like lekythoi and saucers would have either remained intact in the destruction or broken into relatively few fragments and so been more recognizable in the field. But such larger vases as kraters, basins, hydrias, and the like would have been reduced to a mass of sherds which would have required significant effort to restore. If not painted with figural decoration and saved for that reason, therefore, these shapes are probably under-represented in the publications. In these early excavations, sieving, flotation, faunal analysis, and other more modern and sophisticated methods of retrieval were not generally employed. The fact that only recognizably complete vases and figured sherds were saved

History and Archaeology at Olynthus

63

in excavation, and that the microstratigraphy of the houses was not explored or recorded, means that we can reconstruct the stratigraphy of the houses in only the most general way. It also means that most of the pottery from strata other than the destruction level was discarded, since this would have consisted primarily of sherd material. This is regrettable, but for the purposes of this study it is not an irretrievable loss. It does, however, make it difficult to compare domestic assemblages from Olynthus to assemblages from other sites. Scholars have noted, for instance, that relatively few ceramics were recovered from the excavations at Olynthus compared to some other Greek domestic sites, such as Halieis, where the recovery and recording of artifacts was more complete and systematic.121 The largest number of pottery vessels found in any single house at Olynthus is 106 (from the House of Many Colors). By contrast, the house in Area 7 at Halieis yielded ‘‘nearly 4,200 ceramic objects . . . strewn across and embedded in the latest living surfaces,’’ representing a minimum of 497 vessels.122 Some scholars have therefore suggested that the data from Olynthus are too incomplete to draw significant conclusions. Such a comparison between the assemblages from Olynthus and those of other sites is not entirely fair. Only 8 of the 4,200 objects from the Halieis house represent complete vessels; the rest consist of ‘‘primary refuse,’’ discarded broken sherds which were simply not saved or noted at Olynthus. Halieis, like most sites, was peacefully abandoned, and usable household equipment carried away, whereas Philip’s violent destruction of Olynthus left many implements in situ on the household floors. Many of the artifacts from Olynthus, therefore, belong to the destruction level and are probably close to their primary contexts (although cf. below, ‘‘Domestic Assemblages and the Pompeii Premise’’). Although analysis of refuse may reveal a great deal about household organization, such secondary deposits are a palimpsest of materials from different periods, modified by a variety of cultural and noncultural processes, and are far more complicated to interpret than material from primary contexts.123 A smaller, even incomplete, selection of artifacts from primary contexts is arguably more informative about the use of household space than the large quantity of sherds from mixed and redeposited contexts. Olynthus, despite the deficiencies in excavation and recording, is almost unique in the Classical Greek world in its large assemblages of domestic material from primary, rather than secondary, contexts. Because even the clearest household contexts, namely, destruction deposits, are rare and therefore rather poorly understood in the Greek world, we have few models or comparanda to which to compare the Olynthian assemblages. How much order was there in most Greek houses? How much refuse was strewn through different rooms, how many broken or unused artifacts lying around, some waiting to be mended or reused for some other purpose, others waiting to be discarded? Many rooms at Olynthus contained only a few scattered artifacts 64

History and Archaeology at Olynthus

whose use or significance is difficult to evaluate. Are these the paltry remains of more complete groups of objects, or do they represent litter and debris in rooms which were, for practical purposes, empty of artifacts (or at least imperishable artifacts)? Some houses were probably more cluttered and disordered than others, but without very careful excavation with these kinds of questions in mind, it is very difficult to evaluate this and related issues. More recently, Greek archaeologists have returned to Olynthus. House B vii 1 was excavated in 1987, and further work is planned.124 We can hope that more careful and scientific excavation will answer some of the questions left by Robinson’s work fifty years ago. Reanalysis of Robinson’s Discoveries Although rapid and somewhat crude by today’s standards, Robinson’s excavations were far more extensive than would be possible now; and we may capitalize on the sheer quantity of information he recovered to begin to make up for the lack of detail. The excavation of Olynthus produced huge masses of artifacts, a treasure trove of information about household assemblages, the kinds of work that went on in Greek dwellings, and about the organization of houses and of the city. For the most part, the artifacts were numerous but mundane: aside from a few stone and metal sculptures, the only objects of any artistic or intrinsic worth were mass-produced red-figure vases, terracotta figurines, and the like. The sheer mass of artifacts and other information from the site made it difficult for Robinson and his crew to analyze their results thoroughly. The publications, though admirably complete, are organized in a very traditional manner: volumes on architecture, on vases, lamps, coins and other materials. Robinson and Graham’s volume The Hellenic House (Olynthus 8) presents general conclusions about the layout and organization of the houses. It discusses some of the larger equipment found in them, such as pithoi, mortars, presses, and the like; but lists relatively few smaller artifacts. The first and third volumes on the architecture (Olynthus 2 and 12) present lists of finds, but without general synthesis. Concordances following many of the volumes are of only limited use to the reader, and many mundane types of artifacts, such as loomweights or storage vessels, were not systematically published. Robinson had intended to complete the series with a fifteenth volume summarizing the results of the excavations, but this was never written.125 This book is a reanalysis of Robinson’s results, working from the publications and unpublished fieldbooks. The only way to deal with this mass of information was with a computer. My first task was to go through the publications and field notes to create a comprehensive database of all the artifacts—both published and unpublished—rooms, installations, houses, graves, and other results of the excavation. The goal was not to capture detailed information about each artifact, but History and Archaeology at Olynthus

65

simply to record a general description, context, and references to the field notes and publication (if any). Descriptions of objects were standardized to conform to current usage.126 This allowed me to reconstruct the artifact assemblages for each room, to plot where different types of artifacts were found, and to investigate questions which would be impossibly time-consuming if one had to answer them from the printed records alone. After entering all the data from the publications, I went through all the fieldbooks, checking the published records against the original notes and adding to my database information which had not been published.127 This process of checking the final publications added a great deal of information. Many objects recorded in the fieldbooks were never published, such as coarse and plain pottery, loomweights, grindstones, storage amphoras, hardware, architectural fragments, and terracottas—a total of more than 8,132 artifacts.128 Additional contextual information for published objects was often recorded in the fieldbooks. There were also typographic and other, more systematic errors which crept in between excavation and publication.129 The fieldbooks and other notes also preserved unpublished information about house architecture and the city plan: sketches, dimensions, and unpublished plans whose results were not included in the site surveys. Published and unpublished plans and dimensions formed the basis of a new CAD plan of the site, from which the figures presented here were created. This digital plan was then linked to the database through a Geographical Information System, so that the distributions of different sorts of artifacts, room types, and the like could be analyzed and plotted. The database and plans will be made available on the Internet.130 Stratigraphy Once the information from the publications and fieldbooks had been entered into the database, I went through the houses to do a preliminary analysis of the stratigraphy. The stratigraphy of the houses was generally not explicitly recorded. There are often notes recording the presence of a burned floor, or that groups of objects were found on a floor, in upper fills, in ‘‘deposits,’’ and the like, but frequently the only indication of stratigraphy is the depth below surface at which an object was found. Although it is not explicitly stated, it seems that the surface level is not a single datum for an entire house or room but the ground level at some point near where the object was found. So in houses built on a slope, where the surface on the uphill side may be much higher than the surface on the downhill side, objects found at the same absolute level but on different sides of the house can be recorded as coming from very different depths below surface. This makes reconstructing the stratigraphy of rooms and assigning objects to floor levels rather more difficult. Room by room, therefore, I looked at the assemblages and tried to determine where the floor levels were and which objects had been found on the floors and 66

History and Archaeology at Olynthus

which in other contexts. In general I have tried to be cautious about attributing objects to floor levels. But it is important to recognize that artifacts lying on even a fairly level earth floor can be displaced above or below it by various processes: they may lie on top of other objects or furniture and so rest above the floor; they may have originally been set into depressions in the floor, and so lie below the floor level around them; they may have been shifted by natural earth movements or by disturbances, human or animal. Moreover, the modern surface often slopes, so the depth below surface depends on which point you are measuring from. Objects lying on the same absolute level could lie at different depths below a sloping modern surface. In most cases, therefore, I have allowed a leeway of 10 or 15 cm above or below the average floor level. Some objects can be associated with the floor simply by their intrinsic nature: for instance large, fragile objects like storage amphoras, if found complete and broken in situ, probably belong with the floor deposits, and these also helped to establish floor levels for objects found nearby. Some houses had a second story over some of their rooms, and this introduces another complication into the interpretation of the stratigraphy. I looked for architectural evidence in the form of stairbases, or stairbases in neighboring houses since all the houses in one row generally had the same roofline, and examined the covered rooms in these houses more carefully for evidence of assemblages which might have fallen from the second-story rooms. Lacking other information, we have to assume that objects belonging to an upper room would have been separated from the ground floor and its associated assemblages by some amount of fill or building debris—and this seems to be the case in the few houses where we can check, such as the House of Many Colors. But in almost every case it was almost impossible to distinguish second-story assemblages, and I can only begin to tackle the problem in this book. My attempts to reconstruct the stratigraphy are necessarily rather primitive because the records are not explicit. A number of complex and interesting problems are involved: questions of human disturbance and looting; of natural processes, soil shifts, rain runoff, slumping and the like; of the sequence of the collapse of the houses after they were abandoned. More careful excavation of other parts of the site could help us better understand some of these issues and let us interpret Robinson’s results with more clarity. Domestic Assemblages and the Pompeii Premise The violent destruction of Olynthus by Philip of Macedon left thousands of vases, terracottas, loomweights, grindstones, pithoi, and other equipment on the house floors—objects which at other, peacefully abandoned sites would have been carried off. Olynthus thus offers a much richer corpus of domestic implements than nearly any other Greek site, and one which belongs largely to a single destruction deposit. Other Classical and Hellenistic sites have significant destruction deposits, for instance Priene, Himera, Selinus, and Eretria; but none of these have History and Archaeology at Olynthus

67

published the household contents as extensively as Olynthus. Recognizing the importance of this destruction deposit, Robinson dubbed Olynthus the ‘‘Greek Pompeii.’’ 131 The groups of artifacts recovered in the excavation do not, however, represent complete household assemblages. Archaeologists sometimes speak of the ‘‘Pompeii Premise’’—the notion that the assemblages of artifacts found in a room directly reflect its original contents, frozen at a single moment of time like Pompeii, buried by the eruption of Mt. Vesuvius.132 Archaeological assemblages are subject to a variety of cultural and natural processes which remove or displace some artifacts and introduce others; they cannot be interpreted simply as ‘‘fossilized representations of past activities.’’ At Olynthus, however, many of the inhabitants probably escaped before the destruction with their more precious belongings, and Philip’s soldiers probably looted whatever was of real value. The destruction of the city and collapse and burning of the houses further modified the household assemblages, for instance mixing the contents of upper and lower floors. The houses were further picked over by the survivors after the destruction and by casual looters down to Robinson’s day, when villagers were still plundering the site for both antiquities and building material. Erosion, plowing, and other such noncultural formation processes have taken their toll as well. And finally, the excavators did not save, mend, and record every scrap they uncovered, nor are the notes often specific about the precise contextual situation of each artifact. None of these problems are limited to Olynthus, of course; but it is rare that enough remains to even attempt to consider complete domestic assemblages, so the problems are perhaps felt more acutely here than elsewhere. We are left with a great number of artifacts, but by no means a full picture of what was in each room. Can we then use these groups of artifacts as assemblages with real significance? Although abandoned and eventually more or less forgotten, the houses and domestic assemblages at Olynthus were still subject to many natural processes. Although it is difficult to judge the effects of looting, we can try to estimate the impact of erosion on household assemblages. Some of the houses at Olynthus were buried quite deeply, up to 2 m below the modern surface, while others were covered by 20 cm or less of fill. In general, the more deeply buried houses were better protected from erosion and pillaging; but to what degree have erosion and pillaging destroyed the artifact assemblages of the less well protected houses? We might begin by investigating the relation between the total number of finds in a house to the depth to which it was buried. If there is a clear relation, it would warn us that the size and contents of the domestic assemblages excavated by Robinson were significantly affected by erosion. Figure 11 demonstrates the entirely unsurprising result that houses buried under only a shallow layer of fill tend to preserve fewer artifacts—with two exceptions, no more than 100 or so artifacts were found in any house buried under less than half a meter of earth. But it is 68

History and Archaeology at Olynthus

350 A viii 7

Total finds, excluding coins & sherds

300

250

200

150

100

50

0 0

0.5 m

1.0 m

1.5 m

2.0 m

Average depth of fill over house

figure 11. Depth of Burial of Houses vs. Total Number of Artifacts (excluding coins and sherds)

also clear from the contents of even some of the most shallowly buried houses that erosion has not completely destroyed the assemblages. House A viii 7, for instance, represented by the mark in the upper left-hand corner of the graph, was buried under only about 25 cm of earth, yet its pastas contained seven vases, five storage amphoras, four thymiateria, three terracotta female protomes, two plastic vases, a fibula, a lead astragalos, miscellaneous objects of bronze, lead and stone, and 247 loomweights (fig. 55). Although other objects may have been salvaged or washed away, much still remains. In most houses buried under less than half a meter of fill, however, there have probably been substantial losses. In houses buried under more than about half a meter of fill, there is very little pattern in the distribution of artifacts. Some houses have many finds, up to 268 objects; others, more deeply buried (and just as carefully excavated), contain far fewer. This suggests that above about half a meter, depth of deposit does not significantly affect the preservation of the assemblages. We should therefore be wary History and Archaeology at Olynthus

69

of our interpretations of houses whose floors lay very near the modern surface, but the more deeply buried houses seem less damaged by later erosion and robbing. Household Assemblages in Literary and Epigraphic Sources Literary and epigraphic documents reveal a richness and variety in the contents of Greek estates which archaeologists rarely discover. The Attic Stelai, for instance, record the sale at auction of property confiscated from citizens who had been convicted of profaning the Mysteries in 415–414 b.c.133 Most of the objects listed are items, such as real property, slaves, agricultural products, and livestock, which would not appear in the archaeological record. Other objects listed in the Stelai, such as metal vessels, textiles, furniture, and other forms of wealth, as well as the inhabitants and livestock themselves, were the expected booty of a victorious army and would have been taken by Philip’s soldiers. Still other objects, such as furniture, cloth, and wooden articles, were perishable and so leave few or no traces. And the most commonly found archaeological remains—ceramics and terracottas—are surprisingly under-represented in the Stelai. The majority of the vases and other vessels listed on the inscriptions are large, often coarse vessels such as pithoi, amphoras, mortars, troughs, and the like, while the smaller table vases commonly found in excavation are hardly listed. These discrepancies in themselves should warn us not to take the archaeological assemblages as direct, unproblematic reflections of the contents of ancient estates. Determining Activities The study of artifact assemblages and room functions in a complex urban settlement like Olynthus is a demanding and formidable task. Only recently have these questions been addressed in a systematic way in the Classical world, using both the architecture and contents of houses.134 A number of studies of this sort have been carried out in the New World and for prehistoric cultures of the Old, but these too are not without their problems. Some recent studies of assemblages have used sophisticated statistical techniques to understand and explain the distribution of artifacts, and to help establish that elusive link between behavior and its material remnants. One can study correlations between different types of artifacts to try to distinguish groups or ‘‘toolkits’’ of objects which seem to be used together in some specific activity. Patterning in the distribution of such toolkits in different rooms and houses helps reveal the spatial organization of houses and sites.135 At Olynthus, however, I have found these kinds of methods unrewarding. First, the site is very large and complex, moreso than most of the sites where such analyses have been applied successfully; and the variety of objects is tremendous. In the study of rooms at Grasshopper Pueblo, for instance, artifacts and permanent features were categorized into 30 different classes.136 By contrast, my initial (and 70

History and Archaeology at Olynthus

already simplified) database of the objects distinguished 1,048 different sorts of artifacts and 58 types of permanent features. By grouping functionally similar objects into more general categories, I could resolve artifacts into roughly 66 types; the features could be summarized similarly. Classical Greek material culture is more diverse and complex than Late Mogollon and that much more difficult to understand with statistical methods. Moreover, just as rooms can serve more than one function, a single type of artifact can be used for more than one purpose. For instance, large shallow bowls on stands made of marble or terracotta are usually identified as louteria for washing, shown on vase paintings; and indeed some of them are found together with ritual artifacts, altars, and the like, and with amphoras which could have been used to store water for ritual ablution (below, chapter 3: the House of Many Colors, the Villa of the Bronzes, and house A iv 9). Other examples of these artifacts, however, are found together with grinding assemblages and might have been used for kneading dough—and so perhaps to be identified as the Greek kardopos (below, chapter 4). Pithoi were used for storing food in closed rooms and also for collecting rainwater in courtyards. Worn-out eating and drinking vessels could be reused for other purposes in other parts of the house. Just as the use of space in ancient houses was flexible and varied according to season or need, the use of artifacts could change, and our models for interpretation must be flexible enough to allow for such changeability. But most statistical models are unable to take such alternative functions into account. Our interpretation of the use of an individual object or a collection of objects should depend on the whole collection and a flexible understanding of what each component could have been used for rather than a rigid typology. There are also difficult issues surrounding the definition, quantification, and recovery of artifacts. For instance, although loomweights are considered individual artifacts, in antiquity they were only useful when a group of a dozen or more were strung on a loom. Only when such a group was found together should we suspect, therefore, that the room was used for weaving (see below, chapter 4, ‘‘weaving’’). Small numbers of loomweights and similar artifacts found in a room probably imply little about the use of that space. In any analysis, therefore, we must try to define and distinguish between significant assemblages, which have behavioral implications, and small numbers of artifacts, which may be secondary deposits and tell us little about the use of the room. Again, the lack of detailed stratigraphic recording hinders such an analysis, but it does not make it impossible. Biases introduced by incomplete collection and recording of artifacts can also seriously distort statistical analyses, creating seemingly significant associations while obscuring less obvious but more culturally meaningful patterns. For example, Nevett’s correlation analysis of a sample of houses from Olynthus revealed a strong positive correlation between loomweights and coins.137 Does this History and Archaeology at Olynthus

71

imply a functional relationship between weaving and trade or other monetary activity? I suspect not. In fact, only three rooms from the whole site had both sufficient loomweights to outfit a loom and an unusual or significant number of coins.138 The actual correlation between weaving and coinage is therefore very low, although the apparent correlation between artifacts is high. The statistical correlation results in part from her coding the data in terms of presence or absence rather than quantities, so that one loomweight or coin is, for the purposes of the analysis, the equivalent of fifty, and in part from inconsistent collection and recording of artifacts during excavation. Both loomweights and coins are common artifacts, unbreakable, easily recognizable in excavation, and were considered interesting enough to save and record. Small numbers of loomweights and small numbers of coins were therefore recorded in many rooms at Olynthus, leading to a positive statistical correlation.139 But if other, less easily recognizable and notable classes of artifacts had been collected, mended, and recorded as consistently, we would certainly see positive correlations with these, too. For instance, nearly every room almost certainly contained at least a few coarse pottery sherds—such is the nature of every ancient Greek archaeological site, where pottery is ubiquitous. But as described above, much of the sherd material, and many of the small groups of loomweights and coins, might belong to secondary contexts, not to be confused with the destruction debris. Such mixing of contexts in the statistical analysis inevitably results in skewed and uninformative results. Without very careful screening of the data, statistical correlations like these reflect vagaries in excavation and recording methods more than ancient Greek behavior. Statistical methods like cross-tabulation (used by Nevett) are useful in exploratory analysis, but after working with these and other, more complex statistical methods, including factor and cluster analysis, at some length, I was forced to conclude that such analyses almost inevitably fail to take into full consideration all the issues of primary versus secondary deposits, the varying quality of preservation and recording, and the different uses which one artifact could serve. Instead, I found that a close study of each room and assemblage within the context of the entire house was more profitable than applying statistical analyses to the entire database. Examining the architecture, contents, and complete records of each room, I tried to reconstruct the activities for which there was evidence. I could then compare the distribution of activities in different houses, see how rooms of the same architectural type were used, or see how similar activities could be carried in different types of rooms. If this method seems more anecdotal than rigorously analytic, it at least attempts to consider the houses in their entirety and in human terms rather than as mathematical abstractions. Further statistical analysis could doubtless turn up interesting patterns I have overlooked, but all too frequently the human relevance of such abstractions is either meaningless or trivial.

72

History and Archaeology at Olynthus

c on c lu s i on s The excavation of Olynthus was rapid and in some respects careless, and the quality of its records is not what we would ideally wish for. But to balance out the sometimes bleak picture of those hundreds of men with big picks and Decauville railroad cars, we should remember that today we could never afford to excavate the number of houses and the extent of the city plan that Robinson with his rough-and-ready methods and great resources was able to clear. Certainly we can find faults with Robinson’s methods—but those shouldn’t obscure the fact that his excavation remains an achievement that could not be reproduced today, with all our methodologies and technologies. What he accomplished remains unique in Greek archaeology, or the archaeology of any land: a city extensively excavated, with not just one or two houses from which to derive ‘‘types,’’ but a statistically significant sample of the city; with tens of thousands of artifacts carefully excavated, restored, recorded, and published in detail. Robinson could not have begun to digest all the implications of this mass of information, but his perseverance made it possible for future scholars to do so. Another almost unique aspect of the excavation at Olynthus was the attention paid to even the most mundane details. Few Classical archaeologists in the 1930s would have persisted as long as Robinson did in excavating houses, built of rough fieldstone and usually standing less than half a meter high. And the site produced few of the rewards which have traditionally enticed Classical archaeologists: in his four seasons, Robinson found only two minor pieces of stone sculpture and no temples, theaters, or towering city walls. Yet he did not give up or become bored or careless: instead, the excavation and record-keeping become progressively more careful with every season. And the publications were quick and remarkably complete. In the fourteen volumes of the Olynthus series, Robinson published catalogs of some 11,221 artifacts; three volumes on the buildings of the site; and volumes on the graves, the Neolithic settlement, a detailed study of the coinage, and other basic information. The deficiencies in Robinson’s excavation and record-keeping will cause problems in any interpretation of his work. The quality of information varies from house to house and year to year; there are great gaps in our knowledge, about the architecture, about the stratigraphy, about finds which were not recorded or saved, or which were mended and published but not in the detail we would like. If we make mistakes because we do not have all the information we would like or because the records are incomplete or faulty, we can only remember that the sheer quantity of information may help make up for some of the lack of quality. We may not always be right but we will, I hope, gain more truth than falsehoods. The excavation gave us an unprecedented overview of a Greek city and its houses, and despite its many failings in detail, we should not lose sight of that big picture.

History and Archaeology at Olynthus

73

chap ter three

The Houses Described

The houses at Olynthus take a justifiably important place in the history of Greek domestic architecture. More than 100 houses were completely excavated and published, more than at almost any other Greek site, documenting a range of house layouts and organizations. Moreover, because the city was destroyed so soon after the anoikismos, the houses on the North Hill and the Villa Section underwent less remodeling, and so are more coherent in plan than houses at sites with a longer history. We thus get a clearer look at the partition of the land, at the original designs of the houses and blocks, and at the organization of the city than we do at other urban sites. And finally, the assemblages of artifacts left by Philip’s destruction offer unique insight into the kinds of activities that actually went on in rooms and on how household space was organized. In this chapter I will discuss thirteen of the better-preserved dwellings, including houses of regular and irregular plans, houses which seem to be primarily domestic in nature, and houses in which other activities—trade, small-scale industry, and the like—took place. My goal here is not to describe and exhaustively analyze every or even a large sample of the houses at Olynthus, but to give an account of the variety of ways the houses could be organized, the range of artifacts they contained, how they were distributed and the activities they could encompass, and to raise and try to answer some issues about the organization of Olynthian houses. Rather than generalize about the Greek house, as if there were a single norm for all citizens, rich or poor, farmer or merchant or craftsman, we should look at the range of possibilities, at how general principles, such as the pastas plan, and generally accepted types of rooms, such as the andron or the kitchen-complex, could be adapted to different needs and purposes. The accounts and illustrations offered here are necessarily somewhat schematic, limited in part by the information available in the field notes and

74

publications. For a list of all the objects recorded from the site, the interested reader is referred to the database on the Internet.1

p r i n c i p l e s of de s i g n Houses at Olynthus belong to the ‘‘pastas’’ type, a design of house which is widespread in Classical Greece (fig. 12). This type is commonly contrasted to houses of the ‘‘prostas’’ type, found at Priene, Colophon, Abdera, and other sites.2 The standard house plot at Olynthus is roughly square, averaging about 17.2 m across. Two main axes cross the house from east to west. One axis near the midpoint of the house divides it into two nearly equal parts, the second divides the northern half into two portions. These axes govern the placement of most eastwest walls and pillars, which had to be aligned to support the common roof which ran over the northern half of the entire row of houses. A number of principles govern the disposition of rooms. As so commonly in the ancient Mediterranean, most houses at Olynthus are centered about an open courtyard. The court was nearly always located in the southern half of the house, with the pastas and main suites of rooms to its north. Xenophon explains this arrangement: ‘‘When one means to have the right sort of house, must he contrive to make it as pleasant to live in and as useful as can be?’’ And this being admitted, ‘‘Is it pleasant,’’ he asked, ‘‘to have it cool in summer and warm in winter?’’ And when they agreed with this also, ‘‘Now in houses with a south aspect, the sun’s rays penetrate into the pastades in winter, but in summer the path of the sun is right over our heads and above the roof, so that there is shade. If, then, this is the best arrangement, we should build the side facing south loftier to get the winter sun and the side facing north lower to keep out the cold winds.’’ (Xenophon, Memorabilia 3.8.9–10)3 Such houses are a hallmark of the civilized Greek, compared to primitive men who lived in caves and ‘‘had neither knowledge of houses built of bricks and turned to face the sun nor yet of work in wood; but dwelt beneath the ground like swarming ants, in sunless caves’’ 4 The principle of orienting a house towards the south was perhaps not simply an empirical method of making the most livedin parts of the house as comfortable as possible, as Xenophon implies, but may also have been related to the same sorts of medical and philosophical theories as the principles governing the orientations of cities towards the south and towards cleansing winds.5 The courtyard was important as a source of light for the enclosed rooms of the house. Greek houses were closed off from the public, presenting essentially blank

The Houses Described

75

North Room

“Flue”

pillar partition

Bath North Room

KitchenComplex

Axis Pastas

Pastas Room Axis

Shop

Anteroom Courtyard stairs

Andron

0

5 m

figure 12. Schematic Plan of House, Based on A vii 4 walls to the streets. In most rooms, the few windows that may have opened out onto the street would have been small and high off the ground, to keep prying eyes out (fig. 13). Such windows have been found at a few sites, such as Selinus, where houses were built of stone rather than mudbrick; but windows in mudbrick walls such as those of Olynthian houses were probably similar.6 Because most rooms could not be easily lit from outside, the courtyard and other well-lit spaces would have been both important workplaces and sources of light for other rooms. The conflicting goals of light and privacy forced the designers and inhabitants of houses to compromise. Spaces that were well-lit tended not to be private, while enclosed rooms away from the street, courtyard, and other sources of light would have been dark for the everyday household tasks. We will examine these conflicts and compromises further in the next chapter. Opening onto the north side of the court is frequently a long portico, identified by Graham as the ‘‘pastas’’ mentioned in ancient sources.7 This sometimes runs the entire width of the house; more often, however, one or both ends are divided off into separate rooms. The roof of the pastas was supported by pillars along the central axis of the house. These two open spaces, court and pastas, formed uni76

The Houses Described

Distribution of Early Hellenistic Macedonian Royal Coins

6 coins 8 coins 18 coins

single coin 2 coins 3-4 coins 5-8 coins 9-18 coins

2 coins

0 4 coins 4 coins

2 coins

Agora

1 coin , surface

50

100 m

[To view this image, refer to the print version of this title.]

figure 13. Apulian Calyx Krater Showing Slit Window in Upper Left, by the Painter of Athens 1714. Cambitoglou and Trendall, The Red-Figured Vases of Apulia (Oxford, Clarendon Press, 1978) 212, no. 152 fying elements of the house, physically linking and illuminating the rooms around them and socially mediating among the different activities and groups that made up household life.8 Most of the other rooms or room complexes of the house open directly onto either the court or the pastas. The organization of rooms in typical Olynthian houses is thus paratactic or nonhierarchical: that is, there are relatively few ‘‘back rooms’’ on the ground floor, to which access is restricted by intermediate spaces. Nor is there a single dominant room or axis, as there is in the prostas type of house, as well as in the typical Roman house.9 The Houses Described

77

Rooms A number of types of rooms were arranged around the court and pastas. These are normally identified by archaeologists on purely architectural grounds: by their size; placement in the house; access to light, air, and the exterior; the treatment of floors and walls; and by specific architectural features. They have been discussed at length in Olynthus 8 and elsewhere, and extensive description of their architectural characteristics is unnecessary. It is important, however, to distinguish between the identification of a room as defined by its architecture—courtyard, pastas, andron, kitchen, flue, etc.—and the use to which a specific space was actually put. In a modern American house, for instance, the room we call the ‘‘dining room’’ may well be used primarily as a work and play space, while the family both prepares food and eats in the kitchen. Ancient houses were probably just as flexible in their use of space. Study of the artifacts demonstrates that architecturally similar rooms could be used in many different ways: cooking was not always done in the ‘‘kitchen-complex,’’ for instance, and the pastas and courtyard could be used for many different purposes. By designating a particular (architecturally defined) type of room as a ‘‘kitchen,’’ therefore, we should not conclude that each specific room of this type was used to prepare food. Moreover, the use of space probably changed according to the season and weather. A brief review of the different types of rooms may help prepare for a more detailed account of some selected houses and their contents, which gives a more realistic picture of the variety of layout and artifact assemblages at Olynthus. A more general summary of rooms and the organization of household activities will follow in chapter 4, which will take account of both architecture and contents of the houses and try to consider both the norms and the variety of uses of household space. The Courtyard The courtyard formed the focus of the house. Normally the largest single space in the house, it served to link the different parts of the house together and to illuminate them. As Greek houses had few windows to the outside, the courtyard provided light to the enclosed rooms as well through windows. Courtyards at Olynthus ranged in size from very small spaces of 10–15 m2 to large areas of 100 m2, or roughly 3 to 34 percent of the total area of the house. Some houses with particularly small courtyards, such as the House of Many Colors or A iv 9, would have been rather dark and so made special arrangements to light specific rooms which would otherwise be only dimly lit. The court was often paved with cobbles or pebbles; a few houses have cement or even mosaicked courtyards, however. They were often equipped with drains to lead rainwater to the street; some houses had cisterns in the court or nearby 78

The Houses Described

rooms, and others had pithoi in the corners of their courtyards to collect rainwater from the eaves for washing and other purposes. Pastas and Porticoes Most Olynthian houses had at least one portico opening onto the court, providing a space sheltered from sun and rain yet well lit by the courtyard. Many houses had only a single, wider portico north of the court, conventionally referred to as the pastas. Some nineteen houses had additional porticoes facing the open court, and a few had complete peristyles. The roof of the portico was generally supported on wooden pillars; occasionally columns were used, and at least one house, A 3, had stone pillars. The floors of the pastas were usually of earth, occasionally pebbled, but five houses had cement-floored pastades and one had a fine figural mosaic. Although the opening to the court provided the main source of light to the pastas and rooms in the northern half of the house, this opening was deliberately diminished in some houses by walling up the intercolumniations between the pillars or by raising the pillars on low foundations. This would have made a more sheltered and enclosed space. It is unclear whether this was done for reasons of privacy, for more shelter from the elements, or for other reasons. Bright but sheltered, the pastas would have been a convenient working space, as is indicated by the quantities of artifacts found in the pastades of a number of houses. North Rooms, South Rooms, and Pastas Rooms A row of rooms regularly occupied the northern quarter of the house, opening in a row onto the pastas. They were generally about 4.5–5.0 m from north to south, this dimension determined by the line of the roofline of the block. Some of these were equipped with special architectural features which seem to mark them for particular purposes: cement floors with raised borders for an andron, a pillar-partition dividing one or two spaces off the short end for a kitchencomplex, etc. Other rooms are not specialized, however, and cannot be identified or classified purely on architectural grounds. They can vary in width from narrow spaces, almost closets, to some of the largest rooms in the house, and in appointment from very plain to quite fancy, with painted walls and cement floors. Graham described these rooms as ‘‘living apartments’’ or διαιτητήριαι.10 I prefer the more neutral ‘‘North Room’’ because the size, appointment, and contents of these rooms attest a wide range of uses. Likewise, many houses have rooms in the southern half of the house which are best described by their position in the house, rather than their function—hence ‘‘South Rooms.’’ These seem to be more often shops and work areas rather than living rooms; but there is a great deal of variation. Finally, some houses have a room at one end of the pastas, which I have designated the ‘‘pastas room.’’ The Houses Described

79

The Andron Perhaps the most distinctive type of room in Olynthian houses is the andron or men’s dining room.11 This was typically square, measuring about 4.75 × 4.75 m, although larger and smaller ones are found as well. The door of the room was located off center, in order to accommodate dining couches around the room, with the short end of a couch along the wall on one side of the door, and the long side of a couch on the other. The floor was cement or occasionally mosaic, with a raised platform around the sides for the klinai or dining couches. Indeed, it is the presence of this cement floor and platform which definitively identifies these rooms as androns. Other rooms without these features might have served the same purpose, but if so, it is impossible to identify them. Androns were the most lavishly decorated rooms in the house, with painted walls, often in more than one color, and occasionally mosaic floors; if only one room in a house had a mosaic, it was almost inevitably the andron. The andron was often entered through an anteroom, making it one of the few exceptions to the general principle that rooms were accessible from either the pastas or the courtyard. The andron was usually placed adjacent to an outside wall and often at the corner of the house, presumably so that it can be lit through larger windows.12 Again, this makes an exception to the usual principles of private, inward-facing organization and emphasizes the semipublic character of these rooms. Kitchen-Complex A second distinctive suite of rooms characteristic of many Olynthian houses was the kitchen-complex (fig. 14). This typically consisted of a large room (averaging 4.6 × 5.6 m), sometimes with a built stone hearth, and one or two smaller rooms located off one of the short sides, sometimes separated from the main room by a row of pillars. One smaller room was often paved with stone slabs and usually had a separate door to the court or pastas; the other, if present, often contained a bathtub. The preserved examples differ in many respects, but there is enough similarity to recognize this complex as a specific architectural type although, interestingly, one which is almost exclusively confined to this site. These rooms have been given various names and interpretations by the scholars who have studied them.13 The conventions are complicated by the fact that the rooms have no proper modern equivalent (‘‘kitchen’’ does not properly describe the use of any of them), and the ancient terminology is ambiguous. I will refer to the largest room (which Mylonas and Graham label room I, and Mylonas as the ‘‘oecus’’) as the ‘‘kitchen’’; the smaller room often separated from the kitchen as the ‘‘flue’’ (Mylonas and Graham, room II; Mylonas, ‘‘kitchen’’); and the smallest room, which often contained a bathtub, as the ‘‘bath.’’ The names are rather arbitrary; actual uses of these rooms will be examined in greater detail below. The kitchen was often separated from the flue by a row of closely spaced pillars, 80

The Houses Described

bath

hearth

“flue”

“kitchen”

A

A

plan

isometric cut

section A-A

0

5 m

figure 14. Reconstruction of Kitchen-Complex in House A vi 6 sometimes supported on or partly blocked by a rubble foundation. This ‘‘pillarpartition’’ is found in almost two-thirds of the excavated kitchen-complexes and usually in houses which preserve other evidence for two stories, such as a stairbase. As Graham argued, the pillars must have supported the wall of a secondstory room over the main room of the kitchen-complex, while the flue was open through the second story of the house (see figs. 14, 18, 23). Shops A number of houses had rooms which opened directly onto the street, and these have been identified as shops or workshops. In some cases the shops also have doors to the interior of the house (for instance in A vii 4); in others, there is no direct connection. These rooms probably served purposes somewhat separate from the normal household activities, for instance, as shops or workrooms (ergasteria). It is not always certain even that they belonged to the same household which owned the rest of the house: they could have been rented out or sold to other households.14 Any room which opens directly onto the street, and which is not the main entrance into the house, therefore, I have called a ‘‘shop,’’ although The Houses Described

81

again, I do not mean to imply that all these rooms were necessarily used for retail trade. The issue is discussed further below (chapter 6, ‘‘shops’’). The Second Story Many houses seem to have had a second story over at least part of the house. Evidence for the second story consists primarily of stone stairbases preserved in many houses, and, less directly, of the pillar-partitions in kitchen-complexes, which must have supported a wall above the suite.15 All the houses on a row or block tended to have the same number of stories, in order to carry a common roofline (fig. 15). For instance, four of the five houses of the southern row of block A vii preserved stairbases; this row seems to have had a second story, while the row to the north did not. Five of the houses in block A vi had stairbases, and five had pillar-partitions in their kitchens; most probably all the houses on this block had a second story. In block A v, however, only houses A v 6, A v 9, and A v 10 had stairbases. A v 6 was extensively remodeled, and the second story was probably added at this time. The two houses at the east end of the block, like house A iv 9 across the street, were in a slight hollow in the hill here, where the land drops off steeply enough that a common roof could not have been carried across. These houses therefore could have a different number of stories from their neighbors. Issues such as determining the number of stories in a block or row of houses would have to be agreed on by the builders before construction could begin; this and other evidence points to social ties among the inhabitants of a block (below, chapter 5). The second story is referred to fairly frequently in literary sources, and it is commonly associated in modern literature with the women’s quarters, the gynaikonitis. The ‘‘Type House’’ The general similarity among many houses at Olynthus has led many scholars to deal with the ‘‘Olynthian house’’ in an abstract way, using a single house as a type example, or even constructing from those general principles a hypothetical original design which all the houses followed. Walter Graham chose house A vii 4 and the Villa of Good Fortune as typical Olynthian houses, although he discussed variations and changes in design at some length. George Mylonas chose the House of Many Colors as his ‘‘type house.’’ 16 In their influential book Haus und Stadt im klassischen Griechenland Hoepfner and Schwandner take this principle to extremes, arguing that Classical Greek houses were originally identical or differed in only minor ways, this identity reflecting the Greek virtue of isonomia, equality of all citizens before the law. They attribute the standardized house plans to civic ordinances which required citizens to construct their houses according to a single type figure 15. Distribution of Two-Story Houses 82

The Houses Described

Distribution of Two-Story Houses Stairbase certain Stairbase uncertain

P

Pillar-partition

0

50

P P P P

P P P

P P

P P P P

P

P P

P

P

P

P

P

100 m

plan, and explain differences among the houses as the result of later modification and rebuilding.17 In spite of their many common features, however, the houses at Olynthus are a diverse group. Some adhere to the ‘‘standard plan’’ while others never did. Certainly many of the houses were remodeled, some extensively, but this does not begin to account for all the variety found among them. Robinson excavated many of the houses to bedrock, including ‘‘irregular’’ houses such as A 11–A 13, and found no trace of earlier, more regular construction from which these houses might have diverged. And even among the ‘‘standard’’ houses there is considerable variation in the number and types of rooms, the amount of space allocated to different sorts of rooms, special features and installations, decoration of walls and floors, and so forth. Moreover, this variation is not simply random but reveals patterns in the physical makeup of the city and in its social organization. For instance, many of the ‘‘irregular’’ houses cluster around the square at the south end of the North Hill—the agora of the city—and there are significant differences in the architecture of houses on different blocks, which suggests how land was divided and distributed, blocks laid out and built. To study only a single ‘‘type house’’ flattens the richness and variety of this city. At most urban sites, the meager number of excavated houses forces this normative approach upon scholars. But with more than a hundred excavated houses, Olynthus is one of the very few Greek cities where we can look at a range of alternatives rather than a single standard, and put the scholarly construct of the ‘‘type house’’ to the test. Moreover, patterns in the architecture and contents of houses are related to social and economic patterns among the households that lived in them. By looking at the houses as individual dwellings which served the needs of particular households, as places where real people lived, worked, and interacted in and with the spaces around them, rather than as expressions of abstract cultural and political ideals, we can get a fuller idea of how houses were actually organized, how spaces were used. This variety among the houses, households, lifestyles, and economic strategies will be the focus of much of this book.

de s c r i p t i on s of h o u s e s The thirteen houses described here represent a small fraction of the variety of household organization and equipment at Olynthus. The selection of houses was chosen to include examples of both regular and irregularly planned houses, houses with different sorts of artifact assemblages and facilities, houses where different sorts of activities and patterns of organization are represented. By analyzing not only the architecture but also the finds, we can see patterns in how household space was actually used, rather than relying on conventional identifications of rooms and type houses (fig. 16). 84

The Houses Described

The House of Many Colors The House of Many Colors (F −ii 9), in the Villa Section of Olynthus and so one of the later houses at the site, was chosen by Mylonas as the typical Olynthian dwelling (figs. 17, 18).18 Although the house is somewhat unusual in design, the good preservation of both its architecture and finds, and its careful excavation and recording by Lloyd Daly in 1938 make it an excellent introduction to the Olynthian houses and their artifact assemblages. The house was built partly on a terrace cut into the slope of the hill. To the west of the house, the bedrock rises almost to the modern ground surface, more than a meter above the ancient floor level. As a result, the west part of the house was quite deeply buried while the eastern part was shallowly buried and partly eroded away. Numerous traces of fire, which blackened the stucco of the walls, reddened the mudbricks, and left layers of ash in many of the rooms, attest to the violent destruction of the house, while large numbers of complete vases, stone and metal objects, pithoi, and other relatively valuable objects found on the floors comprise an unusually large destruction deposit. The field notes are specific about the stratigraphic position of most artifacts, so that floor assemblages can be reconstructed with some confidence. The house contained a rich assemblage including some ninety-eight complete vases, eight lamps, twenty-seven coins, seventy-six loomweights, two grindstones, five terracottas and fragments of at least fourteen others, two portable altars, and a variety of hardware and other miscellaneous items. These allow us to consider the kinds of activities that went on in the house and to gain a fairly detailed picture of the organization of this dwelling. The house was named the ‘‘House of Many Colors’’ after the quantity and variety of painting on its walls and floors—the most in any house excavated at the site. The walls of the court were molded and painted yellow. The pastas had red walls with a white baseboard. Room c, the cement-floored North Room, was also painted red with a white baseboard, and its cement floor was yellow, except for the red catchbasin in the southeast corner of the room. The lower walls of the exedra (l) were painted blue with incised vertical lines, the upper walls (or upper story?) yellow. Even the kitchen was plastered and painted, with a white baseboard separated from the red upper walls by a projecting blue band—one of only two painted kitchens among all the excavated houses. The andron had the most elaborate decoration, although unfortunately preserved only in fragments. It had a yellow baseboard, divided into panels by incised lines, and red upper walls, separated from the baseboard by one or more projecting blue bands. Fragments of a floral band with palmettes are preserved, one of the very few examples of any figural decoration from the site.19 The anteroom was similarly decorated but painted in yellow and orange. In all, this is the most elaborately painted house excavated at Olynthus. The house was entered from Avenue G through a ‘‘prothyron’’ entrance, with the door set back from the street behind a shallow vestibule.20 To the south of The Houses Described

85

These illustrations do not attempt to represent every artifact found in the houses or the exact positions of the artifacts. The artifacts shown are simplified symbols; see the text and database for more complete descriptions. Symbols are not to scale.

Architectural Key House wall

Cement floor

Low foundation

Cobble floor

Worked block or stairbase

Symbols for Artifacts: Cluster of loomweights, probably fallen from loom

Metal vessel (usually only handles preserved)

Portable altar

Cluster of loomweights, probably in storage or debris

Coin (only shown if in unusual quantities)

Pithos lid or table top (stone or terracotta) disk)

Terracotta figurine or plastic vase

Cluster of bosses, from furniture or door?

Grindstone, saddle quern (lower, upper, or both)

Mold for terracotta figurine

Louter

Cluster of nails

Weight (bronze or lead, sometimes inscribed)

Fibula, other jewelry; mirror

Olive crusher (only orbes found)

Pithos

Scales or scales pan?

Storage amphora

Askos; duck askos

Mortarium or coarse spouted basin

Krater

Kantharos; skyphos

Cooking pot

Hydria

Cup, bolsal, etc.

Jar

Oinochoe or jug

Plate

Olpe

Fishplate

Thymiaterion

Table amphora/ pelike

Shallow bowl, bowl

Lamp

kothon

Lekanis & lid

Lekythos; guttus

Saucer or saltcellar; miniature cup

Stone mortar, usually embedded in floor

M

figure 16. Key to House Plans

Brazier

rooms under construction: amphoras of sand and lime; red and blue pigment & grinders; pebbles for mosaic

a

Street -i b

c

d

e

h

g

f

i

Avenue G

bedrock

plus epinetron and spindle whorl

j

ash & bone

k

l

staircase

altar

0

m

5 m

figure 17. The House of Many Colors the rather small court was a deep portico or exedra (l), which Robinson labeled a ‘‘summer living room.’’ Three North Rooms open onto the pastas (rooms a, b, and c). An andron occupied the northeast corner, entered from an anteroom (d, f ); and a kitchen-complex occupied the southwest (rooms k, g, and h). Court Like almost every Greek house, the House of Many Colors was centered on its courtyard. This served as a common working and social area, as well as a principal source of light for the private rooms of the house, which focused inwards onto this open-air but enclosed space. The court of this house, however, was among the smallest at Olynthus (3.58 × 5.90 m), and with two stories of rooms to both the north and south, it cannot have been as light and airy as larger courtyards in other houses. At the west end was a built altar, covered by a canopy supported by The Houses Described

87

et

re St -i

Avenue

G

figure 18. Reconstruction of the House of Many Colors two bases, which further reduced the available space and light. The court was thus one site of household cult, as implied by literary sources, but two portable altars from the pastas show that the court was not the only cult site in the house.21 There were relatively few finds in the courtyard, suggesting that unlike some other houses, this was not an important workspace. The finds include the mouth of a storage amphora with stamped handles; a fine askos with molded decoration from the south end of the court; seven loomweights (plus four more from fill above the floor); and various hardware and spikes, probably from the woodwork of the colonnades. None of this makes up an easily interpretable assemblage, and some of it may be refuse rather than usage deposits. Pastas Opening onto the north side of the court was a long portico, the pastas (room e). This was semienclosed: rather than being completely open to the court, the pillars supporting the second-story gallery were set on a low foundation or screen wall, leaving a door at the northwest corner of the court.22 The pastas was thus more sheltered from the elements than an open portico would be, but also correspondingly darker. As was common in Olynthian houses, this sheltered but well-lit room was a main locus of domestic activity, to judge from the quantity and types of artifacts 88

The Houses Described

found there. In the southwest corner were two portable altars (arulae), suggesting that like the court, this room was used for household ritual. A marble louter was found nearby, together with a storage amphora, perhaps used both for ritual ablution and ordinary household washing and food preparation. Louters are found in association with portable altars in at least five houses at Olynthus, but they also seem to be used for the preparation of food in other cases; the same object could probably serve a variety of purposes. Four more storage amphoras in the east end of the pastas perhaps stored water or foodstuffs. A number of table vases were found in the pastas.These included a black-glazed fishplate, three black-glazed plates, a shallow bowl, red-figured bell krater, a redfigured pelike, a skyphos, a miniature krater, an unslipped lid, and a lekythos. Many of these were found in the western part of the room, probably stored there on shelves. Some twelve iron bosses and three iron spikes, some from the western end of pastas, probably belonged to a piece of furniture, perhaps a chest for storage. Most of the objects found in the pastas were at the more dimly lit ends of the room, and they were probably being stored there, convenient to the better-lit areas that were more suitable for household activities. Kitchen-Complex In the southwest corner of the house was a group of three rooms forming a ‘‘kitchen-complex’’ (rooms k, g, and h). In the center of the large room (k) was a built hearth, set slightly skew to the walls of the room. The hearth was filled with ashes but contained no other finds, bones or pottery. Three storage amphoras were found in the western end of the room, perhaps used for storing water and other materials for cooking, but otherwise the room contained only a scatter of artifacts: a lead disk with lotus and palmette decoration, a loomweight, a saucer, three coins, and an arrowhead. A bowl and two red-figured sherds came from levels slightly above the floor; they may belong with the destruction deposit or have fallen from the second story. Adjoining the kitchen was a long narrow room (h), of the sort that Graham called the ‘‘flue.’’ This was separated from the kitchen (k) by a low foundation, on which were set bases for three pillars—a pillar-partition.23 The room was entered from the courtyard of the house rather than from the kitchen. The flue was of an unusual design: instead of the earth or stone slab floor found in most kitchencomplexes, the floor was of bedrock with a trench running down the center of the room. This trench was filled with a deep layer of ash containing many bones of cows, sheep, goats, pigs, and deer. Robinson suggests that the trench formed a kind of broiling pit, which was filled with coals to cook meat on spits.24 Presumably the cook sat or squatted on the rather narrow ledges on either side of the trench while cooking, although it still seems a somewhat hazardous arrangement. Fires may instead have been limited to cooking braziers or other devices, and the pit used to dispose of the ashes afterwards. In any event, the flue was apThe Houses Described

89

parently used for cooking in this house as in many others, while the hearth in the ‘‘kitchen’’ was used primarily for heating the room in the winter, as these hearths rarely contain bones or cooking equipment.25 The flue contained a rich assemblage of artifacts. A few of them, such as a spit support, are easily understood as cooking equipment, but there were also a great many other table vases, hardware, terracottas, and other miscellaneous finds which are not so easily explained. Some of this material was quite fragmentary. Twelve terracottas found in the flue, for instance, were all broken pieces rather than intact objects and so had perhaps been discarded in the ashes, while the eleven loomweights were also probably discarded here. But among the more than twenty vases found in this small space were two intact duck askoi, one intact and two nearly complete plates, a nearly intact fishplate (battered around the edges), an intact, unslipped table amphora, a complete (?) storage amphora, three intact or nearly intact saucers, a lid, a juglet, a one-handled cup, a fragmentary redfigured askos, and six almost complete or intact lamps, including a large, open lamp with two nozzles.26 The room also contained a large bronze lid, perhaps for a cooking pot, a fibula, and a spiral ornament. So many intact and complete vases, as well as the bronze and jewelry, must have been stored, rather than discarded, in this room and so form a use (or storage) assemblage.27 As was common, the flue was not used exclusively for cooking but also for storing kitchen and other household equipment. Iron and bronze hardware found in the flue, including bosses, spikes, and metal straps or bands, might have belonged to shelves or fixtures on the walls, forming a convenient place to store kitchen equipment used in cooking. The scarcity of cooking pots and other such coarse vessels is probably due to lack of interest in collecting these vases during excavation. The flue did have a larger proportion of plain and battered vases than other rooms of the house, and few figured or specialized vessels.The difference between the ceramic assemblage of these rooms and that of suite a-b is striking. Next to the kitchen and flue was a small room (g) with an earth floor. In other kitchen-complexes, corresponding rooms usually contained bathtubs or holes where the tub had been removed. This room, however, contained only small fragments of a bathtub. The tub might have been set on top of the floor, rather than into it, and later removed leaving no traces; or perhaps it was never installed, and the fragments are unrelated to the use of the room. The room contained a set of grindstones, an olpe, a pelike, and a lamp, equipment for use in the kitchen, stored here out of the way but convenient for use in the main room. Again, we see multiple uses of these rooms, both for their primary purpose (cooking, bathing) and for storage of household equipment. Suite a-b Two of the North Rooms at the House of Many Colors, rooms a and b, were not separated by a wall but by a pillar-partition like that in the kitchen-complex. 90

The Houses Described

Room a was roofed, but when excavated, the fallen rooftiles extended only partway into room b, showing that this room was at least partly unroofed. Robinson and Mylonas offer no explanation for the joining of these two rooms. The suite is not large enough to have needed internal supports to hold the roof, and the pillars would not be otherwise needed. Graham believed that the suite served as a second kitchen, with b as the flue, rising through the second story to the roof.28 It is, however, distinct from the kitchen-complex in both decoration and contents. Graham’s reconstruction seems basically correct, though, in that the smaller room probably was open to the sky. But rather than for letting smoke out, this could have been to let light into the larger room. This smaller room could have served as a light well, a sort of a miniature courtyard to let light pass through the second story and into the main room. This house has one of the smallest courtyards excavated at Olynthus, so very little light would have reached the center of the house. Moreover, the pastas had a relatively short opening onto the court, which was further reduced by raising the pillars up onto a low foundation, effectively blocking off the lower part of the pastas. The house would therefore have had significantly less light from the usual sources—court and pastas—than other houses at Olynthus. To compensate for this, an alternative source of light was built in the form of a light well, drawing on an already familiar room type, the kitchen-complex (fig. 18). Fires on portable braziers in room b could heat the main room as well, and the smoke evacuated through the opening above. These suites with light wells are unusual, identifiable in perhaps seven houses, of which this and the suite in the neighboring Villa of the Bronzes are among the clearest examples. As preserved, the suite had earth floors and white plastered walls. But it was apparently being redecorated with fancy painted walls and a mosaic floor when the city was captured by Philip. The main room contained an amphora full of sand for making cement; red and blue pigments stored in a small vase and piled on the floor, probably for painting walls; two small pigment grinders; and a pile of blue stone pebbles, probably intended for a mosaic floor. The light well contained another storage amphora, this one full of lime, and a pile of lime lying on the floor nearby. A terracotta ‘‘pithos lid or table top’’ and a terracotta tray partly under the lime pile might have been used for mixing mortar, and a pile of black pebbles for mosaic. When finished, this suite would have been the best-lit and most elaborately appointed space in the house. The analogous suite in the Villa of the Bronzes (rooms a and b) was also painted and mosaic-floored (see below). But even while it was being refurbished, the suite served as an important workplace. The larger room contained forty-one loomweights, about the right number for a loom, an epinetron and a spindle whorl, and thirty-four more loomweights were found in the light well. Well lit and airy yet sheltered, the larger room would have been a comfortable and convenient room for weaving. This room also contained many table vases: two fishplates (one red-figured), two plates, two hydrias The Houses Described

91

[To view this image, refer to the print version of this title.]

figure 19. Red-Figured Lekanis Lid, from the House of Many Colors, Room a. Women in the Gynaikonitis. Olynthus 13, no. 64, pl. 87 and a pelike, a lekanis and four lekanis lids (three of which were red-figured: fig. 19), another red-figured lid, two lekythoi, a guttus, a group of twelve miniature cups, six saucers, a skyphos (?), and two odd unslipped vessels with broad flat flanges around the bottom. The room also contained one complete terracotta female head and a female protome. There is a high proportion of red-figured and generally fine vases here, very different from the group in the flue. Most of the vases were for women’s use, lekanides and the like, as opposed to sympotic vases, such as kraters, which were intended for men. The miniature cups are similar to votives found in sanctuaries, and may have had a ritual function, another activity associated with women. 92

The Houses Described

Room b contained a similar but smaller assemblage: a hoard of thirty-four more loomweights, an olpe, a cup, a lamp, a saltcellar, and two unslipped vases. Two iron and one bronze spikes might belong to furniture or to the woodwork of the pillar partition. This well-lit area then would seem to be a locus of female work, including weaving and ritual. The twelve miniature cups, terracotta protome, and female head probably are to be interpreted as another ritual assemblage, again primarily female activity. A close parallel is found in the corresponding suite in the Villa of the Bronzes (below). The table vases might indicate that this area was also used for eating, or this assemblage of fine and red-figured pottery may be special equipment for festivals, stored safely here as Ischomachos advises, while day-to-day tablewares were kept in the pastas. It is tempting to identify this suite as the gynaikonitis or women’s quarters described by Xenophon and others; we will return to this issue below (chapter 4). Room c was also finely decorated, with cement floor and painted walls, but it contained only a few miscellaneous objects: hardware, a few coins, bronze fragments, and the rim of a pithos. Like so many North Rooms, the function of this space remains uncertain: its fancy decoration might suggest that it was another room for socializing, akin to the andron, or perhaps it served as a bedroom, whose perishable contents—wooden furniture, clothing, and bedding—have all disintegrated. The catchbasin in the southeast corner suggests that the room was originally designed to be used and cleaned regularly, rather than simply as a storeroom. Andron and Anteroom In the northeast corner of the house was the andron (d), a square room 4.71 m on a side. This distinctive type of room is easily identified by its cement floors with a raised border about 0.9 m wide around its edges. Seven dining couches or klinai would have been set on this border around the walls of the room. The border of the andron of the House of Many Colors was of yellow cement, and the center portion of the floor was made of pieces of marble set in cement. A shallow gutter ringed the central part of the floor to carry off spills and water for cleanup. The andron was entered from a relatively large anteroom (room f, 3 × 4.7 m). The entrances to both these rooms were not preserved, but the reconstruction offered here seems the only possible one, since the wall between the andron and room c is well preserved. This corner of the house was rather shallowly buried, and perhaps partly for this reason few finds are preserved here. But in most other Olynthian houses, androns contained relatively few artifacts. The andron itself contained only two iron spearheads, reminding one of the weapons in the andron of Croesus’s palace (Hdt. 1.34) or of the Persian king (Hdt. 3.77). At the south end of the anteroom was found a cluster of sixteen bronze decorative bosses with carbonized wood adhering to some of them, probably the remains of a chest or cabinet used to store equipment for the symposium, for in this area were found The Houses Described

93

a fishplate, two plates, and fragments of a krater.29 The scarcity of cups here and elsewhere in the house is notable, and might suggest that drinking vessels of bronze or more precious metals were used in favor of pottery, and that these were looted when the house was destroyed. Exedra The exedra (l) south of the court is unique at Olynthus. Open to the courtyard through a colonnade, it formed a sort of second pastas, perhaps replacing the pastas proper in some respects, since the semienclosed pastas was relatively poorly lit. Its cement and pebble floor was almost bare of artifacts. The room is distinguished by an unusual deep pit at its southern end, whose mouth flares like a funnel. The court of this house might have originally extended all the way to the south wall, and the exedra (l) might be a later modification, divided off from the court to form a separate, closed space. The cement floor of the room is slightly higher than the cobbles of the courtyard, as if it had been laid on top of the cobbles. In its original design, then, the house would have been similar to the Villa of the Bronzes (fig. 20). This feature therefore might have originally been a cistern located in the southern part of the original courtyard, which was left open even after this part of the court was roofed.30 A door in the south wall of this room might be the remains of a door leading to the drainage alley. Houses in the north row of a block frequently had access to the alley, either through a door or simply by omitting part of the south wall of the house. Storeroom In the southeast corner of the house was a room (m) whose floor was about 0.9 m below that of the court and exedra. At least four pithoi in this room attest its use as a storeroom or πιθεών. The contents of the pithoi were not preserved, but they most likely held agricultural products. This partly subterranean, cool room would, as Xenophon points out, be suitable for storage of grain and wine. Such storerooms are common among houses in this part of the city (below, chapter 6). The Second Story The house almost certainly had two stories. A rubble foundation spans the space between the east wall of the exedra and the easternmost colonnade base. This could have carried a stair running from this foundation southwards along the west wall of the exedra, supported at its south end by a pillar set on the base against the exedra’s south wall.31 Pillar-partitions like those in the kitchencomplex and the light well a-b imply a second story; the pillars must have been placed to support some heavy weight, and it is difficult to imagine what else this could be besides a second-story wall. Finally, over the cement and pebble floor of the exedra were found fragments of another mosaic floor, one piece about 1.3 m 94

The Houses Described

in length, which probably fell from this second story.32 The two floors were separated by about 20 cm of earth and mudbrick, and mudbrick adhered to the upper floor; an iron nail was found in this intermediate mudbrick layer. Common practice according to the literary evidence was to have the northern side of the house higher than the southern, and this is the more usual situation at Olynthus, to judge from the positions of stairbases. Here, however, the second story seems to have covered both north and south halves of the house, although the heights of the rooms may have differed. The first story seems to have been about 2.5 m high, to judge from plaster fallen in the pastas wall; this corresponds well with the run of the stair in the exedra. Few objects, however, can be attributed to this second story, and its reconstruction and use remain problematic. Except for the fragments of mosaic in room l, the stratigraphy of most rooms is quite simple, and there was no recorded architectural remains of a second story. Below the topsoil and a deep layer of fairly sterile fill was a dense layer of rooftiles. Directly under the tiles was a burned layer, which rested directly on the floor. Fragments of roofing or ceiling materials were found in the pastas, consisting of burned clay with impressions of reeds, but their exact location and stratigraphic position are uncertain. The majority of the objects found in the house are explicitly described as lying on the room floors. Some might have fallen from the second story and become mixed with the first-floor assemblage, but for most there is no reason to think that they fell from above, and indeed the excellent preservation of fragile objects like vases suggests the opposite. A few finds were noted as coming from the layer of rooftiles, but I suspect that most of them belong with the assemblages of the ground floor and rested somewhat above the floor because they had been stored on shelves or in furniture. For instance, a number of vases were found in the tile layer in the west part of the pastas, including a red-figured krater, a skyphos, a plate, and a lid. Although these might have dropped from the second story, it seems more likely that these vases belong with the floor assemblage of the west part of the pastas itself, to which they are quite similar, than that they were being stored in the open gallery above the pastas, coincidentally above a similar assemblage in the floor below. A number of vases, including an intact table amphora, a saucer, a one-handled cup, and a terracotta standing female, were found in a layer of tiles in the flue, and an intact duck askos and part of a juglet were found above the tile layer. But this room had no second story above it, and it is unlikely that vases would have survived such a fall. The tiles may have slid from adjacent rooms when the roof collapsed, or the flue might have been covered with a raised, tiled section of roof; the method by which these rooms were covered is uncertain (below, chapter 4, ‘‘Specialized Spaces: The Kitchen-Complex’’). The same phenomenon of objects lying above the floor and tile layers is found in other houses, such as the Villa of the Bronzes. This situation is typical of the houses at Olynthus for which there is architecThe Houses Described

95

tural evidence for a second story. In no case is there clear evidence in the stratigraphy for artifact assemblages which might have fallen from the upper rooms. Fragile objects like vases may well have been broken, scattered, and not recognized or mended by the excavators, but the absence is still remarkable. The use of second-story rooms is therefore problematic, given the paucity of artifacts which can be securely assigned to them. But this house was well preserved and fairly carefully excavated, and if there had been really obvious assemblages or architectural evidence, they would have been observed and recorded. The negative evidence, therefore, suggests that the rooms were of light construction and were not for storage of household equipment, for instance, or for many domestic activities.33 Graham points out that this house has a notable lack of rooms which might have been permanent bedrooms, and suggests that these might have been located on the second story.34 While it is possible that there was no established sleeping areas and that the household slept wherever was convenient, some literary sources, for instance Lysias’s famous description of the ‘‘little house’’ of Euphiletus in On the Murder of Eratosthenes or Xenophon’s Economics, suggest that there were more or less established bedrooms, even if these could be rearranged at will to accommodate different and changing household situations (below, chapter 4). Beds and other furniture leave few traces in the archaeological record here—indeed there are no traces even of the klinai which were ranged round the andron—and neither would the sorts of bedding, textiles, clothing, and other items which might have been stored in rooms used primarily for sleeping.35 In the absence of any evidence for their use, therefore, we may conclude that the upper rooms of the House of Many Colors were generally lightly constructed and contained mainly perishable furnishings and contents, and they were perhaps not used nearly as intensively as the ground-floor rooms. Some may have been used for storage of perishable equipment, bedding and the like; others may have been used for sleeping. Summary Although the House of Many Colors is architecturally unusual, with sophisticated features like the light well and the exedra south of the court, it still serves as a good introduction to the organization of space in Olynthian houses. The courtyard is very small, the pastas unusually enclosed, and the second-story rooms on the south side of the house will have further blocked light to the court and pastas, making the court and pastas less well-lit spaces than in most houses. Perhaps to compensate for this, there are an unusual number of alternative sources of light, particularly the suite of rooms a-b and the exedra south of the court, and these may have taken over some of the activities which were done in the courtyard and pastas in other houses. The range of activities which can be documented in the house is perfectly consistent with what literary sources would lead us to expect in a Greek house: 96

The Houses Described

processing food, cooking, eating, weaving, ritual, storage of foodstuffs, entertainment, and socializing. These activities seem to be fairly strictly spatially organized, as documented by the assemblages of artifacts. But many rooms shared more than one function: washing, domestic storage, and cult in the pastas; weaving, ritual, and more domestic storage in the North Room a; cooking and food preparation in the flue and kitchen; large-scale storage, probably of agricultural products, in the storeroom or pitheon; bathing and storage of food-processing equipment in the bath g. It is difficult to determine where the household ate and slept. Literary sources describe eating in the courtyard of the house, and the andron may have been used by the household as well as by special guests. The house had no workrooms or ergasteria not involved with normal household production like weaving and food preparation. While the owner might have had such workshops outside his house, it is most likely that his main source of income was agriculture, the primary and favored occupation of most Greek citizens. The large storeroom in the southeast corner of the house probably stored the produce of the farm. A gendered division of space may also be distinguished in this house. Some spaces were set aside for women’s work—the kitchen-complex and suite a-b— while the andron was probably primarily intended for symposia, restricted to male guests and their female entertainers. The entrance to the andron seems to have been directly from the entrance hall (j), leaving the rest of the house more private and enclosed. This need not, however, constitute a formal distinction between the women’s and men’s areas of the house, the andronitis and gynaikonitis, at least in the sense used by Xenophon, Lysias, and others. Most of the ground floor seems to have been used for women’s work, and these areas are not closed off from the rest of the house. Although outsiders coming to symposia here did not need to pass through the rest of the house to reach the andron, the house itself is not strictly divided. Such a distinction between men’s and women’s areas is even less apparent in other Olynthian houses (see chapter 4: ‘‘Gendered Divisions of Space’’). The Villa of the Bronzes The Villa of the Bronzes, located two blocks south of the House of Many Colors, is similar to it in a number of respects (fig. 20). It too is well built and well preserved, and had been heavily burned, particularly in the northern part. The number of weapons found in the court and other rooms, including a shield, a sword, three knives, two spearheads, five arrowheads, and thirteen slingbullets, attest to heavy fighting here during the capture of the city, and led to the naming of the house. It was excavated in 1938 by William A. McDonald, then a graduate student at Johns Hopkins on his first trip to Greece.36 When first built, the house probably lacked a neighbor on its west, since the west wall, like the south street wall, was built of cut stone, and a small exThe Houses Described

97

a

b

c

d

shallow square granite basin

ash on floor

e

f

blocked door?

i h

g

reused?

scattered

j shield, sword, 3 knives, 7 slingbullets, 2 spearheads and animal skeleton in courtyard

k

Street -iii

0

5 m

figure 20. The Villa of the Bronzes

terior door led into the storeroom (g). The main door of the house, opening onto Street −iii, was a double door, with a narrow door on the west for people and a wider door, 1.9 m wide, on the east for wheeled traffic; the east threshold was rutted by cart wheels. The court takes the common position in the center of the south side of the house, and the pastas opens to the court through the more usual colonnade of two columns and two engaged pilasters, rather than the semienclosed design of the House of Many Colors. Two stone Doric capitals and a pilaster capital belonging to this colonnade were found near the bases. To the north of the pastas are three rooms, two connected by a pillar-partition and forming a suite with a light well like that in the House of Many Colors. A fourth North Room in the northeast corner of the house (d) was entered through an anteroom (f ) from the pastas and might have served as an andron. In the southeast corner of the house was a kitchen-complex with flue and bath, and in the southwest was a storeroom. The house was well built and appointed. Its south and west walls were built of drafted ashlar masonry, rather rare at Olynthus, and the court paved with cement and large stone slabs, instead of the usual river cobbles. Six of its eleven rooms

98

The Houses Described

were painted, some had molded plaster, and room b had a mosaic floor.The colonnade was fitted with stone capitals, as was the pillar-partition between a and b. The artifact assemblage of the house was rich, although not quite as rich as that of the House of Many Colors. Some forty-nine vases were recorded from the house; the field notes mention ‘‘many vases’’ in rooms a and e, which were not mended or cataloged. In addition to the weapons from the court, the house contained an interesting variety of metal and stone objects, including bronze vessels (a bronze basin and handles from another), implements, jewelry, and tools, two portable altars, and a fine louter. Court The court is of about average size, measuring about 6.4 × 5.9 m. It was drained to the street through a channel and terracotta pipe. Most of the finds here seem to be remains of the final battle for the city: a shield, sword, three knives, two spearheads, and seven slingbullets. The skeleton of a large calf or small cow was also found on the floor of the court, perhaps another casualty of war. Other than the weapons and architectural fragments, however, the court contained only miscellaneous finds: a composite ‘‘dumbbell’’ vase with one closed and one open section, seven coins, a small bronze pick or other tool, a terracotta base or perhaps a portable altar, and fragments of a marble louter. The altar and louter suggest household cult here, as in the court of the House of Many Colors; and likewise we find evidence for ritual in the pastas as well. Pastas As at the House of Many Colors, the pastas was apparently an important workplace. In many respects the assemblages of the two pastades are remarkably similar. In the northeast corner of the pastas was a ritual assemblage: a marble portable altar, and a fine marble louter and base, nearly complete and mended in antiquity (fig. 21).37 The altar was found ‘‘as if it had fallen from a stand against the wall, and it may be noted that we found several chunks of charred wood in this immediate neighborhood which may have come from a wooden stand. The altar obviously was meant to be against a wall for it is plain on one side while on others it has a molding and is smoother.’’ 38 Another portable altar was found towards the center of the pastas, in front of the door to room b. Two stone bases in this corner might have supported the altar or perhaps some more complicated arrangement, although they might also have supported the stairs. A great many vases were found along the north wall of the pastas between the doors to rooms b and c, including an askos, a fishplate and two plates, six saucers, and many other fragments, ‘‘usually broken in large pieces and lying on the floor as if they fell from a shelf or from somewhere above rather than as if this were any sort of a discard dump.’’ 39 This might have fallen from a shelf like that at the

The Houses Described

99

[To view this image, refer to the print version of this title.]

figure 21. Louter from the Villa of the Bronzes. Olynthus 12, pl. 218 west end of the pastas of the House of Many Colors. This chest or shelf also held a hollow bronze instrument with clawlike projections and a few other bronze objects, and other miscellaneous metal objects were scattered through the pastas, most notably a bronze basin or brazier near the westernmost base of the pastas, a finger ring, and a heavy hook. North Rooms Rooms a and b form a suite with light well similar to that in the House of Many Colors. The pillars had no bases but rested on a low rubble foundation; three pillar capitals found in room b must belong to this partition. The main room (b) had red stuccoed walls and a pebble mosaic floor, while the light well (a) had an earth floor, and only its south wall was plastered white. The suite gives a sense of how the corresponding rooms in the House of Many Colors would have appeared had they been completed. Ashes and traces of burning were found in the light well, apparently not only 100

The Houses Described

from the fire of the destruction but from ‘‘continuous fires here’’ during the use of the house.40 These fires probably heated the main room of the suite in the cold Chalcidic winters, while the light well both let in light and allowed the smoke to escape. The suite might also have served as a cooking area during part of the year, although the house had a kitchen-complex as well; many unslipped sherds, perhaps cooking wares, were found on the floor of the light well. A chest or other furnishing stood at the north wall of the main room, attested by eight iron bosses in two sizes. The furnishing probably held perishable objects, perhaps cloth and bedding. The only objects found nearby were two elegant lamps with central handles, one with double, the other with quadruple nozzles. Otherwise, the room contained a rather mixed assemblage: a couple of saucers, two heavy hooks (perhaps attached to the door, where they were found, or holding curtains over the door), a pin with a filleted head, an arrowhead, and three coins. We should perhaps interpret this as a more formal living or entertaining room, lit with fancy lamps and equipped with decorated furniture. In the center of the west wall of the light well was a cluster of twenty nails belonging to another piece of furniture and nearby, two more fancy double-nozzled lamps like those in the main room and two black-glazed plates. Several small iron bosses found in the light well were apparently the remains of a third piece of furniture, as well as personal articles such as a bone spatula and a finger ring with a decorated bezel. At least twenty-five saucers were found near the pillar-partition together with another plate, while many other vases were scattered in this room, and in the southwest corner was a large shallow pot.41 In the northwest corner was an odd egg-shaped pithos sunk into the floor. There are clear similarities between the assemblages of this suite and the corresponding suite in the House of Many Colors. Both contained furniture and probably also the textiles and bedding that formed some of the household’s most valued possessions. They contained other objects usually associated with women: finger rings, spatulas, unguent containers, and miniature vases perhaps used in ritual: twenty-five saucers in the Villa of the Bronzes, twelve miniature cups in the House of Many Colors. The walls of room c were plastered with rough grey plaster, and it was almost bare of finds, except for a pithos in the southeast corner. It probably served as a storeroom. Andron and Anteroom? Two rooms in the northeast corner of the house (d and f ) correspond in position to the andron and anteroom of the House of Many Colors and might have served the same function. The ‘‘andron’’ does not have the usual cement floor with raised border, and its door is not off center as most andron doors are. The rooms were, however, finely plastered, and room d contained part of ‘‘a finely carved rectangular shallow basin of beautiful granite,’’ perhaps sympotic furniture.42 Other The Houses Described

101

than this, however, the rooms contained only a few objects: three coins in the andron, and two coins and a weaving utensil (?) in the anteroom. Androns at Olynthus tend not to preserve significant quantities of sympotic furniture, and so its absence here does not mean that this was not an andron; but in the absence of definitive architectural evidence it is safer not to jump to conclusions. Kitchen-Complex The kitchen-complex in the southeast corner of the house consisted of a large kitchen, a flue separated by a pillar-partition, and a cement-paved bathroom with a tub still in situ (rooms i, j, and k). The flue and bath were separated by only a light partition, which left its impression on the yellow plaster of the bath. Like the kitchen of the House of Many Colors, the main room of this suite contained only a scatter of artifacts. A fragment of a lower grindstone was found here, but this may have been reused, rather than in use here.43 A hinge, a ‘‘keyhole reinforcement,’’ and other hardware are perhaps the remains of storage chest or box for storing foodstuffs. But other finds were scarce: six slingbullets, five coins, six loomweights—not enough for a loom—and a large chunk of lead, worked at one end. The floor of the flue was covered with a layer of ash, charcoal, burned earth, and fragments of animal bones, up to 3 cm deep; this was apparently a cooking room like the flue in the House of Many Colors. But unlike the flue in the House of Many Colors, there were virtually no other finds here, only a coin and a bead. Kitchen utensils were apparently stored elsewhere in the house; if there had been a concentration of vases, even if they were coarse and not saved or mended, McDonald would certainly have noted them. The bathroom (k) was separated from the flue by only a light partition which left no trace, except for the junction of the cement floor of the bathroom and earth floor of the flue. A bathtub was found in situ in the bathroom, and nearby was a large terracotta spouted basin full of ashes, perhaps a makeshift brazier for heating water. Storeroom The other corner of the house was taken up by a large storeroom (g), like that in the House of Many Colors. This contained a huge pithos, 1.7 m in diameter, whose lid was found nearby. The only other finds in the room were a small ring handle and eight coins; but perishable foodstuffs might also have been stored in sacks or baskets which have left no trace. A gap in the outside wall of this room is probably a blocked door, which originally led out not to Avenue F, which was further west, but to an undeveloped house plot, perhaps used as a garden. Later, when a house was built here, the door was blocked. The storeroom in the Villa of Good Fortune had a similar door to the out-

102

The Houses Described

side, although leading directly to the street. These rooms might be interpreted as shops, but they are more likely to be domestic storerooms with outside entrances to make loading and unloading produce more convenient.44 The Second Story As with the House of Many Colors, there is architectural evidence for a second story at the Villa of the Bronzes. A stone stairbase was found along the south wall of the court, shifted out of place; and the kitchen and light well suite had pillarpartitions to support walls on the second floor. The stairs might have climbed the south wall of the court where the base was found, or it might instead have been located in the pastas, supported by the two bases in the northeast corner.45 Very few tiles were found in rooms g, i, j, or k, and the storeroom g might have had a flat roof, which led to a second-story balcony over the pastas from which the rooms over b, c, d, and f could be entered. But as with the House of Many Colors, no objects can be securely attributed to the second-story rooms. Summary In many respects the Villa of the Bronzes is similar in plan and use of space to the House of Many Colors. Although the architecture of the open and semiopen spaces is quite different, the Villa having a larger court and more open pastas, the two pastades have quite similar assemblages and activities, with ritual paraphernalia, vases for eating and drinking. Both houses have well-decorated suites with light wells, which contained furniture, collections of miniature vases, and articles associated with women’s work, although the assemblage in the House of Many Colors was rather richer. But other architecturally similar spaces were used quite differently in the two houses. The flue in the Villa, for instance, contained no pottery or other artifacts at all, only ash and bones. Although used for cooking, like the flue at the House of Many Colors, it was not used to store household equipment. The Villa of the Bronzes does not have a typical andron with cement floor and raised platform, and it is difficult to determine from the artifact assemblage whether rooms d and f served as an andron-anteroom suite. Both houses have large storerooms, and both apparently had second stories although with similar problems in interpretation. House A vii 4 Walter Graham chose this house as his type house for his discussion of the dwellings at Olynthus (figs. 22, 23). The house contains a full range of specialized rooms: a well-defined pastas and cobbled court, a kitchen with bath and flue in the northeast part of the house, an andron and anteroom in the southeast, two

The Houses Described

103

a

c

b

e

d

ash

?

f

M M M M

h

g

j

i k

? M M M

Street vii 0

5 m

figure 22. House A vii 4 North Rooms, a small storeroom off the pastas, and a shop in the southwest part of the house. The house was entered through a prothyron door which led into the court. It had two stories.46 Court The courtyard had about the same area as that of the Villa of the Bronzes (7 × 6.2 m) and was cobbled and drained to the street. The court contained a variety of finds and, unlike the courts of the House of Many Colors and the Villa of the Bronzes, seems to have been heavily used. Sixteen loomweights were found scattered on the floor. These might be the remains of a loom, or perhaps were debris from a loom previously set up in the court. The court also contained a fair amount of domestic equipment: vases, including bowls, a coarse spouted bowl or mortarium, a guttus, a lamp stand, three lekythoi and fragments of many other vessels; a bronze cup or vessel; fragments of four female protomes; hardware and other miscellaneous finds. The exact locations of these finds were not recorded, but in other houses they tend to be set on shelves or pegs on the courtyard walls. Two bronze weights (one inscribed Τ for τέταρτον or τεταρτημόριον, one-fourth of a mina) and a ‘‘lead loomweight’’ (probably another weight) from the court could 104

The Houses Described

et

re

St vi figure 23. Reconstruction of House A vii 4 be connected with commercial activities in the house, as are other weights from this house. Pastas Like the court, the pastas (f ) of A vii 4 was the site of a variety of activities, and their assemblages are generally similar. The pastas contained a number of eating and drinking vessels: a hydria, two kantharoi, an olpe, and a bowl, as well as many other vases, not mended or published. Swinging handles from two bronze vessels were found nearby. Small pouring or unguent vessels included a guttus and two squat lekythoi. Most of these were found in the west part of the pastas, probably stored on shelves or in chests along the wall, as in the houses discussed above. A bronze leaf found with two clamps, and five bronze bosses from the western part of the pastas probably belonged to a chest or other piece of furniture in this part of the room. A stone ‘‘pithos lid or table top,’’ 42 cm in diameter, was also found in the western part of the pastas. These disks could have served a variety of uses, not just as pithos lids but perhaps as portable working surfaces (such as the one found in room b of the House of Many Colors, probably used to mix mortar). Jewelry including an earring and a fibula were found in the western part of the pastas as well. The pastas also contained four more weights, these made of lead rather than bronze but one with two inscriptions ( Τ and ΤΕ ), and nearby was a The Houses Described

105

bronze disk, pierced at the edges, and a rod with a ring at one end. These might have been part of a scales, the pierced disk being a scales pan. North Rooms The two North Rooms (a and b) were almost identical in size, and both had plain walls and an earth floor. Room a contained only a single coin; like so many North Rooms, its function remains difficult to determine, but furnishings and bedding would have left few traces. Room b, on the other hand, contained twentythree loomweights, twelve of them clustered closely, the others scattered nearby. Together with the court, then, this room was probably used for weaving, although the loomweights might also have been stored here for use in the pastas or court. Like the pastas and court, this room also contained tablewares, including a redfigured pelike, two plates, a lekythos, a cup, a saltcellar, and fragments of other black-glazed vases. Like the pastas, this room contained jewelry, another fibula and another earring (not a match to the one from the pastas). Other finds from this room included an odd cylindrical lead bar and a number of fragmentary terracotta figurines. This room was therefore a multipurpose space, used for weaving, storage of household goods, and other functions. Kitchen-Complex The kitchen-complex shares with the other houses on this row the anomaly of lacking stone bases for its pillar-partition (at least at the level preserved), despite the house’s two stories. The wall or foundation between the kitchen and flue is thickened at the south, perhaps to form a kind of platform or counter; the pillars supporting the second-story wall may have been set directly on this platform without stone bases. A large stone mortar was set towards the west end of the kitchen, showing that this room was used for processing food. The room also contained a piece of furniture, which burned leaving a scatter of nails and ash. Two lamps and tablewares, including a pelike, another red-figured vase, a lekythos, and a saucer might have been stored in this furnishing, although their exact situation was not noted. A bronze pendant, few pieces of hardware, and three coins complete the contents; like many other kitchens, it was relatively bare of finds. No finds were recorded in the flue, but a layer of ash and burning on the stone slab pavement show that the space was used for cooking, on fires built on the floor of the flue. The bathtub had been robbed out of the bathroom, leaving a gap in the cement pavement. Two red-figured squat lekythoi found in the bathroom might have been for oil or perfumes used in bathing. Storeroom A small room at the east end of the pastas contained a small pithos sunk into the floor, with a stone disk serving as a lid. The pithos was about 0.8 m in diame106

The Houses Described

ter, and had a capacity of some 190 liters.47 It is much smaller, therefore, than the pithoi in the House of Many Colors or the Villa of the Bronzes. Unlike the large storerooms in those houses, this house had only limited storage capacity —one small rather than multiple large pithoi. This distinction is important in understanding the different economic strategies pursued by different households and will be explored further below (chapter 6, ‘‘Household Storage’’). The pithos contained a black-glazed fishplate and a bronze needlelike instrument, but these probably fell in during or after the destruction of the house. Andron and Anteroom The andron is of the usual seven-couch size (4.4 × 4.5 m), entered through an anteroom from the court. As usual, these two rooms were the most highly decorated in the house. Except for the bathroom, whose walls were plastered, all the other rooms of A vii 4 had plain walls; but the andron and anteroom were painted in three colors. The andron had a white baseboard, yellow surbase, and red walls, while the anteroom had a black baseboard, red surbase and yellow walls. Both rooms had cement floors; the platform around the andron, on which the dining couches were set, was painted bright yellow. And as so commonly at Olynthus, both rooms were all but bare. The neck of a red-figured pelike was found in the anteroom, and nothing at all in the andron. Shop The large room in the southwest corner of the house (h) had doors to both the courtyard and the street, and probably served as a shop or workshop belonging to the owner of the house. It contained only miscellaneous finds: one coin, a lekythos, a saucer, a bronze swinging handle embedded in the wall, three loomweights, and hardware. However, the domestic portion of the house contained a number of artifacts which seem related to retail trade: four weights and scales in the pastas, and two bronze weights and ‘‘lead loomweight’’ in the court. Although these do not help us identify what might have been sold in the shop, they fit with the presence of a shop connected with the house. Relatively few shops at Olynthus preserved assemblages which would help determine what was made or sold in them (below, chapter 6, ‘‘Shops’’). Summary While very regular in plan, house A vii 4 offers a few surprises in its distribution of artifacts and the apparent use of space in the house. Tablewares and related objects were found in the court, in the pastas, in North Room b, and a few in the kitchen. The North Room b and perhaps the court as well were used for weaving. Weights were found in the court and pastas. This suggests a more homogeneous use of architecturally distinct spaces than in, for instance, the House of Many Colors, where few activities were duplicated in different rooms. There is, The Houses Described

107

Street v c

d staircase

b

drain

a

f

Avenue B

e

h

g h

M

k

j

much ash

l MM MM

?

i m MM

0

5m

figure 24. House A iv 9 however, considerable specialization in other kinds of spaces. The house has a storeroom and an andron and anteroom, all of which seem to be single-purpose spaces. The contrast between the two North Rooms (a and b) is particularly striking: when excavated, a was empty while b was a multipurpose room housing a variety of activities which left many material remains. House A iv 9 A iv 9 lies in the center of downtown Olynthus, near the agora and facing onto Avenue B. It was well preserved under 1 to 1.8 m of fill (fig. 24). The hill slopes off to the east here, so the house is terraced at two levels, the eastern part of the house being about a meter lower than the west. The eastern third of the house was taken up by three shops (d, f, l), of roughly equal size. The rest of the house conforms more or less to standard layout: a small court, a pastas extending the width of the domestic area of the house; one North Room (a) and a suite of rooms to the north (b-c); a kitchen-complex south of the court, and two rooms, g and i, in the southwest corner of the house.48 The construction is better than usual, with 108

The Houses Described

drafted ashlar masonry on the east and part of the north wall, two rooms with painted walls, and a number of stone architectural elements. The entrance to the house leads into room c. The door was of the prothyron type, whose four bases are still preserved; but extending from the southeastern base is another series of worked blocks. This is probably a stairbase for a set of stairs leading to the upper story, but here opening to the outside of the house, rather than to the interior. The upper floor, therefore, may have been either a separate dwelling with its own entrance from the street, or some other kind of separate space.49 As elsewhere, there are few finds which can be distinguished as secondstory assemblages; the finds from above the floors are a scant and mixed lot. Court The court of this house was rather small, but well cobbled and drained by a pipe leading out the front door of the house. Pithoi were set into the floor in two corners to catch water from the gutters. A marble louter base, inscribed – –]ΟΣ / – –]ΗΝΗ / – –]ΗΝΙΟΣ, was set against the north wall of the court; nearby was a plastic vase of a Silenus. A terracotta mold for a figure of a comic actor was found in the court; this and a second mold, found in room a, suggest that the owner of the house was connected with coroplasty, although no kiln or other equipment was found (below, chapter 6, ‘‘Coroplasty’’). Other finds seem more random: seven loomweights in the southwest corner suggest that the area might have been used occasionally for weaving, leaving these as debris, but are not enough to indicate that a loom was set up here when the house was destroyed. A guttus in the center of the court, a red-figure lekanis lid in the southeast corner, and a few sherds and miscellaneous finds might suggest occasional dining and other activities here. But like the court in the House of Many Colors, which was also relatively small, the courtyard may not have been an important locus of activity. Pastas The pastas was originally divided from the court by a colonnade, but the intercolumniations were later partly walled off, perhaps to a height of only about a meter or so, leaving a doorway just west of the central pillar. The effect would have been similar to the House of Many Colors, but here the semienclosed design is a later modification.50 Its walls were painted red with a white baseboard. Against the south wall of the pastas was a terracotta louter base in situ, whose basin was found across the room near the door to room a. A storage amphora rested nearby against the south wall, as at the House of Many Colors, and perhaps for the same purpose—to hold water for washing. There were no portable altars in the pastas, however, and this louter may have been used for domestic washing and other tasks rather than for ritual. The distribution of finds near the walls suggests that shelves or other furniture were built against the east and west walls. Part of a red-figure pelike was found in the northeast corner, a one-handled cup The Houses Described

109

in the southeast corner, and three lamps were found against the west wall. Three shallow bowls were also found in the pastas, as well as three loomweights and a spindle whorl, a swinging handle attachment from a bronze vessel, and seven coins. Although the pastas does not show the quantity and variety of finds and so, presumably, of activities as the pastades discussed above, its uses seem generally similar, including washing and shelving of vases and equipment. North Room The North Room (a), painted red with a buff baseboard, was apparently used for weaving: it contained thirty-one loomweights, or about one loom’s worth. This forms a parallel with room b of house A vii 4, another North Room used for weaving. The room also contained the second terracotta mold from the house, of a plump boy. Other miscellaneous finds included a bolsal, a lamp, a strap crossing, a small terracotta disk molded with a head of Athena, like an imitation coin, two slingbullets, and five coins. Suite b-c Graham considered rooms b and c as a second kitchen, in addition to the complex on the south side of the house.51 However, there was no trace of ash in the ‘‘flue’’ (b), the room is rather well appointed for a kitchen, with stone pillar capitals, and the main entrance to the house would lead directly into this room, a situation unparalleled at Olynthus (below, chapter 4). I prefer to interpret this as another suite with a light well, like those found in the House of Many Colors and the Villa of the Bronzes. The relatively small court might have created a need for more sheltered and well-lit space, as in the House of Many Colors. The light well could also have lit room a through windows, making this a more convenient workspace. It is peculiar, however, to have this suite as part of the main entrance; but domestic space was at a premium because one-third of the house was taken up by shops, and the owners may have had to compromise. The light well contained a saucer, a bead, and a pendant—a small collection, but the individual objects are like those found in other light wells. Room c, on the other hand, had a cluster of jewelry and perfume bottles near its east wall, probably from shelves under the stairs. This included an earring, another pendant, one red-figured and one black-glazed lekythos, a lead pestle, and a bronze spatula. These feminine articles accord well with the assemblages found in similar suites described above. In the southeast corner of room c was a peculiar hollow rectangle of stone and mudbrick, plastered in places and drained by an open drain of reused rooftiles. Graham suggested first that the rectangle might be the foundation of a cupboard, and later that it might have supported the stairs to the upper story; but it is far too flimsy to have supported the weight of a staircase. Within this rectangle were found the base of a marble louter, standing upright, and a portable altar. 110

The Houses Described

The reconstruction of the structure is problematic, but it might have formed or supported a kind of domestic shrine, the louter here used for ritual rather than domestic washing, and the drain, which in any case could not have handled large amounts of runoff, intended for liquid offerings. The group is thus similar to groups of portable altars and louteria found in the pastades of the House of Many Colors and the Villa of the Bronzes, and with the groups of miniature vessels, also probably associated with household ritual, from light-well suites at those houses. Kitchen-Complex The kitchen-complex (rooms j, k, and the unlabeled bath) lay in the southern part of the house. The large room (j) contained domestic equipment in three clusters, perhaps originally sitting on shelves. In the northwest corner of the room were two one-handled cups and a lead loomweight or weight. In the southwest corner was a storage amphora, another one-handled cup, fragments of a redfigured bowl, a fishhook, and a terracotta figurine depicting Hermes standing with a ram. Towards the center of the pillar-partition at the east end of the room were three lamps, two saucers, and a coin. The exact positions of other household equipment are not known; these include a bronze swinging handle, a needle, two astragaloi, a large iron object shaped like a figure eight, and two spikes, a lamp, and a miniature skyphos. These assemblages are all rather small and mixed, but this is probably because kitchen equipment was also stored in the adjacent room i. The flue (k) was accessible from the court, and its slab floor was covered by a heavy deposit of ashes, showing that it too had been used for cooking. In addition to ashes, the flue contained a cooking pot full of shells, table pottery, including two one-handled cups, a guttus, an oinochoe, a saltcellar and two saucers, a terracotta plaque of a man with a horse, and a variety of small bronze objects: a vase handle, five bronze rings (handles or attachments from a vessel or other utensil?), and three coins. Most of these finds were found in a small area at the north end of the flue, probably, like the finds from the flue of the House of Many Colors, stored there on a shelf or other such fixture. Such storage of domestic equipment in the flue is a fairly regular feature of Olynthian houses. To the south of the flue was a small space, corresponding to the bathroom in other houses. It was empty, like the bathroom in the House of Many Colors, although a bathtub could have been removed. Workroom and Domestic Storage Room In the southwest corner of the house were two rooms divided by a low wall (g and i). The northern room (g) was entered from the pastas, and there may also have been a wider door from room i into the kitchen. It could have been lit by a window from the courtyard and so formed a convenient working space. Two groups of loomweights—one of thirty-five weights in the northeast corner, The Houses Described

111

the other of twenty-six weights—were found in the room, with ten more loomweights scattered about. These might belong to two looms set up in the room when the house was destroyed. Near the door was a lower grindstone. A terracotta louter base, the fourth from this house, was lying on its side near the southeast corner of the room. Such louters could be used for food preparation, and probably belong with the grindstone and cooking assemblage (below, chapter 4). The room also contained a variety of women’s objects: two bronze mirrors with relief decoration and a bracelet; and a few pottery and metal vessels, including a plate, a saucer, and a miniature skyphos, the swinging handle from a bronze vessel, a fishhook, an arrowhead, a bronze boss, and five ‘‘keyhole reinforcements,’’ perhaps hardware from chests in which some of the other finds were stored. This room was seemingly a working area for the women of the household, used for weaving, food preparation, and such activities; the mirrors and jewelry belong with these activities as well. The back room (i) was smaller and probably less well lit. In it were found a large number of vases including coarsewares (an unslipped duck askos; another large jug or askos; a large spouted mortarium or basin; and a coarse, thick-walled open vessel with three holes in its lower walls, possibly a press), small pouring or unguent vessels (a guttus and a red-figured lekythos), a black-glazed plate, four saucers, and two lamps. In addition, the room contained a ‘‘pithos lid or table top,’’ two lead weights with letters cast in relief, a boss, and seven coins. Full of pottery and other artifacts, this room probably served as a storeroom adjacent to the kitchen and workroom. Shops The three shops on the east of the house (d, f, and l) lie about a meter lower than the rest of the building, and probably were not accessible from the rest of the house. The shops are virtually identical, of the same dimensions and each fitted with two doors, the narrower about 1.03 m wide, the wider 1.65 m wide. Graham proposed that the wider openings were used as display windows. That these rooms were used for retail trade is suggested by the numbers of coins found in them: twenty-eight coins in f, twenty-one in d, and eleven in l. This is a very large quantity of coins for an Olynthian room: f and d are the two rooms with the most coins anywhere in the site (below, chapter 6, ‘‘The Distribution of Coins’’). As in house A vii 4, however, there is no clear sign of what might have been sold or traded in the shops. In addition to coins, shop l contained four D-shaped lead objects (weights?), vases including a thymiaterion, a juglet, a lekanis and lid, and a plate, and a lead bar, cross-shaped in section—but these do not immediately help identify the merchandise. It seems likely, although not proven, that the three shops belonged to the domestic unit to their west. The whole house plot, shops and domestic area, was 112

The Houses Described

originally one οἰκόπεδον and was designed and built as a unit, with careful terracing and bonded walls. And although the shops might have been sold or rented, there is no reason to think that they were. The finds in the domestic portion of the building, moreover, seem more compatible with the household of a shopkeeper (κάπηλος) or someone of similar occupation than with the household of a farmer. There is no provision in the domestic part of the house for large-scale storage of foodstuffs, as is found in nearly every house in the Villa Section, for instance. There are, however, a number of finds which suggest production for sale or consumption outside the household. Three looms is more than the average household would probably need, and their presence suggests weaving on a relatively large scale, for sale rather than internal consumption. The two molds might suggest that the owner was involved with coroplasty. We will see other houses, not far away, where terracottas and cloth were produced for sale. Whether the owner also owned the shops and himself engaged in trade, or simply gained the income from rental of the shops, the household was more involved with the extra-domestic economy than were the households in, for instance, the House of Many Colors or the Villa of the Bronzes. (This issue will be treated further in chapter 6.) House A iv 9 thus displays a diverse and organized use of space. We have a wide range of domestic activities, weaving, processing and cooking food, washing, domestic cult, and the like, all in the western, domestic portion of the house and arranged in a very orderly fashion. Almost one-third of the available area of the house was taken up by shops, which probably document close involvement with retail trade. The second story was separated from the rest of the house, and might have belonged to a separate, perhaps a related household. House A v 10 Across the street, house A v 10 is rather similar in plan (fig. 25). It too has a welldefined pastas and court, a kitchen-complex, and three shops fronting onto Avenue B. The house was entered through double doors on Street v, like those in the Villa of the Bronzes: the entrance on the right is about 2.1 m broad, probably for carts and wheeled traffic, while a second, narrower door on the left, only 1.2 m broad, was probably for pedestrians.52 Another door led into the shop (f ) on Avenue B. The house may have had a second story: a rectangular gap in the pavement of the court against the west wall might once have held a stairbase, supporting a stair towards the south. Court The open court was fairly small, only 3.5 × 3.8 m, and was drained to the south by a drain made of stone slabs. South of the paved court was an unpaved area which must have been covered, its roof supported by a single base at its north end. The area northeast of the court was also roofed. The roofed areas were not distinThe Houses Described

113

Street vi c

a

d

b e

g

f

staircase

bronze vessel?, pigment

j

i

h

k

m

M

?

Avenue B

M

l

a

b

d

odd lead and bronze objects

hoard of bronze coins

c stylus

M M

e

f

g

i

olive crusher, upper stone

sales inscription

2 styli

staircase

k

j

M

h l

MM

room uncertain:

Street v

38 loomweights 23 loomweights 0

figure 25. Houses A v 9 and A v 10

5 m

guished from the court in excavation, however, so the finds from open and roofed areas are mixed together. Together, though, these three areas seem to have seen a wide range of activities. A large fragment of an olive crusher or orbis was found in the courtyard. The stone was broken and there was no trace of the rest of the crushing apparatus, and although we cannot be certain that this stone was part of a complete machine set up here, the wide double doorway into the house, appropriate for carts and heavy loads and found in many houses where agricultural products were processed, and the findspot of the piece on the floor of the court or exedra, where such a machine would have been set up, help confirm the significance of this fragment. Probably the less damaged pieces were salvaged and reused after the destruction, as most such equipment was (below, chapter 6, ‘‘Agricultural Processing’’). Two terracotta louters found here might also have been used in olive crushing. The court contained a variety of other objects. Artifacts perhaps concerned with trade and manufacture included two inscribed lead weights and a figurine mold, equipment similar to that found in A iv 9 across the street; a second mold was found in this house as well. There was also some domestic equipment: handles from two bronze vessels, nine loomweights and a needle, three glass beads, a long bronze instrument, fragments of terracottas, and other finds. Most of a Panathenaic amphora was found here as well, perhaps set up on display.53 Many of these latter finds were probably kept in the roofed area rather than in the open court. An area in the southeast corner of the court was walled off (rooms k and l), and this perhaps was used to store equipment for use in the court: it contained three almost identical red-figured squat lekythoi, three fishhooks, two terracotta figurines, two slingbullets, and other miscellaneous objects. Two groups of loomweights, one containing twenty-three weights, the other thirty-eight, document weaving somewhere in the house, but the room(s) in which they were found were not recorded. Pastas Unlike the houses described above, the pastas of A v 10 contained only a few objects and may have been used for a more limited range of activities. The only finds were an arrowhead, five coins, a pottery lid, two lamp fragments, and a bronze ring. Such a small, mixed assemblage looks more like refuse than the remains of specific activities. Perhaps the covered areas of the court took on some of its functions. Kitchen-Complex The kitchen-complex (rooms b, c, and the unlabeled bathroom) is slightly irregular in design, the wall dividing the bath from the flue being skew, and the flue having a door into the kitchen rather than into the pastas or court, as is usually the case. The Houses Described

115

As so often, the kitchen contained only a few, rather miscellaneous finds: a fragment of a bronze grater, a lamp, a saucer, a few pieces of hardware. A number of fragments of a bathtub were found in the bath and more in the flue. No evidence of burning or cooking was found in the flue, and no coarse or cooking vessels were recorded, suggesting that unlike the flues described above, this area was not used for cooking. Rather, it contained a number of pieces of jewelry and other women’s objects: a fibula, a finger ring, beads of bone and glass, and a bronze weaving implement. The flue also contained two lamps, a terracotta plastic vase, a green stone celt, an iron key, and two more lead weights. The kitchen-complex thus may not have been used in the same way as those in other houses. The fragments of bathtub suggest that at one point at least it conformed to the usual scheme, but when the house was destroyed, cooking may have been done elsewhere, and the flue used for other purposes. North Room The North Room (a) contained a number of odd lead and bronze objects, few of them easily identifiable: a triangular bronze sheet, a D-shaped lead object with holes for nails (?), a large folded lead sheet, and a lead bar 11 cm long, as well as a lamp and two lamp fragments, two arrowheads and three slingbullets, and nine coins. The cluster of lead and bronze objects probably indicates some specific activity here, but interpreting what that might have been is difficult. Other Rooms Rooms g and j take the form of an andron and anteroom, and the placement of the door to j near the main door of the house is appropriate for such a function, but these rooms were not painted and had only earth floors. The artifacts from the rooms are sparse: a lekythos, a bronze handle (?), two coins, and a sales inscription (below) were found in g; only three coins and a bead in j. Shops Like house A iv 9, the eastern part of this house was taken up by three shops facing on Avenue B. The central shop had doors to the street, to the domestic part of the house, and to the shop to its north, while the southern shop had only a door to the street. The connection between the main house and the northern two shops shows that these at least belonged to the owner of the house. This conclusion is supported by the sales inscription from the house. The northern shop (d) contained a hoard of ten bronze coins and sixteen other scattered coins. As in the shops of A iv 9, this is a large number of coins for a single room: excluding the coins in the hoard, the room ranks seventh of all the rooms at Olynthus in quantity of coins. A bronze stylus is perhaps also evidence of trade. In addition, however, the room contained a number of pieces of jewelry and vases, including an earring, a finger ring, a bronze bead; unguent containers 116

The Houses Described

such as a red-figured askos and a red-figured lekythos; a jug and a saucer, and five loomweights. The middle shop contained few finds and was perhaps more of an entrance and anteroom than a shop. The southern shop, however, contained two more styli, eight coins, and a lead weight, an assemblage somewhat similar to that of the northern shop. The shop also contained a number of tablewares and coarse vases: four plates, a guttus, and an amphoriskos; a bronze swinging handle and many fragments of bronze, perhaps the remains of bronze vessels; a large terracotta female head; seven loomweights; a knife and a spearhead; and other domestic and miscellaneous objects. Such a mixture of objects which could be associated with trade, such as weights, styli, coins and the like, and domestic artifacts, unguent vessels, jewelry, and loomweights, is not uncommon at Olynthus. It may be that the assemblages of different rooms have become mixed, or that the domestic material fell in from an upper story over the shops. But in other cases, such as the shops in A vii 10, the position of domestic equipment suggests that this belongs with the first-story shop rather than a fallen second-story assemblage (see below, chapter 6, ‘‘Shops’’). Sales Inscription A sales inscription found in room g records that when Aristoboulos son of Kallikrates was priest, Dionysius son of Ithyras bought this house, together with the storeroom (ὁ πιθεών) and ‘‘all the things that bring income’’ (τ. ὰ. μι . [σθ ]οφόρα 54 πάντα), for 5,300 dr. The price is the highest documented value for a house at Olynthus or, for that matter, in the Chalcidice, and it is among the highest in the Greek world. No pithoi were found in the house, and so the storeroom cannot be identified: either they were salvaged from the house after the destruction, or removed between the time the house was sold and when it was finally destroyed, or perhaps the πιθεών was located outside the house. But the reason for the unusually high price is probably the inclusion of ‘‘the things that bring income,’’ which might include the three income-bringing shops on the east side of the house and the olive-crushing equipment in the court of the house. The fact that they were specifically mentioned in this sale shows that in other circumstances they might not be included with the house—shops might be rented out or sold and other equipment sold separately—but included with the house, they probably increased its value considerably (see below, chapter 6, ‘‘Sales and Prices of Houses’’). Summary Domestic functions are not as well documented in this house as in some others, perhaps because it was excavated more hastily than the houses described above. But the scarcity of equipment in the pastas and kitchen-complex is probably not due simply to poor preservation or excavation; rather, it suggests a different form of organization. Instead of the pastas, the court and roofed exedra took over some The Houses Described

117

domestic functions. The site(s) of weaving are uncertain, but the household certainly produced its own cloth, as was normal in a Greek household. Most interesting, however, is the variety of other sorts of work that went on in the house. An olive crusher is too great an investment for most houses to have made or needed; just like today, only a few specialists in the community actually owned such expensive machinery, and most olive growers would have rented time on the equipment when their olives were harvested. This house, like a number of houses in Row A and elsewhere, seems to have specialized in such agricultural processing and been built accordingly, with wide double doors able to accommodate carts and beasts of burden. The shops might have been involved in this agricultural processing or perhaps involved with another side of the family business. The sales inscription associates ‘‘the storeroom’’ with ‘‘all the things which bring income,’’ and it is tempting to speculate that one of the shops served as a storeroom for olives or oil (its pithoi later salvaged) and so could belong with the olive-crushing enterprise. In any case, these features seem to have been considered among the most important in the house, enough to rate special mention in the sales inscription and probably to increase the price of the house to the most expensive known from the site. House A v 9 The house just north of A v 10 was, in its final form, rather irregular in plan, with a very small court (i), a short pastas, a kitchen-complex, and a number of rather irregular rooms which do not fit the standard plan (fig. 25). It seems to have been remodeled: in the walls between f, j, e, and i were built three reused olive crushers and a reused threshold block. The house may have been more regular in its original form.55 Like A v 10 it had two stories, with a stairbase in the pastas (room e). Once more, however, there are no useful assemblages which can be attributed to the upper rooms. Court and Portico The small paved area (i) seems to have been the only part of the house open to the sky; it is among the smallest courtyards at Olynthus, only 8.6 m2. In the northwest corner was a jar set into the floor for catching rainwater, like the two in A iv 9; there were no other finds. The area to the south of this was not paved, and two stones along its southern side could have served as rough pillar bases supporting a roof over this area. The space would thus be open both to the court and to the alley, into which rainwater from the roof would drain. In this roofed area were found nineteen loomweights at floor level, suggesting that a loom had been set up here. Not far away was a bronze ‘‘probe’’ with pointed ends, perhaps actually a weaving tool, and a few miscellaneous finds. The alley here was paved, probably by the owners of this house since the pave118

The Houses Described

ment ends at the east end of the house. Because it essentially formed part of the household space, it seems to have been kept fairly clean and not used for the disposal of refuse, unlike the alleys in some other houses, which contained quantities of rubbish. The only large finds were a pillar capital, a terracotta female head, a complete oinochoe, and a louter fragment under the stone at the east end of the alley. Weaving Rooms Besides the area south of the court, three other rooms contained large numbers of loomweights. Room h contained twenty-two loomweights, as well as an inscribed lead weight. Room j had eighteen loomweights, and room e, the pastas, twenty-four loomweights. Each group probably represents the debris from one loom. All these rooms adjoined the court and could have been lit by large windows; they would be suitable for weaving, like the workroom in house A iv 9. Four looms is more than would be needed for household use, and with every well-lit space adjoining the court, amounting to about a third of the area of the house, employed for weaving, this house seems to have been producing textiles on a rather large scale (below, chapter 4, ‘‘Textile Manufacture’’). Pastas Room e corresponds to the pastas of the house, although in its final form it was entered from the court through a door rather than a colonnade. As in other houses, this room was an important work area, housing a variety of activities in its sheltered and relatively well-lit space. Near the center of the room was a deposit of artifacts on the floor, including a lower grindstone, seven vases, and a bronze swinging handle; five glass eye beads and a bone pendant; and a bronze instrument like a tweezers. These might have been stored in a chest standing here, although it is somewhat unusual for such furniture to be in the center of the room rather than against a wall. Also found in the room were two bronze vessel feet, six more eye beads, a bronze stamp, a calyx cup, and a lamp. In the southeast corner was a large stone mortar, buried up to its rim in the earth floor. A terracotta louter base was found in the room as well. This might actually be part of a kneading trough, used to prepare dough from grain cleaned and ground with the mortar and grindstone. In addition to weaving, then, the pastas seems to have been used for a variety of domestic activities: preparing cereals and bread, domestic storage, and the like. Kitchen-Complex The kitchen-complex on the north side of the house seems regular in design: a large room (c) separated by a pillar-partition from the ‘‘flue’’ (b), with a small room (a) to its north as a bath. Like the complex in A v 10, however, these rooms contained only a few finds, and its use remains somewhat problematic. Few or no The Houses Described

119

vases were found in the kitchen or in the flue, and no ash is recorded in the flue; there is no evidence for cooking or food preparation in the complex. The kitchen contained a couple of pieces of jewelry, an earring and a bracelet, as well as a spearhead, a terracotta figurine, and a few miscellaneous objects. In the flue was an unslipped bowl on a stand, possibly a thymiaterion, a female protome and a fragment of another, and many fragments of bronze, probably from some disintegrated object. The protomes and thymiaterion suggest cult rather than cooking. No bathtub was found in room a, only a bone doll with attached, movable arms and a lead pendant. There is thus no sign that the kitchen was being used as such, at least at the moment the city was destroyed; the few finds might be remnants of a former use by women of the household. Shops The shop in the southeast corner of the house (k) had two doors to the street, but none to the rest of the house. It contained fourteen coins, a lead weight, and a collection of bronze fragments, including a disk with four holes pierced at its rim, perhaps a small scales pan. The assemblage is similar to those in shops of A iv 9 and A v 10, with a relatively large number of coins, and quite compatible with a function of retail trade. The door to the room in the northeast (d) was not preserved. Twenty coins found here suggest that it too served as a shop, but the room also housed domestic activities as well: among its contents were a lower grindstone; vases, including a fishplate, a squat lekythos, a saucer, and an amphoriskos; a storage amphora; and two fibulas. Here we see another example of the rather common mixing of domestic activities with retail trade at Olynthus. The central ‘‘shop’’ and entrance to the house, room g, contained a rather mixed group of finds: a bronze ladle with a swan’s head handle, two large but disintegrated bronze objects, and an area of red pigment. These rooms therefore might have mediated between the private, domestic areas of the house where textiles were produced and the public consumers outside. Other Rooms Room f, a small intermediate space between the entrance room g and the pastas, contained only a few, rather miscellaneous finds: a bronze pendant, an arrowhead, two coins, a juglet and a plate, an iron boss. In the open space fronting onto the alley (l) were found two bronze mesomphalic phialai, a finger ring, a heart-shaped ornament (?), a ‘‘keyhole reinforcement,’’ a saucer, and an astragalos. Both assemblages are difficult to evaluate; the phialai might have been hidden here for safekeeping during the siege.

120

The Houses Described

Summary Despite its irregular design, in A v 9 we see a fairly strict organization of space and specialized work areas. All four rooms adjoining the court were devoted to weaving, three of them apparently exclusively given over to it. The shops facing the street perhaps sold finished textiles, which have left no trace. The household must have included a number of women, free and/or slave, to keep at least four looms busy. It is no surprise, therefore, to find an unusual amount of jewelry in this house: eleven eye beads in the pastas, the earring and bracelet in the kitchen, two fibulas in room d, an earring in g, a finger ring in l, pendants in the ‘‘bath’’ and alley, as well as another fibula and another eye bead from uncertain rooms. As in house A v 10, domestic functions in this house are somewhat restricted. The kitchen-complex contained no signs of food preparation, cooking, or the storage of eating or drinking equipment. Instead some of these functions are documented in the pastas and room d. This may be in part an accident of excavation and recording, but it suggests a use of space rather different from that in the houses described above, and more similar to that of its neighbor A v 10. There are no provisions in A v 9 for large-scale storage of foodstuffs or other produce. One storage amphora in room d is the only large container recorded, and the pithos in the court probably collected rainwater. In this respect the house is similar to A iv 9 and A v 10. House A v 1 The houses on block A v were generally fairly plain. In striking contrast to the neighboring block A vi, most had few specialized rooms, painted walls, mosaics, or other such features (below, chapter 5, ‘‘Variations on the Type House’’). Houses A v 9 and A v 10, just described, are actually exceptions to this rule: they had kitchen-complexes, shops, and other specialized rooms. This is likely because of their position at the east end of the block, along Avenue B, where houses tend to be differently organized (below, chapter 5, ‘‘Cluster 4: Houses with Shops’’). The northwestern house on the block, A v 1, is more typical of the houses on this block (fig. 26). Buried under only 0.3–0.4 m of fill, the house was not very well preserved and was less rich in artifacts than other, more deeply buried houses; but the assemblage is an interesting one, with a relatively large number of storage amphoras and terracottas found in various parts of the house. Like all the houses on this block other than A v 6, which was rebuilt, and A v 9 and 10, built where the ground dropped off and so not included in the common roof over the block, this house had only one story.

The Houses Described

121

11 amphoras in northern half of house

Street vi a

b

c

bronze eyelashes many storage amphora sherds

e

Avenue A

d

weaving bodkin

f

g

h

2 bosses; spike; many pieces of lead

j i

many pieces of lead; fragment of horse bit

basin containing 3 storage amphoras

0

5 m

figure 26. House A v 1 Court and Pastas The courtyard of the house had a pebbled floor, rather than a cobbled or paved surface. A low wall or foundation running across the north side of the court separated it from the pastas. Breaks in the center of the foundation indicate where the bases of the pillars supporting the pastas have been robbed out. The courtyard was partly open to the alley, as in a number of houses on the north side of Olynthian blocks, making the alley into an additional living and work space (compare house A v 9). The neighboring house A v 3 also had an opening to the alley, so that the two houses could communicate. Such linking of adjoining houses was not uncommon, and allowed more private interchanges, bypassing the public streets. In the southeast corner of the court was the bottom part of a large basin or pithos (described by the excavator as a krater), perhaps used to collect rainwater 122

The Houses Described

from the eaves.56 In this were found fragments of three storage amphoras, perhaps also used for collecting and storing rainwater. Other than this, the court contained only miscellaneous finds: the swinging handle from a metal vessel, a hinge, two terracotta figurines, two loomweights. A similar assemblage was found in the alley, including another swinging metal handle, more loomweights, a basin, and an olpe. The pastas contained another storage amphora but otherwise only a few scattered finds: another terracotta figurine, three loomweights, an arrowhead, a saucer. North Rooms Rooms a and b contained relatively few objects. A bronze eyelash-plate from a life-size bronze statue was found in room a.57 Perhaps found by an inhabitant of the house at a sanctuary and carried home as a souvenir, such odd finds (like an Athenian dikast’s ticket from house A viii 3) add spice and curiosity to the often somewhat mundane household assemblages. Room a also contained five coins and a squat olpe. Room b contained a small pottery lid, a lead clamp, a bronze bar, and two coins. Room c is a large room, about 7.5 × 4.9 m, with a patch of cobble pavement about 2.5 m square in its southwest corner. The published photo and drawing suggest that the room had two doors in its south wall, one in the southwest and one near the center of the wall.58 Although poorly preserved, this might be reconstructed as a kitchen-complex, with cobbled flue at the west entered through a separate door. In a one-story house like this, pillars would not be needed to support the wall of a second story above the kitchen; the flue could therefore have been separated from the kitchen by only a light partition which left no trace. But other interpretations are possible, and it may well be simply a general-purpose space. This room or complex of rooms contained many artifacts, more than any other room in the house, suggesting that it formed an important locus of work. Many storage amphora sherds were found here; and although the total for this room was not recorded, the northern half of the house contained some eleven amphoras, many of which may have come from here. These may have been used to store water or foodstuffs, although nearby, room e also contained a pithos for food storage. The number of amphoras is unusual—no other house had as many storage amphoras. This might reflect an unusual strategy for storing food adopted in this house, or it might be that the amphoras were used not for food but for some other purpose. A bronze bodkin, or large needle with eyes at either end, attests to weaving in the house, as do thirteen loomweights found scattered through the house, although no room in this house had a concentration of weaving implements. Other bronze artifacts from the room include a tweezers, a Homeric-type key, The Houses Described

123

and a ring. Two large spikes might belong to a furnishing here. Besides the storage amphoras were found a few small vases—a saltcellar and another, unidentified vase. Three terracotta figurines—a female protome, a standing woman, and a fragment of an animal—were found in the room as well. The standing woman was found in the southwest, and so in the flue if this is to be reconstructed as a kitchen-complex; the findspots of the others are not known. Again, this is not a striking concentration of objects in this room, but with nine terracotta figurines, the house as a whole has one of the richer assemblages. Storage Room e East of the pastas is a small room which contained fragments of a pithos (room e). The size of the pithos was not recorded, but it was probably similar to the pithoi found in houses A vii 4, A 8, and elsewhere on the North Hill (see below, chapter 6, ‘‘Small-Scale Household Storage’’). At only 3 m square, this room is hardly large enough for the large-scale storage found in houses in the Villa Section, like the House of Many Colors or the Villa of Good Fortune. Other Rooms Two or three rooms open onto the courtyard: rooms f (whose entrance is not preserved), h, and j. Room f is relatively large and contained a furnishing of some sort, attested by a few bosses and other hardware. Together with room j, this room contained many pieces of lead, not described in detail by the excavator. A spoon, lekanis lid, and bowl were also found here. Room h contained part of an upper grindstone and may have been used for food preparation. Room j also contained a variety of different sorts of artifacts: a tiny pitcher or cooking pot, an olpe, a fragment of an iron horse bit, a terracotta pair of standing women, two loomweights, and many pieces of lead like those in room f. Summary The assemblage in A v 1 is scattered and diverse: it is neither as rich nor as coherent as artifact assemblages in some other houses, such as the House of Many Colors, and it is not as spatially organized. Loomweights, terracottas, vases, coins, and other finds are found in small numbers in nearly every room; and the concentration of objects in room c, for instance, is mixed and not particularly distinctive from the overall range of finds in the house. This is undoubtedly due in part to the shallow depth to which the house was buried, which may have led to more disturbance of the destruction debris; but it may also reflect a less organized use of space in this house.

124

The Houses Described

vases, amphora & bathtub fragments in pithos

Also in room a:

a

b

c

shallow stone basin bronze chain

Fortification Wall

staircase

d

lead disk with lotus & palmette

(unexcavated)

M lead pestle bronze implement round lower grindstone and upper saddle quern

e

many vase sherds bronze grater

vase sherds; shell

burned area

Avenue A

door?

spatula; red pigment

f

g

h i

0

5 m

figure 27. House A 8 House A 8 House A 8 shares many similarities in plan with its neighbors, A 9 and A 10 (figs. 7, 27). Like those houses it had a double entrance, with a wider opening for carts and a narrower for foot traffic, leading into a vestibule and then to the court.59 Its court, centered in the southern half of the house, was separated from the pastas by three pillar bases and a partition wall which enclosed the eastern part of the pastas. A stairbase on the north wall of the pastas is evidence of a second story in this house, like most houses of Row A except for the southernmost three. South of the court was the usual drainage alley which separated houses on this row. Three North Rooms were exposed, and a fourth, in the northwest corner of the house, was left unexcavated. The eastern two rooms, however, were used as a suite of shops, as was a third room to their south. It is rather modest, with earth floors except in the courtyard and no painted rooms. Courtyard and Pastas The courtyard (room g) was paved with pebbles and was drained by a pipe below the floor into the alley between A 8 and A 9. A large number of artifacts The Houses Described

125

were mostly scattered along the walls of the court as if fallen from shelves. In the southwest corner of the court was a group of vases including a ‘‘flower pot,’’ two bowls, a duck askos, a plate, a miniature lamp, and a terracotta ‘‘brick.’’ In the northwest were more vases, including three lekythoi and a lid, three fragmentary figurines, a round lower grindstone (an unusual object, unfortunately not further described in the publications), and a boat-shaped saddle quern. Also in the west were a spatula or stylus and two fragments of red pigment, perhaps sinopis. On the north towards the pastas were found a bronze chisel (?) and a redfigured vase. In the northeast corner of the court were three worked blocks, the base for some uncertain object or installation. The only finds in this area were another lekythos and a shell with a hole pierced in its ‘‘snout,’’ perhaps to be used as a vessel or ladle. Also found in the court, although without a more exact findspot, were a number of pieces of jewelry, including a bronze chain, a finger ring and a bead, and a bronze swinging ring handle. The many finds suggest a variety of household activities here: food preparation, grinding grain, and the like. Most of the artifacts document women’s work and equipment: grindstones, lekythoi, jewelry, and terracotta figurines. The pastas contained rather fewer artifacts than the courtyard. In the southeast corner, where the pastas was separated from the court by a wall, was a lead pestle, suggesting food preparation or some similar activity. Also in the east and scattered in the pastas were eight loomweights; not enough to equip a loom, but documenting weaving in some part of the house. In the western part of the pastas were a complete storage amphora, two fragmentary plastic vases: an ‘‘Attis head’’ and a negroid head vase. North Room: Kitchen-Complex? Room a was a large room (4.35 × 7.3 m) with a number of interesting features. A single base was preserved at the north wall, some two-thirds of the way across the room. The position of this block suggests that this is the sole surviving base from a pillar-partition, dividing the space into the common kitchen-flue complex. A door may have led through the south wall into the flue, as the wall is much less well preserved here than elsewhere. A long, narrow block was uncovered in the main room in 1928; this may have belonged to a square stone hearth of the sort often found in kitchens.60 Fragments of a bathtub rim were found in the room as well, suggesting the presence of a bathroom here. This room therefore may originally have been a kitchen-complex, and the blocks of the pillar-partition and hearth were either robbed out with the destruction of the city, or the room had been remodeled and unified before the destruction. Alternatively, it may have been a suite with a light well at the east, like those in the House of Many Colors and elsewhere; the evidence is not conclusive in either case. A pithos was buried to its neck in the floor of the western part of the room, 126

The Houses Described

one of the best preserved pithoi at Olynthus. This vessel is relatively small, with a capacity of only some 110 liters, perhaps a phidakne rather than a pithos proper.61 The smallest pithos in the Villa of Good Fortune held almost ten times as much. Within the pithos were found a jug, a bolsal, and fragments of roof tiles and sherds, but these probably fell into the pithos in the destruction; it may have been empty at the time of destruction, its contents perhaps consumed during Philip’s siege. This type of small-scale storage is typical of houses on the North Hill, for instance at house A vii 4 (below, chapter 6, ‘‘Small-Scale Household Storage’’). Just west of the pithos was a large collection of vases and other objects, again probably stored on shelves or furniture here. These included a red-figured bell krater depicting a four-horse chariot driven by Nike, a red-figured pelike depicting a bride and Eros, a tiny cooking pot, a guttus, a saltcellar, a terracotta mold in the shape of a horse, a terracotta Silenus, a small, roughly carved stone basin or mortarium, four saucers, and other objects. Near the door were two bronze arrowheads and a bronze boss; two more arrowheads were found in the western part of the room. Other finds from the room include a large round-bottomed bowl, another guttus, a bronze needle, a lead double herm statuette, and many coarse vase fragments in the burned layer on the floor. No artifacts were specifically recorded in the eastern part of the room, probably the flue—in other houses a room quite rich in pottery and other material. The partial excavation of this part of the house in 1928, when the recording was often quite sketchy, probably recovered more objects whose provenience is now lost. This complex seems to have served a number of functions. The pithos documents food storage, relatively rare in kitchen-complexes. Only two other houses had pithoi in their kitchens, and both of these were above the floor, not set into it. The stone basin, cooking pot, and other coarsewares suggest that the complex was used for cooking and food preparation, in addition to the courtyard and pastas. The fancy tablewares, such as the red-figured krater and pelike, show that it was used for the storage of tablewares and perhaps also for eating. In all, if this is to be identified as a kitchen-complex, it is one of the few which contained significant quantities of domestic artifacts, the others being the House of the Comedian and house A iv 9, both described elsewhere in this chapter. Shops Three rooms opening onto Avenue A are identified as shops: the suite of rooms b-c, and room e. Rooms b and c were excavated in 1928 and few finds were noted. Room c, however, contained a burned area in the southeast, perhaps the remains of a piece of furniture, and a number of vases, including a guttus, a bowl, a saucer, and many fragments of unslipped pottery. A bronze grater, a disk with lotus and palmette decoration, and eight coins were also found. As so often with shops, it is difficult to distinguish what if anything was being sold or made here. The Houses Described

127

Summary The courtyard and kitchen of this house are both unusually rich in artifacts and seem to have been used for a variety of domestic chores. The pastas, on the other hand, was relatively bare, with only eight loomweights, a storage amphora, and a few vases. While the houses we have considered above tended to concentrate artifacts and, presumably, household work in the better-lit and sheltered rooms of the house, such as the pastas, this house had a richer assemblage in the kitchen. Like A vii 4 and other houses on the North Hill, this house had limited storage facilities, a single small pithos or phidakne. House A 10 House A 10 is the most southerly of the regular houses in Row A, and quite similar in plan to A 8 and A 9 (fig. 28). To the south, houses A 11 , A 12, and A 13 are significantly more irregular in plan, less well constructed, and probably had only one story rather than two as did the houses to the north; there seems to be a distinct change in planning here (below, ‘‘House A 11’’). The sales inscription found in house A 11 describes that house as adjoining those of Polyxenos and Pythion, and Robinson suggests that A 10 belonged to Polyxenos, perhaps the magistrate of that name attested on Chalcidian coinage.62 Like A 8, it was partly excavated in 1928 and the notes for the court, andron, and pastas are sketchy. Court and Pastas The arrangement of court and pastas was somewhat unusual. The pastas was separated from the court by four pillars, two of whose stone capitals were found. The cobbled surface of the court ended in a neat line about 1.6 m short of the pillar bases, however, leaving an uncobbled area which must have been covered by an unusually wide overhang of the roof above, supported by a pillar at the corner of the andron. No cobbles were preserved in the southern part of the court, which may also have been roofed. The total open area of the court may therefore have been quite small, only 4.8 × 2 m, but the sheltered areas would have provided well-lit work areas. In the western part of the court was a built altar, part of whose superstructure was found in room h to the west of the court. The court and other rooms of the house contained many stone objects, including unfinished pieces, which suggest that the owner was a mason, like the owner of the nearby house A 5. These included five pilaster, anta, and column capitals from the court of the house and from room h, which do not seem to belong to the architecture of the house itself; three portable altars from the court, one of which was unfinished, and another cubical block which may be an unfinished altar; an odd block with two large rectangular cuttings; a round stone weight and a stone bowl (a mortar?); and other worked blocks.

128

The Houses Described

2 TC pithos lids

a

b

c

pottery tray; marble object; stone “loomweight”

netting needle

e

d

staircase

f

tripod base?

double door

l g i k window; marble baseboard?

h

m 3+ unfinished portable altars

worked block

a

b

c

d

Avenue A

2 Doric capitals; 3 worked blocks; altar crown; base; etc.

Fortification Wall

House A 10

e

loan inscription

f

g

h

i

cache of 8 terracottas

House A 11

j k

l

hoard

m

House contained 96 coins, including one hoard of 10 bronze coins

figure 28. Houses A 10 and A 11

0

5 m

More stone objects were found in the rooms along the west side of the house, e and h. Room e contained a block with three holes, perhaps a tripod base;63 while room h to its south contained three worked blocks and two Doric capitals, as well as a block which was probably the crowning member of the altar in the court, an upper grindstone, and other objects. These rooms may have been workrooms or, perhaps more likely, used to store stones awaiting use or reuse. The pastas (f ) contained a few objects in the eastern, more secluded area, perhaps stored under the stairway here. These included a saddle quern for grinding grain, two lamps, and a lead loomweight or weight. Like house A 8, the pastas does not seem to have been the site of as much activity as in some other houses; although this may be due to incomplete excavation records. Andron The andron was set in the southwest corner of the courtyard, an unusual arrangement. In most houses the andron was placed next to a street to allow this more public room to be better lit; in this house, the andron may be a later addition. It is a small room, only 3.6 × 3.5 m, holding four rather than the usual seven couches. Although small and floored with cement rather than with mosaic, it is otherwise one of the more elaborate androns at Olynthus. A set of three small window capitals was found in the courtyard, which probably belonged to a window with a central support opening from the andron onto the court. Similar large, elaborate windows with central supports are known from androns at Eretria, the Piraeus, and elsewhere.64 The andron also contained ‘‘many thin slabs of white marble’’ which may have covered the walls as a revetment or baseboard.65 Other houses had wooden baseboards; a marble baseboard, particularly in the house of a mason, is not impossible. The 1928 excavations of the andron do not record any finds for the room. North Rooms The northwest room of the house (a) was only partially preserved, but it was probably a storage room. It contained two terracotta pithos lids and the base of a large vase, perhaps a storage jar. The room also contained two terracotta figurines and a few other miscellaneous finds. The next room (b) was used for weaving: it contained twenty-four loomweights and a netting needle, as well as a bead and two vases. Forty more loomweights are recorded from this house, equipment for a total of two or more looms (below, chapter 4, ‘‘Weaving’’). The room was presumably entered from the pastas, although the door is not preserved. Room c, however, contained a number of odd objects: a terracotta tray 11.5 cm square and 2.5 cm deep; a marble ‘‘bar’’ 31.7 × 6.5 × 2 cm, with raised projections on one side, and a stone object (described by the excavator as a loomweight) 11 × 9 cm, both found in the southwest corner of the room. These might all be implements for processing food, although the fancy 130

The Houses Described

decoration of the room, with a red baseboard and yellow walls (and/or ceiling), might suggest less mundane activities. Room d contained only a few, miscellaneous objects. Shop Room m, in the southeast corner of the house, seems to have been a shop, although the door to Avenue A was not well preserved. The excavation notes do not record any finds from the room, but like shops in other houses on the North Hill, it probably mediated between the domestic production in the main part of the house and consumers. Second Story House A 10 offers good evidence for the second story at Olynthus. A stairbase set against the south wall of the pastas led to a first flight of stairs; this then turned a corner on a landing in the northeast corner of the pastas to a second flight along the north wall.66 The landing was supported by a mudbrick base in the corner of the pastas, which left its mark in the plaster on the wall. But as in other houses, it is not possible to distinguish assemblages of artifacts from the upper floors and so reconstruct their use. Summary At least two households living in Row A engaged in stonecutting and sculpting. To the north, house A 5 contained unfinished altars, relief sculptures, and other blocks (see below, chapter 6, ‘‘Sculpture and Stonecutting’’). In both these houses, worked and unworked blocks were found in the courtyard, which was apparently the common site for such work. The rooms west of the court were also apparently used to work and store stones. The double door into the house might have been useful in hauling raw materials and finished products in and out of the court. The house has no identifiable kitchen-complex, and its andron was a later addition. House A 11 The southern three houses in Row A, houses A 11–A 13, are irregular in plan but rather similar to one another (fig. 28). Like the other houses in Row A, house A 11 was larger than the average Olynthian house, measuring 17.7 × 21.3 m, including the alley between it and A 12, and was built against the west fortification wall of the city. It was quite different in plan from the ‘‘regular’’ houses: instead of the usual courtyard-pastas arrangement, it consisted of a series of rooms arranged around a courtyard (g) and had no kitchen, andron, or other specialized spaces. The house had no pastas; all its rooms seem to have been enclosed. Half of the rooms were roughly the same size: a, d, e, h, i, and j were about 4.9 × 4.5 m; the The Houses Described

131

other rooms somewhat smaller. The rooms were also arranged in a more hierarchical fashion than those of typical houses. For instance, the only entrance into room e is through rooms h and i, and the only entrance into room k is through l.67 The walls were unplastered and unpainted; the floors all earth. The location of the main entrance to the house is not certain because its east wall was not preserved above its foundations, but was probably into the ‘‘alley,’’ room m. There is no evidence for a second floor. The house thus shows a less organized arrangement of architectural space than the more regularly planned houses we have considered so far. To judge from the distribution of artifacts in the rooms, the use of space was less organized as well.68 Some activities were restricted to specific rooms: for instance, weaving was done in room d where twenty-five loomweights were found. But other types of artifacts and activities are distributed all over the house. The rooms in the northwestern part of the house, for example, contained storage amphoras in roughly equal numbers: five rooms (c, d, e, h, and k) each contained two amphoras, while a sixth (i) contained four. Whatever these amphoras were used to store, no single space seems to have been designated for storage, but instead that function was distributed through many rooms. Likewise, no room had a great concentration of vases. Tablewares are found, in moderate numbers, in rooms e, f, g, h, k, l, and m, but no single room seems to have served specifically as a storeroom for household equipment, as was the case in house A iv 9 or the House of Many Colors. Other notable artifacts in the house include a lower grindstone in room h, ‘‘pithos lids or table tops’’ in e and h, a louter base in b, and a louter and base in j. In the western part of the courtyard was a cache of at least eight plastic vases and terracotta figurines, including a head of Dionysos and a faun head, two male heads, a female protome, and three female figures. These caches are not uncommon in the courts and porticoes of houses at Olynthus, and will be discussed further below. Finally, this house, like its neighbors A 12 and A 13, contained a large number of coins. Room j had a hoard of ten bronze coins, and the whole house contained some ninety-six coins, including six silver coins—second only to house A iv 9, which contained one hundred coins (below, chapter 6, ‘‘The Distribution of Coins’’). It is more difficult to distinguish specific and easily interpreted activity areas in irregular houses like A 11 than in the more organized houses described above. Its spaces were less specialized, and except for the court, most of the rooms were architecturally equivalent, many of them being about the same size and shape. The similarity among the assemblages from different rooms suggests that the spaces could be conceived of and used in a more or less equivalent fashion as well. The activities attested at the house, such as weaving, grinding, and storage of tablewares, are primarily domestic, and there is no reason to think that this was not a dwelling: the explanation for the difference in organization is not, I suspect, 132

The Houses Described

that this was not a house per se. But the quantity of storage amphoras in so many rooms suggests nondomestic activities here as well, which seem to have occupied a large portion of the house. Loan Inscription An inscription found in the courtyard records that this house was used as security on a 2,000-drachma loan to Archidamus, son of Metrichus.69 The inscription lists the neighbors as Polyxenus the son of Telagrus and Pythion the son of Diodorus; Polyxenus and Pythion also served as witnesses to the transaction. Men named Archidamus and Polyxenus are among the magistrates of the Chalcidic mint. If these are the same men as those named in this inscription, it is interesting that these prominent men would own such irregular and architecturally undistinguished houses. House D v 6 (The House of Zoilos) D v 6 is another house of somewhat irregular plan, with a small court (j), a pastas and portico (g-h), and other rooms ranged around these spaces (fig. 29). The house has some seventeen rooms, an unusually large number. The southwestern part of the house plot is taken up by two separate suites which are not connected to the main domestic area: rooms e, i, l, and m form one suite, and room n, divided into two subspaces by a spur wall, forms a second. These seem to be shops, but there are obvious differences in plan between this house and houses like A iv 9 or A v 10, which allocate a large amount of space to shops but are otherwise fairly regular in plan. In its organization, the house is more similar to the irregular houses A 11–A 13 and those in Trench 7 than to the regular houses with shops. The main part of the house was entered from room p, which led immediately into the court. Room g probably formed a portico on the west side of the court, and room h the pastas on its north, although no bases were preserved. These spaces led to a series of rooms along the north side of the house (a-d), and to rooms f and k. Room b could be entered only by passing through room c, and room a only from b, so the organization of this area, as in house A 11, is more hierarchical than in more regular houses. There is no evidence of a second story. Most of the rooms had earth floors and unplastered walls. The court was cobbled, and rooms d and k had plastered walls. Room q, in the southeast corner, was painted red with a white baseboard, the only painted room in the house. Court and Pastas The court, pastas and portico (g) of the House of Zoilos did not contain many objects and may not have been important areas of work. The court and portico were entirely bare. The pastas contained four loomweights and a spindle or distaff, and might occasionally have been used for weaving, but many rooms in this The Houses Described

133

vase w. green pigment a

b

d

c

pithos rim

many vases

e

f

large basin

h

spindle? tray/trough red pigment

i

g

j

k

n

many vases l

many vases

m

knife, spearhead

p

q

o

1 vase 11 amphoras sales inscription

Street v 0

5 m

figure 29. House D v 6 (The House of Zoilos)

house contained a scatter of loomweights, so these might be random losses. A low poros stone tray or trough in the middle of the pastas might have been used for grinding or food preparation. An olpe, a saucer, and an earring were also found in the pastas. Room c was apparently open to the pastas and court, and so it would have been another well-lit work area, but this room too was almost bare. Other Domestic Rooms Room b, a small room on the north side of the house, contained fourteen loomweights; it may have been used for weaving.Unlike most weaving rooms, however, it is far from the courtyard or other source of light. A window opening onto the alley north of the house seems like the most likely means of illuminating this area, although it is also possible that the loomweights were simply being stored in this room. A vase containing green pigment was found near the door, and red pigment was found on the floor. Since so few of the rooms were plastered it is unlikely that the pigment was intended to paint walls, like the pigments in the House of Many 134

The Houses Described

Colors. In the northeast corner of the room was a pithos on its side, presumably for food storage. Room a, in the northwest corner of the house, was accessible from the court only by passing through rooms b and c. Only six scattered loomweights, a guttus, four coins and a pithos rim were found in this room; the rest of the pithos was not preserved. In the northeast corner of the house, room d had a very rich assemblage of some fifty-two artifacts. It contained more than twenty-seven vases, including a cooking pot, a very large coarse basin, two storage amphoras, five bowls, a red-figured hydria, a red-figured pyxis lid, an oinochoe, an olpe, two gutti and an askos, three saltcellars, and three saucers. The room also contained part of a bronze unguentarium and the swinging handle from another bronze vessel. These were mostly found near the west and south walls, where they had fallen from shelves. A marble louter was also found here, as well as a terracotta louter or another basin, a slate pithos lid or table top, a bronze grater, ten loomweights, artifacts of lead and iron, and various other finds. This secluded room was probably relatively poorly lit, at best with a window facing onto the alley, and like the domestic storage room in house A iv 9 (i), it was probably used to store domestic equipment for use in the pastas and elsewhere in the house. To the west of the court, room f contained another pithos and its terracotta lid and was apparently a storeroom for foodstuffs, as well as for a few table vases, such as a lekanis with a cup-handle, a bolsal, and an unslipped jug. The other domestic rooms, however, were quite bare. Room k contained only a coin, a spike, and a lead object. Room o was equipped with a drain into the street fashioned from a reused millstone, suggesting a specific function; its only contents were a single unslipped vase. Room q, in the southeast corner of the house, was the best-decorated room in the house, with red walls and a white baseboard. Perhaps it served as a reception room or even an andron. It contained a single loomweight and an uninscribed stele, with only the rule lines finished. Shops/Workrooms The two groups of shops in the southwest corner of the house seem to have been involved with food preparation. The western complex (rooms e, i, l, and m) contained three mortaria with spouts (in e)—an unusual concentration of equipment for preparing food; a large pithos (in i) and terracotta lid; red paint on the floor around the pithos, two more ‘‘pithos lids or table tops’’ (one of terracotta, in room l; and one stone disk in room e) which could have been used as surfaces for processing food; and many vases, including a bowl, an oinochoe, and fragments of many other pots (not mended, perhaps because they were coarsewares). Room i contained two clusters of hardware, one cluster of nails, and other objects in the back of the room, perhaps the remains of a table, chest or some other furnishing; the other consisting of a nail and two large spikes from near the center of the The Houses Described

135

room. A foundation just inside the door to the complex might have supported a counter. The other shop (n) contained another slate ‘‘table top,’’ a marble basin, another mortarium, two louters (one of stone, one of terracotta), and at least eleven storage amphoras, a red-figured pyxis lid, as well as more unmended vases and seven loomweights. The quantities of coarsewares, mortaria, storage vessels, and ‘‘table tops’’ in these shops seem like equipment for some kind of food processing. It is odd, therefore, that there were no grindstones and no obvious cooking areas in the house. These ‘‘shops,’’ then, might have been used for preparing food either for the main part of the house or, more likely, for sale outside the house. Sales Inscription An almost complete sales inscription was found broken and scattered around the door into room p, parts of it leaning against the wall east of the door where it had apparently been set up as a public record.70 The text records that when Aristoboulos son of Kallikrates was eponymous priest, Zoilos, the son of Philokrates, bought the house from Diopeithes for 1,200 dr. The house thus sold for less than a quarter the price of house A v 10, although the two sales took place in the same year and the houses are generally comparable (below, chapter 6, ‘‘Sales and Prices of Houses’’). Another stone, with incised lines to guide a text which had not been written, was found in the door between rooms p and q, and a fragment of a second sales inscription was found in the street in front of the next house to the west.71 Summary Although irregular in plan, the House of Zoilos shows clear organization in its use of space, although this is quite different from the organization of other houses. The best-lit areas of the house adjoining the court, rooms c, g, h, k, and p, oddly enough contained relatively few finds and have little direct evidence of use. Instead, room b seems to have been used for weaving (although it is also possible that the loomweights were simply stored in this room and the loom would have been set up and worked in another space like the pastas). Room d probably was a pantry for storing domestic equipment handy to the pastas. There are no obvious cooking areas; this may have been done on portable braziers in the court or porticoes. By contrast, the two shops, which do not communicate with the rest of the house, had very rich assemblages of artifacts and might have been small restaurants or other food-processing businesses. They may have belonged to Zoilos or been sold or let separately. If involved with food processing, they might have served the household as well as producing goods for sale outside the house and so displaced some of those activities inside the domestic portion of the house. 136

The Houses Described

original limit of house

a

c

3 TC pithos lids (position unknown) b

e

lead disks M

d

spit? & 2 spit supports (hearth dismantled)

g

RF vase

vase sherds

f

vase sherds

basin

j

h

more than 24 TC figurines, many vases at south wall

i

eroded

k

M MMM MMM

67 loomweights, 29 at N wall

Street -iii 0

5 m

figure 30. The House of the Comedian

The House of the Comedian The House of the Comedian is one of the best-built and most elaborately decorated houses at Olynthus, with three mosaics, four rooms with painted walls, cutstone masonry on its street wall, stone capitals in at least two rooms, and other luxurious features (fig. 30). Situation Because the street plan was not completely clarified in this area, the situation of the House of the Comedian is not certain. It was very likely within the original city walls, and therefore it probably belongs in the grid of the North Hill rather than the grid of the Villa Section. It seems to be one of the earlier houses of the site; its pebble mosaic is dated by Salzmann to the end of the fifth or beginning of the fourth century.72 Therefore it probably belongs not with the later houses to the east in the Villa Section, but with the original layout of the North Hill in 432 b.c., and it may be house E −iii 6. The Houses Described

137

Plan The western part of the house has eroded away, leaving only the rooms on the east. Robinson and Graham assume that the courtyard was placed in the center of the house, and that in its final phase the house was larger than normal.73 As originally laid out, however, the house was probably of normal size, since a jog or seam in its north wall occurs at the standard plot width. It may have expanded later to acquire part of the plot to the west—as did house A v 6—to create a house about half again as large as normal; or the house may have been of normal size, and Robinson and Graham’s reconstruction (followed here) incorrect. The court had at least three porticoes and perhaps a full peristyle. To the north was a kitchen-complex, a single North Room, and perhaps more rooms in the northwest corner (reconstructed by Robinson and Graham as another andron, this one with an anteroom). Along the east are three rooms: room g, an andron j, and a partly preserved room in the southwest corner of the house, k. No stairbase was preserved, but the presence of a pillar-partition in the kitchen-complex, albeit an odd one, suggests that the house had two stories. Court The courtyard is unusually well built, with a pebble mosaic pavement, three porticoes or a full peristyle with stone pillar capitals (five of which were found), and painted portico walls. A hole in the center of the court was probably left when a built altar was removed. There were few finds in the courtyard: a lead weight, two lamps, a fishhook, an earring, three loomweights, miscellaneous hardware. Pastas and Porticoes As is common at Olynthian houses, the pastas (f ) and one preserved portico (i) contained more artifacts than the court. At the east end of the pastas were a number of vases including a basin; the others were not mended or saved. These might have been stored on shelves or in a cupboard here. Other objects were scattered along the north wall: a bowl or saucer, a plastic vase in the shape of a woman’s head, a strainer, and four loomweights. The assemblages here are not as complete or coherent as at some other houses. Kitchen-Complex The house has a well-defined kitchen-complex with bathroom and flue (rooms c, d, and e). Three marble blocks were scattered around the room, probably from a hearth in the center of the room, although this would be the only example of marble being used for this mundane purpose. The most unusual feature of the kitchen, however, is the alcove in the southeastern corner, separated from the kitchen by a row of four bases along the east wall. The base in the center of the alcove carried a column with a Doric capital, which was found in the eastern part 138

The Houses Described

of the pastas. This alcove is probably a later addition to the house, as the bases continue the line of the real east wall of the house, while the wall of the kitchen and the back wall of the alcove are both offset to the east. Graham identified this as a second flue in addition to the one on the west side of the room (d) and later restored a large window in the back wall of this alcove facing out onto the plain.74 But the kitchen-complex already has a ‘‘flue’’ at its western end for cooking, and a pillar-partition seems unnecessary here simply to provide an additional means to exhaust smoke from the room. This alcove might have served as a light well similar to those in the House of Many Colors, the Villa of the Bronzes, and other houses, making the kitchen a better-lit workplace. It is possible that in its original phase before the addition of the alcove, the kitchen did have a window here, as Graham reconstructs. When a neighboring house was built to the east, this window would have been blocked, and a light well was added instead to illuminate the kitchen, slightly encroaching onto the neighboring lot. A movable partition could have closed the light well in inclement weather. The kitchen contained many household objects, mostly in two clusters. In the southeast corner was a group of vases and terracottas including two bowls and a one-handled cup, a lekanis and lid, a squat lekythos, a lamp, one complete and two fragmentary terracotta figurines, three loomweights, and two coins. These might have fallen from a furnishing set against the south wall of the room. Another cluster of objects was found in the northeast corner, including a fishplate, a lead ‘‘loomweight,’’ and two lead disks, convex on one side and flat on the other. These last three may all be weights of some sort. These too were probably stored in a cupboard or chest, as a number of iron clamps and other fragments of hardware were found among them. The location of these two groups of vases at the eastern end of the room might suggest that this area was better lit by the light well and so served as a workspace. In the northwest corner of the kitchen were two intact storage amphoras, for storing foodstuffs or for water for the adjoining bath. In the center of the room was another group of objects: a bronze rod 0.14 m long, perhaps a spit; two terracotta objects identified as spit supports; and two lamps. If properly identified, the spits and supports are among the rare pieces of evidence for cooking in the built hearth in the center of these rooms. The floor of the flue (d) was paved with slabs, as was common in these rooms. It had a door to the pastas, and a low foundation separated it from the kitchen, but without preserved bases of a pillar-partition. The fieldbook records no trace of burning, but this may simply not have been noted. The flue contained a fine red-figured calyx krater depicting a victorious charioteer, a black-glazed fishplate, and a lamp.75 This is an unusual assemblage of fine tableware for what in most houses seems to have been primarily a cooking area; it may not have been in use as a cooking area at this time, but rather stored dining and drinking vessels. The Houses Described

139

The bathroom (c) was also unusually well appointed, with a pebbled floor, a bathtub in situ, and a drain running through the door into a basin in the corner of the kitchen. Other than the tub, the bathroom was empty. The presence of so many objects in this kitchen is somewhat unusual at Olynthus. In many houses, the flue contained vases and other finds, while the kitchen was relatively bare of artifacts.The situation seems to be reversed here: the kitchen had clusters of objects as if stored in cupboards or the like, while the flue had only three vases—albeit one very fine one. The kitchen in this house was probably well lit from both the flue and the light well alcove on the east wall, making it a more attractive workplace than the kitchens of most houses; and work that in other houses might have been done in the pastas and porticoes might have been done here in the kitchen. North Room(s) Room b on the north contained three ‘‘pithos lids,’’ two of terracotta (0.26 and 0.35 m in diameter) and one of stone (0.88 m in diameter). If these are indeed pithos lids, they might have belonged to three pithoi which were either shattered and not recorded in excavation, or which had been salvaged after the destruction. The room might therefore have served as a storeroom like those nearby in the Villa Section. Andron The andron (j) was smaller than the usual size (3.65 × 3.35 m), probably a fivecouch rather than the standard seven-couch size. Its mosaic floor and red and yellow painted walls make it among the best-appointed rooms of the house. The artifact assemblage from this room is strikingly different from that of any other andron at Olynthus. Scattered on the floor along the south wall was a group of twenty-four more or less complete and many more fragmentary terracotta figurines. Nineteen of these were standing female figures and female protomes; there were also a male nude, two Pan figures, a Silenus, and a rooster. Also along the south wall were some sixteen vases, including two plates, a cup-kantharos, a lid, an unslipped bowl, a juglet with its neck trimmed off, three lekythoi (one redfigured), six saucers, and the toe of a storage amphora. A pair of glazed terracotta beads in the shape of grotesque heads, perhaps amulets, was also found near the south wall. Clusters of terracotta figurines were not infrequent at Olynthus, but were most common in courtyards, pastades, and porticoes, in the open areas of the house. The situation is, however, paralleled at Priene, where thirty-five terracotta figures and ten small marble statuettes were found in the andron of a house destroyed in the second century b.c.76 These latter terracottas, however, depict mostly Dionysiac subjects and Aphrodite, more in keeping with the functions of the andron —drinking and carousing. The female heads, protomes, and figures from the 140

The Houses Described

andron of the House of the Comedian are not easily reconciled with a sympotic function in the room. The group of vases is also somewhat unusual for an andron. Other androns at Olynthus are relatively bare. Nor does the assemblage contain what we would expect from a sympotic group: no kraters and only a single cup. These may represent the subsidiary vessels from sympotic assemblages: saucers for the game of kottabos, for instance. This absence does not necessarily mean that the room was not used for symposia because such vessels are largely missing from the archaeological record here; it does suggest however that the room had other functions as well. It is therefore difficult to reconcile the assemblage of this room with its architecture. Perhaps the room changed function during the life of the house and was converted from an andron to a cult area, with the andron reconstructed by Robinson and Graham in the northwest corner of the house used for symposia; or it may have served a double function, as both a ritual and a sympotic space. South Room: Shop? In the southeast corner of the house, room k was only partly preserved. Its south wall, the street wall of the house, was largely eroded away, as were parts of its west wall. The entrance to the room was not preserved; it may have had a door to the street rather than to the house. This would tend to identify the room as a shop rather than a domestic room. This room contained a large and varied group of artifacts. Against the north wall was a group of twenty-nine loomweights, and thirty-eight others were also found in this room (their exact findspot not recorded), bringing the total to sixty-seven. It seems unlikely that the room was used for weaving: it was small, crowded, probably poorly lit, and perhaps open to the street. The loomweights may have been stored in the room rather than set up at a loom, or they might have fallen from a second story. There was also a group of lead objects, most of which may be weights: three rectangular weights (one of them inscribed Μ, indicating that it weighed one mina), two loomweight-shaped lead weights (one conical, one pyramidal, found with the rectangular weights in the center of the room), a sheet or weight inscribed Χ on both sides, and a pair of long lead objects. Pottery included three storage amphoras and such tablewares as a bowl, a jug, a juglet, an olpe, a plate, and six more saucers. Other shops, such as those in A iv 9, had relatively large numbers of weights (although weights are not more commonly found in shops than in other types of rooms); and the domestic equipment in a shop otherwise apparently unconnected to the house is also paralleled at other shops. Again, however, it is difficult to determine with confidence whether the room was actually being used as a shop or workshop—particularly when the architectural evidence is ambiguous—and if so, what might have been produced or sold in it. The Houses Described

141

Summary Although basically regular in plan, the House of the Comedian offers a number of singular features. The unusual external light well in the kitchen-complex may be the result of later modification, added when the house to the east was built. The kitchen itself has a rich assemblage of artifacts, which is undoubtedly related to its superior lighting, making this one of the more important places in this house for women’s work. The pastas, porticoes, and courtyard, on the other hand, though unusually well decorated, with painted walls and a mosaic in the court, have relatively meager assemblages of artifacts. The andron has an unusual collection of terracotta figurines and vases, unparalleled at this site but perhaps documented at Hellenistic Priene. And finally, the room in the southeast corner of the house may have been a shop but contained a large number of domestic artifacts as well as weights perhaps used in trading.

The House of the Tiled Prothyron The House of the Tiled Prothyron, F −iii 9, lies one block south of the House of Many Colors, built on a terrace cut into the sloping hillside (fig. 31). The north wall of this block continues to the west, but excavation to the west failed to reveal internal house walls, and the adjacent plot may have been left undeveloped. Similarly, there was probably no house to the west of the House of the Twin Erotes, just to the south of the Tiled Prothyron. Perhaps because it had no neighbors to its west and so did not have to conform to the usual east-west axes that structure most blocks, the layout of the house was somewhat unusual. The two North Rooms, a and b, are very deep (6.52 and 6.32 m N-S, rather than the usual depth of about 4.8 m); the pastas quite shallow, the court relatively small and located in the very southeast corner of the house, leaving no rooms south of the court. In the southwest corner is an unusual complex of rooms, f, g, and k. And finally, two partly subterranean rooms, c and d, were probably used as storerooms. The northwest corner of the house was apparently not terraced, as bedrock was encountered just below ground surface. Robinson and Mylonas place the entrance to the house at room j, which had a tile floor, although there is no break in the east house wall here. There may, however, have been an earlier door into room a: a seam in the north wall of the room about 1.3 m from its northeast corner is probably the original edge of a door leading in from Street −ii.77 There is no sign of a second story. Courtyard, Porticoes, and Pastas The cobble-paved court was surrounded on at least three sides by porticoes. Robinson and Mylonas suggest that it had a complete peristyle, like the House of the Comedian, although capitals were found only for the western portico. In the 142

The Houses Described

Street -ii

blocked door?

b bedrock

rough trough or basin

3 lower, 2 upper grindstones; louter

position uncertain

e

coarse vase sherds

d lead dish

f

bases only

h

lid?

g

j

door?

M M

Avenue G

c

3 TC pithos lids (position unknown)

i

k

l

0

5 m

figure 31. The House of the Tiled Prothyron

center of the court was the base for an altar. The southeast part of the court had been destroyed, probably in the removal of some feature or installation. Because the court was buried under about a meter of fill, this gap cannot be the result of simple erosion.78 A pithos in the northwest corner of the court stored rainwater from the eaves. A slate pithos lid, heavily burned and broken, was found nearby. Along the north side of the court were many potsherds, perhaps coming from the pastas rather than the court itself, since they lay under tile fall from the pastas roof. Other than that, the courtyard and porticoes produced only a few miscellaneous finds: a perfume pot, fragmentary terracottas, three glass eye-beads, some hardware, and three capitals from the west colonnade. The assemblages offer no clue to the nature of the missing installation in its southeast corner. At the eastern end of the pastas were two marble louter bases on their sides and the base of a brazier resting in situ. The floor level seems to have dropped off here because the louter bases rested ca. 23–33 cm below the courtyard pavement. However, the louters were probably not buried below the earth floor, since their tops would have protruded well above floor level; they most probably belong with The Houses Described

143

the destruction assemblages. Louters found in the pastades of other houses may be associated with household cult, with food processing, or with both. The brazier suggests that the pastas and court were used for cooking, as this house has no specialized kitchen. A quantity of coarse vases found in the northwest part of the pastas may also have been cookingwares. Other finds from the pastas include a bronze probe and a bronze spatula, two bosses, perhaps the remains of furnishings in the western part of the room, a fragment of a female terracotta figurine, and two loomweights. North Rooms On the north side of the house were two large rooms of about equal size (a and b). Room a, which had been painted red, contained four portable altars, two of terracotta in the southern part of the room at floor level, a third of terracotta in the northern part of the room, found in fill about 0.75 m above the floor, and a fourth, fragmentary stone altar, its exact findspot not recorded.79 Some sort of furniture in the room is probably attested by hardware found near the center of the room, including bosses, a spike and other metal objects, and a Lakonian key. Nearer the door was a ‘‘keyhole reinforcement’’—a ring with two prongs on the back to attach it to a wooden object, perhaps a chest or the door itself.80 The altars suggest a ritual purpose for this room, probably related to the ritual artifacts from the three-room suite f-g-k (below). Neatly laid out on the floor in the northwest corner of the other North Room, b, were two upper and two lower grindstones. These were arranged with two lower grindstones one on top of the other, flanked on either side by the upper stones to form a north-south line about 0.7 m from the west wall of the room. Just to the east were a fine marble louter and its base, disassembled with the basin lying flat on the floor (fig. 32). A third lower grindstone lay beneath the louter base. Beneath these objects was grey and black ash. This was probably not the remains of a table, as the grindstones were too neatly arranged to have fallen from a raised wooden table. It may have been the remains of a very low wooden structure meant to make it easier to move the grindstones from one room to another, and to help collect the meal as it was ground (below, chapter 4, ‘‘Food Preparation’’). The louter here was probably used together with the grindstones for food preparation, perhaps for kneading dough. Louters were often found together with grindstones and other apparatus for food preparation and could serve a variety of purposes, not only for washing but also for kneading dough. In the northeast part of the room was a rough rectangular basin or shallow trough, perhaps also used in food processing. Among these large objects and between them and the west wall were a number of vases: a red-figured askos, a bolsal, a ladle, a lekanis and an oinochoe, and fragments of other vases. Iron fragments from this area are probably from shelv-

144

The Houses Described

[To view this image, refer to the print version of this title.]

figure 32. Grindstones and Louter in Situ in Room b of the House of the Tiled Prothyron. Olynthus 12, pl. 186:2 ing or furniture. More hardware, including four bronze bosses, nails, and other fragments, was found scattered in the northwestern part of the room. This room therefore seems to have been used for food preparation, while cooking was done in the more open areas of the house, the pastas and perhaps the courtyard. Storerooms In the final phase, room c was a storage room. Three terracotta pithos lids were found in this room; as in the House of the Comedian, the South Villa and other houses, the pithoi to which they belonged may have been looted, or broken and unrecognized among the many coarse sherds scattered about. Two more upper grindstones were found tumbled towards the south end of the room, a storage amphora was smashed near its center, and two lead weights, a number of vases, and many more unidentified sherds were scattered through the room. The anteroom of this storage room, room d, contained fragments of another storage amphora, the spout of a terracotta ‘‘urinal,’’ and a few other miscellaneous objects. Burning in its northwest corner, mixed with bronze and iron hardware, might represent the remains of perishable supplies stored here. But the storeroom was unusual in having red painted walls with a white baseboard, and this was probably not the original function of the room.

The Houses Described

145

Southwest Suite In the southwest corner of the house was a suite of rooms consisting of two rooms of approximately equal size (f and g), entered from a wide transverse anteroom (k). At first glance, this arrangement resembles in plan the standard pastas– North Room complex in most Olynthian houses. But whereas the pastas is a semiopen space which mediates between the open courtyard and the enclosed North Rooms, the anteroom here is roofed and enclosed, separated from the rest of the house by a door, and limits access to the two back rooms. The suite is almost unparalleled at Olynthus, but it is very similar in plan to the three-room suite commonly found at Eretria and other sites (fig. 45).81 Rooms f and g were both painted with blue, red, and yellow paint, perhaps with the main part of the wall red, a yellow baseboard, and blue surbase as in the House of Many Colors. These rooms then are among the best decorated in the house, and they may have had a public character akin to an andron, which seems to have been the function of some of the corresponding suites at Eretria. The artifact assemblages from these rooms, however, suggest very different uses. The western room (f ) contained only a squat lekythos with a net pattern, the arm of a terracotta doll, the head of a female figurine, and a fragment of glass. The eastern room (g) contained two portable altars. One, of stone, lay on the floor in the southwest corner of the room; the other, of terracotta and fragmentary, was found near the center of the room. A louter stood in the southwest corner of the room. In the northwest corner were found two miniature cups and a lead dish with a ribbon handle found filled with ash.82 Room k, the broad anteroom through which f and g were entered, was almost empty, containing a bronze disk with a scalloped edge, perhaps a game token, a stamped amphora handle, a slingbullet, and a red-figured sherd. This architecturally unusual suite is probably associated with ritual rather than domestic or sympotic functions. The combination of louter and portable altars has been noted elsewhere as ritual equipment, at the House of Many Colors, the Villa of the Bronzes, and house A iv 9. This suite f-g-k, however, is much more secluded than the pastas and other open spaces in which similar assemblages are found in other houses. Summary Notable in this house is the relative scarcity of specialized domestic spaces. Room b was used for food processing, while the brazier, louteria, and coarsewares in the pastas suggest further food processing and cooking areas. Seventeen loomweights were found in the house, but scattered a few to a room or without a secure findspot; the household engaged in weaving, although we cannot determine just where. But there was no specialized kitchen, no andron. This need not imply, of course, that the building did not serve as a residence; but in the use of space, other concerns seem to have occupied its owners. 146

The Houses Described

A large proportion of space and many artifacts seem related to ritual. The house contained one built and six portable altars, one louter associated with altars and three others used in food preparation, and other apparently ritual objects, the largest quantity found in use in any house in Olynthus. Rooms a and g do not seem to have served domestic functions, and their architectural arrangements— the blocked door to the street in room a, and the unusual grouping of rooms f, g, and k, also suggest special purposes for these areas. This house therefore combined domestic and special purpose ritual spaces. It might have served as a sanctuary with quarters for attending staff and cooking facilities, but it is more probably the house of a household particularly concerned with ritual. Robinson and Mylonas suggest that it was the house of a priest.83

c on c lu s i on This account of a few houses at Olynthus should give a sense of the variety and range of activities and modes of organization found in the city. Despite the general similarity among the houses, they used space very differently. Architecturally similar spaces—courtyards, pastades, androns, kitchen-complexes—had widely differing assemblages of artifacts found in them. These assemblages do not, of course, perfectly reflect the activities going on in these spaces when the city was destroyed; but they are related to the uses of these spaces. We can use these assemblages and patterns to investigate how different activities and parts of the Greek house were organized, which is the focus of the next chapter.

The Houses Described

147

chapter four

The Houses Organized

The thirteen houses described in chapter 3 illustrate the versatility of what seems, at first glance, like a very standardized house design. In spite of the general uniformity in plan, individual households arranged and used their spaces very differently. Architecturally similar spaces, such as courtyards, pastades, kitchen-complexes, and the like, could be used in different ways, depending on the specific needs of the household. This diversity is primarily reflected in the assemblages of artifacts found in different spaces, which help us reconstruct how space was used at the end of the household’s life. To reduce this diversity to a few simplified generalizations ignores the fundamental fact that houses and households are unique and obscures patterns which can reveal important structures within the city. In this chapter I will therefore try to consider the diversity among Olynthian houses, in addition to what is ‘‘normal’’ in this city.

t e x t s , a r c h i t e c t u r e , a n d a r t i fa c t s : i de ol o g y, i n t e n t, a n d p r a c t i ce Greek literary sources offer a number of insights into the organization of the ancient house. In the most complete preserved description of a Greek house, Ischomachos, the ‘‘Ideal Gentleman’’ in Xenophon’s Economics, describes his house to Socrates, leading him through it as he led his new wife through the house: ‘‘‘Well, I thought it was best to show her [Ischomachos’s wife] the possibilities of our house first. It is not elaborately decorated, Socrates, but the rooms are constructed in such a way that they will serve as the most conve148

nient places to contain the things that will be kept in them. So the rooms themselves invited what was suitable for each of them. Thus the bedroom [θάλαμος], because it was the safest possible place, invited the most valuable bedding and furniture. The dry storerooms called for grain, the cool ones for wine, and the bright ones for those products and utensils which need light. I continued by showing her living rooms for the occupants, decorated so as to be cool in summer and warm in winter. I pointed out to her that the entire house has its façade facing south, so that it was obviously sunny in winter and shady in summer. I also showed her the women’s quarters [γυναικωνῖτις], separated from the men’s quarters [ἀνδρωνῖτις] by a bolted door, so that nothing might be removed from them that should not be, and so that the slaves would not breed without our permission. For, generally, honest slaves become more loyal when they have produced children, but when bad ones mate, they become more troublesome. ‘‘‘After we had gone through these rooms,’ he said, ‘we sorted out their contents by type. We first began putting together the things that we use for sacrifices. After that we separated the fancy clothing that women wear at festivals, the men’s clothing for festivals and for war, bedding for the women’s quarters, bedding for the men’s quarters, women’s shoes, and men’s shoes. Another type consisted of weapons, another of spinning implements, another of bread-making implements, another of implements used for other food, another of bathing implements, another of kneading implements, another of dining implements. ‘‘ ‘And we divided all this equipment into two sets, those that are used daily and those used only for feasts. We set aside the things that are consumed within a month, and stored separately what we calculated would last a year. That way we shall be less likely to make a mistake about how it will turn out at the end of the year. When we divided all the contents by types, we carried each thing to its proper place. After this, we showed the slaves where they should keep the utensils they use every day—for example, those needed for baking, cooking, spinning, and so forth, and we handed these over to them and told them to keep them safe.’ ’’ (Xen. Ec. 9.2–11, trans. Pomeroy) Ischomachos organizes his household space and goods by various criteria. The use of space corresponds to the physical qualities of the room, such as security, dryness, warmth, and access to light. The house also reflects the social divisions of the household: there are separate female spaces and male spaces, the gynaikonitis and andronitis. Interestingly, there is no such division of the house into free and slave quarters. Although the Greeks considered slaves to be less than human, slaves and free members of the household inhabit the same spaces in Ischomachos’s house, with female slaves in the women’s quarters and male in the men’s. The equipment of the household is likewise divided according to the gender of The Houses Organized

149

the user, its purpose, the occasions on which it would be used, and other criteria. He describes a close correspondence between the forms and the uses of space, so that the rooms of this ideal house are perfectly suited for the activities that go on in them, and a perfect organization of space, so that everything had its proper place. Maintaining this organization was demanding and time-consuming; indeed, much of the dialogue is concerned with teaching the wife and slaves how to create and maintain order in the house. The idealized, normative view of the use of space in a house presented here and elsewhere in Greek literature has led scholars to treat the archaeological remains of Greek houses as if they closely reflected this ideal situation. But as we have seen, real Greek houses were much more messy than Xenophon’s utopian estate. A particular space did not have to serve a specialized purpose, and indeed, one space could be used for many different functions: ‘‘For it is not the lodgings and the houses which give their names to the men who have lived in them, but it is the tenants who give to the places the names of their own pursuits. Where, for example, several men hire one house and occupy it, dividing it between them, we call it an ‘apartment house’ [συνοικίας], but where one man only dwells, a ‘house’ [οἰκίας]. And if perchance a physician moves into one of these shops [ἐργαστήρια] on the street, it is called a ‘surgery.’ But if he moves out and a smith moves into this same shop, it is called a ‘smithy’; if a fuller, a ‘laundry’; if a carpenter, a ‘carpenter’s shop’; and if a pimp and his harlots, from the trade itself it gets the name of ‘brothel.’’’ (Aeschines 1.123–4) Much of the domestic space at Olynthus was not architecturally specialized— that is, the building alone need not significantly constrain how the space was used or how it would be perceived by an outsider or household member. Moreover, the same types of rooms, courtyards, pastades, kitchen-complexes, or androns, had very different assemblages of artifacts in different houses, suggesting that they were used for different purposes. They could be and probably were used differently according to the season or the current composition and specific needs of the household. We must be careful to distinguish between the ideology of the Greek house reflected in Greek philosophical and legal texts; the architectural layout of houses, which reflected necessary compromises between an ideal and what the household needed and could afford; and how space was used in practice, which may have been quite different from both the ideal and from what the designer of the house intended. The actual use of space is reflected in the archaeological record primarily by the assemblages of artifacts found in the rooms, not by the architecture. Here Olynthus, alone of excavated Greek cities, allows us to compare the architectural layout of space with how the spaces were used.

150

The Houses Organized

g e n de r e d d i v i s i on s of s pa ce : l i t e r a r y s o u r ce s In a speech dating to the early fourth century, the orator Lysias gives us an intimate glimpse into a Greek house. Two citizens, Simon and the unnamed defendant, were rivals for the affections of a Plataean youth, probably a slave, named Theodotus. Their antagonism reached the point of violence: Simon tried to abduct the youth, was wounded by the defendant in the process, and later brought charges of assault with intent to kill against him. The defendant describes how early in their rivalry, Simon, ‘‘hearing that the boy was at my house, he came there [to the defendant’s house] at night in a drunken state, broke down the doors, and entered the women’s rooms [ gynaikonitis]: within were my sister and my nieces, whose lives have been so well-ordered that they are ashamed to be seen even by their kinsmen. This man, then, carried insolence to such a pitch that he refused to go away until the people who appeared on the spot, and those who had accompanied him, feeling it a monstrous thing that he should intrude on young girls and orphans, drove him out by force’’ (Lysias 3.6). Lysias plays on a series of strongly held Athenian mores: the privacy of the home, which must not be violated by uninvited strangers, especially at night; the good breeding and extraordinary modesty of the defendant’s womenfolk; the support of the defendant’s neighbors, who recognized and acted on Simon’s insolence; and the existence of an especially private area of the house, the women’s quarters or gynaikonitis. In relating how Simon transgressed all these boundaries, Lysias seeks to demonstrate his outrageousness and audacity, qualities which he will dwell on through the rest of the speech. This division of Greek houses into male and female areas is alluded to in other Greek and Latin sources. Xenophon, in the passage quoted above, describes how the women’s quarters were separated from the men’s quarters by a locked door, both to protect its valuables and to prevent female and male slaves from cohabiting. One of the women in the Thesmophoriazeusai of Aristophanes complains that thanks to Euripides’ portrayal of treacherous women in his plays, husbands are now jealous and alert to the tricks they used to play, and so ‘‘we are incessantly watched, we are shut up behind bolts and bars in the gynaikonitis, and dogs are kept to frighten off the adulterers’’ (Aristoph. Thes. 414–17). Vitruvius, in his problematic description of the layout of Greek houses, describes the gynaikonitis as ‘‘the large rooms in which the mistresses of houses sit with their wool-spinners. To the right and left of the prostas there are chambers, one of which is called the ‘thalamos,’ the other the ‘amphithalamos.’ All round the colonnades are dining rooms for everyday use, chambers, and rooms for the slaves. This part of the house is termed ‘gynaeconitis.’ ’’ 1 Perhaps the most famous description of gendered spaces in Greek houses comes from another speech of Lysias, On the Murder of Eratosthenes.2 The defendant,

The Houses Organized

151

Euphiletos, is accused of murdering Eratosthenes; he defends himself by claiming that Eratosthenes was having an affair with his wife, and that he had caught the two in flagrante; his murder was therefore legal under Athenian law, which allowed the husband to kill an adulterer. In his defense, Euphiletos gives the jury an intimate description of his ‘‘little house:’’ ‘‘Now in the first place I must tell you, sirs (for I am obliged to give you these particulars), my dwelling is on two floors, the upper being equal in space to the lower, with the women’s quarters [γυναικωνῖτις] above and the men’s quarters [ἀνδρωνῖτις] below. When the child was born to us, its mother suckled it; and in order that, each time that it had to be washed, she might avoid the risk of descending by the stairs, I used to live above, and the women below’’ (Lysias 1.9). These accounts, although often taken more or less literally as describing actual Athenian houses, are surely rhetorical, idealized houses mirroring households as ideal, and as unreal, as that of Ischomachos. But they must have been plausible to the judges and readers, and show that the seclusion of women was at least publicly acknowledged. The gynaikonitis was not, however, entirely a fiction created by lawyers and philosophers. This part of the house could, for instance, be rented out separately from the rest of the house. Accounts of the temple wardens on Delos in the third century b.c. record separate rental income from andrones and gynaikonitis of a house.3 This does not necessarily imply that these spaces were being used by men and women respectively; the gynaikonitis might have been a convenient term for a specific part of the house, regardless of its actual use; but it does prove a general recognition of the term in a bureaucratic rather than a rhetorical or philosophical context. This seclusion of Greek women in the gynaikonitis has been the subject of much scholarly debate. A number of studies have tried to identify the gynaikonitis in the architectural remains of Greek houses. In her pioneering article on Classical houses, Susan Walker offered reconstructions of Athenian houses with separate men’s and women’s areas, which have been widely reproduced.4 Others have also located the gynaikonitis in specific parts of the house: in the kitchen-complex of Olynthian houses or in the back rooms or ‘‘oikos’’ of houses at Priene.5 Other scholars, however, have found little or no evidence in the archaeological remains for separate men’s and women’s quarters. Jameson argues that ‘‘with the exception of the andron, which is at once distinctively male and oriented to the outsider, the architecture of the Greek house does not reflect the powerful social and symbolic distinctions between the two genders. Attempts to divide space along these lines are arbitrary and obscure the flexibility of use and a broader unity.’’ 6 Nevett follows his conclusions: ‘‘a true example of the gunaikonitis has yet to be found. It is a fact that the lack of upper storeys for investigation is a handicap in a situation where literary evidence has often been used to suggest that a female area existed in an upper storey, but even where houses have been assumed by the 152

The Houses Organized

excavators to have been single-storey constructions, no gunaikonitis has yet been convincingly identified.’’ 7 Faced with a rather deafening lack of architectural evidence for specific, separate men’s and women’s quarters in the ground plans of Greek houses, many have proposed that the women’s quarters were generally on the second floor of the house.8 This reconstruction is based almost entirely on the description of Euphiletos’s house in Lysias 1, quoted above. But rather than proving that women’s quarters were normally on the upper stories of Greek houses, Lysias’s speech shows how the arrangement of space could be adapted to fit the changing needs of the household. When the husband was first married and did not yet feel confident in his wife’s self-control, he arranged the household with the women’s quarters in the more secure, restricted second story; and he emphasizes this to the jury to demonstrate the care he took to preserve his household. However, ‘‘when the child was born to us, its mother suckled it; and in order that, each time that it had to be washed, she might avoid the risk of descending by the stairs, I used to live above, and the women below’’ (Lysias 1.6). And thus his wife was able to deceive him and carry on an affair with an adulterer. The ease with which Euphiletos claims to have reorganized his household space should warn us against looking for architecturally specific women’s quarters. Nor do other references to the gynaikonitis suggest that it was on the second story. The account in Lysias 3 quoted above, in which Simon ‘‘broke down the doors, and entered the women’s rooms’’ looking for Theodotus, makes no mention of climbing up to the second story; nor does Xenophon refer to a stair separating the men’s and women’s quarters, but rather a barred door. The second story may well have been used by women, but there is no justification for the common claim that it was normally the women’s quarters or that women were in any way restricted to it. Attempts to identify a more restricted women’s quarters in the architecture alone have been notably unsuccessful. Virtually none of the discussions of male and female space in Greek houses, however, deals in any depth with artifacts as evidence of how rooms were used, although many lament the absence of such data.9 To understand the use of space, we should look not only at the architectural setting of household space, but also at the areas of the house where tasks traditionally performed by women, including weaving, cooking, childcare, and so forth, were actually carried out. We may then consider what architectural features these spaces share, what features may be considered essential in looking for women’s quarters, and how household space could be organized around these tasks.

the k i t c he n - c o mp l e x Preparing food and cooking were among the main duties of Greek women and the female slaves of the house. The houses at Olynthus are unusual in having an archiThe Houses Organized

153

tecturally specific complex of rooms which seems to be associated with cooking: the kitchen-complex. This suite of rooms is not exactly paralleled at other Greek cities. Some houses at other sites had large, rather plain rooms associated with bathrooms and sometimes additional spaces similar to the Olynthian ‘‘flue;’’ but none have the pillar-partition so characteristic of the Olynthian houses, nor the evidence for cooking in the flue.10 This kitchen-complex was found, in some form, in some forty-four houses at Olynthus, nearly half of those excavated. The basic layout of the kitchen-complex was outlined above, and individual complexes were described with the houses in chapter 3. Again, the names for individual rooms are essentially arbitrary, as there are no proper English equivalents. Rather than label rooms with meaningless letters or numbers, I will refer to the main room as the ‘‘kitchen,’’ the small area on one end as the ‘‘flue,’’ and the even smaller room which sometimes occupies one portion of the flue, and which frequently contained a bathtub, as the ‘‘bath’’ (fig. 14). Some possible Greek terms are suggested below. Although the type is fairly coherent and recognizable, the arrangements in different houses varied considerably. Some kitchen-complexes lacked separate bathrooms, although a bathtub could occasionally be placed in the flue or in another room of the house. About a quarter of the kitchen-complexes had stone-built hearths in them, and the cut stones may have been robbed from others, as they probably were, for instance, from house A 8. The floor of the kitchen was always earth; the walls of about two-thirds of these rooms were unplastered, but six had plastered walls, and two kitchens—in the House of Many Colors and B vi 4—had painted walls. In general, though, these were unpretentious rooms, undecorated even when other rooms of the house were painted or equipped with cement or cobbled floors. The most distinctive feature of some of these complexes was the pillar-partition which in many houses separated the flue from the main room of the complex. This was often found in houses with other evidence for a second story, and in these cases, the pillars undoubtedly supported the wall of a second-story room above the kitchen, while the flue continued through the second story to the roof. The opening of the flue through the roof must have been closable in some manner because there are no provisions to carry off rainwater, which would have flooded the room if it were open to the weather. Tiles found in the flue of the House of Many Colors and a few other houses suggest that it was at least partly covered by a raised section of tiled roof, but other arrangements are also possible. Graham and others suggested that the pillar-partition was intended to help evacuate smoke from the hearths of the main room. But this is an inefficient way of ventilating a room and cannot be the sole explanation for these constructions.11 If only small openings at the top of the pillar-partition were left between flue and main room, as Graham reconstructs, it would have been much simpler to build a solid wall pierced by windows. The use of wooden pillars here must have aimed 154

The Houses Organized

instead to open the space between kitchen and flue as much as possible. The partition could have been open nearly to the floor, leaving at most a low foundation separating the main room from the flue. The flue essentially formed a bay at the short end of the main room, rather than a completely separate space, although it could be entered through its own door from the court or pastas. This arrangement probably served two purposes. Fires built in the flue would have heated the main room during the cold Macedonian winters, and light from the shaft would have illuminated the main room. The flue thus acted both to exhaust smoke from cooking and heating fires and as a light well open through the second story. Not all kitchen-complexes had pillar-partitions separating the main room from the flue, however. It would have been unnecessary in one-story houses, as there would have been no second-story wall to support. And not every complex in a twostory house had this distinctive feature. For instance, four of the five houses in the southern row of block A vii had stairbases, so this row of houses must have had two stories. All houses had kitchen-complexes, but none of them preserved pillar bases. The pillars could have been set on the foundation without bases, or the wall of the second-story room might have been made from lighter materials, carried simply on a beam without support from below. This probably reflects different building traditions in different blocks, a theme which will be discussed further in chapter 5. Even in rather fancy two-story suites, stone bases were not always used. The suite a-b in the Villa of the Bronzes, for instance, was similar to a kitchencomplex in having a light well at one end of a larger room, separated by a pillarpartition. But while the pillars had stone capitals here—a feature never found in kitchen-complexes—they were set directly on a rubble foundation without bases. The lack of stone bases does not imply that there was no pillar-partition, or that the house did not have two stories. Assemblages and the Use of Kitchen Space The artifact assemblages from kitchen-complexes vary greatly from house to house; there is no single typical kitchen assemblage. But there are some significant patterns which help us understand how these spaces were actually used. The flues of some sixteen of the forty-four kitchen-complexes excavated at Olynthus contained ash, traces of burning, and other evidence of fires on the floor. They sometimes contained cooking equipment, and bones and other remains of foodstuffs were occasionally noted. Considering the often haphazard recording of coarse cooking pottery and ash, especially in the early years of excavation, this pattern is significant. The flue was therefore an important location for cooking in many houses. Fires were probably built directly on the floor, which was occasionally paved with stone slabs or cobbles and showed traces of burning, and food was cooked in cooking pots, lopades, barrel-ovens, on spits, and the like.12 But the flues of a number of houses showed no evidence of burning or cooking The Houses Organized

155

despite careful excavation and recording.13 Although such negative evidence is not conclusive, some households may well have cooked in spaces other than the flue, at least at the time when Olynthus was destroyed, and the flue may have been used for other purposes. This is the common pattern in houses at other Greek sites, which seem to lack specialized cooking areas. In some houses, such as the House of Many Colors and A iv 9, the flue also contained tablewares and other domestic equipment, and so seems to have been used as a storage pantry as well as for cooking. Storage amphoras and other storage vessels were sometimes found in flues, suggesting that they could be used to store small quantities of foodstuffs and water for cooking as well. Although the main room of the kitchen-complex was sometimes equipped with a built hearth, these hearths were almost never used for cooking. They never contained bones or other remnants of cooking, only pure ash; and only in the House of the Comedian was any cooking equipment found in the main kitchen. These hearths were therefore probably used to heat this room rather than to cook food. The kitchen, well lit and heated from the light well/flue, has commonly been identified as an important workroom. It was usually among the largest rooms in the house, averaging 26.3 m2. A few kitchens have permanent installations in them. The kitchen of A vii 4, for instance, had a large stone mortar in it, while stone foundations, perhaps for cupboards, were found in other kitchens (although no artifacts were found with them).14 Yet the kitchen was not a formal entertaining or reception room: it was earth-floored and generally unpainted. Mylonas identified this room as the oecus mentioned by Vitruvius, ‘‘the place where the women of the household spent a good deal of their time in the performance of their daily tasks’’; and this function, if not the ancient name, has been accepted by most scholars.15 One would therefore expect to find the apparatus belonging to women’s tasks in these rooms, including loomweights and tools for weaving, such equipment as grindstones and mortars for processing food, containers for storing food, and other domestic implements. In fact, however, these rooms were remarkably bare, usually with only a few, fragmentary artifacts which do not form coherent assemblages. The kitchen of the House of Many Colors, for instance, the house with the best-preserved household assemblages at the site, contained only a scatter of artifacts: three storage amphoras, a saucer, a loomweight, two fragments of terracotta figurines, three coins, and a few other scattered artifacts (fig. 17). Compared to the rich and coherent assemblages from other rooms of the house, this meager group is not what one would expect in an important workplace. The flue of this house, for instance, contained twenty-two complete vases and many other artifacts, and the suite a-b contained fifty-two pots, seventy-five loomweights, hardware from furniture and a variety of other objects. This suite, rather than the kitchen, was where women carried out many daily tasks such as weaving. Of the forty-four kitchen-complexes

156

The Houses Organized

excavated at the site, only one, in house A xi 10, had definitive evidence for weaving, and only the kitchen in A viii 4 contained grindstones.16 The artifact assemblages, or lack thereof, therefore, refute the common claim that the main room of the kitchen-complex was the oecus of the house or generally an important workplace for the women of the household.17 There are exceptions to this pattern, of course. A few houses had richer artifact assemblages in their kitchens, but these houses tended to have limited domestic spaces. Houses A 8, A iv 9, and A xi 10, for example, had relatively large numbers of artifacts in their kitchens, including vessels for storage, tablewares, and other finds (although no loomweights or grindstones). But shops and workshops took up much of the area of these houses, leaving relatively little space for domestic purposes. The kitchen of the House of the Comedian contained two clusters of artifacts, including vases, storage amphoras, cooking equipment, weights, and other objects, while its flue contained a red-figured krater, a fishplate and a lamp, but no cooking utensils and no trace of burning or ash (above, chapter 3, ‘‘The House of the Comedian’’). This is an unusual situation, and the architecture of this complex is unusual as well, with an extra bay or light well at the opposite end of the kitchen. This made the room somewhat similar to the suites with light wells in the House of Many Colors and the Villa of the Bronzes, both of which also contained many artifacts for women’s work, compared with the kitchens of those houses. Significant collections of vases and domestic equipment were also found in the kitchens of A vi 2, A vii 4, A vii 6, B vi 2, and the kitchens of A 8, A vi 2, A vii 10, and B vi 4 contained pithoi for storage of relatively small amounts of foodstuffs (below, chapter 6, ‘‘Household Storage’’). But in general, most houses preserve little or no evidence for food preparation, weaving, or other women’s household tasks in their kitchens. Second Kitchens or Special-Purpose Suites? A number of houses had two suites of rooms with a small space set off from a larger room by a pillar-partition, presumably forming a light well or flue. Graham argued that these and other houses had two independent kitchen-complexes, used at different times according to the seasons, prevailing winds, or other factors.18 The cases of the House of Many Colors, the Villa of the Bronzes, and A iv 9 have been discussed above, where it was argued that the second suites were not the equivalents of the kitchen-complex but were special-purpose work or reception rooms lit by light wells. Other possible examples include A vi 7, A viii 2, and B vii 2. Although these suites do resemble kitchen complexes architecturally, they are quite distinct in other respects. Some were much better appointed: the suite in the Villa of the Bronzes had a mosaic floor, painted walls and stone capitals, and the rooms in the House of Many Colors were being redecorated to look similar. None

The Houses Organized

157

a

b

6 eye beads

d

e

c

coin meathoard, hook 379 BC

finger ring

2 bronze, 5 eye beads

f

1 upper grindstone, 1 saddle quern

Avenue F

coin hoard, 379 BC

g

fragment h

h´ - flue?

j

l

mortar lead bar k

i m

Street -i

coin hoard, buried 348 BC

0

5 m

figure 33. House ESH 4

of them had an associated bathroom or fixed hearth, although these are not indispensable features of the kitchen-complex. The suite in A iv 9 served as an entranceway, a position otherwise almost unprecedented among kitchen-complexes. And the artifact assemblages from these suites were distinct from those in kitchen-complexes as well. The rich assemblage of weaving and other equipment from the suite in the House of Many Colors is very unlike the range of objects found in kitchens, here and elsewhere. These rooms also had equipment associated with household cult: large numbers of miniature vessels in the House of Many Colors and the Villa of the Bronzes, a portable altar, louter base, and other materials in the suite in A iv 9. All these features are unattested in the kitchencomplexes of other houses. Despite the similarity in plan, therefore, these suites are different enough in appointment and contents to consider them separately from kitchen-complexes. A more plausible case of a house with two kitchens is ESH 4 (fig. 33). The northern suite of this house (rooms b, c, and d) had a pillar-partition separat-

158

The Houses Organized

ing the main room from a cement-floored bathroom and a flue. This suite corresponds closely to the most characteristic arrangement of kitchen-complexes. The finds from the kitchen of this complex included a louter base, a meathook, three bowls, a pelike and a saucer, and two fragmentary terracottas. The assemblage suggests that this room was used at times for cooking and food preparation, although perhaps not at the moment of destruction. A bathtub was found in situ in the bathroom (d), while the flue (e) was empty. A hoard of silver coins was found in the western part of the main room, and two more silver hoards were found in this house—the only house with so many coin hoards (below, chapter 6, ‘‘Coin Hoards’’). The southern suite, by contrast, was architecturally quite distinctive. It consisted of a relatively small space (h), 3.8 × 4.8 m, with two rooms or alcoves on the west side. The southern room (i) had a cement floor and a bathtub in situ. The northern alcove (h') might correspond to a flue, although no ashes or cooking artifacts were noted during excavation. If only the northern half of the house had a second story, no pillar-partition would have been needed to support a wall above this suite. This space might have been partly open to the sky to serve as a flue or light well. With its bath and a possible flue or light well, this suite resembles the standard kitchen-complex more closely than other suites like those discussed above. It has a number of peculiarities, though. If it is a kitchen-complex, the main room would be the smallest kitchen at the site, at 18.2 m2 compared to an average of 26.8 m2. A door from the street led directly into the main room h, a situation almost unprecedented at the site, as kitchen-complexes tend to be removed from the entrance of the house (below, ‘‘Conclusions’’).19 The flue and bath adjoin an outside wall of the house, while in virtually every other house, the flue is located in the interior.20 The irregular plan of the house must have affected the layout of rooms; but the spaces could have been arranged differently if this had been desired, with the bath and flue placed away from the exterior walls. This suite contained a large collection of artifacts used for weaving, food processing, and other women’s tasks. A stone mortar with a lid was set into the floor near the north wall of the main room, and an upper and lower grindstone were found nearby. In the south, just in front of the bathroom, was a group of thirty-nine loomweights, probably fallen from a loom and attesting weaving in this space. A stone disk 0.56 m in diameter, a hydria, a terracotta figurine of an actor, and other artifacts were found in the space as well. This assemblage is just what one would expect to find in other kitchencomplexes if they had been important workspaces for the women of the household. Yet this complex was architecturally atypical, unusually small and public. The kitchen-complex in the northern part of the house was larger, more private, and more typical in arrangement, and like so many other kitchen-complexes, it

The Houses Organized

159

had a much sparser artifact assemblage and seems to have been a less important workspace when the house was destroyed. This example proves the general rule suggested by the distribution of artifacts, then, that kitchen-complexes tend not to be primary loci of household work. Moreover, house ESH 4 may have been engaged in producing food on a larger scale than simply for the household. The other rooms in the south half of the house had doors opening onto the streets, and were probably shops or workshops. The double kitchen-complex may be related more to a commercial production in this house than to its domestic functions (below, chapter 6, ‘‘Cooking and Baking’’). The Use of the Kitchen-Complex Although the sophisticated architecture of these large but plain kitchen-complexes might suggest that they were important spaces for women’s work, the contents of most excavated complexes do not support this interpretation. The situation is paradoxical; there is a discrepancy between the architectural evidence for these kitchen-complexes and the evidence of the artifacts actually found in them. One solution is that the use of space in houses may have changed from season to season. The built hearths in the kitchens, which were used for heating rather than cooking, and their proximity to the flues with their open fires, would have made these rooms among the warmest in the house during the cold Chalcidic winters. Indeed such built hearths are only found in rooms otherwise identified as kitchens. Since they were relatively well lit by the light well/flue, they would have been more comfortable workplaces in cold weather than the more exposed courtyard or pastas. In the summer, though, this would not have been an advantage: the better-lit and airier pastas, porticoes, exedras, and courtyard would have been more comfortable areas for preparing food, weaving, and other household tasks, while other enclosed rooms might have been cooler since they were further from the cooking fires in the flue. Since Philip destroyed the city in late summer, the distribution of objects reflects how the houses were organized in that season, with most household equipment moved out of the kitchen into the more open spaces of the house. The kitchen-complex itself might have been arranged differently in different seasons. Wooden shutters installed in the pillar-partition could have opened or closed the division between the flue and main room. In the winter, the partition could be opened, and the kitchen would be heated by the cooking fires in the flue in addition to the fixed hearth. In summer, when one would want the main room to remain cool, the partitions could be closed and smoke and heat confined to the flue. Nevertheless, even during the summer months the flue probably continued to be the primary cooking area in houses so equipped. This explains the common access arrangement of the kitchen and flue, the flue being accessible from the 160

The Houses Organized

pastas or court where most food preparation was done, but not necessarily from the kitchen. There may be other reasons why the kitchen-complex was not as intensively used as we might expect, however. The engendering of household space was not simply a practical matter but strongly ideological as well; these issues will be discussed further below (‘‘Conclusions’’). The Kitchen-Complex: Greek Terminology A number of Greek terms may be proposed for these spaces.21 In Aristophanes’ Wasps, Philocleon tries to escape from his son, who has imprisoned him in his own house, by dodging into the cooking area (ἰπνός: Wasps 139 ff.). His son warns the slave not to let the father escape through the bath drain, demonstrating the connection between cooking and bathing areas. The father then tries to climb out the chimney or smoke hole (κάπνη), but the son and slave trap him by clapping a cover or board (τηλία) over the chimney. The ἰπνός was therefore a cooking area, large enough for a person to enter (and escape through), but small enough to be shut up with a board and near to the bath. This corresponds closely to the flue in Olynthian houses. The relation between exhausting smoke and admitting light is reinforced by a story in Herodotus, that tells how Perdiccas, an ancestor of the Macedonian kings, took as wages the sunlight which streamed in through the smoke vent (καπνοδόκη) in the roof (Hdt. 8.137). The sense of the scene is that the smoke vent provided much of the light in the room. Although the architecture of the kitchencomplex is not exactly paralleled at other Greek sites, some of the principles seem to be common elsewhere.

h o u s e h ol d ta s k s o u t s i de the k i t c he n Instead of focusing on specific architecturally defined room types and trying to link them with particular activities—the usual approach to the study of domestic space since Xenophon—we might instead look at the distribution of artifacts throughout the house to determine where activities were carried out in practice. Archaeologists frequently draw conclusions about the use of space from the architecture alone because most sites do not preserve significant assemblages of artifacts, and those which do are often not studied with these questions in mind. But as we have seen, architecturally similar spaces had very different groups of artifacts associated with them, implying that they were being used for different purposes. Although the study of artifact assemblages has its own difficulties, some of which were outlined in chapter 2, the objects found in the destruction debris offer independent and often compelling evidence for the use of space in practice.22 The The Houses Organized

161

relative lack of large assemblages of artifacts in kitchen-complexes should serve as a warning not to jump to conclusions about how space was used based solely on architectural criteria and hypothetical ancient terminology. Cooking Outside the Kitchen-Complex Most houses at other Greek cities lacked the permanent, well-defined cooking rooms which were so common at Olynthus. Instead, cooking was done in different rooms of the house, probably depending on weather, the number of guests, and other factors.23 Identifying specific cooking areas is therefore likely to be difficult or impossible, a problem compounded by the sparse recording of cooking and coarsewares at Olynthus. More than half the rooms at Olynthus which show evidence of cooking are flues. But cooking could be done in other rooms as well. Braziers could be used for cooking as well as for heating, and they were found throughout the site. They were particularly concentrated, however, in block A viii, where they were found in five of the ten houses.24 The large number of braziers in this block is probably not simply an accident of excavation. Two different excavators dug these houses, and their recording is by no means the most complete at Olynthus. If braziers were common in other houses, they would have been noted at least occasionally, particularly by some of the more observant excavators. Most of the houses with braziers in this block lacked kitchen-complexes, however, and might have used braziers for cooking and heating in part because they did not have specialized spaces for the purpose. Although this is undoubtedly due in part to incomplete recording of coarsewares, the concentration of braziers in block A viii suggests that its inhabitants had somewhat different cooking habits from people elsewhere in the city, and shared particular habits and customs among themselves. This is another way in which social ties among inhabitants of a block might be documented (below, chapter 5). In other houses which lack kitchen-complexes, there is little or no evidence for how the household cooked food. It is unfortunate that we do not have more detailed evidence for cooking in different houses, since this might have revealed further distinctions. Other than the flues in kitchen-complexes, there are few or no rooms where cooking was done on open fires on the floor.25 Instead, the flue seems to have been conceived as an integral part of the kitchen-complex and was not built as its own independent room, without the adjoining large space. While the flue, once built, could be used in different ways, it apparently was only built as part of a larger complex with specific features of design. There is thus a conceptual uniformity of architectural space which was not well reflected in the actual use of that space. The disjunction may result from changes over the years in the way spaces were used: over time rooms which were originally built in the same manner for the 162

The Houses Organized

[To view this image, refer to the print version of this title.] [To view this image, refer to the print version of this title.]

[To view this image, refer to the print version of this title.]

figure 34. Grindstones from Olynthus. Left: Saddle Querns. Right: ‘‘Hopper-Rubber’’ and Reconstruction of Hopper-Rubber in Use. Left: Olynthus 2, fig. 186. Right: Olynthus 12, pl. 189 from the House of the Tiled Prothyron, and Moritz, Grain Mills and Flour 45, fig. 4. same purpose came to be used differently, and their contents at the time the city was destroyed do not reflect their original purpose and use. Food Preparation Cereals constituted up to three-quarters of the caloric intake of ancient Greeks, and processing raw grain was an essential and time-consuming task of Greek women.26 Two types of grindstones were used for grinding grain at Olynthus: the saddle quern and the later and more advanced ‘‘hopper-rubber’’ (fig. 34). The saddle quern was used as early as the Neolithic period. It consisted of two grindstones, a rectangular lower stone and an oval upper stone, whose underside was often striated with narrow grooves to grind the grain more efficiently and to help direct the flow of grain and meal through the grindstones. The hopper-rubber was a more sophisticated machine, with a reservoir in the upper stone to hold unprocessed grain. This trickled through a slot in the center of the upper stone, so that the user did not have to lift the upper stone periodically to add more grain. The upper stone was equipped with a handle, so that it was moved side-to-side in an arc, rather than back and forth. The hopper-rubber was the more common type at Olynthus: thirty-nine exThe Houses Organized

163

amples were found, compared to fifteen saddle querns.27 Moreover, seven of the saddle querns from the site—almost half the total—come from a single house, A 6, which also had five upper stones of the more advanced type as well as an orbis for crushing olives. This household engaged in grinding grain and processing agricultural produce on a large scale, probably employing slaves for much of the work (below, chapter 6, ‘‘Agricultural Processing’’). Both types of grindstones seem to have been in use at the same time in domestic contexts, however, and sometimes in the same household. In addition to A 6, houses A 10, A vii 9, and ESH 4 had both types in use simultaneously, and in A vii 9 they were used side by side. Grindstones were made of imported lava or basalt and were relatively expensive. The Attic Stelai preserve one price of 7 dr 1 obol for an upper stone alone; another is restored as 9 dr 2 obols.28 Houses in Anatolia, Egypt, and elsewhere in the ancient world often had permanent grinding benches to raise the grindstones off the ground for more comfortable use.29 Saddle querns were usually set at a slight angle in these benches to allow gravity to carry the ground meal away from the grindstones. Hopperrubbers were set on a table of some sort since they were worked with a long handle which would have been quite inconvenient at ground level. Illustrations of this machine in use show it set on a table.30 Some Greek terracottas show women kneeling on the ground to grind grain, and others seem to show them standing; in these representations, however, it is difficult to distinguish between grinding grain and kneading dough.31 But houses at Olynthus lacked such permanent grinding installations. Grindstones were found on the floor, with no traces of stands or benches. The fine set of three lower grindstones and two hopper-rubbers was found neatly arranged on the floor in room b of the House of the Tiled Prothyron, resting on a layer of gray and black ash (above, chapter 3, figs. 31, 32). The ash might be the remains of a low wooden platform, but it cannot have been a proper table because the grindstones are too neatly laid out to have fallen from any height. The ash could also be the remains of a cloth or basketry to collect the meal, rather than a wooden platform or table. Nearby was a louter, carefully disassembled but probably used with the grindstones for kneading dough. The northern shop (d) of house A vii 9 also had a well-preserved grinding emplacement against its west wall. This consisted of a rather rough and deeply worn lower grindstone set into the floor, surrounded on the south by flat cobbles (fig. 35). Just to the south was a hopper-rubber, found neatly set by the wall, and nearby was a saddle quern; both these types were used side by side.32 It was not determined whether the lower grindstone was used with the hopper-rubber or the saddle quern, or perhaps both. The cobbled surface surrounding the grinding emplacement would have made it easier to collect the meal. These two examples seem typical, in that the grindstones were set directly on the ground. In most houses, however, the specific setting of grinding implements 164

The Houses Organized

Street viii b

c

d MM

upper & lower grindstones; saddle quern

catchbasin

e

f h g bronze & lead objects hardware?

including coin of Alexander

Avenue B

a

j i bronze vessel earlier building

0

5 m

figure 35. House A vii 9 was not noted. This in itself is significant, however: it shows that grindstones were stored like other movable household objects rather than set into permanent benches. A few rooms had low rubble foundations or benches in them, but they were never associated with grindstones. Olynthian women might have set the grindstones on wooden tables in place of benches; but these too would have been portable rather than fixed like emplacements in the Near East. One advantage of not having permanent grinding emplacements was that the grindstones could be moved to wherever it was convenient to use them. Some grindstones were found in storage areas rather than in spaces where they could be used. For instance, the set of grindstones found in the bath (g) of the House of Many Colors was probably stored there for use in another room. Two upper grindstones were found in the storeroom (c) of the House of the Tiled Prothyron, described above. But in most cases, grindstones were found in rooms where they would have been convenient to use. They were commonly found in open, well-lit spaces such as courts and pastades. In twelve of the twenty-eight houses which contained The Houses Organized

165

grindstones, they were found (and presumably used) in the court, pastas, or in an adjoining open space (plates 1, 2).33 In other houses, they were often found in rooms which adjoined the court, which could have been lit by windows from the courtyard. It is difficult to tell in these cases whether they were being stored here for use in the adjacent court or pastas, or whether they were actually used in these spaces.34 The two examples cited above are atypical, therefore, in that the emplacements were found in more enclosed rooms. Only house A viii 4 contained a set of grindstones in its kitchen, although a set of grindstones and a stone mortar were found in the possible second kitchen in house ESH 4, described above.35 Grindstones were never found in flues, which would have been rather cramped spaces for grinding grain. Grain processing was part of the larger process of food preparation and was usually done in the same spaces as other household tasks. In general, grindstones were not found alone in a room; they were usually found together with other finds, with table and plain vases, and also with louters, storage amphoras, clusters of loomweights, and the like. For example, in house A v 9 the pastas (e) was not only used for weaving, but for grinding, kneading dough, and other domestic activities (above, chapter 3, ‘‘House A v 9’’). The pastades of B vi 5, the House of the Twin Erotes, and the Villa of Good Fortune were also used for both processing grain and other domestic activities, as were the space opening onto the court in A viii 4 (m), the courtyards of A 6, A 8, A vii 10, the court and the second kitchen (h) of house ESH 4, and many other rooms. Other types of grinding implements included stone, wood and terracotta mortars and mortaria (ὅλμοι). Mortars were used with long wooden pestles to break up grain to free the chaff, before grinding it with a hopper-rubber or saddle quern; their use is depicted in a number of vase paintings and terracottas.36 They could also be used to crush other kinds of foodstuffs as well, such as vegetables, herbs, dried peas and legumes, fish, and so forth, although a different kind of mortar (a θυεία or ἴγδος) was more commonly used for these lighter tasks.37 Stone mortars were rarer than grindstones: only twelve were preserved at Olynthus, perhaps partly because they were larger and therefore easier to salvage than grindstones, or perhaps because mortars were commonly made of wood, as they are today. Like grindstones, stone mortars were fairly expensive. The one stone mortar listed in the Attic Stelai sold, secondhand at auction, for almost 9 drachmas, while a wooden mortar sold for between 3½ and 14 dr (the price is not completely preserved). Most households would have used less expensive pottery versions, one of which is listed in the Attic Stelai for less than 2 dr.38 Since mortars were primarily used for crushing grain before it was ground, one would expect them to be found together with grindstones. Yet they are found together in only two houses: in the pastas (e) of A v 9 and in room h, the ‘‘second kitchen’’ of ESH 4. A possible third example is the court of house A 10, which contained an upper grindstone, a saddle quern, and an oval stone bowl, perhaps 166

The Houses Organized

a mortar. Of course wooden mortars left no trace, and they might have been used more commonly with grindstones. Unlike grindstones, mortars were found more often in the more recessed rooms of the house, such as kitchen-complexes, rather than in open spaces, such as the court or pastas. A mortar was found in the kitchen of house A vii 4 and in ‘‘second kitchens’’ of two more, A vi 7 and ESH 4.39 In these latter two cases, it is notable that the mortars were found in the southern, more subsidiary suite of rooms rather than in the more conventional kitchen-complex on the north side of the house.40 The reason why stone mortars were used in more recessed spaces was probably that they were permanent installations and could not be moved around the house according to the weather or seasons the way grindstones could. In a sheltered room like a kitchen-complex they would have been usable all year round, whereas if they were placed in a more open space, inclement weather would have interfered with their use during the winter months. The issue will be discussed further below. Relatively few houses, however, seem to have had large stone mortars, and in a number of cases they seem to have been used for purposes other than purely domestic food processing. The stone mortar in house A vi 10 (i) was probably used in agricultural processing in that and adjoining rooms (j, k; fig. 53 and below, chapter 6, ‘‘Agricultural Processing’’). Another, in house A viii 8, was associated with a bakery (fig. 54 and below, chapter 6, ‘‘Cooking and Baking’’). Instead of expensive stone implements, most households probably used wooden or terracotta mortaria for everyday crushing and dehusking. Many more terracotta mortaria were found at Olynthus: twenty-three examples or possible examples are published or mentioned in the notes, from fourteen houses in different parts of the city, although like most coarsewares they are probably greatly under-represented. Unlike stone mortars, they were only rarely found in kitchencomplexes: one was found in the flue of house A vi 2 and another in the kitchen of B vi 2, but more often they were found in architecturally unspecialized rooms. Interestingly, they were never found together with grindstones, which raises doubts about the identification and use of some of these objects.41 Kneading troughs (κάρδοποι or μάκτραι) could be made of pottery, stone, or wood. Like mortars, stone versions were significantly more expensive than wooden ones and were probably not normal household belongings. A used stone kardopos sold in the Attic Stelai for 7 dr 3 obols, another probably for 7 dr 5 obols, and the base of a broken kardopos sold for 6 dr 3 obols; together they would probably cost 13–14 dr. A ceramic kardopos, on the other hand, is listed for 2 dr.42 The most likely example of such a stone kneading trough from Olynthus was found together with a stone mortar in house A viii 8, a house which seems to have had a bakery; this was probably used in large-scale bread making rather than for domestic consumption (below, chapter 6, ‘‘Cooking and Baking’’). A shallow stone basin was found in the kitchen (a) of house A 8, together with a pithos, many vases and other artifacts; this might also have been a kardopos. Other sorts of The Houses Organized

167

[To view this image, refer to the print version of this title.]

figure 36. Red-Figured Lekanis Showing Women Preparing Sesame Cakes in a Louter or Kardopos. St. Petersburg St. 1791

stone basins were found, but not in contexts which suggest bread making or food processing. Some objects identified as louters, however, may have been used for both washing and kneading (fig. 21). Vase paintings and terracottas show women kneading dough on pedestalled basins indistinguishable from louters (fig. 36).43 Louters are found associated with grinding assemblages in a number of houses at Olynthus. The well-preserved grinding assemblage found with a stone louter in the House of the Tiled Prothyron is described above. On the other side of this room was a large stone with a concave upper surface, which could also have been used as another kneading or working surface, and a number of vases were found nearby.44 Louters were also found with grinding equipment in the pastas of the House of the Twin Erotes, in the court of house B ii 3, in the exedra south of the court of house A viii 4 (room m, an area which might have served as a summer kitchen or food-processing area), in the pastas of A v 9, and in house A iv 9 (g). Most of these spaces are open or semienclosed: pastades, courtyards, and the open exedra would have been well-lit but sheltered workspaces. We should never underestimate the number of different purposes an object like a louter could serve. They could be used in religious ritual, as at the House of Many Colors or the Villa of the Bronzes, where they were found together with portable altars and other ritual equipment; they could be used for washing and are sometimes found together with storage amphoras but without evidence of cult instruments, as in the pastas of A iv 9; or they could be used as basins for working dough 168

The Houses Organized

and other food processing. The implements associated with bread making were no rougher or more massive than those associated with equipment for other purposes, such as washing or ritual. Probably the same utensil could serve all functions, louter, perirrhanterion, and kneading trough as needed. Food Storage A Greek household would have had a wide variety of containers for everyday storage of different kinds of foodstuffs. Pithoi, tubs or bins for grain and meal (probably the Greek σιπύη), lidded basins or λεκάναι, storage jars (including reused wine amphoras), and related vessels, as well as less well-documented vessels of wood, basketry, and wickerwork, were all common household items and are listed in the Attic Stelai in some quantities. Many such coarseware vessels are recorded in the notebooks from Olynthus, although few were mended or published. Such vessels were found in most houses at Olynthus, and where absent, it is probably due more to poor recording than to differences in household organization. Because so few were actually mended, it is impossible to quantify the assemblages of small-scale storage vessels in any one house or to compare the use of such vessels in different houses.45 Some houses have unusual numbers of storage amphoras (for instance house A 11, which contained fourteen amphoras, and A v 1 which contained fifteen), but in these quantities they may have been used for nondomestic use rather than for household storage. Many Greek households seem to have aimed to store a year’s supply of food— an undertaking which would require more extensive storage facilities than smaller storage vessels such as lekanides, bins, and the like could have provided. In practice, however, many households seem not to have had such facilities. The choice between storing enough foodstuffs for the household’s needs and relying on other sources, such as the market, is one of the most fundamental economic decisions which a household can make; and the distribution of these two strategies reveals much about the domestic economy of Olynthus. It will therefore be discussed further in the chapter on the Olynthian economy (chapter 6). Weaving Weaving was another basic task of ancient Greek women, learned at an early age. In Xenophon’s Economics, Ischomachos says that his wife was ‘‘not yet fifteen when she came to me, and had spent her previous years under careful supervision so that she might see and hear and speak as little as possible. Don’t you think it was adequate if she came to me knowing only how to take wool and produce a cloak, and had seen how spinning tasks are allocated to the slaves?’’ 46 Most households produced their own cloth for garments, bedding, and the like, and weaving equipment was found in most Olynthian houses. The Houses Organized

169

[To view this image, refer to the print version of this title.]

figure 37. Red-Figured Skyphos Showing Penelope Weaving. Chiusi 1831; Furtwängler and Reichhold, Griechische Vasenmalerei pl. 142 Unlike grinding grain and processing other foods, which used mostly portable equipment for relatively short periods, most weaving required a loom which could not be moved easily. The Greeks wove cloth on vertical looms, the warp threads weighted with terracotta loomweights (fig. 37). When not in use, the loom could be dismantled and stored out of the way, but once set up, it could not be shifted to another location until the textile was completed. Unlike grinding grain, which could be moved to different rooms depending on the weather or other factors, a loom had to be placed carefully with an eye to available light, shelter, comfort, and security. Literary sources offer only slight help in understanding where in the house women wove. Penelope wove in an upper chamber (ὑπερῷον) of the house (Hom. Od. 1.361, 15.517), and Andromache was weaving in the innermost room ( μυχός) of the house when Hector is killed (Hom. Il. 22.440). A passage of Menander’s Girl from Samos mentions a ‘‘loom-room’’ (ἱστεών): And I myself, too, helping, giving this and that, Into the storeroom, as it chanced, had gone, from whence I did not come directly, busy laying out More food than common and inspecting many things 170

The Houses Organized

Within. Just then, while I was there, a woman came, Descending from an upper storey, from above, Into the store-room’s antechamber.—For with us, There’s an apartment, as it happens, for the looms, So placed that through it is the entrance to the stairs And to the store-room. (Menander, Samia 13–21) In this play, weaving was done in an open room on the ground floor, through which there was access to both the upper story and to the storeroom. Weaving was done both by the free women of the household and female slaves alike, often side by side. ‘‘And fifty slave-women he had in the house, of whom some grind the yellow grain on the millstone, and others weave webs, or, as they sit, twirl the yarn, like unto the leaves of a tall poplar tree; and from the closelywoven linen the soft olive oil drips down’’ (Hom. Od. 7.103–6). Evidence for weaving at Olynthus, as from most archaeological sites, comes almost entirely from loomweights and very occasionally other weaving and spinning paraphernalia. No remains of the framework of looms and only a few possible weaving tools were found, as these were usually made of wood. Loomweights were, however, very abundant. Robinson and Graham concluded that ‘‘loom-weights have been found in nearly every room of every house excavated so that the only reliable evidence for determining where the loom originally stood is the discovery of large numbers in one room . . . we can reach no more definite conclusion than that any large plain room might be used to accommodate a loom.’’ 47 But with closer investigation we can build a more detailed picture than this. In one house, a loom was definitely set up on the ground floor when the city was destroyed. Forty-three loomweights were found in a line 1.1 m long in the southeastern room of Villa CC, a house otherwise very poorly preserved (fig. 38). The notes do not record whether the line was parallel or perpendicular to the wall of the room, although this would help reconstruct how the loom was set up in the room, whether it was leaning against the wall or was freestanding. But it attests to weaving on the ground floor of the house and gives the width of the loom and the minimum number of loomweights which equipped it. A bronze hooked instrument, perhaps a weaving tool, was found in this room as well. The field records of loomweights in other houses are much less detailed than this. Groups of loomweights are sometimes described as piled against a wall or at one side of a room, where they probably fell from a loom leaning against the wall. Some fieldbooks record the 1 × 1 m square in which the loomweights were found, giving a rough distribution. In other cases, we have only the total count of loomweights found in the room—numbers which may be reduced by incomplete collection of these mundane objects. Without more precise records of how they were found, it is difficult to determine whether they had fallen a short distance The Houses Organized

171

Loomweights definitely in situ (1 example)

Group of Loomweights

Select Houses with Weaving Areas

Illuminated Space (Courtyard, Flue or Light Well)

0

20 m

In pastas or court (2 probable, 8 possible examples) a

a

b

c

d

e

a

f

b

c

e

d f

g

h g

courtyard?

h

l b

c

d

i

j

j i

k

k

m

House B vi 5 19 loomweights against south wall of pastas, near foundation

Villa CC 43 loomweights in a single row

House A vii 4 16 loomweights scattered in courtyard; 23 clustered in room b

In rooms adjoining courtyards (10 probable, 2 possible examples) a

c

d

a

g

e

b

c

d

a

b

c

b e

f

f

d

g h

j

i

e

f

k

h g

h

i i

m

j

House A v 9 groups of 18-24 loomweights in 4 rooms around court

k m

l

House A viii 1 32 loomweights against east wall of “exedra” k

House A v 5 18 loomweights in “exedra” h

In rooms adjoining light wells / flues (5 probable, 1 possible examples; cf. house A vii 4, above) a

b

c

e g

d

a

f

e

h i

k l

j

g

b

c

f h

a

b

c

e

h k

m

f

i

g

j

l

j i

House of Many Colors 41 loomweights, epinetron & spindle whorl in room a; 37 in light well b

d

m

House A iv 9 30 loomweights in room adjoining light well b

figure 38. Selected Houses with Weaving Areas

House A xi 10 57 loomweights in kitchen adjoining flue

d

from a loom on the ground floor, like those in Villa CC, or had fallen from a loom set up in a second-story room, or were stored in a cache for use somewhere else, or were simply random losses. More careful attention to just how the loomweights had fallen, to the remains of burned wood and other perishable materials, might add considerably to our understanding of Greek weaving. But these assemblages nonetheless offer a great deal of evidence for the organization of weaving in Greek houses. The number of loomweights found together is one indicator of whether they represent the debris of a loom or trash. A very small number of weights was of no use at all: it takes a substantial number to equip a loom. The loom in Villa CC had 43 weights, but looms could be equipped with smaller numbers of weights. A group of 21 weights was found in a line 1.59 m long, clearly in situ on the floor of a ninth-century building at Gordion in Phrygia. (The same and neighboring buildings contained hoards of up to 500 loomweights, spindle whorls, and textiles in various stages of manufacture.) The four groups of loomweights from different rooms of house A v 9 each consisted of about 20 loomweights, suggesting that these represent four individual looms (above, chapter 3, ‘‘House A v 9’’). Extant Norwegian warp-weighted looms could have as few as 13 weights or as many as 59, but most have between 20 and 40 loomweights (although these were heavier than the Greek examples). Barber estimates that as few as 6 loomweights might have formed a set in some cases.48 To try to determine the minimum number of loomweights which suggest actual weaving rather than random losses for Olynthian looms, we can plot the number of loomweights found together in a group against how commonly that quantity was found (fig. 39). We would expect a relatively small number of rooms to have been used for weaving, while many rooms contained trash and debris. Through this sort of analysis, and considering the architectural setting and what other artifacts were found together with the loomweights, we may begin to distinguish use groups from trash.49 The plot shows that eighty-four rooms had 1 loomweight in them; forty-nine rooms had 2 loomweights; only thirty-eight had 3 loomweights, and so forth until only one room contained 74 loomweights. Because a solitary loomweight or just a pair was of no use, very small groups of loomweights must be interpreted as trash or debris (although perhaps from a loom set up here previously). Larger clusters probably either fell from looms or were stored in the room as the equipment belonging to one or more looms. Twenty-five rooms had between 10 and 25 loomweights in them, and seven rooms had between 25 and 43 loomweights, including the cluster from Villa CC which had clearly fallen from a single loom. All these probably represent the remains—in many cases probably incomplete—of single looms. Only five rooms had more than 43 loomweights in them. These included A iv 9, room g, which had 71 loomweights in two distinct clusters, probably representing the detritus or equipment of two looms (above, chapter 3, ‘‘House The Houses Organized

173

100

Number of rooms

80

60

40 more than 1 group? looms in Av9

20

0

0

10

loom in Villa CC

20 30 40 50 Number of loomweights

trial trench

60

70

80

figure 39. Clusters of Loomweights A iv 9’’). The House of the Comedian, room k, had 29 loomweights scattered along the north wall, perhaps a single loom’s worth, and 38 elsewhere in the room, perhaps belonging to another loom. This room, however, was small and open to the street and is unlikely to have been used for weaving. The groups of loomweights therefore do not necessarily belong to single looms. The kitchen (c) of house A xi 10 contained a tight cluster of 57 loomweights, together with a variety of other domestic artifacts. These were found so tightly clustered, however, that it seems unlikely that they had fallen from a loom; rather, they were probably being stored in a bag here. The largest single group, of 74 loomweights, was found in a trial trench west of house ESH 4, together with a netting needle, four red-figured lekythoi and other vases, a fine female protome, and other objects. As the context was not very clear, it is impossible to say whether they belong to a single set or more than one. Finally, house A viii 7/9 contained two large hoards, one of 50, and one of 247 loomweights. In these quantities, it is likely that they were used for household industry rather than purely domestic weaving; they are therefore considered separately in chapter 6. Given the great range of preservation of different houses, one would not expect to find totally clear patterns in the numbers of loomweights from different rooms. But the clusters of loomweights found at Olynthus agree very well with known groups which clearly fell from a single loom, such as those from Villa CC or Gordion, and with the ethnographic parallels in Norway and elsewhere. Most 174

The Houses Organized

clusters fall in the range of 10 to 40 loomweights, just what others have suggested for an ancient warp-weighted loom. It does not follow, of course, that all these were set up in looms; some may have been complete sets, stored away for use in another space. But we can still get a general impression of what sorts of spaces were used for weaving, where they were located in the house, how they were lit and what else was going on in these spaces. A key factor in choosing a room for weaving would have been access to light. Weaving requires a sheltered but welllit space, and this must have strongly affected the location where women chose to set up their looms. About forty-three rooms, in thirty-five houses, seem to have been used for weaving at Olynthus, depending on how many loomweights we believe constitute credible evidence (plates 1, 2; fig. 38). Most room types are represented in table 3. The only types of rooms which were never used for weaving were androns (which, not surprisingly, never contained more than 3 loomweights), flues and bathrooms, which were too small for a loom, and entrance areas.50 Although Greek has a word for a ‘‘loom-room,’’ this seems to be applicable to any space where weaving was done rather than designating a specific type of room. The pastas would have been bright and comfortable, and collections of loomweights were found in the pastades of a number of houses, including A 3, A 8, A vi 2, B vi 5, and B vii 2. But rather few of these turn out to be completely convincing. The pastas of B vi 5 is the most certain (fig. 38). It contained 19 loomweights, at least 8 of which were found scattered within an area about 2 × 2 m in the more enclosed east end of the pastas. The loom may have been leaned against the south wall here, set on the foundation 0.75 m in front of the wall. The pastas also contained two grindstones and a saddle quern, and a number of vases, most of which were not mended and may therefore have been coarsewares; this seems to have been a space used for a variety of work. The pastas of house A vi 2 contained 8 loomweights in the eastern end, 2 in the west and southwest, and 2 in undetermined locations; this seems a doubtful case, as the numbers are small and distribution scattered. Other pastades had smaller even scatters of loomweights, less than a dozen. These scatters may be the incompletely preserved remains of looms used in the pastas, but they might also be losses or trash. The hoard of 247 loomweights found in the pastas of house A viii 7 was probably being stored for use here or elsewhere in the house; they do not indicate where the looms would have been set up. Courtyards could also be used for weaving, although there would have been a risk of inclement weather. Clusters of loomweights were found in the courts of A vii 1, A vii 4, A vii 10, A viii 10, and B vi 9. As with the pastas, though, these groups were relatively small and their interpretation ambiguous. The largest group, from house A vii 4, consisted of 16 loomweights, apparently scattered around the courtyard rather than concentrated in one area (fig. 22). Another room (b) of this house, however, contained 23 loomweights, at least half of which The Houses Organized

175

table 3: Rooms Used for Weaving House A −1 A 10 A 11 A 12 A3 A8 A iv 9 A iv 9 Av2 Av2 Av4 Av5 Av6 Av9 Av9 Av9 Av9 A vi 2 A vi 4 A vii 1 A vii 3 A vii 4 A vii 4 A vii 6 A vii 9 A vii 10 A viii 1 A viii 7 A viii 7 A viii 10 A viii 10 A xi 10 B vi 5 B vi 9 B vii 2 Dv6 ESH 4 H. Comedian H. Many Colors H. Wash Basin S. Villa V. Bronzes Villa CC

Room i b d a e d a g a g d h b e h j m c e h b b i j c h k b d a g c h m g b h k a f a i d

Room Type room N. Room room room pastas pastas N. Room S. Room N. Room room pastas room exedra N. Room room room room exedra pastas kitchen court N. Room N. Room court kitchen room court room N. Room pastas room court kitchen pastas court pastas room room shop N. Room room N. Room kitchen room

Number of Loomweights 7 24 25 9 8 7 31 71 15 14 18 18 9 24 22 18 19 12 5 12 15 23 16 7 8 7 32 50 247 many 7 57 19 8 9 14 39 67 41 13 7 6 43

Source of Light adjoins court uncertain adjoins court NW corner of house; dim? court/pastas court/pastas light well b adjoins court uncertain adjoins court adjoins pastas, but probably dim adjoins court N wall broken—open to alley? adjoins court adjoins court adjoins court adjoins court court/pastas flue court no pastas, adjoins court? flue adjacent court flue N. Room, problem; but only 8 court exedra or kitchen? open to court cache—weaving done elsewhere? court/pastas; large cache uncertain court flue court/pastas court court/pastas uncertain flue at west space probably not used for weaving light well uncertain uncertain: window to alley? flue uncertain

were clustered within a small space. It seems likely that this room was used for weaving and that the loomweights from the courtyard were scatter or detritus. The courtyard of the nearby house A vii 1 contained 12 loomweights and a spindle whorl; these may well be the remains of a loom, although their disposition was not noted. The other courtyards contained 7 or 8 loomweights at most, evidence which must be considered ambiguous. But even allowing the most generous interpretation, assuming that relatively small numbers of loomweights in courtyards and pastades are the incomplete remains of looms rather than losses or trash, these open spaces of the house only account for about one-quarter of the rooms used for weaving. More frequently, women set up their looms in enclosed or semienclosed spaces adjoining the courtyard or other source of light rather than in the court or pastas itself. The most dramatic example of this pattern was house A v 9, where all four of the rooms adjoining the courtyard contained clusters of 18 to 24 loomweights and were very probably used for weaving (fig. 25). One room was the pastas and was lit from the small courtyard; the others probably had windows facing into the court and so would have been relatively bright. Other households used similar enclosed spaces for weaving, including houses A 11, A iv 9, A v 2 (room g), A v 5, A vii 3, A vii 6, A viii 1, A viii 10, and possibly A −1, the House of the Wash Basin, and others. In house A v 5, a loom was set up in an open ‘‘exedra’’ at the southwest corner of the courtyard. Eighteen loomweights were found in this small space, which would have been particularly open and well lit. The room in the southeast corner of house A viii 1 (k) was also mostly open to the courtyard. Thirty-two loomweights were found at the foot of the east wall, probably fallen from a loom leaning against the wall, well lit by the wide opening to the courtyard. The courtyard was the main source of light in the inward-looking Greek house, and weaving rooms located adjacent to the courtyard were probably illuminated by windows, perhaps relatively large ones, opening onto the court. Other sources of light were available in some houses, however. Some weaving spaces were lit by a light well or flue rather than from the courtyard. Room a of the House of Many Colors, for instance, was lit by a light well at its east end, and contained 41 loomweights, a spindle whorl, and an epinetron for carding wool—a particularly complete assemblage of weaving equipment (above, chapter 3, ‘‘The House of Many Colors’’). The light well (room b) also contained 37 loomweights, mostly found in the southeast corner. This suite may well have housed two looms, together with other household equipment. The ‘‘second kitchen’’ (room h) of house ESH 4 contained a group of 39 loomweights, found near the wall of the bathroom (room i). Like a kitchen-complex, this suite might have been lit by a light well/flue in its northwest corner. The kitchen of house A xi 10 contained 57 loomweights, the only kitchen which contained a significant collection of loomweights (fig. 50). These were discovered The Houses Organized

177

in a dense cluster near the center of the north wall, together with two figurines— one of the loomweights was found inside a figurine. The tight packing described in the fieldbook sounds like this might have been a group of loomweights stored for use here or elsewhere in the house, although this is uncertain. The kitchens of A vi 4, A vii 6, and the Villa of the Bronzes contained small scatters of loomweights—fewer than 10 in each room. Although looms were apparently not set up in these rooms when the city was captured, the loomweights found in them may have been lost from looms set up previously, perhaps in the winter as suggested above. In two houses, A iv 9 (a) and A vii 4 (b), clusters of loomweights were found in rooms adjacent to a flue or light well, but not open to it through a pillar-partition (figs. 22 and 24). Neither of these rooms had another obvious source of light. They are far from the courtyard (which in A iv 9 was quite small in any case), and if the court had been the only source of light, they would have been quite dim for weaving. They may instead have been lit through windows opening into the adjacent light well. The light well thus illuminated two rooms rather than just the one. Although the courtyard and pastas were the best-lit spaces in the house, they were by no means the most commonly used rooms for weaving. There could be many reasons why women would have preferred to set up their looms in smaller, more enclosed rooms. It would have been easier to keep unfinished work safe from children or other interference. These rooms might have been cooler in the hot summer, when the city was destroyed, and more easily warmed in the winter. They were more predictable, remaining usable no matter what the weather, whereas the court or pastas might turn uncomfortable or inconvenient as the weather changed. The use of these rooms for weaving does not, however, seem to result from a desire to restrict this activity to a more private or secluded part of the house. Indeed, weaving areas were sometimes conspicuously close to the entrance of the house, hardly more removed than the court or pastas. It is curious, however, that in these cases, the women of the household did not set up their looms in the adjacent, better-lit spaces, such as the kitchen of A vii 4, which was open to the flue through the entire pillar-partition. Although Greek households generally wove their own cloth, not every house preserved evidence of weaving. Very few houses in block A vi had more than a few loomweights, for instance. The only house on this block with a significant group was A vi 2, which had 12 loomweights in its pastas. Next door, house A vi 4 had 5 loomweights and a spindle whorl in its kitchen (e), perhaps debris left from weaving in this room at some time before the destruction. Overall, though, the average for the block was only 6 loomweights per house; and most of these were in houses A vi 2, 3, and 4. The other houses in this block had either a few loomweights scattered in various rooms, or none at all. House A 9, the Villa of the 178

The Houses Organized

Bronzes, the House of the Twin Erotes, and a number of other houses had too few loomweights to equip even a single loom, although these houses were relatively well preserved, carefully excavated, and contained many artifacts of other sorts. This is probably not solely due to faulty recording since some loomweights were recorded, and if there had been large deposits they would have been noted. These rather luxurious houses presumably belonged to more wealthy families, who may have relied more than average on commercially woven cloth; or the household situation may have been more complex, for instance with part of the household living elsewhere but supplying cloth to the residents of these houses. Three or four houses seem to have woven textiles on a large scale, including A v 9, which had four looms set up in rooms around the court, the double house A viii 7/9, and perhaps A iv 9 and A vii 9. These are discussed under ‘‘Textile Manufacture’’ in chapter 6. A variety of different shapes of loomweights were used at Olynthus, but two sorts predominate in the destruction debris: pyramidal weights (Type 7) and conical weights (Type 8).51 We would expect weights from a single loom to form a matched set, but interestingly, this is generally not the case. The loomweights found in a line in Villa CC and so definitely used together consisted of 19 conical weights, 8 pyramidal, and 16 rectangular loomweights. These loomweights were not weighed; they may have been fairly closely matched in weight, if not in shape.52 Other sets of loomweights were also unmatched: of the 57 loomweights in A xi 10, 45 were conical, 11 pyramidal, and 1 was a truncated pyramid. The group of 39 loomweights from ESH 4 included 29 pyramidal weights, 9 conical, and 1 squat round weight. The weights from house A iv 9 (a), A viii 1 (k), and most other houses were also a mixture of conical and pyramidal types.53 The mismatching in shape, of course, did not matter as much as differences in weight. Even modern warp-weighted looms can have loomweights of different sizes. Variations in weight could be compensated for by attaching more warp threads to the heavier weights, for instance; and the exact tension of the warp threads may not have been considered critical. In a few houses the weights were more uniform, though. The loomweights from the two looms in the House of Many Colors, for instance, were mostly conical, with a very few pyramidal loomweights, as were the set of 24 weights from house A 10 (b). The sets of loomweights from houses which seem to have manufactured textiles on a large scale, A v 9 and A viii 7/9, also formed more closely matched sets. Virtually all the loomweights from house A v 9 were conical (only 5 out of the 85 in the house were pyramidal or squat), while those from the enormous hoard in A viii 7/9 contained both pyramidal and conical types, but of very uniform and unusually light weight. These professionals apparently did not make do with mismatched sets of loomweights, but for normal household weaving, it apparently did not make as much difference whether warp-threads had slightly different tensions. The Houses Organized

179

the a n d r on The women’s quarters are contrasted in Greek written sources to the men’s area, the andronitis or andron. Although this word could be used in a very general sense of men’s living quarters (as it is used in Lysias’s On the Murder of Eratosthenes, quoted above, or in Xenophon’s Economics), it often has a more specific meaning as the formal dining room for that quintessentially Greek institution, the symposium, and this is how archaeologists generally refer to it. The symposium was an occasion when friends, acquaintances, and even less closely related outsiders could enter the house in an intimate setting. It could be an occasion for the host to display wealth and taste in his choice of guests, food, wine, conversation, music, entertainment, and furnishings. And it was to some extent a public event, not only for the participants but also for onlookers (and listeners) in the streets; the rowdiness of these parties was legendary. Most of the distinctive architectural features of the andron result from this special use for formal dining parties. The rooms had cement or mosaic floors, with a raised border 0.85 to 1.0 m wide around the edge of the room to support dining couches for symposiasts to recline on (fig. 40). Most rooms were square, about 4.75 m on a side—a standard size holding seven couches, as specified, for instance, in a sales inscription from the site.54 Larger and smaller rooms are found at the site as well—one house, A vi 5, had two androns, one holding three and one holding eleven couches, the largest and smallest number of couches found at the site. The door to the andron was often set off center to accommodate the odd number of couches that characterized these rooms. These were without exception the most highly decorated rooms in the house. The walls were almost always painted, and frequently the cement floors and platforms were painted as well. Nine of the thirty-four androns at the site had mosaic floors, some of them figural. The rooms were often equipped with drains leading out to the street, so that the room could be sluiced out after a particularly messy and raucous party. The andron was sometimes entered through an anteroom. This isolated the room somewhat from the rest of the house, and it probably also provided a convenient area to prepare dinner and store dining equipment. Symposia demanded special furnishings and vessels: kraters for mixing wine and water; cups, jugs, ladles, and other vessels for pouring and drinking; eating utensils; lamps and lamp stands; and the like. Much of the best black- and redfigured pottery of sixth and fifth centuries was made for these occasions and was frequently decorated with sympotic scenes. Households with androns must have had special sets of sympotic vessels, as Ischomachos describes in Xenophon’s dialogue quoted above (Xen. Ec. 9.2). Such sets have been identified in Athens and elsewhere.55 Unlike plain cooking and household pots which were not collected or noted systematically in the excavations, fancy symposium pottery was carefully mended 180

The Houses Organized

[To view this image, refer to the print version of this title.]

figure 40. Attic Red-Figured Cup Depicting a Symposium. Berlin F 2298

and published, and we have a fairly complete record of what survived. We might expect to find some of the finer vases and other artifacts from the site associated with androns. To judge from drawings on Greek vases and other sources, the most characteristic vessel in the symposium was the krater, used for mixing wine with water (fig. 41). Some forty-four pottery kraters were found at the site (see table 4). A few of these are early and were probably out of use by the time of destruction: these include at least two black-figured kraters from early levels on the South Hill and a fifth-century red-figured column krater found in a grave.56 Single sherds or small portions of some twenty-one more were found, whose very fragmentary state suggests that they had been broken and discarded before the destruction. Nearly all the kraters from the site were red-figured; only a single black-glazed bell krater was published, although the fieldbooks mention a few other vessels initially identified as kraters, which were apparently not mended and published. Some of these may have been unfigured because figured vases would probably have been saved and mended even if they were very fragmentary.57 The total is still a relatively small number, considering the very large area excavated. A single deposit of dining debris from the Athenian Agora, for instance, mostly dating to the mid-fifth century b.c., contained eighty-seven figured kraters—more than half of the figured vases in the deposit.58 A number of these kraters were found in assemblages together with other redfigured and glazed tablewares and probably belonged with dining sets. For inThe Houses Organized

181

[To view this image, refer to the print version of this title.]

figure 41. Red-Figured Bell-Krater from the South Hill at Olynthus (Section N). Olynthus 13, no. 29, pl. 38 stance, the red-figured bell krater from the kitchen (a) of house A 8 was found together with a red-figured pelike, a bolsal, a bowl, a black-glazed oinochoe and a black-glazed jug, two gutti, four saucers, and a saltcellar, as well as coarsewares, storage amphoras, and a pithos. The collection of tablewares stored in the pastas (e) of the House of Many Colors included a red-figured bell krater and a redfigured pelike, four black-glazed plates, a black-glazed skyphos and other vessels (above, chapter 3, ‘‘House A 8’’ and ‘‘House of Many Colors’’). A small alcove or ‘‘cupboard’’ (m) on the south side of the courtyard of house A iv 5 contained a large number of pots, terracottas, and other artifacts, including pieces of a redfigured bell krater and a red-figured lebes gamikos, two olpai, lekythoi, an askos, a juglet, storage amphoras, coarse vessels, and two thymiateria. However, there are a number of peculiarities about these assemblages and the distribution of kraters. The scarcity of drinking cups is striking. Only nine redfigured cups were found at the site, and three of these came from graves and another was found in a street.59 The assemblages which contained kraters gener182

The Houses Organized

table 4: Kraters from Olynthus Room Type Where Found

Andron in House? Shape

A −1

uncertain

no

krater

A −i 9 d

room

?

krater

A8a

kitchen

no

bell krater

A 13

uncertain

no

krater

A iv 5 m

cupboard

no

bell krater

Av2g

room

no

krater

Av3h

courtyard

no

A vi 3 b

andron

yes

bell krater bell krater

A vii 7 A viii 8 b A viii 8 i

uncertain no North Room no kitchen? no

A viii 10 g

courtyard

no

B vi 7

uncertain

yes

bell krater

B vi 7 c1

North Room yes

bell krater

Findspot

Description, References

North Hill

krater krater bell krater bell krater

‘‘Red-figured,’’ unpublished; only mentioned in fieldbook notes RF, frags. Symposium; youths. Group G (ARV 2, 1469.152). Olynthus 13, no. 30 RF, A: Nike driving chariot; B: 3 youths. York Reverse-Group (ARV 2, 1450.2). Olynthus 5, no. 114 RF, frags. Youths. Painter of the Louvre Centauromachy (ARV 2, 1092.72). Olynthus 5, no. 161 RF, repaired. Dionysiac scene, with maenads and satyrs; 3 youths. ‘‘Symposium Painter.’’ Olynthus 13, no. 36 RF, sherds of rim. Maenads and satyrs. Olynthus 13, no. 32 BG, Olynthus 13, no. 250 RF, goats; athlete and youth at altar. Painter of Olynthos 5.156 (ARV 2, 1507.7). Olynthus 5, no. 162 RF, sherds. Olynthus 13, no. 31 RF, frags. of base and belly RF, symposium. Olynthus 13, no. 39 RF, Dionysos, Eros, maenads, satyr; 3 youths. Black-Thyrsus Painter (ARV 2, 1431.4). Olynthus 13, no. 38 RF, Dionysos, maenads, and satyr; 3 youths. Telos Group and RT Painter (ARV 2, 1425, 1427). Olynthus 13, no. 35 RF, Grypomachy; 3 youths. Group G (ARV 2, 1468.141). Olynthus 13, no. 37

table 4 (continued) Findspot

Room Type Where Found

Andron in House? Shape

B vi 8 g

shop

no

krater

B vi 8 g

shop

no

bell krater

Dv2k ESH 1

uncertain uncertain

? yes

krater krater

Tr. 13 (ESH 1–3)

uncertain

?

H. Comedian d

flue

yes

nr. H. Comedian

uncertain

?

kalyx krater kalyx krater krater

St. −iii near H. Twin Erotes St. iv E of Ave. B

street

?

krater

street

no

bell krater

St. v and C v 4 St. v and C v 4

street street

no no

krater krater

H. Many Colors e

anteroom

yes

bell krater

H. Many Colors f

pastas

yes

krater

Sec. G Area 18

room

no

krater

Sec. G Area 19

room

no

bell krater

Sec. G Area 30

room

no

bell krater

Description, References Unpublished; ancient repair. From the same room as, but not Olynthus 13, no. 40 RF, satyrs and maenads; 3 youths. Painter of Rodin 966 (ARV 2, 1449.2). Olynthus 13, no. 40 sherds RF, large. Poseidon and Nereids; heroes? to R. Olynthus 5, no. 131 RF, repaired. Olynthus 5, no. 129 RF, quadriga and Nike. Olynthus 5, no. 112 RF, sherds. Painter of Olynthos 5.156 (ARV 2, 1508.13). Olynthus 13, no. 201D RF, base only, repaired. Standing youths. Olynthus 13, no. 34 RF, Herakles and 3 youths; 3 youths. Near Painter of London F 64 and F 1 (ARV 2, 1421). Olynthus 13, no. 27 RF, unpublished RF, scattered sherds, unpublished

Villa Section RF, symposium; 3 youths. Painter of Munich 2335 (ARV 2, 1164.56). Olynthus 13, no. 41 frags. in f and j, unpublished

South Hill RF, frags. Dionysos, woman, tripod. Olynthus 5, no. 281 RF, symposium; 3 youths. Group G (ARV 2, 1469.149). Olynthus 5, no. 137 RF, symposium; 2 youths. Group G (ARV 2, 1469.153). Olynthus 5, no. 133

table 4 (continued) Findspot

Room Type Where Found

Andron in House? Shape

Sec. G road

street

no

krater

Sec. G road

street

no

krater

H. of Pan c

room

no

krater

S. Hill road Sec. J/K 16e

street room

no no

krater kalyx krater

Sec. J/K 5n Sec. N

room public area

no no

krater bell krater

Sec. N

public area

no

bell krater

Tr. 10

public area

no

bell krater

Tr. 3

uncertain

no

Tr. 6

uncertain

no

bell krater krater?

Description, References RF, frags. A: Poseidon and Amphitrite, Dioskouroi, Athena; B: wedding procession? Olynthus 5, no. 130 RF, repaired. Satyr; maenad? Olynthus 5, no. 306 RF, frags. 2 Erotes; standing woman; seated man. Olynthus 5, no. 290 RF, fragment. Olynthus 5, no. 280 RF, youth pursuing woman; Nike and warrior. Deepdene Painter (ARV 2, 500.27). Olynthus 13, no. 26 RF, laurel leaves, frags. RF, Dionysos, maenad, and satyr; 3 youths. Group G (ARV 2, 1468.138). Olynthus 13, no. 28 RF, symposium; 2 youths. Group G (ARV 2, 1469.154). Olynthus 13, no. 29 RF, A: Nike driving chariot; B: 3 youths. York Reverse-Group (ARV 2, 1450.3). Olynthus 5, no. 113 RF, repaired. Olynthus 5, no. 142 BF, Olynthus 5, no. 23

Early Apotheke G8

(early fill)

Apotheke G8

(early fill)

Grave 279

grave

col. krater krater

col. krater

BF, sherds. Quadriga; hoplite and men. Olynthus 5, no. 24 local; painted in 3 registers with ivy leaves, etc. Olynthus 5, no. P69 RF, A: woman offering phiale to youth. Florence Painter (ARV 2, 542.17). Olynthus 5, no. 105A

Note: RF = red-figured; BG = black-glazed; BF = black-figured.

ally did not contain significant numbers of drinking cups. The fifth-century sympotic debris from the Athenian Agora contained roughly ten times as many drinking cups as it did kraters; the Olynthian assemblages often contain one or two cups, or none at all. A number of red-figured kraters were together with women’s artifacts, rather than with assemblages which seem related to men’s dining. Two were found in rooms used for weaving, for instance, in houses A −i 9 and A v 2, and one in the flue (d) of the House of the Comedian. None of these rooms contained pottery which looked like vessels for the symposium; the kraters seem anomalous here. Most of all, the spatial distribution of kraters is remarkable (plates 1, 2). One would expect that these wine-mixing vessels would be found in sympotic contexts in androns or in houses with androns. But most of the houses in which redfigured kraters were found did not have an andron, and most houses with androns had no kraters. A surprising number of kraters were found in the older part of Olynthus, the South Hill, where houses were smaller, less planned, and possibly less wealthy. Fourteen red-figured kraters came from this part of the site, while twenty-seven were found on the North Hill and Villa Sections.60 Since the area excavated on the South Hill was only one-quarter as large as the excavated area on the North Hill and Villa Sections, and only one andron was discovered there (with no kraters associated), the distribution is unexpectedly heavily weighted towards the South Hill. It is possible that houses which lacked proper androns nevertheless held symposia in other rooms, and that the kraters were used in these somewhat less formal symposia. However, the Greeks drank wine at normal meals, not only at symposia; and these kraters found in houses without androns might just as well have been used to mix wine for everyday eating. Although we think of ceramic kraters as sympotic vessels, they undoubtedly could be used in other contexts as well. And what of houses with androns? Of the thirty-one houses with androns excavated, only five had kraters in them, accounting for seven of the kraters from the site. And of these seven, only two were found in or near the andron. With very few exceptions, the androns at Olynthus were all but empty when excavated. Fragments (unmended, and perhaps therefore black-glazed rather than red-figured) of a krater were found in the anteroom to the andron of the House of Many Colors, together with two plates, a fishplate and a lid, bronze bosses from a door or piece of furniture, and a few other artifacts. This is the closest to a sympotic assemblage found in an andron, and it does not resemble assemblages from, for instance, the public dining debris from the Athenian Agora in quality, quantity, or variety. Fragments of a rather poor quality red-figured bell krater were found on the Bellerophon mosaic in the andron of A vi 3, the only krater actually found in an andron. A few other androns had artifacts associated with them. The andron of house B vi 7 was the smallest on the site, measuring only 2.65 × 3 m,

186

The Houses Organized

but a marble statuette of Asklepios was found in its doorway, one of only two marble sculptures from the site, as well as a fragment of a red-figured pyxis lid and fragments of large unslipped vases.61 By far the largest assemblage of artifacts from an andron was that from the House of the Comedian, where more than twenty-five terracotta protomes and figurines were found fallen from the south wall of the andron (above, chapter 3, ‘‘The House of the Comedian’’). But no mixing vessels and only a single cup-kantharos for drinking were found among the vases in the room. Smaller numbers of terracotta figurines, protomes, and plastic vases were found in the androns of houses A 5 and A vi 6. But other than these few cases, androns throughout the site were remarkably bare. Other ceramic vessels used at symposia, such as red-figured cups, pouring vessels, and the like, followed similar distributions. They were generally not found in androns or in houses with androns but with purely domestic pottery assemblages in houses without androns. It is unlikely that painted ceramic kraters and other vessels were carried to safety by fleeing inhabitants or looted by Philip’s troops. Ceramics were fragile and relatively inexpensive, and the presence of other sorts of artifacts worth much more in Olynthian houses shows that objects of such minimal value were mostly abandoned. None of the accounts of ancient looting and booty mentions pottery.62 Although a great deal, perhaps the majority, of unfigured pottery was not mended or published in the excavations of the 1920s and 1930s, figured pottery was of course meticulously collected and published; so we have a good record of what survived. If ceramic kraters and cups were commonly used in symposia at Olynthus, therefore, they surely would have been found in larger quantities, either in the androns themselves or stored in other rooms of houses with androns. Although arguments from silence may be inherently problematic, in this case the silence is deafening. Like the dog which failed to bark in Sherlock Holmes’s Adventure of Silver Blaze, the absence of kraters, cups, and other sympotic furnishings in these houses is itself important evidence and demands explanation. The most likely explanation for the lack of symposium equipment in androns is that most of the kraters, cups, and other vessels used at Olynthian symposia were made of metal rather than pottery, and were carried off by fleeing Olynthians before the destruction or were looted by Philip’s troops. Metal was much more expensive than pottery, less fragile, and could be melted and reused. The relative importance of metal and pottery vessels has been a subject of considerable debate in recent years. Vickers, Gill, and others have argued that ‘‘the richest members of Athenian society regularly used plate at their symposia’’ at least as early as the period after the Persian Wars, and that archaeologists have grossly underestimated its importance because the Greeks did not bury such wealth with the dead but instead reused or melted it down, and so very little is preserved. They further argue that Athenian black- and red-figured vases, long prized possessions

The Houses Organized

187

of museums and intensely studied by connoisseurs, are cheap substitutes, even slavish copies of metal originals.63 Although other scholars have pointed out some of the flaws in Vickers and Gill’s arguments, particularly concerning the dependence of pottery on metal prototypes, the place of metal vessels in Greek household assemblages remains an open question. Literary and epigraphic evidence refers mostly to the use of metal among the very wealthiest class of Greek citizens, Macedonian royalty and nobility, Eastern potentates, and other foreigners who lived a life of luxury hardly imagined by most Greeks; to dedications in Greek sanctuaries which may have been made by these elite; and to later periods when silver and gold were in much more common use. Literary sources do not describe the practices of more ordinary citizens of the fifth and earlier fourth centuries b.c.; and Vickers and Gill focus their attention primarily on the wealthiest Athenians for the Classical period. In the mid-fourth century the situation of metal vessels in Greek daily life was apparently changing rapidly. One of the few explicit ancient accounts of the relation between pottery and metal vessels relates that ‘‘down to Macedonian times people at dinner were served from utensils of crockery’’ and another associates the common use of silver and gold tableware with the seizure of Delphi by the Phocians in the Sacred War of 355 b.c., when gold and silver dedications worth more than 10,000 talents (262,000 kg) of silver were melted down and released into circulation.64 ‘‘Even those who were reputed to be very rich used to drink from bronze cups.’’ 65 But gold and silver vessels and jewelry appear increasingly often in Macedonian tombs from the second half of the fourth century on, however, and Athenaeus’s accounts of banquets among the Macedonian nobility of this period are replete with extravagant sets of precious tablewares, often given away to the guests.66 The booty from Olynthus brought Philip and his army enormous wealth: ‘‘After the capture of Olynthus, he celebrated the Olympian festival to the gods in commemoration of his victory, and offered magnificent sacrifices; and he organized a great festive assembly at which he held splendid competitions and thereafter invited many of the visiting strangers to his banquets. In the course of the carousals he joined in numerous conversations, presenting to many guests drinking cups as he proposed the toasts, awarding gifts to a considerable number, and graciously making such handsome promises to them all that he won over a large number to crave friendship with him’’ (Diod. 16.55). None of the preserved literary sources help us determine how widespread the use of metal vessels was at any time during the Classical period. Even if plate of precious metal was in use among the wealthiest Athenians of the fifth century, or among the Persian and Macedonian royalty, we can hardly conclude that it was commonly used by the citizen body of a Greek city or that ceramics were merely a cheap imitation. The archaeological evidence from Olynthus must be considered independently, as it pertains to a population whose habits are not preserved in 188

The Houses Organized

[To view this image, refer to the print version of this title.]

figure 42. Lampstand from House ESH 6, in the Shape of a Comic Actor. Olynthus 10, no. 1, pl. 1 the literary tradition. And here, the absence of kraters and other drinking vessels in houses which must have hosted symposia must imply that they were looted or removed before the destruction. Since ceramic vessels were hardly worth looting or carrying to safety, we must conclude that they were made of metal; not necessarily silver or gold, of course, but at least made of bronze. Although they are not preserved in the archaeological record, the gaps left by their removal from contexts which must have had special drinking vessels allow us to infer their original existence. Relatively few complete metal vessels were found at Olynthus. House ESH 6, on the eastern slope of the North Hill, contained a group of interesting bronze vessels and objects, including a lamp stand (?) in the shape of a comic actor, a goblet or thymiaterion, a bowl on a wide bronze stand, and a dish (figs. 42, 43).67 Fragments of a vase, unpublished but described by the excavator as ‘‘red-figured,’’ were found in the room as well, together with a bowl, a cup-kantharos, three lekythoi, a juglet and a lamp, four terracottas, and other objects. This building, however, was only partly built, consisting of a series of rooms on the north side of an otherwise undeveloped house plot. The rooms were domestic in function: one room next door to the findspot of the bronze objects had a bathtub installed in its southeast corner; but the house has no andron and lacks the specialized rooms or features The Houses Organized

189

[To view this image, refer to the print version of this title.]

figure 43. Metal Vessels from House ESH 6. Olynthus 10, nos. 574, 575, pl. 44

found in other houses at Olynthus. Perhaps this house escaped complete looting by virtue of its very modesty. If so, however, it shows that even less wealthy households owned bronze vessels; wealthier houses must have owned many more, and probably vessels of more precious metals. Although few complete metal vessels were found at the site, fragments of at least twenty-one bronze cups, fourteen jugs or oinochoai, nine bowls and other fragments, plus swinging handles from other sorts of bronze vessels, attest to the fairly common use of bronze drinking implements. Bronze vessels were found somewhat more commonly in houses with androns than were ceramic kraters and cups; but because of the extensive looting, there is no close correlation. Only a single silver vessel was found at the site—a pyxis lid with a relief of Cybele in a chariot drawn by a lion, its exact provenience unknown.68 We must not forget that the looting of Olynthus was thorough, and that precious objects were nearly all removed, either by the inhabitants who escaped before Philip’s capture or by Philip’s soldiers. But we need to remember that many of the most valuable household possessions were not those which we prize today. Mundane implements like pithoi, grindstones, and the like were considerably more expensive than ceramic vases and were probably removed more frequently before or after the abandonment of the city. 190

The Houses Organized

c on c lu s i on s A close look at the architecture and contents of houses at Olynthus reveals paradoxes at a number of levels. Literary sources lead us to expect separate men’s and women’s quarters in Greek (or, properly, Athenian) houses, either on different floors or in different sections of the house, with restricted access between them. Most recent studies of Classical domestic architecture, however, have concentrated on the unity of the house and ease of communication between different parts of the dwelling and conclude, for instance, that ‘‘there is no clear evidence on the ground floor for the existence of a specific women’s area or gunaikon, such as is mentioned in the written sources.’’ 69 Rather than see the house in terms of an opposition between men’s and women’s quarters, Jameson, Nevett, and others find an asymmetrical arrangement of space, in which there is a special men’s area, the andron, reserved (at least on occasion) for the men and male guests of the household (and lower-status entertainers, both female and male), but no distinct women’s areas. Nevett concludes that ‘‘the term gunaikon may have referred to the areas to which they [male authors] were not admitted when visiting, and which would have been used by the women of the house.’’ 70 In general, there are no identifiable sectors of the houses at Olynthus to which access was obviously restricted. Instead, houses are mostly of fairly open plans, with virtually all rooms accessible from the court or pastas. Moreover, the seclusion of women is generally considered a phenomenon more of the upper classes, who could afford to own slaves for work outside the house; the poor could not afford the luxury of secluding productive members of the household. Yet it is the more finely built houses, with mosaics, painted rooms, cut stone, and the like, which tend to have more regular and open plans, while the less pretentious houses, such as A 11–A 13, have more rooms to which access was restricted. Although Jameson, Nevett, and others do not recognize a gynaikonitis in the archaeological record, many Olynthian houses did have specific rooms which seem to have been intended for women’s activities: the kitchen-complex. The evidence for cooking and food preparation in these rooms associates them with women, and the plain treatment of their walls and floors contrasts strongly with the mosaics and fancy painted walls of the andron. We probably cannot determine whether an ancient Olynthian would have called this suite a gynaikonitis or by some other name; but whatever the terminology, it seems to be a particular region of the house primarily associated with women. To what degree was access to these suites limited? There is no evidence at Olynthus for the locked doors which separated the gynaikonitis from the andronitis in Xenophon’s Economics or Aristophanes’ Thesmophoriazeusai. Like other rooms of the house, the kitchen and flue were entered from the pastas or the courtyard, without intermediate spaces or other architectural restrictions. The doorway may have been closed with a curtain, but there were no thresholds or obvious signs of The Houses Organized

191

fixed doors. The kitchen-complex, however, was usually located on the side of the house which did not face onto the street: in the southern half of houses on the northern row of a block, in the northern half of houses on the southern row, and on the western side of houses on Row A (plates 1, 2). Of about forty-three kitchencomplexes excavated, thirty-two were located on the side of the house farthest from the street, while only nine adjoined it. There are no obvious environmental or other factors to explain this placement; this arrangement must have been deliberate, to locate the suite far from the entrance to the house. Since prevailing cultural norms would have prevented such a room from having windows opening onto the street, households may have wished to devote space fronting onto the street for other types of rooms, such as androns, which did communicate more with the outside world. The intent seems to have been to locate these suites away from the entrance of the house and keep them relatively private. By contrast, androns were normally located adjacent to a street wall, and ideally on the corner of a block.71 Of the thirty-four androns uncovered at the site, twentyone adjoin a street wall of the house; but the twelve androns on the interior of the house are mostly later additions to the house (plates 1, 2). This placement, the opposite of the kitchen-complex, was probably intended to allow larger windows in the andron walls to open onto the street. Large windows were found in androns in houses A 10 and A vi 5, whose andron had a large window in its south wall, and also at Eretria, the Piraeus, Delos, and elsewhere, although many of these open to the interior of the house rather than to the exterior. Windows facing onto the street might have both illuminated the andron and opened the symposium within to passersby on the streets, who could have heard and perhaps even seen the party in action. Public display played an important role in the symposium: the architecture of the rooms; the fancy mosaics, furniture, and drinking vessels; the food, entertainment, and company proclaimed the wealth and taste of the host. The rest of the house was involved in this public display to some degree as well. Although many houses could have been arranged so that strangers attending a symposium would enter the andron directly from the street or the entrance to the house, in practice the door to the andron was instead placed where guests would have to traverse part of the house to reach the andron. Rather than secluding the rest of the house from guests, it was put on display, if only briefly. This casts doubt on the interpretation of the gynaikonitis as an asymmetrical space, ‘‘the areas to which they [male authors] were not admitted when visiting.’’ As we have seen, however, architecture does not tell the whole story. The distribution of artifacts in houses at Olynthus reveals very different and much more diverse patterns of household organization. Although kitchen-complexes seem to have been used for some activities like cooking and washing, they only rarely contained artifacts associated with other women’s activities, such as grindstones for processing grain or loomweights for weaving. Instead, even households with 192

The Houses Organized

specialized kitchen-complexes seem to have used the court, pastas, and rooms adjoining these open spaces for these household tasks, leaving the specialized suites little-used, at least when the city was destroyed. The lack of evidence for working in these special suites is unexpected and paradoxical: spaces intended for women’s use in a secluded part of the house do not show evidence for such use in practice. A number of possible explanations may be offered. Household space might have been organized very differently in different seasons, as suggested above. The distribution of artifacts reflects the organization of space during the hot summers, when the airier, more open courtyard, pastas, and adjacent rooms would have been more comfortable work areas. In winter, the more sheltered kitchens, warmed by hearths and cooking fires in the flue, would probably have held larger numbers of loomweights, grindstones, and other artifacts used by women in their daily tasks. But this organization of household space, avoiding the secluded kitchen-complex when possible and working in the more public areas of the house, may also reflect a deliberate choice on the part of the women of the household for social rather than purely practical reasons. Even allowing for a great deal of comic exaggeration, Aristophanes’ Thesmophoriazeusai suggests how seclusion in the women’s quarters was an unpopular institution. Houses were designed and built by men and reflect (at least to some extent) the cultural ideals and norms surrounding the organization of house and household—ideals reflected in prescriptive, normative works such as Xenophon’s Economics, in legal speeches of Lysias and others, and other such literary works. But we must distinguish carefully between those ideals and reality. Reconstructions of domestic life based solely on literary and architectural evidence, and neglecting the often mundane and confusing evidence of household artifacts, leads to oversimplified and misleading conclusions.

The Houses Organized

193

chapter five

The Organization of Blocks

e x p l a i n i n g ‘‘t y p e’’ a n d va r i e t y i n ol y n th i a n h o u s e s The greater number of houses at Olynthus share some or all of the organizing principles common to the ‘‘type house’’ and include some or all of the specialized rooms outlined in chapter 4. This general similarity among Olynthian houses has led most scholars to consider only the type house in their treatment of the subject, and to more or less ignore the variety of houses at Olynthus. Most houses at Olynthus can be ‘‘explained’’ with reference to the principles and components of the type; but such explanations consider only their similarity to other houses, not their actual forms. The extreme case of this argument has been made by Hoepfner and Schwandner in Haus und Stadt im klassischen Griechenland. They claim that as originally planned, all the houses at Olynthus were essentially identical, varying only in minor ways, for instance arranging the rooms differently according to the position of the house in the block. Their reconstruction of the city looks like a modern suburban tract-house development, with row upon row of identical, almost prefabricated dwellings (fig. 44). To explain the similarity among Olynthian houses, they propose that not only were the immigrants to Olynthus allotted house plots of identical size, but they were also required by the state to build their houses according to the standard plan and prohibited from building in their own native styles.1 One of their main arguments, then, is that the adoption of the ‘‘Typenhaus’’ is an act of public policy imposed from above. The type house itself becomes a symbol, a material expression of Greek cultural ideals of democracy and isonomia, equality before the law. All the major variations from the ideal type, and the houses which do not conform to the ideal, they explain as later modifications to original type houses. 194

Although such clear connections between cultural ideals and architectural form are seductive in their simplicity and apparent explanatory power, we must again be careful not to mistake ideals for actual practices. Isonomia among citizens may have been a powerful factor in Greek law and custom, but that did not make Greek society completely egalitarian or Greek houses all exactly alike. Variation and ‘‘messiness’’ are inevitable—and revealing—aspects of human existence. Greek cities had to balance such social ideals as isonomia with social realities of inequality. The process by which they reconciled these conflicting demands reveals a great deal about Greek civic and social organization. To ignore the individual and the range of variation and look only at norms and type houses is to discard much of the richness, flavor, and fascination of Greek life. Rather than assume that all houses were originally identical, or to focus on the norm, I propose that from the beginning there was considerable variety among the houses at Olynthus, indeed at any Greek city; and that this variety should not be dismissed simply as deviations from the norm or later remodeling. Here then is one real crux of this study. What does this type house represent? How do we explain this similarity among the houses at Olynthus, or at other Greek or nonGreek cities? Where in a culture is this type house to be found? Through what mechanisms do the characteristic features of the type house come to be realized in the houses when they were built? And, equally importantly, how do we explain the differences between houses? Are they really due primarily to later remodeling, or were houses built differently from the beginning? And if so, are there regularities among those differences in house design which could tell us something about the organization of the city? Remodeling of the Houses Before going further, we must examine the claim that variations in house design are due to later modifications, whereas the original plans were more standard and closer to the type house. One of the unique attractions of Olynthus, paradoxically, is its relative lack of history. Except for the reoccupied area, the gridplanned part of the city was inhabited for not more than eighty-four years, and so the plans and organization of its houses are still relatively close to the situation at the time when they were built. Nevertheless, three generations of Olynthians lived in those houses, bought and sold property, redesigned and remodeled houses as their household situations and needs changed. The final state of the houses was certainly different from their original plans. How much of the original plans are preserved? The easiest alterations to detect are those which altered the original allotments, where one house has encroached on the plot of another. These can also serve as test cases: because we can be fairly sure that these are later changes to the original grid, we can determine to what extent traces of the initial state are preserved in The Organization of Blocks

195

figure 44. Reconstruction of Olynthus by Hoepfner and Schwandner, showing Typenhäuser. Hoepfner and Schwandner, Haus und Stadt fig. 56

the archaeological record. If they are not preserved, we have little hope of distinguishing other kinds of remodeling. But if they are preserved, then in other cases the absence of traces of remodeling would carry more weight, especially since most of the houses were excavated well below the final floors, revealing earlier architectural phases. In three cases, remodeling was accompanied by the incorporation of part of one house plot into a neighboring house: houses A 6 and A 7, A iv 5 and 7, and A v 6 and 8.2 All these houses were substantially remodeled, and their final plans were probably quite different from their original layouts. In all these cases, however, the foundations of the original party walls were preserved and recovered in the excavations. In these houses, then, it seems that the new owners did not root out all traces of earlier construction, but removed the original walls only down to floor level, leaving foundations in place. A few houses seem to have been substantially remodeled, without, however, expanding beyond the original plot. A v 9 seems to have been extensively rebuilt; some of its internal walls contain reused stones such as orbes from olive-crushing machines.3 Some of the irregular features of house A vi 9 are probably also later alterations, such as the wall dividing the pastas from the court (if this is actually a freestanding wall and not simply a foundation for pillars). A viii 3, another irregular house, was probably also remodeled: traces of earlier walls cross the northeastern part of the house. These houses are all rather irregular in their final plans, and in their original states they may have been more regular in design. Later features are found in a number of other houses at the site, and it is sometimes difficult to tell how substantially these changed the organization of the house. Nevertheless, the houses at Olynthus certainly seem closer to their original forms than houses at other Greek cities. Compare, for instance, the houses at Priene, most of which were also inhabited for a relatively short period of time, about two centuries; yet the remains reveal numerous phases of building and rebuilding, and even the original raster was obscured by the many changes the inhabitants made to their dwellings.4 Himera, likewise, shows many phases of construction and considerable irregularity in house design, although it too was occupied for a relatively short time. In contrast, the majority of the houses at Olynthus show much less evidence of later alteration, and except for a few cases outlined above, many of the alterations which can be documented seem not to have greatly changed the organization of the house. And although some of the houses which depart from the ‘‘Typenhaus’’ design were remodeled and may have been more regular in their original plan, others reveal no traces of earlier, more regular planning, even though many were excavated well below the final floor levels. For instance, houses A 11 to A 13, which do not conform to the pastas type, show no evidence of earlier, more regular layouts. In the absence of any evidence to the contrary, it is most reasonable to assume that the houses are fairly close to their

The Organization of Blocks

197

original situations, rather than postulate an original, standard plan which has in many cases been entirely removed, leaving no traces. It is misleading, therefore, to assume that all divergences from the type plan are due to later rebuilding and attribute too great a degree of uniformity to the houses. Certainly a number of houses have been significantly altered from their original plans, but these cases are relatively few, often easily distinguishable, and traces of their original plans often remain. There are other and more compelling explanations for the variety of house designs at Olynthus. Other Explanations for Variations in House Designs The variety among the houses at Olynthus becomes even more striking when contrasted with sites where the houses really are identical. In some Near Eastern cities, for example, the type house takes on an entirely different meaning from Greek cities. Rather than general similarities or principles, the houses there actually are identical, or very simple transformations or mirror images of one another. Egyptian workers’ villages at Kahun or Amarna, for instance, consisted of row upon row of almost exactly identical units. Likewise, the houses of the gridplanned settlement at Zernaki Tepe in Urartu seem to be all of the same plan, built four to a block, each identical to or a mirror image of the others.5 In such settlements where houses really were identical, the most likely explanation is that the houses were planned and built by the state (or perhaps by contractors working for the state), to house workmen or resettled populations. In these cases the ‘‘Typenhaus’’ probably was imposed from above as an act of public policy, but it represents the exact opposite of Hellenic democratic ideals, isonomia and the like: they are the epitome of authoritarian city planning. The contrast to even the most rigidly planned Greek cities is clear. Environmental Factors Instead, we should look to other sorts of explanations for the variety of house designs at Olynthus. A wide variety of factors influence how houses may be designed and built. Rapoport, in his classic House Form and Culture, distinguished a number of physical and sociocultural factors which affected house design, including climate, topography, available materials, level of technology, available economic resources, function, religion, defense, and the like. While physical conditions obviously affect the forms and construction of people’s houses, Rapoport concludes that sociocultural conventions are far more important.6 Besides, at Olynthus we can eliminate or minimize most of the physical and environmental factors: the climate is pretty much the same throughout the city; the same materials and technology would have been available to all citizens, or if they were not, the reasons would have been cultural (in the broad sense) rather than due to literal unavail198

The Organization of Blocks

ability. Function and economic resources vary here more according to cultural than to other factors. Chronology and Topography Likewise, the effects of chronology and topography were probably not significant at Olynthus, at least among the houses excavated by Robinson. Most of the houses on the North Hill were built within a short period of time; chronological development cannot account for the differences between them. The Villa Section is probably a later expansion, but not so much later that chronology per se is a significant factor. The differences in house designs in this part of the city are probably as much due to the changed circumstances of the city and its citizens, now the undisputed masters of the Chalcidice, as to simple development over time. The topography of the site likewise varies only slightly. Some houses were built on the slopes of the hill, but these were not excavated or preserved well enough to consider in this book. For instance, ESH 6 is built into a steep slope, but because of this, it is rather poorly preserved. Otherwise, there is no obvious ‘‘right side of the tracks’’ on the North Hill, no region that is significantly cooler or shadier, breezier or better watered. The city’s organization, like its grid plan, follows a conceptual pattern and was rationally imposed on the landscape, while purely environmental or physical factors played relatively insignificant roles in determining the layout or differentiating regions. Architectural Constraints on House Design Some of the similarity among houses at Olynthus is due to the constraints imposed by the construction of houses in rows and blocks. The most obvious of these is the division of houses along the two major axes. These divisions were imposed by the need to support a common roof running lengthwise over the northern half of the houses. The walls and pillar bases at the midpoint of the house, between pastas and court, supported the southern edge of the common roof; walls on the northern axis, between pastas and the rooms to its north, supported its ridgebeam. Although these axes are placed somewhat differently in different blocks, with very few exceptions they are closely aligned from house to house.7 The placement of these axes and number of stories must have been decided on beforehand, with all the households on a block agreeing to coordinate where these axes would fall, or a common roof would not have been possible. The common roofline does not, however, explain the general structure of Olynthian houses. The houses in Row A and on the East Spur Hill, for instance, were not built in east-west rows and did not share a common roof, but they nevertheless conform to the general pastas type, and the roofs seem to have worked like those of the other houses, with a ridgeline running east-west and rainwater drainThe Organization of Blocks

199

ing into narrow alleys between the houses. The roof could have been oriented north-south instead of east-west, with the pastas facing west rather than south— as, for instance, in house A viii 5—and then shared a common roof running along the length of the row. But other principles, such as that the houses should face south, outweighed the desire for a single roofline. Even free-standing farmhouses often belong to the general pastas type and organize the north part of the house with major axes like houses at Olynthus, although these houses would not have been constrained by a shared roof. Cultural Conventions Cultural conventions can be considered on a number of levels: conventions common to Greeks in general, those shared among citizens of a single region, or a single city, or among smaller groups of people within a city. Cities throughout the Greek world were laid out in a grid plan, with standard house plots distributed among the citizens. This convention of building adjoining houses on lots of identical size was common to most Greeks. Different cities and regions adopted different house types. At Olynthus the pastas type is almost universal, whereas at Priene, Colophon, Abdera, and some other sites the prostas type is most commonly found. Prostas houses and pastas houses were apparently not used together in the same city—citizens built one or the other. These house types seem to be shared among related Greek cities. Ionian towns tended to use prostas houses, or at least adopt some of the principles of that type, such as axiality and hierarchical arrangements of rooms. And the colonies of these cities seem to have used the types of houses of their homeland: the city of Abdera in Thrace was not far from Olynthus, but it was a colony of Ionian Teos and probably for this reason its citizens built prostas houses.8 Pastas houses, on the other hand, are common to both Olynthus and Eretria, not far from the homeland of the Thracian Chalcidians. The relationship between the Thracian Chalcidians and the Euboeans may lie behind their use of similar types of houses, just as the relationship between Abdera and Teos probably lies behind the use of the prostas house at the Thracian colony. Within these general types, cities often have their own local variations. Most of the pastas houses at Eretria include a distinctive three-room complex not common elsewhere, at least in such a standard form (fig. 45). Such local variants are relatively little known, mostly because too few houses have been excavated at most sites to discern patterns in their designs. The kitchen-complex at Olynthian houses is probably such a local variant: although complexes resembling these are occasionally found at other sites, no truly close parallels are known. On a lower level yet, different citizens and groups of citizens will have had different needs and different opportunities, and were subject to different expectations and demands in building their houses. The social positions of some house200

The Organization of Blocks

figure 45. Plan of the House of the Mosaics at Eretria. After Ducrey et al., Eretria 8, fig. 25 holds, for example, demanded that they have spaces for symposia and such events. They therefore devoted space and resource to building androns in their houses, elegantly adorned with painted walls and cement or mosaic floors. The professions of other citizens would require workshops in their houses or special installations in specific rooms. Houses fulfilled the particular needs of their owners, both as individuals and as members of a group within the citizen body.9 Greek house designs were affected by the generally shared cultural conventions of the inhabitants, conventions which acted on different levels, and by the specific needs of the households which built and lived in them. In the variety of architectural forms and the uses of domestic space throughout the city, therefore, we The Organization of Blocks

201

should see a mixture of various influences. We see reflections of generally shared cultural traits, such as the adoption of certain house types as basic models; of distinct ‘‘subcultures,’’ groups of people within the city who share some common ties; and of individual needs. By studying this variety, and trying to discern what factors influenced the design of the houses and the use of space in them, we can address questions about the composite nature of the Greek citizen body and about how a diverse group of citizens and their families organized themselves in a newly founded city. Answers to these questions cannot be obtained directly. We must begin with questions to which the archaeological and historical material can provide at least tentative answers; answers which in turn could shed light on those more general issues. For instance, we must ask who determined the plans and furnishings of individual houses, at Olynthus or any Greek city? Was a standard plan imposed by the state for reasons of public policy, as Hoepfner and Schwandner suggest? Or were the plans determined by professional architects or builders who brought their own specific ideas to the project, perhaps modifying their design to account for the needs of the client but still staying fairly closely within the limits of a single master plan? Or was the house completely planned by the individual owner? In which case, how do we account for the general similarity of most of the houses?

w h o de s i g n e d a n d b u i lt the h o u s e s a t ol y n th u s ? We can learn something about how the houses were built from the construction of their walls. Robinson and Graham observed that the outside walls of each row of houses all bond together, showing that the row, and in most cases probably the whole block, was planned and built as a unit.10 This is implied by the common roof as well. But although each row of houses must have been built all at once, it was not built by a single crew of workmen. The wall foundations belonging to each house are generally built in a slightly different manner from the foundations of the house next door: for instance, one house may be built of small flat stones, while the next is built of larger boulders with small chinking stones wedged between.

seams

These seams between different stretches of wall can be seen today in the north walls of A v 1 and 3; A v 2 and 4; A v 6, 8, and 10; A v 7 and 9; in the south wall of A v 4 and 6 (where the street wall of A v 6 was built of well-cut squared stones, 202

The Organization of Blocks

while A v 4 was built of the usual unworked rubble),11 A vi 6, 8, and 10; and in all the houses in block A vii. Block A viii is not well enough preserved to show distinctions, while B vi and most of the rest of the city is now too overgrown to distinguish much.12 Party walls were probably built by the builders of one house rather than as a communal effort. In Row A, party walls belonged to the house to the south, showing that the row was built from south to north. The regular blocks seem to have been built from west to east, and party walls belonged to the house on the west, as shown on the schematic plan above.13 There are few easily discernible patterns in the bonding of internal house walls to the outer and party walls. In some cases the internal walls are bonded to the outer walls, proving that these rooms belong to the original layout of the house. In others they abut the outer walls, but this does not necessarily imply that the walls are later modifications to the plan. The bonding of the outer house walls shows that the blocks were built at one time, but the differences in construction between stretches of wall demonstrate that different crews of workmen, working simultaneously, built the individual houses. Although it cannot be proven, it seems most likely that the crews of workmen consisted of the households which owned the house, together with friends and relations. If the households who lived in the houses were also responsible for building them, they probably designed them as well, or at least had a strong say in the design. An architect or surveyor was probably responsible for the layout of the grid over the city, planning the streets, public areas, house plots, and the like, and may have helped lay out houses, but the design and construction of houses are more typically the concern of the households that will live in the house, rather than professionals in modern villages and ethnographic parallels. Although professional house builders are attested in literary sources, other sources emphasize the household’s role in designing and building the house.14 Xenophon, for instance, points out that ‘‘some men spend large sums in building houses that are useless, while others build houses perfect in all respects for much less’’—emphasizing both the variety in the design and appointment of houses and the personal nature of their design.15 When Antigonus ordered the synoikism of Lebedos with Teos, he ordered that if the Lebedians had not built houses in Teos on the plots given them, the land should become public. Even in this case where the king is actually sponsoring the synoikism, he chose to provide money and roof tiles to the citizens for them to build their own homes, rather than employ his own troops or other forces. The responsibility and initiative to build thus falls on the Lebedians, not on the king (who, if he had wished, could simply have ordered the houses built).16 Likewise, Pausanias’s description of the building of Messene is one of the whole citizen body collectively building the houses, walls, and public buildings; the easiest way to organize this would be for each family to build its own house.17 The Organization of Blocks

203

The Coherence of Blocks The blocks of houses thus form coherent architectural units: the houses were built all at one time, although by separate crews, with the crews working in concert and cooperating with one another. They could only have been built after a number of matters had been agreed on by all the households. The common roof is one such linking element: all the families would have had to agree on how many stories their houses would have and where the rooflines would fall before construction could proceed. Drainage was a mutual concern as well: roofs and the central alley of the block had to be arranged so that water would be carried away efficiently and not seep into a neighbor’s house. Some blocks, particularly those in the eastern part of the city where the slopes were steeper, would have required landscaping and terracing to control drainage, and this would have been a concern of all the residents of the block.18 The blocks do not simply form basic units of the city plan; they form architecturally coherent buildings constructed in an organized way, probably according to some kind of a ‘‘master plan,’’ by groups of households cooperating with one another.

va r i a t i on s on the t y p e h o u s e Even among houses which conform to the regular type, there are significant differences in the arrangement and use of space, in the appointment of rooms, in the number of specialized spaces, and so forth. Some of these differences are immediately obvious. Compare, for example, the houses in blocks A v, A vi, and A vii (see table 5, fig. 7, plate 1). Although most of these houses conform to the general Olynthian house type, the differences among the houses on these blocks are apparent. Nearly every house in block A vi has a kitchen and rooms with cement floors and painted walls; most have androns; and half of the houses have mosaics, some very elaborate. A relatively large proportion of the spaces in these houses is architecturally specialized and ornate. The houses in block A v, in contrast, are largely without architecturally distinctive or elaborately decorated rooms. They have more generalpurpose rooms, more architecturally unspecific spaces along the north side of the house, and fewer kitchens and androns; and all the androns, one of the kitchens, and most of the cement floors belong to houses which were remodeled. The two rows of block A vii are very different in character. Houses in the southern row are quite regular in plan, although not as well appointed as those of block A vi. By contrast, the houses in the northern row are mostly rather irregular, with few or no specialized rooms, only one certainly identifiable pastas, and irregular early structures that intrude into the raster. Adjoining houses elsewhere occasionally share remarkably similar plans.

204

The Organization of Blocks

table 5: Specialized Spaces and Decorated Rooms in Blocks A v to A vii Cement Floors

Mosaic Floors

Painted Walls

1

6 in 3 houses (5 in A v 6 and A v 8, remodeled)

none

7 rooms in 3 houses (all but one in A v 6 and 8)

8

3

20 in 9 houses

6 in 5 houses

23 rooms in 8 houses

1

5

2

4 in 2 houses

none

7 rooms in 4 houses

none

none

none (1 tub in N. Room)

none

none

none

Block

Androns

Kitchens

Baths

Av

3 (all in remodeled houses)

4 (1 in A v 6, remodeled)

A vi

7 in 6 houses

A vii (south half ) A vii (north half )

Houses A 9 and A 10, for instance, share a great many features in their design and layout, and A 8 is also quite similar (fig. 7). The three houses just to their south, A 11, 12, and 13, however, are completely different, consisting of architecturally unspecialized rooms arranged around courtyards, without pastades or the clear tripartite division along the major axes. They are quite different from the other houses of Row A and indeed from the type house. These comparisons bring us to a series of observations about the houses at Olynthus. First, there is very real variation in the design and construction of houses, even among those which conform to the standard plan. Second, there are no obvious chronological, geographical, or environmental explanations for why some houses should be better appointed and include more specialized rooms than others. All these houses were probably built within a short period, not enough time for chronology alone to be an important factor in the differences. To be sure, the mosaics and painted walls may be later embellishments to the houses, but we still must explain why the houses on one block were so embellished while those across the street were not. And finally, these variations on the type house are not just scattered throughout the city but are localized: houses in one block are different from houses on the block across the street. This grouping of similar houses confirms for us that we are looking at significant distinctions, not random variability.

The Organization of Blocks

205

Cluster Analysis and Groups of Houses Such observations about the planning and appointment of houses on different blocks can lead to interesting conclusions, but a more objective method of comparing houses is needed. Moreover, there are enough houses at Olynthus, and they are complex enough, that we need a somewhat more formal and structured way of comparing houses to one another and finding groups of similar houses. Many such methods have been developed for finding natural groups or classes among a collection of objects. Most of these were developed in the biological and behavioral sciences: they were developed to distinguish species of animals, personality types among people, and for innumerable other such needs. Cluster analysis is a general term for a variety of such techniques. Cluster analysis begins by describing various traits of each individual: for houses, we might tabulate the types of rooms present; the number of androns, kitchens, baths, courtyards, and the like; the areas of those rooms, access, their decoration, floor material, and other such characteristics. Cluster analysis then compares the houses according to those traits in order to determine the similarity or dissimilarity between them, and uses coefficients of similarity to group the entities into groups or ‘‘clusters.’’ These techniques have been used successfully in archaeological research in a number of cases. This is essentially what scholars have done since Aristotle: compare things with one another and group the similar ones together. There are, however, many different ways of determining how similar one thing is to another. How similar things are will depend on which aspects you look at. Is a whale more similar to a fish, for example, because both live in the sea, or to a dog, since both have warm blood and bear live young? Depending on whether you are interested in environment or physiology, the whale may be assigned to different clusters. One must therefore look at a range of different variables and explore each in turn to understand the similarities and differences, not simply postulate one set which you believe will produce a single correct grouping. Cluster analysis facilitates this process and helps make explicit the processes of reasoning by which we automatically group things together as similar. Of the different methods of cluster analysis, I have found ‘‘partitioned’’ or ‘‘Kmeans’’ clustering the most rewarding.19 In this technique, you have to specify beforehand how many clusters you want the houses divided into. The computer then finds the optimum division of houses into clusters, so that houses within each cluster are as similar as possible to one another, and the clusters as different as possible from each other, and produces tables of the coefficients of similarity (appendix 1).

206

The Organization of Blocks

How Many Clusters? One problem is how to decide how many clusters are represented in the data. Positing too few groups conceals variety by lumping dissimilar houses together, but dividing the houses into too many groups introduces confusion and hides similarities. Although there are various mathematical ways to determine the optimum number of clusters, heuristic methods were most effective in this case.20 We may divide the houses into different numbers of clusters and then compare the resulting groups with one another. Some clusters were very stable, remaining together no matter how many groups the houses were divided into, while others gained or lost members when the number of clusters was increased or decreased. By looking at these less stable clusters, watching which houses were joined or separated as the number of clusters changed—and particularly learning why they were being joined or separated—we may decide on a number of clusters that gave the best idea of similarities and differences among the houses. Moreover, as we have seen, houses on different blocks differ in a number of ways, and certain clustering solutions will reveal those differences. Those clustering solutions which show distinct distributions through the city, and which thereby teach us something about the internal structure of Olynthus, are more interesting and elegant than solutions that are randomly distributed—at least for the purposes of this book. To most of these questions, there is no single right answer. But the objective is not simply to classify the houses at Olynthus into groups, but to understand how they differ and try to explain why they do so. Thus the analyses offered below are intended as exploratory investigations, rather than as absolute classifications of houses into different types. There is no reason to think that there is a single ‘‘correct’’ grouping of houses. If we find fairly distinct classes or groups, it is not because there are certain natural ‘‘species’’ of houses, the way there are species of animals. Rather, the owners and builders of similar houses probably shared social and cultural ties and requirements and were influenced by similar factors in the process of designing their houses. Analysis of the architectural features of these houses helps us understand those ties and concerns. Many factors influenced the design of houses, however, and we should not expect that houses can be divided into simple groups. Instead, there will probably be various ways of grouping the houses, depending on which aspects we look at, and these will cross-cut one another, so that a house which is similar to one group of houses in one respect may be similar to other groups in other respects. By grouping the houses in different ways, mapping various groupings, and seeing how houses in each cluster are similar to one another and how they are distributed in the city, we can try to understand some of the factors which influenced citizens to build their houses the way they did and how they reacted to different influences.

The Organization of Blocks

207

The Allocation of Space: Architecturally Specialized Rooms The first aspect of house design to consider is the amount of space devoted to different types of rooms. Some houses have large courts, some small; some have many specialized rooms like androns, shops, or kitchens, others have few specialized rooms but contain many architecturally undistinguished spaces. We can compare the areas of the different types of rooms which make up the Olynthian houses, and so group the houses into clusters which show similar allocation of space. Again, we are looking at the architectural form of the spaces, at room types rather than at their contents or the actual function of rooms in the final moments of the city. To compare houses, the floor area of each type of room was expressed as a percentage of the total area of the house because some houses are larger than others. This analysis included courtyards, pastades, North Rooms (i.e., rooms on the north side of the house which were not otherwise specialized—an andron or kitchen which was located on the north side of the house was classified as an andron or kitchen, not a North Room), androns (including the anteroom if there is one), kitchen-complexes (including flue and bath), other nonspecialized rooms (not on the north side of the house: rooms opening to the court, ‘‘pastas rooms,’’ etc.), and shops. Note that this does not directly take into account the disposition of the rooms in the house. Nor does it consider the appointment of rooms— which houses have mosaic or cement floors, painted walls, and the like. Finally, this analysis only considers the ground floor of houses. The second story remains too problematic to include in actual analyses, but we must keep in mind that some of these houses had much more space available than we see now in plan, and that this must have affected how the household allocated space. The areas of rooms were derived from a new computerized site plan, based on published dimensions, dimensions in unpublished field notes, and where these were lacking, the published plans. A number of houses were too incompletely preserved or excavated to include in the analysis. These included A 7, A iv 1 and 3, B i 5, B vi 6, 8, and 10, ESH 5 and 6, the House of the Comedian, the House of the Wash Basin, the House of the Twin Erotes, and Villa CC. Many of these houses, however, are clear enough in plan to attribute them to clusters defined among the better-preserved houses. These are colored but marked with a question mark in the plan (plate 3). After experimenting with solutions from four to ten clusters, a five-cluster solution seemed to be most effective. If the houses were divided into more clusters, clusters split into two smaller groups which did not seem significantly different, and if they were divided into fewer clusters, houses which do seem different were joined with other clusters. Again, there is no single correct number of clusters; this exploration is aimed at elucidating and understanding some of the patterns in the houses. 208

The Organization of Blocks

The five clusters are mapped in plate 3. The actual results are given in appendix 1. The cluster numbers are arbitrary. Clusters 1 and 5: Regular Houses Cluster 1, 30 houses: A −1, A 3, A 5, A 9, A 10, A iv 5, A v 1, A v 2, A v 3, A v 4, A v 5, A v 7, A vi 1, A vi 3, A vi 7, A vii 3, A vii 5, A vii 7, A viii 1, A viii 6, A viii 7, A viii 8, A viii 9, A xi 9, A xii 9, B vi 7, B vi 8, House of the Tiled Prothyron, ESH 2, ESH 3. Cluster 5, 35 houses: A 1, A 2, A 4, A 8, A v 6, A vi 2, A vi 4, A vi 5, A vi 6, A vi 8, A vi 10, A vii 2, A vii 4, A vii 6, A vii 8, A vii 10, A viii 2, A viii 4, A viii 5, A xi 10, A xiii 10, B v 1, B vi 1, B vi 2, B vi 3, B vi 4, B vi 5, B vii 1, B vii 2, ESH 4, House of the Comedian, House of Many Colors, South Villa, Villa of the Bronzes, Villa of Good Fortune The houses which conform to the standard plan are grouped in this analysis into two clusters. The houses of the smaller cluster, cluster 1, are colored blue, and those of the larger cluster, cluster 5, are colored green in plate 3. These two clusters are rather more ‘‘fluid’’ than the clusters described below: that is, if different numbers of clusters or different criteria are specified at the beginning of the analysis, the membership in these two clusters tends to change somewhat. However, they do not change greatly, and repeated experimentation with different methods suggests that this solution is among the most satisfactory and stable. The major difference between these two groups of houses is the number of specialized rooms in the houses: androns and kitchens. Houses in cluster 5 have more specialized spaces: every one has an kitchen-complex, and fourteen of the thirty-five houses have androns, while only two of the houses in cluster 1 have kitchen-complexes, and eight have androns. Thirteen houses in cluster 5 have both kitchens and androns, while no house in cluster 1 has both types of specialized rooms. Finally, although this analysis does not examine the decoration of rooms, the houses of cluster 5 tend to be better appointed and decorated than those of cluster 1, with more mosaics, cement floors, painted walls, and such embellishments. Most interestingly, these two groups of regular houses are distributed very distinctly throughout the city. Houses belonging to each cluster are not concentrated in specific regions of the city; both clusters are found throughout the North Hill and Villa Sections. But houses on one block or row tend to belong to a single cluster. For instance, those on A v generally belong to cluster 1, while those on A vi belong to the more specialized cluster 5. Houses on southern row of A vii also belong to cluster 5, while those on the northern row are in cluster 1. This agrees with the empirical observations noted above. There are exceptions, of course; but many of these can be explained by inThe Organization of Blocks

209

specting the specific situation of each house. House A v 6 clusters with the more specialized houses, as it has an andron and kitchen-complex. But this house expanded into half of the lot of its neighbor, A v 8, and the kitchen-complex was certainly built after the house was expanded since it lies in the newly added part of the building. The andron may have been added at this point as well; it is one of the few recognizable androns which lacks a raised border around its cement floor, although it has painted walls with a baseboard, an anteroom, and an off-center door. The remaining half-house A v 8 was probably turned into a nonresidential building and clusters on its own (cluster 3). The other two houses on block A v which do not belong to the cluster of less specialized, regular houses belong to the cluster of houses with shops (cluster 4). On the north row of block A vi two houses—A vi 1 and A vi 3—clustered with the less specialized group of houses. But although they lack kitchens, these two houses have fancy androns with figural mosaic floors: in A vi 1, the mosaic in the pastas showed Nereids, and in A vi 3, the andron mosaic showed Bellerophon and the Chimaera, and the court mosaic depicted a Centauromachy.21 They are therefore more similar in many respects to other houses on this block belonging to cluster 5, with androns, kitchens, mosaic floors, and painted walls, than they are to the houses on block A v, which lack mosaic floors and, except where rebuilt, androns. Cluster 2: Irregular Houses Houses A −5, A 11, A 12, A 13, A iv 7, A viii 3, and A viii 10 A third distinctive cluster includes houses of irregular plans, often with a courtyard but without a pastas, with few shops, few androns, and no kitchens. Most of the area is taken up with unspecialized rooms, some of them corresponding in position to North Rooms in the standard plan, but many of them simply opening onto the courtyard. Three of these houses, A iv 7, A viii 3, and A viii 10, were extensively remodeled, and may well have been more regular in their original designs. A viii 3 had an industrial installation in its southeast corner. A iv 7 and A viii 10 were divided into two units; the western half of A iv 7 was joined to A iv 5, while A viii 10 seems to have been converted into two independent houses. But other houses, particularly houses A 11, A 12, and A 13, never seem to have had more regular plans. There are no traces of earlier walls which might indicate extensive remodeling, although these houses were excavated well below the levels of the latest floors. A number of houses can be added to this group. In Trench 7, houses B iii 4, B iii 6, and B ii 2 and 4 were only partly excavated, but they do not seem to be planned around a court and pastas and contain a large number of unspecialized rooms. Had they been fully enough exposed to include in the analysis, these 210

The Organization of Blocks

houses probably would have been grouped here with the other irregular houses. House A 7 is also irregular in plan, but this is because it was largely remodeled after most of its plot was taken by its neighbor, A 6. And the two houses in Section O (houses C −x 5 and 7) were also very irregular in plan. Most of these houses are located in the southern part of the North Hill, close to the agora (below, chapter 6, ‘‘The Agora’’). Relatively few houses were excavated in this part of the city, but those which were tended to have more irregular plans than the houses further north. Many of the houses of this cluster in the northern part of the city, in contrast, owe their irregular plans to later remodeling rather than to their original designs. Cluster 3: Not Residential House A v 8 This cluster consists of only a single house. House A v 8 was probably not residential: in its final form it contained only a court, two androns, and a small room off the entrance. The unusual allocation of space distinguishes this house from all others at the site. Although it is possible that there were living rooms on a hypothetical second floor, its two androns make it more likely that this building served some other purpose, for instance as a clubhouse or center for a hetaireia, or perhaps as a religious building. The artifact assemblage does not suggest any domestic functions (below, chapter 6, ‘‘A Clubhouse and a Hostel’’). Note that this house was remodeled at the time that A v 6 was expanded; it may have been a regular dwelling when originally built. Cluster 4: Houses with Shops Houses A 6, A iv 9, A v 9, A v 10, A vi 9, A vii 1, A vii 9, A xii 10, B vi 9, D v 6, ESH 1. Cluster 4 is a distinctive group of houses. All include shops or workshops: rooms opening onto the street as well as (or instead of ) the rest of the house; these rooms are colored red in plate 3. The houses in this cluster devote an average of one-quarter of their total area to shops. Like the distribution of houses in the two regular clusters, the distribution of these houses is striking. Rather than being grouped by block or row, all but one of these houses are located along a north-south avenue; most of them are on Avenue B, their shops opening directly onto the avenue. The other houses with shops, not included in this cluster, also lie mostly along Avenues A and B; and nearly every house along Avenue B has one or more shops.22 Their position fronting onto an avenue was apparently an overriding factor in the design of these houses. Avenue B was a sort of main thoroughfare for shopping and commerce, more than the other excavated arteries or regions of the city. The Organization of Blocks

211

We will explore the evidence for the use of these shops and for the economic and industrial life and exchange at Olynthus in chapter 6. In other respects the group is rather diverse. Some houses were fairly regular in plan, with a court, pastas, and specialized rooms. Five of them, A 6, A iv 9, A v 9, A v 10, and A vi 9, had kitchens, and two, A 6 and ESH 1, had androns. Other houses, such as A vii 9 and D v 6, were quite irregular in plan, with no court or only a very small one, no pastas, and few other specialized rooms. A vii 9 was built around an earlier building, which may or may not have belonged to the same household; but D v 6 seems similar in its general layout to the irregular houses described above, like A 11 to A 13. House Clusters: Summary Olynthian houses are not simply types. Although they share certain elements of design, each is distinct and different. Moreover, the variety of house designs throughout the city, their differing uses of space and the presence of specialized rooms and of parts of houses devoted to certain activities, are not random but are patterned and interesting, and can tell us something about the organization of the city and of its citizens. The division of houses by cluster analysis is not meant to create a strict typology, but to highlight shared features of houses which have some historical and social significance. The importance of the preceding analysis, then, is not simply to show that houses are different from each other and that we can group them into clusters of similar houses, but that those clusters show patterned distributions in the city. This patterning reinforces the validity of the clustering solution described above. If the houses of these clusters were distributed randomly across the site, it would not mean that the clusters were not valid or did not reflect ancient distinctions; but where the geographic distribution is so distinctive, it suggests that the statistical analysis is discerning significant historical or social differences. The patterned distribution of house clusters operates on a number of levels. The regular houses of clusters 1 and 5 are distributed on the level of blocks or rows of houses. With some exceptions, blocks or rows of houses tend to belong to either one cluster or the other. The houses of some rows share distinctive architectural features, such as the designs of kitchens in the southern row of A vii. Houses characterized by shops in cluster 4, on the other hand, are not distributed block by block or row by row but are primarily along the length of Avenue B, cutting across blocks, with a few along Avenue A and elsewhere. Irregular houses of cluster 2 are mostly located in a specific region of Olynthus: the southern part of the North Hill, around the agora of the city. Some of the main spatial factors which seem to affect house design, therefore, are the general region of the city (especially proximity to the agora), the block to which the house belongs, and the position of the house in the block (i.e., whether the house fronts on an avenue or a street). 212

The Organization of Blocks

Row A includes houses of nearly every cluster, with both irregular and regular plans. But this is not entirely random. The three southern houses are all irregular and seem to have been so in their original designs. A 9 and A 10 are similar in design to one another and cluster together in the less specialized cluster 1. In this analysis A 8 joins the more specialized cluster 5, but in plan it closely resembles its neighbors A 9 and A 10 to the south. To the north of these house A 7 in its final state is quite irregular because much of the house was taken over by A 6, leaving no court or pastas; its original state is impossible to reconstruct. A 6, which like house A v 6 expanded into its neighbor’s plot and was heavily remodeled, clusters with the more specialized houses in cluster 5. The houses to the north of this vary in plan, although they share such features as having an unusual number of porticoes facing the court and having two stories. Nearly all the houses in the Villa Section cluster with the more specialized group, cluster 5.They share other architectural peculiarities which set them somewhat apart from the more ‘‘typical’’ houses of the North Hill. The Villa of the Bronzes and the House of Many Colors, for instance, have suites with light wells. The House of Many Colors has an unusually large covered space south of the court, a sort of exedra with a cistern in the middle. The House of the Tiled Prothyron has an unusual suite of hierarchical rooms f-g-k and unusually deep North Rooms. There is, then, considerable variety in design and allocation of space among the houses of regular plans, and this variety shows distinctive distributions through the city. The division proposed here into two clusters does not capture all the variety, of course; but some patterns seem clear. We see similar patterns in the appointments of different houses. The Embellishment of Rooms Although cluster analysis could be applied to the decoration of rooms, it is easier to discern patterns from maps of different features. Figure 46 illustrates the treatment of floors in different houses; figure 47 maps the number of painted rooms. Many of these features might be late additions to the houses. Rooms could be painted or repainted, and some of the mosaic floors may have been added to the rooms where they were found. Again, patterns in the distribution of painted walls and cement and mosaic floors are easily visible. As noted before, block A vi is by far the best-decorated block excavated in the city: nine of its ten houses have rooms with cement floors, and five houses have mosaics. The same nine houses also have many more painted rooms on average than houses on other blocks. Except for the remodeled A v 6 and 8, by contrast, the houses in block A v have no cement floors, no mosaics, and only one painted room.23 The distinction pointed out above between the north and The Organization of Blocks

213

south rows of block A vii, the southern row being considerably more regular in design than the northern, is also visible in the appointment of the houses in the two rows: none of the houses in the northern row have painted rooms or cement floors, while three houses in the southern row have cement floors, and four have painted walls. Houses in the Villa Section are more richly ornamented overall than the houses of the North Hill. All except the extremely poorly preserved Villa CC have at least two painted rooms, most have rooms with cement floors, and three have mosaics, including the very fine mosaics in the Villa of Good Fortune. The Villa of the Bronzes also has cut-stone masonry walls on two sides. The irregular houses of cluster 2 are extremely modest: only one has any rooms with painted walls, and only two have rooms with cement floors. Again, some of these houses, particularly A iv 7, were probably more regular in their original design, and the well-furnished rooms may date from the period before the remodeling; while the houses that seem irregular in their original layout, such as A 11 to A 13 and the houses in Trench 7, are uniformly plain. Houses with shops, belonging to cluster 4, vary from rather well appointed (A iv 9 and B vi 9, for instance, houses which are similar in design as well), to very plain (A v 9 and 10, A vii 9, D v 6). Again, other than their common feature of including a number of shops, these houses differ in design, some more regular and others quite irregular. Not surprisingly, perhaps, the regular houses in this cluster are much better decorated than the irregular houses. As one might expect, the houses of the two ‘‘regular’’ clusters, clusters 1 and 5, show significant differences: with some notable exceptions, houses of cluster 1 are considerably less well decorated than those of cluster 5. This is easily visible by comparing figures 46 and 47, and plate 3. The distinction is not simply due to the fact that houses in the more specialized cluster 5 tend to have more androns and anterooms, which are nearly always paved with cement and painted. The more specialized houses also have a higher proportion of cement-paved or cobbled courtyards and more painted and cemented rooms which are not otherwise specialized. There are exceptions, of course: house B vi 4, for instance, was called the ‘‘Yellow House’’ because nearly every room was plastered and painted yellow, yet in plan it is more similar to the houses of cluster 1 than those of cluster 5. But the overall pattern is clear.

h o w t o e x p l a i n th i s va r i e t y ? We can assemble some of the observations made above into a clearer picture of the organization of Olynthus. First, although some of the houses were remodeled, figure 46. Treatment of Floors at Olynthus 214

The Organization of Blocks

Cement and Mosaic Floors at Olynthus 1 Room 2 Rooms 3-4 Rooms 5-9 Rooms Mosaic Floor 0

50

100 m

the general plans of most of the houses are relatively similar to their original layout, and most differences in plan between houses cannot be explained simply by later alteration. Second, the design and construction of the houses can be attributed, I believe, to the owner-builders of the houses rather than to city edict or the design of contractors. While a number of factors constrained the way houses were planned, such as the basic plot of land allotted the owner and requirements of the common roof and of drainage, the varying solutions in different blocks suggest that issues of common concern were worked out by more or less informal arrangement among the householders. Within that general framework, an owner was free to build in whatever way suited his needs and wants. Similarities among houses reflect similar social and economic concerns among the households which built the houses. The distinct patterns in the distribution of different house clusters therefore reveal social and economic patterns among the citizens of Olynthus. The coherence of house blocks is thus based not only on architectural but also on social ties: house blocks form not just the physical units of the city, but social units as well, and the physical organization mirrors certain aspects of the social organization of the city, as prescribed by the urban theorists discussed in chapter 1. This coherence between the social and physical organization of the city must have originated when the city was first laid out. The patterns distinguished above belong with the construction of the houses, and are obscured by later rebuilding, not emphasized. They arise from the processes by which citizens come to own and build their houses. The primary means is the original allotment of land, the process by which the state divided and distributed land to new settlers. The original allotment was altered as citizens sold and purchased land. But just as clear traces of the original house plans remain more clearly here than at most other sites, the final situation probably reflects the original allotment and distribution of land more closely at Olynthus than elsewhere. Many of the patterns in house design, particularly in the parts of the city built and occupied shortly after the anoikismos of 432 b.c., such as blocks A v, A vi, and A vii, probably reflect patterns in the initial distribution of land. We must therefore examine more closely this critical process in the creation of a new city. The Distribution of Land The ‘‘original distribution’’ of land in a newly founded city, and the prohibition against redistributing or subdividing those original lots, became important topoi in Greek discussions of land division and ownership of property. Aristotle refers to ‘‘legislation in many states prohibiting the sale of the original allotments’’ (τοὺς figure 47. Treatment of Walls at Olynthus 216

The Organization of Blocks

Treatment of Walls at Olynthus 1 Painted Room 2-3 Painted Rooms 4-5 Painted Rooms 6-8 Painted Rooms 0

50

100 m

πρώτους κλήρους) and ‘‘legislation to preserve the old allotments’’ (τοὺς παλαιοὺς κλήρους διασῴζειν); Plato decrees harsh penalties for anybody who buys or sells his

allotment.24 Some actual prohibitions are preserved in inscriptions. For instance, a sixth-century law from Lokris concerning pasturage (or inheritance) rights in a new community decrees that ‘‘whoever proposes a division (of land) or puts it to the vote in the Council of Elders or in the city or in the Select Council or who creates civil discord relating to the distribution of land, that man shall be accursed and all his posterity, and his property shall be confiscated and his house leveled to the ground in accordance to the homicide law’’ (ML no. 13; trans. Fornara). A third-century decree from Itanos in Crete likewise records an oath not to cause a redistribution of estates or house-plots, or a cancellation of debts.25 And the decree from Korkyra Melaina, described in detail below, preserves such prohibitions as well. This distribution of land was perhaps the single most important process in the creation of a city, for it determined or affected the character of neighborhoods, regions, districts of the town, and the use of space in different regions. The distribution had to be perceived as fair and equitable. Even in the earliest colonies of the eighth century b.c., when governments were anything but democratic, plots of land equal in area were distributed among the colonists, and the importance of equality and fairness is emphasized throughout Greek literature on urbanism. But as even Plato admits, a purely egalitarian state was impossible, and concessions had to be made to social realities. Exactly how these ‘‘first allotments’’ were distributed to colonists is less well understood. It has usually been thought that they were assigned to the settlers in a random allotment. Plato, for instance, orders that the settlers

καὶ δὴ καὶ νῦν τὸν ἀριθμὸν μὲν πρῶτον διὰ βίου παντὸς φυλάξετε τὸν εἰρημένον, εἶτα τὸ τῆς οὐσίας ὕψος τε καὶ μέγεθος, ὃ τὸ πρῶτον ἐνείμασθε μέτριον ὄν, μὴ ἀτιμάσητε τῷ τε ὠνεῖσθαι καὶ τῷ πωλεῖν πρὸς ἀλλήλους· οὔτε γὰρ ὁ νείμας κλῆρος ὢν θεὸς ὑμῖν ξύμμαχος, οὔτε ὁ νομοθέτης. guard throughout your lives the number just mentioned [i.e., the 5,040 original kleroi], and in the next place, do not dishonor the upper limit and size of your property, which you were originally apportioned as a reasonable amount, by buying and selling to one another; for neither will the Lot which distributed (the territory), being divine, be your ally, nor will the lawgiver. (Plato, Laws 741B) Distribution of land by lot is actually specified in Greek sources only very rarely, however, and the process is never described in detail. Its important place in modern works on Greek urbanism and land division is mostly derived from analogies to the distribution of public offices by lot in democratic states, for instance, or

218

The Organization of Blocks

descriptions of the distribution of land in Roman colonies. These are, at least in theory, completely random processes. Hyginus Gromaticus, one of the Roman agrimensores, describes a two-step lottery. In the first lottery, the names of the settlers, grouped in threes, were pulled from an urn to determine the order of allotment; in the second, the centuries of land were pulled from the urn, and each century divided among the corresponding three settlers.26 The resulting distribution would be completely random; a settler would have no control over what land he was allotted. The Distribution of Land at Korkyra Melaina We have few corresponding sources to describe Greek land distribution. The most explicit source for the distribution of land in a Greek city is an inscription from the island of Lumbarda off the coast of Yugoslavia, probably the ancient Korkyra Melaina.27 The right-hand part of the inscription is missing, and its restoration has proved difficult, while interpreting the preserved remainder has not been without its difficulties as well. The text, as restored by Lombardo and others, reads:

Ἀγαθᾶι τύχαι· ἐφ’ ἱερομνάμονος Πραξιδάμου, Μα[χανέος, συνθήκα οἰκισ]τᾶν Ἰσσαίων καὶ Πύλλου καὶ τοῦ ὑοῦ Δάζου. Τάδε συ[νέγραψαν οἱ οἰκισταὶ ] καὶ ἔδοξε τῶι δάμωι· λαβεῖν ἐξαίρετον τοὺς πρώτους [καταλαβόντας τὰν χώ]ραν καὶ τειχίξαντας τὰν πόλιν τᾶς πόλιος οἰκόπ[εδον ἓν ἕκαστον τᾶς] 5 τετειχισμένας ἐξαίρετον σὺν τῶι μέρει, τᾶς δὲ ἔ[ξω αὐτᾶς λαβεῖν τοὺς αὐ]τοὺς καὶ τᾶς χώρας ἐξαίρετον τὸν πρῶτον κλᾶρον [τᾶς μὲν ἀρίστης ἕκαστον] πέλεθρα τρία, τᾶς δὲ ἄλλας τὰ μέρη· ἀναγραφῆμεν δὲ [- - - see below - - -] [π]εῖ ἕκαστος ἔλαχε· κατάμονον δὲ εἶμεν αὐτοῖς καὶ τ[οῖς ἐγγόνοις πέλε][θρο]ν καὶ ἥμισυ ἑκάστωι· λαβεῖν δὲ τοὺς ἐφέρποντας τᾶ[ς πόλιος οἰκόπεδον ἓν] 10 [καὶ τᾶ]ς ἀδιαιρέτου πέλεθρα τέσαρα καὶ ἥμισυ· τὰ[ς δὲ ἀρχὰς ὀμνύναι μηδ ][έποτ]. ε τὰν πόλιν μηδὲ τὰν χώραν ἄνδαιτον ποή[σεσθαι μηδαμῶς. εἰ δέ τί] [κα ἄρχω].ν προθῆι ἢ ἔτας συναγορήσηι πὰρ’ τ[ὰ ἐψαφισμένα, αὐτὸς ἄτιμος καὶ ] [τὰ ὑπάρ]χοντα δαμόσ[ια ἔστ]ω, ἀθῶιο.ς [δὲ ὁ ἀποκτείνας αὐτὸν - - - ] - - - - - - - - - - - - - - - - - -του - - - - - - - - - - - - - - - - 15 - - - - - - - - - - - - - - - - - -ΠΑΤΙ - - - - - - - - - - - - - - - - [- - - - - - - - - - - - - -εἴ κα τῶι δ]άμωι δό[ξηι - - - - - - - - - - ] [Ο]. ἵ.δ[ε] κατέλαβον τὰν χώρ[αν καὶ ἐτείχιξ ]αν τὰν πόλιν. [Δυμᾶ]νες [Ὺλλεῖς] Πάμφυλοι [Ἀρ]χ[έ]λαος Μεσοδάμου Ἡρα[κλείδας Θεοτ]ίμου Ὀνάσιμος Κεφάλου (There follows a list of 200–300 more names.) This may be translated as follows:

The Organization of Blocks

219

Good fortune. When Praxidamos was recorder (hieromnemon), (in the month of ) Ma[chaneus, an agreement of the oikis]ts of Issa and of Pyllos and his son Dazos. [The oikists wro]te this up and the people decreed it. The first (colonists) [who took possession of the la]nd and who fortified the city will take by choice: —of the fortified city, a choice house p[lot, one each,] together with his portion. —of the (land) o[utside it (i.e., the walled city), the same (colonists) will take] a choice ‘‘first allotment’’ of the territory, (consisting of ): —[of the best land], three plethra. —of the other (types of land), the portions. (The magistrates) shall write up [ - - - (see below) - - - a]s each (colonist) received by lot. One and a half plethra (of land) shall be inalienable, for them and [for their offspring. Those who come later shall receive: —of the [city, one house plot, and —o]f the undivided (land), four and one-half plethra. Th[e magistrates shall swear neve]r to make a redistribution of the city or of the territory [in any way. If a magistrat]e proposes or a private citizen advocates anything contrary to wh[at has been decreed, let him be deprived of civic rights, and his prop]erty confiscated, and [whoever kills him will go] unpunished . . . (two lines missing) [- - -if the p]eople dec[ree - - -] These people took possession of the coun[try and fortifi]ed the city: (Tribe of ) [Dyma]nes (Tribe of ) [Hylleis] (Tribe of ) Pamphyloi [Ar]ch[e]laos son Hera[kleidas son Onasimos son of Mesodamos of Theot]imos of Kephalos (List of at least 200–300 more names) The inscription records a treaty between the oikists (?) of Issa and two private citizens, Pyllos and his son Dazos. It decrees how the land in the new polis will be distributed, both to the original settlers and to those who come later; prohibits magistrates and private citizens from ever proposing a redistribution of land; records other provisions which are not preserved; and finally lists the original settlers. First, the word ἐξαίρετον, ‘‘choice, select, not assigned by lot’’ occupies a remarkably emphatic position at the very beginning of the text of the decree and is repeated three times: in line 3, ‘‘the first colonists . . . will take by choice,’’ again in line 5, ‘‘a choice house plot,’’ and in line 6, ‘‘a choice first-allotment of the territory.’’ Choice was therefore an important condition of the distribution of land and seems to be used in two senses as the English word is, both in the sense of 220

The Organization of Blocks

‘‘choice, excellent’’ and in the sense of ‘‘by choice, not assigned by random lot.’’ It modifies both the land which is being distributed, the best land to the first colonists, and the process of distribution, by choice rather than by chance.28 In lines 7–8, it is decreed to inscribe something, which has been broken from the right-hand side of the stone, ‘‘as each colonist was allotted’’ (ἔλαχε). The verb here, λαγχάνω, means ‘‘to receive by lot’’ and implies a process of random allotment, rather than of choice. Previous restorations of the missing words have assumed that the lottery was of parcels of land, for instance τὸν κλᾶρον καὶ τὰ μέρη, ‘‘the lot and the portions.’’ However, the inscription does not list the parcels of land, as it would if this were the fulfillment of the decree. Instead, it lists the names of settlers arranged by tribe, described as those who ‘‘took possession of the country and fortified the city.’’ Moreover, by most reckonings, the restoration [καὶ τὸν κλᾶρον καὶ τὰ μέρη] is too long, sixty-two letters rather than the average of fifty-five or so.29 Rather than restoring ‘‘parcels of land,’’ we can restore ἀναγραφῆμεν δὲ [τὰ ὀνόματα αὐτῶν ὁ π]εῖ ἕκαστος ἔλαχε, ‘‘to write up [their names a]s each was allotted.’’ 30 This gives a more reasonable line length of fifty-five letters and agrees with the general intent of the inscription, which was to list the names of citizens, presumably in the order that they were allotted. Moreover, it explains the triply emphatic ‘‘choice’’ in the earlier part of the inscription. As his name was drawn by lot, each citizen had his choice of a house plot in the city as well as agricultural land and so could decide where in the city he wanted to live. This system therefore mediates between those conflicting goals of fairness and order, by choosing names by random chance, ensuring that the process of distribution will be fair, but then allowing citizens to distribute themselves rationally through the city and countryside as they choose their plots of land. Citizens who wanted plots along the main avenues could choose them, assuming that they hadn’t been already taken. Others who wished to be near the agora could choose plots in that region. Citizens related by birth, origin, profession, or other such ties could select adjoining house plots, or a group of related citizens could select all the plots in one block and then build that block of houses to their own requirements and liking, working out problems of the number of stories, roofing, drainage, and the like as they saw fit. Such an organized, rational distribution of land, and the consequent distribution of citizens through different parts and blocks of the city in a systematic rather than a random manner, could lie behind the systematic distribution of different sorts of houses we see at Olynthus.

c on c lu s i on The houses at Olynthus are not, and never were, identical type houses. To approach them in a purely normative manner, describing the Olynthian house as The Organization of Blocks

221

if one could encapsulate an entire society’s habits in a single dwelling, misses not only much of the interesting variety among dwellings, but important patterns which reveal much about the city’s structure. Moreover, houses with different characteristics were found in different regions of the city; and the distribution of different architectural characteristics works on a number of levels. Some characteristics are distributed block by block; others seem associated with particular streets; others with different regions of the city, such as the area around the agora. Some of this patterning might reflect state policy, the deliberate creation of coherence between the social and physical organization of the city, as discussed in chapter 1. But it may also be due to a more rational system of distributing land, as suggested in the inscription from Korkyra Melaina. Rather than creating clear neighborhoods and districts in the city, this more fluid process would have allowed citizens to arrange themselves according to their individual situations, leading to a more organic fit between the social and physical structures of the city. Other distinctions are visible in the remains of households at Olynthus: in addition to the architectural differences between houses, there are clear differences in household economies. These are examined in the next chapter.

222

The Organization of Blocks

chapter six

The Economies of Olynthus

The debate over the nature of the ancient Greek economy has raged for more than a century, and it remains fundamentally undecided. Depending at least in part on the specific issue in question—maritime trade, banking, subsistence agriculture, exchange, etc.—recent discussions have proposed radically different models. Substantivists, for instance, argue that ‘‘the market seems to have played only a minor, peripheral role in the domestic economy of most Greek peasants’’ and more generally believe that issues we would classify as economic were embedded in social institutions, subordinated to other concerns, such as politics, social status, and independence and self-sufficiency.1 By contrast, a recent book on banking in classical Athens takes an opposing point of view, claiming, for instance, that in the fourth century, ‘‘Athenians functioned through a market process in which unrelated individuals . . . sought monetary profit through commercial exchange’’—a fundamentally disembedded system.2 Recent years have seen a spate of articles, books, and conferences devoted to various aspects of the Greek economy, but without necessarily bringing us closer to a resolution of the issue.3 The most interesting and intractable problems about the Greek economy are ethical: why the Greeks did what they did; how they felt about different occupations and economic systems; how they conceived of the relation between exchange value and use value; the embeddedness of their economy in social and political systems. These are questions which archaeological data from Olynthus do not directly answer. But Olynthus does offer new evidence for what ancient Greeks actually did, as opposed to what they say they did. Too often, debates on the ancient economy are founded solely on literary sources, such as Athenian legal speeches and philosophical dialogues. These ideologically charged and deliberately biased accounts do not try to give accurate accounts; they are inherently utopian and address cultural ideals rather than historical realities. Recent discussions 223

have therefore tended to focus on the accounts as dramatic self-presentations rather than as historical documents. Moreover, Greek literary sources generally concern only the upper strata of society, those involved in legal cases, mortgages of land, and the like. Archaeological evidence provides a useful check on literary approaches. And with its broad coverage of the city, with abundant evidence for household activities preserved in the destruction deposits, Olynthus again provides unique data for an entire urban spectrum, rather than for a select few. There are gaps, of course, in the evidence. The excavations did not focus attention on the by-products of household industry, such as organic or raw materials, slags, and other residues; nor on tools and other implements, nor on analyses of residues in the many installations and platforms in houses. The lack of stratigraphic control on the South Hill in particular makes it impossible to compare this part of the city with the grid-planned North Hill and Villa Section. But Olynthus nevertheless provides an unparalleled body of information which contributes substantially to our understanding of the economy—or economies—of Greece in the fourth century b.c. Perhaps the most important contribution that Olynthus makes to this debate is that the site documents the variety of economic strategies pursued by its households. The economy of Olynthus was not a monolithic entity; all its citizens did not participate equally in the same systems of production and exchange. The majority of the citizens of Olynthus, as at most Greek cities, were probably farmers, and many undoubtedly followed the risk-minimizing economic strategies characteristic of peasants in other cultures. But trade and industry are well attested at Olynthus in household workshops, shops and rooms for retail trade, and equipment for household industry. Some households probably engaged in both agriculture and domestic industry, some on a seasonal basis as suggested for the Greek pottery industry, others by employing slaves so that they could continue production year-round.4 Different systems of exchange were probably in simultaneous use as well. Barter was undoubtedly a widespread means of exchanging goods in this economic system, as Aristotle and others describe.5 By the fourth century, however, the market and money were important economic factors. Participation in a market economy is described in numerous literary sources from the fifth and fourth centuries and is amply attested by the thousands of bronze coins found in every house of Olynthus, as well as by shops and workshops that took up a great deal of space in certain houses.6 Interstate trade is likewise attested through imports and foreign coins. More interesting than the variety of economic strategies is their spatial distribution. Households in different parts of the city tended to follow different strategies and had different relationships with the monetary economy of the city. Households in the Villa Section seem to have stored large amounts of agricultural goods, but they did not engage in household industry. Households on the North 224

The Economies of Olynthus

Hill, on the other hand, frequently produced goods, such as textiles or processed agricultural products, for consumption outside the household, but they lacked the facilities for domestic storage that would have enabled them to maintain selfsufficiency in foodstuffs. In this chapter I shall consider different strategies by which households acquired, stored, and used resources at Olynthus; the articulations between these different economic systems in the city, and the degree to which different social groups had access to and participated in these different economies. As in the discussion of house plans and domestic assemblages, we shall explore the variety of economic strategies adopted at Olynthus and their spatial patterning within the city, the linkages between economy and society and their expression in the physical organization of the city, and we shall avoid, as much as possible, imposing a single theoretical framework or model on these data. No single model for past economic behavior seems to fit the whole range of strategies; we need to think in terms of alternative economic systems and strategies rather than a single Greek economy.

the a gr i c u lt u r a l e c on o m y Ancient Greece was primarily an agricultural society—most ancient historians will agree on this much. In general, Greek farmers seem to have aimed to be selfsufficient and to minimize risks rather than maximize production. A number of strategies helped achieve these goals, including planting a variety of crops over fragmented land holdings to avoid putting all their eggs in one basket, storage, and other buffering mechanisms.7 But, to discuss the agricultural economy of Olynthus is a somewhat frustrating exercise, something like trying to study oceanography in Wisconsin, far from where the real action is. Most agricultural activity naturally took place in the countryside, yet we know almost nothing about the occupation of the Olynthian chora.8 If the settlement pattern was dispersed, as it was at Metapontum, Tauric Chersonesos, and some other regions of Greece, a substantial proportion of the population would be archaeologically unknown; and even if the majority of the population lived in the urban center, there were probably significant buildings in the countryside, about which we know nothing.9 Some households held possessions in villages in the countryside, which were considered part of the household’s estate. A loan inscription found near house A 4 specifies as security for the loan, the estate (οἰκία) except for the ‘‘sevencouch room’’ (i.e., the andron), the property in the village (τὸ κτῆμα τὸ ἐπὶ τὴν κώμην), the storeroom (πιθεῶν), and another exception which is not completely preserved.10 Agricultural activities leave traces in the city center, however. Urban houseThe Economies of Olynthus

225

holds stored the produce of their farms, and storage containers are common at Olynthus. Olives and grapes and other produce were crushed and processed in the city as well as in the countryside, and such workshops are identifiable in a number of Olynthian houses. The sorts of processing and storage facilities in urban houses will help determine the agricultural strategies of those households. Household Storage One of the most important economic strategies employed by ancient farmers was to store enough food to supply the household for a considerable period of time. A number of ancient sources suggest that Greek peasants, like peasants in other cultures, aimed to keep a supply of food sufficient for a year or more on hand. This is supported by ethnographic studies of modern farmers in Greece and elsewhere.11 Such quantities of food took up a great deal of space and required specialized storage facilities to keep it dry and free of vermin over long periods of time. Foxhall and Forbes calculate the yearly cereal requirements of a household of six as about 1,419 kg of wheat, or about 1,838 liters.12 In his analysis, Gallant proposes a model of a smaller household which cyclically increases and decreases in size, composed in different phases of two to four adults, up to three children and two adolescents.13 Based on the caloric requirements of the household, he calculates the yearly amounts of grain, vegetables, and oil/wine needed at different stages of its cycle. The wheat alone would require from about 480 to more than 1,200 liters of storage.14 In addition to food, the family would also have to store seed grain for the next year’s sowing, as well as food for slaves, who are excluded from these estimates. Barley, which has a higher yield per hectare of farmland but provides fewer calories per kilogram, was proportionately more bulky to store. Olives formed an important part of the ancient diet and were an important crop in the Chalcidice.15 One of the pithoi in the Villa of Good Fortune seems to have been used to store these fruits, judging from the olive pits found in it, while crushing stones and press beds were found in a number of houses. The consumption of olives and olive oil in antiquity is difficult to estimate, but modern Greeks in Methana consume roughly 50 liters of oil per year per adult; Forbes and Foxhall conservatively estimate that an ancient household stored about 250–300 liters of oil.16 Moreover, olives produce biennially, so a household would need to store enough in a good year to cover the following lean year.17 The consumption of wine is also difficult to quantify and must have varied widely. Men in the Spartan syssitia were issued 8 choai of wine per month (about 24 liters), or 288 liters per year. Women and children probably drank rather less, so a household of four to six would probably need some hundreds of liters of wine per year.18 If a household made its own wine, it would need to be able to store a number of years’ worth while it aged. Considered in another fashion, a minimal family farm in ancient Greece prob226

The Economies of Olynthus

ably ranged in area from about two to six hectares, although many were much larger.19 While estimates of ancient agricultural yields are notoriously difficult, one hectare of land could produce perhaps 1,000 liters of wheat, more if the Chalcidice was as remarkably productive as ancient sources suggest.20 Again, barley would provide a still greater bulk. The yield of olive orchards in antiquity probably ranged widely; estimates range from 100 kg to 900 kg of oil per hectare.21 A household would need storage facilities sufficient to accommodate the maximum expected yield or risk losing a great deal in a bumper year. To maintain supplies to feed an average household for a year or more, then, and to store the produce of even a small family farm, would require some thousands of liters of storage capacity. Storage Containers Grain and other foodstuffs could have been stored in a variety of containers, both permanent, including pithoi and other pottery vessels, and perishable, sacks and baskets. The classic containers for storing agricultural goods were pithoi, many of which were found in the excavations at Olynthus. Olynthian pithoi seem to come in two basic sizes: a smaller type about 0.9–1.0 m high, with an average capacity of 100–350 liters, and a larger type about 1.7 to more than 2 m high, with a capacity of 1,000 liters and more.22 The two types are rarely found together in the same room, although they occur occasionally in the same house, probably for storing different kinds of foodstuffs. What was stored in these pithoi? The contents of three of the five pithoi in room j of the Villa of Good Fortune were carefully examined by J. Walter Graham. One, found buried up to its neck in the southwest corner of the room, contained olive pits. A second, whose base was found in the southwest corner of the room and was probably set into a shallow pit there, contained ‘‘traces of straw—evidently from some kind of grain.’’ A third, whose base was found upside down in the east part of the room, contained pieces of pine bark fused into lumps, perhaps used to flavor wine to make retsina.23 Other than this, however, the contents of storage jars at Olynthus were not recorded. We are therefore forced to look to comparative archaeological, literary, and ethnographic data to understand what was stored in pithoi. Although Classical Greek texts almost unanimously describe only wet storage in pithoi, primarily wine but also olive oil, honey, and the like, there is archaeological evidence that grain and other dry goods were stored in pithoi, and most modern scholars have concluded that they were used to store both liquid and dry foodstuffs.24 Pithoi at Olynthus were found both buried below floor level and resting or broken on the floor. One of the five pithoi in the Villa of Good Fortune was buried up to its neck; two more were found toppled and so had been set above floor level; and the situation of the last two was unclear, but since their bottoms were not preserved, they too were probably above floor level.25 The storeroom of the House The Economies of Olynthus

227

of Many Colors held at least four pithoi. The bases of two of these were found in situ, suggesting that they were sunk below the floor, while the other two were probably above the floor. The three pithoi in the House of the Twin Erotes and those in A viii 2 and other houses were found shattered on the floor, while pithoi in houses A 8 and elsewhere were found neatly buried below the floor. The pithos in the storeroom (g) of house A vii 4 was buried, while that in the shop (h) of the same house was shattered on the floor. Of six pithoi in the central shop of house B vi 9, two were sunk into bedrock in the southwest corner, and the other four seem to have been above ground level. This distinction may reflect what was stored in these pithoi. A pithos sunk into the floor would have been cooler, less liable to break, and perhaps more convenient to fill and empty than one set above ground level, but it was also more subject to damp and therefore mold and insects. It would therefore be suitable for storing oil, wine, or other liquids, rather than for storing grain which must be kept dry. Of course, a pithos set above ground could also be used for liquid storage, but the converse is probably less true. This is certainly not a hard-and-fast rule, as grain could be and was stored in underground pits and pithoi; but in general, households probably installed their pithoi differently depending on what they intended to store in them. Where more than one pithos is preserved, they tend to be placed differently—some above ground, some below—and were probably used to store different foodstuffs accordingly. When a house had only one vessel, however, it might have served either for wet or dry storage. Pithoi were also the most expensive pottery vessels in the household: preserved prices for the large type range from about 31 to 53 dr—the price of a whole house in a neighboring town.26 Because they were so valuable, pithoi were often salvaged after the destruction, leaving pits in the floor if they had been set below ground level, or leaving just their lids or no trace at all if above ground. Disks of terracotta or slate, ranging in diameter from about 0.24 m to more than 1 m, were found in fourteen houses, sometimes in groups of two or more. Although some of these disks were probably used for other purposes, such as lids for other large vessels or as working surfaces, many were probably the lids of pithoi which were either shattered and not recorded in excavation or had been salvaged after the destruction.27 These may therefore be considered as possible evidence for pithoi. Storing agricultural produce for household consumption was not the only use for pithoi. Some were set in courtyards to collect water from the eaves. Others were used to hold olive oil or grape juice from presses or in other agricultural processing (see below, ‘‘Agricultural Processing’’). Because these were not used primarily for household storage they should be considered part of a different economic system. Others were found in rooms which seem to be shops or workshops and so were probably for nondomestic storage (see below, ‘‘Shops’’).28 We need to look at the contexts of pithoi and other storage containers to determine how they were used. 228

The Economies of Olynthus

Greek households used other types of pottery vessels besides pithoi to store foodstuffs. Although these are relatively secure from pests and damp, they are all much smaller than pithoi. The sipye was used to store barley meal and other grains.29 The vessels identified as sipyai at the Athenian Agora are much smaller than pithoi: the largest has a volume of roughly 120 liters, only one-eighth the capacity of the smallest of the five pithoi in the Villa of Good Fortune, although comparable to the smallest of the small variety of Olynthian pithos.30 They are significantly cheaper, too—3½ to 5 obols in the Attic Stelai, one-fiftieth the cost of a large pithos. Storage and transport amphoras, originally used for wine and other products, were ubiquitous in antiquity, and they were found throughout Olynthus. These too could have been reused to hold grain and other foodstuffs. The largest varieties hold only some 70 liters, and most types are in the 15–25 liter range; the capacity of the Olynthian examples is not known. A household would need a very large number of such smaller storage containers to hold a significant supply of foodstuffs. Because they were not recorded or mended systematically, large, coarse pots are under-represented in the notebooks and final publications. There is no indication, however, that any house at Olynthus had enough coarse pots, other than pithoi, to store grain in the quantities needed to maintain self-sufficiency for a year. Two houses had up to fifteen storage amphoras, amounting to perhaps 350– 400 liters of storage, the equivalent of one or two small pithoi.31 Quantities of coarse pottery were noted in a few houses but not collected or mended in the excavations, so it is difficult to reconstruct or quantify. Baskets, cloth sacks, goatskin bags for wine, pits, and unsealed storerooms were commonly used to store foodstuffs, leaving hardly a trace in the archaeological record. As long as the room remains absolutely dry, grain kept in sacks or baskets will keep for an extended length of time; Roman granaries at Ostia and elsewhere probably stored grain in sacks rather than sealed vessels.32 Although Gallant concludes that ‘‘physical storage of agricultural produce could not provide either a reliable or an extensive buffer against crop yield variability, primarily because of losses due to deterioration while in storage,’’ modern Greek and other peasants do successfully keep stores on hand for many years, and ancient authors discuss steps that can be taken to minimize losses.33 The lack of permanent storage facilities in the archaeological record does not imply, therefore, that houses did not store food. But as we will see, there are clear patterns in storage strategies which demand explanation. Storerooms Households at Olynthus employed a variety of storage strategies. Some houses had single-purpose, large-capacity storerooms which were probably used exclusively to store foodstuffs. Other houses seem to have stored foodstuffs in smaller quantities, usually in rooms which were also used for other purposes: kitchens, The Economies of Olynthus

229

storerooms which also held other household equipment, and the like. Still other houses seem to have few or no storage facilities. These households either stored food outside the house or relied on the market or other sources of food. There were also storerooms capable of holding very large quantities of crops, more than a household could use. These are found in houses which engaged in processing agricultural produce, and although they could have been used to store household supplies as well, most of the goods they held would probably have been consumed outside the household. As outlined above, a house would need some thousands of liters of storage facilities to maintain self-sufficiency and to store the produce of a typical family farm. Because grain and other dry foodstuffs could be stored in perishable containers like sacks, which would leave little or no trace, it will be difficult to distinguish some storerooms without paleobotanical research, which was not conducted in excavations at Olynthus or elsewhere fifty years ago. But some storerooms are clearly identifiable. The storerooms of the Villa of Good Fortune and the House of Many Colors (m), described above, had five and four large pithoi, respectively, and that of the Villa of the Bronzes (g) contained one very large pithos and possibly fragments of others. Although McDonald and others have suggested that the Villa must have been a hostel or some other nondomestic building, in part because they thought this storage capacity excessive for a single household, in fact this is appropriate for the yearly requirements of a household. Storerooms with more than one large pithos are listed in Table 6 (see fig. 48). Most of these rooms shared common architectural features. Most were located in the southern half of the house—perhaps because the North Rooms, being better lit, were more appropriate for working spaces; perhaps because there were larger areas available in this half of the house because the constraints of the roofline were not as critical. In about half these houses, the storeroom was the largest closed room in the house, averaging 29 m2. They were all undecorated and have earth floors, except for the storeroom in the South Villa, whose walls were painted; this room, on the north side of the house, was perhaps originally intended for another purpose. The storerooms in the House of Many Colors, the House of the Twin Erotes, the House of the Tiled Prothyron, and the Villa of Good Fortune were dug below ground level, although one might expect this to lead to problems with dampness.34 Most of the storerooms were located on the side of the house facing onto a north-south avenue. The storerooms in the Villa of Good Fortune and the Villa of the Bronzes (in its original phase) had exterior doors, but in neither case do these doors seem to open onto the street; rather they may have opened to gardens or open spaces belonging to the house (below, ‘‘Gardens’’). This additional access would have made it easier to bring stores into the room. Most seem to be single-purpose rooms, without significant assemblages of other artifacts, evidence of other work, or storage of other household implements.35 A number had wooden furniture or fixtures in them, however, perhaps the remains 230

The Economies of Olynthus

Houses with Large-Scale Storage 0

Houses with multiple pithoi preserved: b

a

c

d

e

a b

f

g

h

i

j m

k

4 or more large pithoi

l

c

d

e

f

h

g

5 pithoi; total capacity ca. 7500 liters.

j

k

l

i

m

n

Villa of Good Fortune

House of Many Colors

a

20 m

c

a

b

c

d

b

d

e

f

e

j f

g

h

i

1 very large pithos with lid

3 large pithoi, 1 with price inscription: 44 dr.

g

h

j

i

Villa of the Bronzes

House of the Twin Erotes

Houses with multiple pithos lids preserved: 3 pithos lids; other artifacts

f

4 pithos lids, much coarse pottery (pithoi?)

4 pithos lids b

a

a

c

e a

d

b

c

d

f f

c

g f

d

e h

f

k

g

h

i

i

j

j

e

h

i

j

k

k l

l

House of the Tiled Prothyron

j

g

House of the Comedian

figure 48. Houses with Large-Scale Storerooms

South Villa

m

table 6: Specialized Domestic Storerooms House

Room

Notes

H. Many Colors

m

H. Twin Erotes

j

V. Bronzes

g

V. Good Fortune

j

storeroom; 4 or more large pithoi, one with price inscription: 53 dr 2 obols storeroom; 3 large pithoi, one with price inscription: at least 44 dr; another pithos in storeroom d storeroom; 1 very large pithos with lid, diameter = 1.7 m; others in fragments? another, small pithos in room c storeroom; 5 or more large pithoi, the smallest holding ca. 1,000 liters; total capacity ca. 7,500 liters

And, less certainly: H.Tiled Pro. H. Comedian S. Villa

c b a

subterranean room with 3 pithos lids 3 pithos lids in North Room 4 pithos lids, stacked up in corner of North Room; much coarse pottery nearby: remains of pithoi?

of shelves, bins, or chests for storing other kinds of foodstuffs.36 The storeroom in the Villa of Good Fortune, for instance, had extensive remains of a wooden structure along the north side of the room—across from the pithoi which were in the southwest and east part of the room. With three to five large pithoi, each holding on the order of 1,000 liters, these rooms would provide enough storage for a year’s supply of grain and other foodstuffs and so agree with the model of self-sufficient, household-based storage of produce and foodstuffs. Buildings on the South Hill also had large numbers of pithoi. Trench 5/6, for instance, incorporated an area of some 45 × 50 m.—roughly the area of four normal-sized houses in the later area of the city—and preserved some thirteen intact pithoi, often in groups of two or three. The houses in Sections F, G, J, and K on the South Hill also contained numbers of pithoi, The dates of these pithoi are uncertain, however. A number of levels from Archaic through the mid-fourth century were apparently mixed, either in excavation or by postdepositional processes. For this reason I have not included them in the distribution plans, but they do attest large-scale household storage in other areas of the city. Small-Scale Household Storage Most houses at Olynthus, however, lacked such storage rooms. Many houses could store a smaller quantity of food, in a single small pithos rather than a number of larger ones.Typical of these smaller-scale storage rooms is that in A vii 4 (g), the house that Graham and others have used as the Olynthian type house (fig. 22). This room measures about 3.5 × 3.5 meters —about one-quarter the size of the storeroom in the Villa of Good Fortune. Its single pithos held perhaps 190 liters, 232

The Economies of Olynthus

sufficient to hold wheat for a household of six for just over a month. The storeroom in the Villa of Good Fortune had about forty times the storage capacity of this house. Examples of such small-scale storerooms are listed in table 7. This group of rooms is architecturally much more diverse than the larger-scale storage rooms. At least four are kitchens, and three more of the less certain examples may be kitchens as well. They range in size from relatively small rooms of 9–18 m2, to rooms as large as the large-scale storage rooms. Like the largescale storage rooms, they are mostly undecorated. There is evidence for activities other than storage in many of these rooms, including weaving, food preparation, washing, and the storage of domestic equipment. Large- Versus Small-Scale Storage The investment of space and resources in large-scale storage rooms, capable of keeping thousands of liters of foodstuffs, suggests that those households aimed to store a year or more’s supply of foodstuffs within the house to ensure selfsufficiency. But the archaeological evidence shows that not all households at Olynthus, indeed not even the majority of households, followed this strategy. The most surprising feature of household storage at Olynthus is immediately obvious from the distribution plan (fig. 49). Every house with large-scale domestic storage facilities is located in the Villa Section. Of the nine houses excavated in this area, four have very clear storerooms set aside for large-scale storage, three more probably had such rooms, although the pithoi have been removed, and the last two houses are too poorly preserved to tell. By contrast, although many houses on the North Hill have pithoi, those used primarily for domestic storage (as opposed to household industry) are small in number and capacity—one or at most two small pithoi holding hundreds, rather than thousands, of liters. No house on the North Hill seems to have pithoi sufficient to store its household food supplies for more than a short period of time. How to explain this striking pattern? A number of possible explanations can be considered and rejected. The difference in storerooms is unlikely to be the result of more extensive looting by Philip’s troops, of differential preservation or postdepositional processes, or of less careful excavation. Pithoi were found and recorded in both areas, but in different household contexts: in large-scale storerooms for food in the Villa Section, but in small-scale storerooms, shops, courtyards where they collected rainwater, or in rooms associated with agricultural processing on the North Hill. There is no reason to think that households on the North Hill were less wealthy than those in the Villa Section and thus less able to afford expensive pithoi. The houses on the North Hill sold for much more than those in the Villa Section and were fitted with mosaic floors, painted walls, and other luxurious and presumably expensive features. Indeed, as we will see, households on the North Hill seem to have more available monetary wealth than those in the Villa Section. Nor is there reason to think that the households on the North Hill simply preThe Economies of Olynthus

233

table 7: Small-Scale Domestic Storage Rooms House

Room

Notes

A8

a

A iv 7

e

Av1 Av3 A vi 2 A vii 1 A vii 2

e k b i b

A vii 4 A vii 7

g b

A vii 10 A viii 2

b d

A viii 9 B ii 1 B vi 8 B vi 9 Dv6

c c b c b

Dv6

f

H.Twin Erotes V.Bronzes

d c

kitchen; 1 small pithos embedded in floor, capacity ca. 110 liters 1 pithos, broken on floor. ‘‘Pastas’’ of half-house, joined to A iv 5, perhaps nondomestic 1 pithos, broken on floor; pastas room, like A vii 4, g fragments of pithos rim, in back room of irregular house kitchen; pithos fragments in NE part of room large South Room; 1 pithos, diameter of rim 0.84 m 1 pithos; basin in SW corner; misc. finds; perhaps involved in agricultural processing 1 small pithos and lid, capacity ca. 190 liters; pastas room 1 jar or pithos, diameter of body 0.9 m; irregular house, perhaps involved in agricultural processing large 4-handled jar (pithos?); kitchen kitchen? fragments of storage jars or pithoi on floor of room 1 small pithos and lid embedded in floor in SE corner 1 pithos embedded in floor; other finds not recorded 2 small pithoi embedded in floor 1 small pithos on floor; possibly a shop 1 small pithos on floor, with incised decoration (Olynthus 13, no. 1119) pithos base embedded in floor, with lid; weaving, other finds 1 small pithos embedded in floor; storage amphora 1 small pithos in SE corner

And, less certainly: A3 A6 A 10 A 11 A 11 A iv 5

b g a e h m

A iv 9

i

A vi 9 A vi 10 A vi 10

a d i

1 pithos lid; 1 broken ‘‘jar’’ 1 pithos lid; pastas of house 2 small terracotta lids 2 pithos lids, 2 storage amphoras, other vases and artifacts pithos lid, 2 storage amphoras, other artifacts 1 pithos lid, many pots (black-glazed and coarse), terracottas, etc.—‘‘cupboard’’ 1 pithos lid, many other finds—domestic storage in back room 1 terracotta pithos lid, few other finds; kitchen 1 pithos lid; pastas room 1 pithos lid, too small for stone mortar in room; room used for agricultural processing?

table 7: Continued House

Room

A vii 4 A viii 5 A xi 10

f i e

B vi 4 B vi 4 B xi 1 Dv6 ESH 4 V. Good Fortune

Notes

1 pithos lid in pastas; many other finds 1 pithos lid; 3 terracotta altars, weights 1 pithos lid found with 1–2 storage amphoras and other vases; pastas of house d 1 pithos lid g pithos sherds in SW portion of house—exact room uncertain, but probably from kitchen g andron 1 pithos lid in andron of house; rest of house not excavated d 1 pithos lid together with many other objects: vases, 2 storage amphoras, louter fragments, loomweights, etc. h 1 pithos lid, mortar, grindstones, 39 loomweights, misc. finds: kitchen? b 1 pithos lid, few other finds (cf. room j); kitchen

ferred different types of storage containers, such as bags or baskets, which were less archaeologically visible than pithoi. While all these households undoubtedly used a variety of containers for their foodstuffs, the absence of pithoi for largescale household storage on the North Hill demands a better explanation. Instead, the distinction in storage facilities between the North Hill and the Villa Section probably reflects differing economic strategies, which will be discussed at the end of this chapter. Communal Storage Many cities must have had facilities for communal storage, although there is little archaeological evidence for them. The Athenian grain-tax law of 374/3 calls for the buyer of the tax to ‘‘heap up the grain in the Aiakeion. The state will make available the Aiakeion in watertight condition and provided with a door.’’ Stroud identifies the Aiakeion as the Rectangular Peribolos in the Agora, a rectangular enclosure 26.5 × 31 m. Roofed, provided with a door and internal divisions, and filled to a depth of 2 m or more with grain, it could have held the quantities of grain specified in the law.37 Subterranean granaries (σιροί ) are attested at the sanctuary at Eleusis, but these are obviously a special case.38 Some sixty bell-shaped pits dug into the streets of the South Hill were identified by Robinson as granaries, but if any were used for grain, it was probably only in the special circumstances surrounding the Persian invasion. There are, however, a number of rooms probably intended for large-scale storage of agricultural produce on the North Hill. Graham suggested that they were ‘‘probably unroofed pens for animals, or possibly were used for storage,’’ but they are associated with

The Economies of Olynthus

235

agricultural processing, rather than purely household storage. They are therefore discussed below (‘‘House A 4’’). Gardens Most Classical cities were densely built up, leaving little room in or between houses for gardens of the type known, for instance, in Roman houses of Campania. Urban gardens in Classical Greece, therefore, are exceedingly rare.39 At Olynthus, the excavated portions of the North Hill revealed no empty spaces which could have been used for gardens, although small areas of courtyards might have been planted. However, as described above, there were considerable open spaces between some houses in the Villa Section. None were excavated extensively, and the trial trenches do not record planting pits or other evidence of use. While some of these spaces might be unclaimed or unbuilt house plots, others might have been used as gardens belonging to adjoining houses. Although this represents a very different approach to the urban planning from the layout of the North Hill, it could also represent a very different economic strategy pursued by these households, related to the difference in storage strategies described above.

h o u s e h ol d i n d u s tr y Ancient Greek writers frequently disdained industry, manufacture, retail trade, and the like, claiming for instance that they are denounced and, quite rightly, held in very low esteem by states. For they utterly ruin the bodies of those who work at them and those of their supervisors, by forcing them to lead a sedentary life and to stay indoors, and some of them even to spend the whole day by the fire. When their bodies become effeminate, their souls too become much weaker. Furthermore, the so-called ‘‘banausic’’ occupations leave a man no spare time to be concerned about his friends and city. Consequently such men seem to treat their friends badly and to defend their countries badly too. In fact, in some cities, especially in those reputed to excel in war, none of the citizens is permitted to work at the banausic occupations. (Xenophon, Economics 4.2, tr. Pomeroy) Aristotle and Plato both forbade citizens of their ideal states to work at industry or retail trade, and Aristotle mentions that at Thebes, ‘‘there was a law that no one who had not kept away from the agora for the last ten years might be admitted to office.’’ 40 In Athens there was even a law against reproaching a male or female citizen for the trade he or she practiced in the agora—which just goes to show figure 49. Distribution of Large-Scale Domestic Storerooms 236

The Economies of Olynthus

Domestic Storage Large-scale domestic storage (>1 large pithos) Large-scale storage, less certain (>1 lid only) Small-scale storage (1 small pithos) Small-scale, less certain (1 lid only) non-domestic pithoi (in shops, etc.) and pithoi for collecting rainwater omitted 0

Agora

pithoi on South Hill omitted

50

100 m

that this sort of insult and slander happened frequently.41 Corinth, the famous mercantile city, was noted for the fact that merchants were less despised there than elsewhere.42 These prescriptions, like Xenophon’s description of the house of Ischomachos, reflect an aristocratic, philosophical tradition and wishful thinking rather than contemporary reality. Literary, epigraphic, and archaeological evidence from Athens and other cities attests a wide variety of ‘‘banausic’’ occupations among citizens, metics, and slaves; and we find significant evidence in Olynthian houses for household industry, manufacture, trade, and other ‘‘illiberal’’ occupations. A considerable number of houses devote a relatively large proportion of their household space to such nondomestic production, and the artifacts found in some houses suggest that the household included a substantial number of slaves. Households at Olynthus engaged in large-scale processing of agricultural produce, weaving, cooking and baking, masonry, manufacture, and many other trades. One difficulty faced by the archaeologist is that many of these are essentially domestic tasks—weaving, grinding grain, or processing foodstuffs—but carried out on a larger scale than they were when undertaken solely for household consumption. Because these household tasks writ large were performed in domestic spaces, it is sometimes difficult to distinguish household industry intended for sale or exchange from household production intended for internal use. When we find unusual quantities of loomweights, grindstones, or other household equipment, do we infer that the household was producing cloth, foodstuffs, or other materials for exchange outside the household, or is this simply a large household producing for its own use? Agricultural Processing Most households ground grain and processed vegetables and other foods for their own consumption with hand mills, mortars, and other such implements. But processing other agricultural produce, such as olives and grapes, requires special and expensive facilities which most households could not afford.43 Some of this work was probably done in the country rather than in the city. Lease and sales inscriptions from other parts of Greece mention mills, storehouses, and other such equipment, and scenes of vintage in vase paintings are normally placed in the rural countryside.44 It was often more efficient to process olives and grapes near the orchards and vineyards rather than transport the unprocessed food into the city. But urban agricultural processing facilities are attested at Olynthus, Halieis, and other Greek sites.45 For an average household, the investment in this equipment, which can process much more than a single family’s holdings, would not have been worthwhile. The owners of such equipment processed oil for a large number of households.46 This is, for instance, the implication of the anecdote about the sixth-century philoso238

The Economies of Olynthus

pher Thales of Miletus, whose understanding of natural history led him to predict a bumper crop of olives one year. He therefore optioned all the olive presses in Miletus and Chios; when the crop ripened, he was able to rent the presses out at a profit and so prove that philosophers could make a fortune, if only they wanted to.47 The anecdote does not tell us whether the presses were in the town or the country; but even if most of them were in the country, as seems plausible, it shows that this kind of processing was not simply a household activity, restricted to the crops produced on a single estate, but served a wider group of households. Some fifteen houses on the North Hill have facilities for processing agricultural produce—a relatively large percentage of the 108-odd houses excavated at the site (plate 4). Many of these facilities were extensively salvaged after the destruction, and even the best preserved are missing essential pieces of equipment. Crushing and grinding stones, which were imported and expensive, difficult to break or burn in the destruction of the city, and large enough to be obvious in the ruins, would have been easy to remove and reuse, and the rest of the equipment for processing olives was multipurpose, relatively portable, and simple in nature, making it easy to remove and difficult to identify even if left behind.48 For example, of the five upper stones of olive crushers (orbes) found at Olynthus, three were found reused in the walls of house A v 9 (a house with no other evidence for agricultural processing); and no lower stones (mortaria) were found: all were probably salvaged.49 Moreover, in the late summer when the city was destroyed by Philip, the olive harvest would have still been on the tree, and processing equipment may not have been set up. In no case do we have the complete equipment for crushing, pressing, and collecting olive oil or wine; we have only parts of the assemblages. But even from an incomplete assemblage we can infer that these households were processing agricultural products within the city, and get a sense of how this fit into the domestic and urban economies (see table 8). House A xi 10 Among the clearest examples of houses with special facilities for agricultural processing is house A xi 10 (fig. 50).50 This house was somewhat irregular in plan, divided along the usual axes and also divided from north to south, leaving a long corridor down the center of the house, and with an unusual hierarchical circulation pattern. A double door led into a wide entrance corridor, allowing carts to enter the house and unload produce. In the southeast corner room of the house was a cement platform with gaps along the south wall. This probably supported a press bed, the gaps to hold the upright beams of a press. The press bed and collecting vessel were not found; they were probably salvaged, together with the press weights and other usable equipment. Two smallish pithoi (or phidaknai?) in the room, with capacities of 150 and 340 liters, must have been used to store the pressed oil or juice. Two more pithoi were found in the pastas (e), and another in the court (h); others may well have been salvaged after the destruction. Since The Economies of Olynthus

239

table 8: Houses Engaging in Agricultural Processing House

Notes

Crushing olives A6 olive crusher (orbis) in courtyard; 12 upper grindstones A v 10 olive crusher (orbis) in courtyard; sales inscription includes ὁ πιθεών, ‘‘the pithos-room’’ and τ. ὰ. μι . [σθ]οφόρα πάντα ‘‘all the things which bring income’’ [A v 9] [3 orbes reused in foundations] Pressing olives or grapes A1 A9 A vi 8 A vi 10 A vii 9 A viii 10 A xi 10 B ii 1

stone press bed in court press in court: boulders with cutting for upright beam; other boulders as weights two cement platforms (press beds or grape treading platforms?) in g; storeroom a room k: cement platform with gap—press? Stone mortar in room i. cement platform in room b; storeroom a room a with cement floor; pithos and lid nearby press in room j; cement floor for crushing; pithoi for storage cement press? floor; pithos in next room

Grinding grain A4

problematic rotary millstone 0.51 m in diameter. Installations in NW, S, SE, SW of court for millstone and other facilities? Large storeroom with 15 pithos lids. 12 upper grindstones; also olive crusher

A6 Uncertain A3 A vii 3 A vii 8 A viii 6

foundation in court. includes irregular structure built before rest of block, with odd basin in SW large storeroom h; perhaps linked to suite of rooms e-f. No finds from those rooms. installation robbed out of court; two drains leading to street; large storeroom d. Very few finds from house.

these were found in work areas of the house, rather than in a storeroom, they were probably used in the extraction of oil or wine rather than for household storage. Another room in the house (g) was probably also used as a workroom; it was equipped with a cement floor, a large drain leading out into the street, and a gap in the northeast corner from which a tub or other installation has probably been robbed out. No crusher was found in this house; either the press was used for 240

The Economies of Olynthus

57 loomweights; 2 TC figurines in deposit; coin of Alexander III b

a

coin of Alexander III in topsoil

c

e

d

f

g

deposit of vases & terracottas h

j

i

2 pithoi, 1 in situ, buried to neck; other on its side

M

spit?

group of 6 coins near wall, including 3 post-348 BC

cement floor for press bed?

0

5m

figure 50. House A xi 10 grapes, crushed perhaps by rollers on the cement floor of room g or else in troughs which have not survived; or perhaps an olive crusher was set up but was salvaged after the destruction, together with the stone elements of the press. This building was not solely an industrial establishment, however. It had a kitchen-complex to the north, in which were found terracottas, vases, and fiftyseven loomweights. The pastas contained more household terracottas and vases, and a fine bronze brazier was buried under the floor in the corner of the press room, presumably to prevent its being looted.51 This then was a house occupied by a household; but a quarter to half of its total area was taken up by special-purpose rooms devoted to household industry. House A 6 An olive crusher and more than twelve grindstones were found in the courtyard of house A 6 (figs. 51, 52). Like A xi 10 and some other houses which were engaged in agricultural processing, this house had a double entrance, with a wider door for The Economies of Olynthus

241

d

a

e stone trough

g f

robbed-out collecting basin?

h

c

Avenue A

Fortification Wall

limit of excavation

b

i j

press bed?

marble female statue head in courtyard

orbis

cistern original south limit of house

slopes

k

l

m

7 hopper-rubbers, 5 saddle querns, lower grindstones

n

0

door to shops?

5m

figure 51. House A 6 carts and a narrower one for foot traffic.52 This led into a paved entranceway and from there into the court. This house is half again as large as normal, because it acquired the northern half of house A 7. The courtyard of this house was unusually large, paved with cobbles but with a cement section about 2 m wide along its western edge. This cement platform rose slightly at its edges, and a wooden (?) rim was probably set into the narrow gap between the cement and cobble pavements, forming a shallow trough draining off to the north.53 This cement strip might have served as a press or crushing bed. The north end of the cement was broken off in a roughly curved edge, per242

The Economies of Olynthus

[To view this image, refer to the print version of this title.]

figure 52. Courtyard of House A 6 with Orbis from Olive Crusher in Situ. Olynthus 2, fig. 184 haps by the removal of some feature like a collecting basin. Part of a rough stone basin or trough with a spout was found in the northeast room of the house (e); this or a similar receptacle could have served to collect the runoff.54 To the south of this cement floor was another, slightly lower cement-paved surface which sloped down steeply to the west. This was where the alley which originally separated house A 6 from A 7 was located, and it may have continued to drain water from the courtyard. Towards the southern end of the press bed was a cistern, fed by a pipe just under the floor which presumably collected water from the eaves. This seems an odd location for a cistern, in the middle of a press bed; its mouth was deliberately raised above the level of the cement by the addition of cut stone blocks, though, so that oil would not flow into the cistern. It may be that the cement floor was a later addition to the house, after the cistern had already been dug, and levels were adjusted to accommodate it. In addition to the pastas on the north side of the court, a portico opened onto its west side, adjoining the cement bed. This was left unexcavated, but it would have been a convenient area to work. On the north side of the house were two small rooms, one with a cement floor, opening onto an unexcavated but probably larger room (a). This was probably a The Economies of Olynthus

243

kitchen-complex, with the usual cement-paved bathroom (b, whose floor had a catch basin and gap left where a bathtub was removed) and flue (c). Foxhall suggests that room b was used as a press room, but in view of its similarity to other bathrooms and its location in the suite of rooms a-c, I would prefer to identify it as a bathroom and place the press bed adjoining the courtyard.55 In addition to pressing olive oil, this household engaged in large-scale grinding of grain. Twelve upper grindstones were found in the courtyard: five of the more sophisticated hopper-rubber type and seven of the older saddle quern types, with a boat-shaped upper stone and a more convex lower stone. ‘‘Several small flat rectangular stones’’ mentioned in the publication must be the lower stones belonging with these, but they were not considered interesting enough to warrant full publication.56 The house was excavated during the first season in 1928, and the finds only sporadically recorded. But other artifacts perhaps associated with agricultural processing were found as well: a large pottery basin with two handles and a spout in the southeast room (n), an elliptical basin in the corner of the andron (h), and a slate pithos lid or table top found in the pastas. This house probably belonged to a miller, as Robinson initially supposed. It is interesting that a professional establishment like this used relatively large numbers of the older saddle querns rather than using the more efficient hopper-rubbers; almost half of the saddle querns from the site come from this one house. Such a large number of grindstones and other processing equipment implies a substantial number of people to work them, and this household may have included a dozen or more slaves to operate the grinding equipment (below, ‘‘Slaves and the Household’’). As so commonly in Greek industry, this ‘‘professional’’ establishment used exactly the same technologies—hand-driven saddle querns and hopper-rubber grindstones—as other households did for everyday use; rather than employing larger or more efficient grindstones, they simply increased the number of grindstones by employing slaves. Houses A vi 8, A vi 10, and A vii 9 A number of other houses had treading or press floors for crushing grapes or olives. These installations were found in A vi 8 and A vii 9, and perhaps in A vi 10, A viii 10, B ii 1, and other houses.57 House A vi 8 had two such platforms in a large room in the southern part of the house (g, fig. 53). One, in the southwest, probably had wooden sides and was drained to the north into a catchbasin, whose capacity was estimated as 110 liters.58 A small drain through the house wall on the south was probably for cleaning after use. A second cement bed was located in the southeast corner of the room; the catchbasin for this press bed was not preserved. This house is unusual in that it contained few or no domestic areas. A large room occupied the entire western third of the house, which was probably a storeroom for agricultural produce (below, ‘‘House A 4’’). When excavated, this was essentially empty. The room in its southeast corner had been taken over by its 244

The Economies of Olynthus

House A vi 8 b

c

d

3 balls or weights

coin hoard

e

a

b

c

stairbase?

spatula pendant; finger ring

TC object

d

f

f

e finger ring; fibula; bronze decorations/ furniture fittings

i

g

j

frags of hardware

i catchbasin

iron hoe

g

h

k

h louter? mended

drain

Street vi

0

5 m

figure 53. Houses A vi 8 and A vi 10 neighbor, A vi 10, for use as an andron. The house had two stairways to the second floor, one in room g and the other in room e, suggesting that the upper floor was divided in use as the lower one was. None of the other rooms contained many domestic implements, only a few vases and figurines in the pastas; but it was in use when the city was destroyed: a hoard of silver coins found in room d included some of the latest issues of Chalcidic silver coins.59 This house may have been purchased and converted to workshop and storage space by the occupants of A vi 10 next door. House A vi 10 also contained equipment for household industry, probably agricultural processing. A cement platform with a rectangular hole in it occupied much of the room in its southeast corner (k). A low foundation or bench separated the room with the cement platform from the room to its north (j). In the next room to the west (i) was a stone mortar or olive-crushing basin, buried to its rim in the floor, with a pithos lid and an iron hoe nearby.60 This suite of rooms was probably used for crushing and processing agricultural products: the cement platform might have been part of a press, the gap left where a stone press bed was robbed out. It is certainly not a typical domestic installation. Another unusual feature of A vi 10 was the external stair in room c, which could have led to an independently owned or inhabited second story over this house. These two houses The Economies of Olynthus

245

Avenue B

a

House A vi 10

pose a complex situation with divisions of space among different households and activities. In the next block to the north, house A vii 9 had a single cement platform or press bed in room b, with a slot cut out of the southern edge for drainage into a jar to the south (fig. 35). Such platforms were probably grape treading platforms, although they could have been used as primitive olive crushers, by rolling a heavy cylindrical stone over the fruits.61 The other facilities in these houses do not help us identify what was being processed. Like A vi 8, house A vii 9 had no obvious courtyard and only scanty domestic space. The large room in the west (a) may have been another storeroom for agricultural produce, like the one in A vi 8; it too was virtually empty, except for two bosses, a lamp, and two coins. The eastern three rooms were shops; the southern shop (j) was built before the rest of the block and perhaps not even owned or used by the same household. The northern shop (d) contained a grinding emplacement, a large basin, perhaps for storing foodstuffs, two inscribed lead weights, eleven vases, and ten coins. The weights and relatively large number of coins suggest commercial activity here, perhaps selling the wine or oil produced in the press. The shop to its south formed a back room to d and contained a number of bronze implements whose use is uncertain. Finally, shop j contained a completely different assemblage: two portable altars, a fragment of a louter base, five bronze bosses from a chest or piece of furniture, a round bronze vessel or container and swinging handles from three other bronze vessels, a plaque depicting a cow suckling a calf, a fibula, a skyphos, a lekythos, and twelve coins. These rooms, although open to the street and not to the house, might have served domestic as well as more public functions as shops because they contained a wide variety of utensils for preparing food, household cult, and the like; their assemblages are unique among the shops at Olynthus. House A 4 The upper and lower stones of an unusual stone object were found in the courtyard of house A 4. The object is problematic: the description and photograph in the publication are indistinct, the field notes from this first year of excavation are sketchy and do not mention it, and the stones themselves are lost; but it is likely a rotary mill or trapetum orbis. A number of scholars have doubted either its identification or context because rotary mills are otherwise not found in the Mediterranean until the late Hellenistic or Roman period. Moritz wonders whether it may not be some other sort of object, such as a door pivot, although from Robinson’s description this seems very unlikely. Runnels, who has done the most thorough study of Greek grinding implements, suggests that it is a later intrusion, perhaps from a Roman villa rustica on the site, since if it were Classical in date, it would be anomalously early.62 246

The Economies of Olynthus

table 9: Large-Scale Agricultural Storerooms House

Room

A4 A vi 8 A vii 3 A vii 8 A vii 9 A viii 6

b a d h a d

Notes 15 slate and terracotta pithos lids no finds miscellaneous objects, seated terracotta female, etc. many bronze objects: vessel handles, etc. a few, miscellaneous finds no finds

However, there is no evidence here for settlement or even occasional use in the late Hellenistic or Roman periods. The absence of clear parallels at other Classical sites is disturbing, but early examples exist in western Europe, and the technology might have been adopted in Greece, at least sporadically, earlier than has been suspected. Arguments from silence—either of comparanda to the rotary mill or of evidence for later occupation of this area—are dangerous, of course. The agricultural processing assemblages from Olynthus, however, demonstrate just how thoroughly these objects were salvaged and reused, leaving few if any traces. Not a single lower stone from an olive crusher was found at the site, for instance, but these devices were clearly used before the destruction, since three upper stones were built into the walls of house A v 9. The absence of comparanda to this rotary millstone therefore may be less problematic than it appears. The court of this house has a circular hole towards its southern end and a larger, rectangular gap in the northwest; both have very neat sides and must be where some feature has been removed or disintegrated. Many terracotta figurines were found fallen against the south wall of the courtyard. The northern third of the house was taken up by a large storeroom, similar to those in A vi 8, A vii 9, and other houses (fig. 7). This room contained fifteen slate and terracotta pithos lids.63 Probably the pithoi belonging to these lids were salvaged after the destruction or shattered and not recorded in excavation. A number of similar rooms are found in other houses in the North Hill. These rooms are about 4–5 m wide and often take up the entire depth of the house, occupying about one-third of its area. They are listed in table 9. Robinson and Graham suggested that these rooms were ‘‘probably unroofed pens for animals, or possibly were used for storage,’’ but there is no indication that any of them were unroofed.64 They are not so wide that they could not have been spanned by roofbeams, and none is cobbled or equipped with a drain; they were in all likelihood covered. Except for the room in house A 4, most were found virtually empty or had only a scattering of miscellaneous artifacts. Most belong to houses which have some other evidence for agricultural processing. These rooms might therefore have been used to store large quantities of olives, grapes, grain, The Economies of Olynthus

247

and other produce, either before or after processing. These households may not only have processed foodstuffs, but also purchased it at harvest time and stored it for later sale on the market. They provide our best evidence for communal storage at Olynthus at a household level. We therefore see a number of different arrangements for processing olives, grapes, and grain in urban houses. Some houses, such as A xi 10, A 6, and A vi 10, have both domestic and industrial areas in the same space, remaining essentially domestic structures. Other buildings are essentially nondomestic in character, such as A vii 9 and A vi 8. They may have been owned or used by neighboring households or perhaps by households further away. Moreover, the households which engaged in agricultural processing seem to have been unusually successful. A number of them acquired part or all of neighboring plots of land. The owners of A 6, for instance, acquired half of A 7 and became one of the grander houses at Olynthus, with a cut-stone facade, cobbled and cement-paved courtyard, and a great deal of worked stone in the architecture. The owners of house A vi 10 took at least the southeast room of A vi 8 for an andron and perhaps owned the whole house. The relation between wealth, trade, and industry will be discussed further below. Cooking and Baking One baking establishment can be identified, in house A viii 8 (fig. 54). Room f of this house contained a large stone mortar and a large stone trough with two basins, probably for kneading or mixing dough. Many fragments of coarse vases, probably belonging to basins, lekanai, and storage bins for flour and the like, were found with the mortar and trough, and also a ‘‘chimney tile’’ with an elliptical opening to let in light and let out smoke. In the other corner of the house was a unique complex of three rooms (h, i, and j). The largest room, j, was cobbled; its floor was covered with ash, particularly in the southeast corner, and contained large amounts of kitchen pottery. The cobble floor, ash, and kitchen pottery resemble a domestic flue for cooking, but the room is significantly larger and does not adjoin a normal kitchen-complex.65 Like a domestic flue, though, this room was probably used for baking and cooking. Greek bread was baked in ovens or barrel ovens, whose broken remains were probably included in the ‘‘kitchen pottery’’ which littered the floor. Like the modern tabun, barrel-ovens were open on top and bottom, and the fire would have been built on the floor of the room, leaving the ash and traces of burning.66 Only one grindstone was found in the house, hardly enough to supply a bakery. The presence of the stone mortar suggests that something was being ground here, but flour may also have been purchased from a miller like those living in house A 4 or A 6. The house is also notable for the number of altars it contained: a built altar was located in the corner of the pastas (a unique placement), and nearby was 248

The Economies of Olynthus

a

b

c

d

red-figured krater

other vases

e

f

h

g

i red-figured krater

j mortar & trough; coarse vases

k

door? ash; coarse pottery: baking area?

Street viii

0

5 m

figure 54. House A viii 8 a portable altar, while room a contained two more portable altars. The association of bread-making equipment and portable altars is also noted in the House of the Tiled Prothyron, which contained five grindstones and seven alters, six portable and one built (above, chapter 3). Many of the rooms in the southern part of house D v 6 were probably also used for large-scale preparation of food, as described above. As these rooms opened to the street as ‘‘shops,’’ it is likely that this house also engaged in preparing food for consumption outside the household. House ESH 4 had two kitchen-complexes, one of which was equipped with a large mortar and was entered directly from the street (room h, fig. 33). The southern half of the house may have been used in part for cooking for outside the household or some related commercial endeavor. A cement-floored workroom (j) and a shop (l) adjoin the southern kitchen, with their own entrances separate from the main entrance to the northern half of the house, which opens from Avenue F. The house has an unusual amount of food-processing equipment including one lower and four upper grindstones and the stone mortar. The southern kitchen was The Economies of Olynthus

249

not used exclusively for nonhousehold food preparation, however: it had an adjoining bathroom with a bathtub—as did the northern kitchen—and thirty-nine loomweights in the southwest corner. A relatively large number of beads and other jewelry were found in ESH 4 as well, suggesting more than the usual number of women in the household: collections of beads were found in room a and the pastas (f ). Finally, three hoards of silver coins were found in the house. Two of these, found in rooms a and b of the house, were apparently closed in about 379 b.c., in the war with Sparta; the third was probably closed in 348 b.c.67 Fishing Living close to the Aegean, Olynthians would have naturally engaged in fishing, and the archaeological evidence documents it throughout the site. Fishhooks were found in small numbers throughout the city. They are found more commonly on the South Hill than on the North Hill or the Villa Section, and more commonly in streets, where they were presumably thrown out or lost, than in houses with the destruction assemblages.68 A cache of 121 lead net weights was found in the west portico (k) of house A v 6, together with a leaden D-shaped object (perhaps also used in netting), three large sea shells, a long bronze rod, two small bronze rings, and a few vases.69 This grand house, which took over part of its neighbor A v 8, had two stories and one of the few androns on the block. The household seems to have engaged in fishing, although whether this was purely for household consumption or more professionally is unclear. Probably most fishing equipment was kept near the coast, though, rather than in the city.70 Textile Manufacture Just as most households ground their own grain, it was expected that the household would weave much of its own cloth; this was a major occupation of ancient women (above, chapter 3, ‘‘Weaving’’). But a number of houses show evidence of weaving on a large scale, more than would be needed for household use; and these households probably wove cloth for sale on the market. The four rooms adjoining the court of house A v 9 each contained looms, which were probably set up and in operation when the city was destroyed (fig. 25). Four looms is probably more than a single household would normally employ, and some of the cloth woven here was probably meant for use or sale outside the household. The greater part of the well-lit space of the house was used for weaving. House A viii 7/9 also seems to have been manufacturing textiles on a very large scale (fig. 55). These two houses were joined by opening the wall between their 250

The Economies of Olynthus

Street ix a

b

c

a

b

c

3 bronze rods w. knife-marks

50 loomweights

d

d

Avenue B

247 loomweights

location?

e

h M

M M

e

f

House A viii 7

g

neolithic figurine

g

f

i

House A viii 9 0

5 m

figure 55. House A viii 7/9

two pastades, so that the pastas formed one long corridor stretching the length of the two houses. In the pastas of A viii 7 were found 247 loomweights, enough for half a dozen to a dozen looms. These were found in a hoard in the northwest corner of the pastas, probably stored in a bag or chest there and so not in use at the moment of destruction. Another 50 loomweights were found in the nearby North Room (b). The situation of these loomweights was not recorded, whether found in a hoard or scattered as if fallen from a loom. The pastas of A viii 7 also contained a variety of domestic artifacts: five storage amphoras; seven plain and black-glazed vases; at least four thymiateria, three terracotta female masks, and two plastic vases; a fibula; a lead astragalos; and various objects of bronze, lead, and stone. Interestingly, two very early stone artifacts were found here as well: a stone axe or celt and a marble Neolithic female figurine.71 Except for the court and pastas, however, most of the other rooms of the house were relatively bare. Interestingly, although the two houses were connected, not a single loomweight was found in A viii 9, and relatively few domestic artifacts were recorded in the fieldbook. The significance of this is unclear: could the eastern part of this double house (A viii 9) have been quarters for slaves who worked the looms in the western portion? The issue will be discussed further below. House A iv 9 also had enough loomweights (133 in all) for at least three looms, The Economies of Olynthus

251

although only one seems to have been set up when the house was destroyed (fig. 24). Houses A v 10 and A vii 9 each contained a total of 88 loomweights, about the same as house A v 9 (which had 85), but the rooms in which these were found were not noted.72 These households also may have woven textiles on a larger scale, perhaps seasonally, if not at the moment the city was destroyed. In many of these houses the uniformity of the loomweights is noticeable. Whereas most domestic looms were equipped with a mixture of different types of loomweights, these houses seem to select with greater care. Nearly all the loomweights from the four looms in A v 9 were conical, as were nearly all the weights from house A iv 9 whose type was noted. The 297 loomweights from house A viii 7 were both conical and pyramidal but were closely matched in weight, and an unusual number were stamped, the conical ones with three circles ( ), the pyramidal ones with a cross ( or ).73 The weights were unusually small, ranging in weight from seventeen to twenty drachms, about half the average weight of loomweights from purely domestic contexts. They were perhaps meant for weaving finer textiles. The loomweights from A v 10 and A vii 9 were not described in the fieldbooks, unfortunately. This uniformity distinguishes these ‘‘professional’’ looms from domestic looms which often employed a rather mixed lot of weights as described above.74 Sculpture and Stonecutting Two houses in Row A engaged in stonecutting. In the court of house A 5 were found eight portable altars, some of them stuccoed but the rest unfinished, two finished and one unfinished anthemion, a relief depicting two griffins, an unfinished stele, five louter bases, and other worked blocks. The altars were in matched pairs, and portable altars are found in pairs in a number of Olynthian houses.75 The notes from the first year of excavation, when this house was dug, do not record the finds from houses A 1 to A 8 separately; however, no obvious mason’s tools were recorded among the metal objects from the whole trench. This house had an unusual design, with a very large andron (the largest at Olynthus, measuring 6.1 × 6.1 m), a large cement-paved courtyard with no pastas, and poorly preserved rooms on the east. The western part of the house was not excavated and may have included more domestic or work rooms. Although no domestic artifacts are recorded as coming from this house in particular, the presence of the large andron suggests that this was not simply a workshop but a dwelling. Nearby, the household in house A 10 seems to have engaged in different sorts of stoneworking (see above, chapter 3, ‘‘House A 10’’). In the court of this house were found five stone capitals which do not seem to belong to the architecture of the house (in addition to more which probably belong to the house); three portable altars, at least one of them unfinished; a rectangular block which was probably an unworked altar; and other stone objects. Room h contained more: three unfin252

The Economies of Olynthus

ished stelai, two Doric capitals, and other worked blocks. A stone tripod base (?) was found in room e. Probably not coincidentally, these houses are quite near to one another. They seem to have specialized, however, in different types of stone artifacts: the owner of house A 5 in stelai, altars, and louters, the owner of A 10 in architectural elements. Notably absent is any evidence for carving grindstones, olive crushers, or other such implements. These objects, of imported andesite or basalt, were probably either worked in the quarries where the stone was extracted or perhaps elsewhere on the site.76 Manufacture Manufacture in Classical cities was generally a small-scale affair, and the Olynthian houses offer some evidence for its organization.77 Coroplasty A number of households were involved in making terracotta figurines. In the southeastern room of house B i 5, east of the agora, were found at least thirteen figurine molds (fig. 56).78 These include two molds of Cybele figures, a female protome, three female figures in various postures (one nude, two clothed), two female heads, an Eros, an actor, a Silenus, a kneeling boy, a plaque showing a horseman, and a monkey. The same molds could be used to make both figurines and plastic vases; a plastic vase made from one of these molds was found in house A 9.79 The house was fairly regular in plan, with pastas, andron (unusually in the center of the house, like the andron of house A 10), and a court with a cistern. Opening onto the court is a rather large exedra (room h), with one pillar base still in situ; this would have made a particularly good work site. No finds are recorded from the exedra itself, but from adjoining rooms came the molds (from i), as well as three marble basins, one with a spout and three lug handles, one perhaps a louter with a square hole in the middle, and one described as a ‘‘millstone’’ with a concave profile and a hole in its center (two from room g in the southwest corner of the house; the provenience of the third not recorded). These might have been used for separating and working clay. The pastas had more domestic sorts of artifacts: a plate, five storage amphoras, a bronze spatula, five loomweights, three terracotta figurines, and a plastic vase. B i 5 was partly excavated in 1928 when finds were not always separated house by house, and other finds from this house were probably inventoried with the general finds from Trench 7. There is no trace of a kiln in this or any house at Olynthus. Kilns may have been built as needed outside the city, and the unfired figurines brought to the kiln after drying. Kilns have been found in urban houses at other sites, however, especially in western Greece at sites such as Herakleia in Lucania and Locri Epizephyrii.80 The Economies of Olynthus

253

Street ii a

b

c

lim

it o

fe

xc

av a

tio

n

d

17 coins

f

e 8 lead “loomweights”

h

i

g cistern stone mortarium?

13 figurine molds

0

5 m

figure 56. House B i 5 Other Olynthian houses also contained figurine molds, although fewer in number. Two each were found in houses A iv 9 and A v 10, not far from B i 5. Both these houses were involved in domestic industry of different sorts, as described above; but there is little evidence for coroplasty other than the two molds. They may therefore be strays. Three more molds were found in house B vi 6. The architecture of this house was very poorly preserved. In the southeast corner were two pits cut into the bedrock; the three molds were found in the westernmost pit.81 These pits may have been left when pithoi were robbed out from the room, but they may also have served some industrial function. Single figurine molds were also found in houses A 3, A 8, the House of the Twin Erotes, and in trial trenches. Although a local red-figured pottery industry has been identified and its products studied, no potters’ workshops have been identified at the site.82 But the equipment of a potter’s workshop is insubstantial and unlikely to be preserved. Potter’s wheels, for instance, are almost unknown in the archaeological record, perhaps because they were made of wood. It is not impossible that households in B i 5 and elsewhere made vases as well, which were fired outside the house.83 254

The Economies of Olynthus

Street -ix a

b

a

b

c

d

e

h

i

‘urinal’

d

c

“impluvium”

e

f

many fragments of lead

f g

4 plastic vases

g

stone trough; many fragments of lead

j

k

M

l

many fragments of lead

p m

h

n

slingbullet mold

q

o

r

s

t i

u

v

House C -x 5

House C -x 7 0

5 m

figure 57. House C −x 5 and C −x 7 The rather casual excavations of 1928, when B i 5 was dug, might well have missed evidence for pottery production. Weapons Lead slingbullets were manufactured in house C −x 5 (fig. 57). This house was quite irregular in plan, and its court had a sunken area in its center surrounded by ‘‘heavy blocks of poros,’’ which Robinson referred to as an ‘‘impluvium.’’ 84 The house had been thoroughly ransacked and its stratigraphy was very disturbed, but a stone mold for slingbullets was found in the court, and rooms e, f, and g contained many amorphous fragments of lead, raw material for making slingbullets.85 Two shallow stone troughs, one in the court and the other in the adjoining area (f ) might have been used in the manufacturing process as well. The neighboring house (C −x 7) is also notably irregular in plan, and although both contained domestic materials in some abundance, it is possible that this region of the city, close to the outskirts, was a more industrial quarter. House B vi 10 was also poorly preserved, but in one room (c) was found part of a bronze mold for making arrowheads. There is no other evidence of bronze The Economies of Olynthus

255

casting in this house, however, and this may be a stray find rather than evidence of industry here. A Clubhouse and a Hostel? House A v 8, at least in its final phase of use, was not a domestic dwelling but was a sort of clubhouse perhaps for hetaireiai or political drinking clubs, or perhaps for ritual meals (above, chapter 5, ‘‘Cluster 3’’). Part of this house plot had been taken over by house A v 6 when that house expanded. After this encroachment by its neighbor, A v 8 consisted of two androns, one of the usual seven-couch size, entered from an anteroom, the other larger and less well preserved, but also with a raised border; a small court; an entrance room; and a ‘‘porter’s room’’ adjoining the entrance (f ). There is no evidence of a second story. Although the house was relatively deeply buried under 0.4–0.75 m of earth, it contained relatively few finds, no loomweights, few vases, and little other domestic equipment. Its court, however, contained a fragment of a marble table top (?), an unusual object which might be the remains of fancier furniture than that found in most houses. Since it has no evidence, either architectural or artifactual, of domestic activities, and has two androns but no other types of rooms, this seems to be a special-purpose building used primarily for eating and drinking, either in a more social or a more ritual context.86 William McDonald suggested that the Villa of Good Fortune was not a house, but a hostel or pandokeion.87 He adduced a number of arguments in favor of his identification. He interpreted the inscriptions and symbols in the mosaics of rooms e and f as referring to particular throws in games of chance and appeals to Lady Luck; the ‘‘wheel of fortune’’ might be another such symbol, or it might be used as a playing field in such games. McDonald concluded that ‘‘these two rooms were set aside primarily for games of chance; and it seems very unlikely that the owner of a private house would sacrifice so much space for this particular purpose.’’ 88 McDonald also calculated that the storage capacity of room j was in excess of what would be needed by even a large family, but would be appropriate for a hostel. He argued from the lack of a bathroom on the first story, the second entrance to the house near the kitchen, and its large size and location near the outskirts of the city that the building would be more appropriate for lodging and dining than as a private dwelling. None of these arguments seems conclusive to me. The size of the building is paralleled in the House of the Comedian, the South Villa, houses A v 6, A 6, and others, all of which are domestic in character. The arrangement of the kitchen and lack of a bath is not unparalleled in Olynthian houses, and as we have seen, the storage capacity is not excessive for a well-off agricultural household. The house may well not be near the edge of the city, since the Villa Section might extend as much as 200 m further east. And while his acute interpretations of the symbolism 256

The Economies of Olynthus

a

b

c

d

e

lease inscription

f h g

i

j

m

hearth? with ashes

l k

many ashes; burned bone; cement basins at N

n

o

Street viii 0

5 m

figure 58. House A viii 3 of the mosaics may well be correct, this does not necessarily imply that the rooms were used solely for gaming (probably as unlikely in a hostel as in a house), nor that the building was public. I therefore believe that the villa is a particularly large and well-built, but nevertheless private dwelling. Uncertain Uses A number of houses have installations or assemblages which seem out of the ordinary line of domestic equipment but whose significance is difficult to interpret. They may have been used in some form of household industry. A viii 3 is irregular in plan and may have been extensively remodeled. In the southeast corner of the house is an unusual complex of rooms with a cement platform (?), two cement basins, and what seems to be a sunken internal area (rooms j, k, l, and o; fig. 58). Robinson interpreted it as ‘‘only a variation on the Olynthian [oecus] unit,’’ i.e., a kitchen-complex; but it bears little resemblance to those rooms.89 The feature Robinson identified as a bathtub is certainly something else. A cement basin in the shape of a quarter-circle is preserved today in room j, not a bathtub, and it was probably one of a pair set below the floor level.90 Room k, The Economies of Olynthus

257

which Robinson identified as the flue, is too large for a normal flue, and its floor level was sunk below the surrounding floors. The walls were carefully built with stackwork masonry. It contained a great deal of ash and charred animal bones, as well as at least two terracottas, eight table vases, including an odd ‘‘composite vase,’’ a lamp, and a bronze Athenian dikast’s ticket. Although it is hard to determine what the area was used for, it does not seem to be simply domestic in nature. A large installation was robbed from or disintegrated in the northwest corner of the court in house A −1. This installation was drained (or filled?) by a pipe set below the pavement of the court, which led to a basin (a settling basin?) in the southwest corner of the house. The installation is unlikely to have been a cistern, which would have been deeper; it might have been a stone or wooden tank set partly below floor level. Again, not enough is preserved to identify what this installation might have been used for; one might guess at fulling or dyeing.91 Other large installations were apparently robbed from the courts and porticoes of houses A vi 3, A vi 5, A viii 2, and the House of the Tiled Prothyron, leaving rectangular gaps in their pavements. A vi 5 had a peculiar heavily cobbled room or space in the southeast corner of the court (h) with a stone drain running across it to the alley. This area may have been open to the court on the west, adjacent to the robbed-out installation. The cobbled area and drain suggest some out-ofthe-ordinary activity here. The installation in A viii 2 collected water through a pipe leading from the north, under the floors of rooms c, d, and f.92 This pipe was probably fed from a gutter along the north side of the roof, through a downspout which has, of course, long vanished; the feature in the court may also have collected rainwater from the inner eaves through conduits now lost. The installation was not deep enough to be a cistern, however, and its interpretation is uncertain. South of the court of house A v 4 was a patch of cement pavement with slots or narrow foundations, ending on the west in a ragged edge as if some feature had been removed. The peculiar exedra (j) opened onto this feature and also had a door to the street, with a well-cobbled area of pavement in front of it. The second entrance and pavement distinguish this area from ordinary domestic spaces and suggest a more public or economic use which required its own door to the street. This might have been the remains of an installation used in agricultural processing, akin to the presses and treading floors in other houses. Houses A 11 and A v 1 had large numbers of storage amphoras in many rooms (above, chapter 3). A 11 had fourteen amphoras, A v 1 fifteen, mostly in the northern rooms of the house. Two rooms in the southern part of A v 1 (f and j) each contained many amorphous lumps of lead, whose significance is also mysterious. There is no suggestion of wine making in either house, and it is unclear what these quantities of amphoras would have been used for, whether for purely domestic storage or for some other activity. 258

The Economies of Olynthus

a

b upper & lower grindstones

Later house division

Original house division

entrance?

sales inscription; various small bronze artifacts; many storage amphoras & vases, TC frags, marble statuette

a

c

b

c

d

many

f

d

e

e

f

g

hoard of 11 flans; earring

many vases

fragmentary steps? later access

g

h

h

j

k j

i

j

k

ashes

i

many storage amphoras

1 silver, 6 bronze coins

l

l

m

n

many vases

bronze objects

m k

hardware & misc. iron objects; many vases

ashes hardware; marble statuette

sherds

wall of stoa

0

base of lebes gamikos; mesomphalic phiale frag.; 2 thymiateria; alabastron

figure 59. Houses A iv 5 and A iv 7 Minting or Counterfeiting? Hoards of bronze flans were found in two houses, both close to the agora. A dish containing about thirty flans was found in house B ii 6, partly excavated in 1928. The room in which the flans were found had a concave cement floor, but little else is known about it or the rest of the house or the context.93 Another hoard of flans was found nearby in house A iv 5/7. A iv 7 in its final phase had been divided and extensively modified. The western half of the house, in which the flans were found, was apparently sold to the owner of A iv 5, who connected the two by means of a stairway (fig. 59). An inscription documenting the sale for 2,000 dr was found in or just outside the house (below, ‘‘Sales and Prices of Houses’’). The Economies of Olynthus

259

5 m

Eleven bronze flans were found in a vase in room e of A iv 7; another flan was found nearby, and two more in the street in front of the room, probably scattered in the destruction, and one flan was found in house A iv 5, room m, reinforcing the connection between these two dwellings.94 Room a is an unusual space, apparently open to the street, or closed only by an insubstantial partition, and open to the house as well. Although open to the outside, it contained a large collection of artifacts in addition to the hoard of flans: at least eight more or less complete and many more fragmentary vases, including a relief lekythos, a fragment of a red-figured open vase, and a thymiaterion or bowl on a stand; many storage amphoras, some with stamped handles; eight terracotta figurines; two bronze vessel handles; many fragments of iron; a strainer; a fishhook; and twelve conical loomweights. Some of these artifacts are fragmentary and might have been discarded here, but a number of the vases are essentially complete, and the deposit is not likely to be purely trash but rather part of the destruction assemblage. The discovery of hoards of flans in private houses rather than a public building is somewhat unexpected. Robinson originally identified both houses as mints, but on reconsideration decided that ‘‘the suggestion that A iv 7 was a mint is scarcely tenable since none of the rooms in the house contained any evidence of more than ordinary household activities.’’ 95 But two such hoards contained in vessels make the identification of these objects as flans very probable. No distinct traces of a furnace for heating the flans before striking was found in A iv 7, but many ashes were found on the floor of room h, and perhaps these mark the remains of a furnace. The house was buried under only 0.4 m of fill, and features like furnaces might not have been preserved. We know very little about how bronze coins were minted and brought into circulation. It is possible that the task of minting at least the less important bronze coinages was farmed out to individuals, much as taxes and other matters dealing with state revenues could be farmed out.96 If so, these houses could belong to citizens who had bid for and received contracts to mint a certain number of bronze coins. However, most of the flans from A iv 7 are larger than even the largest denomination of Chalcidic coins. They measure between 17 and 18 mm in diameter, as opposed to 14–16 mm for genuine coins; and they would have been flattened and their diameters increased when struck. This difference in size makes it unlikely therefore that these flans were blanks for minting Chalcidic bronze coins.97 Instead, these flans might be blanks for counterfeiting coins, a rather common practice in antiquity.98 Two ancient forgeries of Chalcidic silver coins were known to Clement, both made from silver-plated bronze flans; one was found next door to this house, in A iv 9.99 It is perhaps more likely, however, that the forger would not risk counterfeiting local currency but would instead produce less familiar foreign types. One counterfeit Athenian tetradrachm, for instance, was found at Olyn260

The Economies of Olynthus

thus.100 The flans are slightly too large for counterfeiting Chalcidic tetrobols and too small for tetradrachms; they may have been intended, therefore, to be used for a foreign currency. Slaves and the Household From the archaeological remains alone, it is difficult to tell whether work was done by the citizen owners of a house or by slaves. But this would obviously be a critical distinction: a small-time mason or weaver working with his own hands (or the hands of his family) would be the object of scorn, at least to the aristocrats whose writings and opinions survive. A citizen who owned slaves who bring in income, however, would be a very different story. In Xenophon’s Memorabilia, Aristarchus complains to Socrates: ‘‘I am in great distress. Since the revolution there has been an exodus to the Piraeus, and a crowd of my women-folk, being left behind, are come to me,—sisters, nieces and cousins,—so that we are fourteen in the house without counting the slaves. We get nothing from our land, because our enemies have seized it, and nothing from our house property, now there are so few residents in the city. Portable property finds no buyers, and it’s quite impossible to borrow money anywhere: I really think a search in the street would have better result than an application for a loan. It’s hard, Socrates, to let one’s people die, but impossible to keep so many in times like these.’’ When Socrates heard this, he asked: ‘‘How is it that with so many mouths to feed Ceramon not only contrives to provide for the needs of himself and his family, but actually saves enough to make him a rich man, whereas you, with so many mouths to feed, fear you will all be starved to death?’’ ‘‘The explanation, of course, is this: my dependants are gentlefolk, his are slaves.’’ ‘‘And which do you think are the better, his slaves or your gentlefolk?’’ ‘‘My gentlefolk, I think.’’ ‘‘Then is it not disgraceful that you with your gentlefolk should be in distress, while he is kept in affluence by his meaner household?’’ ‘‘Of course his dependants are artisans, while mine have had a liberal education.’’ ‘‘What is an artisan? one who knows how to produce something useful?’’ ‘‘Certainly.’’ ‘‘Are groats useful?’’ ‘‘Yes, very.’’ ‘‘And bread?’’ ‘‘No less so.’’ ‘‘What about men’s and women’s cloaks, shirts, capes, smocks?’’ The Economies of Olynthus

261

‘‘Yes, all these things too are very useful.’’ ‘‘Then don’t the members of your household know how to make any of these?’’ ‘‘I believe they can make all of them.’’ ‘‘Don’t you know, then, that by manufacturing one of these commodities, namely groats, Nausicydes keeps not only himself and his family, but large herds of swine and cattle as well, and has so much to spare that he often undertakes costly public duties; that Cyrebus feeds his whole family well and lives in luxury by baking bread, Demeas of Collytus by making capes, Menon by making cloaks; and most of the Megarians make a good living out of smocks?’’ ‘‘Yes, of course; for they buy foreign slaves and can force them to make what is convenient, but my household is made up of gentlefolk and relations.’’ ‘‘And so, just because they are gentlefolk and related to you, you think they should do nothing but eat and sleep? Do you find that other gentlefolk who live this sort of life are better off and happier than those who are usefully employed in work that they understand? Or is it your experience that idleness and carelessness help men to learn what they ought to know and remember what they learn, to make themselves healthy and strong, and to get and keep things that are of practical use, but industry and carefulness are useless things? When these women learned the work that you say they understand, did they regard it as of no practical use, and had they no intention of taking it up, or did they mean to occupy themselves in it and obtain some benefit from it? Which makes men more prudent, idleness or useful employment? Which makes men more just, work or idle discussions about supplies? Besides, at present, I fancy, you don’t love these ladies and they don’t love you: you think they are a tax on you, and they see that you feel them to be a burden. And the danger in this state of things is that dislike may grow and their former gratitude fade away; but if you exert your authority and make them work, you will love them, when you find that they are profitable to you, and they will be fond of you, when they feel that you are pleased with them. Both you and they will like to recall past kindnesses and will strengthen the feeling of gratitude that these engender; thus you will be better friends and feel more at home. ‘‘To be sure, if they were going to do something disgraceful, death would be a better fate. But in point of fact the work they understand is, as it appears, the work considered the most honourable and the most suitable for a woman; and the work that is understood is always done with the greatest ease, speed, pride and pleasure. So do not hesitate to offer them work that will yield a return both to you and to them, and probably they will welcome your proposal.’’ 262

The Economies of Olynthus

‘‘Well, well,’’ said Aristarchus, ‘‘your advice seems so good, Socrates, that I think I shall now bring myself to borrow capital to make a start. Hitherto I have had no inclination to do so, knowing that when I had spent the loan I should not have the wherewithal to repay it.’’ The consequence was that capital was provided and wool purchased. The women worked during dinner and only stopped at the supper hour. There were happy instead of gloomy faces: suspicious glances were exchanged for pleasant smiles. They loved him as a guardian and he liked them because they were useful. Finally Aristarchus came to Socrates and told him this with delight. ‘‘One objection they have to me,’’ he added: ‘‘I am the only member of the household who eats the bread of idleness.’’ (Xenophon, Memorabilia 2.7) Despite Socrates’ unusual attitude, the biases against working for sale outside the house are obvious here as elsewhere. The prevalence and importance of slavery remain one of the most important unresolved issues in Greek social history. Literary evidence from Athens suggests that much of the labor in workshops there was done by slaves, and the same was probably true at Olynthus.101 Lists of agricultural equipment in the Attic Stelai likewise suggest that substantial numbers of slaves were doing much of the work on larger estates as well.102 Some Olynthian houses contained more equipment for domestic industry than could have been put to use by a standard Greek household.The dozen upper grindstones and the olive processing equipment found in house A 6, for instance, must have been operated by a substantial number of laborers, presumably at least a dozen. The 297 loomweights found in house A viii 7/9—sufficient to outfit half a dozen to a dozen looms—likewise required more adult women to operate than a normal household would have. Although these households might have included relatives and a larger number of free women than usual, it seems more likely that they included slave women who operated the looms and grinding equipment. This would have been both more socially acceptable and more reliable over a long period of time than relying solely on the free women of the household. Both of these houses were larger than normal, providing space for the additional work force, whether slave or free. ‘‘The things that bring income’’ listed in the sales inscription from A v 10 may well have included not only the productive machinery (the olive press and other equipment), but also the slaves to operate it (above, chapter 3, ‘‘House A v 10’’).103 In other cases, it is all but impossible to quantify how many workers engaged in household industry, or whether the work was done by slaves, hired but free workers, or by the free household itself. And agricultural slaves probably lived in the country and so are similarly archaeologically invisible. In general, it seems most likely that households which engaged in domestic industry did so by emThe Economies of Olynthus

263

ploying slave labor, either in addition to free and hired, or exclusively, leaving the free household able to tend to the farmland and other, more morally acceptable occupations. It is also nearly impossible to distinguish where slaves ate, slept, or how they interacted with the household from the archaeological record.104 Literary sources do not distinguish between slave and free quarters of the house as they distinguish between andronitis and gynaikonitis. The usual picture is that slaves inhabited the same parts of the house as free members of the household, working side by side with them on occasion. The reality, of course, may have been quite different. A more detailed record of coarseware, perishable objects, and other artifacts might help us in this regard. But similar problems bedevil the study of Roman household slaves, which are much better documented in literary and epigraphic sources than are Greek.105 Since we cannot even locate the sleeping places of the free Greek household with certainty, the details of slaves’ daily lives also remain clouded for now. Like the equipment for processing agricultural produce, slaves could be an expensive investment, especially skilled artisans which seem to have been common in the fourth century. Prices in Athens in the fifth and fourth centuries average around 150–300 drachmai, and were higher for skilled than unskilled workers.106 A dozen or so slaves therefore involved a substantial investment, as well as continuing living expenses. Summary A number of aspects of household industry at Olynthus defy generally held expectations. First of all, household industry was by no means a marginal activity. Evidence of production of goods for consumption outside the household is found, if not in a majority of houses, at least in a substantial proportion; and we must take into account that these early excavations undoubtedly missed a great deal of evidence. Many of these households may also have owned and farmed agricultural land, but they put a substantial investment into nonagricultural production. Second, industry is not confined to the outskirts of Olynthus or to industrial areas. Instead, the houses which engaged in such production were located in the center of town, on the North Hill, and they were well distributed throughout the hill. Third, domestic industry is closely integrated with other household activities. Many houses have rooms opening directly onto the streets—shops or workshops. If these households had wanted to, they could have kept industrial activities isolated in these separate rooms, away from the domestic portion of the house. Instead, production is generally located in the heart of the household—in the courtyard, and in rooms adjoining the court—while the ‘‘shops’’ were perhaps used primarily as mediating spaces between private production and public sale. This is not always the case, but the degree of integration of domestic industry into 264

The Economies of Olynthus

the rest of the household is striking. Household slaves working on domestic industry would therefore have been closely integrated with the free household, as Xenophon and other literary sources imply. These archaeological realities serve as a warning of how distorted our understanding of ancient industry may be if we base it primarily on idealized, aristocratic literary sources such as Xenophon and Aristotle.

tr a de a n d e x c h a n g e The Agora An open square of some four blocks occupied the south end of the North Hill, the highest point on the plateau. This is flanked on the north by a long stoa-like public building, occupying the southern half of block A iv. At the northeast corner of the square are a public building with an internal colonnade (A iv 10) and a fountainhouse (A iii 9), fed by a pipe which ran diagonally across the northern part of the plaza. The other sides of the square were not tested. Test trenches within the square itself revealed bedrock within 10 centimeters of the surface, and they were therefore discontinued.107 Robinson and Graham identified this area as an open area for military maneuvers, a place where troops could be massed before sallying forth to battle. They rejected its identification as an agora, ‘‘for only two buildings of a public nature seem to have been located in its vicinity, and both of these were relegated to a far corner where they did not even face on the open area.’’ 108 Graham believed that the agora would be found in the Villa Section.109 This same architectural evidence led both Martin and Wycherley to interpret the square as the agora.110 Hoepfner and Schwandner prefer to interpret the square as a sanctuary on the grounds that this highest point on the North Hill would have been reserved for the gods; they hypothetically restore an agora in the hollow between the North and East Spur Hills, in a lower part of the city.111 The location and architecture of the square are more consistent with an agora. There is no trace of a temple, and the location of the square is not the only evidence for its use. The stoa on the north, which had not been discovered when Robinson and Graham first identified the area, does face onto the open square; and although the other public building, A iv 10, faces on to Avenue B, this is probably due in part to the steep dropoff in the hill along the east side of the square, which would make it much more difficult to build a stoa opening to the west. A iv 10 seems to have had some commercial, rather than religious or military, purpose. It contained 218 coins—by far the most found in any building at Olynthus—and a number of weights, as well as a variety of miscellaneous objects. Although Robinson argued that it was used for public assemblies, the large number of coins seems more appropriate for a commercial stoa. Next to this is the only known fountainThe Economies of Olynthus

265

house on the North Hill. Temporary booths and tables could have been set up in the open area as well as in the streets (below, ‘‘‘Street Shops’’’), as described in the Attic orators; these would leave no trace in the shallowly buried square.112 Moreover, the distribution of coins, the prices of houses, and other evidence point to the identification of this area as a commercial agora and offer further information about the spatial distributions of different economic strategies through the city. The Distribution of Coins Some 4,638 coins are recorded from the Olynthus excavations: 4,237 bronze coins, 399 silver coins, and two ancient forgeries of silver coins.113 These came from a variety of contexts: from houses, coin hoards, graves, streets, trial trenches, etc. Of the coins from houses, which are of primary concern here, it is often difficult to distinguish coins which belong with the destruction debris from those which had been trampled into the floor or found in fills or in other contexts. Small and indestructible, coins were collected and saved in excavation, unlike broken potsherds which were generally discarded, thereby ‘‘filtering out’’ most of the pottery which did not belong with the destruction debris. Many of the coins were probably casual losses over a period of time rather than representing the contents of the rooms at the time of destruction. But even if the exact contexts of these coins is often insecure, the distribution patterns are striking. It is immediately clear from the distribution plan that coins are particularly concentrated in the area around the agora at the south end of the North Hill (fig. 60). On average, the houses at Olynthus contain about 20 coins per house. The houses within a block or so of the agora, however, average more than four times as many coins (see table 10). Considered another way, ten of the fifteen houses at Olynthus with the greatest number of coins adjoin the square, and two more are a block away (see table 11). In addition to the coins from houses adjacent to the agora, the public buildings—A iv 10 and the fountainhouse A iii 9—contained large numbers of coins, and Avenue B in this area was littered with coins and other artifacts, far more than are found in other excavated streets at Olynthus. These were probably lost from temporary shops or stalls set up in this area (see table 12, and below, ‘‘‘Street Shops’ ’’). Although numerous, these coins do not represent a lot of money. The overwhelming majority of them are bronze, whose ancient denomination and value are difficult to determine. If they were chalkoi, dichalkoi, and tetrachalkoi, as Clement suggested, they would be worth 1⁄8 , 1⁄4 , and ½ of an obol respectively, or 1⁄48 to 1⁄12 of a drachma.114 Sixty to one hundred of these coins, the largest number that was found in any house, would be worth only a few drachmas, hardly a great figure 60. Distribution of Bronze Coins 266

The Economies of Olynthus

Distribution of Coins 1-4 coins 5-9 coins 10-20 coins 21-60 coins more than 60 coins silver hoard bronze hoard hoard of flans 0

50

buried ca. 421 BC

Agora

2 hoards buried 379 BC; 1 buried 348 BC

hoard buried ca. 479 BC

hoard buried 348 BC

100 m

table 10: Coins Found in Houses Near the Agora Findspot

Number of coins

A 11 A 12 A 13 A iv 1 A iv 3 A iv 5 A iv 7 A iv 9 Tr. 7 (average among ca. 4 houses) total

86 68 66 16 6 12 25 99 144 993

Note: These figures omit coins from topsoil and from hoards. The findspots of coins from Trench 7 were mostly not recorded house by house, but an area equivalent to about four houses was excavated here, giving an average of about 144 coins/house. Of these coins, 76 can be attributed to particular houses or rooms.

table 11: Houses with the Largest Numbers of Coins (in order of decreasing quantity) Findspot Tr. 7 (average of 4 houses) A iv 9 A 11 Av9 A v 10 A 12 A 13 Av7 A vii 9 V. Bronzes Av2 V. Good Fortune

Number of Coins 144 99 86 81 70 68 66 61 46 44 42 39

fortune. The numbers of coins found in a house are therefore not a true index of the wealth of the household. But they may be used as an index of the intensity of monetary trade and participation in commercial economy. The distribution of silver coins, excluding coins from hoards, shows a somewhat similar distribution (fig. 61). Many were concentrated around the agora. Houses A 8–A 13 (with the interesting exception of A 12) have a particular concentration. The houses in Trench 7, which averaged some 144 bronze coins, were 268

The Economies of Olynthus

table 12: Coins from Public Buildings Near the Agora Findspot A iv 10 public building A iii 9 fountainhouse Ave. B between St. iv and v

Number of Coins 218 79 109

also rich in silver coins, although the difference from other houses is not as dramatic, averaging 4.5 coins per house. A iv 9 and A iv 10 are also not as unusually rich in silver coins as they are in bronze, whereas houses in blocks A vi and B vi have relatively more silver coins than bronze. Silver coins are worth much more than bronze and so may offer a somewhat better indication of wealth. But again, outside of hoards, discussed below, no house has enough coins to be worth a great deal. Rather, they are a further indication of participation in the monetary economy. Since shops were frequently rented out, it is worth asking whether the households living in this part of the city were themselves more closely involved with the commercial economy, or whether the concentration is due to the shops whose owners lived elsewhere. Shops are, in fact, no more common in houses near the agora than they are elsewhere in the city, but they are distributed more along the main avenues (below, ‘‘Shops’’ and plate 4). A number of shops, particularly those near the agora, do contain large numbers of coins. For instance, the three shops in house A iv 9 contained 60 of the 99 coins from this house, while A v 9, room d, contained 20 of the house’s 81 coins (figs. 24, 25). But the concentration of coins around the agora remains significant even if we exclude the coins from shops and look only at those found in the domestic areas of these houses (see table 13). As before, even excluding the coins from shops, most of the houses with the largest numbers of coins are within a block of the agora. The large numbers of coins in houses near the agora cannot be explained, therefore, by a preponderance of shops in this area. Houses near the agora do not in fact contain significantly more shops than those in other parts of the city, and even excluding coins from shops, the pattern remains distinct. Rather, the houses themselves in this part of the city contained more coins. The closer involvement with coinage and the commercial economy seems to be a result of the economic strategies adopted by these households, as will be discussed below. Coin Hoards Eight hoards of silver coins were found in houses on the North and South Hills (fig. 60). The distribution through the site is quite different from that of coins from other contexts: hoards are not concentrated around the agora, but found further away (see table 14). The earliest hoard, found on the South Hill, probably The Economies of Olynthus

269

dates to the Persian destruction of 479 b.c., and so it is outside the purview of this discussion.115 Two hoards of silver coins were found in block A vi. One was found in house A vi 8, a house with a large storeroom for agricultural products and two pressing floors for olives or grapes. The house may have belonged to the owners of A vi 10, a house also involved in agricultural processing. This hoard was by far the largest found at Olynthus, containing 34 tetradrachms (136 drachmas). The next largest hoard was found nearby in house A vi 9, with 4 tetradrachms and 59 tetrobols (55⅓ drachmas total).116 The house is the most irregular and unpretentious on its block, however, with no cement or mosaic floors, no painted walls, and only a partially cobbled court. It does, however, have one large and a suite of three smaller rooms (shops?) opening onto Avenue B. What these were used for is unclear; one of them, room c, contained a pithos and a number of large iron objects, perhaps tools of some sort. Block A vi is remarkable for the number of mosaic floors, painted walls, and specialized rooms, and some of the wealthier households in the city may well have built it. It is surprising, therefore, that its two coin hoards were found in its least pretentious houses: A vi 8 may not have had any purely domestic space at all, but been attached to its neighbor. Three coin hoards were found in house ESH 4. Clement dates two of them to 379 b.c. and suggests that their owner perished in the Spartan attack.117 Both of these were fairly small hoards, containing 12⅔ and 6 drachmas. A third coin hoard, however, dating to the final destruction of the city, contained 75 tetrobols (50 drachmas).118 This house had two kitchens and a shop, and may have produced food for sale outside the house. The other hoards from the city are smaller. A hoard on the east side of the South Hill, dating to 348 b.c., contained 1 tetradrachm and 34 tetrobols (total 26⅔ drachmas). The room in which it was found contained two pithoi, and may have been a shop of some kind.119 A final hoard was discovered by clandestine diggers in 1934, apparently between the Villa of Good Fortune and Hagios Nikolaos, that is, in the Villa Section. If this findspot is correct, it documents hoarding of fairly large amounts of wealth (perhaps 136 drachmas, according to local accounts) in this part of the city.120 Four hoards or collections of bronze coins were also found in the excavations. None of these were found in vases or other containers and so are not strictly hoards, but they form coherent groups which were clearly stored together. One, found in house A 11, is notable for the complete absence of Chalcidic coins. The ‘‘hoard’’ consisted of 10 coins of Macedonian kings (Philip and Amyntas) and Thessalian mints (Lamia and Larisa).121 Although the unusual distribution of mints might suggest a date after the destruction of Olynthus, there is no evidence that this house or its neighbors were reoccupied; there are no clearly postdestrucfigure 61. Distribution of Silver Coins 270

The Economies of Olynthus

Distribution of Silver Coins 1 silver coin 2 silver coins 3 silver coins 5+ silver coins silver coin hoard 0

Agora

50

100 m

table 13: Houses with the Largest Numbers of Coins (excluding coins from shops) Findspot

Number of Coins

Tr. 7 (average)a A 11 A 12 A 13 Av7 A v 10 V. Bronzes Av2 Av9 V. Good Fortune A iv 9 Av5 A 10 A9 a Although

144 86 68 66 61 45 44 42 41 39 39 32 32 31

we do not know the exact findspots of most of

the coins from these houses, there are only two possible shops in this region, and even those are not certain. Eight coins were found in the two houses with shops; the exact room was not recorded.

table 14: Coin Hoards Hoard

House

Date (approx.)

Tetradrachms

1 2 3 4 5 6 7 8

S. Hill S. Hill A vi 9 A vi 8 Bv1 ESH 4 ESH 4 ESH 4

479 b.c. 348 b.c. 348 b.c. 348 b.c. 421 b.c. 379 b.c. 379 b.c. 348 b.c.

6 1 4 34

Drachmas

1 2

Tetrobols

Total

4 34 59

26⅔ dr 26⅔ dr 55⅓ dr 136 dr 13 dr 12⅔ dr 6 dr 50 dr

18 7 9 75

tion coins from this house or this part of the city. The unusual makeup of the hoard therefore must be explained in another fashion. Hoarding behavior is notoriously difficult to understand or explain, and it certainly does not necessarily correlate with wealth. The fact, however, that nearly all the silver hoards were found in houses on the North Hill which were involved in domestic industry, but which did not display outward signs of prosperity such 272

The Economies of Olynthus

as mosaics or cut-stone facades, emphasizes the association between monetary wealth and trade and industry rather than with traditional pursuits. Cohen argues persuasively that the importance of the ‘‘hidden economy’’ (ἀφανής οὐσία) has been greatly underestimated by many historians, precisely because it was practiced outside the usual realms of philosophical and legal discourse.122 Coin hoards might be interpreted as evidence of participation in this hidden economy, and provide archaeological evidence for economic strategies more closely tied to the market and with disembedded systems of exchange. The Circulation of Money The distribution of coins reveals other economic information as well. Between one-quarter and one-third of the coins from Olynthus were minted at foreign mints, attesting considerable trade between Olynthus and cities throughout the Greek world. By contrast, only about one-fifth of the fourth- and third-century bronze coins from the Athenian Agora are of foreign origin.123 Athens presumably had a much more productive mint than Olynthus, and this may have restricted the use of foreign coinage; but the numbers remain striking nonetheless. I had hypothesized that analysis would reveal patterns in the distribution of different mints: that certain households or areas of the city might have had closer commercial ties with particular cities or regions of Greece, which could be reflected in higher concentrations of those foreign coins. However, analysis of the distribution of mints reveals no such patterns. The assemblages of coins differ from house to house and block to block, and some houses have a higher proportion of foreign coins; but so far these have turned out to be statistically insignificant or very difficult to interpret.124 It is difficult to prove that there is no pattern at all to the distribution of mints, or to determine whether the lack of meaningful patterns is due to a truly random distribution of coins or to statistically insignificant quantities. Future study might reveal subtle but meaningful patterns. But the tentative conclusion is that unlike silver coins, bronze coinage circulated solely according to size and denomination, while its mint and city of origin were basically irrelevant. A chalkous was worth the same, and was as easily circulated, whether it was a local Chalcidic coin or a coin of another city. The same conclusions have been argued by other numismatists, for instance Kroll on Athenian coins.125 Shops At least twenty-nine houses had rooms with doors opening to the street, which Graham and others called ‘‘shops.’’ The identification of these rooms as shops or workshops is based on architectural rather than artifactual evidence, and therefore is problematic in a number of respects. The street walls of some houses were The Economies of Olynthus

273

not well-enough preserved to determine whether or not there was a door to the street. Some rooms, therefore, might have been shops, but cannot be identified now. On the other hand, the presence of a door to the street does not necessarily imply that a room was a shop or workshop; there are other possible explanations. And finally, the diversity of artifact assemblages from these rooms defies a single purpose or explanation. Most houses with shops produced other evidence for domestic industry (plate 4). For instance, houses A iv 9 and A v 9 produced textiles on a large scale; A 6, A v 10, A vi 10, and A vii 9 processed agricultural products; houses D v 6 and ESH 4 probably cooked food for consumption outside the household. As noted above, however, domestic industry was generally not situated in these shops but in other rooms of the house, particularly in courtyards and the adjoining well-lit rooms. The shops in these houses probably served to distribute textiles, agricultural products, and other materials produced in the inner parts of the houses; they mediated between domestic production and public distribution and consumption. Other rooms opening onto the street were perhaps workshops (ἐργαστήρια) rather than shops for retail trade. As Aeschines points out, workshops could be used for anything from a surgery to a brothel with little change in architectural form.126 There is little evidence for what was produced in these workshops at Olynthus, however. A number of shops have domestic implements in them, such as those in houses A vii 9 and B vi 10. Other such activities would leave few material traces. Workshops could be rented out or sold outright, and it is generally not possible to be sure that the same person owned both house and workshop.127 As noted above, the distribution of shops in the city is noteworthy: nearly every house on Avenue B has at least one shop opening onto the avenue, and many of them devote a third or so of their area to such rooms. A iv 9, A v 9, and A v 10 offer a continuous row of shops facing onto this street. One can easily imagine this avenue as a lively, bustling thoroughfare with significant interaction between the household and the public streets, with the shops acting as intermediary spaces, accessible both to the stranger from without and to the household from within the house. And this axis seems to be distinct from the other economic hub of the town, the agora, as there do not seem to have been significantly more shops near the agora. This avenue thus formed an important commercial artery of the city, a sort of economic axis through the town, similar to the ‘‘Westtorstraße’’ at Priene.128 Finally, this axis must have been planned and recognized as such when the city was laid out. The avenue is wider than the other streets, and many of the shops opening onto it seem to be original constructions.129 Households whose professions required shops or workshops on a main artery, therefore, will have had the opportunity and knowledge to choose plots on this avenue, bringing another level of urban organization in the city. 274

The Economies of Olynthus

‘‘Street Shops’’ Not all shops at Olynthus were permanent rooms, however. There were probably temporary stalls in the agora, but no archaeological evidence for these survives. In a number of spots along the sides of streets and avenues, however, concentrations of artifacts was found which may mark the remains of temporary (or more permanent?) stalls. The interpretation of most of these is problematic because no architectural evidence was found and the exact situation of the artifacts was often not noted. Probably the stalls were light wooden structures whose postholes and other remains were not picked up by the excavators. In some cases the artifacts may be simply refuse thrown into the street or a particularly garbage-laden fill brought in to level out the roadway, or else destruction debris from a neighboring house, thrown into the street by Philip’s soldiers or scattered when the house collapsed. But the assemblages often resemble primary destruction debris: the breakable artifacts are relatively complete, as if they had been lost in the destruction rather than thrown out because they were broken, and there are a fair number of large objects which probably would not have been disposed of in the street. Moreover, there are some interesting and distinctive assemblages, which look more like the contents of specialized shops than like refuse. A great concentration of coins was found in Avenue B near the agora, in front of A iv 9, A iv 10, and the fountainhouse, totaling some 114 bronze and 2 silver coins. This area also contained a large number of bronze implements and artifacts: a bronze fox-like animal head, jewelry including fibulas and finger rings, parts of metal vessels, tools, axe heads, tweezers, weaving implements, fishhooks, and the like, and various miscellaneous metal objects. By contrast there were relatively few vases or figurines, few loomweights or other such domestic objects. Just around the corner, however, in Street v in front of A iv 9, the assemblage was quite different. This region contained many vases, but relatively fewer bronze objects and only 7 coins. These were found in a deposit of black earth, 15–20 cm deep and about 1.5 × 4 m in extent. In front of house A v 6 was a large, nearly complete pottery basin, overturned with several bowls beneath it and more vases nearby; a terracotta ‘‘pithos lid or table top’’ serving perhaps as a counter, two possible scales pans, a piece of lead which might once have been a weight, as well as other artifacts. Again, there were only three coins. The exact situation of these artifacts was not noted and its interpretation is therefore uncertain. But this relatively complete assemblage does not seem to be trash thrown away by the inhabitants of the house, since the vases are relatively intact. It also seems unlikely that it was material discarded by the looters; ceramic vases, a pithos lid, and such mundane artifacts are hardly what Philip’s soldiers would want to plunder from a house. In front of house C v 4 was a terracotta louter leaning against the wall, more than eleven vases (probably fragmentary since none were catalogued), five coins The Economies of Olynthus

275

including a silver coin of Perdiccas II, and the skeleton of a horse or cow. Two worked stones were also set into the earth surface of the street, flanking the outlet of a drain from the house; these could have supported some kind of table, although the presence of a drain underneath makes the interpretation difficult. A deposit of nineteen netting needles, a terracotta stand or altar, one coin and a few other finds was located just in front of house A vi 4, while in front of A vi 2 were sixteen coins, a suspension weight, and two bronze rods, perhaps parts of scales. A group of fishhooks and a deposit of twenty-three loomweights were found in the street outside A xii 9, while ten fragmentary terracotta figurines were found in Avenue B in front of A vi 10. Other deposits are more problematic, but more examples could be adduced.130 Many of these deposits seem quite distinctive, and while their interpretation is by no means secure, I suspect they are the remains of temporary stalls set up in the streets in front of the houses. They need not belong to the house owners, of course; other people could have set them up by arrangement with the owner, or perhaps no arrangement was needed. In any case, they seem quite widespread and perhaps an important but overlooked aspect of urban life. Sales and Prices of Houses Seventeen inscriptions were discovered at Olynthus recording the sale of a house or loan made with the house as surety. They offer unique information and reveal patterns about the values of houses in the city and the processes by which they were sold (appendix 2).131 The inscriptions are fairly formulaic. A typical example, from house D v 6, reads: Gods. Outright sale. In the year when Aristoboulos the son of Kallikrates was priest, in the month of Targelion. Zoilos the son of Philokrates (purchased) from Diopeithes the house which is between the house of Diokles the son of Charon and that of the sons of Apollodoros, for 1,200 drachmas. Polemarches the son of Straton was the guarantor; Diokles the son of Charon, Euxitheos the son of Xanthippos, and Philon the son of Theodotos were witnesses.132 The documents record the date of the transaction (giving the eponymous priest and month), name the seller and purchaser, identify the house by naming its neighbors, and record the selling price, witnesses, and guarantors. Some inscriptions were found where they were originally set up at the door to the house (as this one was); a few others were found inside a house.133 In these cases the inscription probably refers to the house in which it was found. Other inscriptions were found in streets or in other contexts. Although the original locations of these latter inscriptions are not certain, patterns in their findspots and prices suggest that they 276

The Economies of Olynthus

may not be far from the houses whose sales they record, and for the purposes of this discussion I will make that assumption. Most of the inscriptions record the sale of the house. Four specify that the sale is ‘‘outright’’ (οὐνὴ εὐθεῖα); two others simply record that it is a sale (οὐνή). In other texts, the exact nature of the transaction is either specified or not preserved, but these are likely to be sales. Two inscriptions, however, record loans, with the house as security (δεῖνα ἐδάνεισεν . . . ). One seems to be a security in the form of a conditional sale, and the nature of another is uncertain.134 A few inscriptions, for instance those from A v 10 and Avenue A at A −4, specify equipment and spaces to be included or excluded from the transaction, but most only mention the house. All the known sales inscriptions date to a very short period of time. In twentyone inscriptions from Olynthus and neighboring towns which preserve the dating formula, the names of only five different eponymous priests are attested; and seventeen of the transactions took place during the tenures of only three priests. Hatzopoulos tentatively dates the series between the years 355/4 and 350/49 b.c., when the destruction of the city brought an end to buying and selling real estate.135 Even if Hatzopoulos’s absolute dating is not completely secure, it seems clear that the inscriptions date within a short span in the last years of the city’s life. Earlier inscriptions were probably taken down after the transaction had been completed, recognized, and recorded. Just as the time span of the sales documents is very short, so the circle of people buying and selling seems to have been rather limited. In only seventeen documents, some poorly preserved, we find the same men cropping up in more than one transaction. Diopeithes the son of Antipatros, who sold the house in the document quoted above, had received a loan a few months earlier for 4,500 drachmas, using his whole estate as security (ἐπὶ τεῖ οἰκίει ὅλει).136 One of the three men with whom Diopeithes’ loan was deposited was Polyxenos, a neighbor of Archidamos in house A 11, who served as a witness in Archidamos’s loan transaction.137 And neighbors usually serve as witnesses to the sales. Olynthian houses on the North Hill and Villa Section sold or secured loans for amounts ranging from 170 to 5,300 drachmas, a more than thirtyfold range. Such a wide variation in price for houses of the same general size and type distressed many scholars, including Robinson.138 The prices at Olynthus are all the more striking when we consider what houses and property cost in other Chalcidic communities at just this time. A house in the city (ἐμ πόλει) of nearby Stolos sold for 40 drachmas; another house with its upper story sold for 232 dr; four houses in the city of Stolos sold for 1,000 dr, an average of 250 dr per house; and land, vineyards, and houses (of unknown number) in the city sold together for 300 drachmas.139 These houses may have been smaller and less luxurious than those at Olynthus (although probably not so much smaller and poorer, to judge from contemporary houses and country dwellings at other sites), but that probably doesn’t explain why the prices are only one-tenth those of houses in the capital. On the island of The Economies of Olynthus

277

Tenos, the most expensive house preserved in a famous third-century inscription sold for 2,320 dr; the average was about 1,100 dr.140 Houses in Athens and Attica were generally less expensive: the median price of seven houses listed in the Attic Stelai was just 410 dr, while the median value of the houses mentioned by the Attic Orators is 2,000 dr, and the largest recorded price is 5,000 dr, less than the most expensive house at Olynthus.141 Hatzopoulos attributes the wide range in house prices to a financial crisis and panic that shook Olynthus in the face of the growing threat from Philip. He points out that whereas the average house price in the whole series is about 2,300 dr, during the tenure of Antidotos, which he dates to 350/49 b.c., the average price of five preserved sales inscriptions plummeted to about 700 dr. The prices at the nearby town of Kellion were also very low in that year.142 Although the deteriorating political situation in the last years of the city must have had a significant impact on house prices, other factors were at work as well. Take, for example, the sales of houses D v 6 and A v 10. Both took place in the same year, when Aristoboulos was priest, dated by Hatzopoulos to 352/1 b.c.; the sale of D v 6 took place six months after that of A v 10. Both carry the heading οὐνὴ εὐθεῖα, ‘‘outright sale,’’ and so are comparable transactions. They use the same numbering system, whose meaning is certain. The house plots are the same size, although A v 10 had a second story. Neither house is particularly finely built or has mosaic-paved rooms. And yet D v 6 sold for 1,200 drachmas, while A v 10 sold for 5,300 dr. The sale of A v 10 included ‘‘the things that bring income’’ and the storeroom; but this could not have increased the price by more than four times. Chronology is not the only factor which affected the prices of houses. If we map the findspots of these sales inscriptions at Olynthus, we see that the houses surrounding the agora were sold much more frequently and were considerably more expensive than those elsewhere (fig. 62). House A v 10 sold for 5,300 dr—the most expensive sale documented at Olynthus. A sales inscription found behind A iv 3 records a sale for 4,500 dr. An inscription found in B iii 2 records a loan against the house (or rather, the entire estate, ἐπὶ τεῖ οἰκίει ὅλει), also for 4,500 dr; the estate must have been worth at least that. The western half of house A iv 7 seems to have sold for 2,000 dr; the whole house was probably worth 4,000 dr. House A 11 was used as security against a 2,000 dr loan (or a πρᾶσις ἐπὶ λύσει); again, the house must have been valued at least that much. Further away from the agora, however, both the number of sales and prices seem to drop. House D v 6 sold for only 1,200 dr, and the house next door apparently sold for 900 dr. Again, these houses are no smaller or less well built than the more expensive houses near the agora—indeed, houses B iii 2 and A 11 are irregular and undistinguished dwellings. Still further away, a house on the South

figure 62. Distribution of Sales Inscriptions 278

The Economies of Olynthus

Sales Inscriptions Inscription, in situ Inscription, not in situ Prices in drachmas 0

50

100 m

Inscription from Skoinia, in Villa Section? 170 dr. 2500 (loan)

900 (lease)

[2000?]

2000

900

5300 2000 (1/2 house)

1200

4500 4500 (loan)

Agora

410 230

Hill sold for only 230 dr, less than one-tenth the price of A v 10. A hoard of silver coins was found next door, so the neighborhood cannot have been too squalid. The one inscription from the Villa Section was a surface find from the Villa of Good Fortune, which documents a sale for 410 dr, only one-tenth the average price of houses around the agora. It cannot refer to the villa itself, since it mentions two neighbors, and the villa is the last house on a block, with no neighbor to its east. But it may well refer to a house nearby. An inscription found in a field at Skoinia, probably belonging to the Villa Section, documents a sale for a mere 170 dr. Both these sales took place in the tenure of Antidotus (450/49 b.c.), when house prices were lower in general; and the prices of houses here in the lower city may have been particularly depressed in the uncertain political situation of these years. But these lower prices also fit in well with the general decline of prices further away from the agora. The value of a house therefore seems to depend significantly upon its proximity to the agora. Houses near the agora were sold and used as security more frequently and for higher prices than houses farther away.The size of the house, presence of a second story, and other such factors may have played a role, but to judge from the high prices of very plain, one-story houses or even half-houses (including A iv 3, A 11, D v 6, and A iv 7), the quality of construction and decoration of a house did not affect the value nearly as much as its location. Houses in the Villa Section have a significantly higher percentage of mosaics and painted rooms than those on the North Hill, but their values seem to be considerably lower, and more in line with those in outlying communities. Houses on the North Hill near the agora, on the other hand, are generally no larger, fancier, or better equipped than those farther away—quite the opposite—and tend to be irregular in plan (cf. figs. 46, 47, and plate 3). To judge from the architecture of the houses, therefore, living near the agora conferred no particular social or political status on the owner.143 Rather, I suspect that the key factor is economic: that proximity to the market, rather than to the political heart of Olynthus, prompted this increase in house prices. After all, the price of the most expensive house, A v 10, near the agora, was increased by the inclusion of ‘‘all the things that bring income’’ and the storeroom—a clear link between the market economy and high sales prices. The pattern is thus similar to the distribution of small change at Olynthus, and some of the same distinctions may account for both patterns. Households living around the agora seem to be more closely tied to a monetary, market economy than those living further away; and those in the Villa Section seem less dependent on the market and more agriculturally based and self-sufficient. Shops and productive equipment influenced the price and desirability of a house more than fancy decoration or location in a quiet, suburban or ‘‘higher-class’’ neighborhood. And the recurrence of the same names in different inscriptions also suggests that buying, selling, mortgaging, and leasing property was common among only a limited group of citizens. 280

The Economies of Olynthus

Again, this is not to say that these households near the agora must have been primarily the home of traders or merchants. Many of them may have owned and cultivated land (or had others cultivate it for them). But in this area the household and commercial economies were more closely articulated, and this probably made it easier, both economically and socially, for them to buy and sell property, and it made the houses in this area more expensive. It may also have made it easier for these households to liquefy their assets and flee as the conflict with Philip seemed more inevitable; households more closely or exclusively tied to the land might have found it more difficult to pick up and run.

a lt e r na t i v e e c on o mi e s a t ol y n th u s Analysis of the architecture and contents of Olynthian houses reveals a number of patterns in economic behavior. Households in different parts of the city dealt with their agricultural produce differently. On one hand, houses in the Villa Section had special storerooms which could hold a year’s supply of grain or more. Houses on the North Hill, on the other hand, lacked such facilities, and these households must have dealt with produce in a different way. Domestic industry, including weaving, processing olives, wine, and grain, stonecutting, coroplasty, and shopkeeping, was quite common on the North Hill, but there is no evidence for it in the Villa Section. The agora of the city can be identified by its architecture and by the distribution of coins. Houses near the agora contained more coins and engaged in household industry somewhat more frequently than houses farther away, while those fronting onto Avenue B were particularly closely involved with household industry, and devoted much of their household space to shops which mediated between the domestic and public spheres. Such shops and workshops, and traces of domestic industry in general, are rare or unknown in the Villa Section. And finally, houses near the agora were sold more frequently and for far higher prices than houses farther away. Households in different parts of the city therefore followed different economic strategies and their domestic economies had very different relations with higher-level urban economic systems. This distinction seems to be real, moreover, not an accident of excavation or sampling. Although more excavation in the Villa Section might uncover evidence for workshops and household industry there, further work is much less likely to change the picture on the North Hill, where a large sample of well-preserved houses was excavated. Many of these houses had pithoi in them, but in different contexts—used to collect rainwater from the eaves or in shops; so the lack of pithoi is not due to looting or careless excavation. The lack of large-scale storage in this part of the city therefore demands explanation. It is not surprising that such different strategies coexisted in a Classical city. Scholars of economic history of various persuasions, primitivists and modernThe Economies of Olynthus

281

ists, substantivists and formalists, produce ancient evidence to bolster their arguments (and occasionally members of different camps will offer the same evidence in support of very different conclusions); there is some truth to different points of view. Indeed, much of the confusion about the nature of the Greek economy arises from attempting to fit the whole range of ancient economic activity into a single model or theory. At one time or place, though, there must have been a spectrum of economic activities and strategies, some more traditional and others more sophisticated.144 This will have been particularly true in the fourth century b.c., a period of rapid economic growth and increasing complexity. The archaeological evidence from Olynthus bears this out. What is most striking at Olynthus is not the existence of different economic strategies, but their complementary, almost mutually exclusive distribution in space. The layout and facilities of houses in the Villa Section agree well, as we have seen, with the strategy common among ancient and modern peasants of storing a year or more’s supply of grain in their houses. Nearly every house has a dedicated storeroom which can hold some thousands of liters of foodstuffs, enabling the household to store the food from a typical family farm and sustain a typical household for a year. There is at present no evidence in houses in the Villa Section for production of goods which would be consumed outside the household, or for shops which intermediated between public and private spheres. They fit well into the ‘‘primitivist’’ model which emphasizes the self-sufficiency of households and minimizes their involvement with the market. Lacking such facilities, however, the households on the North Hill must have stored or disposed of their produce in some other fashion. Some households may have owned outbuildings or second houses in the countryside where they could have stored produce. We have epigraphic and archaeological evidence for farm buildings with pithoi from other parts of Greece, and intensive survey might discover such remains around Olynthus.145 Households owning such buildings might have left the bulk of their produce on their farms and brought grain into the city in smaller amounts as needed. Or there may have been communal storage facilities, perhaps including the large storerooms belonging to houses which processed agricultural produce. But it is surely no coincidence that the inhabitants of the North Hill were also more involved with household industry and trade than those in the Villa Section: that the spatial distribution of household industry is complementary to the distribution of large-scale storage. Many of the households on the North Hill produced goods for consumption outside the household, and this would have involved these households more closely in the market and the monetary economy. This involvement may have encouraged them to deal with their agricultural produce differently. Rather than store a large supply of food grown on their farms, they may have sold their produce as it was harvested and then bought what they needed on the 282

The Economies of Olynthus

market. The most famous example of this strategy is Pericles. According to Plutarch: During all these years Pericles kept himself untainted by corruption, although he was not altogether indifferent to money-making; indeed, the wealth which was legally his by inheritance from his father, that it might not from sheer neglect take to itself wings and fly away, nor yet cause him much trouble and loss of time when he was busy with higher things, he set into such orderly dispensation as he thought was easiest and most exact. This was to sell his annual products all together in the lump, and then to buy in the market each article as it was needed, and so provide the ways and means of daily life. For this reason he was not liked by his sons when they grew up, nor did their wives find in him a liberal purveyor, but they murmured at his expenditure for the day merely and under the most exact restrictions, there being no surplus of supplies at all, as there were in a great house and under generous circumstances, but every outlay and every intake proceeded by count and measure. (Plutarch, Pericles 16) There are other references to this strategy. The author of the pseudo-Aristotelian Economics argues that: There are four qualities which the head of a household must possess in dealing with his property. Firstly, he must have the faculty of acquiring, and secondly that of preserving what he has acquired; otherwise there is no more benefit in acquiring than in baling with a colander, or in the proverbial winejar with a hole in the bottom. Thirdly and fourthly, he must know how to improve his property, and how to make use of it; since these are the ends for which the powers of acquisition and of preservation are sought. Everything we possess should be duly classified; and the amount of our productive property exceed that of the unproductive. Produce should be so employed that we do not risk all our possessions at once. For the safe keeping of our property, we shall do well to adopt the Persian and Laconian systems. Athenian housecraft has, however, some advantages. The Athenian buys immediately with the produce of his sales, and the smaller households keep no idle deposits in store. Under the Persian system, the master himself undertook the entire disposition and supervision of the household. . . . We have remarked that on small holdings the Athenian method of disposing of the produce is advantageous. On large estates, after the amount for the year’s or the month’s outlay has been set apart, it should be handed to the overseers; and so also with implements, whether for daily or for occasional use. In addition, an inspection of implements and stores should be made periodically, so that remainders and deficiencies may alike be noted. ([Aristotle], Economics 1344b) The Economies of Olynthus

283

The advantages of the Athenian system were its simplicity, security, and control. It allowed Pericles to devote his attention to higher things; and according to [Aristotle] it reduced risk, presumably including risks posed by mold, insects, and so forth, by keeping ‘‘no idle deposits in store.’’ It eliminated the need for expensive pithoi, and it gave the owner more control over everyday expenditures. This is not to say that houses on the North Hill lacked all storage facilities. Many of them had small pithoi, sufficient for a month’s supply of grain or so, serving as a buffer so they did not have to rely completely on the market. The strategies described by Plutarch and Aristotle of keeping no stored supplies in the house but shopping daily in the agora are perhaps an exaggeration. A number of authors in addition to the Aristotelian Economics draw an explicit contrast between products stored for a year and products stored for a month, as if a month-bymonth storage strategy was a recognized alternative to the yearly strategy.146 Patterns in the distribution of storage facilities, household industry, coinage, house prices, and other artifacts and facilities at Olynthus probably reflect such fundamental distinctions in economic strategy. Households on the North Hill seem to be more involved in the market in two ways: they produced goods or services for exchange in the citywide economy, and they relied in turn on that economy to supply their daily needs, rather than store their foodstuffs themselves. Households in the Villa Section, by contrast, seem not to have engaged in household industry, but maintained a greater independence from the market and the urban, monetary economy by storing large quantities of foodstuffs in their houses. How did such differences in economic strategy come about? What distinguished households on the North Hill from those in the Villa Section? A number of explanations may be offered. The economic patterns visible in the distribution of artifacts reflect the situation at the end of the city’s life, and one might wonder whether these patterns developed relatively late in the history of the city, after the original pattern of landholding and the distribution of wealth had changed. The rise of Olynthus in the fourth century as the capital of the Chalcidic League undoubtedly had a profound effect on the city’s economy and social organization. Are we then seeing the results of such economic changes, a sort of flight from the inner city, as (for instance) the original landowners, anxious to avoid the hubbub of the market, sold their houses and moved out to the more peaceful Villa Section, while new settlers flocked to the area around the agora? Does this help explain, among other things, the high rate of sale among houses near the agora? This may be part of the answer, but such patterns seem to date to much earlier in the city’s history. Some of these spatial distinctions were established when the city was first laid out, including the agora and the ‘‘shopping street’’ Avenue B, since people built shops on this street even before the blocks had been built, and the shops of some houses, such as A iv 9, are certainly original construction. Moreover, some houses around the agora are architecturally quite different from 284

The Economies of Olynthus

those elsewhere in the city, not conforming to the usual courtyard-pastas plan and containing significantly more rooms but fewer architecturally specialized ones. These architectural distinctions are probably original to the layout and construction of the city, and therefore explained by the particular needs of households who chose plots of land near the agora in the original distribution of land. If some of the patterns date to early in the city’s life, what other explanations may be offered for their genesis? The most obvious difference between the North Hill and the Villa Section is that the North Hill was settled earlier, after 432 b.c., whereas the Villa Section was probably not laid out until later in the fifth century or the fourth. Perhaps the most important consequence of this involved the distribution of agricultural land. When a city was founded, some land was usually set aside to allot to later settlers. This is a feature of most foundation decrees, such as those from Cyrene and Korkyra Melaina.147 But this land would not be as desirable as that allotted to the original settlers: it is likely to be farther away, of lesser quality, or in smaller plots than the land divided among the first-comers. Aristotle and Plato resolved this problem in their ideal states by giving each household two pieces of land in the chora, one closer to the city and the other near the border. But this inequality was a significant cause of stasis in real states, such as Thurii. The settlers of the North Hill therefore very probably had better land, closer to the city, than those in the Villa Section. We cannot explain the lack of storage facilities on the North Hill, then, by suggesting that citizens there did not own or farm their own land, or that they had smaller plots of land and so less to store. We do not know how large or where these allotments were; they may initially have been somewhat small, since Perdiccas felt it necessary to offer the settlers royal land around Lake Bolbe to entice them to move to Olynthus in 432.148 But it is hard to credit that in the glory days of the Chalcidian League, when Olynthus was the largest city in the region and its territory famously fertile, the well-established inhabitants on the North Hill had inadequate farms or smaller allotments than newer arrivals.149 The equipment for processing olives, grapes, and grain found in many houses on the North Hill further documents the involvement of some of these houses with agriculture. If their land was closer to the city, however, and therefore more defensible than outlying fields, the inhabitants of the North Hill might have felt safe storing produce in rural storage buildings rather than in their urban houses, while the households in the Villa Section might have felt compelled to bring their produce within the city walls because their farms were farther away or less defensible.150 Unfortunately, without a systematic survey of the countryside around Olynthus, such a hypothesis is untestable. This makes the geographical distribution of different economic strategies even more surprising. As we have seen, surviving Greek literature is almost unanimous in condemning trade, industry, and market trading. Farming, we are told again and again, was the most socially legitimate occupation, while manufacture and The Economies of Olynthus

285

trade were best left to metics, noncitizens, and slaves, and were even forbidden to citizens in some states. Although this denunciation is surely more a cultural ideal than a historical reality, it has had a profound effect both on ancient perceptions and on modern interpretations of Greek economy and society. Archaeologists have searched for and found ‘‘industrial districts’’ located near the outskirts or in the poorer parts of town, such as the potter’s quarters of Athens, Corinth, or Locri, or else in a crowded and rather poor city center, like the district south of the Athenian agora, while stressing that most of the elite citizens would have been farmers, not artisans. Yet at Olynthus the situation is just the opposite. Here, the earlier settlers, the households with the best land, those who had the advantages, frequently engaged in ‘‘banausic’’ occupations, while the latecomers followed the more traditional policy of agricultural self-sufficiency, seemingly avoiding trade and concomitant reliance on the market. Households on the North Hill did not engage in household industry out of economic hardship, therefore, nor were they socially or economically marginalized: they did so from a position of strength, owning more and better land closer to the city than the households in the Villa Section. Many of them probably took on these occupations in addition to, rather than instead of farming. Rather than thinking of these houses as belonging to weavers, masons, sculptors, coroplasts, and the like, we may consider them as land-owning citizens who also engaged in nonagricultural production. Some may have done so during seasons when the demands of agriculture were not so pressing; others may have acquired slaves and set them to work in household industry. This would have minimized the alleged social stigma associated with handicrafts. This more diversified economic strategy would have been more secure than relying on agriculture alone, as well as bringing in cash or other goods. It would be interesting to compare the absolute wealth of households in different parts of the city, but this is difficult or impossible to do. A household’s possessions may have included real property, slaves, and other possessions which are archaeologically invisible, but much more valuable than the mundane artifacts recovered in excavation.The architecture and contents of urban houses alone will not reflect the actual economic situation of the household.151 Although house plots are generally identical in size, the architecture of houses should be related to the household’s resources (or rather, a palimpsest of resources over the use-life of the house). Houses with mosaics, painted walls, cut-stone masonry, and other luxurious features presumably belonged to wealthier households. These cluster block by block, as described above, rather than region by region, so that on a large scale there are no architecturally wealthier and poorer sections of the city. The finest houses in the Villa Section, such as the Villa of Good Fortune, have similar numbers of mosaicked and painted rooms as the best-built houses on the North Hill; and although some houses here were larger, they were not more, but much less expensive. 286

The Economies of Olynthus

Household assemblages do not afford an easy index of wealth. The really valuable objects such as metal vessels were mostly looted or burned; and the valuables which remain form such a small and unrepresentative sample that it is hazardous to draw general conclusions. Red-figured pottery and bronze coinage were not expensive commodities and probably do not correlate with wealth. Hoards of silver coins give some indication of a household’s participation in the ‘‘hidden’’ economy, and perhaps of wealth, although even the largest coin hoards contained a fairly small amount of silver relative to the total assets of a well-off household. But it is significant that most of the silver coin hoards were found in houses which engaged in domestic industry. Despite the inconclusive evidence, there is no reason to think that the houses on the North Hill were any less wealthy than those in the Villa Section. If anything, the surviving evidence suggests the opposite. Many of the houses on the North Hill which engaged in domestic industry were successful enough to buy part or all of a neighboring house. House A 6 took over half of A 7 to its south, and A vi 10 took over the andron and perhaps the whole of neighboring A vi 8; all of these were involved with processing agricultural products. A viii 7 and 9 were joined to form a single building where textiles were produced on a fairly large scale. The owner of house A iv 5, who seems to have been involved in some way with producing currency (legitimately or not), purchased the western half of the neighboring house A iv 7 and joined it to his own. The inhabitants of A v 6, again an expanded house, seem to have engaged in fishing, to judge from the net weights and other fishing equipment found there. Nearly all the houses on the North Hill which were larger than normal seem to have engaged in trade of one form or another. The situation is quite different from that in the Villa Section, where although some houses are larger than normal, this is apparently due to an unequal initial distribution of land rather than later expansion of originally equal house plots. The fourth century b.c. saw an increasing awareness of the economic issues. Works such as Xenophon’s Poroi and Aristotle’s Economics, Politics, and Nicomachean Ethics reveal that their authors understand—and often condemn—economic changes which were transforming their poleis. Bankers became increasingly prominent in fourth-century Athens, and the invention of bronze coinage spread this medium of exchange to levels of society and transactions which were outside the original purview of Greek coinage. Other social and political changes may have resulted from a gradual disembedding of the economy from its social context.152 Although it is easy to overstate the case, the fundamental fact that the Athenian economy of the later fourth century was very different from that of the fifth, can hardly be denied, and the situation at Olynthus was probably similar. For instance, Olynthus had an extensive bronze coinage perhaps a generation earlier than Athens, implying a more monetized system of exchange.153 The apparent success of households on the North Hill which engaged in trade and industry can be seen as part of broader economic changes. Important quesThe Economies of Olynthus

287

tions remain difficult or impossible to answer from the archaeological data alone, however. Were the transactions between households which produced goods and those which consumed them carried out primarily through socially defined channels—‘‘embedded’’—or were the transactions atomized, between unrelated individuals without other social relationships? Was there a social penalty for engaging in these banausic occupations, which does not come across in the archaeological record? Olynthus thus offers important evidence in favor of a complex and multifaceted economy, with a variety of economic systems at work simultaneously. Already with the layout of the North Hill, a surprising degree of economic sophistication is visible in the initial construction of shops along Avenue B and the ‘‘rational’’ distribution of house plots around the agora, probably due to the element of choice in the distribution of land. For reasons which are not easily explainable, inhabitants of different regions of the city followed different economic strategies, with the earlier inhabitants of the North Hill apparently following a more diversified and market-based strategy including household production, while the newer settlers of the Villa Section followed the traditional strategy of agricultural selfsufficiency.

288

The Economies of Olynthus

appendix one

Cluster Analysis of Room Areas, Five-Cluster Solution

(K-means,using Systat 5.0 Macintosh)

Summary Statistics for Five Clusters Variable Courtyard % Pastas % North Room % Andron % Kitchen % Other rooms % Shops %

Between SS

DF

Within SS

DF

F-ratio

Probability

1764.757 1612.688 3550.448 875.214 10539.485 12089.247 5586.780

4 4 4 4 4 4 4

3903.612 2846.655 3811.625 2191.956 2404.796 10431.336 2179.980

79 79 79 79 79 79 79

8.929 11.189 18.397 7.886 86.558 22.889 50.615

0.000 0.000 0.000 0.000 0.000 0.000 0.000

Note: These summary statistics compare the variation between clusters (‘‘Between SS’’ = sums of squared distances between clusters) to the variation within clusters (‘‘Within SS’’). ‘‘DF’’ = degrees of freedom. The F-ratio is a measure of the ratio of the variance between clusters to the variance within clusters. If this ratio is large, then houses in different clusters are more different from one another than they are from houses in the same cluster. ‘‘Probability’’ is the probability that the different clusters might actually come from the same population, i.e., that the clustering is simply the result of random variations in house design. These last figures, however, should be interpreted with caution.

289

Statistics for Individual Clusters Members House

Statistics Distance

Variable

Minimum Mean

Standard Maximum Deviation

CLUSTER 1 (‘‘typical plan,’’ fewer specialized spaces) A −1 A 10 A3 A5 A9 A iv 5 Av1 Av2 Av3 Av4 Av5 Av7 A vi 1 A vi 3 A vi 7 A vii 3 A vii 5 A vii 7 A viii 1 A viii 6 A viii 7 A viii 8 A viii 9 A xi 9 A xii 9 B vi 7 B vi 8 H. Tiled Prothyron ESH 2 ESH 3

3.90 2.35 5.32 10.83 2.47 2.56 3.12 3.25 3.67 3.88 7.45 7.27 7.64 5.11 3.62 9.58 5.56 6.70 5.18 8.02 9.16 5.57 3.36 9.55 3.75 4.37 10.26

Courtyard % Pastas % North Room % Andron % Kitchen % Other rooms % Shops %

0.00 0.00 5.18 0.00 0.00 0.00 0.00

18.11 14.95 22.75 2.22 0.55 23.48 0.40

36.70 28.25 32.61 13.88 8.41 41.72 7.72

8.11 6.71 6.32 4.15 2.07 10.31 1.56

Courtyard % Pastas % North Room % Andron %

0.00 0.00 0.00 0.00

12.21 0.00 4.90 1.05

21.63 0.00 15.34 7.35

6.96 0.00 6.27 2.57

5.80 10.52 3.37

CLUSTER 2 (irregular plans) A −5 A 11 A 12 A 13

290

7.01 6.55 6.01 4.08

Appendix 1

Statistics for Individual Clusters Members House

Statistics Distance

Variable

Minimum Mean

Standard Maximum Deviation

CLUSTER 2 (irregular plans)(continued) A iv 7 A viii 10 A viii 3

6.41 4.45 7.64

Kitchen % 0.00 Other Rooms % 45.71 Shops % 0.00

0.00 64.68 0.82

0.00 79.27 5.73

0.00 12.87 2.00

Courtyard % Pastas % North Room % Andron % Kitchen % Other rooms % Shops %

43.93 0.00 0.00 29.55 0.00 17.85 0.00

43.93 0.00 0.00 29.55 0.00 17.85 0.00

43.93 0.00 0.00 29.55 0.00 17.85 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00 0.00 13.00 14.77

8.21 9.29 6.80 1.14 8.79 29.31 25.85

20.91 28.66 12.88 6.79 22.38 54.46 35.65

6.25 7.82 3.97 2.43 10.13 12.21 6.43

13.45 13.92 13.57 4.66 24.38 20.05 4.10

26.89 30.06 28.23 21.64 41.38 46.03 26.54

5.76 4.85 7.87 6.68 5.73 11.27 6.81

CLUSTER 3 (not residential) Av8

0.00

CLUSTER 4 (houses with shops) A6 A iv 9 A v 10 Av9 A vi 9 A vii 1 A vii 9 A xii 10 B vi 9 Dv6 ESH 1

7.61 5.64 5.83 8.76 5.98 5.37 11.65 10.73 5.97 4.53 8.93

Courtyard % Pastas % North Room % Andron % Kitchen % Other rooms % Shops %

CLUSTER 5 (‘‘typical plan,’’ more specialized spaces) A1 A2 A4 A8 Av6 A vi 10 A vi 2 A vi 4 A vi 5

6.31 5.01 10.50 10.68 4.30 7.28 8.76 7.69 8.12

Courtyard % Pastas % North Room % Andron % Kitchen % Other rooms % Shops %

0.00 0.00 0.00 0.00 15.96 0.00 0.00

Appendix 1

291

Statistics for Individual Clusters Members House

Statistics Distance

Variable

Minimum Mean

Standard Maximum Deviation

CLUSTER 5 (‘‘typical plan,’’ more specialized spaces)(continued) A vi 6 5.13 A vi 8 10.92 A vii 10 3.81 A vii 2 5.43 A vii 4 6.80 A vii 6 4.85 A vii 8 7.65 A viii 2 5.99 A viii 4 7.29 A viii 5 5.11 A xi 10 6.18 A xiii 10 6.31 Bv1 8.95 B vi 1 7.48 B vi 2 8.48 B vi 3 5.36 B vi 4 6.29 B vi 5 5.54 B vii 2 7.50 ESH 4 10.99 H. of Comedian 7.62 H. of Many Colors 4.53 S. Villa 6.47 V. Bronzes 4.61 V. Good Fortune 10.03 B vii 1 6.89 Note: For each cluster, the houses which make up the group are listed, together with their distance from the centroid of the cluster (the hypothetical ‘‘average house’’ for that cluster). The distance measurement is an indication of how close each house is to the average for that cluster. Summaries of the minimum, mean, maximum, and standard deviation for each type of room are listed for each cluster to the right. These allow one to compare clusters with one another and to see in which respects they differ.

292

Appendix 1

appendix two

Sales Inscriptions from Olynthus

Priest, Month a

Findspot *A viii 3, a c

A iv 3, l

*A v 10, g

Sales Type

Purchaser

Seller

Witnesses

lease? loan? (καθίεται)

Bakchon son of Pol[– –

Zopyros son of Hipparchos

Eupa]trides D[– –, Perikles, Eu]bou[lides son of Apollodoros

Kleandros son of sale (ὠνὴ Soson, month of εὐθεῖα) Pantheon (354/3 b.c.?, month 12)

Strenios son of Aspia

Pheidippos son of Pheidon

—

Aristoboulos son sale (οὐ . ν. ή. of Kallikrates, [εὐθεῖα]) month of Apatourion (352/1? b.c., month 5)

Dionysios son of Ithuras (?)

Nikandros son of Athenion

Demarchos son of Phanoleus, etc.

Lysis son of Si[– –], month of Hippion (355/4 b.c.?, month 10?)

St. vi @ A vi 4 [Ar]istoboulos [son of Kallikrates], month of Lenaion (352/1? b.c., month 7)

sale?

[Epi]therses son of Thrasylaos

[Polyk]leos? son of Chairilos

Philippos– –, Zopyros– –, Aristoboulos

*D v 6, p

[Aristoboulos] son of Kallikrates, month of Targelion (352/1 b.c., month 11)

sale ([οὐνὴ εὐθε]ῖα)

Zoilos son of Philokrates

Diopeithes son of Antipatros

Diokles son of Charon, Euxitheos son of Xanthippos, Philon son of Theodotos

B iii 2 cistern

Aristoboulos son loan (ἐδάνεισε of Kallikrates, [ν) . month of Pasitheon (352/1 b.c., month 12)

Hermagoras son of – –Nl– – loans

Diopeithes son of Antipatros receives

The whole [loan] deposited with Polyxenos son of Telagros, Aristoteles son of Ktesios, [–]eos son of Theogenos

294

Appendix 2

Guarantors

Neighbors

Notes

Eupatrides son of Kleoboulides

Euboulides son of for five years and Apollodoros and four months Zopyros son of Hipparchos

Herodoros son of Herodoros, Athenodoros son of Aristodemos, St[r]en[ios son of Aspia?]

Telekles

Antibios son of Athenion (brother of seller, Nikandros?)

Demarchos son of Phanoleus

– –(Robinson: Isodoros?]

[Philippo]s, Straton

Polemarches son of Straton

Diokles son of Charon and the children of Apollodoros

house which Pheidippos had bought from polis of Olynthus (after confiscation?)

Price

References b

ἐνακοσίον =

1938 #5, Hatz. 61

900 dr (lease?)

τετράκις χιλίων πεντακοσίω =

1938 #6

4,500 dr

ΨΨΨΨΨ888 = includes ‘‘the pithos room and all 5,300 dr the things which bring income’’ (ὁ πιθεών and τ. ὰ. μι . [σθ]οφόρα πάντα)

1934 #4

[..] (Robinson: ΨΨ = 2,000 dr)

1931 #3

Ψ88 = 1,200 dr

1938 #3

an uninscribed stele found in room q

for the whole house τετράκις χιλίς (or estate) for three [πεν]τακ[ο]σίας years = 4,500 dr (loan)

Olynthus 2, pg. 101; Hatz. 58-59

Appendix 2

295

Priest, Month a

Findspot

Sales Type

Purchaser

Seller

Witnesses

S. Hill, Sec. G, Area 30

Kal[lippides son of Menekles], month of Pantheon (351/0, month 12)

sale?

—

—

—

A2

[Kallipides son of ] Menekl[es] (351/50 b.c.)

sale ([οὐνὴ ε]ὐθεῖα)

Epither[ses] (son of Thrasylaos? Cf. 31 #3, Hatz.)

[– –] son of Hippar[chos]

—

sale (οὐνὴ εὐθεῖ]α)

Xenon [son of Mos]chion?

Euboulides son of Harpaleus

Pharnabazos son of Eukles, Philoxenos son of Chabres, Mysios son of Timanthes

*S. Hill, NE

Antidotos son of part of Tr. 10 Polykles, month (E of Civic of Apatourion (350/49 b.c., Center) month 5)

Villa of Good Fortune (surface find above court)

Antidotos son of Polykles, month of Artemision (350/49 b.c., month 9)

sale (οὐνή)

Ainetos son of Apollodoros Andron son of Chalkon

Derkylos son of Stratios, Theoxenos son of [Menon], Tedesios son of Kalligitos

St. v @ D v 4

Antidotos son of P[olykles], month of Hippion (350/49 b.c., month 10)

sale

—

—

Di[– –

*A 11

Antidotos son of Polykles, month of Targelion (350/49 b.c., month 11)

loan? πρᾶσις Archidamos ἐπὶ λύσει? son of

Soson son of Theodoros

Polyxenos son of Telagros, Pythion son of Diodoros, Nikandros son of Epichares

296

Appendix 2

Metrichos

Guarantors

Neighbors

Notes

Price

References b

—

—

—

—

1931 #4, Hatz. 29 n. 1

—

—

—

—

1928 pg. 232; Hatz. 28

Pytheas son of Philoxenos son of Pythion, Philaina Chabres and his daughter of Heron own (house)

88ΧΧΧ = 230 dr

1928 #1; Olynthus 2, pg. 32

Philokrates son of Aresidemos

Derkylos son of Stratios, Silenos son of Glakios (2 neighbors!)

8888Χ = 410 dr

1934 #3

Kleainetos? (witness or guarantor?)

Theodoros [– – (neighbor?)

of house (οἰκίης τῆ[ς] ἐχο[μένης-]

ΓΗΗΗΗ =

1938 #4

Antidotos son of Theodoros, (Soson’s brother?), Nikon son of Edoneus

Polyxenos son of Telagros and Pythion son of Diodoros

ἐνιαυτὸν ἐπὶ ἀπόλυσιν

900 dr

ΨΨ = 2,000 dr

1931 #2

Appendix 2

297

Priest, Month a

Findspot

Skoinia, near Antidotos son of Olynthus Polykles, month of Pantheon (350/49 b.c., month 12)

Sales Type

Purchaser

Seller

—

Witnesses Phormio son of Drimakos, Chairikles son of Euphraios, Dionysios son of Straton

Findspot of Philip Inscription

Antidotos son of Polykles (350/49 b.c.)

sale (οὐνή . )

—

—

—

Ave. A @ A −4

—

Loan (δανείζει .)

Lykophron son of Theodoros loans

Hegias and Anthes the children of Hegestratos receive

—

nr. B vi 9

—

—

—

—

– –]os son of Kleon, An[–]ros son of Timoxenos

*A iv 7

– –month of Herak[leios] or Herai[on] (Knoepfler)

sale?

Moiragenes?

—

—

a Dates

are given by the eponymous priest and the month (e.g., in the first inscription, ‘‘when Antidotos son of

Polykles was priest, month of Apatourion’’). Years and months according to Hatzopoulos, Actes de Vente 80. b ‘‘1928 #1’’ = Robinson, ‘‘Inscriptions, 1928’’ #1, etc.; Hatz. = Hatzopoulos, Actes de Vente. c * = inscription found in situ in house.

298

Appendix 2

Guarantors

Neighbors

Notes

Lykophron son of Theodoros

Price

References b

ἑκατὸν ἐβδομήκον τα = 170 dr

ADelt 39 (1984) B 226; Bull. Ep. 104 (1991) 504

—

—

—

—

1934 #5

—

—

‘‘except for the seven-couch room and . . . and the posessions in the village and the pithos room’’ (τοῦ . [ἑ]πτακλίνου καὶ . τοῦ [.]. ΝΩΝΟΣ καὶ

δισχιλίας πεντήκοντα = 2,050 dr (loan)

1934 #6; Hatz. 60 and forthcoming

—

1938 #7

ΨΨ = 2,000 dr

1938 #9

τοῦ κτήματος τοῦ ἐ.πὶ . τὴν κώμην ΕΣΣΤ[... καὶ τοῦ πιθε[ῶνο]ς – –son of Bitton, – –son of Demosthenes (guarantors?)

—

—

Ariston

—

Appendix 2

299

Notes

preface 1. Pol. 1253b. 2. Olynthus 1–14.

chap ter 1: greek city planning 1. Among the more general recent studies are Hoepfner and Schwandner, Haus und Stadt; Martin, L’Urbanisme; Sakellariou, Polis-State. 2. Malkin, Religion and Colonization 136. 3. Aristoph., Birds 992–1022. 4. Pol. 1267b–1269a; also Pol. 1330b. Modern works on Hippodamus are legion: see esp. von Gerkan, Städteanlagen 42–61; McCredie, ‘‘Hippodamos’’; Falciai, Ippodamo; Castagnoli, Orthogonal Planning 65–72, 128–38; Wycherley, ‘‘Hippodamus’’; Martin, L’Urbanisme 15–16, 103–6; Bruno, ‘‘Hippodamus and the Planned City’’; Gorman, ‘‘Aristotle’s Hippodamus.’’ 5. A number of ancient sources describe Hippodamus as an architect, such as Strabo (14.2.9), Harpocration, the Lexica Segueriana, Photius, and the Suda (see Falciai, Ippodamo 28). But these sources are all far removed in time from Hippodamus and may reflect later evaluations of his role in the planning of actual cities, rather than as a theorist. Aristotle considers him as at least having pretensions to philosophy, and Aristophanes, if his Meton is really a parody of Hippodamus, considers him a cosmologist, a ‘‘Thales,’’ rather than a mere architect (Birds 992–1020). See Gorman, ‘‘Aristotle’s Hippodamus.’’ 6. Hesychius and Photius record that he emigrated to Thurii, where he is assumed to have been involved with the design of that city. Diodorus, however, does not mention Hippodamus in his description of the founding and layout of Thurii (12.10), and no ancient author specifically attributes the plan of that city to him. Most modern scholars allow that he planned the newly synoikized city of Rhodes in 408/7 b.c., although Strabo says

301

only that the new city was ‘‘founded’’ (ἐκτίσθη) by the same architect who did the Piraeus (Strabo 14.2.9). Hippodamus used to be credited with the replanning of his home town, Miletus, when it was rebuilt in 479 b.c., or at least with taking part in the replanning (as von Gerkan, Städteanlagen 45–46; followed by Hoepfner & Schwandner, Haus und Stadt 17–22, 302). The chronology makes this problematic, however: see, for instance, Bruno, ‘‘Hippodamus and the Planned City.’’ 7. Gorman, ‘‘Aristotle’s Hippodamus.’’ 8. Martin, L’Urbanisme 16, Hoepfner & Schwandner, Haus und Stadt 301–10, and others attribute to Hippodamus a strong Pythagorean influence; cf. also Lévêque and Vidal-Naquet, Clisthène 128 and n. 2. But ideas about numerology and harmony were neither exclusively Pythagorean nor Milesian. They were more widely held, and it seems to me fruitless to attempt to trace Hippodamus’s philosophical background from such meager evidence. 9. Aristotle here speaks only of agricultural land; all citizens probably owned their own house plots in the city. Hippodamus did not state how the public land will be worked to produce the soldiers’ food, and Aristotle criticizes him for this (Pol. 1268a). But Aristotle himself recommends that public land be worked by slaves rather than citizens (Pol. 1329a), and state-owned slaves may have worked the public land in Hippodamus’s state. See Aristotle’s discussion of communal property and of Plato’s Republic, Pol. 1260b ff. 10. Pol. 1268a. 11. Hesychius makes Hippodamus divide (διεῖλεν) the Piraeus; Photius makes him distribute (διένειμεν) it. The different terminology used by these authors is interesting: all three words, ‘‘cut up,’’ ‘‘divide,’’ and ‘‘distribute’’ are commonly used in other contexts to describe different stages in the process of organizing a city and its territory. 12. IG I 3 1101–1115; SEG X, 376–383; Hill, ‘‘Boundary Stones’’; McCredie, ‘‘Hippodamos’’ 97; Garland, Piraeus 140–41, 225–26; Lewis, ‘‘Public Property’’ 250–51. 13. IG I 3 1111–1113. 14. The term νέμησις itself is used elsewhere in connection with land division only by Isaeus 9.17, in describing a quarrel which took place over the distribution of land (νέμησις τοῦ χωρίου). But related words meaning ‘‘to distribute the land’’ are fairly often used, for instance in the Brea inscription IG I 3 46 (ML 49): γεονόμοι . . . hοῦτοι δὲ νεμάντ[ον τὲν γ῀εν (‘‘Land distributors . . . let these men distribute the land)’’: by Plato, Laws 737C, 745E; Aristotle, Pol. 1331a; Aristotle, Ath. Pol. 21.2–6. 15. McCredie, ‘‘Hippodamos’’ 97. 16. IG I 3 1111 is said to have been found in Odos Makra Stoa, in the northern part of the city near the city wall. 17. McCredie, ‘‘Hippodamos’’ 97. 18. Von Eickstedt, Piräus; Hoepfner and Schwandner, Haus und Stadt 22–50; Garland, Piraeus 144–45. 19. On the Laws in general: Morrow, Plato’s Cretan City; Piérart, Platon; Jones, ‘‘Kretan City.’’ 20. Morrow, Plato’s Cretan City 3–13, 95–152, and passim. Plato had some practical experience in lawmaking and politics: he advised Dionysius II of Syracuse for a brief period and may have even helped to draft legislation for Phoebia and Tauromenium. According to various traditions, he was also invited to draw up laws for Megalopolis and Cyrene; and

302

Notes to Pages 3–5

his students at the Academy were certainly involved with the politics and legislation of contemporary states. See Morrow 8 and n. 11. 21. Laws 704A–707D, cf. 745B. 22. Described as a colony (ἀποικία) in 702C, 744B, 753A, 754C, etc., but as a synoikism in 664A, 708A, 739B, 746A, 752E, and other passages, although perhaps not always in the technical sense of the word. Cleinias is described as the oikist, the sanctified leader of a colony, in 753A. 23. On the difficulty of a population accepting a new set of laws, cf. 681B, 752B-C. The question of what sort of laws to adopt was one that any new community, and particularly a mixed community such as this one, would have faced. The laws and customs are occasionally specified in descriptions of colonies. For instance, Himera, founded jointly by Chalcidians and fugitives from Syracuse, had a mixed Chalcidic and Doric language but adopted Chalcidic institutions (Thuc. 6.5). 24. Pol. 1301b, and generally in book 5 of the Politics. 25. Laws 744B-C; see England, Laws, ad loc., and Morrow, Plato’s Cretan City 131–38 on this difficult passage. On ‘‘proportional inequality’’ see 757B-C: this ‘‘truest and best form of equality . . . dispenses more to the greater and less to the smaller.’’ Cf. also Plato, Gorgias 508A-B; Aristotle, Pol. 1301B; Aristotle, Nic. Eth. 1131b, 1158b. 26. Laws 741A-D; 744E. 27. Laws 742A-B. 28. Laws 737E; 746D; esp. 771B: ‘‘We must consider anew the number 5,040, and the number of convenient subdivisions which we found it to contain both as a whole and when divided up into tribes: the tribal number is, as we said, a twelfth part of the whole number, being in its nature precisely 20 × 21. Our whole number has twelve subdivisions (διανομάς) and the tribal number also has twelve; and each such portion must be regarded as a sacred gift of God, conformed to the months and to the revolution of the universe.’’ Aristotle objects that ‘‘a territory as large as that of Babylon will be needed for so many inhabitants, or some other country as unlimited extent, to support five thousand men in idleness and another swarm of women and servants around them many times as numerous’’ (Pol. 1265a). 29. In practice of course neither Plato nor anyone else would expect the population and number of households to stay absolutely stable. Plato does suggest various means of keeping the number of hearths constant (740A–741A). Only one son will be allowed to inherit the allotment of land (kleros); plots cannot be subdivided. He leaves it up to the discretion of the magistrates how to deal with families without male issue or families with more than one heir. He suggests the possibilities of adoption, birth control, and also that surplus population may, in ‘‘that ancient device’’ (τὸ παλαιὸν μηχάνημα), be sent out as colonists; while a shortfall might, in emergency, force the state to accept new citizens. Pheidon of Argos also thought that the number of households and of citizens should remain constant, although the kleroi they held were not equal in size (Pol. 1265b). 30. J. K. Anderson suggests ‘‘levies’’ for ἀγωγάς. 31. Laws 745B-E, my translation. 32. Many commentators interpret the ἀφ’ οὗ as stipulating a radial arrangement, with roads diverging from the center (for instance England, Laws 536, 569; Morrow, Plato’s Cre-

Notes to Pages 6–10

303

tan City 122, 183; Piérart, Platon 21–25; Lévêque and Vidal-Naquet, Clisthène 134–46). It is not clear, however, whether Plato intended the state to be physically laid out in concentric circles, or whether the organization of land was simply conceived as a circle with its ‘‘spiritual center’’ at the acropolis but then executed in a rectilinear fashion, as was standard in Greece from the eighth century on. The moats around Plato’s Atlantis in the Critias are circular, but the city itself is not necessarily so, and except for Aristophanes’ satirical Cloudcuckooland in the Birds (992–1020), no Greek city was ever laid out radially. Mantinea and some other cities have nearly circular walls, but the street plan within the walls was rectilinear, not radial. England and Morrow suggest that ‘‘the boundaries marking the divisions should radiate outward from the center of it [the city]; in this way each division will be a continuous area from the acropolis of the city to the borders of the state, including land within the city proper and in the country outside’’ (Morrow, Plato’s Cretan City 122). But Plato consistently distinguishes between πόλις and χώρα and also between the processes of division of city and territory—and he expressly demands that the urban core (ἄστυ) be divided into twelve sections, a process which would be unnecessary if each section already contained a part of the ἄστυ (see 848E also). These are two separate processes, and I therefore believe it more likely that the twelve μέρη of the city and of the country are parallel, but not contiguous. It is unclear why the idea of dividing land into twelve parts is repeated twice in Laws 745B: μετὰ δὲ ταῦτα μέρη δώδεκα διελέσθαι, and ἀφ’ οὗ τὰ δώδεκα μέρη τέμνειν τήν τε πόλιν αὐτὴν καὶ πᾶσαν τὴν χώραν. Διελέσθαι and τέμνειν seem to be parallel in meaning here, ‘‘to divide’’ and ‘‘to cut.’’ The first phrase seems to introduce the general idea of division of territory, which is pursued until 745E, while the second repeats and specifies what exactly is to be cut up: both the city and the country. The phrasing is a bit awkward, however; one must remember that the Laws is a posthumous and partly unfinished work. 33. This is the usual interpretation and consistent with what follows, for instance, that each tribe will have its own religious festivals. The passage is ambiguous because Plato does not say that ‘‘they must allocate the twelve sections to the twelve gods’’ but ‘‘he must allocate twelve sections to the twelve gods’’—and these could in theory be separate sanctuaries rather than the sections of land. The interpretation offered here and elsewhere is probably the correct one, however. Morrow (121–24) and Piérart (66–68) accept the distribution of tribes to the sections of land, although admitting that it is problematic. A later passage seems also to imply that the tribes are territorially distributed: ‘‘We have marked out the whole country as nearly as possible in twelve equal portions; to each portion one tribe shall be assigned by lot, and it shall provide five men to act as land-stewards and phrourarchs’’ (760B). The manuscripts, however, have κατ’ ἐνιαυτόν after ἐπικληρωθεῖσα: ‘‘to each portion one tribe shall be assigned by lot each year.’’ These words are bracketed by England and others, however, since they are not reconcilable with the following description of the organization of phrourarchs and land stewards, for instance that they serve two-year terms. 34. Laws 848E. 35. Laws 848C-D; 760B-760C. 36. Laws 738D; 771D. 37. Uguzzoni & Ghinatti, Eraclea 165–218, esp. the tables pp. 180, 200. See also the discussion of Metraux, Land-Use 59–75, esp. 69–71. The tablets record the amounts of arable

304

Notes to Pages 10–11

and barren land at some time subsequent to the original division (which was presumably when the colony was founded in 433 b.c.), and the discrepancy between the amount of arable land belonging to the two deities may have arisen after the original division. 38. Polybius 6.45; Busolt, Staatskunde 634. 39. Laws 778C–779D. 40. Aristotle, Nic. Eth. 1170b. 41. Pol. 1326a-b. 42. Pol. 1328a-b. The list of six occupations is different from that given in Pol. 1290b1291a, which enumerates eight classes: farmers, the working class (τὸ καλούμενον βάναυσον), the commercial class (τὸ ἀγοραῖον), serfs, the military class, . . . [a possible lacuna in the text], a propertied class, and those who serve in the magistracies (τὸ περὶ τὰς ἀρχὰς λειτουργοῦν). 43. Aristotle distinguishes between ‘‘parts’’ ( μέρη) and ‘‘indispensable conditions’’ (ὧν ἄνευ τὸ ὅλον οὐκ ἂν εἴη): ‘‘not all the things that are necessary for the state to possess are to be counted as parts of a state’’ (Pol. 1328a). 44. Pol. 1329a. The usual Greek attitude was that farming was the ideal occupation, but other sorts of labor were to be avoided: see for instance Xen., Ec. 4–5; Plato, Laws 743D, 842C-D, 846D-E, 919D-920A, etc.; Isocrates 7.44–45, etc., and chapter 6, below. In the ideal state of the Politics land owned by citizens is to be worked by slaves and foreign serfs (περίοικοι βάρβαροι, Pol. 1329a). Aristotle develops his argument about the exclusion of the working classes (οἱ βάνουσοι) from citizenship throughout book 3, on citizens and citizenship, particularly in Pol. 1277b ff. 45. Plato in the Republic denied the Guardians the right to own property, instead vesting it in the farming class; Aristotle here does just the reverse. 46. Pol. 1331a. 47. Aeneas Tacticus 3.1–5; SIG3 961. See Jones, Public Organization 303; Wilhelm, Beiträge 183–87, Anz. Akad. Wien 1924, 116–17, 149–50. Stratonikeia: Sahin, Inschriften von Stratonikeia 127, no. 1004. 48. Pol. 1330a—1331b. Of modern discussions, see in particular Kondis, ‘‘Εὔτομος’’; McCredie, ‘‘Hippodamos.’’ 49. Pol. 1327a. 50. Pol. 1330a ff. Many of Aristotle’s recommendations seem to be drawn from the Hippocratic Airs, Waters, Places. 51. Pol. 1330a. Aristotle does seem to imply a physical, prescriptive division of the land into different areas rather than a descriptive distinction between public land and private land, without a physical boundary between them. 52. See Martin, L’Urbanisme 21–22; von Gerkan, Städteanlagen 53–54; McCredie, ‘‘Hippodamos’’ 96. Kondis, ‘‘Εὔτομος’’ 256 takes the more traditional view that the Hippodamian and the newer manner are identical. 53. Kondis, ‘‘Εὔτομος’’ 255–58. 54. On the quincunx, see Newman, Politics ad loc.; Kondis, ‘‘Εὔτομος’’ 258–67; and Kondis, ‘‘Thurioi’’ 108–13. 55. Boersma, ‘‘Goritsa’’ 66–68. Boersma suggests that Aristotle is describing an alternating north-south and east-west orientation of the grapevines, which he claims can still be seen today.

Notes to Pages 11–18

305

56. Boyd, appendix to Williams, ‘‘Stymphalos 1983’’ 186; Williams, ‘‘Stymphalos 1984’’ 216–18. 57. Jones, Public Organization. 58. Ehrenberg, ‘‘Foundation of Thurii’’; Rutter, ‘‘Foundation of Thurii’’; Vallet, ‘‘Avenues, quartiers et tribus à Thourioi’’; Lombardo, ‘‘Di Sibari a Thurii.’’ 59. Vallet, ‘‘Avenues, quartiers et tribus à Thourioi’’ is the most complete and balanced treatment of this passage. He points out other difficulties in the text, such as the expression τὴν δὲ πόλιν διελόμενοι κατὰ μὲν μῆκος εἰς τέτταρας πλατείας. Following Kondis, he suggests that this phrase, and the general description of Thurii, are the result of a somewhat confused compression of a longer description; the odd phrase διελόμενοι . . . εἰς τέτταρας πλατείας results from the conflation of two ideas, division into ‘‘sections’’ (εἰς μέρη) and division by means of avenues (διὰ πλατειῶν). On Diodorus’s sources, see Rutter, ‘‘Foundation of Thurii.’’ See also Castagnoli, ‘‘Urbanistica di Thurii’’; Kondis, ‘‘Thurioi’’; McCredie, ‘‘Hippodamos.’’

chap ter 2: history and ar chae ol o gy at olynthus 1. Theophrastus, De Causis Plantarum 1.20.4; Historia Plantarum 8, 11, 7; Pliny, NH 18, 12; Varro, De Re Rustica 1, 44, 3. 2. Olynthus 8, esp. 1–51; subsequent discussions include Hoepfner and Schwandner, Haus und Stadt 68–113 (who, however, reconstruct some features of the houses and architecture for which there is little or no evidence); Nevett, House and Society 53–61. 3. As pointed out by Wycherley, How the Greeks Built Cities 23. 4. Plato, Laws 779B. 5. Olynthus 8, 39–42. 6. Fortification wall on ESH: Olynthus 8, 42. The houses further north appear in the schematic plan, Olynthus 12, pl. 271, but were not added to the detailed fold-out map in that volume (pl. 272). They are recorded in three unpublished plans preserved in the library of the University of Mississippi. 7. Also recorded in the same unpublished plans in the University of Mississippi. 8. Olynthus 12, 310. 9. Olynthus 12, 170; 318–22. 10. Olynthus 8, 43–44. 11. Olynthus 12, 309–12, pl. 251, 2. 12. Olynthus 2, 16–28; McDonald, Political Meeting Places, 231–36. 13. Olynthus 2, 28–29. 14. Olynthus 12, 79–82. The south end of the open area was not precisely located. A trial trench was dug through this area in 1928, but the field notes and publications are meager. 15. Olynthus 8, 21–22; Hoepfner and Schwandner, Haus und Stadt, 78–79. 16. Olynthus 2, 6. 17. Robinson, ‘‘Inscriptions, 1934’’ 103; Olynthus 8, 1 n. 2. 18. Olynthus 2, 16–23. 19. Olynthus 2, 11–14; Crouch, Water Management 171–76; Hoepfner and Schwandner, Haus und Stadt 80–81.

306

Notes to Pages 18–33

20. Olynthus 12, 95–114. 21. Olynthus 12, 125–27. 22. Olynthus 1. A new survey by Bernard Hänsel is aimed at elucidating early remains in this region; results have not yet been published. 23. Neolithic period: Olynthus 1. The date of the earliest Archaic Greek habitation of the area is based on local rather than imported pottery, whose chronology is still somewhat uncertain; see Olynthus 5, 15–61. 24. Hdt. 8.127. 25. Hdt. 8.127, 7.122. On the relationship between the Thracian Chalcidians and the inhabitants of Chalcis on Euboea, see Knoepfler, ‘‘Calendar of Olynthus,’’ who supports the literary tradition (Strabo 7, fr. 11, Strabo 10.1.8, etc.) over Zahrnt’s denial that there was a relationship between Chalcis and the Thracian Chalcidians (Zahrnt, Olynth 12–27). 26. A large number of coins found at the site use the same types as the Chalcidic coinage (Apollo/lyre or Apollo/tripod) but with the legend ΒΟΤΤΙΑΙΩΝ, ‘‘of the Bottiaeans,’’ rather than ΧΑΛΚΙΔΕΩΝ, ‘‘of the Chalcidians.’’ That these coins were minted at Olynthus itself, and not at another site, is proven not only by the great quantities of these coins found there but also by die-links between Chalcidic and Bottiaean coins: Olynthus 9, 300– 304. 27. Olynthus 5, 15; Olynthus 12, 297. 28. Olynthus 2, 27–28; Olynthus 12, 297–300; see also Crouch, Water Management 175; Gallant, Risk and Survival 179. 29. IG I 3 259–272; see Zahrnt, Olynth 31–48; Meiggs, Athenian Empire 538–61. 30. The account of Diodorus (Diod. Sic. 12.34.2) is derived from Thucydides and adds little to our understanding of the event. Diodorus uses the verb συνοικίζειν (‘‘settle together,’’ ‘‘form a single community’’) rather than ἀνοικίζειν (‘‘move upland or inland’’) to describe the move. General discussions of the historical problems of this event include West, History of the Chalcidic League 14–31; Gude, History 10–12; Olynthus 9, 112–29; Gomme, Commentary ad loc.; Larsen, Greek Federal States 62–78; Zahrnt, Olynth 49–66; Moggi, Sinecismi 173–89; Moggi, ‘‘Lo stato dei Calcidesi’’; Demand, Urban Relocation 74–83. 31. See Kahrstedt, ‘‘Chalcidic Studies’’; ATL 3, 322; Zahrnt, Olynth, 49–57. 32. Strabo 7, frag. 11: ‘‘for the Chalcidians of Euboea also came over to the country of the Sithones and jointly peopled about thirty cities in it, although later on the majority of them were ejected and came together into one city, Olynthus; and they were named the Thracian Chalcidians.’’ Strabo does not describe a voluntary ἀνοικισμός but says that they were ‘‘thrown out’’ (ἐκβαλλόμενοι) of their original homes. His account may therefore reflect a different tradition from that of Thucydides and Diodorus, based on the realities of war continuing over many years, forcing people to abandon the smaller coastal towns, and on the subsequent expansion of the Chalcidic League in the fourth century, rather than the single act of rebellion in 432 b.c. 33. Assessments: IG I 3 71, 77; ATL 1 A9, A10; see Meritt, ‘‘Restoration in IG I, 37.’’ In general, West, ‘‘Thucydides V,18,5’’; Zahrnt, Olynth 54–5, 72–9; Moggi, Sinecismi, 178–79. Before the anoikismos, Mecyberna paid one talent (6,000 dr) until 447, two-thirds of a talent from 446 to 436 and one talent again from 436 to 432; Singus paid four talents in 451 and 450, two from 447 to 439, three in 434, two in 433, and one talent in 432; Gale paid half a talent in 433 and 432.

Notes to Pages 33–36

307

34. That the towns were assessed at all in 425 and 421 shows that there were some inhabitants left. In 420 b.c. Mecyberna was in the hands of an Athenian garrison when the Chalcidians attacked and took the town in a sudden onrush (Thuc. 5.39). 35. Thuc. 5.18.5. West proposed that Σαναίους of the manuscripts should be emended to Γαλαίους, because Gale rather than Sane is assessed with the insignificant ten drachmas, because Sane is geographically removed from Mecyberna and Singus while Gale would be logically grouped with them, and because the tribute history of Gale is similar to those of Mecyberna and Singus, both before and after 432 as well as in the assessments of 425 and 421 (West, ‘‘Thucydides V,18,5’’). The emendation is accepted by Zahrnt, Larsen, the authors of ATL, and others. Gomme and Moggi, however, still prefer Σαναίους (Gomme, Commentary ad loc.; Moggi, ‘‘Sane o Gale’’; Moggi, Sinecismi 178–79, Kahrstedt, ‘‘Chalcidic Studies’’ 431 ff. 36. Thuc. 2.70. 37. Thuc. 4.120–23. 38. Thuc. 2.16–17 describes the overcrowding brought by this immigration into Athens. The very strategy of abandoning small coastal towns and concentrating the Chalcidians in one defensible city is analogous to Pericles’ decision to concentrate the defense of Attica on the urban center of Athens and abandon the countryside, and it probably entailed many of the same problems, both political and social. 39. Zahrnt uses the figure for 434 b.c. as typical. In this year the three states paid a combined tribute of 3.5 talents. The average tribute of these three cities from 454 to 432, however, was about 2.8 talents. Moreover, these coastal towns may have had other resources calculated into their tribute assessments, such as fishing, maritime trade, and other opportunities which would not have been available to Olynthus, and therefore their populations may have been slightly lower than the quotas might suggest. Zahrnt does not take into consideration such other cities as Pilorus and Assera, however, nor the influx of country-dwellers or refugees from Potidaia, Mende, or Scione, and these must have contributed significantly to the population increase. 40. Zahrnt suggests that the population of Olynthus might have doubled or tripled in the anoikismos (Zahrnt, Olynth 55). Hoepfner and Schwandner, who include Sermylia together with Mecyberna, Singus, and Gale, estimate a fivefold increase (Hoepfner and Schwandner, Haus und Stadt 76). 41. Nixon and Price argue persuasively that tribute was assessed according to the resources available to a community, including mines, harbors, and the like, rather than population or territory alone (Nixon and Price, ‘‘Size and Resources’’). The lowest estimate, of 750 people per talent of tribute, is from Pounds, ‘‘Urbanization’’ 140–43, based on the carrying capacity of the landscape around the Chalcidice. This is a very dicey form of evidence, however. Zahrnt reaches a figure of 3,000–4,300 people (Zahrnt, Olynth 137–38), while Ruschenbusch suggests that one talent of tribute was levied on 800 adult citizens, for a total population of 6,400 inhabitants (Ruschenbusch, ‘‘Tribut und Bürgerzahl’’ 142; cf. Ruschenbusch, ‘‘Bevölkerung,’’ Ruschenbusch, ‘‘Zahl der Staaten’’). 42. Estimates of the proportion of citizens living in the country range from about 5 percent to about 26 percent. See below, chapter 6, ‘‘The Agricultural Economy.’’ 43. Russell, ‘‘Late Ancient and Medieval Population,’’ followed by Pounds, ‘‘Urbanization’’ 142, Jameson, ‘‘Southern Argolid’’ 85, Hodkinson, ‘‘Mantineia’’ 263, etc. This allows

308

Notes to Pages 36–38

about 5–6 people per house in the planned area of the city, a reasonable average for the Greek household (below, ‘‘The Extent and Population of Olynthus in the Anoikismos’’). A block of 10 houses together with the streets surrounding them occupy ca. 0.38 ha, and so would house about 57 people according to this estimate. In Messenia, the Minnesota expedition calculated modern village densities at about 112 persons per ha, while suggesting that ancient densities were somewhat higher (Minnesota Messenia Expedition 254–55). Dákaris estimates a slightly higher figure of about 200 persons per hectare (Dákaris, Cassopaia 108, para. 419). 44. Discussion in Olynthus 8, 13–14. This interpretation, that at least the initial plan of the North Hill dates to this time, has been accepted by most scholars (e.g., Hoepfner and Schwandner, Haus und Stadt 71–73) and to my knowledge has not been seriously challenged. Martin is more cautious, merely suggesting a date ‘‘avec beaucoup de vraisemblance entre les années 440 et 430 av. J.-C.’’ (Martin, L’Urbanisme 110). 45. Olynthus 6, no. 144; Olynthus 9, 336–38. 46. Lamps of Group VI: Howland, Greek Lamps 63–66; Blondé, Thorikos Lamps 88–94, and similar lamps from Himera predating the destruction of 409 b.c., Himera II 330, nos. 89–92. The deposit contained four lamps of this type as well as two earlier types: Olynthus 14, nos. 12, 30, 33, 35, 38, and Group IV no. a. The terracottas are Olynthus 14, nos. 107, 163, 280, and 326; also found was a pendant, Olynthus 10, no. 445. 47. Hoard 5, Olynthus 9 173–76; Olynthus 8, 13–14. 48. The series of coins of Perdiccas is dated by Raymond to 445–432 and must in any case have been minted before the monarch’s death about 413 b.c. (Raymond, Macedonian Royal Coinage 136–66). The coins of Acanthus are not so closely datable, but were minted on the heavier ‘‘Phoenician’’ or ‘‘Thraco-Macedonian’’ standard rather than the Attic standard used previously by that city. The change of standard has been traditionally attributed to the activities of Brasidas in 424 b.c., although this is at best a guess; Kraay, for instance, suggests that the Athenian Coinage Decree (440s?) would provide as convincing an explanation for the change, although the date of the decree is not fixed either (Gaebler, Antiken Münzen Nordgriechenlands 27–28; Kraay, Archaic and Classical Greek Coins 135–36). Clement dates the Chalcidic coins of Groups B and C to 430–421, based on the number of dies and die combinations among these early issues, their resemblance to coins of Perdiccas, and on his argument that the Chalcidians began to mint coins shortly after 432 b.c. (Olynthus 9, 1–10, 112–29). Westermark, however, suggests that the hoard may date in the early fourth century, based in part on the preponderance of coins of Acanthus in the hoard, and on the relative wear of Chalcidic and Macedonian royal issues (Westermark, ‘‘Coinage’’ 95). These arguments do not seem as compelling to me as the earlier date, and of course they do not show that this house had not been built earlier. 49. See Olynthus 8, 116, 121–22, and pl. 41:1. The north wall of the two-room building runs under the pavement of the court of A vii 5, showing that the eastern part of the earlier structure was torn down when A vii 5 was built. The walls of A vii 3 and A vii 5 abut the north wall of the building. The original purpose of the building is difficult to discern. A jug, a composite ‘‘dumbbell vase,’’ and a marble stand or louter base were found in a deep stratum (at 1.0–1.1 m, about 0.4–0.5 m below the final floors of these houses), but these do not suggest what the use of the structure might have been. 50. Robinson and Graham calculate 320 houses (Olynthus 8, 43). This, however, ex-

Notes to Pages 38–40

309

cludes the valley between the East Spur Hill and the South Hill. Although the trace of the fifth-century fortification is uncertain, it may well have enclosed the East Spur Hill and the valley to its west, closing off the mouth of the valley and joining the defenses of the South Hill. This would enclose about 500 houses. 51. Gallant, Risk and Survival 27–33. 52. Olynthus 8, 43. 53. Hoepfner and Schwandner, Haus und Stadt 72. 54. Raeder, Priene: Funde 14. 55. For the possibility of, for example, ‘‘Singians living in Olynthus,’’ cf. the Teians who fled their homes in Asia Minor and moved to Abdera in Thrace, but could nevertheless be referred to as ‘‘Teians living in Abdera’’ (Hdt. 1.168; Herrmann, ‘‘Teos und Abdera’’). 56. Zahrnt, Olynth 80 ff.; Hampl, ‘‘Olynth’’; Kahrstedt, ‘‘Chalcidic Studies’’ 427 and n. 52. 57. While Xenophon, like other fourth-century writers, such as Demosthenes, speaks of the ‘‘Olynthians,’’ he must mean the state officially known as the Chalcidians. 58. Olynthus 9, 112–29; Larsen, Greek Federal States 58–78; Beck, Polis und Koinon 146–62; West, History of the Chalcidic League 14–31; Moggi, Sinecismi 183; Hampl, ‘‘Olynth’’; Gude, History 18–23; Zahrnt, Olynth 56–79; Demand, Urban Relocation 74–83. 59. Tod, GHI no. 111; Zahrnt, Olynth 122–24; Borza, ‘‘Timber and Politics.’’ 60. Xen., Hell. 5.2.11–5.3.26; IG II 2 36; Tod, GHI no. 158. 61. In the Tribute Lists and Assessments, the Bottiaeans, who had certainly formed a federal league by 422, are nevertheless listed city by city. Likewise, the Chalcidians are listed city by city, even listing cities from which the Athenians could not possibly have expected tribute, rather than acknowledging the existence of a league or other organization. See Kahrstedt, ‘‘Chalcidic Studies.’’ Thucydides describes the state as ‘‘Olynthians’’ twice, at 5.3.4 and 5.39.1; see Gomme, Commentary 204–5. 62. Zahrnt argues that the Chalcidians were organized into a unitary state in the fifth century, which expanded into a federated league in the fourth (Olynth 57–99, 80–90). West, Larsen, and others, in contrast, believe that the anoikismos of 432 resulted in a league rather than a unitary state. 63. Olynthus 9, 112–29; Clement, ‘‘Beginning of Coinage.’’ 64. See Westermark, ‘‘Coinage’’; Rose, ‘‘Coins and History’’; Rose, ‘‘Reconsideration.’’ 65. Tod, GHI no. 68; IG I 3 76. Cf. Gomme, Commentary 1, 207; Larsen, Greek Federal States 76–77. 66. One should not make too much of the investment of time and labor in building houses in one place or another, and conclude that because they built permanent houses at Olynthus, the immigrants intended to remain there forever. Greeks were quite mobile and under certain circumstances were willing to relocate their homes. One striking example is the dissolution and refoundation of Mantinea, during which the inhabitants were forced by the Spartans to destroy their urban homes, build new dwellings in the country (a move which they did not like at first, but which the aristocracy later came to prefer), and then a mere fifteen years later, they abandoned those country homes (or at least many of them) and rebuilt in the city in a second synoikismos (Xen., Hell. 5.2.1–7, 6.5.3–5; Hodkinson, ‘‘Mantineia.’’ In general, Purcell, ‘‘Mobility and the Polis’’; Demand, Urban Relocation). 67. References in Zahrnt’s useful ‘‘Städteliste’’ (Olynth 131–254).

310

Notes to Pages 41–44

68. Xen., Hell. 5.2.14–15, 5.3.1–3. 69. Dem. 19.263. 70. Xenophon’s account of the Olynthian cavalry being driven back to the very walls of the city by Derdas after their raid on Apollonia suggests that this body of 600 were all Olynthians, rather than a mixed body of Chalcidic horsemen (Xen., Hell. 5.3.1–3). And Demosthenes qualifies his account of the relatively small force at the time of the Lacedaemonian expedition (compared to the might of the Chalcidians when Philip attacked, successfully) with ‘‘for the whole Chalcidians had not yet synoikized into one body’’ (οὔπω Χαλκιδέων πάντων εἰς ἓν συνῳκισμένων)—perhaps implying that he is referring to the Olynthians alone. But it is not clear what synoikism he is referring to—certainly not the anoikismos of 432, but perhaps an otherwise unattested change in constitution following their defeat by the Lacedaemonians. And for the later period, Demosthenes adds, ‘‘though all their neighbors were in alliance with them.’’ This might refer to the neighbors of the Chalcidic League: Thracian tribes and the like, rather than the Chalcidic cities neighboring the city of Olynthus. None of these arguments seem conclusive, and the imprecise usage of Xenophon, Demosthenes, and other authors prevents us from deciding the issue one way or the other. Various people have doubted these figures, claiming that they are manifestly too small, and the number of hoplites is disproportionately small compared to the number of cavalry. The 800 cavalry is often amended to 8,000 (e.g., Hammond and Griffith, History of Macedonia 2, 176 n. 1). On the issue, West conjectures that the 800 hoplites represent the contingent of the city of Olynthus rather than of the whole Chalcidic League, while the 600 cavalrymen belong to the whole League, not just the Olynthians (West, History of the Chalcidic League 159–65). Zahrnt accepts that the figures reflect Chalcidic, rather than merely Olynthian, troops, but points out that the cavalry was particularly powerful and therefore probably disproportionately numerous, while the light-armed troops may make up much of the difference between Xenophon’s figures and Demosthenes’ total of 5,000 (Zahrnt, Olynth 93–94). 71. Olynthus 12, 175–76. Hoepfner and Schwandner propose a similar scheme of development, but they suggest that the houses around the perimeter of the city inside the city wall, including Row A and the houses along the East Spur Hill, belong to a separate phase of construction about 410 b.c. (Hoepfner and Schwandner, Haus und Stadt 71–77). There is, however, no evidence that this row is later than the regular blocks. The houses were built against the city wall and so postdate its construction, but one would not necessarily expect them to bond house walls with the defensive circuit wall, so the abutment need not imply a long time lag between the two constructions. The houses in Row A', on the other hand, were built against and over part of the city wall and the round bastion at the northern tip of the hill and so postdate the fortifications (Olynthus 8, pl. 12, and fieldbooks). 72. Olynthus 9, 176–79. 73. Cameron and Tonka, Abandonment of Settlements and Regions. 74. See Cawkwell, ‘‘Defense of Olynthus’’; Gude, History 34–37; West, History of the Chalcidic League 126–37; Hammond and Griffith, History of Macedonia 2, 296–328. 75. Diod. Sic. 16.52.9. 76. Dem. 9.11. 77. Dem. 19.267.

Notes to Pages 44–46

311

78. E.g., Dem. 19.196. 79. Dem. 9.26. 80. Olynthus 10, nos. 1907–1911, 2186–2201, 2228–2240. As Robinson noted, Philip’s slingbullets are somewhat heavier than the Chalcidians’ (averaging almost 30 g as opposed to about 26 g for the Chalcidian bullets). On slingbullets, see Korfmann, ‘‘The Sling as a Weapon.’’ 81. According to Robinson and Graham, there was relatively little evidence of fire in most of the houses excavated up until 1934, and they hypothesized that the attackers deliberately pulled down the houses by removing the pillars which held up the pastas (Olynthus 8, 11–12). But a number of houses excavated later were heavily burned, especially in the Villa Section; and the fieldbooks also record evidence of burning in other houses, such as A v 9 and A v 10, described below. 82. Below, chapter 6, ‘‘Sales and Prices of Houses.’’ 83. A pithos in Building A, room f, on the South Hill was full of carbonized grain, but others were essentially empty when excavated, either because their contents were not carbonized in the destruction, or because they had been used up in the siege. 84. Dem. 9.26. 85. Hatzopoulos, Donation. 86. Diod. Sic. 19.52, cf. Diod. Sic. 19.61, ‘‘Although the Olynthians were very bitter enemies of the Macedonians, Cassander had re-established them in a city called by his own name.’’ Cassandreia was founded on the site of Potidaia. 87. Ph. Petsas, ‘‘Χρονικὰ ἀρχαιολογικὰ 1966–1967,’’ Makedonika 9 (1969) no. 123; Hatzopoulos, Macedonian Institutions 196–97. 88. Robinson, ‘‘Inscriptions, 1934’’ 132–33, no. 8; Hatzopoulos, Macedonian Institutions 197–98. 89. Sparkes and Talcott, Agora 12, 461–62. 90. Olynthus 1, ix. 91. Olynthus 2, xi; 7, vi–vii. 92. Olynthus 9, 370. 93. W. S. Ferguson, review of Gude, History of Olynthus, AJA 39 (1935) 154–55. 94. Olynthus 8, 2 and n. 3. 95. Clement, ‘‘Chronological Notes’’; responding to Bellinger’s review of Olynthus 9, AJP 61 (1940) 102–5. Cf. Clement’s analysis in Olynthus 9, 368–74. 96. Summary in Rotroff, Agora 29, 31–32 97. Dengate’s arguments have not, unfortunately, been published in detail. See Dengate, ‘‘Solution’’; Dengate, ‘‘Abandonment’’; cf. Rose, ‘‘Reconsideration’’; Rose, ‘‘Coins and History’’; Coulson, ‘‘Chatby Reconsidered.’’ 98. Dengate, ‘‘Solution’’ 11–11 bis. I am unsure where he derives his statistic of 154 coins dating between 348 and 316; he apparently includes 108 late coins found at Mecyberna, which are published in the same volumes as the coins from Olynthus. 99. Dengate, ‘‘Solution’’ 12. 100. Dengate, ‘‘Solution’’ 11, 14. 101. Ferguson’s statistics are taken from Gude’s ‘‘Prosopographia Olynthia,’’ in her History, 39–50. Gude lists about forty Olynthians who are attested after 348 b.c., about sixteen of whom may date to after 316 b.c..

312

Notes to Pages 46–52

102. Olynthus 9, 327–330. The Anonymous Bronzes bear the inscription ΒΑ[ ΣΙΛΕΩΣ ], and either the types of the coins of Alexander (Herakles/Bow, quiver, club), or a Macedonian shield with thunderbolt/Macedonian helmet. See Gaebler, Antiken Münzen Nordgriechenlands III 2 193, no. 2; Mørkholm, Grierson, and Westermark, Early Hellenistic Coinage. 103. Olynthus 8, 9–10; Olynthus 9, 370. 104. There are no records in the fieldbooks of the exact findspots or depths of the late coins from A iv 10 or the fountainhouse; the stratigraphy of both was very mixed up, however, probably due to later stonerobbing. In Section N, at the north end of the South Hill, the excavator describes the upper layer of fill, 0.3–0.5 m thick, as dating to after 348, while the lower layer, 0.1–0.4 m thick, is destruction debris from the 348 destruction (Olynthus 12, 309). No architecture was associated with this upper fill. Two coins of Cassander were found at a depth of 0.6 m, but in the upper fill, one of them to the west of the ‘‘arsenal,’’ the other just east of it. The exact findspot of the third late coin, found in 1928, is unknown. 105. The single coin of Megara found at the site came from A' 10; this type has generally been dated later than the middle of the fourth century (Olynthus 9, no. 1099). This mint was traditionally dated by Waage to the third quarter of the third century b.c., but Kroll dates coins of this type (Variety W) to the latter part of the fourth century based on the findspot of this example (‘‘Early Athenian Bronze Coinage’’ 154). One type of Philippi, with a head of Herakles L / tripod, no inscription, is represented by a single coin from A' 7, and this may be later in date as well (Olynthus 9, no. 1247). Two unattributed coins are also restricted in extent to the Northwest Quarter: ‘‘Unattributed 5’’ and ‘‘Unattributed 14.’’ These were not well enough preserved for Clement to identify, and their significance remains questionable (Olynthus 9, 362 and pl. 33, nos. 5–10). Coins of Elaeus in Thrace and Histiaea on Euboea were found in A' 9; the Thracian coins were traditionally dated to the second half of the fourth century, while coins of Histiaea are variously dated by numismatists. One example of each of these coins was found in Trench 7 as well, but since the stratigraphic context of those latter coins is uncertain, they might have come from upper levels (which did produce one late coin), and it is possible that these issues are late as well. No coins of Carystus, Cyme, Hephaestia, Lemnos, Heraea, Iasus, Phygela, or Placia were found on house floors which I would attribute to the 348 destruction, but they were found in parts of the city which do not seem to have been reoccupied. Coins of all the other mints are found in what seem to be destruction contexts in houses which show no signs of reoccupation after 348. It seems most economical, therefore, to assume that those coins belong with the destruction deposits, and that these mints were producing coins before 348. 106. Olynthus 9, 238, 332. The types are: (1) male head R or L / nude horseman R; (2) male head R / horse galloping L; (3) similar, but L / similar, beneath, ‘‘Α’’; (4) similar, R / horse R, legend ΑΛΕΞΑΝΔΡ[ΟΥ ]? or Ο? Clement suggests that the coins of Type 4 could belong to either Alexander II or III, distinguishing between the two kings by the presence of a final Υ in the name of Alexander the Great (Olynthus 9, 326), which is not preserved in the examples assigned to this type. The ‘‘Anonymous Bronzes’’ carry different types and thus must date to the reign of Alexander the Great or later. 107. This includes only the coins from houses which were completely excavated or partly trenched (A' 7–A' 11, A xi 9, A xi 10, A xii 9, A xii 10, and A xiii 10). I have excluded coins from test trenches and so forth.

Notes to Pages 53–57

313

108. This is almost the opposite conclusion from that reached by Bellinger, who suggests that many of the ‘‘wandering’’ Greek coins from outside Macedonia circulated to Olynthus because many of the local mints had been closed by Philip, and the Macedonian royal coinages and remaining autonomous mints were insufficient to meet the needs of the surviving communities. If this were so, however, these ‘‘wanderers’’ should be found together with late Macedonian royal issues, but they are not (Bellinger, ‘‘Notes’’ 185–86, cf. Martin, Sovereignty and Coinage). 109. Rotroff, Agora 29 18–20; Rotroff, ‘‘Athenian Hellenistic Pottery’’; Rotroff, introduction to Two Centuries of Hellenistic Pottery; Rotroff, ‘‘Olynthos and Other Deposits.’’ 110. Sparkes and Talcott, Agora 12 122, n. 65 and 123, n. 68; Rotroff, ‘‘Athenian Hellenistic Pottery,’’ 9, nn. 28–29. 111. Republished as Olynthus 13, no. 510A, pl. 153. It is probably also the cup drawn in Olynthus 13, pl. 185, as no. 510 (which is obviously not a drawing of the kantharos whose photo appears in pl. 188 as no. 510—that cup is probably the one actually excavated in C −x 5). 112. In Olynthus 13, pl. 82, the caption reads ‘‘510A should be 513A,’’ implying that 510A, the kantharos from C −x 5, is not illustrated, while the kantharos from near Grave 377 is illustrated twice, here and in pl. 185. 113. Olynthus 13, no. 42; ARV2, 1507:4; McPhee, ‘‘Red-Figure Vase-Painters’’ 302. The vase is now in the University of Mississippi Museum. 114. Olynthus 13, no. 65; ARV2, 1500:1. The published finds from this pit are a Chalcidic bronze coin (Olynthus 14, no. 227), a shallow bowl with stamped palmettes (Olynthus 13, no. 779); a guttus (Olynthus 13, no. 424); a lamp of Group V (Olynthus 14, no. 21, a fifthcentury type); a lamp of Group VII (Olynthus 14, no. 47); a lamp of Group VIII (Olynthus 14, no. 140); a black-glazed plate (Olynthus 13, no. 846); and two saucers (Olynthus 13, nos. 832, 834). On the pit, Olynthus 12, 154. 115. Rotroff, ‘‘Athenian Hellenistic Chronology’’ 8. 116. The terracottas made from molds attested elsewhere are Olynthus 14, nos. 149 (seated female, probably from same mold as Olynthus 14, nos. 132, 139, 154, Olynthus 7, nos. 220, 221 [same or similar?], 229–40); Olynthus 14, no. 157 (seated female, ‘‘kourotrophos,’’ probably the same mold as Olynthus 4, no. 378), Olynthus 14, no. 17 (female mask, from same mold as Olynthus 14, no. 42; this is probably a daughter mold of that used to make Olynthus 14, no. 15 and Olynthus 7, nos. 18, 36, 37, 38), Olynthus 14, no. 174 (seated female from same mold as Olynthus 14, no. 158), Olynthus 14, no. 250 (Leda and swan, same as Olynthus 4, no. 370), Olynthus 14, no. 394 (plastic vase of grotesque old woman, same as Olynthus 4, no. 398), Olynthus 14, no. 421 (plastic vase of winged male, same as Olynthus 4, no. 321 and Metropolitan Museum 06.1054), Olynthus 14, no. 486 (seated female, same as Olynthus 7, nos. 110–12, Olynthus 14, nos. 160, 162–64, 167). The attributions to molds are taken from Olynthus 4, 7 and 14. 117. This account draws on the published accounts of the houses, on the notes from the unpublished fieldbooks; on the accounts of the excavators with whom I was able to talk, including J. W. Graham, W. McDonald, J. Travlos, P. Clement, and G. Weinberg; and on my own experience in excavating houseshold destruction deposits associated with military capture at Sardis (cf. Cahill, ‘‘Lydian Houses’’). While there are occasional stratigraphic complexities noted in the fieldbooks, the general picture is a relatively simple one.

314

Notes to Pages 57–62

118. Drougou and Vokotopoulou, ‘‘B vii 1.’’ 119. For example, it took a team of two to three conservators more than two seasons to mend the pottery from only three rooms of an Archaic house at Sardis, an area of less than 64 m2 (Cahill, ‘‘Lydian Houses’’). 120. 372 lekythoi, for instance, as opposed to only 52 oinochoai. 121. E.g., Ault, Houses and Households 13; Ault and Nevett, ‘‘Digging Houses’’ 46. 122. Ault and Nevett, ‘‘Digging Houses’’ 49–51; cf. Ault, Halieis 2; Ault, Houses and Households 13 and appendix 1. 123. Schiffer, Formation Processes; LaMotta and Schiffer, ‘‘Formation Processes.’’ 124. Drougou and Vokotopoulou, ‘‘B vii 1.’’ 125. Olynthus 14, v. 126. I have changed the descriptions of pots to conform to the typologies in Sparkes and Talcott, Agora 12 and Rotroff, Agora 29, rather than the often confused terminology in the original publications (Sparkes and Talcott comment that ‘‘the analyses of different shapes found in Olynthus, XIII are a farrago’’). Descriptions of coins follow Clement’s catalog in Olynthus 9. The coins published in Olynthus 14 inexplicably follow a different system. I have changed these to follow Clement’s typology. 127. The fieldbooks are in the museum of the University of Mississippi, together with some of the finds from Robinson’s collection. The University of Mississippi also has an extensive archive of photographs and drawings from the site, including many unpublished photographs and most if not all of the original architectural plans, as well as Robinson’s private papers, many of which relate to Olynthus. Thanks to the generosity of Lucy Turnbull, then director of the University Museum, I was able to spend a week going through their collection and archives; without her help, this study would not have been possible. 128. Many of these records do not quantify the artifacts, simply noting ‘‘many vases’’ or the like. 129. For instance, rooms were given preliminary labels during excavation, and later relabeled more systematically starting from the northwest corner. Sometimes, however, the records of artifacts were not updated to reflect the new room labels, and they were published under the preliminary designations. 130. See http://www.stoa.org/olynthus/. 131. E.g., Olynthus 14, vi. 132. Schiffer, Formation Processes; LaMotta and Schiffer, ‘‘Formation Processes,’’ with further bibliography and more detailed discussions. 133. Pritchett and Amyx, ‘‘Attic Stelai.’’ 134. Ault, Houses and Households; Ault, Halieis 2; Allison, Pompeiian House Contents; cf. Allison, ‘‘Pompeiian House Contents’’; Berry, ‘‘Household Artefacts.’’ 135. On toolkits, see Binford, ‘‘Model Building—Paradigms’’ in his In Pursuit of the Past; Whallon, ‘‘Spatial Distributions.’’ Ciolek-Torello, ‘‘Pueblo Room Function’’ is a good example of the application of these procedures to a fairly complex site. See also Carr, ‘‘Intrasite Archaeological Records’’; Hietala, Intrasite Spatial Analysis. See also Kent, Analyzing Activity Areas; Kent, ‘‘Activity Areas and Architecture.’’ 136. Ciolek-Torello, ‘‘Pueblo Room Function’’ 136. 137. Nevett, House and Society 61–63, appendix 3, pg. 183. The phi-square of the correla-

Notes to Pages 62–71

315

tion is 0.449 (1 would be a perfect correlation; 0 would be random, −1 would be a perfect negative correlation). The chi-square is 0, meaning that the correlation is statistically significant. 138. A room in an incompletely excavated house in Section T, west of house ESH 4, contained seventeen coins and seventy-four loomweights. The courtyard of house A v 3 contained seventeen loomweights and fourteen coins, mostly from a ‘‘deposit’’ which may be a trash midden or other feature rather than destruction debris. House A v 2, room a contained twelve coins and fifteen loomweights. In fact, there may well be a negative correlation between groups of loomweights and groups of coins. 139. 574 rooms (out of 1,274 total) had more than one coin. 294 rooms contained one or more loomweights, but 215 of these had five or fewer—not enough to equip a loom. Coins and loomweights were found together in 220 rooms.

chap ter 3: the houses described 1. These accounts are based on the publications and particularly on the fieldbooks from the excavation and the database of artifacts, rooms, and houses derived from these records. For reasons of space and readability, I will not give publication references to every object mentioned in the text; again, the reader is referred to the Internet site (http://www. stoa.org/olynthus/). The published objects can be found in the Concordance of Proveniences in many of the final publications, such as Olynthus 7 (Terracottas of 1931), 8 (The Hellenic House), 9 (Necrolynthia), 10 (Metal and Miscellaneous Finds), and 14 (Terracottas, Lamps and Coins of 1934 and 1938, with a Master Concordance of Proveniences). 2. See Olynthus 8, 147–51; Graham, ‘‘Origins’’; Drerup, ‘‘Prostashaus und Pastashaus’’; Jones et al., ‘‘Vari House’’ 430–38; Krause, ‘‘Grundformen.’’ The terminology ‘‘prostas’’ and ‘‘pastas’’ is not ancient and may not reflect Greek usage of these words: see Tsakirgis, ‘‘Prostas House.’’ 3. Cf. Xen., Ec. 9.4; [Aristotle] Economics 1345A. The Loeb translation misunderstands the terms τὰ πρὸς μεσημβρίαν and τὰ πρὸς ἄρκτον; cf. Olynthus 8, 144–45 and n. 7. 4. Aeschylus, Prometheus Bound 450–53; see the scholiast and Graham’s comments, Olynthus 8, 145. On this principle, which was recognized at least from the earlier fifth century, see Hoepfner and Schwandner, Haus und Stadt 318–20 and fig. 303. The most notable exception to the southward-facing principle at Olynthus is A viii 5, which has its court in the southwest corner of the house, a short portico to its north (room e), and a longer one on the east (room f ). As Graham pointed out, the house is most intelligible as a normal pastas type, but turned to face the west rather than the south, with a pastas (f ) and a suite of ‘‘North Rooms’’ here on the east. The pastas is also semienclosed, partly blocked off by a low rubble wall (Graham, ‘‘Olynthiaka, 4’’). Graham explains the unusual orientation by suggesting that the builder did not want to have the entrance to the house, from Street viii, take up space from the range of rooms which is usually located in the north side of the house, so he rotated the whole house plan to place these rooms on the east. 5. Aristotle, Pol. 1330a ff. Many of Aristotle’s recommendations seem to be drawn from the Hippocratic Airs, Waters, Places. A southward-facing orientation for the houses and

316

Notes to Pages 72–75

streets of a city is recommended by many authors, and planners seem to have chosen this orientation even in cases where the topography favored another orientation, such as at Goritsa (Boersma, ‘‘Goritsa’’). 6. The very few well-enough-preserved Greek houses have few or no windows facing onto the street: for instance the Peristyle House 1 at Monte Iato, the house at Ammotopos, and the facades of fifth-century houses at Selinus (Monte Iato: Dalcher, Peristylehaus 1 von Iaitas; Ammotopos: Hoepfner and Schwandner, Haus und Stadt 147–50; Selinus: Mertens, ‘‘Greek Architecture in the West’’ 331). In general, see Löhr, ‘‘Hof, Fenster, Türen.’’ 7. The term is used by Xenophon as quoted above, and by Vitruvius in his rather confusing description of the Greek house (Vitruvius 6.7). Vitruvius’s description, however, seems not to apply to houses of mainland Greece but to Greek houses in Sicily and southern Italy: see Tsakirgis, ‘‘Prostas House.’’ Other references in Orlandos and Travlos, Lexicon and Hellman, Vocabulaire de l’architecture grecque, s.v. It is curious that Xenophon uses the plural παστάδες rather than the singular, since houses normally do not have more than one pastas. He may use the term to refer to porticoes on other sides of the court besides the main one on the north. But the word is certainly used in other senses besides the one adopted by archaeologists: in Euripides’ Orestes 1371 the sense demands a locked door, for instance. 8. Cf. Goldman, ‘‘Spatial and Behavioral Negotiation.’’ 9. Compare the arrangement of the core suite of four rooms at houses of the prostas type (e.g., at Priene or Colophon) where one room is regularly secluded behind the ‘‘oikos’’ and access is generally more limited and hierarchical. A similar situation prevails in Greek houses at Eretria, whose houses have a three-room suite with a ‘‘transverse hall’’ restricting access to the rooms behind (fig. 45; Krause, ‘‘Grundformen’’; Ducrey et al., Eretria 8, 40, 47; Reber et al., Eretria 10, 154–69). In Roman atrium-peristyle houses, access to the inner parts of the house around the peristyle is progressively restricted. See Drerup, ‘‘Prostashaus und Pastashaus’’; WallaceHadrill, Houses and Society. 10. Olynthus 8, 170–71. 11. Many references in Olynthus 8, 179–85; Bergquist, ‘‘Sympotic Space,’’ and other articles by Cooper, Bookidis, and Tomlinson in Murray, Sympotica; Hoepfner and Schwandner, Haus und Stadt 327–28; Reber, ‘‘Andrones von Eretria.’’ At Priene, the room opening off the prostas seems to have been the andron, although it lacks the distinctive raised border: Raeder, Priene: Funde 20–25. In literature and art: Murray, Sympotica; Orlandos and Travlos, Lexicon s.v. 12. Graham, ‘‘Olynthiaka, 3.’’ Androns at other sites also make special provisions for lighting, for instance at Eretria (Reber, ‘‘Andrones von Eretria’’). See also below, ‘‘House A 10,’’ and chapter 4. 13. Olynthus 8, 185–204; Mylonas, ‘‘Excursus on the Oecus Unit,’’ Olynthus 12, 369–97; Graham, ‘‘Olynthiaka, 6’’; Svoronos-Hadjimichalis, ‘‘Évacuation de la fumée.’’ 14. Bettalli, ‘‘Case, botteghe, ergasteria.’’ For an interesting and somewhat extreme example of dispersed ownership of rooms in a modern village, see Horne, ‘‘The Household in Space,’’ and Stone, ‘‘Texts, Architecture and Ethnographic Analogy.’’ 15. Olynthus 8, 214–19, 267–80.

Notes to Pages 76–82

317

16. Mylonas, ‘‘Olynthian House.’’ 17. Hoepfner and Schwandner, Haus und Stadt 82–89 (Olynthus), 208–12 (Priene), 312– 25, and passim; Schuller, Hoepfner, and Schwandner, Demokratie und Architektur. 18. Olynthus 12, 183–206; Mylonas, ‘‘Olynthian House.’’ 19. The only other known example is house B i 5, which had painted garlands in the andron (Olynthus 2, 109; 8, 301). 20. Olynthus 8, 154–56. 21. Graham, ‘‘Olynthiaka, 1.’’ One small Doric capital, a fragment of a second, and an iron dowel—all found near the altar—are too small to have fallen from the colonnades of the court itself and so probably belong to the canopy of the altar. Robinson (Olynthus 12, 190) mentions only one capital and postulates that it must have been placed above the door to e, h, or l; but the one complete example is too small for those positions. 22. Only one pillar base is preserved, at the west jamb of the door; those on the low foundation were probably robbed out. But the restoration proposed by Graham, of pillars set on a low foundation, is convincing (Graham, ‘‘Olynthiaka, 4’’). The pastas was apparently designed this way, unlike houses like A iv 9 whose pastades were blocked off later. This may be another indication of a relatively late construction date for the House of Many Colors. 23. Robinson suggests that these stones on top of the foundation were too rough to be bases (Olynthus 12, 200, 375). But the construction of this less formal part of the house need not have been very fine, and the parallel at other houses seems too strong to ignore. See Graham, ‘‘Olynthiaka, 6.’’ 24. A similar trench was found in the flue of house A 2, room e. 25. This interpretation was favored by Robinson and Mylonas, who refer to room k as the ‘‘oecus’’ of the house, and room h—Graham’s ‘‘flue’’—as the ‘‘kitchen.’’ Graham accepted their conclusions, although not their terminology. Olynthus 12, 369–98; Graham, ‘‘Olynthiaka, 6.’’ 26. In Olynthus 12, 200–201, Robinson says these were found in the tile layer of the room; however the fieldbook explicitly notes that nearly all the objects were found below the tile layer on the floor; the only exceptions were a table amphora (Olynthus 13, no. 229), a saucer (Olynthus 13, no. 820), a terracotta female (Olynthus 14, no. 218), a one-handled cup (Olynthus 13, no. 731), and a spike, which were found in the tile layer. 27. Nevett suggests that the flue was commonly used to dump refuse (Nevett, House and Society 66); but the completeness of many of the vases found in flues suggests otherwise. 28. Olynthus 12, 198–99, 392–93; Mylonas, ‘‘Olynthian House’’; Graham, ‘‘Olynthiaka, 6’’ 340. 29. Robinson suggests that the bosses belonged to the door to the room, which is also possible (cf. bosses found in situ at Kassope: Hoepfner and Schwandner, Haus und Stadt 155–56, 314–16). But groups of bosses found elsewhere at Olynthus often come from chests or furniture, for instance, those from the pastas of the House of Many Colors, or from room b of the Villa of the Bronzes. The krater was not mended or inventoried. 30. Robinson suggests that it might have been a cistern or a storage or drainage pit (Olynthus 12, 204). Graham’s reconstruction shows a rather awkward channel leading from the gutter on the south side of the court, diagonally from ceiling to floor through the center of the exedra to drain into this opening, implying that it served as a cistern (perhaps

318

Notes to Pages 82–94

unfinished? Graham, ‘‘Olynthiaka, 5’’ fig. 2, d). If it had been used as a cistern in the final phase, it would have had to be filled via a pipe set into the floor, like, for instance, the robbed-out installation in the court of house A viii 2 (Olynthus 12, 12–14). The interior of the pit was not lined, and it was open and empty at the time of the destruction since two storage amphoras and fragments of mosaic from the upper floor were found at its bottom. See photos Olynthus 12, pls. 160:1, 163:1, 164. 31. Graham, ‘‘Olynthiaka, 5.’’ Note that the base is set slightly to the west of the edge of the foundation. Either the stair turned a corner in the southeast, or the base might be slightly out of place. 32. Olynthus 12, 204, where, however, Robinson declares the evidence ‘‘uncertain and, it seems, rather doubtful.’’ The excavator, Lloyd Daly, on the other hand, declared that ‘‘the conclusion that the broken pieces came from a second story seems almost inescapable’’ (fieldbook LD pg. 76). Three fragments average 0.75 m × 0.5 m, a larger fragment is about 1.3 × 0.5 m, and more fragments were found in the ‘‘cistern.’’ Graham’s reconstruction of a stair along the east wall of the exedra seems convincing: ‘‘Olynthiaka, 5’’ 323–25, and fig. 2. 33. Such an argument from negative evidence is admittedly dangerous because vases and the like might have been too broken to have been collected or mended, while architectural evidence might simply not have been preserved. 34. Graham, ‘‘Olynthiaka, 6’’ 323. 35. Compare the few fragments of beds from Priene (Raeder, Priene: Funde) and the much more common beds from Roman sites like Pompeii and Herculaneum (WallaceHadrill, Houses and Society 96–97, 113–14). 36. Olynthus 12, 235–58. 37. Found in the northeast corner, not in the northwest (as Olynthus 12, 246), see fieldbook WAM 1 pg. 88. 38. Fieldbook WAM 2 pg. 7. 39. Fieldbook WAM 2 pg. 34. 40. Olynthus 12, 253. 41. Some of these artifacts were found rather high in the fill, up to 0.7 m above the floor, but they apparently still belong with the floor assemblage. For instance, three of the saucers were found at a depth of 1.2 m, but less than a meter from where other saucers were found resting on the floor at a depth of 1.9 m. This room was open to the sky, and the artifacts could not have fallen from the second story. As at the House of Many Colors, objects from floor assemblages can be displaced significantly above it, especially if, as seems likely, they had been stored in furniture rather than lying on the floor. 42. Olynthus 12, 249; granite objects are rarely found at Olynthus. 43. McDonald suspected that it was reused because of cuttings and a lead ‘‘filling’’ (not otherwise described, but accepted in Olynthus 12, 251). However cuttings and clamps are not impossible even on lower stones, and the stone (which was found on the floor of the room) might have been still in use. 44. Robinson believed that the room was a shop and separate from the house because there was another storeroom in room c (Olynthus 12, 258). This pithos, however, is of a completely different character from the pithos in room c and does not duplicate its function.

Notes to Pages 94–103

319

45. On the position of the stair see, Olynthus 12, 241, and Graham, ‘‘Olynthiaka, 5’’ 325– 26. 46. There was some confusion in the first stages of excavating this house, and finds from the northeastern part of A vii 2 and the northwestern part of A vii 4 became mixed up. I have done my best to sort it out from the field notes, but some errors may remain. The questionable objects are not numerous or terribly interesting, and they shouldn’t greatly affect our understanding of the use of this house. Finds were not recorded by 1 × 1 m squares as they were in some other houses, so their exact disposition is also unknown. 47. Olynthus 8, 312. 48. Robinson suggests that this house was not domestic in character at all, but was ‘‘more than a private house with its three shops or public stores’’ (Olynthus 12, 67). The purely domestic nature of the rest of the building is proven by the assemblages from its rooms, while such a formal arrangement of shops facing onto the avenue is also attested in A v 9, A v 10, B vi 9, and other houses. I see no reason not to consider this a house in the general sense of the word. 49. Graham does not identify the cut stones in the entrance as a stairbase (Olynthus 8, 279–80). But in a later article, he does suggest that the ‘‘cupboard’’ in the southeast corner of room c supported a stair ascending from the north, but without mentioning the stone base which supported its foot, or the almost unique placement of the entrance outside the house (‘‘Olynthiaka, 5’’ 328). Cf. the sales inscription from nearby Stolos, in which a house ‘‘with the upper story’’ (καὶ τὸ ὑπερῷον) sold for 232 dr. (Hatzopoulos, Actes de Vente 31 no. IV). The specific inclusion in the sale shows that the upper story might have been omitted or sold separately. In most cases, however, the upper floor probably belonged to the same household as the lower, since the stair to the upper floor rose from inside the pastas or other interior space of the lower part of the house. The only other house with a possible preserved external staircase is A vi 10 (and the identification of this base is somewhat dubious). 50. Graham, ‘‘Olynthiaka, 4’’ 204. 51. Olynthus 8, 87–88; Graham, ‘‘Olynthiaka, 6’’ 338. Ashes were found in room c near the center of the colonnade, and Graham wondered whether this could be a hearth, but they are more likely the remains of burned wooden pillars. 52. Graham does not mention this double door or include it among his examples of this type of entrance (Olynthus 8, 97–98, 153, 249–50, and fig. 21). But the reconstruction seems fairly certain from the plan and standing remains. 53. Olynthus 13, no. 11. Compare the group of more than one hundred empty Panathenaic amphoras, sold in lots of ten for 2.4 to 3.7 obols each in the Attic Stelai, possibly the property of Alcibiades (Amyx, ‘‘Attic Stelai 3’’ 178–86). 54. Robinson, ‘‘Inscriptions, 1934’’ no. 4, pp. 127–29. Robinson’s reading of τ. ὰ. μι [ . σθ]οφόρα πάντα is only ‘‘a possible reading’’ (pg. 129) but is generally accepted (e.g., Hennig, ‘‘Kaufverträge’’ 149). Hatzopoulos suggests this or a similar phrase in a rasura of a contemporary inscription from Stolos (Actes de Vente 27). Robinson suggests that τὰ μισθοφόρα refers ‘‘to the slaves or servants or possibly to the income-producing shops.’’ 55. The extensive remodeling also raises the issue of earlier floors and fills whose contents may have become mixed with the destruction assemblages. Most of the rooms were

320

Notes to Pages 103–118

dug well below the final floors, and a number of objects came from below the burned level (which is recorded in the fieldbooks); these can be separated out fairly easily and should not interfere with the interpretation of the floor assemblages. They are not listed here. 56. Olynthus 8, 89. 57. Olynthus 10, no. 12; one of two found at the site (the other from a street in the South Hill, Olynthus 10, no. 11). See Mattusch, Greek Bronze Statuary 183–85. This eyelash plate is 4 cm wide; by Mattusch’s calculations, it could have come from a figure 1.68 m high. 58. Olynthus 8, pls. 26:2 and 95. 59. The northern part of house A 8 was uncovered in the first season of excavation at Olynthus; most of the house, however, was carefully excavated by Walter Graham in 1931, although like many houses in Row A, the western row of rooms along the city wall was left unexcavated. The worked blocks have disappeared from the doors, but an area of cobble pavement is preserved in front of the southern half of the entrance, similar to (although less neatly laid than) the entrance to house A 9 and others. 60. Uncovered in 1928, this had disappeared by 1931 and so does not appear on the final plan of the house: see Olynthus 2, fig. 116 = Olynthus 8, pl. 89; compare Olynthus 8, pl. 90. 61. Amyx, ‘‘Attic Stelai 3’’ 170. 62. Olynthus 8, 80, 83; Robinson, ‘‘Inscriptions, 1931’’ 40–42. Polyxenos was one of the later magistrates in Chalcidian coinage: Olynthus 9, 103, modified by Clement, ‘‘Coinage Epilogue.’’ 63. Olynthus 8, 82 and pl. 81: 2. 64. Eretria: Ducrey et al., Eretria 8, 61–63; Reber, ‘‘Andrones von Eretria.’’ Piraeus: AM 9 (1884) 285 ff., pl. 14; Hoepfner and Schwandner, Haus und Stadt 42, fig. 34. A similar window is depicted on the volute krater by the Primato Painter, showing the struggle for the Delphic tripod: Naples 1762, Trendall, Lucania, Campania and Sicily 165, no. 917, pl. 72:1. 65. Olynthus 2, 91. 66. Olynthus 8, 81, 275–76, 278. 67. Or vice versa; the door into this pair of rooms k-l is not preserved. 68. Many finds in houses A 11–A 13 were mispublished because of confusion in the labeling of rooms. The lists and plan here are corrected according to the notes in the fieldbook. 69. Robinson, ‘‘Inscriptions, 1931’’ 43–49. The inscription includes the words ἐνιαυτὸν ἐπὶ ἀπόλυσιν, apparently indicating that the transaction is a πρᾶσις ἐπὶ λύσει, a ‘‘sale on condition that the seller may release the property from the buyer’s claim on it,’’ or a ‘‘security in the form of a conditional sale,’’ one of the many roundabout ways the Greeks negotiated loans. See below, chapter 6, ‘‘Sales and Prices of Houses.’’ Mint magistrates: Olynthus 9, 103–4. Clement, ‘‘Coinage Epilogue’’ rearranges the order of these magistrates so that their tenure, originally thought to be consecutive, seems now to have been separated by about fifteen years. 70. Not, as published in Olynthus 12, 160–61, at the door into room m. The excavator locates it ‘‘5 m west of the pipe [in D v 8], partly in the street and partly in the house by the door. It was face against the wall on its side against a short piece of wall east of the door’’ (fieldbook FPA pg. 6–7). The plan (Olynthus 12, pl. 136) locates it here as well. Robinson

Notes to Pages 123–136

321

says that ‘‘this is a small sum for a house of seventeen rooms, but it could hardly apply only to the western third, though that is separate’’ (Olynthus 12, 161 n. 2). But to judge from the findspot, the sale must refer to the main part of the house, and may have included the two shops as well. 71. Robinson, ‘‘Inscriptions, 1938’’ 50–51, no. 4.This was found ‘‘4 m west of the House of Zoilos, outside of the wall, in the street’’ (fieldbook FPA pg. 74). This probably indicates the area in front of D v 4, since Albright does not distinguish the shops as distinct from the domestic portion of D v 6. It may be this inscription which Robinson confuses with the sale of the House of Zoilos when he says that the sales inscription was found in front of room m. The inscription probably does not refer to the sale of the shops of D v 6 separately from the house: the sale took place in a different year and refers to the sale of a house ([ὠνὴ εὐθεῖα τῆς] οἰκίης τῆ[ς] ἐχο[ μένης . . . ]) rather than to shops or workrooms (ergasteria). 72. Salzmann, Kieselmosaiken 101, nos. 85–86; cf. Olynthus 8, 65. 73. Olynthus 8, 64, fig. 3. 74. Olynthus 8, 67–68; Graham, ‘‘Olynthiaka, 6’’ 343–44. 75. Krater: Olynthus 5, no. 112. 76. Raeder, Priene: Funde 22–25. 77. Olynthus 12, 215; the seam is not drawn on the plan but is visible in the photo, pl. 180. 78. See Olynthus 12, 211 and photograph pl. 175. In this area were found ‘‘many fragments of pithoi, tiles and other coarse ware’’ and three undecorated simas, but these probably collected here after the area was dug out. 79. Olynthus 12, 215 and nn. 17–18. 80. Olynthus 10, 253–60. 81. Ducrey et al., Eretria 8 40, 47; Reber et al., Eretria 10 136–37 and passim; Krause, ‘‘Grundformen.’’ 82. The terracotta altar is not mentioned in Olynthus 12. The lead dish (Olynthus 10, no. 2538) is tentatively identified as a lamp or a holder for a terracotta lamp, which seems unlikely. 83. Olynthus 12, 209.

chap ter 4: the houses organized 1. Vitruvius 6.7.2. See Reber, ‘‘Aedificia graecorum.’’ 2. See also Morgan, ‘‘Euphiletos’ House.’’ 3. IG XI2 161 A 17, 162 A 14, 199 A 9, 203 A 26, 204 l. 32, 224 A 20, 290 l. 26; Hellman, Vocabulaire de l’architecture grecque 99–100. 4. Walker, ‘‘Women and Housing’’ 82; accepted and reproduced, for instance, in Fantham et al., Women in the Classical World 104, fig. 3.17. 5. Olynthus: Mylonas, ‘‘Excursus on the Oecus Unit’’ in Olynthus 12, esp. 384–89, where he identifies the kitchen-complex with the Greek word οἶκος, but describes it as ‘‘the place where the women of the household spent a good deal of their time in the performance of their daily tasks’’ (p. 384). Priene: Wiegand and Schrader, Priene 285–90 (again identified as the oecus in Vitruvius where the women worked).

322

Notes to Pages 136–152

6. Jameson, ‘‘Domestic Space’’ 104, cf. Jameson, ‘‘Private Space’’ 186–87. 7. Nevett, ‘‘Separation or Seclusion?’’ 103; cf. Nevett, ‘‘Gender Relations’’; Nevett, House and Society 19–20, 71. 8. For instance Hoepfner and Schwandner, Haus und Stadt 218, 328–29, and passim. Despite the lack of architectural or literary evidence, the idea is commonly stated as fact: for instance, Barber, Women’s Work 273–80; Pomeroy, Goddesses, Whores, Wives and Slaves 79–83; Pomeroy, Xenophon Oeconomicus 291–97; Cohen, ‘‘Seclusion.’’ 9. Nevett claims that ‘‘even at Olynthus, which has the largest number of excavated houses and relatively detailed information on finds and their spatial distribution, comparison with the amount of material from other sites suggests that much information is missing, so attempts to discern patterns in spatial layout are disappointing.’’ (Nevett, ‘‘Separation or Seclusion?’’ 110 n. 2). I hope that the present study will help ameliorate this disappointment. 10. A number of houses at Eretria have a three-room complex similar to the Olynthian suite, for instance House IA, rooms g, l, and m; House IB rooms u, A, and B; House II rooms a, a 1 , a 2 , etc. (Reber et al., Eretria 10, esp. 137–39). Houses A, B, and E at Halieis have similar pairings of a large room (‘‘kitchen’’) and bath (Ault, Houses and Households 232–34; Ault, Halieis 2). But none of these show evidence for a specific cooking area or for the distinctive pillar partition. See also Sparkes, ‘‘Greek Kitchen’’ 132; Svoronos-Hadjimichalis, ‘‘Évacuation de la fumée.’’ 11. Olynthus 8, fig. 14; Graham, ‘‘Olynthiaka, 6’’; Svoronos-Hadjimichalis, ‘‘Évacuation de la fumée’’ 497. 12. Mylonas, ‘‘Excursus on the Oecus Unit’’ in Olynthus 12, 368–96; Graham, ‘‘Olynthiaka, 6.’’ 13. Flues in houses A iv 3, perhaps in A v 9 and 10 (see above), A vi 2, 5, 6, and 7 (also A vi 8, whose flue contained a hoard of silver coins), the House of the Comedian, and perhaps B vi 1 have significant quantities of tablewares and other household equipment, but lack recorded evidence of burning. In none of them were substantial quantities of cooking or coarse pottery noted. 14. In A vi 6, A vi 10, A vii 2, and A xi 10. 15. Hoepfner and Schwandner, Haus und Stadt 100–102, 328–29; Nevett, ‘‘Gender Relations.’’ 16. Smaller numbers of loomweights and a few weaving implements were found in other kitchens, however, and a pair of grindstones stored in the bath adjoining the kitchen of the House of Many Colors must have been used elsewhere in the house, perhaps in the kitchen. 17. House IB at Eretria (Bauphase 3, second century b.c.) had a complex of rooms slightly resembling the kitchen-complex at Olynthus. The kitchen was almost entirely bare; ceramics and other finds were stored in room b, fallen from a shelf. Room b also had a stone floor, perhaps for cooking like the flues at Olynthus. See Reber et al., Eretria 10 100–102, 107–11. 18. Graham, ‘‘Olynthiaka, 6.’’ 19. The only other exception is A 4, d; and the ‘‘door’’ in this room may well be due to erosion, with the original door leading into room j (contra Olynthus 8, 136).

Notes to Pages 152–159

323

20. The only other house whose flue adjoins an outside wall is the South Villa. House B vii 2, recently excavated by Vokotopoulou, has a flue-like room in its southeast corner, and if this is to be identified as a flue, it would form another exception. 21. Cf. Svoronos-Hadjimichalis, ‘‘Évacuation de la fumée.’’ 22. Ault and Nevett, ‘‘Digging Houses’’ discuss recent attempts and problems with such analyses. 23. Sparkes, ‘‘Greek Kitchen’’; Reber et al., Eretria 10, 137–39; Jameson, ‘‘Domestic Space’’ 99. 24. Braziers were found in houses A viii 1 (N. Room c), A viii 3 (N. Room b), A viii 4 (pastas f ), A viii 8 (N. Room a, two braziers), and A viii 10 (court g). Outside this block, three braziers (?) were found in house A 1 (two in court h, one from an unknown room), and others in B i 5 (unknown room), the House of the Tiled Prothyron (pastas e, although possibly buried under the floor), Villa CC (workroom? d, with a loom and many finds), and in a trial trench next to the House of the Comedian. A bronze brazier was found under the floor of A xi 10, but this is quite fancy for cooking. Since it was deliberately buried under the floor, its findspot does not tell us about where or how it was used. 25. Ashes, coarse pottery, and bones were found in Villa CC (room c); A viii 3 (k); A viii 8 (j); and A iv 7 (h and k, no bones recorded). Of these rooms, however, that in Villa CC is very possibly a flue, although very poorly preserved. The room in A viii 8 was probably not used for domestic cooking but for larger-scale baking; that in A viii 3 for some unknown activity, possibly industrial; and that in A iv 7 perhaps for striking coins. 26. Foxhall and Forbes, ‘‘Σιτομετρεία’’; Gallant, Risk and Survival 68. On grindstones, Moritz, Grain-Mills and Flour; Runnels and Murray, ‘‘Milling in Ancient Greece.’’ Molleson, ‘‘Abu Hureyra’’ studies the deleterious effects on Neolithic women of working long hours at a saddle quern. On modern experiments with such grindstones, see Samuel, ‘‘Their Staff of Life’’; Foxall in Foxhall and Forbes, ‘‘Σιτομετρεία,’’ 75–81. 27. Counting only the upper stones; from the field notes, at least, it is difficult to distinguish lower stones used with saddle querns from those used with the hopper-rubbers. 28. Pritchett and Amyx, ‘‘Attic Stelai’’ 2, 298–99. 29. Grinding benches in the ‘‘Terrace Buildings’’ at Gordion had large numbers of saddle querns in situ: DeVries, ‘‘Greeks and Phrygians’’ 39 and fig. 10. A house burned in the mid-sixth century b.c. at Sardis had two saddle querns and a clay tray on a low bench in one corner of a room used for cooking and food preparation: Cahill, ‘‘Lydian Houses.’’ Grinding benches at Amarna are discussed by Samuel, ‘‘Their Staff of Life,’’ who reconstructs an emplacement and experiments with the use of these implements. 30. See the ‘‘Homeric’’ bowl from Thebes, illustrated in Moritz, Grain-Mills and Flour 13, fig. 1. 31. Sparkes, ‘‘Greek Kitchen,’’ 125, 134 and pl. 7: 3, 4; pl. 8: 1; Moritz, Grain-Mills and Flour 29–33. 32. Mistakenly identified as coming from room e: Olynthus 8, 327, no. 10. Graham finds that ‘‘the purpose of this apparatus is obscure,’’ but it seems fairly clearly a grinding emplacement (Olynthus 8, 125). 33. In houses A 6, A 8, A 10, A v 9, A vii 10, A viii 4, B ii 3, B vi 5, ESH 1, ESH 4, the House of the Twin Erotes, and the Villa of Good Fortune.

324

Notes to Pages 159–166

34. A 11, A iv 5, A iv 9, A v 1, A vi 7,the Villa of the Bronzes, and other houses. 35. Again, the pair of grindstones found in the bathroom adjoining the kitchen of the House of Many Colors might have been used in the kitchen, or in another space. The grindstone fragment found in the kitchen of the Villa of the Bronzes may well have been reused there. 36. Olynthus 8, 336 n. 6; Moritz, Grain-Mills and Flour 22–28; Sparkes, ‘‘Greek Kitchen’’ 125–26; Amyx, ‘‘Attic Stelai 3,’’ 236–38. On the use of mortars, see Samuel, ‘‘Their Staff of Life’’; Foxall in Foxhall and Forbes, ‘‘Σιτομετρεία,’’ 75–81. They were not used for the same purpose as grindstones, which reduced the cleaned and partly crushed grain to meal or flour. 37. Moritz, Grain-Mills and Flour 22. 38. Amyx, ‘‘Attic Stelai 3,’’ 236. 39. Found in room l of A vi 7, which was originally the main room of a kitchen-complex, later subdivided into three rooms, k, l, and m. 40. Exceptions include the mortar found in the pastas of A v 9, another possible mortar (described as a ‘‘circular stone’’ in the fieldbook) found in the court of house A vii 2, and the oval stone bowl, perhaps a mortar, from the court of A 10. 41. Published examples: Olynthus 13, nos. 1025–1034. In the laconic field notes it is often difficult to distinguish heavy mortaria, used for crushing and grinding, from other types of shallow basins; and it is even more difficult to determine whether these are heavier ὅλμοι or the lighter-weight θυείαι, also used for crushing. The published examples seem heavy enough to be mortaria, and so, perhaps, are some of the other coarse shallow basins or bowls mentioned in the fieldbooks. But they may well have been used for crushing substances other than grain. 42. Amyx, ‘‘Attic Stelai 3,’’ 239–41, 281. 43. Amyx, ‘‘Attic Stelai 3,’’ 221–28, 240; Sparkes, ‘‘Greek Kitchen’’ 126–27; Sparkes, ‘‘Kitchen Addenda’’; Sparkes and Talcott, Agora 12, 218–21; Iozzo, ‘‘Corinthian Basins.’’ Oakley and Sinos, Wedding 23 and fig. 45 on St. Petersburg St. 1791. A marble basin, rougher than those found at Olynthus, was found in what appears to have been a mill or bakery in Boundazeza: Young, ‘‘South Attica’’ 190 and fig. 12. 44. Two more louters were found in the pastas of the house, and another pair of grindstones were found in a storage ‘‘closet’’ (c). 45. The few published storage vessels from Olynthus include coarse lekanai Olynthus 5, nos. 943 and 944; Olynthus 13, nos. 1044 and 1045. More than thirty other such ‘‘basins’’ are mentioned in the notes, and frequent references to ‘‘many coarse vases’’ and the like probably included coarse storage vessels. Other smaller vessels, duck askoi, plain jugs, and the like, could be used for storage of smaller quantities of foodstuffs. Storage amphoras were more completely recorded, probably because even when broken they were more recognizable in the earth: the field notes mention 117 from houses (and many more from graves), as well as mention of ‘‘several,’’ ‘‘many’’ and the like in a dozen or so more instances. In general, see Sparkes, ‘‘Greek Kitchen’’; Sparkes and Talcott, Agora 12, 187–235; Amyx, ‘‘Attic Stelai 3’’; Kanowski, Containers of Classical Greece. 46. Xen., Ec. 7.5, tr. Pomeroy. 47. Olynthus 8, 209.

Notes to Pages 166–171

325

48. Gordion: DeVries, ‘‘Greeks and Phrygians’’ 39. In general, see Barber, Prehistoric Textiles 91–113; Hoffmann, Warp-Weighted Loom; Carington-Smith, Spinning, Weaving and Textile Manufacture. On Olynthian weights, see the chapter by Lillian Wilson, ‘‘Loom Weights’’ in Olynthus 2, 118–28. 49. I have omitted from the graph the pastas of house A viii 7, which contained 247 loomweights because these seem to be used for household industry: below, chapter 6, ‘‘Textile Manufacture.’’ 50. One odd case is the andron of house B i 5, which contained eight ‘‘lead loomweights.’’ These were not published, and their use is not certain; they may have been for some purpose other than weaving. 51. Wilson, in Olynthus 2, 118–28. The loomweights of 1938 were catalogued in a fieldbook, with dimensions and, for some, weights in drachms; this discussion is based largely on that catalogue. 52. Olynthus 12, 207. Wilson says that Olynthian loomweights tend to be closely matched; but since in 1928 loomweights were not collected room by room, she may not be referring to actual assemblages as they were found (Olynthus 2, 121). 53. A Bronze Age loom from Troy had forty-two loomweights, found fallen in several orderly rows, of which eighteen were small, twenty-two large, and two very large: Hoffmann, Warp-Weighted Loom 311. Stamps on loomweights might have designated their weight; but not all loomweights in a set are stamped, and those that are often bear different stamps. 54. Robinson,’’Inscriptions, 1934’’ no. 6; Hatzopoulos, Actes de Vente 60. 55. Rotroff and Oakley, Debris from a Public Dining Place, from a public context; Lynch, ‘‘Communal Drinking.’’ 56. Olynthus 5, 105A; ARV2, 542.17. 57. Unfigured kraters are rare in Athenian pottery as well: Sparkes and Talcott, Agora 12, 55; Rotroff and Oakley, Debris from a Public Dining Place 20. The field notes mention eight unpublished kraters, five of which are described as ‘‘red-figured’’; it is not clear whether these were in fact red-figured kraters, and if so, why they were not mended and catalogued. 58. Rotroff and Oakley, Debris from a Public Dining Place 11, 68–84. 59. Most of the cups from the sympotic debris from the Agora were black-glazed rather than figured: the deposit contained 87 figured and unfigured kraters, but only 30 figured cups of all sorts (including black-figured), compared to between 819 and 1,365 unfigured cups. Rotroff and Oakley, Debris from a Public Dining Place 131–36. 60. Not including the early kraters which clearly predate the destruction. 61. Olynthus 12, 128–38; Olynthus 13, 196C = ARV2, 1497:23. 62. Pritchett, Greek State at War 1, 53–84. 63. Most completely argued in Vickers and Gill, Artful Crafts, the quote from pg. 38. Williams, ‘‘Refiguring Attic Red-Figure,’’ among others, responds to their arguments and concludes that ‘‘the turning point [between the use of pottery and the use of silver and gold] is placed roughly in the middle of the fourth century’’ (pg. 232 n. 17). 64. Juba in Athenaeus 229b; Anaximenes in Athenaeus 231c. 65. Athenaeus 231d. 66. Touratsoglou, ‘‘Dated Gold’’; Athenaeus 128a-130e. 67. Olynthus 10, nos. 1, 574, 575, 577.

326

Notes to Pages 173–189

68. Bronze vessels include Olynthus 10, nos. 633–649, 658–666, 571–585; silver pyxis Olynthus 10, no. 506. More remain unpublished but mentioned in field notes. 69. Nevett, ‘‘Gender Relations’’ 373, who, however, passes over the conceptual identification and practical use of the kitchen-complex. 70. Ibid. 71. Graham, ‘‘Olynthiaka, 2’’ 199–203.

chap ter 5: the or ganization of bl o cks 1. Hoepfner and Schwandner, Haus und Stadt 82. 2. The Villa of Good Fortune and the South Villa are also larger-than-normal houses and so diverge from the raster, but these seem to be part of the original design rather than a later remodeling. The House of the Comedian may also have expanded to the west of its original plot, but this part of the house was poorly preserved and hardly excavated. 3. Some of these might have come from earlier habitation before the anoikismos, but it seems more likely that these walls are secondary features. 4. Wiegand and Schrader, Priene; Hoepfner and Schwandner, Haus und Stadt 188–225. On the destruction date, Raeder, Priene: Funde 11–12. 5. Burney, ‘‘Urartian Fortresses’’ 49–50; Burney, ‘‘Measured Plans’’ 185–88; Nylander, ‘‘Zernaki Tepe’’; Sevin, ‘‘Van/Zernaki Tepe.’’ 6. Rapoport, House Form and Culture; Rapoport, History and Precedent; Rapoport, Mutual Interaction; Kent, Domestic Architecture and the Use of Space; Blanton, Houses and Households. 7. A few houses are not aligned: for instance, A v 7 has an irregular northern axis, and the axis of A v 10 is offset to the north from the other houses in this row. This latter house, though, did not share the common roof of its neighbors. 8. Drerup, ‘‘Prostashaus und Pastashaus’’; Graham, ‘‘Origins’’; Akurgal, Alt-Smyrna. Abdera: Hoepfner and Schwandner, Haus und Stadt 180–87. 9. Wilk, ‘‘Built Environment and Consumer Decisions’’ looks at houses as a ‘‘consumer good,’’ and at the decision-making processes involved in planning and modifying houses. 10. Olynthus 8, 37; Robinson, ‘‘Residential District’’ 120. This is the usual practice in Greek cities: for instance at Priene, Himera, Colophon, New Halos (Reinders, New Halos 185), and many other sites. The seams are particularly evident at Priene. For instance, house 33 and its neighbor house 32k both have well-built ashlar walls fronting Theater Street, while the next house to the west, 32i, has only unworked stone walls (Hoepfner and Schwandner, Haus und Stadt 208–10, fig. 199–203). 11. See the photo and drawing Olynthus 8, pls. 28:3, 30. Since the wall is not preserved further than the original east limit of the house (before it expanded into the plot of A v 8), it is impossible to tell whether this cut stonework is original or belongs with the remodeling. 12. These observations were made on visits to the site in 1985–86. For the most part the published photographs do not show the construction clearly enough to be useful. In houses A vii 3 and A vii 8, the large storerooms are constructed differently from the walls of the rest of the houses. Does this imply that these areas were built and owned (or used) by someone other than the family who owned the rest of the house? Otherwise, however, visible changes in construction all occur at the junctions of house plots.

Notes to Pages 190–203

327

13. Party walls which are attributable to the western house are visible today in the original east wall of A v 6, in the north wall of A v 8 and 10 (where A v 10 abuts A v 8 due to the slope of the land here) and in A vii 1, 2, and 3. 14. For instance, Xen., Symp. 4.4: Socrates asks, ‘‘Do you not see plenty of carpenters, also, and architects that build houses for many another person but cannot do it for themselves, but live in rented houses?’’ (οὐ καὶ τέκτονάς τε καὶ οἰκοδόμους πολλοὺς ὁρᾷς οἳ ἄλλοις μὲν πολλοῖς ποιοῦσιν οἰκίας, ἑαυτοῖς δὲ οὐ δύνανται ποιῆσαι, ἀλλ’ ἐν μισθωταῖς οἰκοῦσι;). 15. Xen., Ec. 3.1. 16. Welles, Royal Letters no. 3, ll. 14–15, and no. 4. 17. Paus. 4.27. 18. For instance, block B vi is built on a steeper slope than the other excavated blocks, and the houses on its northern row seem to have made special provisions for drainage: B vi 3, 5, and 7 have a narrow ‘‘corridor’’ on their south sides, in addition to the alley which separates the two rows of houses, which may have been used in part to control runoff from the roofs on the southern sides of the houses. 19. The other main type, hierarchical clustering, produces a graphic diagram or dendogram of how different houses cluster together at different levels of similarity, but it does not tell you in what ways the houses are similar or different. 20. See, for instance, Kintigh and Ammerman, ‘‘Heuristic Approaches’’; Aldenderfer and Blashfield, Cluster Analysis 53–58, for discussions of various mathematical strategies to find the optimum number of clusters (similar methods used by Altschul, Social Groupings at Teotihuacan 105–6, who, however, finds like me that mapping and common sense generally produce more useful and interpretable results than algorithms). Of a great many experiments, I present here only the most informative results. Interested readers may find the raw data on the Internet site. 21. Olynthus 2, 80–88; Olynthus 5, 2–8; Salzmann, Kieselmosaiken 98 no. 77; 99 nos. 78, 79; Westgate, ‘‘Greek Mosaics.’’ 22. Houses A 8, A 10, A iv 7, A v 7, A vi 2, A vi 10, A vii 4, A vii 10, B i 3, B iii 2, B v 1, B vi 2, B vi 10, B vii 1, B vii 2, C −x 7, and ESH 4 had shops, but shops took up a smaller proportion of the space of these latter houses, and they therefore were grouped with other clusters. Many of these also are located along north-south avenues. The few houses fronting onto Avenue B that do not have rooms which may definitely be identified as shops, such as A viii 9 and 10 and B vi 1, have rooms which may have been shops, but whose walls are not preserved high enough to determine where the doors were. Rooms h and i of A viii 9 and room f of B vi 1 may all have been shops; this would of course affect how these houses were clustered. 23. Note however that the street wall of A v 6 is built of cut stone, and this may be original; if so, it suggests that this was an unusually fine house for this block even when it was built. 24. Aristotle, Pol. 1319a; Pol. 1266b; Plato, Laws 741C. In general, Asheri, Distribuzioni. 25. Lokris: ML no. 13, Vatin, ‘‘Le Bronze Pappadakis.’’ Itanos: I.Cr. III iv 8 = SIG3 526, ll. 21–23. 26. Dilke, Roman Land Surveyors 96–97; Campbell, ‘‘Sharing Out Land’’; Campbell, ‘‘Shaping the Rural Environment.’’ 27. SIG3 141; see Wilhelm, ‘‘Korkyra Melaina’’; Klaffenbach, ‘‘Korkyra Melaina’’; Maier,

328

Notes to Pages 203–219

Mauerbauinschriften no. 57, pp. 204–6; Kirigin, ‘‘Greeks in Central Dalmatia’’; and most recently, Lombardo, ‘‘Lo Psephisma di Lumbarda.’’ The text is mentioned in many discussions of colonization, land division, and related issues: Asheri, Distribuzioni 15; Graham, Colony and Mother City 42–43, 64–65; Jones, Public Organization 155–56; Szanto, Griechische Phylen 5–6. 28. Asheri claims that ἐξαίρετον is not to be understood in its usual sense of ‘‘not assigned by lot’’ but here means only ‘‘choice or excellent’’ (Asheri, Distribuzioni 15 n. 3). But this would not be possible if the first ἐξαίρετον is used adverbially, as I think it must be. Neither he nor other commentators discuss the exceptional stress laid on this word. 29. Lombardo, ‘‘Lo Psephisma di Lumbarda’’ discusses the line length of the inscription. 30. This restoration was suggested to me by R. Stroud.

chap ter 6: the e c onomies of olynthus 1. The quote is from Gallant, Risk and Survival 101. In general, see Finley, Ancient Economy; Millett, Lending and Borrowing; Meikle, ‘‘Modernism, Economics and the Ancient Economy.’’ 2. Cohen, Athenian Economy and Society 4; cf. the review of Morris, ‘‘Athenian Economy.’’ 3. Cf. Morris, ‘‘Athenian Economy’’ 351: ‘‘the debate over the Athenian economy, at least among those who call themselves economic historians, is becoming stale and unproductive.’’ Discussing trading in the Old Assyrian period, Kuhrt suggests that ‘‘the debate [among Graeco-Roman historians] seems to me to have run its course and it has not got a lot more to offer’’ (Kuhrt, ‘‘Old Assyrian Merchants’’ 29). Davies, ‘‘Models and Muddles,’’ injects new life by stressing the lack of a single model, a point of view with which I heartily agree. 4. Arafat and Morgan, ‘‘Pots and Potters.’’ 5. Aristotle, Pol. 1257a; Millett, ‘‘Sale, Credit and Exchange.’’ 6. I do not mean to imply a market driven purely by capitalist supply-and-demand forces; see in general Millett, ‘‘Sale, Credit and Exchange’’; Parkins and Smith, Trade, Traders and the Ancient City. 7. Gallant, Risk and Survival. 8. The survey recently begun by Bernhard Hänsel, although aimed primarily at prehistoric remains in the area, may add to our knowledge of this question. 9. A complex and much-argued issue, which will probably defy generalization over the whole of the Greek world. Estimates of the rural relative to the urban population range from a low of perhaps 5 percent (in the Argolid, Jameson et al., Greek Countryside 553; see also the review by Osborne, ‘‘Survey and Greek Society’’) to a high of perhaps 26 percent (at Metapontum, Carter, ‘‘Metapontum—Land, Wealth and Population’’ 405–12). Work in Chersonesos suggests similar results: Sceglov, Polis et Chora; Pečírka, ‘‘Country Estates’’; Dufková and Pečírka, ‘‘Chora of Chersonesos’’; Saprykin, Ancient Farms and Land-Plots. See also Osborne, ‘‘Buildings and Residence’’; Lohmann, ‘‘Agriculture and Country Life’’; Lohmann, Atene; and a growing number of studies of rural life in ancient Greece. But even at Athens, with its relative wealth of historical sources, it is difficult to estimate how many households did not own land and relied on nonagricultural income (see, for instance, Jameson, ‘‘Agricultural Labor’’ and Foxhall, ‘‘Control of the Attic Landscape’’). In

Notes to Pages 221–225

329

any event, the proportion of citizens who dwelled in the countryside probably varied widely from city to city, and could change quickly and dramatically in a single city with changing historical circumstances (for instance at Mantineia: Hodkinson, ‘‘Mantineia’’). 10. Robinson, ‘‘Inscriptions, 1934’’ no. 6; Hatzopoulos, Actes de Vente 60; Youni, ‘‘Inscriptions olynthiennes’’ 137–39. 11. Gallant, Risk and Survival 94–98; Forbes and Foxhall, ‘‘Ethnoarchaeology and Storage,’’ who point out that households on Methana aim to maintain two years of grain and four of olive oil to cope with crop failures, which are expected on average twice a decade; Garnsey, Famine and Food Supply 53–55; Kramer, Village Ethnoarchaeology. This is well documented in ancient sources. One of the most important duties of a wife, says Xenophon in the Economics, is to see that ‘‘provisions stored up for a year are not spent in a month.’’ Later, her husband takes her on a tour of the rooms of the house, including ‘‘the dry storerooms [that] called for grain, the cool ones for wine, and the bright ones for those products and utensils which need light.’’ She and her husband then ‘‘set aside the things that are consumed within a month, and stored separately what we calculated would last a year’’ (Xen., Ec. 7.36, 9.3, 9.8, tr. Pomeroy). The author of the Aristotelian Economics recommends that ‘‘on large estates, after the amount for the year’s or the month’s outlay has been set apart, it should be handed to the overseers; and so also with implements, whether for daily or for occasional use’’ ([Aristotle] Ec. 1345a). On storage, see also Amouretti, ‘‘Les céréales dans l’Antiquité.’’ 12. See Foxhall and Forbes, ‘‘Σιτομετρεία’’ esp. 71–72 (although they warn that the estimates ‘‘must be viewed only as a maximum’’). Foxhall and Forbes postulate a hypothetical family of six, including one older woman, a husband and wife, a teenage son, a younger daughter, and one child. They calculate that grain composed 70–75 percent of the diet of an ancient Greek. 13. Gallant, Risk and Survival 27–33. 14. Gallant, Risk and Survival 72–75, table 4.6. I have converted his estimates, which are given in kilocalories, to liters based on his estimate of 3,100 kcal/kg (pg. 77) and Forbes and Foxhall’s estimate of 0.772 kg/liter. 15. Xen., Hell. 5.2.38. The decree of Lysimachos granting land to Limnaios, dating to 285/4 b.c., lists large areas (360 plethra near Olynthus, 1,200 plethra near Sermylia, 900 plethra near Strepsa) of γῆ ἐνδένδρος, ‘‘orchards,’’ which are probably olive orchards. The countryside around Olynthus is still densely planted with olive orchards. 16. Foxhall and Forbes, ‘‘Σιτομετρεία’’ 68; Forbes and Foxhall, ‘‘Queen of All Trees’’ 46. See also Forbes, ‘‘Ethnoarchaeological Approach’’; Amouretti, Le pain et l’huile 177–96. 17. Although Theophrastus notes that olive trees around Olynthus bear some fruit every year, the harvest in alternate years is still higher (Theophrastus, De Causis Plantarum 1.20.4). Foxhall speculates that this may be why the earliest rotary olive crushers (trapeta) are found here (Foxhall, ‘‘Oil Extraction’’ 193), since these expensive machines could be used at least to some degree every year, rather than only in alternate years. 18. Plutarch, Lycurgus 12, 2; cf. Foxhall and Forbes, ‘‘Σιτομετρεία’’ 58–59. 19. Gallant, Risk and Survival 82–92; Foxhall, ‘‘Control of the Attic Landscape’’; BurfordCooper, ‘‘Family Farm’’ 168–72; Jameson, ‘‘Agricultural Labor.’’ The plots of land reconstructed in the chora of Metapontum range in size from 6.6 to almost 60 ha; almost half of them are 13.2 ha in area, and another 26 percent are 26.4 ha (Carter, ‘‘Metapontum—

330

Notes to Pages 225–227

Land, Wealth and Population’’ 427). Land plots in the polis of Chersonesos in the Crimea range from 3.9 to 4.5 ha, while those on the peninsula to its south, belonging to the city, are 26.4 ha (Pečírka, ‘‘Country Estates’’ 470–71; Zerebtzoff in Saprykin, Ancient Farms and Land-Plots 121–47). Lohmann concludes that of thirty-three farmsteads discovered in the deme of Atene, eight or nine belonged to estates with about 25 ha of land: Lohmann, ‘‘Agriculture and Country Life’’ 51. After the destruction of Olynthus, Lysimachus made a grant to Limnaios of an estate of 2,480 plethra (248 ha), including 360 plethra (36 ha) just southwest of Olynthus, although this is clearly an exceptional case: Hatzopoulos, Donation. 20. Gallant, Risk and Survival 77, table 4.7, lists the average yields for different regions of modern Greece between 1911 and 1950; these range from 470 kg/ha (609 liters, in Arcadia) to 903 kg (1,170 liters, in Kavala). Garnsey proposes yields of 800 liters/ha for wheat and 1,200–1,600 liters/ha for barley (Garnsey, Famine and Food Supply 95–106; Garnsey, ‘‘Yield of the Land’’). Osborne, Classical Landscape 44–46 estimates 900 kg (1,165 liters) net per hectare in Attica. Sallares doubts whether yields could have exceeded 650 kg/ha except on very good soils (Sallares, Ecology of the Ancient Greek World 372–89). Sanders, ‘‘Reassessing Populations,’’ documents much higher yields on Melos in the seventeenth century, more than 1,900 kg/ha. Cf. the skepticism of Stroud, Athenian Grain-Tax Law 35 and n. 73. The Chalcidice is likely to have been near the high end of the range of ancient yields, whatever they may have been. 21. M.-C. Amouretti and J.-P. Brun, ‘‘Le Rendements,’’ in Amouretti and Brun, Oil and Wine 551–55; Forbes, ‘‘Ethnoarchaeological Approach’’ 93–98, based mainly on evidence from Roman North Africa and the southern Argolid; Amouretti, Le pain et l’huile. Osborne, Classical Landscape 45, estimates a yield in Attica of 400 kg of oil per ha in a good year, 150 kg in a bad year. 22. Olynthus 8, 312–16 divides them on the basis of shape, but I suspect that the distinction in size is more significant. The capacities of only a few pithoi from the site were measured accurately. The others must be assigned on the basis of rough measurements, more casual observations, and photographs. The smaller variety might be a phidakne rather than a pithos: in the Attic Stelai these vessels have an average capacity of about twelve amphora (480 l), whereas the capacities of pithoi in the Stelai average about twenty amphorai (800 l). Amyx, ‘‘Attic Stelai 3’’ 170–73; Olynthus 8, 314 n. 7. For the purposes of this argument I will call them both pithoi, distinguishing simply between large and small. 23. The first pithos contained ‘‘traces of straw and also of the kernels of olives. Of the latter both the outlines and fragments of the husk were actually found.’’ The pithos also contained an iron spearhead and fragments of a lid (1934 fieldbook JWG pg. 77). The second pithos, containing straw and grain, is recorded on the same fieldbook page. The third pithos contained ‘‘pine wood, chiefly the bark . . . a few fragments were charred, but most of the pieces were in their natural state. It was clear from the topsy-turvy arrangement of the structural lines that a great many separate fragments had become fused into solid lumps’’ (1934 fieldbook JWG pg. 88–89). The other two pithoi in this room were broken and their contents not preserved or noted. 24. Classical sources almost exclusively mention wine or other liquids in pithoi (for instance, Hom., Od. 2.340; Hom., Od. 23.305; Aristoph., Peace 703; Plato, Gorgias 493b; Diod. Sic. 13.83.3; Eur., Cyclops 217). But archaeologically documented pithoi often contain grain

Notes to Page 227

331

and other dry foodstuffs (for instance, Constantini, ‘‘Monte San Mauro—Paleoetnobotaniche’’; Jones et al., ‘‘Crop Storage at Assiros’’). 25. Olynthus 8, 61. 26. Olynthus 8, 314–16; Olynthus 12, 205. Even cracked pithoi were sold in the Attic Stelai, as were pithos lids (ἐπιθέματα), both with their pithoi and separately: Amyx, ‘‘Attic Stelai 3,’’ 168–69. 27. If shattered, pithoi may well have been ignored in excavation together with other coarse pottery, while the lids may have been considered more recognizable and noteworthy. Pithoi and lids were found together in A vi 9, c; A vii 4, g; A xi 10, e; B vi 9, g; and D v 6, i; also in Sec. G, Areas 19 and 23. Disks or pithos lids without pithoi were found in A 3, b; A 4, b; A 6, g; A 10, a; A 11, e and h; A iv 5, m; A iv 9, i; A vi 9, a; A vi 10, d and i; A vii 4, f; A viii 5, i; B vi 4, d; B xi 1, andron; D v 6, d, e, l, and n; ESH 1, i; ESH 4, h; the House of the Comedian, b; the House of Many Colors, b; the House of the Tiled Prothyron, c and e; the Villa of Good Fortune, b; the South Villa, a; Sec. J/K, 19e, 9s; and on Street v at A v 6. Of these, the clearest example is A 4, room b, a large room empty except for fifteen slate and terracotta disks. These were probably the lids of pithoi which were salvaged after the destruction, and the room served as a large-scale storeroom (see below, ‘‘Communal Storage’’). In the House of Many Colors, the disk in room b was probably being used to mix cement for the mosaic floor of this or the adjoining room (above, ch. 3). 28. A storeroom (ὁ πιθεών) is mentioned in the sales inscription from A v 10, together with ‘‘the things which bring income’’ (τὰ. μι . [σθ]οφόρα πάντα); the storeroom was probably associated with household industry, in this case probably pressing olive oil, rather than with food storage. See above, ch. 3, ‘‘House A v 10,’’ and below, ‘‘Sales Inscriptions.’’ 29. Amyx, ‘‘Attic Stelai 3’’ 195–97. 30. Sparkes and Talcott, Agora 12, 195–96, 343–45, no. 1542, the volume calculated roughly from their dimensions. 31. House A v 1 contained more than fifteen storage amphoras; A 11, D v 6, and the House of Many Colors had more than a dozen each. The capacity of Olynthian amphoras was not measured; but at 15–25 liters, this amounts to a maximum of only 375 liters of storage, the equivalent of about two small pithoi. The Attic Stelai list empty amphoras (twenty-one sold for 3 drachmas, or a paltry 1⁄7 obol each) as well as empty Chian and Eretrian amphoras, the latter of which sold for at least 3 obols each (Amyx, ‘‘Attic Stelai 3’’ 174–78). The capacity of the large lot of empty amphoras is not specified. If they were full-size transport amphoras, though, they might have had a total capacity comparable to a large pithos (large Corinthian type B amphoras average about 50 liters in capacity, so twenty-one of them would hold about 1,050 liters). The extremely low price suggested to Amyx that they were either smaller than ‘‘full size’’ or else in very poor condition; but they may also have been so common as to be extremely inexpensive. Compare the price of decorated Panathenaic amphoras in the Stelai, at 2.4–3.7 obols (Amyx, ‘‘Attic Stelai 3’’ 178). 32. Rickman, Roman Granaries 85–86. 33. Gallant, Risk and Survival 97–98; Forbes and Foxhall, ‘‘Ethnoarchaeology and Storage.’’ See Theophrastus, Historia Plantarum 8.11: ‘‘For propagation and sowing generally, seeds one year old seem to be the best; those two or three years old are inferior, while those kept a still longer time are infertile, though they are still available as food.’’ Theophrastus goes on to describe a type of earth found at Olynthus which preserves grain.

332

Notes to Pages 227–229

Gallant proposes that plastered rooms adjacent to kitchens were not used as bathrooms but rather as grain stores (pg. 96). He particularly notes rooms in houses A 9, A vi 2, A vi 9, and A vii 1, although of these houses, only A vi 9 had such a room. But the presence of bathtubs and gaps where tubs have been removed, drainage basins set in the cement floor and other features in so many of these rooms, and the lack of clear doors or other means of closing the room off, makes it doubtful that these served as storerooms. 34. The stratification in the House of the Twin Erotes, room j was ‘‘odd, being composed of tile, then beaten and hard earth, then more tile in successive and repeating layers’’ (fieldbook JRC pg. 38). These layers might have been put down to combat the damp. 35. The exceptions are the storeroom of the House of the Tiled Prothyron, which contained two upper grindstones, a storage amphora, and a variety of other artifacts, and that of the South Villa, which had seven loomweights and several plain vases (perhaps for storage of other foodstuffs?). 36. A number of fragments of wood were found in the storeroom in the Villa of Good Fortune, while the House of Many Colors, the House of the Tiled Prothyron, and the House of the Twin Erotes had collections of hardware: nails, spikes, and/or bosses. 37. Stroud, Athenian Grain-Tax Law 84–104. Gallant, Risk and Survival 179, claims that communal storage was exceedingly rare in ancient Greece, citing the facilities on the South Hill at Olynthus among the few known examples. See also Garnsey, Famine and Food Supply, esp. ch. 5. 38. IG I 3 78, 10; IG II 2 1672, 292 (Eleusis). 39. See Carroll-Spillecke, ΚΗΠΟΣ. 40. Aristotle, Pol. 1277b, cf. Pol. 1328b–1329b. 41. Dem. 57.30. 42. Hdt. 2.167. The sentiment is repeated frequently, especially by philosophers and orators: e.g., Plato, Laws 743D, 842C-D, 846D-E, 919D-920A, etc.; Aristoph., Cl. 991; Isocrates 7.48, etc. 43. On agricultural processing in general, Amouretti, Le pain et l’huile; Amouretti and Brun, Oil and Wine; Moritz, Grain-Mills and Flour; Wells, ed., Agriculture; Forbes and Foxhall, ‘‘Queen of All Trees’’; Forbes, ‘‘Ethnoarchaeological Approach’’; Amouretti, ‘‘Oléiculture et viticulture’’; Isager and Skydsgaard, Ancient Greek Agriculture. 44. See Osborne, ‘‘Buildings and Residence’’; IG XII 5, 872, with Étienne, Ténos II. On vintage scenes, Sparkes, ‘‘Treading the Grapes.’’ 45. Foxhall, ‘‘Oil Extraction’’ questions the identification of facilities at Olynthus and Halieis, pointing out that the assemblages are usually quite incomplete, and there is no archaeological evidence for important stages in the processing. But the relatively common discovery of press beds, crushers, and other equipment for processing olives and grapes (found in situ rather than in storage for later use in the country) shows that at least some of the crops were being processed in urban rather than rural houses. See most recently Ault, ‘‘Koprones and Oil Presses.’’ 46. It is difficult to quantify just how many olives an ancient Greek press could accommodate. Most research has been aimed at Roman olive presses, for instance, Mattingly, ‘‘Megalithic Madness’’; Mattingly, ‘‘Technical Specifications’’; Mattingly, ‘‘Maximum Figures’’; see, in general, M.-C. Amouretti and J.-P. Brun, ‘‘Le Rendements,’’ in Amouretti and Brun, Oil and Wine 551–62; Foxhall, ‘‘Oil Extraction.’’ Greek presses were in general simpler

Notes to Pages 230–238

333

and less efficient than later, Roman ones; nevertheless, they remain a very substantial investment, beyond the reach of most farmers. 47. Aristotle, Pol. 1259a. 48. Foxhall, ‘‘Oil Extraction.’’ 49. The orbes and mortaria now at Olynthus illustrated in Isager and Skydsgaard, Ancient Greek Agriculture pl. 3.10 were not found in the excavations but were probably brought to the site from somewhere nearby. Their date is therefore uncertain, but Foxhall believes that they could be from the time of the city (ibid., 60 n. 37). 50. Olynthus 8, 339–43; Graham, ‘‘Olynthiaka, 2.’’ On the question of the reoccupation of this house, see above, chapter 2, ‘‘Olynthus after 348 b.c.’’ Foxhall argues that this facility was not an olive press but some other industrial establishment because neither a collecting basin nor a press bed were found in the room (Foxhall, ‘‘Oil Extraction’’ 187– 88). As Foxhall points out, though, the equipment for processing olives, grapes, and other agricultural produce is mostly lightweight, easily transportable, and multipurpose, and the discovery of a number of postdestruction coins in this house demonstrates that it was at least ransacked (although probably not reoccupied) after the destruction. This may well have been done to salvage valuable and reusable equipment like press beds, pithoi, and the like. 51. Again, these finds almost certainly predate the destruction, although this house also contained a number of coins which postdate 348 b.c. Brazier: Olynthus 10, no. 570. 52. The narrower door is not shown on the plan, Olynthus 2, fig. 182, but inferred from the unpaved strip along the north side of room j and lower foundations of the street wall. Other such double doors are found in houses A 9 (see the reconstruction in Olynthus 8, 250, fig. 21), A v 6, A v 10 (which also had an olive crusher in its courtyard), A viii 4, A xi 10 (reconstruction in Olynthus 8, pl. 69), and the Villa of the Bronzes. Many of these houses engaged in agricultural processing, which made it necessary to bring carts directly into the house. 53. See the photo Olynthus 2, fig. 185, where it looks as if the cement floor slopes down to the north, and the detail of the gap at the north, fig. 191. 54. Olynthus 2, 72–73 and fig. 192 (where it is placed near the bathroom b). 55. Foxhall, ‘‘Oil Extraction’’ 190. 56. Olynthus 2, 69–71; Olynthus 8, 326 nos. 1–7 and 333 nos. 3–7. The lower stones were apparently not saved. See also Moritz, Grain-Mills and Flour; Runnels and Murray, ‘‘Milling in Ancient Greece.’’ 57. See Olynthus 8, 342–43. 58. Olynthus 8, 342. 59. Silver hoard 4: Olynthus 9, 171–73. 60. Foxhall, ‘‘Oil Extraction’’ 190 suggests that the mortar might have been part of an olive crusher. 61. As was probably done in the Bronze Age: Forbes and Foxhall, ‘‘Queen of All Trees’’ 39–40; Foxhall, ‘‘Oil Extraction.’’ Vase paintings usually show grapes being crushed in raised basins, probably wooden, or baskets, from which the juice flows out a spout and into a pithos or basin (Sparkes, ‘‘Treading the Grapes’’). This method would be most convenient in the vineyard, and it is therefore a more suitable subject for paintings of satyrs

334

Notes to Pages 239–246

and their compatriots at work; but in the home a permanent installation might have been more efficient. 62. Olynthus 2, 55 and fig. 146; Olynthus 8, 331. As Robinson describes it: ‘‘a rough smaller round section (1.19 m. in circumference) goes below, and the larger section with a raised broad outer edge and with grooves inside goes above. It has a deep cemented square hole in the middle in which a pivot probably was placed to enable it to turn on the lower section and so grind the grain. It is worn on the under side. [footnote: Ht. 0.34 m., of upper section 0.25 m., upper diameter 0.51 m. hole 0.07 m. square. Cf. for somewhat similar millstones at Delos, Exploration Archéologique de Delos, VIII, p. 229.]’’ See Moritz, Grain-Mills and Flour 34–52; Runnels and Murray, ‘‘Milling in Ancient Greece’’; Runnels et al., Artifact and Assemblage. My thanks to Curtis Runnels for discussing this issue with me. 63. Olynthus 2, 55, figs. 161, 198. 64. Olynthus 8, 111. 65. At 3.35 × 5.32 m, the room is two and a half times as large as the average flue, and half again as large as the largest household flue. 66. Sparkes, ‘‘Greek Kitchen’’ 127–29; Sparkes, ‘‘Not Cooking, but Baking’’; Frayn, ‘‘Home-Baking’’; Cubberley et al., ‘‘Testa and Clibani’’; Cubberley, ‘‘Bread-Baking.’’ 67. Olynthus 9, 176–83, hoards 6–8. 68. Forty-two fishhooks were found on the South Hill, as opposed to forty-four from the entire North Hill (with a much larger excavated area) and seven from the Villa Section. Fifteen of the fishhooks from the North Hill came from streets—a remarkable number considering what a small area of streets was excavated; seven of those from the South Hill came from streets. 69. Olynthus 10, no. 2504. Vases include a lekythos, a skyphos, a plate, two lids, and a miniature skyphos. 70. Gallant argues that netting was the only efficient way to catch fish in antiquity and that the economic importance of fishing has been generally overrated: Gallant, A Fisherman’s Tale. 71. Olynthus 12, 37 and pl. 25.2 (mislabeled). 72. Those from A vii 9 are recorded in groups of 31, 26, 12, 11, and 8. Those from A v 10 are recorded in groups of 38, 23, and smaller groups from rooms b, d, h, and i. It is unclear whether the groups not attributed to rooms were found all together in one room, or were simply the day’s total collected loomweights from that house. In either case, however, the large number of loomweights suggests that weaving was done in these houses on a fairly large scale. 73. 85 loomweights with markings were inventoried and weighed from the hoard in A viii 7 (by Robinson himself, the recorder apparently balking at the enormous pile; Olynthus 12, 34–35 n. 105; 1938 Loomweight Inventory fieldbook). Of these 37 were conical and 47 pyramidal, and 1 was a squat pyramid. Despite the difference in shape, however, most weigh between 17 and 20 drachms (not grams, as published), and only 8 fall outside that range. The pyramidal loomweights are slightly heavier, averaging 19 drachms, while the conical weights average about 17.5 drachms; the difference is statistically significant. They may therefore not have been intended to be used together (perhaps they were stored in different bags).

Notes to Pages 246–252

335

74. See Barber, Prehistoric Textiles 95–96; Hoffmann, Warp-Weighted Loom 42. There may be reasons to use loomweights of different sizes on a loom, for instance, to strengthen the selvedges. In many cases, though, households probably used what was available, perhaps tying more warp threads to the heavier loomweights to make up for the differences in weight. 75. Olynthus 2, 55–68; altars Olynthus 8, 322, nos. 6–7, 8–9, 10–11. Pairs of portable altars were found in houses A 1, A vii 9, A viii 8, B vi 7, the House of Many Colors, the House of the Tiled Prothyron, and the Villa of the Bronzes. 76. Runnels, Millstones from the Argolid; Runnels and Murray, ‘‘Milling in Ancient Greece.’’ 77. See Hopper, Trade and Industry; Burford, Craftsmen. With ‘‘manufacture’’ we could include weaving, stoneworking, and other such occupations; here I use it in a more ‘‘industrial’’ sense, the production of goods which are not normally produced in smaller quantities by the household itself. 78. Olynthus 4, nos. 410–22, all from room i in the southeast corner of the house. On figurines, their manufacture and use, Higgins, Tanagra and the Figurines, 64–70; Uhlenbrock, Coroplast’s Art; Raeder, Priene: Funde, and work in progress on figurines from Priene by Frank Rumscheid. 79. Olynthus 7, no. 397, a seated boy made from mold Olynthus 4, no. 417. 80. Giardino, ‘‘Herakleia’’; Bagnasco, Locri IV; Cuomo di Caprio, ‘‘Ateliers de potier’’; Bagnasco, ‘‘Housing and Workshop Construction.’’ Compare the potter’s quarter at Corinth; Williams, ‘‘Early Urbanization of Corinth’’; Arafat and Morgan, ‘‘Pots and Potters.’’ 81. Olynthus 14, no. 364, and two unpublished mold fragments. 82. McPhee, ‘‘Red-Figure Vase-Painters.’’ 83. Cuomo di Caprio, ‘‘Ateliers de potier’’ 80–81. 84. Olynthus 12, 269. The court is very badly destroyed, though, and not much can be made out from the photographs or the plan. 85. Olynthus 10, 419–20, fig. 23. On such slingbullet molds, see Weiß, ‘‘Schleuderbleie und Marktgewichte.’’ 86. There is no evidence of religious ritual in this building in the form of altars or other equipment, but the possibility of a cult association should nonetheless be kept open. See Lohmann, ‘‘Agriculture and Country Life’’ 35 and figs. 12–13 for a somewhat similar building, also with simply an andron and anteroom, at the Attic deme site of Halai Aixonides. 87. McDonald, ‘‘Villa or Pandokeion?’’ His arguments are accepted by Travlos (see the addendum to his article, pg. 373) and by Kraynak, Hostelries of Ancient Greece 43–49. Cf. also D’Arrigo in Richerche sulla Casa in Magna Grecia e in Sicilia. 88. McDonald, ‘‘Villa or Pandokeion?’’ 369. 89. Olynthus 12, 19. 90. Cf. the photos Olynthus 12, pl. 12:1. 91. Graham does not mention the lost installation in the court, but the pipe in the southwest must have been connected to something, and the settling basin suggests that it was not simply for runoff from the eaves (Olynthus 8, 70–71). He proposes that the basin drained to the north, but in these houses water was drained out through the alley at the south through outlets in the city wall; there is no outlet to the north of the basin. The basin

336

Notes to Pages 252–258

ought therefore have been filled, not drained from the north. The few finds from the house do not help identify its use. 92. Graham insisted that water must have flowed out to the alley instead of in towards the court (Graham, ‘‘Olynthiaka, 6’’ 337; contra Olynthus 12, 12–13). The pipe does slant down from the alley into the court, however, and so cannot be a drain; the joints are sealed with mortar, also suggesting a supply rather than a drainpipe. 93. Olynthus 2, 107; Olynthus 3, 5, 120 and fig. 2. Robinson (Olynthus 12, 73 n. 52) refers to these as coming from house B iii 2 rather than B ii 6; if so, the text of Olynthus 2, 107 is misleading since it clearly refers to houses B ii 4 and 6 (cf. the photo of the room, fig. 257). 94. Olynthus 12, 73–77; Olynthus 14, 403–6. The coin inventory for 1938 records the hoard as coming from square m7, which falls within room e, the pastas of the house; it was published as coming from room a, however. An earring found with the hoard was correctly published as coming from the pastas (Olynthus 10, no. 277). Robinson says ‘‘there was no sign of a container,’’ but the fieldbook records a fragmentary vase in which the flans were found. The vase was not saved, probably explaining its omission from the publication. 95. Olynthus 2, 107; Olynthus 12, 73; Olynthus 14, 403. 96. E. S. G. Robinson and M. Price have suggested, for instance, that Athenian kollyboi were minted by bankers, money-changers, and other private individuals (Robinson, ‘‘Coinage of Athens’’; Price, ‘‘Early Greek Bronze Coinage’’). Kroll however believes that these kollyboi were nonexistent, and that this was the point of the jokes in Aristophanes and Eupolis (Kroll, Agora 26, 24–25). 97. Compare tables Olynthus 14, 404 and Olynthus 9, 299. Of the 15 flans from A iv 7 and the street, only 2 are less than 17 mm in diameter, and 4 are larger; they average 17.8 mm in diameter. By contrast, of the 732 Chalcidic bronzes of the largest denomination (denomination A) measured by Clement, only 47 were 17 mm in diameter, and only 7 were as large as 18 mm. The Bottiaean bronzes are slightly but not significantly larger (Olynthus 9, 303). The flans from B ii 6 are not published as completely as those from A iv 7. According to the photographs in Olynthus 3, pl. 26 nos. 968–79, they vary in size from about 9 mm to about 16 mm, about the right size for Chalcidic bronzes. Other pieces are larger and apparently unusually thick, to judge from the photograph of the whole hoard, pg. 120, fig. 2. Robinson says that some ‘‘are as high [thick?] as 0.012 m, with a circumference [diameter?] of 0.03 m’’ (Olynthus 3, 5). 98. Kroll suggests that these might be counterfeits rather than legitimate coins, and the evidence of size supports his hypothesis. The Athenian law concerning the testing of silver coinage reveals the Athenians’ concern with counterfeiting and the extent of the problem: Stroud, ‘‘Silver Coinage.’’ 99. Olynthus 9, no. 33; the other ancient forgery is in the Metropolitan Museum (ibid. no. 25). Counterfeits could also be made from silver-plated lead: see Stroud, ‘‘Silver Coinage’’ (different terms for counterfeits, ὑπ[όχαλκον], ‘‘bronze underneath,’’ ὑπομόλυβδον, ‘‘lead underneath,’’ and κίβδηλον, ‘‘base’’). The blanks might have been intended to counterfeit bronze coinage, but this hardly seems worth the effort and risk. 100. Olynthus 9, 347. It need not have been made in Olynthus, of course. 101. For instance, compare the views of Jameson, ‘‘Agriculture and Slavery’’ to the reply by Wood, ‘‘Agricultural Slavery.’’ Gallant, Risk and Survival 30–33, takes a position between

Notes to Pages 258–263

337

the two. Cf. also Morris, ‘‘Remaining Invisible.’’ Literary sources are replete with descriptions of industrial slaves: Thuc. 7.27 refers to 20,000 slaves, most of them craftsmen (χειροτέχναι), who fled to Decelea during the Peloponnesian War. Cf. Aeschines 1.97; Dem. 27, etc. 102. Foxhall, ‘‘Oil Extraction’’ 197. 103. Slaves were often sold together with the workshops they worked in: e.g., IG II 2 1122; 2747–2749 = Finley, Land and Credit nos. 88–90; Dem. 27.9–10; Lycurgus 1.22, 1.58. See also Stanley, ‘‘Value of Ergasteria.’’ 104. Cf. Morris, ‘‘Remaining Invisible’’; Jameson, ‘‘Domestic Space’’ 103–4. 105. George, ‘‘Servus and domus.’’ 106. Pritchett and Amyx, ‘‘Attic Stelai.’’ 107. Olynthus 2, 111. 108. Olynthus 8, 21–22; Olynthus 12, 79–114. Robinson tempered his opinion in Olynthus 12, 96: ‘‘the area to the west [of the fountainhouse] was open and probably used for military maneuvers as well as for an agora.’’ 109. In a note to excavators in 1938: ‘‘Study system of blocks in the Villa region. Should be checked by careful measurements—and tried at several points—and then followed up, if it proves correct, by trying to trace out the plan and looking for an agora’’ [margin: ‘‘My hypothesis’’]. 110. Wycherley, ‘‘Olynthus and Selinus’’ 232–33; Martin, L’Agora 386–90. 111. Hoepfner and Schwandner, Haus und Stadt 78–79. 112. Dem. 18.169. 113. Contrast these quantities to the 16,557+ Greek coins of all periods (6th century b.c. through 4th century a.d.) found in the Athenian Agora between 1931 and 1990 (Kroll, Agora 26 xviii–xxvi, 1–3). 114. Olynthus 9, 299–300. 115. Olynthus 10, 165–66, hoard 1. 116. Olynthus 10, 171–73, hoard 4. 117. Olynthus 10, 168–71, hoard 3. 118. Olynthus 10, 176–83, hoards 6–8, respectively. 119. Olynthus 10, 167–68, hoard 2. 120. Olynthus 10, 186–90, hoard 10. 121. Olynthus 10, 191–93, bronze hoard 1. The hoard was found at about floor level of room j, near the east wall. 122. Cohen, Athenian Economy and Society 119 ff., and generally. 123. The proportion at Olynthus varies depending on whether one considers all coins from the site or only those from houses, from stratified contexts, etc. Considering all coins from the site dated between ca. 432 and 348 b.c., the proportion of foreign coins is about 26 percent. But if one considers only coins from stratified contexts in houses and public buildings, excluding coins from hoards, graves, surface finds, and other contexts, the proportion rises to about 37 percent. For the Athenian Agora, Kroll, Agora 26 xviii–xxvi, 168–69. 124. Hypothesizing that some houses would have assemblages of coins similar to one another but different from other houses, I used various clustering routines on the coin assemblages, trying both Q-mode (clustering over findspot) and R-mode (clustering over

338

Notes to Pages 263–273

mint) methods. I also tried grouping coins from each region of Greece together (Thessalian coins, Macedonian coins, coins of the Macedonian kings, and so forth) and clustering the resulting groups. The resulting clusters, however, showed no meaningful spatial patterning and were not significantly different from one another. Other analyses were no more productive. 125. Kroll, Agora 26 168–69. 126. Aeschines 1.123–24. 127. On Greek shops and workshops in general, see Bettalli, ‘‘Case, botteghe, ergasteria’’; Stanley, ‘‘Value of Ergasteria.’’ Horne, ‘‘The Household in Space’’ describes modern distributed holdings. The sales inscription from A v 10 specifies that the sale includes ‘‘the pitheon’’ or storage room, which does not seem to have been part of the house, as no pithoi were found in it; this room might have been a shop in another house elsewhere in the city. 128. Avenue C, which was much less fully explored, might also have had a large number of shops—for instance, house B vi 9 has three very regularly planned shops opening onto the street like those of house A iv 9. This street is also wider than average, but not as wide as Avenue B. 129. The avenue was not widened as an afterthought, as proposed in Olynthus 8. Shops in A iv 9, A v 10, and the early building incorporated into house A vii 10 are certainly original, and other shops probably so, although it is difficult to distinguish a door cut into a house wall later from one which was part of the original design. The workshop in A vii 10 suggests that this was already an important axis before the city was entirely built up. 130. The streets of Olynthus were not systematically excavated, and many of these deposits were exposed only when the excavation of a house was continued a small distance into the street, or by trial trenches to locate the house and block walls. It is impossible therefore to properly quantify the assemblages, or to analyze the distribution of artifacts to try to distinguish garbage from other sorts of assemblages. 131. The basic publications are Robinson, ‘‘Inscriptions, 1928,’’ ‘‘Inscriptions, 1931,’’ ‘‘Inscriptions, 1934’’ and ‘‘Inscriptions, 1938.’’ Important new readings of some inscriptions are published by Hatzopoulos, Actes de Vente, together with new inscriptions from nearby Chalcidic communities. Commentaries on the corpus include Hatzopoulos, Actes de Vente; Hennig, ‘‘Kaufverträge’’; Youni, ‘‘Inscriptions olynthiennes’’; and Nevett, ‘‘Real Estate ‘Market.’’’ On the numeric system, Tod, ‘‘Acrophonic Numerals’’ 248–49; Graham, ‘‘X=10.’’ The new inscription, discovered at Skoinia which borders on the archaeological site, is published by Trakosopoulou, ADelt 39 (1984) B 226 (cf. BCH 115 [1991] 908, w. photo; Hatzopoulos, BE 104 [1991] no. 408, with a reading of the latter part of the inscription; SEG 39 [1989] 617). That this inscription belongs with the city proper and not to a neighboring settlement is suggested by the presence of Lycophron the son of Theodoros among the witnesses of the sale; he also made a loan for a house, attested in the inscription found in Avenue A at A −4. 132. Robinson, ‘‘Inscriptions, 1938’’ 47–50, no. 3. 133. In front of houses A iv 7, A viii 3, D v 6, and on the South Hill, Tr. 10; inside houses A 11 and A v 10. 134. The inscription found in house A 11 specifies ἐνιαυτὸν ἐπὶ ἀπόλυσιν, ‘‘on condition of release (or deliverance) in a year.’’ There is no verb to distinguish whether this is a sale or loan. This seems to be a πρᾶσις ἐπὶ λύσει, a conditional sale with the provision

Notes to Pages 273–277

339

that the property be returned to its original owner within a year, when the owner repays the money given him, amounting to a loan. Other interpretations are possible, however: Robinson argued that the phrase meant that the buyer would not actually take possession for a year (Robinson, ‘‘Inscriptions, 1931’’ 44). The inscription found in house A viii 3 uses the verb καθίεται, ‘‘to let down’’; the usage is unique in sales inscriptions. The price is set at 900 dr for a period of five years and four months. This might be another conditional sale, as Hatzopoulos argues. See Hatzopoulos, Actes de Vente 57–64; Youni, ‘‘Inscriptions olynthiennes’’ 139–40; Finley, Land and Credit 31–37; Étienne, Ténos II 52–58; Millett, ‘‘Sale, Credit and Exchange’’ 176ff.; Millett, ‘‘Attic horai’’; Harris, ‘‘Real Security.’’ 135. Hatzopoulos, Actes de Vente 72–77, table pg. 80. He identifies Kellion with the ancient Stolos and Smixi with the ancient Polichnai, and argues that none of the documents from Stolos can date earlier than 355, when it and other Chalcidic cities were incorporated into the Chalcidic League. The terminus ante quem is 349 when Philip captured the cities. He works out the internal chronology from the number of inscriptions of each year, assuming that the more recent sales will be more frequently preserved, and from changes in spelling. 136. Olynthus 2, 101; Hatzopoulos, Actes de Vente 58–59; Youni, ‘‘Inscriptions olynthiennes’’ 135–36. The inscription was found in the cistern of house B iii 2; Diopeithes might well have owned both houses. 137. Cf. Hatzopoulos, Actes de Vente 58–59. Similar ties are found among the buyers and sellers and neighbors of houses in Stolos and elsewhere. 138. Robinson changed his reading of numerals to increase the price of the inscription from the Villa of Good Fortune, reading 400 χ [ ρυσιοί ] or gold staters, instead of 410 drachmas (‘‘Inscriptions, 1934’’ 125–26). Hennig finds the range ‘‘nicht verständlich’’ and proposes that regardless of the outward form of the sale, many of the transactions were actually loans (Hennig, ‘‘Kaufverträge’’ 151–53). Cf. Hoepfner and Schwandner, Haus und Stadt 317. 139. Hatzopoulos, Actes de Vente nos. I–III. 140. Étienne, Ténos II 60–84. 141. Pritchett and Amyx, ‘‘Attic Stelai’’ 261–76. 142. Hatzopoulos, Actes de Vente, 76–77; also BE 104 (1991) no. 408. 143. Contra Nevett, ‘‘Real Estate ‘Market’ ’’ 337–38. 144. Davies, ‘‘Models and Muddles’’; Parkins, ‘‘Time for Change?’’ 145. Osborne, ‘‘Buildings and Residence’’; Osborne, ‘‘‘Is It a Farm?’’’ Rural survey has become something of a growth industry in the last decade or so, and the important results attained by surveys in other parts of Greece make the absence of such information at Olynthus even more regrettable. On second houses, Plato, Laws 745B-E; Aristotle Pol. 1330a, and commonly in the Attic Orators (e.g., Aeschines 1.97–106; Dem. 42.5, etc.). 146. Xen., Ec. 7.36; Xen., Ec. 9.8. 147. Cyrene: ML no. 5, Graham, Colony and Mother City 29–30, 224–26. Kerkyra Melaina: above, chapter 5, ‘‘The Distribution of Land at Korkyra Melaina.’’ 148. Thuc. 1.58; Hatzopoulos, Macedonian Institutions 174. 149. This is, of course, difficult or impossible to quantify. Compare the situation in Athens, where a motion at the end of the fifth century to deny citizenship to those who did not own land would have resulted in the disenfranchising of only some 5,000 citizens

340

Notes to Pages 277–285

of between 20,000 to 30,000 (Dion. Hal., Lysias 43; Foxhall, ‘‘Control of the Attic Landscape’’; Osborne, ‘‘Exchange and Society’’). The greater agricultural fertility of the Chalcidice might have led to a higher proportion of landowners among Olynthian citizens than in Athens, and most of the houses on the North Hill at Olynthus probably belonged to citizens rather than metics, although this too is difficult to quantify. This does not mean that these citizens did not have other occupations as well. 150. Rural farmsteads often had towers ( pyrgoi) as defensive structures: Xen., Anab. 8.8.12–15, Dem. 47.56; Jones et al., ‘‘Vari House’’; Young, ‘‘South Attica’’; Lohmann, Atene; Saprykin, Ancient Farms and Land-Plots, etc. 151. The Attic Stelai give a reasonably fair account of what was of real worth in a household, and in them, the sorts of artifacts found in archaeological contexts, such as pots, grindstones, terracottas, and so forth, are quite rare and of low value. 152. See Meikle, Aristotle’s Economic Thought; Meikle, ‘‘Aristotle and the Political Economy’’; Cohen, Athenian Economy and Society; Burke, ‘‘Economy of Athens.’’ 153. Athenian bronze coinage does not begin until the mid-fourth century b.c. (Kroll, Agora 26, 24–27). Clement dates the earliest Chalcidic bronzes to very early in the fourth century (contemporary with Group J silver coins, ca. 398–395 b.c.: Olynthus 10, 298–99). Even if his date is not absolutely certain, the sheer quantity and variety of Chalcidic bronze coinage found in the destruction debris proves that bronze had been a common currency for some time.

Notes to Pages 285–287

341

Bibliography

Abbreviations may be found in American Journal of Archaeology 104 (2000) 3–24. Akurgal, Alt-Smyrna

Aldenderfer and Blashfield, Cluster Analysis

Allison, ‘‘Pompeiian House Contents’’

Allison, Pompeiian House Contents

Amouretti, ‘‘Les céréales dans l’Antiquité’’

Amouretti, ‘‘Oléiculture et viticulture’’

Akurgal, Ekrem, Alt-Smyrna I: Wohnschichten und Athenatempel (Ankara, Türk Tarih Kurumu Yayınları, V Dizi, Sa. 40, 1983). Aldenderfer, Mark S. and Roger K. Blashfield, Cluster Analysis. Quantitative Applications in the Social Sciences. Sage University Papers 44 (Beverly Hills, Sage Publications, 1984). Allison, Penelope M., ‘‘A Computer Data Base of Pompeiian House Contents and Its Application,’’ AJA 95 (1991) 305. Allison, Penelope M., ‘‘The Distribution of Pompeian House Contents and Its Significance.’’ (PhD diss., University of Sydney, 1992). Amouretti, Marie-Claire, ‘‘Les céréales dans l’Antiquité: Espèces, moutur, conservation, liaisons et interférences,’’ in M. Gast and F. Sigaut, eds., Les techniques de conservation des grains à long terme. Leur rôle dans la dynamique des systèmes de cultures et des sociétés (Paris, Centre National de la Recherche Scientifique, 1979) 57–69. Amouretti, Marie-Claire, ‘‘Oléiculture et viticulture dans la Grèce antique,’’ in B. Wells, ed., Agriculture in Ancient Greece. SkrAth, 42 (Stockholm, 1992) 77–86.

343

Amouretti, Le pain et l’huile

Amouretti and Brun, Oil and Wine

Amyx, ‘‘Attic Stelai 3’’

Arafat and Morgan, ‘‘Pots and Potters’’

ARV 2 Asheri, Distribuzioni

ATL

Ault, Houses and Households

Ault, Halieis 2

Ault, ‘‘Koprones and Oil Presses’’

Ault and Nevett, ‘‘Digging Houses’’

Bagnasco, ‘‘Housing and Workshop Construction’’

344

Bibliography

Amouretti, Marie-Claire, Le pain et l’huile dans la Grèce antique. Centre de Recherche d’Histoire Ancienne 67. Annales Littéraires de l’Université de Besançon 328 (Paris, 1986). Amouretti, Marie-Claire, and Jean-Pierre Brun, eds., La production du vin et de l’huile en Méditerranée (Oil and Wine Production in the Mediterranean Area). BCH Suppl. 26 (Paris, de Boccard, 1993). Amyx, Darrell A., ‘‘The Attic Stelai. Part III: Vases and Other Containers,’’ Hesperia 27 (1958) 255–310. Arafat, K., and C. Morgan, ‘‘Pots and Potters in Athens and Corinth: A Review,’’ OJA 8 (1989) 31–46. Beazley, J. D., Attic Red-Figure Vase-Painters. 2d ed. (Oxford, Clarendon Press, 1963). Asheri, D., Distribuzioni di terre nell’antica Grecia. Memoria dell’Accademia delle Scienze di Torino. Classe di Scienze Morali, Storiche e Filologiche, series 4a, no. 10 (Torino, Accademia delle Scienze, 1966). Meritt, B. D., H. T. Wade-Gery and M. F. McGregor, The Athenian Tribute Lists (Princeton, American School of Classical Studies, 1939– 53). Ault, Bradley, ‘‘Classical Houses and Households: An Architectural and Artifactual Case Study from Halieis, Greece’’ (PhD diss., Indiana University, 1994). Ault, Bradley, Halieis 2: The Houses: The Organization and Use of Domestic Space. Halieis: Excavations at Porto Cheli, Greece 2 (Bloomington, Indiana University Press, forthcoming). Ault, Bradley, ‘‘Koprones and Oil Presses in Late Classical Halieis in the Southern Argolid,’’ Hesperia 68 (1999) 549–73. Ault, Bradley, and Lisa Nevett, ‘‘Digging Houses: Archaeologies of Classical and Hellenistic Greek Domestic Assemblages,’’ in P. Allison, ed., The Archaeology of Household Activities (London, Routledge, 1999) 43–56. Bagnasco, Marcella Barra, ‘‘Housing and Workshop Construction in the City,’’ in

Bagnasco, Locri IV

Bakhuizen, Greek City

Barber, Prehistoric Textiles Barber, Women’s Work Beck, Polis und Koinon Bellinger, ‘‘Notes’’

Bergquist, ‘‘Sympotic Space’’

Berry, ‘‘Household Artefacts’’

Bettalli, ‘‘Case, botteghe, ergasteria’’

Binford, ‘‘Model Building—Paradigms’’

Blanton, Houses and Households

G. P. Carratelli, ed., The Greek World: Art and Civilization in Magna Graecia and Sicily (New York, Rizzoli, 1996) 353–60. Bagnasco, Marcella Barra, ed., Locri Epizefiri IV: Lo scavo di Marasà Sud. Il sacello tardo-arcaico e la ‘‘Casa dei Leoni.’’ Studi e materiali di archeologia 4 (Florence, 1992). Bakhuizen, S. C., ed. A Greek City of the Fourth Century B.C. BibAr 10 (Rome, ‘‘L’Erma’’ di Bretschneider, 1992). Barber, Elizabeth Wayland, Prehistoric Textiles (Princeton, Princeton University Press, 1991). Barber, Elizabeth Wayland, Women’s Work: The First 20,000 Years (New York, Norton, 1994). Beck, Hans, Polis und Koinon. Historia Einzelschriften 114 (Stuttgart, 1997). Bellinger, Alfred R., ‘‘Notes on the Coins from Olynthus,’’ in G. E. Mylonas and D. Raymond, eds., Studies Presented to David M Robinson, vol. 2 (St. Louis, Washington University, 1953) 180–86. Bergquist, Birgitta, ‘‘Sympotic Space: A Functional Aspect of Greek Dining-Rooms,’’ in O. Murray, ed., Sympotica: A Symposium on the Symposion (Oxford, Clarendon Press, 1990) 37–65. Berry, Joanne, ‘‘Household Artefacts: Towards a Re-Interpretation of Roman Domestic Space,’’ in R. Laurence and A. Wallace-Hadrill, eds., Domestic Space in the Roman World: Pompeii and Beyond. JRA Supplementary Series 22 (Portsmouth, R.I., 1997). Bettalli, Marco, ‘‘Case, botteghe, ergasteria: Note sui luoghi di produzione e di vendita nell’ Atene classica,’’ Opus 4 (1985) 29–42. Binford, Lewis R. ‘‘Model Building—Paradigms, and the Current State of Paleolithic Research,’’ in L. Binford, An Archaeological Perspective (New York, Seminar Press, 1972). Blanton, Richard E., Houses and Households: A Comparative Study. Interdisciplinary Contributions to Archaeology (New York, Plenum Press, 1994).

Bibliography

345

Blondé, Thorikos Lamps Boersma, ‘‘Goritsa’’ Borza, ‘‘Timber and Politics’’

Bruno, ‘‘Hippodamus and the Planned City’’ Burford, Craftsmen

Burford-Cooper, ‘‘Family Farm’’ Burke, ‘‘Economy of Athens’’

Burney, ‘‘Measured Plans’’ Burney, ‘‘Urartian Fortresses’’

Busolt, Staatskunde Cahill, ‘‘Lydian Houses’’

Cahill, ‘‘Olynthus and Greek Town Planning’’ Cameron and Tonka, Abandonment of Settlements and Regions

Campbell, ‘‘Shaping the Rural Environment’’ Campbell, ‘‘Sharing Out Land’’

Carington-Smith, Spinning, Weaving and Textile Manufacture Carr, ‘‘Intrasite Archaeological Records’’

346

Bibliography

Blondé, Francine, Greek Lamps from Thorikos. Miscellanea Graeca 6 (Gent, 1983). Boersma, Johannes S., ‘‘Goritsa, the Residential District,’’ BABesch 58 (1983) 61–82. Borza, Eugene N., ‘‘Timber and Politics in the Ancient World: Macedon and the Greeks,’’ PAPS 131 (1987) 32–52. Bruno, Alfred, ‘‘Hippodamus and the Planned City,’’ Historia 25 (1976) 414–28. Burford, Alison, Craftsmen in Greek and Roman Society. Aspects of Greek and Roman Life (Ithaca, Cornell University Press, 1972). Burford-Cooper, Alison, ‘‘The Family Farm in Greece,’’ CJ 73 (1977) 162–75. Burke, Edmund M., ‘‘The Economy of Athens in the Classical Era: Some Adjustments to the Primitivist Model,’’ TAPA 122 (1992) 199–226. Burney, C. A., ‘‘Measured Plans of Urartian Fortresses,’’ AnatSt 10 (1960) 177–96. Burney, C. A., ‘‘Urartian Fortresses and Towns in the Van Region,’’ AnatSt 7 (1957) 37–53. Busolt, G., Griechische Staatskunde I (Munich, 1920, 1926). Cahill, Nicholas, ‘‘Lydian Houses, Domestic Assemblages and Household Size,’’ NEA (forthcoming). Cahill, Nicholas, ‘‘Olynthus and Greek Town Planning,’’ CW 94 (2000) 497–515. Cameron, Catherine M., and Steve A. Tomka, eds., Abandonment of Settlements and Regions: Ethnoarchaeological and Archaeological Approaches. New Directions in Archaeology (Cambridge, Cambridge University Press, 1993). Campbell, Brian, ‘‘Shaping the Rural Environment: Surveyors in Ancient Rome,’’ JRS 86 (1996) 74–99. Campbell, Brian, ‘‘Sharing Out Land: Two Passages in the Corpus Agrimensorum Romanorum,’’ CQ n.s. 45 (1995) 540–46. Carington-Smith, Jill, ‘‘Spinning, Weaving and Textile Manufacture in Prehistoric Crete’’ (PhD diss., University of Tasmania, 1975). Carr, Christopher, ‘‘The Nature of Intrasite Archaeological Records and Spatial Analytic

Carroll-Spillecke, ΚΗΠΟΣ

Carter, ‘‘Metapontum—Land, Wealth and Population’’

Castagnoli, Orthogonal Planning Castagnoli, ‘‘Urbanistica di Thurii’’ Cawkwell, ‘‘Defense of Olynthus’’ Ciolek-Torello, ‘‘Pueblo Room Function’’

Clement, ‘‘Chronological Notes’’

Clement, ‘‘Coinage Epilogue’’

Cohen, ‘‘Seclusion’’

Cohen, Athenian Economy and Society

Constantini, ‘‘Monte San Mauro— paleoetnobotaniche’’

Coulson, ‘‘Chatby Reconsidered’’ Crouch, Water Management

Approaches to their Investigation,’’ Advances in Archaeological Method and Theory 7 (1984) 103–222. Carroll-Spillecke, Maureen, ΚΗΠΟΣ Der antike griechische Garten. Wohnen in der klassischen Polis 3 (Munich, Deutscher Kunstverlag, 1989). Carter, J. C., ‘‘Metapontum—Land, Wealth and Population,’’ in J.-P Descoeudres, ed., Greek Colonists and Native Populations (Oxford, Oxford University Press, 1990) 405–41. Castagnoli, F., Orthogonal Town Planning in Antiquity (Cambridge, MIT Press, 1971). Castagnoli, F., ‘‘Sull’urbanistica di Thurii,’’ PP 139 (1971) 301–7. Cawkwell, G. L., ‘‘The Defense of Olynthus,’’ CQ 56 (=n. s. 12, no. 1 [1962]) 122–40. Ciolek-Torello, Richard, ‘‘An Alternative Model of Pueblo Room Function,’’ in H. J. Hietala, ed., Intrasite Spatial Analysis in Archaeology. New Directions in Archaeology (Cambridge, Cambridge University Press, 1984) 127–53. Clement, P. A., ‘‘Chronological Notes on the Issues of Several Greek Mints,’’ AJP 62 (1941) 157–68. Clement, P. A., ‘‘The Chalcidic Coinage: Epilogue,’’ in B. Laourdas and C. Makaronas, eds., Ancient Macedonia: Papers Read at the First International Symposium Held in Thessaloniki (Thessaloniki, Institute for Balkan Studies, 1970) 253–55. Cohen, D., ‘‘Seclusion, Separation and the Status of Women in Athens,’’ GaR 36 (1989) 3–15. Cohen, Edward, Athenian Economy and Society: A Banking Perspective (Princeton, Princeton University Press, 1992). Constantini, L., ‘‘Monte San Mauro Caltagirone. Analisi paleoetnobotaniche dei semi contenuti nei pithoi 4 e 6,’’ BdA 64 (1979) 43–44. Coulson, William, ‘‘Chatby Reconsidered,’’ JEA 73 (1987) 234–36. Crouch, Dora P., Water Management in Ancient

Bibliography

347

Cubberley, ‘‘Bread-Baking’’

Cubberley et al., ‘‘Testa and Clibani’’

Cuomo di Caprio, ‘‘Ateliers de potier’’

Dalcher, Peristylehaus 1 von Iaitas

Davies, ‘‘Models and Muddles’’

Demand, Urban Relocation

Dengate, ‘‘Solution’’

Dengate, ‘‘Abandonment’’

DeVries, ‘‘Greeks and Phrygians’’

Dilke, Roman Land Surveyors

348

Bibliography

Greek Cities (New York, Oxford University Press, 1993). Cubberley, Anthony, ‘‘Bread-Baking in Ancient Italy: clibanus and sub testu in the Roman World: Further Thoughts,’’ in J. Wilkins, D. Harvey, and M. Dobson, eds., Food in Antiquity (Exeter, University of Exeter Press, 1995) 55–68. Cubberley, A. L., J. A. Lloyd, and P. C. Roberts, ‘‘Testa and Clibani: The Baking Covers of Classical Italy,’’ PBSR 56 (1988) 98–119. Cuomo di Caprio, Ninina, ‘‘Les ateliers de potiers en Grande Grèce: Quelques aspects techniques,’’ in F. Blondé and J. Y. Perreault, eds., Les ateliers de potiers dans le monde grec aux époques géométrique, archaïque et classique. BCH Supplément, 23 (Paris, École française d’Athènes, 1992) 69–86. Dalcher, Katharina, Das Peristylhaus 1 von Iaitas: Architektur und Baugeschichte. Studia Ietina 6 (Zurich, Archäologisches Institut der Universität Zürich, 1994). Davies, John K., ‘‘Ancient Economies: Models and Muddles,’’ in H. Parkins and C. Smith, eds., Trade, Traders and the Ancient City (London, Routledge, 1998) 225–56. Demand, Nancy H., Urban Relocation in Archaic and Classical Greece. Oklahoma Series in Classical Culture 6 (Norman, University of Oklahoma Press, 1990). Dengate, James, ‘‘A Solution to the Problem of Early Hellenistic Chronology, 350– 260 b.c.’’ Unpub. paper. Dengate, James, ‘‘Do the Coins Suggest the Abandonment of Olynthos in 316 B.C.?’’ AJA 94 (1990) 315. DeVries, Keith, ‘‘Greeks and Phrygians in the Early Iron Age,’’ in K. DeVries, ed., From Athens to Gordion: The Papers of a Memorial Symposium for Rodney S. Young. University Museum Papers 1. (Philadelphia, University of Pennsylvania, 1980) 33–49. Dilke, O. A. W., The Roman Land Surveyors: An Introduction to the Agrimensores (London, Newton Abbot, David and Charles, 1971).

Drerup, ‘‘Prostashaus und Pastashaus’’

Drougou and Vokotopoulou, ‘‘B vii 1’’ Ducrey et al., Eretria 8

Dufková and Pečírka, ‘‘Chora of Chersonesos’’

Ehrenberg, ‘‘Foundation of Thurii’’ England, Laws

Étienne, Ténos II

Falciai, Ippodamo

Fantham et al., Women in the Classical World

Finley, Ancient Economy Finley, Land and Credit

Forbes, ‘‘Ethnoarchaeological Approach’’

Forbes and Foxhall, ‘‘Ethnoarchaeology and Storage’’

Drerup, Heinrich, ‘‘Prostashaus und Pastashaus. Zur Typologie des griechischen Hauses,’’ MarbWPr (1967) 6–17. Drougou, S., and I. Vokotopoulou, ‘‘Ολυνθος —Η Οικια B vii 1,’’ AErgoMak 3 (1989) 339–44. Ducrey, Pierre, Ingrid R. Metzger, and Karl Reber, Le quartier de la maison aux mosaïques. Eretria. Fouilles et recherches 8 (Lausanne, Editions Payot, 1993). Dufková, Marie, and Jan Pečírka, ‘‘Excavations of Farms and Farmhouses in the Chora of Chersonesos in the Crimea,’’ Eirene 8 (1970) 123–74. Ehrenberg, V., ‘‘The Foundation of Thurii,’’ AJP 69 (1948) 149–70. England, E. B., The Laws of Plato. Publications of the University of Manchester, Classical Series 3–4 (Manchester, 1921). Étienne, Roland, Ténos II: Ténos et les Cyclades du milieu du IV e av. J.-C. au milieu du III e ap. J.-C. BÉFAR 263 bis (Paris, 1990). Falciai, F. B., Ippodamo di Mileto, Architetto e filosofo. Università degli Studi di Firenze, Istituto di Filologia Classica. Studi e Testi 2 (Florence, 1982). Fantham, Elaine, Helene Peet Foley, Natalie Boymel Kampen, Sarah B. Pomeroy, and H. Alan Shapiro, Women in the Classical World: Image and Text (Oxford, Oxford University Press, 1994). Finley, Moses I., The Ancient Economy (Berkeley, University of California Press, 1973). Finley, Moses I., Studies in Land and Credit in Ancient Athens, 500–200 BC. The HorosInscriptions (New Brunswick, Rutgers University Press, 1952). Forbes, Hamish A., ‘‘The Ethnoarchaeological Approach to Ancient Greek Agriculture,’’ in B. Wells, ed., Agriculture in Ancient Greece. SkrAth 42 (Stockholm, 1992) 87–101. Forbes, Hamish A., and Lin Foxhall, ‘‘Ethnoarchaeology and Storage in the Ancient Mediterranean: Beyond Risk and Survival,’’ in J. Wilkins, D. Harvey, and M. Dobson, eds., Food in Antiquity (Exeter, University of Exeter Press, 1995) 69–86.

Bibliography

349

Forbes and Foxhall, ‘‘Queen of All Trees’’

Forbes, Hamish A., and Lin Foxhall, ‘‘‘The Queen of All Trees’: Preliminary Notes on the Archaeology of the Olive,’’ Expedition 21 (1978) 37–47. Foxhall, ‘‘Control of the Attic Landscape’’ Foxhall, Lin, ‘‘The Control of the Attic Landscape,’’ in B. Wells, ed., Agriculture in Ancient Greece. SkrAth 42 (Stockholm, 1992) 155–59. Foxhall, ‘‘Oil Extraction’’ Foxhall, Lin, ‘‘Oil Extraction and Processing Equipment in Classical Greece,’’ in M.-C. Amouretti, J.-P. Brun, and E. Eitam, eds., La production du vin et de l’huile en Méditerranée (Oil and Wine Production in the Mediterranean Area) BCH Suppl. 26 (Paris, de Boccard, 1993) 183–200. Foxhall and Forbes, ‘‘Σιτομετρεία’’ Foxhall, Lin, and Hamish A. Forbes, ‘‘Σιτομετρεία: The Role of Grain as a Staple Food in Classical Antiquity,’’ Chiron 12 (1982) 41–90. Frayn, ‘‘Home-Baking’’ Frayn, Joan M., ‘‘Home-Baking in Roman Italy,’’ Antiquity 52 (1978) 28–33. Gaebler, Antiken Münzen Nordgriechenlands Gaebler, H., Die Antiken Münzen Nordgriechenlands 3.2 (Berlin, 1935). Gallant, A Fisherman’s Tale Gallant, Thomas W., A Fisherman’s Tale. Miscellanea Graeca 7 (Ghent, Belgian Archaeological Mission in Greece, 1985). Gallant, Risk and Survival Gallant, Thomas W., Risk and Survival in Ancient Greece: Reconstructing the Rural Domestic Economy (Stanford, Stanford University Press, 1991). Garland, Piraeus Garland, R., The Piraeus from the Fifth to the First Century BC (Ithaca, Cornell University Press, 1987). Garnsey, Famine and Food Supply Garnsey, Peter, Famine and Food Supply in the Graeco-Roman World: Responses to Risk and Crisis (Cambridge, Cambridge University Press, 1988). Garnsey, ‘‘Yield of the Land’’ Garnsey, Peter, ‘‘Yield of the Land,’’ in B. Wells, ed., Agriculture in Ancient Greece. SkrAth, 42 (Stockholm, 1992) 147–53. George, ‘‘Servus and domus’’ George, Michele, ‘‘Servus and Domus: The Slave in the Roman House,’’ in R. Laurence and A. Wallace-Hadrill, eds., Domestic Space in the Roman World: Pompeii and Beyond. JRA Supplementary Series 22 (Portsmouth, R.I., 1997) 15–24. Giardino, ‘‘Herakleia’’ Giardino, Liliana, ‘‘Architettura domestica a Herakleia. Considerazioni preliminari,’’ in

350

Bibliography

Goldman, ‘‘Spatial and Behavioral Negotiation’’

Gomme, Commentary

Gorman, ‘‘Aristotle’s Hippodamus’’

Graham, Colony and Mother City

Graham, ‘‘Olynthiaka’’

Graham, ‘‘Origins’’

Graham, ‘‘X=10’’ Gude, History

Hampl, ‘‘Olynth’’ Harris, ‘‘Real Security’’

Hatzopoulos, Actes de Vente

F. D’Andria and K. Mannino, eds., Richerche sulla casa in Magna Grecia e in Sicilia. Università di Lecce, Scuola di specializzazione in archeologia classica e medioevale; Archeologia e storia 5. (Congedo, Galatina, 1996) 133–59. Goldberg, Marilyn, ‘‘Spatial and Behavioral Negotiation in Classical Athenian City Houses,’’ in P. Allison, ed., Household Archaeology (London, Routledge, 1999) 142–61. Gomme, A. W., A. Andrewes, and K. J. Dover, A Historical Commentary on Thucydides (Oxford, Clarendon Press, 1956–81). Gorman, Vanessa, ‘‘Aristotle’s Hippodamus (Politics 2.1267b22–30),’’ Historia 44 (1995) 385–95. Graham, A. J., Colony and Mother City in Ancient Greece. (Manchester, Manchester University Press, 1964). Graham, J. Walter, ‘‘Olynthiaka. 1: The Altar in the House of Many Colors. 2: An Industrial Establishment on the North Hill. 3: The Andronitis-Gynaeconitis and House Windows. 4: The Semi-Enclosed Portico and Westward Orientation. 5: The Second Story. 6: The Kitchen Complex,’’ Hesperia 22 (1953) 196–207 (nos. 1–4); 23 (1954) 320–46 (nos. 5–6). Graham, J. Walter, ‘‘The Origins and Interrelations of the Greek House and the Roman House,’’ Phoenix 20 (1966) 3–31. Graham, J. Walter, ‘‘X=10,’’ Phoenix 23 (1969) 347–58. Gude, Mabel, A History of Olynthus. Johns Hopkins University Studies in Archaeology 17 (Baltimore, Johns Hopkins University Press, 1933). Hampl, F., ‘‘Olynth und der chalcidische Staat,’’ Hermes 70 (1935) 177–96. Harris, Edward, ‘‘When Is a Sale Not a Sale? The Riddle of Athenian Terminology for Real Security Revisited,’’ ClQ 38 (1988) 351–81. Hatzopoulos, M. B., Actes de Vente de la Chalcidique Centrale. Meletemata 6 (Athens, 1988).

Bibliography

351

Hatzopoulos, Donation

Hatzopoulos, M. B., Une Donation du Roi Lysimaque. Meletemata 5 (Athens, 1988). Hatzopoulos, Macedonian Institutions Hatzopoulos, M. B., Macedonian Institutions under the Kings. Meletemata 22 (Athens, 1996). Hellman, Vocabulaire de l’architecture grecque Hellmann, Marie-Christine, Recherches sur le vocabulaire de l’architecture grecque, d’après les inscriptions de Délos. BÉFAR 278 (1992). Hennig, ‘‘Kaufverträge’’ Hennig, Dieter, ‘‘Kaufverträge über Häuser und Ländereien aus der Chalkidike und Amphipolis,’’ Chiron 21 (1987) 143–69. Herrmann, ‘‘Teos und Abdera’’ Herrmann, Peter, ‘‘Teos und Abdera im 5. Jahrhundert v. Chr.,’’ Chiron 11 (1981) 1–30. Hietala, Intrasite Spatial Analysis Hietala, Harold J., ed., Intrasite Spatial Analysis in Archaeology. New Directions in Archaeology (Cambridge, Cambridge University Press, 1984). Higgins, Tanagra and the Figurines Higgins, Renyold, Tanagra and the Figurines. (Princeton, Princeton University Press, 1986). Hill, ‘‘Boundary Stones’’ Hill, D. K., ‘‘Some Boundary Stones from the Piraeus,’’ AJA 36 (1932) 254–59. Himera II Allegro, N., et al., eds., Himera II: Campagne di scavo 1966–1973 (Rome, ‘‘L’Erma’’ di Bretschneider, 1976). Hodkinson, ‘‘Mantineia’’ Hodkinson, Stephen, and Hilary Hodkinson, ‘‘Mantineia and the Mantinike: Settlement and Society in a Greek Polis,’’ BSA 76 (1981) 239–96. Hoepfner and Schwandner, Haus und Stadt Hoepfner, W., and E.-L. Schwandner, Haus und Stadt im klassischen Griechenland. 2d ed. Wohnen in der klassischen Polis 1 (Munich, Deutscher Kunstverlag, 1994). Hoffman, Warp-Weighted Loom Hoffmann, Marta, The Warp-Weighted Loom; Studies in the History and Technology of an Ancient Implement. Studia Norvegica 14 (Oslo, Universitetsforlaget, 1964). Hopper, Trade and Industry Hopper, R. J., Trade and Industry in Classical Greece. Aspects of Greek and Roman Life (London, Thames and Hudson, 1979). Horne, ‘‘The Household in Space’’ Horne, Lee, ‘‘The Household in Space: Dispersed Holdings in an Iranian Village,’’ American Behavioral Scientist 25 (1982) 677–85. Howland, Greek Lamps Howland, R. H., Greek Lamps and Their Survivals. The Athenian Agora 4 (Princeton, American School of Classical Studies, 1958).

352

Bibliography

Iozzo, ‘‘Corinthian Basins’’ Isager and Skydsgaard, Ancient Greek Agriculture Jameson, ‘‘Agricultural Labor’’

Jameson, ‘‘Agriculture and Slavery’’ Jameson, ‘‘Domestic Space’’

Jameson, ‘‘Private Space’’

Jameson, ‘‘Southern Argolid’’

Jameson et al., Greek Countryside

Jones et al., ‘‘Crop Storage at Assiros’’

Jones et al., ‘‘Vari House’’

Jones, ‘‘Kretan City’’

Jones, Public Organization

Kahrstedt, ‘‘Chalcidic Studies’’

Iozzo, Mario, ‘‘Corinthian Basins on High Stands,’’ Hesperia 56 (1987) 355–416. Isager, Signe, and Jens Erik Skydsgaard, Ancient Greek Agriculture. (London, Routledge, 1992). Jameson, Michael, ‘‘Agricultural Labor in Ancient Greece,’’ in B. Wells, ed. Agriculture in Ancient Greece. SkrAth 42 (Stockholm, 1992) 135–46. Jameson, Michael, ‘‘Agriculture and Slavery in Classical Athens,’’ CJ 73 (1977–78) 122–41. Jameson, Michael, ‘‘Domestic Space in the Greek City-State,’’ in S. Kent, ed., Domestic Architecture and the Use of Space. New Directions in Archaeology (Cambridge, Cambridge University Press, 1990) 92–113. Jameson, Michael, ‘‘Private Space and the Greek City,’’ in O. Murray and S. Price, eds., The Greek City from Homer to Alexander (Oxford, Oxford University Press, 1990) 171–95. Jameson, Michael. ‘‘The Southern Argolid: The Setting for Historical and Cultural Studies,’’ in M. Dimen and E. Friedl, eds., Regional Variation in Modern Greece and Cyprus. Annals of the New York Academy of Sciences 268 (1976) 74–91. Jameson, Michael H., Curtis N. Runnels, and Tjeerd H. van Andel, A Greek Countryside: The Southern Argolid from Prehistory to the Present Day (Stanford, Stanford University Press, 1994). Jones, Glynis, Kenneth Wardle, Paul Halstead, and Diana Wardle, ‘‘Crop Storage at Assiros,’’ Scientific American 254 (1986) 96–103. Jones, J. E., A. J. Graham, and L. H. Sackett, ‘‘An Attic Country House Below the Cave of Pan at Vari,’’ BSA 68 (1973) 355–452. Jones, Nicholas F., ‘‘The Organization of the Kretan City in Plato’s Laws,’’ CW 83 (1990) 473–92. Jones, Nicholas F., Public Organization in Ancient Greece: A Documentary Study: Memoirs of the American Philosophical Society 176 (Philadelphia, American Philosophical Society, 1987). Kahrstedt, Ulrich, ‘‘Chalcidic Studies,’’ AJP 57 (1936) 416–44.

Bibliography

353

Kanowski, Containers of Classical Greece

Kent, ‘‘Activity Areas and Architecture’’

Kent, Analyzing Activity Areas

Kent, Domestic Architecture and the Use of Space

Kintigh and Ammerman, ‘‘Heuristic Approaches’’ Kirigin, ‘‘Greeks in Central Dalmatia’’

Klaffenbach, ‘‘Korkyra Melaina’’

Knoepfler, ‘‘Calendar of Olynthus’’

Kondis, ‘‘Εὔτομος’’

Kondis, ‘‘Thurioi’’ Korfmann, ‘‘The Sling as a Weapon’’ Kraay, Archaic and Classical Greek Coins

354

Bibliography

Kanowski, M. G., Containers of Classical Greece (St. Lucia, University of Queensland Press, 1983). Kent, Susan, ‘‘Activity Areas and Architecture: An Interdisciplinary View of the Relationship between Use of Space and Domestic Built Environments,’’ in S. Kent, ed., Domestic Architecture and the Use of Space New Directions in Archaeology (Cambridge, Cambridge University Press, 1990) 1–8. Kent, Susan, Analyzing Activity Areas: An Ethnoarchaeological Study of the Use of Space (Albuquerque, University of New Mexico Press, 1984). Kent, Susan, ed., Domestic Architecture and the Use of Space: An Interdisciplinary Cross-Cultural Study. New Directions in Archaeology (Cambridge, Cambridge University Press, 1990). Kintigh, Keith W., and Albert J. Ammerman, ‘‘Heuristic Approaches to Spatial Analysis in Archaeology,’’ AmerAnt 47 (1982) 31–63. Kirigin, Branko, ‘‘The Greeks in Central Dalmatia: Some New Evidence,’’ in J.-P. Descoeudres, ed., Greek Colonists and Native Populations (Oxford, Oxford University Press, 1990) 291–319. Klaffenbach, G., ‘‘Zur Siedlungsinschrift von Korkyra Melaina.’’ Studia in honorem Acad. D. Dečev (=Izsledvaniia v. chest na akad. Dimitur Dechev po sluchai 80-godishninata) (Sofia, Bulgarska akademiia na nankite, 1958) 219– 20. Knoepfler, Denis, ‘‘The Calendar of Olynthus and the Origin of the Chalcidians in Thrace,’’ in J.-P. Descoeudres, ed., Greek Colonists and Native Populations (Oxford, Oxford University Press, 1990) 99–115. Kondis, I. D., ‘‘ Ἡ εὔτομος διάθεσις εἰς τὸν Ιπποδάμειου τρόπου,’’ ArchEph (1953–54) 255–67. Kondis, I. D., ‘‘ Ἡ Διαίρεσις τῶν Θουρίων,’’ ArchEph (1956) 106–13, 216–17. Korfmann, Manfred, ‘‘The Sling as a Weapon,’’ Scientific American 229 (1973) 34–42. Kraay, Colin, Archaic and Classical Greek

Kramer, Village Ethnoarchaeology

Krause, ‘‘Grundformen’’

Kraynak, Hostelries

Kroll, Agora 26

Kuhrt, ‘‘Old Assyrian Merchants’’

LaMotta and Schiffer, ‘‘Formation Processes’’

Larsen, Greek Federal States

Lévêque and Vidal-Naquet, Clisthène

Lewis, ‘‘Public Property’’

Lohmann, ‘‘Agriculture and Country Life’’

Lohmann, Atene

Löhr, ‘‘Hof, Fenster, Türen’’

Coins (Berkeley, University of California Press, 1976). Kramer, Carol, Village Ethnoarchaeology: Rural Iran in Archaeological Perspective. Studies in Archaeology (New York, Academic Press, 1982). Krause, Clemens, ‘‘Grundformen des griechischen Pastashauses,’’ AA (1977) 164– 79. Kraynak, Lynn. ‘‘Hostelries of Ancient Greece.’’ (PhD diss., University of California– Berkeley, 1984). Kroll, John H., The Greek Coins. The Athenian Agora 26 (Princeton, American School of Classical Studies, 1993). Kuhrt, Amelie. ‘‘The Old Assyrian Merchants,’’ in H. Parkins and C. Smith, eds., Trade, Traders and the Ancient City (London, Routledge, 1998) 16–30. LaMotta, Vincent M., and Michael Schiffer, ‘‘Formation Processes of House Floor Assemblages,’’ in P. Allison, ed., The Archaeology of Household Activities (London, Routledge, 1999) 19–29. Larsen, J. A. O., Greek Federal States, Their Institutions and History (Oxford, Clarendon Press, 1968). Vidal-Naquet, P., and P. Lévêque, Clisthène l’Athénien: Essai sur la représentation de l’espace et du temps dans la pensée politique grecque de la fin du VIe siècle à la mort de Platon. Annales Littéraires de l’Université de Besançon 65 (Paris, 1964). Lewis, David, ‘‘Public Property in the City,’’ in O. Murray and S. Price, eds., The Greek City from Homer to Alexander (Oxford, Oxford University Press, 1990) 245–63. Lohmann, Hans, ‘‘Agriculture and Country Life in Classical Attica,’’ in B. Wells, ed., Agriculture in Ancient Greece. SkrAth, 42 (Stockholm, 1992) 29–60. Lohmann, Hans, Atene. Forschungen zu Siedlungs- und Wirtschaftsstruktur des klassischen Attika (Cologne, Böhlau Verlag, 1993). Löhr, C., ‘‘Griechische Häuser: Hof, Fenster, Türen nach 348 v. Chr,’’ in W.-D. Heilmeyer and W. Hoepfner, eds., Licht und Architektur.

Bibliography

355

Lombardo, ‘‘Di Sibari a Thurii’’

Lombardo, ‘‘Lo Psephisma di Lumbarda’’

Loomis, Wages

Lynch, ‘‘Communal Drinking’’

Maier, Mauerbauinschriften

Malkin, Religion and Colonization

Martin, L’agora

Martin, L’Urbanisme Martin, Sovereignty and Coinage

Mattingly, ‘‘Maximum Figures’’

Mattingly, ‘‘Megalithic Madness’’

356

Bibliography

Schriften des Seminars für klassische Archäolgie der freien Universität Berlin (Tübingen, Ernst Wasmuth Verlag, 1990) 10–19. Lombardo, Mario, ‘‘Da Sibari a Thurii,’’ in A. Stazio and S. Ceccoli, eds., Sibari e la sibaritide: Atti del trentaduesimo Convegno di studi sulla Magna Grecia (Taranto, Istituto per la storia e l’archeologia della Magna Grecia, 1993) 255–328. Lombardo, Mario, ‘‘Lo Psephisma di Lumbarda: Note critiche e questioni esegetiche.’’ Hesperìa, 3. Studi sulla grecità di occidente a cura di Lorenzo Braccesi. Monografie 3 (Rome, Università di Venezia, Dipartimento di Antichità e Tradizione Classica, sezione storico-archeologica, 1993) 161–88. Loomis, William T., Wages, Welfare Costs, and Inflation in Classical Athens (Ann Arbor, University of Michigan Press, 1998). Lynch, Kathleen, ‘‘Black-Figure, Red-Figure, and Communal Drinking in a Late Archaic Athenian Home,’’ AJA 103 (1999) 298. Maier, F. G., Griechische Mauerbauinschriften. Vestigia 1–2 (Heidelberg, Quelle and Meyer, 1959–61). Malkin, Irad, Religion and Colonization in Ancient Greece. Studies in Greek and Roman Religion 3 (Leiden, Brill, 1987). Martin, Roland, Recherches sur l’agora grecque. Études d’Histoire et d’Architecture Urbaines. BÉFAR 174 (Paris, 1951). Martin, Roland, L’Urbanisme dans la Grèce antique (Paris, A. and J. Picard, 1974). Martin, Thomas R., Sovereignty and Coinage in Classical Greece (Princeton, Princeton University Press, 1985). Mattingly, David J., ‘‘Maximum Figures and Maximizing Strategies of Oil Production? Further Thoughts on the Processing Capacity of Roman Olive Presses,’’ in M.-C. Amouretti and J.-P. Brun, eds., La production du vin et de l’huile en Méditerranée (Oil and Wine Production in the Mediterranean Area), BCH Suppl. 26 (Paris, de Boccard, 1993) 483–98. Mattingly, David J., ‘‘Megalithic Madness and

Mattingly, ‘‘Technical Specifications’’

Mattusch, Greek Bronze Statuary

McCredie, ‘‘Hippodamos’’

McDonald, Political Meeting Places

McDonald, ‘‘Villa or Pandokeion?’’

McPhee, ‘‘Red-Figure Vase-Painters’’

Meiggs, Athenian Empire Meikle, ‘‘Aristotle and the Political Economy’’ Meikle, Aristotle’s Economic Thought Meikle, ‘‘Modernism, Economics and the Ancient Economy’’ Meritt, ‘‘Restoration in IG I, 37’’ Mertens, ‘‘Greek Architecture in the West’’

Métraux, Land-Use

Measurement. Or How Many Olives Could an Olive Press Press?’’ OJA 7 (1988) 177–95. Mattingly, David J., and R. Bruce Hitchner, ‘‘Technical Specifications for Some North African Olive Presses of Roman Date,’’ in M.C. Amouretti and J.-P. Brun, eds., La production du vin et de l’huile en Méditerranée (Oil and Wine Production in the Mediterranean Area.) BCH Suppl. 26 (Paris, Diffusion de Boccard, 1993) 439–62. Mattusch, C., Greek Bronze Statuary from the Beginnings through the Fifth Century B.C. (Ithaca, Cornell University Press, 1988). McCredie, J. R., ‘‘Hippodamos of Miletos,’’ in D. Mitten, J. G. Pedley, and J. A. Scott, eds., Studies Presented to G. M. A. Hanfmann. Harvard University Monographs in Art and Archaeology 2 (Mainz, Philip von Zabern, 1971) 95–100. McDonald, William A., Political Meeting Places of the Greeks. Johns Hopkins University Studies in Archaeology 34 (Baltimore, Johns Hopkins University Press, 1943). McDonald, William A., ‘‘Villa or Pandokeion?’’ in G. Mylonas, ed., Studies Presented to D. M. Robinson I (St. Louis, Washington University, 1951) 365–73. McPhee, Ian, ‘‘Some Red-Figure VasePainters of the Chalcidice,’’ BSA 76 (1981) 297–308. Meiggs, Russell, The Athenian Empire (Oxford, Clarendon Press, 1972). Meikle, Scott, ‘‘Aristotle and the Political Economy of the Polis,’’ JHS 79 (1979) 57–73. Meikle, Scott, Aristotle’s Economic Thought (Oxford, Clarendon Press, 1995). Meikle, Scott, ‘‘Modernism, Economics and the Ancient Economy,’’ PCPS 41 (1996) 174–91. Meritt, B. D., ‘‘A Restoration in IG I, 37,’’ AJA 29 (1925) 26–8. Mertens, Dieter, ‘‘Greek Architecture in the West,’’ in G. P. Carratelli, ed., The Greek World: Art and Civilization in Magna Graecia and Sicily. (New York, Rizzoli, 1995) 315–52. Métraux, Guy P. R., Western Greek LandUse and City-Planning in the Archaic Period.

Bibliography

357

Millet, ‘‘Attic horai’’

Millett, Lending and Borrowing

Millett, ‘‘Sale, Credit and Exchange’’

Minnesota Messenia Expedition

ML

Moggi, ‘‘Lo stato dei Calcidesi’’

Moggi, ‘‘Sane o Gale’’ Moggi, Sinecismi

Molleson, ‘‘Abu Hureyra’’

Morgan, ‘‘Euphiletos’ House’’ Mørkholm, Grierson, and Westermark, Early Hellenistic Coinage

Morris, ‘‘Athenian Economy’’

Morris, ‘‘Remaining Invisible’’

358

Bibliography

Outstanding Dissertations in the Fine Arts (New York, Garland, 1978). Millet, Paul, ‘‘The Attic horai Reconsidered in the Light of Recent Discoveries,’’ OPUS 1 (1982) 219–49. Millet, Paul, Lending and Borrowing in Ancient Athens (Cambridge, Cambridge University Press, 1991). Millett, Paul, ‘‘Sale, Credit and Exchange in Athenian Law and Society,’’ in P. Cartledge, P. Millett, and S. Todd, eds., Nomos: Essays in Athenian Law, Politics and Society (Cambridge, Cambridge University Press, 1990) 167–94. McDonald, William A., and George R. Rapp, Jr., eds., The Minnesota Messenia Expedition: Reconstructing a Bronze Age Environment (Minneapolis, University of Minnesota, 1972). Meiggs, R., and D. Lewis, eds., A Selection of Greek Historical Inscriptions (Oxford, Clarendon Press, 1988). Moggi, M., ‘‘Lo stato dei Calcidesi alla luce del sinecismo di Olinto,’’ Critica Storica 11 (1974) 1–11. Moggi, M., ‘‘Sane o Gale (Thuc. 5, 18, 6),’’ Ann Pisa 3d series, no. 4 (1973) 71–107. Moggi, M., I Sinecismi Interstatali Greci. Relazioni Interstatali nel Mondo Antico 2 (Pisa Marlin, 1976). Molleson, Theya, ‘‘The Eloquent Bones of Abu Hureyra,’’ Scientific American 271 (1994) 70–75. Morgan, Gareth, ‘‘Euphiletos’ House: Lysias I,’’ TAPA 112 (1982) 115–23. Mørkholm, Otto, Philip Grierson, and Ulla Westermark, Early Hellenistic Coinage from the Accession of Alexander to the Peace of Apamea (336– 188 bc) (Cambridge, Cambridge University Press, 1991). Morris, Ian, ‘‘The Athenian Economy Twenty Years After The Ancient Economy,’’ CP 89 (1994) 351–66. Morris, Ian, ‘‘Remaining Invisible: The Archaeology of the Excluded in Classical Athens,’’ in S. R. Joshel and S. Murnaghan,

Morrow, Plato’s Cretan City Murray, Sympotica

Mylonas, ‘‘Olynthian House’’ Nevett, ‘‘Gender Relations’’

Nevett, House and Society

Nevett, ‘‘Real Estate ‘Market’ ’’

Nevett, ‘‘Separation or Seclusion?’’

Newman, Politics Nixon and Price, ‘‘Size and Resources’’

Nylander, ‘‘Zernaki Tepe’’

Oakley & Sinos Wedding

Olynthus 1

eds., Women and Slaves in Greco-Roman Culture: Differential Equations (London, Routledge, 1998) 193–220. Morrow, G. R., Plato’s Cretan City (Princeton, Princeton University Press, 1960). Murray, Oswyn, ed. Sympotica: A Symposium on the Symposion (Oxford, Clarendon Press, 1990). Mylonas, G. E., ‘‘The Olynthian House of the Classical Period,’’ CJ 35 (1940) 389–402. Nevett, Lisa, ‘‘Gender Relations in the Classical Greek Household: The Archaeological Evidence,’’ BSA 90 (1995) 364–81. Nevett, Lisa, House and Society in the Ancient Greek World. New Studies in Archaeology (Cambridge, Cambridge University Press, 1999). Nevett, Lisa, ‘‘A Real Estate ‘Market’ in Classical Greece? The Example of Town Housing,’’ BSA 95 (2000) 329–43. Nevett, Lisa, ‘‘Separation or Seclusion? Towards an Archaeological Approach to Investigating Women in the Greek Household in the Fifth to Third Centuries BC,’’ in M. P. Pearson and C. Richards, eds., Architecture and Order: Approaches to Social Space (London, Routledge, 1994) 98–112. Newman, W. L., The Politics of Aristotle (New York, Arno, 1973). Nixon, Lucia, and Simon Price, ‘‘The Size and Resources of Greek Cities,’’ in O. Murray and S. Price, eds., The Greek City from Homer to Alexander (Oxford, Oxford University Press, 1990) 137–70. Nylander, C., ‘‘Remarks on the Urartian Acropolis at Zernaki Tepe,’’ Orientalia Suecana 15 (1966) 141–54. Oakley, John, and Rebecca H. Sinos, The Wedding in Ancient Athens. Wisconsin Studies in Classics (Madison, University of Wisconsin Press, 1993). Mylonas, George E., Excavations at Olynthus, Part 1. The Neolithic Settlement. Johns Hopkins University Studies in Archaeology 6 (Baltimore, Johns Hopkins University Press, 1929).

Bibliography

359

Olynthus 2

Olynthus 3

Olynthus 4

Olynthus 5

Olynthus 6

Olynthus 7

Olynthus 8

Olynthus 9

Olynthus 10

360

Bibliography

Robinson, David M., Excavations at Olynthus, Part 2. Architecture and Sculpture: Houses and Other Buildings. Johns Hopkins University Studies in Archaeology 9 (Baltimore, Johns Hopkins University Press, 1930). Robinson, David M., Excavations at Olynthus, Part 3. The Coins Found at Olynthus in 1928. Johns Hopkins University Studies in Archaeology 11 (Baltimore, Johns Hopkins University Press, 1931). Robinson, David M., Excavations at Olynthus, Part 4. The Terracottas of Olynthus Found in 1928. Johns Hopkins University Studies in Archaeology 12 (Baltimore, Johns Hopkins University Press, 1931). Robinson, David M., Excavations at Olynthus, Part 5. Mosaics, Vases and Lamps of Olynthus Found in 1928 and 1931. Johns Hopkins University Studies in Archaeology 18 (Baltimore, Johns Hopkins University Press, 1933). Robinson, David M., Excavations at Olynthus, Part 6. The Coins Found at Olynthus in 1931. Johns Hopkins University Studies in Archaeology 19 (Baltimore, Johns Hopkins University Press, 1933). Robinson, David M., Excavations at Olynthus, Part 7. The Terracottas of Olynthus Found in 1931. Johns Hopkins University Studies in Archaeology 20 (Baltimore, Johns Hopkins University Press, 1933). Robinson, David M., and J. Walter Graham, Excavations at Olynthus, Part 8. The Hellenic House. A Study of the Houses Found at Olynthus with a Detailed Account of Those Excavated in 1931 and 1934. Johns Hopkins University Studies in Archaeology 25 (Baltimore, Johns Hopkins University Press, 1938). Robinson, David M., and Paul Augustus Clement, Excavations at Olynthus, Part 9. The Chalcidic Mint and the Excavation Coins Found in 1928–1934. Johns Hopkins University Studies in Archaeology 26 (Baltimore, Johns Hopkins University Press, 1938). Robinson, David M., Excavations at Olynthus, Part 10. Metal and Minor Miscellaneous Finds, an

Olynthus 11

Olynthus 12

Olynthus 13

Olynthus 14

Orlandos and Travlos, Lexicon

Osborne, ‘‘Buildings and Residence’’

Osborne, Classical Landscape

Osborne, ‘‘Exchange and Society’’

Osborne, ‘‘ ‘Is It a Farm?’ ’’

Original Contribution to Greek Life. Johns Hopkins University Studies in Archaeology 31 (Baltimore, Johns Hopkins University Press, 1941). Robinson, David M., Excavations at Olynthus, Part 11. Necrolynthia, a Study in Greek Burial Customs and Anthropology. Johns Hopkins University Studies in Archaeology 32 (Baltimore, Johns Hopkins University Press, 1942). Robinson, David M., Excavations at Olynthus, Part 12. Domestic and Public Architecture. Johns Hopkins University Studies in Archaeology 36 (Baltimore, Johns Hopkins University Press, 1946). Robinson, David M., Excavations at Olynthus, Part 13. Vases Found in 1934 and 1938. Johns Hopkins University Studies in Archaeology 38 (Baltimore, Johns Hopkins University Press, 1950). Robinson, David M., Excavations at Olynthus, Part 14. Terracottas, Lamps and Coins Found in 1934 and 1938. Johns Hopkins University Studies in Archaeology 39 (Baltimore, Johns Hopkins University Press, 1952). Orlandos, Anastasios K., and Ioannos N. Travlos, Λεξικον Αρχαιων Αρχιτεκτονικων Ορων (Athens, 1986). Osborne, Robin, ‘‘Buildings and Residence on the Land in Classical and Hellenistic Greece: The Contribution of Epigraphy,’’ BSA 80 (1985) 119–28. Osborne, Robin, Classical Landscape with Figures: The Ancient Greek City and Its Countryside (London, Philips, 1987). Osborne, Robin, ‘‘Pride and Prejudice, Sense and Subsistence: Exchange and Society in the Greek City,’’ in J. Rich and A. WallaceHadrill, eds., City and Country in the Ancient World. Leicester-Nottingham Studies in Ancient Society 2 (London, Routledge, 1991) 119–45. Osborne, Robin, ‘‘ ‘Is It a Farm?’ The Definition of Agricultural Sites and Settlements in Ancient Greece,’’ in B. Wells, ed., Agriculture in Ancient Greece. SkrAth 42 (Stockholm, 1992) 21–25.

Bibliography

361

Osborne, ‘‘Survey and Greek Society’’ Parkins, ‘‘Time for Change?’’

Parkins and Smith, Trade, Traders and the Ancient City Pečírka, ‘‘Country Estates’’

Piérart, Platon

Pomeroy, Goddesses, Whores, Wives and Slaves Pomeroy, Xenophon Oeconomicus

Pounds, ‘‘Urbanization’’

Price, ‘‘Early Greek Bronze Coinage’’

Pritchett, Greek State at War

Pritchett and Amyx, ‘‘Attic Stelai’’ Purcell, ‘‘Mobility and the Polis’’

Raeder, Priene: Funde

362

Bibliography

Osborne, Robin, ‘‘Survey and Greek Society,’’ AJA 100 (1996) 165–69. Parkins, Helen, ‘‘Time for Change? Shaping the Future of the Ancient Economy,’’ in H. Parkins and C. Smith, eds., Trade, Traders and the Ancient City (London, Routledge, 1998) 1–15. Parkins, Helen, and Christopher Smith, eds., Trade, Traders and the Ancient City (London, Routledge, 1998). Pecírka, Jan, ‘‘Country Estates of the Polis of Chersonesos,’’ in L. d. Rosa, ed., Richerche storiche ed economiche in memoria di Corrado Barbagallo (Naples, Edizioni schentifiche Italiane, 1970) 459–77. Piérart, Marcel, Platon et la cité grecque. Théorie et réalité dans la Constitution des ‘‘Lois.’’ Académie Royale de Belgique, Mémoires de la Classe des Lettres Collection in−8°, 2d series, T. 62, fascicule 3 (Brussels, 1974). Pomeroy, Sarah B., Goddesses, Whores, Wives and Slaves (New York, Schocken Books, 1975). Pomeroy, Sarah B., Xenophon Oeconomicus: A Social and Historical Commentary (Oxford, Clarendon Press, 1994). Pounds, Norman J. G., ‘‘The Urbanization of the Classical World,’’ Annals of the Association of American Geographers 59:1 (1969) 135–57. Price, M., ‘‘Early Greek Bronze Coinage,’’ in C. M. Kraay and G. K. Jenkins, eds., Essays in Greek Coinage Presented to Stanley Robinson (Oxford, Oxford University Press, 1968) 90–104. Pritchett, W. K., The Greek State at War (Berkeley, University of California Press, 1971). Pritchett, W. K., and D. A. Amyx, ‘‘The Attic Stelai,’’ Hesperia 22, 25, 27 (1953–58). Purcell, Nicholas, ‘‘Mobility and the Polis,’’ in O. Murray and S. Price, eds., The Greek City from Homer to Alexander (Oxford, Oxford University Press, 1990) 29–58. Raeder, Joachim, Priene. Funde aus einer griechischen Stadt in Berliner Antikenmuseum.

Rapoport, History and Precedent

Rapoport, House Form and Culture

Rapoport, Mutual Interaction

Reber, ‘‘Aedificia graecorum’’

Reber, ‘‘Andrones von Eretria’’

Reber et al., Eretria 10

Reinders, New Halos Rickman, Roman Granaries

Robinson, ‘‘Fourth Campaign’’ Robinson, ‘‘Inscriptions, 1928’’ Robinson, ‘‘Inscriptions, 1931’’

Robinson, ‘‘Inscriptions, 1934’’ Robinson, ‘‘Inscriptions, 1938’’ Robinson, ‘‘Mosaics’’ Robinson, ‘‘Preliminary Report’’

Robinson, ‘‘Residential Districts’’

Bilderheft der Staatlichen Museen Preufsischer Kulturbesitz 45/46 (Berlin, Gebr. Mann 1984). Rapoport, Amos, History and Precedent in Environmental Design (New York, Plenum Press, 1990). Rapoport, Amos, House Form and Culture. Foundations in Cultural Geography Series (Englewood Cliffs, N.J., Prentice-Hall, 1969). Rapoport, Amos, ed., The Mutual Interaction of People and their Built Environment: A Cross-Cultural Perspective (The Hague, Mouton, 1976). Reber, Karl, ‘‘Aedificia graecorum. Zu Vitruvs Beschreibung des griechischen Hauses,’’ AA (1988) 653–66. Reber, Karl, ‘‘Zur architekonischen Gestaltung der Andrones in den Häusern von Eretria,’’ AntK 32 (1989) 3–7. Reber, Karl, Monica Brunner, Effy Kassapoglou, Ingrid R. Metzger, Philippe Mottet, and Esther Schönenberger, Die klassischen und hellenistischen Wohnhäuser im Westquartier. Eretria. Fouilles et recherches 10 (Lausanne, Editions Payot, 1998). Reinders, H. R., New Halos: A Hellenistic Town in Thessaly (Utrecht, H&S, 1988). Rickman, Geoffrey, Roman Granaries and Store Buildings (Cambridge, Cambridge University Press, 1971). Robinson, D. M., ‘‘The Fourth Campaign at Olynthos,’’ AJA 43 (1939) 48–77. Robinson, D. M., ‘‘A Deed of Sale at Olynthus’’ TAPA 59 (1928) 225–32. Robinson, D. M., ‘‘New Inscriptions from Olynthus and Environs,’’ TAPA 62 (1931) 40–56. Robinson, D. M., ‘‘Inscriptions from Olynthus, 1934,’’ TAPA 65 (1934) 103–37. Robinson, D. M., ‘‘Inscriptions from Macedonia, 1938,’’ TAPA 69 (1938) 43–61. Robinson, D. M., ‘‘Mosaics from Olynthos,’’ AJA 36 (1932) 16–24. Robinson, D. M., ‘‘A Preliminary Report on the Excavations at Olynthos,’’ AJA 33 (1929) 53–76. Robinson, D. M., ‘‘The Residential Districts

Bibliography

363

Robinson, ‘‘Third Campaign’’ Robinson, ‘‘Villa of Good Fortune’’ Robinson, ‘‘Coinage of Athens’’

Rose, ‘‘Coins and History’’ Rose, ‘‘Reconsideration’’ Rotroff, Agora 29

Rotroff, ‘‘Athenian Hellenistic Pottery’’

Rotroff, ‘‘Olynthos and Other Deposits’’ Rotroff and Oakley, Debris from a Public Dining Place

Runnels, Millstones from the Argolid

Runnels and Murray, ‘‘Milling in Ancient Greece’’ Runnels et al., Artifact and Assemblage

Ruschenbusch, ‘‘Bevölkerung’’

Ruschenbusch, ‘‘Tribut und Bürgerzahl’’

364

Bibliography

and the Cemeteries at Olynthos,’’ AJA 36 (1932) 118–38. Robinson, D. M., ‘‘The Third Campaign at Olynthos,’’ AJA 39 (1935) 210–47. Robinson, D. M., ‘‘The Villa of Good Fortune at Olynthus,’’ AJA 38 (1934) 501–10. Robinson, E. S. G., ‘‘Some Problems in the Later Fifth Century Coinage of Athens,’’ ANSMN 9 (1960) 1–15. Rose, Mark, ‘‘Coins and the History of Olynthus,’’ AJA 94 (1990) 315. Rose, Mark, ‘‘A Reconsideration of Coins from Olynthus,’’ AJA 88 (1984) 258. Rotroff, Susan I., Hellenistic Pottery: Athenian and Imported Wheelmade Table Ware and Related Material. The Athenian Agora 29 (Princeton, American School of Classical Studies, 1997). Rotroff, Susan I., ‘‘Athenian Hellenistic Pottery: Toward a Firmer Chronology.’’ Akten des XIII International Kongresses für klassische Archäologie, Berlin 1988. Mainz, 1990) 174–78. Rotroff, Susan I., ‘‘Olynthos and Other Deposits—the Pottery,’’ AJA 94 (1990) 315–16. Rotroff, Susan I., and John H. Oakley, Debris from a Public Dining Place in the Athenian Agora. Hesperia Supplement 25 (Princeton, American School of Classical Studies, 1992). Runnels, Curtis, ‘‘A Diachronic Study and Economic Analysis of Millstones from the Argolid, Greece’’ (PhD diss., Indiana University, 1981). Runnels, Curtis N., and P. M. Murray, ‘‘Milling in Ancient Greece,’’ Archaeology 36 (1983) 62–63. Runnels, Curtis, Daniel Pullen, and Susan Langdon, eds., Artifact and Assemblage: The Finds from a Regional Survey of the Southern Argolid, Greece (Stanford, Stanford University Press, 1995). Ruschenbusch, Eberhard, ‘‘Die Bevölkerung Griechenlands im 5. und 4. Jh. v. Chr.,’’ ZPE 56 (1984) 55–57. Ruschenbusch, Eberhard, ‘‘Tribut und Bürgerzahl im ersten Athenischen Seebund,’’ ZPE 53 (1983) 125–43.

Ruschenbusch, ‘‘Zahl der Staaten’’

Russell, Late Ancient and Medieval Population

Rutter, ‘‘Foundation of Thurii’’ Sahin, Inschriften von Stratonikeia

Sakellariou, Polis-State Sallares, Ecology of the Ancient Greek World Salzmann, Kieselmosaiken Samuel, ‘‘Their Staff of Life’’

Saprykin, Ancient Farms and Land-Plots

Ščeglov, Polis et Chora

Schiffer, Formation Processes

Schuller, Hoepfner, and Schwandner, Demokratie und Architektur

Sevin, ‘‘Van/Zernaki Tepe’’

Ruschenbusch, Eberhard, ‘‘Die Zahl der griechischen Staaten und Arealgröße und Bürgerzahl der ‘Normalpolis,’’’ ZPE 59 (1985) 253–63. Russell, J. C., Late Ancient and Medieval Population. TAPA 48:3 (Philadelphia, American Philosophical Society, 1958). Rutter, N. K., ‘‘Diodorus and the Foundation of Thurii,’’ Historia 22 (1973) 155–76. Sahin, M. Çetin, Die Inschriften von Stratonikeia II, 1, 1 Inschriften griechischer Städte aus Kleinasien 22,1 (Bonn, Habelt, 1982). Sakellariou, M. B., The Polis-State: Definition and Origin. Meletemata 4 (Athens, 1989). Sallares, Robert, The Ecology of the Ancient Greek World (London, Duckworth, 1990). Salzmann, Dieter, Studien zu den antiken Kieselmosaiken. AF 10 (Berlin, Mann, 1982). Samuel, Delwen, ‘‘Their Staff of Life: Initial Investigations on Ancient Egyptian Bread Baking,’’ in B. J. Kemp, ed., Amarna Reports 5. Egypt Exploration Society, Occasional Papers (London, Egypt Exploration Society 1989) 253–90. Saprykin, S. J., Ancient Farms and Land-Plots on the Khora of Khersonesos Taurike: Research in the Herakleian Peninsula, 1974–1990. McGill University Monographs in Classical Archaeology and History 16 (Amsterdam, J. C. Gieben, 1995). Ščeglov, A. N., Polis et Chora: Cité et territoire dans le Pont-Euxin. Espaces et paysages 1. Centre de Recherches d’Histoire Ancienne 118. Annales Littéraires de l’Université de Besançon, 2d ser., 476. (Paris, 1992). Schiffer, Michael, Formation Processes in the Archaeological Record (Albuquerque, University of New Mexico Press, 1996). Schuller, Wolfgang, Wolfram Hoepfner, and Ernst Ludwig Schwandner, eds., Demokratie und Architektur. Der hippodamische Städtebau und die Entstehung der Demokratie. Wohnen in der klassischen Polis 2 (Munich, Deutscher Kunstverlag, 1989). Sevin, Veli, ‘‘Van/Zernaki Tepe: On the Urar-

Bibliography

365

Sparkes, ‘‘Greek Kitchen’’ Sparkes, ‘‘Kitchen Addenda’’ Sparkes, ‘‘Not Cooking, but Baking’’ Sparkes, ‘‘Treading the Grapes’’ Sparkes and Talcott, Agora 12

Stanley, ‘‘Value of Ergasteria’’

Stone, ‘‘Texts, Architecture and Ethnographic Analogy’’

Stroud, Athenian Grain-Tax Law

Stroud, ‘‘Silver Coinage’’ Svoronos-Hadjimichalis, ‘‘Évacuation de la fumée’’ Szanto, Griechische Phylen

Tod, ‘‘Acrophonic Numerals’’ Tod, GHI Touratsoglou, ‘‘Dated Gold’’

Trendall, Lucania, Campania and Sicily

366

Bibliography

tian Grid Plan Once Again,’’ Anatolica 23 (1997) 173–76. Sparkes, B. A., ‘‘The Greek Kitchen,’’ JHS 82 (1962) 121–37. Sparkes, B. A., ‘‘The Greek Kitchen: Addenda,’’ JHS 85 (1965) 162–63. Sparkes, B. A., ‘‘Not Cooking, but Baking,’’ GaR 28 (1981) 172–78. Sparkes, B. A., ‘‘Treading the Grapes,’’ BABesch 51 (1976) 47–64. Sparkes, B. A., and L. Talcott, Black and Plain Pottery. The Athenian Agora 12 (Princeton, American School of Classical Studies, 1970). Stanley, P. V., ‘‘The Value of Ergasteria in Attica: A Reexamination,’’ MünstBeitr 9 (1990) 1–13. Stone, Elizabeth C., ‘‘Texts, Architecture and Ethnographic Analogy: Patterns of Residence in Old Babylonian Nippur,’’ Iraq 43 (1981) 19–33. Stroud, Ronald S., The Athenian Grain-Tax Law of 374/3 BC. Hesperia Supplement 29 (Princeton, American School of Classical Studies, 1998). Stroud, Ronald S., ‘‘An Athenian Law on Silver Coinage,’’ Hesperia 43 (1974) 157–88. Svoronos-Hadjimichalis, Vanna, ‘‘L’évacuation de la fumée dans la maisons grecques des V e et IV e siècles,’’ BCH 80 (1956) 483–506. Szanto, E., Die griechische Phylen. Sitzungsberichte der kaiserliche Akademie der Wissenschaften in Wien, Philosophische-historische Klasse 144, V, Abhandlung (Vienna, 1901). Tod, Marcus N., ‘‘The Greek Acrophonic Numerals,’’ BSA 37 (1936–37) 236–58. Tod, Marcus N., A Selection of Greek Historical Inscriptions. (Oxford, Clarendon Press, 1962). Touratsoglou, Iannis, ‘‘Dated Gold: The Evidence from Hellenistic Macedonia,’’ in D. Williams, ed., The Art of the Greek Goldsmith (London, British Museum Press, 1998) 30– 38. Trendall, A. D., The Red-Figured Vases of Lucania, Campania and Sicily (Oxford, Clarendon Press, 1967).

Tsakirgis, ‘‘Prostas House’’ Uguzzoni and Ghinatti, Eraclea

Uhlenbrock, Coroplast’s Art

Vallet, ‘‘Avenues, quartiers et tribus à Thourioi’’

Vatin, ‘‘Le Bronze Pappadakis’’ Vickers and Gill, Artful Crafts

von Eickstedt, Piräus

von Gerkan, Städteanlagen

Walker, ‘‘Women and Housing’’

Wallace-Hadrill, Houses and Society

Walter-Karydi, Greek House

Weiß, ‘‘Schleuderbleie und Marktgewichte’’

Tsakirgis, Barbara, ‘‘The Universality of the Prostas House,’’ AJA 93 (1989) 278–79. Uguzzoni, A., and F. Ghinatti, La tavole greche di Eraclea. Università degli studi di Padova, Pubblicazioni dell’Instituto di storia antica 7 (Rome, ‘‘L’Erma’’ di Bretschneider, 1968). Uhlenbrock, Jaimee Pugliese, and Rebecca Miller Ammerman, The Coroplast’s Art: Greek Terracottas of the Hellenistic World (New Rochelle, N.Y., The College at New Paltz, State University of New York, 1990). Vallet, G., ‘‘Avenues, quartiers et tribus à Thourioi, ou comment compter les cases d’un damier (à propos de Diod. XII, 10 et 11),’’ in L’Italie préromaine et la Rome républicaine. Mélanges offerts à Jacques Huergon. CÉFR 27. (Rome, École Française de Rome, 1976), vol. 2, 1021–32. Vatin, Claude, ‘‘Le Bronze Pappadakis, étude d’une loi coloniale,’’ BCH 87 (1963) 1–19. Vickers, Michael, and David Gill, Artful Crafts: Ancient Greek Silverware and Pottery (Oxford, Clarendon Press, 1994). Eickstedt, K.-V. von, Beiträge zur Topographie des antiken Piräus. βιβλιοθήκη τῆς ἐν Ἀθήνιας Ἀρχαιολογικῆς Ἑταιρείας 118 (Athens, 1991). Gerkan, A. von, Griechische Städteanlagen. Untersuchungen zur Entwicklung des Städtebaues im Altertum (Berlin, Walter de Gruyter, 1924). Walker, Susan, ‘‘Women and Housing in Classical Greece,’’ in A. Cameron and A. Kuhrt, eds., Images of Women in Classical Antiquity (Beckenham, Kent, Croom Helm, 1983) 81–91. Wallace-Hadrill, Andrew, Houses and Society in Pompeii and Herculaneum (Princeton, Princeton University Press, 1994). Walter-Karydi, Elena, The Greek House: The Rise of Noble Houses in Late Classical Times. The Archaeological Society of Athens Library 171 (Athens, Archaeological Society at Athens, 1998). Weiß, Peter, ‘‘Milet 1994-5: Schleuderbleie und Marktgewichte,’’ AA (1997) 143–56.

Bibliography

367

Wells, ed., Agriculture West, History of the Chalcidic League

West, ‘‘Thucydides V,18,5’’

Westermark, ‘‘Coinage’’

Westgate, ‘‘Greek Mosaics’’

Whallon, ‘‘Spatial Distributions’’

Wiegand and Schrader, Priene

Wilhelm, Beiträge

Wilhelm, ‘‘Korkyra Melaina’’

Wilk, ‘‘Built Environment and Consumer Decisions’’

Williams, ‘‘Early Urbanization of Corinth’’

368

Bibliography

Wells, Berit, ed., Agriculture in Ancient Greece. SkrAth 40, 42 (Stockholm, 1992). West, A. B., The History of the Chalcidic League. Bulletin of the University of Wisconsin 969, History Series 4 no. 2 (Madison, University of Wisconsin Press, 1919). West, A. B., ‘‘Thucydides V,18,5. I. Where Was Skolos (Stolos)? II. Sane or Gale?’’ AJP 58 (1937) 157–73. Westermark, Ulla, ‘‘The Coinage of the Chalcidic League Reconsidered,’’ in A. Damsgaard-Madsen, E. Christiansen, and E. Hallager, eds., Studies in Ancient History and Numismatics Presented to Rudi Thomsen (Aarhus, Aarhus University Press, 1988). Westgate, Ruth, ‘‘Greek Mosaics in Their Architectural and Social Context,’’ BICS 42 (1998) 93–115. Whallon, Robert, ‘‘Unconstrained Clustering for the Analysis of Spatial Distributions in Archaeology,’’ in H. J. Hietala, ed., Intrasite Spatial Analysis in Archaeology. New Directions in Archaeology (Cambridge, Cambridge University Press, 1984) 242–77. Wiegand, Theodor, and Hans Schrader, Priene: Ergebnisse der Ausgrabungen und Untersuchungen in den Jahren 1895–1898 (Berlin, G. Reimer, 1904). Wilhelm, Adolf, Beiträge zur griechische Inschriftenkunde. Sonderschriften des Österreichschen Archäologischen Instituts in Wien 7 (Vienna, 1909). Wilhelm, A., ‘‘Die Landlose der Ansiedler aus Issa auf Korkyra Melaina. Neue Beiträge zur griechischen Inschriftenkunde, III. Teil,’’ Sitzungsberichte der kaiserliche Akademie der Wissenschaften in Wien, Philosophische-historische Klasse 175 (1913) 3–18. Wilk, Richard R., ‘‘The Built Environment and Consumer Decisions,’’ in S. Kent, ed., Domestic Architecture and the Use of Space. New Directions in Archaeology (Cambridge, Cambridge University Press, 1990) 34–42. Williams, C. K., II, ‘‘The Early Urbanization of Corinth,’’ ASAtene 60 (1982) 9–19.

Williams, ‘‘Refiguring Attic Red-Figure’’ Williams, ‘‘Stymphalos 1983’’

Williams, ‘‘Stymphalos 1984’’ Wood, ‘‘Agricultural Slavery’’ Wycherley, ‘‘Hippodamus’’ Wycherley, How the Greeks Built Cities Wycherley, ‘‘Olynthus and Selinus’’ Young, ‘‘South Attica’’ Youni, ‘‘Inscriptions olynthiennes’’

Williams, Dyfri, ‘‘Refiguring Attic RedFigure: A Review Article,’’ RA (1996) 227–52. Williams, E. H., and Thomas Boyd, ‘‘Investigations at Mytilene and Stymphalos, 1983,’’ EchCl 3 (1984) 169–86. Williams, E. H., ‘‘Investigations at Stymphalos, 1984,’’ EchCl 4 (1985) 215–24. Wood, Ellen Meiksins, ‘‘Agricultural Slavery in Classical Athens,’’ AJAH 8 (1983) 1–47. Wycherley, R. E., ‘‘Hippodamus and Rhodes,’’ Historia 13 (1964) 135–39. Wycherley, R. E., How the Greeks Built Cities. (London, Macmillan, 1967). Wycherley, R. E., ‘‘Notes on Olynthus and Selinus,’’ AJA 55 (1951) 231–36. Young, John Howard, ‘‘Studies in South Attica,’’ Hesperia 10 (1941) 163–91. Youni, Maria S., ‘‘À propos de quatre inscriptions olynthiennes: Quelques remarques sur la ‘sûreté réelle’ au 4èmes s. av. J.-C.,’’ TEKMERIA. ΣΥΜΒΟΛΕΣ ΣΤΗΝ

ΙΣΤΟΡΙΑ ΤΟΥ ΕΛΛΗΝΙΚΟΥ ΚΑΙ ΡΩΜΑΙΚΟΥ ΚΟΣΜΟΥ 2 (1996) 135–53.

Zahrnt, Olynth

Zahrnt, M., Olynth und die Chalkidier (Munich, Beck, 1971).

Bibliography

369

Illustration Credits

Figure 13. The Metropolitan Museum of Art, lent by Jan Mitchell, 1988 (L.1988.81.5). Photograph courtesy of the Metropolitan Museum of Art. Figure 19. Copyright © 1950 The Johns Hopkins Press. Reproduced with permission of the Johns Hopkins University Press. Figure 21. Copyright © 1946 The Johns Hopkins Press. Reproduced with permission of the Johns Hopkins University Press. Figure 32. Copyright © 1946 The Johns Hopkins Press. Reproduced with permission of the Johns Hopkins University Press. Figure 34. Copyright © 1946, 1930 The Johns Hopkins Press. Reproduced with permission of the Johns Hopkins University Press. Figure 36. The State Hermitage Museum, St. Petersburg. Figure 40. Copyright © Bildarchiv Preussischer Kulturbesitz, Berlin, 2001. Photo: Jutto Tietz-Glagow. Figure 41. Copyright © 1950 The Johns Hopkins Press. Reproduced with permission of the Johns Hopkins University Press. Figure 42. Copyright © 1941 The Johns Hopkins Press. Reproduced with permission of the Johns Hopkins University Press. Figure 43. Copyright © 1941 The Johns Hopkins Press. Reproduced with permission of the Johns Hopkins University Press. Figure 45. Copyright © The Swiss School of Archaeology in Greece. Figure 52. Copyright © 1930 The Johns Hopkins Press. Reproduced with permission of the Johns Hopkins University Press.

370

General Index

Abdera, 75, 200, 310 Acanthus, 36, 42, 309; coins of, 39 acropolis: in Aristotle’s Politics, 13; in Plato’s Laws, 9 Aeneas Tacticus, 13 Aeschines, 274 Agios Mamas, 34 agora, 32, 84, 108, 211–12, 236, 265–66; free vs. commercial, in Aristotle’s Politics, 18 agriculture, in Aristotle’s Politics, 13. See also countryside (chora); economy: agricultural; trade and industry: agricultural processing agrimensores, 219 Aletes, 2 Alexander III, coins of, 50, 53, 56 allotments, 9–10, 19, 220; inalienable, 8, 220 altar, 71, 87, 128, 130, 138, 143, 248, 252. See also Index of Artifacts: altar, portable Amarna, 198 Ammotopos, 317 Amphipolis, 42 Amyntas III, 42, 270 andron, 80, 85, 93, 97, 101, 107, 116, 130, 135, 140, 149, 152, 175, 180–93, 201, 204–5, 208–12, 225, 244–45, 248, 250, 252–53, 256, 264, 317 Angel, L., 61 anoikismos, 24, 35–44 anteroom (to andron), 80, 85, 93, 101, 107–8, 116, 180, 186, 210, 256

apartment house (synoikia), 150 Aphrodite, 140 Aphytis, 36 Apollonia, 42, 46, 311 Argilus, 36 Argolid, 329 Aristophanes, 2, 151, 161, 191, 193 Aristotle, 3, 12–18, 216, 224, 236, 265, 283, 287 Artabazus, 29, 34–35 Asklepios, 187 Assera, 36–37, 308 Atene, 331 Athena, 9, 11, 110 Athens, 52, 236, 278, 286, 308; coins of, 39; treaty with, 43, 46 Attic Stelai, 70, 164, 166–67, 169, 229, 278, 320, 341 bakery, 167 baking, 248, 261 barter, 224 bathroom, 90, 102, 106, 111, 138, 154, 158–59, 175, 205, 208, 244, 250, 333 bathtub, 49, 80, 102, 106, 116, 126, 140, 154, 159, 250 bedroom, 93, 96 Bellerophon, 210 blacksmith, 150 bones, 155, 258

371

booty, 46, 70, 187–88 Bottiaean League, 44, 310 Bottiaeans, 34–35, 42–44 Brasidas, 37, 43

Demosthenes, 41, 44–46, 49–50, 311 Dengate, J., 51–53, 56–57, 60 destruction: burning in, 48, 95, 97, 106, 127, 143, 145; Persian, 32, 34; by Philip II, 25, 39, 45–49, 67, 85 Dicaea, 36 dining couches (klinai), 80, 93 Diodorus, 19, 45, 50, 307 Dionysus, 11, 132, 140 domestic assemblages; database of, 65; effects of erosion on, 68; statistical analyses of, 70; in Xenophon’s Economics, 149. See also Index of Artifacts door, 149, 158, 191; double, 98, 113, 115, 118, 125, 131, 239, 241; prothyron, 85, 104, 109 drain, 78, 99, 104, 109–10, 113, 140, 240 drainage alley, 27, 40, 94, 118, 122, 204

Carystus, 313 Cassander, 49; coins of, 50, 53 Cassandreia (Potidaia), 49, 51–52, 59, 312 cemeteries, 33 Chalcidic League, 24, 35, 41–44 Chersonesos, 225, 329, 331 Chios, 239 cistern, 34, 94, 243, 318 citizenship, 42, 44, 305, 340 Civic Center (on South Hill), 27, 32–34 Cleisthenes, 2, 4, 11–12 Clement, P., 39, 42–43, 51, 56, 61, 260, 266, 270 clubhouse, 211 coins, distribution of, 266–73 colonization, 1, 5, 19, 219, 303; in Plato’s Laws, 6 Colophon, 75, 200, 317, 327 composite communities, 9, 19, 34, 38 contexts, primary vs. secondary, 64, 66–70 cooking, 89, 101–3, 106, 111, 116, 121, 127, 136, 144, 153–69, 191, 248–49. See also kitchencomplex; trade and industry: cooking and baking Corinth, 238, 286 countryside (chora): in Aristotle’s Politics, 18, 285; buildings in, 225, 285; of Olynthus, 23, 37–38, 48–49, 221, 225–26, 238, 263, 282; in Plato’s Laws, 6, 9, 285; population of, 308, 329; of Thurii, 20 courtyard, 75, 78–79, 87, 99, 104, 109, 113, 118, 122, 125, 128, 133, 138, 142, 160, 165, 167, 175, 178, 193, 208, 228, 242 craftsmen, 10, 264, 286; in Aristotle’s Politics, 12–13; in Hippodamus’s polis, 3. See also trade and industry Croton, 19 Cybele, 190, 253 Cyme, 313 Cyrene, foundation decree of, 285

farmers: in Aristotle’s Politics, 12–13; in Hippodamus’s polis, 3 Ferguson, W. S., 51–52 flue, 80, 89, 95, 102, 106, 111, 119, 138–39, 154–56, 159, 167, 175, 208, 244, 248 food preparation, 48, 71, 119, 121, 124, 126– 27, 135, 144, 146, 149, 156, 159–60, 163–69, 191–92, 238, 244 fortification, 27, 29–30, 37, 39–40, 131; in Aristotle’s Politics, 13, 18; gate on Avenue A, 29; in Homer, 1; in Plato’s Laws, 9, 11; at Thurii, 20; of Villa Section, 45 furniture, 70, 89, 93, 95–96, 101, 106, 109, 112, 119, 124, 127, 135, 139, 144–45, 246. See also Index of Artifacts

Daly, L., 85 Delos, 152, 192 Delphi, 188

Gale, 36–38, 41, 307–8 gardens, 30, 102, 236 gendered division of space, 82, 93, 97, 148–53.

372

General Index

East Spur Hill (ESH), 23, 29, 56, 58, 199, 311 economy: agricultural, 225–48, 285; embedded vs. disembedded, 223; hidden, 273; market, 223–24, 230, 280, 283 Elaeus, coins of, 57, 313 Eleusis, 235 Eretria, 130, 146, 192, 200, 317, 323 Eros, 253 Euphiletos, 153 exedra, 94, 115, 160, 168, 176–77, 213, 253, 258

See also andron; kitchen-complex; women’s quarters ( gynaikonitis) Gill, D., 187 Gordion, 173–74, 324 Goritsa, 17 Graham, J. W., 30, 32, 40–41, 51, 53, 61–63, 65, 76, 79–82, 89, 91, 96, 103, 110, 112, 138– 39, 141, 154, 157, 171, 202, 227, 232, 235, 247, 265, 273 granaries, 34, 229, 235. See also storage: food grinding. See food preparation

Kellion, 278 kiln, 109, 253 kitchen-complex, 80–81, 89, 102, 106, 111, 115, 119, 123, 126, 138–40, 152–61, 162, 167, 191, 200, 204–5, 208–10, 212, 233, 241, 244, 257; second, 91, 110, 157–60, 166, 177 Knossos, 6 Korkyra Melaina, foundation decree from, 218–21, 285 Koroni, 51 Kroll, J., 273

Halieis, 64, 238, 323, 333 Hatzopoulos, M. B., 277 hearth, 80, 89, 126, 138–39, 154, 156, 158, 160, 193, 320 Hephaestia, 313 Heraea, 313 Herakleia, 11, 253 Hermes, 111 Herodotus, 19, 34, 161 Hestia, 9 Himera, 197, 303, 327 Hippodamus of Miletus, 3–5, 12, 15, 19 Histiaea, coins of, 57, 313 Hoepfner, W., and Schwandner, E. L., 194, 265 hostel ( pandokeion), 256 household cult, 33, 71, 88–89, 92, 99, 101, 111, 120, 141, 144, 146–47, 149, 158, 168, 246, 256, 336 household size, 40–41, 226 house prices, 276–81. See also inscriptions: sales and loan house types, prostas vs. pastas, 75, 200

Lake Bolbe, 35, 37–38, 285 Lamia, coins of, 270 land distribution, 7–8, 38, 216–21; in Homer, 1; at Korkyra Melaina, 219–21; at Olynthus, 30, 285; in Plato’s Laws, 8–11; redistribution, 6, 37, 216, 218, 220; at Thurii, 20 land division, 3, 9; in Aristotle’s Politics, 14–18; in Hippodamus’s polis, 3–5; in Plato’s Laws, 8–11; at Thurii, 21 Larisa, coins of, 270 Lebedos, 203 Lecythus (Chalcidic fortress), 43 Lemnos, 313 lighting, 75, 77–79. See also courtyard, light well, windows light well, 90–93, 100–101, 110, 126, 139, 155, 178 Locri Epizephyrii, 253, 286 looting, 48, 187, 190 Lysias, 19, 151, 180, 193

Iasus, 313 inscriptions: funerary, 50; land grant from Cassandreia, 49; on pithoi, 232; sales and loan, 48, 116–18, 128, 133, 136, 180, 225, 238, 259, 276–81; on slingbullets, 46; treaty with Philip II, 32 Ischomachos, 93, 148–49, 152, 169, 180 Issa, 220. See also Korkyra Melaina Itanos, 218 Jameson, M., 152, 191 Kahun, 198 Kassope, 318

Mantinea, 304, 310, 330 Martin, R., 265 masonry, 32, 54, 97–98, 109, 191, 327–28 McDonald, W. A., 61, 97, 102, 230, 256 Mecyberna, 36–38, 41, 46, 61, 307–8 Megara, coins of, 56–57, 313 Mende, 35–37, 41–42, 308 men’s quarters (andronitis). See andron Messene, 203 Messenia, 309 Metapontum, 225, 329–330 Miletus, 239 Monte Iato, 317 mosaic, 91, 94, 100, 137–38, 140, 157, 180, 204–5, 210, 213–14, 256 Mygdonia, 35

General Index

373

Mylonas, G., 34, 61, 80, 82, 85, 91, 142, 147, 156 nemesis (of Hippodamus), 4–5 Nevett, L., 71–72, 152, 191 New Halos, 327 North Hill, 23, 27, 62, 199, 209, 224, 233, 250, 285, 287; layout, 35, 38 Northwest Quarter, 50, 53–60 occupations. See trade and industry oikist, 11, 19, 220, 303 olives and olive oil, 226–28, 239. See also trade and industry: agricultural processing; Index of Artifacts: olive crusher Ostia, 229 Painter of Olynthos 5.156, 58 Painter of Salonica 38.290, 59 Pallene, 36, 49 Pan, 140 pastas, 75–77, 79, 88, 94, 99, 105, 109, 115, 119, 123, 126, 128, 133, 143, 160, 165, 167, 178, 193, 208, 239, 244, 251, 253; semienclosed, 88, 109; used for weaving, 175 Pausanias, 203 Peloponnesian War, 24, 34–35, 38, 41, 43–45 Perdiccas (king in Herodotus), 161 Perdiccas II, 35, 37–39, 285; coins of, 39, 43, 57, 309 Pericles, 283 peristyle, 79, 138, 142 Persian Wars, 34, 235, 270 Phaleas of Chalcedon, 2, 7 Phegetus, 36 Pheidon of Argos, 303 Philip II, 24, 45–46; coins of, 53, 57, 270; treaty with the Chalcidians, 32, 43 Philippi, coins of, 56–57, 313 Phygela, 313 pillar-partition, 81–82, 89–90, 94, 103, 106, 111, 119, 126, 139, 154–55, 158 Pilorus, 36–37, 308 Piraeus, 3–4, 15–16, 130, 192 Placia, 313 Plato, 2–3, 5–12, 236 Plutarch, 283 Polichnai, 340

374

General Index

Polygyros, 23 population: in Aristotle’s Politics, 12; in Hippodamus’s polis, 3; of Olynthus, 37–44; in Plato’s Laws, 8. See also household size portico, 133, 138, 142, 243 ports: in Aristotle’s Politics, 13; in Plato’s Laws, 6 Potidaia, 35–37, 41, 49, 308, 312. See also Cassandreia (Potidaia) press, oil or wine, 239, 242, 244–46, 263 Priene, 75, 140, 142, 152, 197, 200, 274, 317, 319, 327 proportional inequality, in Plato’s Laws, 8, 11 public buildings, 32, 54, 266 Rapoport, A., 198 remodeling of houses, 74, 82, 84, 118, 126, 195–98, 210–11, 214 reoccupation after 348 b.c., 27, 49–61, 270 Rhodes, 3, 16, 52 Robinson, D. M., 27, 29–30, 32–34, 40– 43, 50–51, 53, 61, 65, 68, 73, 87, 89, 91, 128, 138, 141–42, 147, 171, 202, 235, 244, 246–47, 255, 257, 260, 265, 277 roof, common, 199–200, 202, 204 rooftiles, 62, 91, 95 Rose, M., 56 Rotroff, S., 57–58, 60 sanctuaries: of [Artemis], 32; on North Hill? 265 Sane, 36, 46, 308 Sardis, 324 Scabala, 36 Scapsa, 36 Scione, 35–37, 41, 308 seasonal use of houses, 71, 78, 150, 157, 160–61, 167, 193 second story, 82, 94–96, 103, 109, 113, 118, 125, 131, 159, 171, 245, 278, 280, 319; assemblages from, 67; as women’s quarters, 153 Selinus, 76, 317 Sermylia, 35–37, 308 shelves, 95, 100, 105, 109–11, 126–27, 135, 138, 144 shops, 79, 81, 97, 107, 112, 116, 120, 127, 131, 135, 141, 160, 164, 201, 208, 211–12,

226, 246, 249, 273–76; coins from, 269; temporary, on streets, 275–76 Silenus, 109, 127, 140, 253 Singus, 36–38, 41, 307–8 Sithonia, 36 skeleton, animal, 99, 276 Skoinia, 280 slaves, 70, 149, 151, 244, 261–64, 320 Smyrna, 13 South Hill, 23, 27, 34, 54, 186, 232, 250, 270 Sparkes, B. A., 50, 58 Sparta, 11, 42–44, 226, 250 Spartolus, 36, 43 stables, 247 stairs, 67, 94, 103, 109, 113, 118, 125, 131 stasis: in Aristotle’s Politics, 7, 14, 19; in Plato’s Laws, 6–7 stoa, 32, 265 Stolos, 36, 277, 320, 340 storage: communal, 235, 248, 282, 333; food, 48, 94, 101, 106, 124, 126, 130, 132, 135, 156, 169, 226–36, 281. See also Index of artifacts: pithos storeroom, 94, 102, 117–18, 135, 140, 145, 225, 229–33, 244, 246–47, 278, 332 stratigraphy, 60, 62–67, 95–96, 255 Stratonikeia, 13 streets, 48, 250, 266, 274–75; in Aristotle’s Politics, 14; as boundaries of urban nemeses, 4; at Olynthus, 27; at Thurii, 21; of Villa Section, 30, 45; wide and narrow ( plateiai and stenopoi), 5, 16, 21 Strepsa, 36 Stymphalus, 18 Sybaris, 19 symposium, 48, 180, 186, 201; furniture, 180 synoikism, 2, 19, 203, 303, 310–11; in Plato’s Laws, 6 Talcott, L., 50 temples: in Aristotle’s Politics, 18; in Homer, 1. See also sanctuaries Tenos, 278 Teos, 200, 203, 310 textiles, 70, 93, 96. See also weaving; Index of Artifacts: loomweight Thales (of Miletus), 239 theater, 32

Thebes, 236 Theophrastus, 306, 330, 332 Thucydides, 35–36, 41, 43–44 Thucydides (son of Melesias), 19 Thurii, 3, 16, 19–22, 285 timber, 23, 42 Torone, 34–36, 43, 46 trade and industry, 224, 236–76, 281; agricultural processing, 48, 115, 118, 164, 167, 226, 234–35, 238–48, 261; in Aristotle’s Politics, 12–13; carpentry, 150; cooking and baking, 248–50; coroplasty, 253; fishing, 250; fulling or dyeing, 150, 258; manufacture, 253–56; masonry and sculpture, 131, 252–53; minting, 259–61; in Plato’s Laws, 6; pottery, 254; textile manufacture, 250–52 Trapezous, 49 Travlos, J., 61 tribes, 9–10, 13, 22, 220 tribute, 35–37; and population, 37 Troy, 326 Type House, 82–84, 194–214 Vickers, M., 187 Villa Section, 18, 23, 29–31, 40, 45, 62, 199, 209, 213–14, 224, 233, 236, 250, 270, 280, 285, 287. See also individual houses in Index of Houses and Buildings, Blocks, Trenches, and Streets vineyards, 16, 277 Vitruvius, 151, 156, 317 Walker, S., 152 wall painting, 80, 85, 91, 93, 99, 107, 109, 133, 137, 140, 142, 144–46, 154, 157, 180, 204–5, 210, 213–14, 230, 318 water supply, 33; in Aristotle’s Politics, 14 wealth, 286; in Plato’s Laws, 8 weapons, 149. See also trade and industry: manufacture; Index of Artifacts: arrowhead, shield, slingbullet, spearhead, sword weaving, 71, 91, 93, 102, 106–7, 109–10, 112, 115, 119, 121, 123, 126, 130, 132–34, 141, 149, 156–60, 166, 169–79, 186, 192, 250– 42, 261. See also textiles; Index of Artifacts: loomweight Weinberg, G. and S., 61 windows, 76, 78, 110–11, 119, 130, 134, 139, 166, 178, 192, 317

General Index

375

wine, 149, 226–28 women’s quarters ( gynaikonitis), 82, 93, 97, 148–53, 191–93, 264. See also andron; kitchen-complex Wycherley, R. E., 265

376

General Index

Xenophon, 41, 44, 75, 148, 169, 180, 193, 203, 236, 261, 265, 287, 317 Zernaki Tepe, 198 Zeus, 9

Index of Houses and Buildings, Blocks, Trenches, and Streets

houses and buildings A −1, 176–77, 183, 209, 258 A −5, 210 A −i 9, 183, 186 A 1, 209, 240, 324, 336 A 2, 209, 318 A 3, 79, 175–76, 209, 234, 240, 332 A 4, 209, 225, 240, 246, 248, 323, 332 A 5, 128, 131, 187, 209, 252–53 A 6, 164, 166, 197, 211–13, 234, 240, 241–44, 248, 256, 274, 287, 324, 332 A 7, 197, 208, 211, 213, 242–43, 248, 287 A 8, 124–28, 130, 154, 157, 166–67, 175–76, 182–83, 205, 209, 213, 228, 234, 268, 321, 324, 328 A 9, 125, 128, 178, 205, 209, 213, 240, 253, 268, 272, 321, 333–34 A 10, 125, 128–31, 164, 166, 176, 179, 192, 205, 209, 213, 234, 252–53, 268, 272, 324–25, 328, 332 A 11, 39, 84, 128, 131–33, 169, 176–77, 191, 197, 205, 210, 212, 214, 234, 258, 268, 270, 272, 278, 280, 321, 325, 332, 339 A 12, 128, 131–33, 176, 191, 197, 205, 210, 214, 268, 272 A 13, 84, 128, 131–33, 183, 191, 197, 205, 210, 212, 214, 268, 272, 321 A iii 9, 32–33, 54, 265–66, 269, 313 A iv 1, 208, 268

A iv 3, 208, 268, 278, 280, 323 A iv 5, 182–83, 197, 209–10, 234, 259, 268, 287, 325, 332 A iv 5/7, 259 A iv 7, 197, 210, 214, 234, 259–60, 268, 278, 280, 324, 328, 337, 339 A iv 9, 71, 78, 82, 108–13, 115–16, 119–20, 127, 132–33, 135, 141, 146, 156–58, 168, 173, 176–79, 211–12, 214, 234, 251–52, 254, 260, 268–69, 272, 274, 318, 325, 332, 339 A iv 10, 32, 54, 265–66, 269, 313 A v 1, 121–24, 169, 202, 209, 234, 258, 325, 332 A v 2, 176–77, 183, 186, 202, 209, 268, 272, 316 A v 3, 122, 183, 202, 209, 234, 316 A v 4, 176, 202–3, 209, 258 A v 5, 54, 176–77, 209, 272 A v 6, 82, 138, 176, 202, 205, 209–10, 213, 250, 256, 275, 287, 328, 334 A v 7, 202, 209, 268, 272, 327–28 A v 8, 197, 202, 205, 210–11, 213, 250, 256, 327–28 A v 9, 82, 118–21, 122, 166, 168, 173, 176–77, 179, 197, 202, 211–12, 214, 239–40, 247, 250, 252, 268, 272, 274, 312, 320, 323–25 A v 10, 82, 113–18, 119–21, 133, 136, 202, 211– 12, 214, 240, 252, 254, 263, 268, 272, 274, 277–78, 280, 312, 320, 323, 327–328, 332, 334–35, 339 A vi 1, 209–10

377

A vi 2, 157, 167, 175–76, 178, 209, 234, 276, 323, 328, 333 A vi 3, 178, 183, 186, 209–10, 258 A vi 4, 176, 178, 209, 276 A vi 5, 180, 192, 209, 258, 323 A vi 6, 187, 203, 209, 323 A vi 7, 157, 167, 209, 323, 325 A vi 8, 203, 209, 240, 244, 246–48, 270, 287, 323 A vi 9, 197, 211–12, 234, 270, 332–33 A vi 10, 167, 203, 209, 234, 240, 244–45, 248, 270, 274, 276, 287, 320, 323, 328, 332 A vii 1, 175–77, 211, 234, 328, 333 A vii 2, 209, 234, 320, 323, 325, 328 A vii 3, 176–77, 209, 240, 247, 309, 327–28 A vii 4, 81–82, 103–8, 110, 112, 124, 127, 156– 57, 167, 175–76, 178, 209, 228, 232, 234–35, 320, 328, 332 A vii 5, 209, 309 A vii 6, 157, 176–78, 209 A vii 7, 183, 209, 234 A vii 8, 209, 240, 247, 327 A vii 9, 54, 164, 176, 179, 211–12, 214, 240, 244, 246–48, 252, 268, 274, 335–36 A vii 10, 157, 166, 175–76, 209, 234, 324, 328, 339 A viii 1, 176–77, 179, 209, 324 A viii 2, 157, 209, 228, 234, 258, 319 A viii 3, 123, 197, 210, 257, 324, 339–40 A viii 4, 157, 166, 168, 209, 324, 334 A viii 5, 200, 209, 235, 316, 332 A viii 6, 209, 240, 247 A viii 7/9, 69, 174–76, 179, 209, 234, 250–52, 263, 287, 326, 328, 335 A viii 8, 167, 183, 209, 248, 324, 336 A viii 10, 175–77, 183, 210, 240, 244, 324, 328 A xi 9, 54, 209, 313 A xi 10, 49, 54, 59, 157, 174, 176–77, 179, 209, 235, 239–41, 248, 313, 323–24, 332, 334 A xii 9, 54, 209, 276, 313 A xii 10, 211, 313 A xiii 10, 209, 313 A' 7, 54, 313 A' 8, 54 A' 9, 54, 313 A' 10, 54, 313 A' 11, 54, 313 B i 3, 328 B i 5, 208, 253–55, 318, 324, 326

378

Index of Houses and Buildings

B ii 1, 234, 240, 244 B ii 2, 210 B ii 3, 168, 324 B ii 4, 210, 337 B ii 6, 259, 337 B iii 2, 278, 328, 337, 340 B iii 4, 210 B iii 6, 210 B v 1, 39–40, 209, 328 B vi 1, 209, 323, 328 B vi 2, 157, 167, 209, 328 B vi 3, 209, 328 B vi 4, 154, 157, 209, 235, 332 B vi 5, 166, 175–76, 209, 324, 328 B vi 6, 208 B vi 7, 183, 186, 209, 328, 336 B vi 8, 184, 208–9, 234 B vi 9, 59, 175–76, 211, 214, 228, 234, 320, 332, 339 B vi 10, 208, 255, 274, 328 B vii 1, 65, 209, 328 B vii 2, 157, 175–76, 209, 324, 328 B xi 1, 235, 332 C −x 5, 58, 211, 255, 314 C −x 7, 211, 328 C v 4, 275 D iii 4, 58 D v 2, 184 D v 4, 278 D v 6 (House of Zoilos), 133–36, 176, 211–12, 214, 234–35, 249, 274, 276, 278, 280, 332, 339 D vii 4, 39 ESH 1, 184, 211–12, 324, 332 ESH 2, 209 ESH 3, 209 ESH 4, 30, 45, 54, 158–60, 164, 166–67, 174, 176–77, 179, 209, 235, 249–50, 270, 274, 324, 328, 332 ESH 5, 208 ESH 6, 189–90, 199, 208 Fountainhouse (South Hill), 33 House of Many Colors, 30, 64, 67, 71, 78, 82, 85–97, 110–11, 113, 124, 132, 134, 139, 142, 146, 154, 156–58, 165, 168, 176–77, 179, 182, 184, 186, 209, 213, 227, 230, 232, 325, 332–33, 336 House of Pan (South Hill), 185 House of the Comedian, 29–30, 127, 137–42,

156–57, 174, 176, 184, 186–87, 208–9, 232, 256, 327, 332 House of the Tiled Prothyron, 33, 142–47, 164–65, 168, 209, 213, 230, 232, 249, 258, 324, 332–33, 336 House of the Twin Erotes, 142, 166, 168, 179, 208, 228, 230, 232, 234, 324, 333 House of the Wash Basin, 176–77, 208 South Villa, 30, 176, 209, 232, 256, 327, 332–33 Villa CC, 171, 173–74, 176, 179, 208, 214, 324 Villa of Good Fortune, 30, 82, 102, 124, 166, 209, 214, 226–27, 229–30, 232–33, 235, 256, 268, 270, 272, 280, 324, 327, 332–33, 340 Villa of the Bronzes, 29–30, 45, 71, 95, 97– 103, 110–11, 113, 139, 146, 155, 157–58, 168, 176, 178, 209, 213–14, 230, 232, 234, 268, 272, 325, 334, 336

trenches and se ctions Row A (dug as Trench 8), 11, 27, 29, 39–40, 118, 128, 131, 199, 203, 213, 252, 311 Row A', 11, 27, 29, 50, 53, 59–60, 311 Section F/G, 46, 232 Section J/K, 232 Section N, 29, 54, 185, 313 Section O, 211 Trench 3, 185 Trench 4, 58 Trench 5/6, 232 Trench 6, 185 Trench 7, 54, 56, 133, 210, 214, 253, 268, 313 Trench 8. See houses A 1–A 9; Row A Trench 10, 54, 185 Trench 13, 184

streets and avenues blo cks block A iv, 54 block A v, 82, 121, 204, 209, 213 block A vi, 82, 121, 178, 204, 209–10, 213, 270 block A vii, 40, 82, 155, 203–4, 209, 214 block A viii, 162, 203 blocks A xi, A xii, A xiii, 50, 59. See also General Index: Northwest Quarter block B vi, 203, 328

Avenue A, 27, 29, 33, 127, 131, 212, 277 Avenue B, 46, 108, 113, 116, 121, 211–12, 265– 66, 270, 274–76, 281, 284, 288 Avenue F, 30, 40, 102, 249 Avenue G, 30, 85 Street −iv, 98 Street v, 113, 275 Street vii, 39

Index of Houses and Buildings

379

Index of Artifacts

(For complete database see www.stoa.org/olynthus/) altar, portable, 85, 88–89, 99, 110, 128, 144, 146–47, 158, 168, 235, 246, 249, 252, 276 amphora: Panathenaic, 115; storage, 63, 66, 69–70, 89–90, 109, 111, 120, 123, 126, 132, 135–36, 139–41, 145, 156, 166, 168–69, 182, 229, 251, 253, 258, 260, 333; table, 90, 95 amphoriskos, 117, 120 architectural fragments, 66. See also capital arrowhead, 46, 89, 97, 101, 112, 115–16, 120, 123, 127 askos, 88, 90, 99, 112, 117, 126, 135, 144, 182 astragalos, 69, 111, 120 basin, 63, 100, 102, 112, 123, 127, 135–36, 138, 167, 169, 244, 246, 248, 253; stone, 101 bead, 102, 110, 115–16, 119, 126, 130, 140, 143, 250 bodkin, 123 bolsal, 110, 127, 135, 144, 182 boss, 93, 101, 105, 112, 120, 127, 144, 246 bowl, 59, 89, 104–5, 110–11, 120, 124, 126–27, 135, 138–41, 159, 182, 189–90, 275, 314 bracelet, 112, 120 brazier, 49, 100, 102, 143, 162, 241 capital, 98, 100, 128, 252–53 celt, 116 chisel, 126

coin: in Aristotle’s Politics, 13; Bottiaean, 34, 307, 337; Chalcidian, 43; distribution of, 266–73; forgery, 260; hoards, 38–39, 45, 49, 116, 132, 159, 250, 269–73, 287; late, 49, 51, 60. See also General Index: coins, distribution of cooking pot, 63, 90, 111, 124, 127, 135 cup, 90, 93, 95, 104, 106, 109, 111, 119, 139, 180, 182, 190; kantharos, 58–59, 140, 189; miniature, 92–93, 146 dikast’s ticket, 258 dish, lead, 146 doll, 120, 146 duck askos, 90, 95, 112 earring, 105–6, 110, 116, 120, 134, 138 epinetron, 91, 177 eyelash-plate, 123 fibula, 69, 90, 105–6, 116, 120, 246, 251 figurine, 33, 39, 59–60, 66, 85, 95, 106, 111, 115, 120, 123–24, 126, 132, 139, 146, 156, 159, 178, 187, 247, 253, 258, 260; Neolithic, 251 finger ring, 100–101, 116, 120, 126 fishhook, 111–12, 115, 138, 250, 260, 276, 335 fishplate, 89–91, 94, 99, 107, 120, 139

381

flan, 259 flower pot, 126 grater, 116, 127, 135 grindstone, 63, 66, 85, 90, 102, 112, 119–20, 124, 126, 130, 132, 136, 144–45, 157, 159, 163–66, 175, 241, 244, 246, 248–49, 263, 324, 333; rotary, 246 guttus, 92, 104–5, 109, 111–12, 117, 127, 135, 182, 314 hardware, 66, 88, 90, 102, 104, 107, 112, 116, 123, 135, 143, 145 hooks, 100–101 horse bit, 124 hydria, 58, 63, 91, 105, 135, 159 jewelry, 90, 105, 110, 112, 116, 120–21, 126, 250. See also individual pieces jug, 117, 127, 135, 141, 180, 182, 190, 309 juglet, 90, 95, 112, 120, 140–41, 182, 189 kantharos, 58, 105 kardopos, 71, 167 key, 123 knife, 97, 99, 117 krater, 63, 89, 94–95, 127, 139, 180–90 ladle, 120, 144, 180 lamp, 39, 59, 85, 90, 93, 101, 106, 110–12, 115– 16, 119, 126, 130, 138–39, 180, 189, 246, 258, 314 lamp stand, 104, 180, 189 lebes gamikos, 182 lekane, 169, 248 lekanis, 92, 112, 135, 139, 144 lekythos, 63, 89, 92, 104–7, 110, 112, 115–17, 120, 126, 139–40, 146, 174, 182, 189, 246, 260, 335 lid, 89–90, 92, 95, 112, 115, 123, 139–40, 335; lekanis, 59, 92, 109, 124; pithos, 91, 105, 132, 135, 140, 145, 247, 275; pyxis, 135–36, 187, 190 loomweight, 59, 63, 65–66, 69, 85, 88–89, 91, 93, 102, 104, 106–7, 109–11, 115, 117–19, 123–24, 126, 130, 132–36, 138, 141, 159, 166, 169–79, 241, 250–51, 253, 260, 263, 276, 316; shapes and weights, 179, 252. See also

382

Index of Artifacts

General Index: trade and industry: textile manufacture; General Index: weaving lopas, 59 louter, 71, 89, 99, 109–10, 112, 115, 119, 132, 135–36, 143–44, 146–47, 158–59, 164, 166, 168–69, 246, 252, 275, 309 meathook, 159 mirror, 112 mold: arrowhead, 255; figurine, 39, 60, 109– 10, 115, 127, 253–54; slingbullet, 255 mortar, 65, 70, 106, 119, 128, 156, 159, 166, 167, 245, 248–49 mortarium, 104, 112, 127, 135–36, 166–67 nail, 63, 101, 106 needle, 107, 111, 115, 127; netting, 130, 174, 276 oinochoe, 59, 111, 135, 144, 182, 190 olive crusher, 48, 115, 118, 239, 241, 330 olpe, 90, 93, 105, 123–24, 134–35, 141, 182 pelike, 89–90, 92, 106–7, 109, 127, 159, 182 pendant, 110, 119–20 perfume pot, 143 pestle, 110, 126 phiale, 120 phidakne, 127, 331. See also pithos pigment, 91, 120, 126, 134 pigment grinder, 91 pin, 101 pithos, 48–49, 63, 65, 70–71, 93–94, 101–2, 106, 109, 117–18, 122–24, 126, 135, 143, 167, 169, 239, 331; sizes and contents, 227. See also General Index: storage: food plaque, 111 plate, 59, 89–91, 94–95, 99, 101, 106, 112, 117, 120, 126, 140–41, 182, 253, 314, 335 pottery, coarse, 63, 66, 70, 112, 117, 127, 144–45, 229, 232, 248 probe, 144 protome, 69, 92–93, 104, 120, 132, 174, 187, 251 relief, 252 saltcellar, 93, 106, 111, 124, 127, 135, 182 saucer, 63, 89–90, 92, 95, 99, 101, 106–7,

110–12, 116–17, 120, 123, 127, 134–35, 138, 140–41, 156, 159, 182, 314 scales, 106–7, 120, 275–76 shield, 97, 99 sipye, 229 skyphos, 89, 92, 95, 111–12, 182, 246, 335 slingbullet, 46, 97, 99, 102, 110, 115–16, 146, 312 spatula, 101, 110, 126, 144, 253 spearhead, 93, 97, 99, 117, 120 spindle, 133 spindle whorl, 91, 110, 177 spit, 139 spit support, 139 spoon, 124 stamnos pyxis, 59 stamp, 119 statuette, 187 storage bin, 248 strainer, 138, 260

stylus, 116–17 sword, 97, 99 thymiaterion, 69, 112, 120, 182, 189, 251, 260 trough, 70, 248 tweezers, 119, 123 unguentarium, 135 urinal, 145 vase, 119; composite, 99, 258, 309; plastic, 59, 69, 109, 126, 132, 138, 187, 251 vessel: metal, 48, 70, 94, 105, 110, 112, 115, 117, 119, 123, 135, 187–90, 246, 260; miniature, 158; storage, 63, 65, 229 weaving tool, 116, 118 weight, 104–5, 107, 111–12, 115–17, 119–20, 130, 138–39, 141, 145, 246, 265, 275–76; net, 250

Index of Artifacts

383

Distribution of Gendered Spaces and Artifacts (N) Kitchen-Complex Andron & Anteroom Room Used for Weaving Lower Grindstone Upper Grindstone Saddle Quern Uncertain Grindstone Mortar Krater – Room Known House with Krater; Room Uncertain 0

plate 1. Distribution of Gendered Spaces and Artifacts on the North Hill

50 m

Distribution of Gendered Spaces and Artifacts (S) 0

50 m

plate 2. Distribution of Gendered Spaces and Artifacts on the East Spur Hill and Villa Section. Key as in Plate 1.

Distribution of House Clusters Cluster 1: regular, unspecialized Cluster 2: irregular plans Cluster 3: not domestic Cluster 4: many shops Cluster 5: regular, more specialized Incompletely preserved or exc.

? Incomplete, attributed w/o stats.

0

?

? ?

?

plate 3. Distribution of House Clusters

50

100 m

Trade and Industry Agricultural processing Weaving Cooking/baking Coroplasty Fishing Masonry/sculpting Minting/counterfeiting Weapons manufacture Uncertain

0

C -x 5

plate 4. Trade and Industry at Olynthus

50

100 m