Country in the City: Agricultural Functions of Protohistoric Urban Settlements 178969132X, 9781789691320

Citation preview

Country in the City Agricultural Functions in Protohistoric Urban Settlements (Aegean and Western Mediterranean)

edited by

Dominique Garcia, Raphaël Orgeolet, Maia Pomadère and Julien Zurbach

Country in the City Agricultural Functions in Protohistoric Urban Settlements (Aegean and Western Mediterranean)

edited by

Dominique Garcia, Raphaël Orgeolet, Maia Pomadère and Julien Zurbach

Archaeopress Archaeology

Archaeopress Publishing Ltd Summertown Pavilion 18-24 Middle Way Summertown Oxford OX2 7LG www.archaeopress.com

ISBN 978-1-78969-132-0 ISBN 978-1-78969-133-7 (e-Pdf)

© Authors and Archaeopress 2019 Cover image: Chr. W. Eckersberg (1783-1853), View of the Forum in Rome, 1814 This book has received the support of Translitteræ (Ecole universitaire de recherche, program ‘Investissements d’avenir’ ANR‐10‐IDEX‐0001‐02 PSL* and ANR‐17‐EURE‐0025)

All rights reserved. No part of this book may be reproduced, or transmitted, in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of the copyright owners. Printed in England by Severn, Gloucester This book is available direct from Archaeopress or from our website www.archaeopress.com

Contents Introduction���������������������������������������������������������������������������������������������������������������������������������������������������������������������������������1 Raphaël Orgeolet Elites and Farmers in Iberian Iron Age Cities (7th-2nd Centuries BC): Storage and Processing of Agricultural Products����������������������������������������������������������������������������������������������������������������������������������������������������������������6 Natàlia Alonso and Guillem Pérez-Jordà Are There Farmers in Lattara (Lattes, France) during the Iron Age?Plant Resources Acquirement and Management between the 5th and the 1st Centuries BC���������������������������������������������������������������������������������������22 Núria Rovira and Natàlia Alonso The Neglected ‘Fields’ of Proto-Urban Living: A View from Bronze Age Crete�������������������������������������������������������41 Kostis S. Christakis The Management of Agricultural Resources in the Minoan Town of Malia (Crete) from the Middle Bronze Age to the Early Late Bronze Age��������������������������������������������������������������������������������������������������������������������������51 Maria Emanuela Alberti, Sylvie Müller Celka and Maia Pomadère Akrotiri, Thera : Glimpses of the Countryside as Seen Through the Archaeological and Bioarchaeological Data. Whispers of a Dialogue��������������������������������������������������������������������������������������������������������������72 Anaya Sarpaki Feeding Knossos: Exploring Economic and Logistical Implications of Urbanism on Prehistoric Crete�����������88 Todd Whitelaw Beyond City and Country at Mycenae: Urban and Rural Practices in a Subsistence Landscape����������������������122 Lynne A. Kvapil, Jacqueline S. Meier, Gypsy C. Price and Kim S. Shelton Agricultural Self-Sufficiency and Mycenaean Kalamianos on the Saronic Gulf��������������������������������������������������137 Daniel J. Pullen Farming Practice and Land Management at Knossos, Crete: New Insights from δ13C and δ15N Analysis of Neolithic and Bronze Age Crop Remains���������������������������������������������������������������������������������������������������������������������152 Erika K. Nitsch, Glynis Jones, Anaya Sarpaki, Mette Marie Hald and Amy Bogaard Economy and Storage Strategies at Troy�������������������������������������������������������������������������������������������������������������������������169 Diane Thumm-Doğrayan, Peter Pavúk and Magda Pieniążek The Country in and Around the City Looking Back and Forward�����������������������������������������������������������������������������188 Dominique Garcia and Julien Zurbach

i

The editors are grateful for the support of the following institutions, without which the organisation of the Marseille congress in 2014 and the publication of the book would not have been possible: Institut universitaire de France, Aix-Marseille Université, Ecole normale supérieure, Université de Picardie Jules Verne, Centre Camille Jullian (UMR 7299), Laboratoire méditerranéen de préhistoire Europe Afrique (LAMPEA, UMR 7269), TrAme (EA 4284), the Institut national de recherches archéologiques préventives (INRAP), and the LabEx Transfers. The Musée des civilisations de l’Europe et de la Méditerranée (MUCEM) kindly hosted our congress in Marseille.

ii

Introduction Raphaël Orgeolet

Aix Marseille Univ, CNRS, CCJ, Aix-en-Provence, France

The urban and rural worlds are subject to intensive scrutiny and debate by archaeologists and historians, but few ventures attempt to explore the links between these two domains. Studies focusing on the imprints left by agricultural activities within the town are rarer still, particularly for the early periods during which towns emerged. To archaeologists, the title and some of the subjects broached in this book will thus appear to be innovative, or will at least arouse fresh interest. On the other hand, geographers will undoubtedly consider them to be somewhat more banal, but may also see them as an in vogue subject for planners. Indeed, it was in the 1970s-1980s that large towns in emerging countries, which were at that time ideal places for the interaction of the urban and rural domains, became the focus of research on the ‘ruralization of towns’ and the ‘urbanization of the countryside’ (Miossec 1985; Houssay-Holzschuch 1994). Today, this is still a promising field of research and continues to stir up interest, as for example in the geography of the Maghreb countries. Over the past few years, this topic has been taken over by planners and developers, in the light of the growing importance of urban agricultural solutions which now also concern cities in industrialized countries.

Gleason 1994), as shown by two very recent works (Malek 2013; Van Ossel and Guimier-Sorbets 2014). Depending on their location, documentation or association, gardens can be linked to our topics and the methods developed to explore them are identical to those used for the reconstruction of agricultural activities. There are three notable exceptions as regards wider scale strictly archaeological approaches: the Mesoamerican, Singhaler and Khmer cases (Fletcher 2009), which are all part of a specific urban model, characterized by a very low density of buildings in a lowland subtropical environment (Isendahl 2012; Evans et al. 2007; Fletcher 2009), an urban complex defined, sustained, and ultimately overwhelmed by a complex water management network. Since the 1980s that view has been disputed, but the debate has remained unresolved because of insufficient data on the landscape beyond the great temples: the broader context of the monumental remains was only partially understood and had not been adequately mapped. Since the 1990s, French, Australian, and Cambodian teams have sought to address this empirical deficit through archaeological mapping projects by using traditional methods such as ground survey in conjunction with advanced radar remote-sensing applications in partnership with the National Aeronautics and Space Administration (NASA). Apart from these exceptional cases, the issue of agro-pastoral activities in urban settings has hardly been broached by archaeology.

Unlike geographers and sociologists, historians and archaeologists generally steered clear of this topic and studies led by the latter are exceptions to the rule. The Medieval town generated some interest in this regard: for example, the urban 14th century censuses in Montpellier show that a fifth of the town’s populations was made up of agricultural labourers. In Great Britain, Henry French developed remarkable research on the use of ‘commons’ in modern age towns (French 2000). But this is mainly the work of historians, in the same way as the best known works on agro-urban activities, namely the major transformations associated with the Industrial Revolution in the 19th century. Nonetheless, we observe an imbalance, illustrated by an interesting symmetry: in the economic historiography of the West spanning the period from the end of the Middle Ages until the modern era, rural crafts and proto-industry play a vital role, and are also used to account for the Industrial Revolution. Conversely, the agricultural functions of towns are very largely overlooked.

When working at a different scale focusing more generally on links between the urban and rural worlds, and when we incorporate processing and storage activities, the state of research is considerably different. If we examine the early periods, these questions are generally only tackled for historical periods, but not for protohistory. Consequently, the aim of this volume, which represents an international colloquium that took place in Marseille in October 2014, is more to stir up interest in a rather neglected topic, than to establish an overview of the subject. In this way, we did not wish to extend the scope of interest by widening the chronological and geographic fields and we consequently limited our range to the protohistoric Mediterranean region. These questions are clearly innovative, whereas this chronocultural area is particularly well adapted to such a study: the urban phenomenon occurs there at an early stage, at least

Finally, it is important to mention recent renewed interest for the archaeology of gardens (Miller and 1

Country in the City in the eastern zone, whereas intense prior research provides an indispensable critical mass of data for our work.

systems, according to a centre/periphery model in which the town – and especially the palatial system – was seen as a concentration of consumption of agricultural production (Renfrew 1972 and abundant literature since then). Only certain facets of towns generated interest and they were never considered in their entirety. The history of towns, and in particular, of the earliest towns, focuses on pioneering dynamics, such as urbanization (Konsola and Hägg 1986), or on segments or specific roles. It is clear that everywhere, only the central remains are taken into consideration, as they are also the most impressive, and as they are both exclusive and essential, they allow for the most immediate definition of the town compared to the rest. These remains are represented by monumental architecture and the traces left by political, economic or religious institutions. Finally, some rare isolated cases have been studied, but they do not tackle the same issues as those addressed here1.

But, on a Mediterranean scale, there is no single town model, either during protohistory or during Antiquity. Moreover, during the Bronze Age and the Iron Age, the acculturation processes, referred to as ‘Mediterraneanization’ by D. Garcia and J.-C. Sourisseau (Garcia and Sourisseau 2010) are not yet at work. Therefore, a wide variety of situations exist, and in this respect, the Western Mediterranean region during the Iron Age is very different from the Eastern Mediterranean during the Bronze Age. In the latter region, the emergence of a palatial system at the turn of the second millennium, in the Aegean domain, could a priori form a useful marker for differentiating an urban and pre-palatial economic system, which was followed by the palatial system. But this interpretation presents two major inconveniences. Firstly, the Eastern Mediterranean is by no means fully palatial from the end of the second millennium onwards. In addition, certain features of palatial systems go back to the early Bronze Age, such as urbanity and the control of part of agricultural production. Ultimately, rather than a clear chronological and systemic division, it is essential to consider the degree of political and economic integration, and the Minoan and Mycenaean palatial systems represent the highest level of integration at the end of the Bronze Age.

In the West, the proximity of the Iberian and Celtic worlds with classical civilizations (first Greek, then Roman), meant that for a long time, agro-pastoral activities were rapidly classified as ‘traditional’, in comparison to those of contemporaneous classical urban societies. On the other hand, certain themes are better known everywhere. Storage, for example is the focus of attention in the East and the West, as are processing activities (milling and pressing). Nonetheless, as of yet, no studies have focused on their role in the agricultural economy. Ultimately, a systematic approach to agriculture in towns is still missing today. Such an approach should focus on production activities, as well as on storage and processing activities. To this end, we can say that the archaeology of agriculture in urban environments remains to be developed, alongside a reevaluation of the links between the town and the agro-pastoral economy. If, at least for the time being, we leave aside the question of techniques, we can ask: what was produced, where was production based and who was this production for? The advent of major agglomerations against a backdrop of heightened social hierarchisation, the development of specialized classes of craftspeople and state institutions, during the Bronze Age in the Eastern Mediterranean region and the Aegean world and during the Iron Age in the Western Mediterranean, marks at least a partial rupture with the traditional patterns of production and consumption established during the Neolithic. In spite of that, agricultural and livestock farming – we will take both of them into consideration – still hold a central role in these societies.

In the Western Mediterranean region, in the Iberian world and in southern Gaul during the Iron Age, urban systems can be assessed on the basis of contacts between indigenous societies and Greek, and then Roman colonies. Three phases can be differentiated: before processes of ‘Mediterraneanization’ (Garcia and Sourisseau 2010), the pre-classical habitat is polynuclear, archipelagic and not very integrated; it develops as part of a subsistence economy. During the contact phase, these habitats are clustered together into larger urban cores, which are also centres of agricultural production: this is the case for example for Carsac (Albi, France) during the Final Bronze Age (Guilaine et al. 1986). Lastly, the third phase is the urban boom: towns are integrated into the Mediterranean urban network and into the market system; rather than being producers, their main tasks now involve the organization and management of agro-pastoral production. The reasons behind the relative disinterest in these topics in protohistory are multiple and differ from one side of the Mediterranean to the other. In the Aegean world, like in the East, the palatial phenomenon generated several biases: agrarian systems and their links to towns were generally considered through the prism of recent and Middle Bronze Age palatial

1 With the exception perhaps of the reasonably dated work by H. van Effenterre on Malia (Le palais de Mallia et la cité minoenne, Rome, 1980).

2

Raphaël Orgeolet: Introduction

Before attempting to outline the links between the town and agro-pastoral activities, perhaps it is pertinent to define what ‘agriculture in town’ is not, or at least not yet. Our knowledge of pre-classical techniques and the agricultural economy is relatively advanced. This field of research is, in any case, extremely dynamic and has been active for a long time in the West, with the early works of F. Audouze and O. Buchsenschutz or J. Guilaine (Audouze and Buchsenschutz 1989; Guilaine 1991), followed by numerous books (Carpentier and Marcigny 2012) and studies by L. Bouby, Ph. Marinval or P. Ouzoulias. This domain was developed at a later stage in the Aegean world, when it is now vigorous, as shown at Physis 2012 (Touchais et al. 2014). On the other hand, although agglomerations are probably the best known types of habitat in archaeology, we still know relatively little of several aspects of urban reality, as this series of questions illustrates: 1.

2.

3.

dimension, as to define the criteria and tools to evaluate it. Archaeology of productive areas: production in the urban environment A first approach to these questions could proceed from a spatial perspective: this is the archaeology of productive areas or subsistence landscapes, if we use the term coined by K. Shelton and L. Kvapil. The notion of agglomeration is inseparable from the concept of density, although the degree of the latter varies considerably. One result of agglomeration is of course, the pressure exerted on the area, and in the same way as the other activities, urban agro-pastoral production is part of this context. In contemporary agglomerations, many complex schemes of spatial competition denote increased land pressure. However, it would be wrong to try to relate our pre-classical towns to models derived from the modern world as many parameters separate the first from the second and numerous variables make such comparisons difficult. In this way, the economic criteria governing the value of space are not the same, whereas other, often cultural factors create major disparities between agglomerations. We can for example allude to dense agglomerations, as opposed to loose agglomerations, as mentioned above for the Mesoamerican, Khmer and Sri-Lankan cases, but also, nearer to us, to the classical opposition in Aegean protohistory between tells and  ‘flat settlements’, which can probably be assimilated to the duality between concerted urbanism and disorganized aggregation. The work of Erika Nitsch et al. on Knossos will provide ample reflection on this topic. Occupation density within agglomerations, and even the regular displacement of certain parts of the habitat, determine the presence and the importance of free spaces liable to be used for agriculture. Limits are another important factor: some agglomerations are contained by natural boundaries (Mycenae), or by walls (Lattes), whereas others are free of these constraints and can be extended as needed. This has important consequences: land pressure is undoubtedly much higher in the first case and the degree and the perception of urbanity may thus be reinforced. In the Aegean world, open areas in urban settings have received much attention in some cases (the ‘courtyards’ of palaces and Minoan towns), and less in others (in particular in Mycenaean agglomerations). Invariably, these areas are only considered for their political potential and the possibility of any other function is not even envisaged (Cavanagh 2001 refers to the problem without suggesting a new avenue of research). New agglomerations with areas left empty represent a specific case and this question will be tackled in Daniel Pullen’s communication on Mycenaean Kalamianos.

as regards urbanism: in spite of considerable efforts in the Aegean world over the past few years (at Knossos and Mycenae in particular), only the central zones of towns are relatively well known (cf. Malia); little is known of the peripheral zones and their morphology; from an economic point of view: it is vital to assess the degree of food dependency of the town (on a domestic or community level)? By what means can we evaluate this? (storage?); at the end of the Bronze Age in the East, during the Iron Age in the West, an increased hierarchisation and concentration of the habitat point towards several degrees of urbanity: are agro-pastoral activities a pertinent gauge of this new structuration?

For the agro-pastoral economy, the type of production is relatively accessible through paleobotanic (and particularly through carpological analyses) and archaeozoological studies, which are now combined with other data (paleodiet). 4.

5.

but for us, it is essential to identify the original production place in the territory, or at least to distinguish between extensive production outside the town and intensive intramural production; which economic circuits are these productions part of? can we distinguish internal consumption from market consumption?

The last vital question concerns the types of protohistoric towns: 6.

from a social point of view: it is still difficult to represent what a protohistoric urban society may be; the question is not so much to determine whether our protohistoric towns present a rural 3

Country in the City The productive area in agglomerations also includes courtyards and gardens and henceforth, archaeology has a firm grasp of these domains. However, many methodological and technical obstacles encumber this sphere of research. We can, particularly for urban sites with complex stratigraphies, evoke difficulties in identifying undeveloped zones. Today, the enhanced accuracy of excavation techniques allows for the identification of palisades and light developments, whereas a wide range of geochemical and micromorphological investigations provide information on the role of livestock in agglomerations, but also on agricultural food processing activities. Combined with traditional investigation methods (carpology, archaeozoology), documented here by several texts, these techniques raise our expectations for differentiating between intensive productions in urban settings and extensive productions in the countryside. This is the primary aim of Nuria Rovira and Natalia Alonso for the agglomeration of Lattara (Lattes).

2001; Galaty and Parkinson 2007; Pullen 2010) largely disregarded agricultural issues, with the exception of P. Halstead’s contribution on palatial interventionism in Mycenaean agriculture (Halstead 2001). Based on earlier work by J. Killen, the latter author shows that some of the lands belonging to rural communities were cultivated for the benefit of the palace. The Mycenaean case is exceptional as it is backed up by written documentation; but it only provides evidence of links between the state domain and rural communities, with no mention of the private sphere. In addition to qualitative and quantitative information related to the exploited and consumed foodstuffs, two types of remains at the core of agglomerations provide evidence of the links between the land and production management methods. On one hand, recipients and conservation structures allow for the differentiation of several levels of storage strategies (individualhousehold / collective-community / centralized – state) and also enable us to detect the level of economic integration of agro-pastoral production (domestic subsistence production? an allocation from the authorities? access to the market?). The interest and relevance of these data are revealed here by a number of related texts (Kostis Christakis, Maia Pomadère, Emmanuella Alberti and Sylvie Müller-Celka for Crete, Diane Thumm, Peter Pavuk and Magda Pieniazek for Troy). Systematically, these studies highlight links with the land, and in the case of Troy, storage strategies are accompanied by the inflection of agricultural activities.

Agglomerations, land development and production management Mediterranean protohistory, particularly during its most recent phases, is a period characterized by what has commonly, and perhaps too conveniently been called ‘social complexification’, as in my opinion, this term is devoid of meaning when it is not explicitly defined, as is often the case. In the Near East from the 4th millennium onwards, in the Aegean world during the Early Bronze Age (towards the middle of the 3rd millennium), then especially from the second millennium onwards in the southern zone with the development of Minoan, then Mycenaean palatial systems, or during the Iron Age in the West, in the Iberian Peninsula or in Southern Gaul: all these regions undergo structuration processes affecting many domains, to different degrees and according to different systems. On a political level, this is marked by the appearance of state type institutions; on a social level, it is both the affirmation of the elite and the development of specialized classes of craftspeople; lastly, as far as the habitat is concerned, it is the concomitant observation of increasing nucleation and a more intensive hierarchisation of agglomerations.

On the other hand, processing structures (millstones, presses, etc.) provide evidence of the urban management of agricultural production. These are assessed in several papers (Natalia Alonso / Guillem Perez-Jorda) aiming to place them in their operational chains. In this way, the different processing stages can be identified, and the links between the agglomeration and farmlands can be explained. Ultimately, we can ask whether a connection can be established between the type of processing action and its position in the operational chain, and the role of the agglomeration within its political and economic network (Natalia Alonso / Guillem Perez-Jorda). Lastly, if the town is defined by certain non-objective, but nonetheless measurable criteria (building density, population density, presence of institutions, etc.), it is clear that urbanity also depends on less quantifiable and more relative parameters. In this respect, the notion of a hierarchized network is primordial and must be included in the definition of the degree of urbanity.

Most of the models dealing with these transformations rightly place agriculture in a central position, as traditionally, the introduction of new crops and progress in agricultural techniques are used to infer the production of surpluses liable to create wealth (see the works of C. Renfrew, P. Halstead, but also A. Sherratt). Here, we are not so much concerned with the accuracy and pertinence of these models, as with the observation that concrete links between central places and farmland are insufficiently explored. Even recently, interest in the political economy (Voutsaki and Killen

Conclusion: urban societies and rural societies? This brings us to the question of people and society: our starting point was the necessity of a systematic – and 4

Raphaël Orgeolet: Introduction

thus systemic - approach to agriculture, which involves questioning and balancing the role of agglomerations, as they were not only centres of decision and consumption, but integral players in agro-pastoral production. This brings us to another observation: the borders between rural and urban are undoubtedly much less clear than in most of our depictions. Would it be more apt in certain cases to refer to ‘large villages or large rural centres’, as suggested by Natalia Alonso and Guillem Perez-Jorda? Town and country: generally depicted as two antagonistic concepts, are united here and are a lot less detached from each other than we wish to believe. Some authors rightly underline the danger of using models where rural and urban are defined too exclusively of one another. In reality, in the protourban societies in question here, the urban share was reduced and amassed in small-sized agglomerations, in immediate contact with the rural world.

(eds), Archéologie des rivages méditerranéens: 50 ans de recherche: 237‑245. Paris, Errance - Ministère de la culture et de la communication. Guilaine, J. 1991. Pour une archéologie agraire: à la croisée des sciences de l’homme et de la nature. Paris, A. Colin. Guilaine, J., Rancoule, G., Vaquer, J. and Passelac, M. 1986. Carsac, une agglomération protohistorique en Languedoc. Toulouse, Centre d’anthropologie des sociétés rurales. Halstead, P. 2001. Mycenaean wheat, flax and sheep: palatial intervention in farming and its implications for rural society. In J.T. Killen and S. Voutsaki (eds), Economy and Politics in the Mycenaean Palace States. Proceedings of a Conference held on 1-3 July 1999 in the Faculty of Classics, Cambridge: 38‑50. Cambridge, Cambridge Philological Society. Houssay-Holzschuch, M. 1994. La cité sans la ville : Tuléar, Sud-Ouest de Madagascar. Géographie et Cultures 11: 63‑88. Isendahl, C. 2012. Agro-urban landscapes: the example of Maya lowland cities. Antiquity 86, n° 334: 1112– 1125. Konsola, D. and Hägg, R. 1986. Early Helladic Architecture and Urbanization. Göteborg, P. Åström, Studies in Mediterranean Archaeology 56. Malek, A. A. 2013. Sourcebook for Garden Archaeology: Methods, Techniques, Interpretations and Field Examples. Berne - Berlin - Bruxelles, P. Lang. Miller, N. and Gleason, K. 1994. The Archaeology of garden and field. Philadelphia, University of Pennsylvania Press. Miossec, J.-M. 1985. Urbanisation des campagnes et ruralisation des villes en Tunisie. Annales de Géographie vol. 94, n° 521: 38‑62. Van Ossel, P. and Guimier-Sorbets, A. M. 2014. Archéologie des jardins: analyse des espaces et méthodes d’approche. Montagnac, M. Mergoil. Pullen, D. J. (ed.) 2010. Political economies of the Aegean Bronze Age. Papers from the Langford Conference, Florida State University, Tallahassee, 22-24 February 2007. Oxford, Oxford Books. Renfrew, C. 1972. The Emergence of Civilisation: the Cyclades and the Aegean in the Third Millennium B.C. London, Methuen, Studies in prehistory. Touchais, G., Laffineur, R. and Rougemont, F. (eds) 2014. Physis : l’environnement naturel et la relation homme-milieu dans le monde égéen. Louvain, Peeters Publishers, Aegaeum 37. Voutsaki, S. and Killen, J.T. 2001. Economy and politics in the Mycenaean palace states: proceedings of a conference held on 1-3 July 1999 in the Faculty of classics, Cambridge. Cambridge, Cambridge Philological Society.

Bibliography Audouze, F. and Buchsenschutz, O. 1989. Villes, villages et campagnes de l’Europe celtique: du début du IIe millénaire à la fin du Ie siècle avant J.-C., Paris, Hachette. Carpentier, V. and Marcigny, C. 2012. Des hommes aux champs: pour une archéologie des espaces ruraux du Néolithique au Moyen Age. Rennes, Presses universitaires de Rennes. Cavanagh, W. 2001. Empty Space? Courts and Squares in Mycenaean Towns. In K. Branigan (ed.), Urbanism in the Aegean Bronze Age: 119-134. London, Sheffield Academic Press, Sheffield Studies in Aegean archaeology 4. Evans, D., Pottier, C., Fletcher, R., Hensley, S., Tapley, I., Milne, A. and Barbetti, M. 2007. A comprehensive archaeological map of the world’s largest preindustrial settlement complex at Angkor, Cambodia. Proceedings of the National Academy of Sciences, vol. 104, n° 36: 14277‑14282. Fletcher, R. 2009. Low-Density, Agrarian-Based Urbanism: A Comparative View. Insights vol.2 article 4. French, H.R. 2000. Urban agriculture, commons and commoners in the seventeenth and eighteenth centuries: the case of Sudbury, Suffolk. Agricultural history review 48, n° 2: 171‑199. Galaty, M.L., and Parkinson, W.A. 2007. Rethinking Mycenaean palaces II. Los Angeles, Cotsen Institute of Archaeology at UCLA, Monograph 60. Garcia, D. and Sourisseau, J.-C. 2010. Les échanges sur le littoral de la Gaule méridionale au premier Âge du Fer. Du concept d’hellénisation à celui de mediterranéisation. In X. Delestre and H. Marchesi

5

Elites and Farmers in Iberian Iron Age Cities (7th-2nd Centuries BC): Storage and Processing of Agricultural Products Natàlia Alonso

Grup d’Investigació Prehistòrica, 3DPatrimoni – INDEST (SGR2017-1714; HAR2016-78277-R) Dept. d’Història, Facultat de Lletres, Universitat de Lleida

Guillem Pérez-Jordà

GI Arqueobiología (HAR2015-64953-P) Instituto de Historia. CSIC, Madrid

–– larger cities covering an area of about 10 hectares (Ullastret, 9ha; Los Villares/Kelin, 10ha; Tossal de Sant Miquel/Edeta, 14ha). –– cities covering several hectares (La Bastida de les Alcusses, 4ha; Tivissa, 4.4ha). –– major fortified oppida with a complex urban layout covering between 1 or 2 hectares (Turó de ca n’Oliver; MolÍ de l’Espígol). –– fortified oppida and fortresses between 0.5ha (or less) and 1ha (Alorda Park; Mas Castellar de Pontós; Els Vilars d’Arbeca). –– and small rural settlements, at times fortified, less than 0.5ha (Castellet de Bernabé; Els Estinclells).

Introduction The term ‘Iberian Iron Age’ designates a group of territories along the eastern limit of the Peninsula sharing a common past and diverse social processes (Ruiz and Molinos 1998). The terms ‘Iberia’ and ‘Iberians’ were formulated by the Greeks to denote the broad region and its peoples stretching along the Mediterranean coast from the south-east of the Peninsula to western Languedoc. Later, from the 2nd century BC, these terms in Roman sources acquired a broader sense as they initially designated the area extending southwards and eventually came to refer to all the Peninsula (Sanmartí and Belarte 2013: 97).

Some Iberian settlements can therefore be considered cities, in spite of great disparities with regard to Hellenistic, Roman or Phoenician-Punic urban models. The Iberian city is therefore defined not so much by its size but by its social hierarchy and its degree of diversification of economic activities (Sanmartí and Belarte 2013: 97-99; Sanmartí et al. 2015). However, very few of the larger cities have been the object of extensive excavations. Thus this work is necessarily based on the centres of Ullastret, Kelin and Edeta, as well as a few other smaller sites.

Iberian society was far from homogeneous. It was divided into ethnic groups (or tribes) that formed independent territories or city-states very similar to other entities elsewhere in the Mediterranean. The processes of hierarchy and urbanisation followed different regional rhythms and forms, and have been interpreted according to the theoretical positions of different researchers. Certain authors identify the roots of the Iberians as far back as the 2nd millennium BC, as in the case of the Ilergete tribe in Western Catalonia (Junyent 2015: 170). Different views are held regarding the Iberians of the Catalonian coastal zone (Cossetani, Laietani, Indigetes) and the area of Valencia (Moreno 2011; Grau 2002) where profound cultural mutations emerged in the 6th century BC (Sanmartí and Belarte 2013: 98; Sanmartí et al. 2015).

In addition, regarding the geographical area and issues related to the archaeological record, this study focuses exclusively on one region of the Iberian territory, an area now comprised by present Catalonia and Valencia (Figure 1). It is only in this area that archaeobotanical samplings took place systematically during excavations and the studies of settlements and territories applied an interdisciplinary approach. This study area also coincides with the territory corresponding to the first definition of Iberia by the Greeks which some authors consider to be closest to the palaeoethnological reality (Sanmartí and Belarte 2013: 97).

The size of Iberian settlements increased in the 5th and 4th centuries BC, coinciding with a greater variety of urban models and hierarchical divisions. Cities and oppida took on the role of centres that gave structure to both the territory and socioeconomic relations. Yet there were variants to this precept depending on factors of size, rank, territory and ethnic group (Belarte 2013: 81; Junyent 2015: 181-182; Bonet Rosado et al. 2015; Sanmartí et al. 2015). The settlements can be classified in the following manner:

The main objective of this work is to examine how hierarchical processes – among the different settlement types and inside the cities - and urbanisation affected 6

Natàlia Alonso and Guillem Pérez-Jordà: Elites and Farmers in Iberian Iron Age Cities

frame that is subdivided into three periods: 7th-6th centuries BC (Early Iberian period), 5th-4th centuries BC (Middle Iberian period) and 3rd-2nd centuries BC (Late Iberian period). 1. Iberian agriculture The agriculture developed by the Iberians represented a radical break with the traditional territorial model initiated with the arrival of the first of farmers in the middle of the 6th millennium cal BC (Antolín 2016; Pérez-Jordà 2013; Pérez-Jordà and Peña-Chocarro 2013; Albizuri et al. 2011). Two elements exemplify this break. The first is technical and relates to the introduction of iron working and the production of agricultural tools. The second, more directly linked to crops, was the introduction of fruit cultivation. This development is commonly referred to as the Mediterranean Model of agricultural production or ‘Mediterranean triad’ (cereal, olive oil and wine), a traditional system that has survived to present times. 1.1. Crops: cereals, pulses and fruits To present an overview of the diversity and importance of crops in the Iberian period, the different products were grouped into four main groups: cereals, pulses, fruit and either oleaginous plants or craft plants. Cereals comprise common and durum wheat (Triticum aestivum-durum) and hulled barley (Hordeum vulgare subsp. vulgare). Emmer (Triticum dicoccum), einkorn (Triticum monococcum), common millet (Panicum miliaceum), Italian millet (Setaria italica) and naked barley (Hordeum vulgare var. nudum) are present in smaller quantities. Pulses comprise chickpea (Cicer aretinum), grass pea (Lathyrus sativus y L. cicera), lentil (Lens culinaris), pea (Pisum sativum), bitter vetch (Vicia ervilia), broad beans (Vicia faba) and common vetch (Vicia sativa). Cultivated fruit are fig (Ficus carica), apple and pear (Malus sp. and Pyrus sp.), olive (Olea europea), almond (Prunus dulcis), pomegranate (Punica granatum) and vine (Vitis vinífera). The oleaginous or artisanal plants are camelina (Camelina sativa), poppy (Papaver somniferum) and flax (Linum usitatissimum) (Alonso 1999; López et al. 2011; Pérez-Jordà et al. 2007; Pérez-Jordà 2013).

Figure 1: Archaeological sites cited in the text: 1, Pech Mao; 2, Mas Castellar de Pontós; 3, Saus; 4, Emporion; 5, Illa d’en Reixac; 6, Ullastret; 7, Sant Esteve d’Olius; 8, Molí de l’Espigol; 9, Turó de ca n’Oliver; 10, Montjuic; 11, Turó de Font de la Canya; 12, Alorda Park; 13 Vilars; 14, Estinclells; 15, Tivissa; 16, Kelin; 17, El Zoquete; 18, Castellet de Bernabé; 19, La Monravana; 20, Edeta; 21, La Bastida de les Alcusses; 22, La Vital; 23, Alt de Benimaquia; 24, Illeta dels Banyets; 25, Tossal de les Basses.

agricultural production. The study aims to identify the agricultural practices and storage structures that bear evidence of the appropriation (or consolidation) by the emerging elite of harvests and agricultural products, as well as their distribution.

1.1.1. Early Iberian period: 650-500 BC Archaeobotanical analyses were carried out at 13 Early Iberian sites. The trend points to a predominance of cereals, with pulses and fruit (mainly grapes) present in about 10% of the samples (Figure 2A). Moreover, the only evidence of oleaginous plants is camelina. Regarding cereals, hulled barley is predominant. It is on par with naked wheat at settlements situated in zones with better soils. The values of naked barley and einkorn are very low, while emmer, common millet and

The first step to achieve this goal is to characterise Iberian agriculture based on the available data. Subsequently, throughout the text, the study will point out the differences in agricultural systems between the regions south and north of the Ebro River in spite of the fact that none is completely homogeneous. The study will also focus on the diachronic evolution during the Iron Age, between the 7th and 2nd century BC, a time 7

Country in the City Italian millet are more widespread, with values are even higher to the north of the Ebro River. The chickpea is the novelty among the pulses as it makes its appearance in this period. Yet the lentil is the most abundant pulse. Grass pea, peas, broad beans and common vetch are less significant. Fruit, possibly cultivated since the 8th century BC, is recorded throughout the territory with vine being the most remarkable. Fig is found in company of vine to the north of the Ebro while to the south, by contrast, fruit is more diverse: apple or pear, pomegranate and olive. 1.1.2. Middle Iberian period: 500-300 BC There are 349 samples from 12 settlements of the Middle Iberian period. In spite of the increase of fruit and oleaginous plants such as flax and camelina, the trend follows that of the previous phase (Figure 2a). There is a clear difference between the south, that maintained a balance between cereal and fruit, and the north with a predominance of cereals, like the model it followed in earlier times. Hulled barley remains predominant although the values of naked wheat, in some cases, are equivalent. Naked barley and einkorn continue to show low values. It is therefore not possible to determine if they are really cultivated crops or if their presence among the other taxa is accidental. Emmer, common millet and Italian millet show more consistent values mainly in the Catalonian fields to the north of the Llobregat River (Figure 1). Regarding pulses, chickpea disappears from the record while bitter vetch makes its appearance. Yet lentil is the most frequent pulse, especially among the Catalonian sites. Grass pea, broad beans, peas and common vetch also show high values. Common vetch is especially remarkable at the settlement of Bastida de les Alcusses as it systematically appears inside houses, indicating consumption by the inhabitants. Although the almond makes its appearance in this period, the vine remains the most cultivated and widespread fruit. A wider diversity of fruit is observed in the southern study area where certain sites appear oriented towards systematic exploitation of fruit towards commerce. Oleaginous plants are restricted to the sites of Bastida de les Alcusses and Illa d’en Reixac. Camelina is recorded at both sites while flax is only found at the first and poppy in the second.

Figure 2: Evolution of the ubiquity rates of the crop types (si: number of sites; sa: number of samples): a, in the whole area of study; b.1, in sites to the south of the Ebro River; b.2, in sites to the North of the Ebro River.

a reversal of roles (dominance of wheats). There are no specific figures concerning naked barley as a crop as this species was only marginally identified at one site. Einkorn, in turn, only shows relevant values at Castellet de Bernabé while emmer, common millet and Italian millet are more common. The two species of millet are specific to the Catalonian coastal area.

1.1.3. Late Iberian period: 300-200 BC Twelve sites provide the data for the Late Iberian period. The tendency, once again, changes little and two very clear areas with contrasting agricultural orientations stand out (Figure 2b1 and 2b2). In Valencia, cereal and fruit remained balanced, whereas the presence of fruit in Catalonia is subordinate. The prevalence or equivalence between hulled barley and naked wheats continues to represent the general trend. The exception is the case of settlements of the Lleida Plain that reveal

Grass pea, lentils, peas and broad beans are the most common pulses while the values of common vetch are less relevant. A diversity and influence of arboriculture 8

Natàlia Alonso and Guillem Pérez-Jordà: Elites and Farmers in Iberian Iron Age Cities

remains the norm among the sites of the south, while in the north only vine cultivation is recorded.

wealth. The different elements that allow confirmation of agricultural development are present in both rural and urban settlements. In this sense, both types of settlements are part of what was fundamentally a ‘rural’ society.

1.2. Evolution of the agricultural system As noted previously, the fundamental base of the agricultural system in this territory from the middle of 7th BC was extensive farming, and the weight of production fell on two types of products: cereals and fruit. Yet pulses, a series of oleaginous plants and other horticultural crops continued to be grown. Products of horticulture, identified for the most part from carbonised samples, are in fact difficult to detect in the archaeobotanical record.

2.1. The evidence of tools and tillage The Iberians used exclusively two types of force to operate their agricultural implements: human and animal. The archaeological record indicates that rural work was carried out with tools incorporating iron elements, breaking with the ancient tradition of wood, bone and stone implements (Sanahuja 1971; Pla 1968, 1972). It is very difficult, nonetheless, to define the moment when iron was systematically used for their manufacture. It is in fact a type of material that was often reworked, making finds of iron implements rare on archaeological sites. Although iron begins to appear in the settlements since the 8th-7th century BC (Alonso 1999; Le Meaux and Sánchez de Prado 2007), the first tools that can be directly related to agriculture are not recorded until later periods.

The development of this agricultural system may have begun even earlier. Cultivation of fruit is recorded since the end of the 9th (or early 8th) century BC in Andalusia, and later in the south of Valencia (PérezJordà 2013 ; Pérez-Jordà et al. 2017). The introduction of this innovative agricultural feature in the study area is an issue that is still not clear. It must be kept in mind that pottery-based and radiocarbon analyses in this time frame are not accurate.

Another indirect indication suggesting an early use of iron for the manufacture of agricultural implements is the disappearance of flint sickles introduced millennia earlier by the first farmers, and systematically used until the dawn of the first millennium BC. Hence, in all appearance, the early models of sickles with flint blades were abandoned in favour of the new iron sickles.

In the Early Iberian period the two territories show relatively similar behaviours (Figure 2b1 and 2b2). The range of values between 70% and 90% indicate that cereals are the dominant crop. Fruit on the whole does not exceed 20%. It is during the subsequent Middle Iberian period that the different orientations of the two areas begin to take shape. Valencia adopts a diversified arboriculture that will become one of the fundamental bases of its agricultural activity. This new situation is observed at Alt de Benimaquía in the 7th century BC, a settlement with wine production among its main activities (Gómez Bellard et al. 1993).

The first large assemblages of agricultural tools in the study area appear in the 4th century BC (Pla 1968, 1972) and provide insight into the development of work in the fields. Among the finds are ploughshares and bronze miniatures depicting ploughs driven by a pair of oxen (Figure 3). Picks, pickaxes and hoes were used to complement the work of tilling. Hoes, in fact, are characteristic of orchard work. At this time sickles and other tools appear for harvesting cereal, billhooks for pruning fruit trees, and vine billhooks for harvesting grapes. This set of tools is familiar. They show no major modifications through time and are very similar to the tools in use today.

This change will become even more apparent in the 5th century BC when arboriculture takes on a larger role and spreads throughout a large area of the territory of Valencia, and veritable factories surface along the coast specialised in the production and commercialisation of different fruit and their derivatives, mainly wine. A different situation is noted in the territories to the north of the Ebro. There is no sign of an increase of arboriculture, and the diversity of fruit remains low, centred on the vine. The northern area therefore remains geared towards cereals, a fact corroborated notably storage and production features.

Tool finds from this period are not common, except in the case of hoards or settlements that saw a sudden violent end, leaving most of its features and artefacts in situ. Yet, based on the spread of these tools throughout the study area, it is evident that they were present on both small rural settlements and in urban nuclei.

2. Country in the city: farmers in Iberian cities

An exceptional case is that of La Bastida de les Alcusses where a large number were preserved in situ due to the site’s violent end. A recent paper (Bonet and Vives-Ferrándiz 2011a), based on the spread of tools throughout the different houses, has noted that most

It is not simple, in terms of agricultural activity during this period to differentiate the rural and urban worlds. Agriculture not only ensured the livelihood of most of the population, but at the same time, generated 9

Country in the City

Figure 3: Tools and tillage in Iberian sites (modified from Bonet and Vives-Ferrándiz 2011).

of the settlement’s inhabitants took part in agricultural activity. At the same time, the spread of ploughs, serving to assess the extension of the land cultivated by each of the families, showed concentrations in certain richer residences. It is noteworthy that the clusters of ploughs also coincide with the only houses linked to viticulture, as evidenced by the spread of vine billhooks (Vives-Ferrándiz 2013).

units with no atmospheric control, although designed for the shorter-term, contained, at times, large volumes. • Storage pits The presence of underground silos in this territory is not new as they were introduced with the arrival of the first Prehistoric farmers. This storage unit will see a systematic and stable use in certain areas such as Catalonia. In Valencia, in turn, its use was less common.

2.2. The evidence of cereals

2.2.1. Cereal storage

It is currently difficult to explain why silos in Catalonia (especially in eastern Catalonia) were since the 7th century BC one of the defining elements of Iberian settlements while absent at contemporary sites south of the Ebro (Figure 4) (Asensio 2015; Prats 2016). Yet earlier 8th-century BC silos have recently been recorded in Valencia at the settlement of La Vital (García Borja et al. 2013). These new finds suggest that the silo tradition had not been lost in former times, but was abandoned deliberately by the local Valencian communities for economic reasons.

Two types of cereal storage features are recorded in Iberian territory (Sigaut 1988; Alonso 1999). Their classification is based on whether they are airtight or not. Silos controlling the atmosphere allow grain preservation for longer periods of time. Storage in the

On the contrary, the presence of underground silos on coastal and inland settlements north of the Ebro is a reality. Although in most cases they number between one and ten, some sites have more (up to 50, rarely more than 100). The exception is the Middle Iberian

As noted above, cereals were the staple food of the different Iberian communities. It is also certified that in some areas cereal cultivation went beyond that of mere sustenance, with an orientation towards producing a surplus for commerce. Besides the data from the archaeobotanical record, certain storage features reinforce the notion of grain commercial surpluses and their study contributes to quantify their volumes and longevity.

10

Natàlia Alonso and Guillem Pérez-Jordà: Elites and Farmers in Iberian Iron Age Cities

litres (Prats, 2016). The purpose of these features ranged from the small domestic silo to the large, tradeoriented surplus storage unit. • Elevated granaries Elevated granaries are recorded for the first time in the Peninsula in the Iron Age (Rodríguez Díaz et al. 2009). They only appear in the northern and southern ends of the territory to the south of the Ebro at sites devoid of underground silos (Figure 4). The oldest appears to date to the 5th century BC (Gil Mascarell 1976; Alvarez García 1997; Olcina et al. 2009; Bonet et al. 1994). Later, in the 2nd century BC, elevated granaries appear to be predominant in the northern half of Valencia (Cabanes and Vizcaino 2010; Cabanes et al. 2010). These granaries consist of parallel walls supporting a pavement of stone slabs or wooden planks serving as the foundation of a series of wooden or adobe cubicles that contained the grain. It is a type that has survived until today, for example, in northern Morocco (PeñaChocarro et al. 2015) and has a large storage capacity, at times up to 20,000 litres. Like certain underground silos, it often is (mainly between the 5th and 4th century BC) linked to commercial activity. Settlements along the coast incorporate the largest models while those of the inland tend to be smaller. • Other storage features: clay containers (‘trojes’), pottery jars, amphorae

Figure 4: Distribution of sites with storage structures (information about the area north of the Ebro River provided by G. Prats).

Trojes, grain storing cubicles with clay or stone walls, are the most basic. Their base does not appear to have any particular treatment and their capacity varies from a few hundred to several thousand litres. They are at times inside houses in the pantry. In other cases a battery of trojes make up veritable granaries. It is a type of structure that is commonly found on the top floor of houses. This position is at times inferred as conditions of conservation render it difficult to identify archaeologically (Guérin 2003: 238-267). Yet lower floor trojes granaries were abundant in the Iberian world.

site of Sant Esteve d’Olius with more than 120 (Asensio et al. 2001). The location of the silos in the settlements varies. They are both inside (even in houses) and outside or in large concentrations. Camps de sitges (fields of silos) are associated with specialised settlements placed on trade routes either in the inland or along the coast. These sites indicate the existence of centres that hoarded cereal surpluses, possibly due to the system of tribute (Asensio et al. 2002; Asensio et al. 2009; Burch et al. 2010; Sanmartí 2010). The magnitude of these ‘fields of silos’ in settlements along trade routes connecting the inland sites with the great coastline urban centres leaves no doubt as to the productive and organisational capacity of Iberian society (Asensio et al. 2002: 136-137).

The function of these features varied between the simple household granary, the larger-scale communal storage unit, or the storage surplus unit owned by the local elite. It must be noted that in some instances, such as the group of trojes at the sites of La Monravana and Castellet de Bernabé (Guérin 2003; Pérez-Jordà 2013), it is difficult to determine their nature and the ownership. Yet, at the second site there is a clear contrast between the capacity of the modest trojes inside the houses with the capacity of the large trojes in the buildings along the street. A similar case is that of La Monravana where only a small sector of the settlement was excavated with absolutely no indication of a residence. Yet this site comprises trojes with a storing capacity of more

The volume of grains stored in Iberian underground silos also differed greatly. Although the average silo contained about 1000 litres, the smallest held about 100. Some had even higher capacities, up to 80,000 11

Country in the City

Figure 5: Percentage of grain, weeds and chaff plotted sample by sample as a triangular scatter diagram (data from Alonso et al. 2008; Canal and Buxó 2002; López 2008; Pérez 2013; Alonso and López, unpublished).

than 20,000 litres, rotary pushing mill podia and wine presses. It is difficult, however, to ascertain the role of what are clearly production and storage areas independently of the settlement context. The large granary of La Monravana can, in fact, be interpreted either as a communal feature or one owned by the dominant rural elite.

few finds are fortuitous. These vessels are more closely linked to storage other liquid or fatty products. 2.2.2. The evidence of the cereal processing in cities A characteristic of Mediterranean agriculture, due to climatic factors such as hot and dry summers, is that a large part of grain processing did not take place in the settlements. Threshing, winnowing and sifting, before storage, usually took place on threshing floors outside the settlements, either in the fields themselves or in higher areas to take advantage of stronger air currents. This schema coincides with the data obtained from excavations where few elements can be linked to the early stages of processing grain.

The object of the granary of La Bastida de les Alcusses, on the other hand, is clearer (Bonet and Vives-Ferrándiz 2011b). It is evident that this large storage unit in the centre of the city played more that a domestic role. Yet it is harder to pinpoint whether it served as a communal grain reserve or if it was the property, or under the control of, the dominant groups inside the community. Other clearly domestic grain storage units comprised esparto straw or cloth baskets or bags, ceramic vessels and a variety of features made with perishable materials (wood, dung, etc.) that are difficult to detect on an archaeological site (Peña-Chocarro et al. 2015). Although grain assemblages associated with esparto have been identified in different settlements, their poor state of conservation makes them difficult to define.

• Before milling: sieving and cleaning by hand When the common scheme of processing cereals was followed (Hillman 1981 1984), grains arrived at the sites after a first sieving. They only required a later finer sieving to remove the weeds and the finer elements of the chaff. This is observed at the Fortress of Els Vilars d’Arbeca (Alonso et al. 2008) and Castellet de Bernabé (Pérez-Jordà 2013). Most of the samples analysed from Iberian sites are virtually free of weeds and chaff (Figure 5). The exception corresponds to certain smaller settlements where samples with a higher proportion of chaff give the impression of a different type of grain management. The lack of extensive excavations

Among the ceramic vessels there are two types of storage jars: amphorae with their narrow mouths and long bodies, and tinajas (large jars) with wider mouths a greater capacity. Ceramic vessels do not appear have been specifically intended for grain storage and the 12

Natàlia Alonso and Guillem Pérez-Jordà: Elites and Farmers in Iberian Iron Age Cities

and systematic sampling at larger settlements make it difficult to confirm a potential difference of grain management. Yet it is possible to envision that some of the stages of the cereal cleaning process took place beyond the fields in urban nuclei. Furthermore, it is even possible that waste generated in processing areas was transported from the fields to the smaller rural settlements to serve as animal feed. This situation contrasts with that of urban nuclei where animals were not integrated into the settlement, remaining outside the residential quarters. Another interpretation for pre-Roman and Roman sites, developed by English researchers, is that large assemblages of clean grains reflect a centralised management administered by large granaries (Stevens 2003; van der Veen and Jones 2006). This could be the case of La Bastida de les Alcusses. Sites revealing a higher frequency of chaff, by contrast, are those where grain processing is part of the domestic sphere. This hypothesis, although reflecting the current archaeobotanical record of the Iberian world, requires testing in future research. • Milling: the rotary quern and the ‘Iberian rotary pushing mill’ The rotary mill, according to recent research, is an innovation that was probably introduced by the Iberian Culture in the north-east of the Iberian Peninsula and in southern France since the late 6th or early 5th century BC. The largest number of these early mills are found in the north-western Mediterranean at Catalonian sites such as Els Vilars d’Arbeca, Turó de Ca n’Oliver and Alorda Park, or in France at Pech Maho (Alonso 1999; Alonso and Frankel 2017; Alonso and Pérez-Jordà 2014; Longepierre 2014).

Figure 6: Distribution of sites with production structures: rotary querns (only the well-dated examples as rotary querns are present on almost all Iron Age Iberian sites), rotary pushing mill podia, wine presses and oil mills.

the ‘Iberian rotary pushing mill’, different from the smaller hand-driven rotary quern (Alonso et al. 2016; Alonso and Frankel 2017).

It is well established that the introduction of the rotary movement in milling was an extraordinary technological development. With the advent of rotary mills, in particular since the 4th and 3rd centuries BC, two phenomenon are observed. The first is the existence of three types of mills: a) small rotary mills b) mills with either larger (diameter) or thicker (height) millstones c) and mills with both very large and thick millstones. The second is the existence of podia, constructions serving as a base to raise the level of the mills. There are also very thick stones (designated by the term ‘lower stone-bases’) that also elevate the level of the mill (Alonso and Pérez-Jordà 2014; Alonso et al. 2016). Raising the level of the mill allowed driving the large and heavy upper stone with a circular movement by pushing a horizontal lever. The characteristics of this specific type of mill known only in the Iberian world has led to establishing a new mill type labelled

Millstone podia have been documented at about fifteen Iberian sites dated from the 4th-3rd centuries BC (Figure 6). Most of these features are cylindrical, made of stone and clay. The better preserved podia have a channel to collect the flour. Their position with respect to other features (such as walls) allows a circular movement free of obstacles (Alonso et al. 2016, Figure 11). Five complete mills found in situ on this type of podia are known. The most notable are at the settlement of Tossal de Sant Miquel (Llíria, Valencia) (Bonet 1995). The total height of these mills (podium plus mill), as well as the great weight of the upper stones, suggest they were driven by a full rotary motion by either one or two individuals pushing a horizontal lever projecting from the side of the stone, a technique that 13

Country in the City clearly differentiates it from the hand-driven rotary quern. This type of mill with its characteristic levers and podia is only recorded in the Iberian Peninsula and not known, at least for the moment, in other parts of the Mediterranean (Alonso and Frankel 2017). In fact, neither rotary querns nor Iberian rotary pushing mills are known in the Greek or Punic western colonies (Alonso 2015).

production surged once again. An exceptional level of production is recorded at no less than two coastal sites (Tossal de les Basses and Illeta dels Banyets). These two settlements marketed their wine, in addition to other products, through their ports. Wine presses also begin to be detected in this time frame at inland Iberian settlements both in Valencia and Catalonia. Although the Valencian sites of Illeta dels Banyets and Tossal de les Basses were abandoned in the 4th century BC, wine continued to be produced inland, possibly on a smaller scale (Pérez-Jordà 2015). The difficulty of transporting this product by land is obviously much greater. Furthermore, at this time there is a notable reduction in the size of the wine presses compared to those of the two coastal factories. There is even a difference in their style of construction as all the inland presses are either carved into the rock or built with earth and lime, as opposed to those of Illeta dels Banyets made with mortar following the traditions identified at sites under Punic influence such as Castillo de Dña. Blanca (Ruiz Mata 1995) and Truncu e Molas en Sardinia (van Dommelen et al. 2010). Therefore, two different wine press building traditions coexisted in the study area.

2.3. Bread production Bread making is a means of cereal consumption characterised by different types of features. Bread can be baked either on a grate (pre-heated in a domestic hearth) or in an oven. These types of small domestic bread making structures are often difficult to identify. Generally they tend to be preserved in the form of grates or, in some cases, in the form of fragments of the oven’s superstructure. Ovens, however, are the more obvious means of bread production. They are usually built on a circular base of stone with a clay coating and covered with a clay dome. Large bread ovens have been found at La Bastida de les Alcusses and Els Vilars d’Arbeca. In both cases they are associated with large granaries (Bonet and Vives-Ferrándiz 2011b; Alonso et al. 2015; JunyentLópez 2016). Other examples are known in a house in the aristocratic sector of the city of Edeta (Bonet et al. 2007) and on smaller rural settlements such as Zoquete (Mata et al. 2009).

Wine presses appear in large cities as well as in middlesized and small settlements. In the territory of Kelin, Valencia (Pérez-Jordà et al. 2013), they are found outside the settlements spread among the fields. A link between grape growing and wine production with the highest classes of society can in certain cases be established. In the city of Kelin, for example, one of the richest houses included a large cellar, presumably for wine storage, filled with amphorae with a total capacity of about 5000 litres (Mata et al. 1997). An aristocratic house in the city of Edeta, among other craft-related structures, had a wine press (Bonet 1995). Another example is Bastida de les Alcusses where, despite the absence of wine press finds, many billhooks, the tool used to prune grapes, were found in the largest house which also contained the richest goods (Vives Ferrándiz 2014).

2.4. Wine production The development of viticulture is recorded as early as the 7th century BC in archaeobotanical samples throughout all of the study area. This period coincides with the earliest wine press finds at Alt de Benimaquia (Gómez Bellard et al. 1993). Wine production will characterise agricultural activity, especially in the area of Valencia. Although there is evidence of this crop to the north of the Ebro, no wine presses have so far be recorded for this chronological period (Figure 6). Wine production in the two study areas will follow the trends observed elsewhere in the Iberian Peninsula (Pérez-Jordà 2015). Its trade during the 7th century BC appears to be a significant commercial activity and as evidenced by the presses at Alt de Benimaquia. Amphorae produced at this site, as well as the many others found in the southern half of the Iberia, were distributed throughout the eastern rim of the peninsula at sites both along the coast and the inland.

Only one site north of the Ebro, the fortified settlement of Estinclells (Asensio et al. 2010), can be linked to viticulture. This site, dating from 3rd century BC, had two contiguous buildings housing Iberian rotary, leveroperated, pushing mills (Asensio et al. 2013). Another building could have potentially had a multi-functional use since the press could have also served to produce oil and there would have been enough space to store grain, as corroborated by the high density of remains of wheat and barley.

Nonetheless wine commerce appears to retract in the 6th century BC. Presses at sites like Alt de Benimaquia were abandoned and no structures, to date, in this moment. But later, in the 5th century BC, wine

Yet fragments of wine presses, reused or discarded, have been recovered in other sites of Catalonia. They date to the late 5th or early 4th century BC in the rural settlement of Saus (Saus-Camallera-Llampaies) 14

Natàlia Alonso and Guillem Pérez-Jordà: Elites and Farmers in Iberian Iron Age Cities

near Emporion (Casas 2010) and the Fortress of Els Vilars d’Arbeca (Junyent and López 2016). They are also in later contexts (4th and 3rd century BC) in the storage and distribution centres of Turó de la Font de la Canya (Avinyonet) (López 2015), Mas Castellar (Pontós) (Pons 2002) and Mas Castellà (Santa Margarida i els Monjos) (Giró 1960-1961). Although none of these sites corresponds to what can be called cities, most are linked to centres of political and/or economic power.

One of these cases is the association of large granaries, Iberian rotary pushing mills and ovens. La Bastida de les Alcusses and the Fortress of Els Vilars d’Arbeca are examples where grain was stored, milled and baked for consumption (Junyent and López 2016) (Figure 7). This type of correlation exemplifies how dominant social groups based part of their power on the control of the grain surplus. These types of settlements appear to confirm the hypothesis for the oppida of Upper Andalusia that recognise these types of features, interpreted as silos (Ruiz Rodríguez 2007: 240), as the element that conditions the construction and layout of the oppida (Bonet and Vives-Ferrándiz 2011a: 249). Yet, at the same time, these sites exemplify that grain storage space took on an added value including among its functions the milling of flour and baking of bread.

2.5. Oil production Although there is evidence of olive oil production in the study area, its scope is far from that of viticulture (Figure 6). The only oil mills identified so far are in the Valencia study area (Pérez-Jordà 2000; Martínez Carmona 2014). Although some are pointed out in Catalonia at Estinclells (Asensio et al. 2010) and Saus (Casas Genover 2010), recent research suggests that these features are actually linked to the wine making (López et al. 2013; Pérez-Jordà 2015).

Another example is found at the eastern end of the Monravana settlement where grain was stored in large trojes and grain milled by five Iberian rotary pushing mills. The area also had two presses. Somehow this area of the settlement could have a similar function to that suggested previously, although this interpretation requires caution due to lack of knowledge of the context of the rest of the site (Figure 7).

The concentration of oil production features in the southern half of the study area is in part conditioned by environmental factors. In the north, with the exception of the coastline, the zone of wild olive growth is restricted, a factor that certainly limited the expansion of this crop. Furthermore, it is more difficult to estimate the productive capacity of oil mills than that of wine presses. Yet the rarity of olive oil works and the small dimensions of the few that are known suggest smallscale workings. Therefore the production of olive oil, in spite of its massive consumption, does not appear to have been oriented towards commerce.

This type of combination of complex houses with grain storage and processing products is not as well documented in the northern cities. However, the potential elite residences share a number of common elements including the concentration of artefacts such as loom weights or mills, artefacts that are either not detected or in small quantities in the humble houses (Belarte 2013: 86). An example is the first phase of occupation of Puig de Sant Andreu de Ullastret (Zone 14) comprising a large porticoed house with a patio housing the podium of an Iberian rotary pushing mill (Martín et al. 2004: 271-275). Wine production is also associated with cereal milling and bread baking on the ground floors of one of the residential areas of the city of Edeta (Bonet 1995; Bonet and Mata 2009). In this case, each of the activities occupies its own space. In one of the houses there is a wine press and a large Iberian rotary pushing mill, while in the other there is a mill and an oven. And, as the cases cited previously, the houses linked to these activities are the richest.

3. Production areas into the settlements Structures of production, storage and agricultural product processing appear in association with certain tools within the different settlements. Agricultural tools, small grain deposits, assemblages of storage vessels, one or more mills, or structures for processing and cooking foods, as pointed out previously, appear to meet the needs of sustenance of the household. Examples are known at Castellet de Bernabé in room 4 (Guérin 2003; Pérez-Jordà 2013) and Estinclells (Asensio et al. 2007) in adjacent houses (nos. 1 and 3) in conjunction with granaries and mills. The size of these structures is small and they appear to respond to the domestic sphere.

4. Elites and peasants in Iberian settlements Agriculture and its products are key elements in both the diet of these communities and the process of establishing a social hierarchy that will result in the shaping of territorial units that have been defined as archaic states (Sanmartí 2004; Bonet et al. 2009; Moreno 2011). Land is the fundamental element in all of the process and the archaeological record validates the existence of very different realities, not in the sense of farming, which is relatively uniform, but in the

At the same time, several larger assemblages stand out and indicate the existence of larger-scale forms of production. The analysis of some of these assemblages demonstrate the link of the ruling classes to agriculture, a fact that has in part been pointed out in the description of the different elements linked to agricultural production. 15

Country in the City

Figure 7: Two examples of sites with different types of production in the same building.

orientation and scale of production. Small farmers coexisted with elites that were capable of exploiting and monopolising vast areas and reserving the best soils for themselves.

Catalonia, cereal production was always predominant, and in some areas only viticulture gained a certain level of importance.

The absence of texts for this period complicates the definition of the forms of land exploitation. It is not known how the ownership of land was organised or the role played by the farmers. It is most often thought that kinship to the community conditioned access to land (Ruiz and Molinos 1998: 295; Bonet et al. 2008). In any case, it is evident that through this historical process some members of the community had the ability of converting the land and its products into a source of wealth and status.

There are nevertheless a number of diverse elements that together facilitated the development of communities that will be closely connected to the Mediterranean world to where they could export agricultural products that were, at certain times, in high demand. Improvements or innovations in agriculture were therefore essential to the shaping and social development of these Iberian communities. Nutrition was a key factor in any society and certain groups took advantage of technological innovations in agriculture to gain wealth.

Advances in technology multiplied the capacity of production, a fact that will enable a significant expansion of the population. Iron tools facilitated agricultural work and, at the same time, expanded the possibilities of exploitation to previously untouched areas. Fruit production was another key element of this process, especially in lands where environmental factors were not particularly conducive to the cultivation of cereals, as in the area of Valencia. In

It is not the first time in these territories that a surplus of agricultural products was concentrated in the hands of certain individuals. Yet the difference with the period under study, once again, is a question of scale. The higher capacity of production allowed expansion of the population, while more populated territories led to the development of more complex hierarchies. Another factor related to the control of the production of the food necessary to maintain these groups was 16

Natàlia Alonso and Guillem Pérez-Jordà: Elites and Farmers in Iberian Iron Age Cities

the supplemental ‘income’ obtained from the trade of agricultural and iron products with domestic, and especially, foreign markets.

cities, although these can act as centres of exchange. The control and exploitation of the land is therefore one of the fundamental elements of power (Bonet et al. 2015: 263-265).

The link of these communities with the land and with agricultural activity was never severed throughout the period of urbanisation that developed throughout this territory at the outset of the 6th century BC. On the contrary, working the land became essential to the elite to consolidate power. The city was in fact nothing more than an enlarged settlement and there was no conflict between it and the surrounding countryside (Osborne 1991). Most of the city’s inhabitants continued to live directly or indirectly off the land. Furthermore, as has been recognised in other territories with a more developed process of urbanisation (as the case of Athens), the dividing line between the countryside and the city was not evident from the political, religious or population standpoints (Humphreys 1978: 131; Meiksins Wood 1988).

As a result of this process two different agricultural realities emerge. A more diversified agriculture was initiated in the area of Valencia with a predominance of cultivation of fruit in all types of settlements. Environmental factors may have helped define this orientation. Yet it also applies to areas with soils suitable for the cultivation of cereals. Hence, guiding agriculture in this direction is the result of a political decision. By contrast, agriculture in Catalonia, in spite of the presence of arboriculture, focused mainly on grain production. The political factors that ended up defining this strategy require reassessment. Without underestimating the aspects of tradition or environmental issues that may favour one or the other production, one must consider to what extent the demands of the market or the influence of the great Mediterranean cultures may have conditioned these contrasting orientations.

The absence of written documents in the Iberian world does not favour the understanding of the social relationship between rural and urban populations. The archaeological record, however, does reveal that there is no difference between the economic activities of the two.

Bibliography Albizuri, S., Alonso, N. and Cachero, J. 2011. Economia i canvi social a Catalunya durant l’edat del bronze i la primera edat del ferro. In S. Valenzuela-Lamas, N. Padrós, M. C. Belarte and J. Sanmartí, J. (eds), Economia agropecuària i canvi social a partir de les restes bioarqueològiques. El primer mil·lenni aC a la Mediterrània occidental, Arqueo Mediterrània 12 : 11-36. Alonso, N. 1999. De la llavor a la farina. Els processos agrícoles protohistòrics a la Catalunya Occidental. Lattes, CNRS, Monographies d’Archéologie Méditerranéenne 4. Alonso, N. 2015. ‘Moliendo en ibero, moliendo en griego’: aculturación y resistencia tecnológica en el Mediterràneo occidental durante la Edad del Hierro. Vegueta. Anuario de la Facultat de Geografia e Historia, Universidad de las Palmas de Gran Canaria: 23-36. Alonso, N., Bernal, J., Castellano, A., Gonzàlez, S., Junyent, E., López, J., Martínez, J., Mazuque, J., Nieto, A., Oliva, J. A., Piquès, G., Prats, G. and Vila, S. 2015. La fortalesa dels Vilars (Arbeca, les Garrigues): noves aportacions sobre la primera edat del ferro i l’època ibèrica (2010-2015). Balaguer-Lleida, Primeres jornades d’arqueologia i paleontologia de Ponent: 88-89. Alonso, N. and Frankel, R. 2017. Survey of ancient milling systems in the Mediterranean. Revue Archéologique de l’Est supplément 43:461-478. Alonso, N., Junyent, E., Lafuente, A. and López, J. 2008. Plant remains, storage and crop processing inside the Iron Age fort of Els Vilars d’Arbeca (Catalonia, Spain). Vegetation History and Archaeobotany 17: 149158.

There is only one type of settlement where it is not possible to confirm the presence of farmers, in spite of the fact that its activity depended largely on agriculture. These were small coastal sites that had the infrastructure to moor boats and an industrial network capable of producing metal objects and amphorae to contain a series of products. One of the products sealed in the amphorae, of agricultural origin, was wine. Examples of this type excavated in the southern half of the study area are Illeta dels Banyets (Olcina et al. 2009) and Tossal de les Basses (Rosser and Fuentes 2007), enclaves where the cultural affiliation of its inhabitants is difficult to determine due to the clear links with the Punic world (Pérez-Jordà 2013; Aranegui and Vives Ferrándiz 2014). In Catalonia, in turn, a site that could share this character is a little known coastal settlement with silos of Montjuic, Barcelona (Asensio et al. 2009). There are different elements that confirm that agricultural activity continued to be developed in both small rural nuclei and in cities. The houses of the urban aristocratic groups are not isolated from agricultural activities. In fact, they developed agricultural activities and profited from these products and their derivatives as a source of wealth (Bonet et al. 2007; Bonet et al. 2015). The differences cannot therefore be drawn along the lines of the countryside and the city, but within each community, where both the socially powerful classes and the lower classes are present (Belarte et al. 2013). The power is scattered and in no case centralised in 17

Country in the City Alonso, N. and Pérez-Jordà, G. 2014. Molins rotatius de petit format, de gran format i espais de producció en la cultura ibèrica de l’est peninsular. Revista d’Arqueologia de Ponent 24: 239-255. Alonso, N., Pérez-Jordà, G. and López Reyes, D. 2016. Les moulins rotatifs poussés du monde ibériques: caracteristiques et utilisation. In L. Jaccottey and G. Rollier (dir.) Archéologie des moulins hydrauliques, à traction animale et à vent des origines à l’époque médiévale. Actes du Lons-le-Saunier, du 2 au 5 novembre 2011: 559-578. Besançon, Presses universitaires de France-Comté. Alvarez García, N. 1997. El Almacén del Templo A: aproximación a espacios constructivos especializados y su significación socio-económica. In M. Olcina (ed.), La Illeta dels Banyets: (El Campello, Alicante): estudios de la Edad de Bronce y época Ibérica: 133-170. Alicante, Museo arqueologico provincial de Alicante. Antolín, F. 2016. Local, intensive and diverse? Early farmers amb plant economy in the North-East of the Iberian Peninsula. Advances in Archaeobotany, 2. Ed. Barkhuis, Groningen. Aranegui, C. and Vivez Ferrándiz, J. 2014. More than neighbours: Punic-Iberian connections in southeast Iberia. In J. C. Quinn and N. C. Vella (eds) The Punic Mediterranean. Identities and Identification from Phoenician Settlement to Roman Rule: 243-256. Cambridge: Cambridge University Press. Asensio, D. 2015. Ceràmiques importades, béns de prestigi, pràctiques socials i sistemes productius entre les comunitats ibèriques del nord-est peninsular. In C. Belarte, D. Garcia and J. Sanmartí (eds) Les estructures socials protohistòriques a la Gàl·lia i a Ibèria. Homenatge a Aurora Martín i Enriqueta Pons: 237-249. Barcelona, Universitat de Barcelona. Asensio, D., Cardona, R., Ferre, C., Morer, J. and Pou, J. 2001. Tipus d’assentaments i evolució del poblament ibèric a la Catalunya central (eix Llobregat-Cardener). In A. Martín and R. Plana (eds) Territori polític i territori rural durant l’edat del ferro a al Mediterrània occidental, Actes de la Taula Rodona celebrada a Ullastret del 25 al 27 de maig de 2000: 183201. Girona, Museu d’Arqueologia de Catalunya. Asensio, D., Cardona, R., Ferre, C., Morer, J., Pou, J. and Saula, O. 2003. El jaciment ibèric del Estinclells (Verdú, Urgell): un assentament fortificat ilergeta del segle III aC. Revista d’Arqueologia de Ponent 13: 223-236. Asensio, D., Francés, J. and Pons, E. 2002. Les implicacions econòmiques i socials de la concentració de reserves de cereals a la Catalunya costanera en època ibèrica. Cypsela 14: 125-140. Asensio, D., Cardona I Colell, R., Ferrer I Alvarez, C., Morer, J., Pou Vallés, J., Saula, O. and García-Dalmau, C. 2010. Una almàssera del segle III AC dins del nucli ibèric dels Estinclells (Verdú, Urgell). Urtx: Revista cultural de l’Urgell: 55-76.

Asensio, D., Cardona, R., Ferrer, C., García-Dalmau, C., Morer, J., Pou Vallés, J., Saula, O. and Belarte, M. C. 2009. L’arquitectura domèstica en el nucli fortificat ilergeta dels Estinclells (Verdú, l’Urgell), Segle III aC. L’espai domèstic i l’organització de la societat a la protohistòria de la Mediterrània occidental (1er millenni aC): actes de la IV Reunió internacional d’Arqueologia de Calafell (Calafell-Tarragona, 6 al 9 de març de 2007). Barcelona, Universitat de Barcelona. Asensio D., Cela, X., Miró, C., Miró, M. T. and Revilla, E. 2009. El nucli ibèric de Montjuïc. Les sitges de Magòria o de Port. Barcelona. Quarhis: Quaderns d’Arqueologia i Història de la Ciutat de Barcelona 5: 1585. Belarte, M. C. 2013. El espacio doméstico y su lectura social en la Protohistoria de Cataluña (s.VII-II/I aC.). In S. Gutiérrez and I. Grau (eds) De la estructura doméstica al espacio social. Lecturas arqueológicas del uso social del espacio: 77-94. Alacant, Universidad de Alicante. Belarte, M. C., Malgosa, A., Noguera, J., Olmos, P. and Piga, G. 2013. Las necrópolis protohistóricas tumulares de Cataluña meridional: el ejemplo de Sebes (Flix, Tarragona). Trabajos de Prehistoria 70: 295-314. Bonet, H. 1995. El Tossal de Sant Miquel de Llíria : la antigua edeta y su territorio. Valencia, Diputación de Valencia. Bonet, H., Guérin, P. and Mata, C. 1994. Urbanisme i habitage ibérics al País Valenciá. Cota zero 10: 115130. Bonet, H. and Mata, C. 2009. El urbanismo ibérico en el área valenciana. El oppidum como centro de poder y reflejo del tejido social. Butlletí Arqueològic V: 107144. Bonet, H., Mata, C. and Moreno, A. 2007. Paisaje y hábitat rural en el territorio edetano durante el Ibérico Pleno (siglos IV-III a. C.). In A. Rodríguez Díaz and I. Pavón (eds) Arqueología de la tierra: paisajes rurales de la protohistoria peninsular: 247-276. Cáceres, Universidad de Extremadura. Bonet, H., Mata, C. and Moreno, A. 2008. Iron Age Landscape and Rural Habitat in the Edetan Territory, Iberia (4th-3rd centuries BC). Journal of Mediterranean Archaeology 21: 165-189. Bonet, H. and Vivez Ferrándiz, J. 2011a. De la fundación al abandono: Trayectoria histórica del poblado y de sus ocupantes. La Bastida de les Alcusses, 1928-2010: 239-255. Valencia, Museu de Prehistòria de València. Bonet, H. and Vivez Ferrándiz, J. 2011b. El poblado: Murallas, puertas y organización interna. La Bastida de les Alcusses, 1928-2010: 63-71. Valencia, Museu de Prehistòria de València. Bonet, H., Grau Mira, I. and Vivez Ferrándiz, J. 2015. Estructura social y poder en las comunidades ibéricas de la franja central mediterránea. In C. Belarte, D. Garcia and J. Sanmartí (eds) Les estructures socials protohistòriques a la Gàl·lia i a Ibèria. Homenatge 18

Natàlia Alonso and Guillem Pérez-Jordà: Elites and Farmers in Iberian Iron Age Cities

a Aurora Martín i Enriqueta Pons: 251-272. Barcelona, Universitat de Barcelona. Burch, J., Nolla, J. M. and Sagrera, J. 2010. Le système de stockage en silos sur le territoire ibérique aux environs d’Emporion. In H. Tréziny (ed.) Grecs et indigènes de la Catalogne à la mer Noire, Actes des rencontres du programme européen Ramses2 (2006-2008): 391-401. Paris, Ed. Errance, Centre Camille Jullian. Cabanes S., Hernández García, F. J. and Vizcaíno, D. 2010. El poblado fortificado de la Lloma Comuna. In D. Vizcaíno (ed.) La Lloma Comuna (Castellfort, Castellón). Un poblado fortificado de la Edad del Hierro: 21-50. València, Ein Mediterraneo. Cabanes S., and Vizcaíno, D. 2010. Complejos de hábitat y almacenaje en el Pla d’En Balaguer. In D. Vizcaíno (ed.) En Balaguer I (Portell de Morella, castellón). La evolución de una comunidad rural desde la Castelló Edad del Hierro hasta la Romanización. València: 29-46. Canal, D. and Buxó, R. 2002. L’explotació dels recursos vegetals: les anàlisis carpològiques. In E. Pons (ed.) Mas Castellar de Pontós (Alt Empordà). Un complex arqueològic d’època ibèrica (Excavacions 1990-1998) : 443-476. Museu d’Arqueologia de Catalunya, Girona. Casas, J. 2010. Prensas para la elaboración de aceite en el establecimiento rural Ibérico de Saus (Gerona). Notas sobre la explotación del campo en el territorio de Emporion. Archivo Español de Arqueología 83: 67-84. García Borja, P., Carrión Marco, Y., Guttierez Neira, C., Iborra, P., Lopéz, D., Miret, C., Montero, I., Pascual, J. L., Pérez-Jordà, G., Rovira, S., Valero, A. and Vivez Ferrándiz, J. 2013. Nuevas aportaciones al Horizonte del Bronce Final de La Vital (Gandía, València). Saguntum (PLAV) 45: 79-99. Gil Mascarell, M. 1976. Excavaciones en la Torre de Foios (Llucena, Castelló). Cuadernos de prehistoria y arqueología castellonenses 4: 305-313. Giró, P. 1960-1961. El poblado prerromano de ‘Mas Castellà’ (Monjos, Vilafranca del Penedés). Ampurias, XXII-XXIII: 159-182. Gómez Bellard, C., Guérin, P. and Pérez-Jordà, G. 1993. Témoignage d’une production de vin dans l’Espagne préromaine. In M. C. Amouretti and J. P. Brun (eds), La production du vin et l’ huile en Mediterranée, Actes du Symposium International, (Aix-en-Provence et Toulon, 2022 novembre 1991): 379-395. Athènes, Ecole Francaise d’Athènes. Grau, I. 2002. La organización del territorio en el área central de la Contestania Ibérica. Alicante, Universidad de Alicante. Guérin, P. 2003. El Castellet de Bernabé y el horizonte ibérico pleno edetano. Serie Trabajos Varios 101, Diputación de Valencia. Hillman, G. C. 1981. Reconstructing crop husbandry practices from charred remains of crops. In R. E. Mercer (ed.) Farming Practice in Prehistoric Britain: 123-162. Edinburgh, Edinburgh University Press. Hillman, G. C. 1984. Interpretation of archaeological plant remains: the application of ethnographic

models from Turkey. In W. C. Van Zeist and W. Casparie (eds) Plants and ancient man: studies in palaeoethnobotany: 1-41. Rotterdam, A.A. Balkema. Humphreys, S. C. 1978. Anthropology and the Greeks. London, Routledge. Junyent, E. 2015. L’evidència arqueològica en la definició de la societat estatal arcaica ilergeta. In C. Belarte, D. Garcia and J. Sanmartí (eds) Les estructures socials protohistòriques a la Gàl·lia i a Ibèria. Homenatge a Aurora Martín i Enriqueta Pons: 165-191. Barcelona, Universitat de Barcelona. Junyent, E. and López, J.B., 2016. La Fortalesa ibèrica dels Vilars d’Arbeca. Terra, aigua i poder. Catàleg de l’Exposició, Museu de Lleida, Universitat de Lleida. Le Meaux, H. and Sánchez de Prado, M. D. 2007. Le mobilier non céramique. In P. Rouillard, E. Gailledrat and F. Sala (eds), Fouilles de la Ràbita de Guardamar. L’établissement protohistorique de La Fonteta (fin VIIIefin VIe siècle av. J.-C.): 319-337. Madrid, Casa de Velázquez. Longepierre, S. 2014. Les moulins de Gaule méridionale (450-1 av. J.-C.): types, origines et fonctionnement. In N. Alonso (ed.) Molins i mòlta al Mediterrani occidental durant l’edat del ferro, Revista d’Arqueologia de Ponent 24: 289-309. López Reyes, D. 2008. Arqueobotànica de l’ordi vestit (Hordeum vulgare L.): el graner fortificat de Sant Esteve d’Olius (Olius, Solsonès) (s. III ane). Cypsela, 17: 201-218. López Reyes, D. (Coord.) 2015. Jaciment ibèric de la Font de la Canya (Avinyonet del Penedès). Un centre de mercaderies a la Cossetània ibèrica origen de la vinya. Sant Sadurní d’Anoia, Institut d’Estudis Penedesencs. López Reyes, D., Jornet, R., Morer, J. and Asensio, D. 2013. La viticultura prerromana al Penedès: indicadors arqueològics a Font de la Canya (Avinyonet del Penedès, Alt Penedès, Barcelona). In D. Sancho (ed.), El món de la viticultura, els vins, caves i aiguardents al Penedès i al Camp de Tarragona. Actes del V seminari d’Història del Penedès: 29-50. Institut d’Estudis Penedesencs. López Reyes, D., Valenzuela, S. and Sanmartí, J. 2011. Economia i canvi socio-cultural a Catalunya durant l’edat del ferro. In S. Valenzuela, N. Padrós, M. C. Belarte and J. Sanmartí (eds), Economia agropecuària i canvi social a partir de les restes bioarqueològiques. El primer mil·leni aC a la Mediterrània occidental. Actes de la V Reunió Internacional d’Arqueologia de Calafell (Calafell, 16-18 d’abril de 2009), Arqueomediterrània 12: 71-91. Barcelona, Universitat de Barcelona - ICAC. Martín, A., Casas, S., Codina, F., Margall, J. and De Prado, G. 2004. La zona 14 de l’oppidum del Puig de Sant Andreu d’Ullastret: l’hàbitat de tipus polinuclear. Cypsela 15: 265-284. Martínez Carmona, A. 2014. Una almazara ibérica en el yacimiento de la Illeta dels Banyets (El Campello, Alicante). In M. Olcina and J. A. Soler (eds), Arqueología en Alicante en la primera década del siglo 19

Country in the City XXI. II Jornadas de Arqueología y Patrimonio Alicantino: 247-253. Alicante, Marq. Mata, C., Moreno, A., Pérez-Jordà, G., Quixal, D., VivesFerrándiz, J. and Belarte-Franco, M. C. 2009. Casas y cosas del campo: hábitat agrícola y estructura social en los territorios de Edeta y Kelin (siglos V-III a.n.e). In M. C. Belarte (ed.), L’espai domèstic i l’organització de la societat a la protohistòria de la Mediterrània occidental (1er millenni AC): actes de la IV Reunió internacional d’Arqueologia de Calafell (Calafell-Tarragona, 6 al 9 de març de 2007): 143-152. Barcelona, Universitat de Barcelona - ICAC. Mata, C., Pérez-Jordà, G., Iborra, M. P. and Grau Almero, E. 1997. El Vino de Kelin. Una introducción a las prácticas agrícolas y ganaderas en la comarca de Requena-Utiel. Valencia, Universidad de Valencia. Meiksins Wood, E. 1988. Peasants-citizen and slave. The foundations of the Athenian democracy. London, Verso Books. Moreno, A. 2011. Cuando el paisaje se convierte en territorio: Aproximación al proceso de territorialización íbero en la Plana d’Utiel, València (ss. VI-II ane). Archaeopress, British Archaeological Reports International Series 2298. Olcina, M. H., Martínez Carmona, A. and Sala, F. 2009. La Illeta dels Banyets, (El Campello, Alicante): épocas ibérica y romana I, historia de la investigación y síntesis de las intervenciones recientes (2000-2003). Alicante, Museo Arqueológico de Alicante Osborne, R. 1991. Pride and prejudice, sense and subsistence: exchange and society in the Greek city. In J. Rich and A. Wallace Hadrill (eds), City and Country in the Ancient World: 119-146. London, Routledge. Peña-Chocarro, L., Pérez-Jordà, G., Morales, J. and Zapata, L. 2015. Storage in traditional farming communities of the western Mediterranean: Ethnographic, historical and archaeological data. Environmental Archaeology 20: 379-389. Pérez-Jordà, G. 2000. La conservación y la transformación de los productos agrícolas en el Mundo Ibérico. Saguntum Extra 3: 47-68. Pérez-Jordà, G. 2013. La agricultura en el País Valenciano entre el VI y el I milenio a.C. Valencia, Universitat de València. Pérez-Jordà, G. 2015. El cultivo de la vid y la producción de vino en la Península Ibérica durante el I milenio ANE. In R. Francia (ed.), Historia y arqueología en la cultura del vino: 47-56. Logroño, Instituto de Estudios Riojanos. Pérez-Jordà, G., Mata, C., Moreno, A. and Quixal, D. 2013. Stone wine presses and cellars in the Iberian Iron Age territory of Kelin (Utiel-Requena, València) (6th-2nd centuries BC). In A. Martínez Valle (ed.), Paisajes y patrimonio cultural del vino y otras bebidas psicotrópicas: 149-158. Requena, Ayuntamiento de Requena. Pérez-Jordà, G. and Peña-Chocarro, L. 2013. Agricultural production between the 6th and the 3rd millennium

cal BC in the central part of the Valencia region. In M. Groot, D. Lentjes and J. Zeiler (ed.), Barely surviving or more than enough?: 81-100. Leiden, Sidestone Press. Pérez-Jordà G., Peña-Chocarro L., García Fernández M. y Vera Rodríguez J. C., 2017. The beginnings of fruit tree cultivation in the Iberian Peninsula: plant remains from the city of Huelva (southern Spain), Vegetation History and Archaeobotany, 26, 5, 527-538. Pla, E. 1968. Instrumentos de trabajo ibéricos en la región de Valencia. Estudios de Economía Antigua de la Península Ibérica: 143-190. Pla, E. 1972. Aportaciones al conocimiento de la agricultura antigua en la región de Valencia. Rivista di Studi Liguri, XXXV, 319-354. Pons, E. 2002. Mas Castellar de Pontós (Alt Empordà). Un complex arqueològic d’època ibèrica (Excavacions 19901998). Girona, Museu d’Arqueologia de Catalunya. Prats, G. 2016. Caracterització socio-econòmica dels sistemes d’emmagatzematge de productes agrícoles al Mediterrani nord-occidental des del neolític fins a època ibèrica. PhDissertation, Universitat de Lleida. Rodríguez Díaz, A., Pavón, I., Duque, D., Domínguez García, A., Girón, M. and Criado, A. 2009. El asentamiento. El caserío de Cerro Manzanillo (Villar de Rena, (Badajoz) y la colonización agraria orientalizante en el Guadiana Medio. Mérida, Junta de Extremadura, Consejería de Cultura y Turismo. Rosser, P. and Fuentes, C. 2007. Tossal de les Basses. Seis mil años de historia de Alicante. Alicante, Ayuntamiento de Alicante. Ruiz Mata, D. 1995. El vino en época prerromana en Andalucía occidental. In S. Celestino (ed.), Arqueología del vino. Los orígenes del vino en Occidente: 157-212. Jerez, Consejo Regulador de las denominaciones de Origen Jerez-Xeres-Sherry y Manzanilla, Sanlucar de Barrameda. Ruiz Rodríguez, A. C. 2007. La tierra y los iberos del alto Guadalquivir. Arqueología de la tierra: paisajes rurales de la protohistoria peninsular : VI cursos de verano internacionales de la Universidad de Extremadura (Castuera, 5-8 de julio de 2005): 225-246. Cáceres, Universidad de Extremadura. Ruiz Rodríguez, A. and Molinos, M. 1998. The Archaeology of the Iberians. Cambridge, Cambridge University Press. Sanahuja, M. E. 1971. Instrumental de hierro agrícola e industrial de época íbero-romana en Cataluña. Pyrenae VII: 61-110. Sanmartí, J. 2004. From local groups to early states: the development of complexity in protohistoric Catalonia. Pyrenae 35: 7-41. Sanmartí, J. 2010. Demografia y cambio socio-cultural: el caso de la Iberia septentrional. Arqueología de la Población, Actas del VI Coloquio Internacional de Arqueología Especial, Arqueología Espacial 28: 91-108. Sanmartí, J. 2015. Long Term Social Change in Iron Age Northern Iberia (c. 700-200 BC). In B. Knapp and P. Van Dommelen (eds), The Cambridge Prehistory 20

Natàlia Alonso and Guillem Pérez-Jordà: Elites and Farmers in Iberian Iron Age Cities

of the Bronze and Iron Age Mediterranean: 454-487. Cambridge, Cambridge University Press. Sanmartí, J. and Belarte, M. C. 2013. La matérialisation du pouvoir dans l’organisation de l’espace : modèles hétérarchiques et modèles centralisés en Ibérie du Nord. In D. Garcia (ed.), L’habitat en Europe celtique et en Méditerranée préclassique: 97-125. Arles, Errance. Sanmartí, J., Plana, R. and Martín, A. 2015. Les estructures socials en els estats ibèrics de la costa de Catalunya. In C. Belarte, D. Garcia and J. Sanmartí (eds) Les estructures socials protohistòriques a la Gàl·lia i a Ibèria. Homenatge a Aurora Martín i Enriqueta Pons: 119-136. Barcelona, Universitat de Barcelona. Sigaut, F. 1988. A method for identifying grainstorage techniques and its application for European agricultural history. Tools and Tillage 6: 3-32. Stevens, C. J. 2003. An investigation of agricultural consumption and production models for prehistoric

and Roman Britain. Environmental Archaeology 8: 6176. Van der Veen, M. and Jones, G. 2006. A re-analysis of agricultural production and consumption: implications for understanding the British Iron Age. Vegetation History and Archaeobotany 15: 217-228. Van Dommelen, P., Gómez Bellard, C. and Pérez-Jordà, G. 2010. Produzione agraria nella Sardegna punica fra cereali e vino. L’Africa Romana XVIII: 1187-1202. Vives Ferrándiz, J. 2013. Del espacio doméstico a la estructura social en un oppidum ibérico. Reflexiones a partir de La Bastida de les Alcusses. In S. Gutiérrez Lloret and I. Grau Mira (eds), De la estructura doméstica al espacio social. Lecturas arqueológicas del uso social del espacio: 95-110. Alicante, Universitat d’Alacant.

21

Are There Farmers in Lattara (Lattes, France) during the Iron Age? Plant Resources Acquirement and Management between the 5th and the 1st Centuries BC Núria Rovira

ASM, Archéologie des Sociétés Méditerranéennes, UMR5140, Univ Paul-Valéry Montpellier, CNRS, MCC, F-34000 Montpellier, France

Natàlia Alonso

Grup d’Investigació Prehistòrica, 3DPatrimoni – INDEST (SGR2017-1714; HAR2016-78277-R) Dept. d’Història, Facultat de Lletres, Universitat de Lleida

1. Introduction

resources (wild and cultivated)? What are the imported plant products and from where? If we can certify local culture, where are the fields? What agricultural production activities can be attested in the city?

The ancient port city of Lattara (Lattes, Hérault, France) was founded toward 500 BC and abandoned around the 2nd c. AD (Py 2009; Lebeaupin 2014; Daveau and Py 2015). It was established directly on the edge of the stagnum latera (actual Méjean lagoon), between the two branches of the river Lez. This fluvial-lagoon environment characterised by both fresh and salt water, as well as wet and dry lands, dictated the wild types of fauna and flora the city exploited throughout its history (Figure 1).

To answer these questions we have taken into account several sources in addition to seeds and fruits studies: other archaeobotanical analysis (charcoal and pollen) and several archaeological evidences, such as agricultural implements, production structures and storage techniques. 2. Lattara and its surroundings: archaeological research

It is noteworthy that Mediterranean merchants (Etruscan, Greek or Iberians) and local authorities and people (probably depending on the oppidum of Sextantio, located 7 km to the north) founded this commercial enclave of Lattara in order to protect and regulate exchanges between the Mediterranean maritime space and the inland hinterland (Daveau and Py 2015). Is this main commercial activity meaning that all the inhabitants get their foodstuff from trading? Or are there producers (farmers) living in the city and supplying the residents?

Archaeological research at Lattara starts about the 1960s when H. Prades, a village schoolteacher, and some of his schoolchildren discovered a wealth of archaeological remains in particular at the place known as ‘Saint-Sauveur’ (Py 1988, 2009). H. Prades and the ‘Archaeological Group of Painlevé’ will then carry out almost two decades of archaeological work, especially inside the Iron Age fortified city but also in the Roman extra muros quarters and necropolis.

This article aims to characterise the acquirement, management and use of plant products by the Lattara’s residents during the second Iron Age (5th-1st c. BC) mainly based on the analyses of seeds and fruits collected during the archaeological work. These plant macro-remains are the most direct evidence of plant consumption, as well as of procurement practices (agricultural or any kind) developed.

M. Py, a researcher at the CNRS, is the first professional archaeologist to take over the excavations since the early 1980s what will set off systematic research conducted in collaboration with an international team of archaeologists constituted by, for instance, ceramics specialists, archaeobotanists, zooarchaeologists and geomorphologists. At the same time, since the end of the 1990s, archaeological survey operations have been conducted in the surroundings of Lattara (Figure 2). During the Iron Age, although the extension of the city’s hinterland has yet to be confirmed by research, some older and

Assuming the presence of both local and imported plant products, whether from the Mediterranean or the hinterland, four main issues will be addressed: what was actually the economic weight of the local 22

Núria Rovira and Natàlia Alonso: Are There Farmers in Lattara

Figure 1: Location of the ancient city port of Lattara (Lattes, Hérault, southern France) on the edge of the stagnum latera cited by Pliny the Elder in his Natural History (Reworked map from Jorda et al. 2008, Figure 3)

contemporaneous occupations are already documented (Daveau 2007; Py 2009; Newman and Silvéréano 2010; Daveau and Py 2015; Daveau et al. 2015).

Etruscan small bronze discs with pearl patterns around the rim (4th c. BC) and a ‘L-shaped’ building (probably ritual).

One of the earliest settlements known is ‘La Cougourlude’ (excavated in 2010 and directed by I. Daveau, Inrap), located one kilometre to the northeast of Lattara and dated from the 7th c. to 475 BC. As is the case for Lattara, foreign populations (firstly Etruscan and then Greek) seem to be strongly present in this village (Daveau and Py 2015; Daveau et al. 2015). Since the 6th c. BC ceramics show the importance of relations with Marseille. Ditches surrounded the village and half underground huts were made of wooden posts and of wattle-and-daub; pits and hearths were also documented. The residents of La Cougourlude progressively abandoned the village for the nearby-fortified city during the beginnings of the 5th c. BC.

The site of Port Ariane (excavated in 1999 and directed by I. Daveau, Inrap), located less than one kilometre to the north, presents archaeological levels dating from the Middle Neolithic to the Late Medieval period (Daveau 2007). The Iron Age levels are dated from the 7th to the 2nd c. BC and concern some ditches and many pits, especially for plants (grapevines). Finally, the excavations at La Céreirède (conducted in 2002 and directed by V. Bel, Inrap), located about three kilometres to the northeast, revealed archaeological remains dated from the Neolithic to modern times. The most remarkable discovery is a path around which burials dated from the 6th c. BC to the 3rd c. AD are placed.

As an extension to the south of this village and closer to Lattara we also find ritual and living areas on the place called ‘Mas de Causse’ (excavated in 2008 -directed by Ch. Newman, Oxford Archéologie Méditerranée- and 2010 -directed by I. Daveau, Inrap-). The site is globally dated from the 7th c. BC to the Roman times (Newman and Silvéréano 2010; Daveau and Py 2015). Among the most relevant evidences from the Iron Age we can note a ditch and several pits (7th-6th c. BC), a deposit of

3. Economic cultivated and wild plants: local and exotic resources Regarding seed and fruit analysis, we have at present 480 samples studied, representing 370 stratigraphic units and around 20,450 litres of sediment for the whole occupation layers of Lattara. A total number of 308,704 seed and fruit remains have been analysed. 23

Country in the City

Figure 2: Location of Lattara and the main archaeological sites found in its surroundings for the Iron Age period (Reworked map from Daveau et al. 2015: 110, Figure 20)

If we have a detailed view to the distribution of the archaeobotanical analysis, we can see that several sectors of the city with different chronologies (some contemporary) are represented. However, the beststudied areas to date are chronologically the earliest (Figure 3) and will be subject to a more detailed analysis in this work.

concerned: cereals, pulses, cultivated and wild fruit, together with some oleaginous and spices (Figure 4). Different preservation conditions (charring, waterlogging and mineralising), as well as the recovery of concentrations, can partly explain the presence/ absence and abundance of certain taxa. In general, most of the seed and fruit remains are charred, excepting for the 5th c. BC levels of the Quarter 1 (we find the three preservation conditions) and the 1st c. BC - 2nd c. AD samples coming mainly from wells. Concentrations are also quite common and concern mainly cereals or pulses for the earliest levels and grapes for the recent ones.

A great variety of cultivated and wild plants are identified at Lattara (Buxó 1992, 2003, 2005; Rovira and Chabal 2008; Alonso et al. 2008; Alonso and Rovira 2010, 2014, 2016; Rovira and Alonso 2010, 2018). We will only present here those having an economic role particularly related to consumption. Many plant groups are so 24

Núria Rovira and Natàlia Alonso: Are There Farmers in Lattara

Figure 3: Location and dating of the Quarters (‘îlot’) and wells (‘puits’) of Lattara with seeds and fruit analysis. Restitution extracted from the website ‘Lattes en Languedoc: Les Gaulois du Sud’ (www.lattara.culture.fr). Infographic: interactive LaForme / I. and S. Mottola Valletta under the direction of M. Py and T. Janin © Ministry of Culture and Communication Figure 4: Summary of the main economic plants identified in Lattara during the different centuries of occupation based on the absolute number of remains

Cereals Avena cf. sativa Hordeum vulgare Hordeum vulgare var nudum Hordeum sp. Triticum aestivum/durum/turgidum Triticum a/d/t type compactum Triticum dicoccum Triticum monococcum Triticum sp. Hordeum/Triticum Panicum miliaceum Panicum/Setaria Setaria italica Secale cereale

seed seed seed seed seed seed seed seed seed seed seed seed seed seed

5th c. BC 159 6731,3 215721 146232 32 89487 29 1 25388 622 4108 32 2436 21679 1895 81 419 23

Chaff cereals Hordeum (vulgare + sp) Hordeum/Triticum Secale cereale Triticum (a/d/t) Triticum (dicoccum + sp) Cerealia

varia varia varia varia varia varia

6353 1010 1127 2 977 3025 212

793 9 4

Pulses Lathyrus cicera Lathyrus sativus Lathyrus sp. Lens culinaris Medicago sativa Pisum sativum Pisum/Lathyrus Pisum/Lens Vicia ervilia Vicia faba Vicia sativa Vicia/Lathyrus Vicia/Lens Vicia/Pisum Fabaceae

seed seed seed seed seed seed seed seed seed seed seed seed seed seed seed

1792 7 105 27 373 2 744 4 1 4 87 23 18 2 12 383

332 17 48 7 67 1 76 3 3 37 16 1 15

36 1

5 36

2 1

Number of SU Number of litres Number of remains (NR)

4th c. BC 94 4971,8 22190 16797 64 8243 51 1 3153 221 438 11 275 3980 270 3 87

3rd c. BC 48 3499 2062 843

2nd c. BC 30 1016 1180 90

1st c. BC 19 1540 58895 66

1st c. AD 19 2211 8495 1106

2nd c. AD 1 480 161 154

510

29

44

517

32

152 9 3 1

42

18

509

22

4

67

2

146 4

19

13

98

18

15

16 760 4

Hulled barley + barley Barley/wheat Rye Naked wheat Emmer + wheat Cereals

15

3 1

2 16

1

1

38

5

22 3 7

5

5

2

2 6

1

1

1 1 1

25

2

Oat Hulled barley Naked barley Barley Naked wheat Naked wheat Emmer Einkorn Wheat Barley/wheat Common millet Millet Foxtail millet Rye

Red pea Grasspea Grasspea Lentil Alfalfa Pea Pea/grasspea Pea/lentil Bitter vetch Broad bean Common vetch Vetch/grasspea Vetch/lentil Vetch/pea Pulses

Country in the City

Number of SU Number of litres Number of remains (NR) Cultivated fruit Ficus carica Juglans regia Juglans cf. regia Olea europaea Phoenix dactylifera Pinus pinea Prunus avium Prunus avium/cerasus Prunus cerasus Prunus domestica Prunus dulcis Prunus persica Vitis vinifera

Cultivated oleaginous/spices Allium sativum Coriandrum sativum Linum usitatissimum Wild edible fruit Arbutus unedo Celtis sp. Cornus mas Corylus avellana Pistacia lentiscus Prunus spinosa Prunus sp. Pyrus communis/pyraster Quercus ilex/coccifera Quercus sp. Quercus sp. Rubus caesius Rubus fruticosus Rubus idaeus Rubus sp. Sambucus nigra Sorbus sp. Sorbus/Malus

seed shell fruit fruit kernel fruit fruit stone stone stone stone stone stone stone fruit pedicel pip

clove seed seed

fruit seed stone shell seed stone stone seed cupule cupule fruit seed seed seed seed seed seed seed

5th c. BC 159 6731,3 215721 18985 1456

4th c. BC 94 4971,8 22190 926 9

47

1 5

4

3rd c. BC 48 3499 2062 1020

2nd c. BC 30 1016 1180 1029

1st c. BC 19 1540 58895 58819

2

1st c. AD 19 2211 8495 1217 24 187

2nd c. AD 1 480 161 2

66 8 5 228 377 16

Palm date Pine nut

Sweet cherry Cherry Sour cherry Plum Almond Peach

2 19 51

1 2

35 5 410 17060 35 18 17 345 1

45 864

11 1007

5 1024

44 7 58768

2 199

2

1

1 1

1

19

3

1

149 114

3 63 1 4 1 1

6

1

7

1

1

26

1 4 3 1 218 1 3 40

5

1 1

1 3 1 2

1

1

Other wild plants

varia

39292

3756

158

54

8

5957

Undetermined

varia

5014

220

2

3

1

12

Cereal grains are the most usual remains of this group even so chaff remains are also abundant in the 5th and 4th c. BC levels (Figure 5). Hulled barley (Hordeum vulgare) and naked wheat (Triticum aestivum/durum/ turgidum), followed by emmer wheat (Triticum dicoccum) and millets (Panicum miliaceum and Setaria italica) are the major cereal crops. It is also to note the presence of probably cultivated oat (Avena cf. sativa), einkorn (Triticum monococcum) and rye (Secale cereale). During the earliest periods, we still find some naked barley (H. vulgare var. nudum) but it tends to disappear in the current 4th c. BC. Emmer spikelet forks and glume basis, together with barley and naked wheat rachis and node segments, internodes and glumes, mainly compose chaff remains. The abundance of chaff remains in several samples will be evidence of particular agricultural/ consumption practices that will be discussed later on.

Fig Walnut Walnut Olive

Grape 2

Garlic Coriander Flax Strawberry tree Mediterranean hackberry Cornelian cherry Hazelnut Mastic tree Sloe Sloes Pear Evergreen oak Oak Acorn European dewberry Blackberry Raspberry Berry Elderberry Rowan Rowan/apple

Among pulses, lentils (Lens culinaris), peas (Pisum sativum) and grass peas (Lathyrus sativus) tend to be predominant, followed by broad beans (Vicia faba) and bitter vetch (Vicia ervilia). Few findings of red peas (Lathyrus cicera), alfalfa (Medicago cf. sativa) and common vetch (Vicia sativa) show the quite great variety of this group. Compared to cereals, they usually present a lesser number of remains during the whole periods. Cultivated fruit are undeniably better represented at Lattara than wild ones. Grapes (Vitis vinifera) are the predominant taxa during the whole sequence, followed by figs (Ficus carica) and in a lesser extend olives (Olea europaea) for the earliest levels. Fruit remains are very scarce (except for grape pips) until the 1st c. AD when the Romans introduce a great variety of species, some exotic, such as peaches (Prunus persica), walnuts (Juglans 26

Núria Rovira and Natàlia Alonso: Are There Farmers in Lattara

Figure 5: Evolution of the different categories of economic plants (cereals, pulses, cultivated fruit and other plants) attested in Lattara during the Iron Age phases based on the absolute number of remains

regia), cherries (Prunus avium/cerasus), plums (Prunus domestica), pine nuts (Pinus pinea) and dates (Phoenix dactylifera). It is also noteworthy the presence of almonds (Prunus dulcis) during the 5th c. BC maybe as an exotic product coming from the eastern Mediterranean or Italy.

or even partly in the water, which would explain the waterlogging preservation of plant remains in the lower levels of several zones of the site such as Quarter 1. The lagoon is progressively filled with fluvial sediments and around the 3rd-2nd c. BC the occupation surface will probably extend to the south, to the new delta, with the construction of a port quarter (with a quay, a lighthouse and several warehouses) in front of the city walls at least at the end of the 1st c. BC (Garcia and Vallet 2002; Py 2009).

Hazelnuts (Corylus avellana), blackberries (Rubus fruticosus) and elderberries (Sambucus nigra) are the most abundant wild edible fruits for the earliest periods, while fruit stones of Celtis sp. (probably the Mediterranean hackberry) and Prunus spinosa (sloes) are the best found at the end of the sequence. A few remains also show the casual consumption of acorns (Quercus sp.), pears (Pyrus communis/pyraster), raspberries (Rubus idaeus) and rowanberries or apples (Sorbus/Malus).

A fertile delta floodplain, with deep and rich soils, surrounds the city to the west and north. Different limestone massifs certainly covered by oaks’ forests encircle this area. The lagoon is located to the south and small pebble hills (10-20 m above sea level) close the eastern side of the city. Landscape is therefore varied and easily allows agriculture.

Finally, scarce evidence of linseed (Linum usitatissimum), coriander (Coriandrum sativum) and garlic (Allium sativum), show however the use of other edible seeds/ bulbs as oil sources or spices.

The urban frame of the city, that covers around 3.5 ha, also presents differences between the foundation phases and the successive Iron Age periods (Py 1990, 1996, 1999, 2009; Garcia 1994; Janin 2010; Lebeaupin 2014; Gailledrat and Vacheret 2014). The earliest occupation levels of Lattara (5th c. BC) are at present known in particular through two search windows, Quarters 27 and 1, located against the southern and the northeaster walls respectively. The first one, called the ‘Etruscan Quarter’ because of the predominance of this type of ceramics, presents three houses partially excavated as well as a narrow alley against the city wall. In turn, zone 1, presents a different urban frame, looser, with the particular presence of a house in apse and areas dedicated to animals, which however will quickly evolve into the construction of an orthonormal quarter with quadrangular houses and exterior spaces for living, breeding and circulation (courtyards, streets, alleys...). The building in apse is similar to those found

4. Urban frame and surrounding landscape: where are the fields? As we have already said above, the city of Lattara was established directly on the edge of a lagoon, on a small peninsula (delta) situated between the two branches of a Mediterranean coastal river (Figure 6). Geomorphology and malacology studies highlight four phases of morphogenesis previous to the foundation of the Iron Age city, with on the base lagoon vases dated to the Final Bronze Age, followed by fluvial sands leading to the landing of the area, and finally a thin soil formation before the first anthropogenic deposits attributable to the foundation of the city (Jorda et al. 2008; Bagan et al. 2010). The city walls were directly built on the lagoon shore, probably next to the water 27

Country in the City

Figure 6: Lattara is situated in a fluvial-lagoon environment between two branches of the coastal river Lez (Reworked map from Bagan et al. 2010, Figure 1)

at the closer site of La Monédière (Bessan) and supposed to have been built following a Greek schema (Nickels 1976; Gailledrat 2014).

that cultivated fields were located intra muros. An undetermined number of animals seem to have been kept stabled in the city at least during the 5th c. BC and they have also walked through the streets at least until the 4th c. BC as shown by numerous hoof prints found in Quarter 1 (Gailledrat and Vacheret 2014) or the presence of bedding residues and coprolites in Square 123 (Cammas 2003; Buxó 2003).

It is especially from the end of the 5th c. BC onwards that the urban structure becomes tighter and follows different axes of streets and squares organising the circulation into the city walls (Py 2009). Some houses still have courtyards, usually smaller, and the ones in the central area generally have few rooms (from 1 to 3 for a single house). Several bigger houses (with 4 or more rooms and a central courtyard) are instead located against the city walls or over the eastern part of the city.

The close city surroundings were strongly disturbed by human activities (related for instance to trade, fishing or animal husbandry). Pollen analyses show two wasteland areas next to the city to the east and west, as well as cereal fields about 1-2 km to the north and a vineyard about 5 km to the west (Puertas 1998). Another vineyard has been found at the site of Port Ariane, situated less than 1 km to the north thus close to

So, the space inside the city walls is quite dense at least from the 4th c. BC onwards. We do not think 28

Núria Rovira and Natàlia Alonso: Are There Farmers in Lattara

Figure 7: Archaeological traces of the vineyard of Port Ariane (Lattes, Hérault, France) dated to the 3rd-2nd c. BC (Picture: C. Jung, Inrap Méditerranée)

the previous cereal fields (Daveau 2007). The discovery consists on soil traces (plantation pits) occupying more than 2.4 ha and dated to the 3rd-2nd c. BC (Figure 7).

were similarly grown away from regular irrigation in intensively managed plots, judging by the evidence for their watering regime and manuring (Alagich 2015; Alagich et al. 2018).

Seed and fruit analyses also give us information, particularly for the 5th-4th c. BC, about the existence of at least two fields’ locations and several ways of acquiring the crop products (Alonso and Rovira 2010). Charred seeds of wild plants mixed with cereals (weeds) show different environments and many possibilities of plant transport (Figure 8): mostly dry lands, situated towards the north, but also wet lands situated probably around the city (about 0-2 km) and towards the east and west (along the river shores). The first spaces are characterised by evergreen oak’s forests (Quercus coccifera) with strawberry trees (Arbutus unedo), heathers (Erica) and Phillyrea, together with open areas represented by fallow land, pathways, wasteland and dry meadows. The second ones can be related both to fresh waters, with the presence of wet meadows, ash tree (Fraxinus) and elm’s (Ulmus) forests, and to brackish waters related to reed beds (Phragmites, Carex…) and salt meadows with glasswort (Salicornia), ditch-grass (Ruppia maritima) or saltmarsh bulrush (Bolboschoenus maritimus). Charcoal analyses also confirm the abundant presence and use for fuel and construction of the tree taxa just mentioned (Chabal 2005, 2014).

5. Agricultural tools and spaces in the city In general, very few tools related exclusively to agricultural practices have been found until today at Lattara (Figure 9). This is not surprising if we take into account the recycling of these items, especially those made of metal, or the unusual preservation of those made of wood or other perishable materials in dry archaeological contexts (Py 2016: 315). Concerning labour practices, two digging sticks or picks have been found, one made of wood and the other one of bone (-450/-425, Quarter 27). We also know two pickaxes in iron, almost complete, but they are much more recent (-50/-25 and -50/50, Quarter 5); ten other fragments have been found during the whole sequence, slightly concentrate however in the 3rd and 2nd c. BC (Figure 10). The last tools used for labouring the fields found in the site are a possible plough beam made of wood (-500/-400, Quarter 1) and a ploughshare in iron coming from the same quarter and dated to -425/-400 (Figure 11). Tools related to harvesting activities are a little more abundant. We know around twenty-five iron billhooks or sickles, also found during the whole sequence but especially between the 4th and the 2nd c. BC (Figure 10). Both implements are difficult to distinguish on the basis of fragments, but some restorations seem to demonstrate the presence of two different types

Finally, recent carbon and nitrogen stable isotope analyses of charred seeds (cereals and pulses) from Quarters 1 and 27 also reveal two main areas of cultivation around Lattara. Emmer and naked wheat are likely to have grown near the river or in floodplains, but barley would have been grown under much drier or saline conditions further from the river. Pulses 29

Country in the City (Figure 10, see the small inset top right and notice the three tools left which seem different from the others on the right). This question is crucial because billhooks are related to grape harvesting and sickles to cereals, so we would have two different practices concerning two different crops. We have not already mentioned iron or bronze knives, as well as flint blades (some of them used for sickles), potentially used for harvesting too although they might be considered as multi-purposes objects. Unlike bronze blades (it is not sure that all of them belong to knives), iron knives are quite common at Lattara, especially during the 4th, 2nd and 1st c. BC (Figure 9), in quite a lot of quarters, but we cannot associate them directly to harvest. Flint blades and flakes are particularly abundant during the 5th c. BC and regularly found at all times, but there is neither real evidence of their use for agricultural or plant processing purposes. Finally, only a basket maybe used for winnowing (-450/-425, Quarter 27) provides direct information on activities after harvest and before storage and/or consumption (Figure 12). It is not surprising because all these activities (threshing, raking, winnowing, coarse sieving) are carried out in or near the fields, in the threshing floors, although some of the tools could have been kept at home and then found in the city. One of the single open spaces explored for the Iron Age, square 123, show the presence of animals and abundant vegetal litter during the 4th c. BC but very few chaff remains: 26 emmer spikelet forks per 7,806 seed/fruit remains (Buxó 2003).

Figure 8: Summary of the possible plant transport ways to the city of Lattara (Rovira and Alonso 2010)

Figure 9: Chronological distribution of the main categories of tools associated (or that can be associated) with agricultural activities according to the number of items

30

Núria Rovira and Natàlia Alonso: Are There Farmers in Lattara

Figure 10: Spatial distribution of the billhooks/sickles and pickaxes found in Lattara during the whole occupation phases. The small inset top right present some exemples of these tools (Py 2009: 229)

Figure 11: Possible plough beam made of wood, Quarter 1, -475/-450 (Picture: Equipe de Fouilles de Lattara)

31

Country in the City

Figure 12: Burnt basketry possibly for winnowing, Quarter 27, -450/-425 (Picture: Equipe de Fouilles de Lattara)

6. Storage and production structures and spaces

but they increase progressively their volume to reach around 900 litres in the Roman period. During the Iron Age they are usually found in the living spaces (kitchen) or in specific storage rooms.

Storage represented at Lattara is predominantly domestic, based on pits and containers made of different materials. At present, public or communal buildings or spaces are scarce for the Iron Age phases. The only exception could be some rooms identified as granaries or cellars, specialized on grain storage, and dated to the 3rd-2nd c. BC (Figure 13). They are generally situated along or at the edge of two main streets of the city (Py 2009: 216). From the 1st c. BC onwards we also find warehouses in the port but they are especially dedicated to trade purposes.

The majority of the amphorae found in the site during the Iron Age comes from abroad, the only local productions (Gauloise-4) beginning from the 1st c. AD onwards (Py 1993). Briefly and depending on the period, Etruscan (or Italian in general), Greek (mostly from Massalia) and Iberian amphorae originally carried out mostly wine, while Phoenician ones are supposed to transport olive oil or fish products. However, once their primary role accomplished they are sometimes used in household contexts to store cereals or pulses, as several levels of destruction due to fire have highlighted for instance in Quarter 27 (Figure 15).

Storage pits (silos) are not used because of the soil moisture conditions. Containers are in turn very common. Mud and ceramic containers are the best known although some baskets and bags, as well as wooden boxes, have also been found either burnt or waterlogged. Mud containers are especially abundant during the 5th c. BC, decreases on the following century and disappear during the 3rd c. BC. Some archaeobotanical findings show that they contained cereals, pulses and even grape pips (Figure 14).

Direct or indirect evidences of processing practices of plant products before consumption are quite abundant and reveal everyday actions. The usual presence of chaff show that hulled cereals, especially emmer wheat and barley, can be cleaned inside the site in particular for the earliest periods. Weeds often accompany chaff and grain reflecting by-products of fine sieving and hand picking (Figure 16). A little amount of crushed grains of emmer wheat, dated to the 5th c. BC, enable to identify the practice of dehusking to free grains, using a mortar and a pestle or grinding stones (Figure 17). Despite this, it is clear through the example shown in figure 16 that the majority of archaeobotanical samples studied in Lattara usually present cleaned grain.

Amongst storage ceramic containers we especially find pithoi or dolia and amphorae. These first large containers can nevertheless have different dimensions based on the content and function (domestic or not). At Lattara they are supposed to store mainly wine, but the theory of cereals is also admitted (Py 2009). The studies concerning these containers show their appearance around -475 and a peak over the 2nd c. BC explained by the rise of local viticulture (Py 2009; Curé 2014). At the beginning they are quite small (50 to 200 litres)

Concerning processing/culinary practices about cereals, grinding stones are quite abundant in the 32

Núria Rovira and Natàlia Alonso: Are There Farmers in Lattara

Figure 13: Cellar or granary showing several pits for storage vases, Quarter 31, -225/-200 (Picture: Equipe de Fouilles de Lattara)

Figure 14: Burnt mud container with grape pips, Quarter 27, -450/-425 (Picture: Equipe de Fouilles de Lattara)

found in Quarter 5 during the Roman period and has been interpreted as evidence of public bakeries, perhaps existing in Lattara since the 3rd c. BC, if one believes the discovery of three lined up bread ovens in Quarter 4-south. Despite these discoveries coming out of household contexts, a real commercial function of these bakeries is difficult to prove before the 1st c. BC. The only confirmed evidence for public bread production and consumption comes from a building interpreted as a tavern, dated to the Republican period, presenting three grinding structures associated to a three ovens in the same room, to an annexed dining

city. There are several types, mostly back-and-forth, Olynthus and rotary querns (appearing about the beginning of the 4th c. BC), principally made of basalt or sandstone, and almost all of them for individual and domestic use (Figure 18). Importation of finished querns is the norm, in particular from Italy during the 5th c. BC and from the nearby region of Cap d’Agde and the Hérault valley during the whole occupation sequence. However, a local production in the city has also been highlighted, especially around Quarters 1 and 2 from the 4th to the 2nd c. BC (Py 2009). A single fragment of what looks like a Pompeian mill has been 33

Country in the City

Figure 15: Basis of an Etruscan amphora containing cereals (barley) burnt in a fire, Quarter 27, -500/-475 (Picture: Equipe de Fouilles de Lattara)

Figure 16: Distribution of several archaeobotanical samples from Quarters 1 and 27 (5th c. BC) concerning their composition of grain, chaff and weeds (Alonso and Rovira 2010)

34

Núria Rovira and Natàlia Alonso: Are There Farmers in Lattara

room and to different outdoor spaces (Luley and Piquès 2016). As regards to pulses or fruit, few things can be said about processing practices. Only the presence of wine making waste has been identified (Buxó 1992; Alonso and Rovira 2010). The earliest evidence (-450/-425) comes from an assemblage of more than 9,000 grape pips stored in a small mud container probably for later use (to make flour or oil?) (see Figure 14). We can identify these remains as the category A3 given by E. Margaritis and M. Jones (2006): the by-products

Figure 17: Detail of several crushed grains of emmer showing the practice of dehusking (Picture: SRI Universitat de Lleida, Alonso and Rovira 2010)

Figure 18: Examples of grinding stones from the Iron Age: back-and-forth (top) and rotary (bottom) (Pictures: Equipe de Fouilles de Lattara/Py 2009: 217)

35

Country in the City

Figure 19: Location of the wine pressing platform in the courtyard of a house of Quarter 27, -425/-400 (Picture: Equipe de Fouilles de Lattara)

of red/white wine or white must, after the cleaning of the vat or the pithos/dolium used for catching the juice and/or for fermentation. The other archaeobotanical examples of wine making are found from the 3rd c. BC onwards, being then Vitis vinifera the predominant taxa in the site even on cereals (see Figure 5). Occasionally, many thousands of pips, together with stalks, pedicels and sometimes skins form what we can thenceforth consider as grape-pressing waste.

Concerning the origin of the plant resources attested in Lattara during the Iron Age, they can all be grown locally in a temperate Mediterranean climate. However, some taxa such as almonds, coriander or garlic, maybe olives, were possibly imported from abroad. We are unable to determine yet if these plant products (in particular the three first) correspond to luxury or prestige food as defined by M. van der Veen (2003), but their rarity leads us to believe that they were exotic and probably meant for a minority. On the other hand, apart from these few examples, we have also noted that a part of the foodstuffs of plant origin, in particular cereals and grapevines were cultivated in the immediate vicinity of the city, as certain weeds assemblages, pollen analysis and agrarian traces have shown. Moreover, concerning cereals, the existence of different wild plant communities growing mainly in two separate environments corresponding to wet and dry fields, both found in charred and waterlogged assemblages, is indeed the evidence for a wide agricultural exploitation including various areas more or less close to the city.

Additionally, wine making is not much attested at Lattara through production structures or spaces. The only element in place is a pressing platform made of stone located in the courtyard of a 5th c. BC house of Quarter 27 (Lebeaupin and Séjalon 2010) (Figure 19). Chemical analyses of residues show the presence of tartaric acid/tartrate proving its use for wine making (McGovern et al. 2013). Five other fractioned elements of wine/oil presses have been found reemployed in houses’ walls mostly dated to the 3rd-2nd c. BC (Garcia 1992) (Figure 20).

It is precisely the presence of weeds associated with wet and saline environments (mainly reed beds and salt marshes), as well as the results of stable isotope analyses (Alagich 2015; Alagich et al. 2018), which allows us to consider the existence of crop fields around the city and thus the presence of farmers in town. To this is added the discovery of agricultural tools linked particularly with the grain or the grape harvest, especially during the 3rd and 2nd c. BC. These tools are concentrated in the central quarters of the city where we also find the smallest houses. We can then consider the presence of at least two groups of people living in Lattara: traders in larger houses located against the wall and farmers at the centre.

7. Farming at Lattara during the Iron Age As we have seen in the previous sections, plant resources acquirement and management between the 5th and the 1st c. BC in Lattara are currently quite well known, especially for the earliest Iron Age periods. Cultivated resources are predominant while gathering of wild plants seems minority. Cereals (hulled barley, emmer, naked wheat and common millet), pulses (lentils, peas, grass peas, as well as broad beans and bitter wetches) and grapes (maybe also figs) seem to be the main staple crops of the Lattara’s residents. The question is now to know if farmers living in the city produced all or a part of them. 36

Núria Rovira and Natàlia Alonso: Are There Farmers in Lattara

Figure 20: Spatial distribution of the few elements of wine/oil presses found in Lattara during the whole occupation levels. The small inset top right present a detail of the wine-pressing platform of Quarter 27 (Picture: Equipe de Fouilles de Lattara)

but not oil production. At the same time, pollen and anthracological analyses do not allow to affirm the presence of trees on site (Puertas 1998; Chabal 2005, 2014).

Most of the archaeobotanical or archaeological remains concerning processing practices are found in the courtyards or outdoor spaces. They mostly represent cleaning (fine sieving and hand picking) and consumption waste of plant products because most agricultural practices after harvest were certainly near fields outside the city. At present, very few production spaces have been found in the city or its surroundings for the Iron Age periods. The only notable example is the wine pressing platform found during the 5th c. BC in Quarter 27, although some presses fragments were also found during the 3rd and 2nd centuries BC testifying to this activity somewhere in or around the city. According to this, M. Py and R. Buxó (2001) have speculated that the production of wine, booming business, as many evidence of archaeobotanical remains of grapes show, moves from the city to its surroundings from the 3rd century. This is evidenced in particular by a decrease in the presence of dolia in town.

We have focused here on plant data but we cannot forget that farmers also breed animals. We have mentioned in a previous section that many archaeological evidences (hoof prints, pens or barns) testify the presence and circulation of livestock in the city at least during the earliest periods (Figure 21). Archaeozoological analyses show that cattle, goats, sheep and pigs are the basic livestock, together with barnyard animals (Colomer and Gardeisen 1992; Gardeisen 1999, 2003). Stable isotope analyses of the animal bone collagen from samples of Quarters 1 and 27 (5th c. BC) show that especially cattle would have consumed plants or crop by-products (after harvest and during fallow) coming from both wet and dry/saline meadows, while ovicaprids’ diet was more varied implying a larger number of more broadly distributed flocks (Alagich 2015; Alagich et al. 2018). So, all these data allow not only to affirm the existence of farmers in the city, but also to better understand the use of animals in agriculture.

Concerning other current Iron Age plant products, such as olive oil, we do not have any evidence at Lattara. Olive stones are not very abundant and they are mostly whole (Buxó 1992, 1999, 2003; Alonso and Rovira 2010). We can so only confirm the consumption of olives 37

Country in the City

Figure 21: Detail of hoof prints (several types of animals) between an alley (above) and a house (below), Quarter 1, -475/-450 (Picture: Equipe de Fouilles de Lattara)

Archaeological research and rescue surveys have not yet found farms or other production areas or structures in the Lattara surroundings for the Iron Age. Considering this and all that has been discussed above, we can say that the city of Lattara housed a population of farmers at least during the earliest periods of the Iron Age and probably until the arrival of the Romans. These farmers lived together with artisans and traders in a rich and diverse economic dynamics. It is still early to fully identify all the features of this phenomenon and its evolution over a vast period, given the cognitive limitations of the scope of the archaeological excavations in the site. However, current available data allow starting to have a good preview of this question on the presence of farmers in the proto-historic cities of the north-western Mediterranean. For now, we do not have other urban sites with similar archaeological, archaeobotanical and archaeozoological studies in order to compare the ideas and hypotheses set out in this work. We trust that new systematic and interdisciplinary studies will be carried out in order to better know the economic and social dynamics of Iron Age big cities in the Mediterranean basin.

partial fulfilment of the requirements for the degree of Master of Science in Archaeological Science, University of Oxford. September 2015. Alagich, R., Gardeisen, A., Alonso, N., Rovira, N., Bogaard, A. 2018. Using stable isotopes and functional weed ecology to explore social differences in early urban contexts : The case of Lattara in mediterranean France. Journal of Archaeological Science 93 : 135-149 Alonso, N. and Rovira, N. 2010. Consommation et traitement des produits végétaux à Lattara entre 475 et -350. In Janin T. (ed.) Premières données sur le cinquième siècle avant notre ère dans la ville de Lattara. Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes  : Lattara 21 : 329-388 Alonso, N. and Rovira, N. 2014. Stockage de céréales et consommation de végétaux divers attestés dans les maisons du quartier étrusque de Lattara (Lattes, Hérault) au début du V siècle av. n. è. In Lebeaupin, D. (ed.) Les origines de Lattara et la présence étrusque. Les données de la zone 27. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes: Lattara 22: 183-200 Alonso, N. and Rovira, N. 2016. Plant uses and storage in the 5th century BC Etruscan Quarter of the city of Lattara, France. Vegetation History and Archaeobotany 25: 323. Alonso, N., Buxó, R. and Rovira, N. 2008. Archéobotanique des semences et des fruits de Lattara: bilan des recherches. In Janin, T. and Py, M. (ed.) La ville portuaire de Lattara (Lattes, Hérault) et son territoire: nouveaux acquis, nouvelles questions. Gallia 65 : 193-200 Bagan, G., Gailledrat, E. and Jorda C. 2010. Approche historique de la géographie des comptoirs littoraux à l’âge du Fer en Méditerranée occidentale à travers

Acknowlegements We thank Isabelle Daveau (Inrap Méditerranée), Christophe Jorda (Inrap Méditerranée) and Michel Py (CNRS) for kindly providing us maps and some pictures that we have used and reworked in this article. References Alagich, R. 2015. A Diachronic Investigation of Animal Husbandry and Farming Practices at Iron Age Lattara Using Stable Isotopes. Dissertation submitted in 38

Núria Rovira and Natàlia Alonso: Are There Farmers in Lattara

l’exemple du port de Lattara (Lattes, Hérault). Quaternaire 21 (1) : 85-100. Buxó, R. 1992. Cueillette et agriculture à Lattes: les ressources végétales d’après les semences et les fruits. In Py, M. (ed.) Recherches sur l’économie vivrière des Lattarenses. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 5: 45-90. Buxó, R. 2003. Etude carpologique de la place 123: problématique de la presence de restes de semences et de fruits dans un espace urbain non construit. In Buxó R., Chabal L. and Gardeisen A. (ed.) La place 123 de Lattara, recherches pluridisciplinaires sur un espace urbain du IVe siècle avant notre ère. Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes : Lattara 16:193-217 Buxó, R. 1999. Première approche des plantes exploitées au IVe siècle avant notre ère à Lattes. In Py, M. (ed.) 1999. Recherches sur le quatrième siècle avant notre ère à Lattes. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 12 : 525-535 Buxó, R. 2005. Etude carpologique des puits de Lattes: evaluation et comparaison avec l’habitat. In Piquès, G. and Buxó, R. (ed.) Onze puits gallo-romains de Lattara (Ier s. av. n. è. - IIe s. de n. è.), Fouilles programmées 19862000. Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes  : Lattara 18: 199-219 Cammas C. 2003. Microstratigraphie de la place 123 (secteurs 2 et 3) : activités et rythmes de sédimentation au IVe s. In Buxó R., Chabal L. and Gardeisen A. (ed.) La place 123 de Lattara, recherches pluridisciplinaires sur un espace urbain du IVe siècle avant notre ère. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 16: 67-86. Chabal, L. 2005. Charbons de bois et bois gorgé d’eau : des jardins de Lattara aux forêts du delta du Lez. In Piques, G. and Buxó, R. (ed.) Onze puits galloromains de Lattara (Ier s. av. n. è. - IIe s. de n. è.), Fouilles programmées 1986-2000. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 18: 221-234 Chabal, L. 2014. Le bois dans l’architecture et le mobilier (500-475 av. notre ère) : les boisements aux origines de Lattara. In Lebeaupin, D. (ed.) Les origines de Lattara et la présence étrusque. Les données de la zone 27. Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes: Lattara 22: 115-165 Colomer, A. and Gardeisen, A. 1992. La consommation des animaux d’élevage et de chasse dans la ville de Lattara (fin du IVe s. av. n. è.-milieu du Ier s. de n. è.). In Py, M. (ed.) Recherches sur l’économie vivrière des Lattarenses. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 5: 91-110.

Curé, A.-M. 2014. La céramique de cuisine tournée et les pratiques culinaires à l’âge du Fer en Gaule méditerranéenne. PhD Dissertation, University Paul Valéry-Montpellier 3, Montpellier Daveau, I. (ed.) 2007. Port Ariane (Lattes, Hérault). Construction deltaïque et utilisation d’une zone humide lors des six derniers millénaires. Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes : Lattara 20. Daveau, I. and Py, M. 2015. Grecs et Etrusques à Lattes: nouvelles données à partir des fouilles de la Cougourlude. In Roure R. (ed.) Contacts et acculturations en Méditerranée occidentale. Hommages à Michel Bats. Editions Errance-Centre Camille Jullian, Maison méditerranéenne des sciences de l’homme, Aix-en-Provence-Paris, Collection Etudes Massaliètes 12-Bibliothèque d’archéologie méditerranéenne et africaine 15: 31-42. Daveau, I., Chardenon, N., Da Costa, C., Dubesset, D., Henry, E. and Py, M. 2015. Lattes avant Lattara ? Le village du Premier âge du Fer de ‘La Cougourlude’ (Hérault). In Olmer, F. and Roure, R. (ed.) Les Gaulois au fil de l’eau. Actes du 37e Colloque international de l’Association Française pour l’Etude de l’Âge du Fer, Montpellier 2013, Vol. 1. Ausonius éditions, Mémoires 39, Bordeaux : 87-114. Gailledrat, E. 2014. Espaces coloniaux et indigènes sur les rivages d’Extrême-Occident méditerranéen (XeIIIe S. avant notre ère). Presses Universitaires de La Méditerranée, Montpellier, Coll. ‘Mondes Anciens’ Gailledrat, E. and Vacheret, A. 2014. L’occupation de la première moitié du Ve s. av. n. ère (Zone 1). In Garmy, P., Gailledrat, E., Ciesielski, E., Compan, E., Duboscq, S., Jorda, C., Liottier, L., Piques, G., Rivalan, A., Rovira, N. et al. 2014. Lattara (Lattes, Hérault) 2014 : La zone 1. Rapport final d’opération 2011-2014, CNRS-SRA Languedoc-Roussillon. Garcia, D. 1992. Du grain et du vin, à propos des structures de stockage de l’agglomération portuaire de Lattes. In Py, M. (ed.) Recherches sur l’économie vivrière des Lattarenses. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 5: 165-182 Garcia, D. (ed.) 1994. Exploration de la ville portuaire de Lattes, les îlots 2, 4-sud, 5, 7-est, 7-ouest, 8, 9, 16 du quartier Saint-Sauveur. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 7 Garcia, D. and Vallet, L. (ed.) 2002. L’espace portuaire de Lattes antique. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 15 Gardeisen, A. 1999. Économie de production animale et exploitation du milieu à Lattes au cours du IVe siècle avant notre ère. In Py, M. (ed.) 1999. Recherches sur le quatrième siècle avant notre ère à Lattes. Edition de l’Association pour le développement 39

Country in the City de l’archéologie en Languedoc-Roussillon, Lattes  : Lattara 12 : 537-567 Gardeisen, A. 2003. Contribution de l’archéozoologie des grands mammifères à l’étude d’un espace ouvert en contexte urbain. La zone 123 (Lattes/Saint-Sauveur, Hérault). In Buxó R., Chabal L. and Gardeisen A. (ed.) La place 123 de Lattara, recherches pluridisciplinaires sur un espace urbain du IVe siècle avant notre ère. Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes  : Lattara 16: 169-184 Janin, T. (ed.). 2010 Premières données sur le cinquième siècle avant notre ère dans la ville de Lattara. Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes : Lattara 21  Lebeaupin, D. (ed.) 2014. Les origines de Lattara et la présence étrusque. Les données de la zone 27. Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes: Lattara 22 Jorda C., Chabal L. and Blanchemanche P. 2008. Lattara entre terres et eaux. Paléogéographie et paléoboisements autour du port protohistorique. In Janin, T. and Py, M. (ed.) La ville portuaire de Lattara (Lattes, Hérault) et son territoire: nouveaux acquis, nouvelles questions. Gallia 65 : 11-21. Lebeaupin, D. and Sejalon, P. 2010. Evolution d’un groupe d’habitations du Ve siècle dans l’îlot 27. In Janin T. (ed.) Premières données sur le cinquième siècle avant notre ère dans la ville de Lattara. Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes  : Lattara 21 : 135-202 Luley, B.P. and Piquès, G. 2016. Communal eating and drinking in early Roman Mediterranean France: a possible tavern at Lattara, c. 125–75 BC. Antiquity 90 : 126-142. Margaritis, E. and Jones, M. 2006. Beyond cereals: Crop processing and Vitis vinifera L. Ethnography, experiment and charred grape remains from Hellenistic Greece. Journal of Archaeological Science 33 (6) : 784-805. Mcgovern, P., Luley, B.P., Rovira, N., Mirzoian, A., Callahan, M.P., Smith, K.E., Hall, G.R., Davidson, T. and Henkin, J. M. 2013. Beginning of viniculture in France. Proceedings of the National Academy of Sciences of the United States of America 110 (25) : 10147–10152. Newman, C. and Silvereano, S. (ed.) 2010. Lattes–Mas de Causse, Hérault, région Languedoc-Roussillon. Rapport final d’opération, fouilles archéologiques préventives, Oxford Archéologie Méditerranée, Mauguio. Nickels, A. 1976. Les maisons à abside d’époque grecque archaïque de La Monédière, à Bessan (Hérault). Gallia 34 (1) : 95-128. Puertas, O. 1998. Palynologie dans le delta du Lez. Contribution à l’histoire du paysage de Lattes. Edition de

l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes : Lattara 11 Py, M. (ed.) 1988. Mélanges d’Histoire et d’Archéologie de Lattes. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 1. Py, M. (ed.) 1990. Fouilles dans la ville antique de Lattes, les îlots 1, 3 et 4-nord du quartier Saint-Sauveur. Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes : Lattara 3 Py, M. 1993. Dictionnaire des Céramiques Antiques (VIIème s. av. n. è. - VIIème s. de n. è.) en Méditerranée nord-occidentale (Provence, Languedoc, Ampurdan). Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes  : Lattara 6 Py, M.  (ed.) 1996. Urbanisme et architecture dans la ville antique de Lattes. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 9 Py, M. (ed.) 1999. Recherches sur le quatrième siècle avant notre ère à Lattes. Edition de l’Association pour le développement de l’archéologie en LanguedocRoussillon, Lattes : Lattara 12 Py, M. 2009. Lattara (Lattes, Hérault). Comptoir gaulois méditerranéen entre Etrusques, Grecs et Romains. Editions Errance, Paris. Py, M. 2016. Dictionnaire des objets protohistoriques de Gaule méditerranéenne (IXe-Ier siècles avant notre ère). Edition de l’Association pour le développement de l’archéologie en Languedoc-Roussillon, Lattes  : Lattara 23 Py, M. and Buxó, R. 2001. La viticulture en Gaule à l’âge du Fer. In Brun, J.-P. and Laubenheimer, F. (ed.), La viticulture en Gaule. Gallia 58 : 29-43 Rovira, N. and Alonso, N. 2010. Thanatocoenoses of seeds and fruits from Zone 1 at Lattara (Lattes, France) during the 5th-4th centuries BC: the preliminary results. In Bakels, C., Fennema, K., Out, W., Vermeeren, C. (ed.), Of plants and snails: A collection of papers presented to Wim Kuijper in gratitude for forty years of teaching and identifying, Sidestone Press: 217-228. Rovira, N. and Alonso, N. 2018. Crop growing and plant consumption in coastal Languedoc (France) in the Second Iron Age: new data from Pech Maho (Aude), Lattara (Hérault) and Le Cailar (Gard), Vegetation History and Archaeobotany (2018) 27: 85–97 Rovira, N. and Chabal, L. 2008. A foundation offering at the Roman port of Lattara (Lattes, France): the plant remains. Vegetation History and Archaeobotany 17 (1) : 191-200. Van Der Veen, M. 2003. When is food a luxury ?. World Archaeology 34(3) : 405-427.

40

The Neglected ‘Fields’ of Proto-Urban Living: A View from Bronze Age Crete Kostis S. Christakis

Knossos Curator (British School at Athens)

To all those, past and present, whose studies have contributed to the making of Mirabello.

Introduction

on Cretan societies is still controversial (for an overview see Brogan 2011). If life expectancy during prehistory is assumed to be 25-30 years, then what is described here lies within the time-frame of the developing circle of a single family.

At the dawn of the second millennium BCE, a series of complex socioeconomic developments led to the emergence of large proto-urban centres (Whitelaw 2012; Whitelaw, this volume). Presumed seats of political institutions controlling large social segments, these centres marked the Cretan scape for at least eight centuries. Although it has been repeatedly stressed that urban populations strongly relied on surpluses provided by the countryside, the economic interdependence between cities and their immediate and wider hinterland has not always received proper attention in the context of Bronze Age Crete. The main focus of interest for many decades has mostly been the ostentatious material expressions of socioeconomic stratification. Consciously or otherwise, this has meant that priority was given to the contexts most likely to produce imposing buildings and impressive luxury finds, thereby neglecting the urban hinterland and limiting its academic study. Rural communities have been seen as passive suppliers of foodstuffs and raw materials to urban households and as consumers of goods produced in the cities. According this approach, high-ranking government officials controlled most facets of an economy organized on the basis of redistribution. Among the most noteworthy exceptions is the pioneering study of Dewolf, Postel, and van Effenterre on the Maliote economy (1963).

As far as spatial extent is concerned, I discuss information from the sites of Mirabello Bay, East Crete. There are three reasons for this. First, the region has many different sources of wealth: fertile land-seascapes and rich mountain regions. This, the narrowest point of Crete, known as the Isthmus of Ierapetra, is a trade corridor between the north and south coasts of the island, with maritime links to the Aegean and the whole of the Eastern Mediterranean. So it is interesting to investigate how local communities functioned against such a multi-faceted economic background. Second, the major centres of the region have been excavated, although not to their full extent. The region of the Isthmus is the only part of the island with so many excavated sites within its borders. The quality and quantity of testimonies varies from case to case, despite the fact that the area has provided more and fuller publications than any other part of Crete. This variability is due to complex site bibliography and biases in the documentation of old excavations. Moreover, the systematic application of analytical methodologies from retrieving bio-archaeological data in most of the recently excavated contexts provides crucial information. And third, almost the whole of the region has been intensively surveyed, allowing us to trace the human web in great detail.

This contribution considers the relationship between urban centres and their hinterland by examining artefacts and ecofacts concerned with subsistence economy. Deeply rooted in the cultural substrate of society, dietary habits reflect a certain way of approaching and exploiting the natural environment, investing its products in a variety of symbolisms (Atalay and Hastrof 2006). The discussion is limited in time and space. The temporal framework is that of the last phase of the Neopalatial period, known as Late Minoan IB (1520/1510-1440/1430 BCE in absolute terms), this being the best convergence of data for Bronze Age Crete. The beginning and end of this period, which lasted about 80 years, are marked by catastrophic events, whose impact

Datasets from 148 contexts, mostly residential units, excavated at Gournia, Pseira, and Mochlos have been analysed. A considerable part has been fully excavated - albeit very few units have actually been published. The elaborate architecture and finds in some of these indicate that they may have been used by wealthy groups, while the rest were inhabited by more ordinary family segments. Architectural layout and standardization in size suggest that most were inhabited by nuclear or minimally-extended stem families of about 4-5 individuals (Whitelaw 2001). Of 41

Country in the City course, the existence of co-residential corporate groups cannot be excluded, especially in the case of large and complex residences (Driessen 2010). However, crossculturally and diachronically, the nuclear family is the most prevalent domestic unit (for a critical discussion see Apostolaki 2014: 180-202; 2015).

Whitelaw 2001; Branigan 2001; Watrous and Schultz 2012: 58; for land tenure see Christakis 2008: 33-4). If this were the case, then the rural sector must have supplied agricultural surpluses. The volume and consistency of demand for goods would vary from centre to centre - Mochlos and Pseira were located in agriculturally marginal settings, as opposed to Gournia, which was surrounded by more extensive arable lands, especially after the abandonment of some Protopalatial sites in the vicinity of the town (Watrous and Schultz 2012: 55). Even the farmland surrounding Gournia, though, is too limited to sustain the town’s population (Hayden 2004: 111; Christakis 2008: 135; Watrous and Heimroth 2011).

Excavated testimonies have been supplemented with information from the survey projects carried out at Vrokastro (Hayden 2004; 2005), Gournia (Watrous et al. 2012), Kavousi (Haggis 2005); Chrysokamino (Betancourt 2006a), and Pseira (Betancourt et al. 2005). Although the region is not subject to a systematic survey, important information is also at our disposal from Mochlos (Soles 2003). A complex picture of a human-constructed scape, consisting of town-sized settlements, villages, hamlets, and farmhouses, has emerged. The serious drawback, though, is the absence of a significant corpus of archaeological evidence from these sites: actual black dots on topographic maps. How did villages and secondary centres interact with their immediate hinterland and/or hintersea? What were the relationships of rural communities with the major urban centres of the region? These alluring questions are unresolved; we can only advance working hypotheses whose validity can only be confirmed by future research.

Locating cultivated fields is an issue open to debate. Proposed estimates illustrate the size of sustaining areas but not the exact configuration of the arable plots. Moody, discussing a vast array of island-wide data, has shown that prehistoric Cretans exploited lands mostly located on the lower slopes of a region, between 50 and 400 m and further than 1 km from the shore, followed by middle slopes of 400-800 m even further from the sea (Moody 2012). Continuous and intensive use of the countryside through the centuries has completely erased artificial field boundaries, if such boundaries actually existed. Plot marking does not necessarily consist of durable human-made structures; territorial boundaries could be natural and even cognitive, resulting from a knowledge of the land passed down from generation to generation and/or natural constraints (Betancourt 2006b; Bennett 2007).

Production, storage and processing of subsistence goods Discussing the production capabilities of the Mirabello region and their impact on the subsistence needs of the urban population is rather a complex issue. The relevant information, such as macrobotanical, faunal, chemical and palynological data, is depressingly meagre despite some significant exceptions. The physical features of the region, however, may provide some indirect insights. Although the landscape is subject to natural and anthropogenic transformations, these usually occur at a rate much slower than that of political and economic cycles (Asdrachas 2007: 45). Thus, the types of cultivatable staples and agrarian technology would obviously have changed over time, but the geophysical setting and the necessity of access to arable land would not have done so.

Gournia, Pseira and Mochlos, however, were not the only exploiters of natural wealth in the Isthmus region. Survey data show a densely populated countryside during the Neopalatial period, although not as dense as in the preceding era, with a pattern of settlement nucleation and population centralization (Hayden 2004: 112; Haggis 2005: 74; Watrous and Schultz 2012: 55-6). Small town-sized settlements, villages, hamlets, and farmhouses created a complex network. The inhabitants of these sites would have used part of the local production, both to cover their subsistence needs and to gather a surplus, which in the hands of rural elites, would have provided the means of fulfilling social aspirations. If we had sufficient data at our disposal to estimate the farmland needed to sustain the inhabitants of each site, a large degree of overlap would probably be evident between the catchment areas of sites in the Isthmus region. If this picture is correct, we may assume that there would be conflicts of interest, especially in periods of subsistence stress.

The methodology frequently adopted for mapping the relationship between population and available farmland in cultural contexts without relevant data is to multiply the estimated population of each site by the minimum amount of land needed for subsistence (e.g. Chisholm 1968). Assuming a population estimate of 1000 people for Gournia, 750 for Mochlos and 300 for Pseira, and a minimum of 0.6 ha as the amount of arable land needed per capita, I suggest that the land directly available for farming could not have entirely sustained the urban population (for population estimates see

Agriculture intensified in the Late Minoan I period, as indicated by various types of evidence (Betancourt 2005: 290-94). The megalithic farmsteads that now appear in many areas of the region, located on hilltops above fertile alluvial valleys and linked to routes, 42

Kostis S. Christakis: The Neglected ‘Fields’ of Proto-Urban Living

indicate the systematic and intensive exploitation of surrounding farmlands (Hayden 2004: 116-17; Haggis 2005: 75, 77; Watrous and Schultz 2012: 51-6; Betancourt 2006c: 268-70). Annexes to some of these structures might have been residences of extended households and/or labourers, or used for the storage of crops, tools and animals (Hayden 2004: 119). Their architecture and location indicate that these complexes might have belonged to an emerging rural elite class (Haggis 2005: 77). Landscape facilities, including dams and terracing, in less favourable and marginal areas, suggest systematic farming (Haggis 2005: 76; Betancourt 2005: 286-92; Soles 2003: 123-24). The construction, use and maintenance of such facilities indicate an integration of individuals above household level (Adler and Wilshusen 1990). Small farmhouses scattered across the region were used by households who preferred to live close to their land, even in remote areas, to defend their territory, crops and water sources (Hayden 2004: 116-17, 119; Haggis 2005: 75). High off-site distribution of sherds from cooking pots has been interpreted as an indication of agricultural intensification; farmers carried their hot meals with them due to the long hours spent in the fields (Betancourt 2005: 290; Watrous and Schultz 2012: 53). The large number of sherds scattered in areas where potential fields are located suggests that manuring practices were more intensive than before (Betancourt 2006b: 290).

(Christakis 2008: 113-16). To explain this striking picture, one could argue that households might have stored their goods in outbuildings or second houses, or kept their staples exclusively in containers of perishable materials. Although storing produce in field stores has been ethnographically attested, this is the exception rather than the rule; it is a practice which does not guarantee the security of the stored goods, is mainly adopted by small rural communities rather than urban dwellers, and is impractical because goods must be constantly moved. Moreover, no such buildings have been identified in Bronze Age Crete to date. The exclusive use of perishable storage containers such as baskets, cloth sacks, wooden boxes or goatskins is not attested in the historically recorded past, especially in cultural milieu with a long tradition in the production of large coarse-ware. Perishable containers were always associated with pithoi and mostly used for transport and short-term storage; excavated pictures from the Late Cycladic I town of Akrotiti, Thera are particularly informative in this respect (Polychronakou-Sgouritsa 2000).

The basis of agricultural production in the region would have been grain, olive and viticulture, all suitable crops for marly soils dependent on rain for irrigation. Vegetables and fruit orchards would have been grown in areas with more abundant water, with flax for linen being cultivated in riverine soils. A wide variety of crops could have been cultivated in this fertile farming region, depending on local water availability and requirements, mainly subsistence cultivation, with the surplus being used for barter or appropriated by the central and private administration. Macrobotanical remains found in many domestic units of the region are very informative on the agricultural products cultivated in and collected from the fields (e.g. Snowden and Jones 1995; Sarpaki and Bending 2004; Jones and Smith 2009). The body of data could be supplemented by testimonies on subsistence staples collected from other contemporary Cretan contexts (for an overview see Christakis 2008: 19-32).

Small and medium-sized pithoi were found in 21 domestic units; their capacity ranged from 200 to 800 litres, in a few cases reaching 1000 litres (Christakis 2008: 85-92). Pithoi were often supplemented by large amphorae and jugs. This storage strategy occurs in small domestic units of simple design and construction. Building AF Nord at Pseira, House C.3 at Mochlos, a domestic unit whose owner specialized in the metal trade, and Building B.2 at Mochlos, perhaps used by the ruling group of the town, provide evidence for higher storage potentials ranging from 1200 to 2000 litres (Christakis 2008: 87-8, 90). Means of storage also include limited-capacity installations, such as pits, sometimes plastered, dug in the floor (e.g. Hawes et al. 1908: 26) or built cists inside storerooms (e.g. Soles 2003: 12-5). It must be noted, however, that there is no safe indication of their use, given that associated finds are unknown or have not been uncovered or preserved, except in the case of Mochlos. The excavated context with the highest storage capacity in the Mirabello region is the central building at Gournia. Although at least 15 pithoi were found, the stores were designed to accommodate a total of 80 large pithoi (Christakis 2008: 51, 53). This discrepancy has been explained as the result of human intervention.

How much, though, of the agricultural production was stored in the pantries of urban houses? The testimonies reveal significant differences in the distribution of storage containers and installations. Of the houses excavated in the region, only very few produced evidence for staple storage. I have suggested elsewhere that the absence of storage containers does not result from taphonomic processes or research biases but reflects actual patterns of economic behaviour

Of the commodities kept in the pantries, the most important would be the cereals, the mainstay of the daily diet through the ages (Christakis 2008: 19-21). Macrobotanical remains were found in some units, but the picture is not entirely representative since very few contexts have been analysed (Sarpaki and Bending 2004). Stone mortars and querns, usually associated with cereal processing, are another source of indirect evidence. Wooden mortars and other implements made 43

Country in the City of perishable materials would have been used to process cereals, along with pits in the bedrock. Mortars were found in 33 houses, whereas querns are mentioned in just 14 cases (Apostolaki 2014: 275-82, table 7.2). The impressively low number of cases with querns is certainly the result of research biases. The absence of mortars from most of the houses remains unexplained; one could argue that cereals were processed using implements made of perishable materials, or that few households had access to arable land and subsistence goods. Equally possibly, however, mortars served purposes other than cereal processing in which not all households participated (Apostolaki 2014: 359).

Building A (Soles 2003: 24; Sarpaki and Bending 2004). Whether olives were cultivated round the town or imported from another site is uncertain, although the first scenario seems more likely. Olives have also been reported at Pseira and would certainly have existed at Gournia (Jones and Smith 2009). What evidence do we have of the other goods, which would have come to the towns from the fields, the sea and the non-cultivated areas of the region? The picture is just as fragmented as in the case of cereals, olive oil and wine. The data from a few analysed contexts and the array of testimonies from other contemporaneous Cretan sites are indicative of the wealth of information that has been lost from contexts excavated early in the 20th century. Legumes (fava, lentil, grass pea), figs, almonds and various wild plants and herbs have been recovered at Mochlos and Pseira (Sarpaki and Bending 2004; Jones and Smith 2009). The balance in the occurrence of cereals and pulses, and the crop variability observed at Mochlos, make it very likely that crop rotation was practised (Sarpaki and Bending 2004). These agricultural strategies point to a shortage of arable land and environmental restrictions. Plants with aromatic and/or pharmaceutical properties have also been identified by residue analysis on pottery from Mochlos (Koh 2006; see also Martlew and Beck 1999; Beeston et al. 2006). Based on the consumption of edible plants, herbs, fruit and nuts in pre-industrial Crete, the prehistoric Mediterranean diet would probably have been similarly varied (Christakis 2008: 19-25; Moody 2012).

The same dearth of data is also observed in the case of installations used for the production of olive oil and wine. Wine-presses were found in Houses Ac, Bd and Dd at Gournia, and an installation, probably used for the processing of liquids, was excavated in House C.7 at Mochlos (Hawes et al. 1908: 27-8; Fotou 1993: 64, 75; Soles and Davaras 1996). Fragments of wine-presses and remains of grapes are reported in some units at Gournia, Mochlos and Chalinomouri, while wine flavoured with pine resin and various herbs, to add taste or for medicinal purposes, was detected in a basin from Mochlos (Hawes et al. 1908: 28; Barnard and Brogan 2003: 56-7; Beck et al. 2008). The limited evidence of wine production within urban centres indicates that grapes were pressed either in the countryside or in implements made of perishable materials (cf. Palmer 1994: 22; Tsipopoulou and Dierckx 2006). Wine-presses in the towns of the region may have served the needs of many households or been used for the production of limited quantities of wine designated for special use. A fact in support of this argument is that the storage containers associated with wine-presses are mostly small pithoi of low storage potential; only in the case of Mochlos were medium-sized specimens used. This suggests that the must was stored either elsewhere or in containers which did not left traces in the archaeological record. The picture from Mirabello Bay contrasts with that observed at Zakros (Kopaka and Platon 1993). Wine-presses were associated with containers of a considerable storage potential (Christakis 2008: 101-02, 107). The juxtaposition of these two cases shows how different the wine production strategies in two contemporaneous urban centres can be.

Faunal remains have only been collected from few domestic units at Mochlos and Pseira. Sheep, goats, pigs and cows are reported at Mochlos but, quite surprisingly, pigs and cows are absent at Pseira (e.g. Reese 1998; 2009). Cows and sheep/goats were slaughtered at a relatively advanced age, suggesting greater interest in their secondary products, while pigs seem to have been kept for their meat (Moody 2012). Bones from hunted animals are almost absent from the analysed assemblages. Game resources were rare, but this is arguably due to the scant data available. The fish represented in the analysed assemblages are mostly small fish from shallow coastal waters (Rose 1994: 302-17; Mylona 2004). Medium-sized fish also occur, although in the lowest percentages. Small species seem to be preferred in the Artisan’s Quarter at Mochlos, while households living in the central settlement consumed medium/larger species. The percentages of fish bones found at Pseira match to those of animal bones, highlighting the importance of marine resources in the Pseirian diet (Moody 2012).

The published evidence on olive oil production is poor. Macrobotanical remains of olives were found in relatively high percentages at Mochlos (Sarpaki and Bending 2004). Wild and cultivated trees, some infested with insects, were reported, indicating collection of olives from wild and domesticated trees. Fragmented olives, the by-products of olive oil extraction, also occur in high percentages. Olives associated with a large basin of calcareous sandstone, which probably served for olive crushing, were found in room 2 of

To summarise, communities in the Isthmus exploited a vast array of terrestrial and marine subsistence goods. Some of these were detected among the collected 44

Kostis S. Christakis: The Neglected ‘Fields’ of Proto-Urban Living

environmental data, but many others could be posited based on evidence from other contemporaneous contexts; despite regional variation in the production and consumption of certain goods, the existence of common subsistence trajectories among contemporaneous Cretan communities is beyond doubt. The subsistence autonomy of each site is determined by the relationship between population distribution and available agricultural land, and there are sound grounds to suggest that the large urban centres of the region may have received surpluses from surrounding villages and secondary centres. The quantities of goods required by each centre would have varied in each case. I would suggest, though, that rural communities were not passive providers of agricultural wealth, as has traditionally been argued, but dynamic groups with their own needs and aspirations. The varying investment of space and resources in the storage of foodstuffs is indicative of differences in access to arable land among urban households, and the adoption of different coexisting economic activities and strategies within each site.

Davaras 1992; 1994; 1996). Household industries were uncovered in Building BY, whose household engaged in obsidian processing, and Building BN East (Dierckx 1999; Betancourt 1999). The widespread distribution of artefacts usually associated with crafts at Gournia led Watrous and Heimroth to suggest that almost all the town might have served as a manufacturing centre (2011). However, although the presence of households engaged in crafts is not in any doubt, the extent of such activities seems to have been overemphasized. The recent re-examination of data by Apostolaki has highlighted some methodological shortcomings in the definition of household industries at the site (2014: 332-36). For instance, the presence of more than ten specimens of a specific pottery shape in a house was interpreted as evidence of potting production, while the proposed coexistence of different craft activities does not take the necessary infrastructure and human resources into account. A more in-depth examination of data shows that secure evidence for household industries indicative of production above and beyond the needs of a household is limited (Apostolaki 2014: 335-36).

Farmers, artisans and traders

Most of these contexts, which would obviously not have been the only industrially active ones, lack substantial storage facilities; pithoi were either absent or, where they were used, had a maximum capacity of 700 litres (Christakis 2008: 85-91). Even if one were to argue for the exclusive or supplementary use of storage containers made of perishable materials, the picture of limited storage potential remains unchanged. This limited self-sufficiency does not match households dependent on workshop activities. Their involvement in craft production presupposes that they had the economic means to obtain the raw materials and cover their subsistence pursuits, allowing them to exercise their craft unimpeded. Although these households may have engaged in many activities, at the level of handicraft specialization the occupation would have been largely exclusive (Clark and Parry 1990; Soles 2003: 98; Apostolaki 2014: 334). Discussing craft activities at Mochlos, Soles argues that artisans produced goods for exchange in the local markets and that they relied in turn on markets to cover their subsistence needs, a view also adopted by the present author (Soles 1997; 2003: 98-9; 2004; Christakis 2008: 138-39; cf. also Watrous and Heimroth 2011). The reliance on markets fits well with the absence of sizeable domestic storage space, since only short-term storage would have been necessary in communities engaged in trade and exchange (Smyth 1989).

Although securing subsistence and creating surpluses to ensure self-sufficiency for a year or more were cornerstones of pre-modern household economies around the globe, this was not always the rule. Evidence from the Isthmus indicates that not all households, indeed not even the majority, followed this strategy, and only a few had stored goods capable of sustaining a typical family for a year. This pattern is by no means confined to the Isthmus but is echoed across Late Minoan IB urban settings. Interpreting this economic behaviour presupposes the interplay of many parameters (Christakis 2008: 113-18). Some households had storage inadequate for self-sufficiency and this may have created dependency relations with the central administration or wealthy groups. The life of the lower social classes is frugal in many cultural contexts. Believing otherwise is, in my view, more indicative of arbitrary stereotyping of Cretan Bronze Age societies rather than grounded evidence from actual demonstrable facts. There are cases, however, where low subsistence reserves are associated with contextual settings indicating economic distress; this leads us to investigate other paths of inquiry. Some houses yielded evidence for craft activities related to the processing of raw materials, either from the hinterland or imported from other areas, even overseas. The Artisans’ Quarter at Mochlos was occupied by households engaged in the production of stone vases, pottery, textiles, and bronze working (Soles 2003: 91-9; 2004). Specialists in the production of stone vases and seals were active in House C.7 and House C.2 respectively in the same town (Soles and

The existence of markets in the Aegean did not enjoy scholarly support for many decades, mostly due to the wide and uncritical acceptance of the Polanyian paradigm of ancient economies. Palatial institutions were seen as the only regulators of any economic 45

Country in the City transaction. Acting as redistributive agents, they were thought to draw upon raw materials and labour from the hinterland in order to produce and distribute specialized artisanal goods (e.g. Killen 2008). In recent years, however, new perspectives have effectively challenged not only the redistributive role of governing institutions but also the negative attitude towards the existence of market and market-like systems in the Aegean (Galaty et al. 2011; Parkinson et al. 2013). Once people started looking for them, markets have begun to appear in much earlier chronological horizons, shaping the emergence of state administrative institutions (see various contributions in Garraty and Stark 2010). Although there are some constrains in recognising and understanding markets in prehistoric societies, the market is an economic process which should developed further in Bronze Age Crete.

acknowledge that products such as ceramics could not have been effectively supplied through redistribution (Stark and Garraty 2010). The market seems to be the most plausible mechanism by which these goods were channelled to their users, although we do not know whether this distribution only took place through a central market or via retail and small-scale vendors and peddlers. Apart from all these utilitarian goods, which imply transactions on a regular basis, there is also the testimony of archive documents, directly certifying dealings between households. Sealings were found in Houses Fg and Cf at Gournia and at House C.1 at Mochlos (Hawes et al. 1908: 24; Fotou 1993: 63, 89; Soles and Davaras 1992). The testimonies, though few, are evidence of exchanges among certain local households. To these must be added the issue of weights and measures (e.g. Fotou 1993: 89; Petruso 1992: 40-2, 50-1). Weights surely indicate the establishment of equivalent values for commodities, whether for market exchange or reciprocal exchange (Michailidou 1999). The commercial activities in which some households engaged, such as the trade in copper at House C.3 at Mochlos and imported pottery at Building AD Centre at Pseira, would necessitate some form of recordkeeping (Soles 2008; Betancourt 1995, 137-38). The interpretation of these pictures, as well as similar ones at Chania and Zakros where record keeping documents, such as Linear A tables and sealings, were found in some domestic units (Hallager 1978; Palmer 1995), depends on the theoretical viewpoint adopted in our approach to Neopalatial polities. Those who see the palaces as the only economic agents may argue that these households were related to the central administration. If, however, we reduce the involvement of palaces in economic affairs - and I think that there are sound grounds for doing so - these finds may allude to private administrative practices (cf. Schoep 2002: 198). Private economies were truly alive in proto-urban and urban settings (Peyronel 2010). Late Minoan IB public and private commercial activities were closely interwoven; rather than constituting completely different spheres, they generally formed two sides of the same coin.

Testimonies bear witness to the existence of a complex exchange network at the level of ordinary households. Households would have acquired all their goods for daily use - pottery is the best documented in the archaeological record - which they obviously did not make themselves. Vessels used at Pseira, for instance, were not made locally but brought from the GourniaKalo Chorio coastal region (Day 1997). The distribution of agricultural produce from the hinterland and even other regions beyond Mirabello would have been on an even larger scale; amphorae produced at Zakros and Palaikastro and found at Pseira may have contained agricultural products, although most such goods were most likely shipped in perishable containers. How, then, were the raw materials used in craft industries or the imported finished products, such as the Knossian pottery found at Pseira or the Syro-Palestinian mortars found at Gournia and Mochlos, distributed? The movement of these goods and, to a far greater degree, of goods not yet identified in the archaeological record, necessarily presupposes the existence of a kind of mechanism that would ensure their steady distribution to local households. Could these exchanges lie exclusively in the hands of palatial authorities? Although state-centric approaches assign such a role to central authorities, more recent contributions argue that these institutions only controlled some specialized crafts, whereas other goods, covering sizeable segments of the economy, were produced and distributed by networks largely or completely independent of the palaces (e.g. Halstead 1992; 2007; Sjöberg 2004). These views have been put forward with reference to the economic organisation of Mycenaean kingdoms, but I believe that, by broadening our horizons, some potential lessons might be learnt from them. It seems highly unlikely that exchange of mundane utilitarian goods would have been operated by the palace. It would have been far simpler for households to obtain the goods they needed through market transactions. Let us

Conclusions Analysis reveals complex socioeconomic patterns arising from the interaction of the city and its nearer and wider hinterland in order to cover everyday needs. Due to constrictions on the amount of land available for farming and the high concentration of people not directly involved in the production of subsistence wealth, towns largely lack autarky, necessitating the constant flux of goods from the hinterland to the core. Households are in constant need of resupply, not so much with regard to the specific type or amount of foodstuffs and raw materials provided, but with reference to 46

Kostis S. Christakis: The Neglected ‘Fields’ of Proto-Urban Living

the systematic need for their provision. Patterns of distribution of storage containers, household industries and other potential sources of information within urban scapes offer a great diversity of alternative options by which domestic economies might be organized. On the one hand are the households that follow a strategy of subsistence self-sufficiency, while, one the other hand, at the other end of the spectrum, are those which do not achieve this end, either due to poverty or because they are constantly resupplied with goods from markets in exchange for crafted goods or services. Households deal with their needs and their economies may intersect with central urban economic systems in different ways. Both households and individual members are mutually dependent elements of a large, complex and unified economy. Here as elsewhere, the economic environment of urban centres creates and sustains social inequality and systems of power. The ‘haves’ sustain their prestigious lifestyle and enhance their status by exploiting the labour of the ‘have-nots’, who are forced to compete for work and patronage (Smith 2003).

University of Athens. For a online version see http://phdtheses.ekt.gr/eadd/handle/10442/34656 Apostolaki, E. 2015. On the household structure of Neopalatial society. In S. Cappel, U. Gükel-Maschek and D. Panagiotopoulos (eds), Minoan ArchaeologyChallenges and Perspectives for the 21st Century. Proceedings of the International PhD and Post-Doc Conference at Heidelberg, 23-27 March 2011 (Aegis 08): 223-35. Heidelberg. Asdrachas, S. 2007. Greek Economic History: 13th-19th Centuries. Athens, Piraeus Bank. Atalay, S. and Hastorf, C. A. 2006. Food, meals, and daily activities: food habitus at Neolithic Çatalhöyük. American Antiquity 71.2: 283-319. Barnard, K. A. and Brogan, T. M. 2003. Mochlos IB: Period III. Neopalatial Settlement on the Coast: The Artisans’ Quarter and the Farmhouse at Chalinomouri. The Neopalatial Pottery (Prehistory Monographs 8). Philadelphia, Pennsylvania, INSTAP Academic Press. Beck, C. W., E. C. Stout, K. M. Wovkulich, and A. J. J. Phillips. 2008. Absorbed organic residues in pottery from the Minoan settlement of Pseira, Crete. In Y. Tzedakis, H. Martlew, and M. K. Jones (eds), Archaeology Meets Science. Biomolecular Investigations in Bronze Age Greece: 48-73. Oxford, Oxbow Books. Beeston, R., J. Palatinus, C. Beck, and E. C. Stout. 2006. Organic residue analysis of pottery sherds from Chrysokamino. In P. P. Betancourt, The Chrysokamino Metallurgy Workshop and its Territory (Hesperia Supplement 36). Princeton, American School of Classical Studies. Bennet, J. 2007. Fragmentary -geo-metry: early modern landscapes of the Morea and Cerigo in text, image, and archaeology. In S. Davies and J. L. Davis (eds), Between Venice and Istanbul: Colonial Landscapes in Early Modern Greece (Hesperia Supplement 40): 199217. Princeton, The American School of Classical Studies at Athens. Betancourt, P. P. 1995. Comments and conclusions, Building AD Centre. In P. Betancourt, and C. Davaras (eds), Pseira I: The Minoan Buildings on the West Side of Area A (University Museum Monograph 90): 137-138. Pennsylvania, The University Museum. Betancourt, P. P. 1999. Building BN East: comments and conclusions. In P. Betancourt and C. Davaras (eds), Pseira IV. Minoan Buildings in Areas B, C, D, and F (University Museum Monograph 105): 110-111. Philadelphia, The University Museum. Betancourt, P. P. 2005. Discussion and Conclusions. In P. P. Betancourt, C. Davaras, and R. H. Simpson. Pseira IX. The Archaeological Survey of Pseira Island. Part 2. The Intensive Surface Survey (Prehistory Monographs 12): 275-306. Philadelphia, Pennsylvania, INSTAP Academic Press. Betancourt, P. P. 2006a. The Chrysokamino Metallurgy Workshop and its Territory (Hesperia Supplement 36). Princeton, The American School of Classical Studies.

Townspeople engaging in any economic activity will inevitably come into contact with a host of other individuals. As they organize, move around and trade, so the geographical limits of their transactions change, expanding to match the goods on the market brought from furthest afield. From the point of view of both suppliers and supplied, the boundaries of domestic activity extended beyond those of the home. Those who would have had to resort to the harbour, the fields or the market in order to sell their products or acquire necessities, obviously associated and dealt with others. It is this everyday communication that ensures social cohesion, ultimately constituting the city itself. Acknowledgements I would like to thank D. Garcia, R. Orgeolet, M. Pomadère, and J. Zurbach for their kind invitation to participate in the Country in the City colloquium and for their generous hospitality in Marseille. Thanks are due to E. Apostolaki for discussing relevant issues and giving me a copy of her PhD thesis and to Todd Whitelaw for seminal discussion. Many thanks to R. Tzanaki for editing matters. References Adler, M. A and Wilshusen, R. H. 1990. Large-scale integrative facilities in tribal societies: crosscultural and southwestern US examples. World Archaeology, 22: 133-46. Apostolaki, E. 2014. Η δυναμική του οικιακού χώρου. Παραδείγματα νοικοκυριών από τη νεοανακτορική κοινωνία της Κρήτης. Unpublished PhD Thesis, 47

Country in the City Betancourt, P. P. 2006b. Land use on the Chrysokamino Farmstead. In P. P. Betancourt, The Chrysokamino Metallurgy Workshop and its Territory (Hesperia Supplement 36): 241-256. Princeton, The American School of Classical Studies. Betancourt, P. P. 2006c. Survey conclusions. In P. P. Betancourt, The Chrysokamino Metallurgy Workshop and its Territory (Hesperia Supplement 36): 257-278. Princeton, The American School of Classical Studies. Betancourt, P. P., Davaras, C. and Simpson, R. H. 2005. Pseira IX. The Archaeological Survey of Pseira Island. Part 2. The Intensive Surface Survey (Prehistory Monographs 12). Philadelphia, Pennsylvania, INSTAP Academic Press. Branigan, K. 2001. Aspects of Minoan urbanism. In K. Branigan (ed.), Urbanism in the Aegean Bronze Age (Sheffield Studies in Aegean Bronze Age Archaeology 4): 38-50. London and New York, Sheffield Academic Press. Brogan, T. M. 2011. Introduction. In T. M. Brogan and E. Hallager (eds), LM IB Pottery. Relative Chronology and Regional Differences. Acts of a Workshop held at the Danish Institute at Athens in Collaboration with the INSTAP Study Centre for East Crete, 27-29 June 2007 (Monographs of the Danish Institute at Athens 11.1): 39-53. Athens, Danish Institute at Athens. Clark, J.E. and Parry, W. 1990. Craft specialization and cultural complexity. Research in Economic Anthropology 12: 289-346. Chisholm, M. 1968. Rural Settlement and Land Use (2nd ed.). London: Hutchinson. Christakis, K.S. 2008. The Politics of Storage. Storage and Sociopolitical Complexity in Neopalatial Crete (Prehistory Monographs 25). Philadelphia, Pennsylvania, INSTAP Academic Press. Day, P. M. 1997. Ceramic exchange between towns and outlying settlements in Neopalatial East Crete. In R. Hägg (ed.), The Function of the ‘Minoan Villa’. Proceedings of the Eighth International Symposium at the Swedish Institute at Athens, 6-8 June, 1992 (SkrAth 4o 46): 219-228. Stockholm, Paul Åströms Förlag. Dewolf, Y., Postel, F. and van Effenterre, H. 1963. Géographie préhistorique de la région de Malia. In H. van Effenterre and M. van Effenterre (eds), Fouilles exécutées à Mallia. Étude du site (1956-1957) et exploration des nécroploles II (1915-1928) (ÉtCrét 13): 2853. Paris, Paul Geuthner. Driessen, J. 2010. Spirit of place. Minoan houses as major actors. In D. J. Pullen (ed.), Political Economies of the Aegean Bronze Age: 35-65. Oxford and Oakville, Oxbow Books. Dierckx, H. M. C. 1999. Stone tools from Building BY, Area BB, and Area BR by room and context. In P. P. Betancourt and C. Davaras (eds), Pseira IV. Minoan Buildings in Areas B, C, D, and F (University Museum Monograph 105): 217-21. Philadelphia, Pennsylvania, The University Museum.

Fotou, V. 1993. New Light on Gournia: Unknown Documents of the Excavation at Gournia and Other Sites on the Isthmus of Ierapetra by Harriet Ann Boyd (Aegaeum 9). Liège, Université de Liège and University of Texas at Austin. Galaty, M. L., Nakassis, D. and Parkinson, W. A. 2011. Introduction: Why redistribution. In M. L. Galaty, D. Nakassis, and W. A. Parkinson (eds), Redistribution in Aegean Palatial Societies. AJA 115.2: 176-77. Garraty, C. P. and Stark B. L. (eds). 2010. Archaeological Approaches to Market Exchange in Ancient Societies. Boulder, University of Colorado Press. Haggis, D. 2005. Kavousi I. The Archaeological Survey of the Kavousi Region (Prehistory Monographs 16), Philadelphia, Pennsylvania, INSTAP Academic Press. Hallager, E. 1978. Two Linear A tablets from the GreekSwedish Excavations, Kastelli Khania 1977. SMEA 19: 35-47. Halstead, P. 1992. The Mycenaean palatial economy: making the most of the gaps in the evidence. PCPS 38: 57-86. Halstead, P. 2007. Toward a model of Mycenaean palatial mobilization. In M. L. Galaty, and W. A. Parkinson (eds), Rethinking Mycenaean Palaces II. 2nd rev. ed. (Cotsen Institute of Archaeology Monograph 60): 66-73. Los Angeles, Cotsen Institute of Archaeology. Hayden, B. J. 2004. Reports on the Vrokastro Area, Eastern Crete. Volume 2: The Settlement History of the Vrokastro Area and Related Studies (University Museum Monographs 119). Philadelphia, University of Pennsylvania. Hayden, B. J. 2005. Reports on the Vrokastro Area, Eastern Crete. Volume 3: The Vrokastro Regional Survey Project: Sites and Pottery (University Museum Monographs 123). Philadelphia, University of Pennsylvania. Hawes, H. B., Williams, B. E., Seager, R. B. and Hall, E. H. 1908. Gournia, Vasiliki and Other Prehistoric Sites on the Isthmus of Hierapetra, Crete, Philadelphia, The American Exploration Society. Jones, G. and Smith, I. 2009. Plant remains from Block AF. In P. P. Betancourt, Pseira X: The Excavations of Block AF (Prehistory Monographs 28):125-126. Philadelphia, Pennsylvania, INSTAP Academic Press. Killen, J. T. 2008. The Mycenaean Economy. In Y. Duhoux and A. Morpurgo Davies (eds), A Companion to Linear B: Mycenaean Greek Texts and Their World, Vol. 1 (Bibliothèque des Cahiers de l’Institut de linguistique de Louvain, Antiquité 120): 159-200. Louvain-la-Neuve, Peeters. Koh, A. 2006. The Organic Residues of Mochlos, Crete: An Interdisciplinary Study of Typological and Spatial Function at an Archaeological Site. Unpublished PhD thesis, University of Pennsylvania. Kopaka, K. and Platon, L. 1993. ΛΗΝΟΙ ΜΙΝΩΙΚΟΙ. Installations minoennes de traitement des produits liquides. BCH 117: 35-101. Martlew, H. and Beck, C. W. 1999. Minoan and Mycenaean drinking habits: wine with herbs. In Y. Tzedakis and 48

Kostis S. Christakis: The Neglected ‘Fields’ of Proto-Urban Living

H. Martlew (eds), Minoans and Mycenaeans. Flavours of Their Time: 163-164. Athens, Ministry of CultureKapon Editions. Michailidou, A. 1999. Systems of weight and relations of production in Late Bronze Age Crete. In A. Chaniotis (ed.), From Minoan Farmers to Roman Traders. Sidelights on the Economy of Ancient Crete: 87-113. Stuttgart, Steiner. Moody, J. 2012. Hinterlands and hinterseas: resources and production zones in Bronze Age and Iron Age Crete. In G. Cadogan, M. Iacovou, K. Kopaka, and J. Whitley (eds), Parallel Lives: Ancient Island Societies in Crete and Cyprus (British School at Athens Studies 20): 233-271. London, The British School at Athens. Mylona, D. 2004. The fish remains. In J. S. Soles, C. Davaras, J. Bending, T. Carter, D. Kondopoulou, D. Milona, M. Ntinou, A. M. Nicgorski, D. R. Reese, A. Sarpaki, W. H. Schoch, M. E. Soles, V. Spatharas, V., Z.A Stos-Gales, D. H. Tarling, and C. Witmore. Mochlos IC. Period III. Neopalatial Settlement on the Coast: The Artisans’ Quarter and the Farmhouse at Chalinomouri. The Small Finds (Prehistory Monographs 9): 121-25. Philadelphia, Pennsylvania, INSTAP Academic Press. Palmer, R. 1994. Wine in the Mycenaean Palace Economy (Aegaeum 10). Liège, Université de Liège and University of Texas at Austin. Palmer, R. 1995. Linear A commodities: a comparison of resources. In R. Laffineur and W.-D. Niemeier (eds), POLITEIA: Society and State in the Aegean Bronze Age. Proceedings of the 5th International Aegean Conference, Heidelberg, 10–13 April 1994 (Aegaeum 12): 133-56. Liège, Université de Liège and University of Texas at Austin. Parkinson, W. A., Nakassis, D. and Galaty, M. L. 2013. Introduction. In W. A Parkinson, D. Nakassis, and M. L. Galaty (eds), Crafts, Specialists, and Markets in Mycenaean Greece. AJA 117: 413-22. Peyronel, L. 2010. Ancient Near Eastern economics: The silver question between methodology and archaeological data. In P. Matthiae, F. Pinnock, L. Nigro, and N. Marchetti (eds), Proceedings of the 6th International Congress on the Archaeology of the Ancient Near East May, 5th-10th 2008, ‘Sapienza’ - Università di Roma Volume 1: 925-48. Rome, Harrassowitz Verlag -Wiesbaden. Petruso, K. M. 1992. Keos VIII. Ayia Irini: The Balance Weights. Mainz on Rhine, von Zabern. Polychronakou-Sgouritsa, N. 2000. Storage and safekeeping of goods in the LC I settlement at Akrotiri, Thera. ArchEph: 65-94. Reese, D. S. 1998. Investigations in Building AC: The faunal remains. In P. P. Betancourt and C. Davaras (eds), Pseira II. Building AC (the ‘Shrine’) and Other Buildings in Area A (University Museum Monograph 94): 131-32. Philadelphia, The University Museum. Reese, D. S. 2009. Faunal remains from Block AF. In P. P. Betancourt, Pseira X: The Excavations of Block AF

(Prehistory Monographs 28):131-42. Philadelphia, Pennsylvania, INSTAP Academic Press. Rose, M. 1994. With Line and Glittering Bronze Hook: Fishing in the Aegean Bronze Age. Unpublished PhD Thesis, University of Indiana. Sarpaki, A. and Bending, J. 2004. Archaeobotanical assemblages. In J. S. Soles, C. Davaras, J. Bending, T. Carter, D. Kondopoulou, D. Milona, M. Ntinou, A. M. Nicgorski, D. R. Reese, A. Sarpaki, W. H. Schoch, M. E. Soles, V. Spatharas, Z. A. Stos-Gales, D. H. Tarling, and C. Witmore, Mochlos IC. Period III. Neopalatial Settlement on the Coast: The Artisans’ Quarter and the Farmhouse at Chalinomouri. The Small Finds (Prehistory Monographs 9):126-131. Philadelphia, Pennsylvania: INSTAP Academic Press. Schoep, I. 2002. The Administration of Neopalatial Crete. A Critical Assessment of the Linear A Tablets and Their Role in the Administrative Process (Minos Supplement 17). Salamanca, University of Salamanca. Smith, M. L. 2003. Introduction: the social construction of ancient cities. In M. L. Smith (ed.), The Social Construction of Ancient Cities. Washington and London, Smithsonian Books. Smyth, M. P. 1989. Domestic storage behavior in Mesoamerica: an ethnoarchaeological approach. Archaeological Method and Theory 1: 89-138. Soles, J. S. 1997. A community of craft specialists at Mochlos. In R. Laffineur and P. P. Betancourt (eds), TEXNH. Craftsmen, Craftswomen, and Craftsmanship in the Aegean Bronze Age. Proceedings of the 6th International Aegean Conference, Philadelphia, Temple University, 18-221 April 1996 (Aegaeum 16): 425-31. Liège, Université de Liège and University of Texas at Austin. Soles, J. S. 2003. Mochlos IA. Period III. Neopalatial Settlement on the Coast: The Artisan’s Quarter and the Farmhouse at Chalinomouri. The Site (Prehistory Monographs 7), Philadelphia, Pennsylvania, INSTAP Academic Press. Soles, J. S. 2004. New Construction at Mochlos in the LM IB Period. In L. P. Day, M. S. Mook, and J. D. Muhly (eds), Crete Beyond the Palaces: Proceedings of the Crete 2000 Conference (Prehistory Monographs 10): 153-62. Philadelphia, Pennsylvania, INSTAP Academic Press. Soles, J. S. 2008. Metal hoards from LM IB Mochlos, Crete. In I. Tzachili (ed.), Aegean Metallurgy in the Bronze Age. Proceedings of an International Symposium Held at the University of Crete, Rethymnon on November 19-21, 2004: 143-56. Athens, University of Crete. Soles, J. S. and Davaras C. 1992. Excavations at Mochlos, 1989. Hesperia 61: 413-45. Soles, J. S. and Davaras, C. 1994. Excavations at Mochlos, 1990–1991. Hesperia 63: 391-436. Soles, J. S. and Davaras, C. 1996. Excavations at Mochlos, 1992-1993. Hesperia 65: 175-230. Sjöberg, B. L. 2004. Asine and the Argolid in the Late Helladic III Period: A Socio-Economic Study (BAR-IS 1225). Oxford, Archaeopress. 49

Country in the City Snowden, A. and Jones, G. 1995. The plant remains, Building AD centre. In P. P. Betancourt and C. Davaras (eds), Pseira I: The Minoan Buildings on the West Side of Area A (University Museum Monograph 90): 132. Pennsylvania, The University Museum. Stark, B.L. and Garraty, C. P. 2010. Detecting marketplace exchange in archaeology: a methodological review. In C. P. Garraty and B. L. Stark (eds), Archaeological Approaches to Market Exchange in Ancient Societies: 3358. Boulder, University of Colorado Press. Tsipopoulou, M. and Dierckx, H. 2006. Υστερομινωικό ΙΑ σπίτι Ι.1 στον Πετρά Σητείας: δομή, λειτουργία, και κατανομή των ευρημάτων. Πεπραγμένα Θ΄ Διεθνούς Κρητολογικού Συνεδρίου, A1: 298-315. Heraklion, EKIM Watrous, L. W., Haggis, D. Nowicki, K. Vogeikoff-Brogan, N. Schultz. M. 2012. An Archaeological Survey of the Gournia Landscape. A Regional History of the Mirabello Bay, Crete, in Antiquity (Prehistory Monographs 37). Philadelphia, Pennsylvania, INSTAP Academic Press. Watrous, L. V. and Schultz, M. 2012. Middle Minoan IIILate Minoan I periods: the rise of a regional state. In L. W. Watrous, D. Haggis, K. Nowicki, N. VogeikoffBrogan, M. Schultz. An Archaeological Survey of the Gournia Landscape. A Regional History of the Mirabello Bay, Crete, in Antiquity (Prehistory Monographs 37): 51-63. Philadelphia, Pennsylvania, INSTAP Academic Press.

Watrous, L. V. and Heimroth, A. 2011. Household industries of Late Minoan IB Gournia and the socioeconomic status of the town. In K. T. Glowacki and N. Vogeikoff-Brogan (eds), ΣΤΕΓΑ: The Archaeology of Houses and Households in Ancient Crete (Hesperia Supplement 44): 199-212. Princeton New Jersey, The American School of Classical Studies. Whitelaw, T. 2001. From sites to communities: defining the human dimensions of Minoan urbanism. In K. Branigan (ed.) Urbanism in the Aegean Bronze Age (Sheffield Studies in Aegean Bronze Age Archaeology 4): 15-37. London and New York, Sheffield Academic Press. Whitelaw, T. 2012. The urbanisation of Prehistoric Crete: settlement perspectives on Minoan state formation. In I. Schoep, P. Tomkins and J. Driessen (eds) Back to the Beginning. Reassessing Social and Political Complexity on Crete during the Early and Middle Bronze Age: 114-76. Oxford and Oakville: Oxbow Books. Whitelaw, T. 2019. Feeding Knossos: exploring economic and logistical implications of urbanism on Prehistoric Crete. In D. Garcia, R. Orgeolet, M. Pomadère, and J. Zurbach (eds), Country in the City. Forms and Functions of Agro-pastoral Activities in Mediterranean Pre-Classical Cities (Aegean and Western Mediterranean Protohistory): 91-124.

50

The Management of Agricultural Resources in the Minoan Town of Malia (Crete) from the Middle Bronze Age to the Early Late Bronze Age Maria Emanuela Alberti

Università degli Studi di Firenze, Dipartimento SAGAS

Sylvie Müller Celka

Université de Lyon, CNRS, UMR 5133-Archéorient

Maia Pomadère

Université Paris 1-Panthéon Sorbonne, UMR 7041-ArScAn

For most of the 20th century, discussions concerning the emergence of palaces and urbanisation in Crete put much emphasis on the question of agricultural surpluses and storage and palatial centralisation and redistribution (Renfrew 1972; Halstead 1981, 1997; Barret and Halstead 2004; Nakassis et al. 2011: 175-184; Privitera 2014). These questions have recently been supplanted by new interest in the elite ideology and its use for the control of power as illustrated by feasting contexts (Carinci 2001; Hamilakis 2002; Driessen 2009). This new interest often resulted in an analysis of Minoan urbanism focused on sacred dimensions and ritual activities, to the detriment of more prosaic aspects of daily human and animal sustenance (Orgeolet and Pomadère 2011, 2015). However, it is clear that the urban population’s food supply and the processing of agricultural produce are major issues and should be central for understanding how these societies operated (Hanson 1995). Investigations should therefore concentrate on the archaeological traces that these activities have left in the towns, on the role of these activities among other urban functions and on the way the domestic processing and storage structures of agricultural resources compare with the collective and palatial ones.

However, at the present time, our reconstruction is bound to be partial and preliminary: the data from the old excavations at Malia lack information about ‘ecofacts’ as these were not systematically recorded from the urban areas explored during the 20th century. The analyses of the data from the latest excavations are still ongoing (Gomrée, Langohr and Pomadère 2012; Pomadère et al. 2014; Devolder 2015-16) and future results will undoubtedly complement those presented here. The relevance of examining the relationship between the countryside and the town of Malia has been highlighted by recent studies on the natural environment of the site and the agricultural potential of the surrounding region. They show that the plain in which the palatial centre is located has major agricultural potential provided it is irrigated. Furthermore, the coastal marsh to the west of the site, which proved to have been there since the Neolithic period, provided a freshwater reservoir for consumption and irrigation, as well as a wetland which could have been cultivated after drainage or used for grazing, as was the case at Malia until the end of the 1990s. The slopes of Mount Selena, rising to an altitude of 1500m only 5km from the coast, provided altitudinal zones with clear farming possibilities for a variety of crops and for pasture (Dalongeville et al. 2001; Lespez et al. 2003; Müller Celka, Puglisi and Bendali 2014; Lespez, Müller Celka and Pomadère 2016).

Our study will focus on the relationships between the town and the countryside of Malia as far as urban supplies are concerned. It includes the management of flows of agricultural produce towards the palatial centre but also, to some extent, the agrarian practices themselves (Müller Celka 2007; Müller Celka et al. 2014; see Warren 2004; Bevan et al. 2013; van Gijn et al. 2014). Based on recent results from ongoing research at Malia (urban and regional survey, study of the palace, excavations in area Pi and the ‘Magasins Dessenne’), we can now address these questions anew and propose a tentative interpretation of the relationship between the town of Malia and the surrounding region.

In this contribution, we will approach the question of the importance of agriculture in the economy of the town of Malia from two perspectives. First, we will investigate possible evidence for intra-urban agriculture (horticulture, open cultivation, but also stabling and livestock farming) according to spatial and bioarchaeological analysis; then we will try to identify traces of the transformation and storage of agricultural 51

Country in the City

Sherd density map of the MM town of Malia

Postpalatial

Potamos

Haghia Varvara bay

Beach of the Pointe du Moulin

Chrysolakkos

Haghia Varvara House

Villa Alpha House Theta

Marsh

Mu

Pi Palace

Sherd density / m² 0,000000 - 0,695505

Epsilon

0,695506 - 1,909600 1,909601 - 3,673469 3,673470 - 6,145614

0 © EFA/Archéorient - S. Müller Celka (DAO G. Fantino - E. Régagnon)

100

200 m

6,145615 - 12,276400 Survey unit : 50mx50m

Figure 1: Sherd density map of the urban area at Malia (S. Müller Celka, © EFA and Archéorient)

to a denser settlement pattern during MM I, with the appearance of residential neighbourhoods and large public buildings (palace, Chrysolakkos building I) linked by a network of paved streets and squares (Devolder 2016). This process continued until the end of MM II, as attested by Quartier Mu. Intentional town planning is made manifest through the layout, the maintenance and the stability of the road network and large paved esplanades (Van Effenterre 1980: 265-279; Gomrée 2013: 923-924; Gomrée and Pomadère forthcoming; Pomadère et al. 2015-16).

produce within the town as indicated by archaeological remains. The development of urban space from Proto- to Neopalatial times and the possible role of intraurban agriculture Urbanisation during Middle Minoan I-II The process of urbanisation in Crete is still poorly known, as researchers have concentrated so far primarily on the monumental aspects of town planning. The results of excavations and surveys carried out over the past century show that the town of Malia was densely built up during the MM period (Poursat 1988; Schoep and Knappett 2003; Müller Celka 2007) (Figure 1). They now allow us to better follow the development of the site from EM IIB, when several habitation clusters were scattered over the whole area between the beach of the ‘Pointe du Moulin’ to the west and the Haghia Varvara bay to the east, including a denselyconstructed area under the palace and to the west of it,

In the town west of the palace, there is no clear evidence for unbuilt areas (the few open spaces are espanades with paved or coated floor, particularely to the west, the east and the north of the palace (‘Agora’), cf. Deshayes and Dessenne 1959: 3-5, Van Effenterre 1969 and 1980: 189), nor did the excavated parts of the town provide archaeological traces of gardens, orchards or herding activities around or between the houses. It means that only limited agricultural activities took place in the Protopalatial town. Ovine and caprine livestock 52

Kostis S. Christakis et al.: The Neglected ‘Fields’ of Proto-Urban Living

Figure 2: Map of Protopalatial sites in the surveyed area

agricultural exploitation and animal husbandry in the immediate vicinity of the town to meet the needs of the growing urban population (for an early diversification of plant products, cf. Dalongeville et al. 2001: 87). The growing number of rural sites in the plain and in the surrounding mountains must have satisfied these new needs. In facts, the number of sites identified during the regional survey sharply increased between the end of the Early Minoan and the end of the Middle Minoan period. It reached a peak during MM II, the only phase of the Minoan period during which the Selena slopes between 300 and 800m were densely inhabited (Figure 2).

breeding, which is well attested by archaeozoological data (Pomadère et al. 2011), must therefore have been practised outside the town. The archaeozoological study of Quartier Mu suggests that there were no stables in this neighbourhood (Vila and Helmer 2013: 151, 155). The numerous stone containers (‘gournes’) from the south-east part of the ground floor of Quartier Mu are usually described as ‘troughs’ for livestock but this is not corroborated by micromorphological analysis. However, animals may have been kept there occasionally before slaughter (Schmid and Treuil 2017: 264-266). Area Pi also appears to be a place of slaughter and consumption with little evidence of onsite breeding as only a few bones of perinatal ovi-caprids were retrieved from the MM II levels in court 25 and room 17. Some records from the hieroglyphic ‘archives’ of Quartier Mu and the palace display high figures suggesting the administration of large quantities of goods that may have been stored onsite, or controlled from a distance in the case of herds (cf. Oliver and Godart 1996: 142, 146, 152, 156, 170; at Mu, #090 (7000 units), #095 (270 and 890 units); Palace: #105 (210 units), #118 (240 units), #120 (300 units); cf. also Pelon 1983: 702-703 for the MM III date of the ‘dépôt hiéroglyphique’ of the palace).

Urbanisation must have entailed an increase in the proportion of the population not, or not exclusively, involved in farming. In Quartier Mu, the production of seals, pottery, metal objects and stone vases is based on non-agricultural raw materials and implies the presence of craftsmen who were dependent on food supply structures. This holds true for other areas: discoveries in area Pi also include stone craft (production of vases and seals), which confirms the hypothesis made in 1985 by J.-Cl. Poursat: ‘the coexistence, at the same period in Malia, of several similar points or zones of activity’ (Poursat 1985: 299) covering a significant surface of the town during MM IIB.

The urban expansion of Malia during the Middle Minoan period probably involved an intensification of 53

Country in the City This brief overview of the archaeological record shows that, although part of the urban population may have been engaged in farming during the Protopalatial period, Malia mainly was a craft centre and a harbour town open to trade with the rest of Crete and, to a lesser extent, the eastern Mediterranean (Poursat and Loubet 2005; Carter and Kilikoglou 2007; Müller Celka 2007). This vibrant town was totally destroyed by fire at the end of MM IIB. This destruction constitutes a major, or at least most visible, breakdown in the history of the site.

us to refine the picture by confirming that a major break took place during MM III, leading to the conclusion that the Neopalatial period should not be considered as monolithic. O. Pelon had already identified the distinctive character of MM III in Quartier Epsilon (Pelon 1970: 167-168: phase II). The emerging picture, although still incomplete and provisional, is one of an increasingly loose urban pattern from MM  III to LM  I (Pelon 1970, 161: 168). Even though this process seems to be gradual, we probably underestimate the possibility of disruptions, as suggested by the destruction traces in Epsilon alpha and in area Pi, which could be related to one or several seismic events during MM III (Pelon 1970: 161). Surface sherds show that activity during LM I concentrated on the central area, including the palace, the northeast outskirts and the Epsilon house, but also on the eastern edge of the town in the area of the Haghia Varvara house and around the Haghia Varvara bay (Pelon 1966; Müller 1991), as well as on the western side of the plateau, east of the present-day beach of the Pointe du Moulin. In between, there may have been open spaces with more or less ruined buildings.

Urban pattern changes during the Neopalatial period (MM III-LM I) The Neopalatial town offers a different picture, although changes did not affect the whole urban area in the same way. Some of the neighbourhoods were rebuilt following the orientation of older buildings (Quartier Delta; Epsilon building; Quartier Nu, see Devolder 20122013: 61) without any major modification to the urban structure (for a summarizing presentation, see Driessen and Macdonald 1997: 186-197). However, a significant number of them were not, or not completely, rebuilt, as demonstrated by the following non-exhaustive list:

What happened to these vacant spaces during MM III and LM I? A hypothesis is that they were transformed into cultivated areas (gardens, orchards) or grassland with stables or sheepfolds. Likewise, the northwest quarter of the Neopalatial palace (length 31-38m, width 25m) has long been considered to have been replaced by gardens and this hypothesis is backed up by the fact that room III,7a opened onto it through a colonnade (Chapouthier and Demargne 1962: 35, 37, 64; Pelon 1980: 235-242) – of course these gardens could have been ornamental rather than horticultural. So far the interpretation of the available record remains problematic and the presence of archaeological ‘voids’ cannot systematically be attributed to an agricultural or other non-residential use of the area. Invisible frontiers depending on political, symbolic and cultural parameters may have existed without leaving any archaeological traces. In this way, the ruins of Quartier Mu, which was a monumental complex and the seat of an administrative authority during the Protopalatial period, could have been deliberately left visible to preserve the memory of this, according to the hypothesis of damnatio memoriae. In contrast, some wasteland or ruined areas were used as dumps (for instance the space between the area Pi and the house Delta alpha, Pomadère et al. 2015-16: 924-928). At any rate, the analysis of faunal remains and micromorphological samples invalidated the hypothesis of a reuse of the abandoned areas for stabling (P. Mylona, pers. com.). For the time being, the carpological and anthracological remains from the excavations in the urban areas provide information on the cereals, fruit, vegetables and trees discovered in the town but do not tell us whether they were grown there

–– the southern and eastern edges of the Delta alpha house, as well as the north and northeast parts of area Pi; –– Quartier Mu (it remained buried beneath a crust of burnt bricks); –– the Hypostyle Crypt (it undoubtedly functioned in relation to the ‘Agora’ and it underwent a very limited Neopalatial reoccupation, perhaps restricted to MM III, cf. Amouretti 1970: 79- 80); –– the area north-west of the palace (it was abandoned, in contrast with the north-east area reflecting ‘the growing hold of the Palace on its immediate surroundings’, cf. Darcque 2014: 180); –– the ‘Magasins Dessenne’ to the south-west of the palace (Devolder 2015-16); –– the Epsilon beta house (it was abandoned, while the surface of the Epsilon alpha house was first reduced, then progressively abandoned during MM III and LM IA, cf. Pelon 1970: 160-162, pl. XXXIV; Bradfer- Burdet 2004: 65). –– the Sanctuary of the Horns and the MM II Sanctuary; –– Building Beta The building complex Gamma is probably to be added to the list of those abandoned after MM II, or possibly after MM IIIA (Demargne and Gallet de Santerre 1953: 23-39, pl. XIV). The results of the archaeological survey also show a decrease of the urban reach during the Neopalatial period, at least in terms of density, if not in terms of surface (Puglisi 2007). Recent research in area Pi allows 54

Kostis S. Christakis et al.: The Neglected ‘Fields’ of Proto-Urban Living

or imported. Textual data provide similar information but rarely specify the type of produce (cf. supra for the references in Olivier and Godart 1996). The botanical macro-remains identified in the town can thus only lead us to imagine that vines, vegetables and legumes (lentils, beans, peas) and aromatic herbs were grown in the gardens, shaded by olive, almond, fig, pear and palm trees (non-exhaustive list) (Sarpaki 2007; Pomadère 2012-2013: 649, 869; M. Ntinou, pers. com.).

go hand in hand with an evolution of the settlement pattern in the whole region. Nearly all the settlements above 300m in altitude, consisting of Protopalatial small isolated habitation units, were abandoned after MM II and all the Neopalatial settlements were found in the lowlands below 300m in altitude (Puglisi 2007; Müller Celka, Puglisi and Bendali 2014) (Figure 3). Therefore, the decline in the overall number of sites in MM II must be weighed against the concomitant growing number of settlements in the plain, and it is equally essential to determine urban development in relation to the dynamics of regional settlement. The ‘descent’ towards the plain could have resulted from a climate change (Macgillivray, Sackett, and Driessen 2007; Moody 2000; Kaniewski et al. 2013; Flood and Soles 2014; Vokotopoulos et al. 2014) that would have made agriculture in the mountains difficult and/ or from the simultaneous development of irrigation techniques that would have enabled the exploitation of the agricultural potential of the plain (Müller Celka et al. 2014; Müller Celka 2015; Müller Celka 2017). An alternative or complementary hypothesis is that the slopes were reserved for herds after a move towards a specialised pastoral economy possibly controlled by the palace (as already suggested for the LM III period, Zurbach 2007: 871). In both cases, archaeological and geoarchaeological data are still lacking to substantiate these hypotheses. We should also take into account the fact that secondary large settlements presenting similarities to the Malia town emerged in the plain, as indicated in particular by the discoveries on the Bouffo Hill at Sissi, less than 5 km from Malia (Driessen et al. 2009; Driessen et al. 2011; Driessen et al. 2012; Driessen 2016; Driessen et al. 2018). Lastly, the Neopalatial town was perhaps physically more open onto the countryside than the Protopalatial one, without any well-defined limits, since isolated houses similar to the Haghia Varvara House (Pelon 1966) were spotted during the survey in the eastern part of the town (Müller Celka 1992: 747). However, the question of a circuit wall surrounding the Protopalatial town is still under debate (Zielinski 1998; Alusik 2007: 87-93). The walls along the coast to the west of Chrysolakkos and the large ‘oblique wall’ at the eastern end of the beach (Van Effenterre 1976: 2) served as retaining walls, and nothing indicates that they rose above the terraces that they supported. They appear to have been abandoned after the Middle Minoan period (Van Effenterre 1980: 266). Moreover, no circuit wall was identified on the southern border of the town. All this means that boundaries between town and countryside were perhaps equally loose during the Protopalatial period.

These gardens or orchards within the town, or close to it, had to be watered, which was possible thanks to freshwater springs located in the coastal marsh and on the coast, in front of the Theta house (now underwater) and to the east of the Chrysolakkos promontory. Water would have been carried by men or donkeys (the latter are poorly represented in the  archaeozoological  remains and were therefore probably not eaten). As no Bronze Age well or cistern has been identified in the town so far, water for human and animal consumption, as well as for irrigation of urban gardens, must have been transported and stored in leather, wood or terracotta containers. In brief, from MM III onwards, and especially during LM I, the lack of reconstruction in several areas indicates a major change (Driessen and Macdonald 1997: 186193; Cunningham and Driessen 2003: 105; Poursat 2010: 259-267). For the time being it cannot be interpreted straightforwardly as ‘ruralisation’. However, agriculture appears to have supplanted craft activities in the urban economy. The demographic decline appears to have gone hand in hand with a weakening of urban crafts (with the notable exception of a MM III metallurgical workshop north of the palace, Pelon 1987: 269-271). Unlike the preceding phase, craft activity is very poorly represented in the Neopalatial town, especially during LM I, except for the processing of agricultural produce (we will return to this point below). The decline of craft activities could be related to difficulties in raw material procurement (Van Effenterre 1980: 75-79; Carter and Kilikoglou 2007). Nonetheless, this decline is only relative as it does not affect the whole of the urban area in the same way. Indeed, Neopalatial buildings, in particular the new palace, but also the houses of the new Quartier Zeta, the Epsilon building and the Neopalatial walls underneath Quartier Nu, demonstrate the permanence of an urban elite (Demargne and Gallet de Santerre 1953: 73; Deshayes and Dessenne 1959: 18-19; Devolder 2012-2013: 55, 60, 63-64). Note also that the road network developed during this period, which indicates that public infrastructures were not abandoned, but maintained and adapted (Gomrée 2013).

To conclude this section, it appears that the urban nature of the settlement at Malia during LM I is questionable. Although the variety of activities and the population density decreased after the Middle Minoan period, the maintenance of town planning and written

The reduction of the settlement area and the modifications of the urban landscape during MM IIILM I must be placed in a wider perspective, as they 55

Country in the City

Figure 3: Map of Neopalatial settlements on the surveyed area (S. Müller Celka, E. Régagnon)

From processing to storage: continuity of the equipment and working methods through MM-LM

administrative records as well as the hierarchic place of Malia in the regional settlement pattern lead to a positive answer (contra Darcque 2014: 177; about the criteria defining a town, May and Steiner 2014; Orgeolet and Pomadère 2014). It is clear, indeed, that Malia was still controlling, at least partly, the surrounding countryside during MM III and LM I, although probably according to new modalities.

Processing agricultural produce in the town of Malia The processing of agricultural produce and secondary products from animal husbandry is well attested in the town of Malia during the Protopalatial and Neopalatial periods. We will briefly review the data related to oil and wine making, milling, honey extraction and weaving, which constitute most of the currently available evidence on this topic.

The transformation of Malia into a rural site becomes perceptible, in the current state of research, only during the LM III or Postpalatial period. The town was then almost completely abandoned. It was only concentrated around two areas, between the Quartier Nu and the coastal marsh to the west, and around the Haghia Varvara bay to the east, and it can no longer qualify as a ‘town’. Note that at that time, the presence of caprine and pig husbandry is attested in Quartier Nu through archaeozoological data (with the presence of animals of perinatal age, Vila and Helmer 2007: 886), and appears to have been new development in this area.

Domestic production of oil and wine in town The information from the earlier excavations we have to rely on is rather vague. Containers such as wine or oil presses have been identified in several excavations at Malia but only a few of them can really qualify as such. We consider that oil presses in Minoan Crete have a very low press bed and can be associated with counterweights (which are not however necessary for small quantities), whereas wine presses are basespouted deep vats (λήνος), often located on a low wall for discharge into a large storage jar or pithos (collector) (Kopaka and Platon 1993: 61-63, fig. 3, 5; Blitzer 1993; Hamilakis 1996; Shaw 1996: 117-118, 126129, pl. 2.165-166, 2.176, 2.178, 2.181; Foxhall 2007; Hadjisavvas and Chaniotis 2012; for parallels from the

In the following, we will examine whether these changes in the morphology and economy of the town over a long period (MM-LM I) are also perceptible in the processing and storage patterns of agricultural produce. In a diachronic perspective, we will first define the type and distribution of processing and storage activities in town before raising the question of centralisation. 56

Kostis S. Christakis et al.: The Neglected ‘Fields’ of Proto-Urban Living

ancient Mediterranean, Hadjisavvas 1992 and 2003; Eitam and Heltzer 1996; Brun 2003 and 2004; McGovern 2009-2010; Rowan 2015). According to these criteria, only two cases of wine presses can be accepted for Malia in the inventory published by K. Kopaka and L. Platon. The one from Quartier Gamma is a complete device with a collector in situ dating to the Protopalatial period (Kopaka and Platon 1993: 44; Demargne and Gallet de Santerre 1953: 28; Van Effenterre 1980: fig. 238. The date is MM II rather than MM I); the other one comes from room XXVIII,1 in the palace and could date to MM III or LM I (Van Effenterre, 1980: 368-370; Pelon 1992: 24). Whether the grapes were pressed in the room where this container was discovered is doubtful as the room is said to have a ‘beautiful stucco floor’. Was the container simply stored there? Chapouthier argued that the vat was used together with a large pithos sunk into the floor, but the pithos dates to the Protopalatial period (Chapouthier and Demargne 1942: 18, 47, fig. 24). No residue or bioarchaeological marker allows us to confirm the function of these containers, and we could thus assign them other uses in local artisanal activities. The same is true for other types of large containers reported from several Protopalatial (Quartier Mu, building Epsilon) and Neopalatial houses (various specimens in the Quartier Zeta), some of them with a pouring spout and/or a drain hole at the base. They were identified as presses by the excavators and could correspond to a simpler way of producing oil, without a press, by placing previously crushed olives in water (this process was attested in Crete until the beginning of the 20th c. AD, cf. Brun 2004: 78-79; in Lebanon, Chanesaz 2006: 18). On account of their shape, they could also suit various activities using water or other liquids for maceration, such as washing or dyeing (Poursat and Knappett 2005, n°241; Demargne and Gallet de Santerre 1953: 90, n°25-26, pl. XLI.7-8; 91, pl. XXXIX.1-2, LVIII.4: a furnace adapted for dyeing; Alberti 2007 and 2008).

Figure 4: Area Pi, room 12 (EFA, C. Gaston)

weathered and the walls partly reinforced by limestone slabs (dimensions of the vats: length 0,53m, width 0,520,53m, depth 0,45-0,49m); they were probably inserted in a low platform. We can reconstruct the presence of a small hearth near the paved area in the southeast angle. The absence of olive remains in this area is not necessarily problematic as olive stones could have been retrieved and recycled for other needs (fuel or olive pomace to keep pests away, Russel 2003). However we cannot rule out uses in other processing activities for these installations (cf. infra).

The archaeological record concerning oil presses is also ambiguous. At Malia, no apparatus for crushing olives could be identified, yet it is the first necessary operation in oil production, to be carried out with a pestle or a wooden or stone roller (Chanesaz 2006: 1831; Callot 1987: 203). After crushing, the olive paste had to be pressed, but no press elements have been identified at Malia so far. A marble ‘pierre à cupules’ (length 0.58m, width 0.34m, thickness 0.17-0.20m) with a central hollow that was discovered recently in area Pi could have served in this chaîne opératoire: it was secured by wedging stones in the small room 12 (LM IA) and it bears traces of wear (Figure 4). However, we consider it to be working surface rather than a press element. The table was associated with two stone vats made of calcarenite (the sandstone locally called ammouda) that were half-buried in the floor and could also have been used for crushing. The bottom of these containers was

Similar questions arise about the MM II space 29 in area Pi, where another sandstone vat (diameter 0.35m, depth 0.55m) was sunk into the floor and carefully wedged at the end of a small wall. This wall could have supported a press table, according to an apparatus well-known in the Minoan world, and the stone container could have received the juice from the press; the V-shaped cut on the rim of the stone container might have been intended for pouring liquids (Figures 5a-b). On the floor immediately to the south of the vat, several charred olive stones were discovered. Very few objects were found in this room. However, it is worth noting that the vat was part of a fixed, permanent installation. The other numerous sandstone containers discovered in the houses of Malia do not seem to be part of press installations. But they may have served a wide range of 57

Country in the City

Figure 5a: Area Pi, vat in space 29: receptacle from an oil press? (EFA, M. Pomadère)

Figure 5b: Oil press at Vathypetro, LM I (© Asta Adomaityte, http:// www.minoancrete.com/ vathypetro.htm)

daily activities, including husking cereals, as far as the deepest vats/mortars are concerned (Procopiou, pers. com. and see below).

period. One of them comes from the south part of the Epsilon alpha house that was abandoned after MM II (Pelon 1970: 141, 157-160, pl. VII.2). Two further specimens were discovered by A. Dessenne in room IV.14 of Quartier Mu and published as hypothetical ‘votive anchors’ (Poursat 1992: 24; unfortunately these objects have disappeared). Although their function as anchors cannot be ruled out, their presence in one of the largest rooms of Building B with channels in the floor suggests

Some perforated blocks of calcarenite can be interpreted as counterweights of the type attested at Ugarit and Cyprus during the Late Bronze Age (Schaeffer 1978: 371-378; Callot 1987: 197-212). However, the specimens from Malia are older as they all date to the Protopalatial 58

Kostis S. Christakis et al.: The Neglected ‘Fields’ of Proto-Urban Living

that they possibly belonged to a lever press. One of the walls of this room is equipped with a very large recess, 0.21 m above the floor, 2.32 m wide […]’, which could have served to anchor a wooden apparatus including a lever (Detournay, Poursat and Vandenabeele 1980: 235-238, fig. 314.h; Chanesaz 2006: 22, fig. 1.11-1.13). An alternative interpretation is a large weighing scale, with weights similar to the famous octopus-decorated example from Knossos, sometimes also identified as a ‘sacred anchor’ (Petruso 1992: 38, pl. 7, n. 73; Peyronel 2011, scale type 2). However, no press element or tank was found in this unpaved area.

First, a sandstone grindstone was found on the floor of room16 near a stone bench on which it could have been used originally (Figure 6). This association reminds us of the grinding areas in Akrotiri (where, however, grindstones were fixed, Moundréa-Agrafioti 2002: 103106, fig. 6). A tripod pot and three stone hammers were also found on the earthen floor of the room, which otherwise was practically empty apart from a store in an angle containing loom-weights and conical cups. At least one of the stone hammers was associated with stone knapping (H. Procopiou, com. pers) as it was found next to numerous partly knapped obsidian blades (T. Carter, unpublished report, cf. Pomadère 2010: 589). Fragments of olives, almonds, and grape pips were also associated with the hammers, but no cereals or legumes. Several hypotheses can account for the absence of the latter in this room: the very alkaline nature of the Malia sediments, which does not allow for their preservation; the absence of destruction by fire in this area; the use of the grindstone for other purposes than food preparation, such as grinding mineral matter.

Identifiable structures clearly linked to oil and wine production are thus rare in the town of Malia, especially compared to the ratio observed at Ugarit during the Late Bronze Age (an oil-mill in one of five houses, Callot 1994: 196). This could be due to the use of simple techniques suited to domestic production, with non-specific structures (such as stone and terracotta containers) and perishable materials (canvas bags, wood; cf. for instance a seal from Chrysolakkos said to represent a fuller, CMS II,1, 420; Militello 2011, 246). We can also presume that olives or grapes were pressed in the countryside near the place they were grown, as suggested by a press fragment/ληνός identified in one of the rural settlements of Omales, on the northern slopes of Selena (Müller Celka et al. 2014: 434 and pl. CXXVIIIa).

Second, in Room 12 (that was accessed from room 16 to the east) were discovered the two large sandstone vats of unknown function mentioned above (Figure 4). The carpological remains do not allow us to confirm whether they were used for crushing (olives? Cf. supra) or for dehusking, as both olives and cereals are well

Cereal processing in the Pi area We will not comprehensively list the grinding tools found in the town of Malia, as such a list has already been partly drawn up (Procopiou 1998) and is beyond the scope of this study. The tools for milling and grinding form a detectable ‘background noise’ in all the excavated buildings and surveyed areas of the town, implying that grain processing was widespread in urban and rural houses. We should, however, keep in mind that some of these toolkits were possibly not used for food preparation but for processing mineral matter (Procopiou 2013). Cereals (particularly hulled cereals) formed the major part of the Minoan diet, which is why it is essential to focus on how they were processed, dehusked and ground. Now according to the study of the tools from Quartier Mu (MM II) by H. Procopiou and the preliminary results from A. Sarpaki’s carpological study of area Pi, cereals were mainly eaten as whole or crushed grains without systematic de-husking (Procopiou 2013: 57; Procopiou 2003 and 2014: 243-246; also in the nearby site of Sissi, Isaakidou et al. 2011: 217). Let us now concentrate on the material from area Pi. Unfortunately most of it was found in secondary position but two installations from the LM IA levels of the Neopalatial building are nonetheless interesting.

Figure 6: Area Pi, room 16 (EFA, C. Gaston)

59

Country in the City recorded in area Pi. The same type of device has only been identified in the palace (Procopiou 2013: 57; large mortars with the same destination have been found at Galatas: Procopiou 2014b: 242; for ethnoarchaeological and experimental studies, see Peña-Chocarro and Zapata, 2014: 226-232). As in room 16, obsidian blades provide evidence for the whole knapping chaîne opératoire. This area thus appears to have been multi -purpose but especially related to food production, with possible evidence of two different stages of cereal grinding (dehusking and crushing). Moreover, the large vats/mortars could be associated with a change in cereal consumption during the Neopalatial period, as they demonstrate greater efforts to refine grain. This change has been identified by H. Procopiou in palatial contexts and could thus characterize urban houses, perhaps following a new trend in food preparation that was instigated by the palatial elite (Procopiou 2014a: 239 and 2014b: 244-245). At any rate, these large vats point to activities exceeding domestic needs.

beta, and from the Villa Alpha (Pelon 1966: 568-572, fig. 13-14). The distribution of loom-weights in the town sharply differs from that in the surrounding region. During the survey, only a dozen isolated loom-weights were found around the Minoan town (in contrast to more than 200 in the urban area), mainly in the lowlands and in the foothills of Mount Selena and Arkovouno. No weights were found in the rural settlements on the slopes of Mount Selena (Figure 7) (Müller Celka et al. 2014: 435). This complete lack of loom-weights in the mountainous area could suggest that animal husbandry was geared towards meat (rather than wool) production during the MM. This hypothesis is further supported by the analysis of the faunal remains from Quartier Mu, which indicates the preferential consumption of young animals (Vila and Helmer 2007: 887). However, it is also possible that a significant part of animal husbandry was geared towards wool production if wool was transported to town to be woven. At any rate, this is what the faunal data from area Pi suggest, as they reflect the focus on secondary, milk-wool type products (A. Gardeisen, pers. comm.): animals were slaughtered when they were relatively old and could no longer supply milk and wool (‘cull’ animals). It is thus more likely that weaving took place mainly in town, as well as at a few sites in the plain.

Weaving in the urban area Weaving was carried out in all the houses of the town, in different types of rooms, during both the Proto- and Neopalatial periods. Concentrations of various types of loom-weights have been reported recurrently in the archaeological literature concerning Malia, pointing to the domestic production of various wool or linen textiles. In Quartier Mu, large quantities of the same type of weights were found in Buildings A, B and D, which could indicate, beyond daily domestic activity, the presence of a specialised, industrial production (Cutler et al. 2013; Poursat et al. 2015; on the evidence of textile in Hieroglyphic script, see Younger 2005: 405409). The presence of weights with seal impressions in several MM II contexts is noteworthy. In several cases, impressions of the same seal were found on weights from different places, which raises the question of the extent of administrative control on textile production at that time (in one case, the same impression was found on loom-weights from Quartier Mu, Quartier Epsilon and Abords Nord-Est and, in another, on loom-weights from Quartier Mu and Abords Nord-Est, Poursat 2013: 94). Information about weaving is more elusive in the Neopalatial buildings, where it appears to be limited to domestic scale. In area Pi, only two batches of similar loom-weights come from primary contexts: the first consists of 15 truncated pyramidal weights, mainly intended for production with threads of medium thickness; it comes from a Protopalatial exterior paved area (area 22); the second (mentioned above about the LM IA area 16) is composed of 12 discoid weights suited to fine threads for producing dense fabrics (Figures 8a-b). In the Neopalatial House of Haghia Varvara, a sandstone rectangular base with four cavities could constitute the base of a weaving loom; similar bases come from the palace, from houses Zeta alpha and Zeta

Preliminary carding operations are very difficult to detect from archaeological remains, and spinning is poorly attested both in town and in rural areas. Spindle -whorls only appear very sporadically at Minoan sites, probably because spinning was based on techniques and/or materials leaving no remains (Tzachili 1997: 126-129; Cutler et al. 2013: 102). However, three spindles found together with loom-weights in the Protopalatial level of room 3.2 on the north-eastern outskirts of the palace show that spinning and weaving were sometimes associated (Darcque 2014: 65, 74). We do not take into account here the looped-bowls or ‘spindle-bowls’, which could have been used for the processing of flax fibers (Barber 1991: 73-77) as well as for other functions (e.g. dippers). This picture of the processing of agricultural produce in the town of Malia must be completed by apiculture, even though the identification of large conical basins as beehives is relatively new and still the object of much debate (Melas 1999; D’Agata and De Angelis 2014; Müller Celka, Puglisi and Bendali 2014: 434; Harissis 2017: 22-24). In Minoan Crete, these conical basins mainly appear in rural settlements, including in the region of Malia on the slopes of Mount Selena, but considerable numbers of them also appeared in the town of Malia. Whole specimens come from Quartier Mu and from the Haghia Varvara House (Pelon 1966: 573-4, fig. 17, n° 3; Poursat and Knappett 2005: 209, n° 222-226; Darcque 60

Kostis S. Christakis et al.: The Neglected ‘Fields’ of Proto-Urban Living

Figure 7: Distribution map of loomweights on the surveyed area (S. Müller Celka, E. Régagnon).

Several conclusions emerge from this brief review of the evidence for farming related activities within the town of Malia:

2014: pl. 73, n° 0616-002). The evidence from area Pi is particularly rich, with beehive sherds found in several Protopalatial and Neopalatial layers. A single whole specimen (n°1018.28) comes from a destruction deposit in secondary position dating to the end of MM IIIB (Figure 9). However, these containers do not necessarily signify ‘urban’ apiculture: the hives could have been brought to town for extracting honey, or produced and stored in town before being brought to the countryside (Melas 1999). At any rate, their high frequency in town and in the surrounding countryside seems to point to the leading role of apiculture in the local economy.

–– During the MM and the LM I periods, urban houses seem to host a relatively wide range of onsite activities linked to farming and to the economy of food production, but the production of oil and wine intra muros is poorly attested whereas cereal processing and weaving appear more widely. –– whatever the context (private house or monumental building with large storerooms), the processing of 61

Country in the City

Figure 8a: Batch of pyramidal loomweights, MM II, area Pi, space 22 (EFA, C. Gaston)

Figure 8b:. Batch of discoid loomweights, LM IA, area Pi, room 16 (EFA, C. Gaston)

62

Kostis S. Christakis et al.: The Neglected ‘Fields’ of Proto-Urban Living

Storerooms with storage jars aligned on low benches against the walls and recovery channels leading to collecting pots are well evidenced at Malia. However, they were restricted to public buildings or to the residences of the elite during MM II (Palace, Crypt, Mu but also ‘Magasins Dessenne’). In simple houses and in large Neopalatial residences, jars or pithoi were simply placed on the floor against dividing walls (only one collecting jar could be of LM I or LM II-III date in house Epsilon, room XIII, cf. Deshayes and Dessenne 1959: 104; Bradfer-Burdet 2004: 62).

Figure 9: Beehive n°1018.28, MM IIIB, area Pi (EFA, Drawing B. Konnemann)

Small basement rooms, which were probably accessed through trapdoors, were also used as storage rooms by the Minoans (Treuil 1996: 80-83). They cannot compare with silos as only a few of them were suitable for long-term bulk storage. They could not be sealed hermetically and therefore did not offer a confined atmosphere, which was the only way to keep insects away – insects constitute a real danger in the Mediterranean climate due to humidity, which was particularly the case at Malia because of its coastal location (Sigaut 1979: 31; contra Privitera 2014: 438: the deposit of bulk products in spaces like the Bastione at Aghia Triada should be ruled out). Moreover, at least some of these small rooms are likely to have played a structural role in the architecture (Bradfer-Burdet and Pomadere 2011: 105). Therefore, these storerooms could have served as cellars for foodstuffs such as dried fruit, vegetables or grain only if these products were stored in closed containers. In some Neopalatial houses, pithoi with non-determined content were discovered in this type of room (e.g. room IV and XIV in Zeta beta, in this case on a ground floor accessible from the first floor: Deshayes and Dessenne 1959: 13, 16). In the centre of the Pi Neopalatial building, two small underground rooms of this type were identified, each of them with a height limited to 1m and an estimated capacity of 3 m3 to 4m3, but they were filled with destruction material during LM I and did not retain traces of their original function.

agricultural produce was carried out on a domestic scale (for similar conclusions on the production of chipped stone tools at Malia, Carter 2013: 23-24). In addition, the techniques do not appear to have changed much from one phase to another, except perhaps for more developed cereal dehusking during LM I. –– Only weaving was subject to more centralised management, as attested as early as MM IIB in Quartier Mu. It seems to have been partly controlled by the Mu authorities during the Protopalatial period, then by others, or perhaps by the palace, in the Neopalatial period. Apart from this production, which could have been partly intended for trade or exchange, there is no sign of large scale production either within or outside the palatial structure. On the contrary, the storage of foodstuffs was organised at different scales but centralisation significantly increased during the Neopalatial period, as we will see in the following. Agricultural storage in the town Estimating the storage capacities of the domestic units at Malia is out of reach as we lack reliable information on the type of storage techniques and containers used, on the final destination of grain (seed, animal or human consumption) and on the form in which cereals were kept (as cut cereals, straw, cobs or as hulled or naked grain) (Sigaut 1979: 31; Privitera 2010: 60-61; Privitera 2014; Keßler 2015: 139 -143). The remains of shelling discovered at the neighbouring site of Sissi could indicate that cereals were at least partly kept as spikelets (Isaakidou et al. 2011). Pre-cooking cereals (by steaming or parboiling, as for bulgur), would have enabled preserving them for a whole year after the harvest in closed pots, even if not perfectly airtight (Sigaut 1981; on pre-cooked cereals, see Valamoti et al. 2008). Following K. Christakis, we consider that the Minoans mainly kept their agricultural produce in terracotta containers for long-term storage, but other containers in perishable materials such as wicker baskets, cloth and leather bags, wooden casks and chests are likely to have been used too (Christakis 2008: 12-13).

Based on the pottery containers found in the houses of Malia, their storage capacity does not appear to have covered annual requirements either in the Protopalatial or in the Neopalatial period. This is in agreement with the results established by K. Christakis for the Neopalatial period in Crete (Christakis 2008: 109-118). Most of the assemblages in Malia comprise several medium size jars and amphorae (height 3060cm, capacity up to 65l but generally lower), several larger pithoid jars (height 70-80cm, capacity up to 95l) and one or two large pithoi (90-95cm high, with a capacity of up to 155l). Larger pithoi are exceptional and the pithoi of the palace of Malia as a whole are smaller than those of Knossos (Christakis 2008: 15; see also Keßler 2015: 145). In addition, it is important to keep in mind that some of these jars and pithoi may 63

Country in the City

Storage Patterns (K. Christakis 2008)

Protopalatial (calculations M.E. Alberti)

Neopalatial (K. Christakis 2008)

SP IV (2000 – 4000 lt)

Mu Bâtiment A

SP III (1000 – 2000 lt)

Mu Bâtiment B; Bâtiment Dessenne (?)

Maison Epsilon (mainly on architectural ground)

SP II (250 – 1000 lt) (do not cover one year’s need/household)

Mu Bâtiment C Mu Atelier de Potier Mu Bâtiment E Mu Atelier de Sceaux Nu (deep soundings) ANE Bâtiment I

Maison Zeta beta Maison Delta alpha Maison Delta gamma (ST II-III) Maison Epsilon alpha Maison Zeta alpha Maison Zeta gamma (or SP III?) Others Bâtiment Pi (hypothetic, based on the few primary contexts)

Figure 10: Compared storage capacities from Protopalatial and Neopalatial houses (M.E. Alberti). Non-domestic structures are not included (Hypostyle Crypt and Palace).

have been used for water storage, as the houses were not provided with water adduction or with wells or cisterns. Only the largest building of each area had greater storage capacity: for the Protopalatial period, these are buildings A and B in Mu, and most probably the ‘Bâtiment Dessenne’ (Devolder et al. 2015-2016); in LM I, we can mention houses Epsilon and Zeta beta (Figure 10) (Christakis 2008, table 10). No figures can be proposed for building Pi, as primary Neopalatial contexts are too few. However, the composition of the storage assemblages is much the same as in the other houses (many jars and very few pithoi).

(Driessen and Macdonald 1997: 101, 185). However, their spatial proximity to the palace suggests that they were under the control of the palatial authority, whatever its type, personal or collective (Christakis 2014: 210). Two Linear A tablets from room III.8 in the palace record large quantities of *180, which could also point to centralisation (GORILA 1, Ma 4 and Ma 6; the same product is mentioned on roundel MA Wc 7, from the same place). In addition to their practical function, the kouloures were possibly meant to impress locals and visitors by emphasising the agricultural wealth of the palace.

Let us now try to compare the storage capacities of the Malia houses with those of the palace. Although it is impossible to evaluate the storage capacities of the first palace, there is a general agreement on the idea of increased centralisation during LM I at Malia, based on the disappearance of all large storage structures outside the palace (Mu, Hypostyle Crypt, ‘Bâtiment Dessenne’) and on the construction of eight kouloures to the southwest of the palace. These are cylindrical constructions, perhaps of oriental or Egyptian influence, and now widely interpreted as granaries (Pelon 1980: 221-226; Strasser 1997: 84-89; Carinci 2001; BradferBurdet 2005; Privitera 2014; Keßler 2015: 158-160). Each kouloura was 3.70m-4.00m in diameter and about 3m to 4m high and therefore had a capacity of about 32-50m3 (Van Effenterre 1980: 460, n. 62). When the granaries were in use, their storage capacity was about 220,000 to 335,000 litres (Pelon 1980, p. 226; Keßler 2015: 152, 161). This quantity of grain or legumes is much greater than the volumes calculated for the houses and it was sufficient to feed several hundred people for a year, but not to ensure the subsistence of the urban population. Some researchers considered this to be the sign of independent collective storage as these granaries are located outside the palace, with no direct access to it

It is noteworthy that these granaries do not date to the Middle Minoan expansion of the town. On the contrary, they are contemporary with what appears to be a period of decline or, at least, dismantling of the core area of the urban centre. They were built in a late phase of the second palace and may result not only from increased centralisation, but also from more intensive farming in the plain and/or increased pressure on the farmers (through taxes or labour duties?). Conclusion After this brief review of the evidence concerning the management and storage of agricultural produce in the palace, the town and the countryside of Malia, we can now turn to the question of the underlying economic model. Archaeological remains present a dichotomy between the relatively low storage capacity of the urban houses and the relative abundance of tools for processing agricultural produce, other than oil and wine. The latter clearly belong to a group of products with separate management methods. Several explanations could account for this. First of all, it probably means 64

Kostis S. Christakis et al.: The Neglected ‘Fields’ of Proto-Urban Living

that the harvested produce was not brought to town in its entirety but rather was at least partly processed and stored near the fields, especially as regards the grain to be used as seed (Van Effenterre 1980: 461). Equids appear to have been uncommon at Malia, which is why transport from the fields possibly depended upon cattle (see for Messara, Privitera 2014: 437). Medium -sized pithoi have been discovered in several of the rural settlements (Müller Celka et al. 2014: 435) and pit storage in the fields is attested ethnographically (Halstead 2014, 159; Halstead 2014, 159). However, it is not possible to evaluate the storage capacity or the status of these rural sites as long as they have not been excavated. Were they autonomous farming units, as suggested by their fortified aspect, or were they dependent on urban and/or palatial authority, as could be inferred from the fact that most of them were in visual contact with the palace? At any rate, the abandonment of these settlements at the dawn of the Neopalatial period indicates that local production, if the highlands were still cultivated, was transported to the town and stored there during the MM III and LM I periods. It is more likely, however, that farming was concentrated on the lowlands during the Neopalatial period, due to improved techniques (including irrigation) leading to higher crop yields from the coastal plain of Malia (Müller Celka et al. 2014; Müller Celka 2015; Müller Celka 2017). If so, agricultural produce was at least partly stored near the fields and/or in the lowland settlements.

transport, the massive and regular importation of cereals and other foodstuffs to Malia does not seem likely. c) The authorities centralised part of the production and controlled its redistribution by allocating rations to dependent workers. This system is recorded in the Aegean world on the Linear B tablets only in the LM III phase, but much earlier in the Near and Middle East (Van de Mieroop 2004: 27). The redistributive nature of palatial economies, as accepted by most researchers of the second half of the 20th century, has been recently challenged, but the main issue here is the degree of centralisation by the palatial authority (Liverani 2005; Privitera 2010; Nakassis et al. 2011; Christakis 2011; Halstead 2011). The storage capacities of the palace were not sufficient for feeding the inhabitants of the town for more than a few days, especially during the Protopalatial period when the population may have reached 10,000-20,000 people, even according to low estimations (Müller Celka 2007). If redistribution did occur, it could not have supplied all the inhabitants on a daily basis (Keßler 2015: 163) but must have been occasional, for instance in the case of collective events organised by the palace, or restricted to ‘salaries’ paid to people working for the palatial authorities. However, the storage management of foodstuffs in town, particularly cereals, has been demonstrated to have always been overseen by authorities (Sigaut 1979: 2; Trantafyllidou-Baladie 1979). This could have been the case already during the Protopalatial period since the dispersal of large storage units in several ‘masons’ with at least a partially autonomous economic system implies that they were managing part of the agricultural produce beside the palace, to use H. Van Effenterre’s model and its recent reinterpretation according to the concept of ‘heterarchy’ (Van Effenterre 1980: 461-464; Schoep 2002). However, the presence of loom-weights with similar seal impressions in some of these houses suggests that the latter depended on the same authority and did not belong to ‘competing factions’. The kouloures represent the increased capacity of the palace to centralise and mobilise subsistence foodstuffs during LM I, but a smaller Neopalatial kouloura was also identified in the northern part of the rich Epsilon building, which appears to have been the residence of a local elite group during LM I. This group undoubtedly controlled, at least partly, its own supply. The kouloura is a stone-lined pit, 2,50m in diameter, outside the house, in a position reminiscent of the kouloures-granaries of the palace (note that its description in Keßler 2015: 154 is inaccurate), and it plausibly functioned as a granary (BradferBurdet 2005). Thus, during the Neopalatial period the palace did not centralise all the agricultural storage; but the kouloures and other storage facilities

Other complementary, rather than exclusive, hypotheses can be proposed to account for small stock levels in town: a) Some of the inhabitants of the town had easy access to the harvest and primary storage sites located near the fields because they were farmers themselves (at least 70% of the population? Whitelaw, this volume). If so, there would have been a large proportion of on-farm consumption, processing on a domestic scale and autonomous subsistence. This would imply the regular transport of small quantities of foodstuffs from their primary storage site towards the town, as also corroborated by the small number of cattle and donkey bones in the town. b) An urban market supplied the urban population with grain and other foodstuffs. This ‘modernist’ hypothesis is not clearly backed up by archaeological discoveries but cannot be ruled out completely in view of the large number of crafts people in the town during MM II (Feinman 2013). Large open spaces such as the ‘Agora’ and the esplanades surrounding the palace could have served as market areas, in town or on the outskirts. In this case, agricultural products would have come from the surrounding countryside or from more distant regions (for probable importations, Christakis 2008: 135). Nonetheless, on account of the constraints of land 65

Country in the City of the palace, together with the – few – written records, form a body of evidence pointing to the appropriation and management by the palace of the larger part of the production. We cannot identify what brought about such a development: did a change in the land ownership system lead to the concentration of arable land in the hands of the palatial authority during MM III, or did farmers have to pay taxes or labour duties? It has already been pointed out how difficult it is to identify land property without textual evidence (Jusseret et al. 2013) and it is impossible to settle this question for the time being.

Conference on Ancient Textiles, held at Lund, Sweden, and Copenhagen, Denmark, on March 19-23, 2003: 59-63. Oxford, Oxbow. Alberti, M. E. 2008. Textile industry indicators in Minoan work areas: problems of typology and interpretation. In C. Alfaro, L. Karali (eds), Vestidos, textiles y tintes. Estudios sobre la producción de bienes de consumo en la Antigüedad. Actas del II Symposium Internacional sobre Textiles y Tintes del Mediterráneo en el mundo antiguo (Atenas, 24 al 26 de noviembre, 2005): 25-36. València, Universitat de València (Purpureae Vestes II). Alusik, T. 2007. Defensive Architecture of Prehistoric Greece. Oxford (British Archaeological reports 1637). Barber, E. J. W. 1991. Prehistoric Textiles. Princeton, University Press. Barrett, J. C. and Halstead, P. 2004. The Emergence of Civilisation Revisited. Oxford, Oxbow (Sheffield Studies in Aegean Archaeology 5). Bevan, A., Conolly, J., Colledge, S., Frederick, C., Palmer, C., Siddall, R. and Stellatou, A. 2013. The long-term ecology of agricultural terraces and enclosed fields from Antikythera, Greece. Human Ecology 41.2: 255272. Blitzer, H. 1993. Olive Cultivation and Oil Production in Minoan Crete. In M.-C. Amouretti, J.-P. Brun and D. Eitam (eds), La production du vin et de l’huile en Méditerranée: 163-175. Athens, École française d’Athènes (BCH Supplément 26). Bradfer-Burdet, I. 2004. Le petit palais de Malia. Une étude archéologique du quartier Epsilon. Unpublished Mémoire de troisième année de l’Ecole française d’Athènes présenté à l’académie des Inscriptions et Belles-Lettres. Bradfer-Burdet, I. 2005. Une kouloura dans le ‘Petit Palais’ de Malia. In I. Bradfer-Burdet, B. Detournay and R. Laffineur (eds), Kris Technitis. L’Artisan Crétois: Recueil d’articles en l’honneur de Jean-Claude Poursat, publié à l’occasion des 40 ans de la découverte du Quartier Mu : 39–49. Liège and Austin (Aegaeum 26). Bradfer-Burdet, I. and Pomadère, M. 2011. Delta bêta at Malia: Two houses or One Large Complex?. In K. Glowacki and N. Vogeikoff-Brogan (eds), ΣΤΕΓΑ. The Archaeology of Houses and Households in Ancient Crete: 99-108. Princeton, The American School of Classical Studies at Athens (Hesperia Suppl. 44). Brun, J.-P. 2003. Le Vin et l’huile dans la Méditerranée antique: viticulture, oléiculture et procédés de transformation. Paris, Errance. Brun, J.-P. 2004. Archéologie du vin et de l’huile, de la préhistoire à l’époque hellénistique. Paris, Errance. Callot, O. 1987. Les huileries du Bronze Récent à Ougarit, premiers éléments pour une étude. In M. Yon (ed.), Ras Shamra-Ougarit III, Le centre de la ville, 38e-44e campagnes (1978-1984): 197-212. Paris, ERC-ADPF. Callot, O. 1994. La tranchée ‘Ville Sud’. Études d’architecture domestique. Paris, ERC (Ras Shamra-Ougarit X).

In any event, all these possible variables of food acquisition involve an almost continuous circulation of agricultural produce between the countryside and the town. No major, clearly palatial, storage place has been identified in the surveyed countryside, but the palace may have controlled the small Protopalatial rural depots and farms of the whole Malia region. The distribution of Neopalatial sites indicates a shift towards a more intensive type of agriculture practised on lower and flatter arable land and to the concomitant development of a few secondary centres, including storage facilities, in the Malia plain. Let us conclude with a change of scale, by considering the role of Malia in Neopalatial Crete. Today, it is customary to consider that the town and the palace of Malia were in the Knossian orbit, or even subjugated, during LM IA (Wiener 2007; Poursat 2010). If so, it is reasonable to suppose that Knossos was responsible for the intensification of farming and the construction of palatial granaries. Moreover, the new status of Knossos as a dominant authority during LM I could account for the fact that the increasing flow of agricultural produce to the town of Malia did not result in urban growth nor in the development of written administrative records (Weingarten 1990). The people farming the Malia land were then working for a non-local power, leading to demographic decline in the urban and rural territory as well as to the weakening of craft production in the town. The same scenario could explain the appearance or development of secondary centres during MM IIILM IA, such as the Bouffo hill at Sissi, where the ‘court centered building’ (CCB) was erected according to the same orientation as that of Malia, facing the plain and the fields, rather than the sea. Bibliographie Alberti, M. E. 2007. Washing and Dyeing Installations of the Ancient Mediterranean: towards a Definition from Roman Times back to Minoan Crete. In C. Gillis, M. L. Nosch (eds), Ancient textiles: Production, Craft and Society, Proceedings of the First International 66

Kostis S. Christakis et al.: The Neglected ‘Fields’ of Proto-Urban Living

Carinci, F. 2001. Per una diversa interpretazione delle kulure nei cortili occidentali dei palazzi minoici. Creta Antica 2: 43-62. Carter, T. 2013. Chipped stone. In Poursat 2013: 23-24. Carter, T. and Kilikoglou, V. 2007. From Reactor to Royalty? Aegean and Anatolian Obsidians from Quartier Mu, Malia (Crete). Journal of Mediterranean Archaeology 20.1: 115-143. Chanesaz, M. 2006. Le Matruf, le Madras et le Bequf. La fabrication de l’huile d’olive au Liban. Essai d’anthropologie des techniques. Lyon (Travaux de la Maison de l’Orient et de la Méditerranée 44). Chapouthier, F. and Demargne, P. 1942. Fouilles exécutées à Mallia : Troisième rapport : exploration du palais, bordures orientale et septentrionale, 1927, 1928, 1931, 1932. Paris, Geuthner (Etudes crétoises, 6). Chapouthier, F. and Demargne, P. 1962. Fouilles exécutées à Mallia : quatrième rapport : exploration du palais, bordure méridionale et recherches complémentaires, 1929-1935 et 1946-1960. Paris, Geuthner (Etudes crétoises, 12). Christakis, K. 2008. The Politics of Storage. Storage and Sociopolitical Complexity in Neopalatial Crete. Philadelphia, INSTAP Academic Press (Prehistory Monographs, 25). Christakis, K. 2011. Redistribution and Political Economies in Bronze Age Crete. American Journal of Archaeology 115: 197-205. Christakis, K. 2014. Communal storage in Bronze Age Crete: re-assessing testimonies. Κρητικά Χρονικά 34: 201-218. Cunningham, T. F. and Driessen, J. 2003. Site by site: combining survey and excavation data to chart patterns of socio-political change in Bronze Age Crete. In J. F. Cherry and S. E. Alcock (eds), Side by Side Survey: Comparative Regional Studies in the Mediterranean World : 57–73. Oxford, Oxbow. Cutler, J., Andersson Strand, E. and Nosch, M.-L. 2013. Textile production in Quartier Mu. In Poursat 2013: 95-120. D’Agata, A. L. and De Angelis, S. 2014. Minoan Beehives. Reconstructing the Practice of Beekeeping in Bronze Age Crete. In Touchais, Laffineur and Rougemont 2014: 349-358. Dalongeville, R., with the collaboration of Lespez, L., Poursoulis, G., Pastre J.-F., Keraudren B., Mathieu, R., Prieur A., Renault-Miskovsky, J., Darmon, F., Kunesh, S., Bernier, P., Caron, V., Pelc, V., Le Campion, T., Pantazidou, A., Evin, J., Oberlin, C., Noirel-Schutz, C., Sibella, P., Vallat, M., and Viret, J. 2001. Malia  : un marais parle. Bulletin de Correspondance Hellénique 125: 67-88. Darcque, P. 2014. Les abords Nord-Est du palais I : les recherches et l’histoire du secteur : fouilles exécutées à Malia. Athènes, Ecole française d’Athènes (Etudes crétoises, 35). Demargne, P. and Gallet de Santerre, H. 1953. Mallia. Exploration des maisons et quartiers d’habitation (19211948), 1. Paris, P. Geuthner (Etudes crétoises, 9).

Deshayes, J. and Dessenne, A. 1959. Fouilles exécutées à Mallia. Exploration des maisons et quartiers d’habitation (1948-1954), 2. Paris, P. Geuthner (Etudes crétoises, 11). Detournay, B., Poursat, J. C. and Vandenabeele, F. 1980. Fouilles exécutées à Mallia : le quartier Mu, 2. Vases de pierre et de métal, vannerie, figurines et reliefs d’applique, éléments de parure et de décoration, armes, sceaux et empreintes. Paris, P. Geuthner (Etudes crétoises, 26). Devolder, M., Caloi, I. and Gomrée, T. 2015-2016. Rapports sur les travaux de l’École française d’Athènes en 2014. Malia. Le Bâtiment Dessenne. Étude et nouveaux sondages. Bulletin de Correspondance Hellénique 139-140.2: 939-947. Devolder, M. 2012-2013. Le Quartier Nu (Malia, Crète). L’occupation néopalatiale. Bulletin de Correspondance Hellénique 136-137.1: 1-82. Devolder, M. 2016. The Protopalatial State of the Western Magazines of the Palace at Malia (Crete). Oxford Journal of Archaeology 35: 141-159. Driessen, J. 2009. ‘Daidalos’ Designs and Ariadne’s Threads: Minoan Towns as Places of Interaction. In S. Owen and L. Preston (eds), Inside the City in the Greek World: Studies of Urbanism from the Bronze Age to the Hellenistic Period: 41–54. Oxford, Oxbow (University of Cambridge Museum of Classical Archaeology Monographs 1). Driessen, J., Schoep, I., Carpentier, F., Crevecoeur, I., Devolder, M., Driessen-Gaignerot, F., Fiasse, H., Hacigüzeller, P., Jusseret, S., Langohr, C., Letesson, Q. and Schmitt, A. 2009. Excavations at Sissi. Preliminary Report on the 2007-2008 Excavations. Louvain-la-Neuve, Presses universitaires de Louvain (Aegis 1). Driessen, J., Schoep, I., Carpentier, F., Crevecoeur, I., Devolder, M., Driessen-Gaignerot, F., Hacigüzeller, P., Isaakidou, V., Jusseret, S., Langohr, C., Letesson, Q. and Schmitt, A. 2011. Excavations at Sissi. II. Preliminary Report on the 2009-2010 Excavations. Louvain-la-Neuve, Presses universitaires de Louvain (Aegis 4). Driessen, J., Schoep, I., Anastasiadou, M., Carpentier, F., Crevecoeur, I., Déderix, S., Devolder, M., GaignerotDriessen, F., Jusseret, S., Langohr, C., Letesson, Q., Liard, F., Schmitt, A., Tsoraki, C. and Veropoulidou, R. 2012. Excavations at Sissi III. Preliminary Report on the 2011 Campaign. Louvain-la-Neuve, Presses universitaires de Louvain (Aegis 6). Eitam, D. and Heltzer, M. 1996. Olive Oil in Antiquity. Israel and neighbouring countries from the Neolithic to the early Arab period. Padova, Sargon. Feinman, G. M. 2013. Crafts, Specialist, and Markets in Mycenaean Greece. Reenvisioning Ancient Economies: Beyond Typological Constructs. American Journal of Archaeology 117: 453-459. Flood, J. M. and Soles, J. S. 2014. Water Management in Neopalatial Crete and the Development of the Mediterranean Dry-Season. In Touchais, Laffineur and Rougemont 2014: 79-84. 67

Country in the City Foxhall, L. 2007. Olive Cultivation in Ancient Greece: seeking the ancient economy. Oxford-New York, Oxford University Press. Gomrée, T. 2013. La voirie des villes minoennes en Crète orientale et à Cnossos (Minoen Moyen I-Minoen Récent I). Unpublished Ph. D. thesis, Université Lumière Lyon 2. Gomrée, T., Langohr, C. and Pomadère, M. 2012. Excavations in the Pi Area at Malia (2005-2010). In M. Andrianakis and I. Tzachili (eds), Αρχαιολογικό Έργο Κρήτης 2. Πρακτικά της 2ης Συνάντησης, Ρέθυμνo, 26-28 Νoεμβρίoυ 2010 / Archaeological Work in Crete 2. Proceedings of the 2nd Meeting, Rethymnon, 26-28 November 2010: 89-97. Rethymno, University of Crete. Gomrée, T. and Pomadère, M. Forthcoming. Building Pi and the Neopalatial Period at Malia. In Proceedings of the 11th International Congress of Cretan Studies, Rethymnon, 21-27 October 2011. Hadjisavvas, S. 1992. Olive Oil Processing in Cyprus, from the Bronze Age to the Byzantine period. Nicosia, P. Ǻströms Förlag. Hadjisavvas, S. 2003. The Production and Diffusion of Olive Oil in the Mediterranean, ca. 1500-500 BC. In N.C. Stampolidis and V. Karageorghis (eds), Πλόες... Sea Routes... Interconnections in the Mediterranean, 16th6th c. BC. Proceedings of the International Symposium held at Rethymnon, Crete, September 29th-October 2nd 2002. Athens, University of Crete and the A. G. Leventis Foundation. Hadjisavvas, S. and Chaniotis, A. 2012. Wine and olive oil in Crete and Cyprus: socio-economic aspects. In G. Cadogan, M. Iacovou, K. Kopaka and J. Whitley (eds), Parallel Lives: Ancient Island Societies in Crete and Cyprus. Papers arising from the Conference in Nicosia organized by the British School at Athens, the University of Crete and the University of Cyprus, in NovemberDecember 2006: 157-173. London: The British School at Athens (British School at Athens Studies 20). Halstead, P. 1981. From Determinism to Uncertainty: Social Storage and the Rise of the Minoan Palace. In A. Sheridan and G. Bailey (eds), Economic Archaeology: 187-213. Oxford (British Archaeological report International series, 96). Halstead, P. 1997. Storage Strategies and States on Prehistoric Crete: A Reply to Strasser. Journal of Mediterranean Archaeology 10: 103–107. Halstead, P. 2011. Redistribution in Aegean Palatial Societies: Terminology, Scale, and Significance. American Journal of Archaeology 115.2: 229-235. Halstead, P. 2014. Two oxen ahead: pre-mechanized farming in the Mediterranean. Chichester, West Sussex, Wiley Blackwell. Hamilakis, Y. 1996. Wine, Oil and the Dialectics of Power in Bronze Age Crete: A Review of the Evidence. Oxford Journal of Archaeology 15: 1-32. Hamilakis, Y. 2002. Too Many Chiefs? Factional Competition in Neopalatial Crete. In J. Driessen, I. Schoep and R. Laffineur (eds), Monuments of Minos.

Rethinking the Minoan Palaces. Proceedings of the International Workshop ‘Crete of the hundred Palaces?’ held at the Université Catholique de Louvain, Louvain-laNeuve, 14-15 December 2001: 179-199. Liege, Université de Liège (Aegaeum 23). Hanson, V. D. 1995. The Other Greeks. The Family Farm and the Agrarian Roots of Western Civilization. New York, the Free Press. Harissis, H.V. 2017. Beekeeping in Prehistoric Greece. In F. Hatjina, G. Mavrofidis and R. Jones (eds), Beekeeping in the Mediterranean from Antiquity to the present. Nea Moudania. Isaakidou, V., Livarda, A., Tsoraki, C. and Veropoulidou, R. 2011. Bio-archaeological Assemblages and Ground Stone Artefacts. In J. Driessen et al. 2011: 211-218. Jusseret, S., Driessen, J. and Letesson, Q. 2013. Minoan Lands? Some Remarks on Land Ownership on Bronze Age Crete. In M. Relaki and D. Catapoti (eds), An Archaeology of Land Ownership: 46-69. New York and London: Routledge (Routledge Studies in Archaeology 9). Kaniewski, D., Van Campo, E., Guiot, J., Le Burel, S., Otto, T., et al. 2013. Environmental Roots of the Late Bronze Crisis. PloS One 8.8. Keßler, T. P. 2015. A Royal Gift? Bulk Grain Storage in Protopalatial and Neopalatial Crete. Studi Micenei ed Egeo-Anatolici N.S. 1: 137-170. Kopaka, K. and Platon, L. 1993. Lenoi minoikoi. Installations minoennes de traitement des produits liquides. Bulletin de Correspondance Hellénique 117: 35101. Lespez, L., Dalongeville, R., Pastre, J.-F. and Müller Celka, S. 2003. Le site de Malia et la mer. Approche environnementale. Topoi 11: 613-633. Lespez, L., Müller Celka, S. and Pomadère, M. 2016. Changements environnementaux et impact des sociétés humaines autour de la cité minoenne de Malia (Crète, Grèce): bilan des acquis et nouvelles recherches. In M. Ghilardi (ed.), Géoarchéologie des îles de Méditerranée: 245-258. Paris, CNRS Editions. Liverani, M. 2005. The Near East: the Bronze Age. In J.G. Manning and I. Morris (eds), The Ancient Economy. Evidence and Models: 47-57. Stanford, Stanford University Press. McGovern, P.E. 2009 and 2010. Uncorking the Past: The Quest for Wine, Beer and Other Alcoholic Beverages, Berkley, University of California Press. MacGillivray, J. A., Sackett, L. H. and Driessen, J. 2007. Palaikastro: Two Late Minoan Wells. London, The British School at Athens (BSA Supplementary Volume, 43). May, N. and Steinert, U. 2014. The Fabric of Cities. Aspects of Urbanism, Urban Topography and Society in Mesopotamia, Greece and Rome. Leiden-Boston, Brill. Melas, M. 1999. The Ethnography of Minoan and Mycenaean Beekeeping. In P.P. Betancourt, V.  Karageorghis, R. Laffineur and W.‑D. Niemeier (eds), Meletemata: Studies in Aegean Archaeology 68

Kostis S. Christakis et al.: The Neglected ‘Fields’ of Proto-Urban Living

Presented to Malcolm H. Wiener as He Enters His 65th Year. Vol. II: 485-491. Université de Liège, University of Texas at Austin: Programs in Aegean Scripts and Prehistory (Aegaeum 20). Militello, P. 2011. Immagini e realtà della produzione nella Creta minoica: I temi assenti. In F. Carinci, N. Cucuzza, P. Militello and O. Palio (eds), KΡHΤHΣ ΜINΩΙDOΣ, Tradizione e identità minoica tra produzione artigianale, pratiche cerimoniali e memoria dell passato, Studi offerti a Vincenzo La Rosa per il suo 70° compleanno: 239-258. Padova, Bottega d’Erasmo (Studi di archeologia cretese X). Moody, J. 2000. Holocene Climate Change in Crete: An Archaeologist’s View. In P. Halstead and C. Frederick (eds), Landscape and Land Use in Postglacial Greece: 5261. Sheffield, Sheffield Academic Press. Moundréa-Agrafioti, A. 2002. Moudre et broyer dans l’espace construit d’une ville du début du Bronze récent égéen (Akrotiri, Théra). In H. Procopiou and R. Treuil (eds), Moudre et broyer, II. Archéologie et histoire: 93-108. Paris, CTHS. Müller Celka, S. 1992. Prospection de la plaine de Malia. Bulletin de Correspondance Hellénique 116.2: 742-753. Müller Celka, S. 2007. Terroir et territoire de Malia à l’époque protopalatiale. Bulletin de Correspondance Hellénique 131.2: 855-860. Müller Celka, S., Puglisi, D. and Bendali, F. 2014. Settlement Pattern Dynamics and Natural Resources in MM-LM I Crete: The Case of Malia. In Touchais, Laffineur and Rougemont 2014: 431-440. Müller Celka, S. 2015. Recompositions spatiales du peuplement en Crète minoenne, ArchéOrient – Le Blog (Hypotheses.org), 1 mai 2015. [En ligne]. http://archeorient.hypotheses.org/3999 Müller Celka, S. 2017. Peuplement, climat et ressources agricoles en Crète minoenne, Dossiers d’archéologie 382: 24-27. Nakassis, D., Parkinson, W. A., and Galaty, M. L. 2011. Redistributive Economies from a Theoretical and Cross-Cultural Perspective. American Journal of Archaeology 115.2: 177-184. Olivier, J.-P. and Godart, L. 1996. Corpus Hieroglyphicarum Inscriptionum Cretae. Paris, De Boccard, École Française d’Athènes (Etudes Crétoises, 31). Orgeolet, R. and Pomadère, M. 2011. La nature des premiers Etats en Crète : réactions à l’historiographie récente. In D. Garcia (ed.), L’Âge du Bronze en Méditerranée. Recherches récentes: 27-42. Paris, Errance. Orgeolet, R. and Pomadère, M. 2015. Formes et déformations de la ville égéenne : Akrotiri et les villes minoennes au prisme de l’historiographie récente. In S. Bourdin, A. Reltgen-Tallon et M. Paoli (eds), La forme de la ville: 365-384. Rennes, Presses Universitaires de Rennes. Pelon, O. 1966. Maison d’Hagia Varvara et architecture domestique à Malia. Bulletin de Correspondance Hellénique 90.2: 552-585.

Pelon, O. 1970. Fouilles exécutées à Mallia. Exploration des maisons et quartiers d’habitation (1963-1966), III (Le Quartier E). Paris, Paul Geuthner (Etudes crétoises, 16). Pelon, O. 1980. Le Palais de Malia V,1. Paris, Paul Geuthner (Etudes crétoises, 25). Pelon, O. 1983. L’épée à l’acrobate et la chronologie maliote (II). Bulletin de Correspondance Hellénique 107.2: 679-703. Pelon, O. 1987. Minoan Palaces and Workshops: New Data from Malia. In R. Hägg and N. Marinatos (eds), The Function of the Minoan Palaces, Proceedings of the Fourth International Symposium at the Swedish Institute in Athens, 10-16 June, 1984: 269-272. Stockholm, Swedish Institute in Athens (Series in 4°, 35). Pelon, O. 1992. Guide de Mallia. Le palais et la nécropole de Chrysolakkos. Paris, Ecole française d’Athènes. Peña-Chocarro, L. and Zapata, L. 2014. Parching and Dehusking Hulled Wheats. In van Gijn et al. 2014: 226- 232. Petruso, K.M. 1992. Keos VIII. Ayia Irini: The Balance Weights. Mainz on Rhine, P. von Zabern. Peyronel, L. 2011. Mašqattum Kittum. Questioni di equilibrio: balance e sistemi di pesatura nell’Oriente antico. In E. Ascalone and L. Peyronel (eds), Studi italiani di metrologia ed economia del Vicino Oriente Antico dedicati a Nicola Parise in occasione del Suo settantesimo compleanno: 105-162. Roma, Herder (Studia Asiana, 7). Pomadère, M. 2010. Rapports sur les travaux de l’Ecole française en 2009. Malia, secteur Pi. Bulletin de Correspondance Hellénique 134.2: 589-591. Pomadère, M., with the collaboration of Creuzieux, A., Gomrée, T., Mercier, F. and Orgeolet R. 2011. Rapports sur les travaux de l’Ecole française en 2010. Malia, bâtiment Pi. Bulletin de Correspondance Hellénique 135.2: 601-624. Pomadère, M. with the collaboration of Langohr, C., Alberti, M. E., Montagné, M.-P., Dinou, M. 2014. Rapports sur les travaux de l’Ecole française en 2013. Bulletin de Correspondance Hellénique 138.2: 773-775. Pomadère, M. with the collaboration of Alberti, M. E., Avramut, A. M., Devogelaere, J., Gilon, A., Gomrée, T., Langohr, C., Leclercq, C., Lorenzon, M., Montagné, M.-P., Ntinou, M., Papayanni, K., Rivière, K., and Westlake, P. 2015-2016. Rapports sur les travaux de l’école française en 2014 et 2015. Malia, le secteur Pi. Bulletin de Correspondance Hellénique 139-140.2: 917936. Poursat, J.-C. 1985. Ateliers et artisans minoens. In T. Detorakis (ed.), Πεπραγμένα του Ε΄ ∆ιεθνούς Κρητολογικού Συνεδρίου (Άγιος Νικόλαος, 25 Σεπτεμβρίου - 1 Οκτωβρίου 1981), Vol. A’: 297-300. Heraklio, Institute of Cretan Studies. Poursat, J.-C. 1988. La Ville minoenne de Malia: recherches et publications récentes. Revue Archéologique: 61–82. 69

Country in the City Poursat, J.-C. 1992. Guide de Malia au temps des premiers palais. Le quartier Mu. Athens, Ecole française d’Athènes. Poursat, J.-C. and Knappett, C. 2005. Fouilles exécutées à Malia: Le quartier Mu IV. La poterie du Minoen Moyen II: production et utilisation. Athens, École Française d’Athènes (Études crétoises, 33). Poursat, J.-C. 2010. Malia: palace, state, city. In O. Krzyszkowska (ed.), Cretan Offerings. Studies in Honour of Peter Warren: 259–267. London, The British School at Athens (BSA Studies, 18). Poursat, J.-C. 2013. Vie quotidienne et techniques au Minoen Moyen II. Fouilles exécutées à Malia, le Quartier Mu V. Athens, Ecole française d’Athènes (Etudes crétoises, 34). Poursat, J.-C., Rougemont, F., Cutler, J., Andersson Strand, E. and Nosch, M.-L. 2015. Textile tools from Quartier Mu, Malia, Crete, Greece. In E. Andersson Strand and M.-L. Nosch (eds), Tools, Textiles and Contexts: Investigating Textile Production in the Aegean and Eastern Mediterranean Bronze Age: 229-241. Oxford and Philadelphia, Oxbow (Ancient Textiles Series 21). Poursat, J.-C. and Loubet, M. 2005. Métallurgie et contacts extérieurs à Malia (Crète) au Minoen Moyen II : remarques sur une série d’analyses isotopiques du plomb. In R. Laffineur and E. Greco (eds), Emporia: Aegeans in the Central and Eastern Mediterranean. Proceedings of the 10th International Aegean Conference/10e Rencontre égéenne internationale, Athens, Italian School of Archaeology, 14-18 April 2004: 117–121. Liège and Austin, Université de Liège and University of Texas at Austin (Aegaeum 25) Privitera, S. 2010. I granai del re: l’immagazzinamento centralizzato delle derrate a Creta tra il XV il XIII secolo a.C. Padova, Il Poligrafo. Privitera, S. 2014. Long-Term Grain Storage and Political Economy in Bronze Age Crete: Conceptualizing Ayia Triada’s Silo Complexes. American Journal of Archaeology 118: 429-449. Procopiou, H. 1998. L’outillage de mouture et de broyage en Crète minoenne. Unpublished PhD Thesis, Université de Paris I-Panthéon Sorbonne. Procopiou, H. 2003. Les techniques de décorticage dans le monde égéen. Etude ethnoarchéologique dans les Cyclades. In P. C. Anderson, L. S. Cummings, T.  K. Schippers and B. Simonel (eds), Le traitement des récoltes : un regard sur la diversité, du Néolithique au Présent: 115-136. Antibes, APDCA. Procopiou, H. 2013. Les outils lithiques. In Poursat 2013: 43-66. Procopiou, H. 2014a. Barley Meal Processing in the Aegean World: a Look at Diversity. In van Gijn et al. 2014: 243-246. Procopiou, H. 2014b. Grinding and Pounding Tools in Bronze Age Crete. In van Gijn et al. 2014: 238-242.

Puglisi, D. 2007. Le territoire de Malia à l’époque néopalatiale : premières données de la prospection. Bulletin de Correspondance Hellénique 131.2: 865-869. Renfrew, C. 1972. The Emergence of Civilisation. The Cyclades and the Aegean in the Third Millennium B.C. London, Methuen. Rowan, E. 2015. Olive Oil Pressing Waste as a Fuel Source in Antiquity. American Journal of Archaeology 119: 465482. Russel, M. 2003. Après la récolte. Conserver les fruits dans l’Antiquité gréco-romaine. Expérimentations et analyses. In P. C. Anderson, L. S. Cummings, T. K. Schippers and B. Simonel (eds), Le traitement des récoltes : un regard sur la diversité, du Néolithique au présent, 23e rencontres internationales d’archéologie et d’histoire d’Antibes: 249-258. Antibes, APDCA. Sarpaki, A. 2007. Résultats archéobotaniques préliminaires dans divers secteurs de Malia. Bulletin de Correspondance Hellénique 131.2: 882-884. Schaeffer, C. F. A. 1978. Remarques sur les ancres en pierre d’Ugarit. Ugaritica 7: 371-378. Schmid, M. and Treuil, R. 2017. Architecture minoenne à Malia. Les bâtiments principaux du Quartier Mu. Fouilles exécutées à Malia, le Quartier Mu VI. Athens, École française d’Athènes (Etudes crétoises 36). Schoep, I. 2002. Social and Political Organization on Crete in the Proto-Palatial Period: The Case of Middle Minoan II Malia. Journal of Mediterranean Archaeology 15: 101–132. Schoep, I. and Knappett, C. 2003. Le Quartier Nu (Malia, Crète). L’occupation du Minoen Moyen II. Bulletin de Correspondance Hellénique 127: 49–86. Shaw, W. and Shaw, M. (eds). 1996. Kommos I. The Kommos Region and Houses of the Minoan Town. I.2 – The Minoan Hilltop and Hillside Houses. Princeton, Princeton University Press. Sigaut, F. 1979. La redécouverte des silos à grains en Europe occidentale, 1708-1880. 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, vol. 1: 15-40. Paris, CNRS Editions. Strasser, T. 1997. Storage and States on Prehistoric Crete: The Function of the Koulouras in the First Minoan Palaces. Journal of Mediterranean Archaeology 10: 73–100. Touchais, G., Laffineur, R. and F. Rougemont (eds) 2014. Physis: l’environnement naturel et la relation hommemilieu dans le monde égéen protohistorique. Actes de la 14e Rencontre égéenne internationale, Paris, Institut National d’Histoire de l’Art (INHA), 11-14 décembre 2012. Leuven and Liege: Peeters (Aegaeum 37). Trantafyllidou-Baladie, Y. 1979. Greniers publics et familiaux en Grèce du XIVe au XXe s. In M. Gast and F. Sigaut (eds), Les techniques de conservation des grains à long terme. Leur rôle dans la dynamique des systèmesde cultures et des sociétés, vol. 1: 150-158. Paris, CNRS Editions. 70

Kostis S. Christakis et al.: The Neglected ‘Fields’ of Proto-Urban Living

Treuil, R. 1996. De la cigale à la fourmi : les origines des magasins à vivres dans le monde égéen. Topoi 6: 7183. Tzachili, I. 1997. Υφαντική και υφάντρες στο Προϊστορικό Αιγαίο 2000-1000 π.Χ. Rethimno, University of Crete. Valamoti, S. M., Samuel, D., Bayram, M. and Marinova, E. 2008. Prehistoric cereal foods from Greece and Bulgaria: investigation of starch microstructure in experimental and archaeological charred remains. Vegetation History and Archaeobotany 17 (Suppl. 1): 265-276. van Effenterre, H. and van Effenterre, M. 1969. Fouilles exécutées à Mallia. Le Centre Politique I: L’Agora (19601966). Paris, Paul Geuthner (Etudes crétoises, 17). van Effenterre, H. and van Effenterre, M. 1976. Fouilles exécutées à Mallia. Exploration des maisons et quartiers d’habitation (1956-1960). Quatrième fascicule. Paris, Paul Geuthner (Etudes crétoises, 22). van Effenterre, H. 1980. Le palais de Mallia et la cité minoenne. Roma, Edizioni dell’Ateneo (Incunabula Graeca, LXXVI). Van de Mieroop, M. 2004. A History of the Ancient Near East, ca. 3000-323 BC. Oxford, Blackwell Publishing. van Gijn, A. L., Whittaker, J. C. and Anderson, P. C. 2014 (eds). Explaining and Exploring Diversity in Agricultural Technology. Oxford, Oxbow. Vila, E. and Helmer, D. 2007. Note sur l’exploitation des animaux à Malia : les restes fauniques des Quartiers Mu, Nu, du Palais et des Abords Nord-Est. Bulletin de Correspondance Hellénique 131.2: 885-887. Vila, E. and Helmer, D. 2013. Les vestiges de mammifères. In Poursat 2013: 139-180. Vokotopoulos, L., Plath, G. and McCoy, F. W. 2014. The Yield of the Land: Soil Conservation and the

Exploitation of Arable Land at Choiromandres, Zakros in the New Palace Period. In Touchais, Laffineur and Rougemont 2014: 251-263. Warren, P. 2004. Terra Incognita? The Territory and Boundaries of the Early Neopalatial Knossian State. In G. Cadogan, E. Hatzaki and A. Vasilakis (eds), Knossos: Palace, City and State: 165-166. Athens, London, British School at Athens (Studies 12). Weingarten, J. 1990. Three Upheavals in Minoan Sealing Administration: Evidence for Radical Change. In T.G. Palaima (ed.), Aegean Seals, Sealings and Administration. Proceedings of the NEH-Dickson Conference of the Program in Aegean Scripts and Prehistory of the Department of Classics, University of Texas at Austin, January 11-13, 1989: 105-116. Liège, Université de Liège (Aegaeum 5). Wiener, M. H. 2007. Neopalatial Knossos: Rule and Role. In P. P. Betancourt, M. C. Nelson and H. Williams (eds), Krinoi kai Limenes. Studies in honor of Joseph and Maria Shaw: 231-242. Philadelphia: INSTAP Academic Press (Prehistory Monographs 22). Younger, J. G. 2005. Cretan Hieroglyphic Wool Units (LANA, double mina). In M. Perna (ed.), Studi in onore di Enrica Fiandra: Contributi di archeologia egea e vicinorientale: 405-409. Paris, De Boccard (Studi egei e vicin orientali 1). Zielinski, J. P. 1998. Cyclopean Architecture in Minoan Bronze Age Crete: A Study in the Social Organization of a Complex Society. Buffalo, University of New York at Buffalo. Zurbach, J. 2007. La plaine de Malia entre la fin des palais et l’époque archaïque. Bulletin de Correspondance Hellénique 131.2: 870-872.

71

Akrotiri, Thera : Glimpses of the Countryside as Seen Through the Archaeological and Bioarchaeological Data. Whispers of a Dialogue Anaya Sarpaki

Independent researcher

Introduction

3.

Akrotiri is a type site which could provide windows into various aspects of past urban conglomerations, as it provides three-dimensional data sets and, equally importantly, including all of the abandoned contents of its buildings. More than this, the whole island of Thera has a buried human and cultural landscape which has been kept pristine under thick layers of pumice. This has preserved the man-made landscape as a cocoon, but has, also, hindered us from fully locating and studying it at present. These hidden treasures archaeologists have started –we hope- to be able to study hands-on, which will permit the understanding of the inter-connections between three areas of the landscape: a) the urban core, b) the more agricultural periphery (probably) and c) the countryside.

The first excavator of Akrotiri in the 20th century1, namely in 1967, Spyridon Marinatos, seems to have ‘hit’ on the centre of the ‘urban core’ of the town with its large ‘municipal-cum-religious’ buildings, such as Xeste 3 and 4, as well as well-to-do ‘urban’ dwellings. It is still not known whether the periphery of the town, and as one distances oneself from the centre, the density of habitation follows the same model, i.e. concentrated and dense, as there is no evidence, in the excavated part, of courtyards and/or gardens around the buildings. It seems that the terraces of the houses were used as working spaces, probably due to this lack of private yards and outdoor space, which indicates a scarcity of land.

Would this study of Thera, which is an island in both actual and figurative sense, throw some light on the interconnection of the urban space and the countryside for other areas of the Bronze Age? Thera could surely offer some answers to specific questions but does not provide solutions to general problems. It is not the generalizations of this dialogue that are sought after but, rather, the investigation of the variability of these models, in space and time, which could be identified and need to be the ultimate goal of research for Bronze Age societies and any society for that matter.

Thera: the island - general thoughts Thera is a fairly small island of c. 90 sq.km (Therasia, Aspronisi and two Kamenoi islands) but its exact size and shape in the Bronze Age, is surmised but not exactly known (figure 2). Surface survey, in order to locate archaeological sites, is, furthermore, virtually impossible for mainly two reasons. (1) The first is that as the shape of the island has, radically, changed since the LBA destructive volcanic eruption. Some of the sites, especially the ones near the coast, must have disappeared under the sea or have been covered by rubble and ash. (2) Secondly, layers of pumice have covered the whole island and make it impossible to locate surface finds, except for the mountain top, Prophitis Elias, where pumice has been eroded due to the steep topography and weather elements (rain, wind).

Akrotiri Akrotiri, on Thera island, is an urban site, a large town of the LBA at the time of its destruction (c. 20ha) (Doumas 1983; Forsyth 1997: 52) (figure 1). The model of organization of its space within and outside the city, though, eludes us, as urbanization can be expressed by various systems: 1. 2.

Yet others, which could have been an amalgam of these types, with densities changing between neighbourhoods or even with distance from the centre.

By habitation exemplified by concentrated and dense clusters of buildings in the town; Another which could have not been as dense but more extended, with perhaps small gardens and/or courtyards adjacent to the buildings;

The only source of evidence for archaeological finds, around the caldera, were the sites only uncovered when 1 Gorceix, H. and Mamet, H. (1870). Fouilles à Santorin. Bulletin de L’Ėcole française d’Athènes 10, 187-191, excavated in the area before.

72

Anaya Sarpaki: Akrotiri, Thera

Figure 1: Akrotiri: town plan of the excavated area (after Doumas 1992: 44). In dark grey are the areas where wall paintings were found.

73

Country in the City operations, or else, inland, by farmers falling into holes, when the ceilings of buried buildings collapsed under the extra weight of their animals2 (e.x.donkey/mules) and equipment (ard). At present, we know of eighteen LBA sites, including Akrotiri (figure 3) which are found in a comparatively small area, mostly in the southwest of the island. At least seven of these are close to Akrotiri, but we should consider the available data as somewhat biased in its spatial distribution, although Doumas (1983:129) believes, without enumerating the reasons, that the southwest was more densely populated than the north. It is believed, that it might be a result of where most of research has centred in the 19th and 20th centuries AD, but this needs verification. The distribution of sites on Thera, with their highstatus architecture and artefacts, suggest that the settlement pattern diverts from what is known from Kea and Melos, where scattered small habitation sites in the Early Cycladic gave way to a more highly nucleated settlement pattern from the Middle and Late Cycladic. At Thera in the Late Cycladic, though, the habitation sites which are dispersed in the landscape are more substantial than mere ‘farmsteads’ (Barber 1987:65-66), so the definition of the type of population that was dispersed throughout the Theran landscape, would have, probably, been very different from those of the Kean and/or Melian ‘agricultural population’. It is, therefore, necessary to re-define the nature of the population that inhabited the ‘countryside’ of Thera and, most probably, this detailed redefinition will be able to be more clarified/substantiated when the recently excavated site of Raos (Marthari 2004) will be fully studied and published. It is important to note that even in similar periods on island environments and in such close geographical proximity, the organization of space could have had different social and economic structures to sustain it and this is valuable data for enriching our understanding of the factors leading to this variability. Figure 2: Palaeogeologic and palaeotopographic maps showing how Thera might have, possibly, been prior to the LBA eruption. (after Heiken, McCoy and Sheridan 1990 ).

With our present evidence, it is virtually impossible to say whether the site of Akrotiri was a town inhabited by a principally urban population which depended on other agricultural sites for its subsistence needs or whether it reflects a situation where the core of the town was urban and the periphery agricultural, providing the urban centre with its consumption requirements and probably in close interaction with the rural population. The narrative of the ‘miniature frieze’ found in room 5, south wall, of the West House is particularly eloquent, if we observe some symbolism of language, and what

quarries were exploited, which extracted Theran ash. Extraction was begun due to the building of the Suez Canal in the 19th century (c. AD 1869), so to speak, and whilst extracting the ash, the miners stumbled on antiquities. Therefore, most of the sites of pre-eruption Thera have been found by pure chance in quarrying

2 This is what happened at Akrotiri as indicated to S.Marinatos by the ‘guard of fields’ (αγροφύλακας) Stathis Arvanitis (personal communication).

74

Anaya Sarpaki: Akrotiri, Thera

15 12

8

7 11 10

16 4 17 2

9 A

3

B

15

6 18

14

13 0

1

2

3

PRE LBA SETTLEMENT

4

5 km

AREA OF SOUTH HARBOUR

Figure 3: Map of Thera showing the following: a) The pre LBA island with its possible western and southern harbours (after Pichler and Friedrich 1980.); b) the LBA sites located on the island of Thera; c) the possible extent of the area of the site of Akrotiri, area A and B; d) the carrying capacity - sensu stricto - of Akrotiri, as the whole island - sensu lato - could have represented the carrying capacity. Sites: 1) Katsambas (grave) Sperling 1973 and Doumas 1983; 2) Archangelos (building) Sperling 1973; Doumas 1983; Papadopoulos 2008; 3) Kokkino Vouno (building) Marinatos 1969:35-36; Sperling 1973; 4) Balos (building complex) Fouqué 1879; Mamet 1874; Doumas 1983; 5) Potamos-Kamaras (buildings) Gaertringen 1904; Fouqué 1879; Sperling 1973:22-23; 6) Megalochori (grave and building) Sperling 1973; Doumas 1983; 7) Phira quarries (burial and traces) Sperling 1973; 8) Alaphouzos (Therasia) (building) Fouqué 1879; Sperling 1973:39,56-61; 9) Mavromatis quarry (building) Doumas 1983; 10) Karageorghis quarry (burial?) Doumas 1983; 11) Phtellos (building) Doumas 1973; 12) Oia quarry (building) Doumas 1983); 13) Exomiti (surface finds) Gaertringen 1904, Vol.III:42; Marinatos 1968:4; 14) Perissa (surface find) Marinatos 1968:4; 15) Akrotiri Koloumbo (burial) Simpson and Dickinson 1979; BSA 1956:13; 16) Raos (building) Marthari 2004; 17) Katsades (traces) Marinatos 1968:4; Sperling 1973; 18) Chalarovounia (near the modern village of Emborio)(buildings) Marthari 2004:54-57.

the artist tried to transcribe in picture. (figure 4) The artist depicted a town and a ‘hamlet’/farmhouses separated by a river. However, what is important is that, on either side of the river, a farmer/shepherd stand face to face and seem to have a sort of conversation, at least they have eye contact. This, to our mind, justifies the existence of farmers-cum-shepherds (the dress is in accordance to what we know of their attire) in the

urban compound, which had close relationships with similar populations in the countryside. Therefore, Thera should surely stimulate us into researching on the systems of social and economic organization which would project such a pattern and, thus, enriching our knowledge about the dynamics of interaction which controlled the urban sections of the Bronze Age society 75

Country in the City

Figure 4: West House, Room 5 (first floor), the miniature frieze, south wall (detail). It clearly portrays a town (right) and a ‘village’/mansion (left) separated by a river and on either side are two men, perhaps farmers-cum-shepherds ‘talking’ to each other (after Doumas 1992).

with its, perhaps, semi-agricultural and/or agricultural fractions.

completed, the plans, as published, are quite confusing, as each building has been excavated to a different level. In some buildings, the plan depicts the ground floor, in others the first floor and in some others, even, the second floor. One, in fact, can say that only 10 buildings could be studied to a certain, satisfactory, degree. The only building though which has, nearly, been totally excavated is the West House.

Akrotiri, Thera: the city The excavated area of Akrotiri could be described as the concentrated and dense model of the town and covers approximately 1 ha. To date, 35 different houses have been identified under the sheltered area of the excavation but only 19 entrances are, so far, visible. Most of these buildings have been preserved to the second or even the third floor (Palyvou 2005:45), so the single floor building (i.e. only ground floor) is a minority. As excavation of these houses has not been

Akrotiri represents a very large city for the LBA, (cf. Phylakopi where the whole settlement has been calculated at c. 18.000sq.m/1.8 ha) and in such large cities one would expect large spaces devoted to ‘administrative, commercial, industrial, religious 76

Anaya Sarpaki: Akrotiri, Thera

and other public’ (Sumner 1979:16) buildings. So far, two buildings seem to obviously have some form of religious and administrative function and that is Xeste 3 and Xeste 4 (figure 1). Population estimates One could suggest that the range of living space per person would be greater in towns, possibly reflecting the higher status of some of the population, and, therefore, we should expect a lower density in towns than villages. This is perhaps what one needs to expect to have been the case at Akrotiri, although the evidence is not yet conclusive. In order to be able to apply population models based on built space, it was deemed necessary to work out first a ratio between built area and open spaces, such as streets, squares etc. A sampled area was chosen (figure 5) and it was estimated that the built area was 813sq.m while the open spaces (streets, places etc.) was 912sq.m making a total of 1725sq.m (figure 6) Having come to some idea regarding the open spaces versus the inhabited spaces, then we proceeded to estimate the population size based on the living space. Three models, based on anthropological/ethnographic observation, were considered. (figure 7): a) Narroll’s (1962) estimate of 10m/person; b) Weissner’s (1974) theory of 16m/person and c) Sumner’s (1979) theory of 50-100m/person. Just to have a mental picture, -a working model- we could suggest that Akrotiri, based on the anthropological models could have been a city housing between 3000 and 8000 inhabitants (Sumner’s model), especially as there might have been uninhabited houses (ruins) within the settlement. In order to provide comparative models, evidence from recent history of the island could be useful. The population density figures of Thera/Santorini in 1838, was of the order of 4000 inhabitants, whereas Thera before WWII could feed c. 28.000. Consequently, the suggested figures could be within the rational possibility.

Figure 5: Akrotiri: Sampled area of the site, in order to work out the ratio of inhabited space (grey area) to open/public white areas (by the author).

a) they could have been city dwellers-cum-farmers (sensu lato farmers) who cultivated fields and provided the ‘sophisticated’ ‘urbanised’ population with cleaned agricultural produce; b) they could have ‘bought’ agricultural produce from farmers further afield and cleaned them at Balos, in order to ‘trade’/transport the produce to the city dwellers. Balos in this case, could have been the intermediary between ‘urban’ farmers and farmers in sensu stricto; c) the occupiers of the Balos ‘houses’ might have been salaried by the city dwellers, and, thus, not independent farmers, in order to tend the estate of their owners. In essence, in the third scenario, these ‘estates’ might have, basically, belonged to the town dwellers.

As the excavation at Akrotiri has adhered to investigating the area which was first opened, i.e. the center of the city, it, probably, does not provide an exact indication of what the periphery might have been like, unless, for example, the site of Balos -which is believed by some (Doumas included) to have been the northern edge of the city-, could provide some insight. If it were the periphery of the town, it does comply with this view, namely that the neighborhoods on the edge of the city were more agricultural. At Balos there were storage jars found with archaeobotanical material such as barley, legumes and the key issue is that cereals were found in their chaff (figure 8). In this case, the inhabitants of Balos could be classed under several scenario headings:

A pen was also described at Balos by Fouqué (Fouqué 1998:121) where sheep and goat bones were discovered 77

Country in the City

Area sampled sq.m.

Built area (a)

Open spaces

Multiply X2 (if 2 storey)

1725

813 (b)

913 (b)

1626

10 000

4 713

5 287

c. 200.000 (1) 94 260 (1)= estimate by Ch.Doumas for extent of the site.

9 426

105 743

188 520

Figure 6: Akrotiri: Estimate of inhabited area in sq.m., based on sampling by the author.

Estimated inhabited area in sq.m. 1 storey 813 +

4713 ++

94 260 +++

2 storeys 1626

9426

188 520

Narroll’s 10 m/person 1

2

81

163

9426

18852

471

+ = sampled area ++ = excavated area (1 ha) +++ = estimated c. 20 times excavated area

Weissner’s 16 m/person 1

943

2

Sumner’s 50-100 m/person 1

51

102

16

5891

11782

3770

295

590

94

2

32

188

7540

Figure 7: Akrotiri: Estimates of population size according to Naroll’s, Weissner’s and Sumner’s models, which were based on anthropological/ethnological observations.

Figure 8: Thera: the LBA site of Balos (after Fouqué 1879).

78

Anaya Sarpaki: Akrotiri, Thera

Figure 9: West House model (after Palyvou 2005:Figure56), view from north west.

scattered amongst the straw and a container in the middle of the pen seems to have held barley for their feed. None of these, unfortunately, could be examined as the archaeobotanical material does not exist anymore and yet their descriptions indicate that the material was of a different nature from the one found at Akrotiri. There, the crops had not been cleaned from their chaff and were described as straw, whereas at Akrotiri, crops had reached the site as clean grain, after they had been cleaned from the chaff and glumes and had been processed to the stage nearer consumption, i.e. clean crop with a few by-products and small weed seeds. Nowhere at Akrotiri were located open areas where animals could be kept, such as pens and/or stables. The only areas where the inhabitants of Akrotiri could conduct outdoor chores, such as agricultural processing and/or craft work, would have been on the roofs of their houses which seemed to be used as a terrace, probably for all house chores or even, perhaps, social gatherings.

and floor space on its three levels, as has been finally presented by various specialists (figure 9). The habitation density of the house is affected by the laws of proxemics3, i.e. the area needed per individual in his/her private space. These laws are affected by several factors, such as can be culturally predetermined, but, of course, social and economic indicators have a great deal of effect too. It has been noted that wealthy individuals need more space per person, whereas farmers and lower status individuals need less (Kramer 1979; Sumner 1979). This might be somewhat connected to leisure and chores people do indoors, whereas manual work is rather conducted outdoor, in yards, outside on front doors, terraces as we surmise for Thera. In order to come to some conclusion on floor space, it was necessary to make some calculations of the floor space of the West House but, in order to go beyond these measurements and hypothesize on the floor space of the whole settlement, it was necessary to select a part of the settlement and estimate the inhabited space vis à vis the public space, such as squares, and streets (as explained above). The West House has 345sq.m (Palyvou 2005:31) in total. In order to estimate its population, we looked at Narroll’s (1962), Weissner’s (1974) and Sumner’s (1979) population estimates, and we came with a possible population according to the estimates suggested by the models of the authors at 17-34, 1020 and 3-6 people respectively, depending on which

West House choice and population estimate This problem was explored through two avenues of research. The first was calculation of floor space and habitation density and the second through storage capacity. Out of the 35 buildings from Akrotiri, the West House was chosen as a pilot case in order to make various estimates related to the population capacity of the house. The most important reason is the fact that this urban ‘house’ is the most completely excavated, as well as the most extensively studied. It, therefore, provided the detailed information related to its architecture

Proxemics is a term coined by Edward Hall (1963) and refers to how people structure their microspace and how they interact with each other in this space. This was based on ethnographic observations which, of course, need much tuning. The study of proxemics in archaeology is even more complicated as so many factors are in the realm of the hypothetical.

3

79

Country in the City model one considers more realistic. Personally, I would prefer to accept Weissner’s model4 (10-20 people) (cf. Sumner’s estimate when the whole town is considered) as being nearer to the reality of a sophisticated Bronze Age society.

Room 5 (ground floor)

9

Room 6 (first floor)

4

Room 6 (ground floor) Area 7 (first floor)

Room 3C (ground floor)

West House: estimate of its storage capacity

Room 4 (ground floor)

In order to estimate the storage capacity of the West House, it is necessary to undertake some logical leaps of faith, as the detailed pottery study of the building has not yet been totally completed.5 It is also impossible to know how many of the existing pottery functioned as storage and/or serving ware. We, therefore, had to make the assumption that large pithoi and pottery which were larger than 65cm high were designated for storage. This was based on the thought that large pottery occupied much needed space, so these are kept in buildings only when in use and are mostly devoted to storage (liquid and/or solid). The list (figure 10) shows that 22 such vessels were counted and, all in all, the volume of storage capacity amounted to, at least, c. 3300 litres.

Of unknown location TOTAL

4 2 1 1 1

22 = 3300 liters

Figure 10: West House: large pottery vessels (pithoi) -over 65 cm. high-taken to have been used for storage (minimum number).

The archaeological indicators show that the livingcum-working quarters on the first floor were firmly separated from the ground floor where most of the storage was kept, together with the tools for food preparation. At the base of the main staircase, on the ground floor, there was a door (Palyvou 2005) (Separating area 1,2 -vestibule/entrance- and room 3, ground floor) which separated the area of the ground floor from the first floor and the terrace (second floor). This indicated that the people working on the ground floor were not the same as on the first floor, as the two areas were segregated. The question though remains as to who were the people working on the ground floor. They could have been ‘servants’ as there are two coexisting staircases in the West House. (figure 11, area 7 and main staircase) This clearly indicates a servile population performing in the basement and also, probably, on the second floor terrace. The first floor dwellers did not have to intimately communicate with the people on the ground floor nor those living/ working on the second floor and terrace, as there was a door closing room 3 (figure 11) from the landing of the staircase (also on the first floor).

An ethnographic study in Crete, after WWII (Allbaugh 1953) indicated that, on average, a person needed 458kg of stored food per year and, therefore, we could assume that the house stored, at the very least, for c. seven persons per year. This would represent the minimum number, as, surely, some of the smaller vessels together with containers made of organic material, such as wood, wicker, and/or leather might have been in use too, leaving no trace in some cases. This reinforces Weissner’s model even more and might have raised the ‘houseful’6 to perhaps c. 10-20 people7. Whether all these people lived there (household) or rather came to work (houseful) and retired after work elsewhere, is impossible to say. If they had happened to be more than 9-10 people, then storage was not enough for a whole year or over, which is the usual storage capacity kept by farmers.

There could be several types of servility, as a) they could have been prisoner/slave labour and therefore unpaid, or b) they might have been paid workers, such as peasants/farmers. They might have either lived in the same house too and might also have only worked there and/or slept/lived after work in the periphery of town or even lived further afield in the ‘satelite’ buildings/farmsteads around Akrotiri.

Although the model was built on foraging societies, the Narroll model seemed unrealistic for Bronze Age societies, which need space for storage. Kramer (1979:153) who has investigated these issues in a Kurdish village noted that household size was in the range of 6 persons for houses which were fairly large and had 2 floors, such as most of the houses at Akrotiri. It is hard to know though whether they were occupied by nuclear or extended families and whether some or all of the staff lived in the house too. We can only be sure that families occupied staff other than family members. 5 Mariza Marthari completed her PhD (University of Athens, unpublished) on the Akrotiri pottery of the last phase of the settlement but is also preparing a publication which will be submitted in the series of the Archaeological Society of Athens. 6 The term ‘houseful’ is used here instead of household, as the latter has been used by anthropologists to refer to people living in the house, whereas ‘houseful’ includes the people who might live elsewhere but come to work to the house during working hours. 7 Especially when we consider that 8-10 people alone were needed to work the looms in room 3. 4

They seemed to have fulfilled some chores on the ground floor, one of which was milling (figure 12). In room 3 (ground floor) at least four persons could have worked simultaneously, three of which might have milled, and a fourth could have dehusked the grain (such as glume wheats and barley). This indicates that

80

Anaya Sarpaki: Akrotiri, Thera

the house fed a fairly large number of people, beyond the nuclear/extended family8. The agricultural produce (archaeobotanical material) that was found in the West House seems to have been coarsely sieved away from the house as it had very few by-products of threshing such as rachis and/or spikelet forks for glume wheats. It, therefore, meant that the produce was brought/traded from the periphery as partly cleaned crop which only needed the second stage of preparation such as sieving with fine sieves (some fine weeds were identified) and preparing for consumption which involved dehusking hulled cereals (Sarpaki 1992) and, subsequently, milling them, as well as legumes, into groats and flour. This does imply that the West House inhabitants did not ‘buy’ ready prepared flour, groats and fava (Sarpaki 2001), but they also produced them. It is probable then that they might have owned ‘farms/ fields’ in the ‘neighbourhood’ of the site and the servile population could/would also have been the tenders of these ‘estates’. The female tenders might have served the West House inhabitants, whereas the male tenders could have worked the ‘estates’ as farmers/shepherds. The question that comes to mind is who were the owners of the West House? Might they have been a trade’s-person and his family9? In room 3 (first floor) many loom weights were excavated and their number indicate the presence of 4-5 looms (Tzachili 1990:386)10 in function when the destruction occurred, which indicates that these textiles were not for the household needs alone. In the nearby room 6 (first floor) 26 lead weights were found (Michailidou 2007:187), indicating indirectly the existence of ‘trade’ transactions11. Were members of the family weaving on the looms? Probably yes, but not only, as room 3 is very close to the main living-cum-entertainment area, room 5, and the probable sleeping area, room 4, as well as the water closet (south-west corner of room 4). It seems that whoever was working on the ground floor and probably the terrace was there to provide services, -such as food preparation and house chores- to those living and working on the first floor. It would be reasonable to surmise that they represented a craft/trader ‘elite’.

8 Perhaps we should calculate some 8-10 for the looms (some might have been members of the family) and some 4 people on the ground floor, already reaching some 12-14 people occupied as ‘servants’. 9 As well as textiles, the inhabitants of the West House seem to have exercised metallurgy (Michailidou 2007: 187) as a metallurgical mould (found in room 5, first floor) (idem:231) was found as well as lead weights of various sizes (idem; see Figure6, 200). They range from c. 32gr and up to 3000gr., which indicates a wide range of transactions that ranged between light to heavy products (e.x. textile, threads, metal objects). 10 Tzachili (1990: 386) mentioned the presence of 2 people per loom, so we are referring to 8-10 people working in room 3. 11 We know that threads and textile were weighed when traded.

Figure 11: West House plans (after Palyvou 2005:Figure57), looking east into the main staircase.

81

Country in the City

Figure 12: West House: milling bench and reconstruction (after Moundrea-Agrafioti 2007:87).

Our thesis is that the countryside was closely knit to the site of Akrotiri and there must have been a great deal of exchange between Akrotiri, its periphery and the countryside. The miniature frieze of room 5 in the West House (first floor) provides an insight and a detailed narrative of the occupations taking place in those various niches. Shepherds and farmers are close to towns and water is collected from wells by unsophisticated women12, who are carrying it on their head (figure 13). This too could be an indication for a servile section in the population. Terraces were also near water sources (a well) (figure 13). Another clue provided by the miniature frieze is that towns and ‘villages’ are fairly close to each other and these buildings look on a par with the town houses but are only on a smaller scale (figure 4).13 Some ‘satellite’ sites such as Balos (Fouqué 1879), Raos (Marthari 2004), Archangelos (Papadopoulos 2008), Alaphouzos (Fouqué 1879) indicate that these had a prestigious architecture, similar in quality to Akrotiri and in the case of Raos,

even provided evidence of wall paintings. At Balos, installations were found where a possible ‘cistern’ was identified with stables and straw on the floor, and might have functioned as a ‘private’ farm, connected, perhaps, to a well-to-do household from the city of Akrotiri. These were well within walking distance which could have been covered every day, especially when it came to collect water as there are no cisterns within the houses at Akrotiri, nor have wells been found within the excavated site. Traffic for this commodity though must have been intense. Plants: town and country We know that crops came to the site after having undergone their first processing, that is they had been threshed, most probably on a threshing floor, and coarse sieved. The varieties and types of crops which were chosen to be cultivated such as barley (Hordeum hulled -probably 2-row and 6-row) and wheat, but mainly einkorn and emmer, do not need the same amount of water as naked wheats. It is, therefore, probable that dry-farming was practised on Thera.14

12 Their simple clothing indicated that they were not depicting sophisticated urbanised women. 13 This represents exactly what we find on Thera, in that the ‘countryhouses’ are important mansions.

14

82

By dry-farming we imply that irrigation for annual crops such as

Anaya Sarpaki: Akrotiri, Thera

Figure 13: West House: Room 5 (first floor), miniature frieze, north wall: a well and two women carrying water in pottery vessels, on their heads. The round structure behind them is believed to have been a, possible, threshing floor. The position of a terrace has been conventionally indicated by drawing a row of plants which have been underlined by a line (see also drawing). Behind them, there is a second row of taller plants, closer to the well (after Doumas 1992).

Their high dependence on legume crops in the West House could indicate several things: a) Either that much of their daily protein depended on pulses which might have meant, pulses could have been staples; b) and/or the pressure of land was felt, and they could not leave their fields fallow but, instead, needed to intensify agriculture and produce more in less land. This is one of the reasons to practise crop rotation and alternate sowing between cereals and pulses, thus replenishing the nutrients in the soil in this manner. We know that pulses replenish the soil with nitrogen which is beneficial for the next crop, namely cereals.

archaeobotanical study, that plants grew on terraces. We are lucky that the artist of the frieze of room 5 of the West House has depicted a great deal of the pastoral landscape which was familiar to artist and indicated what could have been happening at the time. It is worth noting too that animals (sheep and goats) were shown as being shepherded on the east of the well, as a rather dry landscape is indicated, which is beneficial to these particular animals. In other words, sheep and goat are depicted in an ecosystem that benefits them. On the west of the well, a herd of bulls and cows seem to be grazing near the coast.15 (figure 14) This might have indicated areas of marshland which often form, especially at certain times of the year, alongside beaches and would have been good grazing ground for cattle. The presence of marshland has been indicated as well from the study of phytoliths (Vlachopoulos and Zorzos 2014; Zorzos forthcoming and private communication)

We are also sure that terraces were used at that time, and it has been discussed before (Sarpaki 1997:665). Their creation was very important as they increased the available agricultural area, and transformed ‘wildscape’ into ‘tamescape’. However, we cannot tell, from our

Unfortunately, this section of the wall paintings had had a great deal of damage.

hulled barley and glume wheats might not have been practised. They, probably, relied on rain water.

15

83

Country in the City

Figure 14: West House: Room 5 (first floor), miniature frieze, north wall: detail to the west of the well with cattle grazing near the coast, which could have indicated marshland (After Televantou, C.A. 1994. Ακρωτήρι Θήρας: οι τοιχογραφίες της Δυτικής Οικίας: 232. Βιβλιοθήκη της εν Αθήναις Αρχαιολογικής Εταιρείας no.143, Athens.

such as Cyperus plants and also Schoenus nigricans have been identified macroscopically.

for fragmented olive stones to have been imported and therefore it is an indication of local cultivation of the tree. Olives in the first 3-5 years of their life need to be watered indicating the probable availability of water on the island.

There was surely a dearth of surface water (rivers, lakes etc.) and, therefore, we can assume that dryfarming was already practised. Most pulses, as well as some cereals, can be grown under dry cultivation regimes, by exploiting rain water to the extreme. Some legumes though such as horsebeans (Vicia faba), sweet peas (Pisum sativum) and chick peas (Cicer arietinum), which are rather rare at Akrotiri, need added watering (irrigation) during their growing season.

In accordance to its urban character not many agricultural tools have been found at the site of Akrotiri. There are some metal sickles and also some chipped stone sickles made from rhyolite, but none of these two categories were found in the West House. There is furthermore a striking absence of areas devoted to technologies such as olive and wine pressing in the settlement, stressing even more the urban character of Akrotiri. In this huge town, only one such installation has been discovered (figure 15) which indicates that these activities were conducted elsewhere. The site of Alaphouzos on Therasia (figure 16) is where the oldest olive beam press has been found, so far, in Greece and dated to the LCI.

Although Thera has a very low rainfall (c. 400mm) nowadays, there are reasons to believe that it either had more rainfall in the LBA and/or the island had several wells and/or springs. Had there been a shortage of water, houses at Akrotiri would have been expected to have had cisterns, which would have collected rainwater, just as the traditional houses of the island today. The presence of water is also indirectly indicated by the high presence of the olive in the archaeobotanical material. The presence of the tree is indicated by the high presence of olive wood, and also fragmented olive stones which had been used as fuel (Braadbaart, Marinova, Sarpaki 2016). It is most unlikely

Epilogue The West House could have held up to c. 20 people (houseful) and, yet, the storage capacity could feed only seven people (for one year). We should, therefore, 84

Anaya Sarpaki: Akrotiri, Thera

Figure 15: The only wine press installation found at Akrotiri, in area of the New Trenches 58B. A basket was found inside the basin which contained lime, possibly for disinfecting the vessel, as wine making requires an environment free of bacteria, otherwise it is turned into vinegar. (Archives of the excavations of Akrotiri).

assume that availability of produce was fairly close by. The archaeological sites from its periphery such as Balos, on the edge of the city of Akrotiri, demonstrate this proximity of the urban to the agricultural character. The inhabitants of Akrotiri would have been in direct contact with the other ‘more’ agricultural sites such as Phtellos. Raos and so forth.

production base, amongst others for olive oil and/or wine. The West House indicates, furthermore, that urban high status families must also have hired/owned human labour which had a subordinate role in the house. It is inconceivable that they only existed on extended families, as the mutually exclusive areas (the ground floor separated by doors from the first floor), the two staircases (official and secondary) would not have had a raison d’être.

There are also indications that households in the urban centre might also have been of differing social backgrounds: 1. 2. 3.

16

Urban and affluent inhabitants as demonstrated by the West House. Poorer inhabitants (Sector Beta north) and perhaps the presence/need for flour stores (Arvanitis area) demonstrate such a population. Surely, the connection of the urban area with the periphery/countryside must have been a tight one where, probably, rich city dwellers also owned houses in the ‘country’16 close to their

Bibliography Allbaugh, L.G. 1953. Crete: a Case Study of an Underdeveloped Area. New Jersey, Princeton.

town and another in the countryside is presently visible in the nearby island, Anaphi, where they used to build these ‘αγροικίες’ (exact translation is houses near fields) in the midst of their fields and preferably near a water source. If they owned fields in different parts of the island which were far apart, they sometimes even owned more than one such house in different parts of the island.

This model of ownership, that is a family owning a house in the

85

Country in the City

Figure 16: Alaphouzos site on the island of Therasia, where the oldest, so far, olive beam press has been discovered (after Fouqué 1879).

Barber, R.L.N. 1987. The Cyclades in the Bronze Age. London, Duckworth. Braadbaart, F., Marinova, E. and Sarpaki, A. 2016. Experimental charring of modern olive stones and olive processing residues: implementing the results to assess the use of residues from the Bronze Age sites of Akrotiri (Greece) and Tell Tweini (Syria), as well as the historical site of Kalaureia on Poros Island (Greece). Vegetation history and Archaeobotany 2016. Doumas, C.1973, Ftellos. Αρχαιολογική Εταιρεία: 161-166. Doumas, Ch. 1983. Thera: Pompeii of the Ancient Aegean. London, Thames and Hudson. Doumas, Ch. 1992. The Wall Paintings of Thera. Athens, the Thera Foundation. Forsyth, P.Y. 1997. Thera in the Bronze Age. New York, Peter Lang. Fouqué, F. 1879. Santorin et ses éruptions. Paris, G. Masson. Fouqué, F. 1998. Santorini and its eruptions. (Translated and annotated by A. McBurney). Baltimore and London, The John Hopkins University Press. Gorceix, H. and Mamet, H. 1870. Fouilles à Santorin. Bulletin de L’Ėcole française d’Athènes 10, 187-191. Hall, E. T. 1963. A system for the notation of proxemics behaviour. American Anthropologist 65:1003-1026.

Heiken, G. and McCoy, F. 1984. Caldera development during the Minoan eruption, Thira, Cyclades, Greece. Journal of Geophysical Research 89: 8441–8462. Heiken, G.H., McCoy F.W., and Sheridan M. 1990. Paleotopographic and paleogeologic reconstruction of Minoan Thera. In A. Hardy (ed), Thera and the Aegean world III, Earth sciences, vol 2. London, The Thera Foundation: 370-378. Hiller von Gaertringen, Fr., 1899-1909. Thera. 4 Vols. Berlin, G. Reimer. Kramer, C. 1979. An archaeological view of a contemporary Kurdish village: Domestic architecture, household size, and wealth. In C. Kramer (ed.) Ethnoarchaeology: Implications of Ethnography for Archaeology: 139-163. New York, Columbia University Press. Mamet, H. 1874. De Insula Thera. Lille. Marinatos, S. 1968. Excavations at Thera I. Athens. Marinatos, S. 1969. Excavations at Thera II. Athens. Marthari, M. 2004. Raos and Chalarovounia. Preliminary evidence from two new sites of the Late Cycladic I period on Thera. Άλς 2: 53-65. Michalidou, A. 2007. Metal finds. In C.  Doumas (ed.) Ακρωτήρι Θήρας, Δυτική Οικία -Τράπεζες-ΛίθιναΜετάλλινα-Ποικίλα : 187-243. Athens, Archaeological Society. 86

Anaya Sarpaki: Akrotiri, Thera

Moundrea-Agrafioti, A. 2007. Λίθινα εργαλεία. In C.  Doumas (ed.) Ακρωτήρι Θήρας, Δυτική Οικία: Τράπεζες-Λίθινα-Μετάλλινα-Ποικίλα : 73-112. Athens, Archaeological Society. Narroll, R. 1962. Floor area and settlement population. American Antiquity 27:587-589. Palyvou, C. 2005. Akrotiri, Thera: an Architecture of Affluence 3,500 Years Old. Philadelphia, INSTAP. Papadopoulos, A. 2008. Αρχάγγελος: μία σημαντική θέση στη νοτιοδυτική Θήρα. Άλς 6: 89-94. Pichler, H. and Friedrich, W.L. 1980. Mechanism of the Minoan eruption of Santorini. In C. Doumas (ed), Thera and the Aegean World II: 15-30. London, The Thera Foundation. Sarpaki, A. 1992. A Palaeoethnobotanical study of the West House, Akrotiri, Thera. BSA 87:219-230. Sarpaki, A. 2001. Processed cereals and pulses from the late Bronze age site of Akrotiri, Thera; preparations prior to consumption: a preliminary approach to their study. BSA 96: 27-40. Sperling, J.W. 1973. Thera and Therasia (Ancient Greek cities 22). Athens, Doxiades. Sumner, W.M. 1979. Estimating population by analogy: an example. In Kramer, C. (ed.) Ethnoarchaeology:

Implications of Ethnography for Archaeology, pp. 164174. New York, Columbia University Press. Tzachili, I. 1990. All-important yet elusive: looking for evidence of cloth-making at Akrotiri. In D. A. Hardy, C. Doumas, C.G., Sakellarakis, J.A., and P.M. Warren (eds) Thera and the Aegean World III, Vol.I. London, The Thera Foundation. Vlachopoulos, A. and Zorzos, L. 2014. Physis and Techne on Thera. Reconstructing Bronze Age environment and land-use based on new evidence from phytoliths and the Akrotiri wall-paintings. In G.  Touchais, R. Laffineur and F. Rougemont (eds), L’environnement naturel et la relation homme-milieu dans le monde égéen protohistorique, Aegaeum 37: 183-197. Leuven-Liège, Peeters. Weissner, P. 1974. The functional estimator of population from floor area. American Antiquity 39: 343-350. Zorzos, L. (forthcoming). Reconstructing the Bronze Age Landscape and Environment of Santorini: A Phytolith Analysis approach. In C. Doumas (ed.), Ακρωτήρι Θήρας: Σαράντα χρόνια έρευνας (19672007). Αρχαιολογική Εταιρεία.

87

Feeding Knossos: Exploring Economic and Logistical Implications of Urbanism on Prehistoric Crete Todd Whitelaw

The Institute of Archaeology University College London

Urbanism, agricultural production and provisioning: bringing the country into the city

2011; Müller-Celka et al. 2014; Whitelaw 2018). Such approaches are challenging the long-standing assumption that central Crete was divided among the three principal palatial centres of Knossos, Phaistos and Malia from the start of the palatial period, with stable natural, topographically-defined dependent territories (Whitelaw 2014; 2018).

Despite being one of the defining characteristics of Minoan civilisation, urbanism in prehistoric Crete has received surprisingly little attention, and then, largely descriptive, focusing on individual components (palaces, grand houses, shrines, courts and roads), rather than exploring its nature, characteristics or significance. Surface surveys, both rural and urban, are beginning to at least define the basic dimensions of the phenomenon, but remain under-analysed. On prehistoric Crete, as in other examples of early complex societies, urban centres developed in interaction and competition with others (Renfrew and Cherry 1986; Nichols and Charlton 1997; Hansen 2000; Wright 2005), though in Crete this was extremely insular, involving a very small number of centres, in only distant interaction with similar configurations in the Eastern Mediterranean. This latter provides one of the dynamics in Minoan research, in which we can explore interpretive models based on the far better documented palace-centred societies further east, while critically considering their relevance, local characteristics and development in the Aegean (Whitelaw 2017). The diffusionist/isolationist polarisation of 20th century scholarship (Renfrew 1972; Gale 1991; Cline and Harris-Cline 1998; Parkinson and Galaty 2007) has in the past two decades developed into significantly more critical and challenging explorations of the nature and significance of such interactions (Sherratt 1993; Sherratt 2001; Parkinson and Galaty 2009; Legarra Herrero 2011; Broodbank 2013).

To date, regional surveys, while exploring varied parts of the island, have played only a limited role in such discussions. This is primarily because the initial surveys explored small, often marginal areas, which were largely peripheral to the lowland focus of population and the major centres driving cultural development. The Akrotiri, Kommos and West Mesara surveys (Moody 1987; Hope Simpson et al. 1995; Watrous et al. 2004) started to address this, but were 1970s-80s surveys, characteristic for their time in the limited evidence documented for individual sites. The Vrokastro (Hayden 2004), Gournia (Watrous et al. 2012) and Kavousi (Haggis 2005) surveys of the 1990s provide more information (particularly attempting to define site size by period) on which to construct more dynamic regional settlement histories, and preliminary reports on the Malia (Müller-Celka 2007; Müller-Celka et al. 2014) and Galatas (Buell 2014) surveys suggest comparable levels of detail. But there is a wider limitation of survey data for informing on the sort of territorial questions predominantly asked, in that they can rarely document political boundaries, or looser affiliations (Smith 2005; Casana 2009). This relates to the frequent lack of direct correspondence between the extent of political/administrative and economic regional systems, the variable impacts of political changes on underlying economic structures, the differing timescales of political and economic processes, and the fairly broad temporal resolution of most archaeological data (excavation and particularly survey). Survey data are more directly relevant to addressing demographic and economic rather than political questions, though the former structures and patterns may underpin the latter. In this respect, the variety of regional diachronic patterns noted in Cretan survey data (Driessen 2001b) tells us less about diverse settlement system organisation and more about small surveys sampling variable segments of larger systems.

At the same time, recent fieldwork and reassessments of earlier excavations have documented that Minoan urban centres were comparable in scale and probably complexity to their eastern counterparts (Watrous et al. 2004; Amato et al. 2014; Müller-Celka 2007; Whitelaw 2001a; 2004; Whitelaw et al. 2018; supporting and expanding previous estimates: Demargne and Gallet de Santerre 1953: Pl. 1; van Effenterre 1980; Hood 1958; Hood and Smyth 1981; contra Renfrew 1972: 238-44). The broader implications, in terms of the relationship between urban centres and their support hinterlands, have only begun to be explored (Watrous et al. 2004; Bevan 2010; Bevan and Wilson 2013; Vansteenhuyse 88

Todd Whitelaw: Feeding Knossos

That Cretan settlement data can be used for more nuanced analyses is beginning to be documented (Hayden 2004; Whitelaw 2012; Bevan and Wilson 2014; Müller-Celka et al. 2014; Spencer and Bevan 2018).

limited basis (Bennet 1990; Whitelaw 2018; Bevan and Wilson 2013). Rather than assuming what we should be trying to establish, we can explore urban-focused territories by working from the known to the unknown: defining the likely agricultural support zones or hinterlands of the major urban centres (Whitelaw 2018). Such explorations can construct notional minimum support territories (agricultural catchments), dynamically reflecting changes in centre populations through time and in relation to neighbouring centres. These do not define actual political territories, but can suggest the minimum scale of such, as mapped onto the resource and transport constraints of the local environment. Political territories may be smaller than this if exchange systems allow the transfer of bulk agricultural resources between polities, though internalising access to basic resources is usually a fundamental objective of a political system, to ensure the support of its population. Alternatively, political territories may be significantly larger than such minimal agricultural catchments, usually created through the incorporation of comparable and previously independent neighbouring centres with their own hinterlands, by conquest or alliance (Wright 2000). Such larger polities (territorial states) are often relatively unstable, disaggregating under stress into the constituent building blocks of individual city states and their hinterlands (Marcus 1998; Charlton and Nichols 1997).

Our understanding of economic structures in prehistoric Crete is largely text-based, and is therefore limited by the small number and undeciphered status of the Protopalatial and Neopalatial documents, though systematic studies have provided considerable insights into the contents of some Linear A collections (Schoep 1998-99; 2002a; Palmer 1995; Palaima 1994; Militello 1992). The text-based model of Aegean palace-centred redistributive economies established by Ventris and Chadwick (1956), most clearly articulated by Finley (1957) and developed over several decades (Killen 1985), was linked explicitly though in only limited ways to archaeological evidence by Renfrew (1972). A more critical perspective has developed in the past two decades, hinted at earlier (e.g. Bennet 1988), but most effectively articulated by the detailed comparison of textual and archaeological data (Halstead 1992-93). This fruitful synthesis focusing particularly on palatial resource mobilisation and selective control over production, continues to be developed in both breadth and detail (Halstead 1998-99; 1999a; 1999b; 2011; Bennet 2008; Palmer 1989; 1994; 1995; 1998-99; 1999; Schon 2007; Shelmerdine 1985; Whitelaw 2001b; Bendall 2007; Lupack 2008; Killen 2008). More recently, the entire redistribution model has been explicitly challenged by market models (Nakassis et al. 2011; Parkinson et al. 2013), but the evidence requires a more nuanced consideration of the relative roles of redistributive, reciprocal and market processes (Halstead 2011; Killen 2008; Bennet and Halstead 2014). This is a debate which has only started with reference to earlier Minoan periods, limited by the constraints of the textual evidence, but also due to distractions embodied in the under-developed and confusingly invoked models of factions and heterarchy (Hamilakis 2002; Schoep 2002b; 2006; 2012; cf. Day and Relaki 2002; Christakis 2014).

Agricultural catchment analyses draw on the classic geographical model of the isolated state developed by von Thünen (Chisholm 1968; Hodder and Orton 1976: 229-36). This proposed a zonal model of economic landuse expanding out from a nucleated population centre, based on labour investment in productive activities, distance to fields and movement/transport costs. Different agro-pastoral productive activities require different environmental conditions, so will be adapted to local soils, water access, topography, etc., but also take into account the nature, frequency and seasonal timing of human and animal labour investment. The former are context specific, the latter less so, and so can provide a basis for general models. Crops and other resources which require large amounts or frequent labour investment will be situated close to the settlement, facilitating regular labour and minimising the time lost in commuting to the productive location. Regularly, high-investment gardens will be located close to settlements, lower-investment cereal fields further out, and more extensive cropping, grazing, and wild resource procurement further still. This pattern has been widely documented in recent and historical agricultural production systems, and fundamentally underpins agricultural land-use models (Chisholm 1968).

The present chapter does not have scope to explore in detail the intersection of these urban, regional, social, political and economic perspectives, so aims simply to suggest points of relevance. It focuses on the materially grounded economic and logistical implications of the agricultural provisioning of urban populations, as a concrete framework for starting to integrate some of these broad theoretical concerns in the study of the organisation of agricultural production within regional palatially-organised economic systems. Cities and their hinterlands: models and structuring constraints The common assumption of ‘natural’ topographicallydefined political territories in central Crete has only 89

Country in the City Does this generalised model have any relevance to Bronze Age cultivation systems? We can expect so, since it relates to basic properties such as labour investment and travel/transport costs, rather than the economic structure (e.g. subsistence or market economies) of the society. But there is some question as to whether such a formalist cost-benefit perspective applies in command economies, where the value of labour (e.g. client or slave) may not be considered particularly significant by institutional authorities. In general, the long-term shift from institutional to privatised labour and production from the third through first millennia in the Near East suggests labour costs were considered significant (Hudson and Levine 1996; Diakonoff 1982; Renger 1990; 1995-96; 2007). It is also encouraging that empirical support for such land-use zonation in Bronze Age palacebased economic contexts comes from both textual and archaeological sources in the Near East. Property sale documents from various regions and periods in the Near East, describing the location of fields, regularly locate intensively cultivated gardens and orchards near to city walls, in some cases even within the city (Zaccagnini 1979; Mori 2003: 134-44; Fincke 2000; Wilkinson 2003: 119; Stone 1995: 243; Van de Meiroop 1997: 67-8). Areas of intense cultivation around Bronze Age cities are also now widely documented through radial hollow-ways and manuring sherd scatters around sites, matching estimated site populations, throughout northern Mesopotamia and eastern Syria (Wilkinson 1994; 2003: 111-27; 2010; Ur and Wilkinson 2008; Ur 2009; Casana 2013a). Such manuring scatters have been documented for later periods in survey in Greece (Bintliff et al. 2007; Forbes 2013), and appear to be represented in the Gournia and Pseira surveys, with off-site prehistoric material densest in the vicinity of the principal sites and site clusters (Watrous et al. 2012: map 26; Betancourt et al. 2005).

Figure 1: Cretan farmers, 1948: walking time to most distant field, and to market (Allbaugh 1953: Tables A82, A101).

with panniers, carts or sleds/sledges. Human transport alone puts serious constraints on the movement of bulk agricultural produce (Drennan 1984), with equids and bovids able to carry or pull far larger quantities. Animal traction for ploughs or sleds is suggested in Neolithic Crete, based on skeletal traumas among cows at Knossos (Isaakidou 2008). Equid transport has been proposed from the middle of the EM period, based on changes in the size and quantities of pottery vessels transported (Whitelaw et al. 1997; Whitelaw 2015), and the earliest occurrence of equid bones elsewhere in the Aegean (Brodie 2008). While a cart model is known from early MM Palaikastro (Bosanquet and Dawkins 1923: 17) and chariots are documented in the LMIIIIIA Linear B texts from Knossos (Crouwel 1981), roads within Minoan communities are usually too narrow for wheeled transport, open drains will have been difficult to navigate alongside, and no traces of wheel ruts have been reported. If the unexcavated rural roads identified as Minoan are prehistoric (Tzedakis et al. 1989; Müller 1991), regular steps, as in more recent dromakia, suggest direct animal carrying rather than the use of carts or sleds, still the norm in early post-war Crete (Allbaugh 1953: 311-13; Rackham and Moody 1996: 157-58; Blitzer 2004: 206; Naval Intelligence Division 1945: 312). Carts are documented for local transport of bulky firewood in central Anatolia (Gökçek 2006) and in northern Mesopotamia also for other products (Fincke 2000), but agricultural produce seems to have been principally moved more flexibly by donkey.

Travel costs influence the spatial zonation of cultivation patterns in two ways: the regular commute of farmers to their fields, and the transport costs of bringing bulk agricultural products to consumers (home or market) (Chisholm 1968; Carlstein 1982). Data collected in Crete in 1948 (Allbaugh 1953: 539, 556) indicate that early modern subsistence farmers conformed to this logic, making cost-effective decisions about access to their own fields, and taking their surpluses to market (Figure 1). This does not mean ancient Cretans necessarily behaved the same way, but does indicate the relevance of the general models to the specific Cretan landscape with pre-mechanised agricultural production and transport technology, in a subsistence agriculture oriented context. Similar patterns are supported by data collected widely in comparable contexts (Chisholm 1968; Haggett 1965: 161-75).

Without using carts, transported crop volumes will be lower relative to the maintenance costs of the animals involved. With a 90kg load of grain (standard for the

Inland fields will only be accessible through terrestrial transport involving human porterage or animal transport 90

Todd Whitelaw: Feeding Knossos

Old Assyrian pack-trains: Veenhof 1972: 45; Larsen 1976: 102), the donkey and its human guide could consume 13% of the grain during the 100km (2 day) round trip from Phaistos to Knossos, and 20% on the 3 day round trip from Gournia to Knossos, indicating serious terrestrial transport constraints on the scale of effective regional agricultural support zones. Efficiency would be increased slightly by using a pack-train of several animals and only one human driver.

We have less detailed information from the Linear B records on taxation at Knossos, but two points also emphasise the ‘tyranny of distance’. The far western part of the polity appears to have been managed somewhat at arms length, devolved to a degree to the second-order centre at Khania (Bennet 1985; Petrakis 2014). Within the central and eastern parts of the polity, considerably closer to Knossos, there still appear to have been distance effects. Hart (1965) and Bennet (1985; 1988; 1992) have noted that individuals dubbed ‘collectors’ by analysts (Bennet 1988; 1992; Rougemont 2009), who share management responsibilities with the palace, appear to have a greater role in contexts further from Knossos. It is argued on the basis of some geographical localisation of their responsibilities that these individuals were local elites (Bennet 1988; 1992). So it appears that further from the palace, they could, or were expected to take on greater managerial responsibility. This was perhaps a double-edged relationship: because they were further from the centre, they could probably get away with more independent behaviour, and because the palace’s access to information and ability to be directly involved declined with distance, the palace had to rely on them to take a more independent role. Both observations indicate that even over distances measured in tens of kilometres, administrative practices varied, seemingly responding to the ease and effectiveness of communication and transport.

For most settlements, the areas that could be cultivated effectively by resident farmers commuting daily to their fields are usually assumed in site catchment analyses to be within a 1 hour radius of the community (Bintliff 1999). However, production effectiveness declines continuously with distance (estimated by Chisholm 1968: 52-53 at 15-20% loss per kilometre), so declining labour inputs and likely productivity can be expected at 1-2 hours, approximately a 5-10 km radius. This will tend to limit a largely self-sufficient urban population to the maximum which can be supported by the agricultural produce of such a cultivation zone (Wilkinson 1994: 495-99; 2005; 2010; Bintliff 2002; 2006). To expand beyond this limit requires either outlying seasonal hamlets or farms, occupied during the peak labour periods of the agricultural year (e.g. Whitelaw 1991), or the city will need to be supplied by surpluses produced as taxes or tribute from neighbouring subordinate communities (Wright 2000), or through inter-polity exchange. Crucially, transport (particularly terrestrial) costs also put distance constraints on the extent and effectiveness of such more extended supply networks.

The effect of distance on subsistence support ranges could be mitigated by riverine or coastal transport; recent assessments of the relative costs of terrestrial (cart) vs water transport based on Roman and medieval records document considerable differentials (sea: 1/ river: 5-10/ road: 52), though sea costs had probably been reduced considerably in the context of Roman pacification (Scheidel 2014: 9-10). In addition, larger Roman ships will have been considerably more efficient than small Bronze Age vessels (Monroe 2009: 82-9). Exceptionally large ancient cities (e.g. Memphis, Assur, Babylon, Rome, Alexandria, Constantinople) were usually either coastal or riverine, and could sustain their large populations through the bulk import by water of subsistence produce. On Crete, most rivers are seasonal and the few perennial rivers would only have been navigable for a few kilometres inland, at best. Coastal transport was less limited, but the benefits would be significant only with increasing cargo value/bulk and increasing distance. Along the Cretan coast, travel times might be cut by a day (for potential impacts on communication see Bevan 2010: Figure 3), but transport risks and handling (loading/unloading) costs increased. Early modern coastal trade on Crete is documented (e.g. Blitzer 2004: 212; Betancourt 2004), but its scale and significance, particularly for bulk agricultural transport, are not clear. For a coastal centre such as Malia, with a possible port immediately adjacent at Ayia

That these constraints were real and relevant in prehistoric Aegean contexts, and not just a modern economic concern, receives support from the spatial structure of economic behaviour documented in the Linear B records. At Pylos, the kingdom was administratively structured into two provinces, either side of the Aigaleion range, and further sub-divided into some 16 districts, each managed by local officials (Bennet 1999; 2011). While the palace maintained detailed records which noted interactions with individuals or groups scattered through the kingdom, the preserved records document greater interest in individuals closer to the centre at Pylos. Economically, a limited range of non-subsistence resources was collected as taxes from the entire kingdom, organised by (so devolved to) the districts (Shelmerdine and Bennet 2008). In contrast, it is thought that the palace’s own land-holdings were concentrated closer to Pylos, and most of its subsistence support probably derived, directly or indirectly, from these palace and damos lands (Bennet 2011; Killen 1987; 1998; 2008: 165-73). So the bulkier agricultural produce was obtained relatively close to hand, reducing transport costs. 91

Country in the City Varvara (Müller-Celka 2007), marine transport may have been relevant, though the agricultural capacity of the Malia plain and nearby inland valleys appears to have been sufficient to support the population likely for the centre and its dependent communities (Müller-Celka et al. 2014; Whitelaw 2018). For Knossos or Phaistos, land transport over 6 kilometres would still need to be organised from their associated ports at Katsambas and Kommos. This also does not consider the shipping costs, not least financing the construction and maintenance of ships (Monroe 2010; Clark 1958). Given such considerable costs, organised institutional support for maritime trade, as represented by the substantial ship sheds housing small numbers of ships (Shaw and Shaw 1999), would more likely be aimed at high value to bulk goods, e.g. medium to long distance trade in valuables and value-added goods, rather than the short-range movement of basic agricultural produce (though see Day 1997 on intra-island transport of amphorae).

surprisingly, different Roman cities, at different times, seem to conform to a different balance of the models, with Rome and a few other major administrative cities, usually coastal and on major trade routes, falling at the commercialised end of the scale. The Greek city-states fall largely into the consumer, agricultural centre category, being small and replicating each other across the landscape (Kirsten’s ‘villagestate’), particularly where topographic relief does not distort local agricultural potential and transportation (Bintliff 2002; 2006). Most Roman cities present a similar picture, being small, serving as local centres (BekkerNielsen 1989; Bowman and Wilson 2011; Bintliff 2002). Only a relatively small number of Classical, Hellenistic or Roman cities were exceptionally large, usually serving as trade and administrative hubs for very extensive territories. Intriguingly, even Phaistos and Malia are large in comparison with the majority of later cities, with Knossos standing out (Figure 2, excluding exceptional cities larger than a square kilometre, and palimpsests where period-specific estimates are unreliable).

Debates over the scale and significance of transport and trade echo the range of views on the character of Minoan urbanism; whether urban residents were principally engaged in agriculture or craft production and trade. The latter view is very modern, and is relevant to exceptional centres in the pre-industrial past, usually those situated on coastal, riverine or major inter-regional overland trade routes. The vast majority of communities will have been simpler agro-towns, where most residents were full-time or part-time farmers, cultivating fields immediately surrounding their city to support themselves and a limited number of administrative, religious and occupational specialists (Bairoch 1988). There will have been limited crafts and services addressing the needs of the population of the town and neighbouring dependent dispersed villages and hamlets, with non-local raw materials and limited finished goods brought in by trade.

Recent research with both texts and archaeology relating to Bronze Age Near Eastern cities similarly stresses the agricultural role of the majority of the urban population (Renger 1990: 23; Stone 2007: 231; Ur 2013). City size data (while varying considerably by period) indicates a greater degree of nucleation than for the Greek and Roman examples, but relatively few sites exceed 40ha, representing cities with more than 5,000 inhabitants (Figure 2C). These comparisons approach our Minoan case from two different directions, in terms of the local Mediterranean environment and its constraints, for both agricultural production and transport, but also organisationally for Bronze Age palace-centred states. Aegean prehistory has not yet engaged with the consumer-producer city sort of debate. For prehistoric Crete, the general view still seems fundamentally coloured by Thucydides’ assertions about the Minoan Thalassocracy, regardless of the practicalities, either in terms of production or consumption demands. The most explicit argument for a ‘producer city’ model has been advanced for Gournia (Watrous 2007; Watrous and Heimroth 2011; Watrous et al. 2012), a small regional centre (see Christakis, this volume). In contrast, I see the evidence for craft production as representing parttime cottage industries, seasonally supplementing basic agricultural production by some residents (Branigan 1972; Betancourt 1995). But we simply do not have the detailed, reliable and quantified evidence to adequately assess either or other interpretations. By and large, except for pyrotechnical industries requiring kilns, few specialised facilities or working areas were required for generally small-scale prehistoric craft production,

Some of the most detailed models for ancient urbanism have been developed and extensively debated in Roman history. These have been explored for several decades in Roman archaeology (Parkins 1997a; Bowman and Wilson 2011; Erdkamp 2001; 2012; 2015), and are now increasingly being considered for the Greek polis (Hansen 1997; 2004; Alston and van Nijf 2008; Osborne and Wallace Hadrill 2013), in both cases taking advantage of limited and patchy textual information. Two contrasting models formulated by Weber (1958) and developed by Finley (1973), are of the consumer city and the producer city. The former emphasises an agricultural basis for wealth (particularly land rents for elite city dwellers), the latter industry and commerce; both models focus on the bases for wealth, not basic subsistence. This debate has raged for several decades, with variants and additions (Parkins 1997b; Mattingly 1997; Erdkamp 2001; Stone 2013). In fact, not 92

Todd Whitelaw: Feeding Knossos

and until very recently, excavation deposits were rarely sieved, so the bulk of production debris will have been missed. In re-assessing older evidence (e.g. Watrous and Heimroth 2011), it is usually difficult if not impossible to consider the formation processes of the original deposits, and so assess whether individual artefacts represent in situ craft production. The MMII maisonateliers associated with Quartier Mu (Poursat 1996) and the LMI Artisans’ Quarter at Mochlos (Soles 2003) are the only well-documented workshops of the palatial period, other than pottery kilns (Shaw et al. 2001; Tomasello 2011; Puglisi 2011). These are extremely limited examples, especially considering the amount of LMI townscape exposed, and give us little basis on which to assume that Minoan towns were primarily craft or commercial centres. Are there other approaches that could at least give us an idea of the relative scale of the different sorts of occupational activities? The non-agricultural proportion of an ancient population is extremely difficult to estimate through either textual or archaeological evidence. The former is, for any site, period or region, usually limited to a small sample from only particular contexts. For the ancient Near East, we may have temple, palace or household records, but usually representing an unknown segment of the institution or population, and never in quantities and combinations that can be played off against each other for a full or balanced picture. Similarly, archaeologically, excavations are overwhelmingly biased toward major monuments and the institutions they represent, and rarely inform us in any comprehensive or representative way about the vast bulk of the population. Given the bias of textual records toward the urban and elite, and the bias of most archaeological excavations toward elite contexts, it is difficult to achieve any overview.

Figure 2: Site size distributions: A. Cretan survey data (Moody 1987; Hope Simpson et al. 1995; Hayden 2004; Haggis 2005; Watrous 1982; Watrous et al. 2004; Watrous et al. 2012); B. Mainland Greek survey data (Cavanagh et al. 2002; Cosmopoulos 2001; Forsén and Forsén (eds) 2003; Jameson et al. 1994; Mee and Forbes (eds) 1997; Wells and Runnels (eds) 1996; http://classics.uc.edu/prap/static/sites_list.xsl. html; C. EBA South Mesopotamian survey data (Adams 1965; 1981; Adams and Nissen 1972; Gibson 1972; Wright 1981); D. Aegean Archaic-Hellenistic poleis (Hansen and Nielsen 2004); E. Roman cities in Britain, Gaul, Iberia and Asia Minor (Millett 1990; Pounds 1974; Carreras Montfort 1995-96; Hanson 2011).

Looking for a broad understanding of the social composition of early states, there are general expectations that more than 70-80% of the population will be directly involved in agricultural production (Trigger 2003: 155; Bairoch 1988: 497; 80-90% for Rome: Erdkamp 2012; 2015), but very few cases provide reliable textual documentation. Imperial China c. 1900 AD, allows estimates that the royalty, elites and administrators constituted c. 2% of the population, artisans and merchants c. 5-6%, and agricultural peasants c. 92-93% (Bodley 1994: 209-10). Census data for the indigenous population of central Mexico in AD 1560 indicate 12.1% nobles, 13.4% artisans, 4.2% merchants and 70.3% farmers (Sanders and Santley 1983: 281). More directly relevant to our context, records from LBA Ugarit and Alalakh provide limited documentation. At 15th century Alalakh, partial census, army recruitment and supply lists seem to list individuals by social class, indicating approximately 11% elite, 24.0% crafts and service, and 65% farmers 93

Country in the City (including a few with craft or service occupations) (von Dassow 2008). In 12th century Ugarit, estimates vary, with their basis rarely spelled out explicitly, but elites, administrators and specialists probably accounted for 20-35% of the population of the kingdom, with most administrators and specialists resident in Ugarit itself (Liverani 1982; Vita 1999). For Early Dynastic Shuruppak in Mesopotamia, the documents suggest at least 2,000-3,000 administrators and specialists (Foster 2005; Visicato 1995: 26) in an urban population estimated archaeologically at 20,000, though serving a regional population possibly half again as large, so up to 90% farmers and independent crafters. This eclectic sample suggests only broad patterns, but provides a general picture of c. 10-35% elite (however defined), crafters, merchants and service personnel, with the bulk of the population (65-90%) agricultural producers, aggregating urban and rural populations. Approaching this question theoretically through estimates of Bronze Age production rates, yields estimates of 74-86% of the population engaged in agricultural and utilitarian craft production, for Egypt and Cyprus, the difference due to more productive Nile-flood agriculture in Egypt (Padgham 2014: 101). The differential concentration of elites, administrators, crafters, traders and service personnel in larger centres is documented through administrative records in Roman Egypt (Alston 2002: 334-39) and medieval and early modern Europe (Clark and Lepetit 1996).

This would translate to a minimum of 6-9% elite (and some artisan) families in the polity. The Jn series of tablets, recording the issuing of copper/bronze to smiths throughout the polity, is reconstructed as originally referring to up to c. 400 smiths (Nakassis 2013: 170). An archaeological consideration of the scale of pottery production working up from household consumption figures, suggests c. 450-500 potters throughout the kingdom (Whitelaw 2001b: 68). In both cases we can consider these to have been parttime crafters. There will also have been carpenters, masons and probably leatherworkers at a local town/ village level, as documented at Ugarit (Heltzer 1976; 1979; 1982; 1987) and Alalakh (von Dassow 2008: 25657) and towns in Roman Egypt (Alston 2002: 334-37). More complex crafts will have been represented by relatively few practitioners, and the most elite crafts appear to have been controlled by the palace (Voutsaki 2001; Bennet 2008; Killen 2008; Schon 2007). Specialist textile workers are well-documented in the Linear B records from Knossos, but less so at Pylos; even so, we are talking of a minimum of 1,600 palace dependants, many adult women (650), but including some men and sub-adult children. Taking into account all these figures, we might consider on the order of 25-30% of the population as elite, administrators and crafters, at least 5% as dependent labourers for the palace, and the balance, some 65-70%, farmers and herders. The Linear B tablets from Knossos give us an even patchier picture, which we cannot compare so directly to the archaeological record. The tablets, probably from at least two chronological horizons (Driessen 2000), mention some 1000 different individuals by name (c. 800 in the main phase). While initially suggesting broad parity with the smaller kingdom of Pylos, the two polities were organised differently, so there is no expectation that the palace records should be closely comparable. In addition, the preserved Knossos records are far more partial than those from Pylos, and are dominated by sheep management and textile production records. More than 300 of the named individuals are identified as shepherds, but their responsibility in some cases for multiple flocks at different locations, suggests that in parallel with comparable records in Near Eastern institutional archives, some at least were managers, rather than actual shepherds. Some 40 named individuals, including the ‘collectors’ (Bennet 1992; Rougemont 2009) coordinate productive responsibilities in some relation to the palace (Landenius Enegren 2008), and have been identified as administrators and elites. In addition, there were some 65-75 scribal hands represented among the preserved final phase tablets.

While we have no documents comparable to those from Ugarit or Alalakh from the prehistoric Aegean, individual sets of Linear B tablets, particularly when contextualised with archaeological data, provide some suggestive figures. The incomplete archival documents from the palace at Pylos list some sort of palatial involvement with c. 4,100 individuals (Hiller 1988; the tabulated figures add to 4,595-4,654), within a total polity population estimated roughly on the basis of archaeological survey data as c. 50,000 (Chadwick 1976: 68; Whitelaw 2001b: 63-64). Of this regional population, approximately one quarter would be adult males, and of these, 3,000 are referenced in the tablets, c. 24%. This suggests a minimum palatial interest in one-quarter of the adult male population (or families) within the kingdom. This compares to more than 30-45% in the more complete documentation from Ugarit (Liverani 1982; van Soldt 2010; Vidal 2014), at least 30% at Early Dynastic Lagash (Tyumenev 1969; Diakonoff 1969), and c. 50-60% at Early Dynastic Shuruppak (Visicato 1995; Foster 2005), all with more extensive institutional economies than are documented for Mycenaean polities. Nakassis’ recent prosopographical study of the Pylos tablets documents c. 740-1115 named individuals in the texts, most of whom are listed in an active capacity such that he considers them relatively high ranking.

In contrast with Pylos, we have no detailed estimates for the overall population of the Knossian LMIIIA2 polity (Firth 1994-95 makes many unjustifiable 94

Todd Whitelaw: Feeding Knossos

assumptions), but extrapolations from survey data might suggest on the order of 130-150,000 individuals. While the population of Knossos itself at this period is also not clear, present indications are that there was a significant reduction from its Neopalatial maximum of c. 20-25,000. If we propose a notional city population of c. 15,000, this would include some 3,300-3,750 adult males or families. If half of the named individuals maintained a residential presence at Knossos (with some probably also having a local base elsewhere), these would represent c. 10-12% of the city families. For a polity population of 130-150,000, the 800 named individuals would represent c. 2.4-2.8% of families. Given the far more partial preservation of the Knossos Linear B archives, these figures do not obviously represent a situation incompatible with that at Pylos.

palace-centred production of faience (Foster 1979), the Knossian production of carved relief stone vessels (Bevan 2010: 42-43), and probable restriction of the production of pottery in the ‘special palatial traditions’ to major centres (Betancourt 1985: 140-48). Many institutional archives from the Near East, whether from palaces or temples, provide partial documentation of the rations issued to their employees and dependents. These may include service staff, but also workers in institutional workshops and agricultural estates. This makes it difficult to quantify the scale of staffing of the urban institutions themselves, which we might use as models for the scale and nature of the support staff associated with Aegean palaces. Instead, we can only cite individual cases, such as at Girsu in the Ur III period, where the wife of the king had a household of 700 (servants, etc.), with many more staff serving other branches of the royal household (Tyumenev 1969). In extensively excavated cities like Ebla, Mari, Ugarit and Amarna, it is clear both from excavations and archives that there were multiple palaces hosting specific royal activities and associated with different members of the royal family, as well as other elite members of the community. Similarly there were multiple temples with their own often extensive establishments. While the importance of individual institutions varied in time and space, there appears to have been a decline in the importance of temples relative to palaces as one moves later in time and westwards, with smaller temples among Levantine city states, and the near disappearance of distinct temples in the Aegean.

Overall, while we have less information on the composition of society than for some early state cultures, what information we have for the LBA Aegean polities is not obviously anomalous in comparison, nor particularly surprising, and we can expect that some 65-75% of the overall population was engaged in agricultural production; my own inclination is toward the higher end of that range, particularly as most of the more widespread categories of crafters would have been part-time specialists. Simply for comparison, in Crete in 1928, 69% of the working population was engaged in farming and 31% in industry, trade, services and administration (Allbaugh 1953: 63). From this, it seems pretty safe to conclude that Minoan urban communities, like other ancient Mediterranean, East Mediterranean and Near Eastern Bronze Age centres, were agro-towns, with the majority of their population farming their immediate hinterlands. Only a small number of major (palatial) centres can be expected to have had significantly more differentiated populations. Can we develop this further?

We can expect the larger and palatial centres to have been characterised by exceptional occupational diversity and greater non-agricultural specialisation. In particular, the presence of elites at courts in palatial centres will have generated competitive demand for fine crafted goods (cf. Erdkamp 2001; 2015; Bang 2009; Clark and Lepetit 1996), and driven stylistic innovation and elaboration. Interestingly, while the Ugarit documents are said to demonstrate this, with most craft specialisations concentrated within Ugarit (Liverani 1979), the Alalakh texts may indicate a broader pattern, with individuals in the dominantly elite and crafting classes present in many rural communities (von Dassow 2008: 254-55), though as with rural communities at Ugarit (Heltzer 1976; 1982), many specialists are likely to be low level (utilitarian potters, smiths, leatherworkers) and some will be attached to the rural estates of elites or the palace. The seeming contrast may therefore be a consequence of the different kinds and detail of the documents we have from each polity, and differences in the definitions of the categories used to record individuals. The evidence for urban concentration has led some to assume that the urban population of Ugarit was essentially all administrators and specialists (Liverani 1979), but

Drawing upon another geographical generalisation, we should expect the most specialised occupations (administration, religion, service, fine crafts) to be concentrated in the larger urban centres, the cornerstone of models for economic central place hierarchies and networks (Garner 1967; Smith 1974). The rationale is that goods and services need to map on to consumer demand. For basic products (agricultural, everyday utilitarian crafts) this is broadly determined by the distribution of population, but some goods and services are only desired by or available to segments of a population, particularly the wealthy or elite. We can contrast the broad demand for everyday pottery and metalwork, serviced by village potters and smiths, with elite crafts, probably largely controlled by the palace and destined through gifts, or limited by sumptuary laws, to members of the court and client elites. We can see suggestions of this for Neopalatial Crete with the 95

Country in the City this is contradicted by the widespread distribution of agricultural production facilities in the city, and used sickle blades in most houses (Schloen 2001: 33543; McGeough 2007: 265-308). On this evidence, most households are expected to have been engaged in agricultural production to some degree (Schloen 2001: 342; McGeough 2007: 291, 380).

landscape. To do this, we need to estimate urban populations and reconstruct agricultural territories, as they changed through time, assess agricultural production and consumption, and connect these through the practical logistics of provisioning the urban population. All of these calculations involve considerable (though informed) guesswork, but fully justifying and exploring the possible variants is beyond the space available here. For the exploratory purposes of this chapter, I will use conservative values, to consistently aim at minimal estimates. The purpose is to outline a perspective and begin to explore some of its implications for how we understand Minoan agricultural economics and in particular, how we consider the relationship between the city and the countryside.

In prehistoric Crete, Malia in particular provides evidence for multiple major architectural complexes constructed in the Protopalatial period, with the palace, the Hypostyle Crypte, the Centre Politique, Quartier Mu, and potentially comparable institutions represented by the Magasins Dessenne, and any predecessor to Quartier Episilon. Against the wider Bronze Age background, it is not necessary to imagine some sort of unique heterarchical social structure (Schoep 2002b; 2004; 2006; 2012), when multiple institutions, interlinked, nested or working at different scales, are welldocumented as integral to Bronze Age hierarchical political systems. Regardless of their relations with each other, we can anticipate that each of these institutions will have had their own associated staff, even if we cannot quantify these (cf. Davies 2004). Adding to this in the Neopalatial period are the grand houses (Hägg 1997; Betancourt and Marinatos 1997; McEnroe 1982; 2010), particularly common in the palatial centres (Whitelaw 2001a: 19-24). The wealthy families represented by these structures were consumers of fine crafted goods and competition amongst them undoubtedly drove the diversification and elaboration of material culture that particularly characterises the later Neopalatial period. We can therefore expect the occupants of these structures, the personnel who serviced them, and the crafters who embellished and equipped them, to contribute to greater social and occupational differentiation in larger, particularly palatial centres, in turn reducing proportionally the agricultural residents of such centres, a pattern characteristic of metropolitan centres (Clark and Lepetit 1996).

Elsewhere, I have reviewed archaeological approaches to estimating Minoan community populations in general (Whitelaw 2001a), and in detail for Knossos (Whitelaw 2004). While the densely occupied cores of Minoan sites document up to 400 persons/ha, densities arguably drop towards the periphery of communities. Overall, I have suggested an average of 200-225 persons/ha for major Minoan communities; for villages and hamlets, I suggest a lower figure of 100 persons/ha. This distinction receives support from the periphery of Pseira (McEnroe 2001: Figure 1), and survey reports where individual structures have been recognised in loosely aggregated small communities (e.g. Moody 1987; Hayden 2004; Haggis 2005; Watrous et al. 2012). Comparatively, this distinction can be documented for LBA communities integrating archaeological and textual evidence in the territory of Alalakh (Casana 2009; 2013b), and has been argued based on early modern Cretan village data (Price 2011; contra Aurenche 1981, which only considers villages). Overall, these occupation densities are comparable to those documented in other early urban contexts (Bairoch 1988: 22-24).

Viewed both comparatively, and in terms of relevant Aegean data, we can recognise the fundamental agricultural orientation of prehistoric Cretan urban centres. However, without a detailed and substantial textual record, or extensive and representative samples of the archaeological record, it will be difficult to refine our understanding of the composition of urban populations. But we can explore this further, particularly its economic and logistical implications, through considering the relations between the urban centre and the surrounding country.

The recent intensive survey of Knossos documents continuous Neopalatial occupation over some 100ha, with in addition a suburb on the summit of the hill of Ailias to the east (Whitelaw et al. 2018). I would put the overall urban population in the region of 20-25,000 individuals. Consideration of excavation and survey evidence for the urban centres at Phaistos and Malia suggests they each reached their maximum extent in the Protopalatial period, at some 50ha (Whitelaw 2012; 2014; 2017); both may have contracted or declined in occupation density in the Neopalatial period (for Malia see Alberti et al., for Phaistos Amato et al. 2014: 131). Elsewhere on Crete, Palaikastro was an extensive centre in the Neopalatial period, Zakros and Khania were significant though considerably smaller, while the extent of the community surrounding the palace at Galatas remains to be documented in detail but is on

Defining urban support territories in prehistoric Crete To explore these suggestions about Minoan urbanism in its local context requires mapping them on the 96

Todd Whitelaw: Feeding Knossos

that order of magnitude (Buell 2014). A limited number of town sites can be documented in the 4-10ha range, on the basis of outlying soundings well away from their main excavated structures, including Archanes, Poros, Amnisos, Kastelli (Whitelaw 2001a: 28-29) and probably Priniatikos Pyrgos (Hayden et al. 2007). Other centres such as Gournia, Petras, Sissi, Mochlos, Kommos and Ayia Triadha are smaller, between 2 and 4ha, and represent communities which may have been the centres for small polities or second-order sub-centres within larger polities. Knossos is the only polity with documented significant (>4ha) sub-centres within its hinterland, representing a more complex hierarchical regional structure. Figure 3 plots site size data for the major surveys; very few Minoan sites would be considered as urban on any comparative basis (Bairoch 1988: 217-19). To define agricultural hinterlands, we need to know how much land is required to produce the subsistence requirements for the urban residents. While this will vary for different crops comprising potentially variable Mediterranean diets, we can simplify the calculations by assuming cereal equivalents, since cereals provided the major component of Mediterranean diets in the recent past and in antiquity. Individual crops are variably productive in different conditions, particularly depending on soils, water and the technologies available to work different types of soils. We are increasingly able to assess these characteristics, though still most effectively in relatively small study areas, where soils can be analysed and mapped (e.g. Parsons and Gifford 1995; Morris 2002). GIS-based analyses are allowing such approaches to be extrapolated to larger areas (e.g. Fernandes et al. 2011; 2012), though we do not yet have the data to work at the scale of central Crete. However, even very generalised average values allow us to block out gross patterns. In the following exploration, I have excluded as uncultivable, land with > 10 degree slope, which is usually unstable when cultivated without intensive investment in agricultural terraces. While some terraces have been attributed to the BA, the topographic distribution of sites in areas intensively surveyed in Crete suggests that this was an exceptional rather than common strategy. Figure 3: Cretan Neopalatial site size histograms: A. Akrotiri (Moody 1987); B. Mesara (Hope Simpson et al. 1995; Watrous et al. 2005); C. Malia (Müller-Celka et al. 2014); D. Lasithi (Watrous 1982); E. Mirabello (Hayden 200; Haggis 2005; Watrous et al. 2012); F. Crete, major surveys; G. Crete, community populations, 1948 (Allbaugh 1953: Table A10).

The most dramatic transformation of our understanding of Aegean Bronze Age agriculture has been the recognition that extensive plough cultivation was probably highly restricted in the Bronze Age, due to the costs of raising and maintaining plough oxen (Halstead 1999a; 2014). It seems far more likely that the bulk of prehistoric Aegean farmers cultivated their crops through intensive horticulture, taking significantly less land, but working it more intensively and requiring considerably more human labour per hectare. Extensive agriculture with plough oxen may even have been a palatial monopoly (Killen 1993; 1998; 97

Country in the City

Figure 4: Central Crete with Knossos, Malia and Phaistos, estimated minimum agricultural catchment sequences.

Halstead 1999a; 1999b). This emphasis is also apparent in the Bronze Age Near East, with the accounting systems for institutional farms structured around plough teams. With this in mind, I have estimated that c. 0.5ha of cultivable land would be needed for each individual (cf. Halstead 2014: 247-48; Isaakidou 2008: 98-104), to model the area required to produce the basic subsistence resources for prehistoric Cretan populations. This figure incorporates a great deal of uncertainty, including all of the local factors mentioned above, but also characteristics of individual household production strategies, including crops, rotation, fallow, intensity of soil preparation and weeding, and available labour (discussed in Halstead 2014). Despite these major uncertainties, there is still value in working through the broad implications, for a first-order approximation.

and Friesen 2009). It is possible to see how such catchments change as a centre develops through time, and how the support territories of different centres may have impacted on each other. I plot such notional territories for the major central Cretan palatial centres through time in Figure 4.

I have mapped such catchments elsewhere (Whitelaw 2018), and simply summarise several points here. First, these are very much notional subsistence support territories. They do not predict actual territories, but allow us to consider the scale of the territories involved. Second, these are minimal support territories, estimating only absolutely basic subsistence needs, they do not allow for extravagant consumption, palatial ceremonies, feasts at various scales, or surpluses to support trade, etc., on top of the population’s basic subsistence needs. Suggesting that such baseline calculations are relevant, are estimates that some 84% of the population of the Roman empire lived on less than 1.5 times their basic subsistence needs (Scheidel

The early development of the major central Cretan palatial sites has been outlined elsewhere (Whitelaw 2012). Here it can simply be noted that these centres expanded broadly in parallel into the later Protopalatial period, after which Knossos alone continued to expand through the Neopalatial period. During MMIIIA, the territory necessary to support the rapidly expanding population of Knossos will probably have expanded into the Pediada area. This fits with the argument regularly made that the palace at Galatas was established through expansion or conquest by the palatial power at Knossos, 16km away (Rethemiotakis and Christakis 2011). While initially argued on the basis of the architecture and finds in the palace at Galatas, the survey of the surrounding

This approach allows the exploration of the development of polities in central Crete which directly challenges the assumptions noted at the start of this chapter which have dominated Minoan archaeology: that there are ‘natural’ topographically-defined territories which divide central Crete neatly among the three major palatial centres, and that these territories were stable throughout the palatial periods (Whitelaw 2014; 2018).

98

Todd Whitelaw: Feeding Knossos

landscape documents the establishment of a settlement system focused on the palatial centre at the same time as the palace was constructed (Buell 2014). The rapidity of development of dense settlement conceivably represents the organised colonisation of the territory. The abandonment or down-grading of the palace in LMIA seems likely to represent a significant change in the administrative relationship with Knossos, perhaps a shift of second-order centres, as Knossian territory continued to expand.

hamlets. But these rural demographic changes were probably swamped by the presently undefined scale of population decline at Phaistos itself. Our understanding of Malia has also been changing recently. The palace, as visible today, was laid-out in the later Neopalatial period, perhaps as late as LMIA (Driessen 2010). While there is continuing debate over the degree to which the palace was still in use in LMIB (Pelon 2005; Van de Moortel 2011), its final phase of use appears to have been marked by decline (Driessen and Macdonald 1997: 182-86). In the town, many Protopalatial structures continued in use or were adapted in the Neopalatial period, and fewer new mansions were constructed in late MMIII/early LMIA than at Knossos. MMII appears to have been the period of greatest expansion of the town (Müller-Celka 2007), and later occupation may have been less dense, with some house plots abandoned and not rebuilt (Alberti et al.). At least as profound were changes in the wider territory, now documented by the Malia regional survey. This has revealed a massive contraction in settlement toward the coastal plain in the Neopalatial period, with almost total abandonment of the villages and hamlets of the inland basins (Müller-Celka et al. 2014). Most of the loss is in small hamlets (a 64% decline in sites); in terms of aggregate site area (representing population), about 30% of the rural population was lost (estimated from Müller-Celka et al. 2014: Pl. CXXVI-II). This represents a fundamental re-structuring of the rural settlement system and must indicate dramatic changes in the relationship between urban Malia and its hinterland. This may reflect reduced needs for agricultural produce by a declining population at Malia, improvements in irrigation farming in the plain allowing more intensive production (Müller-Celka et al. 2014), or a shift to more extensive cultivation.

Turning south, surveys of a segment of the immediate hinterland of Phaistos by the Kommos (Hope Simpson et al. 1995) and West Mesara (Watrous et al. 2004) surveys, indicate that Ayia Triadha and Kommos may be considered small subsidiary centres, though Kommos was a specialised port community, and Ayia Triadha particularly develops with the construction of the Villa Reale in the period (MMIIIB-LMIA) when, it is now clear, the palace at Phaistos was in ruins (Puglisi 2003; La Rosa 2010a; 2010b). The city at Phaistos itself may have declined severely during this phase, with only limited MMIIIB-LMIA occupation documented (La Rosa 2010a; Girella 2010), though excavation away from the palace has not been extensive. A revival may have been anticipated with the investment in constructing the new palace during LMIB (La Rosa 2010a). It has been suggested that the Villa Reale at Ayia Triadha may have been constructed under strong Knossian influence (La Rosa 2010b; Puglisi 2003; Privitera 2014). This arguably was a strategic development of a new administrative centre, explicitly displaced from the site of the formerly independent palatial authority at Phaistos. The Villa was accompanied by public courts, a major storage complex (the Bastione), and may have been fortified, but the associated settlement remained small and did not take over the earlier urban functions of Phaistos as a major regional centre (Figure 11). The limited number of Linear A tablets recovered, recording activities of up to 2,788 individuals (Schoep 2002a: 185), and impressive quantities of agricultural produce (Palaima 1994; Militello 1992; Schoep 1998-99; 2002a; Privitera 2014), indicate it had replaced Phaistos as the regional administrative centre. It may have been intended to organise the extraction of resources from the local region (Privitera 2014) to support local Knossian political activities, and potentially transfer surpluses to Knossos. This period of political dependency corresponds with the dilapidation of the civic centre and harbour facilities at Kommos, when the kiln was constructed in its collapsed south portico (Shaw et al. 2001), suggesting a focus on exploiting terrestrial resources rather than maintaining the harbour facility. The rural settlement pattern is marked by the loss of 10% of sites between the Protopalatial and Neopalatial periods, though aggregate site area (representing population) only declines slightly (7%), since the sites lost are predominantly small farms and

Further afield, the Vrokastro, Gournia and Kavousi surveys provide a regional perspective potentially contrasting with the picture of prosperity from the local centres (Gournia, Mochlos and Priniatikos Pyrgos), which develop significantly in the later Neopalatial period. Considering all surveys together, site numbers decline by 21%, while aggregate occupied area declines more significantly, by 27%. This suggests the discrepancy is not just due to the differential loss of the smallest hamlets, as in some regions, but represents a more significant re-organisation of the settlement system. But some of the distinction may also be due to the more detailed documentation of sites in these three surveys, allowing the recognition of changes in site sizes between the two phases. An argument could be made linking such changes to a more effective incorporation of this area into central Cretan economic and political systems, either through trade (Betancourt 1995), or even political dominance (Hayden 2004: 113-14; Haggis 2005: 79; Watrous et al. 2012: 55-57). A decline in local 99

Country in the City products, above). The role of expanding urban and elite demand in re-organising and raising agricultural production is being challengingly explored in analyses of the Roman economy (Scheidel 2004; Erdkamp 2015). While only briefly summarised, these settlement pattern changes in the major surveyed regions encourage us to think in terms of a much more dynamic economic and political history for Crete than has been assumed for most of the past century (Whitelaw 2014). Particularly considering the increasingly refined histories of use of the major palaces (Figure 5), it is possible to suggest, at least for central Crete, a sequence of Knossian political and territorial expansions (Figure 6). We can debate the details of when, how and through what process individual regions were brought under Knossian control; played out over nearly three centuries, the processes will undoubtedly have varied. Whether Knossian dominance was tightly managed or more loosely ideological (cf. Knappett 1999) requires consideration, though the significant transformations in settlement patterns and therefore local economic

Figure 5: Phases of documented use of principal Cretan palaces.

population and the development of coastal access could suggest the re-structuring of rural sites to increase agricultural production, either to provide surpluses for the developing local elites, or as taxes or tribute for export by sea (though bearing in mind the cautions expressed about sea transport of bulk agricultural

Figure 6: Central Crete, a possible sequence for Knossian territorial expansion: A. EMIII-MMIA; B. MMIB-II; C. MMIIIA; D. MMIIIB; E. LMIA; F. LMIB..

100

Todd Whitelaw: Feeding Knossos

organisation, should indicate significant and wideranging, though variable, local impacts.

arguably directly linked to the distinct developmental trajectory of Knossos. It is worth turning specifically to Knossos to see if we can define some of the characteristics of this transformation in greater detail.

Traditionally, it has been largely assumed, and occasionally explicitly argued (e.g. Wiener 2007) that Knossos dominated at the least central Crete, particularly in the Neopalatial period. But the principal difficulty, arguing from architectural, fresco, ceramic and other artefact styles, is our inability to distinguish cultural influence, with innovations (as suggested here), particularly driven by the extreme social competition within the largest and most differentiated population centre (Knossos), from political domination. The latter has been the default assumption since the origins of Minoan archaeology, traceable to Thucydides’ remarks on the dominance of Minos.

Feeding Minoan cities: population, territory and agricultural provisioning Producers and consumers in Minoan urban centres In making the case for Minoan cities as agro-towns, I reviewed Aegean and comparative evidence for the composition of early state populations, concluding for the Minoan case that we might expect on the order of 65-75% of the overall population to have been farmers, many of whom would have lived in the urban centres, commuting to fields surrounding the city. Based on comparative evidence, we can expect the proportion of non-farming families to have been higher in the largest urban centres, though on the basis of Near Eastern and Linear B texts, many elites, administrators and crafters would themselves have cultivated land or had it cultivated for them by dependants. Is it possible to be more specific for the urban situation, by looking at the logistics of such urban farmers?

I have argued here that the preeminence of Knossos can now be documented more clearly by its exceptional growth from the later Protopalatial period. This can also be seen in its regional context, since it is so far the only palatial centre with documented significant subsidiary centres. Some of these, such as Archanes and Tylissos, would have been earlier local centres, subsumed within the territory of Knossos as this expanded. Galatas looks increasingly like an explicit foundation as such a subordinate centre (Rethemiotakis and Christakis 2011; Buell 2014). Phaistos and Malia appear to be formerly independent major political centres, incorporated into the Knossos polity as it continued to expand. That incorporation can be argued on the evidence of significant and sustained set-backs at each site, but also the continuing dramatic expansion of Knossos, driven by its expanding supra-regional administrative role, and supported by its territorial and economic expansion. In the past, the development in the Neopalatial period of the rural villas was assumed to represent an intensification of administrative centralisation of the countryside, tying rural areas more directly to palatial centres (Hood 1983; Walberg 1995; Hägg 1997: 229-31; Betancourt and Marinatos 1997). Whatever this innovation represents, can now be seen to correspond broadly to other major transformations in settlement systems, both at the centres, but also as documented in rural settlement patterns. Considering these three broadly parallel transformations, it is unlikely that they are not connected, and the establishment of the larger villas may represent administrative changes, linked in some way, at least in central Crete, to Knossian territorial, economic and political expansion.

As reviewed above, it is expected that farmers will usually not travel more than 1-2 hours to their fields, or too much of the working day is lost in travelling. While such perceived constraints will be culturally and situationally determined, they at least provide starting guidelines for considering the likely urban hinterlands. In Figure 7, I have imposed 5km circles on the catchments calculated for the three major palatial centres, ambitious one-hour walking distances. The catchments defined also allow for the area required for subsistence support of the populations of any significant sub-centres within the catchment, as well as a background distribution of rural hamlets assuming 100 persons/km2 of agricultural landscape (generalised from survey data). For both Phaistos and Malia, their catchments are, or are largely, within the two-hour limit, even if most of the farming in the outer reaches would have been done by residents of outlying hamlets, and subsistence products transferred to the centre. During the earlier Protopalatial period, Knossos would have been comparable to the other two centres, and could have been largely supported by food produced by resident farmers. But by later LMI, with twice the population of the other centres, much of Knossos’ catchment territory was too distant to be cultivated effectively by farmers resident in the city. So feeding the population of Knossos would have required more complex hierarchical organisation to move substantial quantities of food into the city from dependent communities. Possibly 50-60% of the population of Knossos could have farmed and fed itself from the territory within a notional two hour catchment, while

With all of these changes, it seems increasingly apparent that we are not necessarily documenting a universal transformation throughout prehistoric Crete, though some parallel processes may also have been occurring in distinct regions, in the far east and far west of the island. But in the archaeologically best known part of the island, central Crete, these transformations are 101

Country in the City

Figure 7: Central Crete, Neopalatial catchments and 5 km (c. 1 hour walking time) concentric circles.

the remaining 40-50% must have been supported by food produced by others, brought into the city. Bearing in mind that the mapped catchment is a minimum estimate, and allows only for bare subsistence, the real support territory will almost certainly have been larger, and the percentage farmed by city residents less than 50%.

institutional farms at Early Dynastic Larsa (Foster 1986). In comparison, the occupants of free villages were taxed at 10%, the surplus remaining after feeding their families and covering their own running expenses (Liverani 1989). These were the sorts of surpluses which could be produced in Mesopotamia and the northern Levant (the latter dry-farming). Early modern productivity figures for the Aegean tend to be expressed in terms of the amount of crop per cultivated area, but largely report on plough-based agriculture. The need for fodder for plough oxen increases very considerably the quantities of crops which need to be grown, often beyond the capacities of independent producers (Halstead 2014). Without using traction animals, the limit on farming productivity shifts to labour, rather than land, with more labour-intensive hand-tilled horticulture. Because there are so many relevant variables, it is difficult to quantify the labour costs of hand tillage, but regular surpluses will have been considerably lower per farmer than those recorded in most early modern records. In Venetian Crete, farmers were obliged to hand over 1/3 of their crop to landowners, and were frequently in arrears, a situation which persisted into Ottoman times (Tsougarakis and Angelomatis-Tsougarakis 2004: 390, 409). Elsewhere in Greece in the late Ottoman period, farmers might be allowed to retain as little as 1/5 to 1/3 of the crop (McGrew 1985: 32-40), keeping much of the population in abject poverty, and contributing to the conditions which led to the war for independence.

Those resident farmers could have produced small surpluses which could have been used to feed some of that additional, non-agricultural urban population. Various Near Eastern sources provide comparative data for the institutional organisation of farming. Earlier examples document large institutional workforces, maintained by temples and palaces, but by the second millennium BC, these had largely been replaced by smaller institutional farms which relied on corvée labour during the seasons of highest labour demand, so avoiding the need to support large work-forces throughout the year (Diakonoff 1982; Renger 1995-96). A particularly effective (and exploitative) structure was employed in Ugarit, where royal farms were staffed by dependent male labourers (including slaves, debtors and prisoners: Liverani 1979; 1982; 1989; van Koppen 2001). Because the labourers’ families did not have to be supported, the running costs amounted to the rations for the workers, fodder for the plough oxen, seed for the next year and the costs of transport of produce into the city. The surplus sent to the palace is estimated to have been on the order of 40-60% of the harvest (Liverani 1989), approaching the levels of surplus claimed from 102

Todd Whitelaw: Feeding Knossos

Figure 8: Knossos region, hierarchical provisioning from dependent sub-centres.

Less exploitatively, in northern Greece in the earlymid 20th century, farmers might pay 10% of the yield as rent on a field (Halstead 2014: 240, 302), providing an idea of a normal surplus. Most sustainable taxation regimes could not have been anywhere near as extreme as the late Ottoman, recognising the very low margins and chronic uncertainty of Mediterranean subsistence farming (Halstead 2014).

farming families resident in Knossos, and an additional 13-17,000 farming families throughout the polity. The need to support these non-farmers, in addition to the additional expenses supporting the palace’s activities, can be seen as one obvious rationale for political expansion, to increase the absolute amount of surplus available to the palace. Indeed in a region with few distinct localised resources, obtaining such agricultural surpluses may be a principal objective for territorial expansion. The limited surpluses that could be achieved through intensive horticulture would also provide an incentive for major institutions like the palace to extensify production, and maintain plough oxen (Killen 1993; Halstead 1999a; 1999b). Given the exceptional size of Neopalatial Knossos, we need to consider a more complex supply network than for any other Aegean city for nearly a millennium (Figure 8).

From this, we can anticipate that most farmers will have been able to produce only limited surpluses, so the 5060% of Knossian residents who could farm the territory immediately accessible from the city will normally have contributed only a relatively small amount to the subsistence needs of the rest of the urban population. The remainder will need to have been mobilised as taxes from dependent cultivators further afield, transferred up the settlement hierarchy from hamlets and villages within the territory, or brought in to market by free farmers. Some idea of both the scale, and the degree to which this tied together farmers and non-farmers throughout the polity can be gained from some rough calculations. If farmers could reliably produce a surplus of 10%, then 9 farming families would be required to support each non-farming family (more productive estimates of 2.5-4 families supporting another are based on plough-based production: Bang 2009; Erdkamp 2015). Given an overall LMIB city population of 2025,000, 40% would represent c. 1,800-2,200 families. To support these would require tithes from the 2,700-3,300

Who were the non-farming 40-50% of the city’s residents? This is encouragingly close to the nonfarming percentage (25-35%) of the overall polity population suggested on the basis of comparative Near Eastern and Linear B information on elites, administrators, service personnel and craft specialists. As suggested above, that percentage will have been considerably higher for the urban component of the regional population, because larger urban centres will have a more differentiated population, whereas in rural villages, nearly all of the residents will have been farmers. 103

Country in the City The non-farming urban residents will have provided services to the concentrated population of the city, and to a more limited extent, the wealthier residents in its hinterland. But in addition, a major palatial centre will also house the administrators responsible for the wider territory. The 1,800-2,200 families represented by the 40% are possibly even on the low side compared with the number of elites, administrators and crafters documented above in other early states, but this seems likely to be of the right order of magnitude. The logistics of provisioning Minoan centres

Figure 9: Calculations and estimates of Neopalatial household storage capacities (Christakis 2008: Table 10).

With a general idea of the population of Knossos, we can also estimate the approximate quantities of food necessary to provision it, again only in terms of basic subsistence. The Minoan diet will have varied regionally, temporally, and seasonally, and will have been made up of a variety of locally available resources. Christakis has reviewed a variety of sources for evidence on Cretan diet (2008: 19-32). Here, I will consider this in terms of barley equivalents, to simplify all calculations, consistent with the broad parameters I am trying to establish. If an adult annual consumption of barley is reconstructed as 167kg representing 44% of the diet (Christakis 2008: 29-30), 380kg represents the simplified proxy for the entire diet. Assuming a family of 4.5 adult equivalent consumers, a model household would need 1,700kg of barley for basic subsistence each year. An LMIB population of 20-25,000 for Knossos will have consumed some 7,600-9,500 tons of barley a year, 21-26 tons per day. Assuming a donkey load of 90kg, this represents c. 85-106,000 donkey loads a year, or 230-290 donkey loads daily.

on rural farms or estates. Far more donkeys would be required if transport was seasonal, rather than continuous through most or all of the year. A human driver would be needed, though they could lead a train of several donkeys, and older children could be employed, distributing the labour through the family. Depending on how far the crop was brought in from, this will have tied-up labour for hours or even days for round-trips, and in aggregate was a significant labour requirement. While longer distance transport by sea might have been cost-effective, most of the productive land in the catchment of the palatial centres is not itself particularly accessible from the sea, so this is only likely to have been feasible if longer distance provisioning was required, for example if the Mirabello region was a source of supply for Knossos or Malia. For Malia, the probable port facility at Ayia Varvara was conveniently close to the city, but for Knossos, the port at Katsambas was 6 km distant, requiring overland transport to the city, adding both to costs and congestion.

This transport could have been concentrated seasonally, respecting the harvest (staggered somewhat for different crops, since they were not just consuming barley), or undertaken periodically, since the typical storage capacity in Neopalatial houses at Knossos and elsewhere could contain about 3-4 months worth of grain for a family (Figure 9; Christakis 2008: Table 10). On the other hand, seasonal scheduling would have exacerbated very considerably the logistical problems of transporting these quantities to Knossos. If organised individually by a consuming family, they would need a donkey load of barley every 19 days. With a more varied diet including vegetables and fruit in season (with only short shelf-lives), more frequent transport of some components of the diet would be required, and grain could consequently be brought in less frequently.

For Knossos, much of its productive catchment lies to the south, and the major route in the recent past, as today, was up the Kairatos valley, through the bottleneck narrowing at Spilia. This, and the ridges east and west of the city, will have focused transport on the south road, though other routes could have led south-west, along the slopes flanking the Vlychia. But if the main route came up the Kairatos valley, perhaps suggested by the road-side location of the high status Temple Tomb (Evans 1935: 203-4), there would have been a phenomenal volume of traffic, every day (c. one donkey load every 2-3 minutes), across the Vlychia ravine into the city. This may explain the exceptionally massive construction of the Viaduct (Evans 1928: 61, 93102; Figure 10A); the road it supported was an essential lifeline for the city. As Evans noted (Evans 1928: 99), the cyclopean bridge over the torrent bed immediately south of Mycenae looks similarly over-engineered, but

What would such transport entail? As noted, it would almost certainly be by donkey back, not by cart, so numerous donkeys need to be raised, stabled and fed. Few potential stabling facilities have been recognised in Minoan sites, so presumably the animals were maintained either outside the urban boundaries or 104

Todd Whitelaw: Feeding Knossos

Consumption, storage and provisioning

would also have been an important point of access for that city to its principal supply zone. Evans and Mackenzie also excavated ashlar blocks which they thought represented a Minoan bridge and traces of a roadbed, crossing the much slighter stream just beyond the north end of the city (Evans 1928: 154; Hood and Smyth 1981, near KS2.69), facilitating access to and from the north. Possible carbonised animal fodder bins were excavated at the so-called Caravanserai immediately south-east of the Viaduct, and were indeed a central element of the interpretation of this not obviously domestic complex (Evans 1928: 103-7), though Schofield has rightly expressed caution about this interpretation (1996).

Christakis’ systematic evaluation of the evidence for agricultural storage and inferences about storage strategies (2008; 2014) have put discussions of Minoan economic strategies on a far more solid footing, but also highlight two surprising patterns. The first is the very low agricultural storage capacity in most Neopalatial houses. The disjunction Christakis regularly notes between the architectural extent of storeroom space and that actually occupied by pithoi, might suggest the extensive use of organic containers, though he argues that such alternatives are unsuitable for medium-term food storage (Christakis 2008: 12). Potential changes in room use are also relevant, though the near universally low level of storage in the final occupation phase at most sites remains problematic. In some cases, through detailed consideration of excavation reports, Christakis can argue that these low storage volumes are not due to post-depositional disturbance or poor recovery, recording and publication, at least of ceramics. So the challenge is that in a context where one would expect over-storage due to productive uncertainty (Halstead 1990; Forbes 1989), most households did not even have the storage capacity for a year’s basic subsistence (Christakis 2008: 109-18, 131). To meet consumption needs, most households must have been periodically re-supplied, so only had to maintain limited subsistence storage in the house. Such re-supply may have been from rural estates linked to the city dwellers (unlikely to be relevant for the majority of households) or from individual family field stores (e.g. recent spitakia: e.g. Whitelaw 1991), at least possible for the urban families cultivating fields directly from the city, though this leaves questions about the security of stocks from natural threats (damp, mold, insects, animals), or theft. Together, these might account for up to half the city’s households, but not the remainder. Alternatively, households may have been supplied from central palatial stores (rations in the classic redistribution model) or through markets.

On the north side of the Vlychia, opposite the west end of the Viaduct, Evans traced the foundations of the monumental Stepped Portico, which he reconstructed as leading upslope from a bridgehead for the ravine crossing (Evans 1928: 141-55) to the south-west corner of the palace, and he assumed beyond to the city. If so, given the houses immediately south of the West Court, any such route would have led to the larger paved court further west, under the modern car parks, as revealed by test pits in the 1930s (Figure 10A; Hood and Smyth 1981: KS2.210; Warren 1994: 196-99). This large open area could have served as a market or distribution area, central to the city but readily accessible from its southern outskirts. The large court immediately east of the palace at Malia and just inside the MM east gate of the city (Figure 10B; Demargne and Gallet de Santerre 1953: 63-64, Plate 64; Deshayes and Dessenne 1959: 1-5, Plan 1.), could have served a similar function, in both cases near, but not as directly associated with the palaces as the formal and ceremonial West Courts. Also considering transport, vegetables will have been more easily damaged in transport and are considerably bulkier than threshed and dried cereals. Along with the intensive labour investment required for garden cultivation, these characteristics suggest that vegetables were likely to have been preferentially cultivated close to the city. While the present alluvium in the Kairatos valley is probably largely post-Bronze Age (Roberts 1979; Galanidou et al. 2014), the broad valley bottom adjacent to the city probably contained rich alluvium in earlier millennia as well, and this would have been a prime location for gardens, as it is today.

The redistribution model, inspired by the Linear B tablets, originally was thought to find corroboration in the extensive palace storerooms, particularly at Knossos (Renfrew 1972: 291-97). However calculations of the actual storage capacities of palatial storerooms reveal that even at Knossos, they could have held food for only 750-1000 individuals for a year (Christakis 2008: 120), in other words, possibly enough for the support of staff of the palace itself, court meals (e.g. Milano 1989), and perhaps the occasional large feast on the scale now documented by mainland Linear B texts (Bendall 2008). Of course such stores could have been refilled on a regular basis, though the question then is how they were re-stocked. The Linear B documents indicate one process, with contributions for feasts from individuals and communities as well as the palace (Killen 1994),

Given the exceptional size of its resident population, Knossos will have presented provisioning problems unparalleled by any other prehistoric Aegean community. Estimates of the quantity of food moved into the city, and consideration of the scale of the productive hinterland necessary to supply it, highlight the logistical problems which had to be overcome, practically and politically, in integrating the city with its countryside. 105

Country in the City

Figure 10: A. Knossos, Neopalatial site core; B. Malia, Protopalatial site core.

106

Todd Whitelaw: Feeding Knossos

but court meals and provision of rations to palace dependants would have required regular and substantial re-stocking, not just occasional contributions. Another possibility is frequent replenishment from extra-palatial storerooms; the Linear B tablets at Knossos and the earlier Linear A tablets from Ayia Triadha register very substantial quantities of grain, though not necessarily all ever stored together centrally. Certainly it would make sense for produce collected well away from the palace to be deployed locally as rations for permanent or corvée work groups, or to support palace-organised activities or ceremonies, all documented in the Pylos tablets, rather than be transported to the centre and out again. A potential concern is that we have found no such extra-palatial storage facilities on the scale that would be required, at palatial sites or elsewhere. Other than two recently suggested possible examples in the countryside (Buell 2014: 274-75; Christakis et al. 2015), recent analysis suggests that we have an idea of what these might look like at Ayia Triadha. In LHIIIA2, when the site was probably independent of Knossos, following the final destruction of the palace, the Mercato, adjacent storage cells in Edificio P and the rebuilt Bastione arguably represent extensive storage facilities, which Privitera has related to the massive quantity of grain recorded from Da-wo in the slightly earlier Knossos Linear B tablets (Privitera 2014: 438-45; Bennet 1985: 247). But shifting back in time, he has also related the storage potential of the Bastione in LMIB to the large quantity of grain documented in the Linear A tablets from the site (Palaima 1994; Privitera 2014: 43638).

2004; Killen 2008). The recent arguments tend to be framed in terms of all or nothing alternatives, though it has long been recognised that economic systems may incorporate different processes at different scales or in different contexts (Earle 2002; Renger 2007; Halstead 2011; Bennet and Halstead 2014). This debate was played-out several decades ago in Near Eastern studies, eventually recognising very variable balances of redistributive and market processes in different institutional and private contexts (Renger 1994; 2007). The problems with the recognition of extra-institutional exchange in the ancient Near East parallel those in the Aegean, with institutional texts almost exclusively recording the institution’s economic transactions, which are largely internal and redistributive, and the archaeological difficulty of recognising markets, even when we know historically they existed (Feinman 2013; Stark and Garraty 2010). With the texts largely informing us about redistributive behaviour, we need to work out how to make the archaeological evidence inform us about other economic transactions (Halstead 1992-93; 1999a; Voutsaki 2001; Whitelaw 2001b; Galaty 1999; Parkinson 1999; Pullen 2013; Shelmerdine 2013). But with palaces issuing rations to significant segments of their regional populations, redistributive behaviour is clearly a significant component of palatial economies. The challenge is recognising and understanding the contributions of various economic processes through different types of evidence, and calibrating their relative significance in particular contexts and situations (Bennet and Halstead 2014). We also need to beware of expecting all examples to fit one process or framework; the variability in storage provision documented by Christakis may indicate that different households obtained their subsistence (or a different balance of components of subsistence) through different channels. We can anticipate that the households of those regularly issued rations by the palace, documented in the Linear B tablets and probably the Linear A tablets from Ayia Triadha (Uchitel 2003), will have had different storage strategies from those of independent producers. In Near Eastern institutional rationing systems, the rations (often issued monthly) only accounted for a proportion of a family’s subsistence; they were often also given access to land to cultivate themselves to supplement the rations (Foster 1986; Renger 1995-96; van Koppen 2001), and the balance between such subsistence sources will have differed between institutions and among the families dependent on them. Looking at analogous systems can also remind us of the complexities and suggest potential alternative models. As noted above, the Ugarit texts document two distinct but parallel systems of royal and free farms (Liverani 1979; 1982; 1989), which will undoubtedly have required different physical facilities on the ground, reflecting differences in the resident population and the quantity and

At Knossos, Christakis has suggested that the Northeast House represents extra-palatial storage, and notes the possibility of additional storage in the predecessor of the Arsenal alongside the Royal Road (2008: 121), associated in its later phase with palatial storage through its Linear B tablets (Driessen 1995). I would add the Northwest Treasure House as another potential palatial storehouse, and wonder whether the enigmatic narrow, non-domestic structures excavated along the south side of the Royal Road by Hood and Warren, and identified by the latter as ‘grandstands’ (Warren 1994: 192), might represent storerooms like the Bastione. Together, these would make much of the northern surrounds of the palace and its ceremonial access along the Royal Road, a display of palatial wealth and power (Figure 10A), as at LMI Ayia Triadha, itself probably a Knossian administrative sub-centre (Figure 11). The alternative to provisioning households through palatial redistribution focuses on market exchange (Christakis, this volume; 2008: 139). Market models have recently been strongly advocated, coupled with critiques of the Finley/Renfrew redistributive model (Nakassis et al. 2011; Parkinson et al. 2013), already substantively criticised over the past 30 years (Halstead 107

Country in the City

Figure 11: Ayia Triadha, Neopalatial site core and estimated extent.

organisation of storage. Considering such alternatives may provide us with a wider range of potential interpretive models, and help us to recognise when archaeological differences we can document represent variation within one organisational structure, and when different organisational structures are implicated.

this exceptionally clear. But having demonstrated that only a very small number of villas had more storage capacity than that required by a family for a year (2008: 128-31), he still supports a variant of the traditional interpretation of their role as mobilising centres or management nodes in agricultural supply networks (Christakis 2008: 131, 134-37). Several examples with more agricultural storage also have spatial layouts which draw attention to their surplus storage, but even so, they only have the capacity to store annual subsistence resources for a small number of households (c. 1-7 families: Christakis 2008: Table 10). In such a case, these may be more or less self-sufficient agricultural estates (supporting a limited number of dependent personnel), but there is nothing to indicate clearly that they served as major production or collection points for shipping significant bulk agricultural products to the palaces or urban centres.

The second anomaly arising from Christakis’ analyses is the low storage capacity of most villas, traditionally assumed to be administratively linked to the major palaces as the residences or country estates of local elites, acting as local administrators and resource managers (Hood 1983; Hägg 1997; Betancourt and Marinatos 1997; Walberg 1995). A reassessment of the villas identified nearly all as buildings within small communities, rather than country estates (Hägg 1997: 229-31; Betancourt and Marinatos 1997), though isolated, massively constructed complexes have been identified by some surveys, and substantial farms seem to exist (Zielinski 1998). But in whatever guise, these appear not to be the major rural estates that have been assumed to be funneling agricultural resources to the palatial centres (Renfrew 1972: 258-59; Hood 1983; Hägg 1997). The inspiration for this model appears to be Roman latifundia (confused by labeling a wide variety of Roman rural establishments ‘villas’, and a very variable ‘villa economy’: Launaro 2015), but the Minoan examples are clearly not agricultural estates on such a scale; Christakis’ analysis of storage capacities makes

Looking at the larger structures from the perspective of this chapter, it is intriguing that the largest examples, which also usually have considerably greater storage capacities, are found both in palatial centres, and in much smaller communities. Some are certainly (e.g. the palaces at Gournia, Petras, and the Villa Reale at Ayia Triadha) and others possibly (Tylissos, Vathypetro, Archanes Tourkogeitonia, Nirou Chani) the central buildings of their communities. But others are probably just one among several comparable structures 108

Todd Whitelaw: Feeding Knossos

within larger centres (e.g. Protopalatial Quartier Mu and Neopalatial Quartier Epsilon at Malia and the Neopalatial Little Palace at Knossos). In making this comparison, it seems not unreasonable to anticipate that similarly sized and elaborated establishments might require comparable scales of material support to construct and maintain them and their occupants (cf. Davies 2004; Whitelaw 2017). From this perspective, it is intriguing that a single family or institution within Malia (Quartier Mu, the Magasins Dessenne or Quartier Epsilon), could have access to wealth and resources roughly comparable to the ruler of a local regional centre such as Gournia or Petras. This highlights that we should not expect directly comparable economic systems, or the settlement patterns which articulated them, in all regions of the island.

been asking of them. In parallel, questions relating to agricultural production, economic organisation and land-holding have been receiving more attention in Linear B studies, with analogical explorations in Linear A. But so far, these data sources have only been brought together in limited ways. The present chapter has tried to point to areas of tangency, where the analysis and interpretation of one source may help contextualise the other. This study has been driven from the bottom up, by the recognition through fieldwork of the anomalously large size of Neopalatial Knossos, and through considering the social and economic challenges and opportunities this presented to its occupants. This poses an interpretive challenge because we have generally not viewed agricultural provisioning as problematic, and we have not explored relevant models with the Aegean data. An extreme situation can point to problems we have not really addressed, while at the same time encourage us to reflect back on other less anomalous cases, which we have simply assumed were uninteresting or that we understood.

At present, Christakis’ systematic documentation of storage capacities, on the one hand urges us to expect that the structures traditionally classed together as villas, regardless of scale, elaboration or context, probably represent a variety of different phenomena (2008: 133; Betancourt and Marinatos 1997). On the other hand, I think it is clear that we cannot yet situate any of these structures clearly in any model for the agricultural provisioning of the urban centres. Looked at from the logistical considerations, if they were involved, they cannot be assumed to represent the major organisational nodes in the large-scale provisioning system sketched above. Ayia Triadha and Galatas might represent two different components in such networks, or more likely, in different types of networks; both seem more on the sort of scale that would be required than even the largest villas. Given the varied re-structurings we can start to detect through time in different local regions on the basis of survey evidence, it may be that different regions were organised in distinct ways, depending both on their pre-existing systems, and how they could be adapted to the changing economic demands of the unparalleled urban consuming centre which developed at Knossos.

The preliminary explorations in this chapter, painted with a very broad brush, have been intended simply to outline the nature and scale of some of the processes which must have been involved in provisioning Minoan cities: bringing the country into the city. Even rough quantification, to establish the scale of the processes involved, raises three linked sets of concerns. First, I think we can start to recognise just how much of a transformation marks the extended transition between what we have traditionally thought of and artificially polarised as Protopalatial and Neopalatial Crete (Whitelaw 2014), in reality a comparison of the rich datasets provided by two major destruction horizons at the end of MMIIB and LMIB. The Neopalatial period in particular seems to be characterised by continuous processes of transformation involving distinct changes at each major site, which can no longer easily be presented as an island-wide, synchronised picture (Figures 3-6).

Wider implications of changes in central Cretan agricultural landscapes

The second point is how central Knossos may be to many of these changes. Knossos has traditionally been assumed by most researchers to play a central role in cultural development on the island, but this has been more assumed than argued. Originally eschewing the Knossos-centric perspective, I have been brought back to it from a different angle, as the anomalous size and concomitant social diversity of Knossos has become apparent. But considering Knossos central to at least some of these developments does not actually explain its role and significance, we need to define how its development differed from that of other centres which through the Protopalatial period shared at least some aspects of a broadly comparable overall trajectory

Wide-ranging insights have been developed in the past two decades as archaeological and Linear B evidence have been brought together increasingly effectively, easing what for long appeared to be a pretty fundamental divide (Bennet 2014). Here, my intent has been to start to explore additional areas where the textual and material evidence might be mutually informative. In recent decades, on Crete as in the rest of the Aegean, intensive surface surveys have been producing increasingly detailed data, but its analysis has remained only basic. To a degree this has been limited by the quantity, nature and resolution of the data, but it also reflects the limited questions we have 109

Country in the City (Whitelaw 2012). We can then explore how this distinct development impacted on different nearby regions, different segments of society, and different facets of behaviour, in distinct ways. This will encourage us to think in much more local terms, of specific, contingent developments, and how these then interact in wider patterns. That has been the principal focus of this chapter, to start to explore such processes, buildingup from very practical and grounded perspectives, working out from Knossos itself, and exploring some of the ways it was tied into the wider landscape, and indeed expanded outwards through different types of linkages, as its population expanded.

where they extended, and what ideas and materials moved through them. But its excessive size will also have created new tensions, not encountered before or after in the prehistoric Aegean. Traditionally, we have looked for causes for the late LMIB political ‘collapse’ on Crete in terms of outside forces, required because we assume Minoan culture must have been prosperous and peaceful, long recognised as a romantic projection (Bintliff 1984). These outside forces might be a mainland Mycenaean conquest, or, once it was clear the chronology of the events would not allow a direct cause and effect relationship with the Theran eruption, delayed consequences (Driessen and Macdonald 1987; Knappett et al. 2011). But is it necessary to look for outside causes for the LMIB collapse? Was an exceptionally large Knossos-centred polity in central Crete sustainable? We know from the later Linear B tablets that a Knossosbased palatial elite was able to administer a polity encompassing the central and western two-thirds of the island in LMIIIA1-2 (Bennet 1985; 2011). On the other hand, that period of dominance was fairly shortlived, and it may have been difficult to integrate the attenuated island (Driessen 2001a).

A third set of concerns can only be hinted at here, and is based on the recognition that Knossos, so much larger than any other Aegean Bronze Age centre, created a new social environment, and will have generated novel social, economic and political dynamics. The logistical explorations above point to tremendous economic and social stresses as a consequence of its size. But that size also created novel opportunities in a highly competitive social environment, generated by the differentiated population of elites, administrators, service personnel and highly specialised crafters (cf. Erdkamp 2012; Morley 2008; Clark and Lepetit 1996). This social pressure cooker, not paralleled in any other Aegean community for nearly a millennium, will have been an incubator for cultural innovation and elaboration, as documented by other major urban centres through history (Bairoch 1988; Eisenstadt and Shachar 1987), and recognised among modern urban centres (Bettencourt et al. 2007; 2010). This dynamic social context will have driven what we tend to view as the high point of Aegean Bronze Age culture (Warren 2012), with its unparalleled diversification and elaboration of material culture.

Was a socially, economically and politically complex and geographically extended polity sustainable in LMIB? This trial exploration of the logistics involved, even just considering subsistence support, indicates that maintaining the high urban population of Knossos will have been complicated, and therefore potentially fragile or unstable. We have no evidence that there were significant technological innovations in agricultural production or intensification (though see Müller-Celka et al. 2014), or transport technology, which would have enabled more efficient and effective centralisation. With these processes held constant, the growing population of Knossos will have created increasing logistical stresses through the Neopalatial period. The changes in settlement pattern which have been documented in every major region surveyed may represent adaptations for more efficient agriculture, or tributary agricultural extraction, but each region seems to have developed in at least slightly different ways. This does not rule out changes linked to provisioning Knossos, which would have had to be worked out in different ways in each context, given different productive potentials, transport constraints determined by distance and topography, and pre-existing local social structures. To explore these possibilities, we will need to analyse existing survey data in far more detail than has been attempted to date (e.g. Spencer and Bevan 2018).

Even if during LMI Knossos politically dominated only the centre of the island, it appears to have had a much wider cultural impact, as one can argue from the architectural and artistic styles arguably emulated elsewhere, particularly clear for the Mirabello centres. In most cases, we do not have the chronological resolution to demonstrate that these features and practices actually originated at Knossos. It has also been impossible to demonstrate that these represent Knossian political dominance, rather than some form of cultural hegemony, or whether we are just seeing a classic urban/rural innovator/receptor dynamic (Whitelaw 2018). Recognising that the exceptional size of Knossos created a unique social environment fostering innovation, adds strength to the assumption that many of these traits are likely to have been developed at Knossos. Whether they were explicitly imposed on some members of other communities or were more actively or passively adopted may be less important than that they were, giving us some idea of the networks which acted to integrate communities,

Finally, and not independent of the logistics, we can anticipate that as the polity expanded, the bureaucracy to support it did as well, which will have put increasing pressure on the resources and procurement 110

Todd Whitelaw: Feeding Knossos

infrastructure (Tainter 1988). This rarely, if ever, ends well. Did this expanding administration require increasingly exploitative productive relations between the centre and the countryside (Christakis 2008 14446)?

Bairoch, P. 1988. Cities and Economic Development: From the Dawn of History to the Present. Chicago, University of Chicago Press. Bang, P. 2009. Labor: free and unfree. In A. Erskine (ed.) A Companion to Ancient History: 447-61. Chichester: Blackwell. Bekker-Nielsen, T. 1989. The Geography of Power. BAR International Series 477. Oxford: British Archaeological Reports. Bendall, L. 2007. Economics of Religion in the Mycenaean World. Resources Dedicated to Religion in the Mycenaean Palace Economy. Oxford University School of Archaeology. Monograph 67. Oxford, Oxford University School of Archaeology. Bendall, L. 2008. How much makes a feast? Amounts of banqueting foodstuffs in the Linear B records of Pylos. In A. Sacconi, M. Del Freo, L. Godart and M. Negri (eds) Colloquium Romanum: atti del XII colloquio internazionale di micenologia, Roma, 20 - 25 Febbraio 2006, Vol. 1: 77-101. Pasiphae 1. Roma, Fabrizio Serra. Bennet, J. 1985. The structure of the Linear B administration at Knossos. American Journal of Archaeology 89: 231-249. Bennet, J. 1988. ‘Outside in the distance’: problems in understanding the economic geography of Mycenaean palatial territories. In J.-P. Olivier and T. Palaima (eds) Texts, Tablets and Scribes. Studies in Mycenaean Epigraphy and Economy offered to Emmett L. Bennett Jr: 19-41. Suplementos a Minos 10. Salamanca, Ediciones Universidad de Salamanca. Bennet, J. 1990. Knossos in context: comparative perspectives on the Linear B administration of LM II-III Crete. American Journal of Archaeology 94: 193212. Bennet, J. 1992. ‘Collectors’ or ‘owners’? An examination of their possible functions within the palatial economy of LMIII Crete. In J.-P. Olivier (ed.), Mykenaïka. Actes du IXe Colloque international sur les textes mycéniens et égéens, Centre de l’Antiquité Grecque et Romaine de la Fondation Hellénique des Recherches Scientifiques et École française d’Athènes: 65-101. Bulletin de Correspondance Hellénique, Supplement 25. Paris, Diffusion de Boccard. Bennet, J. 1999. The Mycenaean conceptualization of space or Pylian geography (...yet again!). In S. DegerJalkotzy, S. Hiller and O. Panagl (eds), Floreant Studia Mycenaea. Akten des X. Internationalen Mykenologischen Colloquiums in Salzburg vom 1.-5. Mai 1995, Band I: 13157. Wien, Verlag der Österreichischen Akademie der Wissenschaften. Bennet, J. 2008. PalaceTM: speculations on palatial production in Mycenaean Greece with (some) reference to glass. In C. Jackson and E. Wager (eds), Vitreous Materials in the Late Bronze Age Aegean: 15172. Sheffield Studies in Aegean Archaeology. Oxford, Oxbow. Bennet, J. 2011. The geography of the Mycenaean kingdoms. In Y. Duhoux and A. Morpurgo Davies

From this perspective, the changing relations between the city and the countryside were essential for sustaining Minoan cities, and particularly Neopalatial Knossos. I have suggested that these would have varied between communities with different needs, and in some cases, have been far more complicated, and potentially problematic, than we have to date assumed. If so, in developing a distinctive relationship with its sustaining hinterland, Neopalatial Knossos was a unique experiment in the prehistoric Aegean, but possibly an unsuccessful one. Acknowledgements Andrew Bevan, Kostis Christakis and Paul Halstead kindly read a draft, and I thank Paul for his encouragement. I am also grateful to Andrew Bevan for producing the relief base map of Crete and the site catchments. Bibliography Adams, R. 1965. Land Behind Baghdad: a History of Settlement on the Diyala Plains. Chicago, Chicago University Press. Adams, R. 1981. Heartland of Cities: Surveys of Ancient Settlement and Land Use on the Central Floodplain of the Euphrates. Chicago, University of Chicago Press. Adams, R. and Nissen, H. 1972. The Uruk Countryside: the Natural Setting of Urban Societies. Chicago, University of Chicago Press. Allbaugh, L. 1953. Crete: A Case Study of an Underdeveloped Area. Princeton, Princeton University Press. Alston, R. 2002. The City in Roman and Byzantine Egypt. London, Routledge. Alston, R. and O. van Nijf (eds). 2008. Feeding the Ancient Greek City. Leuven, Peeters. Amato, V., Longo, F., Bredaki, M., Rossi, A., Ghilardi, A., Psomiadis, D., Colleu, M., Sinibaldi, L., Delanghe-Sabatier, D., Demory, F. and Petit, C. 2014. Geoarchaeological and palaeoenvironmental researches in the area of ancient Phaistos (Crete, Greece): preliminary results. In G. Touchais, R. Laffineur and F. Rougemont (eds) Physis. L’environment naturel et la relation homme-milieu dans le monde égéen protohistorique: 129-40. Aegaeum 37. Liège, Université de Liège. Aurenche, O. 1981. Essai de démographie archéologique. L’exemple des villages du Proche-Orient ancien. Paléorient 7: 93-105.

111

Country in the City (eds), A Companion to Linear B: Mycenaean Greek Texts and their World, Vol. 2: 137-68. Bibliothèque des Cahiers de l’Institut de Linguistique de Louvain 127. Leuven, Peeters. Bennet, J. 2014. ‘Literacies’ - 60+ years of ‘reading’ the Aegean Late Bronze Age. Bulletin of the Institute of Classical Studies 57.2:127-137. Bennet, J. and P. Halstead. 2014. O-no! Writing and righting redistribution. In D. Nakassis, J. Gulizio, and S. James (eds), KE-RA-ME-JA: Studies Presented to Cynthia W. Shelmerdine: 271-282. Philadelphia, INSTAP Academic Press. Betancourt, P. 1985. The History of Minoan Pottery. Princeton, Princeton University Press. Betancourt, P. 1995. Pseira, Crete: the economic base for a Bronze Age town. In R. Laffineur and W.-D. Niemeier (eds), Politeia: Society and State in the Aegean Bronze Age. Proceedings of the 5th International Aegean Conference, University of Heidelberg, Archäologisches Institut, 10-13 April 1994, Vol. 1: 163-67. Aegaeum 12. Liège, Université de Liège. Betancourt, P. 2004. Coastal trade: the eastern Gulf of Mirabello in the early twentieth century. In P. Betancourt, C. Davaras and R. Hope Simpson (eds), Pseira VIII: The Archaeological Survey of Pseira Island. Part 1:91-7. Philadelphia, INSTAP Academic Press, Prehistory Monographs 11. Betancourt, P., Davaras, C. and R. Hope Simpson (eds) 2005. Pseira IX: The Archaeological Survey of Pseira Island. Part 2: The Intensive Surface Survey. Philadelphia, INSTAP Academic Press, Prehistory Monographs 12. Betancourt, P. and Marinatos, N. 1997. The Minoan villa. Part I: the definition of the Minoan villa. In R. Hägg (ed.), The Function of the ‘Minoan Villa’: Proceedings of the Eighth International Symposium at the Swedish Institute at Athens, 6-8 June 1992: 91-92. Acta Athenensia, Quarto Series 46. Stockholm, Paul Åströms Förlag. Bettencourt, L., Lobo, J., Helbing, D., Kühnert, C. and West, G. 2007. Growth, innovation, scaling, and the pace of life in cities. Proceedings of the National Academy of Sciences 104.17: 7301-6. Bettencourt, L., Lobo, J., Strumsky, D. and West, G. 2010. Urban scaling and its deviations: revealing the structure of wealth, innovation and crime across cities. Plos One 5.11: e13541. Bevan, A. 2010. Political geography and palatial Crete. Journal of Mediterranean Archaeology 23: 27-54. Bevan, A. and Wilson, A. 2013. Models of settlement hierarchy based on partial evidence. Journal of Archaeological Science 40.5: 2415-27. Bintliff, J. 1984. Structuralism and myth in Minoan studies. Antiquity 58: 33-8. Bintliff, J. 1999. Settlement and territory. In G. Barker (ed.), Companion Encyclopedia of Archaeology: 505-45 London, Routledge. Bintliff, J. 2002. Going to market in antiquity. In E. Olshausen and H. Sonnabend (eds), Zu Wasser und

zu Land: 209-50. Stuttgarter Kolloquium 7. Stuttgart, Steiner. Bintliff, J. 2006. City-country relationships in the ‘normal polis’. In R. Rosen and I. Slutter (eds), City, Countryside and the Spatial Organization of Value in Classical Antiquity: 13-32. Leiden, Brill. Bintliff, J., Howard, P. and Snodgrass, A. 2007. Testing the Hinterland: the Work of the Boeotia Survey (19891991) in the Southern Approaches to the City of Thespiai. McDonald Institute Monographs. Cambridge, McDonald Institute for Archaeological Research. Blitzer, H. 2004. Agriculture and subsistence in the late Ottoman and post-Ottoman Mesara. In V. Watrous, D. Hadzi-Vallianou and H. Blitzer (eds), The Plain of Phaistos. Cycles of Social Complexity in the Mesara Region of Crete: 111-217. Monumenta Archaeologica 23. Los Angeles, University of Los Angeles. Bodley, J. 1994. Cultural Anthropology: Tribes, States, and the Global System. London, Mayfield Publishing. Bosanquet, R. and Dawkins, R. 1923. The Unpublished Objects from the Palaikastro Excavations 1902-1906. Part 1. British School at Athens, Supplementary Volume 1. London, Macmillan. Bowman, A. and Wilson, A. 2011. Settlement, Urbanization and Population. Oxford, Oxford University Press. Branigan, K. 1972. Minoan settlements in east Crete. In P. Ucko, R. Tringham and G. Dimbleby (eds), Man, Settlement and Urbanism: 751-59. London, Duckworth. Brodie, N. 2008. The donkey: an appropriate technology for Early Bronze Age land transport and traction. In N. Brodie, J. Doole, G. Gavalas and C. Renfrew (eds), Horizon. Ορίζων: A Colloquium on the Prehistory of the Cyclades: 299-304. McDonald Institute Monographs. Cambridge, McDonald Institute for Archaeological Research. Broodbank, C. 2013. The Making of the Middle Sea: A History of the Mediterranean from the Beginning to the Emergence of the Classical World. London, Thames and Hudson. Buell, M. 2014. The rise of a Minoan city and the (re) structuring of its hinterlands: a view from Galatas. In A. Creekmore III and K. Fisher (eds), Making Ancient Cities: Space and Place in Early Urban Societies: 257-91. Cambridge, Cambridge University Press. Carlstein, T. 1982. Time Resources, Society and Ecology: On the Capacity for Human Interaction in Space and Time. London, Allen and Unwin. Carreras Montfort, C. 1995-96. A new perspective for the demographic study of Roman Spain. Revista de Historia de Arte e Arquelogia 2: 59-82. Casana, J. 2009. Alalakh and the archaeological landscape of Mukish: the political geography and population of a Late Bronze Age kingdom. Bulletin of the American Schools of Oriental Research 353: 7-37. Casana, J. 2013a. Radial route systems and agro-pastoral strategies in the fertile crescent: new discoveries from western Syria and southwestern Iran. Journal of Anthropological Archaeology 32: 257-73. 112

Todd Whitelaw: Feeding Knossos

Casana, J. 2013b. Settlement, territory, and the political landscape of Late Bronze Age polities in the Northern Levant. Archeological Papers of the American Anthropological Association 22.1: 107-25. Cavanagh, W., Crouwel, J., Catling, R. and Shipley, G. 2002. The Laconia Survey: Continuity and Change in a Greek Rural Landscape. Vol. 2: Archaeological Data. British School at Athens, Supplementary Volume 26. London, British School at Athens. Chadwick, J. 1976. The Mycenaean World. Cambridge, Cambridge University Press. Charlton, T. and Nichols, D. (eds). 1997. The city-state concept. Development and applications. In D. Nichols and T. Charlton (eds), The Archaeology of Citystates. Cross-cultural Approaches: 1-14. Washington D.C., Smithsonian Institution Press. Chisholm, M. 1968. Rural Settlement and Land Use. 2nd ed. London, Hutchinson. Christakis, K. 2008. The Politics of Storage: Storage and Sociopolitical Complexity in Neopalatial Crete. Prehistory Monographs 25. Philadelphia, INSTAP Academic Press. Christakis, K. 2014. Communal storage in Bronze Age Crete: re-assessing testimonies. Kritika Chronkia 34: 201-18. Christakis, K, Mavraki-Balanou, M. and Kastanakis, A. 2015. Economic and political complexity in the Aposelemis valley. A preliminary assessment. In P. Karanastas, A. Tzigounaki and C. Tzigounaki (eds), Archaiologiko Ergo Kritis 3: 293-307. Rethymnon, School of Philosophy, University of Crete. Clark, C. 1958. Transport: maker and breaker of cities. The Town Planning Review 28: 237-50. Clark, P. and Lepetit, B. 1996. Introduction. In P. Clark and B. Lepetit (eds), Capital Cities and Their Hinterlands in Early Modern Europe: 1-25. Aldershot, Scolar Press. Cline, E. and Harris-Cline, D. (eds). 1998. The Aegean and the Orient in the Second Millennium: Proceedings of the 50th Anniversary Symposium, Cincinnati, 18-20 April 1997. Aegaeum 18. Liège, Université de Liège. Cosmopoulos, M. 2001. The Rural History of Ancient Greek City-States: The Oropos Survey Project. British Archaeological Reports, International Series 1001. Oxford, Archaeopress. Crouwel, J. 1981. Chariots and Other Means of Land Transport in Bronze Age Greece. Allard Pierson Series 3. Amsterdam, Allard Pierson Museum. Davies, J. 2004. The economic consequences of Hellenistic palaces. In Z. Archibald, J. Davies and V. Gabrielsen (eds), Making, Moving and Managing. The New World of Ancient Economies, 323-31 BC: 117-35. Oxford, Oxbow Books. Day, P. 1997. Ceramic exchange between towns and outlying settlements in Neopalatial East Crete. In R. Hägg (ed.), The Function of the ‘Minoan Villa’: Proceedings of the Eighth International Symposium at the Swedish Institute at Athens, 6-8 June 1992: 219-228.

Acta Athenensia, Quarto Series 46. Stockholm, Paul Åströms Förlag. Day, P. and Relaki, M. 2002. Past factions and present fictions: palaces in the study of Minoan Crete. In J. Driessen, I. Schoep and R. Laffineur (eds), Monuments of Minos. Rethinking the Minoan Palaces. Proceedings of the International Workshop ‘Crete of the hundred Palaces?’ Held at the Université Catholique de Louvain, Louvain-la-Neuve, 14-15 December 2001: 21734. Aegaeum 23. Liège, Université de Liège. Demargne, P., and Gallet de Santerre, H. 1953. Fouilles exécutées a Mallia: exploration des maisons et quartiers d’habitation (1921-1948). Etudes Crétoises 9. Paris, Paul Geuthner. Deshayes, J. and Dessenne, A. 1959. Fouilles exécutées a Mallia: exploration des maisons et quartiers d’habitation (1948-1954). Etudes Crétoises 11. Paris, Paul Geuthner. Diakonoff, I. 1969. The rise of the despotic state in ancient Mesopotamia. In I. Diakonoff (ed.), Ancient Mesopotamia. Socio-Economic History. A Collection of Studies by Soviet Scholars: 173-203. Moscow, Nauka. Diakonoff, I. 1982. The structure of Near Eastern society before the middle of the 2nd millennium BC. Oikumene 3: 7-100. Drennan, R. 1984. Long‐distance transport costs in pre‐ Hispanic Mesoamerica. American Anthropologist 86: 105-12. Driessen, J. 1995. The arsenal of Knossos (Crete) and Mycenaean chariot forces. In M. Lodewijckx (ed.), Archaeological and Historical Aspects of West-European Societies. Album Amicorum André Van Doorselaer: 48198. Acta Archaeologica Lovaniensia Monographiae 8. Leuven, Leuven University Press. Driessen, J. 2000. The Scribes of the Room of the Chariot Tablets at Knossos: Interdisciplinary Approach to the Study of a Linear B Deposit. Suplementos a Minos 15. Salamanca, Ediciones Universidad de Salamanca. Driessen, J. 2001a. Centre and periphery: some observations on the administration of the kingdom of Knossos. In S. Voutsaki and J. Killen (eds), Economy and Politics in the Mycenaean Palace States. Proceedings of a Conference Held on 1-3 July 1999 in the Faculty of Classics, Cambridge: 96-112. Cambridge Philological Society, Supplementary Volume 27. Cambridge, Cambridge Philological Society. Driessen, J. 2001b. History and hierarchy. Preliminary observations on the settlement pattern in Minoan Crete. In K. Branigan (ed.), Urbanism in the Aegean Bronze Age: 51-71. Sheffield Studies in Aegean Archaeology. London, Sheffield Academic Press. Driessen, J. 2010. Malia. In E. Cline (ed.), The Oxford Handbook of the Bronze Age Aegean (ca. 3000-1000 BC): 556-70. Oxford, Oxford University Press. Driessen, J. and Macdonald, C. 1997. The Troubled Island: Minoan Crete Before and After the Santorini Eruption. Aegaeum 17. Liège, Université de Liège. Earle, T. 2002. Bronze Age Economics: the Beginnings of Political Economies. Oxford, Westview Press. 113

Country in the City Eisenstadt, S. and Shachar, A. 1987. Society, Culture and Urbanization. London, Sage. Erdkamp, P. 2001. Beyond the limits of the ‘consumer city’. A model of the urban and rural economy in the Roman world. Historia: Zeitschrift für Alte Geschichte 50: 332-56. Erdkamp, P. 2012. Urbanism. In W. Scheidel (ed.), The Cambridge Companion to the Roman Economy: 241-65. Cambridge, Cambridge University Press. Erdkamp, P. 2015. Agriculture, division of labour and the paths to economic growth. In P. Erdkamp, K. Verboven and A. Zuiderhoek (eds), Ownership and Exploitation of Land and Natural Resources in the Roman World: 18-41. Oxford, Oxford University Press. Evans, A. 1928. The Palace of Minos: a Comparative Account of the Successive Stages of the Early Cretan Civilization as Illustrated by the Discoveries at Knossos. Vol. 2. London, Macmillan. Evans, A. 1935. The Palace of Minos: a Comparative Account of the Successive Stages of the Early Cretan Civilization as Illustrated by the Discoveries at Knossos. Vol. 4. London, Macmillan. Feinman, G. 2013. Re-envisioning ancient economies: beyond typological constructs. American Journal of Archaeology 117: 453-59. Fernandes, R., Geevens, G., Soetens, S. and KlontzaJaklova, V. 2011. Deletion/Substitution/Addition (DSA) model selection algorithm applied to the study of archaeological settlement patterning. Journal of Archaeological Science 38.9: 2293-300. Fernandes, R., Geeven, G., Soetens, S. and KlontzaJaklova, V. 2012. Mapping the probability of settlement location for the Malia-Lasithi region (Crete, Greece) during the Minoan Protopalatial period. In S. Kluiving and E. Guttmann-Bond (eds), Landscape Archaeology Between Art and Science: From a Multi- to an Interdisciplinary Approach: 35368. Landscape and Heritage Series. Amsterdam, Amsterdam University Press. Fincke, J. 2000. Transport of agricultural produce in Arraphe. In R. Jas (ed.), Rainfall and Agriculture in Northern Mesopotamia. Proceedings of the Third MOS Symposium: 147-70. Istanbul, Nederlands HistorischArchaeologisch Institut. Finley, M. 1957. The Mycenaean tablets and economic history. Economic History Review 10: 128-41. Finley, M. 1973. The Ancient Economy. London, Chatto and Windus. Firth 1994-95. Estimating the population of Crete during LM IIIA/B. Minos 29-30: 33-55. Forbes, H. 1989. Of grandfathers and grand theories: the hierarchised ordering of responses to hazard in a Greek rural community. In P. Halstead and J. O’Shea (eds), Bad Year Economics: Cultural Responses to Risk and Uncertainty: 87-97. Cambridge, Cambridge University Press.

Forbes, H. 2013. Off-site scatters and the manuring hypothesis in Greek survey archaeology: an ethnographic approach. Hesperia 82: 551-94. Forsén, J, and Forsén, B. (eds). 2003. The Asea Valley Survey: An Arcadian Mountain Valley from the Palaeolithic Period until Modern Times. Acta Athenensia, Quarto Series 51. Stockholm, Swedish Institute in Athens. Foster, B. 1986. Agriculture and accountability in ancient Mesopotamia. In H. Weiss (ed.), The Origins of Cities in Dry-farming Syria and Mesopotamia in the Third Millennium BC: 109-28. Guildford, Four Quarters Publishing. Foster, B. 2005. Shuruppak and the Sumerian city state. In L. Kogan, N. Koslova, S. Loesov and S. Tishchenko (eds), Memoriae Igor M. Diakonoff: 71-88. Babel un Bibel 2. Annual of Ancient Near Eastern, Old Testament, and Semitic Studies. Winona Lake, Eisenbrauns. Foster, K. 1979. Aegean Faience in the Bronze Age. New Haven, Yale University Press. Galanidou, N., Gaki-Papanastassiou, K., Karymbalis, E., Maroukian, H., Koskeridou, E. and Giangas, C. 2014. The palaeogeographic evolution of the Kairatos drainage basin and its coastal plain during the Holocene. Kritika Chronkia 34: 97-122. Galaty, M. 1999. Wealth ceramics, staple ceramics: pots and the Mycenaean palaces. In M. Galaty and W. Parkinson (eds), Rethinking Mycenaean Palaces: New Interpretations of an Old Idea: 49-59. Monograph 41. Los Angeles, The Cotsen Institute of Archaeology. Gale, N. (ed.) 1991. Bronze Age Trade in the Mediterranean. Studies in Mediterranean Archaeology 90. Göteborg, Paul Åströms Förlag. Garner, B. 1967. Models of urban geography and settlement location. In R. Chorley and P. Haggett (eds), Socio-economic Models in Geography: 303-60. London, Methuen. Gibson, M. 1972. The City and Area of Kish. Miami, Field Research Projects. Girella, L. 2010. Depositi ceramici del Medio Minoico III da Festòs e Haghia Triada. Studi di Archeologia Cretese 8. Padova, Bottega d’Erasmo. Gökçek, L. 2006. The use of wagons (eriqqum) in ancient Anatolia according to texts from Kültepe. Zeitschrift für Assyriologie 96: 185-99. Hägg, R. (ed.) 1997. The Function of the ‘Minoan Villa’: Proceedings of the Eighth International Symposium at the Swedish Institute at Athens, 6-8 June 1992. Acta Athenensia, Quarto Series 46. Stockholm, Paul Åströms Förlag. Haggett, P. 1965. Locational Analysis in Human Geography. London, Edward Arnold. Haggis, D. 2005. Kavousi. The Results of the Excavations at Kavousi in Eastern Crete I. The Archaeological Survey of the Kavousi Region. Prehistory Monographs 16. Philadelphia, INSTAP Academic Press. Halstead, P. 1990. Waste not, want not: traditional responses to crop failure in Greece. Rural History 1: 147-64. 114

Todd Whitelaw: Feeding Knossos

Halstead, P. 1992-1993. The Mycenaean palatial economy: making the most of the gaps in the evidence. Proceedings of the Cambridge Philological Society 38: 57-86. Halstead, P. 1998-1999. Texts, bones and herders: approaches to animal husbandry in Late Bronze Age Greece. In J. Bennet and J. Driessen (eds), A-na-qo-ta. Studies Presented to J. T. Killen. Minos 33-34: 149-89. Halstead, P. 1999a. Towards a model of Mycenaean palatial mobilization. In M. Galaty and W. Parkinson (eds), Rethinking Mycenaean Palaces: New Interpretations of an Old Idea: 35-41. Monograph 41. Los Angeles, The Cotsen Institute of Archaeology. Halstead, P. 1999b. Surplus and share-croppers: the grain production strategies of Mycenaean palaces. In P. Betancourt, V. Karageorghis, R. Laffineur, and W.-D. Niemeier (eds), Meletemata: Studies in Aegean Archaeology Presented to Malcolm H. Wiener as He Enters His 65th Year. Vol. 2: 319-26. Liège, Université de Liège. Halstead, P. 2004. Life after Mediterranean polyculture: the subsistence subsystem and the Emergence of Civilisation revisited. In J. Barrett and P. Halstead (eds), The Emergence of Civilisation Revisited: 189-206. Sheffield Studies in Aegean Archaeology 5. Oxford, Oxbow. Halstead, P. 2011. Redistribution in Aegean palatial societies: terminology, scale and significance. American Journal of Archaeology 115: 229-35. Halstead, P. 2014. Two Oxen Ahead: Pre-Mechanized Farming in the Mediterranean. Chichester, Wiley Blackwell. Hamilakis, Y. 2002. Too many chiefs?: Factional competition in Neopalatial Crete. In J. Driessen, I. Schoep and R. Laffineur (eds), Monuments of Minos. Rethinking the Minoan Palaces. Proceedings of the International Workshop ‘Crete of the hundred Palaces?’ Held at the Université Catholique de Louvain, Louvainla-Neuve, 14-15 December 2001: 179-99. Aegaeum 23. Liège, Université de Liège. Hansen, M. 1997. The polis as an urban centre. The literary and epigraphical evidence. In M. Hansen (ed.), The Polis as an Urban Centre and as a Political Community: Symposium August, 29-31 1996: 9-86. Acts of the Copenhagen Polis Centre 4. Copenhagen, Kongelige Danske Videnskabernes Selskab. Hansen, M. 2004. The concept of the consumption city applied to the Greek polis. In T. Nielsen (ed.) Once again: Studies on the Ancient Greek Polis: 9-48. Papers from the Copenhagen Polis Centre 7. Stuttgart, Franz Steiner. Hansen, M. (ed.) 2000. A Comparative Study of Thirty City-State Cultures. An Investigation Conducted by the Copenhagen Polis Centre. Historisk-filosofiske Skrifter 21. Copenhagen, Det Kongelige Danske Videnskabernes Selskab. Hansen, M. and T. Nielsen (eds) 2004. An Inventory of Archaic and Classical Poleis : an Investigation Conducted by The Copenhagen Polis Centre for the Danish National Research Foundation. Oxford, Oxford University Press.

Hanson, J. 2011. The urban system of Roman Asia Minor and wider urban connectivity. In A. Bowman and A. Wilson (eds) Settlement, Urbanization and Population: 229-75. Oxford, Oxford University Press. Hart, G. 1965. The grouping of place-names in the Knossos tablets. Mnemosyne 18: 1-28. Hayden, B. 2004. Reports on the Vrokastro Area, Eastern Crete, Vol. 2. The Settlement History of the Vrokastro Area and Related Studies. University of Pennsylvania Museum of Archaeology and Anthropology, University Museum Monograph 119. Philadelphia, University of Pennsylvania. Hayden, B., Bassiakos, Y., Kalpaxis, T., Sarris, A., and Tsipopoulou, M. 2007. A new exploration of Priniatikos Pyrgos: primary harbor settlement and emporium of the Vrokastro survey region. In P. Betancourt, M. Nelson and H. Williams (eds), Krinoi kai Limenes. Studies in Honor of Joseph and Maria Shaw: 93-100. Prehistory Monographs 22. Philadelphia, INSTAP Academic Press. Heltzer, M. 1976. The Rural Community in Ancient Ugarit. Wiesbaden, Reichart Verlag. Heltzer, M. 1979. Royal economy in ancient Ugarit. In E. Lipinski (ed.) State and Temple Economy Vol. 2: 459-96. Orientalia Lovaniensia Analecta. Leuven, Departement Orientalistiëk. Heltzer, M. 1982. The Internal Organization of the Kingdom of Ugarit. Wiesbaden, Reichert. Heltzer, M. 1987. Labour in Ugarit. In M. Powell (ed.), Labour in the Ancient Near East: 237-50. New Haven, American Oriental Society. Hiller, S. 1988. Dependent personnel in Mycenaean texts. In M. Heltzer and E. Lipinski (eds), Society and Economy in the Eastern Mediterranean (c. 1500-1000 B.C.). Proceedings of the International Symposium Held at the University of Haifa from the 28th of April to the 2nd of May 1985: 53-68. Orientalia Lovaniensia Analecta. Leuven, Peeters. Hodder, I. and C. Orton. 1976. Spatial Analysis in Archaeology. Cambridge, Cambridge University Press. Hood, S. 1958. Archaeological Survey of the Knossos Area. British School at Athens, Supplementary Volume. London, British School at Athens. Hood, S. 1983. The ‘country house’ and Minoan society. In O. Krzyszkowska and L. Nixon (eds), Minoan Society. Proceedings of the Cambridge Colloquium 1981: 129-35. Bristol, Bristol Classical Press. Hood, S., and D. Smyth. 1981. Archaeological Survey of the Knossos Area, 2nd ed., revised and expanded. British School at Athens, Supplementary Volume 14. London, Thames and Hudson. Hope Simpson, R., P. Betancourt, P. Callaghan, D. Harlan, J. Hayes, J. Shaw, M. Shaw and V. Watrous. 1995. The Archaeological Survey of the Kommos Area. In J. Shaw and M. Shaw (eds), Kommos 1: An Excavation on the South Coast of Crete. The Kommos Region and Houses of the Minoan Town. Part I: The Kommos Region, Ecology, 115

Country in the City and Minoan Industries: 325-402. Princeton, Princeton University Press. Hudson, M. and Levine, B. (eds) 1996. Privatization in the Ancient Near East and Classical World: a Colloquium Held at New York University, November 17-18, 1994. Cambridge, Peabody Museum of Archaeology and Anthropology. Isaakidou, V. 2008. ‘The fauna and economy of Neolithic Knossos’ revisited. In V. Isaakidou and P. Tomkins (eds), Escaping the Labyrinth: The Cretan Neolithic in Context: 90-114. Sheffield Studies in Aegean Archaeology 8. Oxford, Oxbow Books. Jameson, M., Runnels, C. and Van Andel, T. 1994. A Greek Countryside: The Southern Argolid from Prehistory to the Present Day. Stanford, Stanford University Press. Killen, J. 1985. The Linear B tablets and the Mycenaean economy. In A. Morpurgo Davies and Y. Duhoux (eds), Linear B: A 1984 Survey. Mycenaean Colloquium of the VIII Congress of the International Federation of the Societies of Classical Studies: 241-305. Bibliothèque des Cahiers de l’Institut de Linguistique de Louvain 26. Louvain-la-Neuve, Cabay. Killen, J. 1987. Piety begins at home: place-names on Knossos records of religious offerings. In P. Ilievski and L. Crepajac (eds), Tractata Mycenaea. Proceedings of the Eighth International Colloquium on Mycenaean Studies, Held in Ohrid (15-20 September 1985): 163-178. Skopje, Macedonian Academy of Sciences and Arts. Killen, J. 1993. The Oxen’s names on the Knossos Ch tablets. Minos 27-28: 101-7. Killen, J. 1994. Thebes sealings, Knossos tablets, and Mycenaean state banquets. Bulletin of the Institute of Classical Studies 39: 67-84. Killen, J. 1998. The rôle of the state in wheat and olive production in Mycenaean Crete. Aevum 72: 19-23. Killen, J. 2008. Mycenaean economy. In Y. Duhoux and A. Morpurgo Davies (eds), A Companion to Linear B: Mycenaean Greek Texts and their World, Vol. 1: 159-200. Bibliothèque des cahiers de l’Institut de Linguistique de Louvain 120. Leuven, Peeters. Knappett, C. 1999. Assessing a polity in Protopalatial Crete: The Malia-Lasithi state. American Journal of Archaeology 103: 615-39. Knappett, C., Rivers, R. and Evans, T. 2011. The Theran eruption and Minoan palatial collapse: new interpretations gained from modelling the maritime network. Antiquity 85: 1008-23. La Rosa, V. 2010a. Phaistos. In E. Cline (ed.), The Oxford Handbook of the Bronze Age Aegean (ca. 3000-1000 BC): 582-95. Oxford, Oxford University Press. La Rosa, V. 2010b. Ayia Triada. In E. Cline (ed.), The Oxford Handbook of the Bronze Age Aegean (ca. 3000-1000 BC): 495-508. Oxford, Oxford University Press. Landenius Enegren, H. 2008. The People of Knossos: Prosopographical studies in the Knossos Linear B Archives. Boreas. Uppsala Studies in Ancient Mediterranean and Near Eastern Civilizations 30. Uppsala, Uppsala University Library.

Larsen, M. 1976. The Old Assyrian City-state and its Colonies. Mesopotamia 4. Copenhagen, Akademisk Forlag. Launaro, A. 2015. The nature of the villa economy. In P. Erdkamp, K. Verboven and A. Zuiderhoek (eds), Ownership and Exploitation of Land and Natural Resources in the Roman World: 174-87. Oxford, Oxford University Press. Legarra Herrero, B. 2011. New kid on the block: the nature of the first systemic contacts between Crete and the Eastern Mediterranean around 2000 BC. In T. Wilkinson, S. Sherratt and J. Bennet (eds), Interweaving Worlds: Systemic Interactions in Eurasia, 7th to 1st Millennia BC: 266-81. Oxford, Oxbow Books. Liverani, M. 1979. Ras Shamra. II. Histoire. In H. Cazelles and A. Feuillet (eds), Supplément au Dictionnaire del la Bible: 1295-348. Fascicule 53. Paris. Liverani, M. 1982. Ville et campagne dans le royaume d’Ugarit. Essai d’analyse économique. In M. Dandamayev, I. Gershevitch, H. Klengel, G. Komoroczy, M. Larsen and N. Postgate (eds), Societies and Languages of the Ancient Near East. Studies in Honour of I.M. Diakonoff: 250-58. Warminster, Aris and Phillips. Liverani, M. 1989. Economy of Ugaritic royal farms. In C. Zaccagnini (ed.), Production and Consumption in the Ancient Near East: 127-68. Budapest, University of Budapest. Liverani, M. 2005. Historical Overview. In D. Snell (ed.), A Companion to the Ancient Near East: 3-19. Oxford, Blackwell. Lupack, S. 2008. The Role of the Religious Sector in the Economy of Late Bronze Age Mycenaean Greece. British Archaeological Reports, International Series 1858. Oxford, Archaeopress. Marcus, J. 1998. The peaks and valleys of ancient states: an extension of the dynamic model. In G. Feinman and J. Marcus (eds), Archaic States: 59-94. Santa Fe, School of American Research. Mattingly, D. 1997. Beyond belief? Drawing a line beneath the consumer city. In H. Parkins (ed.), Roman Urbanism: Beyond the Consumer City: 210-18. London, Routledge. McEnroe, J. 1982. A typology of Minoan Neopalatial houses. American Journal of Archaeology 86: 3-19. McEnroe, J. 2001. Pseira V: The Architecture of Pseira. Philadelphia, University of Pennsylvania Museum of Archaeology and Anthropology. McEnroe, J. 2010. Architecture of Minoan Crete: Constructing Identity in the Aegean Bronze Age. Austin, University of Texas Press. McGeough, K. 2007. Exchange Relationships at Ugarit. Ancient Near Eastern Studies, Supplement 26. Leuven, Peeters. McGrew, W. 1985. Land and Revolution in Modern Greece, 1800-1881: the Transition in the Tenure and Exploitation of Land from Ottoman Rule to Independence. Kent, Kent State University Press. 116

Todd Whitelaw: Feeding Knossos

Mee, C. and Forbes, H. (eds). 1997. A Rough and Rocky Place: The Landscape and Settlement History of the Methana Peninsula, Greece. Liverpool, Liverpool University Press. Milano, L. 1989. Food and diet in pre-classical Syria. In C. Zaccagnini (ed.), Production and Consumption in the Ancient Near East: 201-71. Budapest, University of Budapest. Militello, P. 1992. Aspetti del funzionamento del sistema amministrativo ad Haghia Triada. In J.-P. Olivier (ed.), Mykenaïka. Actes du IXe Colloque international sur les textes mycéniens et égéens, Centre de l’Antiquité Grecque et Romaine de la Fondation Hellénique des Recherches Scientifiques et École française d’Athènes: 411-14. Bulletin de Correspondance Hellénique, Supplement 25. Paris, Diffusion de Boccard. Millett, M. 1990. The Romanization of Britain: an Essay in Archaeological Interpretation. Cambridge, Cambridge University Press. Monroe, C. 2009. Scales of Fate. Trade, Tradition, and Transformation in the Eastern Mediterranean ca. 13501175 BCE. Alter Orient und Altes Testament 357. Münster, Ugarit-Verlag. Monroe, C. 2010. Sunk costs at Late Bronze Age Uluburun. Bulletin of the American Schools of Oriental Research 357: 19-33. Moody, J. 1987. The Environmental and Cultural Prehistory of the Khania Region of West Crete: Neolithic through Late Minoan III. Unpublished Ph.D. Dissertation. Minneapolis, University of Minnesota. Mori, L. 2003. Reconstructing the Emar Landscape. Quaderni di Geografia Storica 6. Rome, University of Rome, La Sapienze. Morley, N. 2008. Urbanisation and development in Italy in the late Republic. In de Ligt and Northwood (eds) People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14: 121-37. Leiden, Brill. Morris, M. 2002. Soil Science and Archaeology: Three Test Cases from Minoan Crete. Prehistory Monographs 4. Philadelphia, INSTAP Academic Press. Müller, S. 1991 Routes minoennes en relation avec le site de Malia. Bulletin de Correspondance Hellénique 115.2: 545-60. Müller-Celka, S. 2007. Terroir et territoire à Malia à l’époque protopalatiale. Bulletin de Correspondance Hellénique 131.2: 855-60. Müller-Celka, S., D. Puglisi and F. Bendali. 2014. Settlement pattern dynamics and natural resources in MM-LMI Crete: the case of Malia. In G. Touchais, R. Laffineur and F. Rougemont (eds), Physis. L’environment naturel et la relation home-milieu dans la monde egeen protohistorique: 431-40. Aegaeum 37. Liège, Université de Liège. Nakassis, D. 2013. Individuals and Society in Mycenaean Pylos. Mnemosyne Supplements 358. Leiden, Brill. Nakassis, D., Parkinson, W. and Galaty, M. 2011. Redistributive economies from a theoretical and

cross-cultural perspective. American Journal of Archaeology 115: 177-84. Naval Intelligence Division 1945. Greece. Volume III. Regional Geography. London, Naval Intelligence Division. Nichols, D. and T. Charlton (eds) 1997. The Archaeology of City-States: Cross-Cultural Approaches. Washington D.C., Smithsonian Institution Press. Osborne, R. and Wallace-Hadrill, A. 2013. Cities of the ancient Mediterranean. In P. Clark (ed.), The Oxford Handbook of Cities in World History: 49-65. Oxford, Oxford University Press. Padgham, K. 2014. The Scale and Nature of the Late Bronze Age Economies of Egypt and Cyprus. British Archaeological Reports, International Series 2594. Oxford, Archaeopress. Palaima, T. 1994. Seal-users and script-users/nodules and tablets at LMIB Hagia Triada. In P. Ferioli, E. Fiandra, G.G. Fissore and M. Frangipane (eds), Archives Before Writing. Proceedings of the International Colloquium Oriolo Romano, October 23-25, 1991: 30730. Rome, Centro Internazionale di Ricerche Archeologiche Antropologiche e Storiche. Palmer, R. 1989. Subsistence rations at Pylos and Knossos. Minos 24: 89-124. Palmer, R. 1994. Wine in the Mycenaean Palace Economy. Aegaeum 10. Liège Universitè de Liège. Palmer, R. 1995. Linear A commodities: a comparison of resources. In R. Laffineur and W.-D. Niemeier (eds), Politeia: Society and State in the Aegean Bronze Age. Proceedings of the 5th International Aegean Conference, University of Heidelberg, Archäologisches Institut, 10-13 April 1994, Vol. 1: 133-55. Aegaeum 12. Liège, Université de Liège. Palmer, R. 1998-99. Models in Linear B landholding: an analysis of methodology. In J. Bennet and J. Driessen (eds), A-na-qo-ta. Studies Presented to J. T. Killen. Minos 33-34: 223-50. Palmer, R. 1999. Perishable goods in Mycenaean texts. In S. Deger-Jalkotzy, S. Hiller and O. Panagl (eds), Floreant Studia Mycenaea. Akten des X. Internationalen Mykenologischen Colloquiums in Salzburg vom 1.5. Mai 1995. Band I: 463-85. Wien, Verlag der Österreichischen Akademie der Wissenschaften. Parkins, H. (ed.) 1997a. Roman Urbanism: Beyond the Consumer City. London, Routledge. Parkins, H. 1997b. The ‘consumer city’ domesticated? The Roman city in elite economic strategies. In H. Parkins (ed.) Roman Urbanism: Beyond the Consumer City: 81-111. London, Routledge. Parkinson, W. 1999. Chipping away at the Mycenaean economy: obsidian exchange, Linear B, and palatial control in Late Bronze Age Messenia. In M. Galaty and W. Parkinson (eds), Rethinking Mycenaean Palaces: New Interpretations of an Old Idea: 73-85. Monograph 41. Los Angeles, The Cotsen Institute of Archaeology. Parkinson, W. and Galaty, M. 2007. Secondary states in perspective: an integrated approach to state 117

Country in the City formation in the prehistoric Aegean. American Anthropologist 109: 113-29. Parkinson, W. and Galaty, M. (eds). 2009. Archaic State Interaction: The Eastern Mediterranean in the Bronze Age. Santa Fe, School for Advanced Research Press. Parkinson, W., Nakassis, D. and Galaty, M. 2013. Crafts, specialists and markets in Mycenaean Greece: introduction. American Journal of Archaeology 117: 413-22. Parsons, M. and Gifford, J. 1995. Soil and land use studies at Kommos. In J. Shaw and M. Shaw (eds), Kommos: An Excavation on the South Coast of Crete. Vol. 1: The Kommos Region and Houses of the Minoan Town. Part I: The Kommos Region, Ecology, and Minoan Industries: 292-325. Princeton, Princeton University Press. Pelon, O. 2005. Les deux destructions du palais de Malia. In I. Bradfer-Burdet, B. Detournay and R. Laffineur (eds), Kris Technitis. L’Artisan Crétois: Recueil d’articles en l’honneur de Jean-Claude Poursat, publié à l’occasion des 40 ans de la découverte du Quartier Mu: 185-97. Aegaeum 26. Liège, Université de Liège. Petrakis, V. 2014. Some notes on the palace of ku-do-nija in Late Minoan III political geography. Do-so-mo 10: 55-80. Pounds, N. 1974. An Economic History of Europe. Harlow, Longman. Poursat, J.-C. 1996. Fouilles exécutées à Malia: Le quartier Mu III. Artisans minoens: Les maisons-ateliers du quartier Mu. Études Crétoises 32. Athens, École française d’Athènes. Price, S. 2011. Estimating ancient Greek populations. In A. Bowman and A. Wilson (eds), Settlement, Urbanization and Population: 17-35. Oxford, Oxford University Press. Privitera, S. 2014. Long-term grain storage and political economy in Bronze Age Crete: contextualizing Ayia Triada’s silo complexes. American Journal of Archaeology 118: 429-49. Puglisi, D. 2003. Haghia Triada nel periodo Tardo Minoico I. Creta Antica 4: 145-98. Puglisi, D. 2011. La fornace da vasaio TM IB di Haghia Triada. Le ceramiche e il sistema di produzione, distribuzione e consumo. Creta Antica 12: 199-271. Pullen, D. 2013. Exchanging the Mycenaean economy. American Journal of Archaeology 117: 437-45. Rackham, O. and Moody, J. 1996. The Making of the Cretan Landscape. Manchester, Manchester University Press. Renfrew, C. 1972. The Emergence of Civilisation. The Cyclades and the Aegean in the Third Millennium B.C. London, Methuen. Renfrew, C. and Cherry, J. 1986. Peer Polity Interaction and Socio-political Change. Cambridge, Cambridge University Press. Renger, J. 1990. Different economic spheres in the urban economy of ancient Mesopotamia. In E. Aerts and H. Klengel (eds), The Town as Regional Economic

Centre in the Ancient Near East: 20-8. Leuven, Leuven University Press. Renger, J. 1994. Comment on: on economic structures in ancient Mesopotamia. Orientalia 63.3:157-208. Renger, J. 1995-96. Institutional, communal, and individual ownership or possession of arable land in ancient Mesopotamia from the end of the fourth to the end of the first millennium BC. Chicago-Kent Law Review 71: 269-319. Renger, J. 2007. Economy of ancient Mesopotamia. In G. Leick (ed.), The Babylonian World: 187-97. London, Routledge. Rethemiotakis, G. and Christakis, K. 2011. Landscapes of power in Protopalatial Crete: new evidence from Galatas, Pediada. Studi micenei ed egeo-anatolici 53: 195-218. Roberts, N. 1979. The location and environment of Knossos. Annual of the British School at Athens 74: 23141. Rougemont, F. 2009. Contrôle économique et administration à l’époque des palais mycéniens (fin du IIe millénaire av. J.-C.). Bibliothèque des Écoles françaises d’Athènes et de Rome 332. Athènes, École française d’Athènes. Sanders, W. and Santley, R. 1983. A tale of three cities: energetics and urbanization in pre-Hispanic central Mexico. In R. Leventhal and A. Kolata (eds), Prehistoric Settlement Patterns: Essays in Honor of Gordon R. Willey: 243-91. Cambridge, Peabody Museum of Archaeology and Ethnology. Scheidel, W. 2004. Demographic and economic development in the ancient Mediterranean world. Journal of Institutional and Theoretical Economics 160:743-57. Scheidel, W. 2014. The shape of the Roman world: modelling imperial connectivity. Journal of Roman Archaeology 27: 7-32. Scheidel, W. and Friesen, S. 2009. The size of the economy and the distribution of income in the Roman Empire. Journal of Roman Studies 99: 61-91. Schloen, D. 2001. The House of the Father as Fact and Symbol: Patrimonialism in Ugarit and the Ancient Near East. Vol. 1. Winona Lake, Eisenbrauns. Schoep, I. 1998-99. Minoan administration at Haghia Triada: a multi-disciplinary comparison of the Linear A tablets from the Villa and the Casa del Lebete. In J. Bennet and J. Driessen (eds), A-na-qo-ta. Studies Presented to J. T. Killen. Minos 33-34: 273-94. Schoep, I. 2002a. The Administration of Neopalatial Crete. A Critical Assessment of the Linear A Tablets and their Role in the Administrative Process. Suplementos a Minos 17. Salamanca, Ediciones Universidad de Salamanca. Schoep, I. 2002b. Social and political organization on Crete in the Proto-Palatial period: the case of Middle Minoan II Mallia. Journal of Mediterranean Archaeology 15: 101-32. Schoep, I. 2004. Assessing the role of architecture in conspicuous consumption in the Middle Minoan I-II periods. Oxford Journal of Archaeology 23: 243-69. 118

Todd Whitelaw: Feeding Knossos

Schoep, I. 2006. Looking beyond the First Palaces: elites and the agency of power in EM III-MM II Crete. American Journal of Archaeology 110: 37-64. Schoep, I. 2012. Bridging the divide between the ‘Prepalatial’ and the ‘Protopalatial’ periods? In I. Schoep, P. Tomkins and J. Driessen (eds), Back to the Beginning: Reassessing Social and Political Complexity on Crete during the Early and Middle Bronze Age: 403-28. Oxford, Oxbow. Schofield, E. 1996. Wash and brush up at the ‘Travellers’ rest’: the Caravanserai reconsidered. In D. Evely, I. Lemos and S. Sherratt (eds), Minotaur and Centaur. Studies in the Archaeology of Crete and Euboea presented to Mervyn Popham: 27-33. British Archaeological Reports, International Series 638. Oxford, Tempus Reparatum. Schon, R. 2007. Chariots, industry, and elite power at Pylos. In M. Galaty and W. Parkinson (eds), Rethinking Mycenaean Palaces II. Revised and expanded second edition: 133-45. Cotsen Institute of Archaeology at UCLA Monograph 60. Los Angeles, University of California. Shaw, J. and Shaw, M. 1999. A proposal for Bronze Age Aegean shipsheds in Crete. In H. Tzalas (ed.), Fifth International Symposium on Ship Construction in Antiquity, Nauplia, 26, 27, 28 August 1993. Proceedings: 369-82. Tropis 5. Athens, Hellenic Institute for the Preservation of Nautical Tradition. Shaw, J., Van de Moortel, A., Day, P. and Kilikoglou, V. 2001. A LM IA Ceramic Kiln in South-Central Crete. Function and Pottery Production. Hesperia Supplement 30. Princeton, The American School of Classical Studies at Athens. Shelmerdine, C. 1985. The Perfume Industry in Mycenaean Pylos. Studies in Mediterranean Archaeology, Pocket Book 34. Göteborg, Paul Åströms Förlag. Shelmerdine, C. 2013. Economic interplay among households and states. American Journal of Archaeology 117: 447-52. Shelmerdine, C. and Bennet, J. 2008. Economy and administration. In C. Shelmerdine (ed.), The Cambridge Companion to the Aegean Bronze Age: 289309 Cambridge, Cambridge University Press. Sherratt, A. 1993. What would a Bronze-Age world system look like? Relations between temperate Europe and the Mediterranean in later prehistory. Journal of European Archaeology 1.2: 1-58. Sherratt, S. 2001. Potemkin palaces and route-based economies. In S. Voutsaki and J. Killen (eds), Economy and Politics in the Mycenaean Palace States. Proceedings of a Conference Held on 1-3 July 1999 in the Faculty of Classics, Cambridge: 214-38. Cambridge, Cambridge Philological Society. Smith, C. 1974. Economics of marketing systems: models from geography. Annual Review of Anthropology 1974: 167-201.

Smith, M. 2005. Networks, territories and the cartography of ancient states. Annals of the Association of American Geographers 95: 832-49. Soles, J. 2003. Mochlos IA: Period III. Neopalatial Settlement on the Coast: The Artisans’ Quarter and the Farmhouse at Chalinomouri. The Sites. Prehistory Monographs 7. Philadelphia, INSTAP Academic Press. Spencer, C. and Bevan, A. 2018. Settlement location models, archaeological survey data and social change in Bronze Age Crete. Journal of Anthropological Archaeology 52: 71-86. Stark, B. and Garraty, C. 2010. Detecting marketplace exchange in archaeology: a methodological review. In C. Garraty and B. Stark (eds), Archaeological Approaches to Market Exchange in Ancient Societies: 3358. Boulder, University Press of Colorado. Stone D. 2013. Economy. In P. Clark (ed.), The Oxford Handbook of Cities in World History: 127-46. Oxford, Oxford University Press. Stone, E. 1995. The development of cities in ancient Mesopotamia. In J. Sasson (ed.), Civilizations of the Ancient Near East, Vol.1: 235-48. New York, Scribner’s. Stone, E. 2007. The Mesopotamian urban experience. In E. Stone (ed.), Settlement and Society. Essays Dedicated to Robert McCormick Adams: 213-34. Los Angeles, Cotsen Institute of Archaeology. Tainter, J. 1988. The Collapse of Complex Societies. Cambridge, Cambridge University Press. Tomasello, F. 2011. La fornace da vasaio TM IB di Haghia Triada. L’impianto. Creta Antica 12: 131-98. Trigger, B. 2003. Understanding Early Civilizations: A Comparative Study. Cambridge, Cambridge University Press. Tsougarakis, D. and Angelomatis-Tsougarakis, E. 2004. A province under Byzantine, Venetian, and Ottoman rule, AD 400-1898. In V. Watrous, D. Hadzi-Vallianou and H. Blitzer (eds), The Plain of Phaistos. Cycles of Social Complexity in the Mesara Region of Crete: 359-439. Monumenta Archaeologica 23. Los Angeles, Cotsen Institute of Archaeology. Tyumenev, A. I. 1969. The working personnel on the estate of the temple of Ba-U in Lagaš during the period of Lugalanda and Urukagina (25th24th centuries BC). In I. Diakonoff (ed.), Ancient Mesopotamia. Socio-Economic History. A Collection of Studies by Soviet Scholars: 88-126. Moscow, Nauka. Tzedakis, Y., Chrisoulaki, S., Voutsaki, S. and Venieri, Y. 1989. Les Routes minoennes: rapport préliminaire. Défense de la circulation ou circulation de la défense. Bulletin de Correspondance Hellénique 113: 43-75. Uchitel, A. 2003. Local differences in arrangements of ration lists on Minoan Crete. In A. Bowman, A. Cooley and M. Brosius (eds), Ancient Archives and Archival Traditions: Concepts of Record-Keeping in the Ancient World: 139-52. Oxford Studies in Ancient Documents. Oxford, Oxford University Press. Ur, J. 2009. Emergent landscapes of movement in Early Bronze Age northern Mesopotamia. In J. Snead, 119

Country in the City C. Erickson and A. Darling (eds), Landscapes of Movement: Paths, Trails, and Roads in Anthropological Perspective: 180-203. Philadelphia, University of Pennsylvania Press. Ur, J. 2013. Patterns of settlement in Sumer and Akkad. In H. Crawford (ed.), The Sumerian World: 131-55. London, Routledge. Ur, J. and Wilkinson, T. 2008. Settlement and economic landscapes of Tell Beydar and its hinterland. In M. Lebeau and A. Suleiman (eds), Beydar Studies I: 30527. Subartu 21. Turnhout, Brepols. Van de Mieroop, M.1997. The Ancient Mesopotamian City. Oxford, Oxford University Press. Van de Moortel, A. 2011. LM IB ceramic phases at Palaikastro and Malia: a response to Seán Hemingway, J. Alexander MacGillivray, and L. Hugh Sackett. In T. Brogan and E. Hallager (eds), LM IB Pottery: Relative Chronology and Regional Differences. Acts of a Workshop Held at the Danish Institute at Athens in Collaboration with the INSTAP Study Center for East Crete, 27-29 June 2007. Vol. 2: 531-48. Monographs of the Danish Institute at Athens 11.1-2. Athens, The Danish Institute at Athens. van Effenterre, H. 1980. Le palais de Malia et la cité minoenne. Incunabula Graeca 76. Rome, Edizioni Dell’Ateneo. van Koppen, F. 2001. The organisation of institutional agriculture in Mari. Journal of the Economic and Social History of the Orient 44: 451-504. van Soldt, W. 2005. The Topography of the City-state of Ugarit. Alter Orient und Altes Testament Volume 324. Münster, Ugarit Verlag. van Soldt, W. 2010. Landholders in administrative texts. In W. van Soldt (ed.), Society and Administration in Ancient Ugarit. Papers Read at a Symposium in Leiden, 13-14 December 2007: 151-63. Leiden, Nederlands Instituut voor het Nabije Oosten. Vansteenhuyse, K. 2011. Centralisation and the political institution of Late Minoan IA Crete. In N. Terrenato and D. Haggis (eds), State Formation in Italy and Greece: Questioning the Neoevolutionist Paradigm: 61-74. Oxford, Oxbow Books. Veenhof, K. 1972. Aspects of Old Assyrian Trade and Its Terminology. Studia et Documenta 10. Leiden, Brill. Ventris, M. and Chadwick, J. 1956. Documents in Mycenaean Greek. Cambridge, Cambridge University Press. Vidal, J. 2014. On the demography of Ugaritian villages: a research note. Journal of the Economic and Social History of the Orient 57: 44-54. Visicato, G. 1995. The Bureaucracy of Šuruppak: Administrative Centres, Central Offices, Intermediate Structures and Hierarchies in the Economic Documentation of Fara. Abhandlungen zur Literatur Alt-Syrien-Palästinas und Mesopotamiens Bd. 10. Münster, Ugarit-Verlag. Vita, J.-P. 1999. The society of Ugarit. In W. Watson and N. Wyatt (eds), Handbook of Ugaritic studies: 455-98.

Handbuch der Orientalistik. Erste Abteilung, Nahe und der Mittlere Osten 39. Leiden, Brill. von Dassow, E. 2008. State and Society in the Late Bronze Age: Alalaḫ Under the Mittani Empire. Studies on the civilization and culture of Nuzi and the Hurrians, Volume 17. Bethesda, CDL Press. Voutsaki, S. 2001. Economic control, power and prestige in the Mycenaean world: the archaeological evidence. In S. Voutsaki and J. Killen (eds), Economy and Politics in the Mycenaean Palace States. Proceedings of a Conference Held on 1-3 July 1999 in the Faculty of Classics, Cambridge: 195-213. Cambridge, Cambridge Philological Society. Walberg, G. 1995. Minoan economy: an alternative model. In R. Laffineur and W.-D. Niemeier (eds), Politeia: Society and State in the Aegean Bronze Age. Proceedings of the 5th International Aegean Conference, University of Heidelberg, Archäologisches Institut, 10-13 April 1994, Vol 1: 157-61. Aegaeum 12. Liège, Université de Liège. Warren, P. 1994. The Minoan roads of Knossos. In D. Evely, H. Hughes-Brock and N. Momigliano (eds), Knossos: A Labyrinth of History. Papers Presented in Honour of Sinclair Hood: 189-210. Oxford, Oxbow Books. Warren, P. 2012. The apogee of Minoan civilization: the final Neopalatial period. In E. Mantzourani and P. Betancourt (eds), Philistor: Studies in Honor of Costis Davaras: 255-72. Prehistory Monographs 36. Philadelphia, INSTAP Academic Press. Watrous, L. 1982. Lasithi. A History of Settlement on a Highland Plain in Crete. Hesperia Supplement 18. American School of Classical Studies at Athens, Princeton. Watrous, V. 2007. Harbors as agents of social change in ancient Crete. In P. Betancourt, M. Nelson and H. Williams (eds), Krinoi kai Limenes: Studies in Honor of Joseph and Maria Shaw: 101-6. Prehistory Monographs 22. Philadelphia, INSTAP Academic Press. Watrous, V., D. Hadzi-Vallianou and H. Blitzer. (eds) 2004. The Plain of Phaistos: Cycles of Social Complexity in the Mesara Region of Crete. Monumenta Archaeologica 23. Los Angeles, The Cotsen Institute of Archaeology. Watrous, V., Haggis, D., Nowicki, K., Vogeikoff-Brogan, N. and Schultz, M. 2012. An Archaeological Survey of the Gournia Landscape: A Regional History of the Mirabello Bay, Crete, in Antiquity. Prehistory Monographs 37. Philadelphia, INSTAP Academic Press. Watrous, V. and Heimroth, A. 2011. Household industries of Late Minoan IB Gournia and the socioeconomic status of the town. In K. Glowacki, and N. VogeikoffBrogan (eds), Στέγα: The Archaeology of Houses and Households in Ancient Crete: 199-222. Hesperia Supplement 44. Princeton, American School of Classical Studies at Athens. Weber, M. 1958. The City. New York, Free Press.

120

Todd Whitelaw: Feeding Knossos

Wells, B. and Runnels, C. (eds). 1996. The Berbati-Limnes Archaeological Survey 1988-1990. Acta Athenensia, Quarto Series 44. Stockholm, Paul Åströms Forlag. Whitelaw, T. 1991. The ethnoarchaeology of recent rural settlement and land use in northwest Keos. In J. Cherry, J. Davis and E. Mantzourani (eds), Landscape Archaeology as Long-Term History: 403-54. Monumenta Archaeologica 16. Los Angeles, Institute of Archaeology. Whitelaw, T. 2001a. From sites to communities: defining the human dimensions of Minoan urbanism. In K. Branigan (ed.), Urbanism in the Aegean Bronze Age: 15-37. Sheffield Studies in Aegean Archaeology. London, Sheffield Academic Press. Whitelaw, T. 2001b. Reading between the tablets: assessing Mycenaean palatial involvement in ceramic production and consumption. In S. Voutsaki and J. Killen (eds), Economy and Politics in the Mycenaean Palace States. Proceedings of a Conference held on 1-3 July 1999 in the Faculty of Classics, Cambridge: 51-79. Cambridge Philological Society Supplementary Volume 27. Cambridge, Cambridge Philological Society. Whitelaw, T. 2004. Estimating the population of Neopalatial Knossos. In G. Cadogan, E. Hatzaki and A. Vasilakis (eds), Knossos: Palace, City, State: Proceedings of the Conference in Herakleion Organised by the British School at Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities of Herakleion, in November 2000, for the Centenary of Sir Arthur Evans’s Excavations at Knossos: 147-58. British School at Athens, Studies 12. London, The British School at Athens. Whitelaw, T. 2012. The urbanisation of prehistoric Crete: settlement perspectives on Minoan state formation. In I. Schoep, P. Tomkins and J. Driessen (eds), Back to the Beginning: Reassessing Social and Political Complexity on Crete during the Early and Middle Bronze Age: 114–76. Oxford, Oxbow Books. Whitelaw, T. 2014. Political formations in prehistoric Crete. Bulletin of the Institute of Classical Studies 57.2: 143-44. Whitelaw, T. 2015. The divergence of civilisation: Fournou Korifi and Pyrgos. In C. Macdonald, E. Hatzaki and S. Andreou (eds), The Great Islands. Studies of Crete and Cyprus presented to Gerald Cadogan: 41-8. Athens, Kapon. Whitelaw, T. 2017. The development and character of urban communities in Prehistoric Crete in their regional context: a preliminary study. In Q. Letesson and C. Knappett (eds), Minoan Architecture and Urbanism. New Perspectives on an Ancient Built Environment: 114-80. Oxford, Oxford University Press. Whitelaw, T. 2018. Recognising polities in prehistoric Crete. In M. Relaki and Y. Papadatos (eds), From the Foundations to the Legacy of Minoan Archaeology, Studies in Honour of Professor Keith Branigan: 210-55. Oxford, Oxbow Books.

Whitelaw, T., Day, P., Kiriatzi, E., Kilikoglou, V. and Wilson, D. 1997. Ceramic traditions at EM IIB Myrtos, Fournou Korifi. In R. Laffineur and P. Betancourt (eds), TEHNI: Craftsmen, Craftswomen and Craftsmanship in the Aegean Bronze Age. Proceedings of the 6th International Aegean Conference, Philadelphia, Temple University, 18-21 April 1996: 265-74. Aegaeum 16. Liège, Université de Liège. Whitelaw, T., Vasilakis, A. and Bredaki, M. 2018. Prehistoric Knossos: tracing its long-term history through its surface record. In E. Gavrilaki (ed.), Proceedings of the 11th Cretological Congress, Vol. A1.1: 293-307. Rethymnon, Historical and Folklore Society of Rethymnon. Wiener, M. 2007. Neopalatial Knossos: rule and role. In P. Betancourt, M. Nelson and H. Williams (eds), Krinoi kai Limenes: Studies in Honor of Joseph and Maria Shaw: 231-42. Prehistory Monographs 22. Philadelphia, INSTAP Academic Press. Wilkinson, T. 1994. The structure and dynamics of dry-farming states in Upper Mesopotamia. Current Anthropology 35: 483-520. Wilkinson, T. 2003. Archaeological Landscapes of the Near East. Tucson, University of Arizona Press. Wilkinson, T. 2005. Approaches to modelling archaeological site territories in the Near East. In C. Beekman and W. Baden (eds), Nonlinear Models for Archaeology and Anthropology. Continuing the Revolution. Aldershot, Ashgate: 123-38. Wilkinson, T. 2010. The tell: social archaeology and territorial space. In D. Bolger and L. Maguire (eds), The Development of Pre-state Communities in the Ancient Near East. Studies in Honour of Edgar Peltenburg: 55-62. Oxford, Oxbow. Wright, H. 1981. The southern margins of Sumer. Archaeological survey of the area of Eridu and Ur. In R. Adams, Heartland of Cities: Surveys of Ancient Settlement and Land Use on the Central Floodplain of the Euphrates: 295-345. Chicago, University of Chicago Press. Wright, H. 2000. Modelling tributary economies and hierarchical polities. In G. Feinman and L. Manzanilla (eds), Cultural Evolution. Contemporary Viewpoints: 197-213. New York, Plenum. Wright, H. 2005 The polycentricity of the Archaic civilizations. In V. Scarborough (ed.), A Catalyst for Ideas: 149-67. Santa Fe, School of Advanced Research. Zaccagnini, C. 1979. The Rural Landscape of the Land of Arrapḫe. Rome, Dipartimento di scienze storiche archeologiche e anthropologiche dell’antichità Università degli studi di Roma La Sapienza. Zielinski, J. 1998. Cyclopean Architecture in Minoan Bronze Age Crete: A Study in the Social Organization of a Complex Society. Unpublished PhD Dissertation. Buffalo, University of New York at Buffalo.

121

Beyond City and Country at Mycenae: Urban and Rural Practices in a Subsistence Landscape Lynne A. Kvapil Butler University

Jacqueline S. Meier University of North Florida

Gypsy C. Price

Appalachian State University

Kim S. Shelton

University of California, Berkeley

Introduction

for his model of economic change— namely a shift from previous models focused on labor-intensive cultivation to a model that promotes extensive cultivation schemes that made use of more land and animal labor — which has served as a key exploratory mechanism for Mycenaean economic studies (Halstead 1992a). Archives like those at Pylos and Knossos, however, have not yet been found at every Mycenaean palace; therefore models relying heavily on these texts may not apply universally; and, because palatial texts, by definition, record information of interest to the palace, large portions of the Mycenaean economy are absent from them. Halstead himself found discrepancies between textual and archaeological evidence of subsistence that highlight the need for more combined studies (1988, 1992a, 1992b, 1999a, 2001). This discrepancy emphasizes the need to explore different lines of evidence, especially for palaces will less textual evidence, such as Mycenae, that with explain the high level of economic variation currently noted among different palatial areas that shared cultural traits (Shelmerdine 1999; Shelmerdine and Bennet 2008).

Approaches that explicate the role of cities as participants in the agrarian systems of the Late Bronze Age (LBA) Aegean frequently subscribe to paradigms that strongly dichotomize urban and rural populations, and thus inherently contrast the activities that take place within them. While core-periphery models serve as valuable heuristic devices for studying exchange, they have the unfortunate consequence of divorcing the practices of urban and rural agents (Kardulias 1999; Parkinson 2010). This dichotomy is compounded when paired with models that favor redistributive exchange mechanisms enacted at palatial centers during the LBA that are contingent upon an ‘us’ (elites) versus ‘them’ (provincial rural population) mentality (Finley 1957; Killen 1985, 2008; Parkinson 2010; Polanyi 1957; Voutsaki and Killen 2001). Through this lens, resources are often conceptualized as materials to be gathered through taxation or mobilization towards the urban center for consumption by the ruling elite and associates (Halstead 1992a, 1992b, 1995, 1999b; Killen 1985, 1998; Morris 1986; Palmer 1992). The emphasis on redistribution also highlights the current research focus on palatial practices, which has led to exclusive cultural profiles established chiefly on evidence gleaned from urban centers, rather than taking into account the wider rural practices and geographies with which the palace may have interacted.

We aim to reconceptualize Mycenae as an urban center situated within the broader agricultural geography of the northeastern Peloponnese, a view that will be utilized thereafter to refine the picture of subsistence activities at this palatial settlement during the LBA. Our vision of what is considered ‘urban’ for this time period casts off the notion of a built-up city or walled settlement. Instead, our definition of urban is informed by an investigation of subsistence practices enacted across a range of social and political spheres that transcend binary divisions. Mycenae and the surrounding region offer a unique opportunity for this kind of study due to its rich history of archaeological exploration and the wide range of archaeological data therefore available. By combining a variety of independent lines of evidence — settlement patterns

This focus on palatial practices is largely a consequence of available archaeological evidence, especially since palaces have historically been the favored locations for large-scale excavations. Moreover, models for the Mycenaean economy often rely heavily on Linear B texts from palaces where large deposits of clay tablets, usually regarded as archives, have been found. Linear B texts, primarily from Pylos and Knossos, were used by Halstead, along with ethnographic analogies, as the basis 122

Lynne A. Kvapil et al.: Beyond City and Country at Mycenae

and landscape modification, archaeological evidence from various artifact and ecofact categories, and textual evidence — we illustrate that it is more informative to examine the landscape as one of highly integrated social and economic interests with effects extending beyond our perceived confines of ‘city and country’ or ‘palace and hinterland’.

small-scale, diversified, non-palatial production as opposed to large-scale, specialized, palatial production is problematic for several reasons, as Halstead himself acknowledged (Halstead 2011). First and foremost, it positions subsistence strategies as strictly palatial or non-palatial, which is a distinction that does not hold for data from Mycenae, as we will show especially in the realm of animal management practices, wild animal acquisition, and the gathering and cultivation of botanicals.

Background Current models and interpretations

In general, it is difficult to test certain components of this model with the current evidence. For example, given the dearth of faunal remains studied from extrapalatial areas, excluding unique feasting or ritual deposits (Dabney et al. 2004; Hamilakis and Konsolaki 2004), studies of the scale of animal management practices are limited to a narrow part of Mycenaean society. The current zooarchaeological perspective of animal-based subsistence in the Peloponnese is derived from largely spatially aggregated faunal assemblages. Combining animal bones from multiple contexts to comprise a general-Aegean faunal economic baseline is informative (Trantalidou 1990), yet neglects to separate fauna from palatial and extra-palatial areas of the same site in ways that expose differential access to animal resources. In the area of plant use, text-based assessments describing the cultivation and distribution of plants at Mycenaean sites do not reflect a diversity that is visible in the archaeological record (Halstead 1992a, 1995; Hansen 1988). This conflict between textual and archaeological evidence for botanicals is illustrated by the presence of legumes, for instance, in the archaeobotanical record for LBA Greece as opposed to their total absence in the Linear B texts (Hillman 2011; Sarpaki 1992). Texts from Mycenae also suggest that aromatic and condiment plants were not limited to hinterland production and palatial consumption; instead, they likely were consumed in and out of palace centers but also cultivated or gathered and exchanged within and beyond the palatial sphere.

Many models have been constructed to understand complex economic interactions across the Aegean during the LBA. The model most relevant to a reconsideration of the subsistence economy at LBA Mycenae is that of Halstead, who proposed one based on a combination of textual and archaeological evidence for centralized surplus accumulation (1992a). Building upon prior models proposing that agricultural specialization and redistribution drove state formation in the Aegean Bronze Age (Gamble 1979; Renfrew 1972), Halstead emphasized the importance of surplus as a safeguard against agricultural failure (1981). In putting forth this model, he also proposed that extensification was likely due to changes in agricultural land use, sources and deployment of labor, and animal management over time, which previously seemed to support a model of increasing specialization (1992b, 1995). These changes included the practice of loaning labor animals, as suggested in Linear B texts, and the focused production of sheep (Halstead 1999a, 2001: 40-41; Killen 1993, 1998). While sharing labor animals and trading wool would ideally create social capital and reciprocal risk buffering opportunities, clear evidence for these practices was not born out in the data. Halstead (2003) found instead that increasingly specialized production of sheep was not entirely supported by abundances of domesticated animals. These findings do not suggest a complete shift towards large-scale specialized production over time in the Peloponnese, based on the available diachronic studies of faunal remains from Tiryns and Nichoria, which reveal little change in taxa before and after the rise of the major palaces (Late Helladic (LH) IIIA-C) (Halstead 2003, von den Driesch and Boessneck 1990, Mancz 1989).

These issues, which will be discussed below, illustrate that a more holistic approach, such as we present here, is needed to develop the full picture of the subsistence economy. We focus our study on the site of Mycenae to better understand the role of agriculture in its regional context. Our aim is that this study will provide a means of shedding light on the wide variety of possible plant and animal production and acquisition practices that were employed with differing degrees of palatial involvement across the Aegean.

Halstead concluded that the overall evidence did not support a wholesale shift to large-scale herding or specialization for exchange over time as earlier models expected (Renfrew 1972); rather, he proposed that the texts suggest that palaces practiced large-scale herding and specialization, but, in non-palatial contexts, small-scale agriculture was more likely practiced with diversified agricultural goods, which would shield small farmers from risk and which could be mobilized upwards to palaces specializing in the production of wheat (Halstead 1996, 1995, 1999b). This model of

The site of Mycenae The fortified citadel of Mycenae occupies a limestone hill nestled between two low mountains at the northern end of the Peloponnesian plain of Argos (Figure 1). It is 123

Country in the City

Figure 1: Map of study area including sites mentioned in text. (Figure compiled by Price using QGIS software)

uniquely situated to dominate all the area it oversees and to exploit upland pastures, lowland fields, and routes to and from the shore approximately 20 km to the south. The palace of the fourteenth to thirteenth centuries BC (LH IIIA–B) was built and rebuilt during a long and prosperous phase with ever-increasing signs of centralized control, such as organized building programs that would have required a large workforce— used to create the structures on the acropolis and its fortification, on the one hand, and extramural burial monuments and regional infrastructure, on the other (Iakovidis 1983; French 1989; 2002; Iakovidis and French 2003).

and ceramics workshop (Shelton 2010; 2015), and the Ivory Houses, four domestic and industrial structures destroyed in the 13th century BC (Tournavitou 1995), thus spanning most of the palatial period at Mycenae. We utilize new extra-palatial data derived from a large deposit of material, including preliminary analysis of one quarter of the massive animal bone assemblage of over 60,000 fragments total, excavated between 2002 and 2007 from a well, located within Petsas House. Additionally, excavations at the Ivory Houses yielded the largest number of Linear B tablets from Mycenae, which inform our reevaluation of the site economy. While these tablets were not found in a large enough quantity to constitute an archive, the texts inscribed on them included reference to the acquisition of oil, wool, and, more importantly for this study, edible plants (Bennett and Chadwick 1958; Tournavitou 1995, 257276). Because these houses were located outside of the fortified citadel, they may have played some role in the

The dataset we utilize here is derived primarily from houses in the extensive settlement on the lower slopes of the ridges and hills surrounding the citadel to the north and west (Iakovidis and French 2003), including Petsas House, a 14th century BC residence 124

Lynne A. Kvapil et al.: Beyond City and Country at Mycenae

palatially-administered economy, but there is the equal likelihood that they also participated in extra-palatial economic activities.

and micro-regional economic networks that aided in sustaining the palace but did not entirely stem from it and were by no means controlled by it.

Overview of mycenaean subsistence activities

Agricultural geography

Primary production activities at Mycenae illustrate the integrated nature of this landscape in that a series of decisions regarding the management of land, and other resources, were certainly made at the administrative level by officials who may not have been actively involved in farming or herding; yet production itself took place outside the main administrative center. These decisions are attested in the agriculturallyfocused landscape management installations, monuments that must have been built with palatial support and still survive throughout the immediate hinterland of Mycenae (Iakovides and French 2003), and in Linear B texts that discuss the management of animal products and food resources (Bennett and Chadwick 1958; Chadwick et al. 1962; Tournavitou 1995; Ventris and Chadwick 1973). At the same time, we estimate that activities carried out beyond the Cyclopean fortification walls, such as the acquisition of wild game and the gathering and household cultivation of edible plants and aromatics would have been less regulated and therefore, carried out according to choices outside of palatial administrative control. The results of those independent choices would have resonated in and out of the urban core. Such activities also may have set the stage for the circulation and exchange of goods across socio-political boundaries, as well as market-type activities, that may have taken place in houses which functioned independently or semi-independently of the palace center (Aprile 2013; Hruby 2013; Parkinson 2007; Whitelaw 2001). Therefore, an agricultural model centered on extensive palatial production of wheat, barley, and sheep does not fully account for the production of other goods, which were integral to processes of economic change over time.

Before looking closely at these networks, it is advantageous to reexamine what we term Mycenae’s agricultural geography. Agricultural geography, as we define it, includes not only lines of connectivity between inhabited centers of various sizes or site hierarchies but also centers connected through the surrounding productive landscape. In the case of Mycenae, we can envision the whole landscape as one interwoven with diverse social, economic, and political networks but organized on the whole through subsistence activities and defined by competition, cooperation, and negotiation. Competition in the agricultural landscape is most immediately apparent to the west and south of Mycenae over the Argolid Plain. Within this area are located Tiryns, Nauplio, Argos, and Midea, all of which were substantial settlements that matched or rivaled Mycenae for the status of palace center. It has been proposed that, among these settlements around the plain, there was differential access to imported goods (Burns 2010), a variety of exchanges between palace centers and non-palace centers (Sjöberg 2004), and the visibly competitive use of natural resources like architectural stone (Wright 1987, 2009), as opposed to a single, dominant palace that controlled exchange networks in the region (Voutsaki 1995, 2001, 2010). This diversity of interactions suggests that an environment in which competition and negotiation among and between centers would have been much more the norm than dominance by one or more central palaces. By analogy, then, we can speculate that the environment of social and political competition and negotiation apparent in the exchange of goods and the deployment of natural resources extended to the physical environment as well. The arable land of the Argolid Plain, which was located in the periphery of each of the abovementioned centers, would have been a valued natural resource. Geological investigations of the plain suggest that the soil was fertile and welldrained, and therefore would have been readily cultivated (Finke 1988; Philippson 1959; van Andel, Zangger, and Demitrack 1990; Zangger 1992, 1993). Even though the land of the plain may not have been farmed as aggressively in the LBA as it is today, surely each of the Argolid centers would have relied to some extent on the nearby arable fields where cultivators could have sown grain crops that contributed significantly to the agricultural economy. Given that large tracts of fertile flat land are atypical in southern Greece and that the land of the plain was of limited availability,

Because of the broad array of evidence for food production and acquisition available from Mycenae, extending from the palace center to the hinterland, we have the opportunity to view the palace as more than a ‘consumer city’ (Weber 1969; Finley 1973) or the final destination of upwardly mobilized goods (Gamble 1982). Reassessment of the role of the palace beyond consumption is particularly warranted in this case considering the active part the palatial administration seems to have played in supporting agricultural production in the form of landscape infrastructure, which was substantial enough to have controlled water (Iakovides and French 2003), for example, and facilitated movement throughout the hinterland (Jansen 1994). Therefore, we argue that it is more constructive to imagine Mycenae as a whole, palace and beyond, which encompassed a variety of interwoven regional 125

Country in the City

Figure 2: Map of Mycenaean roads. (Figure compiled by Price using QGIS software)

it is reasonable to suggest that the Argolid centers competed for access to these fields or even to crops grown on it.

and M3) beginning at the Lion Gate. Although these roads appear to have headed towards non-palatial settlements in the Corinthia, such as Tsoungiza and Aidonia, their immediate purpose seems to have been to open up access to farmland just outside the settlement of Mycenae and to the upland fields around Berbati. These hilly areas preserve the best evidence for their routes in the form of installations built into the roadways for the purpose of controlling the seasonal flow of water through remata. Road terraces with drainage culverts and weirs, built in Cyclopean masonry, mark the path of the roads. These landscape features would have mitigated soil erosion and aided in the control and conservation of precious water. The

Traces of this sort of competition are certainly difficult to identify, especially since the Argolid Plain remains a valuable piece of agricultural land and for this reason has been deeply plowed for many years. But, the remains of Mycenae’s road system (Figure 2) and associated installations (Iakovides and French 2003; Jansen 1994; Lavery 1990; Mylonas 1966: 86-87) provide some clues to their approach to the agricultural landscape. Three roadways, with preserved traces of ruts and paving, extended to the north and east of the palace (M1, M2, 126

Lynne A. Kvapil et al.: Beyond City and Country at Mycenae

remains of agricultural terraces — probably Mycenaean — were also identified by the Mycenae survey in hills surrounding the acropolis and by the Berbati-Limnes survey (Iakovides and French 2003; Schallin 1996: 172). Terraces would have created flat land for farming but also would have helped drain excess water while retaining moisture in the facade and interior packing of the retaining wall.

While it may not be possible to determine the type of role Mycenae played, the seeming lack of competition, appearance of stability, and increase in cultivation suggest that the settlement pattern changes came with the cooperation of regional inhabitants. Mycenae’s interest in landscape modification appears less significant to the south, where roads stemming from the acropolis headed in the direction of the plain but ultimately towards the other Argolid centers. The M4, for example, seems to have had the purpose of linking Mycenae to Tiryns, in which case it may have been used as much for the competitive exchange of luxury goods between palatial elites or negotiated access to the sea as it may have been for facilitating access to agricultural land. If there was competition between Mycenae, Tiryns, and Midea for the scant arable land of the plain, as suggested above, perhaps there was not enough stability of land tenure in areas beyond the southern Mycenaean roadways to support a serious investment in infrastructure and landscape modification. Instead, roadways connected palaces to palaces rather than palaces to stable landholdings. This difference in the itineraries and ultimate goals of the roads may be a sign that the Argolid Plain was more of a contested landscape between the palaces that surrounded it.

All of these water and soil conservation features would have been critical to successful cultivation, and they also illustrate multiple perspectives on the land north and east of Mycenae. The investment of labor and resources needed for paved roadways, Cyclopean water management installations, and systems of agricultural terraces was substantial enough that they could only have been constructed with palatial resources; and, to invest in these constructions, the palace must have seen land tenure as stable and their need for agricultural goods as somewhat predictable. The farmers, too, who may have contributed labor in return for more productive fields, may also have maintained some confidence in their attachment to the land and their ability to produce enough to support themselves and return something to the palace as well. For these reasons, the heavily managed landscape surrounding Mycenae and the upland basins of Berbati does not seem to have been contested, nor does there seem to have been rivalry for access to it.

In the Corinthia north of Mycenae we can see possible evidence for cooperation rather than competition and indications that aspects of primary production may have been negotiated features of the agricultural economy. At Kalamianos (Tartaron et al. 2011), Tsoungiza (Dabney, Halstead, and Thomas 2004), and Aidonia (Demakopoulou 1996), there are signs of local elites who defined themselves as such in part through the use of Mycenaean material culture, including fine, decorated pottery. The absence of a palace structure in these areas has raised questions (Pullen and Tartaron 2007), however, about the overall status of these sites in a scheme where palatial centers seem to have been more powerful, regional entities. Yet, the palace at Mycenae in particular would have had reason to be involved in these areas. Tsoungiza (Cherry and Davis 2001; Wright et al 1990) and Aidonia (Hachtmann 2013) may have controlled important land routes to the western Corinthia and interior of the Peloponnese, respectively, while Kalamianos had a deep sea port that allowed access to the Saronic Gulf and sea lanes beyond (Tartaron et al. 2011). Mycenae, especially in view of the potential competition for access to similar routes to the south mentioned above, might have relied on negotiated relationships with all of these centers, and those negotiations might have been to some extent structured around and related to agricultural production that, in turn, contributed to palatial development. The agricultural land surrounding each of these Corinthian settlements makes it possible that

The Berbati-Limnes survey also noted an increase in the number of settlements scattered throughout the survey area during the palatial period at Mycenae (Schallin 1996: 170-172). An increase in dispersed settlements, as opposed to a single centralized settlement, suggests that cultivators were living closer to their land. The changes in settlement patterns, in conjunction with the aforementioned investments in landscape modification, indicate changes in the organization of agriculture in the Berbati region at this time. Terracing would have created more land for farming and would have varied the cultivation environments by adding fields that would have been suitable as much for arboriculture and horticulture as wheat or barley. The presence of terracing in areas where farmers had easier access to some fields than others reinforces the idea that land was being farmed in multiple ways (from extensive to intensive or something in between) and that a variety of crops were cultivated to meet the wide range of palatial needs suggested in part by Linear B texts (Kvapil 2012). The promotion of stability in landholding through their palatial support of the construction of terraces and water management installations along with the investment in a road system that made fields more accessible and facilitated the movement of agricultural goods, illustrates the role that the palace played in the development of a regional agricultural network. 127

Country in the City Animal acquisition

they were, as Pullen has suggested for Kalamianos (this volume), agriculturally self-sufficient, which likewise indicates that the offer of surplus subsistence goods from a palace center might not have been a powerful bargaining chip on the part of Mycenae.

Estimates of the proportion of wild game utilized in extra-palatial areas provide an avenue to assess the type of animal access available outside of the palace at Mycenae. Even though there are limited references to wild animals in Linear B texts (Halstead 2003; Ventris and Chadwick 1973), a range of wild species have been recovered with good preservation at Late Bronze Age sites, including Mycenae. Although there are few comparative sources specifically for small wild animals at LBA Aegean sites, the percentage of small game taxa represented in faunal assemblages (hare, wild carnivores, and hedgehogs) is higher from the well in room Π at Petsas House at Mycenae (3.5%)(Meier et al. In Press) than at Lerna (0.3%; Early and LBA), Pevkakia (0.4%; Early, Middle and LBA), and Kastanas (1.9%; Early-LBA) (Halstead and Isaakidou 2013). This suggests that small wild animals were more frequently acquired for a variety of uses, likely for food or skins, in the LH IIIA2 period of Petsas House than at other sites with multiple periods combined. Birds, fish, and shellfish remains comprise an additional 9.3% of Petsas House fauna, indicating that other small wild species were a noteworthy part of animal subsistence at the site. Such supplementation of the diet with hunted or caught wild game is expected to reflect incomplete consumer dependency on state-regulated, specialized animal production with centralized distribution based on models of early state-level agriculture in the Near East (Zeder 1991). Overall, the high proportion of small, wild game at Petsas House suggests that wild animal resources were used to supplement the diet, and, thus, these extra-palatial occupants were not completely dependent on palatially-regulated subsistence resources.

What Mycenae could have offered in return was help in structuring and enhancing the agricultural landscape, with an eye towards increasing overall production to the benefit of local residents and with the possibility that some of the surplus might ultimately be returned to the palace at Mycenae. In return, the palace may have received favorable access to Corinthian land and sea routes. Evidence of this can be found in the draining of the Nemea Valley, which has been noted as a sizeable endeavor that could only have been completed with assistance from an entity like the palace at Mycenae (Cherry and Davis 2001). Additionally, at Kalamianos, the organization and construction of the LBA terraces discovered within and outside the settlement of Kalamianos and at the nearby upland settlement of Stiri reflect knowledge and understanding of largescale construction projects and a broad perspective on the surrounding landscape that has so far been detected only at palace centers (Kvapil 2012); and, of all the Mycenaean palace centers, this sort of largescale program of landscape management aimed at increasing agricultural production appears particular to Mycenae. The high degree of control over land use suggests that Mycenae may also have had a hand in cultivation strategies at these sites to their own benefit. Indeed, the absence of evidence for landscape management around the cemetery site of Aidonia coupled with Aidonia’s decline throughout the LH period (Hachtmann 2013), at a time when settlements like Kalamianos were flourishing, might suggest that agricultural negotiations in this area were not always successful.

We further explore the degree of self-provisioning of subsistence practiced at Petsas House by focusing on shellfish and deer exploitation. Shellfish made up a considerable portion of the wild game diet at Mycenae. Of the shell remains, 24 species have been identified thus far in the Petsas House assemblage and shells make up 5.3% of the assemblage currently under study — the majority of which are the edible Arca noa variety. Interestingly, these shells are as highly represented in the assemblage as cattle, although they were not nearly as calorically valuable. Still, if brought to the site with fresh meat, shellfish represent a salient portion of the diet at Petsas House.

Agriculture and degrees of palatial control If we turn our attention back to Mycenae itself and the immediate hinterland, evidence points to many subsistence activities that were performed outside of, or on the outskirts of, palatial control; and yet these activities had broad resonance on the palace economy and those associated with it. Based on the predominance of Linear B records with a focus on counts of sheep and wheat (Killen 1964; Palmer 1992), palatial control over different sources of subsistence provisioning and redistribution was likely not equal among all of the species utilized, especially when wild taxa are also considered. We explore zooarchaeological, archaeobotanical, and textual evidence to illuminate variation in the degree of control practiced by palatial authorities over the plants and animals to better understand the economies which functioned at Mycenae.

This preliminary observation sheds light on the marine component of the diet available inland at this time, possibly as a supplementary resource used for independent provisioning of Petsas House. Alternatively, it may reflect that Petsas House had access to an organized system that mobilized marine resources and made them available, such as a market. 128

Lynne A. Kvapil et al.: Beyond City and Country at Mycenae

Access to mobilized marine resources also may denote a similar diet at Petsas House to earlier elites from the Grave Circle A and B, as isotopic studies indicate bone collagen from some of the adult burials to be enriched in nitrogen, suggesting a greater contribution of marine resources to their general subsistence diet (Richards and Hedges 2008).

immediately outside of the citadel. Yet, use of these specific wild taxa may also hint at the potential high status of the Petsas House occupants, given that deer and marine foods have been connected to elite Mycenaean hunting activities and diets. This view of the animal economy from Mycenae suggests that some supplementary animal acquisition was practiced outside of the palace, although possibly due to privileged status at the house. Even so, this supports the supposition that not all subsistence acquisition was under control of the palace and therefore, cannot be explained through a traditional center-periphery redistributive model.

Rarer wild species used at Petsas House include the three types of deer most commonly found in the Aegean (Trantalidou and Messeti 2014). Red, fallow, and roe deer were identified as a rare part of the Petsas House assemblage (6‰ with higher applications (3035t/ha), pulses show much smaller increases in δ15N values (~2‰), and only under high or extremely high manure applications (>70 t/ha) (Treasure et al. 2015), a pattern also noted in traditionally managed pulse plots in Evvia, Greece (Fraser et al. 2011).

Stable isotope analysis Stable carbon isotopes Plant δ13C values reflect the water status of the plant during its growth period (e.g. Farquhar and Richards 1984). In C3 plants (trees, legumes and all cereals reported in this study), differences in δ13C values relate to the rate of passage of carbon dioxide (CO2) through stomata, which is regulated according to water availability (Ehleringer et al. eds 1993; Farquhar et al. 1989). As a result, C3 plant δ13C values are strongly correlated with environmental factors such as precipitation, plant water status and relative humidity (Aguilera et al. 2008; Araus et al. 1997, 1999; Ferrio et al. 2005; Fiorentino et al. 2012; Flohr et al. 2011; Fraser et al. 2013; Masi et al. 2014; Riehl et al. 2014; Wallace et al. 2013, 2015). Since plant δ13C values depend on atmospheric CO2 δ13C values, which have decreased over the last 10,000 years (Francey et al. 1999; Indermühle et al. 1999), plant δ13C values are routinely converted to Δ13C values to allow comparison with modern plant values (Farquhar et al. 1982):

There is a high degree of natural variability in plant isotopic ratios: grains from a single ear vary up to 2.0‰ in δ15N (Bogaard et al. 2007) and up to 0.7‰ in δ13C (Heaton et al. 2009), with smaller within-pod variability found for pulses (Fraser et al. 2013; Treasure 153

Country in the City et al. 2015). This variability is reduced when multiple seeds are homogenized. For δ15N values, homogenizing multiple batches of 10-15 cereal grains from the same field produces standard deviations ranging from 0.12 to 0.91‰ (Kanstrup et al. 2012; Nitsch et al. 2015), with manured fields tending to have higher variability (Kanstrup et al. 2012). The variability in δ13C values is typically half that of δ15N (Nitsch et al. 2015).

for consumption, although emmer concentration 1 contained significant amounts of chaff fragments, which may reflect storage in spikelet form, with grains enclosed by glumes. The evidence suggests that Storeroom P served as a larder for crops awaiting the final stages of food preparation. Since its storage capacity was insufficient to supply the long-term needs of the inhabitants of the Unexplored Mansion, bulk storage of staple goods presumably occurred elsewhere (Jones 1984).

Materials We analysed remains of carbonized cereal grains and pulse seeds from Evans’ 1968 excavations of Neolithic strata below the Central Court (Evans 1964: 136, 140), and LM II storage deposits in the Unexplored Mansion (Jones 1984; Popham 1984). Both of these assemblages resemble stored product concentrations.

Analytical methods Samples were pre-screened for contamination from the burial environment using fourier transform infrared spectroscopy (FTIR). Visual inspection under a microscope showed no signs of surface soil contamination, and lack of contamination was confirmed by FTIR, following reference contamination comparisons for humic acids, carbonates and nitrates (Vaiglova et al. 2014b). Since chemical and mechanical degradation can remove seed/grain material and alter δ15N ratios (Vaiglova et al. 2014b) these samples were not pre-treated.

The Neolithic material consists of three samples of freethreshing wheat grain (Triticum aestivum/durum) from Stratum X, now dated securely (by AMS-dating of freethreshing wheat grains) to the early 7th millennium cal BC (Douka et al. 2017; see also Evans 1968: 272), and one sample of lentil (Lens culinaris) from the same context, all originally studied by Hans Helbaek. The wheat grain was stored in two parts, one labelled I and the other II; we infer that this labelling relates to a first and second shipment of the grain material to Helbaek, mentioned in an unpublished report held in the archive of the British School at Athens (Sarpaki 2013).

Samples of 5 or 10 seeds/grains were homogenized in an agate mortar and pestle. The homogenized powders were weighed into tin capsules for IRMS analysis on a SerCon EA-GSL mass spectrometer, with δ13C and δ15N measured separately. An internal alanine standard was used to calculate raw isotopic ratios. For δ13C two-point normalization to the VPDB scale was obtained using four replicates each of IAEA-C6 and IAEA-C7, while for δ15N the standards were USGS40 and IAEA-N2. Reported measurement uncertainties are the calculated combined uncertainty of the raw measurement and reference standards, after Kragten (1994). The average measurement uncertainty for δ13C was 0.07‰, and 0.35‰ for δ15N. Details of the analytical conditions are reported in Tables 1-2. These calculations were performed using the statistical programming language R. (3.0.2). In calculating Δ13C values, the δ13C value of atmospheric CO2 was estimated using reference tables from the AIRCO2_LOESS system (Ferrio et al. 2005). The plant isotope results reported are corrected for the minor effect of charring on δ13C (by subtracting 0.11‰) and δ15N (by subtracting 0.31‰) (Nitsch et al. 2015), except where otherwise indicated.

The Late Bronze Age crop remains analysed here derive from Storeroom P of the Unexplored Mansion, which was destroyed by fire in LM II (Popham 1984). Crop remains were found in concentrations associated with their original storage jars, and each concentration was dominated by a particular cereal or pulse taxon (Jones 1984, 1992). Cereal species predominating in distinct concentrations included emmer (Triticum dicoccum) and hulled barley (Hordeum vulgare). The barley included twisted grains indicative of the six-row variety (H. vulgare var. hexastichon), but the additional presence of two-rowed barley (H. vulgare var. distichon) among the non-twisted grains could not be ruled out. Pulse species predominating in distinct concentrations included winged vetchling (Lathyrus ochrus) and Celtic bean (Vicia faba). Three hulled barley concentrations (sample numbers 4, 5 and 6) were found close together and may represent a single stored product associated with Pot 75 (Jones 1984). Similarly, two concentrations each of emmer (samples 1 and 2) and Celtic bean (samples 9 and 10) could represent products that were originally stored together but their association is uncertain due to a lack of spatial information for the individual samples. The near complete absence of weeds from the crop stores suggests that they were products cleaned

Results Full results of the stable carbon and nitrogen isotope analysis of seed samples from Knossos are reported in Table 3. The C:N ratios of the archaeobotanical cereal and pulse samples (see Table 3) resemble those of modern experimentally charred cereal grains and pulses, respectively (cf. Bogaard et al. 2013: Figure S1). 154

Erika K. Nitsch et al.: Farming Practice and Land Management at Knossos, Crete

Table 1: δ13C values of standards in each carbon isotope run Alanine

C6

C6sd

C7

C7sd

1

-26.83

-10.05

0.37

-31.42

0.44

3

-26.91

-10.18

0.06

-31.7

0.13

2

-26.74

CAFF

CAFFsd

-34.66

N2

0.17

N2sd

-27.73

12.96

Runfile

sd1

150207A

0.54

150324B

0.06

150206B

0.09

Table 2: δ15N values of standards in each nitrogen isotope run Alanine 1

-2.13

3

-1.88

2

-1.57

CAFF

CAFFsd

-3.11

N2

0.06

N2sd

20.54

0.08

Runfile

sd1

150207A

0.48

150206B

0.26

150324B

0.06

d15N sd

d15N (-0.31 charring offset)

d15N (AIR)

d15N raw

%N

Runfile N

d13C sd

d13C (-0.11 charring offset)

d13C (VPDB)

d13C raw

Species

%C

Date

Runfile C

Area

N seeds

ID

Sample No.

Table 3: Results of stable isotope analysis of archaeobotanical samples from Knossos

CN

KNB001 1

Storeroom P LM II

Emmer

10

150207A 48.4 -24.2 -24.9

-25

0.64 150324B 3.4 3.1

3.2

2.9

0.08 16.6

KNB002 1

Storeroom P LM II

Emmer

10

150207A 47.9 -23.8 -24.4

-24.5

0.64 150324B 2.9 3.3

3.4

3.1

0.08 19.3

KNB003 1

Storeroom P LM II

Emmer

10

150207A 47.7 -24

-24.6

-24.7

0.64 150324B 3.4 2.2

2.3

2

0.08 16.2

KNB004 5

Storeroom P LM II

Hulled barley

5

150207A 50.3 -23.8 -24.4

-24.5

0.64 150324B 2.6 4.5

4.6

4.3

0.08 22.9

KNB005 5

Storeroom P LM II

Hulled barley

5

150207A 48.5 -24.2 -24.8

-24.9

0.64 150324B 2.8 5

5.1

4.8

0.08 20.4

KNB006 5

Storeroom P LM II

Hulled barley

5

150207A 49.2 -23.9 -24.5

-24.6

0.64 150324B 3.4 4.7

4.8

4.4

0.08 17.0

KNB007 5

Storeroom P LM II

Hulled barley

5

150207A 50.1 -24.6 -25.2

-25.3

0.64 150324B 2.6 4

4.1

3.8

0.08 22.7

KNB008 7

Storeroom P LM II

Winged 5 vetchling

150207A 47.3 -22.6 -23.2

-23.3

0.63 150324B 6.7 1.3

1.4

1.1

0.08 8.2

KNB009 7

Storeroom P LM II

Winged 5 vetchling

150207A 49.5 -23

-23.6

-23.7

0.63 150324B 6.2 0.8

0.9

0.6

0.08 9.4

KNB010 7

Storeroom P LM II

Winged 5 vetchling

150207A 46.4 -22.5 -23.1

-23.2

0.63 150324B 6.4 1

1.2

0.9

0.08 8.4

KNB011 7

Storeroom P LM II

Winged 5 vetchling

150207A 61.0 -22.8 -23.4

-23.5

0.63 150324B 5.5 1

1.1

0.8

0.08 13.0

KNB012 7

Storeroom P LM II

Winged 5 vetchling

150207A 47.6 -22.2 -22.8

-22.9

0.63 150324B 6.5 1.5

1.6

1.3

0.08 8.6

KNB013 7

Storeroom P LM II

Winged 5 vetchling

150207A 47.5 -24

-24.6

-24.7

0.64 150324B 7.0 1.4

1.5

1.2

0.08 7.9

KNB014 9

Storeroom P LM II

Celtic bean

5

150207A 47.2 -22.9 -23.4

-23.5

0.63 150324B 5.2 2.2

2.3

2

0.08 10.7

KNB015 9

Storeroom P LM II

Celtic bean

5

150207A 46.6 -22.3 -22.9

-23

0.63 150324B 4.7 2.1

2.2

1.9

0.08 11.6

KNB016 9

Storeroom P LM II

Celtic bean

5

150207A 46.4 -22

-22.5

-22.6

0.63 150324B 4.1 2.1

2.2

1.9

0.08 13.1

KNB017 10 Storeroom P LM II

Celtic bean

5

150207A 46.1 -23.2 -23.8

-23.9

0.63 150324B 4.3 2

2.1

1.8

0.08 12.4

155

d15N sd

d15N (-0.31 charring offset)

d15N (AIR)

d15N raw

%N

Runfile N

d13C sd

d13C (-0.11 charring offset)

d13C (VPDB)

d13C raw

Species

%C

Date

Runfile C

Area

N seeds

ID

Sample No.

Country in the City

CN

KNB018 10 Storeroom P LM II

Celtic bean

5

150207A 45.0 -21.2 -21.7

-21.8

0.62 150324B 4.5 2.2

2.3

2

0.08 11.7

KNB019 10 Storeroom P LM II

Celtic bean

5

150207A 45.4 -21.1 -21.7

-21.8

0.62 150324B 4.1 2.2

2.4

2

0.08 13.0

KNB020 4

Storeroom P LM II

Hulled barley

5

150207A 52.3 -23.6 -24.2

-24.3

0.63 150324B 2.9 5.3

5.4

5.1

0.08 20.8

KNB021 4

Storeroom P LM II

Hulled barley

5

150207A 46.4 -23.6 -24.2

-24.3

0.63 150324B 3.5 5

5

4.7

0.08 15.3

KNB022 4

Storeroom P LM II

Hulled barley

5

150207A 49.4 -24.4 -25

-25.1

0.64 150324B 2.7 5

5

4.7

0.08 21.2

KNB023 4

Storeroom P LM II

Hulled barley

5

150207A 49.0 -23.7 -24.3

-24.4

0.63 150324B 2.6 3.8

3.9

3.6

0.08 21.7

KNB024 2

Storeroom P LM II

Emmer

5

150207A 42.5 -23.8 -24.4

-24.5

0.64 150324B 3.4 5

5.1

4.8

0.08 14.7

KNB025 6a Storeroom P LM II

Hulled barley

6

150207A 47.4 -23.2 -23.8

-23.9

0.63 150324B 5.5 5.7

5.8

5.5

0.08 10.1

KNB026 6b Storeroom P LM II

Hulled barley

6

150207A 45.2 -24.5 -25.1

-25.2

0.64 150324B 3.6 4.2

4.2

3.9

0.08 14.8

KNN001

Trench AC, Stratum X

FreeAceramic threshing 10 Neolithic wheat

150206B 49.7 -23.1 -23.4

-23.5

0.13 150324B 3.2 5.4

5.5

5.1

0.08 18.0

KNN002

Trench AC, Stratum X

FreeAceramic threshing 10 Neolithic wheat

150206B 49.2 -22.8 -23.2

-23.3

0.12 150324B 3.1 5.7

5.7

5.4

0.08 18.8

KNN003

Trench AC, Stratum X

Aceramic Lentil Neolithic

150207A 43.4 -22.7 -23.3

-23.4

0.63 150327B 5.9 1.6

1.8

1.5

0.24 8.6

KNN004

Trench AC, Stratum X

FreeAceramic threshing 10 Neolithic wheat

150207A 48.5 -22

-22.7

0.63 150327B 2.9 5.9

6.1

5.7

0.23 19.2

10

-22.6

(Fraser et al. 2011) but are comparable to those observed in traditional manuring regimes of high labourintensity, as in Asturias, Spain (Bogaard et al. 2016; Fraser et al. 2011). Similar δ15N as well as Δ13C values for free-threshing wheat grain samples were observed at Middle-Late Neolithic Kouphovouno, near Sparta (Vaiglova et al. 2014a), where free-threshing wheat appears to have been grown under more intensively manured conditions than hulled (probably two-row) barley.

1. Neolithic material Δ13C values of the three free-threshing wheat samples range from 16.7 to 17.6‰, while the single lentil sample has a value of 17.4‰ (Figure 1). These values are similar to modern plants grown in moderately well watered conditions (Wallace et al. 2013). δ15N values for the three free-threshing wheat samples range from 5.1 to 5.7‰. The variability observed in Δ13C and δ15N values for freethreshing wheat is consistent with a single growing condition (Bogaard et al. 2016; Nitsch et al. 2015). The single lentil sample has a δ15N value of 1.5‰ (Figure 2). The lower δ15N value of the lentil sample is typical of nitrogen-fixers, deriving N from atmospheric N2 with a value of 0‰, but elevation above this value suggests intensive manuring (Fraser et al. 2011; Treasure et al. 2015). The much higher δ15N values of the three freethreshing wheat samples reflect the assimilation of nitrogen with a higher δ15N value, possibly originating from manure. These values are not as high as the most intensively manured cereals in modern experiments

2. Bronze Age material Δ13C values of pulses at the Unexplored Mansion range from 15.7 to 18.8‰, while cereal Δ13C values are generally higher, ranging from 18 to 19.4‰ (Figures 1 and 3). It is striking that the Knossian Bronze Age emmer appears very well watered, while the barley is moderately to well watered (Figure 1). For δ15N, barley values tend to be higher than those of other species, ranging from 3.6 to 5.5‰. The three emmer samples 156

Erika K. Nitsch et al.: Farming Practice and Land Management at Knossos, Crete

Figure : Δ13C values of wheats, barley and cereals at Neolithic and Bronze Age Aegean sites. Data from Kouphovouno reported by Vaiglova et al. (2014). Data from Archontiko and Thessaloniki Toumba are from Nitsch et al. (2017). Dashed horizontal lines represent ‘well watered’ (upper) and ‘moderately watered’ (lower) thresholds based on studies of modern crops in different agronomic conditions (Wallace et al. 2013). The reference lines for barley assume a mixture of barleys comparable to modern two- and six-row varieties.

is consistent with all samples being grown in similar conditions (Bogaard et al. 2016; Nitsch et al. 2015). For emmer, the variability among the three subsamples from sample 1 (average δ15N: 2.7 ± 0.6‰, average Δ13C: 18.9 ± 0.2‰) is also consistent with a single set of growing conditions, while the much higher δ15N value of the single subsample from sample 2 (4.8‰) suggests that this crop was grown in soil with a higher δ15N value, potentially due to organic matter in the form of animal manure. This contrast between the emmer in samples 1 and 2 coincides with a possible spatial separation within Storeroom P, sample 1 deriving from pot 54 and sample 2 from anywhere in the southwest part of the room (Jones 1984).

from sample 1 have much lower δ15N values, between 2.0 and 3.1‰, while the single emmer sample from sample 2 is higher, with a δ15N value of 4.8‰, similar to the barley samples (Figure 3b). For pulses, the δ15N values of winged vetchling are relatively low (ranging from 0.6 to 1.3‰) compared to those of Celtic bean (1.8 to 2.0‰), but both are elevated above 0‰ and consistent with intensive manuring (Fraser et al. 2011; Treasure et al. 2015). For each species, most of the measured values may be subsamples of material from the same deposit (Figure 3b). For barley, subsamples come from sample 4 (n=4), sample 5 (n=4), sample 6a (n=1) and sample 6b (n=1). There are no differences in δ15N or Δ13C values among barley samples, all of which may derive from pot 75 (Jones 1984). The average δ15N value of all barley samples is 4.8 ± 0.6‰, and for Δ13C is 18.7 ± 0.5. This variability

Turning to the pulses (Figure 3b), isotopic analysis of subsamples from concentrations of winged vetchling (sample 7, 6 subsamples) and Celtic bean (sample 9, 3 157

Country in the City

Figure 2: δ15N values of wheats, barley and cereals at Neolithic and Bronze Age Aegean sites. Data from Kouphovouno reported by Vaiglova et al. (2014). Data from Archontiko and Thessaloniki Toumba are from Nitsch et al. (2017).

subsamples; sample 10, 3 subsamples) are also largely consistent with the hypothesis that each concentration derived from the same growing conditions. Assessment of variation in pulse seed δ13C and δ15N values (data from Fraser et al. 2011; Wallace et al. 2013) shows that experimentally grown pulse seed samples from the same treatment have similarly low variability, with an average standard deviation of 0.5‰ for δ13C and 0.25‰ for δ15N.

and straight grains derived from isotopically different growing conditions (e.g. if the straight grains derived from two-row barley, growing in different conditions to six-row barley), larger differences would have been expected. These results are therefore consistent with Jones’ original inference that six-row barley is represented (Jones 1984).

Finally, variability was also assessed between twisted and straight grains of hulled barley, assumed to derive from the same, six-row hulled barley crop, in which each rachis node produces both straight (central) and twisted (lateral) grains. In order to test this assumption two subsamples of twisted grains and two of straight grains were analysed from each of the barley samples, 4 and 5. Figure 4 shows that, for each set of twisted grain values, the straight grain δ13C and δ15N values from the same sample lie within the 95% range. If the twisted

Isotopic evidence for cereal cultivation practices

Discussion

Estimates of modern annual precipitation around Knossos vary from 483mm (Hellenic National Meteorological Service, average for AD 1955-1996, http://www.hnms.gr/hnms/greek/climatology/ climatology_region_diagrams_html?dr_city=Heraklion) to 589mm (Hijmans et al. 2005). Recent palaeoclimatic reconstruction of past annual precipitation across Europe suggests that Bronze Age rainfall in Crete did 158

Erika K. Nitsch et al.: Farming Practice and Land Management at Knossos, Crete

Figure 3: A) Δ13C and δ15N values of crops from the Late Minoan II destruction of storeroom P of the Unexplored Mansion at Knossos, showing (B) subsamples grouped by associated pot (as described by Jones 1984). C) Δ13C and δ15N values from modern pulses grown under a variety of manuring and irrigation regimes in Evvia, Greece (Δ13C originally reported by Wallace et al. 2013; δ15N values originally reported in Fraser et al. 2011). D) Δ13C and δ15N values of pulses from storeroom P at Knossos compared with the 50% confidence interval ellipse of modern pulse values from Evvia.

not differ greatly from recent conditions and was if anything higher (Mauri et al. 2015: Figure S3). Under these environmental conditions unmanured cereals would be expected to have values similar to those in temperate Europe (Bogaard et al. 2013; cf. Hartman and Danin 2010; Styring et al. 2016a). Aridity is therefore unlikely to be the primary driver of elevated soil δ15N at Knossos; lower δ15N ratios, closer to zero, have been observed for archaeological cereals in other parts of the Aegean with similar rainfall levels (Kouphovouno: Vaiglova et al. 2014a; Archontiko: Nitsch et al. 2017; see Figure 2).

Elevated δ15N values of Neolithic cereals and pulses are consistent with intensive levels of manuring, and suggest manuring/middening as a component of high-labour investment in arable land at Aceramic Neolithic Knossos. Comparably high δ15N values in Neolithic free-threshing wheat have been noted at Middle-Late Neolithic Kouphovouno near Sparta (Vaiglova et al. 2014a) as well as for cereals elsewhere in Neolithic Europe (Bogaard et al. 2013; Styring et al. 2016b), consistent with an intensive cultivation strategy. At 6th millennium BC Kouphovouno and at late fourth-third millennium BC Sipplingen, on Lake 159

Country in the City The δ15N values of Late Bronze Age cereals from the Unexplored Mansion are mostly lower than those of Neolithic free-threshing wheat at Knossos (and Kouphovouno); the δ15N values of the Bronze Age emmer wheat are significantly lower than those of Knossian Neolithic free-threshing wheat, while Δ13C values are significantly higher (Pillai’s Trace = 0.88, F = 14.5, df = (4,5), p = 0.015). Of all the cereals recovered from Storeroom P at the Unexplored Mansion, emmer is most likely to correspond to the cereal type in whose production palatial elites at Knossos intervened according to the Linear B evidence (Halstead 1995c). Other LBA cereal δ15N values, especially of barley, are consistent with moderate manuring. The (six-row) hulled barley δ15N values from Knossos are notably higher than those for (two-row) hulled barley at Neolithic Kouphovouno; on the other hand, the Knossos barley δ15N values are moderate to low compared with (mixed two- and six-row) hulled barley values from Bronze Age sites in the northern Aegean (Figure 2; Nitsch et al. 2017). Moderate manuring of the Knossos barley is inconsistent with radically extensive, lowlabour input cultivation, but this is in agreement with the Linear B evidence, which lacks indications of elite involvement in production of the cereal most plausibly identified as barley (Halstead 1995c).

Figure 4: Comparison of twisted and straight barley grains from the same deposit. For each pair of twisted grain samples, the lines show the estimated 67% (solid) and 95% (dashed) range of expected Δ13C and δ15N values for the population of seeds grown under similar growing conditions, based on a 1sigma range of ± 0.5‰ for Δ13C and ± 1.0 for δ15N. For each pair of twisted grains, the values of the two straight grain samples from the same deposit lie within the 95% range. If the twisted and straight grains derived from isotopically different growing conditions, larger differences would have been detected.

The stable carbon isotope data for the Knossos cereals add a further dimension to the discussion of crop growing conditions and land management. The area of Knossos could have sustained rainfed cultivation, although local variation in topography and soil type would have meant some areas would be more prone to drought than others (Halstead 2008). As in modern Crete, completing tillage and sowing as early as possible would allow crops to ripen before the summer drought (Halstead 2014: 25; Isaakidou 2008).

Constance, free-threshing wheat grew in significantly more 15N-enriched conditions than barley (Styring et al. 2016b; Vaiglova et al. 2014a). Further work on the Neolithic crop assemblage from Knossos is needed to explore inter- as well as intra-crop variation in growing conditions. It is clear, however, that factors such as cultural preference and crop variety or landrace could influence strategic manuring of particular crops; in the northern Aegean during the Bronze Age, hulled barley rather than free-threshing wheat was manured preferentially (Figure 2; Nitsch et al. 2017).

Δ13C values of the Neolithic wheat samples are consistent with moderately to well watered conditions, similar to results from elsewhere in the Aegean (Figure 1; Nitsch et al. 2017; Vaiglova et al. 2014a; Wallace et al. 2015). The relatively high Δ13C values of barley and especially emmer from the Unexplored Mansion are striking, given that rainfall around Knossos, though variable, would not necessarily be higher than in the northern Aegean. The implication is that these cereals had privileged access to very well watered conditions. The Δ13C values could in theory reflect irrigation (cf. Müller Celka et al. 2014; Vokotopoulos et al. 2014); the fact that Late Bronze Age emmer appears better watered than earlier ripening, more drought-tolerant barley (Figure 1) is at least consistent with a deliberate water management strategy. Comparably high ∆13C values were determined in wheat and barley samples from Bronze Age Abu Salabikh in southern Iraq, where artificial watering is obligatory (Wallace et al. 2015; Figure 2).

Intensive ‘horticultural’ production of cereals and pulses featuring manual tillage would tend to yield only marginal surpluses, with labour as the key limiting factor (Halstead 2014: 47, 53, 248). The desire for more surplus may have driven political expansion (Whitelaw this volume), and ultimately established the system of extensive production under palatial control that is reflected in Linear B texts (Halstead 1992b, 1999a, 1999b; Killen 1993). The cost of maintaining specialized draught animals suggests that extensive cultivation was likely limited to areas under palatial control managed for the production of relatively stresstolerant cereals, and used to support a system of elites and craft specialists (Halstead 1999a, 1999b, 2014; Killen 1993, 1998). 160

Erika K. Nitsch et al.: Farming Practice and Land Management at Knossos, Crete

Δ13C values of the Knossian pulse samples are consistent with moderately to well watered conditions, similar to results from elsewhere in the Aegean (Figure 1; Nitsch et al. in press; Vaiglova et al. 2014a; Wallace et al. 2015). In comparison with modern irrigated pulses in Evvia, the Δ13C values of the Unexplored Mansion pulse crops suggest that any artificial watering was variable, which would be consistent with hand-watering of pulses (cf. Wallace et al. 2013). Pulse crops in the Bronze Age northern Aegean were similarly variable in their water status but often better watered than the pulses at the Unexplored Mansion, in contrast to the cereals (Figure 1). Modern experiments have shown that well watered pulses tend to have higher Δ13C values than wheat grown under the same conditions (Wallace et al. 2013). The fact that the Unexplored Mansion pulses tend to have lower Δ13C values than the emmer wheat, and also appear to be more intensively manured, therefore suggests that these crops were grown under different regimes. While the Neolithic cereals and lentil sample from Knossos, like the northern Aegean pulses and cereals, were grown in comparable conditions in terms of both carbon and nitrogen isotope values, and so were potentially grown together or in rotation, the Unexplored Mansion pulses were likely grown separately to the wheat, and perhaps also the barley.

In combination, the nitrogen and carbon isotope data for cereals from Knossos suggest that the Neolithic wheat was more intensively manured than the Bronze Age cereals, which instead had a particularly favourable water status. As emmer is considered the most likely candidate for direct palatial production in the Linear B texts (Halstead 1995c), it is notable that its growing conditions reflect mostly low to no manure inputs but very high water status relative to that inferred from isotopic results from all Bronze Age wheat analysed from the northern Aegean (Nitsch et al. 2017). Strategic sowing of wheat in well watered soils is a relatively lowinput way of ensuring adequate yields in a potentially extensive system (cf. Wallace et al. 2015). Isotopic evidence for pulse cultivation practices Intensive manuring of pulses has been observed in traditional agricultural regimes in Evvia, Greece (Jones et al. 1999) and Asturias, Spain (Charles et al. 2002). Manuring rates in Evvia were particularly intensive, resulting in the formation of ‘dung-soil’ (koprokhoma) (Jones et al. 1999). It has been shown experimentally that extremely high rates of manuring (>70t/ha) can increase broad bean δ15N values by up to 2‰ (Treasure et al. 2015), while values in modern Evvia gardens were up to 4‰ or higher (Figure 3c). δ15N values of the very heavily manured pulses are significantly higher (+1.9‰) than those of unmanured pulses in the Evvia study (t(13)4.476, p = 0.001; Figure 3c).

Variability in crop growing conditions Models of agricultural catchment area and settlement size suggest that by the Late Minoan I period at most 60% of Knossos’ population could have been sustained on food grown within a two-hour walk of the settlement (Whitelaw this volume). The implication is that a significant amount of food would have been brought into the city from further away. The storage vessels of the Unexplored Mansion are characterized by low storage capacity but were highly transportable (Christakis 2008: 76). Characterisations of possible supply systems range from palatially controlled redistribution, to market supply or provisioning by individual family rural estates (Bennet and Halstead 2014; Whitelaw this volume).

Late Bronze Age pulses from the Unexplored Mansion have higher δ15N values than unmanured pulses, as does the single lentil sample from Neolithic Knossos, and are comparable to heavily manured examples (Fraser et al. 2011; Treasure et al. 2015). In pre-mechanised Greek farming, pulses required a large amount of labour to harvest (by uprooting rather than cutting), and were most often grown for household consumption, in part also due to the difficulty of transporting them without losing seeds from drying pods (Halstead 2014: 70, 789, 103; Jones et al. 1999). Pulses are not referenced in Linear B texts (Halstead 1992), but their abundance at southern Aegean sites (Halstead 1994: 204-5, table 1; Livarda and Kotzamani 2014; Sarpaki 1992), including the elite Unexplored Mansion and the palace itself (Evans 1928: 54, 1935: 595, 621), suggests that they were an important component of human diet across the social spectrum. They may also have served an important agricultural role in maintaining soil fertility in rotation with other crops (Sarpaki 1992), though in the Unexplored Mansion assemblage it is clear that the heavily manured pulses could not have been grown together, or in rotation with, the moderately to unmanured cereals (Figure 2). It is also apparent that winged vetchling was less intensively manured than Celtic bean, perhaps reflecting differential status and/ or response to manuring.

Large-scale redistribution would be expected to mix harvests from different fields, different growing conditions, and quite possibly different isotope ratios. The natural variability in δ15N values for individual cereal grains grown together under the same conditions (Bogaard et al. 2007) is low compared to the potential variation due to differences in manuring: measurements of 10 homogenized grains from the same field vary by ± 0.12 to 0.91‰ (1 standard deviation, Kanstrup et al. 2012; Nitsch et al. 2015). Halving the batch size (N seeds = 5) increases the expected variability by a factor of √2 (Nitsch et al. 2015). The measured standard deviations in δ15N values for three to four different batches of five barley grains from samples 4 161

Country in the City

Figure 5: Probability density distribution of simulated standard deviations in δ15N values of four measurements of homogenized aliquots of 5 grains (A) or 10 grains (B) based on 1000 simulated 50/50 mixture of two different isotope ratios (differences ranging from 0 to 6‰). Also shown are the actual measured δ15N standard deviation of three to four measurements of grains from samples 4, 5 and 1.

and 5 and ten grains from sample 1 (ranging from 0.41 to 0.63‰) from the Unexplored Mansion are consistent with random variability under the same or similar growing conditions. This supports the hypothesis that the grains in these samples were grown, processed, transported and stored as a single unit. On the other hand, a simulation of two equal mixtures of cereal grains with different δ15N values illustrates that, with the amount of material available for measurement here, it would be difficult to detect slightly greater variation due to a combination of growing conditions (Figure 5). Nevertheless, for mixtures where the difference in isotope ratios is 5‰ or greater, the range of simulated standard deviations (for four measurements of either five- or ten- grain homogenized aliquots) is generally larger than the standard deviations actually measured. Applying a similar simulation to the pulse samples (Figure 6) suggests that concentrations derive from relatively homogeneous growing conditions, a possible exception being sample 10 (Celtic bean), which has more highly variable δ13C values, suggesting a mixture of growing conditions with less similar δ13C values. The same simulation approach to the Neolithic freethreshing wheat δ15N values suggests that these also derived from relatively homogeneous growing conditions (Figure 7).

the same field, the possibility remains that the grains measured represent mixtures of crops grown under similar conditions. Nevertheless, our results suggest that it is unlikely that the Unexplored Mansion cereal and pulse stores derive from large-scale redistribution involving considerable variation in growing conditions and land use history. Conclusions Agriculture was the economic basis of Aegean Bronze Age urbanism, featuring ‘agro-towns’ with a majority of the population directly engaged in crop production (e.g. Whitelaw this volume). Isotopic measurements of Bronze Age crops from the Unexplored Mansion at Knossos suggest distinctive agricultural strategies for the cereals and pulses. The emmer wheat and hulled six-row barley were grown under conditions of low to moderate manuring and good water supply. The pulses were intensively manured and variably watered. Pulse cultivation strategies at Knossos resemble those of recent small-scale cultivators in Evvia (Figure 3cd), and are comparable to those of the Bronze Age northern Aegean (Figures 1-2; Nitsch et al. 2017). These intensive strategies for pulses provide a clear contrast with extensive cereal cultivation recorded in Linear B texts. Whether or not the Unexplored Mansion cereals represent extensive, palace-mediated production is ambiguous: while they appear less intensively manured than the few Neolithic cereal samples from Knossos analysed here, or indeed some north Aegean Bronze Age cereal assemblages (Nitsch et al. 2017), their high water

In sum, while the relative homogeneity of the isotope results within individual concentrations is consistent with the hypothesis that each of the crops subsampled from the Unexplored Mansion was grown under the same set of conditions and potentially derive from 162

Erika K. Nitsch et al.: Farming Practice and Land Management at Knossos, Crete

Figure 6: Probability density distribution of simulated standard deviations of four δ15N measurements (A) and four δ13C measurements (B) of homogenized aliquots of 5 pulse seeds based on 1000 simulated 50/50 mixture of two different isotope ratios (differences ranging from 0 to 6‰). Also shown are the actual measured standard deviations of three to seven measurements of pulse seeds from samples 7, 9 and 10

Figure 7: Probability density distribution of simulated standard deviations in δ15N values of three measurements of homogenized aliquots of 10 grains based on 1000 simulated 50/50 mixture of two different isotope ratios (differences ranging from 0 to 6‰). Also shown is the actual measured δ15N standard deviation of three measurements of Neolithic free-threshing wheat grains.

163

Country in the City References

status could reflect privileged access to well watered land in the Knossos valley (cf. Whitelaw 2011) or elsewhere. In any case, it is very unlikely that the cereals and pulses in the storeroom of the Unexplored Mansion derived from an integrated system of cereal-pulse rotation. Perhaps this elite household was supported by intensively cultivated pulses grown on ‘allotment’ plots within or on the margins of the urban sprawl of Late Bronze Age Knossos, while the cereals were grown in the valley or other well watered areas. Moreover, the relative homogeneity of the isotope ratios within individual storage deposits of the Unexplored Mansion suggests that crops from specific land holdings could well be represented, and excludes mixtures of crops from wide-ranging growing conditions and land use histories, as might be expected if they derived from large-scale palatial redistribution.

Aguilera, M., Araus, J. L., Voltas, J., Rodríguez-Ariza, M. O., Molina, F., Rovira, N., Buxó, R. and Ferrio, J. P. 2008. Stable carbon and nitrogen isotopes and quality traits of fossil cereal grains provide clues on sustainability at the beginnings of Mediterranean agriculture. Rapid Communications in Mass Spectrometry 22(11): 1653–1663. Andrews, M., Lea, P. J., Raven, J. A. and Azevedo, R. A. 2009. Nitrogen use efficiency. 3. Nitrogen fixation: genes and costs. Annals of Applied Biology 155(1): 1–13. Anyia, A. O., Slaski, J. J., Nyachiro, J. M., Archambault, D. J. and Juskiw, P. 2007. Relationship of carbon isotope discrimination to water use efficiency and productivity of barley under field and greenhouse conditions. Journal of Agronomy and Crop Science 193(5): 313–323. Araus, J. L., Febrero, A., Buxó, R., Camalich, M. D., Martín, D., Molina, F., Rodriguez-Ariza, M. O. and Romagosa, I. 1997. Changes in carbon isotope discrimination in grain cereals from different regions of the western Mediterranean Basin during the past seven millennia. Palaeoenvironmental evidence of a differential change in aridity during the late Holocene. Global Change Biology 3: 107–118. Araus, J., Febrero, A., Catala, M., Molist, M., Voltas, J. and Romagosa, I. 1999. Crop water availability in early agriculture: evidence from carbon isotope discrimination of seeds from a tenth millennium BP site on the Euphrates. Global Change Biology 5(2): 201–212. Bendall, L. 2007. Economics of Religion in the Mycenaean World. Resources Dedicated to Religion in the Mycenaean Palace Economy. Oxford University School of Archaeology Monograph 67, Oxford. Bennet, J. 2008. ‘Now you see it; now you don’t!’: the disappearance of the Linear A script on Crete. In J. Baines, S. Houston and J. Bennet (eds), The Disappearance of Writing Systems: 1-29. London, Equinox. Bennet, J. and Halstead, P. 2014. O-no! Writing and Righting Redistribution. In D. Nakassis, J. Gulizio, and S. James (eds), KE-RA-ME-JA: Studies Presented to Cynthia W. Shelmerdine: 271-282 Philadelphia, INSTAP Academic Press. Bogaard, A., Heaton, T. H. E., Poulton, P. and Merbach, I. 2007. The impact of manuring on nitrogen isotope ratios in cereals: archaeological implications for reconstruction of diet and crop management practices. Journal of Archaeological Science: 34: 335– 343. Bogaard, A., Fraser, R., Heaton, T. H. E., Wallace, M., Vaiglova, P., Charles, M., Jones, G., Evershed, R. P., Styring, A. K., Andersen, N. H., Arbogast, R-M., Bartosiewicz, L., Gardeisen, A., Kanstrup, M., Maier, U., Marinova, E., Ninov, L., Schäfer, M. and Stephan, E. 2013. Crop manuring and intensive land

A model of extensification based on Linear B texts would suggest a fundamental shift in agricultural practices underlying the urban expansion of Knossos in the Bronze Age. Yet isotopic comparison of Neolithic and Bronze Age crops from Knossos, and comparisons with emerging Neolithic-Bronze Age production elsewhere in Greece, suggests a more complex scenario in which an elite Late Bronze Age household was provisioned at least in part through intensive management of pulse crops that are unaccounted for in Linear B texts. Nevertheless, ecological contrasts between cereal and pulse growing regimes in this elite Late Bronze Age household suggest an ‘unravelling’ of the intensive subsistence-oriented farming that had emerged in the Neolithic (Halstead 1981, 2000), as reflected in emerging crop isotope results from Kouphovouno (Vaiglova et al. 2014a) and now Knossos, and persisted in diversified form in Bronze Age northern Greece (Nitsch et al. 2017). A disconnect between Late Bronze Age cereal and pulse growing strategies at Knossos would also be consistent with a special status of cereal growing, and perhaps even restricted rights to grow cereals and/or have access to well watered land in the Final Palatial period (Eleni Hatzaki pers. comm.). It is worth noting in this connection that the far more modest Late Bronze Age household recently excavated on the Lower Gypsades hill (Morgan 2015: 34-5), south of the palace at Knossos, has yielded exclusive evidence for pulse crop storage, perhaps reflecting the importance of intensive pulse cultivation across the social spectrum of the urban landscape. Acknowledgements Stable isotope analysis of crop remains in this study formed part of the European Research Council-funded AGRICURB project (grant no. 312785; PI Bogaard). We thank Charlotte Diffey, Elizabeth Stroud, Amy Styring and Todd Whitelaw for helpful comments on the manuscript. 164

Erika K. Nitsch et al.: Farming Practice and Land Management at Knossos, Crete

management by Europe’s first farmers. Proceedings of the National Academy of Sciences 110: 12589–12594. Bogaard, A., Hodgson, J., Nitsch, E., Jones, G., Styring, A., Diffey, C., Pouncett, J., Herbig, C., Charles, M., Ertuğ, F., Tugay, O. Filipovic, D. and Fraser, R. 2016. Combining functional weed ecology and crop stable isotope ratios to identify cultivation intensity: a comparison of cereal production regimes in Haute Provence, France and Asturias, Spain. Vegetation History and Archaeobotany 25: 57–73. Bogaard, A., Styring, A., Ater, M., Hmimsa, Y., Green, L., Stroud, E., Whitlam, J., Diffey, C., Nitsch, E., Charles, M., Jones, G. and Hodgson, J. 2018. From traditional farming in Morocco to early urban agroecology in northern Mesopotamia:  combining presentday arable weed surveys and crop ‘isoscapes’ to reconstruct past agrosystems in (semi-)arid regions. Environmental Archaeology. 23: 303-322. Bol, R., Eriksen, J., Smith, P., Garnett, M. H., Coleman, K. and Christensen, B. T. 2005. The natural abundance of 13C, 15N, 34S and 14C in archived (1923-2000) plant and soil samples from the Askov long-term experiments on animal manure and mineral fertilizer. Rapid Communications in Mass Spectrometry 19(22): 3216–3226. Cadogan, G., Hatzaki, E. and Vasilakis, A. (eds), 2007. Knossos: Palace, City, State. London, British School at Athens. Charles, M., Bogaard, A., Jones, G., Hodgson, J. and Halstead, P. 2002. Towards the archaeobotanical identification of intensive cereal cultivation: present-day ecological investigation in the mountains of Asturias, northwest Spain. Vegetation History and Archaeobotany 11: 133–142. Christakis, K. S. 2008. The politics of storage: storage and sociopolitical complexity in neopalatial Crete (Vol. 25). Philadelphia: INSTAP Academic Press. Douka, K., Efstratiou, N., Hald, M.M., Henriksen, P.S. and Karetsou, A. 2017. Dating Knossos and the arrival of the earliest Neolithic in the southern Aegean. Antiquity 91: 304-321. Ehleringer, J. R., Hall, A. E. and Farquhar, G. D. (eds), 1993. Stable Isotopes and Plant Carbon/Water Relations. San Diego Academic Press. Evans, A.J. 1928. The Palace of Minos at Knossos: A Comparative Account of the Successive Stages of the Early Cretan Civilization as Illustrated by the Discoveries at Knossos. Vol. 2. London, Macmillan. Evans, A.J. 1935. The Palace of Minos at Knossos: A Comparative Account of the Successive Stages of the Early Cretan Civilization as Illustrated by the Discoveries at Knossos. Vol. 4. London, Macmillan. Evans, J.D. 1964. Excavations in the Neolithic settlement at Knossos, 1957-60. Part I. Annual of the British School at Athens 59: 132-238. Evans, J.D. 1968. Knossos Neolithic, part II: summary and conclusions. Annual of the British School at Athens 63: 267-276.

Farquhar, G. D., O’Leary, M. H. and Berry, J. A. 1982. On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Australian Journal of Plant Physiology 9: 121– 137. Farquhar, G. D. and Richards, R. A. 1984. Isotopic composition of plant carbon correlates with wateruse efficiency of wheat genotypes. Australian Journal of Plant Physiology 11: 539-552. Farquhar, G. D., Ehleringer, J. R., and Hubick, K. T. 1989. Carbon isotope discrimination and photosynthesis. Annual Review of Plant Physiology and Plant Molecular Biology 40: 503–537. Ferrio, J. P., Araus, J. L., Buxó, R., Voltas, J. and Bort, J. 2005. Water management practices and climate in ancient agriculture: Inferences from the stable isotope composition of archaeobotanical remains. Vegetation History and Archaeobotany 14: 510–517. Fiorentino, G., Caracuta, V., Casiello, G., Longobardi, F. and Sacco, A. (2012). Studying ancient crop provenance: implications from δ13C and δ15N values of charred barley in a Middle Bronze Age silo at Ebla (NW Syria). Rapid Communications in Mass Spectrometry 26(3): 327–335. Flohr, P., Müldner, G. and Jenkins, E. 2011. Carbon stable isotope analysis of cereal remains as a way to reconstruct water availability: Preliminary results. Water History 3: 121–144. Francey, R., Allison, C., Etheridge, D., Trudinger, C., Enting, I., Leuenberger, M., Langenfelds, R. L., Michel, E. and Steele, L. P. 1999. A 1000-year high precision record of δ13C in atmospheric CO₂. Tellus 51 B: 170–193. Fraser, R., Bogaard, A., Heaton, T. H. E., Charles, M., Jones, G., Christensen, B. T., Halstead, P., Merbach, I., Poulton, P., Sparkes, D. and Styring, A. K. 2011. Manuring and stable nitrogen isotope ratios in cereals and pulses: towards a new archaeobotanical approach to the inference of land use and dietary practices. Journal of Archaeological Science 38: 2790– 2804. Fraser, R. A., Bogaard, A., Charles, M., Styring, A. K., Wallace, M., Jones, G., Ditchfield, P. and Heaton, T. H. E. 2013. Assessing natural variation and the effects of charring, burial and pre-treatment on the stable carbon and nitrogen isotope values of archaeobotanical cereals and pulses. Journal of Archaeological Science 40: 4754–4766. Halstead, P. 1981. Counting sheep in Neolithic and Bronze age Greece. In I. Hodder, G. Isaac and Hammond, N. (eds), Pattern of the Past: Studies in Honour of David Clarke: 307–39. Cambridge, Cambridge University Press. Halstead, P. 1987. Traditional and Ancient Rural Economy in Mediterranean Europe: plus ça change? Journal of Hellenic Studies 107: 77-87. Halstead, P. 1989. The economy has a normal surplus: economic stability and social change among early 165

Country in the City farming communities of Thessaly, Greece. In P. Halstead and J. O’Shea (eds), Bad Year Economics: Cultural Responses to Risk and Uncertainty: 68-80. Cambridge, Cambridge University Press. Halstead, P. 1992. Agriculture in the Bronze Age Aegean: towards a model of palatial economy. In B. Wells (ed.), Agriculture in Ancient Greece: 105–117. Stockholm, Astrom Editions. Halstead, P. 1994. The North-South Divide: regional paths to complexity in prehistoric Greece. In C. Mathers and S. Stoddart (eds), Development and Decline in the Mediterranean Bronze Age: 195–219. Sheffield, J. R. Collis Publications. Halstead, P. 1995a. From sharing to hoarding: The Neolithic foundations of Aegean Bronze Age society? In R. Laffineur and W.-D. Niemeier (eds), Politeia: Society and State in the Aegean Bronze Age: 11– 22. Liège, Universitè de Liège, Aegaeum 12. Halstead, P. 1995b. Plough and power: The economic and social significance of cultivation with the oxdrawn ard in the Mediterranean. Bulletin on Sumerian Agriculture 8: 11-22. Halstead, P. 1995c. Late Bronze Age grain crops and Linear B ideograms *65, *120, and *121. Annual of the British School at Athens 90: 229-234. Halstead, P. 1998. Texts, Bones and Herders: Approaches to Animal Husbandry in Late Bronze Age Greece. In J. Bennet and J. Driessen (eds), A-NA-QO-TA. Studies Presented to J.T. Killen (Minos 33-34): 141-190. Salamanca, Ediciones Universidad de Salamanca. Halstead, P. 1999a. Towards a Model of Mycenaean Palatial Mobilization. In M. Galaty and W. Parkinson (eds), Rethinking Mycenaean Palaces: New Interpretations of an Old Idea (Monograph 31): 35-41. Los Angeles, The Cotsen Institute of Archaeology. Halstead, P. 1999b. Surplus and Share-croppers: The grain production strategies of Mycenaean Palaces. In P. Betancourt, V. Karageorghis, R. Laffineur, and W.-D. Niemeier (eds), MELETEMATA: Studies in Aegean Archaeology presented to Malcolm H. Wiener as he enters his 65th Year (Vol. II): 319-326. Liège, Université de Liège Halstead, P. 2000. Land use in postglacial Greece: Cultural causes and environmental effects. In P. Halstead and C. Frederik (eds), Landscape and Land Use in Postglacial Greece: 110-128. Sheffield, Sheffield Academic Press. Halstead, P. 2001. Mycenaean wheat, flax and sheep: palatial intervention in farming and its implications for rural society. In S. Voutsaki and J. Killen (eds), Economy and politics in the Mycenaean palace states: proceedings of a conference held on 1-3 July 1999 in the Faculty of Classics: 38-50. Cambridge, Cambridge Philological Society. Halstead, P. 2008. Between a rock and a hard place: Coping with marginal colonisation in the Later Neolithic and Early Bronze Age of Crete and the Aegean. In V. Isaakidou and P. Tomkins (eds),

Escaping the Labyrinth: The Cretan Neolithic in Context: 229–57 (Sheffield Studies in Aegean Archaeology 8). Oxford, Oxbow Books. Halstead, P. 2014. Two Oxen Ahead: Pre-Mechanized Farming in the Mediterranean. Cambridge, Wiley-Blackwell. Hartman, G. and Danin, A. 2010. Isotopic values of plants in relation to water availability in the Eastern Mediterranean region. Oecologia 162(4): 837–852. Heaton, T. H. E., Jones, G., Halstead, P. and Tsipropoulos, T. 2009. Variations in the 13C/12C ratios of modern wheat grain, and implications for interpreting data from Bronze Age Assiros Toumba, Greece. Journal of Archaeological Science 36: 2224 – 2233. Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G. and Jarvis, A. 2005 Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25(15): 1965–1978. Indermühle, A., Stocker, T. F., Joos, F., Fischer, H., Smith, H. J., Wahlen, M., Deck, B., Mastroianni, J., Tschumi, J., Blunier, T., Meyer, R. and Stauffer, B. 1999. Holocene carbon-cycle dynamics based on CO2 trapped in ice at Taylor Dome, Antarctica. Nature 398(6723): 121–126. Isaakidou, V. 2008. The fauna and economy of Neolithic Knossos’ revisited. Escaping the Labyrinth. In V. Isaakidou and P. Tomkins (eds), Escaping the Labyrinth: The Cretan Neolithic in Context: 90-114. (Sheffield Studies in Aegean Archaeology 8). Oxford, Oxbow Books. Isaakidou, V. and Tomkins, P. (eds), 2008. Escaping the Labyrinth. The Cretan Neolithic in Context. (Sheffield Studies in Aegean Archaeology 8). Oxford, Oxbow Books. Jones, G. 1984. The LM II Plant Remains, in M. R. Popham (ed.), The Minoan Unexplored Mansion, Knossos: 303306. British School at Athens supp. Volume 17, London. Jones, G. 1987: Agricultural practice in Greek prehistory. Annual of the British School at Athens 82: 115-23. Jones, G. 1988. The application of present-day cereal processing studies to charred archaeobotanical remains. Circaea 6: 91-6. Jones, G. 1992. Ancient and modern cultivation of Lathyrus ochrus (L.) DC in the Greek islands. Annual of the British School at Athens 87: 211-7. Jones, G., Bogaard, A., Halstead, P., Charles, M. and Smith, H. 1999. Identifying the intensity of crop husbandry practices on the basis of weed floras. Annual of the British School at Athens 94: 167–189. Kanstrup, M., Thomsen, I. K., Andersen, A. J., Bogaard, A., and Christensen, B. T. 2011. Abundance of 13C and 15N in emmer, spelt and naked barley grown on differently manured soils: towards a method for identifying past manuring practice. Rapid Communications in Mass Spectrometry 25(19): 2879– 2887. Kanstrup, M., Thomsen, I. K., Mikkelsen, P. H. and Christensen, B. T. 2012. Impact of charring on cereal 166

Erika K. Nitsch et al.: Farming Practice and Land Management at Knossos, Crete

grain characteristics: linking prehistoric manuring practice to δ15 N signatures in archaeobotanical material. Journal of Archaeological Science 39(7): 2533– 2540. Killen, J. T. 1993. The Oxen’s Names on the Knossos Ch Tablets. Minos 27-28: 101-107. Killen, J. T. 1998. The rôle of the state in wheat and olive production in Mycenaean Crete. Aevum 72: 19-23. Killen, J. T. 2004. Wheat, barley, flour, olives and figs on Linear B tablets. In P. Halstead and J. C. Barrett (eds), Food, Cuisine and Society in Prehistoric Greece: 155-73. Oxford, Oxbow Books. Killen, J. T. 2008. Mycenaean economy. In Y. Duhoux and A. M. Davies (eds), A Companion to Linear B: Mycenaean Greek Texts and their World, Vol. 1: 159-200. Bibliothèque des cahiers de l’Institut de Linguistique de Louvain 120, Leuven, Peeters. Kragten, J. 1994. Calculating standard deviations and confidence intervals with a universally applicable spreadsheet technique. Analyst 119: 2161–2166. Ledgard, S. F., Sprosen, M. S. and Steele, K. W. 1996. Nitrogen fixation by nine white clover cultivars in grazed pasture, as affected by nitrogen fertilization. Plant and Soil 178(2): 193–203. Livarda, A. and Kotzamani, G. 2014. The Archaeobotany of Neolithic and Bronze Age Crete: Synthesis and Prospects. The Annual of the British School at Athens 108: 1-29. Masi, A., Sadori, L., Balossi Restelli, F., Baneschi, I. and Zanchetta, G. 2014. Stable carbon isotope analysis as a crop management indicator at Arslantepe (Malatya, Turkey) during the Late Chalcolithic and Early Bronze Age. Vegetation History and Archaeobotany 23: 751-760. Mauri, A., Davis, B., Collins, P. and Kaplan, J. 2015. The climate of Europe during the Holocene: a gridded pollen-based reconstruction and its multi-proxy evaluation. Quaternary Science Reviews 112: 109-127. Morgan, C. 2015. The work of the British School at Athens, 2014-15. Archaeological Reports 61: 34-48. Müller-Celka, S., Puglisi, D. and Bendali, F. 2014. Settlement pattern dynamics and natural resources in MM-LMI Crete: the case of Malia. In G. Touchais, R. Laffineur and F. Rougemont (eds), Physis. L’environment naturel et la relation homme-milieu dans la monde égéen protohistorique (Aegaeum 37): 431-440. Liege, Universite de Liège. Nakassis, D., Parkinson, W. A. and Galaty, M. L. 2011. Redistribution in Aegean Palatial Societies. Redistributive Economies from a Theoretical and Cross-Cultural Perspective. American Journal of Archaeology 115(2): 177–184. Nitsch, E. K., Charles, M. and Bogaard, A. 2015. Calculating a statistically robust δ13C and δ15N offset for charred cereal and pulse seeds. Science and Technology of Archaeological Research 1(1): 1-8 STAR20 15112054892315Y.0000000001.

Nitsch, E., S. Andreou, A. Creuzieux, A. Gardeisen, P. Halstead, V. Isaakidou, A. Karathanou, D. Kotsachristou, D. Nikolaidou, A. Papanthimou, C. Petridou, S. Triantaphyllou, S. Valamoti, A. Vasileiadou and Bogaard, A. 2017. A bottom-up view of food surplus: Using stable carbon and nitrogen isotope analysis to investigate agricultural strategies and diet at Bronze Age Archontiko and Thessaloniki Toumba, northern Greece. World Archaeology. 49: 105-137 Parkinson, W., D. Nakassis, and Galaty, M. 2013. Crafts, Specialists, and Markets in Mycenaean Greece: Introduction. American Journal of Archaeology 117: 413-22. Peoples, M. B., Brockwell, J., Herridge, D. F., Rochester, I. J., Alves, B. J. R., Urquiaga, S., Boddey, R. M., Dakora, F. D., Bhattarai, S., Maskey, S. L., Sampet, C., Rerkasem, B., Khan, D. F., Hauggaard-Nielsen, H. and Jensen, E. S. 2009. The contributions of nitrogen-fixing crop legumes to the productivity of agricultural systems. Symbiosis 48(1-3): 1–17. Popham, M. R. 1984. The Minoan unexplored mansion at Knossos: text. The British School at Athens, Thames and Hudson. Riehl, S., Pustovoytov, K. E., Weippert, H., Klett, S. and Hole, F. 2014. Drought stress variability in ancient Near Eastern agricultural systems evidenced by δ13C in barley grain. Proceedings of the National Academy of Sciences of the United States of America 111(34): 1234812353. Sarpaki, A. 1992. The palaeoethnobotanical approach: the Mediterranean triad or is it a quartet? In B. Wells (ed.), Agriculture in ancient Greece : proceedings of the seventh International Symposium at the Swedish Institute at Athens, 16-17 May 1990: 61–76. Stockholm, Svenska institutet i Athen. Sarpaki, A. 2013. The economy of Neolithic Knossos: the archaeobotanical data. In N. Efstratiou, A. Karetsou and M. Ntinou (eds), The Neolithic Settlement of Knossos in Crete: New evidence for the Early Occupation of Crete and the Aegean Islands: 63-95. Philadelphia, INSTAP Academic Press Sarpaki, A. and Jones, G. 1990. Ancient and modern cultivation of Lathyrus clymenum L. in the Greek Islands. Annual of the British School at Athens 85: 363367. Schoep, I. 2001. Managing the hinterland: the rural concerns of urban administration. In K. Branigan (ed.), Urbanism in the Aegean Bronze Age: 87-102. Sheffield Studies in Aegean Archaeology 4, London. Styring, A.K., Ater, M., Hmimsa, Y., Fraser, R., Miller, H., Neef, R., Pearson, J.A. and Bogaard, A. 2016a. Disentangling the effect of farming practice from aridity on crop stable isotope values: A present-day model from Morocco and its application to early farming sites in the eastern Mediterranean. The Anthropocene Review 3: 1-21. 167

Country in the City Styring, A.K., Maier, U., Stephan, E., Schlichtherle, H. and Bogaard, A. 2016b. Cultivation of choice: new insights into farming practices at Neolithic lakeshore sites. Antiquity 90: 95-110. Styring, A.K., Rösch, M., Stephan, E., Stika, H.-P., Fischer, E., Sillmann, M. and Bogaard, A. 2017. Centralisation and long-term change in farming regimes: comparing agricultural practice in Neolithic and Iron Age south-west Germany. Proceedings of the Prehistoric Society. 83: 357-381. Treasure, E. R., Church, M. J. and Gröcke, D. R. 2015. The influence of manuring on stable isotopes (δ13C and δ15N) in Celtic bean (Vicia faba L.): archaeobotanical and palaeodietary implications. Archaeological and Anthropological Sciences 8(3): 555-562. Vaiglova, P., Bogaard, A., Collins, M., Cavanagh, W., Mee, C., Renard, J., Lamb, A., Gardeisen, A. and Fraser, R. 2014a. An integrated stable isotope study of plants and animals from Kouphovouno, southern Greece: a new look at Neolithic farming. Journal of Archaeological Science 42: 201–215. Vaiglova, P., Snoeck, C., Nitsch, E., Bogaard, A. and LeeThorp, J. 2014b. Impact of contamination and pretreatment on stable carbon and nitrogen isotopic composition of charred plant remains. Rapid Communications in Mass Spectrometry 28(23): 2497– 2510. Valamoti, S., and Kotsakis, K. 2007. Transitions to agriculture in the Aegean: the archaeobotanical evidence. In S. Colledge and J. Conolly (eds), The Origins and Spread of Domestic Plants in Southwest Asia and Europe: 75-92 Walnut Creek, CA, Left Coast Press. Vinther, F. P. 1998. Biological nitrogen fixation in grass– clover affected by animal excreta. Plant and Soil 203(2): 207–215. Virginia, R. A., and Delwiche, C. C. 1982. Natural 15N abundance of presumed N2-fixing and non-N2fixing plants from selected ecosystems. Oecologia 54(3): 317–325.

Vokotopoulos, L., G. Plath and F. W. McCoy (2014). The yield of the land: soil conservation and the exploitation of arable land at Choiromandres, Zakros in the New Palace Period. G. Touchais, R. Laffineur and F. Rougemont. Physis: l’environnement naturel et la relation homme-milieu dans le monde égéen protohistorique. Actes de la 14e Rencontre égéenne internationale, Paris, Institut National d’Histoire de l’Art (INHA), 11-14 décembre 2012. Leuven and Liège, Peeters: 251-263. Wallace, M., Jones, G., Charles, M., Fraser, R., Halstead, P., Heaton, T. H. E. and Bogaard, A. 2013. Stable carbon isotope analysis as a direct means of inferring crop water status and water management practices. World Archaeology 45(3): 388–409. Wallace, M. P., Jones, G., Charles, M., Fraser, R., Heaton, T. H. E. and Bogaard, A. 2015. Stable Carbon Isotope Evidence for Neolithic and Bronze Age Crop Water Management in the Eastern Mediterranean and Southwest Asia. PLoS ONE 10(6): e0127085. Whitelaw, T. 2001. Reading between the tablets: assessing Mycenaean palatial involvement in ceramic production and consumption. In S. Voutsaki and J. Killen (eds), Economy and Politics in the Mycenaean Palace States: 51-79. Cambridge, Cambridge Philological Society. Whitelaw, T. 2004. Alternative pathways to complexity in the southern Aegean. In J. C. Barrett and P. Halstead (eds), The Emergence of Civilization Revisited: 232-256. Oxford, Oxbow Books. Whitelaw, T. 2011. The Urbanisation of Prehistoric Crete: Settlement Perspectives on Minoan State Formation. In I. Schoep, P. Tomkins, and J. Driesse (eds), Back to the Beginning: Reassessing Social and Political Complexity on Crete during the Early and Middle Bronze Age: 114–76. Oxford, Oxbow Books.

168

Economy and Storage Strategies at Troy Diane Thumm-Doğrayan

Institut für Ur- und Frühgeschichte und Archäologie des Mittelalters, Eberhard-Karls Universität Tübingen

Peter Pavúk

Institute of Classical Archaeology, Charles University, Prague

Magda Pieniążek

Institut für Ur- und Frühgeschichte und Archäologie des Mittelalters, Eberhard-Karls Universität Tübingen

Introduction

the most monumental architecture and the most wideranging foreign contacts (Pavúk 2005; Pavúk 2010; Pavúk and Pieniążek 2016, with further literature; Pieniążek 2012; Pieniążek et al. 2018). The dimension of the lower town, used likely for both habitation and industrial activities, estimated in the course of an intensive survey and the many years of excavations, as well as by the position of a surrounding ditch, corresponds to the proposed size of lower towns of Thebes and Mycenae; the Troy VI Late citadel reached dimensions of the palace area in Pylos or the citadel in Midea (Figure 2) (Jablonka 2006; Jablonka 2014; Pieniążek 2016: 519).

The past 25 years have been a period of intense interdisciplinary research at the site of Troy (Figure 1). Thanks to the work of many specialists, a wide range of data is now available, both on Troy itself, but also on its surrounding countryside (Korfmann 2006; Pernicka et al. 2014; Wagner et al. 2003). Thus we would like to discuss one aspect that was rarely addressed in previous studies: the relation between storage and other aspects of economy and environment. The information on the management of surplus production, archaeobotanical and archaeozoological data, the availability of other resources and the settlement structure display recognizable patterns and allow now for a new reconstruction of local developments. What is more, being a tell settlement, Troy permits a study of the economic system for each stratum separately.1

Storage in the Aegean and Anatolia Throughout the Bronze Age both in Anatolia and the Aegean pithoi were widely used as storage containers. Minoan pithoi were mainly standing on the ground but sunken ones into the earth were known as well, among others from Kommos, Palaikastro or Phaistos (MacGillivray et al. 1989: 429, Pl. 65b; Shaw and Shaw 1993: 144, Pl. 22a; Christakis 2005: 53). However, it seems that the latter ones were rather rare. The advantage of Troy in this respect is that it was a tell settlement with relatively soft (clayish) ground, so that digging the pits for pithoi up to 2m was usually much easier than in the case of flat sites or hilltop citadels with thinner cultural layer on rocky surface.

The development of Troy follows a trajectory similar to many Aegean centres: A slow but steady rise to social complexity during the first half of the EBA (Troy I), with a peak of activity and foreign relations in the second quarter of the 3rd Millennium (Troy II and III), a sudden change around 2200 BCE and again a slow start with a period of seeming decline and isolation in the late EBA and during the MBA (Troy IV and V). The phase of Troy VI Early (roughly MH III to LH I) is characterized by the re-establishment of foreign contacts and important changes in the local production. Among others, the manufacture of Anatolian Grey Ware was introduced. In the next phase, VI Middle (ca. LH II), Troy became a fortified stronghold. During this time, or at the latest at the beginning of Troy VI Late, a considerable transformation took place both in the settlement structure of the site and its neighbourhood. The significance of Troy grew and it undoubtedly reached the status of the most important centre in the region. Troy VI Late and VIIa (LH IIIA and IIIB) was the time of

Pithoi were common as storage containers also in Mycenaean palaces such as at Mycenae, Tiryns or Pylos (Giannopoulou 2010: 37‒39) but they were very popular in northern Greece as well (Jones et al. 1986; Margomenou 2005). Mycenaean pithoi usually are of ovoid shape with small base and often without handles. The bases of the pithoi very often were fixed in the ground – in Pylos stabilized by a clay stand around the pithoi (Blegen and Rawson 1966: 135) – while the LH IIIA pithoi in Mycenae were standing on the ground (Giannopoulou: 2010:38). The pithoi of Cyprus were rather standardized and frequently used also for other kinds of purposes besides storage (Pilides 2000).

This paper was prepared as part of the Charles University programme PROGRES Q09 – History: Key to Understanding of the Globalised World and European Regional Development Fund-Project No. CZ.02 .1.01/0.0/0.0/16_019/0000734 (PP), Tübingen University programme ‘Bridging Funds’ (MP). The authors would like to thank the director and colleagues of the Troy project for their support and the organisers for the invitation to contribute to this volume.

1

Pithoi were of course only one of many known kinds of constructions for storage. From Hittite centres beside pithoi various kinds of large silos are known (Seeher 169

Country in the City

Figure 1: Troy with silted up plain and straits of Dardanelles. On the horizon islands of Gökceada (Imbros) and Samothrace behind it. Photo Hakan Öge

Figure 2: Troy with the Lower Town. Distribution of Bronze Age ceramics collected in the survey. Jablonka, 2014: Abb. 26.

170

Diane Thumm-Doğrayan et al.: Economy and Storage Strategies at Troy

Figure 3: Late Troy I/Troy IIa subterranean granary. Blegen et al. 1950: Figure 433.

one of them 3,90m wide – a considerable amount of carbonized wheat was found, in heaps more than 20cm high, resting on remains of wood (Blegen et al. 1950: 170‒172, Figure 179‒182, 433). This impressive underground granary proves an organized food storage system already in EBA Troy comparable to other EBA centres such as Poliochni (Bernabò Brea 1964:199; Ivanova 2008: 203‒204). At the same time, pithoi with bases sunken into the ground were also used for storage purposes throughout the EBA at Troy. Two types of EBA pithoi are known: thick walled pithoi with decorated raised bands of which only sherds are preserved and rather thin walled pithoi which are either undecorated or with simple finger impressions. They are characterized by a clearly set off neck and vertical or parallel handles (Thumm-Doḡrayan 2016: 306-307). Their small bases were embedded into the ground.

2000; Müller-Karpe 2001: 239‒241), also the newest excavations in Kaymakcı (central part of West Anatolia) revealed deep paved circular stone constructions that may have been used for storage (Roosevelt et al. 2014: 248, Figure 7, 13; Roosevelt et al. 2018, 681, Figure 7-9; 14). Storage and food processing at Troy The  type  and  extent  of  food storage  are  usually considered as important aspects for the investigation of economic structures of any society. Indeed, underground storage facilities made of stone or mudbrick as well as pithoi and storage pits were found already in the Early Bronze Age (EBA) strata of Troy (Thumm-Doğrayan 2016). The most impressive EBA storage installation was the so called granary from Late Troy I or rather Troy IIa (EH IIA) citadel (Thater 2015: note 1; Ünlüsoy 2011: 141, Plan 4) uncovered by Blegen in its northeastern part (Figure 3). It consisted of four subterranean cellar rooms with no visible entrances – two of them with a pavement of irregular stones. The lower parts of the walls (about 55cm in height) were made of stones covered with clay, the upper parts were built of mud bricks. In both of the plastered rooms –

Pieces of large storage vessels roughly following the general appearance of the EBA pithoi were still found in Troy V strata (MH I‒II) (Blum 2012: 201‒202, Taf. 220.4, 221.1, 221.2). Somewhat suspiciously, pithoi seem to disappear afterwards, being replaced by clay lined storage pits during Early Troy VI (MH III/LH I) (Becks 2006; Pavúk 2014: 330‒331). 171

Country in the City

Figure 4: Troy VI Late/VIIa. Pithos pits in the periphery of the lower town, dug into bedrock. Troy Project slide 16518

The information about storage in Troy VI Middle (LH II) is rather scarce, but the pithoi certainly reappear again; mostly fragmentary, but recognizable by a characteristic fabric with dark core and red slip or wash, as well as by the presence of vaulted plastic clay bands (Kibaroğlu and Thumm-Doğrayan 2013: 45). Later on, these plastic bands became wider and flat. Only one well preserved pithos was found directly south of the citadel area. The estimated height was 1,50‒1,60m and with the exception of the lowest 30cm the pithos was slipped and showed a fingerprint decoration on the uppermost of the clay bands. The lacking coat of the lowest part illustrates the embedded depth of the vessel.

from magazine rooms in the palace area which are not preserved – to the newly built magazine rooms close to the citadel wall. It is thus hard to answer the question how and where the people and/or officials of prosperous Late Troy VI did store their surplus. Surely pithoi were used, but the amount and spatial distribution is unclear. Organic containers inside the houses in cellars or on the ground floor are also possible, as well as overground facilities or subterranean long term storage granaries comparable to the EBA one mentioned above or similar to the contemporary Hittite structures at Hattusa (Seeher 2000), Kuşaklı or Kaman (Müller-Kape 2001: 239‒241). Each storage system is more or less suitable for certain storage purposes, due to its specific properties. Therefore, a combination of different storage facilities could have been used to meet every requirement, like different storing goods, short or long term storage, public or individual storage (Seeher 2000: 267‒269).

While the EBA Trojan pithoi can easily be differentiated from the LBA ones, and even those of Middle Troy VI differ slightly from the later ones, we cannot typologically distinguish pithoi of Late Troy VI from those of Troy VIIa. Pithoi as such are well represented in Troy VI Late (LH IIIA) by numerous sherds and even several almost complete vessels but their numbers are by no means comparable with those of Troy VIIa. What is more, the stratigraphical date of some crucial pieces remains unclear, such as the pithos magazine found towards the centre of the citadel in square G6 for which a Troy VI date was proposed by Korfmann (1998: 27) but VIIa cannot be excluded, or nine pithos pits and sherds found next to the lower town ditch, (Figure 4) the latter dated by thermoluminescence to 1447 BC ±383a (1σ) (Jablonka 1995: 55‒56, Abb. 11, 13). There is likewise a good chance, that a number of Late Troy VI pithoi were reused during Troy VIIa and moved – probably

The best evidence so far comes from Troy VIIa (LH IIIB). Up to now, more than 200 well and partly preserved LBA pithoi and pithos pits (pithos negatives), as well as numerous fragments were unearthed through excavations by Dörpfeld, Blegen and Korfmann (Blegen et al. 1958: 35‒36; Schmidt 1902: 315). The evidence for the citadel is not exactly balanced, since the centre of the mound was levelled in the Hellenistic period and no LBA remains in this potentially crucial area are preserved, such as any possible palace or other kind of elite structures. With the exception of the above mentioned magazine in G6 and some single pithoi further east the only well preserved deposits are 172

Diane Thumm-Doğrayan et al.: Economy and Storage Strategies at Troy

Figure 5: Troy VIIa. Pithos magazines next to the eastern part of the citadel wall. Blegen et al. 1958: Figure 338.

thus located on the lower terrace, running along the southern half of the citadel. This is where rows of rooms dating to Troy VIIa were found, many of them full of pithoi. (Figure 5) Additionally, a few clay lined square, round or polygonal mudbrick structures of about 1,50m length and different depths were represented in some of the rooms. The fact, that traces of wheat were found in some of them speaks for an additional storage

function (Blegen et al. 1958: 113‒114; Dörpfeld notebook 1894: 24) (Figure 6). Besides magazine rooms for storage purpose only, there are others showing structures of food processing. Behind the citadel wall, directly next to the south gate entrance is the location of house 700 (Blegen et al. 1958: 64‒66, Figure 328). Several structures of production 173

Country in the City

Figure 6: Troy VIIa. Sketch of Wilhelm Dörpfeld showing pithos magazines VII ß and ɣ. Dörpfeld excavation daybook 1894: 24. DAI Athens

were found inside like a paved sink with a drainage hole leading to the street, two hearths, several large jars or pithoi as well as a sort of granary with carbonized wheat inside. A saddle quern was fit into a trough constructed of clay and crude brick, sloping towards a basin-like clay construction, 25cm deep. In this basin a considerable quantity of carbonized wheat and vetch was found and next to this construction two grinding stones (Blegen et al. 1958: Figure 21). This working area was interpreted by Blegen as bakery or kitchen, probably even with counter situated in the disproportionate wide door opening to the street which is coming up from the gate (Figure 7).

numerous pithos pits. This pithos room thus originally contained at least ten pithoi (Figure 8). In the second sub-phase of VIIa, the room was likely enlarged, containing now even more pithoi found further east (Hnila 2012: 61‒64). Therefore, in contrary to the magazines in the citadel described above, these pithoi were placed in a structure which was not a separate magazine but an elongated space built onto and clearly functionally connected with the main room and other chambers of the complex. The purpose of the whole house is not absolutely clear yet but a ritual function at least of one of its small rear chambers (South Room) is very likely, since a bronze votive figurine, fragments of a bull rhyton as well as many pieces of jewellery and many other valuable objects were found there (Mellink and Strahan 1998; Pieniążek and Aslan 2016; Rigter and Thumm 2004). There are also traces of cult activities in the area of the South Room during the period VIIb and the surrounding area became a sanctuary in the EIA (Aslan and Hnila 2015: 202‒205; Pieniążek and Aslan 2016: 428‒431). Further pithoi were found also in other

Adjacent to the citadel in the western part of the Lower town a building termed Terrace House was uncovered by the new excavations (Becks et al. 2006). It had a central room with two pithoi dug into the ground at both sides of the entrance and two more standing on the ground. In an adjacent long and narrow storeroom three further pithoi were found in situ together with 174

Diane Thumm-Doğrayan et al.: Economy and Storage Strategies at Troy

Figure 7: Troy VIIa. House 700 next to south gate with several installations for food processing. Blegen et al. 1958: Figure 328.

175

Country in the City buildings in the neighbourhood, some of them still standing in situ. Additional remains of several pithoi were discovered in the central and southern part of the Lower Town, but not in any high density, apart from nine pithos pits mentioned above which were embedded at the southern edge of the settlement. In Troy VIIb (LH IIIC), which is the transition period to the EIA, only few pithoi and pithos sherds were found. It is likely no coincidence that the houses now show substructures composed of cell-like rooms without entrance, often elaborately paved (Figure 9). These must have been used for storage with access from above, which would indicate a completely different storage concept (Becks et al. 2006: 183‒184; Hnila 2012: 5‒6, 37, 136‒137). Appearance, size and contents of the LBA Pithoi at Troy The pithoi of Late Troy VI and VIIa are characterized by the lack of neck and handles, while the EBA pithoi usually had both. In the LBA they generally show a narrow orifice, a projected shoulder to gain storage space, a pronounced base for better firing resistance, easier emptying and catch area for particulate matter in liquids and many of them have raised clay bands to strengthen the body. They are mostly undecorated but some show a rather simple decoration on the shoulder zone (Figure 10).

Figure 8: Troy VIIa. In situ pithos in Terrace House with vessel inside.

Figure 9: Troy VIIb. House with cubicles of Troy VIIb. Troy Project slide 24701

176

Diane Thumm-Doğrayan et al.: Economy and Storage Strategies at Troy

Figure 10: Late Troy VI/VIIa. Plain and decorated types of pithoi. Photo Thumm-Doḡrayan

pithoi were determined and ranged from 120kg to more than 240kg.

Whereas in earlier times they seem to have been only partly embedded, by the time of Troy VIIa they are usually dug into the ground, at least up to the shoulder zone, sometimes even up to the rim which deeply improves the conditions for long time storage (Seeher 2000: 267‒268; Thumm-Dograyan 2016: 309). That is also why almost no decoration appears in the developed LBA. The thickness of the wall usually extends 2cm and a thickness of about 3cm is still quite common. Very often there is no burnishing visible. Pithoi of different sizes and shapes, with and without clay bands were found together in the same rooms (Figure 11). As every potter has its own ‘handwriting’ represented by special production techniques such as clay bands, certain rim shapes, decoration, these pithoi must have been produced by different potters, probably also at a different date. This underlines the supposition that these valuable containers were used and reused over a longer period of time. Lead clamps for mending which were found on several pithoi show the value of these containers and indicate at the same time that in these cases dry goods must have been stored inside (Figure 12).

According to the ethnographic evidence underground containers were used mainly for the storage of cereals, pulses, olive oil, wine and rarely for other materials (e.g. Christakis 2005: 53‒54). Although many of the pithoi found in various rooms in and around the citadel of Troy were still covered by a stone lid when excavated, only very few of them revealed any botanical residues, while elsewhere various kinds of carbonized botanical remains were found in abundance. Remarkably, botanical residues were found only in pithoi which were not dug into ground. Several botanical samples indicated the presence of vetch, einkorn and emmer wheats, as well as barley (Lindau 1922: 172‒176; Schiemann 1951: 157; Schliemann 1891: 17). However, it is very likely that the contents of the pithoi dug into ground and covered with a stone lid were not carbonized by fire and therefore we do not have macroscopic residues. The original interpretation of the early excavators was that the pithoi might have been filled with liquids. The fact, that some pithoi were repaired with the help of lead clamps is a further argument against the storage of liquids only.

The volumes of the Troy VI Late/VIIa pithoi range from about 200 to about 600 litres, some reaching 1.000 and even more litres. The inner height ranges from about 1m to 1,80m and the biggest diameter from about 70cm to more than 1,20m. The weights of several complete

One of the pithoi found in the western magazines was covered with a stone lid and additionally smeared with loam (Dörpfeld 1894: 66). A cover like this is known from 177

Country in the City

Figure 11: Troy VIIa. Magazine room with in situ pithoi of different appearance and size. Dörpfeld photo 462, DAI Athens

(Blum et al. 2014; Pavúk and Schubert 2014; Pavúk et al. 2014: 115‒119).

the wine-containers called kvevris in Georgia or the pithoi of Koroni in Messenia (Bariasashvili 2011: 24‒26; Giannopoulou 2010: 89). Because there are no organic remains in these storage vessels the only possibility to get information about the contents is residue analysis of the ceramics. Many of them did not reveal good results, some are still under way, but two of them ‒ one pithos from a VIIa-magazine in the western part of the citadel (B7.18) and another one closer to the centre which might be VI Late or VIIa (G6.340) contained palmitoleic acid typical of yeasts (pers. communication P. Grave).

Chemical analyses (NAA) demonstrated that Troy became a major producer and exporter of Anatolian Grey Ware from the very beginning of Troy VI and continued dominating the Northern Troad throughout the rest of the 2nd Millenium. Another important centre producing fine wares emerged at the coastal site of Larisa-Liman Tepe, in the SW part of the Troad. Minor local production could be proved for some smaller settlements, who received also pottery from Troy and/or Larissa (Grave et al. 2012; Mommsen and Pavúk 2007; Pavúk and Schubert 2014).

Localising the production of the Trojan pithoi One question that was not addressed so far is the production of these large storage containers. Various surveys conducted in the area of Troy by many teams over the years, have established a number of settlements from the 2nd Mill BC, most of which were concentrated on the coast (Akarca 1978; Aslan et al. 2003; Blegen et al. 1950: 5‒7, Figure 415; Cook 1973) (Figure 14). Recent analysis demonstrated that both, the beginning and the end of the 2nd Millennium (Troy V and VIIb) were the periods of lowest settlement density. The number of known sites is the highest during following VI Early (especially VIb/c) and VI Middle. Then, however, the density decreased again during 14th‒13th century BC and we see signs of nucleation, especially since Troy seems to be growing in these centuries (VI Late‒VIIa)

However, the size, unwieldiness, and weight of the pithoi require a different way of manufacture than other pottery. Therefore, they were usually made by specialized potters, producing either solely pithoi or other big vessels of similar manufacturing technique, such as bathtubs or beehives. Since pithoi were used not only for storage purposes but also as coffins, there should have been a high demand in and around the city of Troy. On the other hand, most of the pithoi are of large size, heavy and rather exposed to breakage when transported. Thus, the pithoi of Troy were always thought to have been produced in or at least next to the city close to their later place of use, probably rather by settled than itinerant potters, the latter 178

Diane Thumm-Doğrayan et al.: Economy and Storage Strategies at Troy

Figure 12: Troy VI Late/VIIa. Pithos mended with lead clamps. Photo Thumm-Doḡrayan

clay transportation also (Christakis 2005: 58). In our case it seems to be more likely that the production places of at least many of the Trojan pithoi are to be localised in Ezine district, rather than huge amounts of clay were transported to Troy.

documented ethnographically for instance in Messenia and on the islands of Crete or Cyprus (Christakis 1996; Giannopoulou 2010: 53‒54; Hampe and Winter 1962; London 1989: 69; Voyatzoglou 1974). An extensive program of chemical analyses (NAA) on ceramics done not only for Troy but also for the whole Western Troad revealed very important evidence (Grave et al. 2012; Pavúk and Schubert 2014: 878, Abb. 5). Interestingly, both petrography as well as NAA results of a representative number of pithos sherds from Troy independently came to the conclusion that the clay sources of almost all sampled pieces are nonlocal. Petrographic analysis of the EBA and LBA samples points to clay sources around the district of Ezine which is about 20 km further south, up the stream of Scamander river. The provenance of the most frequent fabric group PG1 as well as that of PG2 are namely the fluvial sediments in the banks of the Skamander river northeast of Ezine (Figure 13). Also one pithos sample of the LH IIIA‒IIIB1 cemetery of Beşik-Tepe (Basedow 2000), about 8km westwards from Troy, was analysed and revealed the same result. Only two analysed pieces of subgroup PG 1c show inclusions which also could be sourced closer to Troy (Kibaroḡlu and ThummDoḡrayan 2013: 49‒50). Ethnographic studies confirm that workshops producing very large storage vessels usually were situated in the vicinity of clay sources (e.g. Giannopoulou 2010: 56) although there is evidence for

Economy in and around Troy The following lines will briefly outline other branches of Trojan economy based on the available data. Although various aspects of it were intensively studied during the past 20‒30 years, as in the case of archaeobotany analysed by Simone Riehl (Riehl 1999; Riehl 2006; Riehl et al. 2014), many areas of research are still in progress. Evidence available so far shows a risk-buffering strategy during the period of Troy IV (EBA III): a lot of hunting animals, many bones of young pigs ‒ indicating premature slaughtering, low percentage of cattle and high diversity among cultivated plants. Regrettably, the archaeobotanical results for the period Troy V (Blum 2012: 316‒319) were not fully evaluated yet but the archaeozoological evidence shows a clear increase in cattle – a trend that continues until Troy VIIa. Horse breeding started in Middle Troy VI and became a very strong part of the economy in Late Troy VI (Korfmann and Zidarov 2006). In addition to it elite hunting of wild goat is evidenced for VI Late and VIIa (Riehl et al. 2014: 752; Uerpmann 2003: 256; Uerpmann 2006: 293). 179

Country in the City

Figure 13: Possible location of clay sources for the production of Trojan pithoi. Map Kibaroḡlu After Kibaroğlu and Thumm-Doğrayan 2013: Figure 3.

180

Diane Thumm-Doğrayan et al.: Economy and Storage Strategies at Troy

Figure 14: Map of LBA settlements (full circles) and silver and copper mines (empty circles) in the Troad. Map Pavúk

with hearth, surrounded by a pavement in connection with several pits containing crushed murex shells no installations of dye production have been found in VI Late and VIIa so far, although these subperiods show the highest amounts of crushed snails. It is, however, thinkable that the main production areas of smell intensive murex dye of these settlement phases were rather situated outside of the settlement.

While the Chalkolithic settlements in the vicinity of Troy show an extremely high amount of shellfish by comparison to vertebrate remains it can be observed that there is a constant decrease throughout the habitation periods at Troy. It is likely that this phenomenon has to do with the increased significance of animal husbandry and agriculture as well as the gradual siltation of the lagoon. In the Bronze Age generally the lagoon cockle (Cerastoderma glaucum) was the most abundant species (Çakırlar 2009). High percentages of the common murex snail Hexaplex Trunculus in the LBA – especially Troy VI Late and VIIa – show that mollusks were used now not only because of their calorific value but also as raw material for dye production. In these LBA settlement phases crushed murex snails were found in heaps as secondary deposits in the southeastern area of the citadel next to the fortification wall as well as smaller amounts in the lower town (Çakırlar and Becks 2009: 90‒96; cf. also Pavúk 2012: 127, note 56; 2014: 423, note 43). While for Troy VI Early there is evidence for a small scale purple dye production in the lower town

Archaeobotanical analyses of Simone Riehl show that during the Late Bronze Age the cereals are generally represented with more than 50% ubiquity, emmer and einkorn remain stable but amounts of barley vary throughout the phases. Millet is better represented in Late Troy VI and Troy VIIb what can indicate droughts, or other unfavourable growing conditions. The biggest diversity among cultivated plants was observed for VI Late. Interesting is also that during Troy VIIa hydrophilic wild plants decrease in comparison with open grassland-type vegetation, which is in contrast to both the preceding Troy VI Late, the following 181

Country in the City Troy VIIb, as well as to most other habitation phases at Troy. This means higher and dryer areas were more intensively cultivated or exploited in other ways during Troy VIIa (Pavúk et al. 2014: 113‒114, Pl. 31; Riehl et al. 2014: 748‒755). Conclusions The absence of written documents from Troy makes any reconstruction of its economic system and the ways how the surrounding environment, the country, was exploited, more difficult. Seals as such were found at Troy only sporadically and there is no evidence for any actual sealing practice at all. Troy seems to remain literally unimpressed by the Minoan sealing practices known on the nearby island of Samothrace during the later Middle Bronze Age (Matsas 1993; Matsas 2004). The only period, when a certain type of seal appeared in any bigger quantities is the end of Troy VI Late and VIIa: 5 seals are known from Beşik-Tepe (Basedow 2000: 132134) and 2 (or 3) from Troy itself (Pieniążek 2017). All of them were defined as so called Mainland Popular Group. Only 2 were made of steatite, what is typical for this group, the rest is of bone (Figure 15) and this is unusual since this material was never used for the production of this kind of Mycenaean seals. This as well as the fact that the seals from Beşik-Tepe show some unusual features clearly indicates that they were local products. However, the purpose of these seals is not clear even in the Aegean, since almost no related sealings are known (Krzyszkowska 2005: 274‒275; Eder 2007). They may have been used as signs identifying status, rank or function. It is thus hard to say, whether they had any role in the administrative systems of Troy and BeşikTepe. This means most probably, that the management of goods in Troy was organised differently from what we know from the southern Aegean or Hittite Anatolia.

Figure 15: Seal of ‘Mainland popular group’ found at Beşik-Tepe Pieniążek in press, fig. 4.

the citadel were erected, which clearly points to the emergence of a strong local elite and the beginning of more centralised economy. Sometime during this phase or on the turn to Late Troy VI, the nucleation of the settlement in the Troad took place, the territory of Troy itself grew and the ditch surrounding the lower town was constructed. Some secondary local centres, such as Bozköy-Hanaytepe still flourished during Middle Troy VI and even had a significant own production of storage containers (Blum et al. 2011: 133, Taf. 10.7‒9; Pavúk and Schubert 2014: 887‒889). It is well known, at least since the excavations of Dörpfeld in 1893‒1894 (Dörpfeld 1902), that Troy VI Late was the period of great prosperity, what is most obviously based on the monumental architecture. However, the economic backgrounds of this development and especially the changes that took place during the following VIIa period could not be sufficiently explained yet. In the meantime, it is clear that the 13th century BC was the time of deep structural changes visible in various aspects of the economy (Pavúk and Pieniążek 2016; Pieniążek 2016). The citadel was rebuilt in a completely new style, the structures became smaller, but the building system denser. At this time numerous embedded pithoi, many in separate storerooms appeared. Theoretically speaking it is possible, that many of the pithoi were originally manufactured during Troy VI Late and became reused in VIIa, transferred for example from the centre to magazine rooms adjacent to the citadel wall – or that during Troy VI Late other kinds of containers were used for storage as well. However, it is only during Troy VIIa that the archaeobotanical evidence shows important changes in the agricultural praxis. The traditional system, exploiting predominantly river valley environments was modified and the fields were set up also in drier and higher areas, such as the nearby lower plateau. As pointed out by Simone Riehl already in 1999, these transformations – the appearance of a new storage system and changes in the land use strategy – were probably related to one another and point to a centrally regulated management of resources (Riehl

In order to understand the relation between Troy and the surrounding countryside we must therefore rely mainly on the ‘standard’ archaeological data. Let us summarise what we know and interpret the evidence: The end of the Early Bronze Age and the beginning of the Middle Bronze Age, Troy IV and V, was the time of a new type of agglutinative architecture and of a riskbuffering strategy in the agricultural production (Blum 2012). For this period, a small scale household based way of storage prevailed and there is no indication for any centrally organised surplus protection. However, we must always keep in mind, that the central part of the citadel is not preserved. The structure of agricultural production began to change during the time of Troy V and Early Troy VI, when the husbandry of bigger animals and grain cultivation increased. Interestingly, it is during Middle Troy VI, when a new strategy of storage was introduced and big containers, pithoi, reappeared. This is also the time when the first monumental fortifications and buildings in the area of 182

Diane Thumm-Doğrayan et al.: Economy and Storage Strategies at Troy

1999). Therefore, it is not surprising, that in the crisis years after the collapse of 1200 BC pithoi practically disappeared and crop cultivation came back to the valley, to a traditional way of agricultural production, known throughout the Bronze Age.

they would be accessible from the new house floor. The difference between the elevation of the floors of VIIa and VIIb1 was rather small. One possible reason could be connected with the construction of the buildings: some VIIa houses may have been two-storied, where the magazines in the basement were only (or partly) designated for pithoi whose opening were positioned roughly on the floor level (and consequently the floor could not be used as such). The upper storey could have living or other function. The inhabitants of the citadel during the 12th and 11th century BC probably did not have the resources necessary to construct multistoried buildings and consequently could not use the vessels sunken into the floor.

This peak of storage activity at Troy was surely connected with the elite’s need for increase of control over wealth and resources. Manufacture of large pithoi belonged to the most difficult among ceramic production requiring the knowledge of complicated procedures (Bariasashvili 2011: 8; Casson 1938: 468; Hansen 2001: 13) and it is clear that it could not be household-based. The fact, that the pithoi were made 20 km to the southeast from Troy demonstrates a very advanced system of territorial specialisation of production that functioned probably already during the period of Troy VI Middle but surely in the periods of Late Troy VI and VIIa. Furthermore, this production was not standardised, what speaks for the existence of various workshops. Since simultaneously almost no Troy VI Late/VIIa pithoi seem to be manufactured at Troy itself, it is logical to assume that this activity was at least controlled from Troy. Currently it is difficult to say why pithoi had to be made there and not in the direct vicinity of Troy. Transportation of big pithoi many kilometres over land or sea routes is not unknown in the Bronze Age Aegean and recent times, however ethnographic studies describe also the evidence of transportation of the raw material of especially desirable properties (Casson 1938; Christakis 2005: 57‒58; Rice 1987: 118). There is no indication for any important advantage the clay from the Ezine area would have in comparison with the material accessible at Troy but the availability of appropriate tempering material and enough clay-rich fluvial sediments for the high demand of pithos production may have played a role. The graspability of those pithoi was probably not so important since they were intended to be dug into the earth. One can suspect that either they were heaved with ropes and transported on wagons along the Scamander valley to Troy or the river itself was used for transportation by boat or raft.

Another possibility is that people did not want to do it, maybe the use of big containers became unpractical, when the centrally organised storage system collapsed and the inhabitants came back to the small scale household economy with short-term storage. The costs of use of these big pithoi could have become simply too high: they were difficult to move, clean, special small containers were necessary to scoop the content. It is possible that the goods stored inside were not (totally or partly) produced anymore and so closely connected with the economic system forced upon the society by the central authority. When this authority became weakened or even completely vanished around 1200 BC the remnants of the (previous) system were rejected and forgotten. It is interesting to compare this situation with the development in the field of other ceramic production: The fact that ca. 40-50% of pottery was produced during Troy VIIb more or less in the same style as during Troy VIIa (Hnila 2012: 214) may point towards the explanation that at least some of the workshops producing local pottery at Troy were not fully specialized and rather led by part-time specialists whose subsistence strategy was more flexible and consequently could easier recover after the collapse of 1200 BC. We saved one more important question for the very end: what or rather who was responsible for changes in the 14th and 13th centuries BC at Troy. Certain economic strategies, the ‘know-how’ in the realm of administration could have arrived in the frame of mercantile activities but also as a result of political influences. The best documented trade connections during the 14th and 13th century BC existed with the South Aegean and even more distant areas, such as Eastern Mediterranean regions. Whereas the last one was surely indirect, the direct communication with Mycenaean centres cannot be ruled out (Pieniążek et al. 2018). Therefore, stimuli coming from the southern Aegean may have played a role in the transformations during 14th‒13th centuries BC. On the other hand, we know from the Alaksandu Treaty that Troy (Wilusa) became in a way dependent from the Hittites (Bryce 2006; Hawkins 1998; Latacz and

When the economic system collapsed after 1200 BC, in Troy VIIb, some pottery workshops still existed at Troy, since the production of Anatolian Grey Ware continued, using the same chemical recipe. However, highly specialised workshops producing pithoi did not survive, or/and the transportation system broke down. This can be explained with the collapse of redistributive and centrally organised management. It is also possible that the goods, be it liquids, cereals or whatever was stored inside, spoiled during the crisis, turbulent times. But all this does not explain why these containers were not reused after 1200 BC. Theoretically it would be easy to excavate them, fill the pit partly with earth and dig them in again, so that 183

Country in the City Starke 2006: 63‒68). Although there are no clear Hittite elements visible in the Trojan culture, we cannot completely reject the possibility of Hittite impact (be it directly from central Anatolia or from one of the conquered polities in western Anatolia). According to the text of the treaty, Wilusa had to guarantee certain troops and other military supplies, therefore it is not impossible that the Hittites would support its rulers with certain technologies enabling more successful mobilization of resources. Christakis demonstrated (2011) that the storage capacities in Knossos and other Neopalatial Minoan centres were most probably intended to support a small group of people, the elite and its direct environment: officers, craftsmen and workmen. Kessler (2015) independently confirmed this observation. Any calculation of storage capacity in Troy would be pointless due to the destruction of the central part of the settlement, but we can suspect that large pithoi from the magazines in the citadel served a similar purpose: the backup of the activities of local authority and its followers while surely some of the singular containers could belong to domestic storage.

Aslan, R., Bieg, G., Jablonka, P. and Krönneck, P. 2003. Die mittel- bis spätbronzezeitliche Besiedlung (Troia VI und VIIa) der Troas und der Gelibolu-Halbinsel. Ein Überblick. Studia Troica 13: 165–213. Aslan, C. and Hnila, P. 2015. Migration and Integration at Troy from the End of the Late Bronze Age to the Iron Age. In N. Stampolidis, Ç. Maner and K. Kopanias (eds), NOSTOI. Indigenous Culture, Migration and Integration in the Aegean Islands and Western Anatolia during the Late Bronze and Early Iron Age. Koç University Press 58: 185–210. Istanbul, Koç University Press. Bariasashvili, G. 2011. Making wine in qvevri – a unique Georgian tradition. Tbilisi, Elkana. Basedow, M. 2000. Beşik-Tepe. Das spätbronzezeitliche Gräberfeld. Studia Troica Monographien 1. Mainz am Rhein, Zabern. Becks, R. 2006. Troia im 2. Jahrtausend v. Chr. (Spätbronzezeit bis Früheisenzeit). Stratigraphie, Architektur, Befunde. Unpublished PhD thesis, University of Tübingen. Becks, R., Hnila, P. and Rigter, W. 2006. Das Terrassenhaus im westlichen Unterstadtviertel von Troia. Studia Troica 16: 27–87. Bernabò Brea, L. 1964. Poliochni. Città preistorica nell´ isola di Lemnos, Vol. I, 1. Rome, L’Erma di Bretschneider. Blegen, C. W., Caskey, J. L., Rawson, M. and Sperling, J. 1950. Troy I. General Introduction. The First and Second Settlements. Princeton, University Press. Blegen, C. W., Boulter, C. G., Caskey, J. L. and Rawson, M. 1958. Troy IV. Settlements VIIa, VIIb and VIII. University Press. Blegen, C. W. and Rawson, M. (eds) The Palace of Nestor at Pylos in Western Messenia. Vol I. Princeton, Princeton University Press. Blum, S. W. E. 2012. Die ausgehende frühe und die beginnende mittlere Bronzezeit in Troia: Archäologische Untersuchungen zu ausgewählten Fundkomplexen der Perioden Troia IV und Troia V. Studia Troica Monographien 4. Darmstadt, Zabern. Blum, S. W. E., Aslan, R., Uysal, F. E., Kirschner, S. and Kraus, S. 2011. Archäologische Untersuchungen zur voreisenzeitlichen Kultursequenz des BozköyHanaytepe, Nordwesttürkei. Studia Troica 19: 119– 177. Blum, S. W. E., Thater, M. and Thumm-Doğrayan, D. 2014. Die Besiedlung der Troas vom Neolithikum bis zur beginnenden Mittleren Bronzezeit: Chronologische Sequenz und Siedlungsstruktur. In E. Pernicka, C. B. Rose and P. Jablonka (eds), Troia 1988–2008: Grabungen und Forschungen I. Forschungsgeschichte, Methoden und Landschaft. Studia Troica Monographien 5: 864–907. Bonn, Habelt. Bryce, T. 2006. The Late Bronze Age in the West and the Aegean. In S. R. Steadman and G. McMahon (eds), The Oxford Handbook of Ancient Anatolia, 10,000–323 BCE: 363–375. Oxford, Oxford University Press.

The case of the elongated magazine in the Terrace House might be a different one. As was mentioned above the function of the whole structure is not clear, but one of the small rear chambers, the South Room had most probably a cultic function, it must have been a kind of shrine. However, this room is an exception in comparison with other buildings known from Late Bronze Age Troy, both in terms of the finds and architectural features (there was a clay platform there supposedly an altar). The objects found there are probably too numerous and too valuable as for a simple domestic, house shrine. Consequently, it is possible that the entire Terrace House had a religious function, and this can explain the presence of a big storage magazine. Temples could enjoy economic independence, Hittite temples in Hattusa, for example, had numerous own magazines, which were partly filled with hundreds of pithoi (Neve 1995, 49‒50, Figure 15; Seeher 2000, 287; Schoop 2006, 35‒36, Figs.4; 7). Though it is not known if true ‘temple economy’ was in operation, they demonstrate some autonomy in the system of storage. Which of the connections, Hittite, western Anatolian or Aegean was the crucial one in respect of the changes in storage management at Troy during the Late Bronze Age is impossible to decide. But it was the Trojan elite that transformed step by step a local economic system, using strategies that were available and feasible. The picture of ‘Troy and the country’ is still patchy but it contains some very interesting points. Bibliographie Akarca, A. 1978. Troas´ta aşağı Kara Menderes ovası çevresindeki şehirler. Belleten XLII-165: 1–52. 184

Diane Thumm-Doğrayan et al.: Economy and Storage Strategies at Troy

Casson, S. 1938. The Modern Pottery Trade in the Aegean. Antiquity 12: 464–473. Çakırlar, C. 2009. To the shore, back and again. Archaeomalacology of Troia. Studia Troica 18: 59–86. Çakırlar, C. and Becks, R. 2009. ´Murex` Dye Production at Troia: An Assessment of Archaeomalacological Data from Old and New Excavations. Studia Troica 18: 87–103. Christakis, K. S. 1996. Craft Specialisation in Minoan Crete: The Case for Itinerant Pithos Makers. Aegean Archaeology 3: 63–74. Christakis, K. S. 2005. Cretan Bronze Age Pithoi. Traditions and Trends in the Production and Consumption of Storage Containers in Bronze Age Crete. Philadelphia, INSTAP Academic Press. Christakis, K. S. 2011. Redistribution and Political Economies in Bronze Age Crete. In M. L. Galaty, D. Nakassis and W. A. Parkinson (eds), Redistribution in Aegean Palatial Societies. American Journal of Archaeology 115/2: 197–205. Forum online, http:// www.ajaonline.org/forum/905 (02.02.2016). Cook, J. M. 1973. The Troad. An Archaeological and Topographic Study. Oxford, Clarendon Press. Dörpfeld, W. 1894. Troja 1893. Bericht über die im Jahre 1893 veranstalteten Ausgrabungen. Leipzig, Brockhaus. Dörpfeld, W. 1902. Troja und Ilion. Ergebnisse der Ausgrabungen in den vorhistorischen und historischen Schichten von Ilion 1870–1894. Athens, Beck and Barth. Eder, B. 2007. Im Spiegel der Siegel. Die nördlichen und westlichen Regionen Griechenlands im Spannungsfeld der mykenischen Paläste. In E. Alram-Stern and G. Nightingale (eds), Keimelion. Elitenbildung und elitärer Konsum von der mykenischen Palastzeit bis zur homerischen Epoche. Akten des internationalen Kongresses vom 3. bis 5. February 2005 in Salzburg: 81–124. Vienna, Österreichische Akademie der Wissenschaften. Giannopoulou, M. 2010. Pithoi. Technology and history of storage vessels through the ages. British Archaeological Reports. International Series 2140. Oxford, Archaeopress. Grave, P., Kealhofer, L., Marsh, B., Hnila, P., Aslan, C., Thumm-Doğrayan, D. and Rigter, W. 2013. Cultural Dynamics and ceramic resource use at Late Bronze Age/ Early Iron Age Troy/northern Turkey. Journal of Archeological Science 40: 1760–1777. Hampe, R. and Winter, A. 1962. Bei Töpfern und Töpferinnen in Kreta, Messenien und Zypern. Mainz am Rhein, Römisch-Germanisches Zentralmuseum. Hansen, J. G. 2001. Zur Tradition und zur Fertigung ‚homerischer‘ Pithoi in Süd-Messenien/Peloponnes. In R. Stupperich and Heinz A. Richter (eds), Thetis – Mannheimer Beiträge zur klassischen Archäologie und Geschichte Griechenlands und Zyperns, Band 8: 7–22. Mannheim, Rutzen. Hawkins, J. D. 1998. Tarkasnawa King of Mira: ‘Tarkondemos’, Boğazköy Sealings and Karabel. Anatolian Studies 48: 1–31.

Hnila, P. 2012. Pottery of Troy VIIb. Chronology, Classification, Context and Implications of Trojan Ceramic Assemblages in the Late Bronze Age/Early Iron Age Transition. Tübingen. Holzer, V. 2008. Ein latènezeitlicher Getreidespeicher aus der keltischen Großsiedlung am Sandberg in Roseldorf (Niederösterreich). Germania 86-1: 135– 270. Ivanova, Mariya. 2008. Befestigte Siedlungen auf dem Balkan, in der Ägäis und in Westanatolien, ca. 5000–2000 v. Chr. Münster/New York, Waxmann. Jablonka, P. 1995. Ausgrabungen südlich der Unterstadt von Troia im Bereich des Troia VIVerteidigungsgrabens. Grabungsbericht 1994. Studia Troica 5: 39–79. Jablonka, P. 2006. Leben außerhalb der Burg – die Unterstadt von Troia. In M. O. Korfmann (ed.), Troia. Archäologie eines Siedlungshügels und seiner Landschaft: 167–180. Mainz am Rhein, Zabern. Jablonka, P. 2014. Archäologischer Survey im Stadtgebiet von Troia. In E. Pernicka, C. B. Rose and P. Jablonka (eds), Troia 1988–2008: Grabungen und Forschungen I. Forschungsgeschichte, Methoden und Landschaft. Studia Troica Monographien 5: 262–363. Bonn, Habelt. Jones, G., Wardle, K., Halstead, P. and Wardle, D. 1986. Crop Storage at Assiros. Scientific American 254/3: 84–91. Kessler T.P. 2015, A Royal Gift? Bulk Grain Storage in Protopalatial and Neopalatial Crete, Studi Micenei ed Egeo-Anatolici NS 1: 137–170. Kibaroğlu, M. and Thumm-Doğrayan, D. 2013. Trojan Pithoi: A Petrographic Approach to Provenance and Production Technique of Bronze Age Storage Vessels from Troy. Journal of Applied Ceramic Technology 82: 44–52. Korfmann, M. O. 1998. Troia – Ausgrabungen 1997, Studia Troica 8: 1-70. Korfmann, M. O. (ed.) 2006. Troia. Archäologie eines Siedlungshügels und seiner Landschaft. Mainz am Rhein, Zabern. Korfmann, M. O. and Zidarov, P. 2006. Trensenknebel in Troia. In H.-P. Wotzka (ed.), Grundlegungen. Studien zur europäischen und afrikanischen Archäologie für M. K. H. Eggert: 677–690. Tübingen, Francke. Krzyszkowska, O. 2005. Aegean seals: an introduction. London, Institute of Classical Studies, University of London. Latacz, J. and Starke, F. 2006. Wilusa und die Großen Vier – Troia in der politischen Landschaft der Späten Bronzezeit. In M. O. Korfmann (ed.), Troia. Archäologie eines Siedlungshügels und seiner Landschaft: 57–70. Mainz am Rhein, Zabern. Lindau, G. 1922. Fossile Samen von Troja. Prähistorische Zeitschrift XIII, XIV: 172–176. London, G. 1989. Töpferei auf Zypern damals – heute. Traditional Pottery in Cyprus. Mainz am Rhein, Zabern. 185

Country in the City MacGillivray, J.A, Sackett, L. H., Driessen, J., Bridges, R. and Smyth, D. 1989. Excavations at Palaikastro,1988, BSA 84: 417–445. Margomenou, D. 2005. Food Storage, Surplus and the Emergence of Institutionalized Inequality: A Study of Storage Jars and Food Storage for Central Northern Greece in the Late Bronze and Early Iron Age. Michigan, University of Michigan. Matsas, D. 1993. Samothrake. Archäologisches Museum. In I. Pini (ed.), Kleinere griechische Sammlungen. Lamia – Zakynthos und weitere Länder des Ostmittelmeerraums. Corpus der minoischen und mykenischen Siegel, Supplementum 1B: 303–314. Berlin, Mann. Matsas, D. 2004. Samothrake. Archäologisches Museum. In I. Pini (ed.), Kleinere griechische Sammlungen. Neufunde aus Griechenland und der westlichen Türkei. Corpus der minoischen und mykenischen Siegel, Band V, Supplementum 3.2: 497–500. Mainz am Rhein, Zabern. Mellink, M. and Strahan, D. 1998, The Bronze Figurine from Troia VIIa, Studia Troica 8: 141–150. Mommsen, H. and Pavúk, P. 2007. Provenance of the Grey and Tan Ware from Troia, Cyprus and the Levant. Studia Troica 17: 25–42. Müller-Karpe A. 2001. Untersuchungen in Kuşaklı, 2000. Mitteilungen der Deutschen Orient Gesellschaft zu Berlin 133: 236–250. Neve, P. 1995. Der Große Tempel (Tempel I) in BoğazköyHattusa. Nürnberger Blätter zur Archäologie 12: 41–62. Pavúk, P. 2005. Aegeans and Anatolians. A Trojan Perspective. In R. Laffineur and E. Greco (eds): EMPORIA. Aegeans in central and eastern Mediterranean. Proceedings of the 10th International Aegean Conference / 10e Rencontre égéenne internationale, Italian School of Archaeology in Athens, 14‒18 April 2004. Aegaeum 25: 269‒279. Liège/Austin, University of Liège/University of Texas at Austin. PAVÚK, P. Between the Aegean and Anatolia. The Shifting Character of Troy in the Middle and Late Bronze Age. Bulletin of the Institute of Classical Studies of the University of London 53-2: 128–129. Pavúk, P. 2012. Of Spools and Discoid Loom-Weights: Aegean-Type Weaving at Troy Revisited. In M.L. Nosch and R. Laffineur (eds), KOSMOS. Jewellery, Adornment and Textiles in the Aegean Bronze Age. 13th International Aegean Conference / 13e Rencontre égéenne international, University of Copenhagen, Danish National Research Foundation´s Centre for Textile Research, 19‒23 April 2010. Aegaeum 33: 121‒130. Leuven/Liège, Peeters. Pavúk, P. 2014. Troia VI-Früh und Mitte. Keramik aus den Grabungen 1988–2002. Studia Troica Monographien 3. Bonn, Habelt. Pavúk, P. and Schubert, C. 2014. Die Troas in der Mittelund Spätbronzezeit. In E. Pernicka, C. B. Rose and P. Jablonka (eds), Troia 1988–2008: Grabungen und Forschungen I. Forschungsgeschichte, Methoden und

Landschaft. Studia Troica Monographien 5: 864–907. Bonn, Habelt. Pavúk, P., Pieniążek, M. and Riehl, S. 2014. Troy and the Troad in the Second Millenium: Changing Patterns in Landscape Use. In G. Touchais, R. Laffineur and F. Rougemont (eds), PHYSIS. L´Environnement Naturel et la Relation Homme – Milieu dans le Monde Égéen Protohistorique. Actes de la 14e Rencontre égéenne internationale, Paris. Institut National d´Histoire de l´Art (INHA), 11–14 décembre 2012. Aegaeum 37: 112–120. Leuven/Liège, Peeters. Pavúk, P. and Pieniążek, M. 2016. Towards Understanding the Socio-political Structures and Social Inequalities in Western Anatolia during the Late Bronze Age. In H. Meller, H. P. Hahn, R. Jung and R. Risch (eds), Rich and Poor. Competing for Resources in Prehistory. Tagungsband des 8. Mitteldeutschen Archäologentages in Halle, 22. – 24. November 2015. Tagungen des Landesmuseums für Vorgeschichte Halle, Band 14/I: 531–551. Halle/ Saale, Landesmuseum für Vorgeschichte. Pernicka, E., Rose, C. B. and Jablonka, P. (eds) 2014. Troia 1988–2008: Grabungen und Forschungen I. Forschungsgeschichte, Methoden und Landschaft. Studia Troica Monographien 5. Bonn, Habelt. Pieniążek, M. 2012. Luxury and Prestige on the Edge of the Mediterranean World: Jewellery from Troia and the Northern Aegean in the 2nd Millennium B.C. and its Context. In R. Laffineur and M.L. Nosch (eds), KOSMOS. Jewellery, Adornment and Textiles in the Aegean Bronze Age. Proceedings of the 13th International Aegean Conference / 13e Rencontre égéenne international, University of Copenhagen, Danish National Research Foundation´s Centre for Textile Research, 21–26 April 2010. Aegaeum 33: 501–508. Leuven/Liège, Peeters. Pieniążek, M. 2016. A Polity in Transition: Troy in the 2nd Millennium BC. In M. Bartelheim, B. Horejs and R. Krauß (eds), Von Baden bis Troia: Ressourcennutzung, Metallurgie und Wissenstransfer. Eine Jubiläumsschrift für Ernst Pernicka. Oriental and European Archaeology 3: 513–534. Vienna, Österreichische Akademie der Wissenschaften. Pieniążek, M. 2018. Foreign Influences and Indigenous Transformations: Late Bronze Age ‘Imports’ in the North Aegean. In S. Gimatzidis, M. Pieniążek and S. Votruba (eds), Archaeology across Past and Present Borders: Fragmentation, Transformation and Connectivity in the North Aegean and the Balkans during the Late Bronze and Early Iron Age. Oriental and European Archaeology 9: 113-138. Vienna, Österreichische Akademie der Wissenschaften. Pieniążek, M. und Aslan, C. 2016. Heroic Past, Memory and Ritual at Troy. In E. Alram-Stern, F. Blakolmer, S. Deger-Jalkotzy, R. Laffineur and J. Weilhartner (eds), Metaphysis. Ritual, Myth and Symbolism in the Aegean Bronze Age. 15th International Aegean Conference held at Vienna 2014: 423–432. Leuven, Peeters. 186

Diane Thumm-Doğrayan et al.: Economy and Storage Strategies at Troy

Pieniążek, M., Pavúk, P. and Kozal, E. 2018. The Troad, South Aegean and the Eastern Mediterranean: Longdistance Connections during the Late Bronze Age. In M. Bartelheim, B. Nessel and D. Neumann (eds), Bronzezeitlicher Transport. Akteure, Mittel und Wege. RessourcenKulturen 8: 375-410. Tübingen: University Press. Pilides, D. 2000. Pithoi of the Late Bronze Age in Cyprus. Nicosia, Department of Antiquities. Rice, P. M. 1987. Pottery Analysis. A Sourcebook. Chicago, University of Chicago Press. Riehl, S. 1999. Bronze Age Environment and Economy in the Troad. The Archaeobotany of Kumtepe and Troy. BioArchaeologica 2. Tübingen, Mo-Vince. Riehl, S. 2006. Unser täglich Brot….Pflanzenproduktion und Ernährung in Troia. In M. O. Korfmann (ed.), Troia. Archäologie eines Siedlungshügels und seiner Landschaft: 297–308. Mainz am Rhein, Zabern. Riehl, S., Marinova, E. and Uerpmann, H.-P. 2014. Landschaftsgeschichte der Troas. Bioarchäologische Forschungen. In E. Pernicka, C. B. Rose and P. Jablonka (eds) Troia 1988–2008: Grabungen und Forschungen I. Forschungsgeschichte, Methoden und Landschaft. Studia Troica Monographien 5: 732–769. Bonn, Habelt. Rigter, W. and Thumm, D. 2004. Ein hohlgeformter Stier aus Troia. Studia Troica 14: 87–99. Roosevelt, C. H., Luke, C. and Ünlüsöy, S. 2014. Kaymakçı Akeoloji Projesi: 2014 Yılı Kazı Sonuçları. Kazı Sonuçları Toplantısı 31/2: 243–268. Roosevelt, C., Luke, C., Ünlüsoy, S., Çakırlar, C., Marston, J., O’Grady, C., Pavúk, P., Pieniążek, M., Mokrišová, J., Scott, C., Shin, N. und F. Slim. 2018. Exploring Space,  Economy, and  Interregional Interaction at a Second-Millennium B.C.E. Citadel in Central Western Anatolia: 2014–2017 Research at Kaymakçı. American Journal of Archaeology 2018, 122, 645–688. Schiemann, E. 1951. Emmer in Troja. Neubestimmungen aus den trojanischen Körnerfunden, Berichte der Deutschen Botanischen Gesellschaft 64: 155–170. Schliemann, H. 1891. Bericht über die Ausgrabungen in Troja im Jahre 1890. Mit einem Vorwort von Sophie Schliemann und Beiträgen von Dr. Wilhelm Dörpfeld. Leipzig, Brockhaus. Schmidt, H. 1902. Die Keramik der verschiedenen Schichten. In W. Dörpfeld, Troja und Ilion. Ergebnisse

der Ausgrabungen in den vorhistorischen und historischen Schichten von Ilion 1870–1894: 243–319. Athens, Beck and Barth. Schoop, U.-D. 2006. Assyrer, Hethiter und Kaskäer – Zentralanatolien im zweiten Jahrtausend vor Christus. In M. O. Korfmann (ed.), Troia. Archäologie eines Siedlungshügels und seiner Landschaft: 29–46. Mainz am Rhein, Zabern. Seeher, J. 2000. Getreidelagerung in unterirdischen Großspeichern: Zur Methode und ihrer Anwendung im 2. Jahrtausend v. Chr. am Beispiel der Befunde in Hattusa, Studi Micenei ed Egeo-Anatolici 42-2: 261–301. Shaw, J. W., and Shaw, M. C. 1993. Excavations at Kommos (Crete) during 1986–1992, Hesperia 62: 129–190. Thater, M. 2015. The pottery of EWA 2 in Northwest Anatolia: Troy I Middle, Troy I Late and Troy IIa. In H. Erkanal (ed.), Associated Regional Chronologies for the Ancient Near East and the Eastern Mediterranean, Western Anatolia. Turnhout, Brepols. Thumm-Doğrayan, D. 2016. Storage Strategies in Early Bronze Age Troy. In E. Pernicka, S. Ünlüsoy, and S.W. E. Blum (eds), Early Bronze Age Troy: Chronology, Cultural Development, and Interregional Contacts. Proceedings of an international conference held at the University of Tübingen, May 8–10, 2009. Studia Troica Monographien 8: 308–318. Bonn, Habelt. Ünlüsoy, S. 2011. Die Stratigraphie der Burg von Troia II.

Uerpmann, H.-P. 2003. Environmental Aspects of Economic Changes in Troia. In G. A. Wagner, E. Pernicka and H.-P. Uerpmann (eds), Troia and the Troad. Scientific Approaches: 251–262. Heidelberg, Springer. Uerpmann, M. 2006. Von Adler bis Zahnbrassen – Der Beitrag der Archäozoologie zur Erforschung Troias. In M. O. Korfmann (ed.), Troia. Archäologie eines Siedlungshügels und seiner Landschaft: 283–296. Mainz am Rhein, Zabern. Voyatzoglou, M. 1974. The Jar Makers of Thrapsano, Expedition 16: 18–24. Wagner, G. A., Pernicka, E. and Uerpmann, H. P. (eds). 2003. Troia and the Troad. Scientific Approaches. Heidelberg, Springer.

187

The Country in and Around the City Looking Back and Forward Dominique Garcia INRAP

Julien Zurbach ENS, UMR AOrOc, IUF

The contributions assembled here directly address both fundamental questions of social archaeology: how societies relate to nature, and how spatial differentiation structures society. Moreover, they concentrate on what has been for a long time an obscure corner, a question which did not attract much study, namely the place of agricultural production and animal husbandry in what we use to call cities, that is, the biggest settlements we know.

the Mediterranean because of changing perspectives on different historical situations. The study of Early Iron age urbanism in Late Hallstatt Central Europe has shown a sustained dynamic since the end of the 20th century. Older perspectives based on the comparison with a model of Classical Mediterranean (that is, Greco-roman) urbanism have been abandoned in favour of a more autonomous appraisal of the characters of this first Central European urbanism. This has a methodological side: it goes away from what Fernandez-Götz (2017: 106) calls ‘checklist approaches’, ‘in which the urban character of a site is determined by its supposed similarities with the supposed ‘standard’ model of classical cities’. It has a historical side as well: it is contrary to diffusionist views where the centre, the Classical Mediterranean, is the active place giving stimuli to others and enabling evolution and progress. In the course of this enterprise, FernandezGötz however does not forget to maintain an interest for the important historical question of relations and possible analogies between Central European and Mediterranean urbanization processes (2017: 107). It is in a sense precisely what is at stake here: to define historical situations without teleological premises renders a more fruitful comparison possible. This perspective is clearly opened now. A possible trap here would be to forget that the ‘classical city’ is also in constant evolution, as Whitelaw notes when analyzing the producer-consumer city debate.

Urbanization in itself cannot be said to be quite new as an archaeological and historical problem. The urban revolution in late prehistory and its instrumental role in the transition to early historical societies has been a theme for a long time. However, it does present some fundamental new inflexions in the last years. Many studies have concentrated on the urban space, the differences and hierarchies it reflects, and its spatial form: to say it briefly, these are questions of urbanism. Urbanization, to put it simply, is the other side of the question, the place of bigger settlements in the wider economy, society and culture, their demographic and political weight. This distinction may be somewhat artificial but it is nevertheless useful and in fact essential to give its right place to the second kind of problems. It goes hand in hand with two different definitions of the city. As is well known, there are a good number of different ways to define what makes a city. On the whole there are two ways of seeing things: the city may be defined through a particular way of organizing space and social interaction, or through a position in a network of settlements and in the wider framework of society. Internal and external criteria both play a role, in fact, and the interaction between these two faces of cities may be a crucial part of what is at stake here.

Seen through these lenses, the contributions assembled here address central questions. ‘Country in the city’ is far from a marginal problem of inherited ‘rural’ aspects in the development of cities. The first fundamental result is in fact the great diversity of the first urban experiences of the Mediterranean. That this result appears now so clearly is a consequence of some original aspects of the congress and the scientific perspective on which it was built. Assembling colleagues from the Aegean and Western Mediterranean was per definition going towards a comparative approach. It did more than that, since it appears that it underlined the importance of regional and micro-regional approaches to the problem of agriculture and husbandry in urban settings. Crete, Anatolia, continental Greece, Languedoc and Catalonia all did experience the development

The heuristic efficiency of the question of ‘country’ in the city may appear more clearly now. Not only does the fact that a part of agricultural production and animal husbandry did take place inside bigger settlements question the original organization of urban spaces, it questions also the relation of the city to its hinterland and its supposed centrality inside a dependent territory. In a more precise context, it may be essential today to address questions of protohistoric urbanization in 188

Dominique Garcia and Julien Zurbach: The Country in and Around the City

of ways of life we would certainly call urban, but it is certainly impossible to reduce all these experiences to a unique process of urbanization. The right scale is therefore not a general opposition – which would allow systematic comparison – between ‘the Aegean’ and ‘the West’, but the regional and micro-regional definition of urban realities and their integration or not of agricultural and animal production activities. This attention for the regional / micro-regional level is also a consequence of the accumulation of data and the introduction of new methods. Even if – in Pullen’s words – we are still lacking ‘significant excavations of Mycenaean towns against which to compare’, and the large-scale horizontal excavation of a Mycenaean town remains an objective of field research, the differences in data between Crete and the continent have been diminishing in the last decades with more horizontal exploration at Mycenaean sites, and the intensive exploration of the Troad and other Aegean regions.

These data are a contribution to a general history of the first Mediterranean cities. They give a clear context for the question of rural activities in these settlements. In the Aegean, the question of urban and rural spaces is per force essentially intermingled with the question of palace towards non-palace economy and society. H. van Effenterre was one of the first to argue, at Malia, for an autonomous development of the town, with the palace being external to both town and country. This has led to many discussions on the nature of the contrast between these three terms, with extreme positions coming to light and researchers holding for instance that there was no town at all, but only palatial sites. The debate however now centers on the kind of agricultural strategy followed by the palace, and on the pertinence of a strict opposition between the palace and what is not the palace – the communities and markets, be they local or more remote. A current model first built by Halstead in the 1990s holds that the palace concentrates on a shirt list of cereals and fruits to enable its ration system to function, and that it develops a new kind of agricultural strategy with new fields, relying on more animal labor. This is in opposition with small-scale agriculture on the level of gardens and small fields, relying on a high diversity of crops as risk-buffering strategy and intensive human (family) labor. This model finds a quite striking confirmation in Knossos through the study by Nitsch et al. where isotope analysis on charred crop remains are taken as basis for the theory of extensification of cereal cultivation linked to the Bronze Age urbanization, with cereals on well-watered lands, separated from pulses which were probably cultivated in garden-like areas subject to labor-intensive cultivation around the town. It is extremely important to find such results with charred remains from the Unexplored Mansion in Knossos, probably a part of the central palatial site at the time of its destruction in LM II. Such garden agriculture is by no way limited to farmsteads or modest houses. The possibility should be kept, however, that the crop remains from the Unexplored Mansion were products from taxation, and therefore coming from areas located away from the centre.

It is certainly expected from any collection of archaeological and historical studies nowadays that it shall reflect in some way the transformation, both in quantity and quality, of information available. In the case of topics like agriculture, husbandry, subsistence strategies, and the nature of urban and rural spaces, it allows generalizing work to take new dimensions. All articles rely on the broadest spectrum of sources available. The essential role plaid by the study of plant remains is notable, and it goes together with methods which have now a longer history like pottery studies with chemical analysis and historical criticism of texts when available. This diversification is also linked to an intensification of data at microregional level, as seen in the diverse landscapes which come into the centre of discourse here, from Catalonia to the Troad. The spectrum of tools used here encompasses also notions from the social sciences, as shown most notably by the Whitelaw when discussing site-catchment analysis, Near Eastern social and economic history from cuneiform sources, and the producer/consumer city debate in Roman history. It should certainly be underlined that small-scale analysis is not only useful on a geographical level, it has to be coupled with precise time-scale variations, like the many contrasting histories of sites and regions in Neopalatial Crete tend to show. In that sense ‘country in the city’ is not only a question of long established structures but of changes on all time-scales, as shown by the plant remains from Knossos – contrasting Neolithic data with Bronze Age ones – or the changes in the city of Malia from Protopalatial to Neopalatial times. To these changes one should add the setbacks of urbanization, which was by no means a progressive, linear evolution. The end of the Bronze Age in the Aegean is of course the main episode of de-urbanization but certainly not the only one.

On the continent it seems that one concentrates nowadays on refining the model or discussing its presuppositions. At Mycenae and Kalamianos – on different scales, regional in the first case, local in the second – more diversified and elaborate patterns of interaction are being defined. The main point would be that it is questionable to distinguish the palace and cut it off from other actors when discussing subsistence strategies: there would be no point in attributing one subsistence strategy to the palace and other strategies to other actors; on the reverse, the palace would have a part to play in every strategy, as shown among others by the wild animals or aromatic plants at Mycenae. The 189

Country in the City role possibly plaid by the palace in the organization of the town of Kalamianos is equally very diverse, from an initial input to a possible overview and appropriation of part of the produce from the territory or the agricultural quadrant in the town itself; but the local community and its dynamics are central. One could wonder if it would not be helpful here to question the other part of the opposition, namely the family farm or local community. What where these actors like? The results of surveys around Mycenae, and the households from Kalamianos, do give some hints. But we do not need to attribute to them a subsistence strategy of a riskbuffering nature, that is, as a reaction to environment, and not more active perspectives – it has been shown that family farms were perfectly able to develop real expansion strategies, and the surplus assembled by the palace is perhaps not the only cause and the last word on Mycenaean expansion, not to say growth. As Christakis writes, ‘rural communities were not passive providers of agricultural wealth, as has traditionally been argued, but dynamic groups with their own needs and aspirations’.

It seems clear from the contributions assembled here that we have to distinguish between Cretan and continental societies in the Late Bronze age. An important fact is that one could read the contribution on the Troad as revealing a situation nearer to the Cretan one than to the Late Helladic one, with a denser town and a neater contrast between urban and rural communities and spaces, not excluding of course storage and transformation in urban settings. In the North-Western Mediterranean (Southern Gaul and Eastern Iberia), during the first millennium before our era, relations between habitat and the related agricultural production area shifted noticeably; reference can even be made to a radical change in paradigm. From the coastal and lagoon area to the first mountainous foothills, the placement of the sites at the end of the Bronze Age and the beginning of the Iron Age reflects the wide variety of terroirs covered and farmed by the dwellers of the times. The diversity in the types of sites recognisable is significant: agglomerations in the high- as well as low-lands, dwellings huddled together in small groups, caves, isolated settlements, etc. Though the clusters vary in surface area, generally speaking, they are extremely small: from 0.1 to 30 hectares, with an average of 1 hectare. Some sites offer proto-urban structures to the eye: enclosures in the form of a basic rampart or ditch, retaining walls such as in the Baou-Roux (Bouches-du-Rhône), etc. Others, the most frequently-encountered, reflect a sedentary lifestyle still in the making.

In Crete and the Aegean islands, however, the question seems to be a quite different one. Contributions on Crete and the islands concentrate on three parameters: the storage capacities, the transformation of agricultural products in urban settings, and the social and economic relationship of the city with the agropastoral system. Placing agricultural or pastoral areas at the heart of the city itself is apparently not a part of the research agenda here. We may contrast this to the conclusions drawn at Kalamianos or Mycenae: that, in the first case, ‘because agricultural activities can and do take place within an ‘urban’ setting, the dichotomy of rural versus urban is an inappropriate construct when discussing Mycenaean agriculture’, or, in the second, ‘it is more informative to examine the landscape as one of highly integrated social and economic interests with effects extending beyond our perceived confines of ‘city and country’ or ‘palace and hinterland’’. When discussing Cretan agriculture in the long-term (Nitsch et al.), concentrating on the place of agricultural production in society (Christakis) or on the variation of space organization at Malia (Muller and Pomadère), the contrast between city and country seems to be a useful category for research on Minoan society, even if it has to be qualified in some way. The importance of agropastoral activity at Neopalatial Malia is for instance contrasted to an earlier, Protopalatial, more clearly urban centre, as is the field production of cereals against earlier, Neolithic garden production in the case of Knossos. Christakis finds on Crete a situation that has similarities to the one Sarpaki describes for Santorini, with transformation and storage in the big houses of a dense town where every household probably had links to rural farms or land.

The subsistence economy appears to have been based on agro-pastoral practices, in which livestock dominates. These were most probably societies engaged in a ‘domestic mode of production’. This system was designed to preclude any form of dependency (economic and therefore political) on neighbouring groups and according to ethnological observations, was fundamentally hostile to the formation of surpluses. However, while no market can be said to have formed, groups (neighbouring or remote) found themselves engaged in more or less intense goods-exchange relationships. It has been suggested that these findings, both archaeological and environmental, should be reconciled with the description of the agrarian slashand-burn (also known as ‘long fallow’) system. In this system, crops are grown in varied wooded areas, then planted on land cleared in advance using slashing techniques. The resulting plots were farmed for one, two or three years, rarely more, after which they were abandoned to the forest fallow land for one or more decades before being cleared and cultivated again. This farming practice always brought with it a strong surge in population and continued for as long as there remained accessible, yet uncleared wooded land. In 190

Dominique Garcia and Julien Zurbach: The Country in and Around the City

the long run however, the system brought about a dual crisis – ecological (deforestation, erosion, etc.) and subsistence-related – which was only overcome with the introduction of the fallow land system and light horse-drawn farming, the widespread use of which, in southern France, probably does not pre-date the middle of the 6th century BC. In this type of primitive socio-economic organisation, it is not the ‘city that is in the countryside’, but the habitat that occupies the countryside, and the agro-pastoral spaces that ‘accompany’ the movements – seasonal or otherwise – of the populations.

depending on the season. As we demonstrated for the central Languedoc, it can therefore be hypothesized that the low-hierarchy village territories of the late Bronze IIIb were discontinuous and temporary, included in a landscape marked by the presence of isolated graves and necropolises, each a landmark on a territory more sensed and experienced than dominated, a form of ‘archipelago territory’. The latter can only be a low-density establishment, with a flexible territorial footprint liable to give an external group a right of use on its land. The granting of this right was the responsibility of the king who, thereby truly played a political role: he controlled the ethnic territory and managed the development of its population. By means of selected young women in his lineage, he would periodically initiate a form of organised offspring production. A model of this territorial dynamic, more social than economic, can be put forth. The village is placed at the centre of its terroir: it is a deforested space, developed for exclusively local needs, probably over a radius of less than 5 km, i.e. less than an hour’s walk from the houses. This food production space formed a landscape that could be directly seen from the habitat; on flat land, it is estimated that the visual scope extended 5 km. It is beyond this first territorial ring, i.e. along the edges of the ethnic territory that a new establishment could be created – through offspring. Different types of ties (physical, symbolic, etc.) between the different settlements were then created by pathways and paths drawn already long ago for grazing purposes, by the lineage lines between the different communities and, lastly, by unifying funeral or worship complexes. Gradually, this dynamic would come to structure the region and build the group’s cultural identity.

In the subcurrent societies that practise this type of economy, apart from dwellings, adjoining gardens and possible long-term plantations, which were subject to a much-applied right of use similar to a form of private ownership, the village territory was open to use by all families. Each family was allocated, each year, by the recognised village authority, wooded plots of land suitable for their needs. This right of use lapsed with the last harvest and the land abandoned with longterm wooded fallow returned to the common domain. As long as the village territory was sparsely populated and the land to be cleared was abundant, the right of use was even easily granted to potential newcomers. This temporary right of use often became a permanent right of use when permanent plantations were set up, or when, due to the increase in the population or deterioration in part of the land, the duration of the fallow period was reduced to the point that the exploitation of a plot by the same family tended toward the continuous. This settling model therefore does not suggest a fixed territorial footprint. The communities were groups limited to a few families that farmed a restricted terroir, for at most one generation. This relative sedentariness and fragmentation by community is believed to result from the apparent homogeneity of the cultures of the late Bronze IIIb, despite stretching across vast geographical expanses (from the Ebro to the Rhône).

After a phase of exploration by Phoenician, Etruscan and/or Greek traders in the 8th-7th centuries BC, the shores of the north-western Mediterranean were to be marked by a ‘colonial’ phase characterised by the establishment of coastal sites (Marseille not far from the mouths of the Rhône, Agde at the mouth of the Hérault, Rosas and Emporion on the Catalan coastal plain, etc.). A spectacular development of the sedentary habitat was then observed along a coastal strip around 50 km wide. This type of agglomeration reflected a desire to control and farm complementary terroirs; it marked the abandonment of agriculture on long fallow, in favour of the spread of the set-aside system and light horse-drawn cultivation. It is believed that these sites, often small or medium-sized, were managed by chieftaincy-like communities.

In addition to the area temporarily farmed for agricultural purposes, the authority of the populations, communities of people or cultural groups, was probably exercised more on a vast set of communication channels dotted with specific markers (water points, graves, etc.), raw material mining zones (minerals, salt, etc.) than on a delimited area, encompassing habitat, terroirs and confines. Similarly, it can be deduced that the sense of ownership related more to livestock than it did to land: in such societies, herds formed a source of capital, generally managed collectively, which grew naturally with reproduction, while the few cultivated fields were family properties whose crops constituted income. Indeed, in these pastoral-style societies, ownership did not overlap with territory: the surfaces, being larger, could be occupied by different groups

The indigenous elites were able to surround themselves with clientèle that could be mobilised for collective work, for example the construction of enclosures. In the necropolises, the large number of weapons (iron offensive and bronze defensive) appearing in most male graves indicates that defence was at this stage 191

Country in the City the concern of all the men in the community, rather than the prerogative of a warrior caste. This period was also marked by an increased social hierarchy both in the south-east of France where only exceptional individuals benefited from a tumulus and in Western Languedoc where the study of Catalan or Languedoc necropoles shows ‘an acute pyramid social structure’.

From Provence to southern Catalonia, this system of integrating agrarian activities into the urban space would be in use from the 6th century BC to the 2nd century BC, where under the influence of Roman economic forms, village or urban activities would be less closely associated with the production, storage and processing of agricultural and pasture products.

This network of habitats was instituted in two ways. The first was marked by the grouping of populations in sedentary habitats that occupy spaces with varied potential, generally the edges of foothills or valleys, or even the mouths of rivers. These sedentary settlements tell of the adoption of new agrarian practices and brought with them significant demographic growth. The second movement was triggered by the commercial activity in the Mediterranean – Phoenician, Etruscan but especially Phocaean – which would reinforce the role of lagoon sites and river mouths by conferring an economic role upon them, that of centres for product exchange and handling (storage, transfer, etc.). This would bring about the first movements of product dissemination, and therefore flows of goods and people between the coast and the hinterland, along the natural circulation thoroughfare, in particular the valleys of the main coastal rivers. These two movements resulted in the creation of numerous sites, particularly in the last third of the 6th century BC and the first quarter of the 5th century BC.

On the whole, one may draw the following conclusions and perspectives from the essays collected in this volume. The first one would be that a sound social archaeology may be possible only by taking into consideration economic factors. These are not only economic ‘structures’, as it is often possible to read, but economic trends and conflicts at all scales and time-scales. As E. Hobsbawm one wrote, ‘identity is not enough’: in our case, let us simply state that feasting is not enough, elite strategies and collective cultural practices are not enough to understand what makes up a society. This means also that we should abandon views of societies without conflict, where peaceful negotiation seems to solve every problem. Fr. Sigaut has made the case for this: ‘it is only about taking seriously the fact that humans have to eat every day,, and also to build houses, find clothes, and so on; and that the specific way they do this at a particular time and place gives the most part of the social structures’. The second one is that a pan-Mediterranean perspective is not only possible but necessary. Only through this kind of dialogue may the really important contrasts be defined, the really important trends by analyzed. The third one is the necessity to elaborate one more set of data in the form of a rural archaeology, an archaeological investigation of cultivated areas, gardens and fields, terraces and agricultural equipment, which remains perhaps the last obscure corner in the available data.

During this period, the countryside would take up a place within the urban space. At the foot of the ramparts or the inner city walls, numerous grain storage areas can be found. These fields of silos have been recognised mainly in Spanish Catalonia, Roussillon and Western Languedoc as far as the Lattes region in the Hérault. Cereal reserves were protected and their management (consumption, exchange with Mediterranean traders, seeds, etc.) controlled by the inhabitants of the urban centre. Grain processing – milling or grating – was generally carried out in the dwellings. It was also within the walls of the city, as shown by the excavations of Lattes in Languedoc, that livestock was protected. No specific places are known for this, but streets and small squares appear to have been occupied by sheep and cattle.

Reference Fernandez-Götz, M. 2017. Discussing Iron Age Urbanism in Central Europe: some thoughts. In S. Stoddart (ed.), Delicate urbanism in context: settlement nucleation in Pre-Roman Germany, Cambridge, McDonald Institute Monographs, 105-110.

192

The validity of an opposition between rural and urban spaces is an important question for our societies; this question has been raised since the radical transformations of the 20th century and the so-called ‘end of the peasants’. In this context it becomes also a question for archaeologists and historians. Country in the City: Agricultural Functions in Protohistoric Urban Settlements (Aegean and Western Mediterranean) assembles contributions on the place of agricultural production in the context of urbanization in the Late Bronze and Early Iron Age Mediterranean. The contributions concentrate on the secondmillennium Aegean and the protohistoric northwestern Mediterranean. They offer a reflection on the nature of urbanization and its consequences for rural spaces near cities and on the many ways in which rural spaces and agricultural activities may be intertwined with urban spaces – a reconsideration of the very nature of urbanism. A deliberate accent is laid on the comparative perspectives between different regions and periods of Mediterranean protohistory, and on the integration of all kinds of sources and research methods, from texts to survey to environmental archaeology. Highlighted throughout are the original paths followed in the Peloponnese or in the Troad with regard to the Minoan model of urbanization, and the many aspects of Minoan urbanization, and many regional differences in Languedoc vis-à-vis Catalonia. Thus a new perspective on Mediterranean urbanization is offered. Dominique Garcia is Professor of Archaeology at the University of Aix-Marseille and, since 2014, has been president of the lnstitut national de recherches archéologiques préventives (National Institute of Preventive Archaeological Research). Raphaël Orgeolet is Senior Lecturer in Mediterranean Bronze Age Archaeology at Aix-Marseille University. His main research interests focus on settlement, funeral practices and society. He has taken part in various archaeological projects in the Mediterranean and especially in the Aegean and is now leading the excavations of the Neolithic and Bronze Age site of Kirrha in Mainland Greece. Maia Pomadère is a Senior Lecturer in Aegean Archaeology at the University of Paris 1-Panthéon Sorbonne and member of the UMR 7041-ArScAn. Her research interests encompass Aegean Bronze Age and Early Iron Age archaeology, especially architecture and funerary practices. She has been directing an archaeological excavation in the Minoan town of Malia in Crete since 2005, and is codirecting a geoarchaeological project on the same site. Julien Zurbach is Senior Lecturer in Greek history at the ENS Paris. He is working on agricultural practices, land distribution and workforce in the Aegean world from the Late Bronze Age to the Archaic period. He concentrates particularly on Mycenaean epigraphy and has led field projects in Kirrha (Phocis) and Miletus (Ionia).

Archaeopress Archaeology www.archaeopress.com