MANFRED BIETAK, ERNST CZERNY, IRENE FORSTNER-MÜLLER (EDITORS)

CITIES AND URBANISM IN ANCIENT EGYPT ÖSTERREICHISCHE AKADEMIE DER WISSENSCHAFTEN DENKSCHRIFTEN DER GESAMTAKADEMIE, BAND LX

UNTERSUCHUNGEN DER ZWEIGSTELLE KAIRO DES ÖSTERREICHISCHEN ARCHÄOLOGISCHEN INSTITUTES

HERAUSGEGEBEN IN VERBINDUNG MIT DER KOMMISSION FÜR ÄGYPTEN UND LEVANTE DER ÖSTERREICHISCHEN AKADEMIE DER WISSENSCHAFTEN VON MANFRED BIETAK

BAND XXXV ÖSTERREICHISCHE AKADEMIE DER WISSENSCHAFTEN DENKSCHRIFTEN DER GESAMTAKADEMIE, BAND LX

MANFRED BIETAK, ERNST CZERNY, IRENE FORSTNER-MÜLLER (EDITORS)

CITIES AND URBANISM IN ANCIENT EGYPT

Papers from a Workshop in November 2006 at the Austrian Academy of Sciences Vorgelegt von w. M. MANFRED BIETAK in der Sitzung am 12. Dezember 2008

Gedruckt mit Unterstützung durch das Holzhausen-Legat der Österreichischen Akademie der Wissenschaften

British Library Cataloguing in Publication data. A Catalogue record of this book is available from the British Library.

Die verwendete Papiersorte ist aus chlorfrei gebleichtem Zellstoff hergestellt, frei von säurebildenden Bestandteilen und alterungsbeständig.

Alle Rechte vorbehalten ISBN: 978-3-7001-6591-0 Copyright © 2010 by Österreichische Akademie der Wissenschaften Wien Grafik, Satz, Layout: Angela Schwab Druck: Wograndl Druck GmbH, 7210 Mattersburg http://hw.oeaw.ac.at/6591-0 http://verlag.oeaw.ac.at

Printed and bound in INHALT

Contents ...... 5 Bibliographical Abbreviations ...... 7 MANFRED BIETAK, ERNST CZERNY and IRENE FORSTNER-MÜLLER Preface by the Editors ...... 9 MANFRED BIETAK Houses, Palaces and Development of Social Structure in ...... 11 ERNST CZERNY Fragments of Information. Observations Concerning the Architectural Layout of the Middle Kingdom Settlement at cEzbet Rushdi ...... 69 FLORENCE DOYEN La résidence d’élite: un type de stucture dans l'organisation spatiale urbaine du Moyen Empire . . . . . 81 IRENE FORSTNER-MÜLLER Settlement Patterns at Avaris. A Study on Two Cases ...... 103 ANGUS GRAHAM Islands in the Nile: A Geoarchaeological Approach to Settlement Location in the Egyptian Nile Valley and the Case of Karnak ...... 125 STAN HENDRICKX, DIRK HUYGE and CLAIRE NEWTON The Walls of Elkab ...... 145 MARK LEHNER and ANA TAVARES Walls, Ways and Stratigraphy: Signs of Social Control in an Urban Footprint at Giza ...... 171 WOLFGANG MÜLLER Urbanism in Graeco-Roman Egypt ...... 217 CORNELIUS VON PILGRIM Elephantine – (Festungs-)Stadt am Ersten Katarakt ...... 257 STEVEN SNAPE Vor der Kaserne: External Supply and Self-Sufficiency at Zawiyet Umm el-Rakham ...... 271 KATE SPENCE Settlement Structure and Social Interaction at El-Amarna ...... 289 CAROLA VOGEL Storming the Gates? Entrance Protection in the Military Architecture of Middle Kingdom . . . . 299 JOCHEM KAHL Frauen aus Nilschlamm – ein Beitrag zur Anthropologie der Stadt ...... 321

BIBLIOGRAPHICAL ABBREVIATIONS

Ä&L Ägypten und Levante CASAE Supplément aux Annales du Service des Anti- ÄA Ägyptologische Abhandlungen, Wiesbaden quités de l’Égypte, Cahiers, Cairo ÄAT Ägypten und Altes Testament. Studien zu E&L see Ä&L Geschichte, Kultur und Religion Ägyptens und EA Egyptian Archaeology. The Bulletin of the Egypt des Alten Testaments, Wiesbaden Exploration Society, AASOR Annual of the American Schools of Oriental Research, EEF Egypt Exploration Fund, London New Haven EES Egypt Exploration Society, London ADAIK Abhandlungen des Deutschen Archäologi- EtudAlex Études alexandrines. Institut français d’archéo- schen Instituts, Abteilung Kairo, logie orientale du Caire, Cairo AERAGRAM AERAGRAM. The official newsletter of Ancient FIFAO Fouilles de l’institut français d’archéologie Egypt Research Associates, Cambridge, MA orientale du Caire, Cairo ÄF Ägyptologische Forschungen, Glückstadt, Ham- GHP Egyptology Golden House Publications Egyptology burg, New York Series, London AJA American Journal of Archaeology, Baltimore, ab GOF Göttinger Orientforschungen, Wiesbaden 1897: Norwood HÄB Hildesheimer Ägyptologische Beiträge, Hildes- Antiquity Antiquity: A Quarterly Review of Archaeology, heim Oxford JARCE Journal of the American Research Center in Egypt, AOF Altorientalische Forschungen, Berlin Boston APF Archiv für Papyrusforschung, Stuttgart und JAS Journal of Archaeological Science, London, New Leipzig York ARCER American Research Center in Egypt Report, JEA Journal of Egyptian Archaeology, London Malibu JNES Journal of Near Eastern Studies, Chicago ARCUS ARCUS. Berichte aus Archäologie, Baugeschichte und Nachbargebieten, Potsdam JRA Journal of Roman Archaeology, Portsmouth, Rhode Island ASAE Annales du service des antiquités de I’Égypte, Cairo Kush Kush. Journal of the Sudan Antiquities Service, AV Archäologische Veröffentlichungen des Deut- Khartoum schen Archäologischen Instituts Kairo, Mainz LÄ Lexikon der Ägyptologie, Wiesbaden BAe Bibliotheca Aegyptiaca, Brussels MBPF Münchner Beiträge zur Papyrusforschung (und BAR British Archaeological Reports, Oxford antiken Rechtsgeschichte), München BASOR Bulletin of the American Schools of Oriental Research, MDAIK Mitteilungen des deutschen archäologischen Instituts, New Haven Abteilung Kairo, Berlin/Mainz BCE Bulletin de liaison du group international d’étude de MDOG Mitteilungen der Deutschen Orientgesellschaft, Berlin la céramique égyptienne, Cairo MEES Memoires of the Egypt Exploration Society, BdE Bibliothèque d’Étude, Institut Français d’Archéo- London logie Orientale du Caire, Cairo MIFAO Mémoires de l’Institut Français d’Archéologie BES Bulletin of the Egyptological Seminar, New York Orientale, Cairo BIE Bulletin de l’Institut d’Egypte, Cairo Morgenland Morgenland. Darstellungen aus der Geschichte BIFAO Bulletin de l’institut français d’archéologie orientale, und Kultur des Ostens, Leipzig Cairo MRE Monographies Reine Élisabeth. Fondation BMRP British Museum Research Publications, London égyptologique Reine Élisabeth, Bruxelles (seit 2004, formely British Museum Occasional Papers 1978ff) ÖJh Jahreshefte des Österreichischen Archäologischen Insti- tuts in Wien, BSEG Bulletin de la société d’égyptologie de Genève, Genf OLA Orientalia Lovaniensia Analecta, Leuven BSGE Bulletin de la Société de Géographie d’Égypte, Cairo PAM Polish Archaeology in the Mediterranean, Warsaw BSFE Bulletin de la société française d’egyptologie, Paris PBA Proceedings of the , Oxford CAJ Cambridge Archaeological Journal, Cambridge PdÄ Probleme der Ägyptologie, Leiden, Boston, Köln CdE Chronique d’Egypte, Bruxelles PEQ Palestine Exploration Quarterly, London CRAIBL Comptes rendus de l’Académie des Inscriptions et Belles-Lettres, Paris PSBA Proceedings of the Society of Biblical Archaeol- ogy, London CRIPEL Cahiers de Recherches de l’Institut de Papyrologie et d’Egyptologie de Lille, Lille RdE Revue d’égyptologie, Cairo; vol. 7ff: Paris 8 Bibliographical Abbreviations

SAGA Studien zur Archäologie und Geschichte Altägyptens, VA Varia Aegyptiaca, San Antonio Heidelberg WVDOG Wissenschaftliche Veröffentlichungen der Deutschen SAK Studien zur Altägyptischen Kultur, Orientgesellschaft, Berlin, Leipzig SAOC Studies in Ancient Oriental Civilization, Chicago Wepwawet Wepwawet. Research Papers in Egyptology, Lon- SDAIK Sonderschriften des Deutschen Archäologi- don schen Instituts, Abteilung Kairo, Mainz World Archaeology World Archaeology, London SOAS School of Oriental & African Studies, London YES Yale Egyptological Studies, New Haven TAPA Transactions of the American Philological Associa- tion, Philadelphia ZÄS Zeitschrift für ägyptische Sprache und Altertumskun- de, Leipzig, Berlin UGAÄ Untersuchungen zur Geschichte und Alter- tumskunde Ägyptens, Leipzig, Berlin ZDPV Zeitschrift des Deutschen Palästina-Vereins, Wiesba- den Urk. Urkunden des ägyptischen Altertums, ed. G. STEINDORFF, Leipzig, Berlin, 1906–1958 Zetemata Zetemata. Monographien zur Klassischen Alter- UZK Untersuchungen der Zweigstelle Kairo des Öster- tumswissenschaft, München reichischen Archäologischen Institutes, Wien PREFACE BY THE EDITORS

In 1993 the the Austrian Accademy of Sciences pub- one of the backbones of any urban research in Egypt. lished the volume “House and Palace in Ancient Therefore, it appears in the present volume with Egypt”, which included the acts of a symposium three articles, all written by the editors of the volume, organised for the 30-years jubilee of Austrian archae- respectively. Both the study by M. Bietak and by ological activities at settlement sites in Egypt and I. Forstner-Müller cover the overall area of the site in Nubia. This still essential book provided a more or discussing structural problems. Bietak’s article focus- less complete overview of the activities in settlement es on diffentiations and hierarchy in the develloping archaeology in Egypt of those days. However, settle- structures, whereas Forstner-Müller’s paper is more ment research had only started by then to be regard- concerned with the different urbanistic concepts ed as an equally valuable aspect of Egyptian archae- applied through different phases in the history of the ology besides the traditional research in sacred or town-site. Finally, E. Czerny’s paper provides a funerary spheres. The articles of the volume were detailed insight into the structure of the 12th dynasty mostly concerned with the basic elements of any set- settlement at cEzbet Rushdi, at the north-eastern tlement: houses and palaces. edge of the Tell el-Dabca site. In November 2006, on the occasion of the 40th However, whenever Middle Kingdom settlements anniversary of the Austrian Archaeological Institute’s are discussed, the site of Lahun, excavated already in excavations at the large urban site of Tell el-Dabca, the 19th century, must be considered. It seems to be and triggered by the research project “Stadtentwick- an inexhaustable resource for all kind of questions lung im Nildelta”, project number P 21074-G08 concerning urbanistic research. Only recently, a new financed by the Austrian Science Fund, another sym- survey started at the site by a Hungarian-Egyptian posium was held. Again, an overall survey of settlement team with the aim to continue previous Canadian archaeology in Egypt should be provided, but this activities. The available plan of the site, originally time, not houses, but settlements as such and their published by W.M.F. Petrie in 1890, has often been respective structures should mainly be considered. considered as confusing with little reliability, and Most of the given papers are to be found in the only the 10 large mansions at the northern side of the present volume, which, thus, might again be consid- town repeatedly attracted scholarly attention. Howev- ered a synopsis of ongoing settlement-research in er, in the last years several important studies Egypt. Moreover, it becomes obvious which progress appeared on the architecture of Kahun (Lahun), has been archieved since the previous publication, among them an article by F. Arnold (Baukonstruk- and which topics and questions are in the focus of tion in der Stadt Kahun. Zu den Aufzeichnungen current research. Ludwig Borchardts)1 and one by R. Frey and E. Knud- Most of the papers of the 2006 – conference are to stad (The Re-examination of Selected Architectural be found in the present volume, which, thus, might Remains at El-Lahun).2 In the present volume, F. again be considered a synopsis of ongoing settle- Doyen presents a thorough analyses of the structure ment-research in Egypt. Therefore they focus mainly of different types of houses, making clear, by the way, on site evidence. Only two contributions (M. Lehner that Petrie’s plan is perfectly workable for detailed and W. Müller) deal also with urban theories and studies. She concluded that the size of the large man- models – a subject which will be addressed in a future sions was used as a kind of “modul”, from which all symposium of the Austrian Academy of Sciences. other formats in the town can be deduced, what can The site of Tell el-Dabca, having been thoroughly be shown to be also the case at the town-site of Aby- studied for four decades, can justly be considered dos-south.

1 2 In P. JÁNOSI (ed), Structure and Significance, UZK 25, Wien JSSEA 34 (2007) 24–65. 2005, 77–104. 10 Preface by the Editors

Similar to Lahun, also the famous town site of and Roman times, beginning with the first Greek Amarna has been excavated since the 19th century, settlements already in the 26th dynasty. and continues to be in the focus of archaelogical Further papers are more concerned with special research. K. Spence in the present volume also pro- problems, such as S. Hendrickx’s et al. study of the vided an analysis of structure and interpreted the enigmatic and much discussed walls of El-Kab, or K. “Amarna-house” as an entity reflective of hierarchy Vogel’s paper on the gates and defensive features of and social control. the Middle Kingdom fortresses in Nubia. That typical Egyptian architectural structures are Far away from the Nile valley, west of Mersa consistent with the highly hierarchical society of Matruh, lies the New Kingdom fortress of Zawiet Ancient Egypt can already be seen in the Old King- Umm el-Rakham. It was built in Ramesside times dom. M. Lehner and Ana Tavares, in a detailed against the growing immigration pressure of Libyan study of the “Pyramid-construction-town” at Giza tribes. S. Snape concentrates in his article on the and its satellites, present a most impressive insight problems of provisioning versus self-supply of this into the living conditions of those concerned with westernmost military outpost of the Egyptian Empire. the building of a pyramid. The barracks for the A completely different research was done by A. workmen’s dwellings could clearly be distinguished Graham, who studied the geomorphological condi- from a quarter of those in charge of food-prepara- tions of island-formation in the Nile at Luxor. By tion, storage facilities for animals and cereals for auger drilling he was able to demonstrate that the pri- food supply, and finally a quarter for the adminis- mal Karnak temple was founded on an island, which tration, where high-ranking officials were housed. only later accreted to the eastern border of the Nile. The architecture of the complex is perfectly adapt- It is intriguing to imagine that the site of the present ed to the purpose of entertaining a large construc- Sacred Lake once was a natural depression filled with tion site and shows a very high degree of analytical water, typical for recently emerged islands. planning. Numerous elements providing for social The only non-archaeological article in the vol- control might be considered an essential element of ume is an essay of J. Kahl’s on the “Anthropology of the architectural layout. the Town”. He collected evidence in many texts that Excavations at the town of Elephantine at the 1st towns were allegorised as women, having human cataract provide a unique sequence of urban stages properties such as feelings or memory. from a First Dynasty-fortress through the flourishing The editors hope that the present volume will be town of the Late Period and even later. In C. v. Pil- acclaimed for its fresh and stimulating approaches grim’s paper a strong focus lies on the diachronic to both newly evolved and since long discussed ques- aspect of settlement devellopment. On the opposite tions of settlement archaeology and urbanistic stud- riverbank of the Nile, modern Assuan covers the ies in Egypt. remains of ancient Syene. Important ongoing exca- vations currently produce new information on the town of the Graeco-Roman period. W. Müller pre- Manfred Bietak sents these up-to-date results in the frame of a thor- Ernst Czerny ough study of the urbanism in Egypt during Greek Irene Forstner-Müller HOUSES, PALACES AND DEVELOPMENT OF SOCIAL STRUCTURE IN AVARIS1

Manfred Bietak

I. INTRODUCTION a long time and, even after more than four decades In the following article, I will be endeavouring to of research at Tell el-Dabca, we only have windows of illustrate a number of points which may be of signifi- stratigraphic development at our disposal. In our cance when trying to assess a settlement. This study reconstruction of town development we have to rely will focus on the site of Tell el-Dabca, drawing on the on such windows and an overall view which also fac- wealth of experience gained at this spot in over 40 tors in the results of the surface surveys. years of fieldwork.2 Size and types of buildings, especially when viewed Settlements should always be viewed in a wider in combination with the vertical development, show perspective, looking at the ancient environment, try- the rank and standing of inhabitants in comparison ing to position it within the primeval landscape, to other districts which have larger or smaller build- showing the river-courses, the backwaters, marshes, ings. What is of specific importance is the compara- overflow lakes, roads and, in the case of towns in the tive assessment of compactness of settlement quar- Delta, its position in relation to the sea. Such map-ori- ters. The degree of inner development concerning ented evaluation may reveal to some extent the func- the intensity of using the space can be measured and tion of a settlement. compared. Such studies may help to recognise the Geophysical surveying is a great innovation and social differentiation and shifts in the social develop- gift to archaeology as it reveals relatively fast large ment of a town. It is most interesting to see if evi- areas of settlement structure, its boundaries and dence at the surface perpetuates previous develop- topographical features such as Nile channels and ment at the same spot or if a dynamic development lakes. What it does not show is the date of the struc- has changed the topography of a town. The difficulty tures. Buildings visible in the evaluation records of is that we have only a patchwork at our disposal and multiple strata sites may show architectural features are obliged – especially working with big towns – to from different periods in one image, even if they infer insights from limited area-samples. have the same orientaion. Ancient strata are not nec- essarily horizontal but also covered hills and depres- II. SIZE, POSITION AND FUNCTION OF SETTLEMENT sions. This is also true in the Delta with its turtlebacks The size and geographical position of a settlement and levees. Levelling surfaces for agriculture may shows its importance and, to some extent, the function take away parts of settlements and expose older stra- of a town. It helps to determine the position of the ta side by side with younger ones. We are also town within the hierarchy of settlements. The growth unaware of the development of settlements which of Tell el-Dabca from a moderate provincial town of ca. may have grown according to a horizontal stratigra- 15–25 ha of the 12th Dynasty to a town of c. 75 ha in phy. What cannot be recognized by such surveys is the the late Middle Kingdom and finally to a maximum of vertical sprawl of a settlement and hidden features c. 250 ha in the early Period shows that the such as turtlebacks and water branches covered by place all of a sudden gained major political impor- later settlements. Such details can only be revealed by tance (Fig. 1). Originally we estimated the size of the excavation and the hidden topographical features by town in the Hyksos Period at c. 250 ha, but recent sur- coring. Those activities at a multiple stratum site take veys have added to the precision of topographic fea-

1 For English editing I am indebted to Adrian Melman, for FORSTNER-MÜLLER 2005; 2006; 2007; BIETAK, MARINATOS & footnote editing Irene Kaplan, for providing the illustra- PALYVOU 2007; FORSTNER-MÜLLER 2008; HABACHI 2001; HEIN tions to Nicola Math from SCIEM 2000. & JÁNOSI 2004; KOPETZKY 2009; MAGUIRE 2009; MÜLLER 2008; 2 For the site see BIETAK 1996a; 2001. Since then: ASTON 2004; SCHIESTL 2009. BADER 2009; BIETAK, DORNER & JÁNOSI 2001; BIETAK & 12 Manfred Bietak tures, especially recognizing water basins, old water than in the New Kingdom and can be estimated at a channels and flood areas which are unsuitable for maximum of 2000 m, perhaps only 1500 m. This building.3 We therefore have to correct the size of the would limit the town size to a maximum of 80 ha. settlement to 222–228 ha minimal size when including However, we have to bear in mind that Memphis, as the area of cEzbet Machali4. It is, however, perfectly the name reveals to us, was most likely a twin town possible that c. 20 ha more could be expected east of and we also have to take into consideration that, cEzbet Mehesin on the other side of the old water besides the town around the Ptah temple called ¡wt- branch F/3 and not yet investigated by geomagnetic ka-PtaH on top of the ruins of the pyramid town of survey, but initially explored by walking along the sur- Pepi I (Men-nefer-Pepi) another living quarter was face. If this area could be verified as settling ground of kept alive till our day with the village Saqqara (Fig. the IInd Intermediate Period we would as a result arrive 2A). This remnant of a pyramid town of the late Old at the originally estimated c. 250 ha. Kingdom must have become an integral part of Mem- It is difficult to compare sizes with residential phis, otherwise, the change of name is unthinkable.14 towns as Memphis because we do not know the Its living space could be estimated at a quarter till a approximate extension of this settlement during the third of Memphis. In the late Old Kingdom and in Middle Kingdom.5 According to the map with recon- the Middle Kingdom the size of the Pyramid town structed mounds by B.J. Kemp, the pre-New Kingdom may even have equalled Memphis of that time. This town cannot be traced.6 The limitations of the river would make the name transfer more understandable. on the east side which distinctly delineate the eastern The two towns may have been connected by a dam- edge of the ruin fields of Memphis,7 the smaller road and a ferry across the minor Nile branch, recon- branch explored on its western side8 and the remains structed by the EES team in the West as an active Nile of ruin mounds would accord Memphis between 550 branch. 15 Right until the present day, the villages of to 600 ha in the Late Period – a size it may have Mit Rahina and Saqqara have been connected by a already reached in the Ramesside Period with its dam road. In the Old and Middle Kingdom the dou- enormous temple precinct of Ptah and the precincts ble town could have gained 120 to 150 ha and was of the other temples (Fig. 2B).9 This would be com- most probably the largest settlement in Egypt. parable to other residential towns such as Amarna Abydos covered an area of c. 2 ha16 and was (early phase c. 300 ha, finally c. 600 ha)10 and Pi- enlarged to c. 3.7 ha,17 Elephantine occupied between Ramesse (c. 550–600 ha).11 2 and 2.5 ha in the early Old Kingdom18 then growing Before the New Kingdom, in the Middle Kingdom in the late Old Kingdom and in the Middle Kingdom and IInd Intermediate Period an enclosure wall to over 8 ha.19 Edfu grew to approximately 8–9 ha at defines the eastern limit of the town.12 The West the same time,20 Hierakonpolis (c. 8.5 ha) and El-Kab seems to border on a small Nile branch,13 which in the Old Kingdom covered approximately the same would limit the East-West diameter to about 400 m. space,21 Dendera in the Old Kingdom is difficult to Also the North-South extension was more restricted measure, but seems to have been quite small (2 ha?),

3 9 See contribution of I. Forstner-Müller in this volume. JEFFREYS 1985, 48–56, pl. 2. 4 10 According to the recent survey conducted by I. Forstner- KEMP & GARFI 1993, fig. 10. 11 Müller, T. Herbich and C. Schweitzer. BIETAK & FORSTNER-MÜLLER in print, 51. 5 12 KEMP 1977a, 192–195, fig. 7; JEFFREYS 1985, 68, fig. 20; JEFFREYS SMITH & JEFFREYS 1986, 94–95, fig. 4, resistivity feature; 1996, 287–291; JEFFREYS 2006b, 166; JEFFREYS & SMITH 1988, JEFFREYS & MALEK 1988, 17, figs. 1–2 (FAM). 13 58. GIDDY, JEFFREYS & MALEK 1990, 12–13, p. 3 w. fig. 2. 6 14 KEMP 1977a, 192–195. fig. 7. JEFFREYS 1985, 4–10 dismisses Cf. MALEK 1997, 94–95 w. fig. 1. Petries’ estimate of the extent of Memphis as it was not 15 A canal/waterway separating the city from the necropolis is based on modern survey methods and, although Jeffreys referred to in Late Period and Ptolemaic sources. Cf. has contributed with his survey to important topographical THOMPSON 1988, 10 w. n. 17, pp.12, 21. 16 fixed points, he does not himself commit to new estimates. KEMP 1977a, 192–195. fig. 7. 7 17 SMITH & JEFFREYS 1986, figs.2–3. I am indebted to Claus Jur- See ZIERMANN 1993, 36, fig. 13. 18 man for pointing out to me topographic features about KEMP 1977a, 192–195. fig. 7; 19 Ancient Memphis. ZIERMANN 1993, figs. 3, 8, 22, 41; ZIERMANN 2003, 112–128, 8 Even if this branch may have been active only during the fig. 50. 20 Early Dynastic Period, it remained an old water branch not KEMP 1977a, 186–189, figs. 3, 6; ZIERMANN 2003, 128, fig. 50. 21 suitable for expanding a settlement. See JEFFREYS & GIDDY ZIERMANN 1993, 36, fig. 13. 1992, 7, fig.; JEFFREYS & TAVARES 1994, 155–157, figs. 5–6. Houses, Palaces and Development of Social Structure in Avaris 13 cAyn Asil the remote capital of the Dakhla Oasis in the towns. According to the second stele of King late Old Kingdom ended up as 3.6 ha.22 Abu Ghalib Kamose, there had been moored at Avaris hundreds (early Middle Kingdom) could be estimated at c. of ships which Kamose boasts that he had plundered 3.4–4.2 ha,23 the town of Kahun at 14 ha,24 Tell el- and carried away:30 “… I did not leave a single plank Yahudiya in the IInd Intermediate Period 21 ha (the belonging to the hundreds of ships of new cedar filled with extramural settlement space being unknown).25 Of gold, lapis-lazuli, silver, turquoise and innumerable considerable size is the fortress of Tell Hebwa north- bronze battle axes, apart from moringa-oil [more likely east of Qantara in the New Kingdom, identified with olive oil], incense, fat, honey,…and all precious woods of the frontier town Zaru.26 It covered an area of c. 80 theirs, and all the good products of Retenu. I carried them ha.27 It is not fully explored how large the town had off completely…” According to Papyrus Anastasi III, 7, been in the IInd Intermediate Period but its size in the 5–6, Pi-Ramesse was “…the marshalling place of thy New Kingdom is remarkable indeed. Compared to [i.e. the pharaoh’s] chariotry, the mustering place of thy other towns in Egypt at that period Tell el-Dabca was army, the mooring place of thy ships’ troops…”.31 Avaris absolutely the biggest and it seems that it was also the had also been right up to the 20th Dynasty according biggest settlement in the whole of the Eastern to inscriptions on naos doors, now in the Pushkin Mediterranean region. Royal towns in Syria, like Ebla, Museum in Moscow, a harbour if not the harbour of Hazor, Ugarit and Qatna, measured between 60 ha to Pi-Ramesse with a temple for Amun of its own.32 The 100 ha, albeit still dwarfed by Tell el-Dabca which must texts testify that Avaris and Pi-Ramesse had been figure even without historical records as one of the harbour-towns and naval bases besides being mili- biggest residential towns in the Near East at that time. tary strongholds. In Ramesside times, the towns according to recent We are now in the lucky position of being able to estimates expanded as much as 600 ha.28 verify the harbour by post-mortem. Geophysical sur- We shall see that size also is an indicator for esti- veys conducted by Irene Forstner-Müller, Thomasz mating the importance of palaces. The town was situ- Herbich, Christian Schweitzer and Michael Weissl ated east of the Pelusiac branch of the Nile and was have revealed a huge harbour basin, with an inlet one of the first Egyptian settlements of the Middle from the Nile and an outlet leading back to the river Kingdom known to us, constructed – as it was – east of (Fig. 3).33 This looks like a facility for busy river traf- the Pelusiac (Fig. 1). Till then, settlements could be fic. The paleogeographers Jean-Philippe Goiran and found only behind (west of) this river.29 The site was Hervé Tronchère from the Université 2 in Lyon were protected eastwards by the biggest drainage system of able to verify the harbour function.34 Another lake Egypt, the Bahr el-Baqar which poured at that time within an old water channel was found further in the into huge overflow lakes which used to form a barrier south. It was according to the French paleogeogra- of perennial swamps towards the east. Coming from phers no harbour basin but could have been, howev- the Sinai, any intruder was forced into a narrow loop- er, only a mooring place for the palace of the middle hole formed by those swamps and the Pelusiac. The Hyksos Period, probably belonging to King Khayan.35 town which controlled this entrance was Tell el-Dabca, A second harbour was found at the river branch itself which was for some time Avaris, the capital of the Hyk- (see the contribution of Irene Forstner-Müller in this sos and later the southern part of Pi-Ramesse, the cap- volume). What is not yet settled definitely is the date ital of Ramses II and his successors. of those different harbours. We also have historical information about Avaris The northern basin measuring c. 450 to 400 m is and Pi-Ramesse besides their function as residential rectangular and with its northern edge parallel to a

22 30 SOUKIASSIAN, WUTTMANN & SCHAAD 1990, fig. 1; ZIERMANN HABACHI 1972, 37. 31 1998, 340–343, fig. 1, 355–357, fig. 8. Translation CAMINOS 1954, 101. 23 32 According to the map in LARSEN 1941, 2, Fig. 1. The origi- TURAYEV 1913, with pl. 13; see BIETAK 1975, 30, 205–206. 33 nal estimate of LARSEN 1936, 50 of 600 × 500 m seems to FORSTNER-MÜLLER 2009; BIETAK 2009. 34 have been on the high side. TRONCHÈRE et al. 2008. 24 35 Measured from PETRIE 1891, pl. 14. Within this basin according to Hervé Tronchère and Jean- 25 PETRIE 1906, pls. 2 and 22. Philippe Goiran only one borehole produced sediments 26 ABD EL-MAKSOUD 1998, 111. typical for harbours. The others were negative in this 27 ABD EL-MAKSOUD 1998, 128, fig.1. respect. It could have been that the sediments were washed 28 BIETAK & FORSTNER-MÜLLER, in print, 51. away or that it was not a harbour but only a berth connect- 29 VAN DEN BRINK 1987, 17–19. ed to the time span of the palace. 14 Manfred Bietak long enclosure wall of the Ramesside Period, which more compelling argument to rebut the location of runs in its eastern parts few meters north of and par- Peru-nefer at Memphis. Before the construction of the allel to a fortification wall of Horemheb. The Rames- barrages, the geographers of the French expedition side wall covered the inlet canal, which means that attest that, during the drought period during the the basin was active before and still visible during this months January till June, river navigation became so period. It could even still have been in operation dur- difficult as to come to a halt altogether in the months ing this time if the outlet canal was used as link to the before the coming of the flood.40 The Nile shrank to Nile system (Fig. 3). The fortification wall of Hor- one fifth of the normal volume.41 We should add that, emheb may have also wrapped right around the inlet during January and February when Nile traffic was still canal, but is not preserved at this spot, not having possible, seafaring in the Mediterranean around Egypt been as deeply sunken as the Ramesside wall. Reflect- stopped because of winter gales, fog and navigational ing on the situation during the late 18th Dynasty, we difficulties under such circumstances.42 This would think it would also have been reasonable to secure reduce down to six months the operation of a harbour with a bulwark a harbour not far from the sea, given for more than 200 km (120 miles) inshore, whilst ships the fact that we have from as early as the time of from harbours within the range of the Mediterranean Amenophis III evidence of raids of seafarers at the could also continue sailing and landing during the river mouths36 and that some time later Ramses II drought period (March till the coming of floods in should catch Sherden trying to make some raids.37 July). It is particularly interesting that all harbours in Several authors have addressed already the ques- deltaic landscapes – if in the Rhine Delta or in the tion if the major naval base of the 18th Dynasty Peru- Indus and Ganges Delta – can be found between 5 km nefer had been located at the same site as Avaris and to 40 km inshore. It is the tides which help navigate Pi-Ramesse.38 Until a short while ago no antiquities past the shallows created by sediments at the river which would have supported such a suggestion had mouths. In the Mediterranean where the tides are only been found. The majority of Egyptologists even now moderate easy access to the sea could be created by follow the opinio communis that this naval base was situ- dredging the river channels. This factor would mean ated at Memphis.39 Despite several supporting placing Peru-nefer within range of the sea. grounds, this identification has to be refuted. Firstly, a It is also unthinkable, from a strategic point of palatial compound of royal dimensions (5.5 ha = 13 view, that the major naval base of Egypt should have acres) from the time of Tuthmosis III and Amenophis been situated 200 km from the sea when warfare in II has been found and extensively excavated (Figs. 3, Asia made fast reaction necessary. The conclusion 28). Both pharaohs can be linked to Peru-nefer, espe- drawn from this makes the continuity of Canaanite cially the latter. The palaces together with the afore- cults from Avaris43 via Peru-nefer44 to Pi-Ramesse45 fall mentioned harbour basin are strong contenders for into place. the site of the most important naval base of the Tuth- Seafaring and harbours bring about trade, in this mosid Period. The physiography of the river Nile is a case with the Near East, Cyprus and during some

36 PORTER & MOSS, 1934, 21(196); YOYOTTE, 1949, 63, 67–69; 501; JEFFREYS & SMITH 1988, 61; EDEL 1953, 155; KAMISH 1985; KITCHEN, 1979, 290; 1996, 120; W. HELCK, 1958, 1821 (text); 1986; DER MANUELIAN 1987; SÄVE-SÖDERBERGH 1946, 37–39; HELCK, 1979, 133, no. 5–7; 1984, 272: 1821 (translation). STADELMANN 1967, 32–35; ZIVIE 1988, 107. 37 40 KITCHEN, 1982, 40f. In another translation variant: KITCHEN, LE PÈRE 1822, 240–241. See also CLOT 1840, 495, and RED- 1996, 120. MOUNT 1995, 134. 38 41 Already SPIEGELBERG 1927, 217, was of the opinion that WILLCOCKS 1899, 46–48; pls. 7–8; BAUMGARTEN (ed.) 1981, 21. 42 Peru-nefer was situated in the Delta. Its association with Pi- YARDENI 1994, 69; STAGER 2003, 243. 43 Ramesse because of the similarity of the Canaanite cults at BIETAK 1981, 247–253; BIETAK 1996a, 36–48; 2003a, 13–20; both sites was first suggested by DARESSY 1928–29, 225, 2003b, 155; MÜLLER 2008, 323–351, 381–384 shows that the 322–326; see also GAUTHIER 1929, 141–2. HABACHI 2001, 9, cult practised in the Canaanite temples at Avaris continued 106–107, 121, insisted that Avaris, Peru-nefer and Pi- at least into the middle of the 18th Dynasty. 44 Ramesse represent a continuity. See also ROEHRIG 1990, STADELMANN 1967, 32–47, 99–110, 147–150; COLLOMBERT & 125–6. NAVILLE 1891, 31, pl. 35 [D], found an inscribed COULON 2000, 217. 45 stone of Amenhotep II that mentions a cult of Amun-Ra DARESSY 1928–29, 326; STADELMANN 1967, 148–150; UPHILL “who resides in Peru-nefer”, and he thought that this site 1984, 200–212, 212, 23323–23324, 245 (Anta), 246 should be located at Bubastis. (Astarte), 252 (Reshep), Seth (252–253). There was also a 39 BADAWI 1943; 1948; GLANVILLE 1931, 109; 1932; HELCK waterway at Piramesse called “the […] waters of Baal” 1939, 49–50; 1971,160, 166, 447–448, 456, 460, 471, 473, (papyrus Anastasi III, 2.8; see CAMINOS 1954, 74). Houses, Palaces and Development of Social Structure in Avaris 15 periods with the Aegean. It is therefore understand- ment representing those syncretistic religious fea- able that, at the site of Tell el-Dabca during the era of tures is the so-called 400 Years’-Stela, found in Tanis, the late Middle Kingdom, the Second Intermediate but originating from the Seth temple at Avaris (Fig. Period and also of the New Kingdom more imports in 5).55 the form of pottery can be found than at other places further upstream.46 III. STRATIGRAPHY AND DEVELOPMENT Summing up, the function of Tell el-Dabca can be Knowledge of the vertical dimension of a site is most identified as harbour town, trade centre, naval and important for tracing its development in time. Many military base and, last but not least, as a royal resi- sites grow from a core settlement, shift with its centre dence of the 14th and 15th Dynasties, probably also as in space, shrink again and/or are abandoned when a part-time residence during the reigns of Tuthmosis the environment changes. This happened in the III, Amenophis II and at the end of the 18th Dynasty Delta mainly in connection with changes in the under Horemheb. Together with Qantir, it was also regime of the Nile. Tell el-Dabca is one of the few test the residence town of the 19th and, to a certain cases in Egypt where we know to some extent the ori- extent, of the 20th Dynasties. During the late Middle gin and some steps of the horizontal expansion of the Kingdom and the Second Intermediate Period, the settlement. We have 17 excavation areas56 at our dis- population was largely of Near Eastern origin, as posal, which act like windows for us to judge the ver- borne out by physical anthropological studies.47 This tical stratigraphy at different spots (Fig. 6). Some of applies mainly to the male population, whereas the those excavation areas as A/II, F/I, F/II, H/I, H/III, female population type was different but also shows H/VI are quite large and cover more than 1000 m2 Near Eastern features. The culture was in the 12th some of them 8000 m2. Those areas do not yield all Dynasty purely Egyptian48 and shows from the late the same strata. Some encompass the time of early 12th Dynasty onwards features of the Syro-Palestinian occupation, such as cEzbet Rushdi (R/I), whilst the Middle Bronze Age culture with acculturation to the medium and the late strata are missing after being Egyptian civilisation.49 From now onwards till the end destroyed by constant agricultural levelling. Other of the Hyksos Period, Near Eastern and Egyptian fea- areas (A/V) are lacking in early stratification with the tures merge, but the Near Eastern ones keep their town being much smaller and the excavations miss- identity until the beginning of the New Kingdom, ing its core. From those 15 windows into the earth, especially in religious and funerary respects. What is akin to working on a mosaic where large parts are amazing is that, after the political turning point and missing, we are forced to reconstruct the spatial Avaris was taken by Ahmose, the cultural mix of development of the town and, as we are lacking in ceramic production, typical of the eastern Delta, con- some of the evidence, our conclusions will to some tinues unbroken till the Tuthmosid Period.50 This is extent remain rudimentary. an indicator that the majority of the Near Eastern In Tell el-Dabca we have some 25 phases starting population which carried Hyksos rule right across from the early 12th Dynasty or probably even earlier Egypt was not expelled but stayed on the spot, despite and lasting with intervals right up to the Late Period the town being abandoned at least in part.51 The (Fig. 7). Some phases are easily identifiable even at intermingled cultural features of Avaris were carried excavation sites at long distances from each other. over into the Ramesside period. The most conspicu- They show distinct architectural features and specif- ous part of it had been of course the continuation of ic brick material. The change of focus heralds a the Canaanite cults,52 especially the cult of the major new building phase. Planned settlement- or Canaanite storm god53 under the name of the Egypt- palatial structures can be quickly recognised as strati- ian storm god Seth.54 The most prominent monu- graphic units. Of special importance are emergency

46 53 BADER 2009. CORNELIUS 1994; GREEN 2003. 47 54 WINKLER & WILFING 1991, 120, 139–140. BIETAK 1990; SCHNEIDER 2003; ALLON 2007. 48 55 CZERNY 1999, 129. SETHE 1930; MONTET 1931; STADELMANN 1965; GOEDICKE 49 BIETAK 1981, 283-288; SCHIESTL 2009. 1966; 1981; BIETAK 1990, frontispiece; BIETAK, HEIN et al. 50 BIETAK in print a. 1994, 279–281. 51 56 BIETAK in print a. A/I–V; E/I, F/I–II, H/I–VI, R/I–II. 52 BIETAK 1990.