Expressing New Rule: Seals from Early Islamic Egypt and Syria, 600– 800 Ce

The Medieval Globe Volume 4 | Number 1 Article 6 2018 Expressing New Rule: Seals from Early Islamic Egypt and Syria, 600–800 CE Petra M. Sijpesteijn Leiden University Centre for the Study of Islam and Society, p.m.sijpesteijn@hum.leidenuniv.nl Follow this and additional works at: https://scholarworks.wmich.edu/tmg Recommended Citation Sijpesteijn, Petra M. (2018) "Expressing New Rule: Seals from Early Islamic Egypt and Syria, 600–800 CE," The Medieval Globe: Vol. 4 : No. 1 , Article 6. Available at: https://scholarworks.wmich.edu/tmg/vol4/iss1/6 This Article is brought to you for free and open access by the Medieval Institute Publications at ScholarWorks at WMU. It has been accepted for inclusion in The Medieval Globe by an authorized editor of ScholarWorks at WMU. For more information, please contact maira.bundza@wmich.edu. i THE MEDIEVAL GLOBE Volume 4.1 | 2018 ii THE MEDIEVAL GLOBE The Medieval Globe provides an interdisciplinary forum for scholars of all world areas by focusing on convergence, movement, and interdependence. Contributions to a global understanding of the medieval period (broadly de ined) need not encompass the globe in any territorial sense. Rather, TMG advances a new theory and praxis of medieval studies by bringing into view phenomena that have been rendered practically or conceptually invisible by anachronistic boundaries, categories, and expectations. TMG also broadens discussion of the ways that medieval processes inform the global present and shape visions of the future. Submissions are invited for future issues: please contact the Editorial Board (medievalglobe@illinois.edu). All articles will be evaluated by the editors and by a double-blind peer review process. For more information about TMG, with further details about submissions and peer review policy, please visit the journal’s website: arc-humanities.org/our-series/arc/tmg. The mark of The Medieval Globe was designed by Matthew Peterson and draws on elements derived from six different medieval world maps. Executive Editor Carol Symes, University of Illinois at Urbana–Champaign Editorial Board James Barrett, University of Cambridge Kathleen Davis, University of Rhode Island Felipe Fernández-Armesto, University of Notre Dame Monica H. Green, Arizona State University Robert Hymes, Columbia University Elizabeth Lambourn, De Montfort University Yuen-Gen Liang, Academia Sinica, Taiwan Victor Lieberman, University of Michigan at Ann Arbor Elizabeth Oyler, University of Pittsburgh Christian Raffensperger, Wittenberg University Rein Raud, Tallinn University & Freie Universität Berlin D. Fairchild Ruggles, University of Illinois at Urbana–Champaign Julia Verkholantsev, University of Pennsylvania Alicia Walker, Bryn Mawr College iii THE MEDIEVAL GLOBE Volume 4.1 | 2018 Seals—Making and Marking Connections across the Medieval World Edited by BRIGITTE MIRIAM BEDOS-REZAK iv British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library © 2018, Arc Humanities Press, Leeds This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Licence. The authors assert their moral right to be identified as the authors of their part of this work. Permission to use brief excerpts from this work in scholarly and educational works is hereby granted provided that the source is acknowledged. Any use of material in this work that is an exception or limitation covered by Article 5 of the European Union’s Copyright Directive (2001/ 29/ EC) or would be determined to be “fair use” under Section 107 of the U.S. Copyright Act September 2010 Page 2 or that satisfies the conditions specified in Section 108 of the U.S. Copyright Act (17 USC §108, as revised by P.L. 94-553) does not require the Publisher’s permission. ISSN 2377-3561 (print) ISSN 2377-3553 (online) https://arc-humanities.org Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY v CONTENTS List of Illustrations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii Cultural Transactions: An Introduction to Medieval Seals from a Global Perspective BRIGITTE MIRIAM BEDOS REZAK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Seals as Conceptual and Ritual Tools in Chinese Buddhism, ca. 600–1000 CE PAUL COPP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Imprinting Powers: The Astrological Seal and Its Doctrinal Meanings in the Latin West NICOLAS WEILL PAROT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 A Medieval Solution to an Early Modern Problem? The Royal Animal Seals of Jambi ANNABEL TEH GALLOP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Expressing New Rule: Seals from Early Islamic Egypt and Syria, 600–800 CE PETRA M. SIJPESTEIJN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 The Formulation of Urban Identity on Byzantine Seals CLAUDIA SODE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 The Cloth Seal: A Mark of Quality, Identi ication, or Taxation? JOHN CHERRY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Archaeology and Sigillography in Northern Europe MICHAEL ANDERSEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 Medieval Treaties and the Diplomatic Aesthetic JESSICA BERENBEIM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 vi vi LIST OF ILLUSTRATIONS Figures Figure 2.1a–b. Huangshen yuezhang seal, front and back . . . . . . . . . . . . . . . . . . . . . . 31 Figure 2.2. Padma (Lotus) seal, from a manual for the making and use of Buddhist talisman-seals (ninth to tenth centuries) . . . . . . . . . . . . . . . . . . . . . 36 Figure 4.1. Seal of Sultan Ahmad Zainuddin of Jambi (r. 1742–ca. 1770) . . . . . . . . 81 Figure 4.2. Seal of Sultan Masud Badaruddin of Jambi (r. ca. 1770–ca. 1790) . . . . . . 82 Figure 4.3. Seal of Sultan Taha Saifuddin (r. 1855–1904) . . . . . . . . . . . . . . . . . . . . . . . 83 Figure 4.4. Seal of Sultan Ahmad Zainuddin (r. 1886–1899) . . . . . . . . . . . . . . . . . . . . 85 Figure 4.5. Brick engraved with an eight-petalled lotus design, ca. ninth to fourteenth centuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Figure 4.6. Minangkabau seal of patronage issued to the sultan of Jambi, impressed on a letter from 1799 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Figure 4.7. Seal of Pangiran Suta Wijaya, later Sultan Ahmad Zainuddin of Jambi (r. 1742–ca. 1770) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Figure 5.1a–b. Arabic tax quittance and seal from Khurasan, dated Ṣafar 150 (March–April 767) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Figure 5.2a–b. Poll tax receipt and seal, dated Ramaḍān 196 (May–June 812) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Figure 5.3a–b. Sealed deed recording the gift of an estate and of a female slave, dated in the month Second New-year 478 (June–July 700). The seal that was attached to the closed document depicts the image of a male buste. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104–5 Figure 5.4a–c. Sealed Arabic order of payment, dated 9 Ramaḍān 196 (May 24, 812), front and back, and detail of the seal, with the name of the inance director al-Ḥasan b. Sa`īd . . . . . . . . . . . . . . . . . . . . . . 105–6 Figure 5.5a–b. Umayyad two-sided lead seal imprint recording the poll tax payment made by the Jewish inhabitants of Tiberias . . . . . . . . . . 106 Figure 5.6a–b. Coptic of icial document from the seventh to eighth centuries and seal imprint depicting a sitting hare . . . . . . . . . . . . . . . . . . . . 107 Figure 5.7a–c. Delivery order from and sealed by the governor ‘Amr b. al-‘Aṣ (d. 664) to the administrator of Heracleopolis/Ihnās, front and back . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108–9 vi viii LIST OF ILLUSTRATIONS Figure 5.8a–c. Arabic sealed tax receipt dated 291 (904), front and back, and seal imprint with the name of the tax of icial Yalhawayh. . . . . . . . . . . 109 Figure 5.9a–b. Sealed letter and sealing from the governor Qurra b. Sharīk to Basileios, administrator of Ishqūh/Aphroditō, dated Ṣafar 91 (December 709–January 710). . . . . . . . . . . . . . . . . . . . . . . . . . . . 110–11 Figure 5.10a–b. Two-sided lead seal imprint recording the tribute or poll tax paid by the inhabitants of Seville, early eighth century. . . . . . . . . 112 Figure 5.11a–b. Two-sided lead imprint recording the tribute paid by the inhabitants of the town of Jaen and its hinterland, early eighth century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 Figure 5.12. One-sided lead seal imprint recording an order of ‘Anbasa b. Suhaym, governor of al-Andalus (in of ice 721–725) . . . . . . . . . . . . . . . . 116 Figure 5.13a–b. Two-sided lead seal imprint recording the order by al-Ḥurr, governor of Andalusia (in of ice 715–718) . . . . . . . . . . . . . . . . . . . . 117 Figure 5.14a–b. Sealed Arabic legal document from Khurasan recording the emancipation of a slave, dated Sha’ban 146 (October– November 763) and the clay sealing showing multiple imprints . . . . . . . 118 Figure 5.15a–b. Tax quittance from Khurasan, dated Rabī’ II 147 (June–July 764), with clay seal imprint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Figure 5.16a–b. Two-sided lead seal imprint from the district of ‘Amwās in Palestine, eighth century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Figure 5.17. Textile covering or container with the name of the addressee, ninth century or later . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Figure 6.1. Seal of Constantine, proedros (metropolitan) of Thessaloniki, twelfth century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Figure 6.2. Seal of the koinon of Sinope, seventh century . . . . . . . . . . . . . . . . . . . . . . 154 Figure 6.3. Seal of the koinon of the Dekapolis of Isauria, seventh century . . . . . 154 Figure 6.4. Seal of Apameia and Antioch, sixth to seventh centuries . . . . . . . . . . . 155 Figure 6.5. Seal of John, protospatharios and strategos of Cherson, tenth century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 Figure 6.6. Seal of John Komnenos Dukas (1240–1242) . . . . . . . . . . . . . . . . . . . . . . . 156 Figure 6.7. Seal of Michael, bishop of Charioupolis, eleventh century . . . . . . . . . . 157 ix LIST OF ILLUSTRATIONS Figure 7.1. Lead cloth seal folded over and still attached to a piece of coarse woollen cloth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 Figure 7.2. Drawing of blank two- and four-part seals . . . . . . . . . . . . . . . . . . . . . . . . . 169 Figure 7.3. Obverse of lead seal for the Dutch immigrant community in Colchester . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 Figure 7.4. Lead cloth seal for the United East India Company . . . . . . . . . . . . . . . . . 176 Figure 7.5. Lead cloth seal for Wesel, Germany. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 Figure 7.6. Stone sculpture from Museum De Lakenhal, Leiden, showing guild of icials inspecting cloth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 Figure 8.1. Pilgrim badge in the shape of a seal, from Rocamadour, France; found at Tårnborg, Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 Figure 8.2. Lead seal impression found in the tomb of Bishop Ulger of Angers (d. 1148) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 Figure 8.3. Seal of bell founder on a fourteenth-century bell from Lumby Church, Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Figure 8.4. Seal matrix of Bishop Henrik of Stavanger, Norway (r. 1207–1224), found at the site of the demolished Mejlby Church near Randers, Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 Figure 8.5. Seal matrix of Bishop Magnus of Växjö, Sweden (r. ca. 1292–1319), found in the churchyard of Fakse Church, Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 Figure 8.6. Seal matrix of Bishop Jon of Skalholt, Iceland (r. ca. 1406–1413), found at the site of the demolished Carmelite monastery in Aarhus, Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Figure 8.7. Seal matrix of an unidenti ied Bishop Henrik, found at Hemsedal Church, Norway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Figure 9.1. Rati ication of the Treaty of Windsor (1387), now in England . . . . . . 219 Figure 9.2. Detail of Figure 9.1: initial letter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Figure 9.3. Detail of Figure 9.1: seal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Figure 9.4. Detail of Figure 9.1: notarial mark. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Figure 9.5. Rati ication of the Treaty of Windsor (1387–8), now in Portugal . . . . . . 222 ix x x LIST OF ILLUSTRATIONS Figure 9.6. Detail of Figure 9.5: initial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Figure 9.7. Detail of Figure 9.5: notarial mark. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Figure 9.8. Articles of Agreement for the Treaty of Windsor . . . . . . . . . . . . . . . . . . . 225 Figure 9.9. Rati ication of the Treaty of Troyes, now in England . . . . . . . . . . . . . . . 226 Figure 9.10. Detail of Figure 9.9: seal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 Figure 9.11. Rati ication of the Treaty of Troyes, now in France . . . . . . . . . . . . . . . 228 Figure 9.12. Detail of Figure 9.11: initial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 Map Map 4.1. The Sultanate of Jambi and its neighbours . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Plates Plate 2.1. Eleven-headed and six-armed Guanyin (Avalokiteśvara) standing on a lotus. China, Tang dynasty (618–907), ca. ninth century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Plate 2.2. Detail of Plate 2.1: the hand bearing a stamp seal . . . . . . . . . . . . . . . . . . . . 28 Plate 5.1a–b. Copper seal recording a tribute or poll tax payments made by the inhabitants of Egypt in the year 94 (712/13): front and back . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Plate 7.1. One half of a stone mould, used for casting four lead blanks at a time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 Plate 7.2. Face of the ifteenth-century copper alloy matrix, and a modern cast from it, for the subsidy for Kent . . . . . . . . . . . . . . . . . . . . . . . . . . 178 Plate 7.3. Modern impression from the late ifteenth-century copper alloy matrix for the subsidy for Wiltshire. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Plate 8.1. Seal matrix of Roskilde Cathedral: walrus ivory, twelfth century . . . . . 194 Plate 8.2. Die-sheet for stamping metal ornaments (Lund, Sweden) . . . . . . . . . . . 197 Plate 8.3. Three seal-shaped lead tablets, Ribe Cathedral, Denmark, second half of the twelfth century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 xi LIST OF ILLUSTRATIONS Plate 8.4. Lead seal matrix of Baldwin IV, count of Flanders (r. 987–1035), found at Læborg, Denmark. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Table Table 4.1. Rulers of Jambi, seventeenth to nineteenth centuries . . . . . . . . . . . . . . . . 79 xi xi xi ACKNOWLEDGEMENTS This volume owes its existence to many. Carol Symes assumed the risk of opening her journal, The Medieval Globe, to seals, a subject matter that until recently still belonged to the domain of specialized connoisseurship and technical Historische Hilfswissenschaften. By enabling the work of scholars from different times and places to appear in a single volume informed by the conceptual framework of global history, she has provided both a forum and a challenge, which each contributor has taken up with exceptional scholarship and lair. Mike Richardson and Linda Paulus shepherded the volume to publication. To all I extend my thankful appreciation. Brigitte Miriam Bedos-Rezak xvi 9 EXPRESSING NEW RULE: SEALS FROM EARLY ISLAMIC EGYPT AND SYRIA, 600–800 CE PETRA M. SIJPESTEIJN THE GREAT ARAB conquests of the seventh century CE brought the former RomanoByzantine provinces of Syria and Egypt, along with the whole Sasanian Empire, under Muslim control.1 Contacts between Arabia and these regions had already been intensive in the pre-Islamic period, resulting in the continuous exchange of knowledge, customs, and administrative practices. In luence and imitation did not simply low in one direction; they were generative processes, in which forms and ideas were actively adapted to new political, religious, and cultural contexts. The practices and ideas that the Arabs brought to these conquered areas were similarly transformed through interaction with local models and customs, and in close relation to socio-political developments at the provincial and empire-wide level. These processes of exchange and adaptation re lect the political and social transitions set in motion by the Muslim conquests, which found expression in administrative systems, material culture, and religious-intellectual life—processes manifested in the seals used in of icial and private contexts. Seals continued to exhibit Byzantine, Sassanid, and Arabian habits, but their use and form also re lect the in luence of developments under Islam. By comparing early Islamic examples with pre-Islamic ones, and by tracing the developments that occurred over time, I will explore the continuities and changes in usage, imagery, and linguistic expression in order This work was supported by the European Research Council under Grant number 683194. I would like to thank Nitzan Amitai-Preiss, Stefan Heidemann, and Taw iq Ibrahim for their help with matters concerning the lead seal imprints. I would also like to thank Robert Carter, Ahmad al-Jallad, Derk Kennet, Michael C. A. Macdonald, and Peter Stein for sharing their knowledge on ancient Arabian materials. Mistakes, of course, remain my own. 1 In the period under study in this article, the Muslim world formed one political unit that was ruled by a caliph, based in Medina (632–661), then in Damascus (661–750), and inally in Baghdad (from 750). It is clear that Islam did not exist at this time in the form that it became known later, based on the outcome of later debates. The terms “Muslim” and “Arab” were also understood differently, especially in the earliest period (e.g., Webb, Imagining the Arabs; Donner, Muhammad and the Believers; Hoyland, “Identity of the Arabian Conquerors”). “Islamic” and “pre-Islamic” are used in this article as chronological terms indicating the time before the prophet Muhammad’s preaching and after it, while “Muslim” is used to refer to the political regime in place. As it is clear that the new rulers put in place administrative, iscal, political, economic, and military structures that differed from local traditions as soon as they arrived, I feel it is justi ied to speak of a Muslim empire even if an imperial organization and ideology were still being developed. The term “Byzantine” refers both to the precursor of the Muslim empire in the eastern Mediterranean and to the Byzantine Empire, which continued to exist in Anatolia. 10 100 PETRA M. SIJPESTEIJN to show how these can be linked to the underlying ideologies and ambitions of Muslim authorities. In particular, I will examine how and why different practices unfolded in Egypt and the Levant, and I will compare these to the dissemination of shared forms throughout the Muslim empire, particularly the rich material from Khurasan in the east and al-Andalus in the west. Near Eastern Seals and Sealing Practices before and after Islam The function and meaning of seals in the Near East, widespread from their irst introduction in the ancient empires of Mesopotamia, did not fundamentally change in the thousands of years they were in use. By the late antique period impression and stamp seal matrices, mounted in rings or in conical form suspended from strings, had become ubiquitous.2 By pressing the seal into wet clay, wax, or another mouldable material, an imprint was pressed directly into or fastened onto the object to be sealed. Seals were individualized through the display of names, titles, and functions; or by personalized symbolic illustrations or mottos. They could also be anonymous, mentioning instead institutions or pious formulae or bearing unidenti iable devices. They were widely used by people in all social categories.3 Seals served two main purposes: (1) identi ication and authorization, similar to modern-day signatures; and (2) protection, by preventing or restricting access.4 A sealing representing an of icial institution or its functionary, when attached to a document, assured its validity. Since most actual writing was carried out by professional secretaries,5 the seal’s imprint certi ied the sender’s identity and his licensed supervision of the transactions recorded and added credence to the message, because of its association with the of icial “behind” the seal, his of ice, and his status. The seal’s imprint was thus a real and lasting reference to the presence and authority of the seal’s owner.6 This can be observed both in the sealings at the bottom of papyrus documents and from descriptions in papyrus letters. For 2 Wassiliou, Siegel und Papyri, 2. For seals and sealings from ancient Arabia, see the examples in the online database of the Corpus of South-Arabian Inscriptions (dasi.humnet.unipi.it) dating from the second millennium BCE to the sixth century CE. I would like to thank Ahmad al-Jallad for bringing these objects to my attention. 3 Ritter, Die altorientalischen Traditionen, 100. 4 The use of seals in magic will not be dealt with here. For a discussion of the use of magical seals in the Islamic period, see Dorpmüller, “Seals in Islamic Literature”; and Soucek, “Early Islamic Seals,” 250–52. 5 Platt, “Making an Impression,” 241. 6 Verity Platt speaks of “an ongoing presence (and protective force) in the face of bodily absence”: ibid., 241–42. 10 EXPRESSING NEW RULE example, an early eighth-century Coptic letter from an Arab of icial introduces an Egyptian tax collector to the taxpayers, recording that the tax collector is licensed and ending his letter with the phrase “I have written this letter and attached [an imprint of] my seal.” One can imagine the tax collector carrying the sealed letter of his superior on his visits to Egyptian taxpayers, showing it as a sign of legitimacy when needed. In an early eighth-century Greek letter, ‘Abd al-Malik b. Yazīd, pagarch (local administrator) of the district of Ihnās (Heracleopolite), writes to an Egyptian village community: “Make sure you get a receipt from the tax collector from your village with [an imprint of] his seal and make sure you do not pay anything more unless you receive a letter of mine with [an imprint of] my seal.”7 Similarly, witnesses to legal documents, merchants or other individuals, added their sealings to substantiate their written words and facilitate identi ication. Such sealings were fastened to the sides or bottoms of letters and legal documents and remained visible when the documents were being read (Figure 5.1a).8 When attached to objects, sealings identi ied the addressee or sender, verifying their value but also impacting the handling and delivery of the goods, which were accorded greater priority and importance.9 Closure sealings, by contrast, secured access to the contents of a letter or container. They were attached in such a way as to close off (part of) a document, bag, or box, either by being attached directly to the container or to strings tying the articles up, or on textiles wrapping the objects.10 Opening and displaying the contents inevitably broke the sealing, which was clearly and immediately visible. So the sealing restricted access to the contents, keeping them hidden or protected against tampering, until they were opened by a suitably authorized person. In many 7 Both documents are discussed in Sijpesteijn, “Seals and Papyri,” 167. 8 Weber, Berliner Pahlavi-Dokumente, 239–41, tab. III, doc. 3 – X, no. 10; XIV, no. 14; XV, no. 15; XXVIII, no. 28; Khan, Arabic Documents, 139, no. 24. Seals were suspended from strings at the bottom of Soghdian documents (Huff, “Technological Observations,” 383). For the application of a seal next to Ptolemaic testaments summarizing the contents, see Vandorpe, “Seals in and on the Papyri,” 235. 9 Some Sasanian clay sealings show traces of fabric imprints on the back: Ritter, Die altorientalischen Traditionen, 59. See also the lead sealing that seems to have been fastened to chain mail based on the traces left on the back of it: Ibrahim, “Notas sobre precintos.” The nail-shaped rivet on the back of a lead sealing bearing the name of caliph Hishām (r. 691–743) suggests it was connected to a container, possibly of wood: Amitai-Preiss and Farhi, “A Small Assemblage,” 233. See below, note 37. 10 For a good overview of the kind of containers that could be closed off with a sealing, as discussed in Arabic literary sources, see Robinson, “Neck-Sealing,” 403n6. See also the account of cloth bales being sealed: Frye, “Sasanian Seals,” 160. For similar usages in ancient Near Eastern and ancient Egyptian contexts, see Regulski, Duistermaat, and Verkinderen, Seals and Sealing Practices in the Near East. 101 102 102 PETRA M. SIJPESTEIJN Figure 5.1a. Arabic tax quittance dated Ṣafar 150 (March–April 767), from Khurasan. The Nasser D. Khalili Collection of Islamic Art. DOC 15 (AR 23). © Nour Foundation. Courtesy of the Khalili Family Trust. (Cf. Khan, Arabic Documents, 104–5, no. 7.) Figure 5.1b. A clay seal imprint is attached to the bottom of the document with a leather string tied through a slit in the document. It has the inscription: Shihāb, for God (shihāb li-llāh). Shihāb b. ‘Amr is the inancial administrator of the governor issuing the document. cases, the closure sealing functioned simultaneously to restrict access, secure the contents, and to identify the goods, the owner/sender, or addressee, thereby functioning to authenticate and protect at the same time. In Bactrian and Pahlavi letters found in Central Asia and Iran, the sealings that closed the documents were typically attached to a piece of leather partially cut from the bottom of the document or af ixed to a separate piece of fabric. Once the document was unrolled, the sealing remained attached, hanging from the piece of leather or textile.11 The sealings referred to in the second letter discussed above similarly served a dual function. The seal’s imprint on the closed letter assured the addressee that the contents had not been interfered with, while at the same time identifying the sender and conveying a sense of urgency, as well as corroborating the authenticity of the text. The sealing of the village tax collector would have identi ied the receipt as an of icial document, while simultaneously closing off the bottom part of the document containing a summary of the contents, which would remain secured 11 For Bactrian letters, see, for example, Sims-Williams, Bactrian Documents, vol. 3, pls. 150, 151a, 151b; for Pahlavi documents, see Weber, Pahlavi-Dokumente, tab. XLI, images 3a, 3b. 103 EXPRESSING NEW RULE Figure 5.2a. Poll tax receipt for the baker Aba Kire written in Arabic on papyrus, dated Ramaḍān 196 (May–June 812). The bottom part of the document was sealed. Under the seal impression, the amount, year and kind of taxes are repeated. Austrian National Library. P. Vindob. A. P. 644. Photo © Österreichische Nationalbibliothek. (Cf. Grohmann, “Probleme,” no. 18.) Figure 5.2b. The clay sealing contains the name of the tax collector Yūnus b. ‘Abd al-Raḥmān. until the proper authorities needed to open it (Figures 5.4a, 5.8a)12 (for a full discussion of this practice, see below). Interestingly, the text of such documents selfrefer to their sealed nature, mentioning that they are sealed or that witnesses had added their sealing to the documents.13 Likewise, in an early eighth-century Arabic 12 See Sims-Williams, Bactrian Documents, vol. 3, pl. 72. 13 For example, Sims-Williams, Bactrian Documents, vol. 1, 104 (doc. T). 103 104 104 PETRA M. SIJPESTEIJN Figure 5.3a. Legal document in duplicate, written in Bactrian with one version visible and the other sealed off at the bottom (for an illustration of the document with its bottom half sealed and closed, see Sims-Williams, Bactrian, III pl. 72). The deed records a gift of an estate and of a female slave and is dated in the month Second New-year 478/June-July 700. A clause in the document states that “this contract has been sealed by my indulgence” and that two witnesses were present at the statement. The Nasser D. Khalili Collection of Islamic Art. DOC 10. Photo © Nour Foundation. Courtesy of the Khalili Family Trust. (Cf. Sims-Williams, Bactrian Documents, I, 98–105, no. T.) letter, the head of an administrative district ordered his subordinate, who was responsible for the tax collection in a subdistrict, to consolidate the dinars from the different villages into one shipment and “then seal what you received with the seal which has been transferred to you.”14 The seal transmitted to the local of icial 14 Sijpesteijn, Shaping a Muslim State, 314–16, no. 8, ll. 19–20. 105 EXPRESSING NEW RULE Figure 5.3b. Clay bulla with the image of a male buste. Figure 5.4a–b. Sealed Arabic order of payment, dated 9 Ramaḍān 196 (May 24, 812): front and back. A seal closes off the bottom part of the document, which presumably contains a summary of the most important information in the document. Austrian National Library. P. Vindob. A. P. 1053. Photo © Österreichische Nationalbibliothek. was presumably that of the district’s head, or in some other way recognizable as belonging to the Muslim administration. The sealing functioned, on the one hand, to protect the shipment of gold coins, making sure that no one would be able to interfere with it en route, and, on the other, identi ied it as a government delivery. Although the function of Near Eastern seals did not change much through time and the purpose of Muslim-Arab seals and sealings compares well to other 105 106 106 PETRA M. SIJPESTEIJN Figure 5.4c. Clay sealing containing pious formulae and the name of the inance director al-Ḥasan b. Sa‘īd in whose name the receipt was issued. Figure 5.5a–b. Umayyad two-sided lead seal imprint recording the poll tax payment made by the Jewish inhabitants of Tiberias. a: khātim kūra ṭabariyya; b: yahūd ṭabariyya. No. 89.8.13190. Collection of the Israel Museum, Jerusalem. Photo © the Israel Museum, Jerusalem. (Cf. Amitai-Preiss, “A Poll Tax Seal,” 104–5.) geographical and historical settings, the speci ic documentary contexts as well as the forms, materials, and imprints of the seals and sealings did vary. Two main processes were at play. First, in the Muslim world Arab, Byzantine and Sasanian material-cultural traditions travelled freely, blending into new composite forms, which also affected seals.15 Second, political consolidation, on the one hand, and religious-political self-awareness, on the other, motivated administrative reforms, such as the Arabicization of the chancery, iscal restructurings, and increased 15 The same observation has been made for Islamic material culture in general (Grabar, Formation of Islamic Art) and, more speci ically, for early Muslim weights (Khamis, “A Bronze Weight”). See also Knappett, “Imprints.” 107 EXPRESSING NEW RULE Figure 5.6a. Coptic of icial document from the seventh to eighth centuries. Austrian National Library. P. Vindob. K. 2579. Photo © Österreichische Nationalbibliothek. Figure 5.6b. Clay seal imprint depicting a sitting hare (Wassiliou, Siegel und Papyri, 16–17, no. 3). supervision of the movement of people and property—all of which are re lected in seals. While some of these measures are better documented at the provincial level, other developments can be connected to empire-wide transformations. Public administrative practices obviously interacted with those common in private commercial contexts, as both of icial and personal seals witnessed similar changes albeit at a different pace. Before examining these developments in detail, I will brie ly discuss the material on which these observations are based. Profiling the Sources Political and climatological conditions, in combination with scholarly, especially archaeological, preferences, have resulted in an uneven distribution of attested 107 108 108 PETRA M. SIJPESTEIJN Figure 5.7-a–b. Delivery order written in Greek on papyrus from the governor ‘Amr b. al-‘Āṣ (d. 664) to the administrator of Heracleopolis/Ihnās for maintenance of an Arab army unit: front and back. Austrian National Library. P. Vindob. G. 39724. Photo © Österreichische Nationalbibliothek. (Cf. Grohmann, From the World, 115–16.) seals and sealings from the pre-Islamic and Islamic Near East. It is dif icult to establish the extent to which practices and forms observed in a speci ic geographical context can be applied more generally. First, it cannot be assumed that the presence of seals in a given archaeological site indicates that the act of sealing objects and documents was practised at that site.16 Seals served ornamental 16 Ritter, Die altorientalischen Traditionen, 58. 109 EXPRESSING NEW RULE Figure 5.7c. Clay seal imprint belonging to the governor ‘Amr b. al-‘Āṣ (d. 664) depicting a charging bull. Figure 5.8a–b. Arabic tax receipt issued to a baker named Mūsā and a builder called George, dated 291 (904): front and back. The bottom part containing a summary of the document is rolled up and sealed with a seal attached to a string tied through the papyrus. Austrian National Library. P. Vindob. A.P. 3378. Photo © Österreichische Nationalbibliothek. (Cf. Grohmann, “Einige bemerkenswerte Urkunden,” no. 12.) Figure 5.8c. The seal of the tax-collector Jalawayh is imprinted twice at the bottom of the document, with the two clay impressions at a 90°-angle from each other. Their inscriptions read: seal of Yalahwayh (ṭābi‘ yalahwayh). 109 10 110 PETRA M. SIJPESTEIJN Figure 5.9a. Sealed letter from the governor Qurra b. Sharīk to Basileios, the pagarch (administrator) of Ishqūh/ Aphroditō, dated Ṣafar 91 (December 709–January 710). Oriental Institute Chicago, D. 13296 (E. 13756). Courtesy of the Oriental Institute of the University of Chicago. (Cf. Abbott, Kurrah Papyri, 47–9, no. 3.) purposes too, and mounted seals were worn as jewellery.17 In Arabia, for example, seal matrices have been found, but sealings remain unattested. The cultural integration of Arabia into Near Eastern writing traditions, however, suggests that 17 See also the much later signet ring seals, in which the writing appears not in mirror image, so as to produce a ‘correct’ text when stamped, but in regular script indicating these no longer functioned as proper seals. 1 EXPRESSING NEW RULE Figure 5.9b. The clay seal imprint at the bottom of the letter shows a quadrupled animal with a star above its back. sealing documents and objects was nevertheless practised in the peninsula. References to sealing do occur, moreover, in written texts.18 Byzantine, Sasanian, and Arab seals were all made of durable materials, such as precious stones, rock crystal, and metal, and were used both in metal signet rings and as block stamp seals.19 Several of these seals have been the objects of study.20 These seal matrices alone do not tell us much about their use, however. For this, seal imprints in lead, clay, or wax—the materials in use in this period and region—are needed, preferably attached to the objects that they authenticated. Indeed, there is uncertainty and even outright disagreement among scholars about how seals were used, by whom, in what context, and for what purpose, both in the Islamic and pre-Islamic Near East.21 18 See below, note 75. 19 Wassiliou, Siegel und Papyri, 4. For signet rings, see ibid., 41–43. See also Porter, Arabic and Persian Seals. 20 See, for example, Ritter, Die altorientalischen Traditionen; Platt, “Making an Impression”; and Amitai-Preiss, “Faunal Iconography.” 21 The application and usage of Sasanian clay seals, whether for objects or documents, and the explanation for the inding of multiple clay imprints together are iercely debated. 111 12 112 PETRA M. SIJPESTEIJN Figure 5.10a–b. Two-sided lead seal imprint recording the tribute or poll tax paid by the inhabitants of Seville, early eighth century: (a) bi-sm allāh; (b) ahl ashbīla. Private collection Taw iq Ibrahim. www.andalustonegawa.50g.com/Seals.html. © Taw iq Ibrahim. While many imprints of seals have been found dating to the early Muslim empire — including in al-Andalus, Egypt, Syria-Palestine, Central Asia, Sicily, Sardinia, southern France, and Iran — and pre-Islamic Sasanian and Byzantine seal impressions have been unearthed in large numbers as well, many more undoubtedly remain to be found. The reasons for this are manifold. Covered in layers of deposits, lead seal imprints are dif icult to detect even in an organized archaeological excavation. To the untrained and naked eye they do not look very impressive, and so often remain unnoticed.22 Metal detectors can, to a certain extent, be used to See, for example, Frye, “Sasanian Seals”; Lerner and Skjaervø, “Some Uses of Clay Bullae”; Huff, “Technological Observations”; and Ritter, Die altorientalischen Traditionen. The interpretation of the use of Arabic lead seals stating place names, districts, provinces, or named of icials continues to be debated as well. An example is the discussion about the relation of some of these seals to the Muslim poll tax and whether they were worn by individual taxpayers as proof of their payments or accompanied transports of goods and coins. See, for example, Schindel, “Nochmals”; Robinson, “Neck-Sealing”; Balog, “Dated Aghlabid”; and Amitai-Preiss, “A Poll Tax Seal.” See also Ibrahim, “Additions,” 115. Further, the choice of illustrations on seals has been called “random” (Amitai-Preiss, “Faunal Iconography,” 212) and “intentional” (Gyselen, Sasanian Seals and Sealings, 8; Ritter, Die altorientalischen Traditionen, 109–10; Platt, “Making an Impression”). The interpretation of the function of individual seals continues to be discussed as well. See the contested interpretation of ‘Abd al-Malik’s sealing (Grabar, Formation of Islamic Art, ill. 21), some arguing that it is, instead, a weight (Amitai-Preiss, “An Umayyad Lead Seal,” 233n2; Khamis, “A Bronze Weight,” 151). 22 Note the lead and copper sealings irst deemed to be too unimpressive to take to an antiquities dealer by the Sicilian farmer who found them (Balog, “Dated Aghlabid,” 129). Only 13 EXPRESSING NEW RULE remedy this and have led to discoveries of lead seals in places where their presence was not initially suspected.23 Yet this has meant that most seals are acquired not via controlled excavations, but via private purchases or on the antiquities market with little knowledge of their provenance.24 Furthermore, countless lead seals will never be retrieved because their material lent itself to reuse, either by restriking outdated imprints or by melting and reusing the lead for new seals.25 Even less likely to be found are clay imprints of seals. Sand-coloured and often damaged because of their brittle nature, they can be found and salvaged only through very intensive recovery methods (e.g., sieving). In addition, unbaked clay imprints quickly dissolve when brought into contact with water. Many seals were broken, either intentionally, when the objects to which they were fastened were opened, or accidentally, when the objects were moved from one place to another. Most clay impressions have been recovered when still attached to documents, usually on folded parts of double documents, which are more easily detected, or when they existed in large enough quantities (and of large enough size) to be noticed.26 Even so, the number of known clay imprints is only a fraction of the number of lead seal imprints.27 At the same time, the number of known lead sealings from the Islamic period is very small compared to the number of Byzantine lead imprints. This is to a large extent caused by the status of the ield, since less archaeological work has been conducted on the Islamic period than on other periods in Near Eastern history—a situation that is, nevertheless, slowly improving. Rarest of all are wax sealings, which were less common in the Byzantine Empire than lead when metal detectors were used to examine the waste deposits removed by archaeologists from sites in al-Andalus could lead sealings be discovered (Ibrahim, “Additions,” 115n2). 23 See previous note. 24 Most lead sealings from the Arab Middle East originate in Israel, where the antiquities trade is well developed: Schindel, “Nochmals”; Amitai-Preiss, “An Umayyad Lead Seal.” None of the lead sealings from al-Andalus originate from controlled, scholarly excavations: Ibrahim, “Additions,” 115n2. From North Africa, on the other hand, none have come to light, though the presence of lead sealings in al-Andalus, France, and Sicily strongly suggests that lead sealings were also in use there: Balog, “Dated Aghlabid”; Ibrahim, “Additions.” The more northerly provinces, such as the Jazira, have also so far not yielded any sealings. 25 Ibrahim, “Additions”; Cheynet, “L’usage,” 24; Huff, “Technological Observations,” 373–75; Ritter, “On the Development,” 104. 26 For sealings connected to documents, see, for example, Wassiliou, Siegel und Papyri; Vandorpe, “Seals in and on the Papyri”; Sims-Williams, Bactrian Documents; and Khan, Arabic Documents. For hoards of clay bullae, see Huff, “Technological Observations,” 375–76; and Vandorpe, “Seals in and on the Papyri,” 231. 27 Wassiliou, Siegel und Papyri, 4; Vandorpe, “Seals in and on the Papyri,” 231. 113 14 114 PETRA M. SIJPESTEIJN ones.28 Although some wax sealings have been recovered on documents from the later Byzantine Empire, only two have so far been retrieved from Islamic Egypt.29 The dearth of actual seals and sealings, especially attached to documents, is partially compensated for by texts. Many documents contain information on the use of seals, discussing the sealing of documents or shipments, and referring to signet rings and other seals.30 In the Greek papyri from the Islamic period, the word bulla is used for the sealing that appears on a document. Arabic papyri employ the terms ṭābi’, which seems to have been especially used in Egypt, and khātim, which was the more general term; both could also be used as verbs. These terms also appear on the seals themselves.31 The Shape and Application of Seals Arab-Muslim seals took different forms depending on their function and the context in which they were used. Sealings still attached to documents from Egypt, Iran, and Central Asia are all imprinted on one side only and are, with the exception of two wax imprints attached to papyrus, all made of clay.32 The clay was shaped by hand, as the occasional ingerprints indicate. A small piece of clay was attached to the folded or rolled-up document, after which a string was tied through or around it over the clay and pressed into it. A ball of clay was added on top and a seal was pressed into it. In this way the string located in the middle of the clay seal was secure, and the seal was attached in the strongest way possible.33 As discussed above, seals could also be attached to a partially cut-off part of the writing surface in such a way that they remained attached and visible after the document had been 28 Wax sealings had replaced lead ones entirely by the twelfth century in the Byzantine empire: Cheynet, “L’usage,” 23. 29 Wassiliou, Siegel und Papyri, 8, 37, no. 27. 30 Sijpesteijn, “Seals and Papyri”; Vandorpe, “Seals in and on the Papyri,” 231. 31 See below at note 95. Khātim occurs on seal imprints from al-Andalus and Syria-Palestine (Figures 5.5, 5.16) and in documents from Khurasan referring to the sealing at the bottom of the document (Figures 5.3a, 5.14a). Ṭābi‘ is attested on Egyptian glass stamps and on a seal imprint: see Figure 5.8c and Wassiliou, Siegel und Papyri, no. 29. It also appears on one lead sealing from Ilyā dated 101 (719/20): Amitai-Preiss, “Umayyad Vocabulary,” 282. See also Soucek, “Early Islamic Seals,” 237. 32 Huff, “Technological Observations.” The wax sealings are attached to papyri with Arabic protocol texts on the other side, indicating an eighth-century date: Wassiliou, Siegel und Papyri, 8. For examples from Iran, see Weber, Pahlavi-Dokumente, tabs. XLI–XLIII; for Central Asia, see Khan, Arabic Documents. See also Grohmann, Einführung, 128–30. 33 Vandorpe, “Seals in and on the Papyri”; Huff, “Technological Observations,” 381; Ritter, Die altorientalischen Traditionen, 74–75. 15 EXPRESSING NEW RULE opened up (Figure 5.3a–b). Seals attached to the bottom of documents or to a part of the document that was folded over were attached in the same way, the only difference being that the string was tied via one or two holes in the papyrus or other material (Figure 5.4a–c).34 This practice of sealing folded or rolled-up documents with clay sealings continued pre-Islamic Sasanian and Byzantine practices, and persisted in unchanged form until the middle of the tenth century, when the last such sealings are attested.35 While some of the larger clay bullae may have been used to secure bags and containers, it seems unlikely that the very vulnerable clay sealings were attached to objects on a large scale.36 The imprints of strings, wires, or textile weaving on the unwritten back of clay sealings that have been used to argue for the use of clay sealings on objects more probably derive from their having been fastened to strings tying documents or to (textile) writing material directly.37 The concave form of some larger clay sealings can also be explained by their having been attached directly to rolled-up documents.38 Lead, copper, and other metals were also used for sealings in the Muslim period, albeit not attached to documents. Lead sealings are attested with one-sided and twosided imprints (Figures 5.11a–b, 5.12). The impressions on both sides were made with tongs, sometimes showing the same inscription, but in most cases inscribed differently on each side (Figures 5.5, 5.11a–b).39 In some cases a wire seems to have 34 See, for example, from Sasanian Iraq, Ritter, Die altorientalischen Traditionen, 60; Frye, “Sasanian Seals”; from Iran, Weber, Pahlavi-Dokumente, tab. XXVIII; from Roman and Byzantine Egypt, Vandorpe, “Seals in and on the Papyri”; from Islamic Egypt, Grohmann, Einführung, 128–29; and from Islamic Central Asia, Khan, Arabic Documents; and SimsWilliams, Bactrian Documents. 35 The latest dated Arabic clay seal from Egypt was attached to a paper document with a tax receipt dated 342 (954/55): Wassiliou, Siegel und Papyri, 39, no. 31. A paper document dated 960 was sent from Iraq by Neḥemiah Gaon to Egypt, where it ended up in the Genizah of the Ben Ezra synagogue of Fusṭāṭ. The seal imprint attached to the bottom of the document was made of two colours of clay and contains goat hair. The name of the Gaon appears on the sealing, as well as a wish for his long life. A twisted piece of paper was fastened through the sealing, while an additional piece of paper was fastened to the back of the sealing to fortify the writing support: Olszowy-Schlangen, “Early Babylonian ‘Documentary’ Script.” 36 Scholarly opinions remain divided on this point: see Freye, “Sasanian Seals”; Ritter, Die altorientalischen Traditionen; and Lerner and Skjaervø, “Some Uses of Clay Bullae.” 37 Huff, “Technological Observations,” 379. For documents written on textile, see Weber, Pahlavi-Dokumente. See igure 5.17 in this essay. 38 Vandorpe, “Seals in and on the Papyri.” Rather than sticks or bundles, as suggested by Richard Frye, “Sasanian Seals,” 157. 39 This continued Byzantine practice: Cheynet, “L’usage,” 24. A two-sided lead sealing mentioning the caliph Sulaymān (r. 715–17) was found in Palestine: Schindel, “Nochmals,” 118. For examples from Arab Syria-Palestine, see Amitai-Preiss, “Umayyad Lead Sealings.” 115 16 116 PETRA M. SIJPESTEIJN Figure 5.11a–b. Two-sided lead imprint recording the tribute paid by the inhabitants of the town of Jaen and its hinterland in compliance with the peace treaty concluded with the Arabs in the early eighth century: (a) muṣālaḥa; (b) arḍ jayyān. Private collection Taw iq Ibrahim. www.andalustonegawa.50g.com/Seals.html. © Taw iq Ibrahim. Figure 5.12.One-sided lead seal imprint recording an order of ‘Anbasa b. Suhaym, governor of al-Andalus (in of ice 721–725). The inscription reads: bi-sm allāh hādhā mā amara bihi al-amīr ‘anbasa ibn suhaym. Private collection Taw iq Ibrahim. www.andalustonegawa.50g. com/Seals.html. © Taw iq Ibrahim. 17 EXPRESSING NEW RULE Figure 5.13a–b. Two-sided lead seal imprint recording the order by al-Ḥurr, governor of Andalusia (in of ice 715–718) for the division of the tribute made by the province of al-Andalus: (a) amara al-ḥurr qasm; (b) al-andalus. Private collection Taw iq Ibrahim, www.andalustonegawa.50g.com/Seals.html. © Taw iq Ibrahim. been pressed into the material at the time of striking, but other samples show that a channel was added at the time of striking or that a hole was subsequently drilled through the seal imprint to enable it to be fastened with a metal wire or string. Other sealings have a loop from which they are suspended like a pendant. Onesided lead sealings sometimes have an attached nail on the back to fasten them.40 Such leaden sealings were not actually broken to open the containers, as only the wires or strings used to attach the sealing had to be cut in order to do so. Since a cut string with sealing could have been easily replaced by a new one without damaging the imprint itself, the secretive and protective function was also less pertinent than that of the clay sealings attached directly to the writing material after folding it. The more durable, but not very expensive, lead sealings seem, on the other hand, to have been frequently used to seal bags, containers, and boxes. Lead sealings could also be fastened to textile bags or sacks directly.41 Some of the lead sealings seem to 40 For a hole drilled through the sealing, see the references in Robinson, “Neck-Sealing,” 423n102; for wires inserted before the imprints were made, see Balog, “Dated Aghlabid” and Ibrahim “Notas sobre precintos”; for channels intended for wires made into the sealings, see Cheynet, “L’usage,” 24. Sealings with nails on the back were found in al-Andalus and SyriaPalestine: Amitai-Preiss, “Umayyad Lead Sealings”; Ibrahim “Notas sobre precintos.” 41 Amitai-Preiss, “Early Islamic Lead Seals,” 111–14. 117 18 118 PETRA M. SIJPESTEIJN Figure 5.14a. Arabic legal document from Khurasan recording the emancipation of a slave, dated Sha‘bān 146 (October–November 763). The document ends with the statement that the manumitted slave, the owner and the witnesses have added their seal (khatama qiyā wa-būya ibn muḥammad wa-l-shuhūd). The Nasser D. Khalili Collection of Islamic Art. DOC 35 (AR 12). © Nour Foundation. Courtesy of the Khalili Family Trust. (Cf. Khan, Arabic Documents 158–9, no. 31.) Figure 5.14b. The clay seal imprint bears multiple imprints amongst which are astral images, including a small sixpointed star drawn above and below some names (Yaḥyā ibn ‘Ubayd, Jahm ibn Qays, Mūsā for God, mūsā li-llāh) and pious formulae. have been fastened to armour or clothing, as traces of metal rings or textile fabrics on the uninscribed back sides indicate.42 There is no indication that lead sealings were used to seal documents in the early Islamic period, though they were used on contemporary, earlier, and later Byzantine documents. 42 As the imprints on the back, which seem to have been made by metal rings as part of a ring mail indicate: Ibrahim “Notas sobre precintos.” 19 EXPRESSING NEW RULE Figure 5.15a. Tax quittance from Khurasan, dated Rabī` II 147 (June– July 764) . The Nasser D. Khalili Collection of Islamic Art. DOC 29 (AR 9). © Nour Foundation. Courtesy of the Khalili Family Trust. (Cf. Khan, Arabic Documents, 92–3, no. 1.) Figure 5.15b. Clay seal imprint depicting a ive-pointed star surrounded by four small crescents attached to the bottom of the document. It should be noted that all the functions that were ful illed by seals made of different materials could also be achieved by pen and ink. Indeed, examples of this are attested. Lines and drawings on the outside of folded letters were positioned across folding and overlapping parts of the papyrus, thus forming an uninterrupted pattern as long as the papyrus or paper remained tightly folded.43 Opening the document disrupted the pattern, which was easily discernible. The contents of a bag, container, or document, and the name of the addressee or sender, could also be indicated by writing on the container or document directly, or on a piece of textile covering the object (Figure 5.17). It is clear that the social and, in particular, the political position of the individuals involved determined the use of seals versus simple ink, though availability and personal choice cannot be ruled out either, especially in private contexts. Nevertheless, the use of seals was widespread in public and private domains. Lead sealings may have been used to manage poll tax collection (see below). In exchange for the poll tax, Christians, Jews, and Zoroastrians received protection and the guarantee, at least of icially, of peaceful coexistence under Muslim rule. Literary sources describe how, in late seventh-/early eighth-century Egypt and northern 43 Vandorpe, “Seals in and on the Papyri,” 243. 119 120 120 PETRA M. SIJPESTEIJN Figure 5.16a–b. Two-sided lead seal imprint from the district of ‘Amwās in Palestine, eighth century: (a) khātim kūrat ‘amwās; (b) iqlīm bālū bayt būsim. The inscriptions can be translated as: seal of the district ‘Amwās, the sub district of Bālū, [the town of] Bayt Būsīm (?). No. 2006.33.26184. Collection of the Israel Museum, Jerusalem. Photo © the Israel Museum, Jerusalem. (Cf. Amitai-Preiss, “Islamic Lead Coins.”) Iraq, non-Muslim taxpayers had to carry an identifying sign in the form of a lead seal imprint around their necks showing they had paid their poll tax. A group of ninth- and tenth-century lead sealings attests to this practice in a later period.44 The amounts of 12 and 24 dirhams listed on the sealings compare exactly with the numbers listed in Arabic legal texts corresponding to two of the three categories of the poll tax for non-Muslims. Poll tax payments were adjusted to the economic position of taxpayers, with low, middle, and rich classes paying different amounts. The Arab Conquest and Early Muslim Rule The seventh-century Arab conquests led to greater reliance on documentation and the increased employment of seals, as is visible on documents and preserved items from Egypt and Syria-Palestine. The new rulers did not completely overhaul the local administrative structures. Although government of icials in the highest 44 Robinson, “Neck-Sealing,” 427n125. Chase Robinson suggests that sealings containing references to the inhabitants of a province might have served the same purpose in earlier periods as the individual seals from the ninth and tenth centuries. He acknowledges that the reference to larger areas or regions makes sense only outside those areas and regions (“Neck-Sealing,” 424–25). Porter also only inds evidence for individual poll tax seals in the ninth century (Arabic and Persian Seals, 3–4). Nikolaus Schindel does not accept the interpretation of such earlier sealings according to which names of provinces or areas refer to payments made by individual taxpayers; rather, he considers these payments to have been made by or for non-Muslim communities as a whole (“Nochmals”). 12 EXPRESSING NEW RULE Figure 5.17. Textile covering or container with the name of the addressee, ninth century or later. 12.855, no. 88. © Leiden University Libraries. layers of the administration were replaced by Arab governors, security of icers, and inancial administrators, appointees on lower positions remained in their functions. This irst generation was replenished from the same reservoirs: the highest positions by Arabs; the lower positions, both those in the countryside and those in the central administration, by indigenous non-Muslims belonging to the same local economic and social elite from which such administrators had been recruited in the pre-Islamic period. 45 The same individuals were in charge 45 In Egypt, the Arabs appointed Egyptian pagarchs (local district managers) belonging to the same landholding class that the Byzantines had used: Sijpesteijn, Shaping a Muslim State. For Syria, see, for example, John of Damascus, whose family had served the local 121 12 122 PETRA M. SIJPESTEIJN of the same taxes according to the same structures as before, but change was also part of the picture. As the Arab rulers relied for a large extend on local expertise, experience and personnel, the prevailing sense was one of continuity in the daily life of most subjects. Nevertheless, some of the new measures did have an impact on individuals subjected to Muslim rule, starting soon after the conquest.46 Muslim rule led to an increase in administrative record-keeping. The difference is striking, with many more administrative documents being attested starting in the immediate post-conquest period, as well as more seals having been preserved from the Muslim period than from the previous period.47 The Muslims were keen and active administrators, demanding extensive written documentation from their government of icials and producing at least as much themselves. 48 Such emphasis on writing and written records also spilled over into the private sphere, where the use of seals on letters and legal documents also proliferated. This increased documentary trail ran across multiple languages. The use of Arabic was introduced directly following the conquests to communicate with the subject population.49 Greek, Coptic, and other local languages continued to be used besides Arabic, but Arabic functioned right from the start as an administrative language for of icial writings. The use of Coptic had increased in Byzantine Egypt, entering domains in which it had not been used beforehand, such as legal documents. This expansion continued under the Muslims, when Coptic began to be used also for administrative documents (Figure 5.6a). The earliest documents preserved on papyrus from Egypt, as well as those on leather from Central Asia and Iran, also show that the Muslims introduced their own documentary practices and administrative habits. These practices differed from local traditions, but showed at the same time commonalities across the Muslim empire. In other words, practices observed in documents from al-Andalus, Egypt, Persia, and Central Asia show the same features, which set them apart from locally produced documents, and which were presumably introduced by administration in Damascus in the pre-Islamic period and continued to do so under the Arab rulers: Hoyland, Seeing, 480–89. 46 For these developments in Egypt, see Sijpesteijn, Shaping a Muslim State, chap. 2. Several additional observations appear in this article. For changes taking place in Syria-Palestine following the Arab-Muslim conquest, see Haldon, “Introduction: Greater Syria.” 47 Wassiliou, Siegel und Papyri, 11. 48 See also the observation, based on the information from lead seals as well as chronicles, that the Arabs executed in al-Andalus a “very diligent and ef icient excise system of collection”: Ibrahim, “Additions,” 119. 49 One Arabic and one Greek-Arabic document from Egypt date to the period of the conquest of the province. They constitute the earliest dated Arabic writings known, dating to 22 (643): Sijpesteijn, Shaping a Muslim State, 65. 123 EXPRESSING NEW RULE the Arab conquerors.50 In certain cases, such practices can in fact be connected to Muslim legal and administrative prescriptions and customs, known from the extensive Arabic literary documentary record, including legal treatises, theological tractates, chronicles, and administrative handbooks produced from the ninth century onwards. The question remains, however, whether this early Arab practice in luenced the debates and prescriptions later recorded in these literary accounts, or whether both can be traced to practices introduced and inspired by a theoretical framework based on Islamic rules.51 Bilingual administrative documents from Egypt and Palestine show how the Greek and Arabic parts of the papyri were not directly translated from each other but, rather, that each linguistic part followed its own documentary and administrative conventions.52 The Arabic documents exhibit a new technical vocabulary, different expressions that disclose a full- ledged documentary and managerial tradition, some of which overlaps with the local practices in the newly conquered lands.53 This Arab-Muslim practice did not replace but existed side by side with local traditions. This combination of adaptation and continuity can also be observed in the use of administrative terminology and titles. “ ‘Abd Allāh,” slave or servant of God, which preceded the caliph’s name on coins, seals, and papyrus protocols and in monumental inscriptions, can be directly connected to similar terms used to refer to Byzantine emperors.54 On the other hand, the use of this part of the caliph’s title in Greek transcription, rather than Greek translation, in the Greek and bilingual Arabic/Greek protocol texts on papyrus suggests that the borrowing took place before the conquests.55 The term amīr, used in Arabic literary texts for 50 As observed already by Geoffrey Khan, “The Pre-Islamic Background.” 51 See, for example, the conditions discussed in an early eighth-century Arabic letter concerning the ḥajj, the Muslim pilgrimage. The vocabulary and contents of this letter resonate with later legal texts discussing the conditions in which someone is obliged to undertake the pilgrimage and the rules that apply to the religious journey. Without additional context, it remains dif icult to decide whether the discussions in the letter describe customs and common expectations or religious legal prescriptions: Sijpesteijn, “An Early Umayyad.” See also Sijpesteijn, Shaping a Muslim State, 68–69. 52 Khan, “The Historical Development.” For the differences between the Arabic and Greek texts of bilingual papyri, see Sijpesteijn, Shaping a Muslim State, 67–69. 53 Sijpesteijn, Shaping a Muslim State, 69–71. 54 Servus Christi (Latin) and doulos tou Christou (Greek), as observed by Nitzan Amitai-Preiss, “Umayyad Vocabulary.” While ‘Abd Allah as a personal name is surely widely attested in Arabia, no occurrence of it being used as title for a ruler can be identi ied with certainty (I would like to thank Ahmad al-Jallad for this information). For protocols, see Grohmann, Protokolle. 55 Greek administrative titles do appear in transcription in Arabic texts (such as māzūt for meizoteros and sammāk for symmachos), while Arabic titles were sometimes translated into Greek (such as symboulos for “governor” and prōtosymboulos for “caliph”). 123 124 124 PETRA M. SIJPESTEIJN “army commander” and “governor,” is used in papyri, on glass weights, and on seals to refer to the governor.56 Greek and Coptic papyrus texts, on the other hand, apply this term to local military-administrative of icials, at the level of districts and larger administrative units, but not for the governor. The appearance of amiras, the Greek and Coptic transcription of the Arabic amīr, shows that no Greek or Coptic equivalent existed in the eyes of the scribes.57 As with other administrative practices, the Arabs introduced new sealing practices, while in other respects the use of seals continued pre-Islamic tradition. The irst generation of clay sealings produced by the Arab conquerors in Egypt show pictures rather than texts. This exclusive use of images constitutes a break with Byzantine and Sasanian tradition, in which texts prevailed on of icial seals, including those used by civil servants in their of icial capacity.58 In general, anepigraphic seals were more commonly used in the private sphere in the Sasanian Empire. Impressions in clay with single gem-like imprints as produced in Muslim Egypt continued Sasanian practice, albeit in smaller and more modest formats. No lead imprints used on documents can be dated securely to the immediate post-conquest period, even though these were common in the Byzantine Empire before and after the Arab conquest.59 While the absence of text on administrative seals, as well as the size of the sealings, thus constituted a break, the images on Muslim clay sealings from Egypt of dogs, bulls, hares, birds, and igures of warriors or saints can already be found on pre-Islamic Byzantine and Sasanian sealings (Figures 5.6b, 5.7c, 5.9b).60 Similar patterns can be observed outside Egypt, where the documentary evidence is more scarce, however. Literary texts report that early Muslim of icials in Iraq also used seals decorated with animals and igures rather than texts.61 Late seventh-century clay sealings excavated in Qasr-I Abu Nasr near Shiraz in Iran similarly depict animals, and their size is smaller than those from the Sasanian period, but the background of the seals’ owners is not clear.62 56 For papyri and weights from Egypt, see Sijpesteijn, Shaping a Muslim State, chap. 2; for seals from Palestine, see Amitai-Preiss, “Umayyad Vocabulary” and “Four Umayyad Lead Sealings”; and, for al-Andalus, see Ibrahim, “Additions,” 119. 57 Sijpesteijn, Shaping a Muslim State, 117–24. 58 Ritter, Die altorientalischen Traditionen, 62; Cheynet, “L’usage,” 25; Soucek, “Early Islamic Seals,” 238. 59 Wassiliou, Siegel und Papyri, 4. 60 For Sasanian images, see Potts, The Arabian Gulf, 215; Ritter, Die altorientalischen Traditionen; and Friedenberg, “The Evolution,” 2, igs. 9–11. For Byzantine images, see Wassiliou, Siegel und Papyri. 61 Soucek, “Early Islamic Seals,” 246n57. 62 Ibid., 245. 125 EXPRESSING NEW RULE As expressions of material culture, seals re lected the development of Islamic art more generally. In the Muslim empire iconographic motifs and material applications circulated freely between the Byzantine and Sasanian cultural realms, resulting in new combinations of patterns, forms, and applications in new geographical and administrative contexts.63 Some iconographic motifs attested on sealings in Muslim Egypt and Syria-Palestine can be related to Sasanian designs, especially the use of celestial bodies, stars, the moon, and the sun. The introduction of the Sasanian motif of a beaded border surrounding an image can already be observed in the “decorative frame of knots” around the picture of a charging bull on the seal of the Arab conqueror and irst governor of Egypt, ‘Amr b. al-‘Āṣ (d. 664) (Figure 5.7c).64 While it might be dif icult to classify this as identi iably and speci ically Arab-Muslim, the Muslims’ application of existing motifs, forms, and usages differed from local practice starting immediately following the conquest. Not only the form and images on the seals changed. The Muslims also introduced innovations in the application of sealings in the conquered provinces that also point to Sasanian in luence. One innovation was to use clay sealings to close off securely the bottom part of a document that contained a summary of the contents of the text written out in full in the main part of the document.65 This practice is attested in documents from Muslim Egypt and Palestine. It was mainly applied to tax-demand notes and tax receipts, when the amount of taxes paid (or to be paid), the date, and sometimes the kind of taxes were repeated in the sealed-off part (Figures 5.2a, 5.4a–b, 5.8a–b). It was used in Arabic as well as Greek or bilingual Arabic/Greek documents, starting with a Greek papyrus dated 642 CE. The purpose is clear: in case of disagreement about the reliability of the visible text, the closed-off part could be opened to reveal the untouched, authentic contents. The use of sealings to close off part of a document was introduced by the Muslims, but the practice had a precedent in the Near East. Egyptian Ptolemaic so-called double documents, for example, display a similar practice. These legal documents contained twice the full text of a legal transaction. After the document had been signed and concluded, one part remained visible, while the second text was sealed off.66 Aramaic double documents were produced in Palestine 63 Grabar, Formation of Islamic Art. Note also the circulation of Byzantine bronze weights and the production of Egyptian-style glass weights throughout the Muslim empire: Khamis, “A Bronze Weight,” 149–54. 64 Priscilla Soucek describes the “decorative frame of knots” without identifying it as Sasanian: “Early Islamic Seals,” 248. This Sasanian motif was introduced via the Muslim empire on Byzantine seals in the late ninth century: Walker, “Islamicizing Motifs,” 389. 65 This innovation is extensively discussed in Sijpesteijn, “Seals and Papyri.” 66 Vandorpe, “Seals in and on the Papyri.” 125 126 126 PETRA M. SIJPESTEIJN around the same period, but the practice is already described for the seventhcentury BCE.67 In Ancient Mesopotamia such double legal documents existed as well.68 In Khurasan full double legal documents were produced in Bactrian into the Muslim period (Figure 5.3a). 69 The Muslim custom of protecting only a summary of the text is different and compares best with some procedures in Roman Egypt. This habit had disappeared from Egypt by the late Roman period, however, and there is also no evidence that it was practised elsewhere in the Middle East. Whether the Muslims had maintained an ancient custom in adjusted form, were exposed to it during the conquests, or revived this practice to it their needs remains to be determined.70 The increased use of single-sided sealings for identi ication and authentication in Muslim Egypt and Syria-Palestine may also be connected to Sasanian administrative practice, which spread in the Muslim empire to other areas. The common mention in documents to the sealings that appear on them and to the sealings’ function as a marker of authority and identi ication was a novelty in Muslim Egypt. Although seals were used to identify and verify Sasanian, Bactrian, and Soghdian documents, the practice was almost entirely unknown in pre-Islamic Egypt.71 As the letters quoted above show, the use of clay sealings was common in the Muslim chancery immediately following the conquest of Egypt and closely follows the application of clay sealings in the Sasanian chancery. Single-sided lead seal imprints used for identi ication and authentication outnumber double-sided lead sealings used for closing and securing in Muslim Syria-Palestine. This is opposite to the situation in the province under the Byzantines, and relates also to Sasanian practice, whereby single-sided seals were the only ones in use. 67 Sijpesteijn, “Seals and Papyri.” The prophet Jeremiah (627–586 BCE) took “the sealed deed of purchase containing the terms and conditions and the open copy” (Jer. 32:11). See also Jer. 32:14. 68 With the copy of the tablet text written on the outside envelope. In case of disagreement, the envelope was broken to reveal the tablet inside: Lerner and Skjaervø, “Some Uses of Clay Bullae,” 74n44. 69 A depiction of this document with the bottom part closed off appears in Sims-Williams, Bactrian Documents, vol. 3, pl. 72. 70 The practice of using stamp seals instead of signet rings was reinvented in the Sasanian Empire after several centuries of having been in disuse: Ritter, Die altorientalischen Traditionen. 71 In Roman Egypt the only example comes from tax receipts to be presented at checkpoints on the desert roads out to the oases, which contained authentication seals of the iscal of icials: Vandorpe, “Seals in and on the Papyri,” 250–53. For Soghdian and Sasanian practice, see Huff, “Technological Observations,” 382–83. For Bactrian, see Sims-Williams, Bactrian Documents, vols. 1 and 2. 127 EXPRESSING NEW RULE Arab Seals and Sealing Although no examples of sealings applied to documents are attested in preIslamic Arabia, written references and archaeological inds suggest that seals were regularly used for authentication and to regulate access. By the time inkwritten documents were in common use, from the fourth century CE at the latest, sealings were most probably applied in similar ways in Arabia as in the Byzantine and Sasanian contexts. These practices continued in the early Muslim community in Arabia and in luenced practices in the provinces conquered in the seventh century. Even before documents were in use in Arabia, practices can be observed that paved the way for the use of sealings for authentication and the control of access to written texts. The practice of sealing to close off and lock items was widely known in pre-Islamic Arabia, where seals made out of precious stone and metal have been found. Sasanian (but not Byzantine) seals have been found on the eastern Arabian Gulf coast.72 Himyarite (110 BCE–ca. 520 CE) seals have been unearthed in south Arabia carrying depictions of igurines and animals often accompanied by writing.73 The seals vary from gem-like precious stones for use in rings or worn on necklaces to metal stamp seals with a loop on the back to facilitate stamping.74 Identi ication practices were applied already in writings incised in stone. The use of personalized ways of signing (‘alāma) can be recognized in pre-Islamic rock inscriptions present all over the peninsula, in which authors identi ied their writings through speci ic ways of inscribing their name or adding symbols to their texts. The most explicit reference to the use of seals in ancient Arabia comes from a third-century Sabaic letter written on palm wood found in Yemen. The writer orders the addressee to “seal it [a document] with seal and wax.”75 How the seals would have been attached to the wooden documents remains an open question. The editor of this text, Peter Stein, suggests that the holes that appear on the left 72 In Tarut (Potts, The Arabian Gulf, 215), in the United Arab Emirates (Kennet, “Transformations,” 154), and in Bahrain (Lerner and Skjaervø, “Some Uses of Clay Bullae,” 71). 73 A black stone seal shows a three-faced head of a man and a Himyarite inscription with a name: Beeston, “Old South Arabian Antiquities,” 22–23, pl. III. A sardonyx seal has a depiction of an eagle with open wings standing on a bucranium with a star next to it and a late Sabaic inscription: Beeston, Pirenne and Robin, Corpus des inscriptions, 601–2. 74 See the 156 seals described in the Corpus of South-Arabian Inscriptions (dasi.humnet. unipi.it) dating from the second millennium BCE to the sixth century CE. 75 Stein, Die altsüdarabischen Minuskulinschriften, 32. Other documents written on palm leaves equally mention the use of wax sealings (e.g., ibid., 324, 541). I would like to thank Michael C. A. Macdonald, Ahmad al-Jallad, and Peter Stein for their help locating references to sealing practices in the ancient Arabian material. 127 128 128 PETRA M. SIJPESTEIJN extremity of several of the legal documents might have been to suspend a string and seal. The in luence of Near Eastern writing and sealing practices was facilitated by the introduction of the Arabic script on the Arabian Peninsula in the fourth century CE. Developed out of the Nabatean alphabet, Arabic is not an epigraphic system moulded in clay or carved in stone but, rather, a “pen-and-ink” writing system applied to parchment, leather, and animal bones, and probably also papyrus.76 The ancient Arabs’ interaction with the world around them was regular and intense, with writing habits being one of the many practices exchanged. Already, in the preIslamic period, Arab merchants, pilgrims, and travellers moved around the known classical world, while goods and objects—gifts, merchandise, utensils—from outside the peninsula circulated in Arabia. Documents, as well as the paraphernalia used to produce them, would surely have accompanied some of these movements. One likely venue of cultural exchange was the Arab Christian tribes who worked in the service of Byzantines and Sasanians to defend the desert borders. The Arab Christian Ghassanids and Lakhmids were thoroughly integrated into the Byzantine and Sasanian Empires respectively. The close af inity and interaction between Ghassanids and Arabians explains to a great extent the direction and sequence of the seventh-century Arab conquests. It also elucidates the continued use of Byzantine administrative practice and instruments in Muslim SyriaPalestine and Egypt, as well as their spread throughout the Muslim empire.77 The Ghassanid and Lakhmid use of seals thus most probably in luenced pre-Islamic Arabian usage, and explains the spread of Sasanian and Byzantine practice in the Muslim empire later on. The Ghassanids and Lakhmids were important cultural brokers whose experience with Byzantine and Sasanian administration and rule served the early Muslim empire well. A lead imprint of the last Ghassanid king, which depicts Christian symbols in combination with the Greek language,78 shows what form this in luence might have taken. It is clear that the earliest Arab-Muslim community in Arabia was indeed familiar with seals. When the prophet Muḥammad (d. 632) was allegedly told that foreign rulers would not accept his letters inviting them to join Islam without his personal seal attached, he decided to adopt one for himself. Reports differ as to what image was depicted, some stating it was a lion or human igure, others that 76 Macdonald, “Ancient Arabia.” See similar observations concerning the switch from cuneiform to Aramaic in Mesopotamia: Ritter, Die altorientalischen Traditionen, 108. For the presence of papyrus as a writing material in Arabia, see Sijpesteijn, Shaping a Muslim State, 1. 77 See, for example, the introduction of Aramaic-Greek terms by the Arabs in Egypt: ibid., 70n52. 78 The seal of the last Ghassanid king, Gabala patrikios (d. 640/41), contained crosses and a Greek legend: Shahid, Byzantium, 159. 129 EXPRESSING NEW RULE it was a legend with the words “Muḥammad is God’s messenger.”79 The Qur’an mentions the use of legal documents to ensure the correct recording of a transaction, adding that witnesses should be used when it is impractical to write the transaction down (Q 2:212). The use of seals to close off or lock is also attested in the Qur’an. Unbelievers who refuse to respond to God’s message are described as having their hearts and ears sealed by God (khatama Q 2:7; 6:46; 36:65; 42:24; 83:25). Muḥammad’s identi ication as the seal (khātim) of the prophets (Q 33:40), is generally interpreted as meaning that he was the last prophet sent by God: his prophecy completed and closed the row of Old and New Testament prophets, as a sealing inalizes and secures a container or document. Adaptations and reshapings surely followed the introduction of governmental practices and instruments in Arabia in response to local norms, implementation and usage, with a particular Arabian praxis developing as a result. As no documents dating from the earliest history of Islam have been preserved from Arabia, the only way to deduce such practice is from the changes implemented by the Muslims in the areas they conquered. Arabicization and Islamicization As discussed above, documentary practices, including the application and form of seals used on documents and goods, show that certain changes followed soon upon the Arab conquest in Egypt and Syria-Palestine. Information from other areas of the Muslim empire suggests that such changes occurred elsewhere too. In many ways, however, administrative routine was unaltered, and change is partially explained by the introduction of known practices into new geographical contexts and for different administrative purposes. It took another couple of decades following the conquests before the Muslim rulers implemented a concerted programme of revision in administrative instruments, organization, and institutions. Initiated under the Umayyad dynasty (661–750) from its capital, Damascus in Syria, administrative reforms continued to be regionally expressed as they interacted with local needs, usages, and conventions. Under Caliph Mu‘āwiya (r. 661–680), signi icant changes can be observed in administrative organization and practice in the papyri from Egypt and 79 Alan and Sourdel, “Khaṭām.” The appearance of the latter text stamped with ink on the letters written on leather ascribed to the prophet Muḥammad must date from the time (tenth century CE onwards) when such practice of stamping seals with ink had replaced the custom of using clay seals. The use of an Arabic text rather than an image would set Muḥammad’s seal apart from contemporary use by other Muslims. See above at notes 60 and 61 for examples of igural seals from early Islamic Egypt and Iraq. 129 130 130 PETRA M. SIJPESTEIJN Syria-Palestine.80 Mu’āwiya is the irst caliph whose name might have appeared on a bilingual Arabic/Greek papyrus protocol—preceding an Arabic/Greek note (54/674) from Nessana in Palestine requesting the delivery of oil and wheat to an Arab army unit.81 His name is reportedly also the irst to appear on monumental inscriptions. Mu‘āwiya is, moreover, credited with installing a bureau of seals and putting a special of icial in charge of it, but, as we noted, the Muslim administration also used seals before that period.82 Further changes were implemented at the end of the seventh century and the beginning of the eighth under the Umayyad Marwanid dynasty (684–750). Modi ications in the administration aimed at increased monitoring, centralized control, standardization, and professionalization were enacted throughout the Muslim empire. The goal was surely to increase the iscal income from the areas under caliphal control, which had become especially urgent when proceeds in the form of booty had substantially diminished with the slowing down of the conquests. An ampli ied self-awareness and self-con idence among the rulers, whose regime had survived ifty years of counter-attacks, civil war, and apparent imminent collapse but now seemed certain to stay, was also expressed in a desire for further Islamicization and Arabicization. The narrative sources indeed assign ‘Abd al-Malik (r. 685–705), and his sons who succeeded him, an ambitious programme of reforms. The Marwanid caliphs are more present in the documentary record in and outside Syria, their immediate area of governance. ‘Abd al-Malik is the irst caliph whose name appeared on lead and glass weights from Egypt and Syria-Palestine.83 His name also appears on an administrative papyrus from Nessana, while numerous papyrus protocols are issued on behalf of his sons. Arabicization of the administration was realized in different ways. The language of the chancery was decreed to be exclusively Arabic at the expense of local languages. The numismatic evidence, which is probably the most widely known and studied, shows a slow process of reform, with images being gradually superseded by the exclusive use of Arabic writing. Lead and bronze weights show a similarly gradual evolution from pre-Islamic via mixed to exclusively ArabicIslamic forms.84 It is from this time too that lead sealings bearing Arabic-only 80 Sijpesteijn, Shaping a Muslim State. 81 Kraemer, Excavations at Nessana, no. 60. See, however, the objections to the editio princeps, including the identi ication of this papyrus as a protocol, in Sijpesteijn, “Arabic Script and Language in the Earliest Papyri.” 82 Abbott, The Ḳurrah Papyri, 28. 83 Grabar, Formation of Islamic Art, ill. 21; Khamis “A Bronze Weight,” 151. 84 Khamis, “A Bronze Weight.” 13 EXPRESSING NEW RULE legends are attested, recording contributions by towns, administrative districts (ajnād), or whole provinces (miṣr; ilasṭīn; al-andalus) (Figures 5.5, 5.10, 5.11, 5.12, 5.13, 5.16, plate 5.1).85 These seals show a standardized formulation and use of technical terms.86 Egyptian papyrus protocols too start to appear in bilingual Arabic/Greek (as opposed to exclusively Greek ones) around this time, with the earliest caliph mentioned being al-Walīd (r. 705–715) and the irst entirely Arabic protocol dating to Caliph Hishām (r. 724–743).87 It appears therefore that the introduction of Arabic on public objects transcending local usage was generally applied throughout Syria-Palestine and Egypt in more or less immediate response to the court-initiated programme.88 Local languages continued to be used and learned in the Muslim chancery for a time, appearing on administrative instruments such as seals, coins, weights, and documents. In Egypt Greek continued to be actively used into the ninth century, while Coptic was used for local administration even longer. In Syria-Palestine Syriac and Aramaic disappeared more quickly in the face of Arabic. In Egypt Greek lingered on seals, even on those used by Muslim of icials, into the eighth century. Images too continued to igure on the clay seals of of icials in Egypt (Figures 5.6b, 5.9b). Similarly, in Khurasan depictions of animals and other images persisted on of icial seals into the middle of the eighth century (Figures 5.3b, 5.14b, 5.15b).89 Eighth-century bronze and lead weights from Egypt, Syria-Palestine, and Iraq show both the continued use of Greek alongside Arabic and the application of pre-Islamic imagery.90 In spite of this transition period, a irm trend towards a speci ically Arab(ic)Islamic sealing usage and practice is visible in the clay and metal sealings from the eighth century onwards and from all over the Muslim empire. Some striking features were introduced on the sealings, which cannot be traced back to Byzantine or Sasanian practice. Others show how Sasanian practice, sometimes combined with local “Byzantine” traits, was introduced in Egypt and SyriaPalestine. A irst difference concerns the identi ication of of icials on the seals. Arab(ic)-Islamic seal inscriptions are speci ically characterized by the general absence of titles. While Byzantine and Sasanian seals mention the (name and) 85 Amitai-Preiss, “Umayyad Lead Sealings,” 19–20; Schindel, “Nochmals” and “Eine Umayyadische Bleibulle,” 118–21. 86 Amitai-Preiss, “Umayyad Vocabulary.” 87 Sijpesteijn, Shaping a Muslim State, 110–11. 88 In general on ‘Abd al-Malik’s reforms, see Robinson, ‘Abd al-Malik, chap. 4. 89 Sijpesteijn, “Seals and Papyri”; Wassiliou, Siegel und Papyri; Khan, Arabic Documents, 83. The district manager of the Fayyum (in of ice ca. 720–750), N ājid b. Muslim, used a seal with his name written out in Arabic and Greek. 90 Khamis, “A Bronze Weight.” 131 132 132 PETRA M. SIJPESTEIJN title or of ice of of icials, Arab(ic)-Islamic seals contain only personal names, either written out or symbolized by an image, sometimes in combination with a pious formula (Figures 5.1b, 5.9b).91 This conforms to the practice in use within the papyri in which bilingual Greek-Arabic texts show the common application of titles in the Greek part while the Arabic section lacks such attributions, emphasizing instead lineage and membership via patronymics and other community identity markers.92 Only the governor and caliph are regularly referred to as amīr and amīr al-mu’minīn respectively, possibly an indication of their elevated status (Figure 5.12).93 Two other innovations on Arab(ic)-Islamic seals, which break with Byzantine and Sasanian usage, are related to technical terms and expressions attested on the sealings themselves. The irst concerns the addition of years to lead sealings imprinted on bags or containers (Plate 5.1a–b).94 This practice is irst attested on a series of lead and brass seals recording tax payments “by the inhabitants of Egypt,” which carry dates between 93 and 95 AH (712–714 CE). Such use of dates occurs only on “impersonal” seals—that is to say, on seals referring to institutions, of ices, the Muslim chancery, or administration in a general sense, without mentioning personal names. The use of words referring to the seal on the sealing itself is another practice that is not found on pre-Islamic seals but appears for the irst time on eighth-century Arabic seals. The words “seal” (khātim) and “stamp” (ṭābi’) are attested on the sealings and discussed in the documents on which the sealings appear (Figures 5.3a, 5.5a, 5.8c, 5.14a, 5.16a).95 Another group of eighth-century lead seal imprints listing administrative territorial divisions, but with no names of of icials in charge, is the product of measures to professionalize the Muslim bureaucracy. Mostly from Filastin but also from al-Andalus and Egypt, these sealings combine names of towns, subdistricts (iqlīm), districts (kūra), or entire provinces (Figures 5.5a–b, 5.10a–b, 5.16a–b; Plate 5.1a–b).96 They record annual tax payments and refer to the Muslim taxcollecting authorities as an institution rather than a person, thereby re lecting the 91 For Byzantine seal inscriptions, see Cheynet, “L’usage,” and Amitai-Preiss, “Umayyad Vocabulary”; for Sasanian examples, see Ritter, Die altorientalischen Traditionen, 61–62. 92 Grohmann, From the World, 121; Sijpesteijn, Shaping a Muslim State, 67–68. 93 See also Amitai-Preiss and Farhi, “A Small Assemblage,” no. 1. 94 No dates were used on Byzantine seals: Balog, “Dated Aghlabid,” 127. 95 See above at note 31. Amitai-Preiss, “Umayyad Vocabulary,” 281–82; Alan and Sourdel, “Khaṭām.” For the use of ṭābi’ in the papyri, see Sijpesteijn, Shaping a Muslim State, 317–18. 96 Amitai-Preiss, “Kūra and Iqlīm.” 13 EXPRESSING NEW RULE Plate 5.1a–b. Copper seal recording a tribute or poll tax payments made by the inhabitants of Egypt in the year 95 (713/14): (a) min ahl miṣr, (b) sanat khams wa-tisa`īn. Walker, Catalogue, 295; Porter, Arabic and Persian Seals, p. 28, no. 9; British Museum, 00803873001. © Trustees of the British Museum. administrative reorganization of the early eighth century, when Muslim administrative bureaucrats replaced local Christian and Jewish landholding of ice holders. Again, the Egyptian papyri offer further information, in this case about the territorial division in place and about the process of administrative transformations as referenced on the sealings as well. At the lowest level in the administration stood the villages, rural estates, monastic communities, and urban neighbourhoods. Village headmen and other local representatives were responsible for the collection of taxes from individual taxpaying residents in the communities. Initially operating as an intermediary between the Muslim authorities and the indigenous taxpayers, these local representatives had also been responsible for the division of the taxes imposed in one lump sum on the communities. From the early eighth century onwards tax collectors, still drawn from the indigenous population, lost this autonomy and operated in the service of the Muslim isc, which assigned taxes on individual taxpayers directly. The local collectors brought the taxes gathered from individual taxpayers to a subdistrict, called an iqlīm in Syria-Palestine and ḥayyiz in Egypt (Figures 5.2a, 5.16a–b). From there the taxes of the different communities were forwarded in one lump sum to the district capital (madīnat al-kūra). From the district the payments were dispatched to the next level in the administrative geography of the province. In Egypt the districts were in direct contact with the provincial capital until a subdivision in Upper and Lower Egypt was introduced in the ninth century. In Syria-Palestine the ajnād functioned as supraregional administrative units that also played a military role. Then the taxes advanced to the capital of the province. At the provincial capital the tax taken was used to pay stipends to Muslim inhabitants according to their place on the dīwān (register), to contribute to the maintenance of the land, the upkeep of the road system and other public services, and the sustenance of the governor and his entourage, and, inally, to send as dues to the caliph’s 133 134 134 PETRA M. SIJPESTEIJN court.97 In spite of the continuous complaints that payments from the provinces were missing, insuf icient, or late, the sealings referring to payments made by the (inhabitants of) provinces make it clear that these payments were indeed destined for the caliphal capital (Figure 5.13a–b; Plate 5.1a–b).98 Some of the “impersonal” sealings that re lect the new administrative order put in place in the early eighth century seemingly record the payment of poll tax contributions by non-Muslim subjects. Such sealings are known from Egypt and Palestine from the time of Caliph ‘Abd al-Malik onwards (Figure 5.5a–b; Plate 5.1a–b).99 Papyri report that poll tax payments were in fact already in place, newly introduced by the Arab rulers, directly following the conquest in Egypt, as they were presumably elsewhere.100 In al-Andalus similar seals were used immediately following the Arab conquest in the early eighth century to record payments made by communities on the Iberian Peninsula (Figures 5.10a–b, 5.11a–b, 5.13a–b).101 Such poll tax or tribute payments contain a place name, sometimes accompanied by a reference to the stamp having been issued in the place (e.g., bi-ludd, “issued in Ludd”), or that this is the seal of the town (e.g., khātim ludd, “Ludd’s seal”)102 or district (khātim kūrat ‘asqalan; kūrat ‘amwās).103 Payments for a whole province are also referred to on such seals (qasm al-andalus, “allotment of al-Andalus” [Figure 5.13a–b]).104 In some cases a reference to the (non-Muslim) population appears as well (e.g., min ahl miṣr, “from the people of Egypt” [Plate 5.1a–b];105 bi-sm llāh ahl ashbīla, “in the name of God, the people of Seville” [Figure 5.10a–b];106 khātim kūra ṭabariyya, “seal of Tiberias”; yahūd ṭabariyya, “Jews of Tiberias” [Figure 5.5a– b]),107 or to the peace treaty that gave rise to the payments (e.g., muṣālaḥat libīra, 97 Sijpesteijn, Shaping a Muslim State, 85–91; Amitai-Preiss, “Islamic Lead Coins,” 15. 98 Robinson, “Neck-Sealing.” See also the list of payments made by the province of Egypt in the years 221–278 AH (836–892 CE) recorded on a papyrus found in Samarra in which Egypt igures prominently: Reinfandt, “Administrative Papyri.” 99 Amitai-Preiss, “A Poll Tax Seal”; Robinson, “Neck-Sealing,” 404. 100 Sijpesteijn, Shaping a Muslim State, 72–74. 101 Amitai-Preiss, “A Poll Tax Seal”; Ibrahim, “Additions.” 102 Amitai-Preiss, “Umayyad Lead Sealings.” See also the examples mentioned by AmitaiPreiss, “A Poll Tax Seal,” of seals mentioning, on one side, khātim (“seal”) of a district’s name and, on the other side, iqlīm (“district”), followed by another geographical name. 103 Amitai-Preiss, “Islamic Lead Coins,” 15–16, nrs. 10–11. 104 Ibrahim, “Nuevos documentos.” 105 Four such copper seals are known, dated to 93 and 95 AH (712–715 CE): Schindel, “Nochmals.” 106 Ibrahim, “Nuevos documentos.” 107 Amitai-Preiss, “A Poll Tax Seal.” 135 EXPRESSING NEW RULE “the peace agreement of Elibirri”;108 muṣālaḥat arḍ jayyān, “the peace agreement of the land of Jaen” [Figure 5.11a–b]).109 Sealings discovered in southern France record booty captured from coastal towns raided by the Arabs, 110 and contain Arabic terms known from legal and administrative manuals. These lead sealings from al-Andalus, Egypt, Syria/Palestine, and southern France protected payments intended for the central Muslim administration. Such examples from across the empire all share an avoidance of images and decorations and reveal a common script and technical vocabulary.111 Another administrative change related to the execution of the collection of the poll tax is visible in early eighth-century sealings. As part of the administrative reforms under the Marwanids, centralizing measures were introduced to increase control over the local administrations and their executives, as well as individual taxpayers and their possessions, in order to secure a higher tax uptake. Closer supervision of taxpayers, their possessions, and their movements resulted in a higher tax burden, while the stronger state presence caused disaffection among the subjects. Protest took different forms, from tax evasion by running away from one’s place of residence to outright revolt. The Muslim authorities struck back with a measure that was as effective as it was humiliating: lead sealings fastened around the necks or wrists of non-Muslim taxpayers identi ied them as liable to the non-Muslim poll tax, while at the same time disgracing them by associating them with slaves and captives. Although no lead sealings that can be connected to the bodily sealing of individual taxpayers dating to the eighth century have been identi ied, such objects do exist, but from a later period.112 Contemporary literary sources from Egypt and northern Iraq give detailed descriptions of the practice. The custom of putting lead sealings around the necks and hands of captives and slaves was, moreover, practised in the Sasanian Empire for the purpose of humiliation, punishment, and identi ication. The Muslims were the irst in the Near East, however, to apply this debasing practice in a iscal context. The measure was not applied continuously and generally but, rather, represented extraordinary punitive or overzealous measures, which, for logistical and moral reasons, could not be maintained in the long term.113 108 Ibrahim, “Additions,” 116, no. 1. 109 Ibrahim, “Nuevos documentos,” 153–55, no. 10, with corrections in the addenda. 110 See also maghnūm ṭayyib qusima bi-arbūna, “licit booty distributed in Narbonne”: Marichal and Sénac, “Sceaux arabes.” 111 Amitai-Preiss, “Umayyad Vocabulary.” 112 Robinson, “Neck-Sealing,” 427. 113 Robinson emphasizes that neck sealing, which initially served only a humiliating and stigmatizing function, from the early eighth century to the tenth served an additional, iscal 135 136 136 PETRA M. SIJPESTEIJN Archaeological material from the most western part of the Muslim empire, alAndalus, shows how the new Muslim administrative standards introduced under the Marwanids spread throughout the empire. The Arab conquerors who crossed over into the Iberian Peninsula in 711 were closely related to the Umayyads of Egypt and Syria-Palestine. They introduced the same kinds of lead sealings, recording (in Arabic) and protecting the payments made by towns and communities to the new rulers. While the speci ic terminology referring to payments made as part of the peace agreements (ṣulḥ, muṣālaḥa) and the status of the conquered land ( fay) is not attested elsewhere, the invading armies had obviously already incorporated the linguistic and administrative changes initiated further east (Figures 5.10a–b, 5.11a–b, 5.13a–b).114 Some of the vocabulary and expressions in use on these early eighthcentury Arab(ic)-Islamic seals compare well with those in use elsewhere in the empire and point to a common administrative culture (Figures 5.5a–b, 5.16a–b; Plate 5.1a–b).115 No clearly identi iable images appear on the Andalusian lead seals at all.116 Global and Local Trends The creation of one Muslim empire uniting the former Byzantine eastern Mediterranean provinces and the Sasanian Empire facilitated the sharing of practices and forms across a large area. Falling under one political system and sharing many linguistic, economic, legal, and social features meant that, transcending local speci icities, local material repertoires, and administrative practices, they were distributed more widely and more rapidly than before. In this way, Sasanian and purpose, even though it was always secondary to the humiliation (“Neck-Sealing,” 417ff.). The measure might also be connected to administrative changes aimed at strict centralized control, which were introduced early in the eighth century and led to censuses and land surveys, but which were not maintained because of logistic dif iculties: Sijpesteijn, Shaping a Muslim State, 94–102. Extraordinary circumstances in eighth-century Middle Egypt motivated increased record-keeping of the movement of people: Delattre, “Checkpoints”; Boud’hors, “L’apport.” See also the Arabic safe conducts and tax receipts that had to be shown to of icials upon request: Frantz-Murphy, Corpus Papyrorum Raineri, 105–10. 114 Ibrahim, “Additions,” 116, no. 1; 118, no. 3; Ibrahim “Nuevos documentos,” no. 14. Other terminology so far only attested in the Andalusian material is qasm: Ibrahim, “Additions,” 121, 125–26, nos. 6–8; Figure 5.13a in this essay. 115 See the use of arḍ, “land,” with a provincial or place name, which appears on lead seals from al-Andalus and Syria-Palestine (Ibrahim, “Nuevos documentos”; Amitai-Preiss, “Arḍ and Jund,” 134–41) and which is also a frequently attested term in papyri from administrative and iscal contexts: Sijpesteijn, Shaping a Muslim State. The reference on seals from al-Andalus and Egypt to payments made by the inhabitants (ahl) of the province mirrors the use of that term in the papyri to refer to the taxpaying subjects (ahl al-arḍ), and might be compared to the identi ication “Jews” (yahūd) on a sealing from Tiberias. 116 Ibrahim, “Nuevos documentos.” 137 EXPRESSING NEW RULE Byzantine shapes, structures, and procedures extended more easily beyond their respective imperial borders. The newly introduced forms and practices also underwent adjustments as they were put to local use and itted to local needs. Seals bear witness to this active process of adjustment, reshaping, and expansion, which resulted in a shared but locally de ined administrative material culture.117 In their form and application, Arab(ic)-Islamic seals show an in lux of Near Eastern in luences that were given new meaning and functions in the Muslim empire and implemented locally. The introduction of the common practice in the Sasanian Empire of adding a lead seal to the neck or hands of captives and slaves, also occasionally performed in the Byzantine Empire, was given a new function and application in the Muslim empire. Signi icantly, the Muslims were the irst to use identifying lead neck-seals in a iscal context; in doing so, they adjusted a Sasanian practice and implemented it throughout the Muslim empire.118 Decorative elements such as stars and crescents, frequently encountered on Sasanian seals, found their way onto the seals of the Egyptian governor Qurra b. Sharīk (in of ice 709–715), where they appear together with the animal igures commonly used on Byzantine seals (Figure 5.9b).119 Stars are also used on early eighth-century seals and coins in al-Andalus and Palestine.120 In eighth-century Khurasani documents star-covered clay sealings inalize the transaction (Figures 5.14a–b, 5.15a–b). An increased presence of one-sided sealings, in clay and lead, might also point to the in luence of Sasanian sealing practice (which exclusively used one-sided seals, as opposed to the double-sided ones commonly used in the Byzantine Empire) upon the whole Muslim empire (Figure 5.12).121 The Byzantine use of lead seal imprints arrived in al-Andalus with the Arab conquerors, who spread this practice westwards.122 Other administrative practices and measures newly introduced by the Muslims found their ways onto the seals in different parts of the empire. Arḍ, “land,” is used in papyri from Egypt, as well as on sealings from al-Andalus and Palestine, with the technical meaning of land under the control of Muslims and subject to taxation. Another technical term related to taxes is ahl, “people,” attested on papyri and 117 For a similar pattern of sharing of Byzantine and Sasanian forms to the development of an Islamic form with local expressions in the domain of material culture, see Grabar, Formation of Islamic Art. See also Knappett “Imprints.” 118 Robinson, “Neck-Sealing,” 405. 119 Sijpesteijn, “Seals and Papyri,” 173. The introduction of stars on seals does not have to be related to eastern practice, as evidenced in eighth-century Khurasani documents: Khan, Arabic Documents, 87–88. 120 Ibrahim, “Nuevos documentos”; Amitai-Preiss, “Four Umayyad Seals,” no. 2. 121 Amitai-Preiss and Farhi, “A Small Assemblage,” no. 1. 122 See above, note 63, for the distribution of glass, lead, and bronze weights throughout the different provinces of the Islamic empire. 137 138 138 PETRA M. SIJPESTEIJN sealings from al-Andalus, Syria-Palestine, and Egypt to designate the taxpaying subjects in lands conquered by the Arabs. Other examples include the term ṣulḥ, “amicable treaty,” which appears on sealings from al-Andalus and southern France recording tribute payments and booty. Local practices thrived as well, with many of the forms, expressions, and technical terms found on seals being speci ic to a particular province or district. Uniformity was imposed from the Marwanid court at the beginning of the eighth century, with Arabic-only seals becoming dominant, albeit not exclusive everywhere. Administrative hierarchies became settled and were represented uniformly and consistently even if this was initially most visible at the level of single provinces.123 Regional differences continued to exist, with local political powers and customs maintaining an in luence. In the ninth century population movements from the eastern part of the Muslim empire to the west introduced new administrative terminologies and practices from the eastern provinces westwards. Armies and government of icials were sent from Baghdad to subdue Egypt, bringing it (temporarily) back under the wings of the Abbasid caliphate. The diminishing economic situation in Baghdad, moreover, drew civil servants westwards to the more prosperous Egyptian capital. These administrators introduced practices from the eastern part of the empire to Egypt and Syria-Palestine, which generally fell under Egyptian in luence, or even outright political control, at this time.124 The application of multiple imprints of the same seal on one piece of clay was introduced into Egypt at this time. For example, the jahbadh (paymaster) Yalahwayh imprinted his seal twice at the bottom of a document dated 291 (904) as proof that he had received the taxes mentioned in the papyrus (Figure 5.8c).125 This practice can be connected to similar practices common in legal documents from the eastern part of the Muslim empire. The custom of applying multiple seals to one lump of clay had been practised by Sasanian legal of icials.126 Their placement and size indicated the hierarchy of the parties involved. Sometimes multiple imprints were impressed by the same seal belonging to the main of icial present.127 A similar development can be observed in Ptolemaic Egypt, where the use of multiple seals on one piece of clay had been replaced by the practice of printing the seal of one of icial on clay and the signatures of additional witnesses surrounding it.128 Arabic legal documents from mid-eighth-century Khurasan have large seals with multiple 123 Amitai-Preiss, “Umayyad Vocabulary” and “Kūra and Iqlīm.” 124 Khan, Arabic Documents, 83, and “The Pre-Islamic Background.” 125 Wassiliou, Siegel und Papyri, 38, no. 29. 126 Macuch, “The Use of Seals.” 127 Ritter, Die altorientalischen Traditionen, 63. 128 Vandorpe, “Seals in and on the Papyri,” 234. 139 EXPRESSING NEW RULE imprints from seals and thumbnail marks belonging to the witnesses to the transaction (Figure 5.14b).129 In its move westwards, the practice of applying multiple seals to one lump of clay thus shifted to the documentary domain, appearing in Egypt on administrative rather than legal documents. Although no clay sealings appear on legal deeds, the eastern practice of witnesses adding their personal seal to their name might have impacted the practice of witnesses using a personalized written sign (‘alāma) behind their name, which becomes commonly practised in Arabic papyri from the ninth century onwards.130 The introduction of paper as a writing material profoundly changed the writing culture in the Middle East.131 Introduced from China via Central Asia, the production of paper was much cheaper than that of parchment, leather, or papyrus. Fabricated from used textiles, paper could be produced everywhere and anywhere. The introduction of this writing support was accompanied by other practices. One such related innovation was the use of ink-stamped seals, which led to innovations in seals and signet rings as well. The irst paper mills appeared in Baghdad in the eighth century, and paper reached Egypt by the ninth century, becoming dominant by the tenth. Clay seal imprints can be found on paper documents from the tenth century, but by the eleventh century ink stamps had replaced clay sealings.132 Conclusion The use and format of sealings in Syria-Palestine and Egypt show an interesting and revealing blend of continuities and changes under the Muslims. With seals and seal imprints dating from directly after the conquest of these areas, the employment and appearance of seals can be connected to practices introduced by the new rulers as well as subsequent changes taking place under a developing and changing Muslim state. The meaning and function of seals as a means for identi ication and authority, on the one hand, and closure and protection, on the other, persisted throughout this period. How these functions were expressed, what materials were used, how the seals were applied to objects, and what they looked like differed depending on local conditions and governmental measures, at both the regional and central levels. Practical considerations, such as the degree of interaction between scribal and administrative cultures before and after the conquests and the level of interaction between subjects and rulers as a result of settlement patterns, played a role. Symbolic 129 Khan, Arabic Documents, 83. The use of ingerprints in clay seals is known from Ptolemaic Egypt (Vandorpe, “Seals in and on the Papyri,” 236) and from the ancient Near East: Ritter, Die altorientalischen Traditionen, 105. 130 Sijpesteijn, “Seals and Papyri.” 131 Bloom, Paper before Print. 132 Grohmann, Einführung, 129. 139 140 140 PETRA M. SIJPESTEIJN considerations, such as the contacts (hostile, diplomatic, and economic) with the Byzantine Empire and the relation between Muslim and non-Muslim populations, as well as processes of Arabicization and Islamicization, were also important.133 Finally, political relations between the provinces and the centre of the caliphate, as well as the dynamics of multicentrism, imperial ambition, and the formation of an Islamic empire, also impacted the way the Muslims expressed their rule and ideology in administrative measures, building projects, and material artefacts such as seals. The exclusive use of the Arabic language by the administration was one example of this. Even taking into account accidents of preservation, a pattern of change disseminating from the political centre to the periphery with faster implementation in the public domain, in which central control was greater, is visible in the material record. The oldest protocol papyrus that includes Arabic is dated to the caliphate of Mu‘āwiya (r. 661–680), and it was used in Nessana, Palestine.134 In Egypt, located further from the centre, the earliest bilingual Arabic/Greek protocols date to the caliphate of al-Walīd (r. 705–715).135 Arabic, though present from the beginning of Muslim rule, did not monopolize the chancery (dīwān), with local languages being actively promoted for at least a century following ‘Abd al-Malik’s decree ordering the exclusive use of Arabic. The oldest known Arabic lead weight was struck in Palestine with ‘Abd al-Malik’s name and additional Arabic legends on it, and a contemporary bronze weight belonging to ‘Abd al-Malik’s governor in Iraq, al-Ḥajjāj (d. 714), has Arabic struck over an original Greek legend.136 Arabic sealings mentioning the name and title of Caliph ‘Abd al-Malik (r. 685–705), his sons and Caliphs al-Walīd (r. 705–715), Sulaymān (r. 715–717), and Hishām (r. 724–743), as well as Caliph Marwān II (r. 744–750), are found in Syria-Palestine.137 Sealings with Arabic start to be used for institutions and administrative domains particular to Muslim rule, such as the poll tax or tribute payments by communities of non-Muslims and the division of booty gained in raiding from the 710s onwards. The use of local languages and the continued use of igures—rather than writing—on seals, even on those belonging to government of icials, continues into the eighth century in Egypt, Iran, and Khurasan. In Armenia no Arabic seals have been found before the ninth century.138 On the other hand, no administrative seals with depictions are known from al-Andalus or SyriaPalestine. Syria-Palestine, being at the centre of the Umayyad Empire, was where 133 See also the exoticizing motive at play in the application of Islamic motifs on ninth- and tenth-century Byzantine seals: Walker, “Islamicizing Motifs.” 134 Kraemer, Excavations at Nessana, no. 60, but see above at note 81. 135 Grohmann, Protokolle. 136 Khamis, “A Bronze Weight.” 137 Amitai-Preiss, “Four Umayyad Lead Sealings,” 176. 138 Ninth-century layers in Dvin yielded Arabic seals: Huff, “Technological Observations,” 376. 14 EXPRESSING NEW RULE policy changes, including the change to Arabic, were irst implemented. While distance thus played a role, the production of seals used by provincial of icials was not subject to central control and scrutiny the way that seals for public use were.139 Individual governors or governments especially closely connected to the caliphal court, such as Qurra in Egypt, al-Ḥajjāj in Iraq, or even the Umayyad conquerors of al-Andalus, implemented administrative customs more readily. Separate traditions continued to exist on both sides of cultural-political borders. In Anatolia and Syria—the areas that the Byzantines recaptured from the Muslims in the eleventh and twelfth centuries—Arabic-speaking Christians were drawn into the Byzantine administration. Their seals in Arabic, often in combination with Syriac and Greek, and bearing prominent Christian symbols such as crosses and igures of saints, are preserved in the dozens. No such imagery is attested on administrative seals of the area under earlier Muslim rule.140 Christian-Arab of icials did of course work in substantial numbers in the Muslim administration, and their seals mentioning their names in Arabic have been preserved in Egypt (Figures 5.2b, 5.8c).141 Conversely, only a very small number of bilingual Arabic/Greek Byzantine seals are attested from the earlier period belonging to Arabs working in the Byzantine administration, while Islamicizing motifs found their way onto some Byzantine seals of the ninth and tenth centuries.142 Byzantine lead sealings for documents continued to be imprinted on both sides and identi ied the of icials mentioned on them by their titles, contrary to the practices in Arab(ic)-Islamic seals as described above. As opposed to public inscriptions, protocol texts, or coins, which serve a longterm function transcending temporal and con ined spaces, seals serve immediate and localized needs while still operating in the public sphere and participating in the large-scale administrative and political processes of the time. These Arabic/ Greek Christian Byzantine seals show how malleable and lexible the medium was, and how scribal and administrative habits, cultural and religious preferences, languages and applications were mixed, rejected, or revived in the service of new masters in new historical realities. 139 Compare the much slower introduction of Arabic-only legends on Umayyad copper coins, which were locally produced beyond central supervision, as opposed to silver and gold coins, the minting of which was centrally controlled. I would like to thank Kees Uitenbroek for pointing this out to me. 140 A small number of Byzantine seals containing Arabic, and in some cases images of animals or Christian symbols, date to the seventh/eighth centuries: Heidemann and Sode, “Metallsiegel,” 48, and “Christlich-orientalische Bleisiegel,” nos. 1, 2, 3. An Arabic seal bearing a saint’s depiction on one side and an Arabic legend on the other belonged to a Byzantine family in the service of the Seljuks in Asia Minor: Heidemann and Sode, “Metallsiegel,” 48. 141 See, for example, Antūna ibn Karīl, whose seal appears on a document dated 862: Wassiliou, Siegel und Papyri, 38, no. 30. 142 Walker, “Islamicizing Motifs.” 141 142 142 PETRA M. SIJPESTEIJN Bibliography Abbott, Nabia. The Ḳurrah Papyri from Aphrodito in the Oriental Institute. Chicago: University of Chicago Libraries, 1938. Alan, James, and Dominique Sourdel. “Khaṭām.” In Encyclopaedia of Islam, 2nd ed., edited by Emeri van Donzel, Bernard Lewis, and Charles Pellat, vol. 4, 1002. Leiden: Brill, 1998. Amitai-Preiss, Nitzan. “A Poll Tax Seal of Tiberias.” In Hammath Tiberias, edited by Moshe Dothan, vol. 2, 104–5. Jerusalem: Israel Exploration Society, 2000. ——. “An Umayyad Lead Seal with the Name of the Caliph Marwān b. Muḥammad.” Al-Qantara 18 (1997): 233–42. ——. “Arḍ and Jund.” Israel Numismatic Journal 19 (2016): 133–41. ——. “Early Islamic Lead Seals: Evidence for Commercial Connections.” Israel Numismatic Journal 19 (2016): 106–14. ——. “Faunal Iconography on Islamic Seals.” Studies in Byzantine Sigillography 9: 207–12. ——. “Four Umayyad Lead Sealings from Palestine.” Israel Numismatic Journal 16 (2008): 176–78. ——. “Islamic Lead Coins, Weights and Seals in the Israel Museum.” Israel Museum Studies in Archaeology 6 (2007): 13–20. ——. “Kūra and Iqlīm Sources and Seals.” Israel Numismatic Journal 19 (2016): 120–32. ——. “Umayyad Lead Sealings.” In The 2nd Simone Assemani Symposium on Islamic Coins, edited by Bruno Callegher and Arianna D’Ottone, 18–21. Trieste: Edizioni Università di Trieste, 2010. ——. “Umayyad Vocabulary on Administrative Objects from Palestine.” In The 3rd Simone Assemani Symposium on Islamic Coins, edited by Bruno Callegher and Arianna D’Ottone, 280–87. Trieste: Edizioni Università di Trieste, 2012. Amitai-Preiss, Nitzan, and Yoav Farhi. “A Small Assemblage of Lead Sealings, Weights and Coins from the Early Islamic Period.” Israel Numismatic Journal 17 (2010): 233–37. Amitai-Preiss, Nitzan, and Oren Tal. “A Lead Bulla with the Place Name Arsūf from Apollonia-Arsuf.” Israel Numismatic Research 10 (2015): 191–205. Balog, Paul. “Dated Aghlabid Lead and Copper Seals from Sicily.” Studi Maghrebini 11 (1979): 125–32. Beeston, Alfred Felix Landon. “Old South Arabian Antiquities.” Journal of the Royal Asiatic Society of Great Britain and Ireland 84 (1952): 20–23. Beeston, Alfred Felix Landon, Jacqueline Pirenne, and Christian J. Robin. Corpus des inscriptions et antiquités sud-arabes, vol. 1, Inscriptions. Leuven: Peeters, 1977. 143 EXPRESSING NEW RULE Bloom, Jonathan M. Paper before Print: The History and Impact of Paper in the Islamic World. New Haven: Yale University Press, 2001. Boud’hors, Anne. “L’apport de papyrus poetérieurs à la conquête arabe pour la datation des ostraca coptes de la tombe TT29.” In From al-Andalus to Khurasan: Documents from the Medieval Muslim World, edited by Petra M. Sijpesteijn, Lennart Sundelin, So ía Torallas Tovar, and Amalia Zomeño, 115– 30. Leiden: Brill, 2007. Cheynet, Jean-Claude. “L’usage des sceaux à Byzance.” In Sceaux d’Orient et leur emploi, edited by Rika Gyselen, 23–40. Bures-sur-Yvette: Groupe pour l’Étude de la Civilisation du Moyen-Orient, 1997. Delattre, Alain. “Checkpoints, sauf-conduits et contrôle de la population en Égypte au début du VIIIe siècle.” In Authority and Control in the Countryside: From Antiquity to Islam in the Mediterranean and Near East (6th–10th Century), edited by Alain Delattre, Marie Legendre, and Petra M. Sijpesteijn, 530–45. Leiden: Brill, 2018. Donner, Fred M. Muhammad and the Believers: At the Origins of Islam. Cambridge, MA: Harvard University Press, 2010. Dorpmüller, Sabine. “Seals in Islamic Literature.” In Seals and Sealing in the Near East, ed. Regulski et al., 189–208. Frantz-Murphy, Gladys. Corpus Papyrorum Raineri: Arabic Agricultural Leases and Tax Receipts from Egypt 148–427 AH/765–1035 AD. Vienna: Brüder Holinek, 2001. Friedenberg, Daniel M. “The Evolution and Uses of Jewish Byzantine Stamp Seals.” Journal of the Walters Art Gallery 52/53 (1994): 1–21. Frye, Richard N. “Sasanian Seals and Sealings.” In Mémorial Jean de Menasce, edited by Philip Gignoux and Ahmad Tafazzli, 155–62. Louvain: Imprimerie orientaliste, 1974. Grabar, Oleg. Formation of Islamic Art, rev. ed. New Haven: Yale University Press, 1987. Grohmann, Adolf. Einführung und Chrestomathie zur arabischen Papyruskunde. Prague: Státní Pedagogické Nakladatelství, 1954. ——. “Einige bemerkenswerte Urkunden aus der Sammlung der Papyrus Erzherzog Rainer an der Nationalbibliothek zu Wien.” Archiv Orientální 18 (1950): 80–119. ——. From the World of Arabic Papyri. Cairo: Ma‘aref Press, 1952. ——. “Probleme der arabischen Papyrusforschung II.” Archiv Orientální 6 (1934): 381–94. ——. Protokolle. Vienna: Burgverlag Ferdinand Zöllner, 1924. Gyselen, Rika. La géographie administrative de l’Empire Sassanide: les témoignages sigillographiques. Paris: Groupe pour l’Étude de la Civilisation du MoyenOrient, 1989. 143 14 144 PETRA M. SIJPESTEIJN —— . Sasanian Seals and Sealings in the A. Saeedi Collection. Leuven: Peeters, 2007. Haldon, John F. “Introduction: Greater Syria in the Seventh Century: Context and Background.” In Money, Power and Politics in Early Islamic Syria: A Review of Current Debates, edited by John F. Haldon, 1–20. Farnham: Ashgate, 2010. Heidemann, Stefan, and Claudia Sode. “Christlich-orientalische Bleisiegel im orientalischen Münzkabinett Jena.” ARAM 11/12 (1999): 533–93. ——. “Metallsiegel in der islamischen Welt: ihre Forschungsgeschichte und orientalische Bleisiegel aus einem Siegelfund in Konstantinopel .” In Sceaux d’Orient et leur emploi, edited by Rika Gyselen, 41–60. Bures-sur-Yvette: Groupe pour l’Étude de la Civilisation du Moyen-Orient, 1997. Hoyland, Robert. “Re lections on the Identity of the Arabian Conquerors of the Seventh-Century Middle East.” al-‘Uṣūr al-wusṭā 25 (2017): 113–40. ——. Seeing Islam as Others Saw It: A Survey and Evaluation of Christian, Jewish and Zoroastrian Writings on Early Islam. Princeton: Darwin Press, 1997. Huff, Dietrich. “Technological Observations on Clay Bullae from Takht-i Suleiman.” Mesopotamia 22 (1987): 367–90. Ibrahim, Taw iq. “Additions to the Lead Seals of the Umayyad Conquest of the Iberian Peninsula.” In Monnaies de haut moyen âge: histoire et archéologie (péninsule Ibérique – Maghreb VIIe–IXe siècle), edited by Philippe Sénac and Sébastien Gasc, 115–32. Toulouse: Presses universitaires du Midi, 2015. ——. “Notas sobre precintos y ponderales I: varios precintos de sulh a nombre de ‘Abd Allah ibn Malik: correciones y una posible atribución. II: Adiciones a Ponderales andalusíes.” Al-Qantara 27 (2006): 329–35. ——. “Nuevos documentos sobra la Conquista Omeya de Hispania: los precintos de plomo.” In 711: Arqueología e historia entre dos mundos, edited by Louis A. Garcia Moreno, vol. 1, 145–61. Madrid: Comunidad Autonoma Madrid, 2015. ——. “Un precinto a nombre de ‘Abd al-Rahman I.” Al-Qantara 16 (1995): 143–46. Kennet, Derek. “Transformations in late Sasanian and Early Islamic Eastern Arabia: The Evidence from Kush.” In L’Arabie à la veille de l’Islam: bilan clinique, edited by Jérémie Schiettecatte and Christian Robin, 135–61. Paris: De Boccard, 2009. Khamis, Elias. “A Bronze Weight of Sa‘id b. ‘Abd al-Malik from Bet Shean/Baysân.” Journal of the Royal Asiatic Society 12 (2002): 143–54. Khan, Geoffrey. Arabic Documents from Early Khurasan. London: Nour Foundation, 2007. ——. “The Historical Development of the Structure of Medieval Arabic Petitions.” Bulletin of the School of Oriental and African Studies 53 (1990): 8–30. ——. “The Pre-Islamic Background of Muslim Legal Formularies.” ARAM 6 (1994): 193–224. 145 EXPRESSING NEW RULE Knappett, Carl. “Imprints as Punctuations of Material Itineraries.” In Mobility, Meaning and Transformations of Things, edited by Hans Peter Hahn and Hadas Weiss, 37–49. Oxford: Oxbow Books, 2013. Kraemer, Casper J. Excavations at Nessana, vol. 3, Non-Literary Papyri. Princeton: Princeton University Press, 1958. Lerner, Judith, and P. Oktor Skjaervø. “Some Uses of Clay Bullae in Sasanian Iran: Bullae in the Rosen and Museum of Fine Arts Collections.” In Sceaux d’Orient et leur emploi, edited by Rika Gyselen, 67–78. Bures-sur-Yvette: Groupe pour l’Étude de la Civilisation du Moyen-Orient, 1997. Macdonald, Michael C. A. “Ancient Arabia and the Written Word.” In The Development of Arabic as a Written Language, edited by Michael C. A. Macdonald, 5–28. Oxford: Archaeopress, 2010. Macuch, Maria. “The Use of Seals in Sasanian Jurisprudence.” In Sceaux d’Orient et leur emploi, edited by Rika Gyselen, 79–88. Bures-sur-Yvette: Groupe pour l’Étude de la Civilisation du Moyen-Orient, 1997. Marichal, Rémy, and Philippe Sénac. “Sceaux arabes, Ruscino et la conquête musulmane.” L’Archéologue 112 (2011): 49–51. Olszowy-Schlanger, Judith. “Early Babylonian ‘Documentary’ Script: Diplomatic and Palaeographic Study of Two Geonic Letters from the British Library Cairo Genizah Collection.” In Manuscrits hébreux et arabes: mélanges en honneur de Colette Sirat, edited by Nicolas De Lange and Judith Olszowy-Schlanger, 177– 95. Turnhout: Brepols, 2014. Platt, Verity. “Making an Impression: Replication and the Ontology of the GraecoRoman Seal Stone.” Art History 29 (2006): 235–36. Porter, Venetia. Arabic and Persian Seals and Amulets in the British Museum. London: The British Museum, 2011. Potts, Daniel T. The Arabian Gulf in Antiquity, vol. 2, From Alexander the Great to the Coming of Islam. Oxford: Clarendon Press, 1990. Regulski, Ilona, Kim Duistermaat, and Peter Verkinderen, eds. Seals and Sealing Practices in the Near East: Developments in Administration and Magic from Prehistory to the Islamic Period. Leuven: Peeters, 2012. Reinfandt, Lucian. “Administrative Papyri from the Abbasid Court in Samarra (AD 836–892): A First Report.” In Actes du 26e Congrès international de papyrologie, edited by Paul Schubert, 639–45. Geneva: Droz, 2012. Ritter, Nils. Die altorientalischen Traditionen der Sasanidischen Glyptik: Form – Gebrauch – Ikonographie. Vienna: LIT, 2010. ——. “On the Development of Sasanian Seals and Sealing Practice: A Mesopotamian Approach.” In Seals and Sealing in the Near East, ed. Regulski et al., 99–114. Robinson, Chase. ‘Abd al-Malik. Oxford: One World, 2005. 145 146 146 PETRA M. SIJPESTEIJN ——. “Neck-Sealing in Early Islam.” Journal of the Economic and Social History of the Orient 48 (2005): 401–41. Schindel, Nikolaus. “Eine umayyadische Bleibulle des Kaliphen Yazid.” Mitteilungen der Österreichischen Numismatischen Gesellschaft 473 (2007): 116–22. ——. “Eine umayyadische Bleitessera mit dem Namen al-Hağğāğ.” Mitteilungen der Österreichischen Numismatischen Gesellschaft 474 (2007): 166–71. ——. “Nochmals zu umayyadischen Bullen und Siegeln.” Mitteilungen der Österreichischen Numismatischen Gesellschaft 481 (2008): 31–39. Seibt, Werner. “A Hebrew Lead Seal from the Period of the Sasanian Occupation of Palestine (614–629 AD).” Revue des études juives 140 (1981): 303–17. Shahid, Irfan. Byzantium and the Arabs: Late Antiquity. Brussels: Byzantion, 2005. Sijpesteijn, Petra M. “The Arabic Script and Language in the Earliest Papyri: A Multi-layered Signi icance.” Forthcoming in Jerusalem Studies in Arabic and Islam 46 (2019). ——. “An Early Umayyad Papyrus Invitation for the Hajj.’ Journal of Near Eastern Studies 73 (2014): 179–90. ——. “Seals and Papyri from Early Islamic Egypt.” In Seals and Sealing in the Near East, ed. Regulski et al., 171–82 ——. Shaping a Muslim State: The World of a Mid-Eighth-Century Egyptian Of icial. Oxford: Oxford University Press, 2013. Sims-Williams, Nicholas. Bactrian Documents, 3 vols. London: Nour Foundation, 2000–12. Soucek, Priscilla. “Early Islamic Seals: Their Artistic and Cultural Importance.” In Leaving No Stones Unturned: Essays on the Ancient Near East and Egypt in Honor of Donald P. Hansen, edited by Erica Ehrenberg, 237–60. Winona Lake: Eisenbrauns, 2002. Stein, Peter. Die altsüdarabischen Minuskulinschriften auf Holzstäbchen aus der Bayerischen Staatsbibliotek in München, vol. 1, Die Inschriften der mittel- und spätsabäischen Periode. Wiesbaden: Harrassowitz, 2010. Vikan, Gary, and John Nesbitt. Security in Byzantium: Locking, Sealing, and Weighing. Washington, DC: Dumbarton Oaks Byzantine Collection Publications, 1980. Vandorpe, Katlijn. “Seals in and on the Papyri of Greco-Roman and Byzantine Egypt.” In Archives et sceaux du monde hellénistique, edited by MarieFrançoise Boussac and Antonio Invernizzi, 231– 91. Athens: École française d’Athènes, 1996. Walker, Alicia. “Islamicizing Motifs in Byzantine Lead Seals: Exoticizing Style and the Expression of Identity.” Medieval History Journal 15 (2012): 381–408. Walker, John. A Catalogue of the Arab-Byzantine and Post-Reform Umayyad Coins. London: British Museum Press, 1956. 147 EXPRESSING NEW RULE Wassiliou, Alexandra-Kyriaki. Siegel und Papyri: Das Siegelwesen in Ägypten von römischer bis in füharabische Zeit. Vienna: Österreichische Verlagsgesellschaft Dworak, 1999. Webb, Peter A. Imagining the Arabs: Arab Identity and the Rise of Islam. Edinburgh: Edinburgh University Press, 2016. Weber, Dieter. Berliner Pahlavi-Dokumente. Wiesbaden: Harrassowitz, 2008. 147 148 148 PETRA M. SIJPESTEIJN Petra M. Sijpesteijn (p.m.sijpesteijn@hum.leidenuniv.nl) is professor of Arabic at Leiden University and director of the Leiden University Centre for the Study of Islam and Society. Her research concentrates on recovering the experiences of Muslims and non-Muslims living under Islamic rule, using the vast stores of radically underused documents surviving from the early Islamic world. Abstract This article explores the usage, imagery, and linguistic expressions found on seals produced in the early Muslim empire and reveals how these developed from the seventh century to the ninth. Comparing Islamic and preIslamic samples exposes continuities and changes in sealing practices among Byzantine, Sasanian, and Arabian cultures and shows how these developments can be linked to the underlying ideologies and ambitions of Muslim authorities. In particular, it explains how and why different practices unfolded in Egypt and the Levant, and compares this phenomenon to the dissemination of shared forms throughout the Muslim empire, with particular reference to the rich material from Khurasan in the east and al-Andalus in the west. Keywords Islam, Arabic, administration, taxes, Islamicization, Levant, Egypt, Iberia, Khurasan, al-Andalus