Inter duo Imperia: Palmyra between East and West 3515127747, 9783515127745

Citation preview

Inter duo Imperia Palmyra between East and West Edited by Michael Sommer

ORIENS E T OCCIDENS Studien zu antiken Kulturkontakten und ihrem Nachleben | 31 Franz Steiner Verlag

Oriens et Occidens Studien zu antiken Kulturkontakten und ihrem Nachleben Herausgegeben von Josef Wiesehöfer in Zusammenarbeit mit Pierre Briant, Geoffrey Greatrex, Amélie Kuhrt und Robert Rollinger Band 31

Inter duo Imperia Palmyra between East and West Edited by Michael Sommer Chapters VIII & IX translated by Diana Sommer-Theohari

Franz Steiner Verlag

Umschlagabbildung: Man with camel, tomb relief from Palmyra, limestone, c. AD 160. Ny Carlsberg Glyptotek, Copenhagen IN 2833. © Professor Jørgen Christian Meyer Bibliografische Information der Deutschen Nationalbibliothek: Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über abrufbar. Dieses Werk einschließlich aller seiner Teile ist urheberrechtlich geschützt. Jede Verwertung außerhalb der engen Grenzen des Urheberrechtsgesetzes ist unzulässig und strafbar. © Franz Steiner Verlag, Stuttgart 2020 Druck: Memminger MedienCentrum, Memmingen Gedruckt auf säurefreiem, alterungsbeständigem Papier. Printed in Germany. ISBN 978-3-515-12774-5 (Print) ISBN 978-3-515-12776-9 (E-Book)

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I.

7

Inter duo imperia. The Palmyrenes between East and West (Michael Sommer) . . . . . . . . . . . . . . Palmyra as a trading hub . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Palmyra’s symbolic patchwork . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Palmyra’s institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9 12 14 15

II. Oasis-polis. Tadmur-Palmyra as a Greek city in the Roman world (or not) in five episodes (Ted Kaizer) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pre-Roman Palmyra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Episode 1: the early stages of Palmyra’s integration into the Roman world . . . . . Episode 2: the long heyday of Palmyra’s glorious civilization . . . . . . . . . . . . . . . . Episode 3: Palmyra as a Roman colonia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Episode 4: a King of Kings in charge (but still a Roman colonia) . . . . . . . . . . . . . Episode 4a: the immediate aftermath of Zenobia’s defeat . . . . . . . . . . . . . . . . . . . Episode 5: late-antique Palmyra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

23 23 24 26 29 31 33 35 35 36

III. Gateway Tadmur. The beginnings of Palmyra’s long-distance trade (Michael Sommer) . . . . . . . . . . . . . . . . . . . . ‘The Hellenistic City’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . From Seleucid to Parthian and Roman: shifting patterns of exchange . . . . . . . . Inter duo imperia: Palmyra as a gateway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

37 38 41 44

IV. Palmyra. Marginal agriculture in a marginal landscape? (Jørgen Christian Meyer) . . . . . . . . . . . . . . . . . . .

47

V.

Palmyrene merchants and the Red Sea trade (Matthew A . Cobb) . . . . . . . . . Palmyrene merchants in Egypt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Palmyrenes in southern Arabia and on Socotra . . . . . . . . . . . . . . . . . . . . . . . . . . . Palmyrene participation in the Egyptian-Red Sea trade: explanatory factors . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

65 66 70 75 82

VI. The rise of the merchant princes? Scale, status, and wealth in Palmyrene trade (Eivind Heldaas Seland) . . . . . . . . . . . . . . . . . . . . . . . . . . A caravan city? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Palmyrene trade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

85 85 86

6 · Contents Elite involvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Merchants in Palmyra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The rise of new elites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

87 90 92

VII. On the edge of empire. Elite representation and identity building in Roman Palmyra (Ann-Christine Sander) . . . . . . . . . . . . . . . . . . . . . . . . . . . Palmyrene caravan inscriptions and elite status . . . . . . . . . . . . . . . . . . . . . . . . . . . Sources of power according to the caravan inscriptions . . . . . . . . . . . . . . . . . . . . Imagining Palmyra’s elite: camels, horses and warriors . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

95 96 97 103 107

VIII. Μείζονα πράγματα. Persians, Romans and Palmyra in Zosimus’ historiography (Georg Müller) . . . . . . . . . . . . . . . . . . . . . . . . . . . Zosimus’ historiographic technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zosimus’ Palmyra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Palmyra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Persians . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Roman Empire I — nationalities and loyalties . . . . . . . . . . . . . . . . . . . . . . . . The Roman Empire II — emperor and charisma . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

109 109 111 118 123 124 125 126

IX. The political powers and Syria. War aims, scenarios and perspectives for a country in civil war (Alexander Will) . . . . . . . . . . . . . . . . The stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Russia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Israel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . United States and European Union . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

129 130 132 134 137 139 139

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

141

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

143

List of illustrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

161

Index locorum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

163

General index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

165

Preface While this manuscript goes into print, the world is in lockdown due to the Corona pandemic . The virus has caught us all off guard . Abruptly and unexpectedly, the benefits of a globalised world, taken for granted by billions nowadays, are no longer effective: largescale mobility of people, intercontinental supply chains, division of labour on a global scale . The crisis has also shattered the perception of being invulnerable so common in modern, high-tech societies . The lesson may well be that we have not left our pre-modern past that far behind . The spirit of invulnerability also prevailed in the ancient oasis city of Palmyra . By the first century AD, the Syrian city had become the major trading hub between the East and the Mediterranean West . It also had become fabulously rich . Until recently, the ruins of Palmyra’s temples, public buildings and tombs have been the silent witnesses of the city’s wealth and power . What made Palmyra’s business model so unique? Why could the city successfully outcompete other players in the intercontinental long-distance trade? The chapters collected in this volume revisit a number of areas fiercely debated in recent scholarship on Palmyra: the city’s early history, the organisation of its trade, patterns of its elite and social stratigraphy, and Palmyra’s short stint on the stage of world history in the 260s and 270s . The concluding chapter looks into the the area’s dire present and near future, and hence into developments which will determine how and when fieldwork in Palmyra may become feasible again . The papers in this volume originated from the conference “Palmyra . Orient . Okzident”, held at the St Lamberti Church in Oldenburg . I am most grateful to Pastor Dr Ralph Hennings, our most gracious host, and to Professor Werner Brinker for a substantial contribution towards the conference’s cost . I also owe a great debt of gratitude to the series’ editor-in-chief, Josef Wiesehöfer, and to the anonymous reviewers . Thanks are further due to my wife, Diana Sommer-Theohari, for the translation of chapters 8 and 9 and for the copy-editing of the other chapters . Alfred Klemm took care of the layout, for which I remain also very grateful . Oldenburg, April 2020

Michael Sommer

I. Inter duo imperia. The Palmyrenes between East and West Michael Sommer Palmyra, urbs nobilis situ, divitiis soli et aquis amoenis, vasto undique ambitu harenis in­ cludit agros, wrote Pliny the Elder, and: velut terris exempta a rerum natura, privata sorte inter duo imperia summa Romanorum Parthorumque est, prima in discordia semper ut­ rimque cura. According to Pliny, Palmyra was privileged because of its location (nobilis situ), and the availability of arable land (divitiis soli) and water (aquis amoenis); it was remote from the rest of the world (terris exempta) by means of the “nature of things” (rerum natura), and hence protected from the vicissitudes of conflict between the great powers . It was subject to its own fate (privata sors) and “between the two empires” (inter duo impe­ ria) of the period: the Roman and the Parthian one .1 Pliny’s short sketch of Palmyra’s geographical and geo-political situation is sufficiently clear . The polyhistor draws a vivid picture of Palmyra as a wealthy city in a remote location; he also suggests that it was in a position of, at least approximate, political equidistance from the neighbouring imperial powers . This leaves the question open as to which chronological horizon is being referred to by the passage . In the source section of his book 1, Pliny mentions “[Cn .] Domitius Corbulo” and “[C .] Licinius Mucianus”,2 who are both cited as authorities for the upper Euphrates in book 5 .3 Corbulo and Mucianus served as governors of the province of Syria from 60 to 63 and 67 to 69, respectively . Corbulo probably died in 66 or 67, while Pliny collected the sources for his NH in the late 60s and early 70s . As far as Palmyra was concerned, Pliny was certainly up to date .4 Yet the information he provides is usually discarded by modern scholars, as anachronistic at best or downright wrong .5 Most researchers are convinced that by the time when Germanicus visited the oasis city, or at least by the Flavian period, Palmyra had become “officially part of ”6 or was “firmly within”7 the Roman Empire and its province of Syria . This view is supported by six pieces of epigraphic evidence, namely:

1 2 3 4 5 6 7

Plin . HN 5,88 . Ibid ., 1,III auct . Cfr . Sallmann 1971, 44–47 . Ibid ., 5,83 . Thus also Edwell 2008, 44: “Pliny’s possible sources on Palmyra and its relationship to the Roman Empire could not have been more contemporary .” Bowersock 1973, 135–36, Dirven 1999, 20, Hartmann 2001, 49, Jones 1971, 267, Sartre – Sartre 2016, 37, Smith II 2013, 24, Starcky – Gawlikowski 1985, 37 . Smith II 2013, 5343; similarly Kaizer 2002, 37: “part of the empire” . Millar 1993, 34 .

10 · Michael Sommer 1 . The oldest of the so-called caravan inscriptions, from the Temple of Bēl at Palmyra and dating from AD 18/19, mentioning a Palmyrene named Alexandros, whom Germanicus sent on diplomatic missions to Spasinou Charax in Mesene and to Emesa .8 2 . An undated votive inscription for the emperor Tiberius, Germanicus and Drusus, erected by Minucius Rufus, legatus of the legio X Fretensis, whose statues had been put up in the cella of the Temple of Bēl .9 3 . A military tomb stone, discovered in 2006 and published in 2010, which appears to indicate that Roman troops were garrisoned in Palmyra on a permanent basis as early as Tiberius’ time, in September AD 27 .10 4 . A boundary stone found ca . 75 km northwest of Palmyra at Khirbet el-Bilaas marking the limits of a regio Palmyrena . The stone is dated to the year 153, but refers to a boundary that had been established under the governor Q . Caecilius Metellus Creticus Silanus (AD 12–17) .11 5 . The references to Germanicus, Corbulo and Mucianus in Palmyra’s so-called tax tariff of AD 137 .12 6 . A milestone of AD 75 found at Erech, 27 km northeast of Palmyra, believed to belong to a Roman military road connecting Palmyra and Sura on the Euphrates . The stone is generally believed to be indicative of Palmyra being connected to the network of Roman military roads .13 Taken together, this evidence is thought to be conclusive inasmuch as it confirms a strong and permanent Roman presence in the oasis by the mid, if not early, 1st century AD . But is it really strong enough to disprove Pliny’s statement about Palmyra? Was the polyhistor wrong in assuming that the city was located inter duo imperia? Some of the arguments against Pliny can be dismissed without further ado . First, while the tax tariff of AD 137 mentions earlier laws issued by Germanicus and the governors Corbulo and Mucianus, it does not say explicitly that such laws applied to Palmyra . Why should the council of Hadrianē Palmyra, which in the meantime had been absorbed into 8 9 10 11 12 13

PAT 2754: [--- d]y mtq rt ʾlksndws | [---] tdmryʾ dy hw ʿbd | [--- h] lqdmyn wšdrh grmnqs / [--- m]lkʾ myšny [ʾ w]lwt ʾrbz | [---] hʾ mn sṭ[---] lyswdy | [--- šm]šgrm mlk [ḥmṣ ml]kʾ ršyʾ | [---] wlwt [---] . Seyrig 1932, 274–75, no . 1: [Dr]uso Caesari Ti. Caesari divi Aug. f. Augusto divi Iuli nepoti Ge[rmanico Caesari] | imperatoribus posuit | [Min]ucius T. f. Hor. Rufus, legatus X Fretensis . Gawlikowski 2010b: Μαβογαιον | Δημητρίοθ σ | τρατιώτη[ν] σπεί | ρης Δαμασκηνῶ | ν Αναμος καὶ Θαιμ | ος οἱ \επίροποι α | ὐτοῦ ἀρετῆ | ς ἕνεκαν ηλτ ´| Γορπίου . See Kaizer in this volume . Schlumberger 1939, 61–63, no . II: pp fines regionis Palmyrae | constitutos a Cretico Silano | leg Aug pr pr ex sententia Di | vi Hadriani patris sui restitu(i)t . PAT 0259,73, 103 and 121 . Seyrig 1932, 276, no . 2: [Imp Vespasia] | [nus Caesar Au]g | [pontif max] | [tribun pot]est VI | [Imperat--] cos VI | [de]sig VII | [ett] Caesar Aug | [Ve]spasian [p]on | [tr p VI Imp­­ co]s IIII | [sub] | [M Ul]pio [Tr]aiano | leg Aug pro | pr | XVI . For Jones 1971, 267, this is the central argument against Pliny .

I . The Palmyrenes between East and West · 11 the Roman sphere of power much further than previously, not make reference to any laws that had been issued for Syria earlier on? It is rather likely that such laws were put in effect in Palmyra by means of the decree of 137 . Similarly, it is far from certain whether or not the original boundary created by Creticus Silanus early in the 1st century AD matched the one marked by the boundary stone set up in AD 153 . There is no conclusive evidence that a regio Palmyrena had been in existence back in the days of Creticus Silanus . Even if this had been the case, the inscription is no proof that direct Roman authority extended into the regio Palmyrena, especially since Emesa, whose territory would have been on the other side of the boundary, was an autonomous client kingdom throughout most of the 1st century AD and thus not subject to direct Roman rule . No more compelling is, thirdly, the argument that makes the Palmyrene man called Alexandros, who had reportedly been “sent” (wšdrh) by Germanicus to Spasinou Charax and Emesa, a key witness for Palmyra being “part of ” the Roman Empire . The mission in itself does not imply any relationship of direct authority between Rome and Palmyra . On the contrary, the choice of a Palmyrene envoy instead of a Roman one could indicate that a native of the oasis was more qualified for the job, due to linguistic and cultural skills, but possibly also because he was not perceived, within the Parthian Empire, as a representative of Rome . The strongest case against Pliny — one that cannot be disputed away so easily — is provided by the milestone from Erech and the recently unearthed tomb inscription . The milestone was probably found prope situm, if not in situ . Its dating into the sixth year of Vespasian’s reign and hence the year 75 is all but certain . Despite its fragmentary state of preservation, the inscription seems to indicate the presence of some Roman military infrastructure in the region around Palmyra . Much earlier, in the time of Tiberius, Roman soldiers appear to have been deployed in Palmyra, as the tomb inscription confirms . However, this does not automatically prove direct Roman control of the Syrian Desert . Roman military infrastructure was in place in many parts of the Germania Magna well beyond the defeat of Varus in AD 9 .14 Similarly, the Roman military was apparently committed to securing the desert to the north of Palmyra in some way or other by the Flavian period . Pliny’s statement that Palmyra was located inter duo imperia is to be taken with a pinch of salt and not literally: therefore, privata sorte should not be confused with “equidistant from the two empires” or “no Roman presence at all” . The main problem with using the inscriptions under scrutiny here as evidence of Palmyra being “part of ” the Roman Empire, is an epistemological one . It lies precisely 14 Corbulo, for instance, had a fort built in the territory of the Frisii in AD 47 (Tac . ann. 11,19: ac ne iussa exuerent praesidium immunivit) . According to Cass . Dio 72,20, Roman garrisons were deployed deep in the territories of the Marcomanni and Quadi . Elton 1996, 36, remarks that little difference was made between provincial soil, client kingdoms and the territories of tribes and kingdoms with which the Romans merely had struck peace treaties .

12 · Michael Sommer in the assumption that any city could become “part of ” or even “officially part of ” the empire at a given point in time, by means of a formal act of annexation . Such a binary proposition implicitly uses the modern nation-state as a heuristic model . In a nation-state, the intensity of governance is equally spread across the state’s surface, which is limited by ideal stationary borders, there where the state’s authority abruptly ends . Any geographical coordinate is therefore either within or outside the state’s borders . In cases when the borders shift, the point may “become part” of the state . The opposite applies to empires: their intensity of governance is subject to changes and variations across space and time . Their power does not end at once . It rather fades out little by little, across wide frontier zones with no neatly defined political and/or cultural affiliation . A city at the edge of an empire does not suddenly become “part of ” that empire: it gradually grows into the imperial sphere of power, sometimes at a speed which makes the process hardly noticeable for the contemporary observer . Such were the conditions in the Roman Near East . A good way of describing Palmyra’s status is probably that the city “had been put on a sort of ‘waiting list’”15 — in other words: that Palmyra, over the 1st and possibly early 2nd centuries AD, was in a very gradual process of being assimilated into Rome’s architecture of power in the East .16 Such a model reconciles Pliny’s statement fairly well with the epigraphic evidence . *

*

*

The story of Palmyra as a centre in its own right, which was located in a twilight between the two empires, yet gradually moving towards Rome, can best be told in three different histories: first, the history of Palmyra as a trading hub between the East and the West; second, the history of Palmyra’s symbolic world with its cults, deities and forms of artistic and architectural expression; and third, the history of Palmyra’s institutions, of the set of patters and practices that shaped society in the oasis . None of these histories can be told in detail in the present chapter . But even brief sketches suffice to explain why and in which respects Palmyra differed from any other city in the Roman world: why it was a typological monolith whose history truly unfolded privata sorte .

Palmyra as a trading hub Owing to the Syro-Austro-German excavations in the so-called Hellenistic City to the south of the imperial period’s city centre, we now have available substantial data on the city’s early history in the 2nd and 1st century BC . The record of imported pottery, patchy as it is, seems to suggest that Palmyra’s trade with Mesopotamia increased in volume to15 Yon 2010, 239 . 16 Similarly Kaizer in this volume .

I . The Palmyrenes between East and West · 13 wards the middle of the 2nd century BC and again in the second half of the first century BC . This is surprising, considering that the trade took an upturn each time the region was affected by warfare . Normally, war and unrest are the natural enemies of commerce, but in Palmyra’s case things were different . The city obviously benefitted from political fissures tearing the region apart . Why the Palmyrenes were such wartime profiteers can be understood from their activities as highlighted by the caravan inscriptions of the imperial period . Once a year, a caravan left the oasis for Spasinou Charax in Mesopotamia, the Persian-Gulf port in which luxury items from India landed . The logistics of the inter-imperial trade was not trivial, though the Syrian Desert and the Euphrates valley provided for relatively easy travel . But nomadic tribes and Parthian authorities had to be dealt with . Navigating through this political jungle required skills, conections and, in particular, trust . Unlike people who came from the Roman Empire, which, since Crassus’ abortive Parthian campaign in 53 BC, was in a latent state of war with its eastern neighbour, the Palmyrene traders enjoyed the trust of the Parthian officials and the merchant communities in the Mesopotamian cities, where they traded and where some of them established themselves permanently . Mutual trust as a key resource repeatedly emerges from the epigraphic record: the Palmyrene Alexandros, whom Germanicus had sent to Spasinou Charax as an envoy, has already been mentioned . The Palmyrene Yarḥai served as governor of the island of Thilouana (Bahrain) for the king of Charakene, who himself was a vassal of the Parthian king;17 another Palmyrene, Yarḥibōl, was sent to the kingdom of Elymais as an envoy .18 Several Palmyrenes sponsored the building of sanctuaries in Parthian cities: a certain Aqqayh paid for an altar and a temple in Vologesias;19 and one Šo‘adu, who was a very prominent figure in the mid-2nd century AD, even established a temple for the imperial cult in the same city .20 Yarḥai, Yarḥibōl and Šo‘adu were all representatives of a lofty elite, far above the rank and file of the army of merchants treading the caravan roads year after year . On Parthian soil, such high-ranking Palmyrenes provided various services for the caravan traders and the Palmyrenes of the Mestopotamian trading diaspora, such as declaring goods and paying duties .21 What transpires from this evidence is the fact that at least the Parthians did not lump the Palmyrenes with their Roman adversaries . The caravan trade seems to have reached its peak in the 130s, 140s and 150s, and thus in a period when the relationship between the two powers was somewhat less hostile . Yet trust was the resource that guaranteed the Palmyrene merchants the leeway and legal security they required for their trade, and which 17 PAT 1374 (AD 131): [Ἀδ]ριανὸν Παλμυρηνόν σατρά | [π]ην Θιλουανῶν Μεερεδάτου | βασιλέως Σπασίνου Χάρακος . 18 PAT 1414 (AD 138): […] καὶ π[ρεσβεύσαντα] αὐθαιρέτως | [πρὸς Ὀρώδην τὸν β]ασιλέα τῆς Αἰλ[υ] | [μήνης ---] . 19 PAT 0263 (AD 108): […] bʾlgšyʾ ḥmnʾ | klh hw wʾtrh wʾp ṭll ʾdrwnʾ . 20 PAT 1062 (AD 145/46): [ἐ]ν Ὀλογα[σίᾳ ναὸν τῶν Σε]βαστῶν κ[αὶ] κ[α]θι | [ερώ]σαν [τα--] . 21 E .g . PAT 2763 (paying duty), PAT 0263 (being helpful) and PAT 1584 (help with the construction of a temple) .

14 · Michael Sommer others would have lacked . And trust went incredibly far, considering that Yarḥai, who came from Hadrianē Palmyra, held office as the virtual Parthian governor of Thilouana . People like Yarḥai were well-networked inter-imperial commuters whose scope of action dwarfed anything we know from local elites in the Roman Empire . If nothing else, this at least should explain why Palmyra was different .

Palmyra’s symbolic patchwork The cosmopolitanism of Palmyra’s elite is mirrored by the iridescent equivocality of the symbolic universe inhabited by Palmyra’s motley population . Whether or not the manifest incongruities in Palmyra’s divine world are due to a supposedly ethnically, culturally and linguistically diverse population, is subject to an ongoing debate .22 What is more tangible is Palmyra’s material culture which, at first glance, seems to be an amalgamation of ‘western’ and ‘non-western’ elements . The city’s largest sanctuary, the Temple of Bēl, dedicated in AD 32, was in every detail inspired by a sanctuary in Asia Minor that was 250 years older: Hermogenes’ Artemision of Magnesia on the Maeander . The Palmyrene temple, however, featured eye-catching oddities: instead of the pitched roof of the ‘ordinary’ classical temple, the Temple of Bēl had a terrace on its top, which was surrounded by a row of pinnacles and accessible via spiral staircases from the inside . The entrance to the cella was on the western long side . And on both the northern and the southern end, lay thalamoi, which were elevated by several metres against the level of the central part . We know precious little about the rituals performed here, but it is evident that they differed considerably from ‘western’ religious practices . The sanctuary reveals the methodology applied by the Palmyrene builders: they borrowed the architectonic and artistic vocabulary from the Greeks and the Romans, but they adapted its semantics for their own purposes . The result is a temple that looks strikingly classical and non-classical at the same time . The Palmyrenes had hardly any choice . Their nascent urban society disposed of an inventory of religious and social practices, of deities and possibly stories and myths . But they lacked any sort of tools which could serve to monumentalise this inventory . The only way forward was to borrow the tools from the undisputed masters of monumentalisation . Accordingly, the Palmyrenes adopted the Tyche of Antioch as the divine emblem of their city . They constructed a colonnaded street, which represented the epitome of urbanism in the East, but which was conspicuously absent in any ‘Western’ city . Next to it, they built a theatre, the symbol of Greekness in Roman Syria, but in the Roman style . What dramas were performed on stage and in which language, we do not know . They carved stone portraits of their dead in the Hellenistic style, but what mattered to them was social rank and status rather than individuality . Finally, in the 2nd century AD, they

22 For an in-depth discussion Kaizer 2002, 35–66 and 261–62, Smith II 2013, 58–68 .

I . The Palmyrenes between East and West · 15 decided to abandon the customary tower tombs in favour of structures that looked like Graeco-Roman temples . In all such cases, they borrowed ‘Western’ types, techniques and symbols, but put them to use in their very own ways . In Palmyra, Hellenisation and Romanisation came with a ‘twist’ . The contact with the Roman world had a massive impact on local society and brought about numerous radical changes . The Greco-Roman look-and-feel of many monuments and works of art was far more than a façade, a veneer .23 It was a means of expression cautiously and consciously adopted and adapted by the Palmyrenes, whose oasis society was a newcomer in an environment of mostly long-established urban centres . The urge to be pari passu with them and possibly outcompete such lighthouses of the Hellenistic oecumene as Antioch and Apamaea was a powerful driving force in early imperial Palmyra .24 Yet the canon of Hellenism into which the Palmyrenes integrated was modular and flexible: it was like a big warehouse they could explore and from which they could chose what served their needs and leave behind what did not . What emerged from this process is the idiosyncratic symbolic patchwork that was Palmyra, not quite inter duo imperia, but certainly privata sorte .

Palmyra’s institutions Palmyra’s third history needs to be viewed backwards, beginning with its demise as a trading hub and power centre, rather than as a city in itself . When the emperor Aurelian captured the oasis city in AD 272, a dozen years lay behind the Near East that saw Palmyra’s rapid rise to a veritable power factor between the empires of the West, Rome, and of the East, Persia, and its likewise sudden downfall . Following Valerian’s defeat at Edessa in the summer of AD 260, the eastern Roman provinces turned into a large-scale power vacuum with only dispersed Roman troops being present and alive with intensified Persian raiding activity .25 In this desperate situation, one man, Odaenathus of Palmyra, assembled what forces were available and, near the Euphrates, won a decisive victory over a retreating Persian army . A few years later we find the same Odaenathus besieging the Persian capital of Ctesiphon twice before he suddenly died under mysterious circumstances, probably in late 267 . By that date, Odaenathus had become the undisputed political and military leader in the Near East, awarded with an array of pompous Roman titles . His death caused a major succession crisis tightened by two different interpretations of Odaenathus’s role in the Near East . From a Roman point of view, Odaenathus was just one 23 This is the point of view of Ball 2000, 396: “Roman frills might be piled onto it, but the real architecture remained what it had always been: Near Eastern .” 24 This is the convincing case made by Andrade 2013, 148–214 . 25 On the following Burgersdijk 2007, Gawlikowski 2010a, Hartmann 2001, Hartmann 2016, Sommer 2008, Sommer 2018a, 139–170 .

16 · Michael Sommer Roman official whose prerogatives fell back to the emperor at the moment of his death; from the Palmyrene perspective, Odaenathus’s heirs — his minor son Waballat and his widow Zenobia – were not only entitled, but downright obliged to succeed him as ruler of Palmyra . In 270, this inextricable conflict prompted a civil war, which was ignited and lost in the end by the Palmyrenes . What matters here, is the combination of military capability and dynastic tradition which becomes problematic if we follow the hypothesis that Palmyra had been just like any other Greek city in the Roman Empire . In the crisis of AD 260, no ordinary Greek city would have had the resources and the manpower to fight back an entire Persian army, let alone pursue the enemy deep into his own territory and put their capital under siege . Some resort to an ominous Palmyrene “militia”, which, however, is not attested by any single source .26 Palmyra’s uniqueness against all other cities in the Roman world, however, is the consequence of its origin from, and proximity to, nomadic tribes . The people of the desert viewed themselves, and were viewed by the Palmyrenes, as kinsmen of the urban dwellers . It is hardly far-fetched to assume that Palmyra was in no short supply of military manpower because of the permanent availability of tribal warriors in its immediate neighbourhood . The tribal, as it were polymorphic, patterns of Palmyrene society also explain the vitality of the dynastic principle in the event of Odaenathus’ death . In this critical moment, the ruler’s next of kin had no choice but to assume power for themselves . As far as we can see, Odaenathus’ monarchic power as expressed in his Palmyrene title of rʾš (“head”) was a completely new feature in Palmyra’s political landscape . One scholar has hypothesised the  “re-orientalisation” of Palmyra in the crisis prompted by Romano-Persian conflict since the 230s,27 but it seems more likely that Palmyra had never fully embraced the constitutional order of a Greek polis . To be sure, the nomenclature of a full-fledged polis is not absent from Palmyra’s epigraphic record . It appears in Greek and in Palmyrene transliteration: the dēmos as well as a boulē are mentioned,28 and among other magistrates, a stratēgos held office .29 Finally, Palmyra is explicitly referred to as a polis .30 Yet, the honorands of most of the honorific inscriptions do not seem to derive their lofty social status from any offices or magistracies . Rather than by such roles, the men are identified by their ancestors, of whom normally two generations, in some cases as many as four, are mentioned . The only function frequently brought up by the inscriptions is that of caravan leader, synodiarchos . Services to Palmyra’s trade seem to outweigh the usual civic magistracies by several degrees . Only once, in the inscription for Aelius Borā 26 27 28 29 30

The extensive scholarship on the “militia” has been collected by Hartmann 2001, 54 . Schmidt-Colinet 1997, 163–64, Schmidt-Colinet et al . 1992, 40–41 . Similarly Hartmann 2001, 95 . PAT 0197 (AD 132), 1414 (AD 138), 1062 (AD 145/46), 1063 (AD 198) and 1378 (AD 199) . PAT 1063 . PAT 0269 . The Greek nomenclature has been used by Sartre 1996 for making his case that Palmyra was a “Greek city” (cité grecque) .

I . The Palmyrenes between East and West · 17 (AD 198), the honorand is identified as stratēgos, and only in the Greek text . Here, to the noun στρατηγός, the genitive object τῆς εἰρήνης is added . This could be owing to an emergency situation before the background of Septimius Severus’ Parthian War .31 Less than half of the inscriptions are marked as decrees of the dēmos and the boulē of Palmyra . Overall, the epigraphic record seems to reflect the outstanding importance of kinship as opposed to the ‘civic’ institutions of a Greek polis . *

*

*

To conclude, the Palmyra we know would have been unthinkable without the presence of the Roman Empire . Palmyrenes were sufficiently open-minded to perceive the opportunities that came with Rome: opportunities that improved the standard of living of the vast majority of the oasis dwellers and hugely enhanced options for upward social mobility and representation of social status . But the Palmyrenes’ open-mindedness towards what Rome had on offer did not lead them to importing Roman culture and its symbolic baggage wholesale into their oasis . The result was a hybrid civilisation in its own right, where ‘local’ features cannot be distinguished from ‘imported’ ones . The importance of Palmyra by far transcends the oasis and even the Roman Near East . Ultimately, any study of Palmyra is a case study in Romanisation, in Hellenisation and, more generally, in acculturation, paradigms that have been popularised long ago by Francis Haverfield’s ground-breaking study in the Romanisation of Roman Britain .32 Such analytical prisms are now under heavy pressure in Anglo-Saxon academia, while they continue to flourish on the continent . Across the Atlantic and the British Channel, the anti-Haverfieldian zeal is nourished by the currently prevailing orthodoxy of post-colonial theory .33 The concepts are suspected to betray the colonial mindsets of their creators, from Mommsen through to the present . New, more fashionable analytical tools, such as ‘globalisation’, have been applied, more or less successfully .34 However, concepts such as ‘Romanisation’ and ‘globalisation’ are mere labels, ideal types in the Weberian sense, that do not exist in any given reality, but only in the minds of the researchers . There is no such thing as Romanisation or Hellenisation in the real world, just rather chaotic events, processes and structures that need be arranged in proper order by scholars in order to make sense . What ultimately matters, is not how we call our concepts, but how we define them . Therefore, any academic occupation with society is also the constant defining and refining of ideal types . Studying the Palmyrenes and their interaction with Rome helps towards a fine-tuned understanding of Hellenisation and Romanisation, an understanding that takes into account the agency of those who are 31 32 33 34

Ibid . Haverfield 1923 . For a critique see Sommer 2011 . See Hingley 2005 and the contributions to Pitts – Versluys 2015 .

18 · Michael Sommer Hellenising or Romanising themselves, causing often perplexing and sometimes contradictory outcomes . Such an understanding is not entirely novel: it was more present in Francis Haverfield’s work than his modern critics are inclined to accept . *

*

*

Regrettably, there is a second reason why, in recent years, Palmyra has been higher on the agenda of classical scholars than it used to be . In May 2015 — the Syrian civil war went into its fourth year — the fighters of Daesh, the so-called Islamic State, were about to bring havoc to Palmyra, the UNESCO world heritage site, a much-visited tourist destination before the war and the location of a high-security prison where the Baʿth party regime used to lock away political opponents . To conquer Palmyra made sense to the Islamists for a number of reasons . One of them was the importance the ancient city had for the Baʿthist charter myth of the modern nation-state of Syria: in a manner of speaking, Zenobia and her clan with their alleged revolt against Rome were forerunners to the Assad family . They were regularly brought back to life in the regime’s struggle for creating a Syrian national identity that went beyond Islam . But the destructions brought upon Palmyra by the Islamist militia were also motivated by the repercussions such acts of barbarism would have on western media coverage of the war . Since at least the 18th century, Palmyra had been a celebrity among the ancient sites of the Near East, visited, since the time of Lady Hester Stanhope, by people who had been celebrities themselves . Capturing and mutilating the city for a second time since antiquity would be a display of the militia’s power . It would hurt the west where it was most vulnerable: in its faith in the world as a place where, essentially, common sense prevailed . The images of the destruction of the Tripylon, the Temple of Bēl and other landmark buildings sparked disbelief, anger and sorrow across the world . Academics, some of whom had been researching Palmyra for a lifetime, wrote newspaper articles and gave interviews in which they drew attention to the immense cultural loss, but also to the suffering of Syria’s civilian population .35 Predictably, the media hype about Palmyra was short-lived . Attention was soon drawn to the refugee crisis and away from Syria’s cultural heritage . It is not the intention of the editor of the present book to put up yet another tombstone for Palmyra .36 Rather, the volume wants to flag recent research in the oasis city, research that has been pursued, cumbersome conditions notwithstanding, in numerous fields, re-

35 In the aftermath of Daesh’s conquest of Palmyra, the author of the present introduction has published an article on Der Spiegel online: Michael Sommer on SPON 24 May 2015 (http://www .spiegel .de/kultur/gesellschaft/ is-bedroht-palmyra-nimrud-hatra-aufsatz-michael-sommer-a-1035171 .html (5 February 2019) . He was, by some debaters in the forum section, criticised for caring about “a few old stones” while people were suffering, starving and dying . Setting off violations of human rights against acts of vandalism misconceives the strategic thinking of the terrorists for whom both serve the same objective . 36 See, for istance, Delplace 2017, Sartre – Sartre 2016 and, most notably Veyne 2017 .

I . The Palmyrenes between East and West · 19 search that helps to write the three histories outlined in the beginning more clearly and more vividly . In the first place, this is partly owing to new research data having been made available through fieldwork, which has been conducted in Palmyra and in the Palmyrene right up to the war . This has produced a wealth of new information: most notably the excavations in the so-called Hellenistic City conducted by a Syrian, Austrian and German team,37 the Norwegian Palmyrena project,38 and the Syro-Italian Pal .M .A .I .S . project on dwellings in the northern neighbourhood .39 Secondly, new insight is provided by new historical research attempting at synthesising the data collected by archaeologists .40 Yet this process has only just begun . To show what paths it might take, is the purpose of this volume . Ted Kaizer’s chapter investigates the history of Palmyra’s integration into the Roman Empire from the city’s earliest, pre-Roman times, all the way down to post-Zenobian Palmyra . He shows how Palmyra gradually moved closer to Rome and became absorbed into the Roman sphere of power, with Germanicus’ and Hadrian’s visits to the oasis both as important steps, and the rise to the rank of colonia, Odaenathus’ monarchic rule and the transformation into a late Roman fortress city as further stages in the process . Despite Palmyra’s gradual and long-term absorption into the imperial power of Rome, the oasis city remained, according to Kaizer, a place unlike any other in the Roman world . Michael Sommer looks into the data obtained from the fieldwork in the Hellenistic City and attempts at explaining the seemingly odd coincidence that in the 2nd and 1st centuris Palmyrene trade with Mesopotamia expanded each time the region was ravaged by war . The chapter argues that trust and the ability to mediate between the two powers was seminal for Palmyra’s rise to the prime hub in the trans-Eurasian long-distance trade . The division of western Asia between two latently or openly hostile powers — first the Seleucids and the Parthians, later the Romans and the Parthians — paved the way for the Palmyrenes, whose caravans could permeate the frontier, because they were trusted by all other agents in the arena . How did the Palmyrenes feed their people? In his chapter, Jørgen Christian Meyer explains how the Palmyrenes managed to cultivate land in the Syrian Desert, which was marginal only from a political point of view, but provided ecological conditions that could be exploited, provided societies made investments in the respective infrastructure . Maintaining such an infrastructure depended on wealth pouring into the urban centre in the oasis . Once the centre was in full decline, in the post-Umayyad period, floodwater farming was no longer an option, and soon the steppe was reclaimed by nomadic tribes .

37 Schmidt-Colinet 2003a, Schmidt-Colinet 2003b, Schmidt-Colinet – al-Asʿad 2000, Schmidt-Colinet – alAsʿad 2002, Schmidt-Colinet – al-Asʿad 2013 . 38 Meyer 2017, Meyer – Seland – Anfinset 2016 . 39 Grassi 2010, Grassi 2011, Grassi 2017, Grassi – Zenoni – Rossi 2012 . At the conference, Lilia Palmieri has given a presentation on the preliminary results oft he Pal .M .A .I .S . excavations . Regrettably, her paper is not included in the present proceedings . 40 Most recently: Andrade 2018b, Grassi 2017, Seland 2016, Smith II 2013, Sommer 2018a .

20 · Michael Sommer “The personal circumstances and motivations that influenced individual Palmyrenes” to become engaged in the Red-Sea trade via Egypt are “beyond recovery” . Thus Matthew A. Cobb summarizes his considerations on the Palmyrene diaspora in Egypt and its role in the exchange of commodities between Egypt and India . Potential explanations include the growth of a kinship and ethnicity-based networks of Palmyrenes in the Nile country, higher profit margins in the trade via Egypt and the attempt, on behalf of the Palmyrenes, to create an alternative route while the Syro-Mesopotamian trade was increasingly compromised by unrest and warfare . Given the nature and extent of the evidence, none of these explanations is conclusive, Cobb points out, but the fact that Palmyrene traders, from the later 2nd century AD onwards, increasingly diversified their activities, is an interesting feature in itself and seems to suggest that by then the bonds between Palmyra and the Roman Empire had become much tighter . Two chapters deal with the Palmyrene elites . Eivind Heldaas Seland investigates the role of the upper echelons of Palmyrene society in the city’s long-distance trade . In 20th century scholarship, Seland distinguishes three models of elite-involvement: the “merchant prince” (M . Rostovtzeff), the “tribal leader” (E . Will) and the “Roman aristocrat” (G . K . Young), criticizing their “static” character . As an alternative, he suggests a dynamic model that takes into account the archaeological and epigraphic evidence from Palmyra: in this model, the tribal organisation of Palmyrene society was, from the 2nd century onwards, increasingly challenged by a new class of wealthy merchants who claimed rank and social prestige for themselves . Ann­Christine Sander searches for a Palmyrene aristocrat’s sources of power . In her investigation of the epigraphic and iconic evidence, she highlights the paramount importance of military power that distinguished the Palmyrene elite from other local elites in the Roman world . Not accidentally are Palmyrene aristocrats commonly represented as warriors, accompanied by horses and camels . Likewise, military services dominate the caravan inscriptions as the prevailing motivation for honours being conferred on members of the elite . According to Sander, Palmyra’s military aristocracy is the decisive factor creating a unique social setting in the oasis . An important source for assessing Palmyra’s role in the Roman crisis of the 260s and 270s is Zosimus’ Nea Historia, scrutinized here by Georg Müller . The chapter first explores Zosimus’ historiographical toolbox and his historiograhical perspective, before revisiting his Palmyra episode, its protagonists and background actors, Zosimus’ sources and different ‘foci’ . By the time Zosimus wrote, it was agreed that Palmyra was an alien factor, intrinsically hostile to the Roman cause . Its dodgy role is epitomized by Zenobia’s, a woman’s, rule . To Zosimus, the episode is pivotal for understanding the third century crisis and Rome’s recovery from Aurelian’s reign onwards . All future research in and around Palmyra will invariably depend on how the political situation in Syria develops over the next few years: reason enough why the final chapter of this volume focuses on the present civil war, the political interests of its main players and its possible outcomes . As Alexander Will observes, all information coming from

I . The Palmyrenes between East and West · 21 Syria is currently filtered by interested parties and hence by definition unreliable . Yet, it is foreseeable that the West will have little say in a post-war order . As Will points out, Israel, Iran, Turkey and Russia each pursue their own interests in the Levant . While Israel’s and Turkey’s strategies aim at holding down regional adversaries — the Kurds in the case of Turkey, Iran in Israel’s case — Russia and Iran are pursuing more ambitious targets: for the Mullahs in Tehran, Syria is their prime base of operation against their main enemy in the region, Israel . For Russia’s Putin, the Assad government is a strategic partner helping Moscow to get a foot in the door to the Near East and to project power into the Mediterranean . Little imagination is required to anticipate that access to a post-war Syria will have to be negotiated via Russia . This holds true for any archaeological fieldwork as well . As of 2019, Russian archaeologists are keeping an eye on Palmyra, while the Russian government is investing substantial amounts of money in the country’s archaeological research institutions .41 Once again, a great power’s archaeological and political ambitions are becoming closely intertwined .

41 https://archaeologynewsnetwork .blogspot .com/2018/06/russian-archaeologists-eyeing .html#hUsqrUtgJ6oZ7RYp .97 (5 February 2019) .

II. Oasis-polis. Tadmur-Palmyra as a Greek city in the Roman world (or not) in five episodes1 Ted Kaizer

Introduction This paper will engage with the oft-discussed question of whether or not Palmyra, located at the oasis of Tadmur in the heart of the Syrian steppe, can be bracketed with the Greek cities of the Roman Empire .2 It will do so not by testing the evidence systematically against the framework of the ‘ideal type’ of the Greek city in the Roman world, but by merely providing an impressionistic sketch of a selection of relevant source material spread out over five partly uneven episodes . By establishing five episodes I may give the impression of following the classic reconstruction first proposed by Henri Seyrig and abided by many scholars working on Palmyra,3 but despite some outward appearances there is not meant to be much overlap . As emphasised by Sommer, Seyrig’s five stages of Palmyra’s history — 1) an independent city state between the empires of Rome and the Parthians or, in 1

2

3

This paper originated in an invitation by Georgy Kantor and Neil McLynn to talk about Palmyra in their seminar series on the Greek city in the Roman world, at Oxford in June 2017, where an earlier and longer version was presented . I am grateful to them, and to Michael Sommer for inviting me to the stimulating conference in Oldenburg in March 2018 . I should also like to thank the Leverhulme Trust for the award of a Major Research Fellowship (2014–17), during which the original version was conceived . The debate is framed by the positions taken up by Maurice Sartre and Michael Sommer . Sartre 1996, 385 . 397, argued that Palmyra conformed to the model of a ‘Greek city’ in every way, and that there should be no doubt about the nature and functioning of its civic institutions following Palmyra’s integration in the Roman world . In contrast, for Sommer 2005a, 294, “Palmyra was no Greek city at all, it was a city of the Near Eastern steppe frontier with a blinding, ingeniously ‘borrowed’ Greek façade .” Cf . Sommer 2005b, 170–183 . It should be taken into account, however, that the nature of Palmyra’s society will have developed over time in part because of the veritable effects of that so-called façade on the quality of its constituent parts . In the post-scriptum following the republication Sartre 2014, 319, reacted against his critics . I had previously dealt with the issue of Palmyra as a so-called Greek city in Kaizer 2007, which argued that Palmyrene coins were not systematically connected to those cults that can be said (based on epigraphic evidence) to have played a certain civic function in Palmyra . The indigenous factor is being emphasised in Graf forthcoming . Seyrig 1941, 170–173, Starcky – Gawlikowski 1985, 33–69, Will 1992, esp . 40: “le mieux est de suivre le schéma d’évolution tracé jadis par H . Seyrig .”

24 · Ted K aizer the words of Franz Cumont and later Ernest Will, a merchant republic;4 2) from the early Julio-Claudians onwards a tributary city in the Roman provincial system; 3) a ‘free’ city from Hadrian onwards; 4) a Roman colonia; 5) a vassal and later independent principality — are, “though suggestive, highly speculative” and in some cases hardly recognisable in the source material .5

Pre-Roman Palmyra That there had been some sort of settlement at the site in much more ancient times is now hardly disputed . The name Tadmur is first mentioned in Assyrian cuneiform tablets from Mari,6 and the stratification of the hill below the Temple of Bēl goes back to the third millennium .7 To what degree any such early settlement was uninterrupted, or rather limited to particular periods, can certainly not be answered with the evidence that is currently available, but it must be emphasised that the one constant factor in this context is Palmyra’s quality as the main oasis in the large steppe zone between the urban centres in western Syria and the Euphrates river in the eastern half of the Syrian lands . The point is emphasised by Pliny the Elder in his Natural History, whose Near Eastern section mostly describes the situation under Augustus, and sometimes even earlier .8 In the relevant passage, Palmyra is referred to as “a city famous for the beauty of its site, the riches of its soil, and the delicious quality and abundance of its water; its fields are surrounded by sands on every side, and are thus separated, as it were, by nature from the rest of the world .”9 Whereas Pliny had once been described by Ernest Will as “poorly informed” about the Palmyra of his own time, having described Palmyra as a typical oasis rather than as a specific place, it has now been stated, by Eivind Seland, that “the Roman naturalist was perhaps less off target than some have allowed .”10 Palmyra did indeed have a number of springs and fields, which were surrounded by a large steppe area, and Seland argued that the description only errs with regard to the degree of the oasis’ isolation: “the Syrian Desert was not empty, as Pliny thought, but constituted the economic and sociopolitical hinterland of the Palmyrene polity .” For the Roman period that argument can be made with reference both to the archaeological remains of the villages in the region to the northwest of Palmyra, the Palmyrena,11 and to the fact that the Palmyrene tax law 4 5 6 7 8 9

Thus Cumont 1926, XXXII, Will 1985, 268: “république de marchands” . Thus Sommer 2005a, 287 . Scharrer 2002, esp . 301–318, on the Mari tablets . See al-Maqdissi 2000 for a re-examination of the 1960s exploration of the stratification of the tell . Detlefsen 1909, 81, Murphy 2004, 163 . Plin . HN 5,88: Palmyra urbs nobilis situ, divitiis soli et aquis amoenis, vasto undique ambitu harenis includit agros, ac velut terris exempta a rerum natura . 10 Seland 2016, 25, Will 1985 . 11 Meyer 2017, Schlumberger 1951 .

II . Tadmur-Palmyr a as a Greek city in the Roman world · 25 (inscribed under Hadrian, but containing regulations from the first century AD) points to a surrounding territory that very much contributes to the city’s economic and social life .12 For the earlier period, however, there is simply insufficient evidence about the Palmyrena . It is now at least clear, from the soundings in the area to the south of Diocletian’s wall, that some building work went on at the oasis settlement itself in the third century BC, and that trade was reaching it at the beginning of the Hellenistic period . Its popular scholarly label of ‘Hellenistic city’, however, seems to imply more than has been revealed, and over the course of the next centuries, the area south of the wadi came to form an integral part of Tadmur-Palmyra and continued to do so after the central colonnade had been built as an extension to (rather than a replacement of) the existing centre .13 There is in any case nothing to support the idea that Palmyra originated as a Seleucid foundation — even though the acceleration of societal and cultural development at the oasis must have taken place before its unique civilization suddenly becomes visible in the sources by the middle of the first century BC . There is, in fact, one fragment of an inscription which may record an interaction with a Seleucid king, the first of two possibly Hellenistic inscriptions from Palmyra:14 [----------] | [--- βα]σιλέα [---] | [--- ἐπι]φανῆ κα[ὶ ---] . It was found in foundations of a structure within the temenos of the Temple of Bēl, together with other ‘archaic’ material, and the letter type has been interpreted as going back to the pre-Roman period . The epithet of the king in question, Epiphanes (‘Manifest’), was worn by a number of Seleucids, the last of whom (Antiochus XII) was defeated by the Nabataeans in 84 . Henri Seyrig, who published the inscription, therefore chose to interpret the unknown king as a member of the Arsacid dynasty instead, since Epiphanes was used by all Parthian kings from the late second century onwards .15 However, an interpretation of Tadmur-Palmyra as ultimately under the authority of the Arsacid King of Kings before the Roman period is not necessarily more compelling than to view the settlement as initially forming part of the Seleucid realm . In its formative phase the isolated location of the oasis seems to have hindered any form of integration — be it in the Seleucid or Parthian world — that would have left reminders in our sources . The fragment referring to a king Epiphanes will in any case have been older than the earliest dated inscription from Palmyra . That honour goes to a console from October 44 BC, later reused within the temenos wall of the Temple of Bēl, which records how “in the month of Teshri in the year 269 the priests of Bēl set up this statue for Garimay, son of Nabuzabad, who is of the tribe 12 Matthews 1984 . 13 Schmidt-Colinet – al-Asʿad 2013, with Kaizer 2015 . For a brief discussion of the sources for Hellenistic Palmyra, see Kaizer 2017, 33–39 . 14 IGLS VII .1,1 . 15 Seyrig 1939, 322–23, no .28 .

26 · Ted K aizer of the benē Kahnebo” .16 The centrality of the cult of Bēl and its priesthood is therefore clear from the very moment that Tadmur-Palmyra appears in our dated sources, and so is the tribal aspect of the local society of the oasis . In addition it may be relevant to note that, from the word go, it was the Seleucid calendar that was used in Palmyra, as it was in most other places in the Roman Near East .17

Episode 1: the early stages of Palmyra’s integration into the Roman world To what degree Palmyra at this time can actually be said to have formed part of the Roman Near East has of course been much debated . With these early days in which the imperial system only gradually spread its influence over the Syrian lands, roughly twenty years after Pompey the Great established the first Roman province in the Levant, we have arrived at the first of my five episodes . At first glance it seems fortunate that a major literary source immediately comes to comment on what went on at the oasis around the same time that the priests of Bēl set up the statue for Garimay . Appian, in his Civil Wars, describes an eventually abortive plan by Mark Antony in 41 BC to raid the oasis: When Cleopatra had sailed homewards, Antony sent his horsemen to the polis Palmyra, not far from the Euphrates, to plunder, accusing them of something insignificant, that they — being on the frontier between the Romans and the Parthians — showed tact to both sides (being merchants, they carry Indian and Arabian goods from the Persians and they dispose of them in the territory of the Romans), but in fact he had in his mind to enrich his horsemen . As the Palmyrenes learned about this beforehand and carried their essentials to the other side of the river and to the riverbank, preparing themselves with bows — with which they are by nature excellent — in case anyone would attack them, the horsemen, seizing the city empty, turned around, not having met anyone, not having taken anything .18

Olivier Hekster and I have argued that the passage does not in fact provide that much valuable information about the oasis settlement in 41 BC, since it is clear that contempo16 PAT 1524, Healey 2009, 144, no .28: byrḥ tšry šnt 2|69 ʾqym kmryʾ | dy bl ṣlmʾ dnh lgrymy | br nbwzbd dy mn pḥd | bny khnbw . 17 As emphasised e .g . by Millar 1993, 321 . 18 App . BC 5,9: Ἀποπλευσάσης δὲ τῆς Κλεοπάτρας ἐς τὰ οἰκεῖα, ὁ Ἀντώνιος ἔπεμπε τοὺς ἱππέας Πάλμυρα πόλιν, οὐ μακρὰν οὖσαν ἀπὸ Εὐφράτου, διαπράσαι, μικρὰ μὲν ἐπικαλῶν αὐτοῖς, ὅτι Ῥωμαίων καὶ Παρθυαίων ὄντες ἐφόριοι ἐς ἑκατέρους ἐπιδεξίως εἶχον (ἔμποροι γὰρ ὄντες κομίζουσι μὲν ἐκ Περσῶν τὰ Ἰνδικὰ ἢ Ἀράβια, διατίθενται ό ἐν τῇ Ῥωμαίων), ἔργω ό ἐπινοῶν τοὺς ἱππέας περιουσιάσαι . Παλμυρηνῶν δὲ προμαθόντων καὶ τὰ ὰναγκαῖα ἐς τὸ πέραν τοῦ ποταμοῦ μετενεγκάντων τε καὶ ἐπὶ τῆς ὄχθης, εἴ τις ἐπιχειροίη σκευασαμένων τόξοις, πρὸς ἃ πεφύκασιν ἐξαιρέτως, οἱ ἱππέες τὴν πόλιν κενὴν καταλαβόντες ὑπέστρεψαν, οὔτε ἐς χεῖρας ἐλθόντες οὔτε τι λαβόντες .

II . Tadmur-Palmyr a as a Greek city in the Roman world · 27 rary opinions and events helped to shape Appian’s description of Antony’s alleged raid .19 But according to Eivind Seland, our approach was “unnecessarily critical” .20 He reasoned that “the existence of caravan trade in the region” is “well established” by a passage in Strabo’s Geography which describes the transport of goods through the Mesopotamian desert .21 For Seland, the geographer’s passage provides the framework for the intensification of the long-distance trade for which the oasis became the core in a swiftly developing network: “political fragmentation and predatory authorities along the Euphrates made the difficult, expensive and uncomfortable desert crossing competitive versus the Euphrates valley, which had been the traditional corridor of communication since the Bronze Age .”22 Be that as it may, Appian’s description of the attempted raid by Antony’s troops contains some serious inconsistencies, and the passage has now been discussed again by Michael Sommer, who agreed that “Appian’s Palmyra episode turns out to be a single anachronism, from which little to nothing can be learnt about pre-imperial Palmyra .”23 He also made the important observation that “for Strabo, Palmyra is not even worth a footnote”, which can only mean that the oasis did not yet play a key function in the caravan trade at the time of writing (or at least at the time in which Strabo’s own sources were written) . In the context of a discussion of Tadmur-Palmyra as a Greek city in the Roman world (or not), it is of course interesting that Appian explicitly refers to the settlement as a polis . With the evidence that is currently available, it cannot be known whether the historian’s word choice is adequate when it comes to the time of Antony .24 It would take another century before the term first appears in Palmyrene epigraphy, when in AD 51 “the polis of the Palmyrenes’ (Παλμυρηνῶν ἡ πόλις)” erected a statue for a citizen praised for his benefactions towards the city and its gods .25 It concerns a bilingual inscription, and the Aramaic equivalent has gbl tdmryʾ, conventionally rendered as the ‘community of the Tadmurenes’ . This term gbl, ‘community’, appears also earlier, in inscriptions from the first and second decennia of the first century AD, in AD 25 as the equivalent of the Greek term δῆμος, ‘assembly’ .26 Could the town have styled its assembly as a δῆμος without viewing itself as a πόλις? It depends of course what was meant in Palmyra by the word δῆμος, whether it was simply a new way of labelling a native organizational structure, or an actual proper reorganization along Greek lines . The second part of the double structure of the standard ‘Greek city’, the βουλή (‘council’) is first attested in AD 74, thus nearly fifty years after the first appearance of the civic assembly . It remains disputed whether this is evidence for a

19 20 21 22 23 24 25 26

Hekster – Kaizer 2004 . Seland 2016, 76 . Strab . 16,1,27, cited and discussed by Seland 2016, 75–77, building on Gawlikowski 1996, Millar 1998 . Seland 2016, 76 . Sommer 2018a, 66–68 . For discussion, see Yon 2010, 229–233 . PAT 0269, IGLS XVII .1,18 . PAT 1353, IGLS XVII .1,17 (AD 25) . The other inscription is in Aramaic only, PAT 2632 (AD 10/11) .

28 · Ted K aizer gradual introduction in Palmyrene society of the constituent parts of the new constitution, or whether it is simply a chance survival of evidence that wrongly gives the impression that one body predates another . It is commonly accepted that Palmyra’s Greek city constitution was implemented under Roman influence, which is often taken to have heralded proper integration of the oasis in the empire’s governing structures in the early years of the reign of the emperor Tiberius, whose adopted son Germanicus is mentioned in three inscriptions from Palmyra .27 Implementation of a Greek city constitution is indeed what the available evidence seems to support, but for the sake of argument it is worth stressing that the epigraphic evidence for the local society around this time is far from complete, and it is theoretically not impossible that the civic institutions — or rather, their Greek labels — dated back to the period before Palmyra came to be firmly part of the provincial system . It was noted above that according to Henri Seyrig the king Epiphanes from the archaic Greek fragment from the Temple of Bēl was a member of the Arsacid dynasty, and it is in principle possible that the oasis settlement briefly functioned as a Greek city in the Parthian world . Needless to say, that scenario can only be a hypothesis, and it must be clear that even if the applied terminology originated from within the Parthian or Hellenistic world, it only became visible at Palmyra by the Julio-Claudian period with the rise of the local epigraphic habit . A related problem is that concerning the so-called ‘four tribes of the city’, a somewhat artificial construct that over time came to occupy central place in Palmyrene society, most likely a social and partly religious reorganization of the existing indigenous tribes, clans and families, based on a new territorial subdivision of the oasis settlement . There are many more than four groupings in Palmyrene inscriptions which have been identified as tribes, and the ‘four tribes’ never fully replaced those pre-existing ones (even though the names of three of the four tribes were simply taken from the indigenous groupings) . The issue is still debated, but the majority of scholars agree that this scheme was also implemented under Roman influence early in the first century AD, alongside the new-styled council and assembly .28 Michael Sommer urged us to move away from a view that looks for ‘the moment’ in which Palmyra became properly integrated into the imperial system, as “the pattern of a smooth provincialisation of Palmyra […] presupposes a sharp separation between Roman and non-Roman territory in the Near East .”29 For Sommer, the region in which Palmyra was located was “a twilight zone with amorphous patterns of territory and power: The

27 Seyrig 1932, Yon 2010, 238–239 . 28 The most informative discussion of the relevant sources is Yon 2002, 66–72 . See also Smith II 2013, 121–143 . Christiane Delplace has argued, in Delplace – Dentzer-Feydy 2005, 210–214, that the civic re-organization of the “four tribes of the city”, for which evidence is only attested in a brief period in the second century AD, was merely “ephemeral” . 29 Sommer 2005a, 287, where he also argued that “the power exercised by Rome […] in the Syro-Mesopotamian steppe decreased gradually with the increase of distance .”

II . Tadmur-Palmyr a as a Greek city in the Roman world · 29 question is therefore not: was Palmyra a city belonging to the Roman Empire, but rather: how far did Roman influence and control reach in Palmyra?”30 With regard to this latter question it is relevant that a military tomb stone which was published in 2010 suggests that imperial forces in the form of an auxiliary unit were stationed at the oasis as early as AD 27 .31 This new evidence for early military involvement in Palmyra on the part of Rome certainly fits well with the fact that Germanicus is mentioned in three inscriptions, and thus with Seyrig’s thesis that the oasis became a proper part of the empire under the Julio-Claudians .

Episode 2: the long heyday of Palmyra’s glorious civilization The answer to Sommer’s question (“how far did Roman influence and control reach in Palmyra?”) is: seemingly a lot further by the time Hadrian made the oasis a stop on his tour through the eastern provinces, as witnessed by a bilingual inscription from AD 130/1 which records how “during the visit of the divine Hadrian” a local notable provided both citizens and foreigners with oil, provided accommodation to the army that accompanied the emperor, and built the cella of the Temple of Baʿal-Šamen .32 We have now arrived at the second of our episodes, the long heyday of Palmyra’s glorious civilization, financed mostly by the city’s control over an intricate network of long-distance trade, and a period when the oasis appears easiest to be bracketed with the Greek cities of the empire . At the time of the imperial adventus, the oasis city underwent a formal change of name and became known as ‘Hadrianē Palmyra’ or — as we know it from the tax law — ‘Hadrianē Tadmur’ (hdrynʾ tdmr) . According to Stephanus of Byzantium the new honorific name was a direct result of a re-foundation by Hadrian: “they called themselves citizens of the city of Hadrian because the city was re-founded by the emperor .”33 Shortly after Hadrian’s visit a prominent citizen is specified as being Ἁδριανὸς Παλμυρηνός, a ‘Hadrianic Palmyrene’, according to an inscription set up at the Agora .34 But the label stuck so well that it remained in use by the first half of the third century, long after the reign of Hadrian and in fact at a time that Palmyra had become a Roman colonia, which was supposedly worth much more of a mention: in two inscriptions set up by Palmyrenes in Koptos in Egypt (the

30 31 32 33

Ibid . Gawlikowski 2010b, see Sommer in this volume p . 10–11 . PAT 0305, IGLS XVII .1,145 . St . Byz . Ethnika, s .v . Πάλμυρα, ed . A . Meineke 1894, 498: οἱ δ’ αὐτοὶ Ἀδριανοπολῖται μετωνομάσθησαν ἐπικτισθείσης τῆς πόλεως ὑπὸ τοῦ αὐτοκράτορος . There is no good evidence, however, as Millar 1993, 324– 325, has emphasised, to support the idea (going back to Seyrig 1941, 171, and his own five phases which I mentioned at the outset) that the oasis at this time was granted the status of a free city, civitas libera or ἐλευθέρα καὶ αὐτόνομος πόλις . 34 PAT 1374, IGLS XVII .1 245 .

30 · Ted K aizer first from AD 216 and the second also from the third century) and one from Rome itself (from AD 236), they continue to identify themselves as ‘Hadrianic Palmyrenes’ .35 The tax law itself is the document that most clearly shows Palmyra as a Greek city – albeit one whose public inscriptions were bilingual .36 Setting out to revise earlier regulations in order to address the persistent disagreements between merchants and customs officials about dues to be collected, the tariff records a decree of the council proclaimed at a statutory meeting in April 137 and provides evidence for a whole series of civic offices . As John Matthews has shown in a classic article, what is often considered as a caravan city ought to be firmly located also within its own civic territory, with the tax law differentiating between loads transported to the urban centre from the villages and from outside the boundaries of the Palmyrena . In contrast to this more ‘typical’ set-up, entries on grazing rights for animals serve as a reminder that we are still concerned with “an oasis settlement surrounded by the economy and customs of the desert .”37 Around the same time Palmyra’s leading position in the long-distance trade reached its zenith, according to the list of ‘caravan inscriptions’, in particular during the peaceful half-century between the Parthian campaigns of Trajan and of Lucius Verus .38 They take the form of honorific inscriptions engraved on the statue bases on the columns lined up throughout the city, applying some of the standard formulations known from the Graeco-Roman world at large, but honouring the local benefactors — in contrast to members of the elite in more classical cities of the empire — for the ‘Oriental business’ of financing caravans and for securing their safe passage through the desert .39 Funded by the profits made in the trade in luxury products, the civic centre was completely transformed in the course of the second century .40 The first part was built of the central colonnade, which over time came to function as a second, more monumental thoroughfare between the western section of the town and the area near the great sanctuary of Bēl . The Agora sector, first planned and built from the Flavians until the first decennium of the second century, now received a number of additional buildings and adjustments also with a view towards the new parts of the city towards the north .41 The Temple of Bēl, whose cella was built and then enlarged in the first century AD, received its propylaea and a series of new porticoes, and many of the existing shrines elsewhere in the town underwent a process of monumentalisation that turned them also into splendid classical 35 For the inscriptions from Koptos, see AE 1984, 925, and SEG 34,1585, respectively . For the inscription from Rome, see PAT 0247 . 36 See above all Millar 1993, 325 . 470 . 37 Matthews 1984, 173 . See Meyer 2016, 95–96, for Safaitic inscriptions from elsewhere in the Near East referring to Tadmur, Meyer 2017, 210–211 . 38 Gawlikowski 1994, Gawlikowski 1996, Gawlikowski 2016a . 39 Millar 1998 . On the elite honoured in these inscriptions, see the discussions by Andrade 2012, Sommer 2016, Yon 2002, Sander in this volume . 40 See now Delplace 2017, 89–105 . 41 Delplace – Dentzer-Feydy 2005 .

II . Tadmur-Palmyr a as a Greek city in the Roman world · 31 temples, such as the places of worship for Baʿal-Šamen, Nebu and Allāth . No gymnasium has been unearthed at Palmyra, but its existence is proven by an Aramaic inscription from the first half of the second century AD that makes reference, in Aramaic transliteration, to a gymnasiarchos, the official responsible for the maintenance and general sponsoring of a gymnasium . As a Greek gymnasium was traditionally considered to be under the protection of the gods Heracles and Hermes, one may link with Palmyra’s lost gymnasium two reliefs from the Roman period that were reused in the Arab bastion that came to incorporate the monumental entrance to the great Temple of Bēl — one showing Heracles with his club and the other Hermes with his purse .42

Episode 3: Palmyra as a Roman colonia The third episode has been referred to as “a real transformation”, the introduction of a new constitution when the oasis was elevated to the status of a Roman colony at some point in the Severan period - as attested both in Ulpian’s book on the census and in the epigraphy from Palmyra itself .43 But does that really count, as Udo Hartmann has argued, as the “turning point in the history of Palmyra”?44 Was it really “the most dramatic change for the Palmyrene society”? The characteristic pair of annual chief magistrates known as duumviri found expression in Palmyra only (with one possible exception45) through the Greek term στρατηγός and its transliteration in Palmyrenean Aramaic . The colonia, therefore, uses the term that in Greek cities designated a general or a civic magistrate . The term already appears on the second of the two above-mentioned possibly ‘Hellenistic inscriptions’ from Palmyra, another archaic fragment found under the flooring of the Temple of Bēl, thus drawing attention to its long history in Palmyrene epigraphy .46 As a colonia, Palmyra remained publicly bilingual - Aramaic and Greek . The few attempts at public trilingualism, with Latin, were now left behind . As Millar put it: “Contrary to what might have been expected, Latin, never very common, now disappears altogether from the inscriptions of the city .”47 Latin is absent too from the one Palmyrene funerary relief (now in the Louvre) depicting a colonist from Berytus, modern-day Beirut, who is identified 42 For the inscription, see PAT 1406, IGLS XVII .1,221, Sartre 1996, 392 . For the reliefs, see Daubner 2015, 39, Seyrig 1944/45, 75–76 and pl . IV . 43 Millar 1990, 42–46, Millar 1993, 326 . Ulp . De cens. 1 (Dig. 50,15,1,4–5): sed et Emisenae civitati Phoenices im­ perator noster ius coloniae dedit iurisque Italici eam fecit. est et Palmyrena civitas in provincia Phoenice prope barbaras gentes et nationes collocata, “Our emperor [i .e . Caracalla] bestowed upon Emesa, a city of Phoenicia, the title and the rights of an Italian colony . The city of Palmyra, situated in the province of Phoenicia, and adjoining barbarous peoples and nations, enjoys the same right .” 44 Hartmann 2016, 54 . 45 PAT 0280, IGLS XVII .1,75: δυα[νδρικὸν] . See also Millar 1990, 44 . 46 IGLS XVII .1,2: Δίων στρα[τηγὸς ---] | τὴν ΑΥΤΟ[---] . 47 Millar 1993, 327 . See now Yon 2008 .

32 · Ted K aizer as such not in the official language of his hometown, but both in Aramaic and in Greek .48 The latter is remarkable in itself, because Greek was not normally used in this medium . Like the public bilingualism, also “city-patriotism”49 continued into Palmyra’s colonial episode . The typically Palmyrene god Yarḥibōl, who had already acted alongside Bēl and ʿAglibōl as a divine recipient of the first part of the new temple in AD 32, now seemed to come to the fore more than ever . A series of inscriptions from the Severan period onwards record how the god, labelled as ‘ancestral’ (πατρῷος), provided testimonials on behalf of deserving citizens, and did so occasionally in partnership with an important Roman official .50 Together with Iulius Priscus, ‘most eminent and sacred praefectus praetorio’, Yarḥibōl witnessed the good actions of a citizen who had served as duumvir (obviously written as στρατηγὸς) at the time the emperor Alexander Severus stopped by Palmyra, the second time that the oasis was honoured by an imperial visit .51 Public service and public generosity were now a reason to receive honorific inscriptions more than ever, at a time that involvement in long-distance trade was mentioned less and less (without this specific Palmyrene form of euergetism disappearing completely) . And where Palmyrene notables who had wanted a ‘career abroad’ in the second century AD acted for example as satrap of Bahrein (on behalf of the king of Spasinou Charax) or as chief magistrate of Mesene,52 the leading citizens of colonia Palmyra now managed to get on the ladder of the Roman order, with membership of the equestrian or even the senatorial order . In the meantime, the transformation of the civic centre continued from the previous episode .53 The middle and eastern sections of the central colonnade were built and — possibly, according to a typically brilliant hypothesis recently put forward by Michał Gawlikowski54 — an oval plaza, as we know it from Gerasa, to negotiate what would otherwise have been too sharp and awkward an angle between the final stretch of the colonnade (east of the arch) and the propylaea of the Bēl temple . But Palmyra’s location as an oasis in the middle of the Syrian steppe remained of course the same, and this fact did not escape Ulpian, as is clear from his above-cited observation that the new colonia was “adjoining barbarous peoples and nations” (prope barbaras gentes et nationes collocata), no doubt meant as a reference to the nomads inhabiting the arid lands surrounding the city .55

48 For the funerary relief of the citizen of colonia Berytus, see Dentzer-Feydy – Teixidor 1993, 162, no . 166; PAT 0761, IGLS XVII .1,551 . 49 Millar 1993, 327 . 50 Kaizer 2002, 146–147 . 51 PAT 0278, IGLS XVII .1,53 . 52 For the σατράπης of the citizens of Thilouos, i .e . Bahrein, see PAT 1374; IGLS XVII .1,245 . For the ἂρχων of Mesene, see IGLS XVII .1 160 . See Sommer in this volume, p . 37 . 53 Delplace 2017, 107–128 . 54 Gawlikowski 2016b . 55 See also Hartmann 2016, 55 .

II . Tadmur-Palmyr a as a Greek city in the Roman world · 33

Episode 4: a King of Kings in charge (but still a Roman colonia) In what is commonly viewed as the final phase in the history of Palmyra’s civilization, our fourth episode, the oasis is often thought to have turned into a ‘kingdom’ or at least a place under one-man (and later one-woman) rule . Udo Hartmann gave the wrong impression, however, when he chose to formulate that the “[colonial] constitution of Palmyra remained intact until the elevation of Odaenathus as king .”56 Firstly, Palmyra was not to give up its status as a colonia any time soon, as is clear from milestones set up under Diocletian and his successors .57 Secondly, Odaenathus was never ‘elevated as king’, in any case not as king over Palmyra (unless we want to believe the Historia Augusta58) . As Michał Gawlikowski has convincingly argued, the undeniable assumption of the title King of Kings by Odaenathus (jointly with his son Ḥairan/Herodian) was a direct challenge to Shapur following a specific victory over the Sasanian army .59 That specific victory seems to have taken place ‘near the Orontes’, between 259 and 262, at least according to Gawlikowski’s interpretation of an inscription on the arch at the end of the central colonnade which was set up in honour of Odaenathus’ eldest son, who must have received the title together with his father .60 The inscription, however, was set up by two men who identified themselves as major office holders of the Roman colony, as procurator ducena­ rius and centenarius, respectively, and both as duumviri . One of them, Septimius Vorodes or Worōd, is known to have been Odaenathus’ main henchman, who in a number of inscriptions is designated as argapet .61 But rather than viewing him in this latter capacity, as Udo Hartmann proposed, as “governor of the king” in a new political system,62 the fact that one and the same person could boast to be a high civil magistrate through both Roman and Persian titulature shows how both imperial and Eastern governmental jargon had permeated the city’s ruling structures . So, contradictory as it may seem, there was a King of Kings (or rather, even more contradictory, two Kings of Kings) in charge of a Roman colonia . It could be argued that there had already been a step towards such a peculiar situation with the introduction in the 250s — similarly at a time that the city enjoyed colonial status — of the innovative position of rš tdmr, ‘head of Tadmur’ . In the same way, therefore, that Odaenathus and Ḥairan/Herodian were concurrently referred to as King of Kings from ca 260, a decennium earlier they seem to have jointly occupied 56 57 58 59

Ibid ., 55 . But see ibid ., 65 . See ibid ., 54 n . 6 for all references . SHA Gall. 10,1: rex Palmyrenorum, SHA trig. tyr. 15,2: adsumpto nomine primum regali . Gawlikowski 2007a, 307, Gawlikowski 2010a, 474–475 . For a different interpretation, see the illuminating study by Gnoli 2007, 92, who argued that it concerned “an extreme action organized by Gallienus in an attempt to stop the breakup of the Oriental provinces after the devastating Persian attacks … [by] claiming the Arsacid throne in the face of the Sasanian usurpation .” 60 Gawlikowski 2007a, 295 . For further references, see IGLS XVII .1,61 . 61 On this title, see Gnoli 2007, 95–113 . 62 Thus Hartmann 2016, 64 .

34 · Ted K aizer the then new and equally non-colonial position of ‘head of Tadmur’ (rendered in Greek as ἔξαρχος Παλμυρηνῶν) .63 In any case, the assumption of the title King of Kings did nothing to deny Odaenathus’ loyalty to Rome .64 The leading citizen of the oasis city, even more from a position of strength as victor over Shapur, took full advantage of the Roman system from within it, though in a rather non-Roman fashion . When he and his son defeated the Sasanian army they could legitimately call themselves King of Kings — as a bit of an inside joke (like the heavy-weight title belt in boxing it is to be gained by beating the incumbent) .65 It certainly did not turn Odaenathus into a vassal king of Rome . The situation was of course to change swiftly following the death of Odaenathus and his son, with Odaenathus’ widow Zenobia in charge on behalf of her still underage son with her late husband, Vaballathus . This most famous part of the fourth episode will not be discussed here, but suffice it to say that the so-called ‘Palmyrene Empire’ was in essence a usurpation attempt very much from within the Roman realm . As regards her own royal title, as Annie and Maurice Sartre have recently pointed out, Zenobia was Queen at Palmyra but never Queen of Palmyra .66 The perhaps not altogether very trustworthy ancient sources on the matter emphasise the classical intellect she assembled at her ‘Hellenistic court’, of which the best-known names are the philosopher Longinus, the sophist and historian Callinicus of Petra, and bishop Paul of Samosata .67 Their, and others’, association with Zenobia and Palmyra is in some cases highly doubtful, but it is clear that the sources wanted to present Zenobia as attempting to emulate Iulia Domna .

63 The relevant evidence is collected and discussed in the classic article by Gawlikowski 1985 . A damaged inscription in Palmyrenean Aramaic has been restored as recording the dedication of a ‘throne’ (mw[tb]ʾ) to Odaenathus, see PAT 2753 with Gawlikowski 1973, 78 . For Sartre – Sartre 2016, 142–143, the position of ‘head of Tadmur’ indicates a purely military position . 64 A very fragmentary Greek inscription on a column drum may suggest the involvement of one of his associates in the so-called imperial cult . See IGLS XVII .1,120, Kaizer 2002, 149 . Gnoli 2007, 88–89 , proposed to link the reference in the inscription to Helios with the language used in relation to Odaenathus in the Thirteenth Sibylline Oracle (on which see the next footnote) . On the link between emperor worship in the Roman Near East and the cult of the sun god, see now also Bru 2016, 61 . 65 The unique depiction of Odaenathus and Herodian on a mosaic floor in a banqueting room just behind the central colonnade could easily be interpreted along similar lines . Combining classical imagery with local iconographic detail, the panels of Bellerophon riding Pegasus and slaying the chimaera, and of the horseman hunting Persian tigers, have been convincingly interpreted as allegories commenting on contemporary third-century events . The depiction of Odaenathus and Ḥairan/Herodian on the mosaic can therefore be viewed as a visualization of the poetic language used in the Thirteenth Sibylline Oracle which had ‘predicted’ the Palmyrene victory . For presentation and ingenious discussion of the panels and the issues involved, see Gawlikowski 2005, Gawlikowski – Żuchowska 2010 . For the relevant text of the Oracle, see Dodgeon – Lieu 1991, 71, no . 4 .3 .2 (lines 164–171), Potter 1990, 177 . 66 Sartre – Sartre 2016, 147 . 67 See Stoneman 1992, 129–132, for what he described as “a cabinet of intellectuals” . For discussion of ‘the Hellenism of Zenobia’, see Bowersock 1987 .

II . Tadmur-Palmyr a as a Greek city in the Roman world · 35

Episode 4a: the immediate aftermath of Zenobia’s defeat It remains unclear how much, or how little, changed in the immediate aftermath of Aurelian’s victory . It is hard to imagine that there would have been an immediate “ban” on Palmyrene culture,68 either then or after the emperor had crushed a brief uprising a year later under the new ‘leader’ Apsaios, a Palmyrene citizen styled as προστάτης in a Greek inscription and mentioned also, uniquely, by Zosimus .69 But some potential key pieces of evidence have proven harder to date than was initially thought . A late Aramaic inscription from one of the necropoleis providing information about the common Palmyrene practice of transferring tomb space from one family to another has been dated to either 274 or 279, but the bilingual Aramaic-Greek inscriptions recording building works in honour of the Palmyrene god Arṣu which had initially been dated to at least six years after the fall of Palmyra, the end of the 270s, are according to Jean-Baptiste Yon from the second half of the second century AD instead .70 In any case, before too long the sources for the local language and its script, for the local deities, and for the local art form all dried up .

Episode 5: late-antique Palmyra In the final, fifth episode a whole new world seems to appear . Gone are the ‘exotic elements’ from our evidence .71 With the oasis now having become a part of the strata Diocletiana, a military camp was built in the west of the ruins (incorporating notably the Temple of Allāth) and the building inscriptions that attest to activities here are now all in Latin .72 If worship of the indigenous goddess continued (which is of course not impossible), it is certainly no longer made explicit in the sources . The well-known classical Athena statue from the temple was in fact, as Michał Gawlikowski has demonstrated, imported into Palmyra only after the building became incorporated into the Roman camp .73 The same seemed to 68 Correctly noted by Hartmann 2016, 66 . 69 IGLS XVII .1,77: Σεπτ(ίμιον) Αψαιον τὸν πολείτην καὶ προστάτην ἡ πόλις, ‘the polis to Septimius Apsaios, citizen and patron’ . See Zos . 1,60,1: ἐχομένου δὲ Αὐρηλιανοῦ τῆς ἐπὶ τὴν Εὐρώπην ὁδοῦ, κατέλαβεν ἀγγελία τοιαύτη, ὡς τῶν ἐν Παλμύρᾳ καταλειφθέντων τινὲς Ἀψαῖον παραλαβόντες, ὃς καὶ τῶν προλαβόντων αὐτοῖς γέγονεν αἴτιος, “As Aurelian continued his journey into Europe, news came that some of those left behind in Palmyra had won over Apsaios, who was responsible for the earlier events” (translation by Ridley 1982, 19) . 70 For the late example of the so-called cession inscriptions, and the problem with dating it securely, see PAT 0055, with Hartmann 2016, 66 n . 69 . For the Arṣu inscriptions, see IGLS XVII .1,80–81, Yon 2013, 342, no . 25–26 . The original dating, to the late 270s, had previously been widely followed, see e .g . Kaizer 2002, 122–123, Millar 1993–336 . 71 The evidence (literary and epigraphic) for Palmyra in Late Antiquity, from the end of the third until the seventh century, is conveniently collected by Kowalski 1997 . See now above all the archaeological and historical study on this period by Intagliata 2018 . 72 See Millar 1993, 182–183 . 73 Gawlikowski 2008 .

36 · Ted K aizer happen at the Temple of Baʿal-Šamen: in 302 an optio princeps made a dedication, in Greek, to Zeus the Highest and Listening, without any mention of the Lord of the Heavens under his Oriental name .74 Another Greek inscription, dated between 293 and 303, records building works at the local baths (designated as ‘of Diocletian’) under Sossianus Hierocles, prae­ ses in Syria Phoenice .75 According to the Notitia Dignitatum, Palmyrene units continued to form part of the empire’s forces, in the form of a ‘wedge’ squadron of armoured cavalry’ .76 Various churches from the fourth century onwards have been excavated,77 and Marinus bishop of Palmyra is listed amongst the participants of the Council of Nicea in 325 .78

Conclusion It has been shown, through what is very much a mere selection of the existing sources, that throughout Palmyrene history there is enough material to argue both that the polis Palmyra can be bracketed with other cities of the empire, and that the oasis is above all an exceptional case where everything seems different . The only time that Palmyra genuinely is ‘like other places’ is the fifth episode, when the oasis settlement has come to function as a military station and a bishop’s see in Late Antiquity . But it is ‘like other places’ only since by then there is no longer evidence available for the unique local civilization expressed through the Palmyrenean script and language, the peculiar art form, the local divine world, typically Palmyrene nomenclature and the peculiar tower tombs . However, the latter would continue to provide the backdrop to life at the oasis, and (with the bilingual inscriptions on the consoles of the standing colonnades and other visible reminders) will surely have helped the inhabitants of fifth-episode-Palmyra to realise that once, the oasis was a city like no other .

74 IGLS XVII .1,154: Διὶ ὑψίστῳ καὶ [ἐ]πηκόῳ Ἀουεῖτο[ς] ὀπτίων πρίνκιψ εὐχὴ[ν ἀ]νέθηκεν ἔτους 613 Γο[ρπ]ιαίου 25 . For recent archival research into the post-Roman occupation of one of the courtyards of the temple, see Intagliata 2016 . 75 IGLS XVII .1,100: ὑπὲρ σωτηρίας καὶ νείκης τῶν κυρίων ἡ[μῶν αὐτοκρατό]ρων καὶ Καισάρων καὶ τῆς ΚΟ [----] ἐτελέσθη τὸ Διοκλητιανὸν βαλανῖον, διακοσμοῦντος τὰ μέγ[ιστα --- ---] τοῦ διασημ(οτάτου) καὶ σπουδῇ Σοσσιανοῦ Ἱερ[οκλέους --] Σεραπίω[νος ---]Υ Ἀντων[---], “for the safety and victory of our lords the emperors and the Caesars and … . who has made the baths of Diocletian, having decorated the large … … the perfect attitude and the zealousness of Sossianus Hierocles --- Serapion ---- .” 76 Not. Dign. [or.] 7,34: cuneus equitum secundorum clibanariorum Palmirenorum, on which see Schmidt Heidenreich 2016, 231–232 . 77 See, e .g ., Jastrzębowska 2013, Majcherek 2013 . 78 See Kaizer 2010, 115 n . 9, for references .

III. Gateway Tadmur. The beginnings of Palmyra’s long-distance trade Michael Sommer In April AD 131, the Palmyrene merchants resident at Spasinou Charax put up a honorific inscription for their fellow-countryman Yarḥai, Ἁδριανὸς Παλμυρηνός .1 The document gives us no clue whatsoever as to why Yarḥai had been honoured . Nor do we learn anything about the precise relationship between the merchants and the honorand . This is remarkable in itself, as it sets this inscription from the Palmyrene Agora apart from the countless honorific decrees we know from the Greek-speaking eastern Roman provinces . But this is not what matters in the present context . What is truly remarkable about the inscription for Yarḥai is the administrative position held by the man from Palmyra: he is the governor, σατράπης, of the Thilouanoi, the inhabitants of the island of Thilouana, the ancient Dilmun and present-day Bahrain . Yarḥai performs this duty for Meeredates, the king of Spasinou Charax, the ruler of the Parthian territory of Charakene .2 Yarḥai, a citizen of Hadrianē Palmyra, a major city with strong ties to the Roman Empire, is the official of a Parthian rex . In AD 131, Rome and Parthia are not at war and are probably going through one of the calmer periods in their somewhat bumpy relationship .3 Yet, a citizen of Palmyra performing duties as a high-ranking official in the service of a Parthian sub-king seems weird enough to seek for an explanation . Such a line-up is conceivable only under a number of circumstances: from a Parthian point of view, a Palmyrene citizen must have appeared as sufficiently loyal or at least harmless to be entrusted with such an important position; to make a Palmyrene the governor of Bahrain was only an obvious choice, if Palmyrenes had business in the Persian Gulf; this was, of course, the case, as Palmyrenes worked the trade route connecting their oasis with the west coast of India; and this trade route ran via the Persian Gulf and Spasinou Charax, the capital of Charakene . All this explains the presence of Palmyrenes in the neighbourhood, but the choice of Yarḥai as governor of Thilouana is plausible only if we accept that this man had a capacity for creating personal networks spanning across imperial frontiers . Yarḥai, the Palmyrene, was a global player of the Hadrianic age . While he served as governor of Bahrain, his statue stood almost 1,500 kilometres afar on the Agora of Palmyra .

1 2 3

PAT 1374 . See Butcher 2003, 185, Gawlikowski 1996, 141–42, Hoffmann-Salz 2011, 424, Merkelbach – Stauber 2005, Nr . 702, Potts 1988, 143–44, Schuol 2000, 56–57, Sommer 2018a, 190–191, Yon 2002, 105 . PAT 1374: [Ἀδ]ριανὸν Παλμυρηνόν σατρά | [π]ην Θιλουανῶν Μεερεδάτου | βασιλέως Σπασίνου Χάρακος . Edwell 2008, 46–54, Timpe 1963, Zecchini 2005, Ziegler 1964 .

38 · Michael Sommer There is no doubt that second century AD Palmyra was a prime hub of the intercontinental long-distance trade . Nor can it be disputed that the oasis city’s elite was outstandingly cosmopolitan, a group of inter-imperial networkers who could rely on Palmyrene diasporas scattered across the Mediterranean and the Near East, at least from Rome to Socotra in the south-east and Babylonia in the east, possibly beyond .4 It is also sufficiently clear that the commerce with luxury items was key to Palmyra’s wealth and power in the second and third centuries . In contrast to most cities of the Roman Empire, Palmyra came close to Max Weber’s ideal-typical Produzentenstadt, with an urban economy supporting the bulk of the city’s livelihood .5 In the context of the Roman world, Palmyra was a giant anomaly .6 While all this is relatively straightforward — controversial as the details may be –, we are almost completely in the dark as to where the anomaly that was Palmyra originated . How did Palmyra’s upper crust develop into a bunch of networkers operating on an enormous scale? Why could the Palmyrenes permeate the frontier between two rivalling empires? Where did the manpower the trade relied upon come from? Why did a substantial portion of the Indian trade at some point shift towards the north, away from Egypt and the Red Sea, making Palmyra its bottleneck and hub? And when did all this happen in the first place? Until recently, there were absolutely no answers to such questions . However, this has changed with the fieldwork that has been conducted by Austrian, German and Syrian archaeologists in the so-called Hellenistic City in the area of the wadi . The two trenches opened in this area, a geophysical survey and remote sensing through satellite imaging, have revealed a wealth of information on the early stages of Palmyra as a settlement and its involvement in commercial activity .7

‘The Hellenistic City’ Calling the area to the south of the wadi the ‘Hellenistische Stadt’ is, strictly speaking, misleading . Neither was this the only part of Palmyra inhabited in pre-Roman times, nor is its occupation limited to the Hellenistic period . On the contrary, settlement activity in the Hellenistic City, as elsewhere in Palmyra, seems to have reached its apex in the Roman imperial period . So far, the chronology of the Hellenistic City seems in perfect consonance with the rest of Palmyra . The difference lies in the accessibility of strata predating

4

5 6 7

There were several Palmyrene diasporas we can trace in the archaeological and epigraphic records: on DuraEuropos Dirven 1999; on the Euphrates Gawlikowski 1983, Gawlikowski 2007b; on Charakene Schuol 2000, 380–88; on Egypt Cuvigny – Briquel-Chatonnet 2012, Speidel 1984 and Cobb in this volume; on Socotra Gorea 2012a, Gorea 2012b; on South Arabia Robin 2012a . Weber 2000, 4 . See Nippel 2000, 22 . Thus Sommer 2015, 179–82, Sommer 2016, 14–16, Sommer 2018a, 194–203, Sommer 2018b, 175–226, and similarly Veyne 2005, 259–344, Veyne 2017 against Sartre 1996 and now Sartre – Sartre 2016, 59–70 . Schmidt-Colinet 2003b, Schmidt-Colinet – al-Asʿad 2000, Schmidt-Colinet – al-Asʿad 2013 .

III . The begin nings of Palmyr a’s long-distance tr ade · 39 the Roman period . With the exception of one sounding in the courtyard of the Temple of Bēl dug under Comte Du Mesnil du Buisson and a first century BC necropolis at the site of the Temple of Baʿal-Šamen, the pre-Roman occupation of the oasis had been a lacuna . Du Mesnil’s sounding had revealed, in the 1960s, some evidence on a sanctuary predating the Temple of Bēl, going back at least to the first century BC .8 Now, the recent excavations in the Hellenistic City have literally boosted our knowledge of early Palmyra and its urbanism . According to the geophysical survey and remote sensing, the Hellenistic City was a densely populated neighbourhood filling the entire eastern part of the triangle between the wadi and the modern road leading to the Efqa spring . The neighbourhood is shaped by two major streets, running roughly parallel to the wadi from west-south-west to eastnorth-east . The northernmost of these two streets, called the ‘main street’ has been the area’s main traffic artery . To the north of this main street, next to the wadi, the patterns of large, monumental buildings have emerged in the record . They appear to be have fallen in disrepair, possibly by the repeated flooding of this area . By contrast, the southern neighbourhood was dominated by small, densely packed building units stretching along the two major axes and their respective secondary roads .9 In this area, where the main traffic artery meets one of the secondary streets, the excavators dug sounding I . The oldest architectural remnants in this trench date back to the third century BC . Mud-brick buildings with quarry-stone foundations were erected here shortly before 200 BC . One of the buildings featured plastered walls with some wall-paintings . Six different phases of occupation can be identified: phase 1 (200–170 BC), phase 2 (170–150 BC), phase 3 (150–100 BC), phase 4 (100–30 BC), phase 5 (30 BC–AD  100), phase 6 (AD 100–300) .10 Of particular interest is the origin of pottery across the various phases . While in the early periods of occupation local wares were mixed with pottery imported from western and northern Syria, as well as (very few) items from mainland Greece, the image changes with time: from occupation phase 2 onward, some fragments of Parthian glazed pottery make their appearance alongside other wares; their proportion increases sharply in phase 4, until it reaches a climax in phase 6 . In addition, starting in the third century BC, there are numerous locally produced vessels imitating Mediterranean pottery .11 When it comes to architecture, the most conspicuous structure in the northern quarter is a large building with numerous rooms, called the ‘Khan’ by the excavators . Here, sounding II was dug . The building, erected in the first half of the first century BC, burnt down around AD 200, but continued to be in use throughout the third and fourth centuries . The structure is a house with a central courtyard surrounded by rooms of different sizes . 8 9 10 11

al-Maqdissi 2000, Du Mesnil du Buisson 1966, Du Mesnil Du Buisson 1967 . Fassbinder – Linck 2013, Linck 2016 . Plattner 2013 . Römer-Strehl 2013a, 8–20 .

40 · Michael Sommer The main part of the building dates to the Augustan period . A number of rooms feature wall paintings and stucco friezes . According to the excavators, the central location of the building, its dimensions and lavish interior, as well as the distribution and quality of the findings suggest that this has been the dwelling of a very wealthy family . A number of small finds (coins, scales), as well as the range of pottery wares directly point towards an involvement in the long-distance trade .12 Six phases of occupation can be identified for this building, the first four dating back to the time before the blaze, the final two after the destruction . Of the diagnostic pottery dating to the first four phases, the majority is Parthian-glazed from Babylonia . Aside from local wares, there is a large amount of Eastern Sigillata A and some traces of pottery imported from the Arabian Peninsula, from Asia Minor and some ‘eggshell’ pottery also coming from Babylonia .13 The ceramic evidence suggests that Palmyra was in touch, from the early third century onward, with the Mediterranean and, by the middle of the century, with Mesopotamia .14 That much of the pottery points towards the Levant, is hardly surprising . But the rise in wares imported from southern Mesopotamia around 150 BC and again at some point during the mid-first century BC is undeniable and also confirmed by other categories of findings . The first Parthian amphorae date back to the mid-first century BC, around the same time when amphorae from the western Mediterranean first appear in the Hellenistic City . By far the oldest piece from the Mediterranean is a stamped amphora from Rhodes, from the workshop of a certain Damokrateus, dating to around 200 BC . Lamps have been imported from the Levant since the early first century BC, but there are no lamps from Mesopotamia .15 The archaeological record from the Hellenistic City solves a number of problems and helps us see clearer about the whens and hows of Palmyra’s urban growth . It confirms that there was a settlement as early as in the late third century . Nothing can be said about the size of this settlement, but it seems sufficiently clear from the evidence that it was in touch with the Levant and, to some degree, involved in the exchange of commodities . Ties between the oasis and Mesopotamia, Babylonia in particular, seem to have intensified twice: for the first time around the middle of the second century BC and for a second time in the second half of the first century BC . It is probably safe to say that the scale and scope of Palmyra’s trade increased over these two centuries and was, by the time of Augustus, substantial . It then reaches its maximum level after AD 100 . Yet, the Hellenistic City does not reveal how precisely this growth proceeded: did Palmyra start out from nothing in order to expand organically into the city it was in the first and second centuries AD? Or did it grow incrementally, in stages, with each stage being tied to events on a larger scale? 12 13 14 15

Ertel – Ployer 2013, Linck 2016 . Römer-Strehl 2013a, 21–30 . Römer-Strehl 2016, 111–12 . Laubenheimer – Römer-Strehl 2013, Römer-Strehl 2013a, Römer-Strehl 2013b .

III . The begin nings of Palmyr a’s long-distance tr ade · 41

From Seleucid to Parthian and Roman: shifting patterns of exchange How did changes in the political landscape of Syria and Mesopotamia affect Palmyra and its trade? Can the record from the Hellenistic City be in any way correlated to the change of imperial tides in the Near East? If so, how can the rise of Palmyra and the blatant cosmopolitanism of its elite be explained? How did the peculiar political stratigraphy develop that turned out to be such a fertile biotope for Yarḥai and the likes of him? Until the reign of Antiochus III, the oasis of Tadmur belonged to the wider Seleucid world, though the intensity of governance was probably low in a marginal area such as the Syrian Desert .16 After Antiochus’ death, the eastern and Central Asian satrapies became detached from the empire for good, but for the time being the Seleucids stood their ground in western Iran .17 This changed when the Parthian king Mithradates I launched a series of attacks on Seleucid Media, Mesopotamia and Persis . Around 148, the Parthians captured the Median capital of Ecbatana; by 141, they had conquered Seleucia on the Tigris, where coins for Mithradates were issued, thus making Mesopotamia a frontier province . After that, they turned south, overrunning the kingdoms of Elymais and Charakene . Much of this catastrophic defeat was due to fact that the Seleucid empire had been debilitated by civil war . In 139, Mithradates captured and imprisoned the Seleucid king Demetrius II .18 However, the once all-powerful Seleucid Empire recovered . In 138, the new king Antiochus VII launched a series of counter-attacks into Babylonia and Iran . He was at least partly successful . Control of Mesopotamia and Western Iran was disputed between the two great powers for the remainder of the decade until, in early 129, the Parthians under their new ruler Phraates II undertook a surprise attack on Antiochus’ army, killing the king in action . From this point onward, the Parthians were in firm control of Babylonia, where they established, with Seleucia and Ctesiphon, two of their royal residences .19 Further up the Euphrates, the Seleucids still remained in control . The latest trace of Seleucid rule in Dura-Europos is a contract (P . Dura 34) dating to 116 BC . What happened to Tadmur is unknown . Whether or not the Seleucid Empire exerted some kind of notional suzerainty over the Syrian Desert is, after all, inconsequential .20 Trying to establish the importance of Mesopotamia for the intercontinental long-distance trade in the Hellenistic period is rather difficult . There is no doubt that there was commercial exchange between India and the Seleucid Empire, as there was trade between 16 Cohen 2013, 36–43, Kosmin 2014, 93–119, Millar 1987, Shipley 2000, 275–279, Sommer 2018a, 39–46, Yon 2003 . On the Partho-Seleucid relationship under Antiochus IV Mittag 2006, 318–27 . 17 Lerner 1999, 13–62 on the development in eastern Iran in the 3rd century BC . 18 Ehling 2008, 127, Sherwin-White – Kuhrt 1993, 223–29 . On Charakene Schuol 2000, 291–300 . 19 Ehling 1998, 141–144, Fischer 1970, Lerouge-Cohen 2005 . 20 Bellinger 1949, 65, Boiy 2004, 180, Gregoratti 2016a, 17 .

42 · Michael Sommer China and the Parthian Empire . A Chinese embassy sent to the empire of Anxi (Parthia) in 115 BC is confirmed through a Chinese chronicle .21 Slightly later, a land-based caravan trade was put in operation between the two empires . Which routes were used by the Seleucids, and whether they were seaborne or land-based, is unknown .22 However, the availability of eastern luxury goods in the western Satrapies is an undeniable fact .23 On the other hand, the scale of trade between mainland Greece and the Seleucid Empire probably declined with time .24 Towards the end of the third century BC, the massive importation of Greek pottery into the empire gave way to local products imitating the Greek wares .25 Most indicators point towards a vibrant domestic trade within the Seleucid Empire . To be sure, most products were consumed locally, as in all imperial economies before Hellenism . But the distribution of coins suggests that currency was moving relatively unhampered across the full expanse of Seleucid territory . Coins from Babylon and Seleucia were found in Asia Minor and Syria, for instance . If the distribution of coin hoards can be taken as evidence for trade routes, then a dense network of such routes covered the entire Fertile Crescent, with the main route connecting Syria and Mesopotamia running probably via Beroia-Ḥalab-Aleppo . The unequal distribution of resources — lack of raw materials, but an abundance of food production in Mesopotamia, and vice versa in most parts of Syria — further suggests that the two regions within the Seleucid Empire must have been connected to each other .26 We would of course like to know in what ways the precise patterns of production, consumption and exchange shifted, once Mesopotamia and Syria were no longer united under the same political outfit . In their still invaluable monograph on the Seleucid Empire, Amélie Kuhrt and the late Susan Sherwin-White point out that there was “no commercial or cultural iron curtain between the Greek-ruled east […] and Parthia” .27 While this is probably true, it remains a hypothesis; and so far, there has been no solid data to verify it . Politically, three periods in the relationship between Mesopotamia and Syria can be distinguished: first, both regions were under firm Seleucid domination (period 1) . This period lasted until the 140s, when Babylonia was first occupied by the Parthians . Next came a phase of unstable political control, initiated by Antiochus’ counter-attack (period 2) . This ended with the Seleucid king’s death in 129, giving way to full-scale Parthian control of Babylonia (period 3) . Period 1 strikingly coincides with the arrival of Mesopotamian pottery, the so-called Parthian glazed pottery, in Palmyra around 150 BC, at the end of 21 Schuol 2000, 149 . 22 Schuol, ibid ., 268, conjectures that there was a Seleucid fleet operating in the Persian gulf . This may have been the case, but there is no evidence for such a fleet . 23 Barisitz 2017, 28–29 . 24 On the severing of ties under Antiochus I who was born in Asia and never visited the Greek homeland Kosmin 2014, 85–87 . 25 Sherwin-White – Kuhrt 1993, 159–60 . 26 Pirngruber 2017, 71, van der Spek 2007, 426–430 . 27 Sherwin-White – Kuhrt 1993, 66 .

III . The begin nings of Palmyr a’s long-distance tr ade · 43 the second period of occupation of sounding I in the Hellenistic City . According to the excavators, the emergence of Parthian glazed pottery in Palmyra marks by far the earliest occurrence of this type in Syria .28 The number of diagnostic fragments is quite low at first (4 and 8 in phases 2 and 3, against a total of 93 and 98 respectively), but it rises considerably in phase 4 (14 against 97) and especially phase 5 (23 against 109) .29 The percentage (Fig . III .1) increases steadily from 0% in phase 1 to 4% (phase 2), 7% (phase 3), 14% (phase 4) and 17% (phase 5), before reaching a staggering 27% (phase 6) . Furthermore, local production in phase 4 increasingly imitated wares from Babylonia and from further east .30 Palmyra’s interaction with

III.1: Proportion of ‘Parthian’ glazed pottery among diagnostic shards from sounding I, Hellenistic City.

southern Mesopotamia thus seems to have accelerated during phases 4 (c . 100–30 BC) and 5 (after c . 30 BC) . This correlates nicely with an increased Roman activity in the Near East, resulting in the province of Syria being established as Rome’s bridgehead in western Asia by Pompey in 64/63 BC .31 In particular, it coincides with the military crisis of Crassus’ 28 29 30 31

Römer-Strehl 2013a, 12 . Ibid ., 11–19, charts 3–7 . Ibid ., 14 . Bridgehead: Millar 1993, 27 . See also Edwell 2008, 7–16, Sartre 2005, 31–44, Sommer 2018b, 62–69 .

4 4 · Michael Sommer abortive Parthian campaign in 53 . Both events fall into phase 4, which sees the proportion of Mesopotamian pottery doubled, as compared to phase 3 .32 Correlating the political history of the Near East with the chronology of pottery imported into Palmyra yields a surprising, yet unambiguous result: twice in the course of a century, Syria became politically detached from Mesopotamia; twice, a political frontier between two rivalling empires was established across the Fertile Crescent . Each time this happened, the ties between Palmyra and Babylonia, as reflected in the ceramic record of the Hellenistic City, became more intense . This can hardly be a coincidence, the only plausible explanation being that instead of obstructing Palmyra’s trade, the political division of its environment greatly facilitated it .

Inter duo imperia: Palmyra as a gateway How could that happen? Which factors allowed Tadmur to benefit from political tensions? My solution to the mystery will be a model, with all its limitations: a model admittedly constructed by myself and not by any ancient source, since no direct answer to such questions can be expected from the evidence . The Palmyrenes themselves were hardly interested in the historical anomaly that was their home . To them, whatever empowered them to run their trade, seemed perfectly normal . As to the Roman observers, despite Palmyra’s economic importance, they were hardly interested in the remote oasis, except when it came to the epic clash of Aurelian and Zenobia in the aftermath of Valerian’s defeat at Edessa in AD 260 . With one exception: around AD 70, Pliny the Elder mentions Palmyra in the fifth book of his Naturalis Historia, referring to it as a city that lay privata sorte inter duo imperia summa Romanorum Parthorumque .33 Inter duo imperia, between the two empires: this was undoubtedly true in the first century BC, at the time when Pompey created the bridgehead of Syria and still when Mark Antony allegedly looted Palmyra .34 But in the second half of the first century AD, when Pliny wrote? Scholars have duly dismissed Pliny’s description as anachronistic at best, if not downright wrong . Many agree that by AD 19, when Germanicus visited Palmyra, the city was firmly integrated into the Roman province of Syria . The first assumption and the cornerstone of the model proposed here is that this view is simplistic .35 However, the 32 On Carrhae Arnaud 1998, Mattern-Parkes 2003, Timpe 1962 . 33 Plin . HN 5,88 . 34 On the episode included in App . BC 5,9, see Hekster – Kaizer 2004, Sommer 2018a, 64–68, and Kaizer in this volume . 35 Will 1992, 39–40: “Pourtant les preuves de l’annexion de Palmyre par Rome ne font pas défaut”; Hartmann 2001, 49: “Inschriften belegen, daß die Oase bereits seit Tiberius unter der Kontrolle der syrischen Statthalter stand […] . Somit war Palmyra wohl seit Tiberius, spätestens aber seit der Mitte des 1 . Jahrhunderts Teil des Römischen Reiches”; Smith II 2013, 24: “Thus this evidence, while it records direct Roman involvement in the affairs of Palmyra, also suggests the likelihood that Palmyra at the time [AD 19] had become officially part of

III . The begin nings of Palmyr a’s long-distance tr ade · 45 Palmyrenes were operating in a world of vast frontier zones, not clearly demarcated territories with neatly drawn boundaries . What matters here is not the question whether Palmyra was part of the Roman Empire, but rather to what extent it was .36 Here comes the second assumption: instead of hypothesising a formal annexation of Palmyra through Rome, I suggest that ties between the two bodies gradually became stronger, as a result of the intensity of Roman governance increasing slowly, but steadily . If the hypothesis is correct, Palmyra was, in the second and first centuries BC, a third party at a spot where the spheres of interest of two great powers — Seleucids and Parthians; Romans and Parthians — converged . Or to be more precise: it had the potential to become such a third party, but in order to act as such, it had to contain a substantial population . Here comes into play the concept of the gateway-city: a gateway city is, according to the geographer Andrew F . Burghardt, “an entrance into (and necessarily an exit out of) some area” .37 The entrance is a narrow bottleneck, which, at a given time, outcompetes potential other entrances . The gateway commands the connections between a tributary area, of undefined size, in its own hinterland and the outside world . By doing so, it unblocks an obstruction, like a port opens the sea . Gateway-cities are usually found next to economic, ecological or political shear lines . Normally, gateway-cities are located at sites “of considerable transportational significance” .38 The ideal-type has been designed, by cultural geographers, for the American frontier with its vastness of ‘empty’ space, yet it can be applied to all kinds of marginal environments where connectivity across the frontier is a resource in high demand and in extremely short supply . Its ability to provide this resource prompts the gateway city to grow at a rapid pace . The third assumption is, that Palmyra had all the potential to become a gateway-city par excellence, ecologically and politically . Ecologically, the oasis of Tadmur is the one spot that facilitates the passage between the Euphrates and Western Syria via the Syrian Desert .39 This is certainly not the shortest route, yet one of the easiest . But what exactly made Palmyra a bottleneck, outcompeting any alternative passage? The distribution of Seleucid coin hoards suggests that the trade between Mesopotamia and Syria had previously taken the route along the middle Euphrates and from there across Northern Syria via Beroia .40 Why did this change? The reason for this can only be a political one . The events of the 140s and 130s BC turned Mesopotamia into a conflict zone . The two empires and their respective combatants could

36

37 38 39 40

the Roman Empire”; Sartre – Sartre 2016, 37: “Toutefois, à l’époque où écrit Pline l’Ancien […], Palmyre a été annexée à l’Empire romain depuis au moin un demi-siècle .” Spatial dimension of empire and intensity of authority in general: Doyle 1988, 30–47, Geiss 1991, Geiss 1994, Geiss 1999, Münkler 2007, 49–58 . Frontier and empire in the Roman Near East: Isaac 1992, Isaac 1998 . For the author’s point of view see Sommer 2009, Sommer 2012, Sommer 2017b, Sommer 2018a, 106–112, Sommer 2018b, 58–62 . Burghardt 1971, 269 . Ibid ., 270 . Gawlikowski 1994, Meyer – Heldaas Seland 2016, Mior 2016, 49–50, Seland 2016, 45–55 . van der Spek 2007, 427, map 15 .2 .

46 · Michael Sommer no longer trust each other . In all likelihood, there was little scope for trade in an area like the middle Euphrates, where the two spheres of influence directly overlapped . Yet, the basic patterns of production and consumption across Mesopotamia and Syria did not change . Both regions were dependent on the exchange of goods and continued to be so . This situation opened the door for a third party: the largely nomadic, tribal groups populating the Syrian Desert . Over the cities dotting the northern route via Beroia, they had the advantage of neither side perceiving them as antagonistic . In the Near East, the involvement of nomads in the long-distance trade was anything but new: in the Fertile Crescent, it reached back at least to the Mari period in the 19th and 18th centuries BC .41 Nor was the partial sedentarisation, resulting in urbanisation, of such groups an element of innovation: again, Mari is the most instructive example . So, here is the fourth assumption: the urban growth of a pre-existing, but rather insignificant settlement at Tadmur owes its rise to the trade between Mesopotamia and Syria, as well as to the nomadic groups getting involved . Probably, some people who were attracted by the oasis and its gateway potential came from the immediate vicinity . Others may have originated from further afield, as is suggested by Palmyra’s linguistic and cultural diversity . Perhaps, some came from as far as Iran or the Mediterranean . This does not really matter . The horizon of such people was, from the outset, a lot wider than that of most urban dwellers in Classical Antiquity . And it continued to grow . Imperial frontiers for them were virtually non-existing . In particular the elites commanded social networks that reached far beyond anything the notables of an average Roman city could conceive in their wildest dreams . In the second century BC, Yarḥai’s ancestors paved the ground for a commercial empire that drew its revenues from guarding the gateway between East and West . Imperial antagonisms created an infinitely fertile biotope for such brokers of the intercontinental long-distance trade . Ultimately, however, precisely these antagonisms dealt the final blow to Palmyra . But this is a different story .

41 Klengel 1972, 49–73 .

IV. Palmyra. Marginal agriculture in a marginal landscape? Jørgen Christian Meyer Agriculture in arid landscapes with sparse precipitation unsuitable for traditional dry farming has attracted an enormous scholarly attention . This is partly due to the fact that regions in North Africa and the Middle East which, in the modern period, have been classified as ‘desert’ or ‘dry steppe’ dominated by a Bedouin strategy of survival, were occupied by agricultural settlements in the ancient period, in the Middle East even up to the early Islamic period . Agriculture in arid zones is often classified as ‘marginal’ or ‘fragile’, compared to Europe and the Mediterranean where stable agriculture with wheat and barley has been the main source of nutrition since the Neolithic . Palmyra is located in an arid zone on the Syrian dry steppe, the bādiya, and is often called the ‘Queen of the desert’, stressing the hostile environment . Regional average annual precipitation during the period from 1934 to 1966 was 125 mm, which is not enough for the cultivation of barley, which needs 200 mm . The huge flat dry steppe to the south, al-Ḥamād, has an average annual precipitation below 100 mm . The mountainous area north of the city lies within the 200–250 mm isohyet, but very often there are drier spells, 3 to 4 years on end, with much lower rainfall . The rainy season starts in October and ends in the beginning of May, with peaks in the cool winter months, but with great variation from year to year .1 Opportunistic dry farming has been practised north of Palmyra at the end of the Ottoman period .2 If the rain was sufficient and fell in the right months, the yields were good; if it failed, the seeds were wasted, so the small population of the oasis was compelled to import wheat and barley from the eastern part of Syria in exchange for salt from the salty plain southeast of the oasis .3 Traditional dry farming cannot sustain a large population in Palmyra . A successful harvest even in years with low precipitation can only be guaranteed in zones with an average of 400 mm annual rainfall, and even then there might be years with bad harvests and food shortages . The settlements on the Syrian dry steppe relied on other strategies to secure enough water . At some places, the ancients constructed qanat systems, subterranean aqueducts leading water from springs, or extracting water from aquifers in the mountains .4 This is a common method in Iran, the oases on the Arabian Peninsula and the Tarim depression in northwest China, though

1 2 3 4

Wirth 1971, 88–92 . Musil 1928b, 147 . Ibid ., 145–146 . al-Dbiyat 2009, Geyer 2009, Lightfoot 1996 .

48 · Jørgen Christian Meyer

IV.1: Drainage from the mountains north of Palmyra and the distribution of villages.

with different names (fogara, karez, falej) .5 In Palmyra, the Efqa spring is famous, but also other neighbouring springs fed the oasis through qanats .6 The settlement of Arak northeast of Palmyra had several qanats .7 All the qanats in and around Palmyra were, 5 6 7

English 1968 . Barański 1997, Hammad 2010, 25–35 . Meyer 2017, 41–43 .

IV . Marginal agriculture in a marginal landscape? · 49 however, used to water gardens with dates, olives and fruit trees, and only to a very limited degree fields with barley and wheat . In the mountains north of Palmyra there was a high concentration of villages from the Roman to the early Islamic period, on Jebel Chaar, Jebel Merah and Jebel Abu Riĝmen (Fig .  IV .1) .8 They relied on cisterns with a sophisticated system of supply channels harvesting the water from the mountainsides during showers .9 In the final publication of the Syrian-Norwegian survey north of Palmyra from 2017, I presumed that the land was owned and the villages founded by the population of Palmyra, but this notion probably has to be revised . The divine universe of the villages, as already noted by Daniel Schlumberger, differed from the cults in Palmyra itself .10 In the villages, Baʿal-Šamen, ʿAglibōl, Malakbēl and Allāth are represented on reliefs and in inscriptions, but otherwise it is Arabian gods related to the southern dry steppe like Abgal, Maʿnū, Saʿdū, Ašar and Šalmān that dominate . Abgal was the most popular, and a sanctuary on Jebel Chaar was dedicated to him .11 Inscriptions on altars, reliefs and architectural fragments are all in Aramaic, not bilingual Greek-Aramaic, as it is common in the public sphere in Palmyra . The population in the villages obviously belonged to another social sphere . I now suggest that at least some of the villages were centres for semi-nomadic groups, a phenomenon well attested in the history of the Middle East .12 They had strong social and economic connections to both the city of Palmyra and the dry steppe south of Palmyra, and they differed from other nomadic groups, among them Safaitic speaking nomads, who brought in their flocks from outside the Palmyrene territory and who were liable to special taxes on grazing rights, according to the Palmyra tariff .13 Some members of the families or clans followed the seasonal migrations with their livestock from al-Ḥamād to the mountainous area with better pastures and access to water from the cisterns in the hot season, when the vegetation on the southern dry steppe had wilted . Other members lived more or less permanently in the village, where they practised horticulture and some agriculture, as shown by pollen analysis of mudbricks, which reveal that barley was cultivated in the northern territory .14 Others may have lived more or less permanently in the city of Palmyra .15 The villages contributed to the city’s food supply, for the Palmyra tariff mentions that foodstuffs im-

8 9 10 11 12 13 14 15

Ibid ., 14–16, 28–35 . Ibid ., 26 . Drijvers 1976, Schlumberger 1951, 124–128, Teixidor 1979, 80–85 . Schlumberger 1951, 13–22 . 51–62 . Haiman 1995, Musil 1928a, 44–45, Salzman 1980 . Greek 233–237 . Fox – Lieu – Ricklefs 2005, 46, Meyer 2017, 62–69 . Krzywinski – Krzywinski 2016 . Gertrude Bell visited a sheikh in his camp with 150 tents in the Qaʾara basin in the northwestern part of Iraq in 1914 . He had his base in Kerbela, 100 km south of Bagdad near the Euphrates, where he owned gardens with palm trees, but he spent half the year on the dry steppe with his flocks and his clan . Bell 2000, 121, 121; Bell Archives, Diaries, April 22, 1914 (Newcasle University) .

50 · Jørgen Christian Meyer ported from and exported to the villages were exempt from taxation .16 Water from cisterns, however, was primarily used for human and animal consumption, as well as for watering small gardens, especially for newly planted trees in the dry season before they had become rooted in the soil,17 and the villages cannot have been the main source of nutrition for the city of Palmyra .18

IV.2: Altar in the at-Tarfa depression (2005).

When the Czech explorer Alois Musil travelled west of Palmyra in 1912, he passed the at-Tarfa depression covered with sorrel, and there he made some very important observations .19 At several places, there were traces of foundation walls of ruined houses at the edge of the depression, and he also registered an olive press . The district was uninhabited in the Ottoman period, so the houses date back to the ancient period . In the middle of the depression, he passed two monumental altars with reliefs showing a hand grasping ears of corn and a hand holding a tree, both with bilingual inscriptions in Greek and in Aramaic (Figs . IV .2 and IV .3) . The altars were dedicated at the spring equinox in AD 114 16 17 18 19

Palmyrene 109–113 . Fox – Lieu – Ricklefs 2005, 52 . Lancaster – Lancaster 1999, 142, Mayerson 1961, 36–37 . Meyer 2017, 38–41, 54 . Musil 1928b, 134–135 .

IV . Marginal agriculture in a marginal landscape? · 51

VI.3 : Relief and inscription on one of the altars in the at-Tarfa depression.

to “He whose name is blessed forever’”, Baʿal-Šamen, Lord of Heavens and bringer of rain and fertility, by the city and the treasurers of Palmyra .20 The altars show that at-Tarfa was an important agricultural district in antiquity, and that the city of Palmyra had an interest in a successful harvest . Since the 1960s, the area has been the most important agricultural district of Tadmur, producing wheat and cotton of high quality, but the modern fields and gardens are primarily irrigated by pumping up the groundwater and a high yield is dependent on the use of fertilizers and on fallowing . At-Tarfa does not fall within the 200 mm isohyet and the fields receive much less water than the northern mountainous area with the villages . Where did the ancient farmers get enough water for the cultivation of cereals? Before he entered at-Tarfa, Musil made another important observation . When he crossed one of the wadis at the edge of the depression, he registered “ruins of several old dams designed to hold back the runoff and prevent the soil from being washed away” .21 Successful agriculture at at-Tarfa was obviously dependent on extra water from the surrounding territory to supplement to the annual precipitation, and the source for this is to be found in the mountains north of Palmyra . All the drainage from the southern part of the mountains runs towards at-Tarfa, before it flows through Palmyra and Wadi al-Qubur towards the salty plain, Sabkhet al-Mouh, 20 Meyer 2017, 45–46, Pillet 1941 . 21 Musil 1928b, 134 .

52 · Jørgen Christian Meyer

VI.4 : Small wadi leading to at-Tarfa after a short shower in the Jebel Abyad area (2009).

southeast of the oasis (Fig . IV .1) . The catchment area is huge, covering about 1400 square kilometres . If the annual precipitation is 200 mm, it will receive 280,000,000 cubic meters of water . Some of the water will be absorbed in the ground and in the wadi beds, but the surface of the dry steppe is generally very hard and impervious, baked by the sun and often reinforced by a thin layer of grassroots below the surface . Even during short showers with only a few mm of precipitation, much of the runoff converges into the wadis, where it can develop tremendous momentum, rather than penetrate the surface (Fig .  IV .4) .22  The  floods  in the wadis form the basis of a special agricultural technique called floodwater  farming or runoff water agriculture .23 By constructing rows of walls across the wadis, as mentioned by Musil, the farmers try to control the floods, letting the moist sediments settle between the walls, ready for cultivation . This type of agriculture has been practised until recently in some wadis northwest of Palmyra (Fig . IV .5), and even by the Bedouin in tributary wadis of Wadi al-Miyah and Wadi al-Swab in al-Ḥamād beyond the 100 mm isohyet .24 Tributary wadis are preferred, as it is impossible to control 22 Meyer 2017, 36–37 . 23 Gilbertson – Hunt 1996, Gilbertson – Hunt – Gillmore 2000, Mayerson 1961, 23–31, Rosen 2007, 161–167, Wilkinson 2003, 170–172 . 24 Tucker 2009, 5 .

IV . Marginal agriculture in a marginal landscape? · 53

IV.5: Former floodwater farming at Hwesys in the valley of Sorrate ech Chéfé 60 km northwest of Palmyra (2006).

the torrents in the main wadis . According to local informants, some Bedouin were until recently engaged in the cultivation of barley and wheat of high quality in the branches of Wadi Abyad north of Palmyra, while avoiding the main wadi bed . Cross-wadi walls are  registered at several places north of Palmyra, but only a few have survived from antiquity to the present .25 If the walls are not maintained, they will be washed away by the torrents . It can also be difficult to decide whether they were retaining walls for fields or for gardens . How productive was floodwater farming around Palmyra in antiquity? According to Pliny, Palmyra was famous for the richness of its soil and its agreeable springs, while its fields were surrounded by the desert .26 Pliny’s description is very general, and we do not know if he and his source meant fields outside the oasis, or fields in the oasis watered by the springs . According to Jibrail S . Jabbur, who came from the small city of al-Qaryatayn 100 km southeast of Palmyra, the Palmyrene area is one of the most fertile districts in Syria, if enough water is available:

25 Meyer 2017, 40 . 125 . 26 Plin . HN 5,88 .

54 · Jørgen Christian Meyer Conditions in the domain of the Syrian Desert differ from one part to another, but generally speaking, it is possessed of very good soil, especially in the volcanic regions around Hawran, in the outlying districts of al-Qaryatayn, Tadmur, and Jazira, and in the heartlands of the Hamad . I have seen lands subject to flooding by torrential streams that benefit some people engaged in agriculture by making their lands more productive than any irrigated lands of the finest agricultural districts .27

Eugen Wirth has confirmed the potential fertility of the soil in the Palmyrene area .28 At alQuaratayn in Wadi Raouda wheat fields produced a yield of 60–100 fold in some optimal years . In dry years, however, the yield was as low as 2–4 fold, but he gives no information about the agricultural technique employed . Foodstuff was, however, imported from outside into the Palmyrene territory in antiquity . According to the old section from the beginning of the first century AD in the Palmyrene tariff, foodstuff was liable to a tax, regardless of whether it was imported from outside the borders or exported, but unfortunately without specifying the goods .29 In the new section of the tariff from AD 137, one denarius per trip was exacted for each camel load of wheat, wine or chaff .30 Does this imply that food production in the Palmyrene countryside was insufficient to feed the population, and that the city was dependent on an extensive import of grain to sustain them, as was the case in the late Ottoman period,31 or was it due to temporary food shortage in Palmyra owing to a failed harvest? The tariff does not mention grain as such, σῖτος, or barley, κριθή, only wheat, πῦρός/ḥṭʾ . We may presume that the new section of the tax law was very precise in terminology . Was wheat imported as a supplement to the local production of barley? The new entry does not specify whether wheat or wine was imported or exported in relation to the borders, as was the case with the foodstuff in the old section . Was wheat also exported, for example, to the southern nomadic groups on al-Ḥamād, where cereals made up an essential part of the diet of the nomadic population? It is also surprising that chaff, ἂχῠρον/ḥmrʾ, is included in the entry . Chaff is important both as fodder for animals and for building and maintaining mud brick houses, but it is highly unlikely that chaff left after threshing was imported to Palmyra from outside its borders all the way from the eastern part of Syria or from the region around Euphrates . It must be a local produce from the harvesting season closer to the city, and taxed when it entered the city . The evidence from the tariff is thus ambiguous .

27 28 29 30

Jabbūr 1995, 63 . Wirth 1971, 441 . Greek 187–189, Palmyrene 109 . Fox – Lieu – Ricklefs 2005, 45 . 52 . Ὁ αὐτὸς πρά[ξ]ει γόμου πυρικοῡ, οἰνικου, ἀχύρων καὶ τοιούτου γένους ἑκάστου γόμου καμηλικοῡ καθ’ ὁδὸν ἑκάστην Ӿά (Greek 89–91) . [y]gb’ mks’ lṭ‘wn’ dy ḥṭ’ wḥmr’ wtbn’ w[k]l mdy dm’ [lhwn lk]l gml l’rḥ ḥd’ d l (Palmyrene 59–60) . Fox – Lieu – Ricklefs 2005, 41 . 49 . 31 Ruffing 2016, 193–194 .

IV . Marginal agriculture in a marginal landscape? · 55 Fortunately, we have other data for the productivity of floodwater farming in the arid zones of the Middle East in antiquity . During the excavations of Nessana in the Negev desert, on the border between Israel and Egypt, two archives from the 6th and 7th centuries were found in two churches . Apart from religious and literary texts, they contained about 200 papyri dealing with daily life in the settlement: accounts, requisition of foodstuff, contracts, letters, legal documents and receipts .32 Nessana was founded in the Hellenistic period, the Romans built a garrison at the site and in late antiquity, it became a flourishing Christian center with three churches and a population of perhaps 1500 inhabitants .33 In the Islamic period, it came under Umayyad control, and it was abandoned perhaps in the ninth or the early tenth century .34 Nessana lies at the northwestern edge of the Negev highland, and several wadis pass through the territory before they enter the sand dunes southeast of the Gaza . Like Palmyra, Nessana relied on the concentration of water in the wadis flowing from the highland in the rainy season to employ floodwater farming .35 The farmers constructed terrace walls across the wadi beds in the tributary wadis to control the floods in the rainy season . They also constructed diversion dams, leading water from the main wadis where the torrents are strong to fields and gardens above the wadi bed .36 The papyri show that they cultivated wheat, barley, grapes, olives, figs and dates . One of the papyri gives us a list of the exact amount of seeds sown, and the exact amount of wheat and barley reaped at different locations near the settlement .37 The yield for barley ranges between 1:8 and 1:8 .7 from two locations; the yield for wheat between 1:6 .75 and 1:7 .2 from three locations .38 From comparative evidence from modern Bedouin agriculture in the same area, this seems to have been fairly close to an average yield .39 The papyri also show two other interesting things . First, wheat seems to have been the dominant crop at Nessana . Wheat needs at least 250 mm of water and is much more vulnerable to drought than barley . Second, the Umayyad governor in Gaza regularly required wheat and oil from Nessana in the end of the 7th century .40 This indicates a stable agricultural regime with the ability to generate a surplus . This is very surprising, as the average annual precipitation at Nessana and the mountains in the hinterland is 100 mm, compared to the 200–250 mm in the mountains north of Palmyra . Even in years with very low precipitation in the highlands, modern Bedouin have been able to obtain a good yield in the wadis .41 This emphasises the fact that the assessment of the agricultural potential of arid lands should not be based on the 32 33 34 35 36 37 38 39 40 41

Kraemer 1958 . Magness 2003, 177, Mayerson 1961, 19 . Magness 2003, 177–185 . Mayerson 1961, 12–14 . 23–31 . Ibid ., 31–36 . Kraemer 1958, 237–240 (document 82) . Mayerson 1961, 17–18 . Ibid ., 18 . Kraemer 1958, 175–197 (documents 60–67) . Mayerson 1961, 30 .

56 · Jørgen Christian Meyer amount of the annual precipitation, but rather on the ability of the landscape to concentrate the runoff in the wadis, the size of the catchment area and the intensity of the showers . The yields stated in the Nessana papyri are surprisingly high compared to Europe and the Mediterranean in the preindustrial age . In Denmark, the yields in 1837 for barley ranged from 1:3 to 1:8 with an average from 1:5 to 1:6, and that was after the agrarian reforms in the end of the 18th century, that increased the productivity of the fields .42 These figures seem to fit well with the scanty evidence from the rest of Europe in the preindustrial age .43 From Cicero, Columella, and Varro we have an idea of the seed yields from the Mediterranean in the Roman period . The city of Leontinoi on western Sicily is located in a very fertile volcanic region south of Etna, and early series of coins issued by Leontinoi show the head of a lion surrounded by grains of barley stressing agricultural fertility .44 According to Cicero, the yield under favourable conditions was 1:8, in some years even 1:10, but that was very seldom .45 Varro mentions yields in some parts of Etruria as 1:10, in some

IV.6: Harvest on field with high tiller production. Ash Sholah, 30 km southwest of Deir al-Zor (2004).

42 43 44 45

Rømer 2004 . Erdkamp 2005, 38–40, Sallares 1991, 375 . Boehringer 1998, 43–53, table 10 . Cic . Verr. 2,3,112–13 .

IV . Marginal agriculture in a marginal landscape? · 57 instances 1:15,46 but this is obviously not normal for the Mediterranean as such . Columella has a more conservative figure for larger parts of Italy, 1:4 .47 A yield of 1:3 is regarded as a minimum to sustain human life . The productivity of the fields has certainly varied a lot from region to region due to differing fertility and management of the soil, but the yields from Nessana obviously fall in the upper end of the scale .

IV.7: Harvest on field with low tiller production. Çatak, southeastern Turkey (1987).

The actual yield is determined by the tiller production from the main stem and the survival of tillers developing spikes with kernels .48 Good tillering also means that you can reduce the amount of seeds, even to the half (Fig . IV .6) . If the soil has low nutrition, is exposed to drought or too much water, the tiller production and their survival will be

46 Varro rust. 1,44,1 . He also mentions districts with a yield of 1:100 around Sybaris in Italy, Gadara in Syria and Byzacium in Africa . Pliny mentions 1:100 in Leontinoi, the whole of Andalousia (sic!) and Egypt (Plin . HN 18,21) . It most probably means very fertile, but according to Eugen Wirth, comparative high yields have been registered at al-Quaratayn southwest of Palmyra in exceptional years (Wirth 1971, 441) . See the discussion in Sallares 1991, 376–378 . 47 Colum . 3,3,4 . 48 I am indebted to Prof . Ole Gjølberg at the Norwegian University of Life Sciences, Ås in Norway, for information about tillering and the importance for the cereals output .

58 · Jørgen Christian Meyer

IV.8: Seeder plough, Red Sea coast, Egypt (2003).

hampered, or they will not develop spikes (Fig . IV .7) . In Europe and the Mediterranean, fallowing and manure are essential to maintain the fertility of the soil . In floodwater farming, moist silt replaces the soil every year . The runoff also contains new nutritious, finegrained material from the frequent dust storms in the open landscape . In contrast to fields irrigated with water from dams and reservoirs exposed to evaporation, there is no salinization of the fields . The water and the silt brought in during the floods have low salinity, and the floods wash out the salt accumulated in the soil from the previous season . The sowing pattern is also important for the size of the yield . The plants compete for nutrition, and if the seeds are sown too densely, tiller production and their survival will be low . In Europe and the Mediterranean, broadcast seeding is the normal method, and this makes heavy demands on the skill of the sower . In the Middle East, the seeder plough, which enables much more control over the spacing of the seeds, is known already in the Early Dynastic period,49 and it has been used until recently in Egypt outside the Nile valley (Fig . IV .8),50 but we do not know if it was employed in the classical period in Negev or Palmyra . 49 Potts 1997, 78–79 . 50 When I visited the Red Sea coast and Sinai in 2003, the traditional seeder plough was a common sight at many farms, but they had not been in use for several years .

IV . Marginal agriculture in a marginal landscape? · 59

IV.9: Flooding in Tadmur (May 2011).

All types of agriculture are vulnerable to recurrent extreme changes in the weather pattern . In central and northern Europe, heavy floods during the summer may reduce the yield . In the Mediterranean, drought poses the greatest menace . Food shortage has been a common phenomenon in European history, sometimes leading to famines, very often due to inefficient public institutions unable to deal with the crisis or speculation in the grain market .51 At the end of the 18th century, Aleppo in Syria experienced recurrent periods of food shortage and even famine .52 Agriculture and horticulture in arid lands based on proper floodwater technique are much more resistant to variations in annual precipitation, which, in areas with rainfed agriculture, means the difference between a successful harvest and a failed crop .53 A reduction of precipitation from 200 mm to 100 mm in the mountains north of Palmyra does not mean a proportional reduction in the water available in the fields, as only part of the runoff is utilized in years with normal precipitation . It depends on how the rainfall occurs . Insignificant rainfall evenly distributed over the rainy season will not generate sufficient momentum in the floods to bring the water over long distances, and extreme drought will also have severe consequences for food production, 51 Alfani – Ó Gráda 2017, Garnsey 1988, 8–39 . 52 Marcus 1989, 123–124 . 53 Rosen 2007, 45, Wilkinson 2003, 172 .

60 · Jørgen Christian Meyer but even a short shower in the mountains north of Palmyra will fill the wadis leading to the at-Tarfa depression . Paradoxically, the greatest threat to floodwater farming is water itself . During heavy or long showers, the floods in the wadis gain tremendous momentum, carrying not only silt, but also stones . The wadis can change their course and the landscape overnight, and even destroy modern asphalt roads and concrete bridges across the wadis . Tadmur has experienced frequent flooding, and in 1993, a dam was constructed across Wadi al-Qubur west of the Valley of the Tombs to protect the ruins and the city .54 Notwithstanding these precautions, Tadmur was flooded in May 2011 and all traffic in the surrounding countryside obstructed for several days (Fig . IV .9) . Floods may wash away fields with planted grain, even in the tributary wadis, and if they occur in May, may destroy the crop immediately before the harvest .

IV.10: Wadi Abyad basin with modern dam (2008).

To minimize the risk, the ancients built dams, not to store the water for the hot season as an irrigation reservoir, but to regulate and control the water flow as a barrage . In the Wadi-Abyad basin north of Palmyra, a series of canals from the western mountainside to the main wadi bed leads the water to a shallow depression behind the modern dam constructed in 1984–1986, called Bîr el Hafîré by Musil, as on old French maps (Figs . IV .1, IV .10

54 I am indebted to Waleed al-Asʿad, former director of the museum in Palmyra, for the information .

IV . Marginal agriculture in a marginal landscape? · 61

IV.11: Canals in the Wadi Abyad basin.

and IV .11) .55 ‘Hafir’ in Arabic means ‘water reservoir’ or ‘dam’, and it is highly probable that there have been an ancient forerunner, now covered by the modern dam . The catchment area behind the dam is 600 square kilometres and all the drainage runs towards the at-Tarfa depression . The huge ancient Harbaqah dam, 69 km southwest of Palmyra near Qasr al-Heir al Gharbi, probably had the same function, but its exact date and function is disputed .56 There have also been much smaller dams, as mentioned by Musil,57 and as registered close to the settlements in the mountains (Fig . IV .12),58 which have controlled the normal flooding, but most of them have been washed away due to lack of maintenance . There were, of course, natural limits on how much land the farmers could bring under the plough, depending on the size of the catchment area, the drainage and the wadi systems, but grain production was of high quality and very resistant to variations in yearly precipitation, and, in the longer chronological perspective, also to climatic fluctuations . Some scholars have tried to explain the abandonment of settlements in the arid zones of the Middle East and North Africa in terms of climatic changes and consequent desertification,59 but this does not make sense .60 The expansion of agricultural settlements 55 56 57 58 59 60

Meyer 2017, 39–40 . 91–93, Musil 1928b, 148; French map: Palmyre . Levant 1:200 000, N 1 37-XV (1944) . For a discussion with references see Meyer 2017, 46–54 . Musil 1928b, 134 . Meyer 2017, 199–200 . Issar – Ginat – Zohar 2011, Issar 2003, 25–28, Issar – Zohar 2004, 213–217 . Gilbertson 1996, Haiman 1995, Rosen 2007, 150–171, Rosen 2000 .

62 · Jørgen Christian Meyer

IV.12: Remains of dam across small wadi, Jebel Merah (2011).

in the Negev desert seems to coincide with a relatively dry period, and the final abandonment in the 10th century with a climatic amelioration .61 The abandonment of stable agriculture in the Palmyrene hinterland was of course partly due to the decline of Palmyra as an important center on the Syrian dry steppe and the dwindling of the population in the oasis after the Umayyads, but one question remains . Why didn’t the population in Tadmur continue growing cereals, instead of importing them from the eastern part of Syria in exchange for salt from the salty plain southeast of the oasis? The answer to this is to be found in the relationship between Tadmur and the nomadic groups of the dry steppe . In the Roman period there were close social and economic ties between Palmyra and the nomadic groups .62 The Palmyrenes descended from these groups and the seminomadic population in the mountains north of Palmyra were an integrated part of the city . There were no conflicts between the desert and the sown . After the fall of Zenobia these ties disappeared, but the Romans continued to have control of the area by constructing a series of forts along the Strata Diocletiana .63 In the Byzantine and Umayyad periods, powerful tribes on the dry steppe played a significant role, but 61 Avni – Porat – Avni 2013, Magness 2003, 215–216, Rosen 2000, 56 . 62 Sommer 2005b, 170–183 . 63 Butcher 2003, 416–421, Gawlikowski 1997 .

IV . Marginal agriculture in a marginal landscape? · 63 they were an integrated part of the political landscape at the edge of the great empires .64 The Ottomans, however, lost control of the dry steppe, where several tribes were then competing for power and territory .65 When Alois Musil visited Palmyra in March 1912, he gave the following description of the relationship between the desert and the sown at the settlement of Arak, watered by several qanats mentioned above: All of the fifteen huts of the hamlet of Arak were deserted . The inhabitants had suffered much from the Bedouins camping in the neighborhood and still more from numerous raiding bands; therefore, as they always do under such circumstances, they had moved in a body with their supplies to Tudmor . The Turkish Government some time in the seventies had ordered a strong barrack to be built halfway between the hamlet and the springs, with five gendarmes for a regular garrison; but this was now deserted, because the Bedouins only made fun of it .66

From a political point of view, Palmyra was located in a marginal zone on the edge of empires, but from an agricultural point of view, the Palmyrene area was not marginal, if the farmers and the community invested time and resources on the construction and maintenance of retaining walls and large and small dams across the wadis . This presupposes a well-organized society and peaceful relations between the city and the nomadic groups on the dry steppe . The fields undoubtedly had high productivity, but we do not know if the Palmyrenes also regularly imported cereals from outside their borders as a supplement to the local production, not just in periods with food shortage .

64 Genequand 2012, 30–31, Intagliata 2018, 97–108 . 65 Lewis 2000, Marcus 1989, 140–144, Reilly 2005 . 66 Musil 1928b, 85–86 .

V. Palmyrene merchants and the Red Sea trade Matthew A . Cobb International and inter-regional trade has long been recognised as a major facet of Palmyrene economic life, which has led many to designate Palmyra a “caravan city” .1 The city’s location by the Tadmur Oasis near the edge of the Syria Desert left it quite wellplaced to act as a conduit between Mesopotamian-Persian Gulf trade networks and the markets of the Eastern Mediterranean .2 This commercial activity is evidenced by a rich array of material finds, such as pearls, agate/carnelian beads and silks (particularly from tombs), as well as by a few dozen inscriptions that refer to the successful arrival of camel caravans .3 These inscriptions, in particular, underlie the importance of connections with sites in central (Babylon, Seleucia, and Vologesias) and southern Mesopotamia (Spasinou Charax and Forat), where imports from India, southern Arabia and beyond could be acquired .4 A few of these inscriptions even indicate that some Palmyrene merchants sailed directly to regions like “Scythia” (northwest Indian subcontinent) .5 Given the importance of the traditional Mesopotamian-Persian Gulf trade networks, it may perhaps be surprising to find that some Palmyrene merchants eschewed these tradi1 2

3

4

5

See for example, Gawlikowski 2016a, 25, Millar 1998, Rostovtzeff 1932a . Seland 2016, 3, notes that the designation of Palmyra as a caravan city remains largely unchallenged . For a description of the Syrian Desert region see Seland 2015, 46 . He notes that the characterisation of the region as a desert is somewhat misleading, since it is more of a dry-steppe with a fair amount of winter rains that could sustain some vegetation . For discussion of the material and artistic evidence attesting to the range of imported products see Gawlikowski 2016a, 22 . 35–36, Seland 2013, 66 . 71, Seyrig 1950, 2, Żuchowska 2013, 383–384 . Seland 2013, 71, also suggests that Palmyrene merchants who sailed to northwest India likely brought back the types of products that the author of the Periplus mentions could be acquired in that region . These connections to sites in central and southern Mesopotamia seem to primarily relate to the Persian Gulf branch of the Indian Ocean trade . Millar 1993, 330, argues that there is little evidence for involvement with the overland “Silk Road(s)” trade (see also Andrade 2018a, 200–201, Gawlikowski 2016a, 23 . 29); although on this possibility see Seland 2013, 69–70 . For a discussion of Palmyrene communities at sites in Babylonia, southern Mesopotamia, and also further into the Persian Gulf (Bahrain) see, Brokaert 2017, 7, Gawlikowski 1996, 142–143, Sartre 2005, 269, Schörle 2017, 149, Seland 2013, 69–70, Żuchowska 2013, 381–382 . PAT 1403, PAT 2763 . For these inscriptions see Andrade 2018a, 4–5, Schwartz 1960, 31, Seland 2013, 70, Sommer 2015, 173–174 . For a boat on a funerary relief of Iulius Aurelius Marona (tomb no . 150) dated to AD 236 see Sartre 2005, 269, Schörle 2017, 150, Seland 2013, 70, Sidebotham 2011, 203 . Delplace 2003, 158–159 . 166–167 has argued that the reconstructed word Choumana, which appears in the form ([Χ]ου[μ]ανων) on the Greek version of a bilingual inscription of Marcus Ulpius Yarḥai: Inv . X .87 (Palmyrene Aramaic), 88 (Greek) = PAT 0306, should be understood as an allusion to the land of the Kushans, although this interpretation is not universally shared: for example, Seland 2016, 39 no . 251, prefers its identification with a site in Mesopotamia .

66 · Matthew A . Cobb tional routes in favour of the Egyptian-Red Sea branch of the Indian Ocean trade . The aim of this paper is to explore why this might have been the case . In order to do this, the discussion is split up into three main sections . The first considers the available evidence for Palmyrene participation in the Egyptian-Red Sea trade routes and whether it is possible to identify any major phases for this activity . The second section considers whether it is possible to connect the evidence for Palmyrene presence at a few sites in southern Arabia and on the island of Socotra with activity via either the Egyptian-Red Sea or MesopotamianPersian Gulf trade routes . The third section analyses the potential “push” and “pull” factors which may have encouraged participation in the Egyptian-Red Sea trade . The plausibility of the following suppositions is considered: that Roman-Parthian/ Sasanian conflict may have made the Mesopotamian-Persian Gulf route more difficult to use; that Palmyrene military presence in the Egyptian Eastern Desert in the latter second and early third century AD may have encouraged some merchants to move to the region; and whether the Egyptian-Red Sea branch offered greater opportunities for profits . Given the limited quantity and nature of the evidence no attempt is made here to prove any of these suppositions beyond doubt . Rather, the intention is to outline the limits of the evidence and assess the extent to which we can support particular claims .

Palmyrene merchants in Egypt The most important evidence attesting the presence of Palmyrene merchants and shippers in Egypt consists of two inscriptions, one connected to Koptos (modern Qift), and the other to Tentyris (modern Denderah) .6 The former inscription comprises a laudation of Zabdalas, son of Salmanos, by fellow members of the community of Red Sea merchants from Hadrianē Palmyra . This was done to celebrate the fact that Zabdalas had spent his own funds on the construction of προπύλαια (a gateway), three στοαί (porticos), and θυρώματα (chambers) from his own funds .7 The inscription is normally dated to the midlate second century AD, the reference to ‘Hadrianē Palmyra’ giving it a terminus post quem of AD 129 (the potential significance of the dating of this inscription is discussed in section 3) .8 Reference to the collective celebration of Zabdalas by his fellow Palmyrene merchants, and to communally owned property, has reasonably been interpreted as evidence for a cultic/community association — perhaps something akin to a collegium .9

6 7 8 9

For an overview of the evidence see Table 1 . See Map 1 for the sites mentioned . I. Portes 103 = AE 1912, 55 no . 171, Schörle 2017, 151–152, Young 2001, 80–81 . PAT 0305 = IGLS XVII .1,245; see also PAT 0259 . Andrade 2013, 177, Sidebotham 1986, 95–96 . See Evers 2017, 129, Grout 2016, 259–260, Seland 2016, 41–42, Sidebotham 1986, 95, Smith II 2013, 161 . For a possible comparison with these kinds of communal activities among the Palmyrene community at Vologesias, see an inscription dating to AD 108 – Inv . IX .15 = CIS 2 .3917, Terpstra 2016, 43 . The associated find of 12 stelai had in the past been assumed to connect with this Palmyrene community association, because it was sup-

V . Palmyrene merchants and the Red Sea tr ade · 67 The second major inscription that seems to attest to Palmyrene participation in the Egyptian-Red Sea branch of the Indian Ocean trade was found at Tentyris, about 5 kilometres from ancient Kainopolis (modern Qena) .10 The fragmentary bits of text that survive are in Greek, but it is still possible to make out Palmyrene script on the left-hand side, indicating that it was originally a bilingual dedication . It records a Iulius Aurelius, who either had the cognomen or patronymic Maqqay/Makkaios . He seems to have died at the site, being buried with a modest tablet to honour him . Mention is also made of merchants (ἔμπο[ροι]) and possibly of a caravan (συ[νοδίαν]), though the latter reconstruction is hardly conclusive .11 The inscription is usually dated to between AD 160 and 212 (again the significance of this dating will be considered in section 3) .12 Seland has observed that Tentyris was near to the terminus of the road that led from Abu Shaʾar on the northern Red Sea coast to the Nile . He notes that the port may have offered a convenient crossing point for a short sail over to the port of Leuke Kome (on the northern Arabian coast of the Red Sea), and from their an overland crossing on the Via Nova Traiana would allow them to reach Syria .13 It is not beyond the realms of possibility that this group of Palmyrene (merchants) was travelling to Abu Shaʾar . Indeed, a small station at Abu Shaʾar el-Qibli (approximately 4 .5–5 kilometres west of the larger Tetrarchic fort) seems to be connected to the Kainopolis-Abu Shaʾar route from at least the second century AD, allowing for the potential of some kind of northern Red Sea-Nile commercial activity .14 However, a few issues should be noted . Much of the traffic on the roads leading from Kainopolis seem to relate to quarrying activity at Mons Claudianus and Mons Porphyrites (especially in the second century AD), while the evidence for wouldbe commercial activity is largely inferred . Secondly, Tentyris is on the west bank of the Nile, unlike Kainopolis, which would seem to make it a less convenient starting point for crossings into the Eastern Desert . Thirdly, if this group was travelling from Tentyris to Palmyra, this trip should not be understood as part of regular direct commercial activity, since this seems less economically viable (Seland does not imply this was the case) .15 Evers

10 11 12

13 14 15

posed that the figures were in typical Palmyrene frontal pose and (alongside frescoes on the walls and two altars) is usually cited in support of this view . For example, Schwartz 1960, 30, interpreted some of the stelai as representing Palmyrene archers (see also Smith II 2013, 161–162) . However, the interpretation of this art as Palmyrene has been disputed . For example, Young 2001, 81 notes that a number of scholars have argued that the stelai are all of Egyptian workmanship . PAT 0256 = I. Portes 39 = CIS 2 .3910 . For discussion of the inscription see Schörle 2017, 152, Seland 2016, 42 . For dating see Sidebotham 1986, 96, Young 2001, 81 . See Smith II 2013, 162, who notes that ‘[s]ince most Palmyrene inscriptions that bear the same imperial nomina date to the early third century A .D ., we may assume that this ones does as well .’ Seland 2016, 42–43 . On Abu Shaʾar and this route see Sidebotham – Hense – Nouwens 2008, 53–60, Sidebotham – Zitterkopf – Riley 1991, 571–622 . That is to say, goods imported into Egypt via the Red Sea ports would be taken to Alexandria for taxation  and then potential redistribution elsewhere in Egypt or in the wider Mediterranean world .

68 · Matthew A . Cobb has suggested that the Tentyris inscription may attest to ‘Palmyrene caravan ventures in the Eastern Desert’, suggesting that they were ‘carrying specifically Palmyrene products into Egypt’ .16 It is not clear, however, what specifically Palmyrene products would be worth bringing to Egypt via such a route .17 The most logical reason for a Palmyrene presence at this site is that suggested by both Seland and Schörle .18 They note that Tentyris was an important stopping point for traffic heading south, where it might be necessary to wait for better wind conditions to allow vessels to sail against the current (i .e . upstream) .19 Schörle suggests that a Palmyrene presence at this site could be seen to parallel the Mesopotamian practice of having diaspora set up at key centres on the Euphrates to help facilitate the wider commercial interests of the community .20 However, the inscription may not even attest to a more permanent Palmyrene merchant presence at Tentyris, but could just simply indicate a group of Palmyrene merchants that happened to be travelling between Alexandria and Koptos (possibly after paying the tetarte, selling their goods, and acquiring new ones for export), and due to reasons entirely lost to us, Iulius Aurelius passed away on the journey . The fact that this inscription was written on a modest tablet, rather than a more substantive commemorative funerary monument, may lend weight to this interpretation .21 This is speculative, of course, but no more so than any of the other theories proposed . The Koptos and Tentyris inscriptions are not the only evidence for Palmyrene connection to the Egyptian-Red Sea trade networks . The port of Berenike (located on Foul Bay, several kilometres south of Ras Banas) which, alongside Myos Hormos (modern Quseir al-Qadim), was one of the primary commercial sites on the Egyptian Red Sea coast, has revealed evidence for Palmyrene presence . Palmyrene Aramaic is one of the numerous

16 17

18 19

20 21

Likewise, goods brought to Palmyra (which were not consumed in the city) would be transported to other sites in  the  eastern  Mediterranean . There is no reason to think that goods already imported (and taxed at 25%)  in  Egypt or Syria, would then be transported via a lengthy overland, sea, and overland (again) route between  Tentyris to Palmyra given the added costs involved — this would be a journey well-exceeding 1,000 kilometres . Such a journey would also entail exiting and re-entering Roman external customs posts . Evers 2017, 130 . For a critique of the notion that ‘such an extraordinarily circuitous route was ever in use’, see Young 2001, 81 . Żuchowska 2013, 384, suggests that on the traditional Mesopotamian-Persian Gulf route, Palmyrene merchants may have exported purple murex dyes, and raw and worked glass available from the Eastern Mediterranean . It seems likely, however, that such products could more cheaply be brought to Egypt via Alexandria . Schörle 2017, 152, Seland 2016, 57 . On the issue of the Nile’s currents and summer wind patterns see Cooper 2014, 125–142 . He notes that the intensity and duration of the northerly winds could vary and that consequently sometimes crews had to resort to using oars, punts or tow ropes to fight against the currents . With regard to Denderah, Cooper, ibid ., 130, observes that the ‘mean wind speeds maintain a much smaller positive differential above the Nile’s mean current speed throughout the year .’ Essentially making sailing conditions much more difficult . Schörle 2017, 152 . On the modest nature of the fragmentary tablet from Tentyris see Seland 2016, 42 .

V . Palmyrene merchants and the Red Sea tr ade · 69 types of scripts found at the site (alongside Greek, Demotic, Egyptian hieroglyphs, Hebrew, Latin, Hadramawti, Tamil-Brahmi, Prakrit, and Sanskrit) .22 Moreover, a cultic structure, referred to by the excavators as the Shrine of the Palmyrenes, was established around the late Antonine or Severan period (on top of earlier Ptolemaic/early-Roman strata of no religious significance) .23 The shrine was evidently used by Palmyrene soldiers as evinced by two inscriptions (discussed in more detail in section 3), although its use by merchants as well seems highly probable . Some of the ostraka found at Berenike may also open up the possibility of Palmyrene presence in the first century AD . However, these texts do not necessarily provide clear-cut proof of Palmyrene commercial activity in the Red Sea at this time . The name Hierabole (Ἱεραβολή), probably a feminine variant of the masculine Hierabolos, and seemingly connected to the Palmyrene deity Hierobol/Yarḥibōl, appears on one of mid-first century AD ostraka from Berenike . It is unclear from the text if the individual is a merchant or someone engaged in another profession .24 The mid-first century Berenike ostraka also reveal a number of other broadly Semitic names (though some of these are in doubt), but it in most cases any obvious Palmyrene connection, as opposed to some other SyrianLevantine ancestry, is not easily distinguishable .25 Four of the Berenike ostraka dating to around the third quarter of the first century AD appear to contain one or more Semitic scripts, of which Palmyrene cursive script is a possibility;26 one ostrakon faintly shows the names ʾAbgar and Abū Magdi or Malki .27 At Myos Hormos a fragment of a Palmyrene amphora has been found, on which appears an ink dipinto of Palmyrene Aramaic cursive script . It is unclear whether the text indicates an owner’s graffito, or its origin, weight, or content .28 The amphora shows parallels with second-third century vessels found at Palmyra but probably dates to the late first or 22 Sidebotham 2007, 164 . 23 On this structure and finds associated with it see Sidebotham 2011, 64–65, Sidebotham 2014, 611–613 . It should be noted that this site was not exclusively connected to the Palmyrene deity Yarḥibōl/Hierobol . Evidence relating to other divine figures like Harpokrates, and objects that may relate to practices connected with Dionysos or the mysteries of Isis, have been found . 24 O. Ber. 97, Bagnall – Helms – Verhoogt 2000, 27 . 30 . 65 . The editors suggest that the possible reference to denarii (δηνάρια) in the text may indicate a military or imperial connection . 25 For Berenike ostraka with Semitic names see O. Ber . 137 – 140a? . 183 . 189? . 198? . – see Bagnall et al . 2005 . A few Semitic names (Bargates, Dosarion, Mambougaios, Thaimos?, and Zaneos) appear in the ostraka from the 2009–2013 seasons, although in the case of Bargates, Dosarion and Zaneos it is thought that these are probably Nabataean, while it is uncertain if Thaimos is actually a Semitic name — see Ast – Bagnall 2016, 12 . In the case of the 1996–1998 seasons, besides Hierabole, the possibly Semitic names Rhobaos and Chennas appear, but in the case of the latter two any obvious Palmyrene connection is not apparent, see Bagnall – Helms – Verhoogt 2000, 27 . 26 O. Ber. 254–257, Bagnall et al . 2005, 104–105 . See also Sidebotham – Zych 2012, 38, for an ambiguous wooden tag piece with Greek text on one side and some Semitic language on the other . 27 O. Ber. 256 . 28 Evers 2017, 129, Tomber 2011, 7 .

70 · Matthew A . Cobb early second century AD .29 Overall the evidence for potential first to early second century AD Palmyrene activity at Myos Hormos and Berenike is limited . Absence of evidence is not evidence of absence, but as it currently stands the available material seems to point to a far greater level of Palmyrene commercial activity in Egypt from the mid-second to third century AD . What, if any, links Palmyrene merchants and shippers based at Koptos had with other prominent Palmyrenes based in Alexandria, or indeed, in Palmyra remains unknown .30 Seland muses on the possibility that someone like Firmus, who is reported in the Historia Augusta as being an ally of Zenobia and undertaking a failed rebellion in Egypt in AD 273, might represent the sort of powerful patron and financier that Palmyrene merchants based in Egypt could have relied on . This Firmus is said to have been a native of Seleucia (probably from Syria), who had built his wealth on trade with India and had maintained good relations with the Blemmyes and the Saracens .31 This premise is not unreasonable, especially as we know that influential members of the Alexandrian economic and political elite (and probably also wealthy investors from Italy and elsewhere) financed Indian Ocean trade ventures .32 Moreover, as Seland notes in his chapter for this volume, highly successful merchants appear to have become more prominent in Palmyrene society by the second to third centuries AD . Unlike the traditional “warrior aristocracy” who were bound by familial and friendship links to nomadic groups in the Syrian Desert region, extremely wealthy merchant families may have been more willing to set themselves up in Egypt as merchants or financiers .33 Of course, this is reasoned speculation . Unfortunately, the fact remains that we do not know for sure whether such links existed, even if we suspect that they might have .

Palmyrenes in southern Arabia and on Socotra It is not only in Egypt where we find the presence of Palmyrene merchants . Evidence from Southern Arabia and Socotra may also attest to direct or indirect participation in Indian Ocean trade networks .34 The evidence primarily takes the form of two inscriptions from 29 Tomber 2011, 7–8 . 30 Schörle 2017, 152, argues that the reference to ‘the merchants from Hadrianē Palmyra’ in the Koptos inscription implies a direct link to the city . While this is a possibility, it cannot be assumed beyond doubt that this implies financial or business links to individuals in Palmyra, as opposed to a desire to express a cultural affinity with their homeland on the part of these merchants — which need not actually attest to any specific business connections . 31 SHA Aurel . 32, SHA Firmus 2–6, Seland 2016, 43, Smith II 2013, 162 . 173, notes that it is unclear if the Palmyrene merchants of Egypt were truly independent or bound by the economic interests of powerful patrons living in Palmyra . 32 For a discussion of this evidence see Cobb 2018 . 33 See also Grout 2016, 261–262, who suggests that the Egyptian community of Palmyrene merchants was established ‘on the back’ of successful Mesopotamian-Persian Gulf trade . 34 See Map 2 .

V . Palmyrene merchants and the Red Sea tr ade · 71 southern Arabia and one from the island of Socotra (usually referred to in Graeco-Roman sources as Dioscorides) .35 Dealing with the southern Arabian inscriptions first, one of these has been found in the main temple of Shabwa, capital of the kingdom of Hadramawt, likely dating to the early third century AD . It mentions an Azi, son of Abianas and a Rab (son of ?), who were Palmyrenes . They are recorded as being residents of somewhere, but only the very first part of the word survive . One reconstruction has them come from the region of Mesene (Myšn, that is the client kingdom of the Parthians based in southern Mesopotamia), but as Robin has noted, Egypt (Mṣr) may also be a possibility .36 Another inscription was found at Al Uqla, 15 kilometres from Shabwa . It records a royal religious ceremony in which two Palmyrenes participated, alongside two Chaldeans and two Indians (the inscription seems to have been set up around AD 220) .37 One of the Palmyrenes was called Aziz, which has led Robin and Seland to suggest that he may have been the same person mentioned in the inscription from Shabwa . They also suggest, quite plausibly, that while these Palmyrenes appear to be envoys, wider trade interests most probably explain their presence .38 A third major piece of epigraphic evidence to be considered is a wooden tablet with Palmyrene text from the Hoq Cave on Socotra .39 The text refers to an ʾAbgar who came ‘in the pain of his soul’, followed by a blessing of the god who installed him there and a request for those who read the tablet to bless him and leave it in its place .40 On the basis of a textual reading the dedication of the tablet can probably be dated to July, AD 258 — since it refers to the month of tummûz and the year 569 (most likely a reference to the era of

35 Schwartz 1960, 27–28 notes the extremely speculative claim that the templum Augusti, which is featured on the Peutinger Table (a medieval map based on an earlier Late Antique version, which probably had its final revision around AD 425) next to Muziris, was built by the Palmyrenes in parallel to the temple constructed at Vologesias in Mesopotamia . Specifically it is stated: ‘[c]ertains trafiquants s’établirent aux Indes même et la Table de Peutinger mentionne à Muzyris un temple d’Auguste (au iiie siècle), que des auteurs modernes attribuent à des Palmyréniens, à l’instar de celui qu’ils ont construit à Vologésias (en Mésopotamie) .’ Whether the templum Augusti which features on this map actually reflects the genuine existence of an imperial cult set up by a diaspora merchant community at Muziris is heavily debated . Beyond the very weak parallel drawn to Vologesias, there is, however, no reason to assume that such a cult space/temple, if it did exist, was built primarily or exclusively by Palmyrene traders . On the Peutinger Table see Rathmann 2011/12, Rathmann 2016, Rathmann 2017, Talbert 2010 . On the question of the how the templum Augusti representation should be understood, see Ball 2000, 131, Fauconnier 2012, 91, Parker 2008, 246, Ray 1994, 66 . 36 RES 4691, Robin 2012a, 488–489 . 37 RES 4909, Bron 1986, 95–98 . 38 Robin 2012a, 491, Seland 2016, 79–81 . 39 This wooden tablet measures 50 cm by 20 cm, has a single handle with a drilled hanging hole, and contains a text 11 lines long that covers about half the tablet . Gorea 2012a, 448, suggest the text was probably written near the Hoq cave because of the use of the adverb ‘here’ (paralleling other usages) . 40 Ibid ., 451 . Of potential relevance here is a Greek inscription also found in the Hoq Cave, where an individual rendered his name in Greek as Aukar (ΑΥΚΑΡ), which could be a rendering for the Palmyrene-Aramaic ʾAbgar . On this see Strauch et al . 2012, 205–206 .

72 · Matthew A . Cobb the Seleucids, which was in widespread use by various peoples in Syria and Babylonia) .41 Radiocarbon dating of this wooden tablet indicates that it came from a tree cut down within the period AD 78–239, presumably indicating that the wood used was obtained from a tree cut down at least twenty years prior to the writing of the text .42 The purpose of ʾAbgar’s visit to Socotra, assuming he was not accidentally shipwrecked, may have been traderelated — given that this island was a well-known entrepôt in Antiquity (at least from the last few centuries of the first millennium BC to the early-mid first millennium AD) .43 It is worth noting that none of the three inscriptions just discussed explicitly connect the Palmyrenes mentioned with trade; as we have seen, it is largely inferred that they were directly or indirectly concerned with such activities . It is also not definitively clear whether these inscriptions attest to Palmyrenes arriving in southern Arabia or Socotra via the Egyptian-Red Sea route or whether they had originally set out from the Persian Gulf (or in the case of southern Arabia via overland routes) . Given the limitations of the evidence, most scholars have quite rightly avoided asserting one or the other view as unquestionably correct, though one interpretation might be favoured .44 Both routes are perfectly plausible means by which Palmyrenes might have reached the inland sites of Shabwa and Al Uqla . The author of the Periplus notes that the port of Kane (very likely Qana’/Qāni, a few kilometres away from modern Bir ʿAlī) was connected to the inland metropolis of Saubatha (Shabwa) .45 So there is certainly no specific reason that the Palmyrenes mentioned in the inscription had to come overland from the north (although this is also a possibility, see below) . Indeed the port of Qanaʾ shows clear trade links with Mesopotamia, India and the Mediterranean world from the first century BC to the seventh century AD . It is worth noting that the relative quantity of Mediterranean material at the site is greatest in the early period (first century BC to first century AD), while the proportion of Mesopotamian and Indian material becomes more significant in the middle period (second to fifth century AD), the site’s heyday, and also the timeframe within which the inscriptions were set up .46 But this fact on its own does not necessarily

41 Gorea 2012a, 452–453 . Gorea notes that tummûz was a more common term for the month and would have been more widely recognisable than the regional Palmyrene variant qnyn . 42 Dridi 2012, 461–462 . 43 For literary references to trade activity at this site extending from the Ptolemaic to Late Antique period see, Agath . 5,105a + b = (a) Phot ., Cod . 250,103,459b = (b) Diod . Sic . 3,47,8 f ., Periplus 30–31, Cosmas Christian Topography 3,178–79 . More generally for epigraphic and archaeological evidence see the papers in Strauch 2012a . 44 For example, Seland 2016, 41 (on the possible Mesopotamian-Persian Gulf connection), Schiettecatte – Arbach 2016, 185 (probably by sea), Evers 2017, 130 . 136 (Red Sea, Persian Gulf and overland via Arabia, all possibilities), Andrade 2018a, 186 . 199–200 (itinerary may reflect Red Sea route, but with Persian Gulf connections) . 45 Periplus 27, Casson 1989, 161–62 . 46 See Sedov 1996, 16–19, Sedov 2007, 76–79 . 92 . 104, Sedov 2010b, 374–375 . For an overview of the site see Sedov 2010a . On the recovery of Mediterranean material from maritime surveys near Qanaʾ see Davidde 2017, 591–593 .

V . Palmyrene merchants and the Red Sea tr ade · 73 tip the scales in favour of a Mesopotamian-Persian Gulf connection . Of additional relevance, is what appears to be a sherd from a Koan amphora (c . first-third century AD) with the word Achaia (Ἀχαία) that has been transliterated in Palmyrene script as ’Ky .47 At what stage the text was written on the vessel and how it finally ended up in Qana’, however, remains insoluble . In the case of the inscription from Al Uqla, the two ‘Chaldeans’ mentioned in the inscription could allude to individuals originating from central Mesopotamia, though some suggest that in this context Mesene (southern Mesopotamia) might make more sense . This has led Seland to propose that, by implication, the Palmyrenes mentioned in the inscription may also have come via the Persian Gulf (either by rounding the south-eastern coast of Arabia or by two stages: the Persian Gulf to northwest India and then from their to southern Arabia) .48 He notes that the author of the Periplus alludes to links between the Persian Gulf and northwest India, as well as northwest India and southern Arabia .49 It may be the case that the Palmyrenes mentioned in the Shabwa and Al Uqla inscriptions did not reach southern Arabia by sea at all . A recently discovered inscription from Jabal Riyām (located in what was the territory of the Sabaeans) and dating to the third century AD (sometime between AD 223–300) mentions an envoy who travelled overland through various tribes in western, central and northern Arabia, reaching as far north as Tadmur (Palmyra) and Tanūkh (Euphrates Valley) .50 It is not beyond the realms of possibility that the (probable) envoys mentioned in the Al Uqla and Shabwa inscriptions (of which the Aziz mentioned in both, could be the same individual) came via an overland route, although this can only be speculation . Coming now to the inscription from the Hoq Cave on Socotra, it has been noted that the purpose of ʾAbgar’s voyage (if not accidental shipwreck) to the island remains unknown . Whether he came via the Egyptian-Red Sea or Mesopotamian-Persian Gulf trade networks is also an open question . The author of the Periplus indicates that Dioscorides (Socotra) had trade links with the southern Arabian coast, northwest India (principally Barygaza), and with the ports of Limyrike (south-western coast of India) .51 Links with western and north-western India are certainly attested by 192 Brahmi epigraphs (and one Kharosthi epigraph) from the Hoq Cave, which date mainly between the second to fourth

47 Sedov 1992, 119–120, Sedov 1996, 15 . 48 Seland 2016, 41 . On connections between the southern Arabian coast and sites on the coast of the Persian Gulf (like ed-Dur) see Rutten 2007 . 49 Periplus 27 – the port of Kane (likely Qana’) had trade connections with Barygaza (Bharuch, northwest India), Scythia, Omana (possibly located on northern coast of the Gulf of Oman) and the coast of Persis (south-eastern Iran), 36 – connections between Apologos (mentioned as being near Spasinou Charax) and Omana . For a commentary see Casson 1989; see also Seland 2013 . On the evidence for potential Mesopotamian and East African trade links see Seland 2016, 202–204 . 50 Inscription Riyām 2006 – 17, Schiettecatte – Arbach 2016 . See also Evers 2017, 136 . 51 Periplus 30–31 .

74 · Matthew A . Cobb centuries AD . Strauch has noted that the styles of writings seem to mostly point to people of a Gujarati or Western Kshatrapa origin .52 Those arguing for a Persian Gulf connection might note that the southwest monsoon (broadly May/June to late-August/early September) caused navigational problems around Socotra due to the heavy winds and currents, making it an unattractive destination for those from the Red Sea heading eastwards .53 However, it should be noted that the mountainous nature of the northern coast means that certain areas can be quite sheltered from the southwest monsoon winds .54 Indeed, it would be a mistake to rule out the possibilty that ʾAbgar arrived via the Egyptian-Red Sea route . As can be inferred from the Periplus, it is possible that ʾAbgar could have set out from one of the southern Arabian ports (having arrived there via the Red Sea) in the winter months . Moreover, his arrival on Socotra (accidentally or intentionally) could connect to a return journey from southwest India to the Red Sea (as noted, the author of the Periplus refers to trade links between Limyrike and the island) .55 I note both possibilities, not to argue that either was necessarily the case, only that they cannot be ruled out . Indeed, a few Greek authors (such as the author of the Periplus and Cosmas), refer to Greek-speaking peoples coming from Egypt (temporarily) residing on Socotra .56 This Greek connection to the island seems to stretch from the Hellenistic down to the Late Antique period, meaning we should be cautious about ruling out arrival on the island via the Egyptian-Red Sea route .57 The notion that the ʾAbgar inscription might allude to regular links between Palmyrene merchants setting out from Palmyra, via Mesopotamia, the Persian Gulf, northwest India and then arriving at Socotra also has its problems . As Seland has noted in his reconstruction of the most likely schedules followed by Palmyrene merchants setting out for northwest India, they probably departed Palmyra around July or August, heading for the central Mesopotamian bank of the Euphrates, and from there used rafts to reach the mouth of the 52 See Strauch 2012a, Strauch 2016, Strauch et al . 2012 . 53 On these difficulties see Evers 2017, 136, Strauch 2012b, 380–381 . On this issue and how the monsoon winds affected Socotra more generally see Jansen van Rensburg 2016, 87–107 . Jansen van Rensburg notes that the safest periods to try and land on Socotra are during the inter-monsoon seasons — March to early May being the safest time to arrive on the island, though August to early September can be difficult for small vessels . More generally arriving with the northeast monsoon has historically been perceived as safer than arriving with the southwest monsoon . 54 Ibid ., 92–93 . 55 For a pertinent discussion of Philostorgius’ (Ecclesiastical History) description of the Christian missionary Theophilus’ journey from Socotra to places like Arabia and Axum see Andrade 2018a, 76–79 . Of additional interest is the presence in the Hoq Cave of six definite, two probable, and three possible Axumite inscriptions — evidence which appears to further attest to links between the Red Sea and Socotra — see Robin 2012c, 439 . 56 It is noteworthy that Cosmas speaks of the Christians on Socotra speaking Greek, even though they have Persian bishops — see Andrade 2018a, 135 . On Sasanian commercial activity in the Indian Ocean see Daryaee 2003 . 57 Agath . 5 .105a + b = (a) Phot ., Cod . 250 .103 . 459b = (b) Diod . Sic . 47 .8–9, Periplus 30–31, Cosmas Christian Topography 3 .178 f . For the evidence of Mediterranean contact with Socotra see Bukharin 2012a, Bukharin 2012b .

V . Palmyrene merchants and the Red Sea tr ade · 75 river, then setting out around September would sail via the Persian Gulf, Gulf of Oman, and northern Arabian Sea to reach northwest India, arriving around late September . They could then make a return journey from between mid-October until March (probably most arrived back in Palmyra during the Spring) .58 Our hypothetical Palmyrene merchant who wishes to go to Socotra, however, will use the northwest monsoon to set off for Socotra . A study of some seventeenth- and eighteenth-century inscriptions from Socotra reveal that it took on average around 50 days for Gujarati merchants to reach the island — probably arriving in December–January . The normal period of the return journey seems to be May– June (early in the Southwest monsoon season) .59 Clearly then our Palmyrene would need to wait in northwest India until mid-October to November, before undertaking the return journey to southern Mesopotamia . Our Palmyrene would only make it back to Palmyra after the conclusion of two trading seasons . In economic terms this seems redundant, especially since ports like Barygaza had strong trade links with many regions in India, Central Asia, and elsewhere, allowing for a rich variety of goods to be acquired .60 There is no obvious economic imperative for a Palmyra — northwest India — Socotra — northwest India — Palmyra circuit undertaken by an individual merchant . Ultimately, whether these inscriptions allude to the Red Sea, the Persian Gulf or overland routes, it is reasonable to accept that some Palmyrene merchants did travel via the Red Sea to reach ports in East Africa, and probably also the Gulf of Aden, if not also the (south)western coast of India . This is made emphatically clear from the Koptos inscription which mentions a community of Red Sea shipowners from Hadrianē Palmyra (…Ἁδριανῶν Παλμυρηνῶν ναυκλήρων Ἐρυθραικῶν…) .61

Palmyrene participation in the Egyptian-Red Sea trade: explanatory factors Patchy though the evidence is, the Koptos and Tentyris inscriptions, in particular, reveal the existence of a Palmyrene merchant and shipping community in Egypt that took part in the Red Sea branch of the Indian Ocean trade . On the basis of the dating of these two inscriptions, this activity is most apparent from the mid-second century into the third century AD . The inscriptions from the Hoq Cave, Shabwa and Al Uqla also appear to (in)directly attest to wider Palmyrene activity in the western Indian Ocean in the third century AD; although whether this evidence can be directly tied to the Egyptian-Red Sea route remains open for debate . Evidence for first to early second century AD Palmyrene 58 59 60 61

See Seland 2011, Seland 2016, 45–61 . Shelat 2012, 431 . See Periplus 27, 31–32, 36, 40–49 . I. Portes 103 = AE 1912, 55 no . 171 .

76 · Matthew A . Cobb commercial activity in Egypt appears rather weak . This consists of one mid-first century ostrakon from Berenike which records the name Hierabole, of likely Palmyrene origin (most of the other Semitic names mentioned in the collection of ostraka cannot be tied more precisely to a Palmyrene background), as well as four ostraka which may contain Palmyrene cursive script . Besides this, we also have a Palmyrene amphora (late first to early second century AD) discovered at Myos Hormos which has cursive script on it . None of this evidence, however, provides the kind of explicit attestation of commercial activity that is given by the Koptos inscription . The available evidence would seem to suggest that a sufficiently sizable Palmyrene community had become established in Egypt around the mid-late second century — sizable enough to have a communal association in Koptos with property, which was almost certainly used for celebrating specific cultic rituals and festivals connected to their ancestral gods; this association probably also helped arbitrate business/community disputes . This begs the question: what factors may have contributed to the creation of this Palmyrene merchant diaspora?

a) Roman-Parthian/Sasanian conflict? One theory that has been advocated is the notion that conflict between the Roman and Parthian (and later Sasanian) empires, made life more difficult for Palmyrenes wishing to use the traditional Mesopotamian-Persian Gulf route .62 We know of at least half-a-dozen major conflicts between the 160s to early 240s AD . This includes a “Parthian War of Lucius Verus” that involved conflict over Armenia and Upper Mesopotamia (AD 161–166); an invasion of Mesopotamia by Septimius Severus that included the sack of Ctesiphon (AD 197) and the establishment of a Roman province in northern Mesopotamia (AD 195–197/98); a “Parthian War of Caracalla”, with military operations largely in northern Mesopotamia, including the battle of Nisibis, followed by a stalemate and the payment of reparations by the Romans (AD 216–218); and clashes under Alexander Severus and Gordian III .63 On the face of it, this seems to parallel, in a quite striking way, the appearance of a Palmyrene merchant community in Egypt from the mid-late second to third century AD, and also the 30 odd inscriptions from Palmyra which attest to activity in Mesopotamia, the Persian Gulf, and northwest India . Most of these date between AD 130–161 (at least 16, but possibly another four that roughly date to this period) .64 Whereas there is a dearth 62 For discussion see Schwartz 1960, 31, Andrade 2018a, 198–199, Cobb 2015, 368–369 . 371, Gawlikowski 2016a, 25–26, Healey 1996, 35–36, Schiettecatte – Arbach 2016, 184, Schörle 2017, 149–150, Young 2001, 80–86 . 151–154 . 63 For an outline and references see Table 2 . 64 It need not be assumed that the concentration of inscriptions in this period can be connected to the notion that Mesene/Characene had become a client-kingdom of the Romans after Trajan’s Parthian campaign (AD 115– 117); remaining so up until AD 151 – for this view see Bernard 1990, Bowersock 1989, Gawlikowski 2016a, 2–3,

V . Palmyrene merchants and the Red Sea tr ade · 77 for the following three decades, followed by three inscriptions dating to the Severan era (Spasinou Charax is no longer mentioned from this period, although a journey to Vologesias is recorded in the early third century AD), and then an absence of inscriptions from the period AD 211 to 247 .65 As Sartre rightly notes, it would be a mistake to assume that the absence of inscriptions for certain periods meant cessation in trade, but it is not hard to believe that sporadic conflicts and their aftermaths made these ventures riskier — loan rates as high as 30 or 32 percent seems to attest to this .66 It does not have to be assumed that the Parthian (or later Sasanian) state tried to directly hinder Palmyrene commercial activity in times of open conflict . But it is possible that these conflicts impinged in some way on the various nomadic groups in the northern Arabian Peninsula, Syrian Desert and the region west of the Euphrates (the area which is referred to as Tanūkh in the aforementioned Sabaean inscription) . This is only speculation since our rather poor literary sources do not relay much information about these groups in the third century AD . However, we do know from later historical periods that some of these groups (such as the Ghassanids and Lakhmids) were backed by the Byzantines and Sassanians, who often used them as proxies in their conflicts .67 It has been convincingly argued that the urban elite and the merchants of Palmyra had close relationships with some of the nomadic peoples of the steppe, which would allow them to mobilise large numbers of animals for journeys between their city and the Euphrates (some of the nomads acting as guards, guides and animal handlers) .68 However, such dynamics are complex, and it appears that in some instances certain nomadic groups or sub-groups acted in a predatory manner towards these trade caravans (as inscriptional evidence attests) .69 Again it can only be conjecture, but it is worth at least pondering whether shifting dynamics brought about by wider regional conflicts impacted on Palmyrene caravan ventures . The notion that the chronological pattern suggested by the surviving inscriptions can be used to suggest levels of Palmyrene commercial prosperity has, however, been challenged . It has been proposed that the inscriptions may actually attest to difficulties that

65 66

67 68 69

Young 2001, 143–148, Żuchowska 2013, 382 . As Gregoratti 2011, 219–224, has noted, after the Romans withdrew (AD 117) the Parthian ruler Vologases III punished his erstwhile vassal, replacing the Hyspaosinid rulers with members of his own Arsacid family . Gregoratti 2019, also notes that Palmyrene activity in central and southern Mesopotamia in this period, no doubt, suited the economic interests of the kingdom of Mesene and their Parthian overlords . Inv. III .29 = CIS 2 .3949, Gorea 2012b, 464–465, Young 2001, 173–175 . Sartre 2005, 270–272 . 352–353 . See also Southern 2008, 32–33, who notes that the inscriptions may actually indicate the opposite, with merchants requiring more support as a result of a more aggressive stance by the Parthians . For a discussion of this see Bowersock 2013, 106–119, Fisher 2008, 311–334, Liebeschuetz 2015, 288–322 . Seland 2015, 49–50 . On enclosed nomadism see Sommer 2016, 14–15 . PAT 1378, Millar 1993, 333, Sommer 2005b, 206–207 . Smith II 2013, 28, actually suggests that Roman-Parthian conflict in the Severan period increased regional insecurity, requiring the Palmyrenes to strengthen their own military capabilities . He notes that threat of banditry persisted during the third century AD, perhaps explaining the increased significance of the role of stratēgos .

78 · Matthew A . Cobb were overcome with the help of benefactors (which may be better understood as an act of support by warrior elites, rather than a form of traditional Graeco-Roman civic euergetism), and that caravan trips that ended in disaster or were uneventful would simply not have been recorded on stone inscriptions .70 Such cautious treatment of the inscriptional evidence is certainly warranted, and as noted above, simply assuming that the absence of inscriptions meant a cessation in trade is unwise . That said, it is equally worth not discounting the potential impact that Roman-Parthian/Sasanian conflict may have had on the wider region .

b) Palmyrene soldiers in the Eastern Desert Another factor which may have influenced the establishment of a Palmyrene merchant community in Egypt is the concurrent presence of Palmyrene soldiers stationed in the Eastern Desert region .71 The evidence is primarily epigraphic and includes a dedication to the Emperor Caracalla set up in Berenike by an auxiliary soldier Marcus Aurelius Mokimos (September 8, AD 215) . This port site has also revealed a dedication to the Palmyrene god Yarḥibōl paralleling one found at Koptos, both probably dating around the late second to early third centuries AD (c . AD 180–212) — the latter alluding to a unit of Palmyrene archers (Hadriani Palmyreni Antoniniani sagittarii) .72 In the case of the Berenike dedication, the bilingual Greek and Palmyrene inscription seems to originally have formed the base of a statue of the god Yarḥibōl/Hierobol . The text also informs us that it was set up when Aemilius Celer was Roman governor of the region and commander of the Ala Herculiana, and when Valerius Germanon was chiliarch (military tribune/commander of a thousand men) . The individual who made the statue is also named as Berichei (very likely a Palmyrene name) .73 Besides these inscriptions, the fortlet of Didymoi also reveals the presence of Palmyrene military units in the region . This fortlet was one in a sequence along the main Koptos-Berenike route (ὁδὸς Βερενίκης) . The evidence consists of an inscription and a few ostraka suggesting that soldiers were either stationed at this site or at least stopped briefly during a crossing between Koptos and Berenike .74 This partially preserved sand70 See Seland 2016 . 68 . 78 . See also Terpstra 2016, 43 . 46 . On the Palmyrene elite as a military aristocracy see Sommer 2015, Sommer 2016 . 71 For an outline of the evidence relating to Palmyrene soldiers in the Eastern Desert region see Table 3 . On the recruitment of Palmyrene military units from the reigns of Trajan and Hadrian onwards see Millar 1993, 333, Smith II 2013, 165–166 . See also Andrade 2018a, 121–122, Cobb 2015, 371 . 72 Alston 1995, 188, Alston 2007, 4, Schwartz 1960, 31, Seland 2016, 42, Sidebotham 2011, 63–66, Sidebotham 2014, 612–613, Sidebotham 2017, 63–64, Speidel 1984, 221, Verhoogt 1998, 193–198, Young 2001, 81 . 73 Sidebotham 2011, 66 . 74 . 74 Besides the Koptos-Berenike route (ὁδὸς Βερενίκης), another major route running across the Eastern Desert went from Koptos to Myos Hormos (ὁδὸς Μυσορμιτική) . For these routes and the fortlet established along

V . Palmyrene merchants and the Red Sea tr ade · 79 stone inscription seems to have been originally oval in form and was found in a small chapel area inside the fortlet . The text is in Greek and refers to a few individuals who are Palmyrene soldiers — specifically Asth…, and (B)arlaas (possibly son of [Bara]thēs), and by implication also Asklas and Maximus . The text dates roughly between 176/177 and AD 219 .75 Another piece of evidence consists of a small, fragmentary text which is a list of soldiers belonging to a Palmyrene unit ([Παλ]μυρηνοί) — dating approximately to the end of the second, beginning of the third century AD . Cuvigny notes that only two of the names appear to have a Semitic origin (five appear to be Egyptian) .76 An ostrakon found at Didymoi (third century AD), refers to a Bassos, a Palmyrene horseman, who went up to Aphrodite (Wadi Menih el-Heir, another fortlet on the Koptos-Berenike route) with Classicus .77 In addition to this, two containers found at the fort have tituli in Palmyrene cursive script (dating around the early to mid-third century AD) — though whether these should be connected to soldiers is unknown .78 These inscriptions and ostraka show that Palmyrene military units were stationed at Koptos and Berenike (as well as possibly at some of the Eastern Desert fortlets like Didymoi) around the late second and early third century AD .79 The contemporaneous appearance of a Palmyrene merchant community in the same period makes it tempting to suggest that individuals or groups with familial or friendship connections to these soldiers decided to accompany/follow them to Egypt, though, admittedly, this is speculation . It has even been suggested that some of these soldiers actually invested (as part of larger consortia) in Red Sea commercial ventures .80 One theory holds that the soldier Marcus Aurelius Mokimos was able to afford his dedicatory inscription as a result of investment in the trade .81 Soldiers erecting inscriptions and statues is not an unusual phenomenon in the wider Roman Empire, so it need not be assumed that Mokimos had to pay for his dedication from the proceeds of trade . That said, the premise that some Palmyrene soldiers

75 76 77 78

79

80 81

them see Brun 2006a, Brun 2006b, Brun – Reddé 2011a, Brun – Reddé 2011b, Reddé – Brun 2006  . For a summary of Palmyrene evidence relating to Didymoi and to the region see Cuvigny 2012 Cuvigny 2012b, 14–15 . I. Did . 5, Cuvigny 2012c, 47–50 . No . 71, Cuvigny 2012d, 135–136 . No . 39, Cuvigny – Gagos 2012, 103–104 . Nos . 285–286, Cuvigny – Briquel-Chatonnet 2012, 219–220 . No . 285 has the text TYM’; no . 286 has two tituli, one in Greek, the other in Palmyrene Aramaic cursive script — in the case of the latter it reads YD’ (first line), ML (second line) . A dipinto (dated on palaeographic grounds to about AD 100) found at the Temple of Hatshepsut in Thebes, and which refers to a tesserarius called Athenodorus (possibly a rendering of the Palmyrene name Waballat), who is recorded as serving at Koptos, has led Speidel 1984, 222–223, to speculate that this could allude to the presence of Palmyrene soldiers at Koptos at an earlier stage (this is, in part, on the basis of a reference to vex­ illationes) . However, this is not conclusive proof for the presence of Palmyrene units as early as AD 100 . Seland 2016, 42 . Less explicitly, see also Evers 2017, 30 . Smith II 2013, 163, suggests that Palmyrene units abroad may have at least facilitated access to new markets . See also Grout 2016, 262 . McLaughlin 2010, 105–106, Sidebotham 2011, 253, Sidebotham 2014, 613 .

80 · Matthew A . Cobb directly or indirectly profited from their compatriot’s involvement in the trade is perfectly plausible . Just as support and security were provided by (para-)military units in the Syrian Desert, it is quite likely that Palmyrene merchants in Egypt drew upon their shared linguistic and cultural heritage to obtain the support of Palmyrene soldiers stationed in the Eastern Desert region . Indeed, as was the case in the Syrian Desert where nomads could sometimes pose a threat to commercial caravans, so too was the Eastern Desert a potentially dangerous region, especially in the late first and second centuries AD .82 It has actually been proposed, that one of the reasons that Palmyrene soldiers were stationed at Koptos and Berenike at this time was their expertise in desert warfare .83 Ultimately, it cannot be proved beyond doubt that the establishment of Palmyrene military units at Koptos and Berenike encouraged a number of Palmyrenes to relocate their commercial activity to the Egyptian-Red Sea route, but it is certainly a far from implausible theory .

c) The comparative profitability of the Egyptian-Red Sea route? The third major theory to be considered is the notion that the Egyptian-Red Sea branch of the Indian Ocean trade offered comparatively greater profit margins than the Mesopotamian-Persian Gulf route . Gawlikowski has argued that many of the goods that came via Palmyra were mostly sold in the markets of Emesa, Damascus and perhaps Apamea . He argues that a much greater bulk of goods entered into the Mediterranean world via Alexandria in Egypt and Antioch in Syria (i .e . the major centres of northern Syria were being supplied via Zeugma) .84 This view is, in part, based on the limited carrying capacity of the camels that would return to Palmyra in the Spring .85 Seland has estimated that the overland journey between Palmyra and the northern Persian Gulf equates to about 1,400 kilometres .86 By contrast, the journey between the Red Sea ports and Koptos, through the Eastern Desert, is shorter . On the route from Berenike to Koptos, the main 82 A number of the ostraka from the fortlets of Krokodilō (al-Muwayh), Dios (Abu Qurayya) and Didymoi (Khashm al-Minayh) reveal the potential (and likely increased) threat posed by some of the indigenous groups of the Eastern Desert of Egypt during the second-third centuries AD . On this evidence see Cuvigny 2005, Cuvigny 2006, Cuvigny 2011a, Cuvigny 2012a, Cuvigny 2014 . 83 Speidel 1984, 221 . 84 Gawlikowski 2016a, 24–25 . Contra Sidebotham 2011, Terpstra 2016 . 85 Seland 2015, 49, notes that even a single shipload of eastern merchandise (perhaps some 100–150 metric tonnes) corresponds to numerous 180 kilogrammes camel loads (around 550–840 camel loads) . And that in the seventeenth and eighteenth centuries, the annual Aleppo to Basra and Baghdad caravans could number up to 5000 animals . 86 Seland 2011, 399–400 . As Seland rightly notes, the rhythms of the Southwest and Northeast monsoon winds means that merchants plying either the Egyptian-Red Sea or the Mesopotamian-Persian Gulf routes would be restricted to one round journey a season .

V . Palmyrene merchants and the Red Sea tr ade · 81 sources (Pliny’s Natural History, the Antonine Itinerary and the Peutinger Table) give a distance of 257–258 Roman miles (roughly 380 kilometres) — a journey which Pliny reports would take 12 days .87 The six- to seven-day journey (the estimate given by Strabo) from Myos Hormos to Koptos would be around 170 kilometres .88 The river-borne journey from Koptos to Alexandria on the Nile would in actuality be about 850 kilometres (as the crow flies, the distance between Alexandria and Koptos [Qift] is in fact about 640 kilometres), the various bends in the Nile elongating the journey .89 Leaving the riverine transport along the Nile aside, the relative overland journey time is almost four to one when comparing the route to Palmyra with the crossing from Berenike to Koptos, and around eight to one for the Myos Hormos to Koptos comparison .90 Seland accepts Gawlikowski basic point that profits for the Red Sea route may have been higher than the traditional route to Palmyra, but unlike Gawlikowski he argues that it was still sufficiently profitable to export goods acquired via Palmyra to Rome . He also notes that this route allowed goods to be brought to the Mediterranean markets in the Spring, at the start of the sailing season, whereas many eastern goods would arrive in Alexandria in the Summer period .91 While our price data for the cost of various goods in the markets of the Indian Ocean and what they might be sold for in the Roman world is very patchy, Seland’s premise appears reasonable . Pliny mentions that goods from India could be acquired and then sold back in the Roman Empire for one-hundred times their original cost (the specific context of this statement is a discussion of the Egyptian-Red Sea route to India) .92 Similarly, the Hou Hanshu claims that there was a ten to one profit

87 See Sidebotham 1986, 60–61 . Additionally, Bülow-Jacobsen 2006, 52, Cuvigny 2011b, 5–7 . Plin . HN 6,102 . 88 Strab . 17,1,45 . Maxfield 1996, 11–12, notes that Bedouin accounts and the British Army Camel Corps training manual indicate a ladened camel could comfortably travel 24 to 32 kilometres in around six to eight hours per day . This suggests that the journey times mentioned by both Strabo and Pliny are realistic, as these averages indicate a journey from Koptos to Berenike would take 12–16 days (c . 380 kilometres), and five to seven days for Myos Hormos (c . 170 kilometres) . See also Cooper 2014, 165 . 89 See Rougé 1986, 47 . While describing the route taken by merchants setting out from Alexandria, Pliny (HN  6,102) claims that departing from Iuliopolis (just south of Alexandria), the merchants would use the Etesian wind (north wind) to sail down the Nile to Koptos . A journey he claims took 12 days and was 309 Roman miles long (457 kilometres) . This calculation would suggest an average travel time of about 38 kilometres per day: a modest speed with a favourable wind (although with opposing currents), but as we have noted the actual distance is longer than Pliny suggests, so a 12 day journey down the Nile would equate to more like 71 kilometres per day . For a more recent discussion of historic journey times on the Nile see Cooper 2014, 155–166 . Information from accounts dating between the eleventh to nineteenth centuries suggests that a journey time between Cairo to Qus (ancient Apollonos Mikra), about 10 kilometres SSW of Qift, would be around 14 .5 to 22 days (ibid ., 157–159) — suggesting that Pliny’s travel time between Juliopolis and Koptos may be at the optimistic end . 90 On the comparison of the overland route to Palmyra with the crossing from Berenike to Koptos see Seland 2011 . 91 Seland 2016, 53 . See also Seland 2011 . 92 Plin . HN 6,101 .

82 · Matthew A . Cobb ratio for trade links between Da Qin (Roman Empire), Parthia and northwest India .93 Both statements are rather generalised, the profit margins will likely have fluctuated depending upon the type of goods being purchased . There is also probably a degree of hyperbolic exaggeration in these claims, although the basic assertions need not be doubted . Thus it is not hard to imagine that items like pearls, semiprecious stones and silks, all (in)directly attested at Palmyra, could have been profitably brought to the major markets of the Eastern Mediterranean for consumption at these sites, and for redistribution further west . It is also worth bearing in mind that other factors, beyond transport times and costs, will have impacted on the prices for different commodities . Variations in levels of production or availability of certain commodities in East Africa, southern Arabia, or India (perhaps due to climatic conditions or socio-political turmoil), high losses at sea, piracy and banditry, and the level of stockpiles in the warehouses of cities like Rome, will all have impinged on the market value of commodities in ways we are not in a position to fully appreciate .94 The main point is that both the Egyptian-Red Sea and the MesopotamianPersian Gulf (via Palmyra, Zeugma or other sites) routes were major conduits for the import of eastern commodities . That the profit margins may have been higher for the Egyptian-Red Sea route is a distinct possibility . That this may have encouraged some Palmyrenes to participate in the Egyptian-Red Sea route is also a possibility, though our evidence does not allow us to assert this definitively . It could also be, as Schörle suggests, that once Palmyrene trade networks in the Mesopotamian-Persian Gulf region had become well-established, it was became easier to take the skills and finance that had been developed, and laterally expand (horizontal integration) into the Egyptian-Red Sea sphere .95

Conclusion It is not the intention of this paper to assert that any one of the factors just discussed can single-handedly explain the presence of Palmyrene merchants in Egypt and their choice to participate in the Egyptian-Red Sea branch of the Indian Ocean trade . Indeed, none of these factors is mutually exclusive, nor do they offer an exhaustive explanation of potential motives . The personal circumstances and motivations that influenced individual Palmyrenes are beyond recovery — it is only possible to speculatively consider, at an abstract level, the types of issues that may have affected people . The purpose of this paper has instead been to consider the nature of the available evidence (limited as it is), as well as considering the extent to which this evidence allows us to interpret any patterns or correlations that might be informative in assessing the various theories proposed . 93 Hou Hanshu 88,12 . Specifically it is noted that ‘They trade with Anxi (Parthia) and Tianzhu (Northwestern India) by sea . The profit margin is ten to one .’ — Translation from Hill 2015 . 94 For a discussion of some of these issues see Cobb 2014 . 95 Schörle 2017, 152–153 .

V . Palmyrene merchants and the Red Sea tr ade · 83 From the review just undertaken, a few things stand out . The most explicit evidence we have for Palmyrene involvement in the Egyptian-Red Sea branch of the Indian Ocean trade connects to the mid-second to early third century AD . This seems to correlate quite closely with evidence for the presence of Palmyrene soldiers based at Koptos, Berenike, and maybe also in the fortlets of the Eastern Desert . We do not know what kind of familial or friendship connections may have existed between these merchants, shipowners and soldiers, and whether any mutually beneficial arrangements were established . But that such connections could have existed seems highly plausible . It is quite easy to imagine that some merchants accompanied or followed in the footsteps of Palmyrene soldiers who were brought to Egypt . There also appears to be a correlation between the comparative dearth of late second and early-mid third century inscriptions from Palmyra which attest to participation in the traditional Mesopotamian-Persian Gulf route, the various incidents of Roman-Parthian/ Sasanian conflict between the 160s to early 240s AD, and the appearance of a Palmyrene community in Egypt at this time . The suggestion that these incidents of conflict may have encouraged some merchants to participate in the Egyptian-Red Sea trade route is, however, controversial . It has also been quite reasonably pointed out the inscriptional evidence should not be used as an absolute indicator of fluctuating levels of Palmyrene commercial activity along the Mesopotamian-Persian Gulf route . That said, it is worth at least considering the possibility that Roman-Parthian/Sasanian conflict may have caused problems in the wider region, including in relation to the various nomadic tribes of the Syrian Desert and Arabian Peninsula, that created (not insurmountable) challenges for Palmyrene merchant caravans . Finally, it has been noted that the profit margins may have been higher for the EgyptianRed Sea route (although both routes were clearly profitable) . Whether this encouraged some Palmyrenes to shift their activities to Egypt remains open to question, though it is a possibility . Ultimately the limited nature of our evidence leaves us to engage in wellfounded conjecture .

VI. The rise of the merchant princes? Scale, status, and wealth in Palmyrene trade Eivind Heldaas Seland

A caravan city? Modern observers take it for granted that the source of Palmyra’s wealth, as it is reflected in the spectacular ruins of the Syrian city, and also evident in the central role she was able to assume in the third quarter of the third century, largely stemmed from her role as an intermediary in trade between the Roman Empire and the East — the Mediterranean and the Indian Ocean .1 Even if Palmyra had other resources as well, including a large pastoral and even an agricultural hinterland,2 and a political position at the height of her power that allowed her to channel resources in her direction,3 there is no real doubt that trade was important in bringing wealth to the city . That said, our impression of Palmyrene grandeur and importance is intrinsically linked with the two aspects — the story of Zenobia’s rebellion — and the, until recently, exceptionally well-preserved ruins . These had largely survived because the city was gradually reduced to a small village in the Islamic period and because looting and reuse of building material was more limited than in most corresponding cases due to Palmyra’s desert location . Compared to Jerash, Dura-Europos, Hatra, Balbeek/Heliopolis or Apamea, Palmyra is perhaps not that exceptional . Taken at face value, our evidence of Palmyrene long-distance commerce is actually not that impressive . We have a single passage in Appian’s Civil Wars claiming that the Palmyrenes were merchants conveying Arabian and Indian goods from the Persian Empire, selling them in Roman lands .4 Then there are the approximately 30 caravan inscriptions, a little more than half of which actually mention caravans, most of the others referring to Palmyrene merchant communities in Mesopotamia .5 A third group of evidence is some thirty additional inscriptions from outside Palmyra,6 distributed geographically from 1 2 3 4 5 6

So e .g . Drexhage 1988, Sartre-Fauriat – Sartre 2008, Seland 2016, Will 1992, Young 2001 . Meyer 2017 . Hartmann 2001 . App . BC 5,9 . Gawlikowski 1994, Yon 2002, 263–264 . Ibid ., 270–273, Seland 2016, 80 .

86 · Eivind Heldaas Seland Yemen to Rome, most of which should arguably be read in the context of Palmyrene trading activities, although only two of them actually mention that .7 This would perhaps not count for much, had it not been for the fact that even this meager evidence far outweighs what we have available for the rest of the Roman Empire altogether .

Palmyrene trade Palmyrene commercial activities seem to have started shortly after the Parthian conquest of Mesopotamia and the partial breakdown of Seleucid rule in Syria .8 The Seleucid crisis left a power vacuum along the Euphrates and the Syrian Desert rim . Erstwhile nomadic or semi-nomadic groups seem to have taken advantage of this situation to settle down on good agricultural land on the edge of the desert and establish their own polities in geographical locations where they were able to profit both from pastoral and agricultural economies .9 This is also the likely context for the emergence of Palmyra . Other examples include the Nabataeans, Itureans, and Emeseans . The Roman geographer Strabo describes how this political fragmentation caused problems for traders, who were used to operate inside imperial borders, but now needed to deal with multiple political authorities, all levying taxes in return for protection . In this situation, it became economically viable to operate caravans across the Syrian Desert in order to avoid taxation, rather than to move along the Euphrates valley, as had been customary in the past .10 In the first century AD, when the inscriptions start to provide more secure data points, it seems that Palmyrene merchants were active in cities in Middle Mesopotamia, namely Seleucia and Babylon . By the first century AD, they were established in Southern Mesopotamia, primarily in Spasinou Charax, a port city and the capital of the small Parthian vassal-kingdom of Mesene . By the second century at the latest, they had taken up maritime trade in the Persian Gulf as well as in the Red Sea, and travelled to Northern India and probably other destinations in the western Indian Ocean .11 By the third century all this continued, but there are also three inscriptions from present-day Yemen, attesting Palmyrenes active in the kingdom of Hadramawt and on the Island of Socotra .12 These latter inscriptions are not explicitly related to trade or merchants, but it is hard to imagine other motivations of travel to such locations . Trade seems to have continued after the Sasanian takeover in Mesopotamia, but all evidence disappears after the Roman sack of Palmyra in 273 AD . We cannot rule out that it continued for a while or in other 7 8 9 10 11 12

I. Portes 39 and 103 . Gawlikowski 1996, 119, Millar 1998, 126–127 . Myers 2010, Sartre 2005, 12–25, Sommer 2005b, 58 . Strab. 16,1,27 . Gawlikowski 1994 . RES 4691, 4859, Gorea 2012a, Robin 2012b . See also Cobb in this volume .

VI . Scale, status, and wealth in Palmyrene tr ade · 87 forms,13 but almost thirty years later, when the Roman and Sasanian rulers sign a peace treaty, it was Nisibis in northern Mesopotamia that was designated as the only legal place for cross-border transactions,14 not Palmyra . If Palmyra can be described as one of the hubs around which world commerce revolved during the Roman period, then surely the caravans going out from the city were the spokes of the wheel . Every late summer caravans would gather and head towards cities on the Euphrates . There, goods would be loaded on large rafts that travelled either to the large Mesopotamian cities or to the ports at the head of the Persian Gulf .15 If the latter was the case, caravans would connect with ships doing the journey to the west coast of India,16 returning in the late winter months due to the seasonality of the monsoon .17 From southern Mesopotamia, caravans would be organized for the homeward journey to Palmyra, and many of the inscriptions commemorate help, which was lent to these caravans by members of the Palmyrene elite .

Elite involvement In Les inscriptions caravanières de Palmyre (1932), the study which arguably started modern scholarship on Palmyrene trade, Mikhail Rostovtzeff summed up that while the inscriptions give quite a lot of information on the history of Palmyrene caravan trade, they reveal precious little on the internal organization of Palmyrene caravans .18 In essence, this observation holds true more than 85 years later . To understand the organization of Palmyrene trade we need to find analogies from elsewhere, be it other parts of Roman history, later and better-documented caravan trade in the Near East, Medieval and early modern trade between the Mediterranean and the Indian Ocean, or modern economic theory . This has resulted in a number of partly overlapping, but nevertheless competing and seemingly incompatible models of Palmyrene caravan trade, which carry implications far wider than the study of Palmyra, because they rest on assumptions about the nature of the ancient economy more in general . The complexity of the issue might be demonstrated by one of the so-called caravan inscriptions preserved in situ in the Agora of Palmyra . It is cited here in Fox, Lieu, and Ricklefs’ translation of the Greek version .

13 14 15 16 17 18

al-Asʿad – Gawlikowski 1986–1987 . Petr . Patr . fr . 13–14 (FGH IV 188–189) . Gawlikowski 1983, Seyrig 1963 . Gawlikowski 1988, Healey 1996 . Seland 2011 . Rostovtzeff 1932b, 805 .

88 · Eivind Heldaas Seland By decree of the council and people the four tribes . Ogelos of Makkaios of Ogelos of Agegos of Sewiras, who assisted them with all his valour and courage, particularly because of his frequent military expeditions against the nomads . He has always assured the safety of the merchants and caravans in all his superintendence of these caravans [or rather ‘caravan-leadership’ I would propose] . For this he has spent large sums out of his own resources and has conduct[ed] his whole political career with brilliance and glory . In his honour . In the year [5]10 .19

Others have discussed the content and significance of the inscription in detail .20 For the present purpose, it is sufficient to point out the sheer number of actors, institutions and social interactions that are evident in this relatively short text . There are a council (boulē/bwl’) and the people (dēmos/dms), as we would expect in any city-state of the Roman East . Then four tribes are mentioned (phylai/pḥzy’), which according to some commentators would be civic tribes,21 like in Athens and Rome, and in the opinion of others would be based on perceived common ancestry like tribes in the ethnographic sense of the word .22 The honorand, Ogelos (ʿOgēlu), is mentioned with no less than four paternal ancestors, highlighting the importance of lineage . The interconnected reasons for honoring him are fourfold: his commands (in plural) against the nomads, his success in keeping the merchants safe during his tenures, again in plural, as caravan leader, his use of his own resources towards these ends, and finally his unsullied civic record . This is not a typical inscription, most of them are shorter and more general, but the text is representative in that it gathers several of the issues that are at the core of the debate on the nature of elite involvement in Palmyrene trade . Together with the other caravan inscriptions, it has fostered at least three distinct, but overlapping models of interpretation . The most time-honoured model is that of the merchant prince . Palmyra, since the work of Theodor Mommsen and especially Mikhail Rostovtzeff, was seen as a republic dominated by a circle of rich merchants, “Großhändler” or “merchant princes”, who were involved in financing, organizing, protecting and conducting trade .23 The likely historical analogies would be late medieval and early modern Europe`, especially the city-states of northern Italy, but also Holland, and perhaps the German Hanseatic League . In all three cases, merchants constituted the civic elites of their home communities . Family-businesses engaged in long-distance trade . Merchants formed expat communities abroad, and armed forces were employed in order to protect trade . The model is sometimes explicit, like in Ernest Will’s book-title La Venise des sables,24 sometimes implicit in the use of analogies, such as factory, funduq or comptoir, terms belonging to early modern trade, for Palmyrene 19 20 21 22 23 24

PAT 1378, Fox – Lieu – Ricklefs 2005, 99 . Andrade 2013, 125–146, Young 2001, 161–163 . Sartre 1996, Yon 2002, 66–69 . Smith II 2013, 37–47, Sommer 2017a, 215–217 . Mommsen 1902–04, Bd . 5, 428–429, Rostovtzeff 1932b . Will 1992 .

VI . Scale, status, and wealth in Palmyrene tr ade · 89 merchant communities in Mesopotamia .25 While adopting parts of this terminology does not necessarily imply acceptance of Rostovtzeff ’s model of Palmyrene trade, the merchant prince or trade lord is still very much alive in modern scholarship on Palmyra .26 Ernest Will was also the scholar who made the breakthrough in the interpretation of the caravan inscriptions, and provided an alternative to Rostovtzeff ’s model of the merchant prince .27 Will noted that there are three main categories of actors in the inscriptions: the merchants, always acting as a group, the honorands, always named individuals, and in some inscriptions the caravan leader, synodiarchēs or in two late instances the head merchant, archemporos, who are also named, but not necessarily identical with the people who are honored . Will saw the honorands as caravan patrons, persons of traditional and tribal authority — he explicitly used the modern Arabic word sheikh — who used the resources at their disposal: manpower, animals, and also money, to invest in and protect the caravans . These people, however, were no merchants . They were aristocrats and experts on the desert and on violence . They were acting as patrons, extending their protection to their community . The merchants were relatively modest men, commercial experts, and the caravan leaders were technical experts specialized in the practical aspects of operating a caravan in a desert environment .28 Will’s model, surely based both on his thorough understanding of classical history, as well as on his many years of first-hand experience from life in the Middle East, introduced a consistent reading of the epigraphic corpus . It combined ancient evidence with ethnographic analogy, and it solved the seeming anomaly of Palmyra as the city of merchant princes, by dismantling the concept altogether . Palmyra was not the only city in the ancient world governed by merchants after all . It was dominated by traditional elites, like Rome was and any other classical city, too . The economic foundation of these elites was not trade, but agricultural and pastoral estates . It is surely a coincidence that Will’s article was published in the same year as Karl Polanyi’s Trade and markets in the early empires, which revived the primitivist model of the ancient economy and that inspired Moses Finley’s Ancient Economy appearing 16 years later, but the timing was nevertheless perfect .29 It comes as no surprise then, that the essentials of Will’s model, which could perhaps be called the nomadic or tribal model, have stood the test of times, even if most commentators now recognize that the synodiarchs were men of considerable standing and resources .30 A different take on Will’s model is represented by the work of Gary K . Young, who emphasized the civic position of the people honored in the inscriptions over their traditional 25 26 27 28 29 30

Gregoratti 2015, Teixidor 1984, Yon 2002 . Palmyra e .g . Celentano 2016, Gregoratti 2015 . Will 1957 . Ibid . 1957 . Finley 1985, Polanyi – Arensberg – Pearson 1957 . Seland 2014, Seland 2016, Sommer 2005b, Sommer 2016, Yon 1998, Yon 2002 .

90 · Eivind Heldaas Seland or tribal authority . Young agreed that the honorands were not merchants or directly involved in commercial operations, but he rejected the notion of caravan patrons, suggesting instead that the services commemorated in the caravan inscriptions were general acts of euergetism found in any city of the Roman Empire, with the important difference that in Palmyra the notables would support caravans instead of or in addition to public fountains, baths, or the restoration of temples .31 Young’s civic model is even closer to the primitivist paradigm of the ancient economy than the nomadic model, in not even ascribing active participation, but rather reactive response, in guarding and protecting the caravans from the side of the Palmyrene elite . This approach has recently been developed by Nathaniel Andrade, who convincingly argues that the Palmyrene elite needed to balance tribal and civic identities in order to succeed both at home and on the regional and imperial stage .32 These three models of elite involvement in Palmyrene trade — the merchant prince, the tribal leader and the Roman aristocrat — are all used and combined in current scholarship . While the merchant prince lacks support in the epigraphic record — no honorand, only dedicands of inscriptions are described as merchants, the tribal and civic models are both consistent with our data . This neat picture, however, has two shortcomings . It is static, and does not account for development during the three centuries of attested Palmyrene participation in caravan trade, and also it largely ignores the merchants, the people who were actually doing the dirty work of trading and presumably reaping some of the profits of it .

Merchants in Palmyra The problem is that there are no Palmyrene reliefs or inscriptions authored by a self-professed merchant with one possible exception addressed below . Who were these people, who were invariably referred to as a collective, and who must have been so important to the city, but who never stood out in public, at least not in ways that we easily recognize? Indeed, what would we expect from a merchant? Even if subscribing to a modernist view of the ancient economy, in that production as well as trade to a large extent responded to market mechanisms, there is no doubt that trade had low status in the ancient world . Elite involvement in trade was if not exactly masked, then rarely boasted of, and as in all pre-modern settings, it was normal to invest profits from trade in more secure, even if less lucrative assets such as agricultural land . Judging from the Palmyrene iconographic and epigraphic records, there were four things that gave status . One was family, including in the wider sense of lineage . Investing in funerary towers and hypogea for the extended family was evidently considered a worthwhile investment for the Palmyrene elite, and adding anything from two to five generations of paternal ancestors to your name in inscriptions was also deemed prudent, whereas most Romans would do with one, or none . 31 Young 2001, 136–186 . 32 Andrade 2012, Andrade 2013 .

VI . Scale, status, and wealth in Palmyrene tr ade · 91 A second source of status was religion . As Rubina Raja has shown, the priestly profession is perhaps the most visible and widespread in funerary art .33 Martial and nomadic virtues were certainly respected, as we see in the comparably many reliefs showing men posing with camels and weapons .34 Finally, there was private benevolence towards the community, for instance in assisting the caravans, which was certainly a way of accumulating honor . Trade, however, was an attractive potential source of income . The figure of Trimalchio in Petronius’ Satyricon, who made a fortune buying ships and exporting wine to Gaul, is intended as satire, but even if it was exaggerated, it needed to resonate with reality in order to work . Pliny in his Natural History writes that Indian goods could be sold in the Roman Empire for one hundred times their purchase cost .35 This only reveals that profits were potentially very high . Our only documentary figure is from the so-called Muziris Papyrus, dealing with the customs clearance of the cargo of the ship Hermapollon, which has returned to Egypt after a journey to Muziris in southern India . Domenic Rathbone has calculated the value of the cargo to be some seven million sesterces, more or less corresponding to the price of an attractive agricultural estate in Italy .36 Each caravan coming to Palmyra would contain the cargo of several ships, and a successful venture will have made at least some people, and not only the tax collector, extremely wealthy . We do have evidence of Palmyrene shipowners . This consists of one well-preserved and one quite fragmentary inscription from the mid-second century AD, reporting Palmyrenes going to Scythia, present-day Pakistan, and returning on ships owned by named individuals .37 Both the onomastics and the fact that the names were considered worth mentioning, imply that these are Palmyrenes, perhaps from the longstanding Palmyrene community in the port city of Spasinou Charax, at the head of the Persian Gulf . There is also a famous inscription from Koptos in Egypt, probably from the second century as well, which commemorates the restoration of the clubhouse belonging to the association of Palmyrene Red Sea ship-owners,38 implying that they were a sizeable group with some measure of continuous presence . What did it imply to be a ship-owner? Ships were the largest and most complex pieces of technology in pre-modern economies . Unfortunately, prices or costs for ships are hardly preserved from the Roman period . Jones estimated the value of a well-sized merchant ship in late antiquity at 600–1000 solidi . This would correspond the annual pay of 120–200 fourth-century infantrymen,39 not at all the people worst off in the ancient world, and it would be on par with the annual revenue of the largest estates mentioned in the biography

33 34 35 36 37 38 39

Raja 2015, Raja 2017 . Seland 2017 Plin ., HN 6,20 . Rathbone 2000 . PAT 1403 and 2763 . I. Portes 103 Harl 1996, 283 .

92 · Eivind Heldaas Seland of Sylvester in the Liber Pontificalis, detailing the properties given to the churches of Rome by Constantine .40 Ships were thus major investments, made by individuals from Palmyra who desired to profit from trade . No doubt, the continued operation of Palmyrene trade for almost 150 years after the Palmyrene shipowners first appear, implies that at least some of these people were successful . We would expect that some of them also wanted to turn economic assets into social standing in their hometown .

The rise of new elites How would people who had made their fortune from trade represent themselves to their communities . Unfortunately there is a lack of data also on this point . From Porta Maggiore in Rome we have the peculiar tomb of Eurysaces, who owned what was apparently a very successful bakery business, and was so proud of that, that he had his tomb fashioned as an oven .41 The Igeler Säule von Trier in Germany is perhaps the closest possible parallel to a Palmyrene magnate .42 The apparent family-firm depicted there seems to have been involved in the production, transport and retail of textiles on the border of the empire, although to the northeast rather than to the southeast . What would a funerary monument like this cost? Again, the data is underwhelming, but Duncan-Jones in his collection of quantitative data from Roman Italy, has several examples of monuments costing 100 .000 sesterces or more,43 corresponding to the annual wages of 111 soldiers,44 and thus comparable to the cost of a ship, which seems not entirely unreasonable . My suggestion is that we have new elites that become visible in Palmyra in the second and third centuries, that these challenge the position of traditional elites, and that their wealth and status were connected with trade . I will offer four pieces of corroborating evidence for this . First, we get the development of a new kind of funerary monument, the so-called temple- or house tomb . These emerge in the second century and flourish in the third .45 They are way more conspicuous than the underground hypogea, and also boast more richly decorated exteriors than the funerary towers . While these certainly did not all belong to merchant families, the architectural change arguably reflected outside impulse and a change of ideology that allowed the announcement of monetary wealth to a larger degree than the case was with the somber funerary towers and the invisible hypogea . It is probably no coincidence that our only identifiable iconographic depiction of a merchant, 40 41 42 43 44 45

Duchesne – Vogel 1884, Seland 2012 . Petersen 2003 . Dragendorff – Kryger 1924 . Duncan-Jones 1982, 127–131 . Harl 1996, 216 . Gawlikowski 1970, Schmidt-Colinet et al . 1992 .

VI . Scale, status, and wealth in Palmyrene tr ade · 93 a man standing with a ship on his left and a camel on his right, is from the tomb of Iulius Aurelius Marona, a house tomb dedicated in 236 AD . Interestingly we only have his own name preserved, not the names of two to five generations of forefathers, as was common in Palmyrene inscriptions, reflecting perhaps that he lacked a family-background worth boasting .46 Second, the caravan inscriptions change . In the first and second centuries they mention the title of synodiarch — caravan leader . In a few cases, he is clearly a man of social standing, like in the inscription cited above . In others, he is simply mentioned along with the merchants offering the inscription, leading Will to conclude that he was a subordinate technical expert . Anyway, he is distinct from the merchants, and the skills required to lead a caravan through a desert inhabited by sometimes hostile nomads were different from those needed by a merchant . The synodiarch was not a merchant-prince, but a man with professional know-how and social standing among the members of the caravan, among authorities on both sides of the imperial border, and among the people that the caravan would encounter in the desert . He fits better with the tribal model of elite involvement .  In two late inscriptions, however, we encounter the archemporos, that is “headmerchant” .47 Taken at face value a head-merchant is not the ad hoc leader of a caravan expedition, but a standing office representing the merchants of Palmyra . I suggest that this reflects the increased status and importance of merchants at the expense of the old tribal elite . By this time, not only the caravans were important, but also the merchants . I think it is not unlikely that the drift towards monarchy in the third century is connected with a replacement of old, aristocratic elites with new people willing to support an upwardly mobile strongman like Odaenathus . This brings me to my final argument . I believe we do have a literary portrait of the kind of man who was successful in Palmyra in the third century . To be frank, it is not a very flattering one . In the fourth-century Historia Augusta, we encounter the Syrian usurper Firmus, who headed a brief and unsuccessful revolt against Aurelian in Alexandria after the defeat of Zenobia . Among other, less probable things, Firmus is said to have been a friend of Zenobia, and on good terms with the Blemmyes of the Egyptian and the Sarcens of the Arabian Desert . He was a trader sending ships to India, and he was marvelously rich . He also indulged in drinking contests, ostrich riding, circus-strongmanship and other pastimes not befitting a Roman aristocrat .48 Commentators are unanimous that Firmus never lived and that the author of the Historia Augusta invented him for sport .49 That, however, does not express criticism, for his portrait of a tremendously rich trader, working together with political authorities and 46 47 48 49

See Sander in this volume . PAT 0282 and 0288 . SHA Firmus . Bowman 1976, 158–159, Hartmann 2001, 403–410, Poignault – Hunzinger – Kasprzyk 2001 .

94 · Eivind Heldaas Seland nomads, and with an outspoken distaste for traditional decorum, is just the kind of upwardly mobile person we would expect to find in Palmyra in the third century . In conclusion, I argue above that Rostovtzeff was wrong in describing Palmyra as a city of merchant-princes and that Ernest Will and the scholars following him were right in emphasizing the traditional, tribal and nomadic nature of authority of the elite members honored in the caravan inscriptions . However, Rostovtzeff was not all wrong and Will, even in the revised versions developed by Young, Yon, Sommer, Andrade and myself, was not all correct . I argue that the tribal or nomadic model for the organization of caravan trade was challenged in the second and third centuries, when a self-confident and wealthy group of people who had made their money from maritime trade, but who lacked inherited aristocratic authority, rose to power . I claim that this reconstruction is supported by our, albeit limited, evidence, that it contributes towards our understanding of the architectural and political changes evident in the city in the third century, and that it lends a much needed dynamic element to our mostly static models of Palmyrene trade . 

VII. On the edge of empire. Elite representation and identity building in Roman Palmyra Ann-Christine Sander

Palmyra — a city of talking stones . From no other city in the frontier area between Rome and Parthia have so many inscriptions survived to posterity as from the oasis of Tadmur . They evoke a vivid picture of the local population of Palmyra from the 1st to 3rd centuries AD . Numerous inscriptions tell us about deeds and merits of members of the local elite, which has prompted many historians to attempt at reconstucting an overall picture of Palmyra’s upper crust from such individual mosaic pieces . Ernest Will, for example, argues that the local elite was a group of “grands patrons” and “protecteurs”, and that their habitus reflected an attitude “de cheikhs selon le terme arabe”,1 while Jean-Baptiste Yon draws a different picture, labelling the local elite an “aristocratie marchande” .2 In his study Les notables de Palmyre, he implicitly identifies the ruling class of Palmyra as a regime of notables, and therefore similar to other local elite groups in the Roman Near East . Yon follows — as did many before and after him — the line of interpretation according to which, under Roman influence, an institutional assimilation of Palmyra into the structures of a Greek polis took place . Recently, Michael Sommer has emphasised that Palmyra’s elite did not only provide financial assistance and access to other members of their social networks, but also military service .3 Thus, he questions the understanding of the Palmyrene elite as a leisure class of notables, and consequently of Palmyra’s institutional assimilation, by Rome, as a polis . Although all such classifications of the Palmyrene elite are predominantly based on information that can be obtained from the famous “caravan inscriptions”, most of them create a rather blanket image on the one hand, while, on the other, they show little awareness of the underlying concepts . Based on epigraphic and archaeological sources, this paper will attempt at drawing a more differentiated picture of the great men of Palmyra . It will explore their positions and functions in Palmyrene society, the mechanisms used to establish such positions, and their habitual representation .

1 2 3

Will 1957 . Yon 2002 . Sommer 2015, 179–82, Sommer 2016, 16 .

96 · An n-Christine Sander

Palmyrene caravan inscriptions and elite status Many of the inscriptions preserved to this day are related to the caravan trade, dating back to the second century AD, mainly to the 130s to 150s . Only three inscriptions date from the 190s AD .4 Altogether, eleven men are mentioned in the corpus . In the first phase, between AD 130 and 161, eight men were honoured . While three inscriptions were dedicated to Šoʿadū, two honourific inscriptions were erected for Nešā, and as many as eight inscriptions were dedicated to Marcus Ulpius Yarḥai .5 For each of the other five men only one inscription has been dicovered .6 This also applies to the three inscriptions from the second phase, from AD 193 to 199 . Therefore, as compared to the former, the latter honorands remain somewhat more obscure . For two men — Iulius Maximus and Marcus Aemilius Marcianus Asklepiades — a non-local origin is likely .7 The remaining nine were indigenous members of the local elite and have been described by scholars as either ‘sheiks’, ‘notables’, or trading aristocrats . The attentive viewer of these mainly bilingual inscriptions soon realises that the men have been honoured for different achievements .8 An explanation for this could be that their social position depended on more than one factor . In general, five types of power can be differentiated, by which members of an elite can consolidate and expand their societal position: (1) social, (2) economic, (3) political, (4) military and (5) ideological . Because explicitely or implicitely, the fifth type is never mentioned in the caravan inscriptions, this investigation will focus on the other four types of power .9

PAT 0294 (inscription for Taimarṣu), PAT 1063 (inscription for Aelius Borā) and PAT 1378 (inscription for ʿOgēlu) . 5 Šoʿadū (PAT 0197, IGLS XVII .1,127, PAT 1062), Nešā (PAT 0262, PAT 1419), Marcus Ulpius Yarḥai (PAT 0274, 1411, 1399, 1403, 1400, 0306 and 1409, Schuol 2000, no . 25) . 6 Yarḥai (PAT 1374), Julius Maximus (PAT 1397), Yarḥibōl (PAT 1414), Forat (PAT 1412), Asklepiades (PAT 1373) . 7 Iulius Maximus (PAT 1397), Iulius Maximus Asklepiades (PAT 1373) . 8 Sommer 2015, 180–81 . 9 Power means “the chance of a man or of a number of men to realize their own will in a communal action even against the resistance of others who are participating in the same action” (Weber 1968, 926) . In line with this very general definition, power can be generated from different sources (Earle 1997, 4) . Michael Mann suggests that these sources of power are overlapping networks of social interactions and they represent a means of attaining human goals (Mann 2012) . He identifies four types of power: Ideological (Mann 2012–23), economic (ibid ., 24–25), military (ibid ., 25–26), and political (ibid ., 26–27) power . In his typology of power, Mann neglects the power that an individual can generate from social relationships, which, however, is an important source of power (Earle 1997, 4–6) . For this reason, the theoretical framework of this analysis based on the typology of Mann’s sources of power is extended to ideological (Mann 2012, 22–23, similarly: Earle 1997, 8), economic (Mann 2012, 24–25, similarly: Earle 1997, 6–7), military (Mann 2012, 25–26, similarly: Earle 1997, 7–8), political (Mann 2012, 26–27) and social (Earle 1997, 4–6) power . Nevertheless, it is important to note that the importance of individual sources of power varies from one society to the next, although multiple sources exist in all societies (Earle 1997, 4) .

4

VII . Elite representation and identity building · 97 Social power refers to the social bonds that a person builds and maintains during their life . 10 In tribal societies such as Palmyra, lineage, kinship and belonging to a tribe defined one’s social position in society .11 Hence, ambitious individuals tried to manipulate such relationships in order to centralise and expand their power .12 Economic power includes financial resources and the ability to control access to goods through trade .13 By contrast, military power implies the potential of violently enforcing one’s own interests,14 which can be in line with the interests of the community at large . In this case, military actions aim at third parties . The fourth source of power involves political action, that is, the power to influence territorial regulation by institutional means, for example by holding public offices .15 Such a typology of the various sources of power provides a flexible analytical tool that can be applied to the evidence from Palmyra and the various roles in which members of the Palmyrene elite appear .

Sources of power according to the caravan inscriptions Economic power was the most frequent power source used by elites to consolidate their status in the Roman Empire . As to be expected, this can also be demonstrated for members of the Palmyrene elite . Some of the inscriptions indicate financial expenditures in order to help caravans or to subsidise the construction of temples . One example is the inscription set up in honour of Yedʿibēl, son of ʿAzizu, in AD 19 . The Palmyrenes, who lived in Seleucia, erected a statue for him because of his generosity towards their construction of the Temple of Bēl .16 A similar honour was given five years later to Maliko, son of Nešā .17 He was honoured by Palmyrene traders residing in Babylon for his financial support for the building of a Bēl-temple . In both inscriptions, the Palmyrene version emphasises the financial commitment of the honorand, while the Greek versions speak less specifically of efforts and benevolence .18 Another type of financial aid can be found in the honorific inscription for Taimarṣu, son of Taime .19 He was honoured by merchants returning with him from Spasinou Charax, as he generously offered them 300 denarii, probably as a customs 10 11 12 13 14 15 16 17 18

Ibid ., 5 . Palmyra as a tribal society: Sommer 2005b, 97, Sommer 2017a, 166–167 . Earle 1997, 5 . Mann 2012, 24; similarly: Earle 1997, 6–7 . Ibid ., 7–8 . Mann 2012, 26–27 . PAT 0270 . PAT 1352 . This observation also applies to PAT 0263 . Aqqayh is honoured by the benē Gaddibol for building a bronze gate and a bronze door, and he donated an altar, a temple and a roof of sanctuary in Vologesias . The Greek version, on the other hand, does not give any explicit reason for the award . 19 PAT 0294 .

98 · An n-Christine Sander duty .20 The honouring of a certain ʿOgēlu in the year AD 199 belongs to the same context .21 He was primarily honoured for his efficiency and his military campaigns, but also because he spent part of his own money for the community . In Palmyra, as throughout the Roman Empire, financial expenditures served as a source for elite status . However, in the caravan inscriptions, such economic power is explicitly addressed only in a few cases .22 More prominently, the inscriptions refer to the importance of social networks as a source of power as, for example, in the case of Marcus Ulpius Yarḥai . The inscriptions vaguely tell us about Yarḥai’s achievements . In most cases, it is only reported that he helped the merchants and caravans in various ways . Only one inscription refers to a specific service, namely to his assistance with the customs declaration of Palmyrene commodities .23 It is particularly interesting that this specific instance is mentioned only in the Palmyrene version . The Greek version only refers to his helpfulness in general . This suggests that the information that he helped with the customs declaration was only relevant to the local addressees of this inscription, because it was an important action for the caravan and at the same time underpinned Yarḥai’s social position . It emerges that the ability to provide financial and practical help was a marker of elite status . Other inscriptions also indicate that Yarḥai was an important individual . Except for two, all other inscriptions relating to him were located in the Agora .24 Here, Yarḥai’s services for the Palmyrene merchants and the long-distance trade were showcased to a wide audience of traders . Unlike in the inscriptions for Šoʿadū, no Roman or Parthian official appears in the inscriptions honouring Marcus Ulpius Yarḥai . Instead, the eight inscriptions relating to him seem to project the image of a family business engaged in the caravan trade .25 An inscription from AD 135 shows that Yarḥai’s brother ʾAbgar was also active in the caravan trade .26 One of Yarḥai’s last inscriptions from June AD 159 also mentions Yarḥai’s son, also named ʾAbgar, as the leader of a caravan returning to Palmyra from Spasinou Charax .27 For no other of the eleven men a similar period of operation can be proven . As far as the inscriptions tell us, Yarḥai’s exceptional endurance as a member of the elite was owing to his capacity for social networking .28 20 Schuol 2000, 81 . 21 PAT 1378 . Similar PAT 0197 in which a certain Šoʿadū is honoured along with other merits for his financial expenses . 22 If someone was honoured for helping the caravans “in all things” this may well refer to financial services . This assumption is unprovable, but it seems at least plausible . 23 PAT 1400 . 24 Agora: PAT 1411, 1399, 1403, 0306 and 1409: Temple of Bēl: PAT 1400 and Schuol 2000, no . 25, unknown location: PAT 0274 . 25 Gregoratti 2016b, 535 . 26 PAT 1397 . 27 PAT 1409 . 28 The inscription (PAT 1400) does not explicitely tell us in which way he helped the merchants in expelling the goods . Therefore, it may have been by helping financially or by accessing personal contacts .

VII . Elite representation and identity building · 99 A typical example of an individual holding significant political power is Yarḥibōl, son of Lišamšū . Yarḥibōl was honoured in AD 138 for his services as an ambassador .29 He had volunteered for an embassy to Orodes, king of Elymais . In addition, the inscription mentions decrees of the Roman governors Bruttius Praesens and Iulius Maior . This shows how crucial Yarḥibōl’s engagement was for, and beyond, Palmyrene society . Yarḥibōl maintained networks with influential people in both empires and could, as the inscription shows, act as a political gatekeeper . As an ambassador, he had the opportunity to engage in political affairs, but also to expand both his own and his city’s range of influence and connections . Yarḥibōl’s inscription highlights the importance of political power and shows that one source of power usually does not appear alone . High-ranking individuals commonly built their position upon multiple sources of power, each with varying degrees of importance . For Šoʿadū, three inscriptions have been preserved that honour him for maintaining security and order in the area stretching from Palmyra to the Euphrates .30 The first inscription, of AD 132, tells us of members of a caravan erecting, on behalf of the boulē and the dēmos, four statues for Šoʿadū, because he was generous to the caravans and to the merchants in Vologesias and had recently saved a caravan arriving from Vologesias from great danger .31 A second inscription for Šoʿadū, dating from AD 144, testifies to the erection of four other statues to be positioned near the ones set up by the first caravan .32 These statues were put up by the leaders of the caravan of all the Palmyrenes returning from Vologesias . According to the inscription, Šoʿadū protected this caravan, along with a large group of men, from the robbers led by ʿAbdallāth of ʾAḥitayʿ, who had imperiled the caravans for a long time . Just over a year later (AD 145/46), Šoʿadū is honoured by the boulē and the dēmos with four statues in Palmyra, and one statue each in Spasinou Charax, Vologesias and Genna .33 The reason given by the inscription found in Umm al-ʿAmad was his repeated and frequent benevolence, especially for the caravans, the founding of a temple for the Augusti in Vologesias, and the reliability and generosity he showed . Looking at the three inscriptions, several things stand out . On the one hand, Šoʿadū did not hold any official position . Nevertheless, he acted in the interest of Palmyrenes in Vologesias and in Palmyra itself . He protected the caravans that frequented Palmyra and Vologesias and assisted them . The inscriptions show Šoʿadū as an influential man who possessed much prestige . His prestige does not seem to have spatial limitations . The inscription from AD 145/146 shows that the influence of Šoʿadū not only reached into the Palmyrene hinterland, but extended to Spasinou Charax and thus deep into Parthian 29 30 31 32 33

PAT 1414 . PAT 0197; IGLS XVII .1,127; PAT 1062 . PAT 0197 . IGLS XVII .1 127 . PAT 1062 .

100 · An n-Christine Sander territory . This influence can be seen in the second inscription, where it is said that Šoʿadū assembled a large group of warriors to fight off the robbers . Šoʿadū had the necessary means for this: social contacts, personal prestige and material resources . In addition to economic power, Šoʿadū’s position was based, particularly, on his military power . Military skills and the necessary social connections to acquire, as well as project manpower, allowed him to develop and strengthen his position . His pedigree and social rank gave his actions credibility; there was no need to hold political offices . On the other hand, there are indications in all three inscriptions that Šoʿadū networked with Roman magistrates and even with the Roman emperor; all three highlight the fact that Šoʿadū’s behaviour received encouragement from the Roman side . Leonardo Gregoratti even suspects a regular exchange of instructions .34 Such an assumption is difficult to prove . However, the inscriptions show that, from the Roman side, Šoʿadū’s position was tolerated and even actively supported . Gregoratti rightly points to the use of the word δυναστεία in the third inscription for Šoʿadū .35 According to Strabo, δυναστεία refers to the control of an area and its resources by Arab tribal leaders .36 Since the inscription was erected by the boulē and the dēmos, it is questionable whether δυναστεία can be understood here as evidence for an informal position of power supported by Rome . It is undeniable, however, that Rome tolerated Šoʿadū’s influence on the steppe and its population . Reasons will have included economic — securing the caravan trade — as well as political considerations . The uprising in Judea and the need for Rome to concentrate troops in the western Levant reduced Rome’s capacity to protect the eastern frontier .37 Because of his position in Palmyrene society, Šoʿadū was able to exercise effective control over a vast territory, on which Rome — as the inscriptions show — relied . Another bilingual inscription, which was found in Byzantine masonry near the Temple of Bēl, dates to AD 198 and honours Aelius Borā, son of Titus, son of Aelius ʿOgēlu .38 Unlike Šoʿadū, Aelius Borā held Roman citizenship . The inscription is interesting for several reasons . First, it shows very clearly that bilingual inscriptions in Palmyra did not necessarily consist of translations of the same content .39 Furthermore, it demonstrates that each of the Greek and Palmyrene versions of one and the same bilingual inscription could have been intended for a different group of recipients . Both versions mention that Aelius Borā was a stratēgos . The Greek version calls him a stratēgos “for peace”, while the Palmyrene version honours the stratēgos Aelius Borā for securing peace within the city . However, in neither version is there any direct evidence for his involvement in the caravan trade . Instead, both versions praise Aelius’ dedication 34 35 36 37 38 39

Gregoratti 2015, 143 . Ibid ., 143–146 . Strab . 16,1,27 and 3,2, see Gregoratti 2015, 144–46, Gregoratti 2016b, 532 . Gregoratti 2015, 147–48 . PAT 1063 . Taylor 2003, 10–14 .

VII . Elite representation and identity building · 101 and his merits . However, while the Palmyrene version states that his peacekeeping on behalf of the city is the main reason for him being honoured, the Greek version refers more specifically to his achievements . Not only was Aelius Borā a stratēgos, he also had courage and efficiency, was faithful to his god Yarḥibōl and loyal to the Roman governors and his fatherland . The Palmyrene version of this inscription names the tribe of the benē Komarē as donors, who received permission to erect the statue through a decree of the boulē and the dēmos . The Greek version, however, reports an equestrian statue to have been set up by the Roman governors Manilius Fuscus and Venidius Rufus, and four other equestrian statues that the four tribes are said to have erected at their own expense .40 In this version, the honour was given by order of the boulē and the dēmos .41 It is noticeable that the Palmyrene version has a purely local frame of reference: a Palmyrene man is honoured by the local population, in return for achievements on behalf of his hometown . The Greek inscription, on the other hand, has a somewhat wider spatial bearing . In this version, emphasis is put on Aelius Borā’s services to his fatherland,42 although it remains unclear which geographical area is meant by the expression πατρίς . In this context, it is striking that the Greek version names two Roman governors as donors . Mentioning them could refer to a broader, regional frame of reference . The word fatherland, for which there is no equivalent in the Palmyrene version, implies a larger geographical realm than just the city of Palmyra itself . Consequently, Aelius Borā’s services must have been of interest not only to the local population, but also to the two Roman governors and to Rome as well . This raises the question why one version has a purely local focus, while the other implies a broader, at least provincial scope . On inspection, the inscription helps to resolve the conflict . Aelius Borā is referred to as a stratēgos or protector of peace in both versions . Especially the Palmyrene version focusses on peacemaking within the urban area . Unlike Šoʿadū, Aelius did not secure peace along the caravan routes, but within Palmyra . This assumes that there was a need to provide peace within the city of Palmyra . Since both versions of this inscription do not mention any specific occasions, as in Šoʿadū’s inscription from AD 144, it seems likely that the disturbances must have been of a common, everyday nature . However, as the Greek version at least suggests, those disturbances must have been of interst to the Romans, too . Aelius Borā’s inscription was put up in February AD 198 . At that time, Septimius Severus was campaigning against the Parthians in Mesopotamia .43 The Palmyrene version of the inscription does not specify why peacekeeping was necessary in the city . So, trying to 40 Ti . Manilius Fuscus was first governor of the newly created province of Syria Phoenice from AD 194; Q . Venidius Rufus Marius Maximus L . Calvinianus from AD 198 onwards . 41 It should be noted that the first three lines — assumingly containing the words boulē and dēmos — of the Greek version have been reconstructed based on the Palmyrene version . 42 καὶ ὑπὸ τῆς πατρίδος (l . 12–13) . 43 Millar 1993, 118 . 121, Sartre 2005, 350–51, Sommer 2005b, 219, Young 2001, 155 .

102 · An n-Christine Sander identify Septimius’ campaign as a reason for unrest in the city might seem somewhat farfetched . The Greek version, on the other hand, implies precisely such a connection . First, in this version, Aelius is honoured for being responsible for peace and for his merits in the service of the fatherland . In addition, two Roman magistrates honour him for policing the area, which, one should assume, would have been rather the task of the Roman government . Now one could argue that Aelius held a post as a stratēgos: was it not then his job to ensure peace? But then what was the reason for putting that much emphasis on this aspect? Here, the Palmyrene version provides a clue . In Palmyrene, Aelius is referred to merely as a stratēgos (ʾsṭrṭgʾ) . Unlike in the Greek version, there is no dynamic component . It seems to be a personal attribute rather than the description of an office . However, the Greek version, by indicating that Aelius Borā held the office of stratēgos more than once, has a diachronic dimension . Despite this wording in the Greek version, stratēgos seems to have been more of an ad hoc created position than an institutionalised office . Consequently, the inscription reports that because of his position, Aelius Borā was able to enforce order and security on Palmyrene soil and maybe even beyond, thus relieving the governors, whose forces were thinly stretched due to the Parthian campaign . Hence, the donation of this statue must be understood as a recognition of his position of power . However, things are different when it comes to the honours given to ʿOgēlu in AD 199 .44 The inscription lists ʿOgēlu’s valour, his many military campaigns, and his financial expenses for the caravans among the reasons for honouring him . Economic and military power are explicitly mentioned side by side . Unlike in the inscriptions for Šoʿadū or Aelius Borā, no Roman magistrate is mentioned . This could suggest that ʿOgēlu’s achievements had a purely local significance . However, they must have had considerable local relevance, since this inscription was the only one of the caravan inscriptions mentioning boulē and dēmos without referring to Roman magistrates, too . The Greek inscription explicitly speaks of κατὰ τῶν νομάδων στρατηγίας . However, the addition “against the nomads” is not found in the Palmyrene version of this inscription . In this respect, it is questionable whether ʿOgēlu’s military campaigns were actually directed against nomads . By no means can this wording be used to imply a separation between the sedentary population and the nomads of the steppe . However, the inscriptions for Šoʿadū, Aelius Borā and ʿOgēlu reveal what the local population expected from the members of the local elite . They should ensure peace within the city, as well as in the surrounding area, as can be seen from another inscription from August AD 21 .45 In this inscription, Hašaš, son of Nešā, is honoured for having mediated between members of the benē Komarē and the benē Maṭṭabōl . Such actions would have been unthinkable, unless they projected social and military power . What was expected from people like Hašaš is

44 PAT 1378 . 45 PAT 0261 .

VII . Elite representation and identity building · 103 consistent with local traditions as reflected in the documents from Mari .46 Members of the elite, who presided over the local tribes, acted as judges and ensured peace in situations of unrest and discontent . The inscriptions clearly show that the members of the Palmyrene elite based their elevated societal rank on a diverse range of power resources . Unlike many other local elites in the Roman Empire, whose status depended on economic power alone, which was then converted into prestige, Palmyrene elite membership resulted from a combination of economic, military, social, and political power . However, the inscriptions convey the picture that military achievements could only establish a position of power if there was a specific threat .

Imagining Palmyra’s elite: camels, horses and warriors While the inscriptions primarily allow conclusions to be drawn about how the elite was perceived in the community, archaeological findings from the funerary context offer insight into the self-image of its members . Particularly revealing in this context are the depictions of camels and horses, which can be found in various contexts in Palmyra and the Palmyrene hinterland . A fragmentary camel representation can be found on a sarcophagus from the early 3rd century AD, which was discovered during excavations in the temple tomb no . 36 in the Western Necropolis (Fig . VII .1) .47 According to Andreas Schmidt-Colinet, the building was the final resting place of Odaenathus’ second in command, Septimius Worōd, and his family .48 The camel representation on the sarcophagus was taken from the local repertoire of pictures . At the centre of the pictorial arrangement is a man in Parthian costume . He wears a belt, but no armament . His clothes match the representation of other Palmyrene elite members and identify him as a member of their circle . Whether the man is indeed Worōd cannot be said for sure . The man’s body is turned slightly towards a camel on the left, whose rein he holds in his left hand . The camel, whose legs are no longer preserved, wears a plaid saddle pad, which is set off with a decorative border . The saddle is only partially preserved and fastened with a wide girth and three girth straps . At the level of the rear flank, a shield is attached to the saddle . With the right hand, the man leads a horse, which stands to his right, the forehand facing him . The legs of the animal are no longer preserved, yet it can be observed that the horse was originally depicted with a saddle . The tail strap as well as the decorated girth, are recognisable . Just behind the saddle is a decorated quiver with arrows attached . Undoubtedly, this horse was a mount with a military 46 Klengel 1972, 114, Sommer 2005b, 207 . 47 Schmidt-Colinet et al . 1992, plate 33a S5 . 48 Ibid ., 40–42 . Similarly: Sommer 2005b, 195–96 .

104 · An n-Christine Sander

VII.1: Sarcophagus from the early 3rd century AD. Found in temple tomb no. 36.

purpose . The same is true for the camel . Saddle and shield are sufficient proof that a function as a pack animal can be ruled out . In addition to camel representations, representations of horses are a similarly common motif on Palmyrene sarcophagi and reliefs . Mostly, they form part of representations of families or are depicted as mounts of male persons . As with the camel portrayals, it is important to distinguish between gods riding horses and men with their horses . The horse representations accompanied by male humans are all consistent in that the horses are always portrayed with magnificent saddlery and, hence, as riding animals in a martial context .49 One such example is the tomb monument of Maqqay (Fig . VII .2) .50 The sarcophagus is almost completely preserved . It is likely that the sarcophagus lid shows Maqqay or a male relative . He is lying on a kline and wears richly decorated clothes typical of Palmyra’s elite . In the background, he is accompanied by three people . Although the heads are missing, their clothes indicate that at least the figures on the left and in the centre are women . The numerous necklaces and the bangle of the woman on the left suggest that she is his wife . The person in the centre also wears jewelry . It can be assumed that she is their daughter . In all likelihood, the figure to the right is their son, as he wears no jewellery . Here the repre49 Regarding the element of standardisation of Palmyrene camel depictons: Seland 2017, 108 . 50 Schmidt-Colinet et al . 1992, plate 69b .

VII . Elite representation and identity building · 105

VII.2: Tomb monument of Julius Aurelius Maqqay. It dates from 229 AD.

sentation as head of the family clearly comes to the fore . The front of the sarcophagus box shows three men . They all wear richly decorated ‘Parthian’ robes and swords . All men have swords hanging from their belts on the left side, while their left hands grasp the sword grips . The man on the left also holds the reins of a horse in his right hand . The horse wears a riding saddle with saddle pad, chest and tail straps . A quiver with arrows is attached to the saddle . Undoubtedly, this horse was used not only as a mount, but also in combat . There is little doubt about the horse’s function: its purpose is to reveal the high social standing of the sarcophagus’ owner . Not only was he the head of his family, but also in possession of military strength and power . His demeanor, as represented by the sarcophagus, demonstrates that he himself wanted to be perceived as an influential warrior and family man . The consistently epichoric costume of the people depicted makes the viewer quickly realise that the deceased was deeply rooted in his local culture . Through the imagery of the sarcophagus, it becomes clear to the viewer that this man was a member of the local elite, who acted as such in the context of local tradition . Family and martial habitus gave his elite status the credibility it required . Camels and horses were indispensable companions for the steppe inhabitants and, as the Palmyrene depictions show, they also were of significant importance to the local elite

106 · An n-Christine Sander in Palmyra . Camels and horses functioned as powerful status symbols in the elitist selfrepresentation .51 They were the touchstones of economic and military power . A military habitus was an important component in the self-representation of Palmyrene men . The majority of Palmyrene camel and horse representations dates to the end of the 1st and 2nd centuries AD . For the time before, no representations of men with camels are known . Depictions of military mounts emerged concomitantly with major urban and social changes in Palmyra . New public and religious buildings transformed the urban landscape of Palmyra for good .52 Along with the reorganisation of public space, changes in the social composition of the city took place .53 The inscriptions evoke the image of a society that was increasingly institutionalised, at least at first sight .54 While the tribes seem to be dwindling from the visibility of public space since the middle of the 1st century AD, boulē and dēmos make their appearance in public inscriptions . Their appearance in the epigraphic record is usually cited as an indication of the formation of Palmyrene citizenship .55 The inscriptions, however, provide no indication as to whether boulē and dēmos were newly established civic institutions, such as they were known from other poleis . However, it seems likely that in Palmyra the terms boulē and dēmos were used to identify already existing bodies .56 The existence of such councils of elders, tribal assemblies and similar gatherings is sufficiently documented for pre-Roman societies in Syria .57 In addition, boulē and dēmos only appear in a few Palmyrene caravan inscriptions .58 While the majority of these caravan inscriptions names merchants and caravans as initiators, boulē and dēmos only appear in the epigraphic record when the honourand possessed military or political power in the first place, which he used in favour of the Palmyrene population . While, for instance, Aelius Borā was honoured for his military accomplishments in AD 198, Yarḥibōl, son of Liṣamšū, was honoured for his service as an ambassador to Orodes, king of Elymais .59

51 Similarly: Seland 2016, 69: “[…] they wanted to be remembered, but either military men, or associated with a nomadic, aristocratic, and martial lifestyle” . 52 Smith II 2013, 25 . 53 Sommer 2017a, 196–207 . 54 Ibid ., 196; also: Sommer 2005b, 171 . 55 Andrade 2012, 75, Andrade 2013, 191–93, similarly: Smith II 2013, 57–58 . 125–132 . 56 Arguing along similar lines: Sommer 2005b, 173, Sommer 2017a, 198–99 . 57 Klengel 1972, 131–34 . 58 PAT 0197, IGLS XVII .1,127, PAT 1414, 1062, 1063, 1378, 0282, 0288 and 1360 . For another inscription (Schlumberger 1961) the mentioning of boulē and dēmos is much likely, because an unknown member of the Aʾby family is honoured for his archonship of Mesene . Hence, he was in possession of political power . The reconstruction of boulē and dēmos in the first line of the inscription would be congruent with the inscription of Yarḥibōl (PAT 1414), who was also honoured for his political achievements . Additionally, Yarḥibōl was also a member of the Aʾby family . Thus, he could have been honoured for being the archon of Mesene (Schuol 2000, 63) . 59 Aelius Borā: PAT 1063, Yarḥibōl: PAT 1414 .

VII . Elite representation and identity building · 107 It is also striking that most inscriptions mentioning boulē and dēmos as the honouring bodies also refer to individuals on whose initiative the honours were conferred: in many cases Roman magistrates .60 Such honourific inscriptions have a different quality from the inscriptions solely initiated by merchants and caravans . While, for example, the services provided by Marcus Ulpius Yarḥai were only of importance to certain groups of the local population, military and political services were of particular interest to all Palmyrenes, and, as shown for Šoʿadū and Aelius Borā, in some cases to Roman governors, too . The significance of such inscriptions went far beyond the local context . It is notable that the manner in which honorands are praised differs between the Greek   and the Palmyrene versions of the inscriptions . The wording of the Greek versions matches the set phrases used in honourific inscriptions of other Greek poleis of that time, in which the “γλυκυτάτη πατρίς” is a central motif of the local elites’ collective identity .61 In the Palmyrene versions, however, an equivalent attribution is missing . This can only mean that the two versions of the inscriptions were aimed at different groups of addressees: an imperial-regional, Greek-speaking community, and a local, Palmyrenespeaking collectivity .62 Thus, if one wants to understand how rank was established inside the local community, one needs to focus on the Palmyrene versions . This would also indicate that, unlike for example in Asia Minor, Palmyra’s honourific inscriptions were not part of a competitive elite system, in which members of local elites competed locally and regionally with each other . Instead, the Palmyrene caravan inscriptions are testimonies of a society organised by kinship, in which social positions were already set by lineage .63 The continuing importance of descent is indicated by the filiations given in the caravan inscriptions until the 3rd century AD . Honourary inscriptions, hence, were a way of displaying the successful fulfilment of the public role by local elite members, as it was expected by the community .

Conclusion Among scholars, there are still various conceptions of the Palmyrene elite . While some historians paint an image of distinguished urban notables, others emphasise combative inclination of Palmyra’s elite . All reconstructions are based on the caravan inscriptions from the oasis, because they provide information about the deeds and merits of individual elite members .

60 PAT 2754, 0197, 1414, 1062 and 1063 . 61 Caravan inscriptions referring to “fatherland lovers”: Yarḥibōl: PAT 1414, Šoʿadū: PAT 1062, M . Ulpius Yarḥai: PAT 1403 and 1400 . γλυκυτάτη πατρίς as a recurring motif: Stephan 2002, 330–31 . 62 Sommer 2005b, 123 . 63 Sommer 2016, 15 .

108 · An n-Christine Sander The underlying principle of elite status is the possession of power . This holds also true for Palmyra . Power can be derived from different sources . Using a theoretical framework based on the typology of Michael Mann’s sources of power and complementing Timothy Earle’s component of social power, the sources of power drawn upon by Palmyra’s great men were examined . The inscriptions paint a colourful, varied picture . While economic and political power is only explicitly mentioned in some cases, the epigraphic record implicitly demonstrates the importance of social connections as a source of power . Particularly revealing is the analysis of those inscriptions that explicitly refer to military power . The phrasing and the dating of such inscriptions clearly prove that military power played an important role both for the local and regional population, especially in times of political unrest . The references of Roman magistrates in the inscriptions for Šoʿadū and Aelius Borā add further proof to this . Equally noteworthy is the fact that the exercise of military power was not necessarily tied to public offices, as the inscriptions of Šoʿadū show . The specific importance of military power as a local particularity is reflected in the self-portrayal of the elite, as shown by camel and horse representations in the funerary context . Unsurprisingly, military and political power did not only possess local appeal . The honour inscriptions for members of the elite, who held such power, often refer to boulē and dēmos and Roman magistrates as initiators . The bearing of such epigraphic testimonies went beyond the local context . Additionally, the way the honorands are praised in the epigraphic record differs in the Greek and Palmyrene versions of these inscriptions . Accordingly, the two versions of the inscriptions were aimed at two distinct groups of addressees: an imperial, Greek-speaking one and a local, Palmyrene-speaking one . Obviously, specific sources of power were more important to one group of addressees than to others . Due to the geopolitical situation, Palmyra’s great men were faced with both the challenge and the chance to develop their elite status from a wide variety of power sources . Unlike many other local elites across the Roman Empire (whose position depended on the convertibility of economic power into prestige), elite membership in Palmyra was based on flexible combinations of economic, political, social, and military power . Military power as a source of elite status was not usually found in the Roman Empire’s local elites of notables; it was a specific feature of the Palmyrene elite . Palmyra’s ruling class was no leisure-class of notables, but rather represented themselves as frontier warriors, and economic, as well as political gatekeepers .

VIII. Μείζονα πράγματα. Persians, Romans and Palmyra in Zosimus’ historiography Georg Müller

The city of Palmyra occupies a special place in Zosimus’ account of Roman history .1 In order to investigate the intended effect on the reader, it is necessary to understand its position in the first book of the Nea Historia . Zosimus’ main interest lies in the facts (πράγματα), which are used in order to show how the decline of Roman power took place .2 In the belief that the facts speak for themselves, he normally restrains from delivering judgements, although he evaluates each emperor’s policy starting from Septimius Severus . Such assessments are based upon moral considerations, while the administrative and political categories, the patterns or guidelines according to which the protagonists have or ought to have acted, are largely absent . Often, they can be deduced e silentio . This is the purpose of the present paper .

Zosimus’ historiographic technique A lot has been written on the editorial condition of the Nea Historia, mainly in a negative tone . Otto Veh, for one, speaks of a “torso” and “clear traces of immaturity”,3 Wilhelm Enßlin assumes an anticipatory technique, as well as “abridgements”,4 Johannes Straub counts “summary notes”,5 which are interspersed in the real text . Whatever form the book’s final version may have had, guidelines and intentions can be identified in the version we know today .

1

2 3 4 5

The purpose of the present paper is not to use Zosimus as a source, in order to gain historical information or prove the author wrong . Rather, it is intended to define the perspective from which he views history . What is important to Zosimus? How does he create ‘his’ history? And above all: how do his readers perceive this history? This paper aims at answering such questions using the story of Palmyra’s meteoric rise and fall as a case-study . To this he hints in the introductory chapters, but the thought is explicitly developed only in the description of Palmyra . 1,57,1: ὅπως ἐν οὐ πολλῷ χρόνῳ σφῇσιν ἀτασθαλίῃσιν αὐτὴν διέφθειραν ἔρχομαι λέξων . Veh 1990, 5 . Enßlin 1949, 19 on 1,27,2 Persians; 23 on 1,26 epidemic and 1,27,1 ‘Scythians’ . Abridgements: 51 and passim . Straub 1952, 49 .

110 · Georg Müller First, let us have a look at Zosimus’ historiographic technique . He uses five focalizing approaches: in a sense, hors catégorie are systematic, structural considerations such as the philosophy of history, political points of view, religious beliefs, etc .6 The wide-angle (zoom level 1), that is, an overall view of the empire, its condition and its perspectives, is rarely used7 . More frequently, however, zoom levels 2–4 are used, highlighting (2) regional hotspots of foreign policy, (3) local events, individual cities and areas, and providing (4)  detailed descriptions of events of local significance . As a rule, Zosimus focuses on one large theatre and rarely resorts to structuring wordings8 to mark a scene change . The description of details has a twofold purpose: first, of course, to provide information about the matter itself, to give plasticity to the description, to make it exciting, to make the historical valuation plausible . On the other hand, it is intended to be exemplary in that it should support the explanations on the higher levels by allowing the reader to make analogies, offering them patterns for the cases in which no details are provided . Already in the programmatic anti-monarchic fifth chapter of the first book, Zosimus points to a phenomenon, which is constitutive of the central government’s policy on the one hand, and of any representation of an imperial history on the other: the enormous size and variety of the Roman Empire,9 which only sporadically allows for considering details . The historian is thus faced with the task of describing events in major lines, in order to simplify complex contexts for the reader, as well as making assessments, in order to emphasize the choices of options for individual agents . The abstract thinking, the meaning and the character of institutions — of the empire, its centre, cities, army and elites – stands on the surface of the text behind the real thinking, above all behind the description and evaluation of the actions of single persons (monarchs and other leaders, councillors, sometimes even persons without office) or groups (the general public, the senate, elites, army, nations, citizens or residents of a city) . Since these people are always given their own decisions, their own initiative and responsibil-

6 7

8

9

Τύχη’s great plan dominates, but Zosimus refers to it only intermittently . The sections of the respective zoom levels are not to be discussed here . This is attributed to the author’s sources and aims . Zosimus reports at length facts on the eastern half of the empire only . When Zosimus boastfully speaks of ‘the empire’ (‘all the provinces’ etc .), he sometimes only refers to the parts in his focus . Cfr . Veh 1990, 8: “Nicht ungeschickt führt er uns zwischen Hauptkriegsschauplätzen im Osten wie Westen  des  Imperiums hin und her […] .” For instance 1,30,3: τὰ […] παροικοῦντα τὸν Ῥῆνον ἐν τούτοις ἦν (“This is how it looked on the Rhine”); 1,40,1: οὔσης τοιαύτης τῆς ἀμφὶ τὴν ἀνατολὴν καταστήσεως (“Such was the situation in the East”) . Significantly, Zosimus presents Gallienus’ macro-perspective who had been forced to keep track of more than one hotspot . Thus, the reader may have got an impression of his strong nerves and strategic foresight, as in one case Gallienus stays where he is, while in the other he swiftly leaves for Italy . 1,5,3: οὐκ ἂν ἀρκέσοι πᾶσιν κατὰ τὸ δέον προσενεχθῆναι, τοῖς πορρωτάτω που διακειμένοις ἐπικουρῆσαι μὴ δυνάμενος ἐξ ἑτοίμου, ἀλλ’ οὔτε ἄρχοντας τοσούτους εὑρεῖν οἳ σφῆλαι τὴν ἐπ’ αὐτοῖς [οὐκ] αἰσχυνθήσονται ψῆφον, οὔτε ἄλλως ἠθῶν τοσούτων ἁρμόσαι διαφοραῖς .

VIII . Persians, Romans and Palmyr a in Zosimus’ historiogr aphy · 111 ity,10 which have an influence on the condition of the polity in question, they should not remain on an abstract level . Zosimus gives the reader all the means to do the abstraction themselves, thus pursuing a very specific goal . Despite, after all, showing a tendency of pursuing failure, he wants to explain Rome’s decline . Instead of morally justifying failures or condemning decisions or decision makers, if something went wrong, he lets the reader form their own opinion on the basis of facts . Perspective is also an important tool in this context . Things are by no means considered from the same point of view . The (1) Roman perspective can be distinguished from the foreign one (Scythian, Persian), but then also (2) the different inner-Roman perspectives stand against the bird’s eye perspective of the centre, or even the grassroot perspective of single individuals .

Zosimus’ Palmyra In his history of Rome, which is primarily based on the succession of its rulers, Zosimus has embedded the Palmyra episode within the period between Valerian and Aurelian (1,29–62), dedicating to it as many as fourteen chapters .11 For Zosimus, imperial policy is primarily foreign policy and therefore it manifests itself in a few hotspots of permanent defensive action . The cobweb in which the central power is trapped, consists essentially of three threads: (1) severe military problems with the raids of neighbours in the north, northeast and east, (2) the frequently mentioned and consequently serious threat of epidemics, which could paralyse the defensive forces of the empire,12 and obviously, (3) the rampant usurpations, which steadily thwarted a regulated defence of the borders and delayed the appropriate countermeasures . The events surrounding Palmyra are only understandable in the global context of the political and strategic situation in the middle of the 3rd century . Zosimus is surely aware of this, as shown by his assessment of the situation in chapter 1,30 . Obviously, his sources 10 Without discussing this systematically, Zosimus, against all stereotypes, attributes autonomous thought and action even to subaltern agents . The army, for instance, displays a surprising and almost altruistic foresight when it removes Aemilianus because he views things from a too soldierly perspective and too little as a politician, 1,29,1: ἀρχικῶς μᾶλλον ἢ στρατιωτικῶς . 11 Hartmann 2001, 26: “Die Ausführlichkeit seiner Berichte über Palmyra legt die Vermutung nahe, dass Zosimus im ‘Aufstand’ der Palmyrener neben den Raubzügen der ‘Skythen’ eine der größten Existenzbedrohungen für das Reich im 3 . Jahrhundert sah .“ 12 Mentioned three times: 1,26 (so far unseen extent: λοιμὸς […] οὔπω πρότερον […] τοσαύτην ἀπώλειαν ἐργασάμενος); 1,36 (annihilates a large part of Valerian’s army; λοιμοῦ […] τὴν πλείω μοῖραν [τῶν στρατοπέδων] διαφθείραντος; unprecedented extent λοιμὸς […], οἷος οὔπω πρότερον […] συνέβη); 1, 37 (likewise unprecedented disaster λοιμὸς ἐπιβρίσας ταῖς πόλεσιν, οἷος οὔπω πρότερον ἐν παντὶ τῷ χρόνῳ συνέβη) . — This is not the place to discuss the epidemics, as they were, to Zosimus, natural phenomena and as such beyond reach for political agents .

112 · Georg Müller do not provide the same amount of information on all fronts, so his descriptions are quite unequal, the first book overemphasizing the events on the Danube and its backdrop, and in Asia Minor (politically and geographically the area may be described as the ‘Scythian frontier’) . This is not the place to investigate the sources used by Zosimus,13 however, it is worthwhile to analyse what the readers of the Nea Historia knew beforehand, what their expectations were, and how far Zosimus went to fulfill them .

The situation prior to the Palmyra episode The enemy: the Persians The Persians appear three times before chapter 1,39, briefly in each case, yet a coherent image is created . After the prosperous14 period of the Severans15 (which was marked only by domestic problems) and their immediate successors, Maximinus and the Gordians, dark clouds shadow the political climate . This is analysed in chapter 18, its root being the beginning of the Sasanian rule with their expansionist policy .16 This is how Zosimus presents the Persian threat as the first destructive external moment in the imperial chronology, which until then had only been threatened by internal discord . The reaction of Gordian III, a massive pre-emptive strike,17 starts auspiciously, only to fail because of problems between Roman army leaders . The fact that Philip the Arab enters into a treaty of friendship with Shapur instead of continuing to vigorously fight the Persians, obviously benefits the consolidation of his usurpation and provides no indication of a particular Persian threat .18 For the second time, the Persians are mentioned in a very different context . Here they are no longer an individual phenomenon, no generis sui problem, but merely one element in a whole new world-order . This is dramatically worse for the Roman centre, since it leads, because of its suddenness, to a catastrophic overstraining of resources . The main problems here, as mentioned in Chapters 1,26–27, are the ‘Scythians’, the plague and the Persians with their incursions into Roman territory the impact of which is exagger-

13 No more relevant are the chronological issues, as discussed by Alföldi and Straub . Cfr . Paschoud 2000–2003, Bd . 1, 154–158 . 14 1,18,1: βασιλείας οὐσης ἐν ὀχυρῷ . 15 Septimius Severus is explicitly praised for his skilled administration; the government of Alexander Severus is likewise seen under a positive light . 16 Zosimus remains silent about the impact, constructed by modern scholarship, of the Severan eastern policy on the fall of the Parthian Empire . 17 1,18,2: παρασκευῇ τῇ πάσῃ . 18 In 1,32,4, the treaty is heavily criticised (εἰρήνην αἰσχίστην) while Zosimus states that the Persians did not make any territorial gains . Thus Enßlin 1949, 17 .

VIII . Persians, Romans and Palmyr a in Zosimus’ historiogr aphy · 113 ated .19 The empire becomes ungovernable .20 In the short subchapter 1,27,2, the Persians simply seize an opportunity: now they appear as a real danger and masters of the situation, who decide strategically21 and thus achieve maximum success . For their part, they avoid an overstretching of power by limiting themselves to the conquest of Mesopotamia and by renouncing Asia (Minor), which they could have easily captured, but very likely could not have defended .22 Instead, they maximize their loot and send out a signal by completely destroying Antioch and by enslaving or killing all of its inhabitants . Undoubtedly, they benefit from the complete paralysis of the Roman administration . But first and foremost, their clever and yet uncompromising attack justifies the assessments expressed in 1,18 . The Persians are invulnerable, terrifying opponents, who disappear after their raids, returning to where they came from . The third act: 1,36 is a combination of 1,18 (Roman strategy and situation at command  level) and 1,27 (worsening conditions) . The Romans have become aware of the Persian danger, which is embodied in Shapur, and mobilize their best troops headed by  the  emperor (= 1,18) . A swift action against the Persians is however prevented by leadership problems23 (= 1,18) resulting from the war on two fronts (= 1 .27), as well as by epidemics breaking out among the troops (= 1 .27) . Shapur bides his time to attack the Romans and occupy the entire east (= 1 .27) . He apparently estimates the propaganda effect24 of capturing Valerian to be of higher value than the pecuniary gain obtained from the tributes offered by the Romans . This humiliation is as much unique as it is spectacular, and Zosimus attributes it to the weak character of the emperor25 (not to his strategy) .

19 The ‘Scythians’ are looting all Roman provinces, 1,26,1: μηδὲ ἕν ἔθνος […] ἀπορθητὸν καταληφθῆναι; the epidemic rampages as never seen before, 1,26,2: ὁ λοιμός […] οὔπω πρότερον τοσαύτην ἀνθρώπων ἀπώλειαν ἐργασάμενος; the Persians destroy Antioch entirely without meeting any (οὐδενὸς παντάπασιν ἀντιστάντος) resistance; they would have conquered the whole of Asia if they had intended to do so . 20 1,27,1: τῶν κρατοῦντων οὐδαμῶς οἵων τε ὄντων ἀμῦναι τῷ πολιτεύματι . That this was possible is due to the irresponsibility (ἐκμέλεια) of Philip the Arab and Trebonianus Gallus, 1,23,1: διὰ τὴν Φιλίππου περὶ πάντα ἐκμέλειαν; 1,26,1: ἐκμελῶς […] τοῦ Γάλλου τὴν ἀρχὴν μεταχειριζομένου; 1,28,3 also on Trebonianus Gallus: τὸ ἐκμελὲς καὶ το ἀνειμένον τοῦ ἀνδρός); the course of events cannot be changed by the exceptional Decius (ἄριστα βεβασιλευκ[ώς] . 21 1,27,2: διενοήθησαν . 22 Enßlin 1949 rightly observes that Zosimus makes an anticipating assessment of Shapur’s conduct of war; however, he goes too far when he concludes with Alföldi 1937, 64, “daß Sapor […] in Wahrheit keine organisierte Besitznahme vorhatte, sondern nur sengen und plündern wollte” . — More differentiated on the Persian strategy and the conditions under which it operated Enßlin 1949, 55: even after 260 the Romans still have available substantial defensive power . 23 Valerian dreads a usurpation . This is why he attacks the ‘Scyths’ . 24 See RGDS 19–36 and the representations of the rock reliefs of Naqsh-e Rustam . 25 1,36,2: Οὐαλεριανοῦ διὰ τε μαλακίαν καὶ βίου χαυνότητα .

114 · Georg Müller The centre: Valerian and Gallienus As far as chapter 18, Zosimus has stated that the situation at the beginning of the Persian campaign by Gordian III had been stable . The following rulers, however, cut a very poor figure . They lose grip,26 and consequently the empire becomes ungovernable .27 Valerian then starts with the bonus of “general approval” (κοινῇ γνώμῃ) . Even the troops of his opponent Aemilianus consider him the better ruler for being less soldierly than their own leader . First of all, Zosimus appreciates his engagement (σπουδή) and approves his goal: the restoration of order (τὰ πράγματα εὖ διαθεῖναι) . This rather trivial assessment is the political credo of the administrator Zosimus . In the basic statements about the crisis, there is constant talk of shock and disorder28 (not of decline, loss of power, collapse, etc .) . Subsequently, it is emphasized that the middle level of the empire resort to active self-help to defend themselves from the Scythians . Thessalonike’s citizens breach the siege all by themselves, Athens rebuilds its walls and the Peloponnesians bar the Isthmus . All these actions appear systematic and well organized, there are no reproaches to the centre for having been inactive and having left the province to its fate . This widens the scope of the centre . Valerian, who has just been extolled for his political virtues, is worthy of the praise and after judging the situation,29 takes action by systematically grading the effort put into defence of individual borders according to the risk factor . The highest priority is given to the Persians’ repulsion, and this task is taken by the emperor himself (see above) . His son Gallienus appears three times before ‘Palmyra’: Having been appointed co-ruler30 by his father and equipped with a comprehensive, independent military command, his task was to protect Europe from the repeated intrusions of barbarians . Gallienus acts as systematically as his father .31 Like him, he assesses the situation,32 and then further divides his sphere of power into two zones: a high-risk zone (Germany), which he takes under his direct command, and a lower-risk area (Italy, Illyria, Greece) where local commanders have to cope by themselves . This should be plausible enough, and the positive image is further enhanced by his defensive successes in 26 1,23,1: διὰ τὴν Φιλίππου περὶ πάντα ἐκμέλειαν . 27 1,27,1: τῶν κρατοῦντων οὐδαμῶς οἵων τε ὄντων ἀμῦναι τῷ πολιτεύματι . 28 1,20,2: πολλῶν […] ταραχῶν, 1,7,2: μικροῦ […] συνεταράχθη, 1,23,1: τῶν πργαμάτων […] ταραχῆς πληρωθέντων, 1,37,3: πάσης τῆς […] ἀρχῆς […] σαλευομένης . 29 1,30,1: Συνιδών τὸν πανταχόθεν ἐπικείμενον κίνδυνον . 30 1,30,1: τῆς ἀρχῆς κοινωνό[ς] . 31 According to Enßlin 1949, 39, the combination of the words πανταχόθεν and ἁπανταχόθεν in two subsequent sentences is clumsy . He concludes that Zosimus must have used an abridged version of his source in two different instances . However, this argument is by no means compelling, as it seems to be Zosimus’ aim to highlight the harmony between the two emperors . Besides, πανταχόθεν is used in the same sentence again in the context of barbarian incursions . 32 The introductory ὁρῶν is conspicuously parallel to συνιδών, which refers to Valerian; the linear-durative aspect indicates that Gallienus was always master of the situation .

VIII . Persians, Romans and Palmyr a in Zosimus’ historiogr aphy · 115 Greece . Great as the danger may be, the reader has no choice but find the planning coherent and in line with the factual necessities . Zosimus does not underpin this surprising assessment with detail: in the narrative, the Germanic danger appears, so to speak, out of nowhere . This may be due to the scarcity of his sources .33 Not a word on the fact that the usurpation of Maximinus happened in connection to the insecure situation on the Rhine . Surely, Zosimus has only poor information about the situation in Germany: this transpires from the very restrained, almost nebulous phrases he uses when speaking of a negotiating partner as τινα τῶν ἡγουμένων ἔθνους Γερμανικοῦ .34 No locations, tribes or individual leaders are named . Nevertheless, he draws — from a distance — a precise picture of the situation and of the capabilities and possibilities of the emperor: a sharp analyst, excellent strategist, smart tactician, energetic and courageous army leader,35 as well as a successful foreign politician with a cool, calculating mind . True to his differentiated assessment of the situation, in Germany he opts for a in three steps defensive tactic, dictated by the striking numeric inferiority of his troops . He (1) tried to stop the Germanic tribes before crossing the Rhine . This was not only preferable from the point of view of the protected people, but also tactically useful showing the emperor’s operational vision . If this was not possible, he would attack (2) the enemy at their most vulnerable point while crossing the river . In order to reduce the pressure from the outset, he adopted (3) a diplomatic strategy in order to enhance the border security which Zosimus did not approve: putting into practice a divide et impera policy, he allied himself with a Germanic prince . Zosimus had commented on similar treaties36 signed by Philip the Arab and Trebonianus Gallus in a much more unfriendly fashion . Gallienus, in contrast to his predecessors, decides to swallow the bitter pill only after careful consideration,37 having a rather modest target in mind: through this alliance, all he expects to reduce is the risk of invasion . His expectations are even surpassed, for the new ally operates within a vast space, thus preventing the other barbarians from crossing the Rhine . In chapter 37, Zosimus offers a brief, but disastrous view of the overall situation, in which Gallienus, however, cuts a marginal,38 though positive figure . This time his per33 In all likelihood, Zosimus here depends on Dexippus, for whom the ‘Scythian’ threat was obviously the most important one . 34 He risks a potential misunderstanding here, as ἔθνος, in addition to the meaning ‘people’ suggested by the translators (Paschoud 2000–2003, Ridley 1982, Veh 1990) can be a technical term for ‘province’ . 35 The personal commitment is emphasised: 1,30,2 (αὐτός) . 36 1,20,2 . Philipp yields to the diplomatic initiative of the ‘Scythians’ too easily (ῥᾷον); 1,24,1 (Gallus): the deal marks a lucky day (εὐημερία) for the ‘Scythians’ . Philipp’s treaty with the Persians is mentioned neutrally by Zosimus 1,19, but he later harshly criticises the emperor (comparing him to Jovian) 3,32,4: Φιλίππου διαδεξαμένου τὴν ἀρχὴν καὶ εἰρήνην αἰσχίστην πρὸς Πέρσας θεμένου . 37 1,30,2 (ἔδοξεν) . 38 Precisely the syntax places him in the periphery . 1,37,2: Γαλλιηνοῦ δὲ τοῖς ἐπέκεινα τῶν Ἄλπεων τόποις ἐγκαρτεροῦντος καὶ Γερμανικοῖς ἐνασχολουμένου πολέμοις […] — The plural τόποις/πολέμοις points to the complexity of the situation in Germany .

116 · Georg Müller sistence and commitment are praised because he does not yield to the increasing pressure from across the Rhine . Even in this short appearance, it is striking that it bears exactly those qualities which Zosimus missed in Valerian in the previous chapter .39 As a result of the new strategic situation caused by the expansion of the ‘Scythians’, however, Gallienus changes his arrangements and leaves his younger son Saloninus in Cologne, probably to consolidate the loyalty of the troops . Although this daring attempt fails dramatically, since Postumus, a charismatic and unscrupulous40 army leader, usurps power on the Rhine, Gallienus remains altogether successful, since he manages to nip in the bud most of the usurpations which had been cropping up far and wide after Valerian’s capture . Zosimus is thus far from dramatising the situation .41 He even represents Postumus’ revolt as a marginal event42 that does not force Gallienus to alter his strategy of flexible response .

The periphery While the Nea Historia offers only a few details (zoom level 4) on the period preceding Severan rule, this is followed by a more detailed account of the following events, with a clear focus on the ‘Scythians’, whose operations are described much more widely43 than those of the Persians, or even the Germans, who sometimes are not mentioned at all . The Germanic danger, for instance, which ultimately led to the fall of Alexander Severus, is not mentioned at all . Zosimus merely remarks that the emperor was in the Rhine provinces .44 Having an abundance of information at his disposal on the operations against the ‘Scythians’, Zosimus plastically illustrates how the inhabitants of the provinces experience the raids of the invaders, showing at the same time structural elements of the imperial organisation, from which the reader can make conclusions, by analogy, on other, less documented areas . Therefore, it is worthwhile having a closer look at the ‘Scythian’ invasions, although they are not directly related to Palmyra . The fates of the cities here provide examples of the mistakes and failures which contributed to the decline of the empire . At the same time, Zosimus demonstrates his literary skills with a classic tricolon .

39 The account’s real intent, however, is to describe the successful mobilisation of all resources of the city of Rome against the ‘Scythians’ . 40 He has Saloninus and his tutor Silvanus killed . 1,38,2: ἀμφοτέρους ὁ Πόστουμος ἀνελών . 41 Cfr . Veh 1990, 291: “Epoche der ‘30 Tyrannen’” and his reference to Paneg. 8 [5], 10, with an account of Gallienus’ reign . 42 1,38,2: τὴν ἐν Κελτοῖς ἐπικράτειαν εἶχε . 43 Chapters 20, 23, 24, 26, 28, 31, 32, 34, 37; Zosimus distinguishes —  in contrast to his portrayal of the Germans – individual nations in chapters 20, 27 and 31 . 44 1,13,2 .

VIII . Persians, Romans and Palmyr a in Zosimus’ historiogr aphy · 117 A first attack of the barbarians on the city of Pityus on the Black Sea coast fails because of the clever tactics of the local commander Successianus, who is explicitly praised for his victory .45 The episode has by no means only local significance, as it shows on the one hand that the balance of power is favourable for Rome, and on the other the importance and the possibilities of the local contingents, depending on the military competence of their leaders . (1) The Scythians dread a possible merging of the surrounding garrisons, which would outnumber their army . (2) The victory of Pityus frees the whole Black Sea region from the Scythian danger . (3) Apparently there is a great need for capable officers, since Successianus is immediately promoted to praefectus praetorio . However, the next Scythian attack succeeds without a hitch . The same military unit, which under the leadership of Successianus has been capable of a very successful counter-offensive, is not even able, without its leader, to offer a decent defence . In the immediately following siege of Trebizond, Zosimus shows the consequences of the lack of discipline rampant in an army without a leader . Chalcedon and Nicomedia are no longer defended . In the first case, the garrison flees from the enemy, in the second case there is no mention even of regular troops . It is obvious why things are running downhill . While a garrison with a capable commander can easily cope with intruders, it turns out that fighters who are left to themselves cannot stand the pressure for long, and that unfortified cities in the hinterland are easy prey . At the beginning of the Palmyra episode, the reader has the following foreknowledge: 1 . The Persians are fast, well-organized elusive opponents who operate flawlessly, although they are keen on looting rather than expanding their empire . The government considers them as the most dangerous enemy and has already twice undertaken unsuccessful massive campaigns against them . 2 . Valerian and Gallienus have made a virtue out of necessity and drafted a new military doctrine . a) Multipole strategy: division of command with priorities, b) hierarchy of risks: no hectic reaction to the eruption of new trouble spots, c) subsidiarity: devolution of responsibility to local authorities; self-empowerment of the lower administrative level is of crucial importance . If it is run properly, results are promising . 3 . The numerous problems (in addition to the three border zones under threat, usurpations and epidemics) still lead to a serious imbalance of the system . The energetic and prudent Gallienus, however, repeatedly succeeds in stabilising the situation .

45 1,32,2: οἱ μὲν τὸν Εὔξεινον πόντον οἰκοῦντες τῇ Σουκεσσιανοῦ στρατηγίᾳ περισωθέντες .

118 · Georg Müller

Palmyra Zosimus divides the description of the ‘Palmyrene Empire’ into three sections . 1,39 is dense with the successful years under Odaenathus . Zenobia’s reign is divided into chapters 1,44 and 1,50–1,56 . In chapters 1,57–58, a prophecy related to Palmyra is told, before chapters 1,59–1,61 describe the time after Zenobia and Palmyra’s downfall .46

Odaenathus 1,39 is a typical regional chapter of zoom level 2 . The description sets out from Gallienus’ perspective,47 maintaining the top-down perspective to the end of the chapter . The emperor has left the Germanic wars to his subordinates, even at the price of further usurpations, and now takes care by himself of the ‘Scythians’, whose rampage is drastically described with three participles .48 So the following remarks refer to events in a secondary theatre, whose protagonist does not occur by chance as the (accusative) object for the first time . He is introduced as a representative of the emperor, who is to help in a very difficult situation: τοῖς δὲ περὶ τὴν ἑῴαν πράγμασιν οὖσιν ἐν ἀπογνώσει βοηθεῖν Ὀδαίναθον ἔταξεν, ἄνδρα Παλμυρηνόν . The reader, who already knows Gallienus’ strategy, must wait for the emperor to call for foreign aid,49 the circumstances being similar to Germany . However, things are different here, since there are no Germanic tribes to be played off against each other and, above all, Rome must not make an agreement, but can order it . The explanation to this can be inferred from the brief biographical notes that follow . Odaenathus is a 46 1,39 Odaenathus; 1,40 (Gallienus’ assassination);1,41 The end of the Usurper Aureolus; 1,42 ‘Scythian’ naval operations fail; 1,43 The ‘Scythians’ devastate Asia Minor; 1,44 Zenobia’s ambition; Zabdas conquers Egypt;1,45 The Scythians in Macedon; 1,46 The death of Claudius; 1,47 The Quintillus episode — Aurelian becomes emperor; 1,48 Aurelian in Pannonia; 1,49 Aurelian in Italy; 1,50 Aurelian’s first victory over Zenobia; 1,51 Zabdas’ ruse, escape to Emesa, Aurelian’s clemency; 1,52 Aurelian’s second victory at Antioch; 1,53 The armies of the warfaring parties; 1,54 Palmyrene council of war, Aurelian besieges Palmyra; 1,55 Escape to the Euphrates; 1,56 Disagreement among the Palmyrenes, Zenobia being put on trial; 1,57–58 Prophecies; 1,59 Zenobia and Vaballathus to Rome, death of Zenobia; 1,60 Attempted usurpation of the Palmyrenes; 1,61 Aurelian destroys Palmyra . 47 Straub 1952, 61: “in gedrängtestem Kurzbericht” — against Straub (who is joined by Kettenhofen 1992, 297) Zosimus does not claim that Gallienus has been victorious . In that case he would have used a different wording . 48 Τῶν δὲ Σκυθῶν τὴν Ἑλλάδα κάκιστα διαθέντων καὶ τὰς Ἀθήνας αὐτὰς ἐκπολιορκησάντων Γαλλιηνὸς μὲν ἐπὶ τὴν πρὸς τούτους μετῄει μάχην ἤδη τὴν Θρᾴκην καταλαβόντας . The perspective changes from a global one on Greece with the adversative clause Τῶν δὲ Σκυθῶν τὴν Ἑλλάδα κάκιστα διαθέντων (in sharp contrast to εὖ διαθεῖναι, which normally characterises a good policy, see below 1,38,2 on Odaenathus); this is followed by the conquest of Athens and the occupation of Thessalonike, which is geographically confusing, but chronologically correct; cfr . on this Kettenhofen 1992, who considers this account as extraordinarily precise and detailed, but objects the chronology . 49 Above pp . 114–15, ἐν ἀπόροις τε ὤν ἔδοξεν ἐν μέρει τὸν κίνδυνον ἐλαττοῦν τῷ σπονδὰς πρός τινα τῶν ἡγουμένων ἔθνους Γερμανικοῦ πεποιῆσθαι .

VIII . Persians, Romans and Palmyr a in Zosimus’ historiogr aphy · 119 Roman magistrate . The numerous offices and titles with which Gallienus and his predecessors50 have heaped him are only mentioned quite generally, such as when section 1,39,1 mentions him to be worthy of an office ‘because of his ancestors’ .51 The cryptic wording leaves open what it might have been, but it is evident that his position was a responsible one . At the same time, the redundant ἄνδρα52 acts as a placeholder for a possible office53 in Palmyra, which does not fit here .54 Consequently, it is on purpose that no exact position in the Palmyrene society is revealed . The result is a deliberately blurred image of the imperial representative Odaenathus and his powers in the Roman state .55 He has some kind of home force (οἰκεῖα δύναμις),56 which he integrates into the Roman troops . This phrasing highlights that despite the Palmyrenes not providing an auxiliary force for the Romans, their soldiers are still part of their legions, fighting as it were under the imperial standard .57 Overall, Odaenathus’s approach unites improvisation and high risk . The improvised character of his efforts transpires three times within a few sentences: (1) δύναμιν ὅτι πλείστην emphasises his daring, since he holds back no reserves for himself, his family or his city . On the other hand, his limited resources become obvious . The reinforcements do not correspond to the needs, but to what is available . Accordingly, (2) his pursuit of Shapur runs κατὰ τὸ καρτερόν, without any delay and without sparing the forces, but also without the last decisive blow . Similarly, (3) Odaenathus’ rearrangement of the war-torn areas follows a hand-to-mouth strategy: ὡς οἷος τ᾽ἤν . Yet he achieves amazing things with his makeshift forces . While the RomanoPalmyrene territorial gains could be regarded as a collateral damage of the Persian strategy of hit and run, the fact that Odaenathus reconquered all the cities is significant . Almost 50 For an overview cfr . Geiger 2013, 128–29, esp . n . 742 with reference to further scholarship; Hartmann 2001, 186–194 . 51 […] ἄνδρα Παλμθρηνὸν καὶ ἐκ προγόνων τῆς παρὰ βασιλέων ἀξιωθέντα τιμῆς; according to Veh 1990, 55 (“ob seiner Vorfahren”) and Paschoud 2000–2003, vol . 1, 36 (“à cause de ses ancêtres”) the ἐκ is causative, for them, the achievements of the ancestors are the reason why Odaenathus has been honoured who, therefore, has no merits of his own so far . A chronological significance should at least be considered which would mean that the family held a certain office regularly . Tιμή must mean ‘office’ here; this follows lexically from ἔταξεν and from ἀξιωθέντα . Otherwise this would be a pleonasm atypical for Zosimus . Hence Ridley 1982, 12 (“honour”), Veh 1990, 55 (“Ehrung”) and Paschoud 2000–2003, vol . 1, 35 (“honneurs”) are imprecise . 52 Next to an ethnic noun this is used rarely by Zosimus . 53 Exarch for example . 54 Theoretically, the situation in the oasis may have been irrelevant or unknown to Zosimus, if one exludes that Odaenathus held no office at home . 55 The reader will have accepted at least two elements: Odaenathus’ Roman citizenship, together with an eminent position in Palmyra . The wording ἐκ προγόνων refers to the third generation at least . 56 The troops belong to Odaenathus’ οἶκος, who hence either considers the entire city of Palmyra as his personal property or belongs to an extremely powerful family . 57 1,39,1: τοῖς αὐτόθι λελειμμένοις στρατοπέδοις δύναμιν ἀναμίξας ὅτι πλείστην οἰκείαν . The legions form (numerically) the core of Odaenathus’ forces .

120 · Georg Müller incidentally, it is remarked here that Shapur does not limit himself only to looting, but has captured important locations in the Roman sphere of influence . The quick destruction of the significant city of Nisibis ἐξ ἐφόδου is not only seen as a military achievement, but as the outcome of a deliberate strategy of psychological warfare . Obviously, the city is being punished for taking a pro-Persian attitude in the preceding events . The twofold siege of the Persian capital of Ctesiphon uses the same effect of deterrence . Here the story culminates in the remark that the Persians have to fear for their lives .58 This somewhat destructive harshness of the Palmyrenes is joined by the constructive feature of careful circumspection . The reorganisation of the areas which have been devastated by the Persians is praised with the predicate εὖ, and in addition to the soldierly qualities, which predominate, Zosimus grants Odaenathus the very important virtue of ἐπιμέλεια .59 And last but not least, one learns that Odaenathus had an entourage, who were involved in decision-making and displayed, like their superior, a healthy quantum of pragmatism .60 As we see, he was open to advice . If yet another proof of his ability to rule were needed, it comes with the violent end of his life ἐξ ἐπιβουλῆς .

Zenobia Since Zosimus’ narration follows the reigns of the individual emperors, it is not easy to gain a clear picture of Zenobia and to pursue the Palmyrene expansion associated with her name (as well as the downfall of the oasis city) . There are three sections dealing with her: Gallienus (1,39), Claudius Gothicus (1,44) and Aurelian (1,50–56 and 1,59) . In perspective, the information about Zenobia in Gallienus’ time is in the range of zoom level 2, while the events during the reign of the two other emperors are zoomed in more closely (zoom levels 3 and 4) . Zosimus’ account of the Palmyrene expansion reaches a degree of detail that is matched only by his description of the ‘Scythian’ attacks . This raises the question of his sources and intentions . The first information in 1,39 differs fundamentally not only in extent (one sentence) and in level from those of the other two sections, but also in character .61 Zenobia is presented there as a successor; in terms of her style of government which is described as ἐπιμέλεια, she is almost a copy of her deceased husband . She relies on her advisors and does not display much ambition in terms of foreign policy . There is not a 58 The conspicuous wording οὐχ ἅπαξ ἀλλὰ καὶ δεύτερον may be directed against a different tradition — the other sources only know of one campaign . It also creates in the reader the image of a gambler who takes any risk; and of a humiliation of the Persian defenders whose not very heroic thoughts are spread with relish: ἀγαπῶντας εἰ παῖδας καὶ γυναῖκας καὶ ἑαυτοὺς περισώσαιεν . 59 This is the prime virtue for leadership, see above e negativo on Philip and Trebonianus Gallus . 60 This is demonstrated to the reader using the twofold πράγματα in the final sentence . 61 This can hardly be connected to a change in character or political insight, since Zosimus would have mentioned this, as can be seen from the transformation of Alexander Severus: 1,12,2: τῷ τε σώματι καὶ τοῖς κατὰ φύσιν λογισμοῖς ἀλλοιότερος ἦν .

VIII . Persians, Romans and Palmyr a in Zosimus’ historiogr aphy · 121 word about military achievements . Zenobia uses the φρόνημα ἀνδρεῖον that is ascribed to her only to consolidate Odaenathus’s achievements . The second Zenobia, on the other hand, is the exact opposite, literally a different person . She is incapable of constructive activity and, in her hubris, tears everything down . As Zosimus laconically notes in 1,44, she strives for higher levels . However, the historian does not go into further detail: whether she holds any posts or titles or what her exact goals were we do not know . Yet it is clear that the Zenobia of Claudius Gothicus’ reign roughly matches the Zenobia of the time when Aurelian was in power, so much so that they cannot be separated without destroying the organic character development that transforms her from Palmyra’s sole ruler into the city’s figurehead and ultimately into a sad figure . In 1,44, she appears as commander-in-chief of the Palmyrene army, giving general Zabdas the command for the Egyptian campaign which would turn out to be a success . She enjoys the same position in 1,50 when Aurelian attacks her in Antioch . She no longer plays an active part, but merely serves, with her large army, as the most important target of the Roman expedition . While describing the overall situation — the central government has lost control over much of the East — Zosimus refers to “the Palmyrenes” in general as the emperor’s opponents in his quest to recuperate the territories, but not explicitly to Zenobia .62 After the devastating defeat of their forces, Zabdas takes the reins in his hand and orders the retreat to Emesa . “He carries along” Zenobia .63 Although this harsh expression could be justified by the military context, where quick action is imperative and Zabdas, as a military specialist, is in command, Zenobia’s further appearances show that she is indeed no longer in charge . A second heavy defeat follows, and Zenobia falls into depression .64 Even grammatically she appears as an object of her own discouragement, and then her position in the council of war is practically demoted . There is no suggestion of her summoning, or presiding over, a council . From now on, she is considered rather as a burden by the other participants . The decision to retreat to the fortified city of Palmyra takes place κοινῇ γνώμῃ, without individuals being mentioned . Even Zabdas does not play any specific role . Here at the latest, it is clear that she is no longer in command . As the siege drags on, the Palmyrenes decide to take refuge with the Persians . Zenobia is obviously no longer involved in the deliberations . Instead, she is put on a camel65 and sent to the Euphrates . Particularly embarrassing is not only her capture at the last moment, but rather Aurelian’s bad mood in finding himself tricked off his triumph, Zenobia being only a woman . 62 63 64 65

Passim, for instance 1,51,1: Παλμυρηνῶν προστασία . 1,51,2: τὴν Ζηνοβίαν ἑαυτῷ συνεξαγαγὼν . 1,54,1: Ἀθυμίας δὲ πολλῆς εἰκότως ἐπὶ τῇ ἥττῃ Ζηνοβίαν ἐχούσης . 1,55,2: καμήλῳ τὴν Ζηνοβίαν ἀναβιβάσαντες . Linguistically, (she is the object) and factually a hardly convincing performance .

122 · Georg Müller A final low is reached by her (self)degradation, when she pretends to be a weak female and victim of her male advisers in the court hearing ordered by Aurelian . She now denounces them to save her skin .66 The opposite character to Zenobia is the philosopher Longinus, who is introduced not only as a man of education and its dissemination, receiving thus the highest credentials, but shows also style and bravery, the qualities that Zenobia is lacking . He accepts, like Socrates, the death sentence and comforts those left behind . Now the circle is complete . The once shrewd, masculine planner of 1,39, who relies on a circle of advisors to continue her husband’s achievement, the consolidation of the eastern provinces, has become a gambler and female traitor . In the process, Zenobia gradually becomes a puppet of her entourage . Zenobia’s attempt to usurp supreme power, which is well documented by the numismatic record, is only hinted at by Zosimus .67 He does not quote any symbolic acts, such as the donning of purple, which he normally uses in order to highlight such demonstrations of power, which are already doomed . Likewise, there is no mention of any other ways of appropriating Roman insignia or of claiming Roman offices . The fact that Zosimus never mentions the name Vaballathus, the heir to the throne, and only incidentally alludes to his existence,68 is certainly not a coincidence in this context . It shows his intention not to associate Zenobia and Vaballathus with usurpation . They are referred to as rebels, their goal was (so it is meant to look like) to get rid of Roman rule as such, and not to compete with the emperor . This intention of the historian is supported by the lengthy narrative of the Palmyrenes showing around Aurelian’s doppelganger in Antioch (1,51,1–2) . They try to destroy the opponent’s charisma since they cannot build one of their own . The message is supposed to be, ‘Look, we have got the emperor!’, and not ‘We are the emperor’ . Consequently, Zosimus pushes the usurpation to the very end of the Palmyra episode (1,60), as a last resort, when Rome’s remaining opponents with one last mad act dig the grave of their state and city .

The Palmyrenes Palmyra is introduced in 1,39 as a considerable regional power with an intact leadership structure . While Odaenathus does not occupy a prominent position, he relies not merely on the Romans’ protection . He surrounds himself — at least in civil matters — with consultants, who count on careful planning and pragmatism . The smooth transition of power to 66 Hartmann 2001, 26: “Schließlich rückt [Zosimos] Zenobia in ein schlechtes Licht . Sie soll beim Prozess in Emesa vor Aurelianus ihre Ratgeber verraten haben .” 67 By means of the word ἐπανάστασις which he uses 1,59 and also for the attempted usurpations against Gallienus 1,38 (ἐπανίσταμαι) . 68 1,59: Αὐρηλιανοῦ […] συνεπάγοντός οἱ Ζηνοβίαν τε καὶ τὸν παῖδα τὸν ταύτης καὶ πάντας ὅσοι τῆς ἐπαναστάσεως αὐτοῖς ἐκοινώνησαν .

VIII . Persians, Romans and Palmyr a in Zosimus’ historiogr aphy · 123 Zenobia is also evidence of the need for, and sense of, continuity . A small shadow falls over the shine of unity when Odaenathus is murdered at a birthday party . This, however, does not blur the impression of homogeneity . Tribal structures of the Palmyrene state and social order or even the competitive spirit of a rivalling aristocracy are not yet to be felt here . Later, namely in the Claudius Gothicus and Aurelian sections, the picture becomes more differentiated . Zenobia’s gradual disempowerment reveals the peculiar patterns of Palmyrene society, and once again Zosimus shows the destruction of their harmony in a three-step process . (1) The unanimous war council (1,54) reflects the aristocratic principle dominating in Palmyra, since the interests of the city are given priority over those of individuals .69 Another indication of such open negotiations between equals is the possibility of a divergent view, which could only have been afforded by a political heavyweight . Twice, Zosimus emphasises the importance of unanimity in the decision-making process, thus pointing to the possibility of precisely such divergent opinions .70 (2) There is the same impression of a collaborative elite in 1,55, here, but it is clear that unanimity is no longer an issue .71 Since, apart from general Zabdas, there is no mention of names of Palmyrene elite members, the impression of homogeneity arises . This does not mean that their interests were in harmony, since (3) after Zenobia’s escape, the elite splits .72 Some favour surrender, others want to continue the fighting even in the face of an absolutely hopeless situation, not even giving up after the final defeat . They persistently look for a usurper who can be positioned against Aurelian . They do this regardless of possible sanctions, which then follow surely enough . Aurelian razes Palmyra to the ground . (1,61)

The Persians Shapur does not play an active part throughout the Palmyra episode, although he previously had a key role as the trigger of the situation . He is only mentioned at the very beginning of 1,39, twice in one sentence, as Odaenathus pursues him and recaptures Nisibis . After that, he is not mentioned again . The Persians in general, as a foreign power, are no longer a threat; they have become irrelevant . This is somewhat surprising after the previous statements . The reticence in 1,39 may be explained by the fact that the chapter deals with Odaenathus, and Shapur is consistently presented as a strategist who remains true to his tactics, tackling only the feasible .

69 For this reason it is unlikely that the council was a body of — possibly even foreign — soldiers who were more interested in the army than in the city . 70 1,54,1: κοινῇ γνώμῃ and πᾶσιν in the first sentence, then οὐδὲν ἦν ἔργου καὶ λόγου τὸ μέσον in the final sentence of the paragraph . 71 1,55,1: γνώμην ποιοῦνται [without κοινήν], 1,55,2 βουλευσάμενοι [likewise without the assurance that no alternative has been suggested] . 72 1,56,1: αἱ γνῶμαι διχῇ διῃροῦντο .

124 · Georg Müller Zosimus highlights the difficult political situation in the East with its diverse loyalties . The inhabitants of Nisibis may not have been the only ones in the region who sided with the Persians .73 This puts Shapur’s success considerably into perspective, and the seemingly infallible Persian king is pared down to a normal size . Yet in the following chapters, the reader may have been waiting for the Persians to capitalise on Aurelian’s quarrels with Zenobia, as the laughing third party . Yet this does not happen . On the contrary, two text passages give the impression that there is no Persian danger at all . The first, a carefully painted miniature, draws on the pattern of ethno-political complexity, which has been introduced taking Nisibis as a model . During the siege of Palmyra, a Persian serving Rome silences an insolent defender by shooting him with an arrow .74 Zosimus emphatically mentions the archer’s nationality twice, stressing that the Persian had offered himself to Aurelian for the task all by himself . The archer’s commitment goes far beyond what is expected of a mercenary, since the act is performed only to defend Aurelian’s honour, not arising from military necessity . The reader is thus given the chance to make up his own mind about the subtle situation . The second passage where Persians are mentioned, tells the story from another perspective and on a new background . Facing final defeat, the Palmyrenes consider seeking help from the Persians, in order to continue their war against Aurelian from the other bank of the Euphrates .75 What Ctesiphon thinks about the proposition remains open, and so do the chances of success of such a coalition . After Zenobia is captured, however, the initiative comes to nothing . The subsequent harakiri of the Palmyrene war party turns the option of an anti-Roman alliance purely theoretical . The carefully constructed image of a hostile and terrifying Persia fades with the appearance of the Palmyrenes .

The Roman Empire I — nationalities and loyalties For the central government, it is obviously crucial to know how loyal the different parts of the empire, the subjected peoples and polities, are in case of a threat from the outside — and how easy it is to win over the subjects of the opponent . Until the beginning of the Palmyra episode, the situation is favourable everywhere . The ‘Scythians’ are seen as enemies everywhere, while Rome can buy friends on the other side of the border .76 However, part of the empire’s population regard the Persians77 as no worse rulers than the

73 1,39,2: τὰ τῶν Περσῶν φρονοῦντα . 74 1,54,2–3: ἤδη δέ τινος καὶ εἰς αὐτὸν τὸν βασιλέα λόγους αἰσχροὺς ἀφιέντος, παρεστώς τις τῷ βασιλεῖ Πέρσης ἀνήρ ‘εἰ κελεύεις’ ἔφη, ‘τὸν ὑβριστὴν τοῦτον ὄψει νεκρόν’ . [3] Ἐγκελευσαμένου δὲ τοῦ βασιλέως, προστησάμενος ἑαυτοῦ τινὰς ὁ Πέρσης τοὺς ἀποκρύπτοντας κτλ . 75 1,51,1 . 76 Above, on the German diplomacy of Gallienus . 77 Above, on Nisibis .

VIII . Persians, Romans and Palmyr a in Zosimus’ historiogr aphy · 125 Romans, and the same is true for the Palmyrenes, who are earmarked as outsiders of the empire, so that Zosimus can demonstrate the integrative power of the idea of an empire e positivo and e negativo . The inhabitants of Emesa, for example, are declared, with a meaningful phrase, to be hostile to Zenobia:78 διὰ τὸ καὶ τοὺς Ἐμισηνοὺς ἀλλοτρίως πρὸς αὐτὴν ἔχοντας τὰ Ῥωμαίων αἱρεῖσθαι, meaning, they “choose the side of the Romans”, ergo they are not from the outset and basically loyal inhabitants of the empire . In Antioch, by contrast, there is fear of acts of revenge by Aurelian because of a prevailing friendly attitude to Zenobia .79 A third example is the rather complicated situation in Egypt . There is talk of an Egyptian named Timagenes, who is supposed to help the Zenobia-Palmyrenes secure control over Egypt,80 a kind of leader of a fifth column, the ‘friends of the Palmyrenes’ ([Αἰγύπτιοι] ὅσοι […] τὰ τῶν Παλμυρηνῶν φρονοῦντες 1,41,2) . These may include the Alexandrians, who are known for their rebelliousness . This is revealed from a later note (1,61,1), in which they are also subjugated after the destruction of Palmyra . Aurelian’s skill is now required to tame the centrifugal forces and to weld together the multifarious parts of the empire, using not only the above-mentioned punitive measures . Zosimus shows this very subtly in the transition from chapter 51 to 52, just before the showdown between the emperor and Zenobia . From a superficial point of view, the passage serves to increase the tension before the clash of two huge armies, yet deeply inside it satisfies the longing for power through greatness . While the forces fighting on Zenobia’s side next to the Palmyrenes are only generally and thus depreciatively summarized as “others who had all decided to participate in the campaign”,81 Zosimus meticulously records the special forces originating from a total of twelve parts of the world, who have been sent by Aurelian into battle . The true message is revealed when the ‘Dalmatian’ and ‘Moorish’ cavalries are transformed into ‘Roman’ troops and all kinds of peoples from the Mysians to Palestinians become the ‘Roman army’ .

The Roman Empire II — emperor and charisma Aurelian’s rule is presented in full detail as a series of military successes over ‘Scythians’, Palmyrenes and assorted usurpers . The message of the final tribute to the emperor is that these triumphs have been achieved through the skill and physical courage of a real sol-

78 1,54: […] διὰ τὸ καὶ τοὺς Ἐμισηνοὺς ἀλλοτρίως πρὸς αὐτὴν ἔχοντας τὰ Ῥωμαίων αἱρεῖσθαι . 79 1,51,3: Εὑρὼν δὲ τοὺς πολλοὺς τὴν πόλιν ἀπολιπόντας δέει τοῦ μὴ κακοῦ τινὸς ὡς τὰ Ζηνοβίας φρονήσαντας πειραθῆναι . 80 What is planned here, is left open by Zosimus . In an extreme case, it could be a mere legitimation through an influential personality who, for instance, calls the attackers “for help” . Strikingly, Timagenes is referred to, like Odaenathus, as ἀνὴρ Αἰγύπτιος (1,44,1) . Hence his position in Egypt remains uncertain . 81 Translation after Veh 1990, 62 .

126 · Georg Müller dier .82 For the readers, however, all these details produce a coherent and consistent character sketch, which makes the emperor’s success only plausible by referring to his charisma . (1) Aurelian is proud, though neither vain, nor haughty . He endures the insults of the besieged Palmyrenes and only pursues them when the lower army ranks ask for it — the episode also proves the soldiers’ esteem for the emperor . He cannot enjoy to the full the victory over Zenobia . (2) Aurelian is cool and diplomatic, and has a sound judgment: he is mild and understanding with the Antiochians (1,51,3) and even with the Palmyrenes (1,56,1–2); he does not feel challenged by the grotesque usurpation of Antiochus (1,60– 61) and leaves him alive . (3) Aurelian’s influence is effective even when he is not present . Being forced by the emperor to the defence, Zenobia’s commander-in-chief Zabdas resorts to a psychological trick, which is significant in the light of recent events . Worrying that Antioch’s inhabitants might attack his troops upon learning that the Palmyrenes were beaten, he has a look-alike of Aurelian be led through the city, pretending to have won the war and to have (re)captured the emperor . This has a twofold effect on the reader: (a) the anecdote sets Aurelian in a positive light compared to Valerian . He does not easily fall into the hands of the Orientals . (b) The emperor has a charismatic personality, and the Palmyrenes appreciate him very highly (otherwise the trick would not have worked) .

Summary A selection of Zosimus’ techniques and intentions are revealed in this wide context of historical events .

Above and below — the sources In the first passage up to the reign of Claudius Gothicus, Shapur and his Persian attacks dominate the Near Eastern theatre . In Zosimus’ report, Shapur appears and disappears in a flash among the Scythians, just like his counterpart in the Roman eastern provinces . In the process, he increasingly becomes more dominant, up to the shocking climax of capturing the emperor . Now headed by Gallienus and aided by Odaenathus, Rome can still cope with the situation . The expected continuation of this struggle ends only seemingly abruptly . Although the Persians disappear almost completely as opponents, the narrative replaces them with the Palmyrenes, who, however, mutate from loyal helpers to arrogant rivals . This takes place with the shift of power from the reliable Odaenathus to Zenobia . The effect is achieved by means of the zoom technique, which brings Zenobia from the distance of 1,39 to the closer view of 1,44 . Nevertheless, the attentive reader will

82 1,62,3: ἐτάφη μεγαλοπρεπῶς […] τῶν ἔργων ἕνεκα καὶ τῶν κινδύνων .

VIII . Persians, Romans and Palmyr a in Zosimus’ historiogr aphy · 127 have noticed that μειζόνων ἐφιεμένη πραγμάτων is only a trick label for an abrupt paradigm shift . The same technique is used to describe the shift of alliances in the West-East antagonism from Romans and Palmyrenes against Persians to Romans against Palmyrenes (/ Persians) . Just because Zosimus describes the Odaenathus passage from a wide angle and very shortly, he can count on the fact that after the change of the foci to a detailed approach, the Persians will disappear from view and will be forgotten by the superficial reader . In this context, it is very important that the two great narrative threads be maintained: ‘the danger comes from the East’ and ‘care, commitment and solidarity bring longterm success’ . Surely Zosimus has not strived to create this superimposition, which is rather the result of his source material, out of which he extracts the best he can . Apparently, he had a large amount of information about Palmyra starting from the reign of Claudius Gothicus . This allowed him to have a detailed internal view of the circumstances and to make a vivid account of the conflict with Aurelian . On the other hand, he is influenced by the sources to depict Palmyra under a negative light, and Zenobia even worse . Hence, the change of his material does not take place, as generally assumed, in 1,46, after the reign of the emperor Claudius Gothicus, but earlier .83 In spite of the fissures resulting from the existing sources, Zosimus’ description takes the reader on a wave trough, from the calm overall situation at the end of the Severan period down into the abyss created by Shapur’s attacks and again up to the consolidation of the situation as a result of Aurelian’s calm and strength .

Balance — the foci and perspectives By repeatedly changing perspective and distance to the object, Zosimus gives the impression that one thing counts on all levels: the πρέπον . This means both the appropriate handling of situations, as well as the recognition and observance of public order, especially the hierarchies . Viewed from afar, Shapur plays his hand perfectly, without overdoing it . Odaenathus is the prime example of the underling who sticks to his duty and can achieve the best performance in his domain . Gallienus does not try the impossible, but keeps calm even in the 83 This is based on a quotation from Photius (Fr . 1 Blockley), according to which Eunapius’ work which was a continuation of Dexippus started here . In fact, the meaning of this section is a different one: ἄρχεται μὲν ἀπὸ τῆς Κλαυδίου βασιλείας ἐς ὅν Δεξίππῳ ἡ ἱστορία καταλήγει . Paschoud 2000–2003, vol . 1, XXXVII (“La seconde édition de l’ouvrage historique d’Eunape commence incontestablement en 270”) argues against Reitemeier according to whom Eunapius’ history began with the accession of Claudius Gothicus (268) and for whom the fissure is after 1,40 . Veh 1990, 10–12, in the introduction to his translation, assumes a fissure after 1,46 . However, Paschoud 2000–2003, vol . 1, LXIX in the second edition of his introduction argues in favour of Zosimus having used only Eunapius and Olympiodorus claiming that the beginning of the Nea Historia is based on the first edition of Eunapius’ history that began with Augustus .

128 · Georg Müller deepest crisis . Even Valerian’s course of action is not flatly criticized: he fails because at the crucial moment his nerves let him down . The close-up view makes the mistakes stand out more sharply: Zenobia succumbs to her hubris, because she misjudges both the resources of the oasis city and the dangerousness of her opponent Aurelian, as well as her own skills . Although Zabdas is prudent in his actions, Palmyra’s internal structure is unstable and the government is led by the wrong people . Aurelian is the only one to be whitewashed . His dealings with subordinates and opponents are foresighted and his assessment of the situation is exemplary . The important message from the Palmyra episode then is: Μείζονα πράγματα are events over which one loses control because one cannot keep track of them .

IX. The political powers and Syria. War aims, scenarios and perspectives for a country in civil war Alexander Will

It is difficult to make predictions, especially about the future. Mark Twain

Civil war is raging in Syria . The outbreak of the conflict took the world by surprise, and has not stopped shocking humanity with the merciless brutality of the fight . At the beginning of the millennium, Syria appeared to be a stable state, certainly a dictatorship, but far removed from the failed states in Africa or Central Asia . This image was based on a bunch of false observations about the country, observations which were and are still circulating, and which will be dealt with later . A detailed reconstruction of the processes and developments is not possible at the present time . This paper will therefore proceed from the state of knowledge available in July 2018, and from this basis, it will try to outline possible lines of development for the conflict in the future . Main sources to be consulted will be press releases, internal inquiries by government agencies, and papers from various think tanks, as well as background talks with decision-makers . The turn of the year 2017/2018 marked a significant change in the Syrian civil war . After the announcement of a partial withdrawal of the Russian expeditionary force, and the intervention of Turkish troops in the northeast of the country, it is possible to speak of a turning point . The most radical jihadist1 forces, known as the Islamic State (IS), were decisively beaten, if not crushed . With the help of Russia and Iran, the central government of Syria was able to regain control of large parts of the country . In the Kurdish areas in the north, Turkey had begun to impose pacification on its terms . The slow establishment of a balance of forces in the country goes hand in hand with a far-reaching marginalization of the internal Syrian forces . This paper’s proposition draws upon the observation that by the beginning of 2018, external players have completely taken control of the events in Syria . Both the central government and the various rebel groups, with the exception of the IS, have become clients of external powers . Possible scenarios and perspectives for the future of Syria are therefore to be thought above all considering 1

For the sake of readibility, this paper does not apply the phonetic transcription of Arabic, Hebrew and Russian terms . Geographic and ethnographic names have been anglicised .

130 · Alex ander Will the interests of the intervening powers . The most important one is Russia . Other contenders are Turkey, Israel, the US, Iran, and, marginally, the European Union .

The stage After the outbreak of the Syrian crisis, many Europeans were astonished when they realized that the country is by no means ethnically and religiously homogeneous . This illustrates, in turn, the effectiveness of the country’s image as it was propagated by the Damascus regime, which has ruled Syria for decades, relying on Arab nationalism and socialist elements, though being sectarian in its character .2 Many people around the world actually took for granted the myth of a homogeneous, largely secularized “Syrian Arab nation” . The reality, though, is completely the opposite: the country is ethnically as well as religiously extremely fragmented . In many areas before the war, religions and ethnic groups were mixed . That was (and still is) the case especially in Damascus, but also in Aleppo . It was precisely in these mixed regions where the fiercest battles of the war have taken place so far . On the other hand, there are other areas which are settled by closed communities of Alawites, Druze, Kurds and Turkmen . These regions remained largely stable for a long time, some of them having adopted forms of self-government, such as in the Kurdish regions, or the central government having managed to maintain control over them . In 2011, at the beginning of the conflict, the vast majority of the 20 million Syrians were ethnic Arabs . The following table shows the ethno-religious division of the country in that year:3

2 3

Ethnic / religious group

Percentage of the population

Sunni-Arabic

60

Alawite-Arabic

12

Druze-Arabic

3

Ismaili (Shiite)-Arabic

2

Christian-Greek-Orthodox

9

Christian-Armenian

4

Sunni-Kurd

9

Turkmen, Circassian, Assyrian, Jewish

1

The view of a Syria, which is essentially a nation-state troubled by only marginal ruptures, is still maintained by EU experts and affects EU policy . See, for instance, Gaub 2016 . Based on figures supplied by the US State Department . US Department of State, Background Note: Syria, March 18, 2011 . US Department of State, 2010 International Religious Freedom Report, September 13, 2011, https://www .state .gov/j/drl/rls/irf/2010_5/168276 .htm (17 .12 .2017) .

IX . The political powers and Syria · 131 In addition, there were about 80,000 Yazidis, who are ethnic Kurds, living in the country . The total number of the Jewish population before the outbreak of the civil war was 40 . Before the Civil War, most key positions in the country were in Alawite hands . This also explains the structural potential for conflict, which eventually erupted in the civil war .4 On the one hand, a religious minority dominated the country, being hardly disposed to cooperate with others . On the other hand, the Sunni majority radicalized in Syria as well, as it did in the entire Islamic world . The key word here is political Islam . Neo-Salafism and jihadism are its manifestations . The so-called Arab Spring, during which opposition forces from all over the Arab world put massive pressure on their ruling regimes, was the fuse which detonated that explosive mixture . It is not the aim of this paper to expand on the course of the civil war and the highly fractioned opposition militias . That much, however, is certain: a serious armed opposition, which is secular and ‘moderate’ as the Western world understands it, no longer exists .5 As of 2018, even the so-called Free Syrian Army is dominated by jihadists . The so-called opposition, on the other hand, is so fragmented and involved in constantly changing local alliances of all colors, that their militias look rather like rival robber gangs than a political movement . The exceptions are still the Islamic State and the Kurdish militia YPG . In this context, as of July 2018, the balance of power in Syria was as follows . Essentially, there were four major blocs of power: the Kurds in the north, the regime, fragmented opposition forces and the IS . All these blocks had one thing in common: without massive support from abroad, they would have fallen victim to their opponents . This could be observed, during 2017 and in the first half of 2018, in the case of the IS, which by this time had lost its base in Iraq after losing support from the Gulf monarchies and Turkey . The Kurdish YPG, which fought without foreign support against the Turkish army and the Ankara-backed ‘opposition forces’, lost, in 2018, large parts of its territory in northern Syria . Similarly, unaided Islamist militias lost several strongholds to the Russian-backed central government in Damascus . In this regard, by the beginning of 2018, the autochthonous military players of the Syrian civil war were either completely dependent on foreign support, or they faced destruction . Forecasts regarding further developments of the conflict, therefore, are closely connected with the interests of political powers outside Syria, which include not only Russia and Turkey, but also the US, Israel, and, to a lesser extent, the EU .

4 5

On patterns of power and conflicts in Syria in recent times: Gerlach 2015 . This has been foreseeable since 2014 . Back then, a Rand Corporation paper observed that the Syrian National Coalition (SNC), which at that point still led the uprising against the central government, had lost control of the Islamist militias . It concluded: “Islamist hardliners will increasingly dominate the rebellion . […] For now, it appears that the jihadists have become the cutting edge oft he rebellion .” Jenkins 2014, https://www .rand . org/content/dam/rand/pubs/perspectives/PE100/PE115/RAND_PE115 .pdf (10 .11 .2015) .

132 · Alex ander Will

Russia At the beginning of the Russian intervention in Syria, in the autumn of 2015, the Russian government introduced a number of official war aims, which still apply .6 In October 2015, Vladimir Putin mapped out a kind of dual approach: on the one hand, the aim was to stabilize the “legitimate government”, to secure the country’s territorial unity, and to create conditions for a political compromise . On the other hand, it was in the interest of Russian security to crush the Islamic State, as well as other jihadist militias . The West did not seem determined enough in this point: Our duty is to stabilize the legitimate government and to create conditions for a political compromise . […] If we do not interfere in Syria, then thousands of people fighting each other with Kalashnikovs will eventually come to us .7

The spokeswoman of the Ministry of Foreign Affairs, Mariya Zakharova, echoed: “We do not support Assad . For us, it is important to preserve the Syrian statehood .”8 Prime Minister Dmitry Medvedev went along with this line too: “We will do nothing that undermines the authority of the legitimate government,” he pronounced in October 2015 .9 Indeed, this is one of Russia’s most important motives behind its intervention in Syria: the fear that Syria’s territorial disintegration and the establishment of Islamic emirates in its place might jeopardize Russia’s internal security . Sergey Markov, a professor of political science at the University of Moscow, expressed this fear as follows: First and foremost, we have to suppress the danger of the spread of the Islamic State . In the event of a victory in Syria and Iraq, this war could flare up in the North Caucasus, the Volga region, and Central Asia .10

Fending off this threat was a central issue for the military establishment in Moscow too . The head of the Academy of National Security, General Arsen Rean, particularly emphasized this matter in an interview in October 2015 . The general justified the support giv6

On the methodology: In Russia, the press is not free, but it has got a certain freedom . There is, for instance, a lively debate in Russian media on the war objectives in Syria . Also, there can be found usually rather up-front and clear analyses from various political and military think-tanks in Moscow and Saint Petersburg . What follows here, is essentially an examination of such material: several hundreds of articles from the daily newspaper “Izvestia”, the business newspaper “Kommersant” and the weekly “Argumenty i Fakty”, all published from September 2015 to February 2018 . 7 Владимир Путин четче обозначил некоторые позиции . Коммерсант, 12 .10 .2015 . Cfr . also: “Wir wollen keine Supermacht sein” Interview mit Wladimir Putin . Bild, 12 January 2016 . 8 Мы не поддерживаем Асада, для нас важно сохранить сирийскую государственность . Коммерсант, 14 October 2015 . 9 Дмитрий Медведев: Россия борется с террористами, а не защищает Башара Асада, Коммерсант, 17 October 2015 . 10 Сергей Марков: Великое сотрудничество континентов? Известия, 1 October 2015 .

IX . The political powers and Syria · 133 en to Assad with two arguments . On the one hand, he was still the best ally against the Islamist threat, and on the other, the foreign powers, such as the West and the Gulf monarchies, were eager to overthrow the Syrian president, which would pave the way for Syria’s Islamist fragmentation: We protect not only Syria’s state unity, we prevent the establishment of a deployment area for [Jihadists] as well, which would unsettle the Russian statehood . […] Russia is home to 20 million Muslims . […] If such a deployment area was created in Syria, the situation would become dangerous . That is why Russia defends not only the interests of its ally, but its own .11

The General blamed the West, with its support of the Syrian opposition, for destabilizing Syria and bringing forth such developments, which threatened Russia’s national security . This narrative, which has also been officially and repeatedly brought forward, was followed by other hard objectives, which, in spite of having never been officially expressed as such, are a matter of open discussion in the internal Russian debate . First of all, the expansion of Russia’s geostrategic position . The publicist Alexander Prokhanov hit the nail on the head when he explained, in September 2015: the Russian naval base in Syria’s Tartus is absolutely necessary to secure Russia’s presence in the Mediterranean . This base is advantageous because it stands closer to the conflict regions in the eastern Mediterranean than the US naval bases in Barcelona and Naples do . And what he meant was not just a question of warships: Syria […] is Russia’s pivotal base in the Middle East . The region contains incredible oil reserves . The veins of worldwide communication cross their ways there, and global conflicts pile up . That’s why Syria is Moscow’s defensive line .12

There is another goal in Russia’s portfolio, and it is a military one . It might be described with the phrase “warfare for warfare’s sake” . Within the frame of Syria’s intervention, the General Staff and political leaders in Moscow aimed to develop and expand Russia’s ability for modern warfare and military expeditions in remote areas . Further than that, Syria was used as a shooting range . The Russian military tested new weapons under operating conditions, including drones, Kalibr cruise missiles, and from summer 2018, the new stealth fighter SU-57 .13 In Russia, therefore, there is open talk of the “Syrian school of modern warfare” . 11 Цель ИГИЛ — раскачать Ближний Восток для дестабилизации России . Известия, 9 October 2015 . 12 Александр Проханов: Москва, Дамаск — рубежи обороны . Известия, 9 October 2015 . 13 In the Russian public the following lessons, learnt in Syria, have been intensely discussed: improvement of the air force’s operational and command abilities, training of pilots, staff and line officers under combat conditions; by means of a rotation system, the majority of officers saw action; such “Syrians” will form the nucleus of a new command elite; combat experience for more than 5000 rank and file, improvement of the logistic

134 · Alex ander Will In addition to these three hard goals of war, a few ‘soft’ ones may be recognized as well . Overcoming the “Afghan trauma” has been a major concern in Russian society and the army . The war in Afghanistan, which was fought and lost by the Soviet government in the 1980s, played a significant role in the public debate at the beginning of the Syrian intervention . There was widespread anxiety about the possibility of another lossy and unsuccessful intervention war, as well as of attacks on the home front . This could notably be observed in the rhetorical question of a military analyst, asking where “is the Reichstag in Syria”, in which to hoist “the victory banner”,14 as well as in demands for an early exit strategy . The unease, however, was ambiguous, reflecting the trauma of defeat in Afghanistan on the one hand and, on the other, the longing for a victory comparable to that of the Soviet Union in the Great Patriotic War . Russia’s war aims were understood along the same lines by American politicians . Shortly after the Russian intervention in 2015, a quick analysis by the Congressional Research Service in Washington read: Moscow’s primary intentions may include safeguarding the Asad regime, preserving Russian naval access to Syria, and challenging U .S . policy toward Syria . Putin’s recent call for an all-out effort against the Islamic State also may stem from the sizable number of jihadist fighters from the North Caucasus fighting in Syria, who may pose a serious problem for Moscow should they return to Russia .15

In view of Syria’s future, however, two of these war goals remain relevant: to unconditionally maintain the country’s territorial integrity, free from Islamist and jihadist forces, and to establish a strong military and political position for Russia in the Middle East, making Syria an unsinkable aircraft carrier . Both targets bring Russia into conflict with the interests of other powers .

Turkey This is true especially regarding Turkey . In recent years, we have repeatedly seen periods of confrontation and cooperation in the relationship between Ankara and Moscow in Syria . While cooperation has been a tactical move, confrontation may ultimately mark

capabilities and generation of the capability to conduct military expeditions; testing ground for high-precision weapons and thus reduction of the Russian backlog to the US in this area; development and testing of new tactical principles . Cfr . Руслан Пухов: Война, которую Россия выиграла . Известия, 13 October 2017 . Сирийская школа современной войны . Известия, 29 December 2017 . 14 Александр Коновалов: Надо выбирать момент, когда уходить из Сирии . Известия, 13 October 2017 . 15 Humud et al . 2015, 2 .

IX . The political powers and Syria · 135 the relationship of the two powers regarding Syria in the long run . Their strategic interests are too contradictory .16 Ankara is pursuing two main goals beyond its borders, a short-term and a long-term one . So as in Russia’s case, the main objective is to ward off an internal threat to the Turkish state, namely, the Kurdish autonomy and secessionist aspirations . Any form of Kurdish statehood on Syrian soil is considered by Ankara as a threat to its own state integrity . Every Turkish government so far, either Kemalist or Islamist, has always dreaded irredentist movements, and has, therefore, fought Kurdish independence efforts even beyond its borders . Statements of the Turkish President Recep Tayip Erdogan regarding “the extermination of terrorists” leave no room for doubt in this respect .17 So far, there is no inconsistency between Ankara and Moscow . Turkey’s second, long-term goal in Syria, however, is incompatible with Russia’s ambitions . Neo-Ottomanism is the fitting concept to characterize the policy of the Turkish regime under Erdogan and his party, the AKP . Ultimately, this foreign policy strategy aims at largely restoring the Ottoman Empire (which collapsed in the wake of World War One) under a Sunni-Islamist regime, the ‘province’ of Syria, with such important cities as Aleppo and Damascus, being part of it . Accordingly, this is not just war against secessionist aspirations, but an imperialist-driven expansionism as well . Annexations of Syrian territory to Turkey, though, run against Russian interests .18 As part of the Astana peace process, which was initiated by Russia, Moscow favors a federal solution to the Kurdish problem within an all-Syrian state . Turkey refuses to accept that .19 Neo-Ottomanism, as well as islamism, characterize the political thinking of the ruling political elites in Turkey . There is an infinity of symptoms to that: Ever since Istanbul is being governed by the AKP, each 29th May, the country commemorates the city’s conquest by Sultan Mehmed II in 1453 .20 The AKP and Erdogan maintain a special relationship with the former ruling house of Osman . In February 2015, Erdogan personally ordered a military expedition to northern Syria, in order to recover the remains of Suleiman Shah, grandfather of the dynasty’s founder, Osman I .21 Using diplomatic channels, Erdogan arranged the “homecoming” of the last heir to the throne, Dündar Abdulkerim Osmanoglu, who had grown up in Damascus, following the banish16 Cfr . on this conflict of interest Sahin 2014 . 17 Erdogan: Türkei will bei Syrien-Offensive “Terroristen ausrotten”, Deutsche Presseagentur, 24 January 2018, 21 .38h . 18 In the meantime, the assumption that the Neo-Osmanism of the AKP regime is essentially peaceful and not aimed at territorial expansion, but at winning favour culturally, has been proven wrong . On such interpretations, for instance, Erdmann – Herzog 2012 . 19 Sahin 2014, 2 . 20 K . Krüger, Türken und Araber werden eins . Frankfurter Allgemeine Zeitung, 16 September 2011, www .faz .net/ aktuell/feuilleton/neo-osmanismus-türken-und-araber-werden-eins .html (20 January 2018) . 21 S . Moubayed, The Turkish president imagines the rebirth of the Ottoman Empire in Syria, 29 May 2017, https:// www .alternet .org/world/turkish-president-imagines-rebirth-ottoman-empire-syria (20 January 2018) .

136 · Alex ander Will ment of the Osman family from the Turkish Republic in 1924 .22 Countless historical TV series and feature film productions also proclaim the glory of the lost Ottoman Empire .23 AKP’s accession to power was followed suit by an intensive cultural and religious-political propaganda abroad . From 2009 on, dozens of Junus Emre Institutes have been established . These are institutes of cultural activities abroad, concentrating on the territory of the former Ottoman Empire . No wonder, that the first one was opened in Sarajevo .24 The Turkish state television TRT has enormously enlarged its broadcasting capacity and with its Arabic channel, as well as the Turkish program, is now in a position to reach more than 350 million Arabic-speaking people .25 Neo-Ottomanism with its intrinsic pan-islamic sense of mission has been, therefore, the ideological foundation of AKP’s rule in Ankara in 2018 . This manifested itself in a wide range of political symbolism, which is based on the alleged “great age” of the Ottoman Empire . In January 2018, for instance, the deployment of Turkish troops was greeted by a Janissary military band dressed in early modern costumes, while the pictures were published by the state news agency Anadolu .26 A monumental portrait of Sultan Abdulhamid II was hung during a pro-Palestinian mass demonstration organized by AKP on 10 December 2017 in Istanbul . This Sultan, who was overthrown by the Young Turks in 1908, is the central figure of pan-Islamism, a political idea aspiring to unite all Islamic nations under the Ottoman leadership . And this is exactly where we find the borderline to a neo-Ottomanism, which, leaving aside all cultural issues, is in the end nothing else than classic imperialism . From time to time, Erdogan makes that perfectly clear . In January 2018, he pronounced in a speech: “We will not be prisoners on 780 000 square kilometers”,27 hinting at Turkey’s current size . In 2017, he challenged the Treaty of Lausanne, which defined the borders of his country after the First World War: “We have not voluntarily accepted our current borders . Our founding fathers were born outside these borders .”28 22 Last heir to Ottoman throne evacuated from Syria upon Erdoğan’s order, Daily Sabah, 29 August 2017, https:// www .dailysabah .com/turkey/2017/08/30/last-heir-to-ottoman-throne-evacuated-from-syria-upon-erdogans-order (24 January 2018) . 23 See on this media-related pan-ottoman approach: Erdogans Lieblingsserie, Die Tageszeitung, 1 August 2017, http://www .taz .de/!5430376/ (24 January 2018), D . Hussain, The revival of Ottomanism in shaping Turkey’s influence in the Muslim world, Foreign Policy Journal, 26 September 2017, www .foreignpolicyjournal .com/2017/09/26/the-revival-of-ottomanism-in-shaping-turkeys-influence-in-the-muslim-world/ (24 January 2018) . 24 Krüger, Türken (note 20) . 25 Ibid . 26 Cfr . Hüriyet Daily News, 22 January 2018, http://www .hurriyetdailynews .com/ottoman-military-band-performs-at-turkey-syria-border-126102 (24 January 2018) . 27 Erdogan will an Größe des Osmanischen Reiches anknüpfen, Die Welt, 10 November 2016, https://www .welt . de/politik/ausland/article159401434/Erdogan-will-an-Groesse-des-Osmanischen-Reiches-anknuepfen .html (24 January 2018) . 28 H . Kazim, Erdogans Traum vom Osmanischen Reich, Spiegel Online, 26 October 2016, http://www .spiegel .de/ politik/ausland/tuerkei-recep-tayyip-erdogan-traeumt-vom-osmanischen-reich-a-1118342 .html (30 June 2018) .

IX . The political powers and Syria · 137 Erdogan’s neo-Ottoman rage targets Greece as well: “In the Treaty of Lausanne, we have given away islands . So close to us, that we can hear your voices . These were our islands . That’s where our mosques stand .”29 In 2017, the public were upset in Israel, when Erdogan suggested that the Israelis remember the good times of Ottoman rule, attributing Israel’s problems today to the downfall of the Ottoman Empire .30 Turkey’s support to Hamas, the governing authority in the Gaza Strip, as well as to various islamist groups in Israel, is also part of this strategy .31 The Israeli government, in turn, supports Kurdish independence efforts, a policy that Ankara is unwilling to accept .32 Ultimately, so much can be said: as long as Ankara pursues its neo-Ottoman imperialist dreams, Turkey will remain involved in conflicts with other powers in Syria . This applies particularly to Russia, but also to any country lying on the territory of the former Ottoman Empire, including Israel .

Israel Israel itself has no immediate interests in Syria . The question of Golan, which has been under Israeli control since 1967 and was later annexed, is considered as finally resolved by the Israeli public and by the majority of the political parties .33 Syria with its outdated and disorganized army was discarded as a real military threat as early as 1973, after the Yom Kippur War . The Israelis have come to terms with Russia .34 Jerusalem and Moscow are not best friends, the two powers, however, have found a modus vivendi . Even massive Israeli air raids on Iranian targets in Syria in the summer of 2018 did not affect that .35 29 Ibid . 30 H . Keinon, Israel to Erdogan after Jerusalem tirade: The days of the Ottoman Empire are over, The Jerusalem Post, 25 July 2017, https://m .jpost .com/Arab-Israeli-Conflict/Israel-to-Erdogan-after-Jerusalem-tirade-Daysof-Ottoman-Empire-are-over-500740 (30 June 2018) . 31 Efron 2018, 26–30 . 32 Ibid ., 32–33 . 33 This became apparent in an editorial jointly written by Moshe Ya’alon, a conservative Likud politician and former Minister of Defence, and Yair Lapid, founder and chairman of the liberal party Yesh Atid . In this article, in view of the desperate situation in Syria both politicians call for the international recognition of the annexation of the Golan by Israel: Y . Lapid, M . Ya’alon, Will the West cede the Golan Heights to a psychopath?, The Times of Israel, 01 July 2018, https://www .timesofisrael .com/will-the-west-cede-the-golan-heightsto-a-psychopath/ (02 July 2018) . 34 Cfr . Defense Minister says Russia is a ‘very pragmatic’ actor in Syria, The Times of Israel, 28 April 2018, https:// www .timesofisrael .com/defense-minister-says-russia-is-a-very-pragmatic-actor-in-syria/ (30 June 2018) . 35 A . Fulbright, Russia OKs Israeli strikes on Iranian targets deep inside Syria — report, The Times of Israel, 01 June 2018, https://www .timesofisrael .com/russia-oks-israeli-strikes-on-iranian-targets-deep-inside-syriareport/ (30 June 2018) . A . Fulbright, Israel, Russia said to reach secret deal on pushing Iran away from Syria border, The Times of Israel, 28 May 2018, https://www .timesofisrael .com/israel-russia-said-to-agree-on-pushing-iran-from-syria-border/ (30 June 2018) .

138 · Alex ander Will In the middle of 2018, Israel’s interest in Syria was utterly centered on Iran . After Russia, Iran had become a major supporting pillar to the Assad regime . In Tehran, there was talk of the “Axis of Resistance”, stretching from Iran across the Shiite areas of Iraq, over to Syria and up to Lebanon . “Resistance” here is meant for the destruction of the “Zionist entity”, namely Israel .36 In the middle of 2018, Israel considered itself to be in a precarious position . In the north, in Lebanon, the Shiite Hezbollah militia, which is supported with weapons by Iran, had an overpowering influence . Their missiles reached far into Israel . In the East, the presence of Iranian military advisers, as well as regular troops, continued to increase . So how would Jerusalem act, if the worst comes?37 The answer should depend on the opportunities . As late as January 2018, it seemed that in the event of a major conflict and a massive attack on those fronts, Israel’s strategy would be exclusively defensive . The counter-attack would then be pointed directly against Iran .38 In the event of a general conflict, this strategy is certainly not excluded, in the first half of 2018, however, the Israeli Security Cabinet opted for a more direct intervention in Syria . This is demonstrated by a series of massive airstrikes, which have been regularly launched against Iranian and Shiite militias in the country since February 2018 . Voices in the Israeli Security Cabinet, the key body for taking security policy decisions, pleaded it was high time to definitely drive Iran out of Syria .39 On the other hand, the Israeli government has repeatedly sought Russia’s support in their endeavors to mitigate Iran’s influence on the neighboring country .40 The topic has been a source of new conflicts with Turkey, however, which, from the middle of 2017, has been increasingly willing to accept Iran as a power factor in Syria .41 Apart from Iran’s threat, Israel will, therefore, remain largely defensive in Syria . Israel’s main concern is the defence of its own country and the containment of Iran’s influence, whereby the Israeli governments will continue to seek both cooperation with Russia and support from the United States .

36 Cfr . Drums of war: Israel and the ‘Axis of Resistance’ = Middle East Report No . 97, 02 August 2010, International Crisis Group . M . Sullivan, Hezbollah in Syria = Middle East Security Report 19, April 2014 . 37 For an investigation of Israel’s reactions to the Syrian civil war by the end of June 2018 cfr . S . J . Frantzmann, Syria challenge: How will Netanyahu act?, The Jerusalem Post, 3 July 2018, https://m .jpost .com/Middle-East/ Syria-Challenge-How-will-Netanyahu-act-561313/amp?__twitter_impression=true (3 July 2018) . 38 Cfr . the guest opinion by the Israeli Ministry’s of Intelligence Services Head of Department for international relations, Arye Sharuz Shalicar, Die Fronten klären sich, Nordwest-Zeitung, 28 February 2018 . 39 Top minister: Israel should eradicate any trace of Iranian entrenchment in Syria, The Times of Israel, 12  May  2018, https://www .timesofisrael .com/top-minister-israel-should-eradicate-any-trace-of-iranian-entrenchment-in-syria/ (30 June 2018) . 40 A . Fulbright, US, Russia reach Syria agreement to distance Iran from Golan Heights — reports . The Times of Israel, 12 September 2017, https://www .timesofisrael .com/us-russia-reach-syria-agreement-to-distance-iranfrom-golan-heights-reports/ (30 June 2018) . 41 Efron 2018, 31–32 .

IX . The political powers and Syria · 139

United States and European Union For the United States, Syria does not present a geopolitical priority . Basically, the Americans have no important interests there . The Donald Trump administration has mainly pursued five goals in Syria since taking office: ȣ the final destruction of the Islamic State and the Islamic terrorist network Al Qaeda . ȣ a political solution for the Syrian civil war through negotiations between the central government and opposition groups . ȣ reducing Iranian influence in Syria, eliminating threats starting from Syria over to neighboring countries . ȣ return of the refugees . ȣ Syria devoid of weapons of mass destruction .42 The Americans maintained a small military presence of about 2,000 men in Syria and on the Iraqi border,43 supporting selected opposition militias .44 Agreements have been made with Russia to avoid military confrontations, though they are of a fragile character .45 In Syria, the Americans only react . Moscow and Ankara have their say . The withdrawal of US forces from Syria announced by President Donald Trump in December 2018 highlighted his administration’s relative lack of interest in the Syrian crisis .46 Roughly the same applies to the European Union, which, apart from the refugee issue, has basically never pursued a consistent policy in Syria . This is witness to the failure of a joint European foreign policy, as well as a failure of foreign policy strategies based on soft power . Prior to the Russian intervention in Syria, there was a real chance for Europe to prove its supposed role as a peace-making power . This, however, would have implied opting for a massive military intervention . Considering the post-heroic character of societies in Europe, the limited military capabilities and internal instability of the Union, this option therefore never came into question .

Perspectives All attempts to predict the further development of the Syrian conflict will inevitably be speculations . They may be, however, at least well-founded speculations . Politicians in Moscow have scored a success in the Syrian question, although they know that in order 42 43 44 45 46

Humud – Blanchard – Nikitin 2018, 17 . Ibid ., 18–21 . Ibid ., 21–23 . Ibid ., 20 . Cfr . USA bereiten Abzug aller ihrer Truppen aus Syrien vor . Deutsche Presseagentur, 19 December 2018, 17 .49h .

140 · Alex ander Will to secure this success and to realize its strategic goals, it is necessary to achieve a political balance for the various ethnic and religious forces within Syria . It is to this end that the Russian government has supported the Astana Process . The Astana Process has more than once contributed to the de-escalation of regional conflicts within Syria . In any case, president Vladimir Putin is not married to Assad . If need arises, or interests change, Russia will drop the regime in Damascus as naturally as it now supports it . The same applies to its cooperation with Iran . Syria’s fate will therefore be decided from the outside . If before 2015 the Syrian state seemed likely to break apart, that possibility is not urgent any more . It is unlikely that new borders will be drawn, or that the post-war order in the Middle East will be changed, although this might have benefited such ethnic groups as the Kurds . The probability is, instead, that under the formal roof of a unitary state, a system of zones of influence might be emerging in Syria . The interested powers — especially Russia, Turkey and Israel — could, in one way or another, mark out and balance their interests, thus restoring peace to the country . The Russians and the British did the same in 1905 in Persia . Ankara has already cemented its presence in the former self-governed Kurdish areas in the north since mid-2018, while Moscow has brought the central government in Damascus in complete dependency . A future with more bloodshed is also possible . A ‘major’ war in the Middle East could be particularly sparked by the Iranian presence in Syria . But even Moscow and Ankara could come to blows, if a reconciliation of interests does not succeed . We know enough about Russian-Turkish wars from the past . In spite of everything, albeit only one is for Syria not in sight: an imminent peace followed by a democratization process of a Western model .

Contributors Matthew Adam Cobb received his PhD from Swansea University in 2012 and currently lectures in ancient history at the University of Wales Trinity Saint David . His research focuses on Graeco-Roman participation in the Indian Ocean trade, as well as cross-cultural engagement between West and East in antiquity . Among other works, he is the author of Rome and the Indian Ocean Trade from Augustus to the Early Third Century CE (Leiden, 2018) and the edited book The Indian Ocean Trade in Antiquity: Political, Cultural and Economic Impacts (Abingdon, 2019) . Ted Kaizer is Professor in Roman Culture and History at Durham University . He is the author of The Religious Life of Palmyra (Stuttgart, 2002), has edited or co-edited five volumes, including Religion, Society and Culture at Dura­Europos (Cambridge , 2016) and, with Margherita Facella, Kingdoms and Principalities in the Roman Near East (Stuttgart, 2010), and has written numerous articles on religious and social history of the Roman Near East . He was awarded a Major Research Fellowship by the Leverhulme Trust for 2014–2017) . Eivind Heldaas Seland teaches ancient history and global history at the University of Bergen, Norway . His research addresses trade and contacts in the western Indian Ocean and the Near East in the pre-Islamic period . He is the author of Ships of the Desert, Ships of the Sea: Palmyra in the World Trade of the first three centuries CE (Wiesbaden, 2016) . Jørgen Christian Meyer is Professor of Ancient History at the University of Bergen . His fields of research include Palmyra and the surrounding territory, nomadic networks on the Arabian Peninsula and the relations between the Middle East and Central Asia . He directed the University of Bergen’s Palmyrena project and is the author of Palmyrena. Palmyra and the Surrounding Territory from the Roman to the Early Islamic Period (Oxford, 2017) . He recently edited, with Eivind Heldaas Seland and Nils Anfinset, Palmyrena. City, Hinterland and Caravan Trade between Orient and Occident (Oxford, 2016) . Georg Müller is a Lecturer in Ancient History at the University of Vechta and a Lecturer in Latin Studies at the Carl von Ossietzky University of Oldenburg . He studied Latin, Greek and History at the Saarland University at Saarbrücken . His research focuses on Zosimus’ historiographical work . Ann-Christine Sander studied Ancient History and Sociology at the Universities of Osnabrück and Münster, as well as Criminology at the University of Leicester . She is now a PhD student at the Carl von Ossietzky University of Oldenburg . Her research, funded by the graduate school “Cultures of Participation”, is concerned with the elite of Palmyra .

142 · Contributors Michael Sommer teaches ancient history at the Carl von Ossietzky University of Oldenburg, Germany . His areas of research include the cultural history of the Hellenistic and Roman Near East, cross-Mediterranean history and the Roman Empire . Among his recent publications are Die Soldatenkaiser (Darmstadt, 2020), Palmyra. A History (London, 2018), Roms orientalische Steppengrenze (Stuttgart, 22018) . With Tassilo Schmitt he is the editor of Von Hannibal zu Hitler. “Rom und Karthago” 1943 und die deutsche Altertumswissenschaft im Nationalsozialismus (Darmstadt, 2019) . Alexander Will is chief political editor of the Nordwest­Zeitung at Oldenburg . He studied Islamic Studies at the University of Hamburg, at the Hebrew University and at the School of Oriental and African Studies in London and took his PhD from the Saarland University at Saarbrücken . He is the author of Kein Griff nach der Weltmacht. Geheime Dienste und Propaganda im deutsch­österreichisch­türkischen Bündnis, 1914–1918 (Wien – Köln 2012) .

Bibliography al-Asʿad – Gawlikowski 1986–1987: Ḫ . al-Asʿad – M . Gawlikowski, New honorific inscriptions in the Great Colonnade of Palmyra, Annales Archéologiques Arabes Syriennes 36–37, 1986–1987, 164–171 . al-Dbiyat 2009: M . al-Dbiyat, Eau et peuplement dans les marges arides . Le cas de la région de Salamya en Syrie centrale, in: M . Mouton – M . al-Dbiyat (ed .), Stratégies d’acquisition de l’eau et société au Moyen-Orient depuis l’antiquité . Études de cas (Beyrouth 2009) 179–200 . al-Maqdissi 2000: M . al-Maqdissi, Note sur les sondages réalisés par Robert du Mesnil du Buisson dans la cour du sanctuaire de Bêl à Palmyre, Syria 77, 2000, 137–158 . Alfani – Ó Gráda 2017: G . Alfani – C . Ó Gráda (ed .), Famine in European history (Cambridge 2017) . Alston 1995: R . Alston, Soldier and society in Roman Egypt . A social history (London 1995) . Alston 2007: R . Alston, Fraying round the Edges . Models of change on the margins, in: Aspects of the Roman East . Papers in honour of Professor Fergus Millar (Turnhout 2007) 1–32 . Andrade 2012: N . J . Andrade, Inscribing the citizen . Soados and the civic context of Palmyra, Maarav 19, 2012, 65–90 . Andrade 2013: N . J . Andrade, Syrian identity in the Greco-Roman world (Cambridge 2013) . Andrade 2018a: N . J . Andrade, The journey of Christianity to India in Late Antiquity . Networks and the movement of culture (Cambridge 2018a) . Andrade 2018b: N . J . Andrade, Zenobia . Shooting star of Palmyra (Oxford 2018b) . Arnaud 1998: P . Arnaud, Les guerres parthiques de Gabinius et de Crassus et la politique occidentale des Parthes Arsacides entre 70 et 53 av . J .-C ., Electrum 2, 1998, 13–34 . Ast – Bagnall 2016: R . Ast – R . S . Bagnall, Documents from Berenike . 3 . Greek and latin texts from the 2009–2013 seasons 36, Papyrologica Bruxellensia (Bruxelles 2016) . Avni – Porat – Avni 2013: G . Avni – N . Porat – Y . Avni, Byzantine-Early Islamic agricultural systems in the Negev highlands . Stages of development as interpreted through OSL dating, Journal of Field Archaeology 38, 2013, 332–346 . Bagnall – Helms – Verhoogt 2000: R . S . Bagnall – C . Helms – A . M . F . W . Verhoogt, Documents from Berenike I . Greek ostraka from the 1996–1998 seasons 31, Papyrologica Bruxellensia (Bruxelles 2000) . Bagnall et al . 2005: R . S . Bagnall – C . Helms – A . M . F . W . Verhoogt – A . Bülow-Jacobsen, Documents from Berenike II . Texts from the 1999–2001 seasons 33, Papyrologica Bruxellensia (Bruxelles 2005) . Ball 2000: W . Ball, Rome in the East . The transformation of an empire (London 2000) . Barański 1997: M . Barański, Western Aqueduct in Palmyra, Studia Palmyrénski 10, 1997, 7–18 . Barisitz 2017: S . Barisitz, Central Asia and the silk road . Economic rise and decline over several millennia (2017) . Bell 2000: G . L . Bell, The Arabian diaries, 1913–1914 (Syracuse, NY 2000) . Bellinger 1949: A . R . Bellinger, The excavations at Dura-Europos . Final report, 6 . The Coins (New Haven/London 1949) .

14 4 · Bibliogr aphy Bernard 1990: P . Bernard, Vicissitudes au gré de l’histoire d’une statue en bronze d’ Héraclès entre séleuccie du Tigre et la Mésène, Journal des Savants 1–2, 1990, 3–68 . Boehringer 1998: C . Boehringer, in: R . Ashton – S . Hurter (ed .), Studies in Greek numismatics in memory of Martin Jessop Price (London 1998) 43–53 . Boiy 2004: T . Boiy, Late Achaemenid and Hellenistic Babylon 136, Orientalia Lovaniensia analecta (Leuven 2004) . Bowersock 1973: G . W . Bowersock, Syria under Vespasian, Journal of Roman Studies 68, 1973, 133–140 . Bowersock 1987: G . W . Bowersock, The Hellenism of Zenobia, in: J . T . A . Koumoulides (ed .), Greek connections . Essays on culture and diplomacy (Bloomington, Ind . 1987) 19–27 . Bowersock 1989: G . W . Bowersock, Social and economic history of Syria under the Roman Empire, in: J .-M . Dentzer – W . Orthmann (ed .), La Syrie de l’époque achémenide à l’avènement de l’Islam (Saarbrücken 1989) 63–80 . Bowersock 2013: G . W . Bowersock, The Throne of Adulis . Red Sea wars on the eve of Islam (Oxford 2013) . Bowman 1976: A . K . Bowman, Papyri and Roman Imperial History, 1960–75, Journal of Roman Studies 66, 1976, 153–173 . Brokaert 2017: W . Brokaert, Going mental . Culture, exchange and compromise in Rome’s trade with the East, in: H . F . Teigen – E . H . Seland (ed .), Sinews of empire . Networks in the Roman Near East and beyond (Oxford 2017) 1–22 . Bron 1986: F . Bron, Palmyréniens et Chaldéens en Arabie du Sud, Studi epigrafici e linguistici sul Vicino Oriente 3, 1986, 95–98 . Bru 2016: H . Bru, Le culte impérial dans l’Orient romain . Mythes, rites et structures, in: A . Kolb – M . Vitale (ed .), Kaiserkult in den Provinzen des Römischen Reiches . Organisation, Kommunikation und Repräsentation (Berlin 2016) 57–77 . Brun 2006a: J .-P . Brun, Chronologie de l’équipement de la route à l’époque gréco-romaine, in: H . Cuvigny (ed .), La route de Myos Hormos . L’armée romaine dans le désert oriental d’Égypte (Le Caire 2006) 187–205 . Brun 2006b: J .-P . Brun, Méthodes et conditions de fouille des fortins et des dépotoirs ou les affres d’un Gallo-Romain en Égypte, in: H . Cuvigny (ed .), La route de Myos Hormos . L’armée romaine dans le désert oriental d’Égypte (Le Caire 2006) 61–71 . Brun – Reddé 2011a: J .-P . Brun – M . Reddé, Le cadre géographique, in: H . Cuvigny (ed .), Didymoi . Une garnison romaine dans le désert oriental d’Égypte . 1 . Les fouilles et le materiel (Le Caire 2011) 9–16 . Brun – Reddé 2011b: J .-P . Brun – M . Reddé, Le fort, in: H . Cuvigny (ed .), Didymoi . Une garnison romaine dans le désert oriental d’Égypte . 1 . Les fouilles et le materiel (Le Caire 2011) 17–113 . Bukharin 2012a: M . D . Bukharin, The Greek inscriptions at Hoq, in: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the Cave Hoq (Bremen 2012) 494–500 . Bukharin 2012b: M . D . Bukharin, Greeks on Socotra . Commercial contacts and early Christian missions, in: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the Cave Hoq (Bremen 2012) 501–539 . Bülow-Jacobsen 2006: A . Bülow-Jacobsen, Toponyms and Proskynemata, in: H . Cuvigny (ed .), La route de Myos Hormos . L’armée romaine dans le désert oriental d’Égypte (Le Caire 2006) 51–59 .

Bibliogr aphy · 145 Burgersdijk 2007: D . Burgersdijk, Rome en Palmyra in de crisis van de derde eeuw na Christus, Phoenix 53, 2007, 5–23 . Burghardt 1971: A . F . Burghardt, A hypothesis about gateway cities, Annals of the Association of American Geographers 61, 1971, 269–285 . Butcher 2003: K . Butcher, Roman Syria and the Near East (London 2003) . Casson 1989: L . Casson, The Periplus Maris Erythraei (Princeton (N .J .) 1989) . Celentano 2016: C . Celentano, Palmyrenes abroad . Traders and patrons in Arsacid Mesopotamia, PHASIS 2016, 30–57 . Cobb 2014: M . A . Cobb, The exchange of goods from Italy to India during the early Roman Empire . The range of travelling times, Ancient West & East 13, 2014, 89–116 . Cobb 2015: M . A . Cobb, The chronology of Roman trade in the Indian Ocean from Augustus to early third century CE, Journal of the Economic and Social History of the Orient 58, 2015, 362–418 . Cobb 2018: M . A . Cobb, Rome and the Indian Ocean trade . From Augustus to the early third century CE 418, Mnemosyne supplements . History and archaeology of classical antiquity, (Leiden 2018) . Cohen 2013: G . M . Cohen, The Hellenistic settlements in the East from Armenia and Mesopotamia to Bactria and India 54, Hellenistic culture and society (Berkeley 2013) . Cooper 2014: J . P . Cooper, The medieval Nile . Route, navigation, and landscape in Islamic Egypt (Cairo 2014) . Cumont 1926: F . Cumont, Fouilles de Doura-Europos (Paris 1926) . Cuvigny 2005: H . Cuvigny, Ostraca de Krokodilô . La correspondance militaire et sa circulation, O Krok . 1–151 51, Fouilles de l’Institut français d’archéologie orientale du Caire (Le Caire 2005) . Cuvigny 2006: H . Cuvigny, La route de Myos Hormos . L’armée romaine dans le désert oriental d’Egypte, Praesidia du désert de Bérénice 1 2(Le Caire 2006) . Cuvigny 2011a: H . Cuvigny (ed .), Didymoi . Une garnison romaine dans le désert oriental d’Égypte . 1 . Les fouilles et le matériel (Le Caire 2011) . Cuvigny 2011b: H . Cuvigny, Le cadre historique, in: H . Cuvigny (ed .), Didymoi . Une garnison romaine dans le désert oriental d’Égypte . 1 . Les fouilles et le materiel (Le Caire 2011) 3–7 . Cuvigny 2012a: H . Cuvigny (ed .), Didymoi . Une garnison romaine dans le désert oriental d’Égypte . 2 . Les textes (Le Caire 2012) . Cuvigny 2012b: H . Cuvigny, Introduction, in: H . Cuvigny (ed .), Didymoi . Une garnison romaine dans le désert oriental d’Égypte . 2 . Les textes (Le Caire 2012) 1–37 . Cuvigny 2012c: H . Cuvigny, Les inscriptions de Didymoi, in: H . Cuvigny (ed .), Didymoi . Une garnison romaine dans le désert oriental d’Égypte . 2 . Les textes (Le Caire 2012) 39–56 . Cuvigny 2012d: H . Cuvigny, Petits documents administratifs, comptables et juridiques, in: H . Cuvigny (ed .), Didymoi . Une garnison romaine dans le désert oriental d’Égypte . 2 . Les textes (Le Caire 2012) 117–178 . Cuvigny 2014: H . Cuvigny, Papyrological evidence on ‘barbarians’ in the Egyptian eastern desert, in: J . H . F . Dijkstra – G . Fisher (ed .), Inside and out . Interactions between Rome and the peoples on the Arabian and Egyptian frontiers in Late Antiquity (Leuven 2014) 165–198 .

146 · Bibliogr aphy Cuvigny – Briquel-Chatonnet 2012: H . Cuvigny – F . Briquel-Chatonnet, Instrumentum inscriptum [Tituli palmyréniens], in: H . Cuvigny (ed .), Didymoi . Une garnison romaine dans le désert oriental d’Égypte . 2 . Les textes (Le Caire 2012) 219–220 . Cuvigny – Gagos 2012: H . Cuvigny – T . Gagos, La correspondance officielle et sa circulation, in: H .  Cuvigny (ed .), Didymoi . Une garnison romaine dans le désert oriental d’Égypte . 2 . Les textes (Le Caire 2012) 103–104 . Daryaee 2003: T . Daryaee, The Persian Gulf trade in Late Antiquity, Journal of World History 14, 2003, 1–16 . Daubner 2015: F . Daubner, Gymnasia . Aspects of a Greek Institution in the Hellenistic and Roman Near East, in: Religious identities in the Levant from Alexander to Muhammed . Continuity and change (Turnhout 2015) 33–46 . Davidde 2017: B . Davidde, The Port of Qanaʾ, a junction between the Indian Ocean and the Mediterranean Sea, in: A . Wilson – A . K . Bowman (ed .), Trade, commerce and the state in the Roman world (Cambridge 2017) 579–597 . Delplace 2003: C . Delplace, Palmyre et l’Inde . A propos de l’inscription Inv . X 88, in: Hommages à Carl Deroux . 3 . Histoire et épigraphie . Droit (Leuven 2003) 158–167 . Delplace 2017: C . Delplace, Palmyre . Histoire et archéologie d’une cité caravanière à la croisée des cultures (Paris 2017) . Delplace – Dentzer-Feydy 2005: C . Delplace – J . Dentzer-Feydy, L’Agora de Palmyre 175, Bibliothèque archéologique et historique (Beyrouth 2005) . Dentzer-Feydy – Teixidor 1993: J . Dentzer-Feydy – J . Teixidor, Les antiquités de Palmyre au Musée du Louvre (Paris 1993) . Detlefsen 1909: S . D . F . Detlefsen, Die Anordnung der geographischen Bücher des Plinius und ihre Quellen 18, Quellen und Forschungen zur alten Geschichte und Geographie (Berlin 1909) . Dirven 1999: L . Dirven, The Palmyrenes of Dura-Europos . A study of religious interaction in Roman Syria (Leiden 1999) . Dodgeon – Lieu 1991: M . H . Dodgeon – S . N . C . Lieu, The Roman eastern frontier and the Persian Wars (AD 226–363) . A documentary history (London 1991) . Doyle 1988: M . W . Doyle, Empires 2(Ithaca 1988) . Dragendorff – Kryger 1924: H . Dragendorff – E . Kryger, Das Grabmal von Igel 1, Römische Grabmäler des Mosellandes und der angrenzenden Gebiete (Trier 1924) . Drexhage 1988: R . Drexhage, Untersuchungen zum Römischen Osthandel (Bonn 1988) . Dridi 2012: H . Dridi, Appendix II . Radiocarbon analysis of splinters from the wooden tablet, in: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the Cave Hoq (Bremen 2012) 461–462 . Drijvers 1976: H . J . W . Drijvers, The religion of Palmyra (Leiden 1976) . Du Mesnil du Buisson 1966: R . Du Mesnil du Buisson, Première campagne de fouilles à Palmyre, Comptes Rendus de l’Académie des Inscriptions et Belles-Lettres 1966, 158–191 . Du Mesnil Du Buisson 1967: R . Du Mesnil Du Buisson, La découverte de la plus ancienne Palmyre, Bibliotheca Orientalis 24, 1967, 20–21 . Duchesne – Vogel 1884: L . Duchesne – C . Vogel, Le liber pontificalis . Texte, introduction et commentaire (Paris 1884) .

Bibliogr aphy · 147 Duncan-Jones 1982: R . Duncan-Jones, The economy of the Roman Empire . Quantitative studies (Cambridge 1982) . Earle 1997: T . K . Earle, How chiefs come to power . The political economy in prehistory (Stanford, Calif . 1997) . Edwell 2008: P . M . Edwell, Between Rome and Persia . The middle Euphrates, Mesopotamia and Palmyra under Roman control (London 2008) . Efron 2018: S . Efron, The future of Israeli-Turkish relations (Santa Monica, CA 2018) . Ehling 1998: K . Ehling, Seleukidische Geschichte zwischen 130 und 121 v . Chr ., Historia 47, 1998, 141–151 . Ehling 2008: K . Ehling, Untersuchungen zur Geschichte der späten Seleukiden (164–63 v . Chr .) . Vom Tode des Antiochos IV . bis zur Einrichtung der Provinz Syria unter Pompeius 196, Historia Einzelschriften (Stuttgart 2008) . Elton 1996: H . Elton, Frontiers of the Roman Empire (London 1996) . English 1968: P . W . English, The origin and spread of qanats in the old world, Proceedings of the American Philosophical Society 112, 1968, 170–181 . Enßlin 1949: W . Enßlin, Zu den Kriegen des Sassaniden Schapur I 1947, 5, Sitzungsberichte der Bayerischen Akademie der Wissenschaften . Philosophisch-Historische Klasse (München 1949) . Erdkamp 2005: P . Erdkamp, The grain market in the Roman Empire . A social, political and economic study (Cambridge 2005) . Erdmann – Herzog 2012: G . Erdmann – O . Herzog, Die Türkei in Afrika . Im Schatten des NeoOsmanismus?, Giga Focus 1, 2012 . Ertel – Ployer 2013: C . Ertel – R . Ployer, Sondage II . Ein späthellenistisch-römisches Hofhaus . Baubefund, Architektur, Chronologie, in: A . Schmidt-Colinet – W . al-Asʿad (ed .), Palmyras Reichtum durch weltweiten Handel . Archäologische Untersuchungen im Bereich der hellenistischen Stadt (Wien 2013) 118–169 . Evers 2017: K . Evers, Worlds apart trading together . The organisation of long-distance trade between Rome and India in antiquity (Oxford 2017) . Fassbinder – Linck 2013: J . W . E . Fassbinder – R . Linck, Geophysikalische Prospektion, in: A . Schmidt-Colinet – W . al-Asʿad (ed .), Palmyras Reichtum durch weltweiten Handel . Archäologische Untersuchungen im Bereich der Hellenistischen Stadt (Wien 2013) 79–88 . Fauconnier 2012: B . Fauconnier, Graeco-Roman merchants in the Indian Ocean . Revealing a multicultural trade, in: M .-F . Boussac – J .-F . Salles – J .-B . Yon (ed .), Autour du ‘Périple de la mer Érythrée’ (Lyon 2012) 75–109 . Finley 1985: M . I . Finley, The ancient economy 2(Berkeley 1985) . Fischer 1970: T . Fischer, Untersuchungen zum Partherkrieg Antiochos’ VII . im Rahmen der Seleukidengeschichte . Diss . phil (Tübingen 1970) . Fisher 2008: G . Fisher, The political development of the Ghassān between Rome and Iran, Journal of late antiquity 1, 2008, 311–334 . Fox – Lieu – Ricklefs 2005: G . Fox – S . N . C . Lieu – N . Ricklefs, Select Palmyrene inscriptions, in: I . Gardner – S . N . C . Lieu (ed .), From Palmyra to Zayton . Epigraphy and iconography (Turnhout 2005) 27–126 .

148 · Bibliogr aphy Garnsey 1988: P . Garnsey, Famine and food supply in the Greco-Roman world . Responses to risk and crisis (Cambridge 1988) . Gaub 2016: F . Gaub, Scheidung auf Syrisch . Warum eine Aufteilung keine Lösung ist, Arbeitspapier Sicherheitspolitik 17, 2016 . Gawlikowski 1970: M . Gawlikowski, Monuments funéraires de Palmyre (Warszawa 1970) . Gawlikowski 1973: M . Gawlikowski, Palmyre VI . Le temple palmyrénien . Étude d’épigraphie et de topographie historique (Warszawa 1973) . Gawlikowski 1983: M . Gawlikowski, Palmyre et l’Euphrate, Syria 60, 1983, 53–68 . Gawlikowski 1985: M . Gawlikowski, Les princes de Palmyre, Syria 62, 1985, 251–261 . Gawlikowski 1988: M . Gawlikowski, Le commerce de Palmyre sur terre et sur eau, in: J .-F . Salles (ed .), L’Arabie et ses mers bordieres . I . Itinéraires et voisinages (Lyon 1988) 163–172 . Gawlikowski 1994: M . Gawlikowski, Palmyra as a trading centre, Iraq 56, 1994, 27–33 . Gawlikowski 1996: M . Gawlikowski, Palmyra and its caravan trade, Annales archéologiques arabes syriennes 42, 1996, 139–144 . Gawlikowski 1997: M . Gawlikowski, The Syrian Desert under the Romans, in: S . Alcock (ed .), The early Roman Empire in the East (Oxford 1997) 37–54 . Gawlikowski 2005: M . Gawlikowski, L’apothéose d’Odeinat sur une mosaǐque récemment découverte à Palmyre, Comptes rendus des séances – Académie des inscriptions & belles-lettres, 2005 .2005 149, 2005, 1293–1312 . Gawlikowski 2007a: M . Gawlikowski, Odainat et Hérodien, rois des rois, in: J .-B . Yon – P .-L . Gatier (ed .), Mélanges en l’honneur de Jean-Paul Rey-Coquais (Beyrouth 2007) 289–311 . Gawlikowski 2007b: M . Gawlikowski, Palmyra on the Euphrates, Mediterraneo Antico 10, 2007, 129–136 . Gawlikowski 2008: M . Gawlikowski, The statues of the sanctuary of Allat in Palmyra, in: Y . Z . Eliav – E . A . Friedland – S . Herbert (ed .), The sculptural environment of the Roman Near East . Reflections on culture, ideology, and power (Leuven 2008) 397–411 . Gawlikowski 2010a: M . Gawlikowski, Odainat of Palmyra between Rome and Persia, in: E . Dąbrowa (ed .), Hortus Historiae . Studies in honour of Professor Jozef Wolski on the 100th anniversary of his birthday (Kraków 2010) 467–479 . Gawlikowski 2010b: M . Gawlikowski, The Roman army in Palmyra under Tiberius, Studia Palmyrénski 11, 2010, 49–54 . Gawlikowski 2016a: M . Gawlikowski, Trade across frontiers . Foreign relations of a caravan city, in: J . C . Meyer – E . H . Seland – N . Anfinset (ed .), Palmyrena . City, hinterland and caravan trade between Orient and Occident (Oxford 2016) 19–28 . Gawlikowski 2016b: M . Gawlikowski, The urban development of Palmyra, ARAM 28, 2016, 485–496 . Gawlikowski – Żuchowska 2010: M . Gawlikowski – M . Żuchowska, La mosaïque de Bellérophon, Studia Palmyrénski 11, 2010, 9–42 . Geiger 2013: M . Geiger, Gallienus (Frankfurt am Main 2013) . Geiss 1991: I . Geiss, Kontinuitäten des Imperialismus, in: W . Reinhardt (ed .), Imperialistische  Kontinuität und nationale Ungeduld im 19 . Jahrhundert (Frankfurt am Main 1991) 12–30 .

Bibliogr aphy · 149 Geiss 1994: I . Geiss, Great powers and empires . Historical mechanisms of their making and breaking, in: G . Lundestad (ed .), The fall of the great powers . Peace, stability and legitimacy (Oslo 1994) 23–43 . Geiss 1999: I . Geiss, Imperien und Nationen – Zur universalhistorischen Topographie von Macht und Herrschaft, in: Neue politische Geschichte (Tel Aviv 1999) 57–91 . Genequand 2012: D . Genequand, Les établissements des élites Omeyyades en Palmyrène et au Proche-Orient, Bibliothèque archéologique et historique 200 (Beyrouth 2012) . Gerlach 2015: D . Gerlach, Herrschaft über Syrien . Macht und Manipulation unter Assad, Schriftenreihe der Bundeszentrale für Politische Bildung 1630 (Bonn 2015) . Geyer 2009: B . Geyer, Pratiques d’acquisition de l’eau et modalités de peuplement dans les Marges arides de la Syrie du Nord, in: M . Mouton – M . al-Dbiyat (ed .), Stratégies d’acquisition de l’eau et société au Moyen-Orient depuis l’antiquité . Études de cas (Beyrouth 2009) 25–44 . Gilbertson 1996: D . D . Gilbertson, Explanations . Environment as agency, in: G . Barker – D . D . Gilbertson – B . Jones – D . J . Mattingly (ed .), Farming the desert . The UNESCO Libyan Valleys Archaeological Survey I (Paris 1996) 291–318 . Gilbertson – Hunt 1996: D . D . Gilbertson – C . O . Hunt, Romano-Libyan agriculture . Walls and floodwater farming, in: G . Barker – D . D . Gilbertson – B . Jones – D . J . Mattingly (ed .), Farming the desert . The UNESCO Libyan Valleys Archaeological Survey I (Paris 1996) 191–226 . Gilbertson – Hunt – Gillmore 2000: D . D . Gilbertson – C . O . Hunt – G . Gillmore, Success, longevity  and failure of arid-land agriculture . Romano-Libyan floodwater farming in the  Tripolitanian pre-desert, in: G . Barker – D . Gilbertson (ed .), Archaeology of drylands: Living on margins (New York 2000) 137–159 . Gnoli 2007: T . Gnoli, The interplay of Roman and Iranian titles in the Roman East (1st–3rd century A .D .) (Wien 2007) . Gorea 2012a: M . Gorea, Palmyra and Socotra, in: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the Cave Hoq (Bremen 2012) 447–457 . Gorea 2012b: M . Gorea, The sea and inland trade of Palmyra, in: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the Cave Hoq (Bremen 2012) 463–485 . Graf forthcoming: D . F . Graf, Palmyra . The indigenous factor, in: A . S . Lewin (ed .), La città romana imperiale (n .n . forthcoming) . Grassi 2010: M . T . Grassi, La romanità orientale di Palmira . Nuove ricerche, Bollettino di Archeologia on line 1, 2010, 1–11 . Grassi 2011: M . T . Grassi, Une nouvelle mission conjointe syro-italienne à Palmyre (Pal .M .A .I .S .) . Observations préliminaires sur les campagnes 2007–2009 dans le quartier sud-ouest, in: Chronique archéologique en Syrie (Damascus 2011) 189–196 . Grassi 2017: M . T . Grassi, Palmira . Storie straordinarie dell’antica metropoli d’Oriente (Milano 2017) . Grassi – Zenoni – Rossi 2012: M . T . Grassi – G . Zenoni – G . Rossi, Tecniche e materiali dell’architettura Palmirena . Il caso dell’edificio con peristilio del quartiere sud-ouvest, in: M . P . Bologna – M . Ornaghi (ed .), Novissima studia . Dieci anni di antichistica milanese . Atti dei Seminari di Dipartimento 2011 (Milano 2012) 53–82 . Gregoratti 2011: L . Gregoratti, A Parthian port on the Persian Gulf . Characene and its trade, Anabasis . Studia Classica et Orientalia 2, 2011, 209–229 .

150 · Bibliogr aphy Gregoratti 2015: L . Gregoratti, The Palmyrene trade lords and the protection of the caravans, ARAM 27, 2015, 139–148 . Gregoratti 2016a: L . Gregoratti, Dura-Europos . A Greek town of the Parthian empire, in: T . Kaizer (ed .), Religion, society and culture at Dura-Europos (Cambridge 2016) 16–29 . Gregoratti 2016b: L . Gregoratti, Some observations concerning trade and territorial control in Roman Palmyra, ARAM 28, 2016, 531–540 . Gregoratti 2019: L . Gregoratti, Indian Ocean trade . The role of Parthia, in: M . A . Cobb (ed .), The Indian Ocean trade in antiquity . Political, cultural, and economic impacts (Abingdon, Oxon 2019) . Grout 2016: J . B . Grout, The role of Palmyrene temples in long-distance trade in the Roman Near East . Diss . phil . (Royal Holloway 2016) . Haiman 1995: M . Haiman, Agriculture and nomad-state relations in the Negev Desert in the Byzantine and early Islamic periods, Bulletin of the American Schools of Oriental Research 297, 1995, 29–53 . Hammad 2010: M . Hammad, Palmyre . Transformations urbaines . Développement d’une ville antique de la marge aride syrienne (Paris 2010) . Harl 1996: K . W . Harl, Coinage in the Roman economy, 300 BC to AD 700 (Baltimore 1996) . Hartmann 2001: U . Hartmann, Das palmyrenische Teilreich, Oriens et Occidens 2 (Stuttgart 2001) . Hartmann 2016: U . Hartmann, What is it like to be a Palmyrene in the age of crisis? Changing Palmyrene identities in the third century AD, in: A . Kropp – R . Raja (ed .), The world of Palmyra (København 2016) 53–69 . Haverfield 1923: F . Haverfield, The romanization of Roman Britain 4(Oxford 1923) . Healey 1996: J . F . Healey, Palmyra and the Arabian Gulf trade, Aram 8, 1996, 33–37 . Healey 2009: J . F . Healey, Aramaic inscriptions and documents of the Roman period 4, Textbook of Syrian semitic inscriptions (Oxford 2009) . Hekster – Kaizer 2004: O . Hekster – T . Kaizer, Mark Antony and the raid on Palmyra . Reflections on Appian, Bella Civilia V, 9, Latomus 63, 2004, 70–80 . Hill 2015: J . E . Hill, Through the Jade Gate . China to Rome . A study of the silk routes during the Later Han Dynasty 1st to 2nd centuries CE . An annotated translation of the chronicle on the ‘Western Regions’ in the Hou Hanshu 2(Charleston South Carolina 2015) . Hingley 2005: R . Hingley, Globalizing Roman culture . Unity, diversity and empire (London 2005) . Hoffmann-Salz 2011: J . Hoffmann-Salz, Die wirtschaftlichen Auswirkungen der römischen Eroberung . Vergleichende Untersuchungen der Provinzen Hispania Tarraconensis, Africa Proconsularis und Syria 218, Historia Einzelschriften (Stuttgart 2011) . Humud – Blanchard – Nikitin 2018: C . E . Humud – C . M . Blanchard – M . B . D . Nikitin, Armed conflict in Syria . Overview and U .S . response, Congressional Research Service 7–5700, 27 February 2018, 2018 . Humud et al . 2015: C . E . Humud – S . Woehrel – D . E . Mix – C . M . Blanchard, Russian deployments in Syria complicate U .S . policy, CRS Insight IN10360, 2015, 1–2 . Intagliata 2016: E . E . Intagliata, The post-Roman occupation of the northern courtyard of the sanctuary of Baalshamin in Palmyra . A reassessment of the evidence based on the documents in

Bibliogr aphy · 151 the ‘fonds d’archives Paul Collart, Université de Lausanne’, Zeitschrift für Orient-Archäologie 9, 2016, 180–200 . Intagliata 2018: E . E . Intagliata, Palmyra after Zenobia 273–750 . An archaeological and historical reappraisal (Oxford 2018) . Isaac 1992: B . H . Isaac, The limits of empire . The Roman army in the East 2(Oxford 1992) . Isaac 1998: B . H . Isaac, An open frontier, in: The Near East under Roman rule (Leiden 1998) 403– 426 . Issar – Ginat – Zohar 2011: A . Issar – H . Ginat – M . Zohar, Shifts from deserted to inhabited terrain the the arid part of the Middle East . A function of climage change, Journal of Arid Environments 86, 2011, 5–11 . Issar 2003: A . S . Issar, Climate changes during the Holocene and their impact on Hydrological systems (Cambridge 2003) . Issar – Zohar 2004: A . S . Issar – M . Zohar, Climate change . Environment and civilization in the Middle East (Berlin 2004) . Jabbūr 1995: J . S . Jabbūr, The Bedouins and the desert . Aspects of nomadic life in the Arab East (Albany 1995) . Jansen van Rensburg 2016: J . Jansen van Rensburg, The maritime traditions of the fishermen of Socotra, Yemen (Oxford 2016) . Jastrzębowska 2013: E . Jastrzębowska, Christianisation of Palmyra . Early Byzantine church in the temple of Bel, Studia Palmyrénski 12, 2013, 177–191 . Jenkins 2014: B . M . Jenkins, The dynamics of Syria’s civil war, Rand Corporation Perspective, Santa Monica 2014 . Jones 1971: A . H . M . Jones, The cities of the Eastern Roman provinces 2(Oxford 1971) . Kaizer 2002: T . Kaizer, The religious life of Palmyra . A study of the social patterns of worship in the Roman period (Stuttgart 2002) . Kaizer 2007: T . Kaizer, “Palmyre, cité grecque”? A question of coinage, Klio 89, 2007, 39–60 . Kaizer 2010: T . Kaizer, From Zenobia to Alexander the Sleepless . Paganism, Judaism and Christianity at Late Roman Palmyra, in: B . Bastl – V . Gassner – U . Muss (ed .), Zeitreisen (Wien 2010) 113–123 . Kaizer 2015: T . Kaizer, On the origins of Palmyra and its trade, Journal of Roman archaeology 28, 2015, 881–888 . Kaizer 2017: T . Kaizer, Trajectories of Hellenism at Tadmor-Palmyra and Dura-Europos, in: B . Chrubasik – D . King (ed .), Hellenism and the local communities of the Eastern Mediterranean . 400 BCE–250 CE (Oxford 2017) 29–51 . Kettenhofen 1992: E . Kettenhofen, Die Einfälle der Heruler ins römische Reich im 3 . Jh . n . Chr, Klio 74, 1992, 291–313 . Klengel 1972: H . Klengel, Zwischen Zelt und Palast . Die Begegnung von Nomaden und Sesshaften im alten Vorderasien (Wien 1972) . Kosmin 2014: P . J . Kosmin, The land of the elephant kings . Space, territory, and ideology in the Seleucid Empire (Cambridge, Mass . 2014) . Kowalski 1997: S . P . Kowalski, Late Roman Palmyra in literature and epigraphy, Studia Palmyrénski 10, 1997, 39–62 .

152 · Bibliogr aphy Kraemer 1958: C . J . Kraemer, Excavations at Nessana (Auja Hafir, Palestine) . 3 . Non-literary papyri (Princeton, NJ 1958) . Krzywinski – Krzywinski 2016: K . Krzywinski – J . Krzywinski, Agriculture in Byzantine Palmyrena, in: J . C . Meyer – E . H . Seland – N . Anfinset (ed .), Palmyrena . City, hinterland and caravan trade between Orient and Occident (Oxford 2016) 171–183 . Lancaster – Lancaster 1999: W . Lancaster – F . Lancaster, People, land and water in the Arab Middle East . Environments and landscapes in the Bilâd ash-Shâm 2, Studies in environmental anthropology (Amsterdam 1999) . Laubenheimer – Römer-Strehl 2013: F . Laubenheimer – C . Römer-Strehl, Hellenistische Amphorenstempel, in: A . Schmidt-Colinet – W . al-Asʿad (ed .), Palmyras Reichtum durch weltweiten Handel . Archäologische Untersuchungen im Bereich der Hellenistischen Stadt (Wien 2013) 106–108 . Lerner 1999: J . D . Lerner, The impact of Seleucid decline on the Eastern Iranian Plateau . The foundations of Arsacid Parthia and Graeco-Bactria (Stuttgart 1999) . Lerouge-Cohen 2005: C . Lerouge-Cohen, Les guerres parthiques de Démétrios II et Antiochos VII dans les sources gréco-romaines, de Posidonios à Trogue/Justin, Journal des Savants 2005, 217–252 . Lewis 2000: N . Lewis, The Syrian Steppe during the last century of Ottoman rule . Hawran and the Palmyrena, in: M . Mundy – B . Musallam (ed .), The transformation of nomadic society in the Arab East (Cambridge 2000) 33–43 . Liebeschuetz 2015: J . H . W . G . Liebeschuetz, East and West in Late Antiquity . Invasion, settlement, ethnogenesis and conflicts of religion 20, Impact of Empire (Leiden 2015) . Lightfoot 1996: D . R . Lightfoot, Syrian Qanat Romani . History, ecology, abandonment, Journal of Arid Environments 33, 1996, 321–336 . Linck 2016: R . Linck, Geophysical prospection by ground- and space-based methods of the ancient town of Palmyra (Syria), in: J . C . Meyer – E . H . Seland – N . Anfinset (ed .), Palmyrena . City, hinterland and caravan trade between Orient and Occident (Oxford 2016) 77–85 . Magness 2003: J . Magness, The archaeology of the early Islamic settlement in Palestine (Winona Lake, Ind . 2003) . Majcherek 2013: G . Majcherek, Excavating the basilicas, Studia Palmyrénski 12, 2013, 251–268 . Mann 2012: M . Mann, The sources of social power . Volume 1 . A history of power from the beginning to AD 1760 2(Cambridge 2012) . Marcus 1989: A . Marcus, The Middle East on the eve of modernity . Aleppo in the eighteenth century (New York 1989) . Mattern-Parkes 2003: S . P . Mattern-Parkes, The defeat of Crassus and the just war, The Classical World 96, 2003, 387–396 . Matthews 1984: J . F . Matthews, The tax law of Palmyra . Evidence for economic history in a city of the Roman east, Journal of Roman Studies 74, 1984, 157–180 . Maxfield 1996: V . A . Maxfield, The eastern desert forts and the army in Egypt during the principate, in: D . M . Bailey (ed .), Archaeological research in Roman Egypt . The proceedings of the Seventeenth Classical Colloquium of the Department of Greek and Roman Antiquities, British Museum, held on 1–4 December, 1993 (Ann Arbor 1996) 9–19 .

Bibliogr aphy · 153 Mayerson 1961: P . Mayerson, The ancient agricultural regime of Nessana and the central Negeb (London 1961) . McLaughlin 2010: R . McLaughlin, Rome and the distant East . Trade routes to the ancient lands of Arabia, India and China (London 2010) . Merkelbach – Stauber 2005: R . Merkelbach – J . Stauber, Jenseits des Euphrat . Griechische Inschriften . Ein epigraphisches Lesebuch (München 2005) . Meyer 2016: J . C . Meyer, Palmyrena . Settlements, forts and nomadic networks, in: A . Kropp – R . Raja (ed .), The world of Palmyra (København 2016) 86–102 . Meyer 2017: J . C . Meyer, Palmyrena . Palmyra and the surrounding territory from the Roman to the early Islamic period (Oxford 2017) . Meyer – Heldaas Seland 2016: J . C . Meyer – E . Heldaas Seland, Palmyra and the trade route to the Euphrates, ARAM 28, 2016, 497–523 . Meyer – Seland – Anfinset 2016: J . C . Meyer – E . H . Seland – N . Anfinset (ed .), Palmyrena . City, hinterland and caravan tade between Orient and Occident . Proceedings of the conference held in Athens, December 1–3, 2012 (Oxford 2016) . Millar 1987: F . Millar, The problem of Hellenistic Syria, Hellenism in the East 2, 1987, 110–133 . Millar 1990: F . Millar, The Roman coloniae of the Near East . A study of cultural relations, in: H . Solin – M . Kajava (ed .), Roman eastern policy and other studies in Roman history (Helsinki 1990) 7–58 . Millar 1993: F . Millar, The Roman Near East . 31 BC–AD 337 (Cambridge, Mass . 1993) . Millar 1998: F . Millar, Caravan cities . The Roman Near East and long-distance trade by land, in: M . M .  Austin (ed .), Modus operandi . Essays in honour of Geoffrey Rickman (London 1998) 119–137 . Mior 2016: P . Mior, The road from Palmyra to Damascus in the Tabula Peutingeriana, in: J . C . Meyer – E . H . Seland – N . Anfinset (ed .), Palmyrena . City, hinterland and caravan trade between Orient and Occident (Oxford 2016) 49–57 . Mittag 2006: P . F . Mittag, Antiochos IV . Epiphanes . Eine politische Biographie (Berlin 2006) . Mommsen 1902–04: T . Mommsen, Römische Geschichte 9/5(Berlin 1902–04) . Münkler 2007: H . Münkler, Empires . The logic of world domination from ancient Rome to the United States (Malden, Mass . 2007) . Murphy 2004: T . Murphy, Pliny the Elder’s Natural History . The empire in the encyclopedia (Oxford 2004) . Musil 1928a: A . Musil, The manners and customs of the Rwala bedouins 6, American Geographical Society . Oriental explorations and studies (New York 1928) . Musil 1928b: A . Musil, Palmyrena . A topographical itinerary 5, American Geographical Society . Oriental explorations and studies (New York 1928) . Myers 2010: E . A . Myers, The Ituraeans and the Roman Near East . Reassessing the sources 147, Society for New Testament studies . Monograph series (Cambridge 2010) . Nippel 2000: W . Nippel, Webers ‘Stadt’ . Entstehung – Struktur der Argumentation – Rezeption, in: H . Bruhns – W . Nippel (ed .), Max Weber und die Stadt im Kulturvergleich (Göttingen 2000) 11–38 . Parker 2008: G . Parker, The making of Roman India (Cambridge 2008) .

154 · Bibliogr aphy Paschoud 2000–2003: F . Paschoud, Zosime . Histoire nouvelle (Paris 2000–2003) . Petersen 2003: L . H . Petersen, The baker, his tomb, his wife, and her breadbasket, The Art Bulletin 85, 2003, 230–257 . Pillet 1941: M . Pillet, Les autels de l’ “el-Karassi” (Syrie centrale), Revue Archéologique 17, 1941, 5–17 . Pirngruber 2017: R . Pirngruber, The economy of Late Achaemenid and Seleucid Babylonia (Cambridge 2017) . Pitts – Versluys 2015: M . Pitts – M . J . Versluys, Globalisation and the Roman world . Perspectives and opportunities, in: M . Pitts – M . J . Versluys (ed .), Globalisation and the Roman world . World history, connectivity and material culture (Cambridge 2015) 3–31 . Plattner 2013: G . A . Plattner, Sondage I . Eine hellenistisch-römische Straßenkreuzung und angrenzende Wohnbebauung . Baubefund, Architektur, Chronologie, in: A . Schmidt-Colinet – W .  alAsʿad (ed .), Palmyras Reichtum durch weltweiten Handel . Archäologische Untersuchungen im Bereich der Hellenistischen Stadt (Wien 2013) 89–117 . Poignault – Hunzinger – Kasprzyk 2001: R . Poignault – C . Hunzinger – D . Kasprzyk, Les usurpateurs du Quadrige des Tyrans dans l’Histoire Auguste . Des personnages de roman?, Collection de la Maison de l’Orient méditerranéen ancien . Série littéraire et philosophique 29, 2001, 251–268 . Polanyi – Arensberg – Pearson 1957: K . Polanyi – C . M . Arensberg – H . W . Pearson (ed .), Trade and market in the early empires . economies in history and theory (Glencoe, Ill . 1957) . Potter 1990: D . S . Potter, Prophecy and history in the crisis of the Roman empire . A historical commentary on the thirteenth Sibylline Oracle (Oxford 1990) . Potts 1988: D . T . Potts, Arabia and the kingdom of Characene, in: D . T . Potts (ed .), Araby the blest . Studies in Arabian archaeology (København 1988) 137–167 . Potts 1997: D . T . Potts, Mesopotamian civilization . The material foundations (London 1997) . Raja 2015: R . Raja, From studying portraits to documenting Syria’s cultural heritage, in: R . Raja – A . H . Sørensen (ed .), Harald Inghold & Palmyra (Aarhus 2015) 10–22 . Raja 2017: R . Raja, To be or not to be depicted as a priest in Palmyra . A matter of representational spheres and societal values, in: T . Long – A . Høyen Sørensen (ed .), Positions and professions in Palmyra (København 2017) 115–130 . Rathbone 2000: D . Rathbone, The ‘Muziris’ papyrus (SB XVIII 13167) . Financing Roman trade with India, Bulletin de la Société archéologique d’Alexandrie 46, 2000, 39–50 . Rathmann 2011/12: M . Rathmann, Neue Perspektiven zur Tabula Peutingeriana, Geographia antiqua . Rivista di geografia storica del mondo antico e di storia della geografia 20/21, 2011/12, 83–102 . Rathmann 2016: M . Rathmann, The Tabula Peutingeriana and antique cartography, in: S . Bianchetti – M . Cataudella – H .-J . Gehrke (ed .), Brill’s companion to ancient geography . The inhabited world in Greek and Roman tradition (Leiden 2016) 337–362 . Rathmann 2017: M . Rathmann, Tabula Peutingeriana . Die einzige Weltkarte aus der Antike 2 (Darmstadt 2017) . Ray 1994: H . P . Ray, The winds of change . Buddhism and the maritime links of early South Asia (Delhi 1994) .

Bibliogr aphy · 155 Reddé – Brun 2006: M . Reddé – J .-P . Brun, L’architecture des praesidia et la genése des dépotoirs, in: H . Cuvigny (ed .), La route de Myos Hormos . L’armée romaine dans le désert oriental d’Égypte (Le Caire 2006) 73–185 . Reilly 2005: J . A . Reilly, Town and steppe in Ottoman Syria, Der Islam 92, 2005, 148–160 . Ridley 1982: R . T . Ridley, Zosimus . New history . A translation with commentary 2, Byzantina Australiensia (Canberra 1982) . Robin 2012a: C . J . Robin, Appendix IV . Les Palmyréniens en Arabie du Sud, in: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the Cave Hoq (Bremen 2012) 488–492 . Robin 2012b: C . J . Robin, Le Palmyréniens en Arabie du Sud, in: I . Strauch (ed .), Foreign Sailors on Socotra: The inscriptions and drawings from the cave at Hoq (Bremen 2012) 488–492 . Robin 2012c: C . J . Robin, Sudarabiques et Aksūmites à Suquṭra . D’après les inscriptions de la grotte de Ḥōq, in: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the Cave Hoq (Bremen 2012) 438–442 . Römer-Strehl 2013a: C . Römer-Strehl, Keramik, in: A . Schmidt-Colinet – W . al-Asʿad (ed .), Palmyras Reichtum durch weltweiten Handel . Archäologische Untersuchungen im Bereich der Hellenistischen Stadt (Wien 2013) 7–80 . Römer-Strehl 2013b: C . Römer-Strehl, Lampen, in: A . Schmidt-Colinet – W . al-Asʿad (ed .), Palmyras Reichtum durch weltweiten Handel . Archäologische Untersuchungen im Bereich der Hellenistischen Stadt (Wien 2013) 109–126 . Römer-Strehl 2016: C . Römer-Strehl, Seleucid, Roman, and Mesopotamian influence in Palmyrene pottery and the question of a Palmyrene identity, in: J . C . Meyer – E . H . Seland – N . Anfinset (ed .), Palmyrena . City, hinterland and caravan trade between Orient and Occident (Oxford 2016) 107–114 . Rosen 2007: A . M . Rosen, Civilizing climate . Social responses to climate change in the ancient near East (Lanham 2007) . Rosen 2000: S . A . Rosen, The decline of desert agriculture . A view from classical period Negev, in: G . Barker – D . D . Gilbertson (ed .), Archaeology of drylands . Living on margins (New York 2000) 45–62 . Rostovtzeff 1932a: M . Rostovtzeff, Caravan cities (Oxford 1932) . Rostovtzeff 1932b: M . Rostovtzeff, Les inscriptions caravanières de Palmyre, in: G . Glotz (ed .), Mélanges Gustave Glotz (Paris 1932) 793–811 . Rougé 1986: J . Rougé, La navigation intérieure dans le Proche Orient antique, in: P . Louis (ed .), L’homme et l’eau en Méditerranée et au Proche Orient . 3 . L’eau dans les techniques . Séminaire de recherche 1981–1982 (Lyon 1986) 39–49 . Ruffing 2016: K . Ruffing, Economic life in Roman Dura-Europos, in: T . Kaizer (ed .), Religion, society and culture at Dura-Europos (Cambridge 2016) 190–198 . Rutten 2007: K . Rutten, The Roman fine wares of ed-Dur (Umm al-Qaiwain, U .A .E .) and their distribution in the Persian Gulf and the Indian Ocean, Arabian Archaeology and Epigraphy 18, 2007, 8–24 . Rømer 2004: J . Rømer, Foldudbyttet i Danmark fra 1600–tallet til 1837, Landbohistorisk Tisskrift 2004, 41–59 .

156 · Bibliogr aphy Sahin 2014: K . Sahin, Die russische Option . Kann die Türkei eine Allianz mit Russland schmieden?, Arbeitspapier Sicherheitspolitik 17, 2014 . Sallares 1991: R . Sallares, The ecology of the ancient Greek world (London 1991) . Sallmann 1971: K . G . Sallmann, Die Geographie des älteren Plinius in ihrem Verhältnis zu Varro . Versuch einer Quellenanalyse 11, Untersuchungen zur antiken Literatur und Geschichte, (Berlin 1971) . Salzman 1980: P . C . Salzman, Introduction, in: P . C . Salzman (ed .), When nomads settle . Processes of sedentarization as adaption and response (New York 1980) 1–22 . Sartre – Sartre 2016: A . Sartre – M . Sartre, Palmyre . Vérités et légendes (Paris 2016) . Sartre 1996: M . Sartre, Palmyre, cité grecque, Annales Archéologiques Arabes Syriennes 43, 1996, 385–405 . Sartre 2005: M . Sartre, The Middle East under Rome (Cambridge, Mass . 2005) . Sartre 2014: M . Sartre, L’historien et ses territoires . Choix d’articles 70, Scripta antiqua (Bordeaux 2014) . Sartre-Fauriat – Sartre 2008: A . Sartre-Fauriat – M . Sartre, Palmyre . La cité des caravanes (Paris 2008) . Scharrer 2002: U . Scharrer, Nomaden und Seßhafte in Tadmor im 2 . Jahrtausend v . Chr ., in: M . Schuol – U . Hartmann – A . Luther (ed .), Grenzüberschreitungen . Formen des Kontakts zwischen Orient und Okzident (Stuttgart 2002) 279–330 . Schiettecatte – Arbach 2016: J . Schiettecatte – M . Arbach, The political map of Arabia and the Middle East in the third century AD revealed by a Sabaean inscription, Arabian Archaeology and Epigraphy 27, 2016, 176–196 . Schlumberger 1939: D . Schlumberger, Bornes frontières de la Palmyrène, Syria 20, 1939, 43–73 . Schlumberger 1951: D . Schlumberger, La Palmyrène du Nord-Ouest . Villages et lieux de culte de l’époque impériale . Recherches archéologiques sur la mise en valeur d’une région du désert par les Palmyréniens (Paris 1951) . Schlumberger 1961: D . Schlumberger, Palmyre et la Mésène, Syria 38, 1961, 256–260 . Schmidt Heidenreich 2016: C . Schmidt Heidenreich, Les unités palmyréniennes de l’armée romaine . Une approche historique, in: C . Wolff – P . Faure (ed .), Les auxiliaires de l’armée romaine (Lyon 2016) 223–235 . Schmidt-Colinet 1997: A . Schmidt-Colinet, “Aspects of romanization” . The tomb architecture at Palmyra and its decoration, in: S . Alcock (ed .), The early Roman Empire in the East (Oxford 1997) 157–177 . Schmidt-Colinet 2003a: A . Schmidt-Colinet, Nouvelles données sur Palmyre hellénistique, in: M . Sartre (ed .), La Syrie hellénistique (Paris 2003) 299–302 . Schmidt-Colinet 2003b: A . Schmidt-Colinet, Untersuchungen zur Urbanistik des hellenistischen Palmyra, in: K . S . Freyberger – A . Henning – H . v . Hesberg (ed .), Kulturkonflikte im Vorderen Orient an der Wende vom Hellenismus zur römischen Kaiserzeit (Rahden/Westfalen 2003) 19–22 . Schmidt-Colinet – al-Asʿad 2000: A . Schmidt-Colinet – Ḫ . al-Asʿad, Zur Urbanistik des hellenistischen Palmyra . Ein Vorbericht, Damaszener Mitteilungen 12, 2000, 61–93 . Schmidt-Colinet – al-Asʿad 2002: A . Schmidt-Colinet – Ḫ . al-Asʿad, Archaeological news from hellenistic Palmyra, Parthica 4, 2002, 157–166 .

Bibliogr aphy · 157 Schmidt-Colinet – al-Asʿad 2013: A . Schmidt-Colinet – W . al-Asʿad (ed .), Palmyras Reichtum durch weltweiten Handel . Archäologische Untersuchungen im Bereich der hellenistischen Stadt (Wien 2013) . Schmidt-Colinet et al . 1992: A . Schmidt-Colinet – U . Heimberg – P . Caselitz – C . Müting-Zimmer – Ḫ . al-Asʿad, Das Tempelgrab Nr . 36 in Palmyra . Studien zur palmyrenischen Grabarchitektur und ihrer Ausstattung 2(Mainz 1992) . Schörle 2017: K . Schörle, Palmyrene merchant networks and economic integration in competetive markets, in: H . F . Teigen – E . H . Seland (ed .), Sinews of empire . Networks in the Roman Near East and beyond (Oxford 2017) 147–154 . Schuol 2000: M . Schuol, Die Charakene . Ein mesopotamisches Königreich in hellenistisch-parthischer Zeit (Stuttgart 2000) . Schwartz 1960: J . Schwartz, L’Empire romain, l’Égypte et le commerce oriental, Annales . Économies, societés, civilisations 15, 1960, 18–44 . Sedov 1992: A . V . Sedov, New archaeological and epigraphic material from Qana (South Arabia), Arabian Archaeology and Epigraphy 3 .2, 1992, 110–137 . Sedov 1996: A . V . Sedov, Qanaʾ (Yemen) and the Indian Ocean . The archaeological evidence, in: H . P . Ray – J .-F . Salles (ed .), Tradition and archaeology . Early maritime contacts in the Indian Ocean (New Delhi 1996) 71–111 . Sedov 2007: A . V . Sedov, The ‚Port of Qanaʾ’ and the incense trade, in: D . P . S . Peacock – D . Williams (ed .), Food for the gods (Oxford 2007) 71–111 . Sedov 2010a: A . V . Sedov, Description du site, in: J .-F . Salles – A . V . Sedov (ed .), Qāniʾ . Le port antique du Ḥaḍramawt entre la Méditerranée, l’Afrique et l’Inde . Fouilles russes 1972, 1985–89, 1991, 1993–94 (Turnhout 2010) 5–10 . Sedov 2010b: A . V . Sedov, Stratigraphy and development of the site . Preliminary remarks, in:  J .-F . Salles – A . V . Sedov (ed .), Qāniʾ . Le port antique du Ḥaḍramawt entre la Méditerranée,  l’Afrique  et l’Inde . Fouilles russes 1972, 1985–89, 1991, 1993–94 (Turnhout 2010) 371–380 . Seland 2011: E . H . Seland, The Persian Gulf or the Red Sea? Two axes in ancient Indian Ocean trade, where to go and why, World Archaeology 43, 2011, 398–409 . Seland 2012: E . H . Seland, Trade and Christianity in the Indian Ocean during Late Antiquity, Journal of Late Antiquity 5, 2012, 72–86 . Seland 2013: E . H . Seland, Ancient Afghanistan and the Indian Ocean . Maritime links of the Kush Empire, Journal of Indian Ocean Archaeology 9, 2013, 66–74 . Seland 2014: E . H . Seland, The organisation of the Palmyrene caravan trade, Ancient West and East 13, 2014, 197–211 . Seland 2015: E . H . Seland, Camels, camel nomadism and the practicalities of Palmyrene caravan trade, ARAM 27, 2015, 45–54 . Seland 2016: E . H . Seland, Ships of the desert and ships of the sea . Palmyra in the world trade of the first three centuries CE, Philippika 101 (Wiesbaden 2016) . Seland 2017: E . H . Seland, The iconography of caravan trade in Palmyra and the Roman Near East, in: T . Long – A . Høyen Sørensen (ed .), Positions and professions in Palmyra (København 2017) 106–115 .

158 · Bibliogr aphy Seyrig 1932: H . Seyrig, L’incorporation de Palmyre à l’empire romain (antiquités syriennes 9), Syria 13, 1932, 266–277 . Seyrig 1939: H . Seyrig, Inscriptions (antiquités syriennes 30), Syria 20, 1939, 302–323 . Seyrig 1941: H . Seyrig, Le statut de Palmyre (antiquités syriennes 36), Syria 22, 1941, 155–175 . Seyrig 1944/45: H . Seyrig, Héraclès-Nergal (antiquités syriennes 39), Syria 24, 1944/45, 62–80 . Seyrig 1950: H . Seyrig, Palmyra and the East, JRS 40, 1950, 1–7 . Seyrig 1963: H . Seyrig, Les fils du roi Odainat, Annales archéologiques de Syrie 13, 1963, 159–172 . Shelat 2012: B . Shelat, Appendix I . The Gujarati stone inscriptions from Rās Ḥowlef (Socotra), in: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the Cave Hoq (Bremen 2012) 407–432 . Sherwin-White – Kuhrt 1993: S . M . Sherwin-White – A . Kuhrt, From Samarkhand to Sardis . A new approach to the Seleucid empire (London 1993) . Shipley 2000: G . Shipley, The Greek world after Alexander . 323–30 BC (London 2000) . Sidebotham 1986: S . E . Sidebotham, Roman economic policy in the Erythra Thalassa . 30 B .C .– A .D . 217 91, Mnemosyne . Supplementum (Leiden 1986) . Sidebotham 2007: S . E . Sidebotham, Excavations, in: S . E . Sidebotham – W . Wendrich (ed .), Berenike 1999/2000 . Report on the excavations at Berenike, including excavations in Wadi Kalalat and Siket, and the survey of the Mons Smaragdus Region (Los Angeles, Calif . 2007) 30–165 . Sidebotham 2011: S . E . Sidebotham, Berenike and the ancient maritime spice route 18, California world history library (Berkeley 2011) . Sidebotham 2014: S . E . Sidebotham, Religion and burial at the Ptolemaic-Roman Red Sea emporium of Berenike, African Archaeological Review 31, 2014, 599–635 . Sidebotham 2017: S . E . Sidebotham, Zwischen Wüste und Meer . Der Hafen von Berenike in Ptolemäisch-römischer Zeit, Antike Welt 48 .2, 2017, 60–71 . Sidebotham – Hense – Nouwens 2008: S . E . Sidebotham – M . Hense – H . M . Nouwens, The red land . The illustrated archaeology of Egypt’s Eastern desert (Cairo 2008) . Sidebotham – Zitterkopf – Riley 1991: S . E . Sidebotham – R . E . Zitterkopf – J . A . Riley, Survey of the ʿAbu Shaʾar-Nile road, American Journal of Archaeology 95, 1991, 571–622 . Sidebotham – Zych 2012: S . E . Sidebotham – I . Zych, Berenike . Archaeological fieldwork at a Ptolemaic-Roman port on the Red Sea coast of Egypt 2011–2012, Sahara 23, 2012, 29–48 . Smith II 2013: A . M . Smith II, Roman Palmyra . Identity, community, and state formation (New York 2013) . Sommer 2005a: M . Sommer, Palmyra and Hatra . “Civic” and “tribal” institutions at the Near Eastern steppe frontier, in: E . S . Gruen (ed .), Cultural borrowings and ethnic appropriations in antiquity (Stuttgart 2005) 285–296 . Sommer 2005b: M . Sommer, Roms orientalische Steppengrenze . Palmyra – Edessa – Dura-Europos – Hatra . Eine Kulturgeschichte von Pompeius bis Diocletian, Oriens et Occidens 9 (Stuttgart 2005) . Sommer 2008: M . Sommer, Der Löwe von Tadmor . Palmyra und der unwahrscheinliche Aufstieg des Septimius Odaenathus, Historische Zeitschrift 287, 2008, 281–318 .

Bibliogr aphy · 159 Sommer 2009: M . Sommer, Imperiale Macht und lokale Identität . Universalhistorische Variationen zu einem regionalhistorischen Thema, in: M . Blömer – M . Facella – E . Winter (ed .), Lokale Identität im Römischen Nahen Osten (Stuttgart 2009) 235–248 . Sommer 2011: M . Sommer, Colonies – colonisation – colonialism . A typological reappraisal, Ancient West & East 10, 2011, 183–193 . Sommer 2012: M . Sommer, Empire, frontier and “Third Spaces” . The Near East under Roman rule, in: M . Cassia – C . Giuffrida – C . Molè – A . Pinzone (ed .), Pignora amicitiae . Scritti di storia antica e di storiografia offerti a Mario Mazza (Catania 2012) 123–138 . Sommer 2015: M . Sommer, Les notables de Palmyre . Local elites in the Syrian Desert in the 2nd  and  3rd centuries AD, in: P . Briks (ed .), Elites in the ancient world (Szczecin 2015) 173–182 . Sommer 2016: M . Sommer, The Venice of the Sands . Palmyrene trade revisited, in: J . C . Meyer – E . H . Seland – N . Anfinset (ed .), Palmyrena . City, hinterland and caravan trade between Orient and Occident (Oxford 2016) 11–17 . Sommer 2017a: M . Sommer, Palmyra . Biographie einer verlorenen Stadt (Darmstadt 2017) . Sommer 2017b: M . Sommer, Texture of empire . Personal networks and the modus operandi of Roman hegemony, in: H . F . Teigen – E . H . Seland (ed .), Sinews of empire . Networks in the Roman Near East and beyond (Oxford 2017) 85–93 . Sommer 2018a: M . Sommer, Palmyra . A history (London 2018) . Sommer 2018b: M . Sommer, Roms orientalische Steppengrenze . Palmyra – Edessa – Dura-Europos – Hatra . Eine Kulturgeschichte von Pompeius bis Diocletian 2(Stuttgart 2018) . Southern 2008: P . Southern, Empress Zenobia . Palmyra’s rebel queen (London 2008) . Speidel 1984: M . P . Speidel, Palmyrenian irregulars at Koptos, The Bulletin of the American Society of Papyrologists 21, 1984, 221–224 . Starcky – Gawlikowski 1985: J . Starcky – M . Gawlikowski, Palmyre 2(Paris 1985) . Stephan 2002: E . Stephan, Honoratioren, Griechen, Polisbürger . Kollektive Identitäten innerhalb der Oberschicht des kaiserzeitlichen Kleinasien (Göttingen 2002) . Stoneman 1992: R . Stoneman, Palmyra and its Empire . Zenobia’s revolt against Rome (Ann Arbor, Mi . 1992) . Straub 1952: J . Straub, Studien zur Historia Augusta, Dissertationes Bernenses . Ser . 1 4 (Bern 1952) . Strauch 2012a: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the cave Hoq (Bremen 2012) . Strauch 2012b: I . Strauch, Socotra and the “Indian Connection” . Facts and Fictions, in: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the Cave Hoq (Bremen 2012) 366–403 . Strauch 2016: I . Strauch, Indian inscriptions from Cave Hoq at Socotra, in: M .-F . Boussac – J .-F . Salles – J .-B . Yon (ed .), Ports of the ancient Indian Ocean (Delhi 2016) 79–97 . Strauch et al . 2012: I . Strauch – C . J . Robin – M . D . Bukharin – N . Sims-Williams, Catalogue of inscriptions and drawings in the Cave Hoq, in: I . Strauch (ed .), Foreign sailors on Socotra . The inscriptions and drawings from the Cave Hoq (Bremen 2012) 25–218 . Talbert 2010: R . J . A . Talbert, Rome’s world . The Peutinger map reconsidered (Cambridge 2010) .

160 · Bibliogr aphy Taylor 2003: D . Taylor, Bilingualism and diglossia in late antique Syria and Mesopotamia, in: J . N . Adams – M . Janse – S . Swain (ed .), Bilingualism in ancient society . Language, contact and the written text (Oxford 2003) 298–331 . Teixidor 1979: J . Teixidor, The pantheon of Palmyra 79, Etudes préliminaires aux religions orientales dans l’Empire romain (Leiden 1979) . Teixidor 1984: J . Teixidor, Un port romain du désert, Palmyre, et son commerce d’Auguste à Caracalla 34, Semitica (Paris 1984) . Terpstra 2016: T . T . Terpstra, The Palmyrene temple in Rome and Palmyra’s trade with the west, in: J . C . Meyer – E . H . Seland – N . Anfinset (ed .), Palmyrena . City, hinterland and caravan trade between Orient and Occident (Oxford 2016) 39–48 . Timpe 1962: D . Timpe, Die Bedeutung der Schlacht von Carrhae, Museum Helveticum 19, 1962, 104–129 . Timpe 1963: D . Timpe, Geschichte der politischen Beziehungen zwischen Römer- und Partherreich . Habilitationsschrift (Freiburg 1963) . Tomber 2011: R . Tomber, Pots with writing, in: D . P . S . Peacock (ed .), Myos Hormos – Quseir alQadim . Roman and Islamic ports on the Red Sea . 2 . Finds from the excavations 1999–2003 (Oxford 2011) 5–10 . Tucker 2009: D . J . Tucker, Tracking mobility in the Syrian Desert . Potential of simple features for mapping landscapes of mobile pastoralists, Computer Applications to Archaeology . Online Proceedings 2009 . van der Spek 2007: R . J . van der Spek, The Hellenistic Near East, in: W . Scheidel – I . Morris – R .  Saller (ed .), The Cambridge Economic History of The Greco-Roman World (Cambridge 2007) 409–433 . Veh 1990: O . Veh, Zosimos . Neue Geschichte 31, Bibliothek der griechischen Literatur . Abteilung Klassische Philologie (Stuttgart 1990) . Verhoogt 1998: A . M . F . W . Verhoogt, Greek and Latin textual material, in: S . E . Sidebotham – W . Z . Wendrich (ed .), Report of the 1996 excavations at Berenike (Egyptian Red Sea coast) and the survey of the Eastern Desert (Leiden 1998) 193–198 . Veyne 2005: P . Veyne, L’ Empire gréco-romain (Paris 2005) . Veyne 2017: P . Veyne, Palmyra . An irreplaceable treasure (Chicago 2017) . Weber 1968: M . Weber, Economy and society . An outline of interpretive sociology (New York 1968) . Weber 2000: M . Weber, Wirtschaft und Gesellschaft . Die Stadt 22,5, Studienausgabe der Max Weber-Gesamtausgabe . Abt . 1, Schriften und Reden (Tübingen 2000) . Wilkinson 2003: T . J . Wilkinson, Archaeological landscapes of the Near East (Tucson 2003) . Will 1957: E . Will, Marchands et chefs de caravanes à Palmyre, Syria 34, 1957, 262–277 . Will 1985: E . Will, Pline l’ancien et Palmyre . Un problème d’histoire ou d’histoire littéraire?, Syria 62, 1985, 263–269 . Will 1992: E . Will, Les Palmyréniens . La Venise des sables (Ier siècle avant – IIIème après J .-C .) (Paris 1992) . Wirth 1971: E . Wirth, Syrien . Eine geographische Landeskunde, Wissenschaftliche Länderkunden 4–5 (Darmstadt 1971) . Yon 1998: J .-B . Yon, Remarques sur une famille caravanière à Palmyre, Syria 75, 1998, 153–160 .

Bibliogr aphy · 161 Yon 2002: J .-B . Yon, Les notables de Palmyre (Beyrouth 2002) . Yon 2003: J .-B . Yon, Les villes de Haute-Mésopotamie et de l’Euphrate à la fin de l’époque hellénistique, in: M . Sartre (ed .), La Syrie hellénistique (Lyon 2003) 193–210 . Yon 2008: J .-B . Yon, Bilinguisme et trilinguisme à Palmyre, in: F . Biville – J .-C . Decourt – G . Rougemont (ed .), Bilinguisme gréco-latin et épigraphie (Lyon 2008) 195–211 . Yon 2010: J .-B . Yon, Kings and princes at Palmyra, in: T . Kaizer – M . Facella (ed .), Kingdoms and principalities in the Roman Near East (Stuttgart 2010) 229–240 . Yon 2013: J .-B . Yon, L’épigraphie palmyrénienne depuis PAT, Studia Palmyrénski 12, 2013, 333–379 . Young 2001: G . K . Young, Rome’s eastern trade . International commerce and imperial policy . 31 BC – AD 305 (London 2001) . Zecchini 2005: G . Zecchini, Il bipolarismo romano-iranico, in: C . Bearzot – F . Landucci – G . Zecchini (ed .), L’equilibrio internazionale dagli antichi ai moderni (Milano 2005) 59–82 . Ziegler 1964: K .-H . Ziegler, Die Beziehungen zwischen Rom und dem Partherreich . Ein Beitrag zur Geschichte des Völkerrechts (Wiesbaden 1964) . Żuchowska 2013: M . Żuchowska, Palmyra and the far eastern trade, Studia Palmyreńskie 12, 2013, 381–387 .

List of illustrations III .1 M . Sommer; IV .1 J . C . Meyer based on Google Earth; IV .2 J . C . Meyer; IV .3 Pillet 1941 fig . 4 and 5; IV .4–8 J . C . Meyer; IV .9 N . Anfinset; IV .10 J . C . Meyer; IV .11 J . C . Meyer based on Google Earth; IV .12 J . C . Meyer; VII .1–2 A . Schmidt-Colinet .

Index locorum 1. Literary sources Agath . 5,105a + b 72, 74 App . BC 5,9 26, 44, 85

Plin . HN 6,20 91 Plin . HN 6,101–102 81

Cic . Verr . 2,3,112–13 56 Colum . 3,3,4 57

SHA Aurel . 32 70 SHA Firmus 93 SHA Firmus 2–6 70 SHA Gall . 10,1 33 SHA trig . tyr . 15,2 33 Strab . 16,1,27 27, 100 Strab . 16,3,2 100 Strab . 17,1,45 81

Hou Hanshou 88,12 88,12 Not . Dign . [or .] 7,34 36 Periplus m . r . 27 72–73 Periplus m . r . 30–31 72–73 Periplus m . r . 31–32 75 Periplus m . r . 36 75 Periplus m . r . 40–49 75 Petr . Patr . fr . 13–14 (FGH IV 188– 189) 87 Phot . Cod . 250,103,459b 72, 74 Plin . HN 1,III auct . 9 Plin . HN 5,83 9 Plin . HN 5,88 9, 24, 44, 53

Ulp . De cens . 1 (Dig . 50,15,1,4–5) 31 Varro rust . 1,44,1 Zos . 1,5 Zos . 1,12 Zos . 1,13 Zos . 1,18

57

110 120 116 112

Zos . 1,20 114 Zos . 1,23 113–14 Zos . 1,26 111, 113 Zos . 1,27 113–14 Zos . 1,28 113 Zos . 1,29 111 Zos . 1,30 115 Zos . 1,32 112, 117 Zos . 1,36 111, 113 Zos . 1,37 114–15 Zos . 1,38 116, 122 Zos . 1,39 118–19, 124 Zos . 1,39–61 118 Zos . 1,51 121, 124–25 Zos . 1,54 121, 124–25 Zos . 1,54–56 123 Zos . 1,55 121 Zos . 1,57 109 Zos . 1,59 122 Zos . 1,62 126

2. Inscriptions and ostraca AE 1984, 925

30

Dentzer-Feydy – Teixidor 1993, 162, no . 166 32 Gawlikowski 2010b

10

Healey 2009, 144, no . 28

26

I . Did . 5 79 I . Did . 39 79 I . Did . 71 79 I . Did . 285 79 I . Did . 286 79 I . Portes 103 66, 75, 86, 91 IGLS VII .1,1 25 IGLS XVII .1,2 31 IGLS XVII .1,17 27 IGLS XVII .1,18 27 IGLS XVII .1,53 32 IGLS XVII .1,61 33 IGLS XVII .1,75 31 IGLS XVII .1,77 35 IGLS XVII .1,80–81 35

IGLS XVII .1,100 IGLS XVII .1,120 IGLS XVII .1,127 IGLS XVII .1,145 IGLS XVII .1,154 IGLS XVII .1,221 IGLS XVII .1,245 IGLS XVII .1,245 IGLS XVII .1,551 Inv . III .29 77 Inv . IX .15 66 Inv . X .87 65 Inv . X .88 65 O . Ber . 97 69 O . Ber . 137 69 O . Ber . 140a 69 O . Ber . 183 69 O . Ber . 198 69 O . Ber . 254 69 O . Ber . 255 69 O . Ber . 256 69 O . Ber . 257 69

36 34 96, 99, 106 29 36 31 32 66 32

PAT 0055 35 PAT 0197 16, 96, 98–99, 106–07 PAT 0256 67, 86 PAT 0259 66 PAT 0259,59–60 54 PAT 0259,73 10 PAT 0259,103 10 PAT 0259,121 10 PAT 0261 102 PAT 0262 96 PAT 0263 14 PAT 0269 16, 27 PAT 0270 97 PAT 0274 96, 98 PAT 0278 32 PAT 0280 31 PAT 0282 93, 106 PAT 0288 93, 106 PAT 0294 96–97 PAT 0304 29 PAT 0305 66 PAT 0306 65, 96, 98

164 · Index PAT 0761 PAT 1062 106–07 PAT 1063 PAT 1352 PAT 1360 PAT 1374 PAT 1378 106 PAT 1397 PAT 1399 PAT 1400

32 14, 16, 96, 99, 16, 100–01, 106–07 97 106 13, 32, 37 16, 77, 88, 96, 98, 102, 98 96, 98 96, 98, 107

PAT 1403 PAT 1406 PAT 1409 PAT 1411 PAT 1414 PAT 1419 PAT 1524 PAT 1584 PAT 2632 PAT 2753 PAT 2754 PAT 2763

65, 91, 96, 98, 107 31 96, 98 96, 98 14, 16, 99, 106–07 96 26 14 27 34 10, 107 14, 65, 91

RES 4691 71, 86 RES 4859 79 RES 4909 71 RGDS 19–36 113 Schlumberger 1939, 61–63, no . II 10 Schuol 2000, no . 25 96, 98 SEG 34,1585 30 Seyrig 1932, 274–75, no . 1 10 Seyrig 1932, 276, no . 2 10 Seyrig 1939, 322–23, no . 28 25

General index Abdulhamid II, Sultan of the Ottoman Empire 1876–1909 136 Abu Qurayya, see Dios 67 Abu Shaʾar 67 Abu Shaʾar el-Qibli 78 Aemilius Celer, praefectus montis Bernicidis 78 Aemilius Marcianus Asklepiades, M . 96 Afghanistan 134 Al Uqla 71–73, 75 al-Assad, Bashar 18, 21, 132–33, 138, 140 al-Assad, Hafiz 18 al-Ḥamād 47, 49, 52, 54 Al-Muwayh, see Krokodilō al-Qaryatayn 53–54 Aleppo, see Beroia Alexander Severus 32, 76, 112, 116, 120 Alexandria 67–70, 80–81, 93, 125 Antioch 14–15, 80, 113, 118, 121–22, 125, 126 Antiochus III Megas, king of the Seleucid Empire 222–187 BC 41 Antiochus IV Epiphanes, king of the Seleucid Empire 175–164 BC 41 Antiochus VII Sidetes, king of the Seleucid Empire 138–129 BC 41–42 Antiochus XII Dionysus Epiphanes, king of the Seleucid Empire 87– 84 BC 25 Aphrodite 79 Apollonos Mikra 81 Apologos 73 Appian 26–27, 85 Arabia 40, 47, 65–67, 70–77, 83, 85, 93 Arak 48, 63 Armenia 76, 130 Asia Minor 14, 40, 42, 107, 112–13, 118 at-Tarfa 50–52, 60–61 Augustus, Roman emperor 27 BC– AD 14 24, 40, 127 Aukar 71

Aurelian, Roman emperor AD 270–275 15, 35, 44, 93, 111, 118, 120–28 Aureolus, Roman usurper AD 268 118 Babylon 42, 65, 86, 97 Babylonia 38, 40–44, 72 Bagdad 49 Bahrain, see Thilouoana Barcelona 133 Barygaza 73, 75 Bell, Getrude 49 Berenike 68–70, 76–83 Beroia 42, 45–46, 59, 80, 130, 135 Berytus 31–32 Bir ʿAlī 72 Black Sea 117 Bruttius Praesens, C ., special envoy to Syria AD 138 99 Caecilius Metellus Creticus Silanus, Q ., governor of Syria AD 12–18 10–11 Cairo 81 Callinicus of Petra 34 Caracalla, Roman emperor AD 211– 217 31, 76, 78 Caucasus 132, 134 Central Asia 41, 75, 129, 132 Chalcedon 117 Charakene 13, 37–38, 41 China 42, 47 Choumana 65 Classicus 79 Claudius II Gothicus, Roman emperor AD 268–270 118–23, 126–27 Cologne 116 Constantine 92 Ctesiphon 15, 41, 76, 120 Cumont, Franz 24 Danube 112 deities Abgal 49 Allāth 31, 35, 49 Arṣu 35 Ašar 49 Athena 35

Dionysos 69 Harpokrates 69 Helios 34 Hierobol, see Yarḥibōl Isis 69 Malakbēl 49 Maʿnū 49 Šalmān 49 Saʿdū 49 Tyche 14 Yarḥibōl 32, 69, 78, 101 ʾAglibōl 32, 49 Demetrius II Nicator, King of the Seleucid Empire 129–126 BC 41 Denderah, see Tentyris Didymoi 78–80 Diocletian, Roman Emperor AD 284–305 25, 33, 35–36, 62 Dios 80 Dioscorides, see Socotra Domitius Corbulo, Cn . 9–11 Drusus 10 Du Mesnil du Buisson, Robert Comte de 39 Ecbatana 41 Egypt 20, 29, 38, 55–58, 66–83, 91, 93, 118, 121, 125 Elymais 13, 41, 99, 106 Emesa 10–11, 31, 80, 118, 121–22, 125 Erdogan, Recep Tayip 135–137 Erech 10–11 Etruria 56 Euphrates 9–10, 13, 15, 24–27, 38, 41, 45–46, 49, 54, 68, 73–74, 77, 86–87, 99, 118, 121, 124 Eurysaces, baker 92 Finley, Moses I . 89 Firmus, Roman usurper AD 273 70, 93 Forat 65, 96 Foul Bay, see Berenike Gallienus, Roman emperor AD 253–268 33, 110, 114–27 Gaza 55, 73, 75, 137 Genna 99 Gerasa 32

166 · Index Germanicus 9–13, 19, 28–29, 44 Germany 92, 114–15, 118 Golan Heights 137–38 Gordian III, Roman emperor AD 238–244 76, 112, 114 Greece 39, 42, 114–15, 118, 137

Licinius Mucianus, C ., governor of Syria AD 67–69 9–10 Limyrike 73–74 Longinus 34, 122 Lucius Verus, Roman emperor AD 161–169 30, 76

Hadramawt 69, 71, 86 Hadrian, Roman emperor AD 117– 138 10, 19, 25, 29, 78 Haverfield, Francis 17–18 Hermogenes 14 Holland 88

Magnesia on the Maeander 14 Manilius Fuscus, Ti ., governor of Syria Phoenice AD 193–194 101 Mari 24, 46, 103 Mark Antony (M . Antonius) 26– 27, 44, 150 Maximinus Thrax, Roman emperor AD 235–238 112, 115 Media 41 Mediterranean 21, 38–40, 46–47, 56–59, 65, 67–68, 72, 74, 80–82, 85, 87, 133 Medvedev, Dmitry 132 Mesene, see Charakene Mesopotamia 12–13, 19–20, 27– 28, 40–46, 65–77, 80, 82–83, 85– 89, 101, 113 Minucius Rufus, legatus legionis 10 Mithradates I, Parthian king 171– 132 BC 41 Mommsen, Theodor 17, 88 Mons Claudianus 67 Mons Porphyrites 67 Musil, Alois 47, 49–52, 60, 63 Myos Hormos 68–70, 76, 78, 81

Illyria 114 India 13, 20, 37, 41, 65, 70–76, 81–82, 86, 91, 93 Indian Ocean 65–67, 70, 74–75, 80, 82–83, 85–87 Iran 21, 33–34, 41, 46–47, 73, 129– 30, 137–40 Iraq 49, 131–32, 138–39 Israel 21, 55, 130–31, 137–38, 140 Istanbul 135–36 Italy 57, 70, 88, 91–92, 110, 114, 118 Iulia Domna 34 Iuliopolis 81 Iulius Maior, Sex ., governor of Syria AD 136–140 99 Iulius Maximus 96 Iulius Priscus 32 Iunius Moderatus Columella, L . 56–57 Jebel Abu Riĝmen 49 Jebel Chaar 49 Jebel Merah 49, 62 Kainopolis 67 Kane (Qana’) 72–73 Kerbela 49 Khashm al-Minayh, see Didymoi Khirbet el-Bilaas 10 Koptos 29–30, 66, 68, 70, 75–76, 78–83, 91 Krokodilō 80 Lapid, Yair 137 Lausanne 136–37 Leontinoi 56–57 Leuke Kome 67 Licinius Crassus, M .

13, 43

Naples 133 Negev 55, 58, 62 Nessana 55–57 Nicomedia 117 Nisibis 76, 87, 120, 123–24 Omana 73 Orodes, king of Elymais 99, 106 Osman I, Ottoman ruler 135 Osmanoglu, Dündar Abdulkerim 135 Palmyra Agora 29–30, 37, 87, 98 Colonnaded Street 14, 25, 30, 32, 34, 36 Efqa spring 39, 48 Hellenistic City 12, 19, 25, 38–45

Temple of Allāth 31, 35 Temple of Baʿal-Šamen 31 Temple of Bēl 10, 14, 18, 24–25, 28, 30–32, 39, 97–100 Temple of Nebu 31 Tripylon 18 Valley of the Tombs 60 Palmyrenes Abū Magdi (or Malki) 69 Aeilius Borā 16, 96, 100–02, 106–08 Alexandros, envoy to Charakene 10–11, 13 Antiochus, usurper 126 Apsaios, προστάτης 35 Aqqayh 13, 97 Asklas 79 Athenodorus, tesserarius 79 Aurelius Mokimos, M ., soldier 78–79 Azi son of Abianas 58, 67–68, 71, 81 Aziz 71, 73 Barlaas 79 Bassos, horseman 79 Berichei 78 Garimay son of Nabuzabad 25– 26 Ḥairan 33–34 Herodianos, see Ḥairan Hierabole 69, 76 Iulius Aurelius Maqqay 67–68 Iulius Aurelius Marona 65, 93 Marinus, bishop of Palmyra 36 Maximus 79 Nešā 96–97, 102 Ogelos, see ʿOgēlu son of Maqqay 88 Rab son of ? 71 Septimius Odaenathus 15–16, 19, 33–34, 93, 103, 118–27 Septimius Worōd 33, 103 Šoʿadu son of Bōlyadʿ 96, 98– 102, 107–08 Taimarṣu son of Taime 96–97 Ulpius Yarḥai, M . 65, 96, 98, 107 Vaballathus 79 Yarḥai, governor of Thilouana 13–14, 37, 41, 46 Yarḥibōl son of Lišamšū 13, 96, 99, 106–07 Yedʿibēl son of ʿAzizu 97

Gener al index · 167 Zabdalas, son of Salmanos 66 Zabdas 118, 121, 123, 126, 128 Zenobia 16, 18–20, 34–35, 44, 62, 70, 85, 93, 118, 120–128 ʾAbgar (at Berenike) 69 ʾAbgar (at Socotra) 71–74 ʾAbgar son of Yarḥai 98 ʿOgēlu son of Maqqay 88, 98, 100, 102 Parthian Empire 9, 11, 13, 19, 26, 41–45, 77–78, 83, 95, 112 Paul of Samosata 34 Persia 15, 124, 140 Persian Gulf 13, 37, 42, 65–66, 68, 70, 72–76, 80, 82–83, 86–87, 91, 131, 133 Persis 30, 41, 73, 77, 123 Petronius, T . 91 Philip the Arab, Roman emperor, AD 244–249 112–115, 120 Phraates II, Parthian king 132–127 BC 41 Pityus 117 Pliny the Elder (C . Plinius Secundus maior) 9–12, 24, 44, 53, 57, 81, 91 Polanyi, Karl 89 Pompey the Great (Cn . Pompeius Magnus) 26, 43–44 Prokhanov, Alexander 133 Putin, Vladimir 21, 132, 134, 140 Qana’ 72–73 Qasr al-Heir al Gharbi 61 Qena, see Kainopolis Qift, see Koptos Qus, see Apollonos Mikra Quseir al-Qadim, see Myos Hormos Ras Banas 68 Rean, Arsen 132 Red Sea 38, 58, 65–83, 86, 91 Rhine 110, 115–16 Roman Empire 9, 11–20, 23, 29, 37–38, 45, 79, 81–82, 85–86, 90– 91, 97–99, 103, 108, 110, 124–126

Rome 30, 82, 88, 92 Rostovtzeff, Mikhail 94 Russia 21, 129–140

20, 87–89,

Sabkhet al-Mouh 51 Saloninus, Roman emperor AD 260 116 Sasanian Empire 33–34, 66, 74, 76–78, 83, 86–87, 112 Saubatha, see Shabwa Schlumberger, Daniel 49 Scythia 65, 73, 91 Seleucia on the Tigris 41–42, 65, 86, 97 Seleucia Pieria 70 Seleucid Empire 19, 25–26, 41–42, 45, 72, 86 Septimius Severus, Roman emperor AD 193–211 76, 101–02, 109, 112 Septimius Vorodes, see Worōd Seyrig, Henri 23, 25, 28–29 Shabwa 71–75 Shah, Suleiman 135 Shapur I, Persian king c . AD 240– 270 33–34, 112–13, 119–20, 123– 24, 126–27 Sicily 56 Silvanus 116 Socotra 38, 66, 70–75, 86 Sossianus Hierocles, praeses of Syria Phoenice 36 Spasinou Charax 10–13, 32, 37, 65, 73, 77, 86, 91, 97–99 Stanhope, Lady Hester Lucy 18 Stephanus of Byzantium 29 Strabo 27, 81, 86, 100 Sura 10 Sylvester, bishop of Rome AD 314– 335 92 Syria Phoenice, Roman province of 36 Syria, Roman province of 9, 11, 14, 44 Syrian Desert 13, 23–24, 32, 41, 45–47, 54, 62, 65, 70, 77, 80, 83, 86

Tanūkh 73, 77 Tentyris 66–68, 75 Terentius Varro, M . 56–57 Thilouana 13–14, 32, 37 Thilouos, see Thilouana Tiberius, Roman emperor AD 14– 37 10–11, 28, 44 Timagenes 125 Trajan, Roman emperor AD 98– 117 10, 30, 76, 78 Trebonianus Gallus, Roman emperor AD 251–253 113, 115, 120 Trimalchio 91 Tullius Cicero, M . 56 Turkey 21, 57, 129–31, 134–40 Umm al-ʿAmad 99 Valerian, Roman emperor AD 253– 260 15, 44, 111, 113–14, 116–17, 126, 128 Valerius Germanon, military tribune 78 Venidius Rufus, Q ., governor of Syria Phoenice c . AD 198 101 Vespasian, Roman emperor AD 69–79 10–11 Vologases III 77 Vologesias 13, 65–66, 71, 77, 97, 99 Waballat, see Vaballathus Wadi Abyad 52–53, 60–61 Wadi al-Miyah 52 Wadi al-Qubur 51, 60 Wadi al-Swab 52 Wadi Menih el-Heir, see Aphrodite 79 Will, Ernest 24, 88–89, 94–95 Wirth, Eugen 54 Yemen 86 Zakharova, Mariya 132 Zosimus 20, 35, 109–127 ʿAbdallāth son of ʾAḥitayʿ 99

oriens et occidens Studien zu antiken Kulturkontakten und ihrem Nachleben

Herausgegeben von Josef Wiesehöfer in Zusammenarbeit mit Pierre Briant, Geoffrey Greatrex, Amélie Kuhrt und Robert Rollinger.

Franz Steiner Verlag

17.

18.

19.

20.

21.

22.

23.

ISSN 1615–4517

Charlotte Lerouge L’image des Parthes dans le monde gréco-romain Du début du Ier siècle av. J.-C. jusqu’à la fin du Haut-Empire romain 2007. 327 S. mit 9 Abb., geb. ISBN 978-3-515-08530-4 Michael Blömer / Margherita Facella / Engelbert Winter (Hg.) Lokale Identität im Römischen Nahen Osten Kontexte und Perspektiven. Erträge der Tagung „Lokale Identität im Römischen Nahen Osten“ in Münster vom 19.–21. April 2007 2009. 340 S. mit 171 Abb., geb. ISBN 978-3-515-09377-4 Ted Kaizer / Margherita Facella (Hg.) Kingdoms and Principalities in the Roman Near East 2010. 453 S. mit 22 Abb., kt. ISBN 978-3-515-09715-4 Josef Wiesehöfer / Thomas Krüger (Hg.) Periodisierung und Epochenbewusstsein im Alten Testament und in seinem Umfeld 2012. 155 S., kt. ISBN 978-3-515-10114-1 Lucinda Dirven (Hg.) Hatra Politics, Culture and Religion between Parthia and Rome 2013. 363 S. mit 38 Abb., kt. ISBN 978-3-515-10412-8 Lawrence J. Baack Undying Curiosity Carsten Niebuhr and The Royal Danish Expedition to Arabia (1761–1767) 2014. 443 S. mit 22 Abb., geb. ISBN 978-3-515-10768-6 Julien Monerie D’Alexandre à Zoilos Dictionnaire prosopographique des porteurs de nom grec dans les sources cunéiformes 2014. 255 S. mit 11 Abb., kt. ISBN 978-3-515-10956-7

24. Joseph Gilbert Manning (Hg.) Writing History in Time of War Michael Rostovtzeff, Elias Bickerman and the “Hellenization of Asia” 2015. 153 S., kt. ISBN 978-3-515-10948-2 25. Rolf Strootman / Miguel John Versluys (Hg.) Persianism in Antiquity 2017. 557 S. mit 79 Abb., geb. ISBN 978-3-515-11382-3 26. Pierre Briant, übersetzt von Amelie Kuhrt Kings, Countries, Peoples Selected Studies on the Achaemenid Empire 2017. 658 S. mit 25 Abb., geb. ISBN 978-3-515-11628-2 27. Melanie Wasmuth Ägypto-persische Herrscher- und Herrschaftspräsentation in der Achämenidenzeit 2017. 392 S. mit 54 Abb., 4 Tab. und 12 Farbtafeln, geb. ISBN 978-3-515-11693-0 28. Helge Bert Grob Die Gartenlandschaft von Pasargadai und ihre Wasseranlagen Topographischer Befund, Rekonstruktion und achaimenidischer Kontext 2017. 384 S. mit 164 z. T. farbigen Abb. auf 112 Tafeln, geb. ISBN 978-3-515-11867-5 29. Frank Schleicher / Timo Stickler / Udo Hartmann (Hg.) Iberien zwischen Rom und Iran Beiträge zur Geschichte und Kultur Transkaukasiens in der Antike 2019. 356 S. mit 21 Abb., kt. ISBN 978-3-515-12276-4 30. Christopher Schliephake On Alexander’s Tracks Exploring Geographies, Memories, and Cultural Identities along the North-West Frontier of British India in the Nineteenth Century 2019. 312 S. mit 8 Abb., geb. ISBN 978-3-515-12400-3

Palmyra – in the Roman imperial period, the marvel of the Syrian Desert was situated at the crossroads of the intercontinental long-distance trade, in a political and cultural twilight between the East and the West: inter duo imperia, “between the two empires”, according to Pliny the Elder. How accurate is Pliny’s description of the oasis of Tadmur? How strongly

ISBN 978-3-515-12774-5

9

7835 1 5 1 2 7 745

was Roman influence felt in the city of Bēl – and how did it develop over the centuries? What was the significance of trade? And how did the close interaction between sedentary and nomadic populations shape society in the oasis? The authors revisit the textual and material evidence on and from Palmyra in the light of recent research, spanning five centuries of Near Eastern history.

www.steiner-verlag.de Franz Steiner Verlag