e-ISSN 2610-9336 ISSN 2610-881X

Antichistica 28 Studi orientali 11

—

# **Corpus of Nabataean Aramaic-Greek Inscriptions**

Giuseppe Petrantoni

**28**

PETRANTONI

**INSCRIPTIONS**

**ARAMAIC-GREEK** 

**CORPUS OF NABATAEAN** 

# Antichistica Studi orientali

Collana diretta da Lucio Milano

28 | 11

# Antichistica Studi orientali

#### Direttore scientifico

Lucio Milano (Università Ca' Foscari Venezia, Italia)

#### Comitato scientifico

Claudia Antonetti (Università Ca' Foscari Venezia, Italia) Filippo Maria Carinci (Università Ca' Foscari Venezia, Italia) Ettore Cingano (Università Ca' Foscari Venezia, Italia) Joy Connolly (New York University, USA) Andrea Giardina (Scuola Normale Superiore, Pisa, Italia) Marc van de Mieroop (Columbia University in the City of New York, USA) Elena Rova (Università Ca' Foscari Venezia, Italia) Fausto Zevi (Sapienza Università di Roma, Italia)

#### Direzione e redazione

Dipartimento di Studi Umanistici Università Ca' Foscari Venezia Palazzo Malcanton Marcorà Dorsoduro 3484/D 30123 Venezia

Antichistica | Studi orientali e-ISSN 2610-9336 ISSN 2610-881X

URL http://edizionicafoscari.unive.it/it/edizioni/collane/antichistica/

Giuseppe Petrantoni

Venezia **Edizioni Ca' Foscari** - Digital Publishing 2021

Corpus of Nabataean Aramaic-Greek Inscriptions Giuseppe Petrantoni

© 2021 Giuseppe Petrantoni per il testo | for the text © 2021 Edizioni Ca' Foscari - Digital Publishing per la presente edizione | for the present edition

**Corpus of Nabataean Aramaic-Greek Inscriptions**

# cb

Quest'opera è distribuita con Licenza Creative Commons Attribuzione 4.0 Internazionale This work is licensed under a Creative Commons Attribution 4.0 International License

Qualunque parte di questa pubblicazione può essere riprodotta, memorizzata in un sistema di recupero dati o trasmessa in qualsiasi forma o con qualsiasi mezzo, elettronico o meccanico, senza autorizzazione, a condizione che se ne citi la fonte.

Any part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means without permission provided that the source is fully credited.

Certificazione scientifica delle Opere pubblicate da Edizioni Ca' Foscari - Digital Publishing: il saggio pubblicato ha ottenuto il parere favorevole da parte di valutatori esperti della materia, attraverso un processo di revisione doppia anonima, sotto la responsabilità del Comitato scientifico della collana. La valutazione è stata condotta in aderenza ai criteri scientifici ed editoriali di Edizioni Ca' Foscari.

Scientific certification of the works published by Edizioni Ca' Foscari - Digital Publishing: the essay published has received a favourable evaluation by subject-matter experts, through a double blind peer review process under the responsibility of the Scientific Committee of the series. The evaluations were conducted in adherence to the scientific and editorial criteria established by Edizioni Ca' Foscari.

Edizioni Ca' Foscari - Digital Publishing Fondazione Università Ca' Foscari Venezia Dorsoduro 3246, 30123 Venezia http://edizionicafoscari.unive.it/ | ecf@unive.it

1a edizione aprile 2021 | 1st edition April 2021 ISBN 978-88-6969-507-0 [ebook] ISBN 978-88-6969-508-7 [print]

The publication of this book was kindly and generously supported with a grant provided by Professor Lucio Milano

Corpus of Nabataean Aramaic-Greek Inscriptions / Giuseppe Petrantoni— 1. ed. — Venezia: Edizioni Ca' Foscari – Digital Publishing, 2021. — 182 p.; 16 cm. — (Antichistica; 28,11). — ISBN 978-88-6969-508-7.

URL https://edizionicafoscari.unive.it/it/edizioni/libri/978-88-6969-508-7/ DOI http://doi.org/10.30687/978-88-6969-507-0

Giuseppe Petrantoni

# Abstract

The aim of the present study is to collect together all the Nabataean Aramaic-Greek epigraphic evidence existing in the Middle-East and Oriental Mediterranean areas and dating from the 1st century BCE to the 3rd-4th century CE. The volume contains 51 inscriptions written in Nabataean and Greek.

The texts, which are mostly engraved on stones, have been accurately identified, transcribed and analysed through an historical and epigraphic commentary.

It is known that, in the Hellenistic and Early Roman Near East, the contact between Greek and Nabataean led the inhabitants to erect statues and write inscriptions in public spaces employing one of the two languages or both. This practice is considered as a Hellenistic influence. In this uncertainly diglossic situation, Greek was employed as a prestige language and *lingua franca* and Nabataean as a vernacular idiom.

To date there exists no comprehensive *corpus*, either in digital or in paper format, bringing together all these Nabataean-Greek inscriptions, of which there are around six thousand (dating from the 2nd century BCE to the 4th century CE).

This collection could open up important avenues for further research in the analysis of the linguistic contact between Nabataean and Greek. It may deepen our knowledge of the linguistic situation of Nabataean in the field of Semitic Philology and Semitic Epigraphy; in addition, the *corpus* allows us to study the modalities of cultural exchange (especially in social and religious contexts) between Nabataeans and Greeks. Moreover, the investigation of onomastics (mainly of Nabataean names transcribed into Greek script) may allow us to know more about the Nabataean phonological system.

**Keywords** Nabataean Aramaic. Greek. Epigraphy. Diglossia. Ancient Near East.

Giuseppe Petrantoni

# Acknowledgements

The initial version of this work was the core of my PhD dissertation *Corpus delle iscrizioni bilingui aramaico nabateo-greche. Approfondimenti onomastici e problemi di ricostruzione del sistema fonologico del nabateo* which I discussed in February 2017 in Sapienza University of Rome under the supervision of Prof. Alessio Agostini. I am very grateful to him for guiding me attentively during its preparation.

I would like to express my deep gratitude to Prof. Lucio Milano for his patient guidance, encouragement and useful suggestions in publishing this research work.

I wish to express my gratitude to Prof. Maria Giulia Amadasi Guzzo for her invaluable advice and assistance in keeping my progress on schedule, from checking readings and literature to discussing the inscriptions.

I am grateful also to Prof. Marco Moriggi who showed a great interest in my research works giving me much helpful advice.

I thank Prof. Adalberto Magnelli who gave me precious suggestions on various issues concerning the bilingualism and the role of Greek in the Ancient Near East.

Finally, I wish to pay homage to the memory of Prof. Gianfranco Fiaccadori who has always shown interest in studying the Greek-Semitic bilingualism; this indirectly steered me toward the argument proposed here.

Giuseppe Petrantoni

# Table of Contents


To Sonia, whose love, support and advice have never failed, this book is dedicated

Giuseppe Petrantoni

# Introduction

**Summary** 1 Introduction. – 2 Nabataean Aramaic and Greek. – 2.1 The Variety of Nabataean Aramaic. – 2.2 Nabataean Aramaic in Contact with Greek. – 3 Nabataean Aramaic-Greek Inscriptions. – 3.1 Research Background. – 3.2 The Numbering and Nature of the Inscriptions.

#### 1 Introduction

The history of the Nabataeans is clearly linked to the history of the Ancient Near East. According to Greek and Roman sources, as well as the epigraphic and archaeological evidence, Nabataeans frequently interacted with Greeks, Romans and Jews. Since the Nabataeans were a nomadic tribal society, there exists no real Nabataean literature. They presumably had an oral tradition that can no longer be reconstructed today.<sup>1</sup>

The Nabataeans constituted an ethnic group in which most literates (a small minority, who nonetheless occupied a dominant social, economic, and political position) used Greek as the language of communication in formal contexts. Aramaic, instead, was considered as an informal and vernacular language dating to a later period.

The Nabataeans were probably bilingual, as witness the inscriptions written in Greek and Nabataean collected in the present volume. Since the times of the Achaemenid Empire (ca. 550-330 BCE), during which the royal chancellery continued with the deeply-entrenched employment of Aramaic in local and provincial administration, there was an expansion of multilingualism and a spread of varieties of Aramaic, Greek and other languages, such as Hebrew and

1 Wenning 2007, 25.

Arabic idioms, which became evident during the subsequent Greco-Roman period and until the 6th-7th century CE.<sup>2</sup>

When Alexander the Great had conquered the lands of the Near Eastern civilizations and the local dynasties were established by his successors, the Hellenization of this geographical area had a strong impact causing wars and social instability, especially in Syria.<sup>3</sup> The real degree of the influence of the Greek language and culture is still debated, and varied from region to region.<sup>4</sup> Furthermore, after the Greek and Roman conquest of the Near East, including Arabia, Augustus established a period of relative peacefulness, which is best known as *Pax Romana*. During this period (from 27 BCE to 180 CE) the Roman Empire reached its peak land mass area and the Roman trade in the Mediterranean Sea increased;<sup>5</sup> the immediate consequence was the emergence of several new city-states (such as Petra and Palmyra) that adopted Hellenistic customs.<sup>6</sup>

It would not be appropriate here to enter into a detailed history of the Nabataeans.<sup>7</sup> We need only note that, from the social and cultural point of view, the Nabataeans were a nomadic Bedouin tribe that roamed the Arabian desert and moved with their herds to wherever


7 For a historical overview, we may refer to several modern works, such as Starcky 1955; Bowersock 1983; Lindner 1997; Wenning 1987; 2007.

<sup>2</sup> The rise of Islam and the consecutive wars of conquest of the Muslim armies transformed the hegemony and the society in the Near East. Islam achieved a rapid success without facing strong resistance, as demonstrated by the defeat of Heraclius in 636 at the battle of Yarmuk (Bowersock 1990, 71). This was probably because the Hellenization of the Near East was or had become, to some extent, superficial. In addition, Greek had ceased to be relevant as an epigraphic medium by the end of the 8th century, but it was still used occasionally, alongside Syriac, for Christian rituals within the new dominant Islamic culture in which Arabic took over as the prevailing language (cf. Di Segni 2009).

<sup>3</sup> There is not a word for *Hellenization* in classical or Byzantine Greek language and the notion of *Hellenism* identifies the language and the culture: "in which peoples of the most diverse kind could participate. […] Hellenism […] represented language, thought, mythology, and images that constituted an extraordinarily flexible medium of both cultural and religious expression" (Bowersock 1990, 7). Apart from the Greek language, the First Book of the Maccabees offers us a historical account of events. Cf. 1Macc 1,1-9. It is usually accepted that the first usage of the term *hellenismos* is found in the Second Book of the Maccabees in which it is narrated that under Antiochus' rule a gymnasium was built in Jerusalem and young men were obliged to wear *foreign* clothes: ἦν δ᾿οὕτως ἀκμή τις ῾Ελληνισμοῦ καὶ πρόσβασις ἀλλοφυλισμοῦ διὰ τὴν τοῦ ἀσεβοῦς (2Macc 4,13).

<sup>6</sup> The Hellenistic influence also affected the Semitic custom of the inhabitants to erect statues and carve honorific inscriptions in public using their local variety of Aramaic or reproducing the text in Greek in order to exhibit their degree of literacy and economic power. So, for instance, in Palestine by the 4th century BCE the shift from Hebrew (which became the holy language) to Aramaic and the spread of Greek had triggered a complex linguistic development in which Aramaic presumably came to dominate (cf. Gzella 2015, 226 fn. 709, who quotes Poirier 2007).

they could find pasture and water. Although the precise origin of the Nabataeans remains uncertain<sup>8</sup> (they were probably an Arab people who inhabited northern Arabia and the southern Levant), we may assert that the Nabataean kingdom, which remained independent from the 4th century BCE until it was annexed by the Roman Empire in 106 CE, emerged as a key player in the region during their period of prosperity. There are substantial doubts about the identification of the Nabataeans with other peoples referred to in the Assyrian<sup>9</sup> and Biblical<sup>10</sup> sources. The most common theory, according to which the Nabataeans were an Arab group, is today supported by three pieces of historical and linguistic evidence. Firstly, when Greek writers mentioned these people they usually refer to them as "Arabs".<sup>11</sup> Secondly, there is the presence of Arabic personal names in the Na-

9 Cf. e.g. the annals of Tiglath-pileser III (745-729) in which the *Nabatu*, among the 36 Aramaic tribes against Babylon, is found (Luckenbill 1926-27, 283 and especially Tadmor, Yamada 2011, nos. 4.5; 40.5; 47. 6; 51.6; 52.6).

10 In Gen 25:13 and in 1Chr 1, 29 the term *nĕbāyōt* is found, but according to Starcky (1966, 900-3) there is a linguistic incompatibility between the forms *nbyt* and *nbṭw* (the latter is the form used by the Nabataeans to call themselves). In fact, it entails the passage of /ṭ/ into /t/ and the loss of /y/. Conversely, Broome (1973, 1-16) supposes that the biblical *nĕbāyōt* are actually the Nabataeans; this assumption is supported by the fact that in Semitic the shift of /t/ into /ṭ/ is possible (cf. Abu Taleb 1984, 3-11) and the root \**nby*, of obscure origin, does not appear to be recorded in the *corpus* of the pre-Islamic inscriptions, while the root \**nbt* is common in Akkadian and North-West Semitic. The suffix /-oṯ/, in *nbyt*, as a plural feminine represents another linguistic problem (Graf 1990, 67-8). See also Coogan, Metzger 2004, s.v. "Nabateans", 248.

11 Among them Diod. Sic. 19.94.1: τὴν χώραν τῶν Ἀράβων τῶν καλουμένων Ναβαταίων. According to some scholars, the Nabataeans quoted by Diodorus were not Arabs. For Retsö (2003, 364-91 and 623-6), the term 'Arab' mainly refers to a social *status* rather than an ethnic one; he also interprets the original reading as *nomádes* rather than *Nabataíoi* on the basis of two manuscripts from chapter 19 of Diodorus: *Parisinus graecus* 1665 (dating back to the 10th century CE) and *Laurentianus* 70.12 (dating back to the late 15th century CE). In the latter manuscript the Nabataeans are not mentioned (Retsö 2003, 283-8 and 1999, 115-16). Cf. also Fisher 1906, 146, no. 1. In another passage, Diodorus mentions the "Arabs who bear the name of Nabataeans": Ἄραβες οὓς ὀνομάζουσι Ναβαταίους (2.48.1). The reliability of Diodorus' narration is supported by the fact that the fundamental source, for the writing of books 18-20, is Hieronymus of Cardia, who was a friend of Antigonus and an eyewitness of the events during the expedition against Petra (Diod. Sic. 19.44.3).

<sup>8</sup> Milik (1982, 261-5) suggests that the Nabataeans were inhabitants of the Arabian Peninsula, their native land, taking into account as proof the obscure phrase (which is found in several Aramaic texts from Palmyra, Petra and Mada'in Saleh): "the God of *ṣ'bw*"; the latter is identified as the god of the "Luck of the Nabataeans". According to Milik, *Ṣa'bū*, which is placed in the Persian Gulf, is the native land of the Nabataeans before they moved to the west, toward Syria and Transjordan. E.A. Knauf argues (1986, 74-86) that the Nabataeans originated from the ancient Arab tribal confederation of the Qedarite and the evidence to corroborate this assumption is that the god Dūšarā was identified with the indigenous deity Qōs, who is the national god of the Edomites (see the bilingual from Bosra no. 20); *contra* D.F. Graf (1990, 45-75) who asserts that the Nabataeans came from Mesopotamia. For an illustrative summary of the debates on Nabataeans' origins, with related bibliography, cf. Parr 2003, 27-35 and *Quellen*, 15-19.

bataean inscriptions. Finally, we might consider the frequent usage of Arabic elements, such as particles, verbs, words and whole sentences, in Nabataean.<sup>12</sup>

# 2 Nabataean Aramaic and Greek

### **2.1 The Variety of Nabataean Aramaic**

Nabataean is an epigraphic language, one of several varieties of Aramaic, belonging to Middle Aramaic (300 BCE-first centuries CE),<sup>13</sup> that was presumably spoken from the 2nd century BCE to the 3rd-4th century CE. As K. Beyer suggests: "Nabataean stands nearer to Achaemenid Imperial Aramaic than does Hasmonaean";14 in fact, various archaic morphosyntactic features distinguish Nabataean from the other Middle Aramaic varieties, such as Palmyrene and Hatraean.<sup>15</sup> Although Nabataean is considered an offshoot of Achaemenid Aramaic: "there is thus no sharp linguistic distinction between the Achaemenid standard idiom on the one hand and its evolving heritage in the local varieties of Aramaic of the Greco-Roman period on the other".<sup>16</sup>

Nabataean was employed as *lingua franca* to ease the communication among the peoples of the Arabian Peninsula and as an international language to facilitate trade and business in the Near East. According to M. Morgenstern (1999, 135) Nabataean Aramaic is formed of three main elements: 1) a *sub-stratum* constituted by an Aramaic literary tradition going back to the Achaemenid Persian era (5th-4th century BCE); 2) an inner-development of the language; 3) the Arabic influence, which can be observed mainly in the vocabulary.

12 Cf. Diem 1973, 227-37.

13 For a detailed description and definition of Middle Aramaic, see in particular Fitzmyer 1979, 61-2 and Beyer 1986, 43-53. For a summary of the main theories with related bibliography, see Moriggi 2012, 279-89.

14 Beyer 1986, 27.

15 In particular, see Healey 1993, 55-9; Healey 2009, 38-40; Morgenstern 1999, 136- 9. Some typical linguistic features of Nabataean are e.g.: 1) the usage of the relative *dy* < *zy* (archaic); 2) the graphic preservation of etymological \**n* before consonants; 3) the usage of the grapheme *š* instead of \*/ś/; 4) the extension of the perfect 3rd m. pl. verbal suffix to the feminine; 4) the masculine plural in -*īn*; 5) the employment of *'-* instead of *h-* in the formation of the causative *ap'el* form; 6) the transformation of /l/ > /n/, like in *mnkw* < *mlkw*; 7) the shift of /ā/ > /ō/, like in *'nwš* </'ēnāš/; 8) the assimilation of *n*, like in *'tt'* < \**'ntt'*, *mṣb* < root *nṣb*; 9) the usage of final -*w* in masculine personal names of probable Arabic origin; 10) the employment of *yt* like *nota accusativi*.

16 Gzella 2015, 213.

The Nabataean inscriptions were written in a local Aramaic variety using a local script. Nevertheless, various Arabic loanwords<sup>17</sup> and grammatical borrowings have been observed.<sup>18</sup> Today most scholars believe that the linguistic presence of Arabic in Nabataean is due to the fact that the Nabataeans spoke Arabic in everyday life but employed Aramaic as a *lingua franca* to write their inscriptions or other documents;19 in fact, according to G. Garbini and O. Durand, the Nabataeans were already sociologically an Arab people, but were still linguistically Aramaic.<sup>20</sup>

Nabataean uses a typical script deriving from the Persian Chancellery, as is evident, for instance, in the 5th century BCE Elephantine papyri. F. Beer, in 1840, was the first to correctly read some graffiti from Sinai, deciphering the Nabatean characters.<sup>21</sup> The Nabataean script could be classified as a monumental script used for public, funerary or religious inscriptions on stones,<sup>22</sup> and as a cursive script adopted for legal, diplomatic and commercial documents.<sup>23</sup> The script used in graffiti does not belong to a specific classification. It is probably that carvers of graffiti always attempted to make their inscriptions more formal than the common calligraphic script.<sup>24</sup>

From a historical point of view the development of the Nabataean script can be classified into three specific stages. The first is the ancient period, dating from the end of the 2nd century to the beginning of the 1st century BCE, during which the letters are quite wide in shape and there are not many ligatures; the second is known as the classical or calligraphic period25 represented by the inscriptions, dating back to the 1st c. BCE-1st c. CE, in which the writing shows elongated characters and a growing tendency to ligature; finally, the last stage depicts the usage of the characters during the period following the end of the Nabataean Empire, but immediately before the Islamic period. It would not be suitable to analyse here the late development


23 These documents were mostly written on papyri or scrolls, as we can see in the Nabataean texts of the Babatha archive in Yadin et al. 2002.

24 Healey 2011, 49.

25 The term calligraphic for the Nabataean calligraphy was first employed by J. Starcky on the basis of the "numerous curves and ligatures due to the quill of the scribes" (Starcky 1966, 931).

<sup>17</sup> For instance: *gt* 'corpse' (< Ar. *ǧuṯṯah*); *wld* 'offspring' (< Ar. *walad*); *l'n* 'to curse' (< Ar. *la'ana*); *nšyb* 'father-in-law' (< Ar. *nasīb*). Cf. Nöldeke 1885; O'Connor 1986; Healey 1995, 78-9.

of the Nabataean alphabet; it is sufficient to underline the fact that the majority of scholars today agree that Arabic writing originated from the Nabataean<sup>26</sup> rather than from the Syriac script.<sup>27</sup> The writing from Hauran shows some different features compared to the genuine Nabataean script. The inscriptions from Hauran, dating back to the end of 1st c. BCE, preserve less elongated and more isolated letters.<sup>28</sup>

The most ancient Nabataean epigraph comes from the town of Haluza dating back to about 170 BCE The geographical area within which the Nabataean inscriptions have been found includes Jordan, south Syria, the Negev, Egypt (the eastern desert and the Sinai Peninsula), the northwestern part of the Arabian Peninsula (Hejaz), the Aegean islands and various sites in southern Italy.<sup>29</sup>

The inscriptions, including the bilingual examples, are of two types: dedicatory and funerary. The former are written on an object, a statue or an altar dedicated to a deity. The main formula is: this is the statue (altar etc.) that X made + the name of the god to whom it is dedicated + the reason (often 'for the life') of the ruling king + the date + the artisan's name (not always available). The latter are engraved on tombs, blocks of stone (in this case the text is longer) and directly on rocks (the extension of these letters is usually less significant).<sup>30</sup> The main structures are the following: if they deal with long texts written on tomb façades, they highlight the owner's name (of the tomb) + the members of the family + formal data about the tomb and the family.<sup>31</sup> By contrast, in the short texts the sequence reports *npš'* ('tomb') + the name of the deceased. The graffiti, which are carved on rocks,32 follow approximately the same pattern, that is: *dkyr* ('let be remembered') + the believer's name frequently followed by *šlm* ('peace') or *bṭb* ('in good').

27 Cf. Starcky 1966, 933 and Noja 2006.

30 Most bilingual inscriptions included in this corpus belong to this category.


<sup>26</sup> For further details see Gruendler 1993 and the more recent work of Nehmé 2010.

<sup>28</sup> A good example is the inscription of Salkhad, dating back to 95 CE (*CIS* II nos. 184 and 183 = Milik 1958, 227-8). Cf. also the bilinguals from Sī', nos. 26-29.

<sup>29</sup> More specifically, about 1,000 inscriptions (partly unpublished) have been found in Petra. They are dedications on statues erected by members of the royal family. In Mada'in Ṣaleḥ, there are monumental tombs decorated with majestic façades in Greek style; the inscriptions are longer than those of Petra, reporting the judicial matters related to the property of the tomb and dating back to the first 75 years of the 1st c. CE. In Bosra, the epigraphs appear to be few. In Mount Sinai, 3,851 short graffiti are engraved (they are included in *RIGP*), dating back to 2nd-3rd c. CE. Bilingual inscriptions (nos. 49, 50 and 51) have been found in the Aegean Sea, in Miletus and on the islands of Delos and Kos. In Italy we have two inscriptions from Pozzuoli (in *Quellen*, 116-19) and three from Rome (*Quellen*, 108-11).

#### **2.2 Nabataean Aramaic in Contact with Greek**

The term bilingualism refers to an individual's alternate use of two or more languages. When defining the alternate employment of two languages in a society, the noun diglossia<sup>33</sup> is more appropriate; in fact, it more precisely identifies a situation in which two dialects or languages are used by a single language community.

Alexander the Great's invasion of the Ancient Near East laid the foundations for the spread of Greek in the new conquered regions. This study focuses on multilingualism in the Hellenistic and Roman Near East, and in particular on the contact between Nabataean Aramaic and Greek. It is known that different varieties of Aramaic were widely employed as vernaculars in the Hellenistic and Roman Near East. It is worth bearing in mind that Greek permeated large parts of the Fertile Crescent becoming the official language of the administration. The impact of Greek on Aramaic in these multilingual settings involved lexical loans regarding architectural and administrative terminology, but did not trigger any phonological nor morphosyntactic interference.<sup>34</sup> Moreover, in the Hellenistic and Early Roman Near East the interaction between Greek and Nabataean caused inhabitants to adopt the so-called Hellenistic epigraphic habit, which mainly consisted in erecting statues and engraving inscriptions in public spaces using one of the two languages or both. In this questionable diglossic situation, Greek was used by the upper classes as the high-register variant and Aramaic as the vernacular of the uneducated masses; in addition, the new written forms of Aramaic acted as prestige languages and as a vehicle of indigenous cultural affinity, as Gzella argues.<sup>35</sup>

The language that spread in the Nabataean territories, besides the local Aramaic, was Koine Greek, a common supra-regional form of Greek spoken and written during the Hellenistic and Roman antiquity and the early Byzantine era, or Late Antiquity. Koine Greek is also known as Alexandrian dialect, common Attic, Hellenistic<sup>36</sup> or biblical Greek37 and it was used between about 330 BCE-330 CE (subsequent-

<sup>33</sup> With the term *diglossia* we refer to a kind of bilingualism in a society in which one of the languages has high prestige and the other one has low prestige (Ferguson 1959).

<sup>34</sup> Gzella 2015, 223.

<sup>35</sup> Gzella 2015, 215-16 and in detail Gzella 2005 and 2006.

<sup>36</sup> It evolved as a result of the spread of Greek following the conquests of Alexander the Great in the 4th century BCE Koine Greek represents the second stage in the development of Greek after the ancient period (about 800-330 BCE).

<sup>37</sup> Koine Greek is the original language of the New Testament, the Septuagint (the 3rd century BCE Greek translation of the Hebrew Bible), and the earliest Christian theological writing by the church fathers.

ly we refer to the medieval period, 330-1453 CE). Oral and written Koine Greek served as the *lingua franca* (also for literary purposes) of many Mediterranean regions and in the Middle East, spreading as far as India. In such a wide geographical context it was inevitable that speakers of different languages, in such distant regions, adopted Greek in a more or less correct form and according to their level of education. Presumably the same may well have been true in the Nabataean realm, where Aramaic speakers used Greek,with a degree of correctness which depended on their background.

Koine Greek was mainly based on Attic, but it was not Attic, or rather it was a local roughly *atticized* dialect. According to Meillet, *foreigners* (here we take into account the Nabataean speakers) spoke a type of Greek that: "a été celui des Grecs avec lesquels ils ont été en rapports, et ces Grecs n'ont été que pour une faible part des Athéniens".<sup>38</sup>

From a historical point of view, Greek was not widely used in the Nabataean realm until about 106 CE as demonstrated in the documents from the Babatha archive. In fact, none of the Greek texts predate 106 CE.<sup>39</sup> Therefore, following the annexation of *Nabataea* to the Roman Empire, Greek began to spread considerably across Petra and the other Nabataean regions replacing Aramaic as the official language of bureaucracy after the 4th century CE, as previously mentioned.

In the Nabataean-speaking territories, as well as in Syria and Mesopotamia, a complex linguistic landscape, focused on bilingualism, took shape, in which the linguistic ability and proficiency of the speakers, the level and the nature of linguistic interference, and their awareness of the diglossic situation, are particularly salient.<sup>40</sup> Although the strong influence of Arabic on Nabataean is clearly evident, as witness the presence of Arabic personal names in the Nabataean onomastics, the lesser use of Greek personal names seems to be due to the fact that Greek was learnt through formal education and was not spoken in domestic environments.<sup>41</sup> In Palmyra, as well as in Petra, Aramaic was employed in religious and domestic contexts, while Greek was spoken in public activities and trading relations with foreign territories.

On the basis of evidence arising from the *Nabataea*, it would appear that the social and linguistic situation was different from that of the other areas in the Near and Middle East where the introduc-

<sup>38</sup> Meillet, A. (1913). *Aperçu d'une histoire de la langue grecque*. Paris, 229, reprinted in Italian, see Meillet [1913] 2003, 375.

<sup>39</sup> Healey 2011, 48.

<sup>40</sup> Taylor 2002, 298.

<sup>41</sup> Taylor 2002, 318.

tion and usage of Greek could be examined from synchronic and diachronic perspectives; we may indeed imagine the impact and the increased density of Greek usage in Palestine<sup>42</sup> and Egypt.<sup>43</sup>

In the 1st century BCE, the Aramaic varieties and Hebrew (in Palestine) were active languages in the Near East being the L1 for the indigenous inhabitants, whereas Greek became the L1 for the social and political elite and the L2 for the indigenous community who employed it in social, administrative and economic environments.<sup>44</sup> Greek was the *lingua franca* from Greco-Roman Egypt to the eastern Mediterranean where Ptolemaic Egypt and Seleucid Syria were created after the Alexander's death.

In the Arabian Peninsula and Mesopotamia, between the 1st c. BCE and the 1st c. CE, local Semitic languages were the L1 for the indigenous population, whereas Koine Greek was not an official language.

As regards the Nabataean Kingdom, a diachronic perspective describes some sociolinguistic environments of Greek usage bolstered by historical and material sources.

The first linguistic contact between Greek and Nabataean Aramaic dates back to 312 BCE thanks to the account by Diodorus Siculus45 who recalls two Macedonian military campaigns, led by Antigonus I Monophthalmus, against the Arabs/Nabataeans in Petra.<sup>46</sup> The Macedonians reached Petra and took prisoners. Afterwards the Nabataeans defeated the Macedonians<sup>47</sup> and wrote to Antigonus 'in Syrian characters'.48 This passage refers to Aramaic as a *lingua franca* of the powers of the Near East. Even though, according to classic authors (Diodorus and Strabo) Ἀσσύρια Γράμματα designates the cuneiform writing, it is sometimes also used in reference to Aramaic script. Although Greek was the official language in Palestine, in the

45 Diod. Sic. 19.94.1-98.1.

<sup>42</sup> Koine Greek became first *lingua franca*, then prestige language and finally a widespread vernacular among the inhabitants in the 1st century CE. The Greek New Testament documents constitute one piece of synchronic evidence for this, as do the significant number of documentary Greek papyri found in a variety of sites, including Masada and different sites around the Dead Sea. See Porter 2016, 212-27.

<sup>43</sup> The documentary papyri found in Egypt show that Koine Greek was not only the prestige language of the Greco-Roman elite, but also the second language of the working class with Demotic and then Coptic being its first language. See Vierros 2014.

<sup>44</sup> The L1 is the first language and the L2 the second language; the L1 is the native language or mother tongue, whereas the L2 is a language learnt in a second moment in relation to the mother tongue.

<sup>46</sup> The expedition against the Nabataeans is believed to have taken place in 311 B.CE following the previous campaign against Gaza (Graf 1990, 51 fn. 30).

<sup>47</sup> Diod. Sic. 19.95.3-5.

<sup>48</sup> Diod. Sic. 19.96.1: πρὸς δ᾽ Ἀντίγονον ἐπιστολὴν γράψαντες Συρίοις γράμμασι.

4th century BCE in Petra the Nabataeans were not continually exposed to Koine Greek and they used their Aramaic variety to write their official letters.

Following the Seleucid dominion and the forced Hellenization, led by Antiochus IV Epiphanes, the Nabataeans were 'clients' of the Seleucids and involved in their affairs regarding Judaea. In this period, until the Hasmonean revolt (116-110 c. BCE), an important linguistic change occurred and Greek became the *lingua franca* and the prestige language in the eastern Mediterranean as well. The Nabataeans, who were one of the numerous nomadic tribes of Bedouins wandering the Arabian desert,49 still remained on the fringes of the Hellenistic territories and their contacts with the Greek world usually took place through the trade routes, when Petra was the last staging point for the caravans who carried spices to send to the European markets through the port of Gaza.<sup>50</sup>

The first Nabataean king, Aretas I (169 BCE), recorded in 2Mac 5, 7-8, is cited in the inscription from Haluza in Aramaic, but not in Greek.<sup>51</sup> This represents further evidence that the language used by the Nabataeans was still Nabataean Aramaic. In the 2nd century BCE the Nabataeans were not in contact with Greek in a diglossic situation, and therefore presumably still used Aramaic for their official purposes.

From the 1st century BCE the Nabataeans started to use Greek alongside Nabataean Aramaic. During the reign of Aretas III (85/84- 62 BCE), who conquered Damascus, the Nabataeans began to coin as a proof of their wide economic and politic independence. Coins were written in Greek and Aretas styled himself as 'Aretas Philhellen'.<sup>52</sup> Under Aretas III's rule the Nabataeans, who were a nomadic tribe, changed their style of life becoming a Near Eastern power allied with Greek culture and language; in fact, Koine Greek was imposed by king Aretas III as a vehicular language and Hellenistic architecture was also promoted, as is especially visible in Petra. During this time the inscriptions were carved in Nabataean Aramaic and also in Greek; but until the middle of the 1st century there is no sign of bilingual Nabataean Aramaic-Greek inscriptions. Presumably for most of the 1st century BCE, in accordance with the wishes of Aretas III, the

<sup>49</sup> They were described as 'Arabic nomads' as reported in Joseph. *AJ* 12.333-335 and 1Mac 5,24-25 and 2Mac 12.

<sup>50</sup> The Nabataeans controlled many trade routes towards the South along the Red Sea shore in the Hejaz desert, and towards the North to Damascus.

<sup>51</sup> *Quellen*, 393-5.

<sup>52</sup> See the coin of Aretas III from Damascus: βασιλέως Ἄρέτου Φιλέλληνος '(coin) of king Aretas, Philellen' (*Quellen*, 142-3). According to numismatic data, the Nabataeans coined until 72 BCE when their rule of Damascus was interrupted in 72 BCE by a successful siege led by the Armenian king Tigranes II.

Hellenized Nabataeans employed Greek as vehicular and official language or as a prestigious language, and this situation persisted until the Roman conquest of the *Nabataea* into the Greco-Roman culture.

Furthermore, it should be noted that, in the Near East, Greek monolingual inscriptions date back mostly to the Roman (1st c. BCE-4th c. CE) and Proto-Byzantine (4th-7th c. CE) period, even if there also exist a significant number of Hellenistic inscriptions (3rd-1st c. BCE).<sup>53</sup>

The Romans adopted the cultural conventions of the Greeks and Koine continued to be the primary idiom of the Greco-Roman east; in addition, Roman officials were often: "not only code-switching between Latin and Greek but also being diglossic in their knowledge of Greek, using a High Attic form and a Low vernacular".<sup>54</sup> This period marked a linguistic and social transition during which the Nabataeans definitely entered into the Greek linguistic sphere.

During the reign of Herod the Great (40 BCE-4 CE), the Greek language prevailed over the other Semitic languages. Herod was educated in Greek language, philosophy and culture, so he imposed Greco-Roman culture throughout his Hellenized territories. Even though he pretended to be Jewish, his policy was to impose Greek on all the strata of society, with a significant decline in Semitic languages, and above all of Hebrew.<sup>55</sup>

During the late antique period, the importance of Greek was different in Mesopotamia and Syria, as compared to the Nabataean Realm.56 In Dura Europos, where no native variety of Aramaic is attested, Greek was apparently dominant, probably in public life. In Palmyra, Greek coexisted, as a written language, with Palmyrene Aramaic, and in Edessa its social usage is clearly evident, with bilingualism becoming more visible after the first records of Classical Syriac.<sup>57</sup> Even though in Dura Europos and Edessa there exists no clear proof of a diglossic or bilingual situation, we do know that Palmyra was strongly influenced by Hellenistic culture. Some twothousand inscriptions written in the local Aramaic of Palmyra and accompanied by a Greek and/or Latin parallel text have emerged to date. This suggests that Greek was neither deliberately relegated to an informal language, nor did it take over from Aramaic as an offi-


<sup>53</sup> The *corpora* of *IGLS* represent a systematic collection of Greek and Latin inscriptions from the Ancient Near East.

<sup>54</sup> Porter 2016, 210.

<sup>57</sup> Gzella 2015, 247. In Edessa the former presence of Greek culture and language can be deduced from Edessa's foundation as a Seleucid colony and from some lexical loans, but all found inscriptions are monolingual and do not reflect Greek syntactic interferences or do not encompass Greek expressions.

cial language, but instead: "it was integrated into a more complex multilingual environment".<sup>58</sup>

It is also difficult to establish whether local varieties of Aramaic in Mesopotamia, especially at Assur, Hatra and their surroundings, were spoken alongside other languages. Greek seems to be less present and restricted to a couple of lexemes concerning economics and Hellenistic architecture with no syntactic interference in Aramaic.<sup>59</sup>

It is intriguing that the first bilingual Nabataean Aramaic-Greek inscriptions were found outside the linguistic borders of *Nabataea* and date back to the 1st century BCE; in particular the oldest inscriptions date back to 9 BCE They were found in Miletus and in Delos<sup>60</sup> and were commissioned by Sylleus during his journey to Rome. Another inscription, found in Sidon dates back to 4 BCE<sup>61</sup> and was probably written by a Nabataean trader in honour of his god Dūšarā. Therefore, the first signs of bilingual inscriptions are found outside Petra and the Nabataean Kingdom, and this suggests that in the 1st century BCE the Nabataeans did not yet use Greek alongside Aramaic; they were neither bilingual, nor, probably, diglossic. In fact, the inscriptions from Delos and Miletus reveal that Sylleus only wanted a simple Greek translation, of the Nabataean text, to leave in a Greekspeaking region. In addition, Littmann (*PPAES* IVA, XV-XVI) gives the story of the inscription from Miletus on the basis of the shapes of the Nabataean engraved letters. In fact, the Nabataean script is cursive and tends to ligature, as compared to the Greek text carved in beautiful and regular letters. It can therefore be surmised that Sylleus arrived at Miletus carrying a Nabataean handwritten copy of the text, drafted by himself on a papyrus or on a parchment. Later, he delivered it to a Greek mason who translated the text, but he did not know Nabataean Aramaic and so he exactly copied the Nabataean part.

The rest of the bilingual inscriptions, carved in the Nabataean regions, date from the 1st century until the 2nd-3rd century CE. Whereas in Palmyra the population appears to have been bilingual, in Petra and in the rest of the Nabataean kingdom the situation was quite different. Although in some inscriptions the Aramaic text is almost of the same length of the Greek, in other bilingual inscriptions the Nabataean part provides more information than the Greek, which is only a summary of the Nabataean text. As regards, Mountain Sinai, the graffiti (dating back to the 2nd-3rd century CE) consist in short and fragmentary texts. They are mostly dedicatory and funer-


<sup>58</sup> Gzella 2015, 249. See also Gzella 2005, 445-58. For a *corpus* of Palmyrene Aramaic inscriptions see *PAT*.

<sup>59</sup> Gzella 2015, 275.

ary inscriptions and provide no noteworthy evidence of the alleged bilingualism of the Nabataeans. The majority of the graffiti are only carved in Nabataean Aramaic, there remaining a mere fourteen examples of inscription engraved in Nabataean Aramaic and Greek.

The bilingual epigraph from al-Ruwāfah, north of the Arabian Peninsula,<sup>62</sup> shows the usage of Greek in honorific and historical contexts; the epigraph was erected by the tribe of the Thamud using Nabataean and Greek that: "would thus both serve as prestige languages for representational purposes among speakers of Old Arabic and Ancient North Arabian vernaculars".<sup>63</sup> So, Nabataean and Greek were a sort of combined *lingua franca* for the North Arabian people.

The bilingual inscriptions show us that the texts were written independently, probably following the same content, but not translating from one language to another. Furthermore, the Greek linguistic influence on Nabataean Aramaic is reflected in a handful of loanwords referring to architecture.

Generally, these are funerary and votive inscriptions, in spite of being short and often fragmentary. They consist in burial stones that often refer to the possessor of the tomb through the sentence *dnh mqbr'*/*npš'* 'this is the tomb', or simple graffiti that record the passage or the death of somebody through the common formula *dkrt* = Gr. Μνησθῇ 'let be remembered'.

The texts exhibit different patterns of content following the typical stylistic tradition of the two languages. So, there are distinct versions of the same content of an inscription within a multilingual environment. Moreover, only nine inscriptions are 'really' bilingual (nos. 9, 16, 19, 22, 25, 32, 33, 37, 39), even if in nos. 32 and 33 the Nabataean text reports the initial formula *šlm* 'peace' and the closing formula *bṭb* 'in good' (in no. 33) and no. 9 only reports the same personal name in both languages.

In the rest of the epigraphs, elements of the texts are distinct and in two cases the Nabataean and the Greek versions are totally different in content (nos. 29 and 34).

Two texts are exclusively in Greek, including a series of Nabataean letters (no. 12) and a Nabataean personal name (no. 18). Conversely, only one inscription is entirely in Nabataean (no. 20), with the exception of a Greek personal name.

As regards the different patterns of the epigraphic habits, in no. 26 the Nabataean opening formula is 'this is the statue of…', whereas the Greek one is 'the people (or council) of… honoured', both reflecting the West Semitic and Greek traditions. Furthermore, nos. 10 and 26 report the Hellenistic expression 'out of affection' and 'of his

<sup>62</sup> No. 47.

<sup>63</sup> Gzella 2015, 242.

piety', whereas some inscriptions contain a more extensive genealogy in Nabataean like in nos. 11 and 28. Only three inscriptions give more information in Greek (nos. 13, 21, 47).

Two epigraphs are not bilingual, but instead contain two completely different texts (nos. 29 and 34) and, curiously, three inscriptions report different personal names (nos. 10, 37 and 44).

Considering the content and the small number of the bilingual Nabataean-Greek inscriptions, we may assume that direct evidence for bilingualism is rare in *Nabataea*. In the first-century Near East there was a complex multilingualism among the various peoples, who used varieties of languages such as Aramaic, Hebrew, Greek and in some strata, Latin.

The discovery of such a large number of unilingual inscriptions written in Greek and in Nabataean Aramaic suggests a sociolinguistic environment in which Koine Greek was used in a diglossic situation as a prestige language, whereas Nabataean, as the ethnic language of the conquered, was used for personal purposes. These unilingual Greek inscriptions were functionally communicative for the Nabataean population.

Some bilingual inscriptions were engraved due to the writer and the place written, the inscriptions out of *Nabataea* are cases in point. Other epigraphs reflect ethnic or religious traditions, regardless of knowledge of Greek. The addition of further information, such as a more extensive genealogy, to the Nabataean version rather than in the Greek is a typical mark of the Nabataean epigraphic habit in which the writer wanted to highlight his Semitic-Aramaic tradition.

In conclusion, we may assert that Greek was not solely the *lingua franca* or prestige language of the Nabataeans, but, as epigraphic evidence would suggest, that it was probably also the vernacular employed by some social strata of the population. So, the Nabataean-Greek bilingual inscriptions are apparently not bilingual, but rather multilingual texts carved within a diglossic linguistic situation among the Nabataeans. Nabataean was the L1 of the indigenous population and possibly even the primary idiom for some in the lower social stratum, whereas Greek was used as the L2 in lower social contexts and as an administrative language.

### 3 Nabataean Aramaic-Greek Inscriptions

#### **3.1 Research Background**

The presence of *corpora* of Semitic inscriptions, and in particular of Aramaic inscriptions,<sup>64</sup> is indicative of the exponential expansion in epigraphic studies in the field of the Near Eastern philology. The interest in Greek epigraphy in collecting Greek inscriptions found in the Ancient Near East65 conveys the desire to further study and analyse the linguistic contacts between the Classical and Semitic worlds.

The Eastern Mediterranean, during the period between the end of the 7th century BCE and the 5th century CE, saw intense cultural and commercial exchanges between the Syro-Palestinian territories and the Aegean area. The contacts between Semites (above all the Canaanites, the Phoenicians and the Syro-Palestinians) and Greek speakers increased in North Africa, Rhodes, Kos, in the rest of the Central Aegean islands, Crete and Greece up to southern Italy and Sicily.

In the field of epigraphy, this intricate network of relationships triggered the intriguing linguistic phenomenon of these bilingual inscriptions; epigraphs written in Greek and in Semitic languages record a contact between various cultures, especially along borders and in prolonged contact areas.

The aim of this study is to collect the bilingual inscriptions carved in Nabataean Aramaic and Greek. Although around 6,000 Nabataean inscriptions, dating to the period between the 2nd c. BCE and the 4th c. CE, have so far been discovered, a complete *corpus*, comprising all Nabataean inscriptions, has yet to be compiled.<sup>66</sup> A wide selection of texts from the entire Nabataean region is recorded in *Quellen* (see bibliography), and a number are also included in Yardeni 2000,<sup>67</sup> while a great

65 Along with the monumental works of the *Inscriptiones Graecae* (*IG*), designed as a continuation of the *Corpus Inscriptionum Graecarum* (*CIG*), which collect all Europe's ancient Greek inscriptions in 49 volumes, it is worth mentioning the great project *Les inscriptions grecques et latines de la Syrie* (*IGLS*) which brought together Greek and Latin inscriptions mainly from Syria, Jordan and Lebanon in 21 volumes. Furthermore, a useful online database (https://inscriptions.packhum.org/), constantly updated, consisting in all the Greek inscriptions of the Mediterranean area, including the Greater Syria and the East is to be found in the *Searchable Greek Inscriptions of The Packard Humanities Institute* (PHI) – Project Centers at Cornell University & Ohio State University.

66 Gzella 2015, 239.

67 See Beyer 2004, 23 for additions.

<sup>64</sup> For Old Aramaic and Official Aramaic, see for instance Gibson 1975; *KAI*, 201-79, 309-20; Schwiderski 2008; Porten, Yardeni 1986-99; Beyer 1984, 29-32 and 1986, 15- 16 including a supplement in 2004, 17. Apart from Nabataean, for the varieties of Aramaic in the Hellenistic and Early Roman period, see e.g. Magen et al. 2004 for inscriptions from Mount Gerizim; the series *Discoveries in the Judaean Desert* (*DJD*, 1955-) and Beyer 1984 and 2004, including Y*TDJD* and Yadin et al. 2002, for Aramaic of the Qumran scrolls; *PAT* for Palmyrene; Beyer 1998 for Hatraean.

many inscriptions remain unpublished or partially described. The *corpus* of M.E. Stone (in *RIGP*) includes the graffiti found at Mount Sinai, one of the regions in which the greatest number of Nabataean inscriptions have been discovered. A *corpus* including the bilingual Nabataean-Greek inscriptions, found in the Eastern Mediterranean and in the Near East dating back to the period from the 2nd century BCE to the 3rd-4th century CE, has also yet to be compiled. For this reason, the assembly of the above-mentioned bilingual texts could deepen our understanding of the morphological, syntactic and lexical aspects of the two languages in question and shed light on the cultural, social, political and religious relationships between the Nabataeans and the Greeks.

This collection is not to be considered as complete, and we hope it will be expanded by further discoveries of bilingual epigraphs.

This work represents the research that I began in my PhD dissertation that also includes a brief history of the Nabataean kingdom and, above all, a close examination of the Nabataean onomastics, and Nabataean names transliterated in Greek, performed with a view to reconstructing the phonological system of Nabataean.<sup>68</sup>

#### **3.2 The Numbering and Nature of the Inscriptions**

This corpus is made up of 51 bilingual inscriptions. The epigraphs are collected following a geographical order, and the numbering of the inscriptions reflects this same pattern; the first assembled epigraphs come from the main region in which the Nabataean society flourished, i.e. the territory of Jordan, around the capital city of Petra, and they have the numbers from 1 to 17. Those hat follow are from Syria and, more specifically, from Hauran (nos. 18-30). Other inscriptions were found in Egypt, two in Safājā, while another epigraph is located in the vicinity of a further station on the road that leads from Qifṭ to al-Quṣayr al-Qadīm; the remaining inscriptions, found in Egypt, were discovered at Mount Sinai (nos. 31-46). Subsequently, we come to the sole and longest bilingual inscription unearthed in Saudi Arabia, at al-Ruwāfah (no. 47). Outside the borders of the Nabataean kingdom we find an epigraph from Sidon, in Lebanon, (no. 48), and three inscriptions from the Aegean Sea: one found at Miletus, Turkey (no. 49) and two on the Greek islands of Delos and Kos (nos. 50-51).

The large geographical area across which the inscriptions were found suggests the presence of Nabataean traders on the caravan routes that led to Egypt in the west, passing through Mount Sinai, and to the Aegean Sea in the north-west.

<sup>68</sup> For a more in-depth the reconstruction of the Nabataean Aramaic phonological system, see Petrantoni 2020.

#### Editorial conventions


As regards the forms of the Nabataean personal names given in translations, they reflect my own close examinations and studies of these during my research<sup>69</sup> in the course of which I attempted to cite them in a vocalized form. In addition, the pronunciation of the names of Arabic origin is rendered as the Arabic form suggests.

The Nabataean graph *š* is here transliterated as *š* even though in some names of Arabic origin, the same letter can be read /ś/. As concerns the six plosive consonants *b*, *g*, *d*, *k*, *p*, *t* they are pronounced with spirantisation following a vowel like in the Biblical Hebrew and in the other varieties of Aramaic;70 but since it is only a phonetic phenomenon, here the spirantisation will not be marked and only the graphematic transcription will be provided.

### Transcription

For the transliteration of the varieties of epigraphic Aramaic and Hebrew we use the transliteration adopted by *SBL Handbook*, 26. For the transcription of Biblical Aramaic we follow Rosenthal (2006, 11 for the consonants, and 14-6 for the vowels).

Syriac Aramaic is transcribed by the ancient and classical variety of *'Esṭrangēlā* script following *SBL Handbook*, 26 for the consonants, whereas the vowels are not written, but they are marked in the transcription using the East Syriac vocalic system. The transcription of Syriac Aramaic vowels and fricative consonants follows that of Muraoka (2005, 4-7).

The Romanization of Arabic is based on The Hans Wehr transliteration system (Wehr 1976, VIII-XV).

For the transcription of Greek we prefer to report the words of the inscriptions without accents and breathing marks. Therefore, when a Greek term occurs in the comment, footnotes and indices, it will be written precisely with accents and breathing marks.

69 Petrantoni 2020.

<sup>70</sup> Spirantisation occurred during the earliest stages of Aramaic and was stabilized as a consonantal feature in Classical Aramaic. Rosenthal (2006, 17, § 15) points out that this feature began to appear from the 6th century BCE

Giuseppe Petrantoni

# Jordan

### **Petra**

**1.** The inscription is located in Wādī Mūsā, near to the Sīq of Petra, in front of the Obelisk Tomb.<sup>1</sup> It is a text, approximately 3,50 metres in length, deeply carved in a sandstone that is perpendicular to the stream, 5 m above the ground and placed at the opening of the cave room BS23;<sup>2</sup> it is written in large letters, which were perhaps originally ochre in colour (Milik 1976, 143). Unfortunately, the text has been mostly ruined. The text consists of five lines, the first three in Nabataean, the other two in Greek. (Plate I, no. 1)

```
Dimensions length 3.50 m 
Dating 1st CE (40-70 CE?)
Bibliography Milik 1976, 143-52; SEG 27 no. 1012; Milik 1980, 12, fig. 9; Zay-
   adine 1984, 64-5; Zayadine 1986, 221-2; Healey 1993, 243-4; IGLS 21,4 no. 
   54; Quellen, 222-4; Atlas, MP5
```
1 The town, in ancient times known as Gaia, is located in the Ma'ān Governorate in southern Jordan. It is called Wādī Mūsā probably because according to tradition the prophet Moses passed through the valley and struck water from the rock for his followers (Num 20,10-11). The Nabataeans built channels that carried water from this spring to the city of Petra.

2 *Atlas*, MP5, 164.

#### Text<sup>3</sup> and translation

1) *mqb[r'] (dn)h b(n)[h 'bdmnkw br] (')kys br*

2) *šly [br] (')t(y)h[w---lnpšh] (w)['] (ḥ)rh w'ḥr*

3) *hm l'[l]m 'l(m)[yn šnt---) l(m)nkw bḥy(why)*

4) Αβδομα[νχ]ος [Αχ]αιου [επ]οιησ[ε]

5) [μ]νημειο[ν εαυτω και υι]οι[ς]

1) (Thi)s burial-monume[nt] bui(l)[t 'Aḇdomankō son of] ('A)ḵayos son of

2) Šollē [son of] ('A)ta(y)h[ō---for himself] (and) his [d](escen) dants and their descendants

3) for [e]ver and ev(e)[r in the year--] of (M)ankō<sup>4</sup> during his lifetim(e)

4) Abdoman[nch]os son of [Ach]aios [m]ad[e]

5) this [f]uneral monumen[t for himself and for his ch]ildre[n]

# Commentary

Line 1. The term *mqb[r']* is attested in inscriptions from Sī', in Ḥauran (*RES* no. 805; *PPAES* IVA no. 2), although it is rarely used in comparison to *mqbrt'* or *qbr'* (Milik 1976, 144; *DNWSI*, 678). The masculine form of the noun means 'burial, tomb'.<sup>5</sup>

The noun *mqbr'* possibly identifies a place, a setting, *strictu sensu*, suitable for the burial, i.e. a grave instead of a complex of tombs; in addition, the expression, with the demonstrative *dnh* 'this',<sup>6</sup> could prove that the monument was located not far from the place in which it was engraved.<sup>7</sup>

After the verb *bnh* 'to build, to construct, to erect' there should be the name of the author, but only the Greek text reports it, that is *'bdmnkw* = Αβδομανχος as suggested by Milik,<sup>8</sup> rather than a possible *'bd'mnw* = Αβδομανος, 9 with the latter being too short for our in-



<sup>3</sup> From hereon the Greek text will be presented without *spiriti* and accents.

<sup>6</sup> This regards a common custom in Nabataean epigraphy; the demonstrative placed before the subject assumes the sense of subject pronoun of a nominal clause, while put after it becomes a demonstrative adjective (Cantineau 1930-32, 2: 58). Such a construction, with a postponed demonstrative, is also attested in two inscriptions from Umm al-Jimāl. Cf. *RES* no. 2064; *PPAES* IVA no. 42.

scription.<sup>10</sup> This personal name is made up of *'bd* 'servant, slave' and *mnkw* (= *mlkw*) 'Malichos' (the name of a Nabataean king) and it occurs other 16 times in the Nabataean onomastics.11 It is worth bearing in mind that the form *mnkw* (= *mlkw*) reflects the phonetic transition *l* > *n*, a typical Nabataean feature observed in many names.<sup>12</sup> The epitaph might have been carved during the reign of Malichus II (39/40-69/70 CE), even though among the Nabataeans the deification of kings was only witnessed by Obodas I (96-85 BCE).13 King or *pseudo* god names as '*bdmnkw*, *'bdḥrtt* or *'bdrb'l* would be bestowed upon sons of the king; as a consequence, if we assume that the inscription was drawn up during the reign of Malichus I, it would mean that the Greek linguistic influence occurred in Nabataea starting from the middle of the 1st century BCE<sup>14</sup>

At the end of the line *'kys* = Gr. Αχαῖος appears. This name is recorded 4 times.<sup>15</sup> It is a Greek name having the shape Αχις, Αχιος<sup>16</sup> and Ἀχαιός. 17

Line 2. At the beginning there is the name of the author's grandfather, *šly*. The root should come from Arabic *salā* 'to neglet, to forget', also 'to console, to comfort'.<sup>18</sup>

This name is widely used in Nabataean19 although it is not reported in the Greek part of the inscription. In Greek it has the form Συλλαιος, 20 in the bilingual inscriptions from Miletus (no. 49) and Delos (no. 50). The father's name of *šly* is *'tyhw*. <sup>21</sup> If the reading is correct we are dealing with a rare name in Nabataean onomastics; it may be an adjective of *pe'il* form (productive in Arabic) from Arabic *'utiha*  'to become stupid', 'idiot, dumb' > *'atīh*. It is an epithet that indicates "a man addicted to annoying another and mimicking his speech"<sup>22</sup> or


16 Negev 1991, 81. In *Quellen*, 222-3 we read Ἀκις, a name also found in Egypt (*SB* I no. 3311).

17 Wuthnow 1930, 30. Αχαιος is more frequent (Pape 1911, 184).

18 Lane, 1417.

19 Cantineau 1930-32, 2: 150; al-Khraysheh 1986, 174; Negev 1991, no. 1137. The name is also present in Hatraean *šly* (Beyer 1998, H 20).

20 Wuthnow 1930, 113. Another form is Σολλου (Wuthnow 1930, 170).

21 Negev (1991, no. 949) suggests that it is a diminutive form < *'thw*.

22 Lane, 1951.

better "intrigant, qui rapporte les paroles des autres dans le but de nuire" as Kazimirski (1860, 169) points out. On the contrary, Milik (1976, 147)<sup>23</sup> vocalizes and reads *'utēh* (< *fu'ayl* in which there is the diphthong *ay* > *ē*).

At the end of the line we read *'ḥrh* 'his descendants', 'qui est après, postérité'<sup>24</sup> with the singular masculine pronoun -*h*. Probably it is of Lihyan origin or a loanword (cf. *CIS* II no. 197), even though in Lihyan the substantive is employed along with words bearing a more detached meaning.<sup>25</sup> In Nabataean this meaning, expressed by the root *'ḥr* 'to come after, to follow, to be late', is more general. After that, there is *w'ḥr* and at the beginning of line 4 we encounter the plural masculine suffixed pronoun -*hm* meaning 'their descendants'. In Nabataean epigraphy it is rather unusual to break graphically a syntagma into two parts.<sup>26</sup>

Line 3. After the suffixed pronoun we read the expression *l'lm llmyn*, literally 'for the eternity of the centuries', therefore 'for ever and ever', 'in *saecula saeculorum*'. In the middle of the line the reading is difficult. We may reconstruct a possible *šnt* 'year' that generally goes with the name of a king in order to date the carving of the inscription. Indeed, at the end of the line we find the name *lmnkw*, referring to the Nabataean king Malichus I or II, while the absence of the title: *mlk' mlk nbtw* 'the king of the Nabataean's kings' is unusual, because it commonly follows the sovereign's name.<sup>27</sup>

At the end of the line the phrase *bḥywhy* comes into view meaning 'during the course of his life'. After the substantive *ḥyw* there is the masculine singular possessive pronoun *-hy*. In Nabataean, as in Biblical Aramaic, Egyptian Aramaic and Syriac, the singular masculine suffixed pronouns -*h* and -*hy* (in Nab. -*w* and -*hw* are also used) differ in use; -*hy* is generally employed before nouns ending in -*w* or -*y*. <sup>28</sup> The origin of the two suffixes is difficult to establish.<sup>29</sup>

Line 4. The letters are in part ruined, but the reading does not seem to be difficult to reconstruct. The name Αβδομανχος and the patronymic Αχαιου appear, 'Abdomanchos (son of) Achaios'. At the end of the line we see the sequence -οιης- that we interpret as [επ]οιησ[ε] 'he built'.

23 *Contra Quellen*, 222-3 and Nehmé (*Atlas*, MP5, 164).


27 In the matter of Malichus, his title is also absent in another inscription, in *CIS* II no. 222, in which we read: *bšnt 17 lmnkw* 'in the year 17 of Malichus', quite akin to the reconstruction of the sentence contained in our epigraph.

28 Cantineau 1930-32, 2: 54-5.

29 See Brockelman 1908, 312 for an explanation.

Line 5. The articles τό and τοῖς, of the nouns μνημεῖον and υἱός, are omitted by the stone-cutter (due to the insufficient space?<sup>30</sup>). As regards the former, the term comes from the verb μιμνῄσκω 'to remind, to put in mind'<sup>31</sup> < μνήμη (Dor. μνάμα) 'memory, souvenir', but above all 'grave, burial',<sup>32</sup> just as μνημεῖον (Ion. μνημήιον) indicating the 'memorial', the 'grave'.<sup>33</sup> Moreover, the substantive would be a diminutive observed in burial memorials in order to highlight the importance of μνῆμα and certain derivatives in the funeral lexicon.34 According to Healey (1993, 243), the term should be reconstructed as μνημóσυνον 'remembrance, memorial' also referring to a commemorative funeral memorial, that is less important than a grave.

**2.** The inscription is carved on a white marble and was found in a robbed tomb. The epigraph, which is an incised graffito, consists of two lines. In the first line, in which two Nabataean words appear, there seems to be a fracture in the stone, while the second line is not damaged.<sup>35</sup> (Plate I, no. 2)

**Dating** unknown **Bibliography** Bowersock 2015, 123-4

Text and translation

1) *[h]grw slyt'* 2) Αγαρη

1) [H]aḡarō *slyt'* 2) Agare

30 As Milik suggests (1976, 147).

31 *LSJ*, 1135.

32 *LSJ*, 1139. Cf. Euripides, *Phoenissae* 1585: ὡς ὥρα τάφου μνήμην τίθεσθαι 'it is time we thought of their burial'.

33 Cf. Plato, *Res publica*, 414a: τάφων τε καὶ τῶν ἄλλων μνημείων μέγιστα γέρα 'the supreme honours of burial-rites and other memorials'.

34 Chantraine 1968, 703.

35 Bowersock (2015, 123) asserts that: "the stone looks abraded, and […] there was a second word of Greek to match the second word of Nabataean".

### Commentary

The graffito reports a woman's name, *hgrw*, which is well-known in tNabataean onomastics and considered as equivalent to the Greek Αγαρη. <sup>36</sup> This Greek form is also well-documented.<sup>37</sup>

The second Nabataean word, *slyt'*, is problematic because the first two letters present difficulties in reading; if they were *šl-* the word might represent another form of the name *šly*. <sup>38</sup> John Healey proposes reading *ṭlyt'* 'the girl',39 however his hypothesis is rejected by Bowersock who asserts that there is no reason: "why a female name would need to be qualified in this way".<sup>40</sup> As regards its etymology probably *slyt'* derives from the Arabic root *slw* 'to neglet, to forget', also 'to console, to comfort'. *Hgrw* may be described as a consolation or a comfort. But this is only a suggestion since, unfortunately, we do not have a parallel Greek word after the mention of Αγαρη. Therefore, it seems to be possible that whoever carved the graffito preferred to record in Greek only the name of the woman without writing anything else.

**3.** This stone was found among the ruins of Petra. The stele is damaged on the left part, while the right side seems to be well-preserved. The six-line text, is less legible on the left side and we can only reconstruct a few lines. It is mainly in Greek except for the last two lines where some illegible signs in Nabataean script appear. (Plate II, no. 3)

**Dimensions** height 41 cm, length 12.5 cm (inscribed surface 33.5 cm × 12.5 cm) **Dating** unknown **Bibliography** *IGLS* 21,4 no. 28

Text and translation

1) [---](ε)πηκοω Δου-2) [-σαρι ---]ς Ολφιος 3) [---]ων[..ε]κ των 4) [---]τω 5)-6) difficult reading. Only Nabataean signs.

37 Through three inscriptions from Arabia (in *IGLS* 13,1 no. 9315, *IGLS* 21,5.1 no. 157, *IGLS* 21,5.1 no. 158) and two from Hauran (in *PPAES* IIIA 3 no. 519; 5 no. 755).

38 Negev 1991, no. 769.

40 Bowersock 2015, 124.

<sup>36</sup> Cantineau 1930-32, 2: 84; al-Khraysheh 1986, 61-2; Negev 1991, no. 293; Healey 1993, H 13, 14, 30. Cf. also Pape (1911, 8) who records Ἄγαρ and Wuthnow 1930, 11. In Palmyrene we find *hgr* (*PAT*, 433 = Stark 1971, 14 and 84). See also *ICPAN*, 608.

<sup>39</sup> In a personal message in replying to Bowersock's questions about this Nabataean word.

1) [---l]istener Du-2) [-sares---]s Olfios 3) [---] of 4) [---].

#### Commentary

Line 1. The text is illegible at the beginning of the line; Sartre (*IGLS* 21,4 no. 28) reconstructs θεῷ ἁγίῳ 'to the Holy God'. After this expression there is a sequence of letters, [---](ε)πηκοω, that is simply interpreted as ἐπηκόῳ < ἐπήκοος 'listener, hearer' referring to the god who listens to the prayers and grants them. Although the reading is uncertain, at the end of the line we see the letters δου- that probably form the first part of the God's name Δουσαρι – *Dusares*.

Line 2. It is almost illegible up to the end where the name Ολφιος comes into view. We do not know if it refers to the author of the inscription; either way, the name probably derives from the Arabic root *ḫlf*. In Nabataean the name *ḥlpw*<sup>41</sup> is found and it is compared with Arabic *ḫalaf*<sup>42</sup> and Greek Αλαφος<sup>43</sup> as Cantineau points out (1930- 32, 2: 96). The general meaning is that of 'successor', also 'enfants'.<sup>44</sup>

The Greek restitution of the name highlights a probable syncope on the penultimate non-stressed vowel: *ḥalp* < \**ḥalap*. In addition, the Greek transcription would exhibit a final etymological -\**ī* represented by ι before the suffix -ος. 45

Lines 3-4. At the end of line 3 there is the plural genitive article τῶν, while in line 4 we reconstruct a supposed ἰδίων 'of own'.<sup>46</sup>

45 See other names of Arabic origin like Αδιος = Nab. *'dy* registered at Umm al-Jimāl (*PPAES* IIIA no. 366) and Μονιος = Nab. *mġny* found in Hauran (Wadd. no. 2153).

46 *IGLS* 21,4 no. 28.

<sup>41</sup> In Hauran (cf. *PPAES* IVA no. 19; *RES* no. 2048), in Hegra (modern Mada'in Saleh or al-Ḥiǧr) and Hejaz (*CIS* II nos. 206, 209; *JSNab* nos. 53, 297); Cf. also al-Khraysheh 1986, 84. As regards Palmyrene, see *PAT*, 434 = Stark 1971, 22-3 and 88.

<sup>42</sup> Cf. *ICPAN*, 198 *ḥlf* > *ḥlīf* 'ally, sworn friend'.

<sup>43</sup> This appellative is frequent in Syria, Phoenicia (*IGLS* 5 no. 2250) and Hauran (*PPAES* IIIA 2 nos. 90 and 185).

<sup>44</sup> Kazimirski 1860, 1: 620; Negev 1991, no. 452. It is also possible that the root is related to *hlypw* (al-Khraysheh 1986, 62), name found in Sinai (*CIS* II no. 2973), that Cantineau (1930-32, 2: 86) renders into Arabic as *hillawf* 'hirsute', *hallawf* 'sanglier'. In the Negev, during the Byzantine period, the name Αλφιου-Αλφειος is recorded (Negev 1982, 40, no. 39:2) as it is in the Nessana papyri (Kraemer 1958, 67, PC21) and in those from Egypt dating back to the 6th-7th century CE (Preisigke 1967, col. 21).

**4.** This inscription was found in the area of the monumental building of el-Deir and it was engraved inside the monument no. 465 on the south wall, to the right of the entrance, 1.30 metres above the ground and 1.90 metres away from the back wall. The epigraph is unrefined, presenting several complications in reading due to the fact that the wall is blackened by soot. The inscription is made up of five lines (if we also include the last two lines of the inscription)<sup>47</sup> in which the Nabataean section is not placed in line 3, as Brünnow e Domaszewski (1904, no. 437) graphically reproduced, but to the right of the Greek sequence of line 2.

The shape of the *aleph* suggests that the Nabataean script here used goes back to a late stage of writing so we can date the inscription to the 1st century CE.

```
Dimensions height of letters A, H, M 10.5 cm; letter O 3.5 cm
Dating 1st-2nd CE (?) 
Bibliography Brünnow, Domaszewski 1904, 335, no. 465 b; CIS II no. 437; 
   IGLS 21,4 no. 35
```
Text and translation


#### Commentary

Line 1. The term τὸ προσκύνημα < προσκυνέω 'to make obeisance to the gods or their images'<sup>48</sup> reveals the cultic character of the epigraph and probably constitutes a rare case of a *proskynema* found out of Egypt.<sup>49</sup>

<sup>47</sup> According to Sartre (*IGLS* 21,4 no. 35) they belong to another text whose further parts have disappeared or perhaps were never engraved at all.

<sup>48</sup> *LSJ*, 1518.

<sup>49</sup> The practice of προσκύνημα was born in Egypt towards the 2nd century BCE in a hellenized environment characterized by a religious syncretism where both Greek and

Line 2. We read ΑΜΙΨΜΙ, but the sequence of letters cannot be interpreted. On its right side there is the Nabataean text: *dkyr* […]*n' bṭb* in which the first element, *dkyr*, denotes the remembrance of someone (it is a common formula in the commemorative Nabataean epigraphy), although we cannot read the name of the remembered person. After a space, consisting of 3 letters, we glimpse a *l* preceding the *n* in order to obtain […]*ln'*, probably the final part of the name of who is remembered. At the end there is the expression of greetings *bṭb* 'in good'.

Line 3. Sartre (*IGLS* 21,4 no. 35) reads -ομου, omitted (or badly read?) by Domaszewski who connected it to the previous sequence ΑΜΙΨΜΙ.

Lines 4-5. The phrase τὸ προσκύνημα occurs again, while in the last line we read in Nabataean *šlm* 'peace'.

**5.** The inscription was found at Little Petra, also known as Sīq al-Bārid, and it was engraved on a rock near a temple that could have been an ancient shrine.<sup>50</sup>

**Dating** unknown **Bibliography** *CIS* II no. 480; Lagrange 1898, 180, no. 70

Text and translation

1) *šlm hbwls* 2) Αβολος 1) Peace! Habōlos

2) Abolos

Commentary

The name *hbwls* seems to be of Greek rather than Nabataean origin51 considering the -*s* ending that recalls the suffix of the nominative case -ος. Furthermore, Αβολος is attested in an inscription from Asia Minor;52 with two other forms, Ἀβóλλας and Ἄβολλα, being attested only by Pape.<sup>53</sup>

Egyptian divinities were associated. For a detailed study, cf. Geraci 1971.

51 al-Khraysheh 1986, 61.

52 Ιn Phrygia. Inscription no. 333, 92 in Haspels 1971.

53 Pape 1911, 3. In the volumes of the *LGPN* and in the database of the *Searchable Greek Inscriptions* of The Packard Humanities Institute the two names are not present.

<sup>50</sup> Lagrange 1898, 179.

# **Jabal Kharazah Ratamah**

**6.** The two inscriptions were found at Jabal Kharazah Ratamah, in the Wādī Ṯalājah, at a dozen of kilometres east of al-Qwayrah between Petra and Aqaba. The only fragment in Greek script is engraved on a cistern-facing rock; it carries the name of the owner, while a short distance from it there are the two Nabataean inscriptions.

In the area, in the clefts on the western side, five Nabataean dams have been discovered. Among the installations the southernmost is the best. The short Greek part reports only a name, Ηλεος, carved in carefully cut lettering and related to the owner of the cistern.

The two Nabataean texts are hammered and not cut into the rock. The first of the two lines, is situated 4.50 metres in front of the dam and 3.50 metres above the riverbed; the second line, badly weathered, is placed below the first one.<sup>54</sup>

```
Dimensions 1st Nabataean inscription 80 cm × 16 cm, height of letters 6 cm; 2nd 
inscription length 105 cm, height of letters 8 cm
Dating 1st CE (32 CE) 
Bibliography Kirkbride, Harding 1947, 19; Milik 1958, 249-51, no. 8; IGLS 21,4 
   no. 136; Farès-Drapeau, Zayadine 2001, 205-16; Quellen, 283-4
```
Text and translation

Inscription A Ηλεος

1) *lšb' br 'lh 'tyd šnt 'rb'yn wḥdh* 2) *lḥrtt mlk nbṭw rḥm 'mh yq'*

Inscription B 1) *dkyr twds br 'lh šlm*

Inscription A Eleos

1) To Šab'a son of 'Eleh, (this dam) was set up in the year 41 2) of Aretas, king of Nabataeans, who loves his people. Yiqā

Inscription B 1) Let be remembered Theudas, son of 'Eleh. Peace.

#### Commentary

#### **Inscription A**

Line 1. The name of the addressee of the manufact, *šb'*, appears denoting the 'lion'.<sup>55</sup> This term is present in Arabic *sab'* meaning 'wild beast, lion',<sup>56</sup> even though in North-West Semitic languages this root primarily means 'to be sated', also 'seven'.<sup>57</sup>

The second name that comes into view is *'lh* = Gr. Ηλεος, 58 the father of *šb'*; the Greek form furnishes the presumable Nabataean vocalization /'ēleh/, a variation of the same transcription that identifies /'īlah/ 'god'.<sup>59</sup>

Successively, we find the verb *'tyd* 'to be set up, to be arranged, to be prepared', the passive participle of *pe'il*, <sup>60</sup> that is also attested in Arabic *'atuda*  (= Heb. *'td*) 'to be ready, to be prepared' representing a secondary formation from *'adda* 'to count, to enumerate', hence the form *i'tadda* 'to consider, to believe, to evaluate' (> *ista'adda*), 'he considered, prepared himself'.<sup>61</sup>

Line 2. We encounter the name *ḥrtt* 'Aretas', Aretas IV Philopatris according to his epithet *mlk nbṭw rḥm 'mh* 'king of Nabataeans who loves his people'; in addition, the mention of the year of the reign of Aretas *'rb'yn wḥdh* '41' suggests that the epigraph was carved in 32 CE.

58 Another form is Ηλειας (Wuthnow 1930, 51). The other attested Greek shapes, Αλειος, Αλεου (in Wadd. nos. 2005, 2520), diverge compared with our inscription. Cf. also Pape 1911, 53.

59 *'lh* is recorded as a personal name in *PPAES* IVA nos. 13, 14; *RES* nos. 2043, 2044. Its meaning comes from Arabic *'alaha* 'to worship' (Negev 1991, no. 88). In *Quellen*, 283 it is read *Alih*. In Palmyrene *'lh* is considered as a hypocoristic name (*PAT*, 430 = Stark 1971, 4 and 68).

60 Cantineau 1930-32, 1: 75. See also *DNWSI*, 897.

61 Klein 1987, 490; Lane, 1969-70.

<sup>55</sup> It is also found in the Nabataean onomastics from Sinai in the form of *šb'w* (*CIS* II no. 891) and in Greek transcription, Σαβαος, in Hauran (Wadd. nos. 1990, 2101). It is also registered as a compound name *šb'[']lhy* (*CIS* II no. 370; *RES* no. 1472) like Arabic *sab'u llāhi* and Hebrew *ărî'ēl* (cf. Cantineau 1930-32, 2: 148). Cf. al-Khraysheh 1986, 170.

<sup>56</sup> Negev 1991, 1099. Cf. also *ICPAN*, 309.

<sup>57</sup> See *DNWSI*, 1101-2. It appears in the Bible as a personal name *šibə'â* (1Sam 20,1 and 1Chr 5,13). In Arabic this semantic passage from 'seven' to 'lion' is explained, according to tradition, by the fact that the root has assumed the meaning of 'place in which mankind shall be congregated on the day of resurrection'; this is related to the story of a shepherd who, while among his flock, saw a wolf that took a sheep, and ran after the wolf, rescuing his sheep. Whereupon the wolf said to the shepherd: *man lahā yawma l-sab'i?* 'Who will be for it (namely a sheep or a goat as defender) on the day of resurrection?' As Lane explains: "who shall be fοr it on the occasion of trials, when it shall be left to itself, without pastor, a spoil to the animals of prey" (Lane, 1296); as a result the animal of pray: "has a fang and tearing claw or canine tooth with which it attacks and seizes its prey as does the lion" (Farid 2006, 378). In the Koran we find the sentence *wa-mā 'akala l-sabu'u* 'what an animal of pray has eaten' (V, 4).

At the end of the line there is *yq'*, probably the name of the author. According to Milik (1958, 251) it is a hypocoristic name from the root *yqh-* (with the variant *wqh*), very frequent in the north and south Arabian onomastics of the 1st millennium BCE Although the root is less used in Hebrew, *yqh* 'to protect, to obey'<sup>62</sup> > proper name *Jāqeh* (Prov 30,1) < *yiqəhāh* 'obedience' (Gen 49,10; Prov 30,17), Akk. *utaqqu* 'to obey',<sup>63</sup> it is productive in Arabic in which the verb *waqā* 'to preserve' (*waqiya* 'to be obedient') is used, as the sentence *waqāhu llahu l-suū'* 'God preserved him from evil'<sup>64</sup> demonstrates. The name could be translated as 'pious'.

# **Inscription B**

In the last short one-line inscription there is *twds*, the name of the other son of 'Eleh, that derives from Greek Θευδᾶς (Θουδᾶς).<sup>65</sup> The fact that a Nabataean bears a Greek name would represent the proof of the spread of Hellenization, at least superficially, into the Nabataean military centres during the reign of Aretas IV.

### **Ḥismā-Wādī Ram**

**7.** It is a painting on a plaster that covered the inner wall of the sanctuary of the goddess Allat situated in Wādī Ram. Wādī Ram, also known as The Valley of the Moon, is a valley cut into the sandstone and granite rock in southern Jordan 60 km to the east of Aqaba. There are about 30,000 inscriptions in the form of rock paintings and graffiti realized first by the Thamud and then by the Nabataeans who installed in Wādī Ram in the 4th century BCE They lived peacefully along with the Thamud worshipping the same deities, including Dūšarā. Besides paintings and graffiti, the Nabataeans also built temples. The Greek text is located on the left part of the inscription, while the Nabataean is on the right side.


65 See *LGPN* I, s.v., "Θουδᾶς" is only encountered in Crete (*IC* II, 46) and Delos (*IG*  XI 2 no. 203 A:65). It is a name recorded in the Near East as the New Testament shows, mentioning a certain Israelite Theudas (Acts 5,36), the leader of a revolt who was subsequently killed; also Josephus (*AJ* 20.5.1) cites the figure of an instigator (Gr. γόης 'imposter, swindler, charlatan') who led the masses against the Ancient Roman *eques* and the procurator of Iudaea Province, Cuspius Fadus, but afterward he was captured and decapitated. The Acts speak about an event that happened before 37 CE, while according to Josephus it took place at the end of the mandate of Fadus, between 45-46 CE.

<sup>62</sup> *DGes*, 487; *KAHAL*, 224 'vorsichtig', 'Gehorsam'.

**Dating** 2nd CE (?) **Bibliography** Savignac, Horsfield 1935, 263-4, no. 1 and p. 265, fig. 19; *SEG* 8 no. 348; Milik 1976, 145, no. 5; *IGLS* 21,4 no. 141; *Quellen*, 289-91

# Text and translation<sup>66</sup>

1) Μνησθη Ουαβαλας ο κα[ι] 2) Αβδομαν[ο]ς Αβδομανου 3) του Αιαλο[υ Φ]αινησιος α(ρχ)[ιτεκτων] 4) *[dkrt ']lt[w w]hb'lhy dy mtqr' 'bd'mnw br 'bd'mnw*  5) *[br] 'ylw [br] 'bd'bdt br qynw pyny bny'* 1) Let be remembered Ouaballas called als[o] 2) Abdoman[o]s (son of) Abdomanos 3) (son of) Aialo[s from F]aino a(rc)[hitect] 4) [remember ']Allat[ W]ahb'āllahā nicknamed 'Aḇd'omanō son of 'Aḇd'omanō 5) [son of] 'Ayalō [son of] 'Aḇd'oḇoḏaṯ son of Qaynō from Faino. The builder.

# Commentary

The author of the painting, *'bd'mnw*, <sup>67</sup> is known in a *proskynema* found near the sanctuary, in which he is mentioned along with the names of other builders who made the sanctuary.<sup>68</sup> According to Milik (1976, 145) he is the same architect who engraved his name in a Thamudene graffito found in the same region in the form of *'bd'mn*; <sup>69</sup> moreover, this name would be the only proof in the Thamudic epigraphy.<sup>70</sup>

As regards the onomastics, the real name of the author of the painting is Ουαβαλας, <sup>71</sup> in line 1, while the nickname, pronounced ὁ κα[ὶ] (= Nab. *mtqr'*, sing. masc. part. of *etpe'el* 'named, nicknamed' *< qr'*), is the above-mentioned *'bd'mnw*. The Greek transcription of Ουαβαλας recalls the Nabataean name in line 4, *whb'lhy*, that is a

67 Cf. Negev 1991, no. 817.

68 The inscription was discovered by R. Savignac (1933, 418, no. 9). Cf. also the inscription no. 8.


71 This name, transcribed in Greek in this way, is found at Umm al-Jimāl (cf. *IG-LS* 21,5.1 nos. 423, 424 and *PPAES* IIIA, 3 no. 476). Other forms are: Ουαβαλλας, Ουαβαλλος, Ουαβελου, Ουαβηλος (Wuthnow 1930, 91).

<sup>66</sup> The transcription is that of Milik 1976, 145, no. 5.

theophoric composed of *whb* 'gift, present' and the name of the deity.<sup>72</sup>

The name Αβδομανος, <sup>73</sup> which corresponds to *'bd'mnw*, is made up of *'bd* 'servant, slave' and the root *'mnw*, and it was recorded in Sinai, meaning 'rester dans (un lieu)?',<sup>74</sup> Ar. *'amana* 'to remain, to stay in a place'.75 The Greek shape of the name probably shows the assimilation of the Aramaic nominative mark -o (=\**u*) into the second element of the name -ομανος; this fact identifies the shape of Αβδομανος as a result of a contemporary transcription compared with the most archaic Αβδοομανου recorded at Ḏībān,<sup>76</sup> in which the second -o is the nominative mark: \*/'aḇdo-'oman/. The name *'bd'mnw* is well documented in Hauran.<sup>77</sup>

The grandfather of Ουαβαλας is a certain Αιαλος who appears in line 3 in the Greek section of the inscription. The equivalent in Nabataean is *'ylw* mentioned in the last line.78 The root is of semitic origin having a correspondence to the Hebrew *'îlay*, a personal name of one of David's heroes who gave him strong support in his kingdom.79 It means 'the higher' < *'l'*, Ar. *'ālin* < *'aliya* 'to be high'. Cantineau (1930-32, 2: 129) connected the Nabataean root to the Arabic *ġayyāl* or *'ayyāl*. <sup>80</sup> The former indicates a wide, ample space,<sup>81</sup> while in the Koranic context it describes the following: "such as one judges to be of little extent, through it is for extending";<sup>82</sup> the latter takes on the meaning of: "that inclines from side to side in gait and is proud, haughty or self-conceited, therein".<sup>83</sup>

72 Negev 1991, nos. 337, 338; al-Khraysheh 1986, 67.

74 Cantineau 1930-32, 2: 132; *ICPAN*, 441: *'mn* 'to remain'.

75 The hypothesis of Littmann (*PPAES* IVA, 79) according to which the second element *'mnw* is identified with the Egyptian god Amun is hard to support.



<sup>73</sup> It is used in Greek as Αβδομανος (Cantineau 1930-32, 2: 126; Wuthnow 1930, 8 and 154); it still survived in the Negev during the Late Roman Period, as a Greek inscription found in the northern part of the acropolis of the city of Avdat/Oboda shows (Negev 1982, 17, no. 3).

<sup>80</sup> Cf. also Negev 1991, no. 882. al-Khraysheh (1986, 139) thinks it is *Ġaiyāl* 'killer', 'Menchelmörder'. In addition, see *ICPAN*, 451.

The Nabataean text provides the most information about Αιαλος, the son of *'bd'bdt*, the latter omitted in the Greek part.<sup>84</sup> *'bd'bdt*, in turn, was the son of *qynw*, as we can read in the last line of the Nabataean inscription. According to Milik (1976, 145) and Cantineau (1930-32, 2: 142), the root relates to Arabic *qayn* (pl. *quyūn*) meaning 'forgeron, mineur, fondeur' < *qāna* 'to forge, to adorn'.<sup>85</sup>

At the end of the two texts we read the ethnic Φαινήσιος, Nabataean *pyny*, that indicates the ancient town of Faina in Hauran, corresponding to the modern village of al-Mismiyah,86 considered the μητροκωμία (lit. 'mother village' < -κώμη<sup>87</sup>) of the ancient Trachonitis,<sup>88</sup> today called al-Lajāh (lit. 'the refuge'), a region to the south of Syria and to the east of the Jordan river.

The ethnonym is recorded in several texts.89 The site, well-documented in ancient topographies,<sup>90</sup> was probably a strategic place to stop over and successively to conquer Kanatha or Bosra.<sup>91</sup>

In the region of Ḥismā, where it is unlikely that the Greeks lived, the usage of Greek as a written language, apart from the other Semitic languages and Latin of Roman soldiers, was of considerable importance; this is because it was very widely known and the sanctuary of Iram was much visited.<sup>92</sup>

It is difficult to date precisely the inscription on the basis of the steps in the building of the temple, and it would be impossible to identify when the plaster was applied and then painted.<sup>93</sup> Instead, in Milik's view (1976, 145), we may date the inscription to the mid-


91 Sartre 1999, 197-8.

92 *Quellen*, 290.

93 *Quellen*, 290-1.

<sup>84</sup> For the recordings see Cantineau 1930-32, 2: 126. Cf. also al-Khraysheh 1986, 131 and Negev 1991, no. 815.

<sup>85</sup> Proto-Afro-Asiatic \**ḳVyVn-* 'forge' (*HSED* no. 1629). Negev (1991, no. 1025) translates "qain, smith, artisan", in relation to Greek Καινος found on papyri (Wuthnow 1930, 61 and 164). The root *qynw* seems to have also the meaning of a female name, as some inscriptions from al-Ḥijr prove (*CIS* II nos. 205, 207), that al-Khraysheh (1986, 160) vocalizes as *Qainā* and translates as 'Sklavin, Sängerin'. The meaning of 'slave' may be noticed in the epigraphs from al-Ḥiǧr and Sinai (cf. *CIS* II nos. 324, 550, 551, 1239, 1699 *et passim*). The name is also recorded in pre-Islamic Arabic, see *ICPAN*, 492.

<sup>89</sup> In *IGR* III no. 1119 in a letter of the governor of Syria to the inhabitants of the town: Φαινησίοις μητροκωμίᾳ τοῦ Τράχωνος. Cf. also *IGR* III nos. 1120, 1123; *PPAES* IIIA no. 800. In the West it is found in Italy, in Rome and Aquileia, and in Salona, in modern day Croatia. Cf. Feissel 1982, 337-8.

<sup>90</sup> In Ierocle (*Synecdemus* 723.1) there is Φαίνα; Dia-Fenis (*Not. Dign.* [*occ.*] [*or.*] 37.23).

dle of the 2nd century CE, since the main painting of the temple reports the date: year 147 of the month of August.<sup>94</sup> The shape of the Nabataean script seems to be late.<sup>95</sup>

**8.** It is a piece of *grès*, intact on the left, but broken in the remaining part. The curved surface presents an oblique incision in which we read, on the top, one line in Nabataean and below two lines in Greek. M. Sartre (*IGLS* 21,4 no. 148) read only the Greek text through a photograph. Unfortunately, the Nabataean line is not reported. (Plate II, no. 2)

**Dating** unknown **Bibliography** *IGLS* 21,4 no. 148

Text and translation

1) *Nabataean text* 2) Μνησθη οι οικο-3) -δομοι και ΟΠ

1) – 2) Let be remebered the build-3) -er and OP

Commentary

The stone comes from the hallway of the temple of the goddess 'Allat and it is related to the Nabataean *proskynema*, as it is reported by R. Savignac (1933, 418, no. 9), in which the builders of the temple are mentioned: *dkrt 'lt bny' pr''* | *wtymw wḥdnw w'bd'mnw bṭb* 'that 'Allat may remember the builders Far'ā | Taymō, Ḥaḏanō and 'Aḇd'omanō, in good'.

As regards the Greek section of the inscription, we find the mention of the architects, οἱ οἰκοδόμοι, after the letters KAIOΠ that may form the beginning of a proper name or be the indication of another category of artisans: καὶ οἱ τ[---], which may be ταμίαι or τέκτονες. 96


# **Wādī Ġuwayr (al-Šawbak)**

**9.** The inscription was found in the Wādī Ġuwayr, one of the numerous streams flowing from the plains of Edom to the Wādī ʻArabah, south of the Dead Sea basin. The Wādī Ġuwayr is located north of the city of al-Šawbak. Unfortunately, we have no precise data and details about the location and the usage of the inscriptions (collected in *IGLS* 21,4 nos. 120-7) found in the Wādī Ġuwayr, neither we do not know if there was a sanctuary in the area. Nonetheless, the contents of the inscriptions make us think that it was a sacred site.<sup>97</sup>

**Dating** unknown **Bibliography** *CIS* II no. 489; Brünnow, Domaszewski 1904, I no. 120e; *IGLS* 21,4 no. 120

Text and translation

1) Κατταβος 2) *gdṭb*

1) Kattabos 2) Gaḏṭaḇ

Commentary

This inscription reports a theophoric name made up of the terms *gd*  and *ṭb* meaning 'the (god) Gad is good', well-known in the Nabataean epigraphy and, more generally, in the Aramaic.98 The root *gd'* becomes the name of a deity to be identified with ἡ Τύχη – 'The Fate' of the Greek divine context. It is found, in the shape of *gdṭb*, not only in Petra, but also in Hegra.<sup>99</sup>

The worship of Gad, who became the god of Fate (Lat. *Eutychus*<sup>100</sup> < Gr. Εὐτύχης), was widespread in Hauran.101 In fact, several sanctuaries, called Τύχεια or Τύχαια, were built<sup>102</sup> becoming *Bayt Gadā*. The Syriac poet Jacob of Serugh (451-521 CE) in his *Homiliae* (Syr. *mem-*

97 *IGLS* 21,4, 154.

98 Cf. al-Khraysheh 1986, 52-3 and Negev 1991, no. 213.

99 *CIS* II no. 236; *RES* no. 1167.

100 It is recorded in the Nabataean-Latin bilingual inscription in Rome, in *CIS* II no. 159.

101 Cf. Mordtmann 1877, 99.

102 See *CIG* nos. 4554, 4555, 4556.

*rē*) writes about *Bayt Gadā* located on the mountaintop and transformed into monasteries during the Christian period.<sup>103</sup>

The Biblical Hebrew name *Gād* is present in the Ancient Testament (Isa 65, 11) with *Mənī* as beneficiaries of food offerings. The Hebrew verb *gādad*, Arabic *ǧadda*, means 'to cut, to divide' and herein probably lies the origin of the idea of the 'fate', defining the destiny of human beings.<sup>104</sup> From the Hebrew verb *gād*, Ar. *ǧadda* and Syr. *gaddā* 'to be lucky, rich', we have the expression *bə-gād* (Gen 30,11) 'Thankfully!' (it was translated in the LXX ἐν τύχῃ, and in the Vulgata *Feliciter*).<sup>105</sup>

As regards Κατταβος, it is a quite rare masculine name in the Greek epigraphy106 and the only reference is to be found in a Latin-Greek bilingual epigraph from Cyrenaica.<sup>107</sup>

#### **Ġūr al-Ṣāfī**

**10.** This inscription was found at al-Naq' cemetery of Ġūr al-Ṣāfī, ancient Zoar,<sup>108</sup> located to the south-east of the Dead Sea, in Jordan. It is an epitaph carved in a rectangular tombstone of whitish sandstone. The letters are painted in red colour and the entire inscription consists of four lines.

The Greek text presents a calligraphic oval script with symmetrical letters except for the tiny *omicron* at the end of the first line. In addition, the author of the inscription uses small dots as word-dividers. The Nabataean section is engraved in an elegant elongated script tending to the *scriptio continua*.

103 Clermont-Ganneau 1898b, 81.

104 Gesenius 1846, 157.

105 In the Palmyrene epigraphy *Gad* is rendered into Τύχη (cf. *VIS* nos. 3, 95; *PAT*, 433, 352 = Stark 1971, 13, 81). In Punic it is noticed in the inscription of Nora, Sardinia (4th-3rd c. BCE): *lrbt ltnt pn b'l wgd* 'Alla Signora, a Tanit, volto di Baal, Fortuna' (Amadasi Guzzo 1990, 73, no. 3) and in the inscription of Ibiza (2nd c. BCE): *lrbt ltnt 'drt whgd* 'To the Lady, to the powerful Tanit and the Gad' (*KAI* no. 72); see also Krahmalkov 2000, 136-7. The root *gdy* is generally attested in Aramaic and *'mgd* in South Arabic. For a close examination of the meaning that it assumes and for its presence in the Semitic languages, cf. *DNWSI*, 212-3. In the Hatraean epigraphy it is recorded in the form of *g(n)d'* (Beyer 1998, 147). Moreover, it survives as *eterogram* in Middle Persian (or Pahlavi) as *GDE*, read *xwarrah* 'lucky' (MacKenzie 1986, 96) and as a loanword in Ge'ez *gadd* 'lucky' (Leslau 1991, 180).

106 Cf. Pape 1911, 637. In the Wādī Haggag, Sinai, the name is written as Γαδος (Negev 1977, no. 184).

107 *CIG* III, 5175. In Latin it is transcribed *L. Vibio L. [F.] Cattabo*. See *SEG* 9 no. 247 with *gamma* (Gattabos).

108 It is mentioned in the Ancient Testament as *Zo'ar* (Gen 14,8, previously called *Bela*), and it was part of the late Roman province of *Palaestina Tertia*.

As regards al-Naq' cemetery, it was probably predominantly of Jewish ritual. Indeed, there are tombstones written in Aramaic, Hebrew and Greek, as well as a Hebrew-Greek bilingual,109 which belonged to Jewish, Christian and pagan people. The inscriptions of deceased Jewish people are dated according to the cyclical agricultural calendar of seven years bearing Jewish symbols; meanwhile the Christian inscriptions are dated starting from the Lord's Day and have crosses as symbols.<sup>110</sup>

The Greek part is shorter than the Nabataean, which contains more information about the deceased woman and her date of death. (Plate III, no. 1)

**Dimensions** height 23 cm; length 40 cm; thickness 13 cm **Dating** 2nd-3rd CE (?) **Bibliography** *IPT* Ib no. 50; Petrantoni 2016, 131-6

# Text and translation<sup>111</sup>

1) Ισμεηλη Αβδαρετου Ζωιλος 2) Αλεβου συνβιω ευνοιας χαριν 3) *d' npš' dy 'bd zyls br 'lbw l'šm'yn* 4) *'ntth ḥbybt brt' 'bšlm byrḥ sywn* 1) For his wife Ismeele (daughter of) Abdaretas, Zoilos 2) (son of) Alebos, out of affection

3) This (is) the tomb which Zoilos, son of 'Aleḇō, made for 'Išma'īn

4) his beloved wife, daughter of 'Abušalem, in the month of Sīwan

### Commentary

In line 1 there is the name of a woman, Ισμεηλη; its Nabataean transcription, *'šm'yn*, is intriguing. The Greek masculine form, Ισμαηλος, is attested in two epitaphs from Jericho,<sup>112</sup> while the name Εσμαηλος is found in a funerary inscription from Busān, in Hauran, dating to

109 Cf. Cotton, Price 2001, 277-83.

110 Ilan 2012, 30. It is also true that some inscriptions are dated on the basis of the era of *Provincia Arabia*, whereas other inscriptions in Greek have no date.

111 In the *editio princeps* the authors only report a translation of the Nabataean text without its transcription, which I myself provide below (see the bibliography).

112 Hachlili 1979, 34-5; *IPT* Ib 4b-c no. 10 and comment on pp. 48-9; Rahmani 1994, 243-4; *SEG* 31 nos. 1407.6. See also Pape 1911, 573.

the middle of the 4th century CE (341/2).<sup>113</sup> *'šm'yn* is probably new in the Nabataean onomastics, since it is not encountered elsewhere in Nabataean, Palmyrene, Hatraean or in pre-Islamic Arabic.<sup>114</sup>

The patronymic Αβδαρετος represents the Greek transcription of the Nabataean *'bdḥrtt*. In the funerary epigraphy from Ġūr al-Ṣāfī, it appears in the variant of Αβδοάρθα<sup>115</sup> (the genitive of Αβδοάρθας), Nabataean *'bdḥrtt*, but here the Nabataean name does not correspond to the Greek. Instead, in fact we find *'bšlm*. Presumably Αβδαρετος was called *'Abdaretas* in the Greek speaking environment of Zoar, while among Nabataean speakers he was known as *'Abušalem* which is a name given to him after the birth of his first son who was called *Šalem* (< *šlm* 'peace'). In fact, he is traditionally cited as the father of a firstborn son according to the Arabic practice, still in use today: *'b*  'father' + son's name.<sup>116</sup> The name *'bšlm* is recorded at Hegra as well.<sup>117</sup>

At the end of line 1 we read the author's name Ζωίλος, usually used in Greek,<sup>118</sup> deriving from the word ζωή 'life',<sup>119</sup> and transcribed as *zyls* in Nabataean, line 3. The first mention of this name in Palestine during the Hellenistic period is found in the Greek-Aramaic votive bilingual inscription of the 3rd-2nd century BCE from Tell Dan<sup>120</sup> and in an amphora from Rhodes.<sup>121</sup>

The patronymic of *Zoilos* is Αλεβος, a Semitic name that corresponds to Nabataean *'lbw*, in line 3.<sup>122</sup> The root probably comes from Arabic *ġālib* 'winner' < *ġalaba* 'to win, to subdue, to conquer'<sup>123</sup> and is present in the Aramaic *'lb* as well (Syr. *'eleb* 'to exceed, to surpass, to cause wrong, to oppress'<sup>124</sup>). Negev's assumption (1991, no. 886 as first hypothesis) of an Arabic origin *'alib* 'to become thick or coarse,

114 Cf. Cantineau 1930-32; al-Khraysheh 1986; Negev 1991; *NABLEX*. For Palmyrene and Hatraean cf. *PAT*; Stark 1971; Beyer 1998. For pre-Islamic Arabic see *ICPAN*.

116 *IPT* Ib, 127.

117 *JSNab* no. 313. Cf. also Negev 1991, no. 29.


122 It is used as a personal name in *CIS* II no. 363; Dalman, 2: nos. 4, 11, 14; *RES* nos. 1383, 1389, 1392.

123 Cf. Cantineau 1930-32, 2: 130.

124 Sokoloff 2009, 1099-100. The root is also found in Old Aramaic (*DNWSI*, 850). In Hebrew the verb *'ālab* 'to be strong' is employed in *hiph'il* form, *he'ălīb*, that means 'to put to shame, insulted, umiliated' (Gesenius 1846, 630).

<sup>113</sup> Wadd. no. 2247.

<sup>115</sup> *IPT* Ia nos. 21 and 79.

rude',<sup>125</sup> 'hard, tough' is unconvincing. The Greek form Αλεβος<sup>126</sup> is exclusively recorded in Hauran in funerary127 and monumental128 inscriptions.

In the last line of the Greek section in the substantive συνβίῳ < σύμβιος 'wife, spouse, partner' the ν, preceding β, is not assimilated. This section ends with εὐνοίας χάριν that, in the same way as ἕνεκεν, is a common expression used in honorific inscriptions in order to indicate the respectful attitude of the honoured person towards the community showing the honour; in the funerary context this phrase reflects the feelings of affection towards the deceased.<sup>129</sup>

As regards the Nabataean part of the epitaph, in the first line we read the term *npš'* <sup>130</sup> followed by the author of the object.

In the last line there is the substantive *'ntth* (<\* *'nth*) 'wife', followed by the singular masculine suffix pronoun -*h* 'of him' = 'his'.<sup>131</sup> Here the assimilation of *n* to the following *t* should be expected in order to obtain *'tt-* (cf. *CIS* II nos. 158, 161, 194), but *'ntt-* is also attested somewhere along with *'tt-*. <sup>132</sup>

The affection of the husband *Zoilos* to his dead wife is manifested by the adjective *ḥbybt* 'be loved', the singular feminine passive participle of *pe'il* form from *ḥbb* 'to love'.

The final expression *byrḥ sywn* 'in the month of Sīwan', the 9th month of the Hebrew calendar, should reveal the date of the engraving of the epitaph or the month when *Ismeelé* died.

127 Wadd. no. 2053a; *SEG* 7 nos. 1144, 1156.

128 *SEG* 46 no. 2073.

129 *IPT* Ib, 126. Regarding references to the expression in Hauran see *SEG* 7 nos. 1072, 1086.

130 It is a noun frequently used in funerary Nabataean inscriptions (Cf. for instance *CIS* II nos. 159, 169, 195, 191, 194, 352, 353, 465 *et alia*) and it may have several meanings. In some inscriptions it is translated as 'tomb' or 'gravestone', while in other cases it refers to 'soul, life, person and body' (for the employment, the meaning and the occurrences of the name cf. *DNWSI*, 744-9). In Nabataean other terms are employed to indicate the tomb within a stylistic and architectural context and, compared to other regions, the term *npš'* has the same meaning both at Hegra and Petra, and in Hauran and in Sinai (Abdelaziz-Rababeh, 2008, 182). Other words meaning 'tomb' are, for instance, *mqbr'*, *qbr'*, *kpr'* (respectively in *CIS* II nos. 350, 184 and 197 to quote only three examples), while *'m'* (*CIS* II no. 173) represents the 'sarcophagus', a 'little case', an 'ossuary', *gwḥ'* (*CIS* II no. 211) a 'burial niche', *ṣryḥ'* (*CIS* II no. 213) a 'niche inside of a tomb', and *wgr'* (*CIS* II no. 205) a 'cavern in a mountain'.

131 For the occurrences of the possessive suffix see Cantineau 1930-32, 2: 65.

132 Cantineau 1930-32, 1: 44-5; *PPAES* IVA no. 82; *RES* no. 2103.

<sup>125</sup> Lane, 2126.

<sup>126</sup> Along with Αλαβ, Αλεβου, Αλβος (Cf. Wuthnow 1930, 16 and 157). The latter is present at Tocra in Cyrenaica (*LGPN* I), at Ephesus (*LGPN* VA) and at Pinara, in Lycia (*LGPN* VB).

### **Madaba**

**11.** The epigraph was found at Madaba,<sup>133</sup> a town located 30 kilometres south-west of Amman in Jordan. The inscription is an epitaph and is engraved on a square stone whose surface is rather ruined; nevertheless, the first editors had no difficulty in deciphering the text. (Plate III, no. 2)

**Dimensions** height 53 cm; length 43 cm; inscribed surface 51 cm × 39 cm; average height of the letters 3 cm **Dating** 2nd CE (108-109 CE) **Bibliography** Milik 1958, 243-6, no. 6; Milik 1980, 44-5; *SEG* 20 no. 494; *IGLS* 21,2 no. 118; *Quellen*, 212-3

# Text and translation


<sup>133</sup> The Moabite city of *Mêdəbā* mentioned in the Bible (Num 21,30; Josh 13,9) was one of the settlements divided by the twelve tribes of Israel during the Exodus. Its name also appears in the Mesha's stele (*CNSI*, 1-2) that was built around 850 BCE by the will of the Moabite king Mesha to commemorate his victory over the Israelites. It was conquered by Alexander the Great and ruled by the Seleucid dynasty. During the Seleucid reign, the town fell under the rule of the Ammonites, Israelites and finally it was part of the Nabataean realm. In 106 CE it was annexed by the Romans and flourished, even though it was not of primary importance.

#### Commentary

Line 1. The incipit is similar to that of the inscription on the tomb of *Itaybel*, a witch of Madaba, and her sons, which was built in 37 CE.<sup>134</sup> The author clearly distinguishes *mqbrt'* (the feminine form of *mqbr'* in no. 1) from *npš'*. If the first term<sup>135</sup> identifies a 'sepulchre', a 'tomb', the second refers to the 'tomb built in the sun', usually a pyramid-shaped tomb covered by a cube; in fact, as we may read in the inscription, it lies on the top of the deceased: *dy 'l'mnh* 'which is on top of him'.

Line 2. Here the name of the author appears. It is a certain *'bgr*, corresponding to Αβγαρ of the Greek text in line 7. It deals with a rare name among the Nabataeans inasmuch it is recorded in the northern Aramaic onomastics.<sup>136</sup> This name derives from \**bgrt* (Ar. *buǧrah*) 'navel'.<sup>137</sup> According to Milik (1980, 46), Abgar had the function of the guide of the Nabataean herd of horses and camels.

Line 3. We read the name of the author of the epigraph, *'yšywn*, transliterated in Greek as Εἰσίων. As Milik points out (1958, 245), it is a name originating from the root *'yšw* which is frequently used among the ancient Arabs (cf. Ar. *'iīās* and Saf. *'yṣ* 'desperation, scepticism' < *'aīsa*) <sup>138</sup> and usually transcribed in Greek as Ἰᾶσος (*RES* no. 463).<sup>139</sup> Here, the name Εισιων simply reflects the Nabataean form of the name.

The patronymic *mn't* = Μονοαθου, line 8, probably vocalized as \*/mono'aṯ/,140 derives from the Arabic *man'ah* 'power, strenght' < *mana'a* 'to ban, to prohibit, to forbid',<sup>141</sup> Heb. *māna'*, Ge. and Am. *mänänä*

134 *CIS* II no. 196; *RES* no. 674.

137 Cantineau 1930-32, 2: 70; Negev 1991, no. 5. The name *'bgr* was found in the graffiti from Sinai (*CIS* II no. 698), while the form *'bgrw* in *CIS* II no. 750; in Greek Αβγαρος (Wadd. nos. 1984, 2046; *PPAES* IIIA, 2 no. 112; Pape 1911, 2; Wuthnow 1930, 7) and in Latin *Abgarus* (*CIS* II no. 159).

138 *ICPAN*, 88: *'ys*, *Iyās*, *'yst*, *'ysn*.

139 Catineau 1930-32, 2: 61. The name ῎Ιασος is found in Attica (*LGPN* II) and in Magna Graecia (*LGPN* IIIA), in ancient Venusia (modern day Venosa in the province of Potenza), Italy.

140 *Quellen*, 213. Al-Khraysheh (1986, 109) reads *Māni'a*. Cf. also Negev 1991, no. 660.

141 Cantineau 1930-32, 2: 116. See also *ICPAN*, 568-9.

<sup>135</sup> It is found in *CIS* II nos. 181, 196, 2033; *PPAES* IVA no. 106; *RES* no. 1090; in construct state *mqbrt* in DM, II no. 18; *RES* 481. It is also present in Palmyrene (Cf. *DN-WSI*, 678).

<sup>136</sup> Cf. al-Khraysheh 1986, 24. Pre-Islamic Arabic *'bjr* (*ICPAN*, 9). Abgar was the name of several kings of the Osroene kingdom of Edessa, in Mesopotamia. There is a possibility that it referred to a title as August or Caesar did, rather than a personal name. (Cf. Ball 2001, 90). It is frequently encountered in Palmyrene (*PAT*, 429 = Stark 1971, 1 and 63).

'to reject, cast aside'. The primary syllable is *n'* which has a negative force,<sup>142</sup> but Syr. *mana'* 'to bring, lead', 'to arrive, come, attain'.<sup>143</sup> The Greek transcription<sup>144</sup> gives us a rough Aramaic vocalization of the name in which we may notice an /o/ before α, the vowel of /'/, and the usage of θ for the final /t/ indicating the actual pronunciation of /-aṯ/.

Line 4. The name of the tribe (*'l* 'family, tribe' = Ar. *'āl* ), to whom the deceased's father belonged, is *'mrt*. <sup>145</sup> It is vocalized by Milik (1958, 245) as *'āmirat* through the Greek Αμιραθου (Wuthnow 1930, 19), and it is the feminine present participle (cf. Ar. *'umayrah* 'subdivision of a tribe' < *'amīr* 'a place inhabited, peopled').<sup>146</sup> The root comes from the Arabic 'to build, to live' (cf. Syr. *'mar* with the same meaning) and it is used as a personal name both in Nabataean and in Safaitic;<sup>147</sup> it is also the name of a Safaitic tribe from the northern Transjordan desert.<sup>148</sup> In one of the Nabataean graffiti found in the area of Burqu'<sup>149</sup> the author describes himself as *dy mn 'l 'mlt*; <sup>150</sup> as Milik argues (1980, 43), *'mlt* is a phonetic variation of *'mrt*. The Greek part of the inscription does not contain the tribe's name.

The deceased son's name is *šlmn* = Gr. Σελαμαν (line 6)<sup>151</sup> < *šlm*  'peace'. Its variations *šlmn* or *ślmn* (cf. Ar. *salamān*) 152 are well-documented in Petra<sup>153</sup> and in Hegra<sup>154</sup> as well as in the Palmyrene<sup>155</sup> and Hatraean<sup>156</sup> onomastics.

145 Cf. Negev 1991, no. 919 and al-Khraysheh 1986, 145.

146 Al-Khraysheh (1986, 143) noted the existence of the name *'myrt* meaning 'headgear', 'Kopfschmuck', the diminutive form of *'mrt*.


150 Milik 1980, 42-3, texts nos. 1-2d.

151 Pape (1911, 1361) reports the form Σελαμιν attested in Galilee. Further forms are: Σελαμ, Σελαμανους, Σελεμα, Σελημ, Σελημα, Σελομανης and Σελυμαιω (Wuthnow 1930, 107 and 171).

152 Pre-Islamic Arabic *slmn*, *Salmān* (*ICPAN*, 326).


<sup>142</sup> Gesenius 1846, 487 = *DGes*, 699; *KAHAL*, 306. In JBA, PTA, Gal., Sam. *mn'* bears the meaning of 'to hold back, stop doing, withhold' (Jastrow 1903, 801; *DNWSI*, 661; *DJPA*, 318b; *DJBA*, 687b; Tal, *Sam*, 478).

<sup>143</sup> Payne Smith 1903, 282; Sokoloff 2009, 784.

<sup>144</sup> With variations Μοναθος (Wadd. no. 2499) and Μονοαθου (*PPAES* IIIA, 7 no. 800). Cf. Wuthnow 1930, 78.

Line 5. After the mention of the year when the text was carved, i.e. *bšnt tlt* 'in the third year', an unusual title appears: *hprk* 'eparch',<sup>157</sup> instead of *hprky'* 'eparchy' as written in the Greek part; in fact, in the last line we read ἐπαρχεία. In the Nabataean epigraphy the term *hprky'*<sup>158</sup> is quite common; it is a Greek loanword < \*ὑπαρχία = ὑπαρχεία 'province, district'.<sup>159</sup> In Nabataean this term is used to mark the territory, in this case that of Bosra, belonging to the *Provincia Arabia*. <sup>160</sup> The year 3 of the eparchy of Bosra corresponds to 108/109 CE, the later date of the era of the City.<sup>161</sup>

Lines 6-7. The Greek section of the inscription starts by mentioning the deceased's name Σελαμαν followed by χρηστέ 'good' that is usually used in the inscriptions along with χαίρειν, as in our case: ἄλυπε 'without pain' and χαῖρε 'hi!'.

Lines 8-9. We find the adjective τειμίῳ < τίμιος 'well-loved, honoured' referred to the dead son and in line 8 the substantive τὸ μνῆμα 'burial monument, gravestone, memory'.

**12.** The inscription is carved on a basalt within a *tabula biansata* that is broken into two parts. The fragment on the left side appears to be worn out and some letters are no longer legible. The epigraph is entirely written in Greek and, according to the first editors, only at the end are we able to distinguish any signs in the Nabataean script.

**Dimensions** height 32 cm; length 55 cm; average height of the letters 3 cm **Dating** 3rd CE (157 CE?)

**Bibliography** Germer-Durand 1895, 590; Clermont-Ganneau 1898a, 12-14; *RES* no. 2021

Text and translation


157 Cf. Cook 1898, 43.

158 See CNSI nos. 231 and 260; *YTDJD* I nos. 297, 323, 326; II, 97, 109, 111. Generally, for Aramaic see Cook 1898, 44.

159 The *h* = ε represents ἐπαρχ- / ὑπαρχ- (Wasserstein 1993, 206).

160 Monferrer-Sala 2013, 106.

161 Milik 1958, 246.

1) Abdallas (son of) Ana[.] this tomb


#### Commentary

The author's name is identified, in the *editio princeps*, <sup>162</sup> as Αβδαλλα and his patronymic corresponds to a supposed Σανα. However, it would seem more plausible to accept the hypothesis proposed by Clermont-Ganneau (1898a, 12) who simply reads Αβδαλλας Ανα[μ] ου as 'Abdallas son of Anamos'.

As regards the author's name, it is a theophoric (= Nab. *'bd'lhy*) attested in Greek and in Nabataean epigraphy,163 as well as his patronymic Αναμος (= Nab. *'nmw*).<sup>164</sup> The latter seems to originate from the Arabic *ġānim* 'qui fait du butin qui réussit sans effort',<sup>165</sup> in Saf. *'nm* and Palm. *'nmw* 'successful, noble'<sup>166</sup> (< Ar. 'to loot, to pillage', 'to rob').

In line 6, after καταστάσεως, according to the epigraphic usage, we should read the noun τῆς πόλεως, in place of the name of the city, that fills the blanks before Ἀντωνείν[ου].<sup>167</sup>

Lines 6-7 provide two important historical references in order to date the inscription; in fact, there is the name of Antoninus (probably the Roman Emperor Antoninus Pius<sup>168</sup>) and the year of his reign, that is XIX corresponding to 157 CE, probably the year in which the inscription was composed. In line 5 we find two further dates, but the first is only visible in the horizontal line above the letters, while

165 Cantineau 1930-32, 2: 134.

166 *PAT*, 438 = Stark 1971, 45 and 106.

167 Clermont-Ganneau 1898a, 13. It is probable that the Σ at the end of the *lacuna*, linked to the A, truly represents the genitive ending of πόλεως.

168 Antoninus Pius was Emperor from 138 to 161 CE.

<sup>162</sup> Germer-Durand 1895, 90.

<sup>163</sup> For Nabataean see Cantineau 1930-32, 2: 126; for the Greek form Αβδαλλας cf. *PPAES* IIIA no. 144 (it is attested in Hauran), Negev 1991, no. 793 and Wuthnow 1930, 7 and 153 in which the form Αβδαλας is recorded. Whereas Αβδαλλα is present in Sinai in the Wādī Haggag (see Negev 1977, no. 42).

<sup>164</sup> Cf. Cantineau 1930-32, 2: 133-4. In Greek we find forms like Αναμος (Negev 1991, no. 924), Ανεμος (Wadd. no. 2053; Wuthnow 1930, 22), Ανναμος, Ανμος, Ωνεμος (Wuthnow 1930, 23, 121, 159).

the second, which is 340, coming after καταστάσεως, indicates the year of the building of the city.

At the end of the last line Germer-Durand recognized some Nabataean characters identifying them as a consonantal sequence such as *mnbrk'*. For Clermont-Ganneau (1898a, 13) it may be the name of the city of Madaba either in the Biblical form of *mydb'* (= Heb. *mêdəbā*<sup>169</sup>) or, according to the original Moabitic diction, *mhdb'* which occurs in the stele of Mesha.<sup>170</sup>

# **Zīzah (Zuwaiza), East of Madaba**

**13.** This epigraph was found during the excavations about 250 metres south-east of Qala'at Zīzah probably near the ruins of an ancient church.<sup>171</sup>

In the inscription, which is engraved on a limestone, the bilingual texts are written on two parallel columns. The stone presents an oblique incision on the base that belonged to another stone. The stone was most likely worked to be embedded in the floor.<sup>172</sup> The two texts are separated by a little central space of about 8 cm. The Nabataean inscription is shorter and more damaged than the Greek and only 12 lines can now be read, though the whole text must have been longer.

**Dimensions** length 70 cm; height of the bigger side on the right 36 cm; height of the smaller side 23 cm; thickness 15 cm **Dating** beginning of the 2nd c. CE (?) **Bibliography** *RES* no. 1284; Jaussen, Savignac 1909, 587-92; *JSNAb* nos. 392 and 21, tab. 71; *IGLS* 21,2 no. 154; *Quellen*, 213-5

Text and translation<sup>173</sup>

1) *[---]* 2) *[---]y bnh* 3) *dms br hll*

169 Cf. Num 21,30; Josh 13,9.

170 *CNSI* no. 1.

171 The ancient town of Zīzah is located on the mountain of Darb al-Ḥaǧ, east of Madaba. When the stone was extracted the reporters (Jaussen, Savignac 1909, 588) saw no buildings and no trace of a wall that might restore the image of the church. They only found at the bottom of a hole and on the edge a set of columns whose forms and dimensions resembled those of the Roman milestones.

172 Jaussen, Savignac 1909, 588.

173 The transcription here presented follows that of *Quellen*, 214.

4) *br dms 'mny* 5) *dy mr byt'[t---*] 6) [Δημ]ας Ελλην[ος] 7) [Παν]αμου μηνος [ω]- 8) κοδομησεν [το] 9) ιερον του Διο[ς τ]- 10) ου εν Βεελφε[γωρ] 11) και τον ναον [α]- 12) [φιερω]σεν σ[---] 1) [---] 2) [---] who made 3) Demas son of Hillel 4) son of Demas from 'Amman 5) who (is) the Lord of the house 6) [Dem]as (son of) Ellen[os] 7) in the month of [Pan]amos [bu]- 8) -ilt [the] 9) sanctuary of Zeu[s] 10) who is in Beelfe[gor] 11) and the temple 12) [sacr]ed [---]

Commentary

The beginning of the epigraph may be completed by the expression: 'This is the temple of Baal of (Mount) Pegor'<sup>174</sup> in order to recall the lines 10-11 of the Greek section.

In line 3 the author's name, *dms*, is of Greek origin and it helps us to reconstruct the name [Δημ]ας occurring in line 6. Macdonald (1999, 274) suggests a cross reference to Safaitic *dmṣ*, *dmṣy*, proposed by Winnett (1973, 54).<sup>175</sup> If this hypothesis is right, Safaitic *dmṣ* may represent the Greek name Δημᾶς-Δαμᾶς as Nabataean *dms* and forms with *nisbah* may reproduce the hypocoristic of Δαμάσιππος. <sup>176</sup> This

174 Such a reconstruction is found in *Quellen*, 214: "Dies ist der Tempel des Baal vom (Berg) Pegor".

176 Cf. Milik, Starcky 1970, 142. This kind of hypocoristic is found at al-Ḥiǧr in the shape of *dmsps* (cf. al-Khraysheh 1986, 60) Greek Δαμάσιππος (Pape 1911, 267).

<sup>175</sup> Pre-Islamic Arabic *dms'* < *damīs* or also *dmṣ* 'to hasten' (*ICPAN*, 243). He wonders whether in north Arabia a tribe named Damaṣī truly existed. The assumption of Macdonald replaces that of Negev (1991, no. 278), who believes that the forms *dms*/*dmsy*  are identical and there is a link with Arabic *damīs* 'hidden, concealed'. Cf. also Wuthnow 1930, 43. In Nabataean this name is present in *JSNab* no. 392 (cf. Cantineau 1930- 32, 2: 83 as well). It is also recorded in Palmyrene (*PAT*, 433 = Stark 1971, 43 and 135) in which the form Δαμας appears.

name has Greek origins and its meaning refers to the form Δημέας that Pape (1911, 288) identifies with 'Volkmann' or 'Dörfler', a clear connection with the people (< Gr. δῆμος).<sup>177</sup>

The father's name is *hll* (= Gr. Ελλην[ος] in line 6).178 If the reading is right, it represents the Biblical name *hillēl* (Jugd 12,13), the father of Abdon the Pirathonite,<sup>179</sup> whose root *hll* means 'to be clear, bright',<sup>180</sup> also 'to praise, to glory, to celebrate' in *pi'el* form.

Line 4. The ethnic *'mny* is omitted in the Greek part. It deals with an adjective referred to 'Amman, Ammanite', the place of birth of the author's grandfather. In the Nabataean section, Amman is presumably used to indicate the name of the god Ba'al that should appear in line 1, according to the reconstruction.<sup>181</sup>

Line 5. The reading is uncertain, but we find the title, probably bestowed on *Demas*, *mr byt*, that means 'Lord of the Temple'.<sup>182</sup> If the reading is right,<sup>183</sup> successively we may read a probable *'tr*, as this is the 'Lord of the Temple of the place'.<sup>184</sup>

Line 6. The Greek part of the inscription begins by quoting the author's name and his father's name. The reading Δημᾶς is prompted by the Nabataean text since here we may only trace the middle

179 In Biblical Hebrew *'Abədōn*, in Greek Ἀβδών, he was the twelfth Judge of Israel.

180 Gesenius (1846, 226) interprets: "used of a clear, sharp tone or sound" and reports that in Ethiopia the women, during the public rejoicing, have the habit of reproducing the sound *ellellell-ellellell*. Cf. *ΜDGes*, 278-9 and *KAHAL*, 129-30 'rühmen, Gott preisen'. Indeed in Ge'ez we find *tahalala* 'jubilate, utter cries of joy'. The Hebrew root is connected with Arabic *hallala* (< *halla* 'to appear, to begin', also 'to pour down [the rain], to peal', therefore 'to make noise') that means 'to praise God, to shout for joy, to rejoice' (also Akk. *alālu*, *elēlu* 'to sing a joyful song, boast, exult, celebrate'. Cf. *AHw* I, 34 and 197; *CAD* I, 331 and ff.; *CAD* IV, 80 and ff. and Klein 1987, 152). In the Aramaic of Qumran, CPA, Sam., Man. *hll*, Syr. *hallel* has the same meaning of 'to pray' (Jastrow 1903, 353; Sokoloff 2009, 344; Tal, *Sam*, 210).

181 *Quellen*, 214.

182 This expression is used in the shape of *mr' byt'* in *RES* nos. 1088, 1111; *CIS* II nos. 235a, 235b; *JSNab* no. 58; *Quellen*, 269-72 with commentary. It is transcribed as *mr byt'*  in *JSNab* no. 59, *RES* no. 1284; while *mr byt* in *JSNAb* no. 392. Nöldeke (1909, 184-5) argued that, with reference to the inscription in *RES* no. 1088, *mr' byt'* was to be identified with *rabb al-bayt* which is recorded in the Koran (CVI, 3). In the Nabataean epigraphy the connection of the title *mr byt* to the deity of al-'Uzzā comes from an inscription from the Wādī Rām (Savignac 1933, 413-5, no. 4), which is engraved on the left of a niche along the road from Jabal al-Kubṯā to Petra (Dalman, 2: no. 46, fig. 42) and in an inscription from Hegra (Nehmé 2005-06, 189-94, no. 12, fig. 134).

183 In fact, in *mr* all the base of the Nabataean *m* was lost and we may only reconstruct the substantive.

184 *Quellen*, 214.

<sup>177</sup> *dyms* is a loanword in Nabataean assuming the meaning of 'people', as well as in Pal., Sam. and Syr. (*DNWSI*, 253).

<sup>178</sup> Jaussen, Savignac 1909, 589. Cf. also Negev 1991, no. 303 and Cantineau 1930- 32, 2: 86.

line of the A and of the M. On the contrary, Ἕλληνος is more legible.

Line 7. At the beginning we read -AMOY<sup>185</sup> and, on the basis of the following μηνός, we may reconstruct the name of the Greek month [Πάν]ημος.

Lines 9-10. Here we may see the Greek equivalent of the dedication to Zeus (= Διός), the god who 'is' in Belfagor, as expressed in line 10: οῦ ἐν Βεελφε[γωρ]. He is a middle eastern deity worshipped by the Moabites; in Hebrew *baʿal-pəʿôr* 'The Lord of mount Peor.<sup>186</sup> With the arrival of Greek culture in the Moab region, *Ba'al* would become God/Zeus worshipped in other places in the same way.<sup>187</sup>

### **Jerash**

**14.** This epigraph is engraved on a red dolomitic limestone (in Arabic *Mizzi aḥmar*),<sup>188</sup> found in May 1931 in the ancient town of Jerash,<sup>189</sup> 48 kilometres north of the capital Amman.

The left part of the inscription is ruined by a vertical incision that partially splits the stone into two parts. So, it is hard to read the fragment because the beginning and the end of the stone were lost.

The first who studied this inscription was Father Luis-Hugues Vincent (1872-1960), from École Biblique et Archéologique Française of Jerusalem in collaboration with his colleague Father A.J. Savignac (1871-1962). Father Vincent gave his contribution to the analysis of the Nabataean text as Kraeling reports (1938, 371). (Plate IV, no. 1)

185 On the contrary, the first editors read -[NE]MOY. Cf. Jaussen, Savignac 1909, 589.

186 The Bible narrates the event in which the Israelites yoked themselves to the Ba'al of Peor triggering the Lord's anger against them (Num 25,3).

187 Jaussen, Savignac 1909, 589.

189 The town is located along the banks of the Wādī Ǧaraš river, an affluent of the Zarqā' river. The first settlement of some importance is that of the Greeks after the conquest of Alexander the Great, presumably around 331 BCE However, Jerash only became really important after the Roman conquest in 63 BCE and it was annexed to the Roman province of Syria; in addition, it joined the Decapolis league of cities. During the following two centuries, Jerash conducted business with the Nabataeans and, thanks to the gains of trade and the wealth obtained through agriculture, it became rich and flourishing. Jerash achieved the peak of prosperity in the 3rd century, but the Persian invasion in 614 CE and that of the Arabs in 636 led to its rapid decline. Moreover, in 749 CE a major earthquake destroyed much of Jerash and its surroundings and its population decreased.

<sup>188</sup> It is common in and around Jerusalem where it has been used in buildings since ancient times. In particular, it was used for *ablāq*-style multi-colored masonry.

```
Dimensions height 39 cm; length 22.5 cm; height of Greek letters about 1 cm; 
height of Nabataean letters 1-2 cm
Dating 1st CE (80-81 CE)
Bibliography Kraeling 1938, 371-3; Bowersock 1973, 139, no. 54; Amadasi 
   Guzzo, Equini Schneider 1997, 55; Quellen, 202-3
```
# Text and translation

1) [---]της 2) [---]νος 3) [---]αυ πο 4) [---]δινετο 5) [---]εις δυσμα-6) [-ς ---]δ εις νοτ-7) [-ον ---]ιων μερω-8) [ν-- τ]αις επαλξεσι 9) [---]μως 10) *dnh ṣlm'[---]* 11) *dy (nkr/dw/py') [---]* 12) *ḥrtt (?) mlk nbṭw [---]* 13) *[---]* 14) *[---]* 15) *[….] 'l ḥyy mr'n' rb'l mlk' [---]* 16) *'šryn wḥd bsywn šnt 'šr wḥd[h]*

```
Lines 1-2-3-4 are hard to read
5) [---]towards the West 
6) [---]towards the South
7) [---]of the parts (?)
8) [--- t]o the shelters 
9) [---]
10) this is the statue (of?) [---]
11) that[---]
12) Aretas king of the Nabataeans [---]
Lines 13-14 are hard to read 
15) [….] for the life of our Lord Rab'ēl, the King [---]
16) 21 of Sīwan of the year 11
```
# Commentary

In the Greek section, the first four lines carry only some final letters; in lines 5 and 6 we read εἰς δυσμάς, referring to a direction towards the west, and in the following line εἰς νότον, in this case indicating a direction towards the south.

The reading of line 7 is rather difficult. On the basis of the facsimile, which was presented by Father Vincent, Kraeling thought the text could be reconstructed as τῶν μερῶν. Subsequently, in a further revision, *Quellen* reads …]ΙΩΝΜΕΙΩΝ<sup>190</sup> not providing a translation.

In line 8 we probably read ταῖς ἐπάλξεσι which is a possible reference to the protective walls (< ἡ ἔπαλξις 'means of defence, parapet, shelter'<sup>191</sup>) of the city.

The Nabataean section is quite damaged. At the beginning of line 10 we may only see the sentence *dnh ṣlm'* 'this is the statue', its addressee remaining unknown owing to the deterioration of the stone.

In line 11 the text is illegible as we can observe through the photograph. Initially, Father Vincent tried to reconstruct a segment beginning with the relative pronoun *dy* (it might also be a *nota genitivi*), some gaps (presumably four) due to the corruption of the stone, and the letters *nk-* followed by some signs that may be interpreted as *r*, *d*, *w*, *p* and *y'*. The possible combinations on the basis of which a new sentence might be read from this are innumerable.

In line 12 the name *ḥrtt* 'Aretas' occurs, which is probably a reference to the king Aretas IV.<sup>192</sup>

In line 15 the name *rb'l* appears, probably referring to the last king of the Nabataeans Rabbel II (70-106 CE). The presence of the latter would corroborate Milik's reading of line 16. Indeed, he reconstructs the sentence *'šryn wḥd bsywn snt 'šr wḥd[h]* '21 of Sīwan of the year 11' of Rabbel II's reign that corresponds to June 81 CE; presumably this indicates the date of the erection of the stele.

The fact that Jerash is the subject of a new state-building plan, after the erection of Zeus's temple in 69 CE,<sup>193</sup> may confirm what has already been mentioned above. The *terminus ante quaem* of the urban renovation project is established by an inscription, found in the northwest walls of Jerash, in which we read the name of the Syrian governor Lucius Ceionius Commodus who served as consul from 78 until 81 CE.<sup>194</sup> Between 69 and 80 CE the urban renovation project was completed and huge defensive walls were erected.195 If the reading ἐπάλξεσι (line 8) is right, the two directions 'west' and 'south' (lines 5-6) may refer to the boundaries of a plot of land that is ad-

190 *Quellen*, 202.

191 *LSJ*, 606.

193 Kraeling 1938, 375-6, no. 5.

194 Kraeling 1938, 397-8, no. 50. Commodus was the first of the *gens Ceionia* to become a consul.

195 Bowersock 1973, 138-9.

<sup>192</sup> According to the palaeography of the text, the predecessors Aretas I and Aretas II (169-96 BCE) are earlier, while Aretas IV (9 BC-40 CE) would be more appropriate for our inscription.

jacent to the walls of the city.196 There might be a coincidence between the year of Rabbel II's reign and the building of the walls of the city and therefore we can date this inscription to the period between 80 and 81 CE.

# **Umm al-Jimāl**

**15.** The two inscriptions are written in an altar that was found on the ground of the courtyard of the so-called 'House no. VI' in the eastern part of Umm al-Jimāl, a village about 17 kilometres away from Mafraq in northern Jordan.197 The column represents a gift to the local deity Dūšarā-A'ara.

As regards the Nabataean text, the shape of the letters could date the engraving of the altar to the 1st or the 2nd century CE, as the editors point out. (Plate IV, nos. 2-3)

```
Dimensions height 140 cm; height of head and base 22 cm and 37.5 cm; height of 
Nabataean letters 6-13 cm; height of Greek letters 6-10 cm 
Dating 1st-2nd CE (147 CE?)
Bibliography VIS no. 120; Levy 1869, 436; RES no. 1096; PPAES IIIA no. 238; 
   PPAES IVA no. 38; CIS II no. 190; Clermont-Ganneau 1906a, 215; Meyer 1906, 
   344; Littmann 1909, 383-6; JSNab no. 39; Cantineau 1930-32, 2: 23; Sourdel 
   1952, 60; IGLS 13,1 no. 9031; IGLS 21,5.1 no. 98
```
# Text and translation


4) son of 'Awī- 4) -anos

196 Kraeling 1938, 373.

197 Umm al-Jimāl rose in the 1st century CE as a rural suburb of the ancient Nabataean capital of Bosra. The Nabataeans are considered to be the first to build permanent homesteads in the area creating a settlement in which there was mainly a farming community and a trading outpost dependent on Bosra. Cf. De Vries 1998.


#### Commentary

The Nabataean noun *mšgd'* or *mśgd'*, absent in the Greek section, pinpoints the object dedicated to a deity that is a stone idol-altar.<sup>198</sup> The name is presumably a loanword or an interference from Arabic into Nabataean<sup>199</sup> < Ar. *saǧada* 'to bend until you touch the ground by your forehead in act of worship'.<sup>200</sup>

In line 3 of the Nabataean text we find the name *mškw* or *mśkw*  corresponding to Greek Μασεχος in lines 1-2. It is a common name that occurs in the Nabataean and Greek inscriptions from Central Syria and Mount Sinai.<sup>201</sup>

The Greek Μασεχος is also found in the form of Μασαχος. <sup>202</sup> It is the abbreviation of a theophoric name, *mšk'l*. <sup>203</sup> This compound form is recorded in Safaitic *msk-'l* and occurs as Μασαχηλω in a Greek inscription from Sī' as well;204 its meaning is '(god) has taken possession'205 and it is a birth name that represents the cultural uniformity of the Syrian regions of Hauran and Ḥarra.<sup>206</sup>

In lines 4-5 the name *'wyd'*, corresponding to Greek Αουειδανου, 207 appears. The insertion of the ν as a suffix is vague and probably

200 Lane, 1308. In Com. *sgd* 'to bow down, to prosternate' (*DNWSI*, 775).

204 Jaussen, Vincent 1901, 572. Cf. also *ESE* I, 337 no. 6.

205 Sartre 1985, 216.

206 De Vries 2009, 179.

<sup>198</sup> This noun, which is formed by the prefix *m-* indicating the place, is used in Official Aramaic meaning 'place of adoration', 'object serving as a permanent sign of adoration of the god to whom it is dedicated', as a monument or an altar dedicated to a god in order to recognise the gratitude for a favour or to obtain one. Cf. *DNWSI*, 663.

<sup>199</sup> Colombo 1994, 73.

<sup>201</sup> al-Khraysheh 1986, 115 and Negev 1991, no. 701. Also, *msk* (*ICPAN*, 545). It exists in Palmyrene (*PAT*, 437 = Stark 1971, 37 and 97).

<sup>202</sup> Cantineau 1930-32, 2: 118. A certain Eros son of Masekos – Ηρος Μασεχου (*IG-LS* 21,5.1 nos. 291, 303) is attested, the name Μασεχος is mentioned in *IGLS* 21,5.1 nos. 348, 349, 350, 351, 352. The form Μασαχος in *IGLS* 21,5.1 no. 347, the feminine Μασαχη in *IGLS* 21,5.1 no. 346.

<sup>203</sup> This name is present in the bilingual text in no. 28, in *PPAES* IVA no. 101 and in *RES* no. 2117.

<sup>207</sup> At Umm al-Jimāl there is an epitaph dedicated to a son of Α]ουιδανου (*IGLS* 21,5.1 no. 512).

Αουειδανος is simply an extended form of the name Αουειδος, <sup>208</sup> used in Arabic as *'Awīḏ*. <sup>209</sup> Therefore, *'Awīḏā* should be the diminutive form of a Sinaitic name, more precisely *'wdw =* Αυδος, hence the corresponding Arabic root *'ūḏ*. <sup>210</sup> The Arabic verb *'āḏa* means 'to take refuge, to take cover'.<sup>211</sup>

The last two lines 5-6 show the name *dwšr'*, equivalent to Greek Δουσαρει. This is the main deity of the Nabataean pantheon, worshipped above all at Petra, Mada'in Saleh, and Bosra. The name *Dūšarā* deserves a brief analysis: does it concern the deity's name or is it simply a god's epithet? From a strictly linguistic point of view, the appellative, in Arabic *ḏū l-šarā*, means 'who (the owner, the Lord) of al-Šarā' referring to the mountain range located near Petra;<sup>212</sup> it is a 'sacred' mountain range that, during the *Jāhiliyya*, was considered *ḥaram* 'prohibited' or *ḥima* 'protected'; the latter is interpreted as a place protecting animals, plants and fugitives.<sup>213</sup> Actually, the substantive *šarā* also has the meaning of 'road, tract of land, mountain'<sup>214</sup> and sometimes it is employed in the context of 'sacred land'.

As a result of the Nabataean cultural expansion, the cult of Dūšarā spread along the Mediterranean Sea and therefore Greek and Latin authors mention this deity in the form of Δουσάρης and *Dusares*. 215 Following the Roman conquest, Dūšarā continued to represent the main deity of the capital Bosra.

Starting from the middle of the 3rd century, four-year-games in honour of Dūšarā were established at Bosra, as can be seen from imperial medals bearing the legend *ACTIA DUSARIA*, along with the representation of a hand press.<sup>216</sup>

A temple dedicated to Dusares was found in Italy, at Pozzuoli, since the ancient Puteoli traded with the Near East; the Nabataean pres-

212 Peterson 2006, 23-4. Cf. also Wenning 2016.

213 Gawlikowski 1990, 2663.

214 Lane, 1545.

215 This Nabataean deity is quoted in the 9th century by the Arabic historian Hišam Ibn al-Kalbī in his *Kitāb al-Aṣnām*: "The Banū al-Hārith ibn-Yashkur ibn-Mubashshir of the 'Azd had an idol called Dū Sharā" (1952, 33); cf. also Healey 2001, 87.

216 *VIS* no. 120. Cf. also Stockton 1971, 52.

<sup>208</sup> It is present at Umm al-Jimāl in *IGLS* 21,5.1 nos. 203, 204, 206, 243. In the shape of Αουιδη, in *IGLS* 21,5.1 no. 205, *PPAES* IIIA no. 288. In addition, a certain Μασεχος, son of Αουειδου, always appears at Umm al-Jimāl in *PPAES* IIIA no. 271.

<sup>209</sup> *PPAES* IIIA no. 138.

<sup>210</sup> Blau 1862, 380. In pre-Islamic Arabic *'wd* 'return' (*ICPAN*, 447). The forms *'wyd'*  and *'wydw* (Beyer 1998, 164) are recorded in Hatraean as well as in Palmyrene (*PAT*, 438 = Stark 1971, 44 and 104-5).

<sup>211</sup> Cf. Al-Khraysheh (1986, 136) who renders *'Uwaiḏū* with 'Zuflucht Suchender' and Negev 1991, nos. 853, 854.

ence is attested starting from the middle of the 1st century BCE.<sup>217</sup>

In the last line of the Greek text there is the name Ααρρα, not present in the Nabataean part of the inscription. We are dealing with a socio-religious name of considerable importance that identifies the deity *''r'* worshipped at Bosra and associated with Dūšarā, as we can see in two inscriptions found at Bosra and Imtān.<sup>218</sup> The inscription coming from Imtān and dating to 93 CE describes Dūšarā, who is assimilated to the local deity Ā'arā. It is worth bearing in mind that Dūšarā always had a significant influence on the Nabataean royal house as is shown, for instance, in a legend depicted on a silver coin of king Obodas III, dated to 16 CE.<sup>219</sup> The transfer of the cult of Dūšarā and contextually the shift of the capital from Petra to Bosra under king Rabbel II (71-106 CE) are further proofs that corroborate the previous assumption.<sup>220</sup> As a matter of fact, the king established closer relations between Bosra and Dūšarā assimilating the latter to the local deity Ā'arā.<sup>221</sup> In addition, Rabbel II yearned to make Dūšarā the Nabataean national god, his own god and that of his city.<sup>222</sup>

The present inscription provides the first Greek transcription of the whole name of Dūšarā with the addition of *''r'* = Ααρρα. As regards its etymology various editors have proposed different theories about its meaning.<sup>223</sup>

Littman's thesis (*PPAES* IVA, 35) seems to be particularly reasonable; he asserted that *Ā'arā* derives from Arabic *ġarā* meaning 'good,


222 In the inscription from Imtān we read, lines 5-11: "Dūšarā and Ā'arā God of our Lord who is at Bosra. In the year 23 of Rabēl the king, the king of the Nabataeans". Cf. *editio princeps* in *VAS*, 169 no. 36.

223 Particularly, Dussaud and Macler (*VAS*, 169-70) proposed associating *''r'* with Hebrew *ēṣer* 'treasure' (the transition /'/ > /ṣ/ is common between Aramaic and Hebrew. Cf. Aram. *'r'* > Heb. *ereṣ* 'earth'), who was the biblical son of Se'ir, in Gen 36,21-30. This latter was a personification of the mountainous region that extends from the Dead Sea to the Red Sea, a territory inhabited by the Nabataeans. In contrast, Lidzbarski (*ESE* I, 330) initially interpreted the term as the equivalent of Latin *abundantia*, that is *ops*, but without philological explanations. At a later stage, he abandoned his first hypothesis proposing to relate *Ā'arā* to Hebrew *'rṣw* = Ar. *ruḍān*, rather than Safaitic *rṣw* (*ESE* II, 93). Clermont-Ganneau (1898c, 374) speculated that *Ā'arā* was the specific name of a god and in particular the form *O'ro* corresponds to the first element of the ancient Arabic deity Οροταλτ mentioned by Herodotus (*Historiae* 3.8).

<sup>217</sup> Museo archeologico dei Campi Flegrei 2008, 60-3. Cf. Lacerenza 1988-89 and *CIS* II no. 157.

beautiful',<sup>224</sup> 'beau, joli, bon',<sup>225</sup> from which, in turn, the name of a stone idol *al-ġariyyu* derives, under which Dūšarā was worshipped at Petra.<sup>226</sup> In fact, as recorded in the Byzantine encyclopedia Suda, Dūšarā (with the title of Θεός Ἄρης) was worshipped in the shape of a rectangular, rough black stone onto which the blood of the sacrificed animals was poured.<sup>227</sup>

The Greek form would be Αρρα(ς) <sup>228</sup> probably identified with Ares, the God of war. This theory may well be supported if the hypothesis, according to which the deity's name derives from the Arabic *af'al*  form *ġry* > *aġrā* 'dyeing' or 'anointing',<sup>229</sup> is true. In fact, *al-ġariyyu* < *ġry*, as stated above, is the name of a stone idol, worshipped by the pagan Arabs and stained or better dyed by the blood of the sacrificed animals; one of its derivatives, *ġariā*, identifies 'a certain red dye'.<sup>230</sup> This is the reason why the connection with the Greek God Ares, the god of war who 'stains himself by blood', would be appropriate both phonologically and culturally.<sup>231</sup>

**16.** The two inscriptions were found separately. The stone on which the Nabataean text occurs was situated in the wall of a house near the central church of Umm al-Jimāl, whereas the Greek part was uncovered in a courtyard not too far from the same church. Although the epigraphs are inscribed on two different stones, they bear the same content. (Plate V, no. 1)

**Dimensions** height of the Nabataean epigraph 28.5 cm, length 62 cm, thickness 17 cm; height of the Greek epigraph 37 cm, l. 57 cm, height of letters 4-4,5 cm (Φ 8 cm) **Dating** 3rd CE

**Bibliography** *VIS* no. 122; *CIS* II no. 192;<sup>232</sup> *RES* no. 1097; Littmann 1909, 386- 90; *PPAES* IIIA no. 2381 ; *PPAES* IVA no. 41; Cantineau 1930-32, 2: 25, no. 13; Sartre 1979, 253-8; *SEG* 29 no. 1604; Robert, Robert 1980, 478-9, 560; De Vries 1998, 33; Mascitelli 2006b, 231-7; *IGLS* 21,5 no. 499; Macdonald et al. 2015, 28-30

224 Lane, 2254.


227 Stockton 1971, 51-2; al-Shorman 2012, 43. In Suda (p. 192): Θευσάρης τουτέστι θεὸς Ἄρης ἐν Πέτρᾳ.


230 Lane, 2254.

231 Wenning 2001, 84-5.

232 In the *Corpus* the reading is uncertain and some letters are doubtful.

#### Text and translation


#### Commentary

The Nabataean script seems to be at a transitional stage towards the Arabic script. Indeed we may notice a tendency towards ligature,<sup>233</sup> although the *š* in line 1 is not attached to the preceding *p* and the *y*  in *gdymt*, in line 2, is not linked to the following *m*. <sup>234</sup> Moreover, we may observe that the *g* has already assumed a similar form encountered in the Kufic script.<sup>235</sup> Littmann (*PPAES* IVA, 38) suggested that the writer was an Arab who knew Nabataean as an archaic literary language since he used the final -*w* in common names, as we may see in *npšw* for npš, line 1, and *rbw* for *rb*.

The tomb is dedicated to a certain *phrw*<sup>236</sup> = Gr. Φερου; his father's name is *šly*, well-documented in other inscriptions.<sup>237</sup> In this case the Greek transliteration Σολλεος does not correspond to the well-known Συλλαιος, who was the minister of the Nabataean king Obodas III, also recurring in the inscription from Miletus (no. 49). It might be a solecism, even if it is recorded elsewhere.<sup>238</sup>

235 *PPAES* IVA, 38.

236 Cf. al-Khraysheh 1986, 151-2 for references; Negev 1991, no. 956, probably from Arabic *fihr* (*ICPAN*, 473). Cantineau (1930-32, 2: 136) translated it as 'pilon en pierre'. In Arabic it is a personal and tribal name, in fact Fihr is remembered as the direct descendant of 'Isma'īl and as another name of the Qurayš tribe.

237 Cantineau 1930-32, 2: 150; Negev 1991, no. 1137; al-Khraysheh 1986, 174-5.

238 In an inscription from Be'er Ševa' (Abel 1903, 428, no. 6), while the form Σολεος in Wadd. no. 1989, *PPAES* IIIA nos. 158 (= Wadd. no. 2003), 212 = Ar. *Sulaiḥ*. Cf. Wuth-

<sup>233</sup> Littmann 1909, 387. This inscription, along with that of Mar'al-Qays of al-Namāra (Louvre Museum, AO 4083; *RES* no. 483) dating to 328 CE, represents an important document of pre-Islamic history.

<sup>234</sup> The tendency to separate the letters is probably due to the *usus* of the monumental inscriptions.

The most important character is Gaḏīmaṯ, the king of the Tanūḥ (*gdymt*<sup>239</sup> *mlk tnwḥ* = Gr. Γαδιμαθου<sup>240</sup> βασιλεύς Θανουηνων). He was a sovereign, documented in the Islamic historiography, who reigned between the Euphrates and Syria as chief of the Tanūkh tribe<sup>241</sup> in the second half of the 3rd century CE (around 275 CE).242 As a consequence the stele should be dated to the end of the century.<sup>243</sup>

In the Greek text we read τροφεύς, a term wrongly inflected in the nominative along with βασιλεύς, whereas a genitive is to be expected; this refers to a title granted to Fehrō. In the Nabataean section τροφεύς is rendered as *rb*, translated by Littmann (1909, 387) as 'Erziehers', rather than 'tutor' (*PPAES* IVA, 38), giving it the significance of 'educator', 'rabbi', or better 'mentor'. Although the first meaning of τροφεύς is 'one who brings up, foster-father',<sup>244</sup> we do not exclude that it indicates the role of 'educator, instructor, teacher'. Sartre (1979, 253-8) thought that it was a late title conferred on members of the court of the Hellenistic kings, such as that of Seleucides or Ptolemaics. To his mind, we are dealing with a parental title of Seleucid influence; as Strabo also wrote (15.4.21), in the Nabataean kingdom there was a king who called ἀδελφός 'brother' his administrator – ἐπίτροπος. In the bilingual inscription of Miletus, the minister presents himself as a ἀδελφὸς βασιλέως, as reported in the Aramaic version *'ḥ malk'* and not *rb*. <sup>245</sup> Therefore, the τροφεύς, instead of σύντροφος 'foster-brother', is to be interpreted as 'père nourricier' (= θρεπτήρ), 'putative-father', as Sartre pointed out.<sup>246</sup> Jeanne and Louis Robert rejected Sartre's hypothesis247 without adding a plausible explanation of the term. So, it would appear to be a title linked to a specific task within the royal court, such as that of the tutor.

now 1930, 111 and 163.

240 Wuthnow 1930, 38 and 133.

241 In the Arabic historiography he is mentioned by al-Ṭabarī, *Kitāb aḫbār al-rusūl wa al-mulūk*, 2: 744-61; Perlmann 1987, 128-43. According to the tradition, Jaḏīmat al-Abraš was one of the first kings of al-Ḥīrah, an enemy of queen Zenobia (Zabbā') of Palmyre by whom he was killed (cf. al-Mas'ūdī, *Murūj al-ḏahab wa ma'ādin al-jawhar*, 222 and ff.; Ibn Qutayba, *Kitāb al-ma'ārif*, 216 and ff.).


<sup>239</sup> al-Khraysheh 1986, 53-4; Negev 1991, nos. 216, 217; Cantineau 1930-32, 2: 77 = Ar. *ǧaḏīmah*. It means 'cut off, amputated' < *ǧaḏama* 'to cut off' (Lane, 398), Syr. *gdam*  and Man. *gdm*; Akk. *gadāmu* 'to cut off hair' < Pro. Afro-As. \**gad-* 'cut, split' (Sokoloff 2009, 206; *CAD* V, 8; *AHw* I, 273; *HSED* no. 868). Cf. also Ge'ez *gadāmit* 'that which cuts, scissors' (Leslau 1991, 182-3).

Finally, the Greek translation of the Aramaic term *npš* with στήλη should be considered. It is worth remembering that the old Nabataean city of Umm al-Jimāl was rebuilt by Christians, as Littmann asserted (*PPAES* IVA, 40), who looted the ancient tombs in order to erect stone buildings; once the *stelae* were extracted from the ground, they were used as shelves or steps of staircases. With reference to the Greek inscription it would be difficult to establish whether it is a stone employed as a stela or a lintel. The Nabataean word *npš* could refer to both a *stele* and a tomb, meaning that we are dealing with a rare case in which στήλη identifies a burial, a tomb.

# **Umm al- Quṭṭayn**

**17.** The epigraph was found at Umm al-Quṭṭayn,<sup>248</sup> about 12 km east of Ṣabḥa on the north border with Syria. The inscription is engraved on a basalt, to be more precise on a grayish slab<sup>249</sup>, broken on the right side where the text is almost illegible. The bottom of the stone is not entirely inscribed.

**Dimensions** height 123 cm, length 33 cm, thickness 14 cm; height of letters 9-10 cm **Dating** last period of Roman Empire (4th-beginning of the 5th c. CE?) **Bibliography** MacAdam, Graf 1989, 191, no. 3; *SEG* 39 no. 1610; *IGLS* 21,5.1 no. 724

Text and translation

1) *[---t] br m[---]* 2) εαθ[---] 3) ηετ[---] 4) λ 1) [---] son of [---] 2) - 3) ag[e ---] 4) 30

249 MacAdam, Graf 1989, 191.

<sup>248</sup> It is an important site in the Hauran. Its ruins show a settlement dating to a period between the Bronze Age and the times of the Ottoman Empire, although its most prosperous period was during the Byzantine era. For a brief history of the village cf. *IGLS* 21,5 no. 319 with a comprehensive bibliography.

#### Commentary

The legible part of the text consists of four lines. In line 1 there are letters in the Nabataean script representing two names: the first is illegible, except for the final -*t*, and the second contains an initial *m-*. The only recognizable term is *br*.

In line 2 the Greek letters are clear, but unfortunately the remaining part of the inscription does not permit us to identify the name of the deceased. A distinct sequence of characters, such as -εαθ- and Εαθ- can be observed.

In line 4 we may reconstruct the age of the deceased: η ἐτ(ῶν) λ, '30 years old'.

After a further analysis of the photograph of the stone, provided by MacAdam, Bader believes that the inscription is totally written in Greek and the rebuttable Nabataean signs are barely more than the initials of the deceased's name:<sup>250</sup> O followed after a space by M, in order to render the feminine name Ομεαθη, attested in the region of Umm al-Quṭṭayn.<sup>251</sup>

According to the facsimile in the *editio princeps* the inscription would not seem to be bilingual, but after a careful analysis we may distinguish a *t*, at the beginning of line 1, followed by signs that resemble *br*, and at the end of the line a round drawing that looks like a final *m*.

In a space into which three or more characters may fit, the presence of a single O followed by a M (after a space)252 seems to be improbable.

The palaeography would date the inscription to the end of the 4th century or to the beginning of the 5th c. CE.<sup>253</sup>

250 *IGLS* 21,5.1 no. 330.

252 Bader (*IGLS* 21,5.1: 330) points out that: "Ce qui a été pris pour du nabatéen se lit d'abord comme un O puis, après un espace, un M".

253 *SEG* 39 no. 1610.

<sup>251</sup> In *IGLS* 21,5.1 no. 669, the inscription was not found and we may only resort to the drawing; in *IGLS* 21,5.1 no. 717 a masculine form Ομεθου occurs with a doubtful μ; in *IGLS* 21,5.1 no. 725 the name is not totally legible, in fact we can only make out ομε- (in line 1) and θεγ[υ] (in line 2) that leave open the possibilities of interpretations.

**Corpus of Nabataean Aramaic-Greek Inscriptions** Giuseppe Petrantoni

# Syria

# **al-Summāqīāt (al-Samāfīyāt)**

**18.** The epigraph was found in the western part of the village of al-Summāqīāt, a little further to the south of the western cemetery. The town, known today as al-Samāfīyāt, is located in southern Syria, in Hauran, about halfway between Bosra and Umm al-Jimāl in the district of Dar'ā. At the time of its discover, the stone lay above the foundations of a well-defined tomb that was partially dug up. The text consists of six lines.

**Dimensions** height 70.5-71.5 cm; length 29.5 cm; height of letters about 8 cm **Dating** unknown **Bibliography** *PPAES* IVA no. 10; *PPAES* IIIA, 2 no. 93

Text and translation

1) Oαρ-2) ε(τ)αθο-3) σ Σαιη-4) λου 5) ετ(ων) κς 6) *š(y)''l* 1) Oar-2) -e(t)atos

3) (son of) Saie-

4) -los 5) ag(e) 26 6) Ša(y)'ēl

#### Commentary

Lines 1-2. We find the deceased's name Οαρεταθος. 1 This is a Semitic name: in fact, the first part οαρετ- may correspond to *wrṭ*[*t*]*-* 2 and we may reconstruct it as *Warṭat*. 3 According to Littmann, it derives from the Arabic *warṭah* 'abyss',<sup>4</sup> also 'a low or depressed piece of ground in which there is no way or road directing to escape'<sup>5</sup> (compare the Hebrew *yaraṭ* with the same meaning<sup>6</sup> ).

Lines 3-4. The father's name, Σαιηλου, 7 corresponds to Nabataean *šy''l*. <sup>8</sup> The name is a theophoric and it is new in the Nabataean onomastics, but well-attested in the Safaitic: *s2''l*. It derives from the root *šy'* meaning 'to follow, to accompany',<sup>9</sup> Arabic *šā'* 'to spread, to become widespread', in the *fa''ala* form *šayya'a* 'to accompany, to follow'.<sup>10</sup> In Hebrew *sî'ā* is considered a loanword that means the 'council' of the Syrians and Caldeans as well as 'congregation, assembly'<sup>11</sup> (< *sy'* 'to accompany, join a caravan, escort'). On the basis of such a premise, it is difficult to establish the real pronunciation of the initial *š* of the name.


8 Negev (1991, no. 1125) suggests that it derives from Arabic *su'āl* 'cough'. Conversely, Macdonald (1999, 285) argues that it is the Safaitic name *s²''l* in which the first element is a divine epithet as in Nab. *šy'-'l-qwm* and Saf. *s²'-h-qm*.

9 Cantineau 1930-32, 2: 150. Negev (1991, no. 1126) compares the Greek form with Nabataean *šy''lhy* < Ar. 'witch, hag'. Here, the hypothesis to translate the root 'the witch of Allah' is absurd (Macdonald 1999, 285). Cf. pre-Islamic Arabic *šy'*, *shai'* 'companion, follower' (*ICPAN*, 364).

10 The term *šay'* also means 'the one who follows', 'companion', the noun *šā'ah* identifies the 'wife', 'because she follows, or conforms with, [the wishes of] her husband'. Cf. Lane, 632. In Syriac *sī'tā* denotes 'company, band, troop, body of soldiers' < *saya'* 'to aid' (Sokoloff 2009, 1006). In OfA, BA and Pal. *sy'* <\**śy'* simply means 'to help' (*PAT*, 392).

11 Gesenius 1846, 585 = *DGes*, 884.

# **al-Ġāriyah**

**19.** The inscription was found in the village of al-Ġāriyah, about 25 km south-east of Bosra, in the district of Dar'a. It is a funerary epigraph in which the deceased's name is followed by the name of his father.

**Dimensions** height 75 cm; length 35 cm **Dating** s.d. **Bibliography** *VAS*, 185-6 no. 59; *RES* nos. 85 and 1095; *PPAES* IIIA, 2 no. 194; *ESE* I, 332; *ESE* II, 254; Clermont-Ganneau 1901b, 172; Littmann 1909, 381-3

Text and translation<sup>12</sup>

1) Pαββα-2) νης Mο-3) (ε)ιθου 4) *rb' [br]* 5) *m'y(t)w* 1) Rabba-2) -nes (son of) Mo-3) –(e)itos 4) Rabbā [son of] 5) Mo'ey(ṯ)ō

# Commentary

The two names are of Semitic origin. The first, in lines 1-2, is Ραββανης13 that corresponds to Nabataean *rb'*, in line 4, but the second appears less legible, judging from the copy of the stone; in fact, at the end of the line there appear a vertical sign (*n* ?) and a detached barely sketched horizontal one over another vertical sign (' ?). Probably the first editors, Dussaud and Macler (*VAS*, 185), read *rb'*[*n'*] in comparison to the name Ραββουνί 'my master', a clear calque from Aramaic which is found in the Gospels (cf. Matt 22,8; Mark 10,51 and John 20,16). Moreover, they exclude the possibility of restoring *rbyb'*<sup>14</sup> or *rfyn'*, as proposed by Clermont-Ganneau.<sup>15</sup> The name *rb'* with the

14 As it is recorded in *CIS* II no. 287.

<sup>12</sup> The transcription here presented follows that of Lidzbarski (*ESE* II, 254).

<sup>13</sup> Wuthnow 1930, 96 and 166.

<sup>15</sup> It is present in Hauran (Clermont-Ganneau 1901a, 122, no. 1); he compared *rfyn'* to the Greek transcription Ῥουφῖνος, a popular name in the Greek-Syrian onomastics.

suffix *-n'* = -ανης could be a simple Greek calque, or transcription, of a Semitic name. It may derive from Hebrew *rbn* < *rab* 'chief, teacher' (in Aramaic too), that is *Rabban*, a title conferred on a scholar, as well as *rabbānā*, a Babylonian title.<sup>16</sup>

In the revised version of the epigraph<sup>17</sup> the Nabataean text is read as follows: *rb'* [*br*] *m'ytw*. If the reading is correct, we get the Nabataean name *rb'* 'chief, teacher'.<sup>18</sup> So, the Greek suffix -ανης would be an extended form of *rb'* as it appears in *'wyd'* = Αουειδανος (cf. no. 15 and *RES* no. 1096) and *rb'* seems to be a diminutive of *rb'l* – *Rab'ēl*, even though this hypothesis is not convincing.

The second name is Μο(ε)ιθου<sup>19</sup> = Nab. *m'y*(*t*)*w* in which the *t* is uncertain; initially Dussaud and Macler (*VAS*, 185) suggested reading *r* in spite of *t* in order to restore the well-attested name *m'yrw*<sup>20</sup> = Μοαιερος<sup>21</sup>, Μογαιρος etc. However, the Greek form with θ excludes the latter possibility assuming the restitution of *m'ytw* which is to be compared with Arabic *muġayṯ* 'one who helps, aids, assists, helper, rescuer'<sup>22</sup> and Greek Μογιτος. <sup>23</sup> The name derives from the Arabic root *ġāṯa* 'to help, to aid, to assist, to rescue'; a particular case is the Nabataean-Arabian Lion-God *y'wt* – *Yaġūt* worshipped during the pre-Islamic period by the Yemenite tribe of Maḏḥij<sup>24</sup> and mentioned both in the Bible, as *īə'ûš* (in the LXX Ιεους),<sup>25</sup> and in the Koran, as *yaġūṯ*. 26

17 It is published in *RES* no. 1095 and *ESE* II, 254.

18 Cantineau, 1930-32, 2: 145; Negev 1991, no. 1051.

19 It is attested at Izr'a or Zorava, in the region of al-Lajāh, southern Syria, in an inscription from the church of Saint George (cf. Wadd. no. 2483). Μοειθος is also present at Umm al-Jimāl (cf. *IGLS* 21,5.1 nos. 144, 366, 367). Other forms are Μοεαθος, Μοεειθον (Wuthnow 1930, 77 and 156).

20 al-Khraysheh 1986, 110 and Negev 1991, no. 667.

21 Cf. *IGLS* 21,5.1 no. 359. The name Μοαιρος in *IGLS* 21,5.1 no. 361, while Μοεαρος in *IGLS* 21,5.1 nos. 363, 363, 364, 365. Wuthnow 1930, 76 and 157.

22 Pre-Islamic Arabic < *''yr* or *mġyr* < *ġyr* (*ICPAN*, 460 and 558-9). Cf. Cantineau 1930-32, 2: 117 and Negev 1991, no. 668. See also *PAT*, 436 = Stark 1971, 34 and 96 for Palmyrene.

23 It is encountered in Wadd. no. 2203c. Cf. Wuthnow 1930, 77 and 156.

24 Strawn 2005, 214.

25 Cf. Gen 36,5-14-18; 1Chr 1,35; 7,10; 8,39; 23,10-11; 2Chr 11,19.

26 LXXI, 23.

Cf. *LGPN* I, II, IIIA, IIIB, IV, VA, also Pape 1911, 1316.

<sup>16</sup> Cf. *rbn*, *rbn'* in Pal., Qumran, Sam. meaning 'chief, head man', 'scholar, doctor' in Gal. and Syr. (Jastrow 1903, 1444; Payne Smith 1903, 526; *DNWSI*, 1055; *PAT*, 409; Tal, *Sam*, 809a), *rabbānan* 'the rabbis', 'our teacher' in JBA. For a close examination of the employment of the term in the Targumic passages see *DJBA*, 1055b. The name *rb'* is also attested in pre-Islamic Arabic (*ICPAN*, 263). In the Hatraean onomastics *rbn* is recorded (Beyer 1998, 58 and 184) meaning 'chief', 'Meister'. In Palmyrene we have *rb'n*  and *rbn* (*PAT*, 440 = Stark 1971, 49 and 111) 'teacher'.

#### **Bosra**

**20.** The inscription was found in the cemetery of Šuhādah, to the southwest of Bosra Citadel.<sup>27</sup> It is engraved in the lower part of the basalt that probably had the shape of an eagle. (Plate V, no. 2)

**Dimensions** height 35 cm, thickness of the base 38 cm, length of the base 91 cm, height 7 cm; average height of the letters 5 cm **Dating** 2nd-3rd CE (?) **Bibliography** Milik 1958, 235-41, no. 3; *IGLS* 13,1 no. 9003; Vattioni 1985, 769; *Quellen*, 183

Text and translation

1) Μοαινος 2) *m'ynw br zbdy 'bd nšr' lqws* 3) *ḥlypw br tym['] 'mn'*


Commentary

The author's name, Μοαινος (= Nab. *m'ynw*), is of Semitic origin meaning 'helper',<sup>28</sup> while his father's name, *zbdy*, is considered an Arabic hypocorism from *zubayd* < *zabd* 'gift, present.<sup>29</sup>

<sup>27</sup> Bosra is located 140 km south of Damascus. It was the first Nabataean city in the 2nd century BCE under king Rabbel II. After the conquest of Cornelius Palma, a general of Trajan, in 106 CE, Bosra was renamed *Nova Trajana Bostra* and was the residence of the *legio III Cyrenaica*. It was made capital of the Roman province of *Arabia Petraea*. The city flourished and became a major metropolis at the crossroads of several trade routes, namely the *Via Traiana Nova*, a Roman road that connected Damascus to the Red Sea. For an in-depth analysis of the story of the Citadel, see Sartre 1985.

<sup>28</sup> Al-Khraysheh 1986, 110; Negev 1991, no. 666; Wuthnow 1930, 77 and 156.

<sup>29</sup> *ICPAN*, 294. Cantineau 1930-32, 2: 91; al-Khraysheh 1986, 72; Negev 1991, nos. 367, 369. Greek forms are: Ζαβδαιος, Ζοβεδος (Wuthnow 1930, 48 and 137). According to Macdonald (1999, 275): "the root *z-b-d* is common in the formation of names in Aramaic, Hebrew and Arabic. There seems no reason to prefer an Arabic or Ancient North Arabian etymology to an Aramaic one". In Qumran, PTA, Sam., Syr. it takes on the meaning of 'to provide, to give' (Jastrow 1903, 377; *DJPA*, 171a; *DJBA*, 397a; Tal, *Sam*, 224). In Palmyrene we have *zbdy*, *zbd*, *zbd'* 'gift'; the latter is found at Dura Europos (*PAT*, 434 = Stark 1971, 17, 85 and 143). The root is also present in Hatra with the same meaning (Beyer 1998, 83 and 174).

The sculptor's name is *ḥlypw*, <sup>30</sup> son of *tym'*, 31 but we must not confuse it with *tym* whose *'* is omitted for haplography, as confirmed by the shape of *m* that is not a final *m*. <sup>32</sup> According to Milik (1958, 236), it is possible that the author of the inscription is the father of a certain Rabbū, son of Ḥulayfū, who also sculpted an eagle found at Hebron.<sup>33</sup>

The fact that the author sculpted an eagle dedicating it to the god *Qws* deserves particular attention. It is a deity worshipped among the Edomites appearing in the form of an eagle in the hellenised capital Bosra of the *Provincia Arabia*. 34 The name *Qws* derives from the Semitic root indicating 'bow'<sup>35</sup> and it would designate the god of war, the archer god.<sup>36</sup> It is also used as a theophoric.<sup>37</sup> In Nabataean there is another mention of the cult of Qōs found in an inscription from Khirbat al-Tannūr (18 km west of Jerusalem) dated to 1st c. BCE-1st c. CE.38 A further proof of the cult of Qōs among the Edomites is provided by Josephus who remembers the figure of Kostobaros, a priest of the god Koze (Gr. Κοζε), a deity worshipped by the Edomites before their conversion to the Judaism by Hyrcanus.<sup>39</sup> The connection Qōs-Koze appears credible even though it has generated some disagreement.<sup>40</sup>

As regards the dating, Milik (1958, 237) proposed to date the inscription to the 2nd or the 3rd century CE.

31 al-Khraysheh 1986, 186; Negev 1991, no. 1212. It is frequently used in Palmyrene as *tym*, *tym'* and *tymw* (Stark 1971, 54-5 and 117) and in Hatrean (Beyer 1998, 167).


35 Cf. Pro. Afro-As. \**ḳawas-*/*ḳayas-* 'bow, arrow' (*HSED* no. 1560). Knauf (1999, 676) believes that an original bi-consonant *qs* became tri-consonat by the addition of -*t* in Akkadic, Ethiopic, Canaanite and Aramaic, but by an infixed -*w* in Arabic.

36 Knauf 1999, 676.

37 See Kelley 2009, 255-80.

38 The inscription is written on a small stele that was found by N. Glueck and published by Savignac (1937, 401-16). It goes as follows: "*[d]y 'bd qsmlk lqs 'lh ḥwrw'*", '[stele tha]t Qūsmilk made for Qūs, god f *HWRW'*' (Kelley 2009, 259 who quotes Bartlett 1989, 200). Conversely Milik (1958, 237) reads "[stele] that Qosmilik made for Qosallah. Ḥūrū (the sculptor)".

39 Josep., *AJ* 15.7.9.253-4.

40 Milik (1958, 239) doubted that the connection between the Greek transcription of the name and the Nabataean one is proper. By contrast Wellhausen (1897, 67, 81, 146) suggested identifying Qūs with the god Quzaḥ, genius of the arch of the sky among the ancient Arabs, but this conjecture is improbable from the phonetic point of view. In Arabic the rainbow is called *qaws quzaḥ*, literally 'arch of the god Quzaḥ', probably referring to a Pre-Islamic cultural heritage. Indeed, in pre-Islamic Arabic we find *qwst* 'bow' (*ICPAN*, 491).

<sup>30</sup> al-Khraysheh 1986, 83. Cf. no. 3.

#### **Jamarrīn**

**21.** The inscription is engraved on the right part of a basalt architrave found by Milik in the modern Arabic cemetery of Jamarrīn, a small village located 3 km to the north of Bosra.

The text is placed in a *tabula ansata* decorated with a metope representing a rose; this is a common element in the monuments of the Roman period.<sup>41</sup>

The inscription is damaged on the left part of the stone in which two or three letters are missing, in fact we may only reconstruct the beginning of the four Greek lines. (Plate V, no. 3)

**Dimensions** height of the stone 39 cm, length 90 cm; *tabula ansata* 60 cm; height of Greek letters 3 cm; height of Nabataean letters 4 cm **Dating** 2nd-3rd CE (?) **Bibliography**: Milik 1958, 241-2, no. 4; *SEG* 20 no. 408; *IGLS* 13,1 no. 9412

# Text and translation<sup>42</sup>

1) [Δη]μητρις Βορδου 2) [επ]οηεεν<sup>43</sup> το ανα-3) -[πα]υμα εαυτω 4) [και] τυς<sup>44</sup> υιοις 5) *nyḥ' dy 'bd dmtrys br [brdw]*

1) [De]metris (son of) Bordos 2) [ma]de (the place of) re-3) -[s]t for himself 4) [and] for his sons 5) (place of) rest that Demetrīs son of [Bordō] made

Commentary

The author is a certain Δημήτρις (< Δημήτριος),<sup>45</sup> a well-documented name in the Greek onomastics;46 on the contrary, the Aramaic equiva-


lent *dmtrys* is new in the Nabataean onomastics.<sup>47</sup> The father's name, Βορδος<sup>48</sup> (= Nab. *brdw*),<sup>49</sup> is of Semitic origin.<sup>50</sup>

In line 2 the author wrote –ΟΗΕΕΝ, reconstructed as [ἐπ]όησεν<sup>51</sup> and followed by the substantive τὸ ἀνά[πα]υμα 'resting place',<sup>52</sup> the latter was rare in Syria during the Roman period.53 In the Nabataean section we find the term *nyḥ'* 'rest', 'calm', mentioned for the first time in Nabataean, but common in other varieties of Aramaic.<sup>54</sup>

#### **al-Mu'arribah**

**22.** The stele was found on top of the door of a farmer's house in al-Mu'arribah, a village a few kilometres to the northwest of Bosra. The text presents a hollow from line 2 and reports two names.

```
Dimensions height 90 cm; length 30 cm; characters of line 1, 10 cm; lines 2-3, 5 
cm; Nabataean lines 11-14,5 cm 
Dating 1st BCE-1st CE
Bibliography RES no. 1094; PPAES IIIA 5, no. 615; PPAES IVA no. 92; Littmann 
   1909, 380-1
```
Text and translation

1) Yφφαλ[ος] 2) Θαι[μ]- 3) ου

47 Negev 1991, no. 280.

48 Wuthnow 1930, 37 and 132.

49 In the *editio princeps* the name is reconstructed since it is not visible because of the damaged basalt.

50 Negev (1991, no. 196) links it to Arabic *Burd*, *Buraid* probably from *baruda* 'to become cold' (see also *ICPAN*, 101). The name is also attested in Hatraean as *bryd'* meaning 'dark', 'spotted, 'dunkel', 'gesprenkelte' (Beyer 1998, 80 and 156, 171).

51 In the photograph of the inscription the lunate sigma blends in with the ε, therefore it should be the case that the author wrote the correct segment -οησεν.

52 *LSJ*, 115.

53 Cf. Wadd. nos. 2300, 2391.

54 The form *nyḥ*, *nyḥ'* \*/nyāḥ, nyāḥā/ is attested in Jewish Palestinian Aramaic, CPA, JBA, Man., Syr. and in Pal. (Jastrow 1903, 904; Payne Smith 1903, 338; *DJPA* 349a; *DN-WSI*, 729); in the latter the feminine form *nyḥt* is present (*CIS* II no. 3907). The root derives from Common Aramaic *nwḥ* 'to rest' and it is used in Hebrew *nûḥ* and Arabic *nāḫa*  'to kneel down as a camel', hence *munāḫ* the place in which the camels lie. In Phoenician the root *nḥ* means 'to rest, to be at peace' and the nouns *nḥt* and *mnḥt* mean 'peace' (see Krahmalkov 2000, 295 and 327-8).

4) *ḥpl[w br]* 5) *ty(m)[w]* 1) Uffalo[os] 2) (son of) Tai[m]- 3) -os 4) Ḥuffālō [son] 5) of Tay(m)ō

#### Commentary

In line 1 there is the name Υφφαλ[ος] that corresponds to Nabataean *ḥpl*[*w*] in line 4. This name is difficult to explain because it can be interpreted in various ways. It seems to be constituted by the Arabic root *ḥafala* 'to gather, to flow, to pour, to course'.<sup>55</sup> According to the Greek transcription of the name, the vocalization suggested by the editors would represent the form *ḥuffāl* meaning 'multitude', also 'numerous, crowded' referred to the name that would have been given to the child by parents hoping for his offspring to be prosperous.<sup>56</sup> If this hypothesis is right, the Greek transliteration uses Y to render the Nabataean vowel /u/, but here some problems of interpretation arise. According to the historical period in which the Greek *koinè* develops, the υ has already changed its phonetic value to /i/ (< Classic Greek /ü/), therefore Aramaic /ḥ/ would have had the vowel /i/ rather than /u/ and the name should be read *ḥiffāl* not *ḥuffāl*. Nevertheless, in Arabic *ḥiffāl* has no meaning; conversely, if *ḥuffāl* is correct, the Greek transcription would be expected to be Ουφφαλ[ος] since in the *koinè* the diphthong ου = /u/ was phonetically productive.

In line 1 there is not any sign that could be ascribed to a possible initial o- for Ου or sufficient space to include it; in addition, the inscription shows the deceased's name with the father's name proving that it is not a fragmentary text.

So, Υφφαλ[ος] should be an atypical case of vocalization; in fact, contrary to Attic-Ionic dialect and *koinè*, only Beotian Aeolic of the 4th-3rd c. preserved *ŭ* and *ū* as back vowels, phonetically /u/ and /ū/ (ου in Attic-Ionic and *koinè*), and not Attic /ü/ or *koinè* /i/.<sup>57</sup> Does this therefore constitute a mistake of the Aramaic speaking author who did not know Greek?

<sup>55</sup> Negev (1991, no. 475) proposes *ḥafūl* 'assiduous', on the basis of *ICPAN*, 195 *ḥfl*, *ḥfīl*. Cf. also Wuthnow 1930, 117 and 143.

<sup>56</sup> *PPAES* IVA, 69.

<sup>57</sup> Meillet 2003, 124.

If we acknowledge the phonetic value /i/ of υ, according to Spielberg the name is a variation of Greek Ἵππαλος 'rider, horseman',<sup>58</sup> that is usually used in Egypt.<sup>59</sup> If we accept this hypothesis, the name has Greek origins and in Nabataean it should be read /ḥippalō/. In addition, a name is registered as *hypalus* by Pliny the Elder (*HN* 6.100), as well as it is recorded in the *Periplus Maris Erythraei*<sup>60</sup> as ἵππαλος indicating the Monsoon; the two forms would be a corruption of the most ancient ὕφαλος. 61

As regards the father's name, Greek θαι[μ]ου62 outstretches between lines 2 and 3 presenting the μ out of the engraved surface that we may restore thanks to the Nabataean equivalent *tym*[*w*].

### **al-Ġāriyah al-Šarqiyah**

**23.** This inscription was found in a wall of the village of al-Ġāriyah al-Šarqiyah,<sup>63</sup> about 24 km away from Dar'a District.

In the *editio princeps* Dussaud and Macler (*VAS*, 205) do not consider this inscription as truly bilingual,64 but rather as simply monolingual Greek. Lidzbarski (*ESE* I, 335), instead, reports a corresponding Nabataean part of the Greek text. Unfortunately, we have no precise knowledge of the real nature of the inscription.

**Dating** unknown **Bibliography** *VAS*, 205 no. 96; *ESE* I, 335 no. 96

58 Pape (1911, 560) translates as 'Rössel'.

61 For a close examination cf. Mazzarino 1982-87 and De Romanis 1997; this latter scholar thinks that ὕπαλος < ὕφαλος is a correct genuine form of the Greek denomination of the wind as a result of a technical term that was employed in Ionic science (particularly cf. De Romanis 1997, 688-9).

62 Cf. Pape 1911, 477 and Wuthnow 1930, 52-3 and 175.

63 The town is 4 km away from the other al-Ġāriyah al-Ġarbiyah that is located to the west, as the name suggests.

64 They assert that the final part of the Greek text, (ε)ἰς θεὸν [Ἀ]μέρου, corresponds to the Nabataean formula *l'lh 'mrw*.

<sup>59</sup> *PPAES* IVA, 69. It is also attested in Greece, at Eretria (*LGPN* I), and in Macedonia (*LGPN* IV).

<sup>60</sup> It is a Greco-Roman periplus written in Greek describing navigation and trading opportunities from Roman Egyptian ports along the coast of the Red Sea, and others along Northeast Africa and the Sindh and South western India. The document has been ascribed to the mid-1st-century CE.

#### Text and translation


# Commentary

The name Γαφαλος, transcribed as *gḥplw* by Lidzbarski, presents difficulties of interpretation because of its rare use in the Greek onomastics<sup>65</sup> and its absence in the Nabataean. It is probably a Semitic name<sup>66</sup> that assumed a Greek form; it might be assumed to from the Arabic *ġafala* 'to pay no attention, to neglect, to omit', but this hypothesis seems to be improbable.

The father's name Χαρητος corresponds to the well-attested pre-Islamic Arabic name *ḥrtw*<sup>67</sup> < Ar. *al-ḥāriṯ* 'farmer, ploughman', also 'lion'.<sup>68</sup>

As regards Αμερος, Nabataean *'mrw* identifies a name found both in Palmyrene<sup>69</sup> and in the pre-Islamic Arabic inscriptions in the shape of *'mr*<sup>70</sup> (< Ar. *'amara* 'to live, to dwell').<sup>71</sup>

The presence of the vowel ε in the Greek transliteration indicates the Arabic active participle *'āmir*, or better, the Nabataean Aramaic active participle /'āmer/ 'the one who is living, dwelling'.<sup>72</sup>

67 *ICPAN*, 182-3. In Syriac *ḥrat* means 'to hollow out, to cut to pieces, to incise', 'to plough' (Sokoloff 2009, 496)

68 Cantineau 1930-32, 2: 100. Negev (1991, no. 493) proposes other two solutions: the first is that the root comes from Arabic *ḫirrīt* 'expert guide'; the second is that it derives from Arabic *ḫurṯiy* 'useless articles of tent furnishing'. In Greek we also have the variation Χαρήτας (Pape 1911, 1670; *LGPN* II); cf. also Wuthnow 1930, 120 and 143.

69 *PAT*, 438 = Stark 1971, 45 and 106. Hatraean records *[']mrt* (Beyer 1998, 102 and 164).

70 *ICPAN*, 436.

71 In Qumran Aramaic and in Christian Palaestinian Aramaic, Syr. and Man. *'mr* 'to dwell, inhabit' (Jastrow 1903, 1090; *DNWSI*, 873). Cantineau (1930-32, 2: 133) compares it with Arabic *'mr* = Gr. Αμρος (Wadd. no. 2028 *et passim*).

72 Aμερου is found in *AGLI*, 35, 49, 111, 112, 113, 117, 121, 130. Αμερος at Umm al-Jimāl in *IGLS* 21,5 nos. 188, 189, 190, 191, 211.

<sup>65</sup> We have an occurrence in Hauran (*ESE* I, 219 no. 53) and two in Arabia (*SEG* 7 no. 1108 and 1115).

<sup>66</sup> Wetzstein (*AGLI*, 349 and 366) points out that it derives from Arabic *ǧaḥfal* 'big army, legion, swarm', in addition it also appears as a variation of Ταφαλος < Ar. *ṭifl* 'child'. In Pre-Islamic Arabic *jḥfl*: *Jaḥfal* is attested (*ICPAN*, 153).

#### **al-Buṯaynah**

**24.** The inscription was engraved on a pillar that was smuggled and subsequently reused as an architrave. This altar was found in the village of al-Buṯaynah, 7 km north of Šaqqā in the province of al-Suwaydā'.<sup>73</sup>

In the upper part of the stone we find four lines written in Greek and in the lower part we may distinguish some damaged Nabataean letters.

**Dating** unknown **Bibliography** Dunand 1932, 410, no. 41

Text and translation


### Commentary

The Greek text could be incomplete. The damaged Nabataean letters represent the names of the author and of his father: *mlkw br bdr*/*bdrw*.

The name Μαλαχος is a variation of the well-known Μαλεχος<sup>74</sup> and Μαλιχος75 from the common North-West Semitic root *mlk* that identifies the king or 'one who rules', also the 'holder', the 'owner' (e.g.

<sup>73</sup> The town is located at Jabal al-Durūz, literary 'Mount of Druze', also called Jabal al- 'Arab. Al-Buṯaynah is placed in the ancient 'Arḍ al-Baṯaniyyah 'land of Batanaea', which is the Hellenized/Latinised form of the biblical Bašān (Heb. *ha-bāšān* 'soft, sandy soil'), hence the name al-Buṯaynah that is, according to Dussaud and Macler (*VAS*, 138), a diminutive form of *baṯnah* 'level and soft soil, soft sand', even though the root *bṯn* does not exist in Arabic. Presumably the toponym is a calque from Hebrew in which /š/ > Ar. /ṯ/. The town has archaeological remains of Nabataean, Roman, Byzantine and Islamic eras.

<sup>74</sup> *IGLS* 21,5 nos. 342, 428, 718.

<sup>75</sup> Wuthnow 1930, 70; *LGPN* II; *IGLS* 21,5 no. 39.

Ar. *mālik*). The α, present in the Greek transcription, expresses the Arabic/Aramaic active participle.

As far as Βαδουρος = *bdrw* is concerned, it is a Semitic name wellknown in the Greek and Nabataean epigraphy76 being of Arabic origin, from *badara* 'to hasten, to hurry', 'to come suddenly',<sup>77</sup> also 'to be full' (referring to the moon)78 and in Common Aramaic *bdr* 'to disperse, to scatter'.79 The presence of α in the Greek transcription permits us to establish that we are dealing with another active participle, from Ar. *bādir* 'one who is hurrying', also 'one who is dispersing' (< \**bāder*) should the root be connected to Aramaic. The diphthong ου in the Greek rendering sounds strange since it does not correspond to the second vowel of the Semitic root; therefore, a probable \*/bādūrō/ may be considered as a variation of the pronunciation of the Nabataean active participle.

#### **al-Suwaydā'**

**25.** This is a lost inscription that was carved into a *cartouche* located on the facade of a memorial, since destroyed.80 It was a mausoleum, placed in the western part of al-Suwaydā',<sup>81</sup> on the north bank of the *wādī*, that was depicted and drawn by all travellers who passed by this place.<sup>82</sup>

<sup>76</sup> Cf. Cantineau 1930-32, 2: 70; *ICPAN*, 97; Negev 1991, nos. 173, 174. Other forms are: Βαδρος (Wadd. no. 2354; Wuthnow 1930, 31 and 170; *LGPN* I) and Βαδαρου (Wadd. no. 2330). At al-Suwaydā' the name *bdr* appears in a basalt altar (cf. Littmann 1904, 93-4).

<sup>77</sup> Lane, 165.

<sup>78</sup> Wetzstein (*AGLI*, 347) translates this name as an improbable 'Vollmond', 'full moon' < Ar. *badr*.

<sup>79</sup> See *DNWSI*, 145.

<sup>80</sup> The building was destroyed with dynamite during the Ottoman period and no fragments were found.

<sup>81</sup> It is the capital of al-Suwaydā' Governorate located in southwestern Syria, close to the border with Jordan. The city was founded by the Nabataeans as *Suada*, lit. 'the black', due to black blocks of basalt that were used in constructing buildings. It became known as Dionysias (Ancient Greek: Διονῡσιάς) in the Hellenistic period and the Roman Empire; the city is situated in a famous ancient wine-producing region.

<sup>82</sup> For further bibliography about the travellers who stopped at the site on their way cf. Sartre-Fauriat 2001, 196-7. The monument, which was built on step blocks with a pyramid on the top, consisted of Doric columns. In the intercolumniation low relief weapon trophies were carved, whereas the trabeation consisted of an architrave with a Doric frieze of triglyphs and smooth metopes on top of it and a contoured cornice.

**Dimensions** of the monument: height 10.50 m, height of the pyramid 5.10 m **Dating** 1st BCE or beginning of 1st CE

**Bibliography** *CIG* no. 4620 (Greek part); Vogüé 1864, 284, no. 1; Wadd. no. 2320 (Greek part); *VIS*, 89-92; Schröder 1884, no. 5; *CIS* II no. 162; Brünnow, Domaszewski 1909, 3: 98-101; Sartre-Fauriat 2001, 1: 197

Text and translation

1) *npšh dy ḥmrt dy bnh lh 'dynt b'lh*

2) Οδαινατος<sup>83</sup> Αννηλου ωικοδομησεν την στηλην Χαμρατη τηι αυτου γυναικι

1) Tomb of Ḥamraṯ who built for her 'Oḏaynaṯ her husband 2) Odainatos (son of) Annelos built the tomb for Chamrate his wife

#### Commentary

In line 1 *npšh*, in the emphatic state, corresponds to Greek στήλη (as at Umm al-Jimāl in no. 16) identifying the memorial. The addressee of the tomb inscription is *ḥmrt* = Χαμρατη, an Aramaic name meaning 'red' (Com. *ḥmr'* 'wine' < *ḥmr* 'to become wine or bitumen colored';<sup>84</sup> Ar. *ḫamr* 'wine').

We cannot be sure of the identity of the husband, named *'dynt* = Οδαινατος. The name, which is found both in Nabataean<sup>85</sup> and in Greek<sup>86</sup> epigraphy (whose meaning is linked to Arabic *uḏaynah* 'small ear' < *uḏn* 'ear' and Com. *'dn'*), should refer to Odaenathus the Arab,<sup>87</sup> probably a phylarch of the region before the Roman occupation; he was chief of *Beni Samayda* or *Beni Amila al-Amālik* as reported in the Arabic sources.<sup>88</sup> This was Waddingotn's hypothesis, who found two

85 Negev 1991, no. 36; al-Khrayshesh 1986, 27.

<sup>83</sup> In *VIS*, 89 and *CIS* II no. 162 the name is written with θ: Οδαιναθος.

<sup>84</sup> The connection with the bitumen is justified by the fact that the root, also used in Hebrew, identifies the boiling, the fermentation, hence *ḥēmār* 'bitumen' (= Gr. ἄσφαλτος). The Dead Sea erupted asphalt from the middle of the lake as if the waters boiled, and once it became hard, given the heat of the sun, it remained on the Dead Sea surface, therefore it is also called *lacus asphaltites*, as Strabo (16.2.42-3) and Tacitus (*Hist.* 5.6) pointed out.

<sup>86</sup> Wuthnow 1930, 87 and 124; Pape 1911, 1035.

<sup>87</sup> We rule out the possibility of connection with Odaenathus, the founder king of the Palmyrene Kingdom (died 267/268 CE).

<sup>88</sup> *VIS*, 91.

Greek inscriptions bearing this tribe's name in the same region.<sup>89</sup> The father's name is present only in the Greek text as Αννηλου = Nab. *ḥn'l* (cf. nos. 27, 28).

In the Nabataean section we notice the use of *b'lh* (with the 3rd p. masc. sg. suffix pronoun -*h*) designating the 'lord', then 'her husband'.

As regards the dating, the epigraph appears to have been carved in the 1st c. BCE or at the beginning of the 1st c. CE if we consider the usage to write ι after ω and η in ὠικοδόμησεν and in Χαμράτῃ τῆι.

#### **Sī'**

**26.** The epigraph was part of a series of inscriptions coming from the temple of Ba'alšamīn,<sup>90</sup> a place of pilgrimage of nomads of the steppe who passed through Sī'. It is a temple built between the 1st c. BCE and the 2nd c. CE. The majority of the inscriptions include a script labelled as 'Aramaic', but with characters similar to the Nabataean, today known as the Hauranitic script.<sup>91</sup> The Nabataean text is engraved at the base of a pedestal, under the portico of the temple, whereas the Greek text is carved into a die of another pedestal, found further down the road, which should have been linked to the same base.

**Dating** 1st BCE

**Bibliography** *CIS* II no. 164; Wadd. no. 2366; Vogüé 1864, 286; *VIS*, 94-6; Cantineau 1930-32, 2: 13-4; Grushevoi 1985, 51-4; *YTDJD*, A: 309; B: [101]; *Quellen*, 174-6; Tholbecq 2007, 288-90; Healey 2009, 115 no. 15

Text and translation

1) *dnh ṣlm' dy 'qymw 'l 'byšt* 2) *lmlykt br m'yrw br mlykt* 3) *lqbl dy hw bnh byrt' 'lyt'* 4) *kdw br 'byšt 'mn' šlm*

1) Ο δημος ο τ-2) -ων Οβαισην-

3) -ων ετειμησε-

89 In Wadd. no. 2308 φυλῆς Σομαιθηνῶν is encountered in an inscription belonging to the period of Commodus, coming from al-Suwaydā', whereas in Wadd. no. 2495 the Arabic form αβαβαιη ἡ Σομαιδαθη is recorded.

90 They are Nabataean and Greek inscriptions found by W.H. Waddington and M. de Vogüé in 1861-62 on the top of a small citadel located 3 km southeast of Qanawāt (Tholbecq 2007, 285).

91 Kropp 2010, 5.

4) -ν Μαλειχαθον

5) Μοαιερου υπερ-


1) This is the statue that the tribe of 'Oḇayšaṯ erected

2) for Malīḵaṯ son of Mo'ayrō son of Malīḵaṯ

3) because he built the upper part of the temple

4) Kaḏō son of 'Oḇayšaṯ, the artisan. Peace.

1) The people of


### Commentary

In line 1 of the Nabataean text we read the name of the 'Obayšat's tribe who erected the statue as a gift for Malīkat. We notice the verb *'qymw*, an *ap'el* form of *qwm* (Sem. 'to erect, to stand up'92), frequently used in Palmyrene, Hatraean, Qumran and Biblical Aramaic with the same meaning. The tribe's name is indicated through the substantive *'l*, known in Arabic (*'āl*) and sporadically used in Hauran.<sup>93</sup> In Aramaic it might be a *hapax*, used only in Nabataean; according to Cantineau (1930-32, 2: 62) it is a clear Arabism.<sup>94</sup> The mention of the tribe of 'Obayšat is important since it is found in Safaitic inscriptions of the desert; here it is identified in Greek as δῆμος of the village of Sī'. Therefore, we may state that they were inhabitants of a village within the *Provincia Arabia*, besides being a tribe of wandering shepherds in the desert during the transhumance.

In line 2 we find Malīkat, a name that probably derives from a diminutive of the root *mlk* 'king';<sup>95</sup> it is present in the Greek part, in


line 4, as Μαλειχαθον. <sup>96</sup> His father's name, *m'yrw* (= Gr. Μοαιερου), may originate from Arabic *muġayr* 97 'one who puts down the furniture of his camel from off him, to relieve and ease him,'<sup>98</sup> even if a derivation from *'ār* 'to wander, to roam'<sup>99</sup> is more likely.

In line 3 the syntagma *lqbl dy* 'because, given that' is presumably a *hapax* since it is not employed in Nabataean with this meaning; the Aramaic usual meaning is 'in front of, opposite, over against'. Here there is a semantic correspondence to Egyptian Aramaic100 and Biblical Aramaic.<sup>101</sup>

The employment of *byrt'*, which in Aramaic mostly identifies a 'fortress',102 is interesting since here it takes on the meaning of 'temple precinct', in accordance with the Greek τὸ ἱερόν. 103 The following expression, *'lyt'*, suggests that it is referring to the upper part of the temple (cf. Syr. *'elitā* 'the upper floor, room', 'a pinnacle'<sup>104</sup>) as the analogous Greek term ὑπεροικοδομήσαντι (< ὑπεροικοδομέω 'to build over or above'105), here in the dative rather than in the accusative (anacoluthon?) case, proves.

The builder's name, *kdw*, does not appear in the Greek section, and it is not attested in other Nabataean inscriptions.<sup>106</sup>

97 Cantineau 1930-32, 2: 117.

98 Lane, 2316.

99 Precisely, al-Khraysheh (1986, 110) translates it as 'traveller', 'Reisender'; cf. also Negev 1991, no. 667. In the past this Semitic root had different meanings among sedentary and nomadic peoples if we consider that also today in Hebrew (language of sedentary people) *'îr* indicates the 'city', whereas in Arabic (language of nomadic people) *'īr* refers to the 'the caravan'. In Palmyrene *m'yr'* is the 'traveller' (*PAT*, 436 = Stark 1971, 34 and 96).

100 Porten, Yardeni 1986-99, B3.10, 17.

101 Ezra 4,16.

102 *DNWSI*, 155-6; Cantineau 1930-32, 2: 70. In Akkadian *birtu* 'fort, citadel, castle' (*AHw* I, 129; *CAD* II, 261), the term is also attested in Egyptian Aramaic (Porten, Yardeni 1986-99, B2.2 R.3), in Biblical Aramaic (Ezra 6,2) and in Syriac (*Pešiṭtā*, 2Chr 17,12).

103 The substantive appears with the same meaning in the main inscription of the temple of Sī' (*CIS* II no. 163).

104 Sokoloff 2009, 1102-3.

105 *LSJ*, 1866.

106 al-Khraysheh 1986, 98. Cantineau (1930-32, 2: 106) asserts that we may also read it as *krw*. In pre-Islamic Arabic *kd* 'work, toil' is recorded (*ICPAN*, 496).

<sup>96</sup> It is attested in the same shape at Ṣūr al-Laǧā (*Quellen*, 145-7), and at Sī', in inscriptions from the ruins of the temple of Ba'alšamīn (*CIS* II no. 163; Wadd. nos. 2367, 2368). See also Wuthnow 1930, 69 and 148.

**27.** The inscription is carved on a lintel of a tomb found among the ruins of built tombs that looked like round towers.<sup>107</sup> These tombs were near the road leading from Sī' to Qanawāt,<sup>108</sup> a village in South Syria located 7 km northeast of al-Suwaydā'.

The Greek text precedes the Nabataean;<sup>109</sup> the entire inscription is engraved in a *tabula biansata* according to the common usage in the Roman provinces. (Plate VI, no. 1)

**Dimensions** height 45 cm; length 151 cm; height of the Greek letters 6.5-9 cm; height of the Nabataean letters 5-8 cm **Dating** 1st CE (25-50 CE) **Bibliography** Buttler, Littmann 1905, 409; *ESE* II, 260; *RES* no. 1093; Littmann 1909, 378-80; *PPAES* IVA no. 105

### Text and translation


### Commentary

The name *ṭnnw* (= Gr. Τανενος) would be an adjective of *fa'īl* or *pe'il* form for *ṭn(n)-'l*, <sup>110</sup> according to the first editors. The name *ṭn- 'l* (Τανηλος), *ṭnn-'l* (Ταννηλος), *ṭn* and *ṭnn* (Ταννος) is frequently at-

#### 107 *PPAES* IVA, 84.

109 According to Macdonald (2003, 45), this Nabataean script belongs to the local Aramaic script called Hauranitic, like that in the previous inscription (no. 26).

110 *ESE* II, 260; Littmann 1909, 379.

<sup>108</sup> Qanawāt is one of the earliest cities in Hauran. It is probably referred to in the Bible as *kənāt* (cf. Num 32,42 and 1Chr 2,23). The town called *Kanatha* (Gr. Κάναθα) is mentioned for the first time in the reign of Herod the Great (1st century BCE) when Nabataean forces defeated the Jewish army. The city remained a bone of contention between the two powers. Under Pompey (106-48 BCE) and Trajan (53-117 CE) Kanatha was a city of the Decapolis; in the 1st c. CE it was annexed to the Roman province of Syria. In the 2nd c. CE it was renamed *Septimia Canatha* by Septimius Severus (145-211 CE) and transferred to the province of Arabia (cf. Burns 2009, 246-7).

tested in Safaitic inscriptions.<sup>111</sup> According to Littmann (*PPAES* IVA, 85) the correct transliteration should be *Ṭāninū*, since in Hauran the names deriving from the stems of *mediae geminatae* do not have contracted forms.

The father's name, *ḥn'l*, is also attested on the altar of Gālis,<sup>112</sup> at Sī', as one of the benefactors of the statue, and we cannot rule out the hypothesis that he is the same person. This name is a theophoric in which the first element *ḥnn* in Aramaic means 'merciful'.<sup>113</sup>

In line 4 we read the name *ḥwrw*, probably Ḥūrō,<sup>114</sup> that is either linked to Ḥūr, son of 'Oḇayšaṯ, who may be the same sculptor who carved his name in a little round altar of Ba'lšamīn's temple, or referred to the brother of a certain Kadō, son of 'Oḇayšaṯ, mentioned in *CIS* II no. 164.<sup>115</sup> The same name that designates the artisan, *'mn'*, is found in other inscriptions from Sī' (see e.g. *CIS* II no. 166) in which a certain *šwdw* (Šūdō?), the sculptor, appears.<sup>116</sup>

The root *'mn'* is productive in Biblical Hebrew *'āmān* 'workman, artificer' and also in Syriac *'āwmanā* 'workman, maker, craftsman'.<sup>117</sup>

**28.** This is a pedestal in the form of an altar that served to bear a statue. It was found among the ruins of a temple. The pedestal is broken in the middle and the four corners of the capital are ornamented with lion-heads linked to wreaths in high relief.

The Nabataean text, consisting of seven lines, is engraved in the front, which has been well preserved, whereas the only three Greek lines are inscribed on the broken side of the pedestal.

114 al-Khraysheh 1986, 78-9. Negev (1991, no. 424) reads *ḥūwār* as 'young camel'.

115 *PPAES* IVA, 85.

116 The same person appears in an inscription published by Savignac (1904, 580): *šwdw br 'byšt 'mn' šlm* 'Šūdū son of 'Ubaišat the sculptor, peace'. The three artisans quoted in the text should belong to the same family (Buttler, Littmann 1905, 409-10, no. 1).

117 Gesenius 1846, 58 = *DGes*, 74 and *KAHAL*, 34 'Handwerker, Werkmeister, Künstler' and Payne Smith 1903, 6 and Sokoloff 2009, 17. Akk. *ummānu*, *ummiānu* < Sum. *um-me-a* (*AHw* III, 1413-4 and *CAD* XX, 108 and ff.).

<sup>111</sup> Pre-Islamic Arabic *ẓnn* > *ẓanūn* 'suspicious' < *ẓn* 'thought, belief' (*ICPAN*, 394). Littmann 1909, 379; Cantineau 1930-32, 2: 102; Wuthnow 1930, 115 and 144. Negev (1991, no. 510) transcribes *Ẓānin* = Gr. Ζοναινος, Ζονενη (Wuthnow 1930, 51 and 144).

<sup>112</sup> The name is also recorded in *PPAES* IVA no. 9, in an inscription from al-Summāqīāt. Cf. also al-Khraysheh 1986, 88 and Negev 1991, no. 469.

<sup>113</sup> In CPA, JBA, Gal., Pal. and Syr.; in BA and Syr. the verb *ḥnn* means 'to show mercy' (Jastrow 1903, 484; Payne Smith 1903, 149; *DNWSI*, 389; *PAT*, 366; *DJPA*, 209a; *DJ-BA*, 474a). Also, Ge. *ḥannā* 'grace, charm, joy' (Leslau 1991, 236).

**Dimensions** height of the Nabataean inscribed surface 33 cm; width 42 cm; height of the left side of the Greek part 49 cm, width of the extant portion, at the top 32.5 cm, at the bottom 30 cm, height of the Nabataean letters 3-6 cm, height of the Greek letters 4.5-6 cm **Dating** 1st CE (29-30 CE) **Bibliography** *PPAES* IIIA 6 no. 768; *PPAES* IVA no. 101

# Text and translation

1) *bšnt 33 lmrn'* 1) [ερ]γον Zαι-2) *plps 'bdw wtrw br* 2) -[δ]ηλου ιερο-3) *bdr wqṣyw br šwdy* 3) -[δ]ουλου 4) *wḥn'l br mšk'l wmn' b[r]* 5) *grmw bwms ṣlm glšw* 6) *br bntw* 7) *'n'm br 'ṣbw 'mn' šlm*



# Commentary

Lines 1-2. The sentence: *bšnt 33 lmrn' plps* 'in the year 33 of our lord Philippos' is important for our understanding of the inscription from an archeological and historical point of view, since Philip, the Tetrarch or Herod Philip II, is mentioned; he was the son of Herod the Great. He ruled in the north-eastern part of his father's kingdom, as recalled by Flavius Josephus (*AJ* 17.8).118 His popularity was noteworthy in the *Nabataea* to such an extent that local inhabitants in the independent territories of Hegra, Petra and Bosra used to calculate the days of the year according to those of his rule. For this reason, Herod Philip was not named *mlk'* by the common people, who scarcely knew the Greek title (τέτραρχος), but *mrn'* 'our lord'. The date mentioned in line 1 presumably refers to the year in which the pedestal was erected, i.e. 29 or 30 CE. In the same line we have the

<sup>118</sup> Josephus wrote that he ruled in Gaulanitis, Trachonitis, Batanea and Paneas. Herod Philip is also quoted in the New Testament (Luke 3,1 and Mark 6,17) as husband of Salome, daughter of Herodias and Herod II.

sequence of the authors' names of the altar followed by their fathers' names. The first is *wtrw* which is transcribed in Greek as Ουιθρος, 119 Οιθρος, <sup>120</sup> coming from Arabic *watar* 'rope, cord, bowstring',<sup>121</sup> *witr*  'alone, unique'122 or from *waṯir* 'bed, carpet'.123 Another possible derivation may be directly from Hebrew, given that the name of Jethro is used (Exod 3,1; 4,18) – *yitərô lit.* 'His Excellence, Posterity' (< *yitərâ*  'abundance, rich', 'Überschüssiges, Erübrigtes').<sup>124</sup>

Line 3. After the father's name *bdr*, the root of which is extensively discussed at no. 24, the name of the second sculptor, *qṣyw*, <sup>125</sup> appears; this is probably a diminutive form (*qāṣin*?) from the Arabic *qaṣā* 'to be faraway',<sup>126</sup> rather than *qaṣiyy* 'faraway, distant'.<sup>127</sup> His father's name, *šwdy* or *šwry*, discovered in two other inscriptions from Sī',<sup>128</sup> seems to be of uncertain origin.129 It presumably derives from Arabic *sadā* 'to offer', 'benefit, favour'130 or a *nisbah* for *suwadī* 'inhabitance of Soada' (modern al-Suwaydā'), as it was originally interpreted by Littmann (*PPAES* IVA, 79).

The names in line 4 are well-known: *ḥn'l* (cf. no. 27), *mšk'l*, a theophoric made up of *mškw* (no. 15) and *'l*, and *mn'* linked to *mn't* (no. 11).

Line 5. The father of *mn'* is *grmw* = Gr. Γαρμος, <sup>131</sup> from Arabic < *ǧarama* 'to cut off, to sever'<sup>132</sup> (cf. Heb. *gāram* and Syr. *gram*) or more

119 Wadd. nos. 92, 2537; *AGLI*, 156; Wuthnow 1930, 92 and 137.

120 Wadd. no. 137; Wuthnow 1930, 87 and 137.

121 Cf. Heb. *yeter*, Syr. *yatrā* and JBA, in addition to Ge. *watr*.

122 Negev 1991, no. 365 and al-Khraysheh 1986, 71-2. The latter reads *Wātiru* as 'einer, der auf Rache sinnt'.

123 Cantineau 1930-32, 2: 90.

124 *DGes*, 516; *KAHAL*, 234. As well as in OfA., Gal., Sam. *ytr* 'remaining part, excess' < verb *ytr* (Jastrow 1903, 605; *DNWSI*, 482; *DJPA*, 248b; Tal,Sam, 369b).

125 It is also found in *PPAES* IVA no. 11.

126 Al-Khraysheh (1986, 161-2) translates 'fern (weit)entfernt' < *Qusaiu* 'Derjenige, der ins Exil geschickt wurde' with a clear connection with an 'exile'. *ICPAN*, 483: *qṣy*.

127 Cantineau 1930-32, 2: 143; Negev 1991, no. 1037. The same form is recorded in Greek as Κασσαιος (Wuthnow 1930, 63), Κασσιος (Wuthnow 1930, 165; Pape 1911, 633), which are very frequent in the inscriptions from Negev (*LGPN* I, IIIA-B, IV, VA), and Κασιος (*IGLS* 21,5 nos. 336, 347).

128 *ESE* II, 260.

129 Cantineau 1930-32, 2: 149.

130 *ICPAN*, 314; Negev 1991, no. 1115. Greek Σοδαιος (Wuthnow 1930, 112 and 169) and Σαδος (*IGLS* 21,5 nos. 319, 455, 457, 704).

131 Wadd. no. 2513 and Wuthnow 1930, 39 and 134.

132 *ICPAN*, 159; Cantineau 1930-32, 2: 78-9; al-Khraysheh 1986, 56-7; Negev 1991, no. 253. Here it takes on the meaning of 'to decide', 'to determine', in Palmyrene *grymy*, *grmy* are recorded (*PAT*, 433 = Stark 1971, 14 and 82) with the same meaning, as *grm'lt*  < *grm* in Hatraean appears (Beyer 1998, 157 and 172).

probably < Heb. *gerem*, Com. *grm'* 'bone' and Ar. *ǧirm* which is used in the sense of 'body', 'beast of burden'.<sup>133</sup>

The artefact is named *bwms* \*/bŭmos/ 'pedestal, altar', a word also used in Syriac<sup>134</sup> and in Christian Palestinian Aramaic;<sup>135</sup> it is a clear loanword from Greek βωμός 'raised platform', 'altar with a base'.<sup>136</sup> The statue on the top of the altar was erected by a certain *glšw*. This name is new in the Nabataean onomastics, probably = Gr. Γαλεσος (Wadd. 2041)<sup>137</sup> and Saf. *gls.* It may derive from Arabic *ǧālis*  'one who is sitting', 'assis'.<sup>138</sup>

Line 6. The father's name, *bntw*, is linked to Greek Βαναθος, 139 coming from Arabic *bannah* 'smell, aroma, fragrance'140 and employed both in male and female names.

Line 7. The sculptor's name *'n'm* is an elative form of Arabic *na'īm*  'calm, relaxed serene' (< *nu'aym* and *nu'm* 'comfortable life', 'wealth, prosperity, tranquility').<sup>141</sup> In Greek the forms Αναμος, <sup>142</sup> Ανεμου, 143 Ανημος and Ανομου<sup>144</sup> are present. The name can be read *An'am* or *An'um* in which the transition from /a/ to /u/ results from the presence of

133 The masculine name *garəmî* 'bony' is also attested in the Bible (1Chr 4,19).

134 In *Pešiṭtā* 2Chr 1,3 we read: *būmsā*.

135 *DNWSI*, 168.

136 Monferrer-Sala 2013, 101. Actually, this Greek term, which generally derives form βαίνω < βῆμα 'altar' (Chantraine 1968, 204; Beekes 2010, 251), might be a loanword from Semitic if we consider that *bāmôt*, 'hills, high places', Jewish places of worship, derives from *bāmā* (< *bûm*) 'fortress, castle', also 'a sanctuary built on a mountain to God or idols' (Gesenius 1846, 124-5 = *DGes*, 155-6 'Anhöhe, Berg-, Höhen- rücken', 'Kulthöhe, Kultstätte, Bama'; *KAHAL*, 71). In Qumran Aramaic, PTA *bmh, bmt'* 'pagan altar, high place' (Jastrow 1903, 176; *DJPA*, 105a) and *byms* 'pedestal, elevated stand, rostrum' in CPA, JBA, Nab. and Syr. (Jastrow 1903, 162; Payne-Smith 1903, 42; *DNW-SI*, 168; *DJPA*, 96a; *DJBA*, 210b). The root has a parallel meaning in Akkadian *bamātu*  'open country, plin, field, slope of a hill' (*AHw* I, 101; *CAD* II, 76-7) linked to Sum. *bà*  'half, share, portion'.

137 Other occurrences in *AGLI*, 59; Wuthnow 1930, 30; *IGLS* 21,5 nos. 67, 163, 369.

138 From pre-Islamic Arabic *jls*, Jals, Jils (*ICPAN*, 165). Cf. Cantineau 1930-32, 2: 78 and Negev 1991, no. 233 as well.

139 Cantineau 1930-32, 2: 72; *AGLI*, 99 and 347-8; *IGLS* 21,5 nos. 51, 260; Wuthnow 1930, 32 and 130.

140 *ICPAN*, 119; Negev 1991, no. 185.

141 *ICPAN*, 80; for *n'm* and derivatives, *ICPAN*, 593-5; Cantineau 1930-32, 2: 121; al-Khraysheh 1986, 40; Negev 1991, no. 113. The same root is productive in Heb. *na'ămâ* (*DGes*, 826; *KAHAL*, 356) that is, besides the meaning of 'pleasant', a proper name of two women in the Bible (cf. Gen 4,22; 1Kgs 14,21-31; 2Chr 12,13) and in Gal. with the same meaning (*DJPA*, 354b). In Palmyrene *'n'm* 'tender' (*PAT*, 430 = Stark 1971, 6 and 70).

142 *AGLI*, 152, 155; Wuthnow 1930, 21 and 152.

143 *AGLI*, 76, 182, 186; Wuthnow 1930, 22 and 159.

144 Wadd. no. 2412.; Wuthnow 1930, 22-3 and 152.

the following /m/.<sup>145</sup> This last reading also identifies a southern Arabic tribe, the An'um, the watchmen of the temple of the god Yaġūṯ in a place called Guraš.146 The father's name, *'ṣbw*, is new in Nabataean147 corresponding to Greek Ασβος<sup>148</sup> and Οσεβος<sup>149</sup>. Initially it was linked to Arabic *ḥuṣayb* (diminutive of *ḥasab* 'pebbles, flints, gravel')<sup>150</sup> or *'uṣayb*  (< *'aṣaba* 'to tie, to tighten, to bind', Syr. *'ṣab* with the same meaning).<sup>151</sup>

In the broken part of the pedestal we find the Greek inscription that was copied by Mr. Magies: ἔργον Ζαιδηλου ἱεροδούλου. It is probable that *Zaid'el*, a theophoric name,<sup>152</sup> built the pedestal or worked as a sculptor among those who erected this altar.

**29.** The inscription is engraved on a tablet that was found by Littmann in 1909<sup>153</sup> in the north wall of the podium of the new temple presumably dedicated to Dūšarā. (Plate VI, no. 2)

**Dimensions** height 38.5 cm; length 42.5 cm; height of the Greek letters 2.5-3 cm; height of the Nabataean letters 4.5-5.5 cm, with the exception of *l* in line 1 that is 8 cm high **Dating** unknown **Bibliography** *RES* no. 1092; Littmann 1909, 375-8; *PPAES* IIIA, 6 no. 767; *PPAES* IVA no. 103; Cantineau 1930-32, 2: 14-5; *SEG* 29 no. 1603

Text and translation

1) Σεεια κατα γην Aυρα-2) νειτιν εστηκυια 3) *d' ṣlmt'* 4) *dy š'y'w*

1) Seia, in the land of Aura-

2) -nitis (is) standing

145 The same name appears in *PPAES* IVA nos. 12, 14.

146 Wellhausen 1897, 19-20.

147 *RES* no. 2117.

148 Wadd. no. 2008; *ESE* I, 329 no. 31; Wuthnow 1930, 26 and 159.

149 Wadd. no. 2412; Wuthnow 1930, 90 and 159.

150 Wetzstein (*AGLI*, 361) translates 'Brennstoff (des Kampfes)' < Ar. *ḥasaba* 'to throw stones', 'to stone'.

151 *PPAES* IVA, 79. It also means 'tense', 'emotional'. Negev (1991, 929) interprets it as *ġaḍab* 'rage, anger'.

152 Wuthnow 1930, 49 and 138. For the etymology see inscription no. 31.

153 On the contrary, *RES* no. 1092 reports that the epigraph was found in 1905 inside the temple of Ba'alšamīn.

3) This (is) the image 4) of Še'ī'ō

#### Commentary

Line 1. The name of Σεεια occurs at the beginning of the line. There are other forms of this name, such as Σεειηνων, <sup>154</sup> Σειηνος<sup>155</sup> and Σαιηνων. <sup>156</sup> After the mention of Σεεια, a segment written in *scriptio continua* appears: καταγηναυρα; it is made up of the preposition κατά followed by the feminine substantive γῆ 'in the land' and Αυρα-, which constitutes the initial part of the toponym Hauran and whose final part ends in the following line.

Line 2. We read Αυρανειτιν, the Hellenised form of the Semitic name Auranitis,<sup>157</sup> Αὐρανῖτις, -ιδος, here in the accusative.<sup>158</sup>

In the last part of the line we see the singular feminine perfect participle ἑστηκυῖα 'she is standing, rising up' < ἵστημι.

Line 3. After the Nabataean demonstrative *d'* the word *ṣlmt'* occurs meaning 'image'.<sup>159</sup> In the Akkadian pantheon *ṣalmu* 'statue' is employed along with the name of a deity to indicate its image. Akkadian *ṣalāmu* 'to be black' would probably originate from the root *ẓl* (*ẓl* > *ṣl*) 'shadow'<sup>160</sup> (cf. Ar. *ẓulmah* 'shadow' < *ẓalama* 'to become dark'; Ge. *ṣallim* 'balck, dark'); according to some scholars, the acceptation of 'image' is based on the aforementioned origins.161 Moreover, in Greek σκιά means 'shadow' assuming the concept of 'reflection, image, phantom'.<sup>162</sup>

Line 4. At the beginning the particle *dy*, as *nota relationis*, is followed by the name *š'y'w*, which is the appellative of the local goddess of Sī'. As regards its etymology, in Jewish Babylonian Arama-

154 Wadd. no. 2367.

155 Wadd. no. 2418.

156 In DM, II no. 250.

157 Josephus wrote that it was a Greek-Roman *provincia* (cf. *AJ* 16.9.1; 17.11.4).

158 We notice the diphthong ει, *usus* of the *koinè*, employed by the author in order to identify ῖ < εῖ.

159 In Com. *ṣlm*, *ṣlm'* 'image (physical)'; JBA 'picture', BA *ṣəlēm* 'form, appearance' (Dan 3,19), Syr. *ṣalmā* 'image, statue, idol' (Jastrow 1903, 1284; Payne Smith 1903, 480; *DNWSI*, 968; *DJPA*, 465b; *DJBA*, 966a; Sokoloff 2009, 1290). In Phoen. *ṣlm* 'statue' (Krahmalkov 2000, 417).

160 Eybers 1972, 29-32.

161 Scagliarini 2008, 63 and no. 6.

162 *LSJ*, 1609. Cf. Soph. *Aj.* 301. The original meaning is 'refuge, covering' < σκιάζω 'to overshadow, to cover' (cf. *LSJ*, 1610; Hom. *Il.* 21.232). Compare the Greek term with Hebrew *sukkā* 'shed, tabernacle, covering', Akk. *sukku* 'shelter'.

ic *š'ī'a* means 'smooth'<sup>163</sup> (Heb. *šā'a'* 'to spread, to stain, to smooth', 'bestreichen, glätten'164); Littmann (*PPAES* IVA, 82) translates it 'levelled square' or 'even space', comparing<sup>165</sup> it to ἡ ἱερὰ πλατεῖα and ἡ πλάτιος ἱερατική (in Wadd. nos. 2034 and 2035) from πλατύς 'wide, spacious and holy space'.166 In Arabic the term is shortened by haplology to *sī'* < *s'y'*; <sup>167</sup> in addition *š'yw* occurs in an inscription from Sinai (Euting 1891, 558, line 3) standing for *si'y* 'man from Sī''.<sup>168</sup>

Ultimately, the Greek transliteration suggests that the final -*w* of *š'y'w* does not correspond to the suffix -*u*, equivalent to the Arabic nominative case, but coincides with the original vowel /ā/, written as /ō/ in various Greek transliterations, probably reflecting the real Aramaic pronunciation of the vowel /ā/;<sup>169</sup> this implies that the deity's name should be pronounced /še'ī'ō/ (?).

As regards the question of the correspondence between the deity's name and the place where the statue was found, Littmann wondered the name of the place appeared before or after the name of the goddess that appeared first.<sup>170</sup> The fact that *š'y'w* is an abbreviation of *rbt š'y'w* 'Lady/Goddess of Š'y'w' might imply that the name of the place existed before. We have to make clear that during the pre-Islamic period, Arabic pagan names of deities and places coalesced;<sup>171</sup> the Semites, in particular, never separated a god from its 'house'.<sup>172</sup> Considering that the temples at Sī' were dedicated to Ba'alšamīn and Dūšarā, we may infer that Še'ī'ō was not the goddess of Sī', but probably an abstract form or the personification of the holy place: a θεὸς σύνναος inside the temple of Dūšarā.<sup>173</sup>


168 al-Khraysheh (1986, 183-4) reads *sā'iyu* 'Herr des Stammes'; Negev (1991 no. 1182) links it to Arabic *sa'īah* = Gr. Σαιος (Cantineau 1930-32, 2: 153).


#### **Ḥarrān**

**30.** The inscription was found in the little village of Ḥarrān, 50 km north of Bosra, by M. Wetzstein;<sup>174</sup> It was revisited at a later date by W.H. Waddington.<sup>175</sup> The two texts are carved on a lintel, today reemployed above a house door; originally, the lintel was garnished with Christian crosses, which were placed at the beginning and in the middle of each inscription.<sup>176</sup>

The text was classified, by the first editors as being among the Nabataean inscriptions; it exhibits handwriting and morphological features that reflect the last stage of the Nabataeo-Aramaic script moving towards the Kufic calligraphic form of the early years of Islam.<sup>177</sup> It is to be noted that there exists a long chronological gap between the latest dated Nabataean inscription and the earliest instances of cursive Arabic script.<sup>178</sup> Although the inscription is considered to be written in Arabic, it seems appropriate to include it in this *corpus* both for further historic and linguistic consideration and to preserve a proof of the (Nabataeo-Arabic) linguistic transition, during the Byzantine and Christian period in southern Syria. (Plate VII, nos. 1-2)

```
Dating 6th CE (568 CE)
```

```
Bibliography AGLI, no. 110; Wadd. no. 2464; VIS, 117-18; Schroeder 1884, 530-
   4; DM, 726-7; Littmann 1912, 193-8; Cantineau 1930-32, 2: 50-1; Abu al-Faraj 
   al-'Ušš 1973, 55-84 (Arabic); Mascitelli 2006a, 183-7, no. 19; Hoyland 2008, 
   55-6 (reproduction); Fiema et al. 2015, 414-5
```
#### Text and translation

1) *'n' šrḥyl bn/br ṭlmw bnyt d' 'lmrṭwl* 2) *snt 463 b'd mfsd* (?) 3) *ḫybr/ḥnyn(?) b'm* 1) Σαραηλος Ταλεμου 2) φυλαρχ[ο]ς εκτισεν το μαρτ[υριον] 3) του αγιου Ιωαννου ινδ[ικτιωνος] A του ετους ΥΞΓ. Μνησθιε ο γραψας

1) I, Šarḥīl son of Ṭalemō, built this *martyrion* 2) in the year 463, after the expedition/defeat?


3) of Khaybar? One year

1) Saraelos (son of) Talemos,

2) the phylarch, founded the *mart[yrion]*

3) of Saint John in the first ind[iction] of the year 463. May the

writer be remembered

#### Commentary

In line 1 the first singular personal pronoun *'n'* already assumed the same shape as the Classic Arabic *ānā*, with its final *mater lectionis '*, and that of the Syriac *ānā*. <sup>179</sup> The author of the epigraph, *šrḥyl* = Gr. Σαραηλος, <sup>180</sup> bears a name that is well-documented in South Arabic like *s²rḥ'l* <sup>181</sup> and only in the Greek part of the inscription he declares himself to be a phylarch; but according to the sources, his name is not recorded elsewhere, and above all we must not confuse him with the Ghassanid phylarches, since at the time of the building of the church of Ḥarrān the Ghassaind phylarch was a certain al-Munḏīr.182 This is a theophoric from Arabic *šaraḥ* 'to uncover, to reveal',<sup>183</sup> so 'God has revealed'.

The father's name, *ṭlmw =* Gr. Ταλεμου, is problematic from the point of view of transliteration and phonetic interpretation.<sup>184</sup> However, as already mentioned, in this stage of graphic transition from Nabataean to Arabic, the alphabet, which was still without diacritical marks, used the grapheme *ṭ* both for /ẓ/ and for /ṭ/, for this reason the name should be read *Ẓālim* 'unjust, unfair, oppressive', from *ẓalam* 'to be unjust, to oppress';185 unfortunately, Greek does not have a grapheme neither for /ẓ/ nor for /ṭ/, and consequently it does not help us in any way.

182 Wadd. no. 2464. Further information in Mascitelli 2006a, 186.

183 *ICPAN*, 345 reports *šrḥyn*.

184 For the Greek form see Wuthnow 1930, 115.

<sup>179</sup> In an Arabic graffito from Jabal Usays (100 km east of Damascus), dating back to 528-529 CE, the 1st sg. pers. is *'nh* (Macdonald 2010, 141-3) = BA *ănâ*; this may prove that still in the 6th c. CE in Arabic -*h* and -*'* interchange as *mater lectionis* as a result of an Aramaic influence.

<sup>180</sup> Waddington reads it Ασαραηλος, as does Cantineau report (1930-32, 2: 51), probably due to a misreading resulting from the presence of the icon of the cross, carved at the beginning of the Greek inscription. Cf. also Wuthnow 1930, 105 and 174.

<sup>181</sup> In Minean (*RES* no. 2999), in Qatabanic (*RES* no. 3902) and in Sabean (Ryckmans 1949, 57-8).

<sup>185</sup> See the root *ẓlm* in *ICPAN*, 393. Cf. CPA, Man., Sam. *ṭlm* and Syr. *ṭlam* with the same meaning (*DJPA*, 225b; Jastrow 1903, 537; Payne Smith 1903, 175; Sokoloff 2009, 533; Tal, *Sam*, 313a).

After the demonstrative particle *ḏā*, we read *'al-marṭūl*, a loanword from Greek μαρτύριον and also a *hapax* in Nabataean and Arabic; the usage of the final *l* in *mrṭwl* is intriguing.<sup>186</sup>

In lines 2-3, after the mention of the year 463, we find several words that present problems of interpretation. After *b'd* (presumably *ba'ad* 'after'), the term *mfsd* is not present in Arabic as a substantive, but as a verb of form *'af'al*, *afsada*, which is employed by Ibn Khaldūn meaning 'to sack, to ransack', as well as the plural form *mafāsid*  'raids, loots, robs' (sing. *mafsadah*<sup>187</sup>); in ancient literature *'al-īfsād fī l-ārḍ* has the same meaning of *fitnah* in a political sense.<sup>188</sup> The last words were read by R. Dussaud and J. Cantineau *ḥyn[y]d* (= *ḥayna'iḏ*  'at the time, at that time, then') *n'm* translating the entire sentence as: "après la corruption, la prospérité (?)".<sup>189</sup> Conversely, Littmann read *ḥyn[y]d* as *ḫybr* (=*Khaybar*) and *n'm* as *b'm* (= *bi-'ām*), so *ba'da mafsadi ḫaybara bi-'āmin*, and translated: "un anno dopo la spedizione di Khaybar".190 To his mind, the sentence is related to a raid, to looting carried out at Khaybar<sup>191</sup> by al-Ḥāriṯ bin Abī Šamir,<sup>192</sup> although Littmann (1912, 195) identified the individual as al-Ḥāriṯ bin Jabalah.<sup>193</sup>

In contrast, the Greek text does not include references to military campaigns, but to the building of the *martyrion* of Saint John – τοῦ ἁγίου Ιωάννου – with the indiction: ἰνδ(ικτιῶνος) A τοῦ ἔτους ΥΞΓ, i.e. the year 463 (of the year of Bosra) that roughly corresponds to the period running from 22 March to 1 September 568 CE. The final sentence in the last line mentions the person who wrote the inscription: Μνησθῖε ὁ γράψας. There is a textual incongruity because the Greek section is of a religious rather than military nature and in addition it clearly refers to the addressee of the martyr's sanctuary. Moreover, we would expect a mention of John in the Nabataeo-Arabic part of the inscription, but the name does not seem to occur. C. Robin, by contrast, suggests another reading of the last Nabataeo-Arabic lines

189 Cantineau 1930-32, 2: 50; DM, 727.

190 Littmann 1912, 194.

191 It is an oasis situated 153 km north of Medina that was inhabited by Jews before the advent of Islam.

192 As Ibn Qutayba suggests (*Kitāb al-ma'ārif*: 414).

193 He is known as [Φλάβιος] Ἀρέθας from an inscription (*IGLS* 5, no. 2553D) and as Khālid ibn Jabalah in later Islamic sources (cf. Shahid 1995, 1: 216-17; Greatrex, Lieu 2002, 102-3); he was a king of Ghassanids from 528 to 569. For further details, see Mascitelli 2006a, 185-6.

<sup>186</sup> Fiema et al. (2015, 414 and no. 209) also quote the explanation of Shahid who reconstructs *mrtwr* [*sic.* for *marṭwr*].

<sup>187</sup> Dozy 1881, 2: 266.

<sup>188</sup> Littmann 1912, 194 on the basis of the consideration that was stated by the Dutch Orientalist Prof. Snouck Hurgronje (1857-1936). Cf. Koran II, 251.

translating them: "for [as is the sense of *b'd* in Dedanite] the holy [assume *mfsd* is a mistake for *mqds*] John [i.e. *Ḥnyn*] vale [i.e. *ni'ma*]".<sup>194</sup> Mascitelli is sceptical about this hypothesis (2006a, 185) and hesitates to state that the final letter of the name *Ḥnyn* is *n*, but rather *r* which is of the same shape of the other *r* occurring in the text; in addition, on the basis of the hagiographic sources, he observed that there was no Saint John visiting the surroundings of Ḥarrān except Saint John the Hesychast<sup>195</sup> who died in 558 (ten years before the building of the *martyrion*), meaning that the chronology does not correspond.196 Taking into account C. Robin's reading, we may reexamine the proposed translation and review the sequence of letters *ḥnyn*. If we check the image better, we may notice that the last grapheme is not *n* but *r*, as Mascitelli also points out (2006a, 185), and if *ḥnyn*  were a proper name, the final -*w* would be expected to appear as in the two previous names, so *Ḥnynw*. Furthermore, we clearly do not see the double marks ىى after *ḥ* and we may simply read *ḥibr* 'prelate, religious authority', a term often used in the Arabic-Christian world.<sup>197</sup> The sentence then assumes the following meaning: 'for the holy prelate. Peace' indicating a clergyman who did a lot of good for the little community of Ḥarrān and was remembered for this reason.<sup>198</sup> If the last reading is right, it is possible that *y* is used as *mater lectionis* to distinguish the substantive from *ḥibr* 'ink'.

194 Hoyland 2008, 66, no. 14.

195 He was also known as Saint John the Silent and he was bishop of Colonia in Armenia (today Köylühisar, in Turkey).

196 Mascitelli 2006a, 187.

197 Kazimirski 1860, 366; Lane, 498; Dozy 1881, 1: 243.

198 In Nabataean the root *ḥbr* (as well as in Hebrew) means 'companion', also 'member', especially of a religious confraternity. In Syriac it means 'to be a companion of', in Pal. 'to make a partner of', in Qumran, Gal., PTA, Sam. (Sokoloff 2009, 409-10; *DJ-PA*, 186a; *DJBA*, 428a; Tal, *Sam*, 243; Cook, *Qumran*, 75). The term appears in an inscription from al-Ḥiǧr (Nehmé 2005-06, 194-200, no. 13, fig. 138). Hoyland (2008, 55) claims, with regard to the figure of John, that he was: "evidently an important man in the local Christian community".

Giuseppe Petrantoni

# Egypt

# **Mt. Sinai**

**31**. The inscription was found between Wādī Haggag and Jabal Mūsā and bears simple names among which, in the Greek part, we read that of the addressee, also mentioned in the Nabataean section along with his father. The same combination of names is attested in another inscription from Sinai.<sup>1</sup> (Plate VIII, no. 1)

**Dating** 2nd-3rd CE<sup>2</sup> **Bibliography** *SEG* 31 no. 1422; Negev 1981, 69

Text and translation

1) Ζεος 2) *zy(d)w br p[ṣyw]*

1) Zeos 2) Zē(ḏ)ō son of Fo[ṣayō]

<sup>1</sup> *CIS* II no. 2278 we read: *šlm, zydw br fṣyw. bṭb* 'Peace. Zaydū son of Foṣayū. In good'.

<sup>2</sup> The approximately 3,000 graffiti from Sinai mostly date from the 2nd and the 3rd century CE (Garbini 2006, 218).

#### Commentary

The Greek Ζεος (which must not be confused with Ζεύς) represents the phonetic transliteration of the Semitic name *zy(d)w*, here presumably written in *scriptio defectiva* like in the Greek form without δ.

The same name, in its original shape, appears in another bilingual from Sinai (cf. no. 42); it is *Zayd*, a widespread name in the Arabic speaking-world, which derives from the root *zāda* 'to increase, to grow'.<sup>3</sup> It is transliterated in Greek using other variants,<sup>4</sup> but in Nabataean there is the assimilation /ay/ > /ē/ reading *Zē(ḏ)ō* = Ζεος.

The father's name *p/fṣyw* comes from Arabic *faṣṣā* 'to deliver from'<sup>5</sup> (Syr. *paṣī* and Heb. *pāṣā* 'to deliver from, to rescue, to save');<sup>6</sup> the theophoric form, *fṣ'l* <sup>7</sup> 'God has saved' corresponding to Greek Φασαηλος (Joseph. *AJ* 14.7.3),<sup>8</sup> is also used.

**32.** From Wādī 'Alayyāt (Wādī Fayrān).

**Dating** 2nd-3rd CE **Bibliography** Euting 1891, 150; *CIS* II no. 2258

Text and translation

1) *šlm bry' br klbw bṭb* 2) μνησθη Αυρηλιος 3) Βοραιος Χαλβου

1) Peace! Borayā[ō] son of Kalbō, in god

2) Let be remembered Aurelios

3) Boraios (son of) Chalbos

3 Cantineau 1930-32, 2: 92; *ICPAN*, 304; al-Khraysheh 1986, 73-4; Negev 1991, nos. 380, 385. In Hatraean the name *zyd'* is recorded in an inscription from Ṭūr 'Abdīn (Beyer 1998, 117 and 158).

4 Ζαεδος, Ζαιδος, Ζοδος (Wuthnow 1930, 49 and 138).

5 Cantineau 1930-32, 2: 137; al-Khraysheh (1986, 153) reads *Fuṣaiyu*, while Negev (1991 no. 971) *Fāṣiya*. The name is also found in Palmyrene *pṣ'* and *pṣy* (*PAT*, 439 = Stark 1971, 47 and 109).

6 Sokoloff 2009, 1219; Jastrow 1903, 1204; *DNWSI*, 930; *DGes*, 1068-9; *KAHAL*, 454. It may be an Akkadian influence, from *puṣṣū* (Kaufman 1974, 84) which means 'to whiten, to clean', hence *pūṣu* 'white spot, fleck, mark' (*CAD* XII, 539; *AHw* II, 883). It is probable that the original Akkadian took on the meaning of 'to save' (< 'to deliver from a spot, to clean'?) in Aramaic, thus serving to translate the name as 'Saviour'. In Hatrean *pṣī*  has the same meaning, 'retten' (Beyer 1998, 182).

7 *CIS* II, 354; al-Khraysheh 1986, 152-3. In Palmyrene *pṣ'l* is employed and translated by Stark (1971, 47 and 109) as "God has opened (the womb)".

8 Cf. also Pape 1911, 1604. Other versions are Φασαειλη (*IGLS* 21,5 nos. 493, 495), Φασεελη (*IGLS* 21,5 no. 704) and Φασηελη (*IGLS* 21,5 nos. 146, 495, 496, 497).

#### Commentary

The name *bry'* is mainly used in the Sinaitic inscription,<sup>9</sup> deriving from Arabic *baraā'* 'free, blameless',<sup>10</sup> also *barī'* 'innocent' < Ar. *bariā'* 'to be innocent'.11 According to Cantineau (1930-32, 2: 74), it also expresses the concept of 'born during the first night of the month' (cf. Ar. *bar'* 'divine creation'). The Greek transliteration, Βοραιος, 12 allows us to read the Nabataean name as \*/burayō/, that reflects the Arabic diminutive form *buray'un*, <sup>13</sup> assuming that the diphthong αι preserves the sound /ai/ and not /e/.

The father's name *klbw* is linked to Arabic *kalb* 'dog'<sup>14</sup> (Com. *klb'*, Heb. *keleb*, Akk. *kalbum*) <sup>15</sup> and is often employed as a personal name. The Greek form Χαλβου<sup>16</sup> precisely reflects the shape of *klbw*.

In the Greek section of the graffito the completed name appears, with the addition of Αὐρήλιος (< Roman *nomen Aurelius*) well-documented in the Greek onomastics.<sup>17</sup>

**33.** From Wādī 'Aǧalah (Wādī Fayrān)

```
Dating 2nd-3rd CE
```

```
Bibliography Euting 1891, 258; CIS II no. 1753; RIGP no. 2969 (Nabataean), 
   no. 2970 (Greek)
```
Text and translation

1) *šlm 'wdw br 'myw* 2) Αυδος Αμ-3) -μοιυ

9 Cantineau 1930-32, 2: 74-5.

10 Negev 1991, nos. 194, 199. In pre-Islamic Arabic *bry* 'fatigued' (*ICPAN*, 103).

11 The root is reflected in *barā'* 'to create' (linking to *barā* 'to form, fashion by cutting'. Lane, 197).

12 Other forms are Βουρεος (no. 39; *CIS* II no. 1195; Wuthnow 1930, 37) and Βορειος (Wuthnow 1930, 132).

13 al-Khraysheh 1986, 49.

14 Cantineau 1930-32, 2: 107; al-Khraysheh 1986, 100; Negev 1991, no. 561. In Palmyrene there is the form *klb'*, *klby* (*PAT*, 435 = Stark 1971, 29 and 92), as well as in Hatraean (Beyer 1998, 177).

15 *DNWSI*, 509; proto-Afro-Asiatic \**kVwVl-* / \**kVyVl-* 'dog, wolf' in which Sem. \**kalb-* 'dog' may be a consequence of \**kal-* with the suffix indicating dangerous animals \*-*b*- (Cf. *HSED* no. 1521).

16 Other forms are Χαλβας, Χαλβης, Χελβες (Wuthnow 1930, 119 and 146).

17 Pape 1911, 117; *IGLS* 21,5 no. 5; *LGPN* I, IIIA, IV, VA, VB.

1) Peace! 'Awḏō son of 'Ammayō

2) Audos (son of) Am-

3) -moios

### Commentary

The etymology of the name *'wdw* has fully been discussed in the bilingual from Umm al-Jimāl (no. 15), in which the form *'wyd'* along with the equivalent elongated Greek form Αουειδανος are recorded. The two names, *'wdw* = Gr. Αυδος, are also found in another bilingual from Sinai (cf. no. 38).<sup>18</sup>

The term *'myw*, 19 written in Greek as Αμμοιυ20 did not initially appear to have an equivalent in Arabic;21 conversely, according to al-Khraysheh (1986, 143), the name may be a sort of Arabic diminutive form, *'Umaiyu*, from *'ā'mā* 'blind'. Closer examination of the Greek transcription provides us with a particular clue; in fact, taking into consideration the diphthong οι = /i/, the name seems to correspond to Arabic *'āmmī* 'common, working-class man', also 'ordinary man'. Another assumption lies in the fact that the name in Aramaic, *'mm*, takes on the meaning of 'gentile, non-Jew, Christian'<sup>22</sup> (cf. also Syr. *'ammayā* 'gentile, pagan' < *'my* 'to blind'23), and is probably more appropriate in a place like Mount Sinai through which many Christians passed between the 2nd and the 3rd century.

20 Other forms are Αμμαιος (*CIS* II no. 1197 and no. 40), Αμεος (Euting 1891, 342).

21 Cantineau (1930-32, 2: 132) wrote: "pas d'équivalent sûr en arabe", suggesting Aramaic origin from *'mm* 'meeting, community'. Cf. also Negev 1991, no. 903.

23 Payne Smith 1903, 417; Sokoloff 2009, 1110 and 1112.

<sup>18</sup> Greek Αυδος is attested in *IGLS* 21,5 no. 241; *PPAES* IIIA no. 445. Aὖδος is also the name of a river in Mauritania (Pape 1911, 175). Other forms are Αυειδος, Αουδηνοι (Negev 1991, no. 851), Αουεδος (*IGLS* 21,5 no. 202), Αουιδος (*IGLS* 21,5 nos. 206, 712) and Αουειδος (*AGLI*, 134, 179; *IGLS* 21,5 nos. 99, 132, 203, 204, 324, 434, 482, 604). Cf. Wuthnow 1930, 29 and 124 as well.

<sup>19</sup> It is found in *RES* no. 489.

<sup>22</sup> In Gal., PTA, CPA, Sam., JBA (Jastrow 1903, 1089; *DJPA*, 411a; *DJBA*, 869b; Tal, *Sam*, 644a).

**34.** From Wādī Muġārah (Plate VIII, no. 2)

**Dating** 2nd-3rd CE **Bibliography** *CIS* II no. 671; *RIGP* no. 1430

Text and translation

1) *šlm 'lktryw* 2) *br hny'w bṭb* 3) το προσκ[υνη]- 4) -μα Γ... 1) Peace 'Alkatrayō 2) son of Honēāō, in good

3) the venera[tio]-

4) -n G…

Commentary

The graffito contains the name *'lktryw*, which is rare in the Nabataean onomastics; it is presumably a theophoric with the Arabic definite article *'l*-. *ktryw* is probably the name of the deity *Kaṯrā*, a pre-Islamic idol<sup>24</sup> that was worshipped by the tribes of Ṭasm and Ǧadīs,<sup>25</sup> meaning that the addressee of the epigraph bears a divine name.<sup>26</sup>

The father's name *hny'w* corresponds to Greek Ονεος<sup>27</sup> and it is the Arabic diminutive *hunay'* < *hānī'* 'servant', 'happy, cheerful'.<sup>28</sup>

The Greek part is less legible and we may only distinguish the term τὸ προσκύνημα followed by various letters.<sup>29</sup>

24 *ICPAN*, 495.

25 Cantineau 1930-32, 2: 108-9; Negev 1991, no. 586. The two names refer to mythological clans of Ancient Arabia who settled in the Yamāma region (in the eastern part of Naǧd). The Ṭasm and the Ǧadīs belong to the so-called 'Disappeared Arabs' (Ar. *al-ʿArab al-bāʾidah*), a label with which genealogists indicate the groups of people who began to speak Arabic first, after the confusion of the languages provoked by the destruction of the Tower of Babel (Fahd 1968, 109; Heinriches 2000, 359-60).

26 According to al-Khraysheh (1986, 36), the name specifies something that is on top, a jag, a hump, 'der Hochgestellte, Höcker'.

27 It is attested in Euting 1891, 229.

28 Cantineau 1930-32, 2: 87; al-Khraysheh 1986, 63; Negev 1991, no. 311. The basic root is *haniā'* 'to cheer up', 'to feel pleasure' < CPA, JBA, Gal., Man., Syr. *hny* 'to be pleasing, benefit, profit', 'to aid someone' (Jastrow 1903, 359; *DJPA*, 166b; *DJBA*, 386a; Tal, *Sam*, 211a; *DNWSI*, 289). In Palmyrene *hn'y* (*PAT*, 433 = Stark 1971, 15 and 84) and in Hatraean *hny* (Beyer 1998, 12-13, 52, 158 and 174) with the same meaning.

29 For the practice of the *proskynema* in Egyptian setting, see Geraci 1971.

#### **35.** From Wādī Mukattab

**Dating** 2nd-3rd CE **Bibliography** *CIS* II no. 985; *RIGP* no. 2177

Text and translation

1) αλλσος 2) ρυοτε 3) μου 4) μ *šlm 'šrw t 'mm* 1) you leapt As-2) -ru with(?) a-3) -mum-4) -m peace 'Ašrō with (?) 'Amum

#### Commentary

This graffito presents some problem of interpretation. In line 1 the Greek letters αλλσος do not provide a solution regarding a possible reading and they do not correspond to the beginning of the Nabataean line in which *šlm* is written, so we could interpret them as ἆλσο<sup>30</sup> (2nd sg. pers. of the third Aorist Indicative < ἅλλομαι<sup>31</sup>) 'you leapt'. This obviously has no connection with the meaning that the term *šlm* assumes in the funerary inscriptions. Furthermore, it seems more improbable to read ἄλσος as 'grove, sacred grove',<sup>32</sup> but either as also 'any hallowed precinct'. The presence of the two λ may be due to a scribal error.<sup>33</sup>

Following the Nabataean text, after *šlm* we see the name *'šrw*(*t*) with a doubtful final -*t* that is unrelated to the *corpus* of the word. If the reading is right, in Greek we have an assimilation between the o of ἆλσο and the α of (α)σρυ representing the phonetic transliteration of the Arabic name *'Āsrā* (a female name used in the Arabic-speaking world) = m. *'Īsrā'* < *sarā* 'to travel by night'.34 The *t* may repre-

30 Cf. *LSJ*, 70 and 73; Homer, *Il.* 16,754. We also have ἇλτο encountered in line 755.

32 *LSJ*, 73.

<sup>31</sup> Cf. Lat. *salio* 'to salt, to salt down, to leap, to jump' to which the etymology of the Greek root is linked (Beekes 2010, 72).

<sup>33</sup> In another graffito from Sinai (in Euting 1891, 328) we read the name αλσορεος, with one λ that is probably the Arabic article *'al-*.

<sup>34</sup> Lane, 1355. In pre-Islamic Arabic the form *'srw* < *srw* 'to be generous, manly' is used. In Com. the root *šry* means 'to loosen, untie' (cf. Jastrow 1903, 1629-30), also 'to

sent the Semitic particle *'t* (cf. Heb. *'ēt* or Akk. *itti*) 'with, alongside'<sup>35</sup> written in *scriptio defectiva*, but employed in *plena* in the Greek οτ (loanword?).

The last name *'mm* = εμουμ comes from a well-attested root, discussed at no. 33.

#### **36.** From Wādī Mukattab

**Dating** 2nd-3rd CE **Bibliography** Levy 1860, 451-2, no. 34; *CIS* II no. 973; *RIGP* no. 2139

Text and translation

1) *šlm*. 2) šlm bry'w br mg[dyw] 3) (Β)ουρεος

1) Peace! 2) Peace, Borayāō son of Mug[dēō] 3) (B)oureos

Commentary

The first name *bry'w* is found in other two bilinguals (nos. 32 and 39). As regards the second name, although the line is damaged, after *br*  two letters, *mg-*, and probably *[*-*dyw]* appear and allow us to reconstruct the name *mgdyw*. It is an Arabic root < *maǧdī* o *maǧīd* 'glorious, famous, worthy'.<sup>36</sup>

dwell', 'to reside' (cf. Payne Smith 1903, 595-6). For more general references see *DN-WSI*, 1192.

<sup>35</sup> In Qumran, Gal., PTA, Man. *'t* assumes the meaning of 'sign, mark', in Com. that of 'miracle', as we may find, for instance, in BA *'ātayyā'* (Dan 3,32), in Syr. *ātā* means 'sign, mark, miracle, example', also 'monument' (*DJPA*, 78b; *DJBA*, 175b; Jastrow 1903, 132; Cook, *Qumran*, 26; Sokoloff 2009, 109).

<sup>36</sup> Cantineau 1930-32, 2: 111-2; *ICPAN*, 528; al-Kraysheh 1986, 104; Negev 1991, no. 608. The name is also attested in other inscriptions from Sinai (*CIS* II nos. 2377, 2388, 2443), while another *mˀgydwˀ* is present in a Nabataean inscription found at al-Ruwāfah (*Quellen*, 300). The root is already known in Aramaic (*mgd'*) indicating 'precious goods' (Jastrow 1903, 249), Syr. *magdā* 'some sort of fruit' (Payne Smith 1903, 249; Sokoloff 2009, 707), Heb. *meged* 'gute Frucht', 'Ertrag an Früchten' (*DGes*, 625 and *KAHAL*, 277). In addition, in Palmyrene and Samaritan Aramaic *mgd* assumes the meaning of 'gift' (*DNWSI*, 592; *PAT*, 377; Tal, *Sam*, 450).

In Greek we see the forms Μουγδεος<sup>37</sup> and Μαγδος. 38

The Greek section of the graffito includes the name Ουρεος which may probably be reconstructed as Βουρεος (see below no. 39; *CIS* II no. 1195).

#### **37.** From Wādī Mukattab.

On the top of the inscription the figure of a man with his arms in the air, presumably in prayer, appears; the figure is apparently barefoot and his sandals are presumably engraved between his legs. This image recalls the order given by the Lord to Moses (Exod 3,5) and Joshua (Gen 5,15), that is to take off their sandals for they were standing on the Holy Land.

```
Dating 2nd-3rd CE
```

```
Bibliography CIG no. 4668a; CIS II no. 1044; Euting 1891, 596; Levy 1860, 469-
   72, no. 47; Chabot 1901, 442-50; ESE I, 339-40; Clermont-Ganneau 1903, 59-
   66; RIGP nos. 2213 (Nabataean), 2209 (Greek)
```
Text and translation

1) *mdkyr 'wšw br ḥršw ṭryw bṭb* 2) μνησθη Αυσος Ερσου 3) καλ(ε)ιται Ο(?)υμαρου 4) εν αγαθοι[ς]

1) Let be remembered 'Āwšō son of Ḥeršō (which is called) Ṭarīō in good

2) Let be remembered Ausos (son of) Ersos


Commentary

Line 1. There is the mention of the name *'wšw*, well-known in the Nabataean epigraphy,<sup>39</sup> which comes from Arabic *'āws* 'gift' < *'aāsa* 

<sup>37</sup> DM, 76; Wuthnow 1930, 79.

<sup>38</sup> Wuthnow 1930, 68 and 147.

<sup>39</sup> Cantineau 1930-32, 2: 57-8; al-Khraysheh 1986, 29; Negev 1991, no. 52. Cf. pre-Islamic Arabic *'ws*, *'ws'*, *'wst* < *'s*, *'āssa* 'to build, found' or *'āws* 'gift' (*ICPAN*, 40 and 84) = OfA, CPA, LJLA *ʾwšʾ*, Qumran *'š* (Jastrow 1903, 35; Cook, *Qumran*, 25; *DNWSI*, 47) and BA *w'ūsayyā'* (Ezra 4,12) 'foundation'.

'to give, to offer'.<sup>40</sup> In Greek, the form corresponds to Αυσος. <sup>41</sup> The second name, *ḥršw*, is of Arabic origin too and according to Cantineau (1930-32, 2: 100) is linked to *ḥirš* 'wood, forest',<sup>42</sup> although this connection is not very convincing. In the Palmyrene epigraphy, though, the presence of *ḥrš*, *ḥršw* 'mute, dumb',<sup>43</sup> as well as *ḥryšw*  in Hatraean,<sup>44</sup> suggests that this name simply means 'mute, dumb', probably deriving from Arabic *'āḫraš* (cf. Heb. *ḥērēš*). Greek Ερσου (in *CIS* II no. 972) is recorded in other forms.<sup>45</sup>

The last name was read differently by the first editors,<sup>46</sup> but on the basis of a close analysis of the text we may read *ṭryw*, <sup>47</sup> a nickname of Arabic origin from *ṭariyy* 'fresh'<sup>48</sup> < *ṭariā* or *ṭaruwa* 'to be fresh (like meat)', also 'to be new, tasty, soft, tender'.<sup>49</sup>

In line 2 the Greek text begins,<sup>50</sup> and we subsequently reach the ambiguous part of the graffito since after καλεῖται almost all the editors interpreted the element ΟΥΜΑΡΟΥ as the name Οὔμαρος = 'Omar (Nab. *'mrw*?).51 The writer probably carved the name declining it in the genitive, Ουμαρου, maybe a nickname, and it is unclear

42 al-Khraysheh (1986, 92) proposes other vocalizations of the name. Negev (1991, no. 491) suggests reading *ḫaršī* 'to hunt lizards' (?), in connection with the hypothesis of Lankester Harding (*ICPAN*, 184).

43 *PAT*, 435 = Stark 1971, 23 and 90.

44 Beyer 1998, 159.

45 Ερσος, Ορσος, Ερασος, Ερας (Wuthnow 1930, 46, 47, 89, 143).

46 After the initial *ṭ* the second letter may be read *b* or *n*, so *ṭbyw* or *ṭnyw* (Chabot 1901, 445-6), the latter may identify a job, that is Ar. *ṭaniyy* 'merchant of dates', or a *non liquet* about its meaning (Clermont-Ganneau 1903, 61-2).

47 It is the first assumption proposed by Euting (1891, 596) and re-suggested by Stone (*RIGP* no. 2213). On the contrary Lidzbarski (*ESE* I, 339) read *ṭbyw*.

48 Cantineau 1930-32, 2: 102; Negev 1991, no. 513.

49 In the Koran (XVI, 14; XXXV, 12) it is an epithet referred to meat or fish; in Hebrew *ṭārî* 'fresh' is employed in connection with the 'wound' (Isa 1,6) and the jaw of a donkey (Jdg 15,15), so *ṭārâ* has the original meaning of 'plucking off' (Gesenius 1846, 324 = *DGes*, 428; *KAHAL*, 199). In Gal. and in LJLA *ṭry*, *ṭryt'* indicates a kind of salty fish sold at the market (Jastrow 1903, 554); presumably Greek borrowed from Aramaic the word θρίσσα (Att. θρίττα) 'twait shad' (*Alosa fallax*).

50 The Greek section was read differently in *CIG* no. 4668a: Μνησθῇ Αὔ[λ]ος Ἔρσου Καλιταίου Μάρου ἐν[θάδε ἐ]λθ[ών].

51 Clermont-Ganneau 1903, 62 and no. 1.

<sup>40</sup> In Palmyrene the form *'wšy* is attested (*PAT*, 429 = Stark 1971, 3 and 66); the root may also indicate the 'wolf' (Chabot 1901, 445, no. 1). The form *'wšw* is recorded in *Quellen*, 144, 171, 348, 361.

<sup>41</sup> Wuthnow 1930, 30 and 124; *AGLI*, 59, 62, 152; *IGLS* 21,5 nos. 152, 378, 432, 467, 522 (?), 572, 661, 666; The name Αὖσος is also present in an inscription of the Late Roman period dating back to the 3rd c. CE from the acropolis of Avdat in which Ausos was the grandfather of Amenos, one of the builders of the Nabataean temple (Negev 1981, 13-14, no. 1e). Another form of the name is Αουσος (Wadd. nos. 2205, 2511).

whether it is referred to Αυσος or Ερσου. According to the formulaic usage, after καλεῖται the employment of ὁ καί or τοῦ καί is expected. If the person mentioned in the inscription bears a second name, this would be put next to the first name, as is typical in the Greek-Palmyrene and Nabataean epigraphy.52 It seems that the author conformed Ουμαρου to the genitive in accordance with Ερσου making a mistake. We may suggest not reading Ουμαρου, but: ὁ (καὶ) Υμαρου<sup>53</sup> 'he who is the son of Umaros'.

**38**. From Wādī Mukattab.

**Dating** 2nd-3rd CE **Bibliography** *CIS* II no. 1194; *RIGP* nos. 2367 (Nabataean), 2371 (Greek)

Text and translation

1) *šlm 'wdw br 'lmb* 2) *qrw 'qry bṭb* 3) μνησθη 4) Αυδος Aλμο-5) - βακκερου

1) Peace. 'Awḏō son of 'Almoba-2) - qqerō 'Aqarī, in good 3) Let be remembered 4) Audos (son of) Almo-5) - bakkeros

#### Commentary

The addressee's name is *'wdw* = Gr. Αυδος, whose etymology is widely discussed in nos. 15 and 33.

The father's name, *'lmbqrw*, corresponding to Gr. Αλμοβακκερου, shows the definite article *'l-* before the personal name *mbqrw* < Com. *bqr* 'regarder avec soin, examiner'.<sup>54</sup> The participle active emphatic

<sup>52</sup> Clermont-Ganneau 1903, 63. It is common to use the formula: A called A son of B, or A son of B called B and never A son of B called A. In Nabataean, as well as in Palmyrene, next to the common Greek expression ὁ καὶ, ἐπικαλούμενος etc., we find *dy mtqr'* 'which is called' (cf. *VIS* no. 123a) with orthographic variations as in Nabataean of Sinai *mqtry* (see *CIS* II no. 1254 = Euting 1891, 567).

<sup>53</sup> In the bilingual no. 22 Υφφαλος is used without the diphthong ου.

<sup>54</sup> Cantineau 1930-32, 2: 61 and 73-4.

*mqbr'* means 'one who examines the victims (i.e. a certain priest)'.<sup>55</sup> Possibly the meaning of the name is 'examiner' – *'al-Mobaqqerō*. <sup>56</sup> The name is common in the area of Mount Sinai.<sup>57</sup>

In the Nabataean section a second name, *'qry*, appears referring to the previous one, which is absent in the Greek part; it is found in another Sinaitic inscription (*CIS* II no. 1010) and it refers to a demonym.<sup>58</sup>

**39**. From Wādī Mukattab.

**Dating** 2nd-3rd CE **Bibliography** *CIS* II no. 1195; *RIGP* no. 2372 (Greek)

Text and translation


Commentary

For the name *bry'w* = Gr. Βουρεος compare the commentaries at nos. 32 and 36. As regards the father's name, *š'd'lhy* = Gr. Σαδαλλου, <sup>59</sup> it is a theophoric in which the first element *š'd* means 'être heureux,

55 *DNWSI*, 187, s.v. "*bqr1* ".

56 al-Khraysheh 1986, 36.

57 There are occurrences in a graffito from the Wādī Haggag (Negev 1977, no. 221) and from other regions of Sinai too (see *CIS* II nos. 1222, 1489, 2214, 2383, 3196).

58 Cantineau 1930-32, 2: 134.

59 Other forms are Σαδδηλος, Σαδε(ι)λος, Σαδαλας (Wuthnow 1930, 101 and 173), Σαδαλλας, Σαδειλος, Σαδελος (*IGLS* 21,5 nos. 699, 454, 572).

propice',60 Ar. *sa'd* 'delight, luck', then *Sa'dallahi* 'luck of God',61 in addition it indicates the name of a tribe.<sup>62</sup>

**40.** From Wādī Mukattab.

**Dating** 2nd-3rd CE **Bibliography** *CIS* II no. 1197; *RIGP* nos. 2374 (Nabataean), 2401 (Greek)

Text and translation

1) *dkyr ḥr[y]šw* 2) *br 'myw bṭb* 3) μνησ- Αρισος 4) -θη Αμμαιος

1) let be remembered Ḥar[ī]šō 2) son of 'Ammayō, in good. 3) let be remember- Arisos 4) -ed (of the) Ammaios

Commentary

As regards the two names *ḥr[y]šw* = Gr. Αρισος and *'myw* = Gr. Αμμαιος, see respectively our inscriptions nos. 37 and 33. In the Greek section, μνησθη is divided into two parts; the final -θη continues in the following line. As far as Αμμαιος is concerned, we should expect Αμμαιου in the genitive case rather than in the nominative case; it is probably due to a scribal error. Ḥarīšō, son of 'Ammayō, should be the priest of T', the goddess who appears in other inscriptions from Sinai.<sup>63</sup>

63 See Cantineau 1930-32, 2: 154.

<sup>60</sup> Cantineau 1930-32, 2: 152-3. In Palmyrene we find *š'd'l* and the forms *š'd*, *š'd'*, *š'dw* and *š'dy* < *š'd* 'Luck' (*PAT*, 441 = Stark 1971, 53 and 115). For Hatraean, Beyer (1998, 162) cites *s'dw* written with /s/ *contra* Vattioni (1981, 118) who interprets it as *š'dw* with the same meaning.

<sup>61</sup> al-Khraysheh 1986, 181. See *s'dlh* in *ICPAN*, 319.

<sup>62</sup> Negev 1991, no. 1169.

**41.** Wādī Mukattab. (Plate VIII, no. 3)

**Dating** 2nd-3rd CE **Bibliography** *CIS* II no. 1032; *RIGP* no. 220

Text and translation

1) *dkyr 'yydw br 'wšw bṭb* 2) *['w]šw br 'yyd[w]* 3) Αυσος

1) let be remembered 'Oēḏō son of 'Āwšō, in good 2) ['Āw]šō son of 'Oēḏ[ō] 3) Ausos

Commentary

The inscription gives the name *'yydw*, possibly the diminutive form of the most frequent and documented *'ydw* (< Ar. *'āy'ḏ* 'qui cherche un refuge');<sup>64</sup> for a more detailed analysis see the inscription no. 15. The form Οαεδος (Wadd. no. 2472) is registered in Greek. As regards *'wšw* = Αυσος, see comments in the inscription no. 37.

**42**. From Wādī 'Aǧalah. This is not a real bilingual inscription since the probable Nabataean section was proposed by Euting.

**Dating** 2nd-3rd CE **Bibliography** Euting 1891, 253; *RIGP* no. 3562

Text and translation

1) Μνησθη 2) Χαλιος Zε-3) -δου 1) *k/ḥlyy* 2) *zydyy* 1) Let be remembered 2) Chalios (son of) Ze-3) -dos

<sup>64</sup> Cantineau 1930-32, 2: 129. In pre-Islamic Arabic *'ydd* 'aid, feast?' (*ICPAN*, 450).

1) K/Ḥalī 2) Zēḏī

Commentary

The name Χαλιος seems to be of Greek origin since the toponym Χάλιος (f. Χαλία), indicating the city of Boeotia, is mentioned by Stephanus of Byzantium.<sup>65</sup> In this case it is an Aramaic name, that is *ḥly* 'sweet', 'to be sweet', used in Egyptian-Aramaic,66 Syriac, Christian-Aramaic, Palestinian-Jewish, Mandaean and Samaritan.67 Therefore, it is probably an adjective 'one who is sweet, one who has a good heart' (cf. Ar. *ḥalwā* or *ḥulū*).

As regards the father's name, Ζεδου = Nab. *Zydyy*, see the explanation in the inscription no. 31. In the Nabataean section the two names appear with a final -*y* rather than with the usual -*w*; we can formulate two hypotheses: according to the first one, the final -*y* could be the adjectival suffix (as in Arabic); according to the second one, it could be the genitive case marked by the author.

**43**. From Wādī 'Aǧalah. (Plate IX, no. 1)

**Dating** 2nd-3rd CE **Bibliography** *CIS* II no. 1719; *RIGP* nos. 2925 (Nabataean), 2952 (Greek)

Text and translation

1) Ανεος 2) šlm hn'w

1) Aneos 2) Peace! Han'ēō

Commentary

The epigraphy gives a name, *hn'w*, already mentioned in the Sinaitic bilingual from Wādī Muġārah (no. 34). The Greek form, Ανεος, <sup>68</sup> is a variation of Ονεος which is preserved in this inscription.

65 Steph. Byz., *Ethnika*, 681 and Pape 1911, 1663.

66 Porten, Yardeni 1986-99, C1.1 (*Aḥiqar*), 148.

67 Cf. Jastrow 1903, 467; Payne Smith 1903, 143; Sokoloff 2009, 455; Tal, *Sam*, 271; *DNWSI*, 374.

68 Another occurrence of this name, though in the genitive case, is recorded in *AGLI*, 14.

### **Bīr Umm Ḍalfah**

**44**. This graffito was found at Bīr Umm Ḍalfah, a station located north-west of Safāǧā, along the road that went to Myos Hormos, an ancient seaport built on the Red Sea by the Ptolemies in the 3rd century CE. The epigraph seems to have been engraved by two different people who wrote down their names. (Plate IX, no. 2)

**Dating** unknown **Bibliography** Littmann, Meredith 1953, 8, no. 13

Text and translation

1) *šmrḥw* 2) Γαρα(μ)ου

1) Šimraḥō 2) Gara(m)os

#### Commentary

We notice that the initial *š* of *šmrḥw* is written in an unusual way. The letter has a quadrangular form and its vertical left edge is longer than the right; in fact, it looks like the Hebrew *q*; above the initial *š* there is a cross, probably symbolising the religious faith of the deceased. The name, mainly recorded in Sinai, is linked to the Arabic *šimrāḫ* 'rameau de palmier ou de vigne chargé de fruits'.<sup>69</sup> The root, also used in Arabic as *šumrūḫ*, may identify a stem of the branch 'of a palm tree originally related to a raceme of dates'.<sup>70</sup>

In the Greek section we notice the letter M of Γαρα(μ)ου cut in two parts; at the end of the line a symbol of obscure meaning appears. The name is the Greek form of Γαραμος, Γαρμος71 = Nabataean *grmw* and also occurs in no. 28 (see this inscription for the philological analysis).

<sup>69</sup> Cantineau 1930-32, 2: 152; al-Khraysheh 1986, 178-9; Negev 1991, no. 1158. The same is found in Litmann 1953, 13, no. 33 and in Litmann, Meredith 1954, 216, nos. 56 and 57.

<sup>70</sup> Lane 1863, 1596. Pre-Islamic Arabic *šmrḫ* shumrukh 'stalk of the palm?' (*ICPAN*, 357).

<sup>71</sup> Wuthnow 1930, 39; Wadd. no. 2513.

### **'Ābū Kū'**

**45.** The graffito was found at 'Ābū Kū', literally 'father of the corner' (also 'curve of the street'), a raised wādī, next to a hairpin bend, slightly west of the Roman station/well (*Hydreuma*) al-Muwayḥ; this station is located between Koptos (modern day Qifṭ) and Leukos Limen (modern day al-Quṣayr al-Qadīm).<sup>72</sup> (Plate IX, no. 3)

**Dating** about 1st-3rd CE **Bibliography** Littmann, Meredith 1954, 217-8, no. 59

Text and translation

1) *šlm 'bn 'lqyny br* 2) *'myw bṭb* 3) Εβναλκαιν-4) -ος Ομα-5) -ι(ο)υ

1) peace. 'Ēbn al-Qaynī son of 2) 'Omayō. In good. 3) Ebn Alkain-4) -os Oma-5) -i(o)u

Commentary

The Nabataean name *'bn 'lqyny* = Gr. Εβν Αλκαινος has been discussed in no. 7, the inscription that comes from the sanctuary of the goddess Allat in Wādī Ram. We notice the usage of Arabic *'bn* and the article *'l-*. The name is mainly recorded in Sinai.<sup>73</sup>

The father's name, *'myw* = Gr. Ομαι(ου), is also recorded in two other bilingual inscriptions from Sinai (nos. 33 and 40); we refer to these for philological explanations.

The Nabataean text is not particularly difficult to read and to interpret apart from the last three letters of *'lqyny*, line 1, since the final *y* seems a *t*.

72 Fournet 1995, 174.

<sup>73</sup> Cantineau 1930-32, 2: 143. The presence of *Qaynō* and *'Ēbn al-Qaynī* in the Nabataean onomastics may refer to the Qeniti, the tribe of Yethro (father in-law of Moses) who settled in Sinai and Negev and probably worshipped the god Yahweh (Zayadine 1990, 160).

As regards the Greek section, we notice a discordant writing. The N of ΕΒΝ shows a horizontal line, probably due to a mistake by the author, and the M of ΟΜΑΙΟΥ is not entirely visible.

In line 4, at the end of the sequence, there is a sign that could seem Greek -ιω. It could form the name Oμαιωου, but it is more probable that it is a symbol of the author (his signature?) and the signs in line 5 represent the suffix -I(O)Y of Ομαιου.

As concerns the dating of this epigraph, we have to consider that the graffito from Abū Kū' was drawn up in a period of time probably ranging from Augustus to Maximinus Thrax (1st-3rd century CE).<sup>74</sup>

# **Wādī Ǧiḍāmī**

**46**. The inscription was found in the Wādī Ǧiḍāmī on the way to Safāǧā on the Red Sea. (Plate X, no. 1)

**Dating** unknown **Bibliography** Littmann, Meredith 1954, 222-3, no. 71

Text and translation

1) Ουααβλας 2) *br ryšw br* 3) *'wšw šlm*

1) Wahballāh 2) son of Rayšō son of 3) 'Āwšō. Peace

Commentary

Curiously enough, the author of the graffito transcribed his name into Greek and his ancestors' names into Nabataean. Probably, the first is a mistake; in fact, we may expect Ουβαλλας οr Ουβαλας (see no. 7) instead of Ουααβλας. In this case the author used A to render Nabataean H.

The father's name *ryšw* is the equivalent of the Arabic *ra'īs* 'boss, leader',<sup>75</sup> while the ancestor *'wšw* is well-attested in the Nabataean epigraphy.

75 *Rā'is* is recorded as the name of an Arabic tribe in Egypt (Littmann, Meredith 1954, 222), but Negev (1991, no. 1071) interprets it as the Arabic *rašš* 'to sprinkle'.

Giuseppe Petrantoni

# Saudi Arabia

#### **al-Ruwāfah**

**47.** The archaeological site of al-Rawwāfah/al-Ruwāfah is placed 75 km southwest of Tabūk in modern day Saudi Arabia.<sup>1</sup>

The first mention of the monumental inscriptions of al-Ruwāfah is found in a report of journey of the 19th century,<sup>2</sup> while the discovery of a bilingual inscription (texts A, B and C), a Nabataean inscription (text D) and a Greek inscription (text E) came about only at the beginning of the 20th century.<sup>3</sup>

The bilingual text (A, B and C) is engraved on a lintel placed at the main entrance of the temple, while the Greek text (C) continues (lines 7-8) on the column that supported the left far end of the epistyle ending (lines 9-10) on the right column of the door. The other Nabataean inscription (text D) was found near the temple, while the

3 A. Musil (1926, 184-9, 258, 291, 312) came to the site on 25 June 1910 saying that he had found the ruins of a sanctuary and the inscriptions at issue. Later, H. St. J.B. Philby (1951, 448-59), staying at al-Ruwāfah from 20 to 24 January 1951, claimed to have found a fragmentary Greek inscription; when he came back the following year, he could no longer findeither the Nabataean inscription or the Greek fragment, which had probably been moved to the collection of Khalil al-Faraj. The Greek inscriptions (texts A, C and E) were published by Seyrig (1957, 259-61; *SEG* 19 no. 899).

<sup>1</sup> The correct spelling of the toponym is al-Ruwāfah (cf. Macdonald et al. 2015, 44 fn. 84).

<sup>2</sup> The description is given by Burton (1879, 1: 239) who says that some beduins showed him the fragment of a Nabataean inscription, which had been brutally broken, and of two other two slices that were lying on the ground.

Greek inscription (E) presumably comes from the capital of the column that supported the other part of the lintel.

**Dimensions** Texts A, B and C (6 lines) length 230 cm 1st line: height of letters 2.5 cm; 2nd line 1.7 cm; 4th line (Nabataean text) 2.2 cm; 5th line 2.5 cm<sup>4</sup> Text C, lines 7-8 height 8 cm, length 35 cm; lines 9-10 height 8.5 cm, length 24.7 cm, height of letters 2.5 cm **Dating** 2nd CE (164-169 CE) **Bibliography** *SEG* 45 no. 1995; Seyrig 1957, 259-61; Altheim, Stiehl 1969, 5: 548-51; Milik 1972, 54-8; Bowersock 1975, 513-22; Graf 1978, 9-11; *Quellen*, 295-302; Macdonald et al. 2015, 44-56

Text and translation<sup>5</sup>

# **A**

1) Υπερ αιωνιου διαμονης κρατησεως (τ)ων θειοτατων κοσμοκρατορ(ω)ν (Σ)εβα(στ)ων μεγιστων (Α)ρμενια(κ)ων Μαρκου Αυρηλιου Αντωνεινου και Λουκιου 2) [Αυ](ρη)λιου Ουηρου (π)[*lacuna of approximately 14 letters*]ν [Θ](αμ)ουδην(ω)ν (ε)[θνος lacuna of approximately 60 letters ] ΣΤΑ καθειδρυσεν μετα προτ(ρο)[πης] 3) [*lacuna of approximately 5 letters?*] και εκ (πει)[*lacuna of 25* 

*letters* Κοι]ντο [υ *lacuna*]

# **B**

4) *'l šlm' dy mt*[lacuna of approximately 30 letters *mrqs*] *'wrlys 'nṭwnyns wlwqys*

*'wrlys [wr]s dy '[---]'* [lacuna] *dnh nws' dy 'bdt (š)rkt tmwdw qdmy šrk(t)h* 

*lmhw' [š](w)h mn ydhm wmšmš(h) [ml'l]m* 

5a) [lacuna] *wḥp(y)t* [lacuna of approximately 12 letters] *['](dw) nṭs (')*

*(d/r/w)[.][.]ty* [lacuna of approximately 10 letters] *w(r)mṣhm*

4 Milik 1972, 55. The dimensions were provided by H.St.J.B. Philby who ignores lines 3 and 6.

5 The transcription and the translation of the present texts follow the reading of Macdonald et al. (2015, 50-1 and 54-5), and are adapted to the transcription conversions here employed.

#### **C**

5b) Επι νεικη και αιωνιω διαμονη αυτοκρατορων Καισα(ρ)ω(ν) [Μ](α)ρκο(υ) [Αυ]ρηλιου Αντωνεινου 6) και (Λ)ουκ(ι)ου (Α)υρηλιου Ουηρου Σεβ(αστων) Αρμ(ε)νιακ(ω)ν [Μηδι]κων (Πα)ρθικ(ω)ν με(γ)[ι](σ)των και του πα(ντος) (οι)[κου α](υ)τ(ω)ν το (τω)ν Θ(α)μου(δ)ηνων εθνος [*lacuna*]

7) τον νεω συνετελε(σε)ν 8) και το ιερον καθ(ε)ιερωσεν 9) [… Κλ]αυδιου Μοδεστου 10) […](β) αντιστρατ(ηγου)

# **D**

1) *(d)nh (by)t' dy [']bd (š)'dt 'pkl* 2) *(')[l](h)' (mg)yd(w) dy mn rbtw* 3) *l'l(h') 'lh […(m)…(k)…]ḥ(p)yt mr'n' [ ] hgmwn'* 4) *mr'(n') (') […] (h)gmwn'* 5) *(.) […] (')mnw*

# **E**

1) (E)/(C)YCΘΑ(Ι)/(T)Ο(Y) (Θ)[αμ](ου)δ(η)νω-2) ν φυλης Ροβαθου οικοδο-3) μησα<ν> το ιερον τουτο

# **A**

1) For the eternal duration of the power of the most divine rulers of the world, the great (A)ugu(st)i, *(A)rme(n)iaci*, Marcus Aurelius Antoninus and Lucius 2) [Au]relius Verus [*lacuna*] (*n)[atio]* of the [T](ham)ud has founded [*lacuna*] with the enco(u)[ra]gment 3) and through [*lacuna* Q]uintus [*lacuna*]

4) For the well-being of [*lacuna* Marcus] Aurelius Antoninus and Lucius Aurelius (Ver)us who [*lacuna*]. This is the temple which the *(n)atio* of Thamud made, the commanders of their *natio*, for the existence of which it was set in place by their hand and their worship [will be there, for ever]

5a) [lacuna] and with the encou(r)agment of [*lacuna*] [A](dv)entus [*lacuna*] and at their request

# **C**

5b) For the victory and the perpetual continuance of the emperors, the Caesa(r)s, [M](a)rc(us) [Au]relius, Antoninus 6) and (L)uc(i)us (A)urelius Verus, Aug(usti), *Arm(e)n(i)aci, [Me] dici, (Pa)rth(i)ci Maximi* and their wh(ole) (ho)[use] the *natio* of the Th(a)mu(d) [*lacuna*]

7) have comple(te)d the temple 8) and ha cons(e)crated the sanctuary.

9) [… of Cl]audius Modestus 10) […] Proprae(tor)

# **D**

1) (T)his is the (temp)le which (Š)a'aḏaṯ, the priest of 2) ('I)[la](h)ā son of (Mu)ḡīḏ(ō) who is from (*the tribe*) Roḇaṯō, [m] ade 3) for 'Ila(hā), the god of[…(m)…(k)…] the en(c)ouragment of

4) o(ur) Lord […] the (g)overnor

5) […] ('A)manō

# **E**

1) (E)/(Ϲ)YϹΘA(I)/(T)O(Y) of the tribe of Thamud

2) of the tribe of Robathos they bu-

3) -ilt this sanctuary

#### Commentary

#### **Bilingual inscription (A, B, C)**

It contains a dedication to commemorate the feats of the emperors Marcus Aurelius Antoninus and Lucius Verus in the Near East; the campaign against Vologases IV, who attacked the Roman Empire in 162-166, is mentioned. The inscription indicates that the temple was erected by the Thamud tribe between the Roman victory over the Parthians and the death of Lucius verus, so between 166 and 169 CE.<sup>6</sup>

The semidivine image of the two winning *Augusti*, protectors of the Armenians (Σεβαστῶν μεγίστων Ἀρμενιακῶν), is highlighted by the label τῶν θειοτάτων κοσμοκρατόρων<sup>7</sup> to whom Milik maintains the Nabataean *mtmkyn*<sup>8</sup> corresponds. Conversely, Macdonald et al. (2015, 50), after a new revision of the epigraph, read only *mt*. As is clear from the two texts, the temple was built by the Thamud tribe. Even though the reference to Θαμουδηνῶν ἔθνος is difficult to read, we have a clear connection with it in the Nabataean part in which the sentence *šrkt tmwdw*<sup>9</sup> is mentioned. The substantive *šrkt* was translated by Milik (1972, 56) as 'fédération' rather than as 'tribe'.<sup>10</sup> On the basis of its mentions by Ptolemy and Stephanus of Byzantium some scholars have suggested that the term *Saracen* originated from the name of the tribe at issue through Greek Σαρακηνοί. 11 Macdonald (1995, 98-100) identified *šrkh* with 'military unit', but this implies that at al-Ruwāfah the Greek term ἔθνος took on the meaning of 'a

6 Graf 1978, 10.

10 Cf. Ar. *šāraka* 'to share, to take part, to participate', Syr. *šrek* 'to be left, to end up in a state' (Payne Smith 1903, 598b; Sokoloff 2009, 1608-9), also Akk. *šarāku* 'to make a votive offering, to grant, to entrust' (*CAD* XVII, 40; *AHw* III, 1183-4).

11 *DNWSI*, 1193; for a detailed examination, cf. Graf, O'Connor 1977, 52-66; O'Connor 1986, 213-29 and Macdonald 1995, 93-101 and especially Macdonald et al. 2015, 46-7.

<sup>7</sup> Other two dedications to Marcus Aurelius and Lucius Verus come from Philadelphia (Amman), in an inscription dating back to 161-169 (*IGLS* 21,2 no. 18), and in a text from Gerasa dating back to 163 (*CIG* no. 4464); in both epigraphs the two leaders are called Aὐτοκράτωρ.

<sup>8</sup> If the reading is right, it is a participle form from the root *tmk* 'the rulers of the whole world (said of two Roman emperors)', 'to support' (*DNWSI*, 1221). The root is used in Hebrew *tāmak* meaning 'to take hold of' (Gesenius 1846, 872 = *DGes*, 1444; *KAHAL*, 659-60), so in Akk. *tamāḫu*, also *tamāku*, with the same meaning (*CAD* XVI-II, 107; *AHw* III, 1312).

<sup>9</sup> Actually, in the Greek version there is not a clear mention of Θαμουδηνῶν ἔθνος since during the recovery of Musil the stone had already deteriorated; in fact, Philby himself did not include this construction in his copy (Seyrig 1957, 260). This is also apparently confirmed by Altheim and Stiehl (1969, 5: 24) who do not mention it in their transcription of the epigraph.

group of people united in some way' as in Pseudo-Hyginus.<sup>12</sup> Milik does not specify that the first letters in line 3 do not begin at the left edge where the words of lines 1-2 are all aligned.<sup>13</sup>

In the Nabataean text, after the references to Antonius and Verus, Milik originally reconstructs *'rmny'* 'Armenians', but Macdonald points out that there is not enough space between *m* and final *'* order to recreate the ethnonym.14 After that there is the term *nws'* 'temple', a clear loanword from Greek ναός; <sup>15</sup> the latter is not visible in section A where only the verb καθείδρυσεν (< \*καθδρύω) 'to settle, establish' (but Nabataean *'bd* 'to do') remains.

At the end of line 4 the verb *šmš* appears, here in *pa'el* form meaning 'to serve'<sup>16</sup> in the sense of 'divine service in the sanctuaries', the root was subsequently used by Christian Assyrians to identify the *diaconus*. 17

In line 5a there is the verbal name *ḥfyt* that must not be confused with the Aramaic root *ḥpy* 'to cover a surface';<sup>18</sup> it is loanword from Arabic *ḥafiy bihi* 'to show someone honour, kindness, to manifest joy' = Gr. προτροπή. <sup>19</sup> In the same line Milik reconstructs *'nṭsṭys 'dwnts* 

13 Macdonald et al. 2015, 50 fn. 120. In addition, Milik, after the quotation of Κοίντος, reconstructs the expression: [Ἀντιστίου Ἀδουεντου πρεσβευτου Σεβαστῶν ἀντιστρατέγου...] identifying the legatus Quintus Antista Adventus.

14 Macdonald et al. 2015, 50 fn. 122.

15 The word is also employed in Palmyrene, Galilean Aramaic and Syriac indicating the citadel, 'a palace area of a tell' (*DNWSI*, 723).

19 Lane, 604-5. Cf. Arabic *ḥifāyah* (pl. *ḥufawā'*) 'care, attention'. In a bilingual Greek-Palmyrene inscription the term *ḥpyw* = Gr. σπουδή is found (*DNWSI*, 395).

<sup>12</sup> In *De munitionibus castrorum* (§29 and §19, 43), a work probably dated to the period of the inscription of al-Ruwāfah, the term *natio* = Gr. ἔθνος is employed with the acceptation of military unit comprising particular ethnic groups. In Wadd. no. 2196 there is στρατηγοῦ νομάδων, a title referred to an Arab leader; these nomads erected a statue for an imperial legatus of the province (Wadd. no. 2203): οἱ ἀπό ἔθνους νομάδων. Ιn this case the term ἔθνους refers to a 'military unity' of nomads, as it is clear from another inscription from Hauran (like the previous two texts) in which we read παρεμβολὴ νομάδων translated as 'escadrons de nomades' (*PPAES* IIIA, 752).

<sup>16</sup> In Com., BA *ysamsûnēh* (Dan 7,10), Pal. (*PAT*, 416) 'to serve'; in Syr. *šameš* also 'to provide, to recite', while in JBA, PTA, and Gal. *šmš* + *'m* 'to lie with, copulate' (Jastrow 1903, 1602; Payne Smith 1903, 585; Sokoloff 2009, 1576; Cook, *Qumran*, 239; *DJ-PA*, 559a; *DJBA*, 1162a; *DNWSI*, 1168).

<sup>17</sup> Cf. Syr. *mšamšayā* and Ar. *šammās* 'deacon, sacristan', he who serves Mass in the Christian oriental churches' rituals. The root, which may be confused with Sem. *šms*/*š*  'sun', is of uncertain origin. It seems without doubt to be of Egyptian origin or even an Egyptian loanword *šamš* 'to serve', rather than from the root meaning 'sun' in the sense of 'to serve the sun!' (Klein 1987, 668). Cf. Sem. \**śamš-* = Pro. Afro-As. \**ŝam-* 'sun' linked to the root meaning 'to burn, to shine' (*HSED* nos. 2328 and 2329).

<sup>18</sup> *DNWSI*, 394-5.

*hgmwn'*, i.e. legatus Quintus Antistius Adventus<sup>20</sup> who is identified through the title of *hgmwn'* 'governor, leader'; *hgmwn'* is evidently a loanword from Greek ἡγεμών. Macdonald does not agree with this reading and, conversely, he points out that there are two *ṭ* separated by a wide *lacuna* and in the space between *ḥfyt* and *'dwnṭs* about 13 letters fit in, too many to reconstruct what Milik proposed.<sup>21</sup>

At the end of the line we read the verb *rmṣ* probably meaning 'to pacify', used only in Nabataean with this meaning;<sup>22</sup> according to Milik, who proposed this interpretation, it is a loanword coming from Arabic *ramaṣa bayna* 'to make peace between', 'réconcilier des ennemis, mettre la paix entre eux'.<sup>23</sup> By contrast, Macdonald suggests that this translation is philologically impossible because here *ramaṣa* is not followed by *bayna*, but by the plural pronoun -*hm* whose translation 'made peace between them' appears to be unacceptable; so, according to Macdonald the form is a *maṣdār*, i.e. *ramṣ* and the sentence *wa-ramṣ-hum* would mean 'and at their request'.<sup>24</sup>

The text C includes the name of L. Claudius Modestus who probably was a governor of Provincia Arabia in the 2nd century.<sup>25</sup>

#### **Inscriptions D and E**

The two texts describe the building of the sanctuary. The Nabataean section (D) mentions *š'dt* (a name that recurs in no. 39), probably a minister who built the sanctuary. In the inscription we notice the term *'pkl* (also *'pkl'*), a loanword from Akkadian *apkallu* (< Sumerian *ab-gal*) used in Nabataean, Palmyrene and Hatraean meaning 'reli-

<sup>20</sup> He was legatus of the legion *VI Ferrata* in Syria Palaestina and commanded a legion, the II *Adiutrix*, in the war against the Parthian Empire (161-166). Probably from 165 to 167 he was governor of Arabia Petraea and he is cited in an epigraph from Bosra (*Corpus Inscriptionum Latinarum* III no. 92) and in another from Gerasa (*SEG* 7 no. 822).

<sup>21</sup> Macdonald et al. 2015, 51.

<sup>22</sup> *DNWSI*, 1078. More generally the root *rmṣ* in used in JBA 'to wink, to drip' and in LJLA and Sam. 'to embroider, to set (gems)' (Jastrow 1903, 1483; *DJBA*, 1088b; Tal, *Sam*, 839).

<sup>23</sup> Kazimirski 1860, 1: 924-5.

<sup>24</sup> See Macdonald et al. 2015, 51 fn. 126 and Macdonald 2009, 11-12.

<sup>25</sup> The name may refer to the consul named in 155 CE in the Acts of Arval as *frater*  (see *Corpus Inscriptionum Latinarum* VI no. 2086). If this identification is right, he would be Claudius Modestus who became consul in 152 CE, as suggested a mutilated inscription from *Beneventum* (*Corpus Inscriptionum Latinarum* IX no. 1574). According to some scholars, on the contrary, he could be the father of the governor of Arabia (for a detailed analysis of the question cf. Camodeca 1983, 207-12).

gious functionary, priest',<sup>26</sup> and 'exorcist-priest'<sup>27</sup> too. The personal name *mgydw* comes from Arabic *muǧīd*, another form of the name *mgdw* and *mgdyw* (see no. 36). The god 'Ilahā, to whom the temple is dedicated, seems to have assumed a special significance for the tribe,<sup>28</sup> whose name *rbtw* (/roḇaṯō/?) = Gr. Ροβαθου may be a literal translation of al-Ruwāfah.<sup>29</sup>

In the inscription E the initial sequence, which Milik originally read ϹIϹΘAΙΟΙ (on the copy of Philby), turns out to be problematic in its interpretation.<sup>30</sup>

<sup>26</sup> *PAT*, 342; *DNWSI*, 95-6

<sup>27</sup> Healey 1995, 78. According to Kaufman (1974, 34) it is an Arabic title rather than Aramaic.

<sup>28</sup> Milik 1972, 58.

<sup>29</sup> Bowersock 1971, 231. The name appears in *Not. Dign.* [*occ*.] [*or*.] 34.27 as *Robatha*.

<sup>30</sup> According to Van den Branden it could be a graphic representation of the Arab custom to cut the hair and to shave the temples, in fact, considering O rather than Θ, he reconstructs the word σισόη translating it as 'manière de couper les cheveux en rond', also 'roll of hair' (*LSJ*, 1601); but this suggestion proves to be contrived (Macdonald et al. 2015, 54 and fn. 138). For further considerations, see Graf 1978, 10.

Giuseppe Petrantoni

# Lebanon

#### **Sidon**

**48.** The inscription comes from Sidon. It was discovered in 1866 in a garden near the castle of St. Louis, in which there were a lot of fragments of columns and residues of marble belonging to the same construction of the stone in question.<sup>1</sup>

The form of the Nabataean letters is typical of the classical period, between the 1st century BCE and the 1st century CE, although here *'* has an altered writing (for example, see the *'* of the third line). (Plate X, no. 2)

**Dimensions** height 15 cm; length 14 cm **Dating** 1st BCE (4 BCE?) **Bibliography** De Saulcy 1867, 9-14; Renan 1868, 537-9; *VIS*, 113-4, no. 7; Levy 1869, 435-40; *CIS* II no. 160; *RES* nos. 482.1, 2092.1; Roche 1996, 75-7; Nehmé 2003, 4-6; *Quellen*, 131

Text and translation

1) *d' rb't' dy[ 'bd ---]* 2) *'srtg' br zw[---]*  3) *ldwšr' 'lh[---]* 4) *5 lḥrtt[---]*

1 De Saulcy 1867, 9. The epigraph was found about 5 metres away from the south door of the city. Littmann (*PPAES* IVA, 4) suggests that the inscription has been engraved on the jamb of a door or on an altar rather than on an architrave.

5) Ζωιλου στρατηγος [---] 6) -σεν 1) This is the *rb't'* that [made ---] 2) the strategist son of Zū[---] 3) in honour of Dūšarā god of[---] 4) 5 of Areta[---] 5) (son of) Zoilos, strategist, [---] 6) –

# Commentary

In line 1 the word *rb't'* appears. It is a noun, found only in Nabataean in five inscriptions,<sup>2</sup> whose meaning has been interpreted in many different ways.<sup>3</sup>

In Palmyrene the term *rbw'h* is registered in an inscription on a slab inside the tomb of Malkō, in the necropolis located south-west of Palmyra, identifying an 'alcove' at the bottom of which there are some burial recesses.<sup>4</sup>

From a philological point of view, in Arabic *raba'a* means 'to gallop', 'to graze the grass in spring'.<sup>5</sup> In the *fa''ala* form *rabba'a* means


<sup>2</sup> At Ḫarabā (*VAS*, 195, 77 = *RES* no. 88), Bosra (*RES* no. 2092), Tell al-Šuqāfiyyah (found in 1982, published in 1988 and republished in 1990 by Jones et al. 1988, 47-57; Fiema-Jones 1990, 239-47), Petra (found in 1981 and published by Hammond 1996, 132) and Kos (see no. 51).

<sup>3</sup> For Renan (1868, 539) the root of the word, which defines a square object, may derive from the Aramaic *rb'*, to compare to a sort of cubic *naos*, a κλίνη 'bed, couch' (Latin *pulvinar* on which the statue of the god was laid down to offer him a banquet, *lettisternio*), but it could also be a synonym of *bayt* 'house', 'temple'. De Vogüé translates it as 'altar', 'base or votive tablet' (*VIS*, 113-4). Levy (1869, 436, he cites a note of Nöldeke) interprets the term as a 'square block' since its presence in the Mishnah (Middoth 3,5): *wrby'yn šl 'rṣ* 'square blocks of cedar'. Clermont-Ganneau (1906a, 216) supposes a derivation from the root *'rb'* 'four', identifying an aedicule characterized by four columns, galleries or doors, exactly as in Greek τετράστυλον, τετράπυλον and τετράστοον (= Latin *atrium*). Littmann (*PPAES* IVA, 4-5), in agreement with Clermont-Ganneau, thinks that *'rb'* followed by the adjectival suffix -*n'* is a nominal stem *'rb'n'* that designates a 'cell', a 'recess'; Cantineau (1930-32, 2: 145) suggests that *rb't'* means 'votive tablet', whereas *'rb'n'* 'recess, cell, square sanctuary'. So, it is a building made up of four elements recognizable in an inner cell of certain Nabataean temples, in particular the two temples of Sī' that, taking into account the drawing of H.C. Butler at the beginning of the century, had a square form (Nehmé 2003, 17 and no. 46: the excavations led by J.M. Dentzer in the sanctuaries of Sī' show that the cell of the temple 2, dedicated to Dūšarā, was rectangular whereas we do not know if the cell of Ba'alšamīn was rectangular or quadrangular). See also *DNWSI*, 1058: 'cell', 'rectangular sanctuary' and *Quellen*, 92 'ein Weiherelief'. As regards the inscription of Tell al- Šuqāfiyyah, the editors have translated *rb't'* as '(quadrangular) shrine' (see the comments in Jones et al. 1988, 49).

'multiply by four' and from its derivatives the idea of 'four, square' originates, as in the other Semitic languages.<sup>6</sup>

The term *rb't'* could be an Arabic loanword from *ribā'ah*, a sort of *ḥamālah* that means 'obligation, duty, responsibility' to return the favour;<sup>7</sup> presumably it refers to a 'building erected in honour of a deity to repay it for the grace'.<sup>8</sup>

According to Levi Della Vida, it is simply a 'rectangular object' referred to the tablet carrying the votive inscription.<sup>9</sup>

In the missing part of the epigraph about ten letters could fit in with the *lacunae* and possibly the first line ends with the verb *'bd* followed by a personal name.<sup>10</sup>

In line 2 there is the term *'srtg'* < Gr. στρατηγός, which is used only in Nabataean in this form<sup>11</sup> assuming the meaning of 'général (d'infanterie)';<sup>12</sup> it refers to the son of Zωιλος (a name occurring in no. 10), presumably a Nabataean official who was present at Sidon.<sup>13</sup> At the end of the line we distinguish two letters, *zw*, that are the initial part of *zw'l*<sup>14</sup> (?).


13 For an explanation regarding the presence of a Nabataean official at Sidon, see Roche 1996, 76-7.

14 In Quellen, 131 the authors read /q/ rather than /z/ in order to restore the name *qw[yl']* or *qw['l]*.

<sup>6</sup> In some varieties of Aramaic *rb'* assumes the meaning of 'to lie down', 'crouch down (in the manner of an animal)' and also 'to make square'. See JBA, PTA, Gal., Sam. (Jastrow 1903, 1444; *DJPA*, 514b and 515a; *DJBA*, 1058a; Tal, *Sam*, 812; Cook, *Qumran*, 218; *DNWSI*, 1055). Also in Hebrew the original meaning of *rb'* is 'to lie down' with interchangeable /'/ - /ṣ/, so much so that the verb *rbṣ* has the same meaning, as well as in Arabic *rabaḍa* (passage /ṣ/ > /ḍ/) means 'to kneel', 'to lay down', 'to sit upon thighs or buttocks' (Lane, 1011; *DGes*, 1215-1216; *KAHAL*, 526-7). In Ugaritic *rb'* means 'to quadruple', 'four' (*CLUC*, 280-1). In Akkadian *rabāṣu* has two meanings: referred to animals it means 'to sit', referred to men it means 'to camp, to lay down' (*CAD* XIV, 10 and ff.; *AHw* II, 933-4). As L. Nehmé points out, the term assumes a specific meaning in Late Aramaic, in fact, in Syriac we find *rba'* 'to recline, to dine' (in Origen's Hexapla in Isa 17,2 the LXX's κοίτη 'bed' was translated *raba''*). Cf. the complete philological explication in Nehmé 2003, 20-3.

<sup>11</sup> *DNWSI*, 92. See the metathesis /rt/ for /tr/. The transcription indicates the original Aramaic spelling of the term; in fact, it is a Greek word that entered Aramaic as a loanword. The same metathesis happens in Jewish Aramaic, Palmyrene and Syriac: *'sṭrṭ''* (Monferrer-Sala 2013, 104-5 and 109). For the role and the characteristics of *'srtg'* in the Nabataean epigraph see Nehmé 2015, 103-22.

<sup>12</sup> Cantineau 1930-132, 2: 66. Unlike *hprk'* (< Gr. ὕπαρχος) that among the Nabataeans designated the 'général de cavalerie' before the Roman conquest (Cantineau 1930- 32, 2: 88), also the 'prefect' (*DNWSI*, 292).

The text dates back to 4 or 3 BCE, in fact in line 4 we read '(year) 5 of Aretas',<sup>15</sup> referred to Aretas IV.

In line 5, after the mention of στρατηγός, there is sufficient space to include a verb. In *editio princeps* De Saulcy (1867, 9) integrates with (ᾠκοδόμη)σεν, while in other editions (ἀνέστη)σεν is preferred.

At Sidon during the reign of the local king 'Ešmun'azar II (Phoen. *'šmn'zr*<sup>16</sup>) the temple complex dedicated to 'Ešmun, the tutelary deity of the city associated with Greek Asclepius, was expanded. The presence of a huge sanctuary represented for a Nabataean, who traded incenses and aromas, an opportunity to stop over and carve a dedication to his gods, in this case Dūšarā, as we may read in line 3.

16 See Krahmalkov 2000, 83.

<sup>15</sup> De Saulcy (1867, 14) read the number '13', while Levy (1869, 437) and Vogüé (*VIS* no. 114) the number '32'. Kammerer (1929-30, 463, no. 7) dated the inscription to 23/24 CE. In the text it is clear that the year is engraved by three oblique and two joined lines tending to the left. The correct reading, 'year 5', is recorded in *CIS* II no. 160.

Giuseppe Petrantoni

# Aegean Sea

#### **Miletus, Turkey**

**49.** The epigraph was found at Miletus in 1905, near the temple of Apollo Delphinios during the excavations conducted by a German archaeological mission led by Th. Wiegand on behalf of the Royal Museum of Berlin (today known as Altes Museum).<sup>1</sup>

In the Nabataean section the script is cursive tending to the ligature, although it is curious considering the nature of the inscription and the dating of its editing.

It is a dedication commissioned by Sylleus, the *epitropos* of king Obodas III, during his journey to Rome. According to Josephus<sup>2</sup> and Strabo<sup>3</sup> the stone could be dated to 10 or 9 BCE. (Plate XI, no. 1)

```
Dimensions height about 13 cm; length 13 cm
Dating 1st BCE
Bibliography Clermont-Ganneau 1906b, 305-29; 1906c, 159-60; 1907, 289-
   91; 1924, 114; RES nos. 675 and 1100; ESE III, 89; Kawerau, Rehm 1914, I.3, 
   387-9; Cantineau 1930-32, 2: 45-6; McCabe 1984, 295; Vattioni 1987-88, 108, 
   no. 21; Dijkstra 1995, 70-2; Roche 1996, 80-2, no. 9; Quellen, 127-8; Healey 
   2009, 120-1, no. 18
```

<sup>1</sup> The text of the inscription was published by Mordtmann, who gave the transcription of the Nabataean section with some wrong additions, but he did not provide a photograph of the stone. Cf. *editio princeps* in Wiegand 1906.

#### Text and translation

1) *šly 'ḥ mlk' br tym[---]* 2) *[---]t' 'l ḥyy 'bdt mlk' byrḥ ṭ[bt]* 3) [Συλλ]αιος αδελφος βασιλ[εως---] 4) ανεθηκεν Διι Δου[σαρει---]

1) Šullay (/Šollē) brother of the king, son of Taym[---]

2) [---] for the life of Obodas the king, in the month of Ṭ[ēbēt]

3) [Syll]aios, brother of the ki[ng---]

4) erected for Zeus Du[sares---]

#### Commentary

In line 1 Sylleus is named as *'ḥ* = Gr. ἀδελφός 'brother',<sup>4</sup> here interpreted as a sobriquet that identifies an official relationship or a vaguer kinship.<sup>5</sup> Sylleus' father's name is *tym-*, probably *tymw* meaning 'servant' (for the equivalent Greek form Θαιμος see nos. 8, 20 and 22).<sup>6</sup>

In line 2 we notice the letters *t* and *'*. Healey (2009, 120-1) restores and reconstructs the word *mdt'* 'city, town' referring to Miletus. At the end of the line only the sign *ṭ* is visible, linked to *ṭbt*, that is the name of the tenth Hebrew month, the fourth of the secular calendar (Esth 2,16); it corresponds to a part of December and a part of January and it was simply called 'the tenth month' (1Chr 27,13).

In line 4 the Greek part presents an important case of cultural syncretism in which the cult of Zeus is associated with that of Dūšarā.

Littmann (*PPAES* IVA, XV-XVI) speculates that Sylleus came to Miletus carrying a handwritten copy of the text, drafted by himself on a papyrus or a parchment. Later he delivered it to a Greek mason who translated the dedication and copied the Nabataean part.

<sup>4</sup> Both Josephus and Strabo call Sylleus "brother" or "*epitropos*".

<sup>5</sup> Healey 2009, 121. See also Starcky 1966, 939.

<sup>6</sup> Sylleus' father's name is recorded in another inscription from Puteoli, in *CIS* II no. 157. The name is also found in Hatraean (Beyer 1998, 167) and Palmyrene (*PAT*, 441 = Stark 1971, 54-5 and 117).

#### **Delos, Greece**

**50.** The epigraph was found on the island of Delos. The inscription is engraved on a small white marble stone that is considerably ruined and broken on both left and right parts.<sup>7</sup> This is another dedication to king Obodas III commissioned by Sylleus.

The difficulty in reconstructing the text is related to the wear of the stone that complicates the reading. Nevertheless, we notice three initial lines written in the Nabataean script and three more in Greek.

**Dimensions** height of the right part 185 cm; length 33 cm; thickness 15 cm; height of the Nabataean letters 2 cm; height of the Greek letters from 18 to 2 cm<sup>8</sup> **Dating** 1st BCE (9 BCE)

**Bibliography** Plassart 1928, 263; *ID* 5 no. 2315; Bruneau 1970, 244-5; Milik 1976, 146, no. 11; Vattioni 1987-88, 106, no. 15; Roche 1996, 83-5; *Quellen*, 124-6

Text and translation

```
1) - r-
2) [---] ḥyy 'b[d]t [---]
3) [---] dn [---]
4) [---]ισια[---]
5) [---]υδο[υσαρει---]
6) [---]ορον[---]
1) - [Dūša]r[a]
2) [---] for the life of 'Obo[d]as [---]
3) [---] this [---]
4) [---](Syllaios ?)[---]
5) [---](to Du)[sares---]
6) [---] ? [---]
```
### Commentary

In line 1 we only read *r* followed by unspecified signs.<sup>9</sup> In line 2 we can distinguish *ḥyy* 'life' and*'b[d]t* 'Obodas'.<sup>10</sup>

7 Plassart 1928, 263.


In the Greek part, the first line is intended as ΙΣΙΑ<sup>11</sup> (in *ID* 5 no. 2315), while in line 5 ΥΔΟ and in line 6 ΟΡΟΝ appear. The inscription was also studied by Starcky and Milik who read in lines 2 and 3: 'for the life of Obodas the king, the king of the Nabataeans'; while in lines 5 and 6: '[D]ii Dou[sarei] Sul[laios]'.<sup>12</sup>

At a later stage Milik suggests reading the inscription as completely as possible through the insertions of many reconstructions and the addition of a seventh line.<sup>13</sup> He rebuilds a dedication made by Sylleus to Obodas III. Moreover, it is probable that the inscription dates back to 9 BCE and was written after that of Miletus. In addition, among the pottery remains in the sanctuary B three lamps of the Imperial era and the bottom of a lamp, today lost, carrying the inscription [Ζ] ωΣΙΜ[ος] were found; these findings are a proof that the site was frequently visited starting from a later period (1st century CE), so the dedication of Sylleus is a testimony to the fact that the sanctuary already operated towards the end of the 1st century BCE<sup>14</sup>

A further proof that the inscription was written by Sylleus is the discovery at Delos of a sculpted-head, representing Obodas III, belonging to the dedications produced by Sylleus during his journey to Rome.<sup>15</sup>

12 Bruneau 1970, 244.

13 Milik 1976, 146, no. 11. He presents the following reconstruction of the epigraph: 1 *['bd šly 'ḥ] m[lk' br tymw]*


Translation: 1 [A fait Šullai frère du] r[oi fils de Taimu] 2 [le Nabatéen] la tablette en l'honneur de Dûša[râ dieu] 3 [de Gaiâ pou]r le salut de 'Ubdat le ro[i] 4 [et cela au jou]r 5 de shebat en l'an 2[1] (février 8 av. J.C.) Syl[laios] 6 à Dou[sares] 7 la ta[blette].

Vattioni (1987-88, 106) believed that Milik's transcription was simply an adaptation of that proposed by Bruneau who conversely does not follow the Nabataean-Greek layout of the original text.

It should be noted that Milik reads in line 2 *lwh'*, a term used in Com., meaning 'stone tablet' on which a text may be engraved (*DNWSI*, 569). In line 3 he identifies *Gay'ā* (Ar. *'al-Ǧī*), the name of the western suburb of Petra.

14 Bruneau 1970, 244-5.

15 al-Salameen 2008, 25.

<sup>11</sup> Plassart (1928, 263) initially reconstructed [ὑ]ψίστ[ῳ], while he transcribed the whole Greek section as follows: (1) -YΙΣΤ - (2) – ΥΙΔΟ – (3) – ΟΙΟΛ -. He did not provide the Nabataean part.

#### **Kos, Greece**

**51.** The inscription was found on the island of Kos,<sup>16</sup> in March 1936, near the little church of Saint George. Mario Segre announced this discovery at a marginal note in the *editio princeps* written by Levi Della Vida.<sup>17</sup> The inscription was not carved in the place in which it was found, in fact, it is possible that the epigraph comes from a sanctuary brought to light dozens of metres away from the above-mentioned church. It was probably a sanctuary dedicated to Aphrodite Pandamos and Pontia. According to Segre, this suggestion is corroborated by the fact that the sanctuary, being located in the port, was frequented by traders who stopped over on the island; this is proved by an inscription mentioning a sacrifice to the ἔμποροι and to the ναυκλᾶροι, travellers and ship owners who sailed away from Kos. Therefore, it is possible that a Nabataean merchant made a vow to his own goddess assimilated to Aphrodite.<sup>18</sup>

The inscription consists of eight lines, of which the first five are written in Nabataean and the others are drafted in Greek, carved onto a white quadrangular marble stone, cut in the rear and upper part. The stone is also damaged on the right side. According to Levi Della Vida on the top of the stone, on the left, we may glimpse a cut decoration in the shape of festoon from which a twirling tendril climbs down; the same figure would appear on the right side.<sup>19</sup> The two decorations, as other funeral altars from Kos show, were probably linked on the top by a garland crown, even though this kind of motif had not previously featured in votive monuments.<sup>20</sup> (Plate XI, no. 2)

**Dimensions** height 51 cm; length 60 cm; thickness 37 cm; the remaining part of the block is 41 cm high and 40.5 cm long; height of the Nabataean letters 3 cm; height of the Greek letters 2 cm **Dating** 1st CE (9 CE) **Bibliography** Levi Della Vida 1938, 139-48; Rosenthal 1939, 91, no. 4; Roche

1996, 78-80; Nehmé 2003, 11-3; *Quellen*, 128-9; *IG* XII 4,2, no. 558


<sup>16</sup> This is part of the Dodecanese islands in front of the Turkish city of Bodrum, ancient Halicarnassus, 4 km off the coast. The island was frequented by Nabataean traders who traded their own products with the West.

#### Text and translation

1) *[---] šnt 18 lḥrtt* 2) *[---]w 'bd 'wšlhy* 3) *[br rwḥw] br 'zqn rb't'* 4) *[---]t' ll'z' 'lht'* 5) *['l ḥyy] ḥrtt mlk nbṭw* 6) Αυσαλλασου Ρουα[ιου] 7) Ναβαταιου Θεα Αφρο[διτη] 8) ευχην 1) [---] year 18 of Aretas 2) [---] made 'Awšallah 3) [son of *RWḤW* (Rūeḥō?)] son of *'ZQN* the altar 4) [---] for al-'Uzzā the goddess 5) [for the life] of Aretas king of the Nabataeans 6) Ausallos (son of) Roua[ios] 7) the Nabataean, to the Goddess Aphro[dite] 8) vow

### Commentary

Line 1. At the beginning we read *šnt* 'year' followed by the numerical symbols 10+5+1+1+1 = 18. Levi Della Vida reconstructed the expression *byrḥ 'b* 'in the month of 'Ab' evaluating no more than six letters that may precede the term *šnt* in the cut part; in fact, *byrḥ* contains four letters and only the mention of the month *'b*, of two characters, may fill the blank space.21 However it seems to be difficult to reconstruct this verse given the absence of epigraphic data and basing our assumption only on the Nabataean *usus scribendi*. 22

At the end of the line there is the mention of Aretas, *lḥrtt* 'for Aretas'. So, the year 18 could be that of the reign of king Aretas III (87- 62 BCE) or Aretas IV (9 BCE-10 CE). The lack of the common epithet of Aretas IV, *rḥm 'mh* 'he who loves his people', corroborates the assumption that the inscription refers to king Aretas III,<sup>23</sup> but it may also concern Aretas IV, as Nehmé points out (2003, 13), since the epithet, though recurring, is not necessary in the inscriptions of Aretas IV.<sup>24</sup>


24 For instance, the inscription from al-Ǧawf, northern Saudi Arabia, includes the name of Aretas IV without his epithet (lines 3-4): *šlm bšnt 13 lḥrtt* 'peace, in the year

As regards the palaeography, we may ascertain that the author writes *'* in different shapes; so *'* occurs in the archaic form (3 times in line 4) and in the late form (4 times in lines 2, 3 and 4). In addition, sometimes the characters appear to be isolated in writing, as in the traditional archaic inscriptions, but on other occasions they show ligature tending to the cursive script; it is a stylistic transition that begins since the 1st century BCE It is also true that the writing is not so archaic and it is also recorded in the few known inscriptions of Aretas III;<sup>25</sup> therefore it could simply belong to the period of Aretas IV. Moreover, Segre pointed out that the shape of the Greek characters dated back to the period between the end of the 1st century BCE and the beginning of the 1st century CE,<sup>26</sup> but Levi Della Vida rejected his hypothesis<sup>27</sup> taking into account what Littmann<sup>28</sup> spotted in the bilingual from Miletus (no. 49); in fact, it shows a more modern kind of script compared to the proposed dating, i.d. 9 BCE Therefore, here the Nabataean characters could have been written by a Greek author who transcribed them from a copy in which they appeared in a cursive and modern shape. In his footnote Segre added that the Greek letters date back to the second half of the 1st century BCE;<sup>29</sup> finally, D. Bosnakis and K. Hallof date the epigraph to 9 CE.<sup>30</sup>

Line 2. After the damaged part, on the right side, we see a final *w*  that is part of the reconstructed common expression referred to Aretas IV: *mlk nbṭw* 'king of the Nabataeans'. At the end of the line the name *'wšlhy*, in spite of *'wš'lhy*, <sup>31</sup> in *scriptio defectiva* appears. This theophoric (= Gr. Αυσαλλασου, in line 6) is based on the root *'wš* 'to give, to offer' (for the analysis see no. 37).

Line 3. At the beginning we may only reconstruct the father's name according to its presence in the Greek part, in line 6, as Ρουα[ιος]. Presumably it corresponds to the Nabataean name *rwḥw* – Rūḥū,<sup>32</sup> from Arabic *rawḥ* 'rest, relief',33 compared to Greek Ρουαιος34 and al-

13 of *ḥrtt*'; it resembles our inscription. Cf. al-Theeb 1994, 34.


so Ραουαος. <sup>35</sup> The name *rwḥw* is attested five times in the Nabataean inscriptions, but it only refers to two individuals.<sup>36</sup> If the reading is right, Greek helps us to demonstrate the presence of another *rwḥw*  outside the *Nabataea*. 37

The following name is that of the author's grandfather. Levi Della Vida proposed to read *'lqn* that is akin to Arabic names *'Ilāqa* or *'Allāq*. <sup>38</sup> Conversely, Rosenthal interpreted the second letter *l* as *z*  reading *'zqn*. <sup>39</sup> The name *'Azqān*<sup>40</sup> is not found in Nabataean and its origin remains obscure;<sup>41</sup> we may suppose that it is a transcription of the Homeric Greek name Ασκάνιος. <sup>42</sup> At the end of the line there is evidence of the dedicated object in the inscription, which is a *rb't'*  (for a closer examination of the term see comments at no. 48).

Line 4. After the broken part of the stone there is the singular feminine suffix in emphatic state, *-t'*, referring to *ṣlmt'* 'statue' followed by the sentence *lb'l' 'lht'* '[and the statu]e (?) to the goddess Ba'alā'.<sup>43</sup> According to Levi Della Vida, this latter was the goddess *al- 'Uzzā* 'mother goddess', similar to the Greek Aphrodite, who assumed several names among the Semites of the North and the Nabataeans. Probably *'lb'ly* 'the Lady' replaced the other epithet of *al-'Uzzā*, which was 'the Powerful'.44 On the contrary, Rosenthal suggested reading the whole expression as follow: *ll'z' 'lht'*; if the reconstruction is right, and the goddess identified by Rosenthal is really *al-'Uzzā*, this will prove to be the only Nabataean inscription in the Mediterranean that quotes a deith other than Dūšarā.<sup>45</sup>

35 Cantineau 1930-32, 2: 146.

36 Macdonald 1999, 284-5. There are only two inscriptions in which the name is present. In one of these (*CIS* II no. 182) two people, both named *rwḥw*, are quoted once in line 1 and other two times in line 2. One of these two men is also mentioned in another inscription (*CIS* II no. 184).

37 Possibly, he could be a descendant of the two *rwḥw*, cited in *CIS* II no. 182, since they lived at the time of Malichus III, son of Aretas IV, in the year 17 of his reign, i.d. in 50 CE (line 3). In Palmyrene *rwḥ'* and *rwḥbl* (*PAT*, 440 = Stark 1971, 49 and 111) are attested.

38 Levi Della Vida 1938, 144.

39 Rosenthal 1939, 91, no. 4.

40 There is no trace of a *'* to read /ā/ and probably it is a *defective* spelling.

41 In Arabic the root *'azaqa* means 'to hoe, to hoe the ground' (cf. Heb. *'āzaq* with the same meaning), but it is not used to express personal names.

42 Roche 1996, 79. The name appears in Homer, *Il.* 2.862-3; 13.792. Cf. also Pape 1911, 158.

43 Levi Della Vida 1938, 144.

44 Levi Della Vida 1938, 149.

45 Roche 1996, 80.

Line 5. Levi Della Vida fills the last line with *'l ḥyy '*for the life of' followed by *ḥrtt mlk nbṭw* 'Aretas, king of the Nabataeans'.

Line 6. The Greek section mentions the name Αυσαλλασου in the genitive case rather than in the nominative. It is possible to find an isolated personal name in the genitive case, but only in sepulchral inscriptions in which the terms μνῆμα or τάφος are implied; for this reason we may speculate that the noun κτίσμα 'building' is omitted here, even if it would be unusual.<sup>46</sup> In addition, we have Αυσαλλας and Αυσαλλου<sup>47</sup> in the genitive, but not Αυσαλλασου < \*Αυσαλλασος (?). This could imply that the author of the epigraph knew a bit of Greek and marked the genitive case as a calque of Nabataean *'wš'lhy*, whose compound name reflects the genitive case. Furthermore, in Αυσαλλασου the intervocalic σ was not elided probably because the author considered it the ending -ας, so he added the genitive marker -ου to it.

46 Levi Della Vida 1938, 140.

47 Wuthnow 1930, 30.

Antichistica 28 | 11 **141** Corpus of Nabataean Aramaic-Greek Inscriptions, 133-142

# Nabataean Glossary

This glossary contains only complete Nabataean words and words restored with certainty. The list is made up of three columns. The first column includes the lemma and the second column contains the English translation of it. The reference to Nabataean Aramaic lemmas is usually to *DNWSI.* Forms that occur in the inscriptions are listed in the last column. Verbs are in the *pe'al* form unless specified otherwise.

#### Grammatical Abbreviations




#### **Giuseppe Petrantoni List of Abbreviations and Bibliography**




#### Ethnic and geographical names

Antichistica 28 | 11 **148** Corpus of Nabataean Aramaic-Greek Inscriptions, 143-154



#### Ethnic and geographical names


### Onomastics in Nabataean script


Antichistica 28 | 11 **150** Corpus of Nabataean Aramaic-Greek Inscriptions, 143-154




#### Onomastics in Greek script

#### **Giuseppe Petrantoni List of Abbreviations and Bibliography**


Giuseppe Petrantoni

# List of Abbreviations and Bibliography

#### **Abbreviations**

#### **Languages and Scripts**



### **Bibliographical abbreviations**


*IG* XI 2 = Dürrbach, F. (ed.) (1912). *Inscriptiones Graecae*. Vol. XI, *Inscriptiones Deli*. Fasc. 2, *Inscriptiones Deli liberae. Tabulae archontum, tabulae hieropoeorum annorum 314-250*. Berlin.

*IG* XII 4,2 = Bosnakis, D.; Hallof, K. (edd) (2012). *Inscriptiones Graecae*. Vol.XII, *Inscriptiones insularum maris Aegaei praeter Delum*. Fasc. 4, *Inscriptiones Coi, Calymnae, insularum Milesiarum*. Pars 2, *Inscriptiones Coi insulae: Catalogi, dedicationes, tituli honorarii, termini*. Berlin; Boston.


*IGLS* 21,2= Gatier, P.-L. (éd.) (1986). *Inscriptions grecques et latines de la Syrie.*  Vol. XXI, *Inscriptions de la Jordanie*. Tome 2, *Région centrale (Amman, Hesban, Madaba, Main, Dhiban)* (BAH 114). Paris.


# **Bibliography**

Abel, F.-M. (1903). "Inscriptions grecques de Bersabée". *Revue Biblique*,12, 425-30.

Abdelaziz, M.; Rababeh, S. (2008). "The Terminology Used to Describe Tombs in the Nabataean Inscriptions and its Architectural Context". *Levant,* 40- 2, 177-83.

Abu al-Faraj al-'Ušš, M. (1973). "Naša'at al-ḫaṭṭ al-'arabī wa-taṭawwuru-hu". *Les annales archéologiques arabes syriennes*, 23, 55-84 (Arabic).

Abu Taleb, M. (1984). "Nabayati, Nebayot, Nabayat and Nabaṭu: The Linguistic Problem Revisited". *Dirasat*, 11, 3-11.

Altheim, F.; Stiehl, R. (1964-69). *Die Araber in der Alten Welt*. 5 Bde. Berlin.

Amadasi Guzzo, M.G. (1990). *Iscrizioni fenicie e puniche in Italia.* Roma.

Amadasi Guzzo, M.G.; Equini Schneider, E. (1997). *Petra. Capitale dei nabatei.*  Milano.

Arbeitman, Y.C. (1994). "Detecting the God Who Remained in Dan". *Henoch*, 16, 9-14.

Ball, W. (2001). *Rome in the East: The Transformation of an Empire*. London.

Bartlett, J.R. (1989). *Edom and the Edomites*. Sheffield: Sheffield Academic Press. Journal for the Study of The Old Testament: Supplement Series 77.

Beekes, R. (2010). *Etymological Dictionary of Greek.* London; Boston.

Beer, F.E.F. (1840). *Inscriptiones Veteres Litteris et Lingua Hucusque Incognitis ad montem Sinai magno numero servate cet.* Leipzig.

Beyer, K. (1984). *Die aramäischen Texte vom Toten Meer*, Bd. 1. Göttingen.

Beyer, K. (1998). *Die aramäischen Inschriften aus Assur, Hatra und dem übrigen Ostmesopotamien*. Göttingen.

Beyer, K. (2004). *Die aramäischen Texte vom Toten Meer*, Bd. 2. Göttingen.

Blau, O. (1862). "Über die nabatäischen Inschriften". *Zeitschrift der deutschen morgenländischen Gesellschaft*, 16, 331-88.

Boisacq, É. (1916). *Dictionnaire Étymologique de la langue grecque*. Paris.

Bowersock, G.W. (1971). "A Report on Arabia Provincia". *Journal of Roman Studies*, 61, 219-42 + pls. XIV-XV.

Bowersock, G.W. (1975). "The Greek-Nabataean Bilingual Inscription at Ruwwāfa, Saudi Arabia". Bingen, J. et al. (éds), *Le monde grec: Pensée, littérature, histoire, documents. Hommages à Claire Préaux*. Bruxelles, 513-22.

Bowersock, G.W. (1983). *Roman Arabia*. Cambridge (MA): Harvard University Press.

Bowersock, G.W. (1990). *Hellenism in Late Antiquity (Thomas Spencer Jerome Lectures)*. Ann Arbor (Michigan).

Bowersock, G.W. (2015). "Bilingual Inscription". *American Center of Oriental Research*, 123-4.

Brockelman, C. (1908). *Grundriss der vergleichenden Grammatik der semitischen Sprachen*. 2 Bde. Berlin.

Broome, E.C. (1973). "Nabaiati, Nebaioth and the Nabataeans: The Linguistic Problem". *Journal of Semitic Studies*, 18, 1-16.

Bruneau, P. (1970). *Reserches sur les cultes de Délos à l'époque hellénistique et à l'époque impériale*. Paris.

Brünnow, R.E.; Domaszewski, A. (1904-09). *Die Provincia Arabia.* 3 Bde. Strasbourg.

Beyer, K. (1986). *The Aramaic Language. Its Distribution and Subdivisions*. Göttingen.

Bowersock, G.W. (1973). "Syria under Vespasian". *Journal of Roman Studies*, 63, 133-40.

Burns, R. (2009). *The Monuments of Syria*. London.

Burton, R.F. (1879). *The Land of Midian*. 2 vols. London.

Buttler, MM.H.C.; Littmann, E. (1905). "Explorations at Si'". *Revue archéologique*, 1, 404-12.

Camodeca, G. (1983). "Le coppie consolari di CIL IX 1574, i consoli del 152 e i Claudii Modesti". *Zeitschrift für Papyrologie und Epigraphik*, 51, 207-12.

Cantineau, J. (1930-32). *Le Nabatéen.* 2 vols. Paris.

Chabot, J.B. (1901). "Sur une inscription bilingue, nabatéenne et grecque, du Sinaï". *Journal Asiatique*, 2, 442-50.

Chantraine, P. (1968). *Dictionnaire étymologique de la langue grecque. Histoire des mots.* Paris.

Clermont-Ganneau, C. (1898a). "Inscription gréco-nabatéenne de Medaba (Moabitide)". *Recueil d'Archéologie Orientale*, 2, 12-4.

Clermont-Ganneau, C. (1898b). "L'inscription nabatéenne de Kanatha". *Recueil d'Archéologie Orientale*, 2, 75-82.

Clermont-Ganneau, C. (1898c). "Les nouvelles inscriptions nabatéennes de Pétra". *Recueil d'Archéologie Orientale*, 2, 370-81.

Clermont-Ganneau, C. (1901a). "Inscriptions grecques du Hauran". *Recueil d'Archéologie Orientale*, 4, 113-22.

Clermont-Ganneau, C. (1901b). "Nouvelles inscriptions nabatéennes". *Recueil d'archéologie orientale*, 4, 169-84.

Clermont-Ganneau, C. (1903). "Inscription bilingue nabatéo-gracque du Sinaï". *Recueil d'Archéologie Orientale*, 5, 59-66.

Clermont-Ganneau, C. (1906a). "L'expédition américaine dans la Syrie centrale". *Recueil d'Archéologie Orientale*, 7, 213-17.

Clermont-Ganneau, C. (1906b). "Un épitrophe nabatéen à Milet". *Recueil d'Archéologie Orientale*, 7, 305-29.

Clermont-Ganneau, C. (1906c). "Un épitrophe royal nabatéen à Milet". *Journal asiatique*, 159-60

Clermont-Ganneau, C. (1907). "Inscription bilingue grecque et nabatéenne découverte à Milet". *Comptes rendus de l'Académie des inscriptions et belleslettres*, 6, 289-91.

Clermont-Ganneau, C. (1924). "La dédicace gréco-nabatéenne de Milet". *Recueil d'archéologie orientale*, 114.

Colombo, V. (1994). "Interferenza araba nel nabateno". Brugnatelli, V. (a cura di), *Sem Cam Iafet. Atti della 7ª Giornata di Studi Camito-Semitici e Indoeuropei* (Milano, 1 giugno 1993). Milano: Centro Studi Camito-Semitici, 63-76.

Coogan, M.D.; Metzger, B.M. (2004). *The Oxford Guide to People and Places of the Bible*. Oxford.

Cook, S.A. (1898). *A Glossary of the Aramaic Inscriptions.* Cambridge.

Cotton, H.; Price, J.J. (2001). "A Bilingual Tombstone from Zo'ar (Arabia)". *Zeitschrift für Papyrologie und Epigraphic*, 134, 277-83.

De Romanis, F. (1997). "Hypalos: distanze e venti tra Arabia e India nella scienza ellenistica". *Topoi*, 7(2), 671-92.

De Saulcy, F. (1867). "Inscription bilingue". *Bulletin Archéologique du Musée Parent*, 1, 9-14.

De Vries, B. (1998). *Umm el-Jimal. A Frontier Town and its Landscape in Northern Jordan*. Vol. 1, *Fieldwork 1972-1981*. Portsmouth (Rhode Island). Journal of Roman Archaeology Supplementary Series 26.


Dozy, R. (1881). *Supplément aux dictionnaires arabes*. 2 vols. Leiden.


Ferguson, C.A. (1959). "Diglossia". *Word*, 15(2), 325-40.

Fiema, Z.T.; Jones, R.N. (1990). "The Nabataean King-List Revisited: Further Observations on the Second Nabataean Inscription from Tell esh-Shuqafiya, Egypt". *Annual of the Department of Antiquities of Jordan*, 34, 239-47.

Fiema, Z.T. et al. (2015). "*Provincia Arabia*: Nabataea, the Emergence of Arabic as a Written Language, and Graeco-Arabica". Fisher, G. (ed.), *Arabs and Empires before Islam*. Oxford, 373-433.

Fisher, Th. (1906). *Diodori Bibliotheca Historica*, vol. 5. Leipzig.

Fitzmyer, J.A. (1979). "The Phases of the Aramaic Language". Fitzmyer, J.A. (ed.), *A Wandering Aramean: Collected Aramaic Essays.* Grand Rapids-Livonia (Michigan), 57-84.

Fournet, J.-L. (1995). "Les inscriptions grecques d'Abū Kū' et de la route Qufṭ-Quṣayr". *Bulletin de l'Institut Français d'Archéologie Orientale*, 95, 173-233.


Gibson, J.C.L. (1975). *Aramaic Inscriptions, Including Inscriptions in the Dialect of Zenjirli.* Oxford. Textbook of Syrian Semitic Inscriptions II.


Healey, J.F. (2011). "Nabataean Inscriptions: Language and Script". Healey, J.F. (ed.), *Law and Religion between Petra and Edessa: Studies in Aramaic Epigraphy on the Roman Frontier*. Farnham; Burlington (Vermont), 45-54.

Heinriches, W.P. (2000). "Ṭasm". Bearman, P.J. et al. (eds), *The Encyclopaedia of Islam*, vol. 10. Leiden, 359-60.


Ibn Qutayba, A.M. (1850). *Kitāb al-ma'ārif*. Ed. by F. Wüstenfeld.Gottingen.


Leslau, W. (1991). *Comparative Dictionary of Ge'ez (Classical Ethiopic).* Wiesbaden.

Levi Della Vida, G. (1938). "Una bilingue greco-nabatea a Coo". *Clara Rhodos*, 9, 139-48.

Levy, M.A. (1860). "Ueber die nabathäischen Inschriften von Petra, Hauran, vornehmlich der Sinai-Halbinsel und über die Münzlegenden nabathäischer Könige". *Zeitschrift der Deutschen Morgenländische Gesellschaft*, 14, 363-484.

Levy, M.A. (1869). "Epigraphisches auf neuerdings gefundenen Denkmälern". *Zeitschrift der Deutschen Morgenländische Gesellschaft*, 23, 434-42.

Lindner, M. (1997). *Petra und das Königreich der Nabatäer; Lebensraum, Geschichte und Kultur eines arabischen Volkes der Antike*. München.


Littmann, E.; Meredith, D. (1953). "Nabataean Inscriptions from Egypt I". *Bulletin of the School of Oriental and African Studies*, 15, 1-28.


Macdonald, M.C.A. (1993). "Nomads and the Hawran in the Late Hellenistic and Roman Periods: A Reassessment of the Epigraphic Evidence". *Syria*, 70, 303-403.

Macdonald, M.C.A. (1995). "Quelques réflexions sur les Saracènes, l'inscription de Rawwāfa et l'armée romaine". Lozachmeur, H. (éd.), *Présence arabe dans le Croissant fertile avant l'Hégire = Actes de la Table ronde international* (Paris, 13 novembre 1993). Paris, 93-101.

Macdonald, M.C.A. (1999). "Personal Names in the Nabataean Realm. A Review Article". *Journal of Semitic Studies*, 44(2), 251-89.

Macdonald, M.C.A. (2003). "Languages, Scripts and the Uses of Writing Among the Nabataeans". Markoe, G. (ed.), *Petra Rediscovered. Lost City of the Nabataeans.* Cincinnati, 36-56.

Macdonald, M.C.A. (2009). "On Saracens, the Ruwwāfah Inscription and the Roman Army". Macdonald, M.C.A. (ed.), *Literacy and Identity in Pre-Islamic Arabia*. Aldershot, 1-26. Variorum Collected Studies Series 906.

Macdonald, M.C.A. (2010). *The Development of Arabic as a Written Language*. Oxford. Supplement to the Proceedings of the Seminar for Arabian Studies 40.

Macdonald, M.C.A. et al. (2015). "Arabs and Empires before the Sixth Century". Fisher, G. (ed.), *Arabs and Empires Before Islam.* Oxford, 11-89.

MacKenzie, D.N. (1986). *A Concise Pahlavi Dictionary.* London.


Mascitelli, D. (2006b). "Il RB di Ğadīma: considerazioni sull'iscrizione bilingue greco-nabatea di Umm al-Ğimāl (CIS A 192)". *Egitto e Vicino Oriente*, 29, 231-7.

al-Mas'ūdī, 'A. [1283h] (1867). *Murūj al-ḏahab wa-ma'ādin al-jawhar*. al-Qāhira. Mazzarino, S. (1982-87). "Sul nome del vento *hipalus* ('ippalo') in Plinio". *Helikon*, 22(27), VII-XIV.

McCabe, D.F. (1984). *Miletos Inscriptions. Texts and List*. Princeton. The Princeton Project on the Inscriptions of Anatolia.

Meillet, A. [1913] (2003). *Lineamenti di storia della lingua greca*. Torino.


Milik, J.T. (1982). "Origines des Nabatéens". Dentzer-Feydy, J. et al. (eds), *Studies in the History and Archaeology of Jordan*, vol. 1. Amman, 261-5.


Morgenstern, M. (1999). "The History of Aramaic Dialects in the Light of Discoveries from the Judean Desert: The Case of Nabataean". *Eretz-Israel*, 26, 134-42.

Moriggi, M. (2012). "Middle Aramaic: Outlines for a Definition". Bettini, L.; La Spisa, P. (éds), *Au-delà de l'arabe standard. Moyen-arabe et arabe mixte dans les sources médiévales, modernes et contemporaines*. Firenze, 279-89. Quaderni di Semitistica 28.

Muraoka, T. (2005). *Classical Syriac. A Basic Grammar with a Chrestomathy.* 2nd ed. Wiesbaden. Porta Linguarum Orientalium 19.


Negev, A. (1991). *Personal Names in the Nabataean Realm*. Jerusalem. Qedem, Monographs of the Institute of Archaeology 32.


Nehmé, L. (2010). "A Glimpse of the Development of the Nabataean Script into Arabic Based on Old and New Epigraphic Material". Macdonald, M. (ed.), *The Development of Arabic as a Written Language*. Oxford, 47-88. Supplement to the Proceedings of the Seminar for Arabian Studies 40.

Nehmé, L. (2012). "Le dieu Obodas chez les Nabatéens: hypothèses anciennes et découvertes récentes". Sachet, I.; Robin, C. (éds), *Dieux et déesses d'Arabie. Images et représentations. Actes de la table ronde tenue au Collège de France* (Paris, 1-2 octobre 2007). Paris, 181-224. Orient & Méditerranée 7.

Nehmé, L. (2015). "*Strategoi* in the Nabataean Kingdom: A Reflection of Central Places?" *Arabian Epigraphic Notes*, 1, 103-22.

Noja, S. (2006). "L'origine della scrittura araba da quella dei nabatei: un'ipotesi da smontare". Borbone, P.G. et al. (a cura di), *Loquentes Linguis. Studi linguistici e orientali in onore di Fabrizio A. Pennacchietti*. Wiesbaden, 535-54.

Nöldeke, T. (1885). "Noten zu den nabatäischen Inschriften". Euting, J. (Hrsg.), *Nabatäische Inschriften aus Arabien*. Berlin, 73-80.

Nöldeke, T. (1909). "Der Gott ביתא מרא und die Ka'ba". *Zeitschrift für Assyriologie*, 23, 184-87.

O'Connor, M. (1986). "The Arabic Loanwords in Nabataean Aramaic". *Journal of Near Eastern Studies*, 45, 213-29.

Pape, W. (1911). *Wörterbuch der griechischen Eigennamen. Dritte Auflage neu bearbeitet von Dr. Gustav Eduard Benseler*. Brunswick.

Parr, P.J. (2003). "The Origins and Emergence of the Nabataeans". Markoe, G. (ed.), *Petra Rediscovered*. London, 27-35.

Payne Smith, J. (1903). *A Compendious Syriac Dictionary: Founded Upon the Thesaurus Syriacus of Robert Payne*. Oxford.

Perlmann, M. (1987). *The History of al-Ṭabarī*, vol. IV. New York; Albany.

Peterson, S.B. (2006). *The Cult of Dushara and the Roman Annexation of Nabataea* [PhD dissertation]. McMaster University.

Petrantoni, G. (2016). "An Epigraphic Note on the Nabataean Text of a Bilingual (Greek-Nabataean) Inscription from Ġūr al-Ṣāfī". *Anabasi*s, *Studia Classica et Orientalia*, 7, 131-36.

Petrantoni, G. (2020). *Onomastica aramaico-greca. Un tentativo di ricostruzione della fonologia del nabateo*. Alessandria. Fonti e studi di Storia Antica 23.


*und nichtsemitischen Menschennamen, soweit sie in griechischen Urkunden (Papyri, Ostraka, Inschriften, Mumienschildren usw.).* Amsterdam.


Retsö, J. (1999). "Nabataean Origins - Once Again". *Proceeedings of the Seminar for Arabian Studies*, 29, 115-18.


Sartre-Fauriat, A. (2001). *Des tombeaux et des morts, monuments funéraires, société et culture en Syrie du Sud du Ier s. av. J.-C. au VIIème s. ap. J.-C.* 2 vols. Beirut.


Seyrig, H. (1957). "Sur trois inscriptions du Hedjaz". *Syria*, 34, 259-61.

Shahid, I. (1995). *Byzantium and the Arabs in the Sixth Century.* 2 vols. Washington.

al-Shorman, Z.M. (2012). "The Assimilation of Dushara-Ḍwšara in Greco-Roman Period". al- Helabi, A. et al. (eds), *Arabia, Greece and Byzantium*, vol. 2. Riyadh, 43-58.

Sokoloff, M. (2009). *A Syriac Lexicon. A Translation from the Latin, Correction, Expansion, and Update of C. Brockelmann's Lexicon Syriacum*. Winona Lake; Piscataway.

Sourdel, D. (1952). *Les Cultes du Hauran à l'époque Romaine*. Paris.


Stark, J.K. (1971). *Personal Names in Palmyrene Inscriptions*. Oxford.

Stockton, E.D. (1971). "Petra Revisited: A Review of A Semitic Cult Complex". *Australian Journal of Biblical Archaeology*, 1(4), 51-73.

Strawn, B.A. (2005). *What is Stronger Than a Lion? Leonine Image and Metaphor in the Hebrew Bible and the Ancient Near East.* Fribourg. Orbis Biblicus et Orientalis 212.

Ṭabarī, A.J.M. (1960-69). *Kitāb aḫbār al-rusul wa al-mulūk*. Ed. by Muḥammad Abū l-Faḍl Ibrāhīm, 10 voll. + Indici. Cairo.

Tadmor, H.; Yamada, S. (2011). *The Royal Inscriptions of Tiglath-Pileser III (744- 727 BC) and Shalmaneser V (726-722 BC), Kings of Assyria*. Winona Lake (Indiana). The Royal Inscriptions of the Neo-Assyrian Period 1.

Taylor, D.G.K. (2002). "Bilingualism and Diglossia in Late Antique Syria and Mesopotamia". Adams, J.N. et al. (eds), *Bilingualism in Ancient Society. Language Contact and the Written Text.* Oxford, 298-331.


Tholbecq, L. (2007). "Hérodiens, Nabatéens et Lagides dans le Ḥawrān au Ier s. av. J.-C.: réflexions au tour du sanctuaire de Ba'alšamīn de Sī' (Syrie du Sud)". *Topoi*, 15(1), 285-310.

Vattioni, F. (1981). "Le iscrizioni di Ḥatra". *Annali dell'Università degli Studi di Napoli 'L'Orientale'*, suppl. 28, 41(3).

Vattioni, F. (1985). "Ai primordi della storia degli Arabi. Appunti sui Nabatei". Sarnelli Cerqua, C. (ed.), *Studi arabo-islamici in onore di Roberto Rubinacci nel suo settantesimo compleanno*. Napoli, 719-72.

Vattioni, F. (1987-88). "Fenici, Siri e Arabi emigrati in area greca". *Annali dell'Istituto Orientale di Napoli*, 9-10, 91-124.

Vierros, M. (2014). "Bilingualism in Hellenistic Egypt". Giannakis, K. (ed.), *Encyclopedia of Ancient Greek Language and Linguistics*, vol. 1. Leiden; Boston, 234-8.

Vogüé, M. de (1864). "Inscriptions araméennes et nabatéennes". *Revue Archéologique*, 9, 284-8.

Wasserstein, A. (1993). "A Note on the Phonetic and Graphic Representation of Greek Vowels and of the Spiritus Asper in the Aramaic Transcription of Greek Loanwords". *Scripta Classica Israelica*, 12, 200-8.

Wehr, H. (1976). *Arabic–English Dictionary: The Hans Wehr Dictionary of Modern Written Arabic, Edited by J M. Cowan*. Ithaca (NY).

Wellhausen, J. (1897). *Reste Arabischen Heidentums.* Berlin.


# Plates

**Plate I**

1 *IGLS* 21,4 54, pl. XXIX

2 Bowersock 2015, 123, fig. 7. XX

Antichistica 28 | 11 **172** Corpus of Nabataean Aramaic-Greek Inscriptions, 171-182

**Plate II**

1 *IGLS* 21,4 28, Pl. XX

2 The photograph of the Greek part (a) *IGLS* 21,4, pl. XLVIII, no. 148. The facsimile of the Nabataean section (b) Savignac-Horsfield 1933, 418, fig. 10

> Antichistica 28 | 11 **173** Corpus of Nabataean Aramaic-Greek Inscriptions, 171-182

**Plate III**

1 *IPT Ib*, 50, pl. XXV

2 Milik 1958, fig. 2 and tab. XIXb

#### **Plate IV**

1 Amadasi Guzzo, Equini Schneider 1997, 55, no. 39

2 *IGLS* 21,5 98: 87

3 De Vries 2009, 181, no. 3

Antichistica 28 | 11 **175** Corpus of Nabataean Aramaic-Greek Inscriptions, 171-182

#### **Plate V**

1 *IGLS* 21,5 no. 499

2 Milik 1958, tab. XIXa

3 Milik 1958, tab. XXa

Antichistica 28 | 11 **176** Corpus of Nabataean Aramaic-Greek Inscriptions, 171-182

**Plate VI**

1 *PPAES* IVA, 105

2 *PPAES* IV, 103

**Plate VII**

1 Fiema et al. 2015, 306, pl. 16

2 Hoyland 2008, 56, fig. 6

Antichistica 28 | 11 **178** Corpus of Nabataean Aramaic-Greek Inscriptions, 171-182

#### **Plate VIII**

1 Negev 1981, pl. X

2 *CIS* II, 671, tab. LXV

3 *CIS* II, 1032, tab. LXXIX

#### **Plate IX**

1 *CIS* II, 1719a

2 Littman, Meredith 1953, 9, no. 13

3 Littmann, Meredith 1954, 217 and pl. II b

Antichistica 28 | 11 **180** Corpus of Nabataean Aramaic-Greek Inscriptions, 171-182

1 Littmann, Meredith 1954, 222, no. 71

2 Nehmé 2003, 5, fig. 2

#### **Plate XI**

1 Kawerau, Rehm 1914, I.3, 387, no. 165

2 Nehmé 2003, 92

Antichistica 28 | 11 **182** Corpus of Nabataean Aramaic-Greek Inscriptions, 171-182

The impact of the Hellenization in the Ancient Near East resulted in a notable presence of Greek *koin* language and culture and in the interaction between Greek and Nabataean that conducted inhabitants to engrave inscriptions in public spaces using one of the two languages or both. In this questionably 'diglossic' situation, a significant number of Nabataean-Greek inscriptions emerged, showing that the *koin* was employed by the Nabataeans as a sign of Hellenistic cultural affinity. This book offers a linguistic and philological analysis of fifty-one Nabataean-Greek epigraphic evidences existing in northern Arabia, Near East and Aegean Sea, dating from the first century BCE to the third-fourth century CE. This collection is an analysis of the linguistic contact between Nabataean and Greek in the light of the modalities of social, religious and linguistic exchanges. In addition, the investigation of onomastics (mainly the Nabataean names transcribed in Greek script) might allow us to know more about the Nabataean phonological system.

**Giuseppe Petrantoni** awarded his PhD in Philology and History of the Ancient World, curriculum in Philologies of the Near and Middle Orient at the Sapienza University of Rome. He is currently Adjunct Professor in Arabic Language at the University of Enna "Kore". He taught Semitic Philology and Greek Epigraphy at the University of Florence as Adjunct Professor. Since 2018 he has been a Subject Expert in Semitic Philology at the Faculty of Humanities of the University of Catania. He has published several articles in high-ranked national and international journals and he is the author of two monographs: *Onomastica aramaico-greca. Un tentativo di ricostruzione della fonologia del nabateo* (2020) and *I Turcarabi. Due popoli sotto l'Impero ottomano* (2009).

PETRANTONI

**INSCRIPTIONS**

**ARAMAIC-GREEK** 

**CORPUS OF NABATAEAN**