Page 1

28

P

E

T

R

A

N

T

O

N

I

C

O

R

PU

S

O

F

NAB

A

T

AE

AN

A

R

A

M

A

IC

-G

R

EEK

INSC

R

IP

T

IO

NS

Antichistica 28

Studi orientali 11

—

Corpus of Nabataean

Aramaic-Greek

Inscriptions

Giuseppe Petrantoni

Edizioni

Ca’Foscari

e-ISSN 2610-9336

ISSN 2610-881X

Corpus of Nabataean Aramaic-Greek Inscriptions

Antichistica

Studi orientali

Collana diretta da

Lucio Milano

28 | 11

Antichistica

Studi orientali

Direttore scientifico

Lucio Milano (Universit� Ca’ Foscari Venezia, Italia)

Comitato scientifico

Claudia Antonetti (Universit� Ca’ Foscari Venezia, Italia)

Filippo Maria Carinci (Universit� Ca’ Foscari Venezia, Italia)

Ettore Cingano (Universit� 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 (Universit� Ca’ Foscari Venezia, Italia)

Fausto Zevi (Sapienza Universit� di Roma, Italia)

Direzione e redazione

Dipartimento di Studi Umanistici

Universit� Ca’ Foscari Venezia

Palazzo Malcanton Marcor�

Dorsoduro 3484/D

30123 Venezia

Antichistica | Studi orientali

e-ISSN 2610-9336

ISSN 2610-881X

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

Corpus of Nabataean

Aramaic-Greek

Inscriptions

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

cb

Quest’opera � 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 pu� 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 responsabilit� del

Comitato scientifico della collana. La valutazione � 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 Universit� 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

URL https://edizionicafoscari.unive.it/it/edizioni/libri/978-88-6969-508-7/

DOI http://doi.org/10.30687/978-88-6969-507-0

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.

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, tran-

scribed 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 Epig-

raphy; in addition, the corpus allows us to study the modalities of cultural exchange

(especially in social and religious contexts) between Nabataeans and Greeks. Moreo-

ver, 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.

Corpus of Nabataean Aramaic-Greek Inscriptions

Giuseppe Petrantoni

Acknowledgements

The initial version of this work was the core of my PhD dissertation Corpus delle iscri-

zioni bilingui aramaico nabateo-greche. Approfondimenti onomastici e problemi di ri-

costruzione 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 al-

ways shown interest in studying the Greek-Semitic bilingualism; this indirectly steered

me toward the argument proposed here.

Corpus of Nabataean Aramaic-Greek Inscriptions

Giuseppe Petrantoni

Corpus of Nabataean Aramaic-Greek Inscriptions

Giuseppe Petrantoni

Table of Contents

Introduction

13

Jordan

31

Syria

73

Egypt

103

Saudi Arabia

121

Lebanon

129

Aegean Sea

133

Nabataean Glossary

143

List of Abbreviations and Bibliography

155

Plates

171

To Sonia, whose love, support and advice

have never failed, this book is dedicated

Antichistica 29

10.30687/978-88-6969-507-0/000

Corpus of Nabataean Aramaic-Greek Inscriptions

Giuseppe Petrantoni

13

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 frequent-

ly interacted with Greeks, Romans and Jews. Since the Nabataeans

were a nomadic tribal society, there exists no real Nabataean liter-

ature. They presumably had an oral tradition that can no longer be

reconstructed today.1

The Nabataeans constituted an ethnic group in which most liter-

ates (a small minority, who nonetheless occupied a dominant social,

economic, and political position) used Greek as the language of com-

munication 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 inscrip-

tions written in Greek and Nabataean collected in the present vol-

ume. Since the times of the Achaemenid Empire (ca. 550-330 BCE),

during which the royal chancellery continued with the deeply-en-

trenched employment of Aramaic in local and provincial administra-

tion, there was an expansion of multilingualism and a spread of va-

rieties of Aramaic, Greek and other languages, such as Hebrew and

1 Wenning 2007, 25.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

14

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

Arabic idioms, which became evident during the subsequent Greco-

Roman period and until the 6th-7th century CE.2

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.3 The

real degree of the influence of the Greek language and culture is still

debated, and varied from region to region.4 Furthermore, after the

Greek and Roman conquest of the Near East, including Arabia, Au-

gustus 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;5 the immediate conse-

quence was the emergence of several new city-states (such as Petra

and Palmyra) that adopted Hellenistic customs.6

It would not be appropriate here to enter into a detailed history of

the Nabataeans.7 We need only note that, from the social and cultur-

al point of view, the Nabataeans were a nomadic Bedouin tribe that

roamed the Arabian desert and moved with their herds to wherever

2 The rise of Islam and the consecutive wars of conquest of the Muslim armies trans-

formed 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).

3 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 cul-

tural 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).

4 Cf. Bowersock 1990.

5 For an economic overview see Hopkins 1980. In general, see Goldsworthy 2016.

6 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 Arama-

ic or reproducing the text in Greek in order to exhibit their degree of literacy and eco-

nomic power. So, for instance, in Palestine by the 4th century BCE the shift from He-

brew (which became the holy language) to Aramaic and the spread of Greek had trig-

gered a complex linguistic development in which Aramaic presumably came to domi-

nate (cf. Gzella 2015, 226 fn. 709, who quotes Poirier 2007).

7 For a historical overview, we may refer to several modern works, such as Starcky

1955; Bowersock 1983; Lindner 1997; Wenning 1987; 2007.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

15

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

they could find pasture and water. Although the precise origin of the

Nabataeans remains uncertain8 (they were probably an Arab people

who inhabited northern Arabia and the southern Levant), we may as-

sert 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 Assyrian9 and

Biblical10 sources. The most common theory, according to which the

Nabataeans were an Arab group, is today supported by three piec-

es of historical and linguistic evidence. Firstly, when Greek writ-

ers mentioned these people they usually refer to them as “Arabs”.11

Secondly, there is the presence of Arabic personal names in the Na-

8 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 Na-

bataeans came from Mesopotamia. For an illustrative summary of the debates on Na-

bataeans’ origins, with related bibliography, cf. Parr 2003, 27-35 and Quellen, 15-19.

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 Tad-

mor, 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 pas-

sage 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

Rets� (2003, 364-91 and 623-6), the term ‘Arab’ mainly refers to a social status rath-

er than an ethnic one; he also interprets the original reading as nom�des rather than

Nabata�oi on the basis of two manuscripts from chapter 19 of Diodorus: Parisinus grae-

cus 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

(Rets� 2003, 283-8 and 1999, 115-16). Cf. also Fisher 1906, 146, no. 1. In another pas-

sage, 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 ex-

pedition against Petra (Diod. Sic. 19.44.3).

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

16

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

bataean inscriptions. Finally, we might consider the frequent usage

of Arabic elements, such as particles, verbs, words and whole sen-

tences, in Nabataean.12

2

Nabataean Aramaic and Greek

2.1 The Variety of Nabataean Aramaic

Nabataean is an epigraphic language, one of several varieties of Ar-

amaic, belonging to Middle Aramaic (300 BCE-first centuries CE),13

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, var-

ious archaic morphosyntactic features distinguish Nabataean from

the other Middle Aramaic varieties, such as Palmyrene and Hatrae-

an.15 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 herit-

age in the local varieties of Aramaic of the Greco-Roman period on

the other”.16

Nabataean was employed as lingua franca to ease the communica-

tion among the peoples of the Arabian Peninsula and as an interna-

tional language to facilitate trade and business in the Near East. Ac-

cording to M. Morgenstern (1999, 135) Nabataean Aramaic is formed

of three main elements: 1) a sub-stratum constituted by an Arama-

ic 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 Fitz-

myer 1979, 61-2 and Beyer 1986, 43-53. For a summary of the main theories with relat-

ed 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 ’- in-

stead 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 assimila-

tion of n, like in ’tt’ < *’ntt’, mṣb < root nṣb; 9) the usage of final -w in masculine per-

sonal names of probable Arabic origin; 10) the employment of yt like nota accusativi.

16 Gzella 2015, 213.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

17

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

The Nabataean inscriptions were written in a local Aramaic varie-

ty using a local script. Nevertheless, various Arabic loanwords17 and

grammatical borrowings have been observed.18 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 em-

ployed Aramaic as a lingua franca to write their inscriptions or other

documents;19 in fact, according to G. Garbini and O. Durand, the Na-

bataeans were already sociologically an Arab people, but were still

linguistically Aramaic.20

Nabataean uses a typical script deriving from the Persian Chancel-

lery, 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.21 The Nabataean

script could be classified as a monumental script used for public, fu-

nerary or religious inscriptions on stones,22 and as a cursive script

adopted for legal, diplomatic and commercial documents.23 The script

used in graffiti does not belong to a specific classification. It is prob-

ably that carvers of graffiti always attempted to make their inscrip-

tions more formal than the common calligraphic script.24

From a historical point of view the development of the Nabatae-

an script can be classified into three specific stages. The first is the

ancient period, dating from the end of the 2nd century to the begin-

ning 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, dat-

ing back to the 1st c. BCE-1st c. CE, in which the writing shows elon-

gated 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

17 For instance: gt ‘corpse’ (< Ar. ǧuṯṯah); wld ‘offspring’ (< Ar. walad); l‘n ‘to curse’

(< Ar. la‘ana); nšyb ‘father-in-law’ (< Ar. nasīb). Cf. N�ldeke 1885; O’Connor 1986; Hea-

ley 1995, 78-9.

18 See al-Hamad 2014.

19 Healey 2011, 46.

20 Garbini, Durand 1994, 51.

21 See the work of Beer 1840.

22 As, for instance, the tomb inscriptions from Mada’in Ṣaleḥ show (Healey 1993).

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).

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

18

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

of the Nabataean alphabet; it is sufficient to underline the fact that

the majority of scholars today agree that Arabic writing originated

from the Nabataean26 rather than from the Syriac script.27 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.28

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 Pen-

insula), the northwestern part of the Arabian Peninsula (Hejaz), the

Aegean islands and various sites in southern Italy.29

The inscriptions, including the bilingual examples, are of two

types: dedicatory and funerary. The former are written on an ob-

ject, 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 lat-

ter are engraved on tombs, blocks of stone (in this case the text is

longer) and directly on rocks (the extension of these letters is usu-

ally less significant).30 The main structures are the following: if they

deal with long texts written on tomb fa�ades, they highlight the own-

er’s name (of the tomb) + the members of the family + formal data

about the tomb and the family.31 By contrast, in the short texts the

sequence reports npš’ (‘tomb’) + the name of the deceased. The graf-

fiti, which are carved on rocks,32 follow approximately the same pat-

tern, that is: dkyr (‘let be remembered’) + the believer’s name fre-

quently followed by šlm (‘peace’) or bṭb (‘in good’).

26 For further details see Gruendler 1993 and the more recent work of Nehm� 2010.

27 Cf. Starcky 1966, 933 and Noja 2006.

28 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.

29 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 fa�ades in Greek

style; the inscriptions are longer than those of Petra, reporting the judicial matters re-

lated 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 en-

graved (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).

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

31 A lot of these inscriptions hail from Mada’in Ṣaleḥ.

32 Especially Mount Sinai.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

19

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

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 diglossia33 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 Ara-

maic 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 administra-

tion. 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 morphosyn-

tactic interference.34 Moreover, in the Hellenistic and Early Roman

Near East the interaction between Greek and Nabataean caused in-

habitants 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 ques-

tionable diglossic situation, Greek was used by the upper classes as

the high-register variant and Aramaic as the vernacular of the un-

educated masses; in addition, the new written forms of Aramaic act-

ed as prestige languages and as a vehicle of indigenous cultural af-

finity, as Gzella argues.35

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 antiqui-

ty and the early Byzantine era, or Late Antiquity. Koine Greek is also

known as Alexandrian dialect, common Attic, Hellenistic36 or biblical

Greek37 and it was used between about 330 BCE-330 CE (subsequent-

33 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).

34 Gzella 2015, 223.

35 Gzella 2015, 215-16 and in detail Gzella 2005 and 2006.

36 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 de-

velopment of Greek after the ancient period (about 800-330 BCE).

37 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 the-

ological writing by the church fathers.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

20

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

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 Na-

bataean 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 Meil-

let, foreigners (here we take into account the Nabataean speakers)

spoke a type of Greek that: “a �t� celui des Grecs avec lesquels ils

ont �t� en rapports, et ces Grecs n’ont �t� que pour une faible part

des Ath�niens”.38

From a historical point of view, Greek was not widely used in the

Nabataean realm until about 106 CE as demonstrated in the docu-

ments from the Babatha archive. In fact, none of the Greek texts pre-

date 106 CE.39 Therefore, following the annexation of Nabataea to

the Roman Empire, Greek began to spread considerably across Pe-

tra and the other Nabataean regions replacing Aramaic as the offi-

cial language of bureaucracy after the 4th century CE, as previous-

ly mentioned.

In the Nabataean-speaking territories, as well as in Syria and

Mesopotamia, a complex linguistic landscape, focused on bilingual-

ism, 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.40

Although the strong influence of Arabic on Nabataean is clearly ev-

ident, as witness the presence of Arabic personal names in the Na-

bataean 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.41 In Palmyra, as well

as in Petra, Aramaic was employed in religious and domestic con-

texts, while Greek was spoken in public activities and trading rela-

tions with foreign territories.

On the basis of evidence arising from the Nabataea, it would ap-

pear that the social and linguistic situation was different from that

of the other areas in the Near and Middle East where the introduc-

38 Meillet, A. (1913). Aper�u d’une histoire de la langue grecque. Paris, 229, reprint-

ed in Italian, see Meillet [1913] 2003, 375.

39 Healey 2011, 48.

40 Taylor 2002, 298.

41 Taylor 2002, 318.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

21

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

tion and usage of Greek could be examined from synchronic and dia-

chronic perspectives; we may indeed imagine the impact and the in-

creased density of Greek usage in Palestine42 and Egypt.43

In the 1st century BCE, the Aramaic varieties and Hebrew (in Pal-

estine) were active languages in the Near East being the L1 for the

indigenous inhabitants, whereas Greek became the L1 for the so-

cial and political elite and the L2 for the indigenous community who

employed it in social, administrative and economic environments.44

Greek was the lingua franca from Greco-Roman Egypt to the east-

ern 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 lan-

guage.

As regards the Nabataean Kingdom, a diachronic perspective de-

scribes some sociolinguistic environments of Greek usage bolstered

by historical and material sources.

The first linguistic contact between Greek and Nabataean Ara-

maic dates back to 312 BCE thanks to the account by Diodorus Sic-

ulus45 who recalls two Macedonian military campaigns, led by An-

tigonus I Monophthalmus, against the Arabs/Nabataeans in Petra.46

The Macedonians reached Petra and took prisoners. Afterwards the

Nabataeans defeated the Macedonians47 and wrote to Antigonus ‘in

Syrian characters’.48 This passage refers to Aramaic as a lingua fran-

ca of the powers of the Near East. Even though, according to clas-

sic 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

42 Koine Greek became first lingua franca, then prestige language and finally a wide-

spread vernacular among the inhabitants in the 1st century CE. The Greek New Testa-

ment documents constitute one piece of synchronic evidence for this, as do the signifi-

cant number of documentary Greek papyri found in a variety of sites, including Masa-

da and different sites around the Dead Sea. See Porter 2016, 212-27.

43 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 work-

ing class with Demotic and then Coptic being its first language. See Vierros 2014.

44 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.

45 Diod. Sic. 19.94.1-98.1.

46 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).

47 Diod. Sic. 19.95.3-5.

48 Diod. Sic. 19.96.1: πρὸς δ᾽ Ἀντίγονον ἐπιστολὴν γράψαντες Συρίοις γράμμασι.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

22

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

4th century BCE in Petra the Nabataeans were not continually ex-

posed 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 Se-

leucids 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 pres-

tige language in the eastern Mediterranean as well. The Nabataeans,

who were one of the numerous nomadic tribes of Bedouins wander-

ing the Arabian desert,49 still remained on the fringes of the Hellen-

istic 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.50

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.51 This represents further evidence that the language used

by the Nabataeans was still Nabataean Aramaic. In the 2nd centu-

ry BCE the Nabataeans were not in contact with Greek in a diglos-

sic 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’.52

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 architec-

ture 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 bilin-

gual Nabataean Aramaic-Greek inscriptions. Presumably for most of

the 1st century BCE, in accordance with the wishes of Aretas III, the

49 They were described as ‘Arabic nomads’ as reported in Joseph. AJ 12.333-335 and

1Mac 5,24-25 and 2Mac 12.

50 The Nabataeans controlled many trade routes towards the South along the Red

Sea shore in the Hejaz desert, and towards the North to Damascus.

51 Quellen, 393-5.

52 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 suc-

cessful siege led by the Armenian king Tigranes II.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

23

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

Hellenized Nabataeans employed Greek as vehicular and official lan-

guage 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 mono-

lingual 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 ex-

ist a significant number of Hellenistic inscriptions (3rd-1st c. BCE).53

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 be-

tween Latin and Greek but also being diglossic in their knowledge of

Greek, using a High Attic form and a Low vernacular”.54 This period

marked a linguistic and social transition during which the Nabatae-

ans definitely entered into the Greek linguistic sphere.

During the reign of Herod the Great (40 BCE-4 CE), the Greek lan-

guage prevailed over the other Semitic languages. Herod was educat-

ed 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.55

During the late antique period, the importance of Greek was dif-

ferent in Mesopotamia and Syria, as compared to the Nabataean

Realm.56 In Dura Europos, where no native variety of Aramaic is at-

tested, 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 bi-

lingualism becoming more visible after the first records of Classi-

cal Syriac.57 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 two-

thousand 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-

53 The corpora of IGLS represent a systematic collection of Greek and Latin inscrip-

tions from the Ancient Near East.

54 Porter 2016, 210.

55 Porter 2016, 210-1.

56 See Taylor 2002.

57 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 in-

terferences or do not encompass Greek expressions.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

24

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

cial language, but instead: “it was integrated into a more complex

multilingual environment”.58

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 pre-

sent and restricted to a couple of lexemes concerning economics and

Hellenistic architecture with no syntactic interference in Aramaic.59

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 in-

scriptions date back to 9 BCE They were found in Miletus and in De-

los60 and were commissioned by Sylleus during his journey to Rome.

Another inscription, found in Sidon dates back to 4 BCE61 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 Arama-

ic; they were neither bilingual, nor, probably, diglossic. In fact, the in-

scriptions from Delos and Miletus reveal that Sylleus only wanted a

simple Greek translation, of the Nabataean text, to leave in a Greek-

speaking 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 beau-

tiful and regular letters. It can therefore be surmised that Sylleus ar-

rived at Miletus carrying a Nabataean handwritten copy of the text,

drafted by himself on a papyrus or on a parchment. Later, he deliv-

ered 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 re-

gions, date from the 1st century until the 2nd-3rd century CE. Where-

as in Palmyra the population appears to have been bilingual, in Pet-

ra 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 Si-

nai, the graffiti (dating back to the 2nd-3rd century CE) consist in

short and fragmentary texts. They are mostly dedicatory and funer-

58 Gzella 2015, 249. See also Gzella 2005, 445-58. For a corpus of Palmyrene Ara-

maic inscriptions see PAT.

59 Gzella 2015, 275.

60 Respectively nos. 49 and 50.

61 No. 48.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

25

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

ary inscriptions and provide no noteworthy evidence of the alleged

bilingualism of the Nabataeans. The majority of the graffiti are on-

ly 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,62 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 lan-

guages for representational purposes among speakers of Old Arabic

and Ancient North Arabian vernaculars”.63 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 in-

dependently, probably following the same content, but not translating

from one language to another. Furthermore, the Greek linguistic in-

fluence 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 pas-

sage 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 environ-

ment. 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 Nabatae-

an text reports the initial formula šlm ‘peace’ and the closing formu-

la 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 differ-

ent in content (nos. 29 and 34).

Two texts are exclusively in Greek, including a series of Nabatae-

an letters (no. 12) and a Nabataean personal name (no. 18). Converse-

ly, only one inscription is entirely in Nabataean (no. 20), with the ex-

ception 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…’, where-

as the Greek one is ‘the people (or council) of… honoured’, both re-

flecting the West Semitic and Greek traditions. Furthermore, nos. 10

and 26 report the Hellenistic expression ‘out of affection’ and ‘of his

62 No. 47.

63 Gzella 2015, 242.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

26

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

piety’, whereas some inscriptions contain a more extensive genealo-

gy 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 complete-

ly 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 sociolinguis-

tic environment in which Koine Greek was used in a diglossic sit-

uation 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 epigraph-

ic evidence would suggest, that it was probably also the vernacu-

lar employed by some social strata of the population. So, the Na-

bataean-Greek bilingual inscriptions are apparently not bilingual,

but rather multilingual texts carved within a diglossic linguistic sit-

uation among the Nabataeans. Nabataean was the L1 of the indige-

nous population and possibly even the primary idiom for some in the

lower social stratum, whereas Greek was used as the L2 in lower so-

cial contexts and as an administrative language.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

27

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

3

Nabataean Aramaic-Greek Inscriptions

3.1 Research Background

The presence of corpora of Semitic inscriptions, and in particular of

Aramaic inscriptions,64 is indicative of the exponential expansion in

epigraphic studies in the field of the Near Eastern philology. The in-

terest in Greek epigraphy in collecting Greek inscriptions found in

the Ancient Near East65 conveys the desire to further study and ana-

lyse 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 Canaan-

ites, the Phoenicians and the Syro-Palestinians) and Greek speakers

increased in North Africa, Rhodes, Kos, in the rest of the Central Ae-

gean 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 in-

scriptions; epigraphs written in Greek and in Semitic languages re-

cord 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 in-

scriptions, dating to the period between the 2nd c. BCE and the 4th c.

CE, have so far been discovered, a complete corpus, comprising all Na-

bataean inscriptions, has yet to be compiled.66 A wide selection of texts

from the entire Nabataean region is recorded in Quellen (see bibliogra-

phy), and a number are also included in Yardeni 2000,67 while a great

64 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 Ara-

maic in the Hellenistic and Early Roman period, see e.g. Magen et al. 2004 for inscrip-

tions from Mount Gerizim; the series Discoveries in the Judaean Desert (DJD, 1955-) and

Beyer 1984 and 2004, including YTDJD and Yadin et al. 2002, for Aramaic of the Qum-

ran scrolls; PAT for Palmyrene; Beyer 1998 for Hatraean.

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 Hu-

manities Institute (PHI) – Project Centers at Cornell University & Ohio State University.

66 Gzella 2015, 239.

67 See Beyer 2004, 23 for additions.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

28

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

many inscriptions remain unpublished or partially described. The cor-

pus of M.E. Stone (in RIGP) includes the graffiti found at Mount Sinai,

one of the regions in which the greatest number of Nabataean inscrip-

tions have been discovered. A corpus including the bilingual Nabatae-

an-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 un-

derstanding of the morphological, syntactic and lexical aspects of the

two languages in question and shed light on the cultural, social, politi-

cal 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 disser-

tation 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.68

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 inscrip-

tions 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.

68 For a more in-depth the reconstruction of the Nabataean Aramaic phonological

system, see Petrantoni 2020.

Giuseppe Petrantoni

Introduction

Antichistica 28 | 11

29

Corpus of Nabataean Aramaic-Greek Inscriptions, 13-30

Editorial conventions

[x]

Lacuna that is reconstructed.

[…]

Lacuna that is not reconstructed, the missing letters are replaced by dots.

[---]

Lacuna that is not reconstructed and of which we have no suggestion

of the exact number of the missing letters.

(x)

Lacuna that is considered as uncertain.

As regards the forms of the Nabataean personal names given in

translations, they reflect my own close examinations and studies of

these during my research69 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 con-

cerns 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 phe-

nomenon, 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 He-

brew 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 tran-

scription using the East Syriac vocalic system. The transcription of

Syriac Aramaic vowels and fricative consonants follows that of Mu-

raoka (2005, 4-7).

The Romanization of Arabic is based on The Hans Wehr translit-

eration 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.

70 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

Antichistica 28 | 11

DOI 10.30687/978-88-6969-507-0/001

Corpus of Nabataean Aramaic-Greek Inscriptions

Giuseppe Petrantoni

31

Jordan

Petra

1. The inscription is located in Wādī Mūsā, near to the Sīq of Petra,

in front of the Obelisk Tomb.1 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;2 it is written in large letters, which were perhaps orig-

inally 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 fol-

lowers (Num 20,10-11). The Nabataeans built channels that carried water from this

spring to the city of Petra.

2 Atlas, MP5, 164.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

32

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

Text3 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)ankō4 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 compar-

ison to mqbrt’ or qbr’ (Milik 1976, 144; DNWSI, 678). The masculine

form of the noun means ‘burial, tomb’.5

The noun mqbr’ possibly identifies a place, a setting, strictu sen-

su, suitable for the burial, i.e. a grave instead of a complex of tombs;

in addition, the expression, with the demonstrative dnh ‘this’,6 could

prove that the monument was located not far from the place in which

it was engraved.7

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,8 rather than a possi-

ble ‘bd‘mnw = Αβδομανος,9 with the latter being too short for our in-

3 From hereon the Greek text will be presented without spiriti and accents.

4 He is Malco. See commentary.

5 The plural form mqbryn is present in CIS II no. 350.

6 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 construc-

tion, with a postponed demonstrative, is also attested in two inscriptions from Umm al-

Jimāl. Cf. RES no. 2064; PPAES IVA no. 42.

7 Atlas, MP5, 164.

8 Milik 1976, 145.

9 The name also appears in the bilingual painting, no. 6, from the sacred place of Rām.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

33

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

scription.10 This personal name is made up of ‘bd ‘servant, slave’ and

mnkw (= mlkw) ‘Malichos’ (the name of a Nabataean king) and it oc-

curs other 16 times in the Nabataean onomastics.11 It is worth bear-

ing in mind that the form mnkw (= mlkw) reflects the phonetic tran-

sition l > n, a typical Nabataean feature observed in many names.12

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 pseu-

do 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 BCE14

At the end of the line ’kys = Gr. Αχαῖος appears. This name is re-

corded 4 times.15 It is a Greek name having the shape Αχις, Αχιος16

and Ἀχαιός.17

Line 2. At the beginning there is the name of the author’s grand-

father, šly. The root should come from Arabic salā ‘to neglet, to for-

get’, also ‘to console, to comfort’.18

This name is widely used in Nabataean19 although it is not re-

ported in the Greek part of the inscription. In Greek it has the form

Συλλαιος,20 in the bilingual inscriptions from Miletus (no. 49) and De-

los (no. 50). The father’s name of šly is ‘tyhw.21 If the reading is cor-

rect 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”22 or

10 Milik 1976, 144-5.

11 Milik 1976, 144-5.

12 Cantineau 1930-32, 1: 45.

13 Cf. Nehm� 2012, 181-224.

14 Atlas, MP5, 164.

15 Milik 1976, 146.

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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

34

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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)23 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 apr�s,

post�rit�’24 with the singular masculine pronoun -h. Probably it is of

Lihyan origin or a loanword (cf. CIS II no. 197), even though in Lihy-

an the substantive is employed along with words bearing a more de-

tached meaning.25 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 Na-

bataean epigraphy it is rather unusual to break graphically a syntag-

ma into two parts.26

Line 3. After the suffixed pronoun we read the expression l‘lm ll-

myn, literally ‘for the eternity of the centuries’, therefore ‘for ever

and ever’, ‘in saecula saeculorum’. In the middle of the line the read-

ing is difficult. We may reconstruct a possible šnt ‘year’ that gener-

ally 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 unu-

sual, because it commonly follows the sovereign’s name.27

At the end of the line the phrase bḥywhy comes into view mean-

ing ‘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 mascu-

line 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.28 The origin of the two suffixes is difficult to establish.29

Line 4. The letters are in part ruined, but the reading does not seem

to be difficult to reconstruct. The name Αβδομανχος and the patronym-

ic 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 Nehm� (Atlas, MP5, 164).

24 Cantineau 1930-32, 2: 60.

25 Milik 1976, 147.

26 Milik 1976, 147.

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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

35

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

Line 5. The articles τό and τοῖς, of the nouns μνημεῖον and υἱός,

are omitted by the stone-cutter (due to the insufficient space?30). As

regards the former, the term comes from the verb μιμνῄσκω ‘to re-

mind, to put in mind’31 < μνήμη (Dor. μνάμα) ‘memory, souvenir’, but

above all ‘grave, burial’,32 just as μνημεῖον (Ion. μνημήιον) indicating

the ‘memorial’, the ‘grave’.33 Moreover, the substantive would be a

diminutive observed in burial memorials in order to highlight the im-

portance of μνῆμα and certain derivatives in the funeral lexicon.34 Ac-

cording to Healey (1993, 243), the term should be reconstructed as

μνημ�συνον ‘remembrance, memorial’ also referring to a commemo-

rative 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.35 (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”.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

36

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

Commentary

The graffito reports a woman’s name, hgrw, which is well-known in

tNabataean onomastics and considered as equivalent to the Greek

Αγαρη.36 This Greek form is also well-documented.37

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.38 John Healey proposes

reading ṭlyt’ ‘the girl’,39 however his hypothesis is rejected by Bower-

sock who asserts that there is no reason: “why a female name would

need to be qualified in this way”.40 As regards its etymology probably

slyt’ derives from the Arabic root slw ‘to neglet, to forget’, also ‘to con-

sole, to comfort’. Hgrw may be described as a consolation or a com-

fort. 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 dam-

aged 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 recon-

struct 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.

36 Cantineau 1930-32, 2: 84; al-Khraysheh 1986, 61-2; Negev 1991, no. 293; Hea-

ley 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.

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.

39 In a personal message in replying to Bowersock’s questions about this Nabatae-

an word.

40 Bowersock 2015, 124.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

37

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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 ex-

pression 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 read-

ing 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 in-

scription; either way, the name probably derives from the Arabic root

ḫlf. In Nabataean the name ḥlpw41 is found and it is compared with

Arabic ḫalaf42 and Greek Αλαφος43 as Cantineau points out (1930-

32, 2: 96). The general meaning is that of ‘successor’, also ‘enfants’.44

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 repre-

sented 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’.46

41 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.

42 Cf. ICPAN, 198 ḥlf > ḥlīf ‘ally, sworn friend’.

43 This appellative is frequent in Syria, Phoenicia (IGLS 5 no. 2250) and Hauran

(PPAES IIIA 2 nos. 90 and 185).

44 Kazimirski 1860, 1: 620; Negev 1991, no. 452. It is also possible that the root is re-

lated to hlypw (al-Khraysheh 1986, 62), name found in Sinai (CIS II no. 2973), that Can-

tineau (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).

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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

38

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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)47 in which the

Nabataean section is not placed in line 3, as Br�nnow e Domasze-

wski (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 inscrip-

tion 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 Br�nnow, Domaszewski 1904, 335, no. 465 b; CIS II no. 437;

IGLS 21,4 no. 35

Text and translation

1) το προσκυνημα

2) ΑΜΙΨΜΙ[---]    dkyr […](l)n’ bṭb[---]

3) ομου

4) [τ]ο (π)[ρο]σκυνημα

5) šlm

1) the veneration

2) -           let be remembered […](l)n’ in good[---]

3) -

4) (t)he (v)[en]eration

5) peace

Commentary

Line 1. The term τὸ προσκύνημα < προσκυνέω ‘to make obeisance to the

gods or their images’48 reveals the cultic character of the epigraph and

probably constitutes a rare case of a proskynema found out of Egypt.49

47 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.

48 LSJ, 1518.

49 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

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

39

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

Line 2. We read ΑΜΙΨΜΙ, but the sequence of letters cannot be in-

terpreted. 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 remem-

bered. 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.50

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 ori-

gin51 considering the -s ending that recalls the suffix of the nomina-

tive case -ος. Furthermore, Αβολος is attested in an inscription from

Asia Minor;52 with two other forms, Ἀβ�λλας and Ἄβολλα, being at-

tested only by Pape.53

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

50 Lagrange 1898, 179.

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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

40

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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 weath-

ered, is placed below the first one.54

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; Far�s-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.

54 Milik 1958, 250.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

41

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

Commentary

Inscription A

Line 1. The name of the addressee of the manufact, šb‘, appears de-

noting the ‘lion’.55 This term is present in Arabic sab‘ meaning ‘wild

beast, lion’,56 even though in North-West Semitic languages this root

primarily means ‘to be sated’, also ‘seven’.57

The second name that comes into view is ’lh = Gr. Ηλεος,58 the fa-

ther of šb‘; the Greek form furnishes the presumable Nabataean vo-

calization /’ēleh/, a variation of the same transcription that identi-

fies /’īlah/ ‘god’.59

Successively, we find the verb ‘tyd ‘to be set up, to be arranged, to be pre-

pared’, the passive participle of pe‘il,60 that is also attested in Arabic ‘atuda

(= Heb. ‘td) ‘to be ready, to be prepared’ representing a secondary forma-

tion from ‘adda ‘to count, to enumerate’, hence the form i‘tadda ‘to consid-

er, to believe, to evaluate’ (> ista‘adda), ‘he considered, prepared himself’.61

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.

55 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 ăr�’ēl (cf. Cantineau 1930-32, 2: 148). Cf. al-Khraysheh 1986, 170.

56 Negev 1991, 1099. Cf. also ICPAN, 309.

57 See DNWSI, 1101-2. It appears in the Bible as a personal name šibə‘� (1Sam 20,1

and 1Chr 5,13). In Arabic this semantic passage from ‘seven’ to ‘lion’ is explained, ac-

cording 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 at-

tacks 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).

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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

42

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

At the end of the line there is yq’, probably the name of the au-

thor. 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’62 > proper name

Jāqeh (Prov 30,1) < yiqəhāh ‘obedience’ (Gen 49,10; Prov 30,17), Akk.

utaqqu ‘to obey’,63 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’64 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).65 The

fact that a Nabataean bears a Greek name would represent the proof

of the spread of Hellenization, at least superficially, into the Nabatae-

an 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, al-

so known as The Valley of the Moon, is a valley cut into the sand-

stone and granite rock in southern Jordan 60 km to the east of Aqa-

ba. There are about 30,000 inscriptions in the form of rock paintings

and graffiti realized first by the Thamud and then by the Nabatae-

ans who installed in Wādī Ram in the 4th century BCE They lived

peacefully along with the Thamud worshipping the same deities, in-

cluding Dūšarā. Besides paintings and graffiti, the Nabataeans al-

so built temples. The Greek text is located on the left part of the in-

scription, while the Nabataean is on the right side.

62 DGes, 487; KAHAL, 224 ‘vorsichtig’, ‘Gehorsam’.

63 KAHAL, 263.

64 Lane, 3059.

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 sub-

sequently killed; also Josephus (AJ 20.5.1) cites the figure of an instigator (Gr. γόης ‘im-

poster, 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 accord-

ing to Josephus it took place at the end of the mandate of Fadus, between 45-46 CE.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

43

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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 translation66

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,67 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.68 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;69 moreover,

this name would be the only proof in the Thamudic epigraphy.70

As regards the onomastics, the real name of the author of the

painting is Ουαβαλας,71 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

66 The transcription is that of Milik 1976, 145, no. 5.

67 Cf. Negev 1991, no. 817.

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

scription no. 8.

69 Harding, Littmann 1952, no. 57 A.

70 IGLS 21,4, 177.

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).

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

44

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

theophoric composed of whb ‘gift, present’ and the name of the deity.72

The name Αβδομανος,73 which corresponds to ‘bd‘mnw, is made up

of ‘bd ‘servant, slave’ and the root ‘mnw, and it was recorded in Si-

nai, meaning ‘rester dans (un lieu)?’,74 Ar. ‘amana ‘to remain, to stay

in a place’.75 The Greek shape of the name probably shows the assim-

ilation of the Aramaic nominative mark -o (=*u) into the second ele-

ment 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,76 in which the second -o is

the nominative mark: */‘aḇdo-‘oman/. The name ‘bd‘mnw is well doc-

umented in Hauran.77

The grandfather of Ουαβαλας is a certain Αιαλος who appears in

line 3 in the Greek section of the inscription. The equivalent in Na-

bataean is ‘ylw mentioned in the last line.78 The root is of semitic or-

igin having a correspondence to the Hebrew ‘�lay, a personal name

of one of David’s heroes who gave him strong support in his king-

dom.79 It means ‘the higher’ < ‘l’, Ar. ‘ālin < ‘aliya ‘to be high’. Can-

tineau (1930-32, 2: 129) connected the Nabataean root to the Arabic

ġayyāl or ‘ayyāl.80 The former indicates a wide, ample space,81 while

in the Koranic context it describes the following: “such as one judges

to be of little extent, through it is for extending”;82 the latter takes on

the meaning of: “that inclines from side to side in gait and is proud,

haughty or self-conceited, therein”.83

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

73 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 in-

scription found in the northern part of the acropolis of the city of Avdat/Oboda shows

(Negev 1982, 17, no. 3).

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

75 The hypothesis of Littmann (PPAES IVA, 79) according to which the second ele-

ment ‘mnw is identified with the Egyptian god Amun is hard to support.

76 IGLS 21,2 no. 183 and Dalman, 2: no. 98.

77 PPAES IVA, no. 79; RES no. 2100.

78 This name is encountered at Hegra and in Hauran. Cf. JSNab no. 344.

79 1Chr 11,29; Cf. DGes, 954 and KAHAL, 402.

80 Cf. also Negev 1991, no. 882. al-Khraysheh (1986, 139) thinks it is Ġaiyāl ‘killer’,

‘Menchelm�rder’. In addition, see ICPAN, 451.

81 The form ġayyāla referred to a woman obtains the meaning of ‘she is tall’.

82 Lane, 2319.

83 Lane, 2213.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

45

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

The Nabataean text provides the most information about Αιαλος,

the son of ‘bd‘bdt, the latter omitted in the Greek part.84 ‘bd‘bdt, in

turn, was the son of qynw, as we can read in the last line of the Na-

bataean 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’.85

At the end of the two texts we read the ethnic Φαινήσιος, Na-

bataean pyny, that indicates the ancient town of Faina in Hau-

ran, corresponding to the modern village of al-Mismiyah,86 consid-

ered the μητροκωμία (lit. ‘mother village’ < -κώμη87) of the ancient

Trachonitis,88 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-docu-

mented in ancient topographies,90 was probably a strategic place to

stop over and successively to conquer Kanatha or Bosra.91

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 Se-

mitic languages and Latin of Roman soldiers, was of considerable im-

portance; this is because it was very widely known and the sanctu-

ary of Iram was much visited.92

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.93 Instead,

in Milik’s view (1976, 145), we may date the inscription to the mid-

84 For the recordings see Cantineau 1930-32, 2: 126. Cf. also al-Khraysheh 1986, 131

and Negev 1991, no. 815.

85 Proto-Afro-Asiatic *ḳVyVn- ‘forge’ (HSED no. 1629). Negev (1991, no. 1025) trans-

lates “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) vo-

calizes as Qainā and translates as ‘Sklavin, S�ngerin’. 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.

86 At almost 50 km south of Damascus.

87 LSJ, 1130.

88 It is mentioned in Luke 3,1.

89 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.

90 In Ierocle (Synecdemus 723.1) there is Φαίνα; Dia-Fenis (Not. Dign. [occ.] [or.]

37.23).

91 Sartre 1999, 197-8.

92 Quellen, 290.

93 Quellen, 290-1.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

46

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

dle of the 2nd century CE, since the main painting of the temple re-

ports the date: year 147 of the month of August.94 The shape of the

Nabataean script seems to be late.95

8. It is a piece of gr�s, intact on the left, but broken in the remain-

ing 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 ’Al-

lat 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 tem-

ple 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 men-

tion 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

94 Savignac, Horsfield 1935, 265.

95 IGLS 21,4 nos. 141, 178.

96 IGLS 21,4, 181.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

47

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

Wādī Ġuwayr (al-Šawbak)

9. The inscription was found in the Wādī Ġuwayr, one of the numer-

ous 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.97

Dating unknown

Bibliography CIS II no. 489; Br�nnow, 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 Nabatae-

an epigraphy and, more generally, in the Aramaic.98 The root gd’ be-

comes 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 on-

ly in Petra, but also in Hegra.99

The worship of Gad, who became the god of Fate (Lat. Eutychus100

< Gr. Εὐτύχης), was widespread in Hauran.101 In fact, several sanctu-

aries, called Τύχεια or Τύχαια, were built102 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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

48

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

rē) writes about Bayt Gadā located on the mountaintop and trans-

formed into monasteries during the Christian period.103

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 proba-

bly lies the origin of the idea of the ‘fate’, defining the destiny of human

beings.104 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).105

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.107

Ġūr al-Ṣāfī

10. This inscription was found at al-Naq‘ cemetery of Ġūr al-Ṣāfī, an-

cient Zoar,108 located to the south-east of the Dead Sea, in Jordan.

It is an epitaph carved in a rectangular tombstone of whitish sand-

stone. The letters are painted in red colour and the entire inscrip-

tion consists of four lines.

The Greek text presents a calligraphic oval script with symmet-

rical 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-di-

viders. 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, Sardin-

ia (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 Krah-

malkov 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 Per-

sian (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 (Ne-

gev 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 Be-

la), and it was part of the late Roman province of Palaestina Tertia.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

49

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

As regards al-Naq‘ cemetery, it was probably predominantly of

Jewish ritual. Indeed, there are tombstones written in Aramaic, He-

brew and Greek, as well as a Hebrew-Greek bilingual,109 which be-

longed to Jewish, Christian and pagan people. The inscriptions of

deceased Jewish people are dated according to the cyclical agricul-

tural calendar of seven years bearing Jewish symbols; meanwhile

the Christian inscriptions are dated starting from the Lord’s Day and

have crosses as symbols.110

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 translation111

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 tran-

scription, ’šm‘yn, is intriguing. The Greek masculine form, Ισμαηλος,

is attested in two epitaphs from Jericho,112 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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

50

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

the middle of the 4th century CE (341/2).113 ’šm‘yn is probably new in

the Nabataean onomastics, since it is not encountered elsewhere in

Nabataean, Palmyrene, Hatraean or in pre-Islamic Arabic.114

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 Αβδοάρθα115 (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.116 The name ’bšlm is recorded at Hegra as well.117

At the end of line 1 we read the author’s name Ζωίλος, usually

used in Greek,118 deriving from the word ζωή ‘life’,119 and transcribed

as zyls in Nabataean, line 3. The first mention of this name in Pales-

tine during the Hellenistic period is found in the Greek-Aramaic vo-

tive bilingual inscription of the 3rd-2nd century BCE from Tell Dan120

and in an amphora from Rhodes.121

The patronymic of Zoilos is Αλεβος, a Semitic name that corre-

sponds to Nabataean ‘lbw, in line 3.122 The root probably comes from

Arabic ġālib ‘winner’ < ġalaba ‘to win, to subdue, to conquer’123 and

is present in the Aramaic ‘lb as well (Syr. ‘eleb ‘to exceed, to surpass,

to cause wrong, to oppress’124). Negev’s assumption (1991, no. 886 as

first hypothesis) of an Arabic origin ‘alib ‘to become thick or coarse,

113 Wadd. no. 2247.

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.

115 IPT Ia nos. 21 and 79.

116 IPT Ib, 127.

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

118 It is above all employed in Greece. Cf. LGPN I, II, IIIA, IIIB, IV, VA, VB.

119 Cf. Pape 1911, 448.

120 Cf. SEG 26 no. 1684 with references and Arbeitman 1994.

121 Cf. SEG 8 no. 237.

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).

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

51

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

rude’,125 ‘hard, tough’ is unconvincing. The Greek form Αλεβος126 is

exclusively recorded in Hauran in funerary127 and monumental128 in-

scriptions.

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 in-

dicate 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.129

As regards the Nabataean part of the epitaph, in the first line we

read the term npš’ 130 followed by the author of the object.

In the last line there is the substantive ’ntth (<* ’nth) ‘wife’, fol-

lowed by the singular masculine suffix pronoun -h ‘of him’ = ‘his’.131

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 at-

tested somewhere along with ’tt-. 132

The affection of the husband Zoilos to his dead wife is manifested

by the adjective ḥbybt ‘be loved’, the singular feminine passive par-

ticiple 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 engrav-

ing of the epitaph or the month when Ismeel� died.

125 Lane, 2126.

126 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 Ly-

cia (LGPN VB).

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 mean-

ings. In some inscriptions it is translated as ‘tomb’ or ‘gravestone’, while in other cas-

es 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 oth-

er regions, the term npš’ has the same meaning both at Hegra and Petra, and in Hau-

ran 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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

52

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

Madaba

11. The epigraph was found at Madaba,133 a town located 30 kilome-

tres 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

1) d’ mqbrt’ wnpš’ dy ‘l’

2) mnh dy ‘bd ’bgr dy mtqr’

3) ’yšywn br mn‘t dy mn

4) ’l ‘mrt lšlmn brh

5) bšnt tlt lhprk bṣr’

6) Σελαμαν χρηστε και

7) αλυπε χαιρε Αβγαρ ο και Εισιων

8) Μονοαθου υιος υιω τειμιω το μνημα

9) εποιησεν ετους τριτου επαρχειας

1) This is the tomb and the monument which is on top

2) of him which made ’Aḇgar who is (also) named

3) ’Eyšīōn, son of Mono‘aṯ, of the

4) tribe of ‘Amirat, for Šelaman his son

5) in the year three of the eparch of Bosra

6) Selaman, good and

7) without pain, hi! Abgar (also) named Ision,

8) son of Monoath, for (his) well-loved son, the monument

9) he made, in the third year of the eparchy

133 The Moabite city of M�dəbā 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 con-

quered 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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

53

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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.134 The author clearly distinguishes mqbrt’ (the feminine form of

mqbr’ in no. 1) from npš’. If the first term135 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 ra-

re name among the Nabataeans inasmuch it is recorded in the north-

ern Aramaic onomastics.136 This name derives from *bgrt (Ar. buǧrah)

‘navel’.137 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, scep-

ticism’ < ’aīsa)138 and usually transcribed in Greek as Ἰᾶσος (RES no.

463).139 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’ < ma-

na‘a ‘to ban, to prohibit, to forbid’,141 Heb. māna‘, Ge. and Am. m�n�n�

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

135 It is found in CIS II nos. 181, 196, 2033; PPAES IVA no. 106; RES no. 1090; in con-

struct state mqbrt in DM, II no. 18; RES 481. It is also present in Palmyrene (Cf. DN-

WSI, 678).

136 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).

137 Cantineau 1930-32, 2: 70; Negev 1991, no. 5. The name ’bgr was found in the graf-

fiti 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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

54

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

‘to reject, cast aside’. The primary syllable is n‘ which has a negative

force,142 but Syr. mana‘ ‘to bring, lead’, ‘to arrive, come, attain’.143 The

Greek transcription144 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.145 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 ‘sub-

division of a tribe’ < ‘amīr ‘a place inhabited, peopled’).146 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;147 it is also the name of a Safaitic tribe from the northern

Transjordan desert.148 In one of the Nabataean graffiti found in the

area of Burqu‘149 the author describes himself as dy mn ’l ‘mlt;150 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)151 < šlm

‘peace’. Its variations šlmn or ślmn (cf. Ar. salamān)152 are well-docu-

mented in Petra153 and in Hegra154 as well as in the Palmyrene155 and

Hatraean156 onomastics.

142 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).

143 Payne Smith 1903, 282; Sokoloff 2009, 784.

144 With variations Μοναθος (Wadd. no. 2499) and Μονοαθου (PPAES IIIA, 7 no. 800).

Cf. Wuthnow 1930, 78.

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

146 Al-Khraysheh (1986, 143) noted the existence of the name ‘myrt meaning ‘head-

gear’, ‘Kopfschmuck’, the diminutive form of ‘mrt.

147 Negev 1991, no. 53. See also ICPAN, 436.

148 Milik 1958, 245 with references at note 5.

149 Macdonald 1993, 359.

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).

153 CIS II no. 426.

154 CIS II nos. 294, 302; JSNab no. 172.

155 PAT, 440 = Stark 1971, 51-2 and 114.

156 Beyer 1998, 166.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

55

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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’,157

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’158 is quite common; it is a Greek loanword < *ὑπαρχία

= ὑπαρχεία ‘province, district’.159 In Nabataean this term is used to

mark the territory, in this case that of Bosra, belonging to the Pro-

vincia Arabia.160 The year 3 of the eparchy of Bosra corresponds to

108/109 CE, the later date of the era of the City.161

Lines 6-7. The Greek section of the inscription starts by mention-

ing 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, hon-

oured’ 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 en-

tirely 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

1) Αβδαλλας Ανα[.]ου το ταφειμα

2) τουτο ε[ποιη]σεν [εξ ουσ]ιων ιδιων θε-

3) [---]εκατερω

4) θεν εκτισεν αμα και [ιε]ρον τερμα

5) […]ετους [---]μτ κατα

6) στα(σ)εως [.........]Αντωνειν-

7) [-ου]Καισαρος ετους ιθ[---]  mnbrk’

157 Cf. Cook 1898, 43.

158 See CNSI nos. 231 and 260; YTDJD I nos. 297, 323, 326; II, 97, 109, 111. General-

ly, 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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

56

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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

2) m[ad]e [at hi]s own expense

3) [---]of the two (side)

4) he built at the same time the [sa]cred boundary

5) […]the year [---]340 the foun-

6) -da(t)ion (of the city ?) [………]of Antonin-

7) [-o]the Emperor the year 19[---]  mnbrk’ (?)

Commentary

The author’s name is identified, in the editio princeps,162 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 pat-

ronymic Αναμος (= Nab. ‘nmw).164 The latter seems to originate from

the Arabic ġānim ‘qui fait du butin qui r�ussit sans effort’,165 in Saf.

‘nm and Palm. ‘nmw ‘successful, noble’166 (< 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].167

Lines 6-7 provide two important historical references in order to

date the inscription; in fact, there is the name of Antoninus (proba-

bly the Roman Emperor Antoninus Pius168) 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

162 Germer-Durand 1895, 90.

163 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 Si-

nai in the Wādī Haggag (see Negev 1977, no. 42).

164 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 (Wuth-

now 1930, 23, 121, 159).

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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

57

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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 Na-

bataean 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.

m�dəbā169) or, according to the original Moabitic diction, mhdb’

which occurs in the stele of Mesha.170

Zīzah (Zuwaiza), East of Madaba

13. This epigraph was found during the excavations about 250 me-

tres south-east of Qala‘at Zīzah probably near the ruins of an an-

cient church.171

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.172 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 translation173

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 Mada-

ba. 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 dimen-

sions resembled those of the Roman milestones.

172 Jaussen, Savignac 1909, 588.

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

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

58

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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’174 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).175 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.176 This

174 Such a reconstruction is found in Quellen, 214: “Dies ist der Tempel des Baal vom

(Berg) Pegor”.

175 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 Mac-

donald 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 Wuth-

now 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.

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).

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

59

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

name has Greek origins and its meaning refers to the form Δημέας

that Pape (1911, 288) identifies with ‘Volkmann’ or ‘D�rfler’, a clear

connection with the people (< Gr. δῆμος).177

The father’s name is hll (= Gr. Ελλην[ος] in line 6).178 If the read-

ing is right, it represents the Biblical name hillēl (Jugd 12,13), the fa-

ther of Abdon the Pirathonite,179 whose root hll means ‘to be clear,

bright’,180 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 presuma-

bly used to indicate the name of the god Ba‘al that should appear in

line 1, according to the reconstruction.181

Line 5. The reading is uncertain, but we find the title, probably be-

stowed on Demas, mr byt, that means ‘Lord of the Temple’.182 If the

reading is right,183 successively we may read a probable ’tr, as this is

the ‘Lord of the Temple of the place’.184

Line 6. The Greek part of the inscription begins by quoting the

author’s name and his father’s name. The reading Δημᾶς is prompt-

ed by the Nabataean text since here we may only trace the middle

177 dyms is a loanword in Nabataean assuming the meaning of ‘people’, as well as in

Pal., Sam. and Syr. (DNWSI, 253).

178 Jaussen, Savignac 1909, 589. Cf. also Negev 1991, no. 303 and Cantineau 1930-

32, 2: 86.

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 re-

ports that in Ethiopia the women, during the public rejoicing, have the habit of repro-

ducing the sound ellellell-ellellell. Cf. ΜDGes, 278-9 and KAHAL, 129-30 ‘r�hmen, 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. N�ldeke (1909, 184-5) ar-

gued that, with reference to the inscription in RES no. 1088, mr’ byt’ was to be identi-

fied with rabb al-bayt which is recorded in the Koran (CVI, 3). In the Nabataean epig-

raphy the connection of the title mr byt to the deity of al-‘Uzzā comes from an inscrip-

tion 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 (Nehm� 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 recon-

struct the substantive.

184 Quellen, 214.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

60

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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

Line 7. At the beginning we read -AMOY185 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 dedica-

tion 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-pəʿ�r ‘The Lord of mount Peor.186 With

the arrival of Greek culture in the Moab region, Ba‘al would become

God/Zeus worshipped in other places in the same way.187

Jerash

14. This epigraph is engraved on a red dolomitic limestone (in Ara-

bic Mizzi aḥmar),188 found in May 1931 in the ancient town of Jerash,189

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 Vin-

cent (1872-1960), from �cole Biblique et Arch�ologique Fran�aise

of Jerusalem in collaboration with his colleague Father A.J. Savig-

nac (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.

188 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.

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. Dur-

ing 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. Moreo-

ver, in 749 CE a major earthquake destroyed much of Jerash and its surroundings and

its population decreased.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

61

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

62

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

The reading of line 7 is rather difficult. On the basis of the facsim-

ile, which was presented by Father Vincent, Kraeling thought the text

could be reconstructed as τῶν μερῶν. Subsequently, in a further re-

vision, Quellen reads …]ΙΩΝΜΕΙΩΝ190 not providing a translation.

In line 8 we probably read ταῖς ἐπάλξεσι which is a possible ref-

erence to the protective walls (< ἡ ἔπαλξις ‘means of defence, para-

pet, shelter’191) 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 ad-

dressee remaining unknown owing to the deterioration of the stone.

In line 11 the text is illegible as we can observe through the pho-

tograph. Initially, Father Vincent tried to reconstruct a segment be-

ginning 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 ref-

erence to the king Aretas IV.192

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, af-

ter the erection of Zeus’s temple in 69 CE,193 may confirm what has

already been mentioned above. The terminus ante quaem of the ur-

ban 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.194 Between 69 and 80 CE the urban renovation project

was completed and huge defensive walls were erected.195 If the read-

ing 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.

192 According to the palaeography of the text, the predecessors Aretas I and Are-

tas II (169-96 BCE) are earlier, while Aretas IV (9 BC-40 CE) would be more appropri-

ate for our inscription.

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

194 Kraeling 1938, 397-8, no. 50. Commodus was the first of the gens Ceionia to be-

come a consul.

195 Bowersock 1973, 138-9.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

63

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

jacent to the walls of the city.196 There might be a coincidence be-

tween 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 be-

tween 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 east-

ern part of Umm al-Jimāl, a village about 17 kilometres away from

Mafraq in northern Jordan.197 The column represents a gift to the lo-

cal 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 ed-

itors 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;

PPAESIVA no. 38; CISII no. 190; Clermont-Ganneau 1906a, 215; Meyer 1906,

344; Littmann 1909, 383-6; JSNabno. 39; Cantineau 1930-32, 2: 23; Sourdel

1952, 60; IGLS 13,1 no. 9031; IGLS 21,5.1 no. 98

Text and translation

1) mšgd’

1) Μασε-

2) dy ‘bd

2) χος Α-

3) mškw

3) ουειδ-

4) br ‘wy-

4) ανου

5) -d’ ldw-

5) Δουσ-

6) -šr’

6) αρει Α-

7) αρρα

1) The cult-stone 1) Mese-

2) which made 2) kos (son of) A-

3) Mašekō

3) ueid-

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 Nabatae-

an 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 com-

munity and a trading outpost dependent on Bosra. Cf. De Vries 1998.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

64

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

5) -ḏā, for Dū-

5) (for) Dus-

6) – šarā

6) -are A-

7) -arra

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.198

The name is presumably a loanword or an interference from Arabic

into Nabataean199 < Ar. saǧada ‘to bend until you touch the ground

by your forehead in act of worship’.200

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.201

The Greek Μασεχος is also found in the form of Μασαχος.202 It is

the abbreviation of a theophoric name, mšk’l.203 This compound form

is recorded in Safaitic msk-’l and occurs as Μασαχηλω in a Greek in-

scription from Sī‘ as well;204 its meaning is ‘(god) has taken posses-

sion’205 and it is a birth name that represents the cultural uniformi-

ty of the Syrian regions of Hauran and Ḥarra.206

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

198 This noun, which is formed by the prefix m- indicating the place, is used in Offi-

cial Aramaic meaning ‘place of adoration’, ‘object serving as a permanent sign of ad-

oration 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.

199 Colombo 1994, 73.

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

201 al-Khraysheh 1986, 115 and Negev 1991, no. 701. Also, msk (ICPAN, 545). It ex-

ists in Palmyrene (PAT, 437 = Stark 1971, 37 and 97).

202 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.

203 This name is present in the bilingual text in no. 28, in PPAES IVA no. 101 and in

RES no. 2117.

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

205 Sartre 1985, 216.

206 De Vries 2009, 179.

207 At Umm al-Jimāl there is an epitaph dedicated to a son of Α]ουιδανου (IGLS

21,5.1 no. 512).

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

65

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

Αουειδανος is simply an extended form of the name Αουειδος,208 used

in Arabic as ‘Awīḏ. 209 Therefore, ‘Awīḏā should be the diminutive form

of a Sinaitic name, more precisely ‘wdw = Αυδος, hence the corre-

sponding Arabic root ‘ūḏ. 210 The Arabic verb ‘āḏa means ‘to take ref-

uge, to take cover’.211

The last two lines 5-6 show the name dwšr’, equivalent to Greek

Δουσαρει. This is the main deity of the Nabataean pantheon, wor-

shipped 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;212 it

is a ‘sacred’ mountain range that, during the Jāhiliyya, was consid-

ered ḥaram ‘prohibited’ or ḥima ‘protected’; the latter is interpreted

as a place protecting animals, plants and fugitives.213 Actually, the

substantive šarā also has the meaning of ‘road, tract of land, moun-

tain’214 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 Lat-

in 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 im-

perial medals bearing the legend ACTIA DUSARIA, along with the

representation of a hand press.216

A temple dedicated to Dusares was found in Italy, at Pozzuoli, since

the ancient Puteoli traded with the Near East; the Nabataean pres-

208 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.

209 PPAES IIIA no. 138.

210 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).

211 Cf. Al-Khraysheh (1986, 136) who renders ‘Uwaiḏū with ‘Zuflucht Suchender’ and

Negev 1991, nos. 853, 854.

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.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

66

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

ence is attested starting from the middle of the 1st century BCE.217

In the last line of the Greek text there is the name Ααρρα, not pre-

sent 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.218 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 roy-

al house as is shown, for instance, in a legend depicted on a silver

coin of king Obodas III, dated to 16 CE.219 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.220 As a matter of fact, the king established

closer relations between Bosra and Dūšarā assimilating the latter to

the local deity Ā‘arā.221 In addition, Rabbel II yearned to make Dūšarā

the Nabataean national god, his own god and that of his city.222

The present inscription provides the first Greek transcription of

the whole name of Dūšarā with the addition of ’‘r’ = Ααρρα. As re-

gards its etymology various editors have proposed different theories

about its meaning.223

Littman’s thesis (PPAES IVA, 35) seems to be particularly reason-

able; he asserted that Ā‘arā derives from Arabic ġarā meaning ‘good,

217 Museo archeologico dei Campi Flegrei 2008, 60-3. Cf. Lacerenza 1988-89 and

CIS II no. 157.

218 Cf. those inscriptions in Savignac, Abel 1905, 592 and ESE I, 330.

219 Healey 2001, 154.

220 Teixidor 1977, 85.

221 Dijkstra 1995, 312.

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 He-

brew 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 hypothe-

sis 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).

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

67

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

beautiful’,224 ‘beau, joli, bon’,225 from which, in turn, the name of a

stone idol al-ġariyyu derives, under which Dūšarā was worshipped

at Petra.226 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 sacri-

ficed animals was poured.227

The Greek form would be Αρρα(ς)228 probably identified with Ares,

the God of war. This theory may well be supported if the hypothe-

sis, according to which the deity’s name derives from the Arabic af‘al

form ġry > aġrā ‘dyeing’ or ‘anointing’,229 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’.230

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.231

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 un-

covered 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;232 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.

225 Kazimirski 1860, 462.

226 PPAES IIIA, no. 138; PPAES IVA, no. 35.

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

θεὸς Ἄρης ἐν Πέτρᾳ.

228 Sourdel 1952, 60.

229 Teixidor 1977, 85-6.

230 Lane, 2254.

231 Wenning 2001, 84-5.

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

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

68

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

Text and translation

1) dnh npšw phrw

1) Η στηλη αυτη Φε-

2) br šly rbw gdymt

2) -ρου Σολλεου

3) mlk tnwḥ

3) τροφευς Γαδι-

4) -μαθου βασιλευς

5) Θανουηνων

1) This is the memorial of Fehrō 1) This (is) the memorial of Fe-

2) son of Šollē, tutor g Gaḏīmaṯ 2) - ros, (son of) Solleos

3) king of Tanūḥ

3) tutor of Gadi-

4) - mathos king of

5) Thanuenos

Commentary

The Nabataean script seems to be at a transitional stage towards the

Arabic script. Indeed we may notice a tendency towards ligature,233

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.234 Moreover, we

may observe that the g has already assumed a similar form encoun-

tered in the Kufic script.235 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 phrw236 = Gr. Φερου; his father’s

name is šly, well-documented in other inscriptions.237 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.238

233 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.

234 The tendency to separate the letters is probably due to the usus of the monu-

mental inscriptions.

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 de-

scendant 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-

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

69

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

The most important character is Gaḏīmaṯ, the king of the Tanūḥ

(gdymt239 mlk tnwḥ = Gr. Γαδιμαθου240 βασιλεύς Θανουηνων). He was

a sovereign, documented in the Islamic historiography, who reigned

between the Euphrates and Syria as chief of the Tanūkh tribe241 in

the second half of the 3rd century CE (around 275 CE).242 As a conse-

quence the stele should be dated to the end of the century.243

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 ‘educa-

tor’, ‘rabbi’, or better ‘mentor’. Although the first meaning of τροφεύς is

‘one who brings up, foster-father’,244 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 Hellen-

istic 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 inscrip-

tion of Miletus, the minister presents himself as a ἀδελφὸς βασιλέως,

as reported in the Aramaic version ’ḥ malk’ and not rb.245 Therefore, the

τροφεύς, instead of σύντροφος ‘foster-brother’, is to be interpreted as

‘p�re nourricier’ (= θρεπτήρ), ‘putative-father’, as Sartre pointed out.246

Jeanne and Louis Robert rejected Sartre’s hypothesis247 without add-

ing 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.

239 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).

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 Pal-

myre 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.).

242 Mascitelli 2006b, 235.

243 Cantineau 1930-32, 2: 25.

244 LSJ, 1827.

245 Cf. no. 49.

246 Mascitelli 2006b, 232.

247 Robert, Robert 1980, 479.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

70

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

Finally, the Greek translation of the Aramaic term npš with στήλη

should be considered. It is worth remembering that the old Nabatae-

an city of Umm al-Jimāl was rebuilt by Christians, as Littmann assert-

ed (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,248 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 slab249, 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

248 It is an important site in the Hauran. Its ruins show a settlement dating to a pe-

riod 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.

249 MacAdam, Graf 1989, 191.

Giuseppe Petrantoni

Jordan

Antichistica 28 | 11

71

Corpus of Nabataean Aramaic-Greek Inscriptions, 31-72

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 remain-

ing 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:250 O followed after a space by M,

in order to render the feminine name Ομεαθη, attested in the region

of Umm al-Quṭṭayn.251

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 re-

semble 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 pres-

ence of a single O followed by a M (after a space)252 seems to be im-

probable.

The palaeography would date the inscription to the end of the 4th

century or to the beginning of the 5th c. CE.253

250 IGLS 21,5.1 no. 330.

251 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.

252 Bader (IGLS 21,5.1: 330) points out that: “Ce qui a �t� pris pour du nabat�en se

lit d’abord comme un O puis, apr�s un espace, un M”.

253 SEG 39 no. 1610.

Antichistica 28 | 11