Language Change in the Wake of Empire: Syriac in Its Greco-Roman Context

Citation preview

Language Change in the Wake of Empire

Linguistic Studies in Ancient West Semitic edited by

Cynthia L. Miller-Naudé and Jacobus A. Naudé The series Linguistic Studies in Ancient West Semitic is devoted to the ancient West Semitic languages, including Hebrew, Aramaic, Ugaritic, and their near congeners. It includes monographs, collections of essays, and text editions informed by the approaches of linguistic science. The material studied will span from the earliest texts to the rise of Islam.  1. The Verbless Clause in Biblical Hebrew: Linguistic Approaches, edited by Cynthia L. Miller  2. Phonology and Morphology of Biblical Hebrew: An Introduction, by Joshua Blau  3. A Manual of Ugaritic, by Pierre Bordreuil and Dennis Pardee  4. Word Order in the Biblical Hebrew Finite Clause: A Syntactic and Pragmatic Analysis of Preposing, by Adina Moshavi  5. Oath Formulas in Biblical Hebrew, by Blane Conklin  6. Biblical Hebrew Grammar Visualized, by Francis I. Andersen and A. Dean Forbes  7. Time and the Biblical Hebrew Verb: The Expression of Tense, Aspect, and Modality in Biblical Hebrew, by John A. Cook  8. Diachrony in Biblical Hebrew, edited by Cynthia L. Miller-Naudé and Ziony Zevit  9. The Syntax of Volitives in Biblical Hebrew and Amarna Canaanite Prose, by Hélène Dallaire 10.  The Relative Clause in Biblical Hebrew, by Robert D. Holmstedt 11.  Language Change in the Wake of Empire: Syriac in Its Greco-Roman Context, by Aaron Michael Butts

Language Change in the Wake of Empire Syriac in Its Greco-Roman Context

A aron M ichael B utts

Winona Lake, Indiana Eisenbrauns 2016

Copyright © 2016 Eisenbrauns All rights reserved. Printed in the United States of America. www.eisenbrauns.com

Library of Congress Cataloging-in-Publication Data Names: Butts, Aaron Michael, author. Title: Language change in the wake of empire : Syriac in its GrecoRoman context / Aaron Michael Butts. Description: Winona Lake, Indiana : Eisenbrauns, 2016. | Series: Linguistic studies in Ancient West Semitic ; 11 | Revision of author’s thesis, University of Chicago, 2013. | Includes bibliographical references, appendixes, and indexes. | Description based on print version record and CIP data provided by publisher; resource not viewed. Identifiers: LCCN 2015050107 (print) | LCCN 2015047546 (ebook) | ISBN 9781575064222 (paperback) | ISBN 9781575064215 (hardback : alk. paper) Subjects: LCSH: Greek language—Influence on Syriac. | Syriac language—History. | Languages in contact. Classification: LCC PJ5415 (print) | LCC PJ5415 .B88 2016 (ebook) | DDC 492/.32481—dc23 LC record available at http://lccn.loc.gov/2015050107

The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials. ANSI Z39.48-1984.♾™

For Sebastian Brock

Contents Preface and Acknowledgments . . . . . . . . . . . . . . . . . . . . . .   xi Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   xii General xii Reference Works  xii Abbreviations and Symbols in Linguistic Glosses  xv Abbreviations and Citations of Biblical Books  xvi Transcription / Transliteration  xvii Chapter 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  

1

Part 1: Prolegomena Chapter 2.  The Contact Linguistic Framework . . . . . . . . . . . . . . 11 2.1. Overview  11 2.2. Contact Linguistic Terminology  11 2.3. Early Typologies of Language Contact  13 2.4. The Typology of Van Coetsem  16 2.5. The Typology of Thomason and Kaufman  20 2.6. Synthesis  21 2.7. Conclusion  24 Chapter 3.  The Sociohistorical Setting . . . . . . . . . . . . . . . . . . 25 3.1. Overview  25 3.2. The Conquests of Alexander  25 3.3. ʾUrhɔy Is Edessa  27 3.4. Analyzing Contact-Induced Changes in Syriac Due to Greek 30 3.5. Conclusion  40 Part 2: Loanwords Chapter 4.  Greek Loanwords in Syriac: The Methodological Framework . . . . . . . . . . . . . . . . . . . 43 4.1. Overview  43 4.2. History of Research  45 4.3. Definition  46 4.4. Corpus  47 vii

viii

Contents 4.5. Lehn- oder Fremdwörter? 48 4.6. Code-Switching  50 4.7. Immediate Source and Ultimate Source  53 4.8. Latin Loanwords in Syriac  54 4.9. Greek Loanwords as Inheritances in Syriac  56 4.10. The Greek Source  61 4.11. Conclusion  63

Chapter 5.  The Phonological Integration of Greek Loanwords in Syriac . . . . . . . . . . . . . . . . . . . . . 64 5.1. Overview  64 5.2. Consonants  65 5.3. Vowels  88 5.4. Conclusion  95 Chapter 6.  The Morphosyntactic Integration of Greek Loanwords in Syriac . . . . . . . . . . . . . . . . . . . . . 97 6.1. Overview  97 6.2. Nouns  97 6.3. Verbs  111 6.4. Particles  117 6.5. Secondary Nominal Derivations Involving Greek Loanwords in Syriac  120 6.6. Structural Consequences of Loanwords  129 6.7. Conclusion  135 Part 3: Grammatical Replication Chapter 7.  Grammatical Replication: The Methodological Framework . . . . . . . . . . . . . . . . . . 139 7.1. Overview  139 7.2. Definition  139 7.3. Change in Frequency of a Pattern  140 7.4. Creation of a New Structure  144 7.5. Grammatical Replication within Historical Linguistics  147 7.6. Alternative Designations for Grammatical Replication  148 7.7. Conclusion  151 Chapter 8.  The Syriac Copula ʾiṯaw(hy) Replicated on Greek ἐστίν . . 153 8.1. Overview  153 8.2. Verbless Clause Formation in Syriac  154 8.3. Extension in the Copulaic Use of ʾiṯ 158 8.4. Late Aramaic Comparanda 164 8.5. The Increase in the Frequency of ʾiṯaw(hy) 170 8.6. Conclusion  172

Contents

ix

Chapter 9.  The Syriac Conjunctive Particle den Replicated on Greek δέ . . . . . . . . . . . . . . . . . . . . . . 174 9.1. Overview  174 9.2. Earlier Aramaic Antecedents of Syriac den 176 9.3. The Replication of Syriac den on Greek δέ  180 9.4. Late Aramaic Comparanda 189 9.5. Conclusion  190 9.6. Excursus: Syriac ger and Greek γάρ  191 Chapter 10.  Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . 195 10.1. Overview  195 10.2. The Transfer of Structure in Situations of Borrowing  195 10.3. The Beginning of Syriac-Greek Language Contact  198 10.4. Syriac in Its Greco-Roman Context  202 Appendix 1.  Greek Loanwords Inherited in Syriac . . . . . . . . . . . 212 Appendix 2.  Citations for Verbless Clauses . . . . . . . . . . . . . . . 223 Verbless Clauses with Substantival Predicates  223 Verbless Clauses with Prepositional Phrase Predicates  224 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 Indexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 Index of Authors  264 Index of Biblical Sources  270 Index of Syriac Words  273 Index of Greek Words  279 Index of Subjects  283

x

Contents List of Tables

Table 2.1.  Van Coetsem’s Typology of Language Contact . . . . . . . . 17 Table 5.1.  Consonantal Inventory of Koinē Greek . . . . . . . . . . . . . 66 Table 5.2.  Consonantal Inventory of Syriac . . . . . . . . . . . . . . . . 67 Table 5.3.  Representation of Greek Consonants in Syriac . . . . . . . . . 69 Table 5.4.  Diachronic Synopsis of Spellings of παῤῥησία in Syriac . . . . 82 Table 5.5.  Vowel Phonemes of Koinē Greek in the Roman Period . . . . 88 Table 5.6.  Reconstructed Vowel Phonemes of 4th-Century Syriac . . . . 89 Table 5.7.  Representation of Greek Vowels in Syriac . . . . . . . . . . . 91 Table 5.8.  Diachronic Synopsis of Spellings of διαθήκη in Syriac . . . . 92 Table 5.9.  Yaʿqub of Edessa’s Preferred Orthography for Greek Loanwords . . . . . . . . . . . . . . . . . . . . . . 93 Table 7.1.  Frequency of Nisba Adjectives (Excluding Gentilics and Ordinal Numbers) . . . . . . . . . 141 Table 8.1.  Existentials and Copulas in Middle Aramaic, Syriac, and Greek . . . . . . . . . . . . . . . . . . . . . . . 163 Table 8.2.  Distribution of Verbless Clauses with Substantival Predicates . . . . . . . . . . . . . . . . . . . . 170 Table 8.3.  Distribution of Verbless Clauses with Prepositional Phrase Predicates . . . . . . . . . . . . . . . . 172 Table 9.1.  Frequency of Syriac den in Early Syriac Prose . . . . . . . . 186 List of Figures Figure 4.1.  Recto of P. Euphrates 6 . . . . . . . . . . . . . . . . . . . . 51 Figure 8.1.  Existential to Copula . . . . . . . . . . . . . . . . . . . . . 160 List of Graphs Graph 7.1.  Diachronic Frequency of Nisba Adjectives . . . . . . . . . 140 Graph 8.1.  Distribution of Verbless Clauses with Substantival Predicates . . . . . . . . . . . . . . . . . . . . 171 Graph 8.2.  Distribution of Verbless Clauses with Prepositional Phrase Predicates . . . . . . . . . . . . . . . 173

Preface and Acknowledgments I would like to thank a number of people who have helped in various ways with this book: Elitzur Bar-Asher (Hebrew University), Adam Becker (New York University), Monica Blanchard (The Catholic University of America), Claire Bowern (Yale University), Sam Boyd (University of Colorado, Boulder), Sebastian Brock (University of Oxford), Dexter Brown (Yale University), Riccardo Contini (University of Naples ‘L’Orientale’), Edward Cook (The Catholic University of America), Stephen Davis (Yale University), Ashwini Deo (Yale University), Ben Foster (Yale University), Steven Fraade (Yale University), Eckart Frahm (Yale University), Simcha Gross (Yale University), Jean Gascou (University of Paris, Sorbonne), Gene Gragg (The University of Chicago), Lenore Grenoble (The University of Chicago), Sidney Griffith (The Catholic University of America), Andrew Gross (The Catholic University of America), Dimitri Gutas (Yale University), Humphrey “Chip” Hill Hardy II (Southeastern Baptist Theological Seminary), Rebecca Hasselbach (The University of Chicago), Christine Hayes (The University of Chicago), Kristian Heal (Brigham Young University), John Healey (University of Manchester), Theo van den Hout (The University of Chicago), Scott Fitzgerald Johnson (The University of Oklahoma), Ulla Kasten (Yale University), Stephen Kaufman (HUC-JIR, Cincinnati), George Kiraz (Beth Mardutho: Syriac Institute), Christina Kraus (Yale University), Bentley Layton (Yale University), Geoffrey Moseley (Yale University), Hindy Najman (Oxford University), Dennis Pardee (The University of Chicago), Aaron Rubin (Pennsylvania State University), Shawqi Talia (The Catholic University of America), Janet Timbie (The Catholic University of America), Ilya Yakubovich (Philipps—Universität Marburg), Robin Darling Young (The Catholic University of America), and Lev Weitz (The Catholic University of America) – knowing all too well that I have left out a number of names that should be recorded here. I am also grateful to Jim Eisenbraun and his staff at Eisenbrauns for their efficient handling of this book. I would specifically like to thank Beverly McCoy for saving me from numerous errors and for making this a more readable book, and Andy Kerr for the attractive cover. My deepest gratitude goes to Lucas Van Rompay (Duke University), who taught me Syriac and so much more. He has supported this project since its inception, and it would never have been possible without him. Finally, Leah Comeau and, more recently, Mary Pearl Comeau have lived with this book on almost a daily basis. I am forever grateful for their steadfast love and support. xi

Abbreviations General

C Curetonianus ms cent. century chap(s). chapter(s) cod. codex d. died ed. editor/edited by ET English translation ext. exterior f. folio fl. flourished int. interior JBA Jewish Babylonian Aramaic JPA Jewish Palestinian Aramaic ms manuscript n(n). note(s) no(s). number(s) NT New Testament OT Old Testament P Peshiṭta pl(s). plate(s) r. ruled/reigned; recto rev. reverse RL recipient language rt. root S Sinaiticus ms SL source language Syr. Syriac TAM Tense-Aspect-Mood Tg. targum

Reference Works AB ABD

AfO AJP ANRW BibOr BK BO BSOAS

Analecta Bollandiana Anchor Bible Dictionary. Edited by D. N. Freedman et al. 6 vols. New York, 1992 Archiv für Orientforschung American Journal of Philology Aufstieg und Niedergang der römischen Welt Biblica et Orientalia Biberstein-Kazimirski, A. de. Dictionnaire arabe-français. Paris, 1860 Bibliotheca Orientalis Bulletin of the School of Oriental and African Studies

xii

Abbreviations CAD CAL CBM CCSG CDG CPD CPG CRAIBL CSCO CSS DCPA DJA DJBA DJPA DNWSI DSA DULAT ELO ETL FC GEDSH GLBRP GOFS GRBS HALOT HSS Hugoye IOS JA JAOS Jastrow

xiii

Oppenheim, A. L., et al., editors. The Assyrian Dictionary of the Oriental Institute of the University of Chicago. 21 vols. (A–Z). Chicago, 1956–2011 Comprehensive Aramaic Lexicon (CAL), http://cal.huc.edu Chester Beatty Monographs Corpus Christianorum: Series Graeca Leslau, W. Comparative Dictionary of Geʿez (Classical Ethiopic). Wiesbaden, 1987 MacKenzie, D. N. A Concise Pahlavi Dictionary. London, 1971 Geerard, M., editor. Clavis Patrum Graecorum. 6 vols. Thurnhout, 1974–98 [vol. 5: ed. M. Geerard and F. Glorie; vol. 3A: ed. J. Noret] Comptes rendus (des séances) de l’Académie des Inscriptions et Belles-Lettres Corpus Scriptorum Christianorum Orientalium Cistercian Studies Series Sokoloff, M. A Dictionary of Christian Palestinian Aramaic. OLA 234. Louvain, 2014 Sokoloff, M. A Dictionary of Judean Aramaic. Ramat-Gan, 2003 Sokoloff, M. A Dictionary of Jewish Babylonian Aramaic of the Talmudic and Geonic Periods. Ramat-Gan, 2002 Sokoloff, M. A Dictionary of Jewish Palestinian Aramaic of the Byzantine Period. 2nd ed. Ramat-Gan, 2002 Hoftijzer, J., and K. Jongeling. Dictionary of the North-West Semitic Inscriptions. Leiden, 1995 Tal, A. A Dictionary of Samaritan Aramaic. Leiden, 2000 Del Olmo Lete, G., and J. Sanmartín. A Dictionary of the Ugaritic Language in the Alphabetic Tradition. Leiden, 2003 Elementa Linguarum Orientis Ephemerides Theologicae Lovanienses Fontes Christiani Brock, S. P., et al., editors. Gorgias Encyclopedic Dictionary of the Syriac Heritage. Piscataway, NJ, 2011 Sophocles, E. A. Greek Lexicon of the Roman and Byzantine Periods. New York, 1900 Göttinger Orientforschungen, I. Reihe: Syriaca Greek, Roman, and Byzantine Studies Koehler, L., W. Baumgartner, and J. J. Stamm. The Hebrew and Aramaic Lexicon of the Old Testament. Translated and edited under supervision of M. E. J. Richardson. 5 vols. Leiden, 1994–2000 Harvard Semitic Studies Hugoye: Journal of Syriac Studies Israel Oriental Studies Journal Asiatique Journal of the American Oriental Society Jastrow, M. Dictionary of the Targumim, the Talmud Babli and Yerushalmi, and the Midrashic Literature. New York, 1886–1903

xiv JEastCS JECS JEOL JJS JNES JNSL JRS JSS JTS KAI

Abbreviations

Journal of Eastern Christian Studies Journal of Early Christian Studies Jaarbericht . . . Ex Oriente Lux Journal of Jewish Studies Journal of Near Eastern Studies Journal of Northwest Semitic Languages Journal of Roman Studies Journal of Semitic Studies The Journal of Theological Studies Donner, H., and W. Röllig. Kanaanäische und aramäische Inschriften mit einem Beitrag von O. Rössler. Wiesbaden, 1969–73 LCL The Loeb Classical Library LD Lewis, C. T., and C. Short. A Latin Dictionary. Oxford, 1969 LLGE Daris, Sergio. Il lessico Latino nel Greco d’Egitto. 2nd ed. Barcelona, 1991 LS2 Brockelmann, C. Lexicon Syriacum. 2nd ed. Halis Saxonum, 1928 [trans. Sokoloff 2009 = SL] LSAWS Linguistic Studies in Ancient West Semitic LSJ Liddell, H. G., and R. Scott. Greek-English Lexicon. Revised by H. S. Jones and R. McKenzie. Oxford, 1996 LSJ Suppl. Liddell, H. G., and R. Scott. Greek-English Lexicon: Revised Supplement. Revised by H. S. Jones and R. McKenzie. Edited by P. G. W. Glare. Oxford, 1996 LSP Schulthess, F. Lexicon Syropalaestinum. Berolini, 1903 MD Drower, E. S., and R. Macuch. A Mandaic Dictionary. Oxford, 1963 MDOG Mitteilungen der Deutschen Orient-Gesellschaft zu Berlin MPIL Monographs of the Peshiṭta Institute, Leiden NDAE Bedrossian, M. New Dictionary. Armenian-English. Venice, 1875–79 OBO Orbis Biblicus et Orientalis OC Oriens Christianus OCA Orientalia Christiana Analecta OLA Orientalia Lovaniensia Analecta OLD Glare, P. G. W. Oxford Latin Dictionary. Oxford, 1982 Or Orientalia PAT Hillers, D. R., and E. Cussini. Palmyrene Aramaic Texts. Baltimore, 1996 PGL Lampe, G. W. H. A Patristic Greek Lexicon. Oxford, 1961 PLO Porta Linguarum Orientalium PO Patrologia Orientalis PS Patrologia Syriaca PTS Patristische Texte und Studien ROC Revue de l’Orient Chrétien SC Sources chrétiennes SD Beeston, A. F. L., M. A. Ghul, W. W. Müller, and J. Ryckmans. Sabaic Dictionary (English-French-Arabic). Louvain-la-Neuve and Beirut, 1982 Sem Semitica SL Sokoloff, M. A Syriac Lexicon: A Translation from the Latin, Correction, Expansion, and Update of C. Brockelmann’s Lexicon

Abbreviations

STDJ StOr TAD TEG TH TSAJ WO ZA ZAC ZAH ZDMG ZPE ZS

xv

Syriacum. Winona Lake, IN, 2009 [2nd, corrected printing, Winona Lake, IN, 2012] Studies on the Texts of the Desert of Judah Studia Orientalia Porten, B., and A. Yardeni. Textbook of Aramaic Documents from Ancient Egypt. Jerusalem, 1986–93 Traditio Exegetica Graeca Théologie historique Texts and Studies in Ancient Judaism / Texte und Studien zum antiken Judentum Die Welt des Orients Zeitschrift für Assyriologie Zeitschrift für Antikes Christentum / Journal of Ancient Christianity Zeitschrift für Althebräistik Zeitschrift der deutschen morgenländischen Gesellschaft Zeitschrift für Papyrologie und Epigraphik Zeitschrift für Semitistik und verwandte Gebiete

Abbreviations and Symbols in Linguistic Glosses All examples in this book that are larger than one word are provided with word-by-word or morpheme-by-morpheme glosses. Some examples are given in a line in square brackets—especially short examples or examples in footnotes. I have followed the Leipzig Glossing Rules as far as possible. I did find it necessary, however, to introduce a number of categories for the Semitic languages. A full list of abbreviations that occur in linguistic glosses is as follows: 1 2 3

first person second person third person abs status absolutus abstract Aramaic abstract suffix *ūt acc accusative act active adj Aramaic adjectival suffix *āy adv adverb aor aorist art definite article c (stem) Aramaic ʾap̄ʿɛl cnd conditional com comparative con status constructus cont continuous ct (stem) Aramaic ʾɛttap ̄ ʿal d (stem) Aramaic paʿʿɛl dat dative dom direct object marker dt (stem) Aramaic ʾɛṯpaʿʿal

xvi

Abbreviations

emp

status emphaticus existential f feminine g (stem) Aramaic pʿal gen genitive gn geographic name gt (stem) Aramaic ʾɛṯpʿɛl imp imperative ind indicative int interrogative marker m masculine n neuter neg negation nml nominalizer, i.e., the Aramaic particle *ðī (Wertheimer 2001b) nom nominative past past tense pl plural part participle particle particle pass passive pre prefix-conjugation pn personal name quot quotative rel relative sg singular suf suffix-conjugation vblz verbalizer > diachronic development / transfer in language contact → synchronic inflection and/or derivation ex

Abbreviations and Citations of Biblical Books Books of the Bible as well as some other ancient texts are abbreviated according to the SBL Handbook of Style for Biblical Studies and Related Disciplines (2nd ed.; Atlanta, 2014), 124–37. The Syriac Old Testament is cited according to the Leiden edition where it exists, and otherwise, according to the British and Foreign Bible Society’s edition (1905–20). The Syriac Gospels are cited according to Kiraz 1996, with the sigla C referring to the Curetonianus ms, S to the Sinaiticus ms, and P to the Peshiṭta. Other texts of the New Testament are cited according to the British and Foreign Bible Society’s edition (1905–20). The Hebrew Bible is cited according to the 4th edition of Biblia Hebraica Stuttgartensia (BHS). The Greek Septuagint is cited according to A. Rahlfs, Septuaginta (1935). The Greek New Testament is cited according to the 4th rev. edition of the Deutsche Bibelgesellschaft / United Bible Societies, The Greek New Testament (1993).

Transcription / Transliteration The Syriac consonants are transcribed ʾ, b, g, d, h, w, z, ḥ, ṭ, y, k, l, m, n, s, ʿ, p, ṣ, q, r, š, and t. The fricative realization of the nonemphatic voiceless stops b, g, d, k, p, and t are transcribed ḇ, ḡ, ḏ, ḵ, p̄, and ṯ, respectively (i.e., “spirantization” or rukkɔḵɔ). The Syriac vowels are transcribed a, ɔ, ɛ, e, i, o, and u. The symbol ɔ is used instead of the traditional ā (or o in West Syriac) because ā implies a quantitative distinction with a that is not synchronically accurate; historic *ā had become ɔ already in early Syriac (Nöldeke 1904: §11; Boyarin 1978: 149). The symbol e is used for the vowel in the first syllable of rešɔ ‘head’ and kepɔ ‘rock’ in contrast to ɛ in ḥɛlmɔ ‘dream’ (on the history of this e, see Blau 1969). The vowels in the Tiberian sublinear system and the Babylonian supralinear system are transcribed the same as Syriac—that is, a, ɔ, ɛ, e, i, o, and u (Babylonian lacks ɛ). When indicated in the Babylonian system, the reduced vowel shwa is transcribed ə; reduced vowels are not transcribed for the Tiberian system. In addition, matres lectionis are not represented in transcription for either system. Mandaic is transliterated according to the system developed by Macuch in the Mandaic Dictionary (Drower and Macuch 1963) and his Handbook of Classical and Modern Mandaic (1965); note that the matres lectionis ʾ, w, and y are transliterated as a, u, and i, respectively. The only departure from Macuch’s system is the use of e for ʿ following Burtea (2004: 92–93; 2011) and Voigt (2007: 150). With the individual Semitic languages, one of the standard transliteration/ transcription systems is generally followed: Hebrew according to Huehnergard 2002a (with one difference, which is that reduced vowels are transliterated ə, ă, ɔ̆, and ɛ̆); Gəʿəz according to Leslau 1987; and Arabic according to Fox 2003: xvi–xvii. The consonants of Proto-Semitic are reconstructed as follows: *ʾ, *ʿ, *b, *d, *ð, *g, *ɣ, *h, *ḥ, *x, *x ̣ (see Huehnergard 2003), *k, *ḳ, *l, *ɬ, *ɬ̣, *m, *n, t t *p, *r, *s, * s, * ṣ, *t, *ṭ, *θ, *θ, *w, *y, and *dz; and the vowels are *a, *i, *u, *ā, *ī, and *ū. Note that the sibilants *s, *ɬ, and *ts correspond to traditional *š, *ś, and *s, respectively (as well as to s1, s2, and s3; for the reconstruction of the sibilants, see Cantineau 1951; Faber 1981; 1985; Steiner 1977; 1982; Voigt 1979; 1992). xvii

Chapter 1

Introduction Language has a setting. (Sapir 1921: 207)

It is well documented that one of the primary catalysts of intense language contact is the expansion of empire. This is true not only of recent history, as in the many examples of Western European colonization in the Americas, Oceania, India, and Africa, but it is equally applicable to the more remote past. An exemplary case (or better: cases) of language contact in the wake of expanding empires is Aramaic. Aramaic is a member of the Semitic language family and is related to modern languages such as Hebrew, Arabic, and Amharic. 1 It is first attested in written records from the tenth century b.c.e. in Syria and Mesopotamia and has continued to be spoken in this region until the present day. Due to the expansions of empires, Aramaic has throughout its long history been in contact with a variety of languages, including Akkadian, Greek, Arabic, and various dialects of Iranian. These include Akkadian under the Neo-Assyrian (10th–7th cent. b.c.e.) and Neo-Babylonian (7th–6th cent. b.c.e.) Empires; Iranian under the Achaemenid (6th–4th cent. b.c.e.), Parthian (3rd cent. b.c.e.–3rd cent. c.e.), and Sassanian (3rd–7th cent. c.e.) Empires; Greek under the Seleucid (4th–1st cent. b.c.e.) and (Eastern) Roman (1st cent. b.c.e.–7th cent. c.e.) Empires; and Arabic beginning with the Arab conquests in the seventh century and continuing until today. Each of these languages—and so also each of these empires—left its imprint on Aramaic in some way. In this book, I focus on one particular episode in this long history of Aramaic language contact: the Syriac dialect of Aramaic in contact with Greek. Syriac is the best documented dialect of Aramaic. It originated in or around Edessa (Syriac ʾurhɔy), present-day Urfa in southeastern Turkey. From there, it spread, as a language of Christianity, over most of Mesopotamia and Syria reaching as far as Ethiopia, India, and central Asia. Syriac is first attested in non-Christian tomb inscriptions that date from the first to the third centuries. 2 1.  Overviews of the various dialects of Aramaic are available in Beyer 1986; Brock 1989; Fitzmyer 1979b; Gzella 2015 (which appeared too recently to be taken into account in the present book); Kaufman 1992; 1997. 2.  Edited by Drijvers and Healey 1999.

1

2

Chapter 1

The majority of Syriac literature, however, stems from the Christian communities that emerged in Mesopotamia and North Syria by the second century. 3 The “Golden Age” of Syriac spanned from the fourth to the seventh centuries and produced a considerable corpus of original prose and poetry as well as translations from Greek and occasionally Middle Persian. After the Arab conquests in the seventh century, Syriac was gradually replaced by Arabic, though it lived on for several centuries and even witnessed a renaissance in the twelfth and thirteenth centuries. 4 Alongside the numerous Neo-Aramaic dialects, Classical Syriac still functions today as a liturgical and literary language for Syriac Christians both in the Middle East and the worldwide diasporas. 5 Throughout its long history, Syriac has been in contact with an array of different languages. In addition to inheriting words from Akkadian, Sumerian (via Akkadian), and different forms of Iranian, 6 Syriac transferred words from a variety of languages, including Hebrew, Middle Persian, and—later in its history—Arabic. The language that has had the most significant impact on Syriac is, however, Greek. It is widely acknowledged that a prolonged period of contact with Greek resulted not only in a large number of loanwords in Syriac but also in changes to Syriac morphology and syntax. In the preface to his classic treatment of Syriac grammar, for instance, Nöldeke states, “The influence of Greek is shown directly, not merely in the intrusion of many Greek words, but also in the imitation of the Greek use of words, Greek idiom and Greek construction, penetrating to the most delicate tissues of the language (bis ins feinste Geäder der Sprache)” (Nöldeke 1904: xxxii). Although it is widely acknowledged that Syriac was influenced by Greek, the specific contours of this interaction remain unclear. In this book, I present a new analysis of contact-induced changes in Syriac due to Greek. More specifically, I intend to show that Syriac is the outcome of a particular sociolinguistic situation in which inherited Aramaic material was augmented and adapted through contact with Greek. Augmentation refers to the fact that speakers of Syriac added a large number of Greek loanwords to their inherited Aramaic vocabulary. Greek loanwords in Syriac are the subject of chaps. 4–6 below. Adaptation, in contrast, refers to instances in which speakers of Syriac repli3.  Unfortunately, there is no up-to-date history of Syriac literature (so also Van Rompay 2000: §1; 2007a: §9); for now, see Assemani 1719–28; Barsoum 2003; Baumstark 1922; Brock 1997; Macuch 1976; Ortiz de Urbina 1958; Wright 1894. In addition, the recently published Gorgias Encyclopedic Dictionary of the Syriac Heritage (GEDSH) contains entries for most authors of Classical Syriac (Brock et al. 2011). For a valuable bibliography of published Syriac text editions, see Brock apud Muraoka 2005: 144–55. 4.  For this renaissance, see recently Teule and Tauwinkl 2010. 5.  Brock 1989a; Kiraz 2007. 6.  For Akkadian (and Sumerian) loanwords in Aramaic (including Syriac), see Kaufman 1974. For Iranian loanwords in Syriac, see Ciancaglini 2006.

Introduction

3

cated inherited Aramaic material on Greek. This type of change, which will be termed grammatical replication in this study, is the subject of chaps. 7–9. The time frame for the present study extends from the earliest attestations of Syriac at the beginning of the Common Era up to Yaʿqub of Edessa, who died in 708. The Arab conquests in the seventh century (Seleucia-Ctesiphon fell in 637) set into action a number of changes that would dramatically affect the Syriac-speaking population, including its interaction with the Greco-Roman world. 7 These changes, however, took time. In the realm of language use, the Syriac Chronicle of 1234 reports that Greek was not officially replaced by Arabic as the language of civil service until 708. 8 Eventually, the context of Syriac and Greek interaction changed due to the Arab conquests. Given the coincidence of the date of the introduction of Arabic as the language of civil service at least according to the Chronicle of 1234 with the date of his death, Yaʿqub of Edessa provides a convenient end point for this study. 9 This is not, however, to imply that Syriac and Greek did not continue to be in contact past the beginning of the eighth century. In fact, the opposite is known to be true. The later contact between Greek and Syriac can be illustrated by the role that Syriac speakers played in the Greco-Arabic translation movement in the early ʿAbbāsid period (8th–10th cent.). 10 Or, to take even later examples, a number of previously unattested Greek loanwords appear in the poems of two fifteenthcentury authors, Isḥaq Shbadnaya of the Church of the East and Dawid Puniqoyo of the Syriac Orthodox Church. 11 These different historical contexts, however, call for separate studies. This study of contact-induced changes in Syriac due to Greek is both comparative and diachronic. It is comparative in that it locates Syriac within the context of its Late Aramaic sister dialects of Jewish Palestinian Aramaic, Christian Palestinian Aramaic, and Samaritan Aramaic in the Levant; and Jewish Babylonian Aramaic and Mandaic in Mesopotamia. 12 Most previous studies of Greek-Syriac language contact have treated Syriac solely as a language of Near Eastern Christianity and only exceptionally compared it with other dialects of the Aramaic language family. 13 This tendency has obscured the fact that Syriac underwent a number of changes—some of which are due to 7.  For Arab conquests instead of Islamic conquests, see Hoyland 2015: 5–6. 8.  The Syriac text is edited in Chabot 1916: 298.28–299.1. 9.  For a similar cutoff date, see Brock 1996: 253. An additional reason to set 708 as an endpoint is that many Syriac texts from the eighth century and onward have not yet been edited (see also Brock 2010: 124). 10.  In general, see Gutas 1998. 11.  For the former author, see recently Carlson 2011: esp. at p. 200 n. 41 (Greek loanwords); for the latter, see Butts, in GEDSH, 177; and (with more detail) Butts 2009b. 12.  For the importance of this, see Brock 1996: 262. 13.  There are several exceptions, including Healey’s comparison of Greek loanwords in Nabatean and the Old Syriac inscriptions (1995), Taylor’s study of Greek-Aramaic

4

Chapter 1

contact with Greek—that are not found in other dialects of Aramaic, whether earlier or contemporary. Thus, in order to describe accurately contact-induced changes in Syriac, we must situate Syriac within its broader Aramaic linguistic context. 14 This book will show that Syriac as well as Christian Palestinian Aramaic differ from their sister dialects of Late Aramaic due to contact with Greek. This study is also diachronic in that particular attention is paid to changes in the way Syriac interacted with Greek over time. While diachrony has played a role in some studies of Syriac-Greek language contact, especially those by Brock, more work remains to be done in this area. 15 By adopting a diachronic approach, this book will be better able to document the changing degrees of contact that Syriac had with Greek over generations of speakers. There are at least two loci for contact between Syriac and Greek. The first is interactions between Syriac speakers, Greek speakers, and bilingual SyriacGreek speakers. Syriac was the native language of a large portion of the population in Late Antique Syria and Mesopotamia; Greek was the language of empire. Given this situation, many native Syriac speakers learned Greek to one degree or another, with some becoming fully bilingual, whereas others had a more limited knowledge of Greek. 16 In addition, even individuals who had no knowledge of Greek would probably have been exposed to the language to some degree. Ephrem (d. 373), the most well-known Syriac author, for instance, is usually said to have known little or no Greek. 17 Regardless of the validity of this traditional understanding, Ephrem must have at the very least seen written Greek, since the baptistery in the church at Nisibis where he was a deacon contains a Greek building inscription dated to 359/360. 18 This Greek inscription illustrates how far Greek had penetrated into the Syriac-speaking world at a relatively early date. The interactions between Syriac speakers, Greek speakers, and bilingual Syriac-Greek speakers provide one locus for the introduction of contact-induced changes in Syriac due to Greek. A second locus for contact between Greek and Syriac is translation. A small body of Syriac literature was translated into Greek, including the Dialogue on bilingualism in Late Antique Syria and Mesopotamia (2002), and Brock’s comparison of Greek loanwords in Palmyrene and Syriac (2005). 14. As Goshen-Gottstein (1970: 239) stated in even broader terms: “It simply is impossible to ignore the prehistory of Syriac as reflected in the dialectology of Aramaic languages.” 15.  In one of his earliest papers on Greek-Syriac language contact, Brock states, “I have mentioned here only some of the more outstanding features a diachronic study of Greek words in Syriac would throw up; it remains a subject that has been almost completely untouched” (Brock 1975: 90). For Brock’s diachronic work on Syriac-Greek language contact, see Brock 1975; 1982; 1990 [diachronic changes more generally]; 1996; 1999–2000; 2003 [diachronic changes more generally]; 2004; 2010. 16.  This topic is examined in detail in §3.4 below. 17.  See, e.g., Pat-El 2006: 43. For additional references, see below at pp. 199–200. 18.  Bell 1982: 143–45 with pls. 70–83; Canali de Rossi 2004: 39 (no. 62).

Introduction

5

Fate attributed to Bardaiṣan (d. 222), works by Ephrem (d. 373), the Apocalypse of Pseudo-Methodius (written in the late seventh century), the works of Isḥaq of Nineveh (late seventh century), as well as various hagiographical texts. A much larger body of literature was translated from Greek into Syriac from the late fourth to the late ninth century. 19 These translations fall into three broad categories: (1) biblical; (2) patristic, including Basil of Caesarea, Cyril of Alexandria, Eusebius of Caesarea, Evagrius of Pontus, Gregory of Nazianzus, Gregory of Nyssa, Severus of Antioch, and Theodore of Mopsuestia; and (3)  so-called secular, including Aristotle, Galen, Isocrates, Lucian, Plutarch, Porphyry, and Themistius. 20 The translation technique from Greek to Syriac changed from “free” reader-oriented translations to “literal” textoriented translations over time. 21 This culminated in the seventh century with translations in which the lexical and morphological material of Syriac was mapped onto the semantic and grammatical categories of Greek producing what resembles a subtype of mixed language called converted language. 22 The translations from the early ʿAbbasid period (8th–10th cent.), associated above all with Ḥunayn b. Isḥāq (d. 873), returned to more reader-oriented translations. The large number of translations from Greek to Syriac provides a second locus for the introduction of contact-induced changes in Syriac due to Greek. While translation serves as a major locus of change, I do not deal in this book with Syriac texts that were translated from Greek, at least not primarily. Instead, I focus almost exclusively on native Syriac compositions. This is because I am interested in changes to the Syriac language itself, even if some of these changes were mediated by Syriac translations from Greek. Innovative features in a translation may always be the result of the translation process and not changes in a language. Native compositions provide a more reliable indication 19.  For overviews with further references, see Brock, in GEDSH, 180–81; as well as Brock 2007a. 20.  These are often termed “secular” in the secondary literature, but this description is misleading, since these texts were decidedly not secular for many translators and readers but, rather, they were Christian, or at the very least useful for Christians. The famous translator Sergios of Reshʿayna (d. 536), for instance, writes, “Without all this [i.e., Aristotle’s logical works] neither can the meaning of writings on medicine be grasped, nor can the opinions of the philosophers be known, nor indeed the true sense of the divine Scriptures in which the hope of our salvation is revealed” (Brock 1997: 204). Not too long after that, the Cause of the Foundation of the Schools (late 6th/early 7th cent.) describes Plato as one who “taught correctly about God and spoke about his only-begotten Son as the word begotten from him according to nature and about the Holy Spirit as the hypostatic power that proceeds from him” (ed. Scher 1907: 363.12–364.1; ET Becker 2008: 133). Finally, it should not be forgotten that the so-called secular texts were translated and transmitted by deacons, priests, and monks in churches and monasteries. 21.  A number of studies are available on Syriac translation technique; for a general orientation, see the classic study of Brock (1979) and the recent monograph of King (2008). 22.  For converted language, see Bakker 2003: 116–20.

6

Chapter 1

that a language has changed. Thus, in this book, I examine a wide range of native Syriac compositions. I will, however, use Syriac translations from Greek in a few ways throughout this book, including to show how certain changes were mediated by translation literature (see, for example, pp. 128–129) and to establish that Syriac speakers equated a particular feature in their language with a feature in Greek (see, for example, pp. 143–144). Like other studies of ancient language contact, 23 this book does not have access to native speakers and must rely entirely on written documents. In the case of Syriac, written documents represent a standardized literary language. 24 Written Syriac is remarkably uniform with little dialectical variation. In the context of a study analyzing contact-induced changes in Syriac due to Greek, it is interesting to note that texts written in the Roman Empire, where Greek was the language of empire, show only a few differences from those written in the Sassanian Empire, where Greek was much less prominent, though not nonexistent. 25 Written Syriac is not only a standardized literary language, but there is also evidence to suggest that it does not reflect, at least not exactly, the spoken variety (or better: varieties) of the language in Late Antiquity. The orthography of written Syriac, for instance, is extremely conservative, resembling the Aramaic of centuries earlier more than its Late Aramaic sister dialects. 26 The fact that written Syriac is a literary language that does not entirely reflect the spoken language(s) has repercussions for this study, since many of the contact-induced changes in the written literary language would have been mediated by the spoken language(s), which remain(s) inaccessible to the modern researcher. This is especially the case for changes in which the locus of change was contact between speakers, though perhaps less so for changes in which the locus of change was translation. Thus, throughout this work, it must be borne in mind that the object of study is not the everyday spoken language of Syriac speakers in Late Antiquity but their standardized literary language. This study is located at the intersection of two fields: contact linguistics and the study of ancient languages. It is based on the premise that these two fields can and should exist in a mutually beneficial symbiotic relationship. The study of one can and should inform the study of the other, and vice versa. 27 It is worthwhile to treat each aspect of this intersection separately. 23.  Three of the most important of which are Adams 2003 on Latin; Yakubovich 2010 on Luwian; and Smelik 2013 on Hebrew and Aramaic. 24.  For Syriac as a standard language, see the influential study of Van Rompay 1994; as well as the more recent remarks in Taylor 2002: 325. 25.  For one potential difference with Syriac in the Sassanian Empire, see the discussion in n. 9 of chap. 7 below. 26.  Beyer 1966. 27.  Contrast this with Johnson’s claim that “[m]odern scientific approaches to contact linguistics, bilingualism, language acquisition, etc., tend to falter in antiquity on the lack of adequate evidence” (Johnson 2015a: xvi). Such a broad statement, made without citing

Introduction

7

First, contact linguistics can inform the study of ancient languages. Following the publication of Thomason and Kaufman’s classic Language Contact, Creolization, and Genetic Linguistics (1988), the field of contact linguistics has seen a surge in research. 28 This has resulted in an increasingly robust theory of language contact that is, inter alia, better able to correlate socio­historical factors with particular types of contact-induced change. This development is particularly useful for the modern researcher of ancient languages, since it is precisely the concrete sociohistorical background of the speakers that often remains opaque due to the passage of time. Within Syriac studies, for instance, it continues to be debated when Syriac speakers first had intense contact with the Greco-Roman world, with proposals ranging from the turn of the Common Era to the fifth century. 29 Notwithstanding the sparseness of the sometimes conflicting sociohistorical information about this question, there is an abundance of linguistic data for contact-induced changes in Syriac due to Greek. If these data are analyzed from the perspective of contact linguistics, it is possible to illuminate the previously hidden sociohistorical context of ancient Syriac speakers. This question is addressed in detail in the conclusion to this book (§10.3). This is but one way in which the current book employs contact linguistics to inform the study of an ancient language as well as the sociohistorical background of its speakers. The study of ancient languages can also inform the field of contact linguistics. The linguist who studies only modern languages often lacks adequate historical data to outline in detail diachronic changes, including contact-induced changes. In the field of contact linguistics, this has proven to be difficult particularly in discussions of contact-induced changes in (morpho‑)syntax. To put the matter simply, the contact linguistic literature contains far too few cases in which a proposed contact-induced (morpho-)syntactic change has been systematically described with the support of convincing historical data. 30 This topic is also addressed in detail in the conclusion to this book (§10.2). It is here that an ancient language such as Syriac can be put to good use. Syriac boasts an extensive written record spanning more than two millennia, a sizable portion of which can be reliably dated. Written records also survive for five sister dialects of Syriac, in addition to more fragmentary evidence for earlier Aramaic dialects. This considerable body of data often enables the historical linguist to supporting evidence or secondary literature, ignores the growing body of research that aims to use contact linguistics to investigate situations of ancient language contact (for several examples, see n. 23 above). Moreover, this angle of research has not been limited to scholars of ancient languages, but contact linguists have explored it as well (see, e.g., Thomason 2004). 28.  For a recent survey, see Hickey 2010b. 29.  For the former, see, e.g., Drijvers 1992; for the latter, see, e.g., Brock 1982a. 30.  So, recently, Poplack and Levey 2010. This holds true outside (morpho-)syntactic contact-induced change as well. In their typological study of loanverbs, for instance, Wichmann and Wohlgemuth (2008: 113) note that the lack of adequate diachronic data limited the definitiveness of the conclusions that could be drawn.

8

Chapter 1

trace changes, including contact-induced changes, step by step from their prehistory through their completion. The sister dialects of Syriac, in turn, provide an important control for determining whether or not a given change is contact induced. Thus, the study of an ancient language such as Syriac, with its large and diverse written record, can inform the field of contact linguistics as well as historical linguistics more generally. Given its location at the intersection of contact linguistics and the study of ancient languages, this study envisions several audiences. The primary audience is individuals in the field of Syriac studies. In particular, I aim to contribute to the ongoing contextualization of Syriac Christianity—one of the primary exponents of which is the Syriac language—within its Greco-Roman milieu. 31 In addition to Syriac scholars, this book addresses, secondarily, contact linguists and scholars in the field of ancient languages, especially Semitic studies. For scholars in these two fields, the study aims to document in detail various types of contact-induced change over a relatively long period of time. Finally, a word about the organization of this book is in order. It is divided into three parts: Prolegomena (chapters 2–3), Loanwords (chapters 4–6), and Grammatical Replication (chapters 7–9). Part 1 sets the background for the book. Chapter 2 develops the contact linguistic framework, and chapter 3 outlines the sociohistorical context for the Syriac-Greek contact situation. Part 2, which consists of the next three chapters (chapters 4–6), comprises an analysis of Greek loanwords in Syriac. Chapter 4 discusses the methodological framework for the study of loanwords. Chapter 5 analyzes the phonological integration of Greek loanwords in Syriac, while chapter 6 treats their morphosyntactic integration. Part 3, which consists of the next three chapters (chapters 7–9), turns to another category of contact-induced change termed grammatical replication, in which speakers of Syriac created new grammatical structures on the model of structures in Greek. Chapter 7 develops the methodological framework for grammatical replication. Chapter 8 presents a case study of the grammatical replication of the Syriac copula ʾiṯaw(h)y on Greek ἐστίν, and chapter 9 presents a case study of the grammatical replication of the Syriac conjunctive particle den on Greek δέ. Conclusions of the book are presented in chapter 10. 31.  In this way it also adds a different dimension to discussions of the “Hellenization” of Christianity more broadly; for which, see recently Markschies 2012.

Prolegomena ✧ ✧ ✧

Chapter 2

The Contact Linguistic Framework It cannot be doubted that contact-linguistics is badly in need of a general theory. (Van Coetsem 2000: 39)

2.1. Overview The field of contact linguistics lacks a unified—or even consensus—terminology. This is at least partly a reflection of the fact that there is no generally agreed upon theory of language contact. Thus, this chapter aims to lay out the contact linguistic framework, both terminology and theory, that is employed throughout this study. The chapter begins with a brief discussion of terminology (§2.2). It then turns to the various typologies of language contact that have been proposed, looking first at early typologies (§2.3), then at the typology of Van Coetsem (§2.4), and finally at the typology of Thomason and Kaufman (§2.5). In §2.6, these typologies are evaluated, and it is argued that Van Coet­ sem’s typology is the most robust. This chapter, thus, aims to provide the linguistic framework upon which this book is built. 2.2.  Contact Linguistic Terminology Before looking at the various typologies of language contact, I need to say a few words about terminology. There unfortunately continues to be no common or standard terminology in the field of contact linguistics. One example is sufficient to illustrate this lack of a unified terminology: borrowing. This seemingly benign term has been used in a multitude of ways throughout the contact linguistic literature, not to mention beyond it. In The Dictionary of Historical and Comparative Linguistics, for instance, Trask defines borrowing as “[b]roadly, the transfer of linguistic features of any kind from one language to another as the result of contact” (Trask 2000: 44). Thus, borrowing is a cover term for any type of contact-induced change, ranging from loanwords to lexical calques to the transfer of morphosyntax. Borrowing is used in this sense by a number of other scholars. 1 In contrast, Heine and Kuteva restrict borrowing to the 1.  See, e.g., Aikhenvald 1996; 2002; Aikhenvald and Dixon 2001b; 2006; Bloomfield 1933: 444; Campbell 1993; Hafez 1996; Haspelmath 2009; Haugen 1950b; Hetzron 1976: 97; Sihler 1995: 1; Wohlgemuth 2009: 52.

11

12

Chapter 2

transfer of “phonetic substance, that is, either sounds or form-meaning units such as morphemes, words, or larger entities.” 2 In this narrower definition, borrowing is restricted to a subset of the various categories of contact-induced change: it includes only the transfer of “phonetic substance,” whether in the form of a morpheme, a lexeme, or multiple lexemes. 3 Thus, Heine and Kuteva exclude changes such as lexical calques and the transfer of morphosyntax, which Trask’s definition would include. Yet a third definition of borrowing is found in Thomason and Kaufman’s classic Language Contact, Creolization, and Genetic Linguistics (1988), in which the term is said to refer to one of the types of language contact—in the sense of a typology of language contact situations—that involves “the incorporation of foreign features into a group’s native language by speakers of that language” (Thomason and Kaufman 1988: 37; see also p.  21). For them, borrowing is to be contrasted with interference through shift, in which “a group of speakers shifting to a target language fails to learn the target language (TL) perfectly” (1988: 38–39). Thus, interference through shift refers to a situation in which nonnative speakers of the recipient language transfer features of their native language into the recipient language. This contrasts with borrowing, in their system, which occurs when native speakers of the recipient language transfer features from another language into the recipient language. 4 Thomason and Kaufman’s definition of borrowing as one type of contact-induced change has been adopted by other contact linguists as well. 5 While Trask’s and Heine and Kuteva’s definitions of borrowing differ primarily in scope, Thomason and Kaufman use the term in an entirely different way to refer to a particular sociohistorical setting for language contact and the changes associated with it. This example involving three very different—and at times mutually exclusive—definitions of the term borrowing illustrates the importance of defining terminology at the outset of any work on contact linguistics. In this book, the broadest category covering all the ways in which one language influences another is termed contact-induced change. This is used similarly by other contact linguists. 6 Rough equivalents found in the (contact) linguistic literature include “interference,” “borrowing,” “transfer,” “transference,” and “diffusion,” to name only a few. 7 Contact-induced change involves the transfer of a feature from the source language (SL) 2.  Heine and Kuteva 2006: 49. See also 2008: 59; 2010: 86. 3.  A similar definition is found in Ross 2001: 145. 4.  This typology is discussed in greater detail in §§2.5–2.6 below. 5.  See, e.g., Hickey 2010b: 11; Winford 2003: 11–12. 6.  See, e.g., Poplack and Levey 2010; Ross 1987; Thomason 1986; Winford 2005. 7. On “interference,” see, e.g., Ciancaglini 2008; Janda 1976; King 2000; Poplack 1996; Rayfield 1970; Thomason 1986; 2003; 2004; 2007; Thomason and Kaufman 1988; Weinreich 1953. On “borrowing,” see n. 1 above. On “transfer,” see, e.g., Silva-Corvalán 1994: 4; Smits 1996; Van Coetsem 1988; 1990; 1995; 1997; 2000; 2003; Weinreich 1953.

The Contact Linguistic Framework

13

to the recipient language (RL). Feature is a cover term for all types of linguistic material from phonology to morphology to syntax to discourse pragmatics (Stolz 2008). Rough equivalents found in the (contact) linguistic literature include “material” and “element.” 8 Transfer is used similarly by a number of scholars. 9 It is roughly equivalent to Johanson’s “copy.” 10 Alternative pairs for source language~recipient language include “donor~recipient” (Wohlgemuth 2009: 51), “originator~adopter” (Winter 1973), and “model~replica” (Heine and Kuteva 2005; Weinreich 1953: 30– 31; Wohlgemuth 2009: 54). 2.3.  Early Typologies of Language Contact It was already recognized in some early works in the field of contact linguistics that particular linguistic features tend to be transferred in certain linguistic contexts. One of the more influential such observations was that of Windisch (1897), who was a student of H. Schuchardt, a well-known figure in the field of areal linguistics. In his paper entitled “Zur Theorie der Mischsprachen und Lehnwörter” (1897), Windisch expounded the following principle: “Nicht die erlernte fremde Sprache, sondern die eigene Sprache eines Volkes wird unter dem Einfluss der fremden Sprache zur Mischsprache” (Windisch 1897: 104). This principle was intended to account for the fact that bilingual speakers often introduce features of a foreign language into their own language, but they do not typically introduce features of their own language into a foreign language. As an example of this principle, Windisch cites the case of Frederick the Great (1712–86), who introduced French lexemes into his native German but not German lexemes into his French. Based on the examples that he cites, we can observe that Windisch had primarily loanwords in mind when formulating this principle, not other features, such as phonology or syntax. 11 Windisch’s principle, which is sometimes known under the moniker Windisch’s Law, was subsequently adopted by a number of linguists. It was, for instance, included almost verbatim in the Grundriss der romanischen Philologie: “Nicht die erlernte fremde Sprache, sondern die einheimische On “transference,” see, e.g., Clyne 1967; Ross 1985. On “diffusion,” see, e.g., Foley 1986; Gumperz and Wilson 1971; Heath 1978. 8.  For the former, see Wohlgemuth 2009: 51; for the latter, see Weinreich 1953: 7. Van Coetsem (2000: 51) combines them. 9. See, e.g., Gołąb 1959: 8; Silva-Corvalán 1994: 4; Smits 1996; Thomason 2003; 2004; 2010; Thomason and Kaufman 1988; Van Coetsem 1988; 1990; 1995; 1997; 2000; 2003; Weinreich 1953: 1, 7; Wichmann and Wohlgemuth 2008: 89; Winter 1973; Wohlgemuth 2009: 51. 10.  See, e.g., Johanson 2002a; 2002b. See also Stolz 2008. 11.  Sandfeld (1938: 61) later argued that Windisch’s principle only applied to loanwords (see Vildomec 1963: 96). Haugen states the problem in a different way by noting that Windisch’s principle “does not apply to the mature language learner” (Haugen 1950a: 280–81).

14

Chapter 2

Sprache wird unter dem Einflusse der Sprache einer überlegenen Kultur zur Misch­sprache” (Gröber 1904–6: 1.404). Similarly, in his Language: Its Nature, Development and Origin, Jespersen cites Windisch before proceeding to explain: When we try to learn and talk a foreign tongue we do not introduce into it words taken from our own language; our endeavour will always be to speak the other language as purely as possible, and generally we are painfully conscious of every native word that we intrude into phrases framed in the other tongue. But what we thus avoid in speaking a foreign language we very often do in our language. Frederick the Great prided himself on his good French, and in his French writings we do not find a single German word, but whenever he wrote German his sentences were full of French words and phrases. (Jespersen 1922: 208) 12

The principle as well as the example in this quotation are a direct reflection of Windisch’s work. Similar applications of Windisch’s Law can be found elsewhere in the linguistic literature. 13 The influence of Windisch’s principle was not restricted to the field of linguistics in the narrow sense but was also employed in more concrete applications. In a still-important article entitled “Grec biblique” (1938), Vergote invoked Windisch’s formulation in order to explain the contact situation of Greek in Egypt. According to Vergote’s analysis, an Egyptian who spoke Greek would not introduce Egyptian loanwords into Greek, though this person’s native Egyptian language might occasionally influence Greek in other domains: “un bilingue égyptien, écrivant en grec, se gardait bien d’introduire des mots coptes dans cette langue étrangère [= Greek], mais il ne pouvait pas toujours se soustraire à certaines réactions de sa langue maternelle dans le domaine de la sémantique, de la syntaxe et de la phraséologie” (Vergote 1938: 1365). Conversely, Egyptian, more specifically Coptic, does contain a number of Greek loanwords: “l’action du grec sur le copte se manifeste en premier lieu par les mots d’emprunt” (1938: 1365). Thus, by recourse to Windisch’s principle, Vergote was able to explain the fact that Coptic contains numerous Greek loanwords, whereas Greek texts in Egypt contain few Egyptian, including Coptic, words. Vergote also augments Windisch’s principle by noting that Greek texts in Egypt do exhibit influence from Egyptian in other linguistic domains, such as semantics, syntax, and phraseology. This marks an important advancement in constructing a typology of contact-induced change, calling to mind later developments such as Van Coetsem’s imposition and Thomason and Kaufman’s interference through shift and imperfect learning. 14 12.  This same paragraph is repeated almost verbatim in Jespersen’s Growth and the Structure of the English Language (1948: §37). 13.  See, e.g., Dillon 1945: 13; Flom 1905: 425. 14.  These are discussed in §2.4 and §2.5, respectively.

The Contact Linguistic Framework

15

The distinction observed by Vergote, building on Windisch, was further abstracted and systematized by Vildomec (1963: 80–86). Citing Windisch 1897, Vildomec notes that “[t]here often is tendency for a multilingual to use words of an Le [= foreign language] when he speaks (or writes) his Lm  [= mother language], but this tendency does not always operate with the same intensity in the opposite direction” (1963: 80). Thus, loanwords are not typically introduced from the “mother language” into the “foreign language”: “If an educated adult has to use an Le [= foreign language] he will usually try to handle it as well as possible; he is unlikely, therefore, to use many words of his Lm [= mother language] when talking the Le [= foreign language]” (Vildomec 1963: 81). What is transferred in this situation, however, is “accent” (1963: 82–84) and especially syntax (1963: 84). Thus, Vildomec distinguishes two broad types of language contact: (1) transfer from “foreign language” to “mother language” that involves primarily lexical features; (2) transfer from “mother language” to “foreign language” that involves primarily “accent” and syntactic features. This distinction is roughly equivalent to what would later be established in the work of Van Coetsem and of Thomason and Kaufman. The same binary noted by Vildomec was observed by several other scholars prior to the late 1980s. In a study of Eastern-European Jewish immigrants in the U.S.A., for instance, Rayfield noted that, when the primary language was the recipient language and the secondary language was the source language, it was lexical material that was typically transferred with the transfer of “structural” and “phonic” material being less prominent (Rayfield 1970: 103–6). 15 When the situation was reversed, however, and the primary language was the source language and the secondary language was the recipient language, the types of change encountered were also reversed—that is, “structural” and especially “phonic” material were transferred, and the transfer of lexical material was less prominent. He attributes this distribution to the fact that “[t]he bilingual retains most persistently the earliest learned systems of his primary language” (Rayfield 1970: 106), arguing that the systems are learned in the following order: phonology, syntax, morphology, and lexicon (1970: 103). This last argument marks an early forerunner of Van Coetsem’s stability gradient of language, which will be discussed in §2.4 below. Prior to the late 1980s, then, a number of scholars noticed that loanwords tended to be transferred in certain contact situations, whereas phonology, syntax, and (rarely) morphology were transferred in others. 16 Notwithstanding these developments, a comprehensive typology of language contact was not 15.  Rayfield never defines primary language and secondary language, and thus it is unclear whether these refer to language proficiency or native~foreign language. This is an important distinction, as will become clear in the discussion of Thomason and Kaufman’s native language versus Van Coetsem’s linguistic dominance (see below, §2.6). 16.  See also Bátori 1979; Janda 1976: 590; Lado 1957: 2; Winter 1973: 145–46.

16

Chapter 2

formulated until 1988 with the publication of two important monographs, one by Van Coetsem and the other by Thomason and Kaufman. These two works commenced a discussion on the typology of language contact that has not yet abated now, two and a half decades later. 17 Though the typologies proposed by Van Coetsem and by Thomason and Kaufman share a number of similarities, we will find it helpful to review each of them individually. 2.4.  The Typology of Van Coetsem Van Coetsem first proposed a typology of language contact in his Loan Phonology and the Two Transfer Types in Language Contact (1988). The basic typology that Van Coetsem espouses in this monograph is unfortunately at times obscured by his complicated argumentation as well as by the many tangential discussions accompanying it. This led Van Coetsem to outline a more concise version of his typology in 1995 in an article that serves as a précis of his earlier monograph. Van Coetsem then revisited the typology of language contact in his General and Unified Theory of the Transmission Process in Language (2000) along with its accompanying summary article (1997). These later two works do not in general depart from his earlier work but, rather, provide an updated, more integrated analysis of the earlier typology. Finally, in 2003, a lengthy article by Van Coetsem was published posthumously in which he treated a variety of issues related to language contact, including some further comments on his typology. In addition to his own work, brief overviews of Van Coetsem’s typology of language contact are available by others. 18 According to Van Coetsem, language contact can be divided into three basic types. First, there is borrowing or recipient language agentivity. 19 Borrowing occurs when the agents of change are dominant speakers of the recipient language. In cases of borrowing, it is a less stable domain of language, such as a lexical item, that is transferred from the source language to the recipient language. To illustrate borrowing, Van Coetsem (2000: 53) refers to the case of a native speaker of French who incorporates an English lexeme into his language. The second broad category of language contact in Van Coetsem’s typology is imposition or source language agentivity. 20 Imposition occurs when the agents of change are dominant speakers of the source language. In imposition, a more stable domain of language, such as a grammatical or phonological 17.  For subsequent work, see especially Guy 1990; Haspelmath 2009: 50–51; Hickey 2010b; Ross 1991; Smits 1996: 29–58; 1998; Thomason 2001: 59–98; 2003: 691–93; Van Coetsem 1990; 1995; 1997; 2000; 2003; Winford 2003: 11–24; 2005; 2009: 283–85. 18.  See particularly Smits 1996: 30–33; 1998: 378–80; Winford 2005: 376–82; 2007: 25–28; 2009: 283–85. 19.  Van Coetsem 1988: 10–11; 1995: 77–80; 1997: 358–59; 2000: 53, 67–73, 137–66. 20.  Van Coetsem 1988: 9–10, 47–76; 1995: 73–77; 1997: 358–59; 2000: 53–54, 73–82, 167–212.

17

The Contact Linguistic Framework

feature is transferred from the source language to the recipient language. Imposition is usually associated with second-language acquisition, though this is not necessarily the case. To illustrate imposition, Van Coetsem (2000: 53–55) refers to the case of a speaker of French who learns English and transfers some articulation habits to English, such as pronouncing the p of the English word pear without aspiration. These two types of language contact are summarized in table 2.1, where A and B refer to different languages, underscoring indicates linguistic dominance, and the > symbol indicates transfer in contact-induced change. 21 Table 2.1.  Van Coetsem’s Typology of Language Contact A is linguistically dominant B>A

A>B

B is linguistically dominant A>B

B>A

Agentivity

recipient lang. source lang.

recipient lang. source lang.

Domain of transfer

less stable

more stable

less stable

more stable

Type of transfer

borrowing

imposition

borrowing

imposition

In any contact situation involving two languages, there are then four basic forms of interaction (Van Coetsem 2000: 54–55): borrowing from B into A, imposition of A into B, borrowing of A into B, and imposition of B into A. Van Coetsem’s distinction between borrowing and imposition is based on what he terms linguistic dominance. 22 Linguistic dominance refers to the greater proficiency that a speaker has in one language as compared with another language. Van Coetsem’s concept of linguistic dominance derives from Weinreich’s claim that “[a] bilingual’s relative proficiency in two languages is easily measured . . . . one of the languages can hence be designated as dominant by virtue of the speaker’s greater proficiency in it” (Weinreich 1953: 75). In many cases, a speaker will be linguistically dominant in her native language or first acquired language. Van Coetsem (1988: 15), however, notes that this is not necessarily the case. 23 This means that an individual’s linguistic dominance can change over time. 24 Van Coetsem illustrates the change in linguistic dominance with the example of an immigrant in the United States whose native language is not English but who over time becomes more fluent in English than in her native language. 25 The fact that an individual’s linguistic 21.  This summary is based on one of the additional synthesizing diagrams at the end of Van Coetsem 2000. 22.  Van Coetsem 1988: 13–17; 1995: 70–72; 1997: 358; 2000: 32, 42, 49, 58–62, 66– 67. See also Smits 1996: 30–31. 23.  So already Weinreich 1953: 75 n. 1. 24.  Van Coetsem 1988: 16–17, 76, 85; 1995: 70–71, 81; 1997: 359; 2000: 52, 81. 25.  Van Coetsem 1988: 16; 1995: 70–71, 81; 2000: 52, 81. See also Weinreich 1953: 76.

18

Chapter 2

dominance can change over time has important implications for the analysis of cases of language attrition, as will become clear below when comparing Van Coetsem’s typology with that of Thomason and Kaufman (see §2.5). At this point, however, it is important to note that in Van Coetsem’s framework linguistic dominance is not necessarily the same as native language (or first language). In addition to a diachronic change in linguistic dominance, it is also possible for a speaker’s linguistic dominance to change according to register or context (Van Coetsem 1988: 16–17). That is, a speaker can be linguistically dominant in one language in one context, but linguistically dominant in another language in another context (Van Coetsem 2000: 84). To illustrate this change in linguistic dominance, Van Coetsem (2000: 84) refers to Weinreich’s example, in which “[a] child learning both languages in its familial and play environment . . . may be equipped to deal with everyday things in both tongues; but if it studies certain subjects in a unilingual school, it will have difficulty in discussing these ‘learned’ topics in the other language” (Weinreich 1953: 81). Finally, it should be noted that linguistic dominance is to be distinguished from social dominance (Van Coetsem 1988: 13; 2000: 57), which refers to the political or social status of one of the languages. Alongside linguistic dominance, the other major factor that leads to the different effects between borrowing and imposition is what Van Coetsem calls the stability gradient of language. 26 The stability gradient of language refers to the fact that certain features of language, such as phonological and grammatical features, are more stable and resistant to change than others, such as lexical items, especially contentive words. The concept of the stability gradient of language has long been recognized in the study of language contact, even if it has not always been termed as such. Already in the late nineteenth century, Whitney (1881: 19–20) proposed a hierarchy of borrowing according to which nouns were transferred before adjectives, adjectives before verbs, verbs before adverbs, adverbs before prepositions and conjunctions, and so forth. Similar hierarchies have been proposed by a number of other scholars. 27 Though the concept of the stability gradient of language has long been recognized, the exact ranking of each feature of language remains controversial, as Van Coetsem recognizes (1988: 34; 1995: 67–68). Despite the lack of a generally accepted ranking of features, the stability gradient of language has important implications for which features of language will be transferred in the different types of language contact. This is because in any given contact situation the stable features of the dominant language will tend to be retained. If the agent of change is linguistically dominant in the source language, then the more 26.  Van Coetsem 1988: 25–34; 1995: 67–70; 1997: 358; 2000: 31–32, 50, 58–62, 105– 34. See also Smits 1996: 31–32. 27.  For citations and discussion, see Campbell 1993: 100; Matras 2010: 76–82; Wohlgemuth 2009: 11–17.

The Contact Linguistic Framework

19

stable elements of the source language, such as phonological and grammatical features, will be preserved and so transferred to the recipient language. This explains why source language agentivity, or imposition, results primarily in the transfer of phonological and grammatical features. In contrast, if the agent of change is linguistically dominant in the recipient language, then the more stable elements of the recipient language will be preserved while the less stable elements, such as lexical items, will be transferred from the source language to the recipient language. This explains why recipient language agentivity, or borrowing, results primarily in the transfer of lexical items. Alongside borrowing and imposition, Van Coetsem recognizes a third type of transfer called neutralization. 28 Neutralization occurs when an individual is equally dominant in two languages. In neutralization, the distinctions between recipient language agentivity and source language agentivity are no longer relevant, and any feature can be transferred. Van Coetsem uses this third category to explain contact situations such as Media Lengua, 29 Mednyj Aleut (also called Copper Island Aleut), 30 Michif, 31 and Ma’a, 32 which are more often termed “mixed languages” in the linguistic literature. 33 In his earlier work (1988: 87–91; 1995: 81), Van Coetsem limits the transfer types to borrowing and imposition and argues that neutralization is the state that occurs when the distinction between these two types is no longer clear. In his later work (1997; especially 2000), however, Van Coetsem follows Buccini (1992: 329–32) in recognizing neutralization as a third transfer type, with the caveat that it is of a different order from borrowing and imposition (Van Coetsem 2000: 43). He illustrates this difference between borrowing and imposition, on the one hand, and neutralization, on the other hand, by invoking the image of a triptych with the two outer panels corresponding to borrowing and imposition and the central panel representing neutralization (Van Coetsem 1997: 360; 2000: 42). To summarize, then, for Van Coetsem, borrowing occurs in situations of recipient language agentivity and results in the transfer of the less stable domains of language, such as lexical items. Imposition, in contrast, occurs in situations of source language agentivity and results in the transfer of the more stable domains of language, such as phonological and grammatical features. Finally, neutralization occurs when the distinction between recipient language agentivity and source language agentivity is no longer relevant—that is, it 28.  Van Coetsem 1988: 87–91; 1995: 81; 1997: 359–66; 2000: 82–99, 239–80. 29.  Muysken 1981; 1994; 1997. 30.  Golovko 1994; 1996; Golovko and Vakhtin 1990; Thomason 1997d; Thomason and Kaufman 1988: 233–38. 31.  Bakker 1994; 1997; Bakker and Papen 1997; Thomason and Kaufman 1988: 228–33. 32.  Mous 1994; 2001; 2003; Thomason 1997e. 33.  See, e.g., Bakker and Mous 1994; Matras and Bakker 2003; Thomason and Kaufman 1988: 223–28.

20

Chapter 2

is neutralized. In situations of neutralization, any feature can be transferred. These three types of transfer are based on the linguistic dominance of the agents of change. The different linguistic effects of each of these three types of transfer are determined by the interplay between the linguistic dominance of the agents of change and the stability gradient of language. 2.5.  The Typology of Thomason and Kaufman In the same year that Van Coetsem published his Loan Phonology and the Two Transfer Types in Language Contact, Thomason and Kaufman published their Language Contact, Creolization, and Genetic Linguistics (1988). In this influential book, they proposed a typology of contact-induced change that distinguishes two primary types. 34 First, there is borrowing, which involves “the incorporation of foreign features into a group’s native language by speakers of that language” (Thomason and Kaufman 1988: 37). In this case, the native language is maintained—that is, there is language maintenance. The primary effect of borrowing is the transfer of lexemes (loanwords), though in cases of “strong long-term cultural pressure” anything can be transferred, including phonology, syntax, and even morphology. While the borrowing of vocabulary can occur quickly, longer periods of intense contact are needed for the borrowing of phonology, syntax, and morphology. To capture this continuum, Thomason and Kaufman propose a Borrowing Scale that on one extreme has casual contact involving loanwords only and on the other extreme has heavy structural borrowing in situations of very strong cultural pressure (Thomason and Kaufman 1988: 74–109; see also Thomason 2001: 69–71). Borrowing, then, occurs in situations in which native speakers maintain their language, and it is primarily associated with the transfer of lexemes, though structure can be transferred as well, especially in situations of longer, more-intense contact. The second type of language contact for Thomason and Kaufman is interference through shift, which is defined as a type of language contact “that results from imperfect group learning during a process of language shift. That is, . . . a group of speakers shifting to a target language fails to learn the target language (TL) perfectly. The errors made by members of the shifting group in speaking the TL then spread to the TL as a whole when they are imitated by the original speakers of that language” (Thomason and Kaufman 1988: 38–39). The primary effect of interference through shift is the transfer of phonology and syntax as well as occasionally morphology. These changes can take place in a relatively short period of time—in fact, as little as a generation. Interference through shift, then, occurs during cases of language shift when shifting speakers have imperfect knowledge of the recipient language, and it is associ34.  The foundation for this typology was already laid in Thomason 1986: 265–74, where a distinction was made between borrowing and substratum interference, or more fully, “interference that results from imperfect group learning during language shift.”

The Contact Linguistic Framework

21

ated primarily with the transfer of structure, such as phonology, syntax, and occasionally morphology. The key variables for Thomason and Kaufman, then, are native language and maintenance versus shift. In cases of language shift, nonnative speakers of the recipient language transfer features of their native language (= source language) into the recipient language. This is interference through shift. In cases of language maintenance, native speakers of the recipient language transfer features from another language (= source language) into the recipient language. This is borrowing. Finally, it should be noted that Thomason and Kaufman’s typology only includes cases of what they call “normal transmission,” excluding mixed languages, pidgins, and creoles, which they argue are the result of different processes. 35 2.6. Synthesis The typology of Van Coetsem and that of Thomason and Kaufman share a number of similarities. In a review of Thomason and Kaufman’s monograph, Van Coetsem notes that their typology “basically agrees” with the one that he espouses (Van Coetsem 1990: 261). Similarly, Thomason (2003: 691; see also 2001: 95) observes that Van Coetsem argues for “a nearly identical distinction” to that proposed in her joint work with Kaufman. The similarities between the typologies of Van Coetsem and of Thomason and Kaufman are also recognized by others working in the field of contact linguistics. Guy (1990) and subsequently Ross (1991), for instance, attempt to combine the two proposals into a unified typology of language contact. Applying the typology of language contact to a practical problem, Louden (2000) also combines Van Coetsem’s proposal and Thomason and Kaufman’s into a single typology. Notwithstanding their many similarities, however, there is a fundamental difference between the typology of Van Coetsem and that of Thomason and Kaufman. Van Coetsem’s typology is based on the variable of linguistic dominance, whereas Thomason and Kaufman’s is based on the variables of native language and of language maintenance versus shift. The difference between the two typologies is most apparent in cases of language attrition. Consider, for instance, the case discussed by Ross (1991; with more details in 1987) in which the interclausal syntax of the Bel Group of Austronesian languages corresponds, not to languages belonging to the same language family, but to the unrelated Papuan languages with which they are in contact. Ross describes the sociolinguistic situation as one in which the socially dominant Papuan speakers did not learn the Bel languages, whereas the socially subordinate speakers of the Bel languages often learned Papuan. 35.  For an important critique of this false dichotomy between normal transmission and whatever its presumable counterpart would be (ab-normal transmission?!?), see Mufwene 2001.

22

Chapter 2

Thomason and Kaufman would classify the changes in the Bel languages as borrowing since they are part of a situation of language maintenance involving native speakers. That is, the Bel languages continue to be spoken by the population. The difficulty, however, is that the contact-induced changes that occur are more in line with their shift-induced interference: systematic changes in syntax. It is in fact this difficulty that prompts Ross to classify this as an instance of imposition within the typology of Van Coetsem (in conversation with Guy 1990). What is important in this case, as Ross notes, is that native speakers of the Bel languages have become linguistically dominant in Papuan. In Ross’s words, speakers of the Bel languages “were already more at home in the Papuan language than in their inherited Austronesian language” (Ross 1991: 122). Thus, speakers of the Bel languages are more dominant in Papuan than in their native language. In Van Coetsem’s typology, then, this is a case of imposition, in which native speakers of the Bel languages have become linguistically dominant in Papuan. In Thomason and Kaufman’s typology, however, the changes in the Bel languages must be analyzed as borrowing, since the Bel languages are maintained and since the changes involve “the incorporation of foreign features into a group’s native language by speakers of that language” (Thomason and Kaufman 1988: 37)—their very definition of borrowing. For Thomason and Kaufman, changes associated with interference through shift would only apply to changes in the Papuan language due to shifting native speakers of the Bel languages. Thus, this case illustrates that the important variable in the typology of contact-induced change is the linguistic dominance of the agents of change (in this case, Papuan) and that the variables of language maintenance versus shift (in this case, language maintenance) and of native language (in this case, the Bel languages) are not viable indicators of the type of change to be expected. 36 More broadly, Thomason and Kaufman’s typology does not provide an economic account of cases of language attrition. 37 In Thomason and Kaufman’s typology, language attrition is to be classified as borrowing, since it occurs technically in situations of language maintenance, and the agents of change are native speakers of the recipient language. Cases of language attrition, however, usually witness systematic changes in phonology and/or syntax, which are more in line with their category of shift-induced interference that occurs in situations of language shift and that involves agents of change who are nonnative speakers. Van Coetsem’s typology, in contrast, does not face the same difficulty, since cases of language attrition are classified as imposition with the expected outcome. Van Coetsem’s imposition, then, is wider than Thomason and Kaufman’s interference through shift, whereas Van Coetsem’s borrowing 36.  See also Ross 1991: 125–26. 37.  So also Smits 1996: 52–54, where similar cases are discussed, including Iowa Dutch and Asia Minor Greek.

The Contact Linguistic Framework

23

is narrower than Thomason and Kaufman’s borrowing. 38 Crucially, Van Coetsem’s typology is able to account for the fact that language shift and language attrition both involve linguistic dominance of the source language and that, consequently, language shift and language attrition both witness the same effects in the recipient language: systematic changes primarily in phonology and syntax as well as more rarely in morphology. As Smits notes, “An important similarity between acquiring a language and ‘losing’ a language is that in both cases the recipient language is the linguistically non-dominant language. That is, in both cases knowledge of the recipient language is imperfect” (Smits 1996: 33). The inability of the typology of Thomason and Kaufman to account for cases of language attrition is indicative of a deeper theoretical problem. The crucial variable for a typology of language contact is not that of native language nor that of maintenance versus shift, but rather, it is that of linguistic dominance. Thus, the typology of Van Coetsem, with its basis on the variable of linguistic dominance, is a more robust typology of language contact. 39 In her more recent work, Thomason points out that the typology in Thomason and Kaufman 1988 needs to be revised since the crucial variable is not whether or not shift takes place but whether or not there is imperfect learning. 40 Thomason’s variable of imperfect learning is a close inverse counterpart to Van Coetsem’s linguistic dominance. Thus, Thomason’s revised typology of contact-induced change closely approximates that of Van Coetsem. In Thomason’s revised typology, borrowing occurs when “the agents of change are fully fluent in the receiving language” and “imperfect learning plays no role in the transfer process” (Thomason 2004: 7). This is similar to Van Coet­ sem’s borrowing, in which the agents of change are dominant speakers of the recipient language—that is, there is not imperfect learning in Thomason’s revised framework. Thomason’s second type of contact-induced change is shift-induced interference, which occurs in situations of imperfect learning. In contrast to her earlier views (see §2.5), her revised shift-induced interference does not necessarily involve language shift: “When imperfect learning enters the picture, I call the process shift-induced interference, though sometimes there is no actual shift of one population to another group’s language because the L2 learners maintain their original L1 for in-group usage” (Thomason 2004: 7). This is similar to Van Coetsem’s imposition, in which the agents of change are dominant speakers of the source language—that is, there is imperfect learning of the recipient language in Thomason’s revised framework. Thomason’s revised category of shift-induced interference without actual shift is able to capture situations of language attrition, such as that involving the Bel Group of Austronesian languages discussed above (pp.  21–22). Thus, in 38.  Louden 2000: 95; Smits 1996: 52–53. 39. Similarly Smits 1996: 52–58; Winford 2005; 2007. 40.  Thomason 2001: 59–98; 2003: 691–93; 2004: 7.

24

Chapter 2

adopting imperfect learning as the key variable, Thomason’s revised typology is very similar to Van Coetsem’s typology, which adopts linguistic dominance as the key variable. 41 2.7. Conclusion In this chapter, I have focused on establishing the contact linguistic framework for the study of contact-induced changes in Syriac due to Greek. After a brief discussion of terminology (§2.2), I turned to the various typologies of language contact (§§2.3–2.5). I argued that the typology of Van Coetsem, with its basis on the variable of linguistic dominance, is the most robust. Thus, following Van Coetsem, I adopt in this book a threefold typology of contactinduced change. First, there is borrowing (recipient language agentivity) in which the agents of change are dominant speakers of the recipient language. Borrowing results primarily in the transfer of lexemes. The second broad category of language contact is imposition (source language agentivity), in which the agents of change are dominant speakers of the source language. Imposition results primarily in the transfer of phonology, syntax, and to a more limited extent morphology. Finally, there is neutralization, in which an individual is equally dominant in two languages. Any feature can be transferred in neutralization. With the contact linguistic framework established, I now turn to the sociohistorical context of contact-induced changes in Syriac due to Greek. 41.  It should be noted that this revised typology has not been consistently implemented in Thomason’s more recent work. In a 2003 paper, for instance, Thomason classifies a case in which a native speaker of Italian began to have an “American accent” in her Italian after spending 12 years in the United States as borrowing. This case should, however, be classified as imposition—or in Thomason’s revised framework, as shift-induced interference without actual shift—since the native Italian speaker arguably became linguistically dominant in the source language, English.

Chapter 3

The Sociohistorical Setting The linguist who makes theories about language influence but neglects to account for the socio-cultural setting of the language contact leaves his study suspended, as it were, in mid-air. (Weinreich 1953: 4)

3.1. Overview In chap. 2, I laid out the contact linguistic framework for this book. In this chapter, I continue to provide background by outlining the sociohistorical context for the contact-induced changes that will be the subject of the study. I begin with a brief historical narrative of Syria and Mesopotamia from the beginning of the Seleucid Empire up to the Roman Empire (§3.2). I then turn to the topic of Greco-Roman influence on early Syriac-speaking culture (§3.3). In the final and longest section of the chapter, I investigate language use among the inhabitants of Late Antique Syria and Mesopotamia with an eye toward establishing how contact-induced changes in Syriac due to Greek are to be classified within the typology of Van Coetsem (§3.4). 3.2.  The Conquests of Alexander In November of 333 b.c.e., Alexander the Great defeated the Persian army led by Darius III at the plain of Issus in northwest Syria. Two years later, Alexander again defeated Darius III, this time in Gaugamela, east of the Tigris near Arbela (modern Erbil, Iraq). The outcome of these battles set into motion a number of changes that would affect the entire Near East. 1 It marked the beginning of the end of the Achaemenid Empire. 2 It also ushered in the Seleucid Empire, which would control Syria and Mesopotamia for the next two centuries. 3 With the Seleucid Empire came the foundation of Hellenistic cities throughout Syria 1. See Briant 1979; 1999. 2.  For an excellent history of the Achaemenid Empire, see Briant 2002. 3.  For Hellenistic Syria and Mesopotamia, see Millar 1987; Sartre 2001: 60–63; 2005: 5–12.

25

26

Chapter 3

and Mesopotamia. 4 In the case of Edessa, which would eventually become the geographic center of Syriac-speaking culture, Seleucus I Nicator transformed the older settlement of ʾUrhɔy (earlier Adme) into a Greek polis in 303/2 b.c.e. and gave it the name of the ancient Macedonian capital. 5 Hellenistic cities were also established at Antioch, Apamea, Ḥarran (Carrhae), Nisibis, Reshʿayna, SeleuciaCtesiphon, and Singara, all of which were Aramaic speaking at the time and would be at least partially Syriac speaking in Late Antiquity. The Seleucids also brought the Greek language to Syria and Mesopotamia as the language of empire. 6 Even though it never fully supplanted Aramaic in Syria and Mesopotamia, Greek became a well-established language of communication and commerce throughout the area. Already by the last quarter of the fourth century b.c.e., then, the Aramaic-speaking population of Syria and Mesopotamia came into contact with the Seleucid Empire and its Greek language. Greek influence in the Near East became even more pronounced with the Roman conquest of the area. 7 The second century b.c.e. witnessed the partial disintegration of the Seleucid Empire. By 133 b.c.e., the region of Osrhoene and its important center of Edessa were ruled by the Abgarid Dynasty. 8 The area survived more or less as an independent state between the Roman Empire in the West and the Parthian/Sassanian Empire in the East until the middle of the third century. By the beginning of the second century c.e., however, Rome began to play a more prominent role in the area. This reached a climax with the Abgarid ruler Abgar VIII (r. 177–212), who maintained close ties with the Roman Empire and was even granted Roman citizenship. Following the death of Abgar VIII and the short reign of his successor, Edessa was declared a Roman colonia in 212/213. Though the Abgarid Dynasty was briefly restored in 239, Rome was again in power by 242. The (eastern) Roman Empire would continue to control Syria and Mesopotamia up to the Arab conquests in the seventh century (Seleucia-Ctesiphon fell in 637). 9 Prior to the establishment of Roman control of Edessa, Greek was the language of international communication and commerce throughout the Seleucid Empire. The Roman Empire did not significantly alter this. 10 In general, the Roman Empire did not force the Greek-speaking provinces to adopt Latin. 4. See Briant 1978; Grainger 1990. 5.  For the connection of Edessa with cuneiform Adme, see Harrak 1992. 6.  It should be noted that Greek was present in Syria and Mesopotamia before the Seleucid Empire, as can be established by the existence of Greek loanwords in Aramaic beginning already in the mid-first millennium b.c.e. (see §4.9). 7.  For the Roman Near East in general, see Millar 1993; Sartre 2001; 2005. 8.  In general, see Millar 1993: 457–67, 472–81; Ross 2001; Segal 1970: 1–61. 9.  Unfortunately, there continues to be no comprehensive study of the history of Syria and Mesopotamia during the Late Antique period. For Edessa during this period, see Segal 1970: 110–216. For the Late Antique world more broadly, see the excellent overviews in Brown 1989; and Cameron 2012. 10.  For the following, see Rochette 2010: 289–90.

The Sociohistorical Setting

27

Rather, Greek remained the official language of empire in the eastern provinces. Latin had a more restricted use, being employed primarily in military matters. The use of Greek and Latin in a Roman city in Mesopotamia from the first centuries of the Common Era can be illustrated by the more than 150 documents discovered at Dura-Europos, an important military outpost on the Euphrates until its destruction in 256 c.e. 11 A majority of the documents from this site are written in Greek or Latin. The documents from the archives of the Cohors Vicesima Palmyrenorum (a Roman military troop) are primarily in Latin (P. Dura 54–150). All of the texts associated with official military business are in Latin, including reports (P. Dura 82–97) and rolls and rosters (P. Dura 98–124). The famous Feriale Duranum, which is a calendar of official religious observances, is also in Latin (P. Dura 54). Correspondence by military officials is primarily in Latin, though some is in Greek (P. Dura 55–81). Finally, judicial business and receipts from the archives of the Cohors Vicesima Palmyrenorum are primarily in Greek, though a few are in Latin (P. Dura 125–29). In contrast to the predominance of Latin in the archive of the Cohors Vicesima Palmyrenorum, a vast majority of the texts found outside this archive are in Greek (P. Dura 1–52). Thus, all of the texts from the registry office are in Greek (P. Dura 15–44), which include individual documents, such as a gift, loans, deeds of sale, deposits, a marriage contract, and divorce contracts. The lists and accounts are also in Greek (P. Dura 47–53), as are the texts associated with civil administration (P. Dura 12–14). Two letters are also in Greek, one of which may be an official letter (P. Dura 45), and the other of which is from a soldier (P. Dura 46). The documents from Dura-Europos, thus, illustrate the degree to which Latin was restricted to the military and, even then, to official military matters. Greek, in contrast, was used by the military in some correspondence, as well as in day-to-day legal matters. Outside the military, Greek was the official language for the majority of tasks. Thus, in Syria and Mesopotamia, Greek would have been the language of the Roman Empire, with Latin restricted more or less to official military matters. 12 3.3.  ʾUrhɔy Is Edessa The Syriac-speaking culture that comes into view in the first centuries of the Common Era is one that had been in contact with the Greco-Roman world for centuries. The effects of these centuries of contact can be seen in various places. The art and architecture of the region, for instance, reflect significant GrecoRoman influence. 13 This is perhaps most clear in the mosaics from the region 11.  All the texts are edited in Welles, Fink, and Gilliam 1959. For language use at DuraEuropos, see Gascou 2011. 12.  For the respective domains of Latin and Greek in the Roman Near East, see also the concise remarks in Millar 2013a: 19, 28. 13. See Possekel 1999: 28; and especially Mango 1982. Images of many of the relevant objects are available in the plates section in Segal 1970.

28

Chapter 3

around Edessa. 14 A recently discovered mosaic, dated to 194, for instance, depicts Orpheus charming wild animals. 15 Another depiction of Orpheus is known from a mosaic dated to 227/228. 16 Finally, a mosaic dated to 235/236 depicts a Phoenix. 17 Each of these mosaics has an inscription in Syriac; each, however, also depicts a clearly Greco-Roman motif. Thus, these mosaics reflect the influence of Greco-Roman culture in Edessa already from the late second century. A further indication of Greco-Roman influence is found in early Syriac literature. One of the earliest texts to survive is the Book of the Laws of the Countries. 18 The text is a philosophical-theological discussion, in the form of a Platonic dialogue, on fate and free will. The main protagonist is Bardaiṣan (154–222), the earliest known author of Classical Syriac, who was active in the court of the previously mentioned Abgar VIII (r. 177–212). The Book of the Laws of the Countries was probably written in Edessa in ca. 220. Based on its form as a Platonic dialogue and its philosophical subject matter, it is a clear example of Greco-Roman influence in Edessa during the first centuries of the Common Era. 19 Moving a little later in time, a more concrete example of Syriac and Greek interaction is found in the earliest extant dated Syriac manuscript (Brit. Libr. Add. 12,150), which was written in Edessa in 411. 20 This manuscript is composed entirely of translations of Greek works, including Against the Manichaeans by Titus of Bostra, selections of the PseudoClementine Recognitions and Homilies, as well as the Theophany, the History of the Martyrs in Palestine, and the Panegyric on the Christian Martyrs, all by Eusebius of Caesarea. This manuscript establishes the existence of a welldeveloped translation program from Greek into Syriac by at least the fourth century in Edessa and is thus a testament to the interaction of Syriac speakers in Edessa with the Greco-Roman world at this time. In their literature and in their art and architecture, then, the Syriac-speaking population of the early centuries of the Common Era shows signs of significant contact with the Greco-Roman world. This contact was not limited, however, 14. See Bowersock 1990: 31; Possekel 1999: 28. 15.  Published in Healey 2006; with further discussion in Possekel 2008. 16.  Image in Segal 1970: pl. 44; Drijvers and Healey 1999: pl. 53. 17.  Image in Segal 1970: pl. 43; Drijvers and Healey 1999: pl. 52. 18.  Edited with Latin translation in Nau 1907. The Syriac text with English translation is also available in Drijvers 1965. In general, see Brock, in GEDSH, 56–57; Drijvers 1966: 217–18; Jansma 1969; Possekel 2004; 2006; 2009; 2012; Ramelli 2009: 54–90; Ross 2001: 119–23; Teixidor 1992: 65–70. 19. Similarly, Bowersock 1990: 31–32; Millar 1987: 160; Possekel 1999: 29; Ross 2001: 119. 20.  For description, see Wright 1870–72: 2.631–33. A color plate is available in GEDSH, 457. For additional discussion of this manuscript in the context of Greco-Syriaca, see Millar 2011a: 104–6; 2013: 123–24.

The Sociohistorical Setting

29

to literature, art, and architecture but also extended to language. It is clear from inscriptions and documents that the Greek language was present throughout the Syriac-speaking world of Late Antiquity. The majority of the inscriptions west of the Euphrates are written in Greek. 21 In addition, a more limited number of Greek inscriptions come from east of the Euphrates, stretching from the Roman provinces of Osrhoene and Mesopotamia to the Sassanian Empire and beyond. 22 Greek inscriptions are, for instance, known from Syriac-speaking centers such as Edessa, Tella, Amid, and Nisibis, to name only a few. As expressions of the so-called epigraphic habit, inscriptions are not necessarily indicative of language use. 23 These inscriptions do, however, at the very least establish that Greek was present in the Syriac-speaking world. More-compelling evidence for the interaction of Greek and Syriac derives from papyrological documents. 24 As already mentioned, more than 150 documents were discovered at Dura-Europos. The majority of these are written in Greek or Latin, though there are also a few in Iranian or Aramaic. 25 In addition, one of the documents is (mostly) in Syriac: P. Dura 28, which is a bill of sale for a female slave dated 9 May 243. 26 The main text of the document is in Syriac, as are most of the signatures; the signature of the στρατηγός Aurelius Abgar, however, is in Greek, as is that of Aurelius Mannos, who is described in Greek as ‘the one in charge of the sacred and civic (archives)’ (ὁ ἐπὶ τοῦ ἱεροῦ καὶ τοῦ πολειτικοῦ). An even higher degree of interaction between Greek and Syriac is found in the third-century cache of texts known as P. Euphrates that likely originated from Appadana (Neapolis), just north of Dura-Europos on the Euphrates. This cache includes two Syriac parchments (P. Euphrates 19, 20), 27 21.  The inscriptions are currently being collected in Inscriptions grecques et latines de la Syrie (1929–). Various inscriptions are also discussed in, inter alia, Bowersock 1990: 29–30; Kennedy and Liebeschuetz 1988: 69–70; Millar 1987; 2007; Possekel 1999: 27–28; Taylor 2002: 304–17; as well as in the Supplementum Epigraphicum Graecum (SEG). 22.  A useful collection is available in Canali de Rossi 2004. A number of these inscriptions are found with additional discussion in Merkelbach and Stauber 2005. 23.  Fraade 2012: 22*–23*. For the “epigraphic habit,” see MacMullen 1982; Meyer 1990. 24.  For a general discussion of papyrology in the Roman Near East, see Gascou 2009. A checklist of papyrological texts is available in Cotton, Cockle, and Millar 1995. 25.  All are edited in Welles, Fink, and Gilliam 1959. 26.  The most accessible version of the text is Drijvers and Healey 1999: 232–36 [s.v. P1]. See also Bellinger and Welles 1935; Goldstein 1966; Healey 2009: 264–75; Torrey 1935; Welles, Fink, and Gilliam 1959: 142–49 with pls. 69, 71. Plates are also found in Moller 1988: 185–86. 27.  P. Euphrates 19 is a transfer of debt dated 28 December 240 (the most accessible version of the text is Drijvers and Healey 1999: 237–42 [s.v. P2]; see also Aggoula 1992: 391– 99; Brock 1991; Healey 2008; 2009: 252–64; Teixidor 1989: 220; 1990: 144–54 [includes two plates]). P. Euphrates 20 is a property lease dated 1 September 242 (the most accessible version of the text is Drijvers and Healey 1999: 243–48 [s.v. P3]; see also Aggoula 1992: 391–99; Brock 1991; Teixidor 1989; 1990: 154–57; 1991–92 [includes two plates]).

30

Chapter 3

as well as 19 Greek papyri and parchments. 28 On several of the Greek documents, there is additional writing in Syriac. P. Euphrates 6, for instance, along with its duplicate, P. Euphrates 7, records the sale of a slave in Greek, which is followed by seven lines of Syriac summarizing the sale. 29 These two caches of documents illustrate the high degree of interaction between Greek speakers and Syriac speakers (at least on the official level of administration) already in the third century. 3.4.  Analyzing Contact-Induced Changes in Syriac Due to Greek Based on inscriptions and documents, we can establish that Greek and Syriac coexisted in Late Antique Syria and Mesopotamia. It is now necessary to investigate how best to classify contact-induced changes in Syriac due to Greek within the typology of Van Coetsem. Based on the arguments presented in chap. 2, I adopt the threefold typology of contact-induced change proposed by Van Coetsem. First, there is borrowing (recipient language agentivity), in which the agents of change are dominant speakers of the recipient language. Borrowing results primarily in the transfer of lexemes. The second broad category of language contact is imposition (source language agentivity), in which the agents of change are dominant speakers of the source language. Imposition results primarily in the transfer of phonology, syntax, and to a more limited extent morphology. Finally, there is neutralization, in which an individual is equally dominant in two languages. Any feature can be transferred in neutralization. The question to be addressed now is how best to classify contact-induced changes in Syriac due to Greek. Is this a situation of borrowing, imposition, or neutralization in Van Coetsem’s typology? I propose that this contact situation is best analyzed as a situation of borrowing, in which speakers who were linguistically dominant in the recipient language, Syriac, transferred features from the source language, Greek. This analysis is supported by the socio­ linguistic context as well as by the linguistic data. The nonlinguistic evidence for the question at hand is unfortunately slim, being almost entirely restricted to literary sources that do not provide unbiased accounts of language use. In addition, the meager evidence that is available is not representative of the population as a whole but, rather, relates exclusively to a restricted subset of the population, particularly male authors and public figures. Indeed, almost nothing is known about the language use of the majority of the population of Late Antique Syria and Mesopotamia. 30 Not28.  These are edited in Feissel and Gascou 1989; 1995; 2000; Feissel, Gascou, and Teixidor 1997. 29.  An image of this document is found in fig. 4.1 on p. 51 below. 30.  Note that Millar concludes his discussion of the interaction of Greek and Syriac in Edessa from 363 to 435 by stating, “We still do not hear anything directly about the language and culture of the mass of the Edessene population” (Millar 2011: 113).

The Sociohistorical Setting

31

withstanding these difficulties, it is important to see what the literary sources can reveal about language use in the Late Antique population of Syria and Mesopotamia. 31 It is convenient to divide the population of Late Antique Syria and Mesopotamia into two categories: those whose native language was Greek and those whose native language was Syriac. 32 The latter group will be discussed first. Among the segment of the population whose native language was Syriac, there was a continuum of knowledge of Greek. On one end of the con­ tinuum, there were those who had little or no knowledge of Greek. A number of individuals are traditionally thought to have belonged to this category. The “Persian sage” Aphrahaṭ (fl. 337–345), for instance, who is the author of 23 Demonstrations (taḥwyɔṯɔ) and who lived in the Sassanian Empire, is traditionally said not to have known any Greek. 33 Similarly, the well-known author and poet Ephrem the Syrian (d. 373), who spent most of his life in Nisibis, is usually thought to have known little or no Greek. 34 Nevertheless, as noted in the introduction (chap. 1), the baptistery in the church at Nisibis where Ephrem was a deacon contained a Greek building inscription dated to 359/360. 35 So, at the very least, Ephrem must have been exposed to written Greek. 36 Moving a little later in time, the influential West Syriac poet Yaʿqub of Serugh (d. 521) is usually said to have had no knowledge of Greek (Brock 1994: 157), even though he studied Syriac translations of Greek writings in Edessa. 37 While it is possible—and perhaps even likely—that these three well-known Syriac authors had little or no knowledge of Greek, there is very little evidence for this: it is based mainly on arguments from silence. There is evidence, however, that a certain Ouranius, who was bishop of an otherwise unknown place in Osrhoene, was a Syriac speaker who never learned 31.  For the following, see the earlier discussion in Brock 1994: 153–58; 1998: 714–17. 32.  Recall from the previous chapter that a native language is not necessarily the same as a dominant language (see pp. 17–18, 21–23 above). 33. So Brock 1975: 81; Van Rompay 1996: 621. For Aphrahaṭ more generally, see Baarda 1975: 2–10; Brock, in GEDSH, 24–25; Bruns 1991: 1.41–47; Parisot 1894–1907: ix–xxi; Pierre 1988–89: 1.33–41; Ridolfini 2006: 14–22; Wright 1869: 1–10. The Demonstrations are edited in Parisot 1894–1907. Several scholars, most notably Fiey (1968), have argued that the text transmitted as Demonstration 14 may have been written by a different author and only added secondarily to the collection, which would then have originally contained 22 (not 23) Demonstrations. For rebuttal, see Owens 1983: 4–9. More recently, the unity of the Demonstrations as they are currently preserved has been convincingly questioned in Walters 2015. 34.  See, e.g., Pat-El 2006: 43. For additional references, see below at pp. 199–200. For Ephrem more generally, see Brock, in GEDSH, 145–47; a guide to Ephrem’s works is available in Brock 2007b; a bibliography on Ephrem is available in den Biesen 2011. 35.  Bell 1982: 143–45 with pls. 70–83; Canali de Rossi 2004: 39 (no. 62). 36. See Millar 2013a: 45–46. 37.  Jacob mentions this in his Letter 14 (ed. Olinder 1937: 58–61). For discussion, see Becker 2006: 52–53; Jansma 1965; Van Rompay 2010: 207 with n. 22.

32

Chapter 3

Greek. Millar (2013a: 120–21) has drawn attention to the fact that in the acts of the councils Ouranius is consistently presented as being unable to speak and to understand Greek. Thus, he would provide an example of at least one Syriac speaker who did not learn Greek. Undoubtedly, a large number of other Syriac-speaking individuals in Late Antique Syria and Mesopotamia had little or no knowledge of Greek. For all these people, Syriac would have been their linguistically dominant language. Among the people whose native language was Syriac, there were also those who learned some Greek but probably lacked a high degree of proficiency in the language. Within this group was probably Philoxenos (d. 523), bishop of Mabbug. 38 Philoxenos was born outside the Roman Empire in Beth Garmai, and he was educated in Edessa at the School of the Persians. 39 Throughout his career, Philoxenos was actively involved in trying to incorporate Greek theological idioms into Syriac and even sponsored new translations of Greek works, including a new translation of the New Testament by Polykarpos (the now lost Philoxenian Version). Philoxenos’s writings survive only in Syriac, but it is clear that he knew some Greek. There are indications, however, that his knowledge of Greek was limited. In his Commentary on John, for instance, Philoxenos discusses the similarity in spelling between the Greek words γένεσις ‘becoming’ (LSJ 343) and γέννησις ‘birth’ (LSJ 344), stating: The reading of the words “becoming” and “birth” are similar to one another in the Greek language, because two nwn’s are placed one after another in “becoming,” but only one in “birth.” (de Halleux 1977: 43.17–19)

Philoxenos is correct to point out that Greek γένεσις and γέννησις are similar in spelling; however, he confuses the two words in claiming that the former has two n’s and the latter only one. 40 It should be noted that this is not an isolated slip but that other mistakes of this sort involving Greek are found in Philoxenos’s writings. 41 A further indication that Philoxenos had a limited knowledge of Greek is that he did not undertake translations from Greek himself but commissioned translators such as Polykarpos. 42 Finally, when writing to Maron of Anazarbus, Philoxenos mentions that his letter would be translated from Syriac to Greek, which was presumably the language that Maron read. 43 This probably implies that Philoxenos was unable to respond in Greek, and so he wrote the letter in Syriac and then had it translated into Greek. 44 Thus, 38.  For Philoxenos, see de Halleux 1963; Michelson 2014. 39.  For the School of the Persians, see Becker 2006. 40.  For discussion, see de Halleux 1963: 22. 41.  See de Halleux 1963: 123–124. 42. So Brock 1994: 157. 43.  The relevant passage is found in Lebon 1930: 55.21.22 (Syr.); 80.12–13 (LT). 44.  For a similar interpretation, see de Halleux 1963: 21; Lebon 1930: 80 n. 2.

The Sociohistorical Setting

33

though he clearly had some knowledge of Greek, Philoxenos seems to have lacked a high degree of proficiency in the language and probably could not write or speak fluently. This is likely the case for a number of other native Syriac speakers in Late Antique Syria and Mesopotamia. People at this point of the continuum were linguistically dominant in Syriac even if they had some knowledge of Greek. There were also native Syriac speakers who learned enough Greek to be able to speak and/or write in the language. One such person was Yuḥanon of Tella (d. 538), who was born in Kallinikos in 482. 45 According to his Vita, Yuḥanon’s father died when he was only two and half years old, but his mother and grandparents “educated him in the writing (sɛp̄rɔ) and wisdom of the Greeks” (Brooks 1907: 39.22). The word ‘writing’ (sɛp̄rɔ) in this passage could refer to writing in the sense of ‘literature’, but it could also refer to writing in the sense of ‘literacy’. 46 His education was facilitated by a ‘teacher’ (Syriac pdgwgʾ < παιδαγωγός [LSJ 1286]), who was charged with instructing him in the pagan Greek authors. Yuḥanon was also employed in the praetorium of the dux in Kallinikos (Brooks 1907: 39.23–24) and had all of the preparation necessary for a profitable secular career. 47 Against his mother’s wishes, Yuḥanon adopted a monastic life and eventually became bishop of Tella. His writings that survive are only in Syriac, and there is no evidence that he ever wrote in Greek. His Vita, however, at least depicts Yuḥanon as being able to speak Greek with a translator of his Persian captors: When the Marzaban heard this, he immediately commanded that he (viz. Yuḥanon of Tella) sit before him on the ground, and he spoke with him through an interpreter. That one said to him in Greek, “How did a man such as you dare to cross into our place without us? Do you not know that this is another polity?” 48 The blessed one said to him through the interpreter in Greek, “It is not the first time that I have crossed into this land.” (Brooks 1907: 71.21–72.2)

This passage suggests that Yuḥanon of Tella could speak Greek. It also provides an interesting glimpse into the use of Greek in the Sassanian Empire in 45. For Yuḥanon of Tella, see Menze in GEDSH, 447–48. An informative Vita of Yuḥanon of Tella survives (ed. Brooks 1907: 29–95; with an English translation in Ghanem 1970). A shorter Vita is found in the Lives of the Eastern Saints by Yuḥanon of Ephesus (d. 589; ed. Brooks 1923–25: 2.513–26). 46.  In a passage of special interest to literacy and gender in Late Antiquity, P. Dura 28 concludes: “I PN confess that I wrote on behalf of PN my wife in the subscription because she does not know ‘writing’ (sprʾ )” (lines 20–22). This clearly establishes one of the meanings of sɛp̄rɔ as ‘literacy’, a definition not found, for example, in SL 1035. 47.  Both words in the Syriac text are ultimately of Latin origin: Latin praetorium (OLD 1448; LD 1436) > πραιτώριον (LLGE 93; PGL 1126–27) ‫ ܦܪܛܘܪܝܢ‬prṭwryn ‘governor’s residence’ (Brooks 1907: 39.23); Latin dux (OLD 582; LD 621) > δούξ (LLGE 41–42; LSJ 447) > ‫ ܕܘܟܣ‬dwks ‘leader’ (Brooks 1907: 39.23). For Latin loanwords in Syriac, see §4.8. 48.  The word translated here as ‘polity’ is Syriac ‫ ܦܘܠܝܛܝܐ‬pwlyṭyʾ < πολιτεία (LSJ 1434).

34

Chapter 3

the early sixth century. 49 Thus, Yuḥanon of Tella provides an example of a native Syriac speaker who may have received a Greek education as a child that enabled him to communicate in Greek later in life. It seems clear that people like Yuḥanon of Tella, who knew enough Greek to communicate, would still have been dominant speakers of Syriac. Moving along the continuum, there were those whose native language was Syriac and who also had a high degree of knowledge of Greek. To this group, one could point to translators such as Pawlos of Kallinikos (first half of 6th cent.) and Sergios of Reshʿayna (d. 536), both of whom clearly had high facility in Greek. 50 Unfortunately, little is known about their biographies, so it is difficult to say much about their language use. More is known, however, about the language use of Rabbula (d. 435/436), the controversial bishop of Edessa. 51 In his Vita, Rabbula is said to have been instructed in Greek “writing” as part of his education: 52 When he grew up, he was instructed in Greek “writing” (sɛp̄rɔ) as a child of rich nobles of the city of Qenneshrin. (Overbeck 1865: 160.25–27)

As in the case of Yuḥanon of Tella (discussed above), the word ‘writing’ (sɛp̄rɔ) in this passage could refer to ‘literature’ or ‘literacy’. A clearer picture of Rabbula’s language use can be obtained from the fact that Rabbula wrote and spoke in both Greek and Syriac. Several of his works related to regulating clergy and monastics were written in Syriac (CPG 6490–92). 53 His Vita relays that he also wrote 46 letters in Greek. 54 In addition, he is said to have translated Cyril of Alexandria’s On Orthodox Faith from Greek to Syriac (CPG 6497). 55 Finally, Rabbula seems to have preached a homily in Constantinople in Greek (CPG 6496). 56 49.  See similarly Millar 2013b: 63. 50.  For the former, see Van Rompay, in GEDSH, 323–24; and (with more detail) King 2007; 2008: 175–77, passim. For the latter, see Brock, in GEDSH, 366 with additional references. 51.  For Rabbula, see Blum 1969. A Syriac Vita of Rabbula survives (ed. Overbeck 1865: 159–209; with an English translation in Doran 2006: 65–105). For additional discussion of Rabbula and his relationship to Greek, see Millar 2011: 109–13; 2013a: 128. 52. See Becker 2006: 11; J. W. Drijvers 1999: 141. 53.  Edited in Overbeck 1865: 210–21; Vööbus 1960: 24–50, 78–86. See Vööbus 1970a: 128–38; 1970b: 307–15. Some of these are of dubious authenticity. 54.  Overbeck 1865: 200.18–23. Several letters that are attributed to him, or selections thereof, are preserved in Syriac (CPG 6493–95; ed. Overbeck 1865: 222–38). For further discussion, see Millar 2011: 111. 55.  For the question of authorship of this translation, see King 2008: 27–28. See also Brock 1998: 716 n. 17; Millar 2011: 112. 56.  The text survives only in Syriac translation (ed. Overbeck 1865: 239–44). For discussion, see Blum 1969: 131–49.

The Sociohistorical Setting

35

The homily that Rabbula preached in Constantinople in Greek provides further evidence concerning his language use. 57 He begins this homily by expressing hesitancy about speaking Greek in front of a native Greek-speaking audience: We are small in our word (mɛllṯɔ) and in our knowledge. You, however, are great in spiritual wisdom and in acuteness of language (lṭišuṯɔ dlɛššɔnɔ). Because of this, who would not be afraid in a church such as this! (Overbeck 1865: 239.5–8)

The contrast between Rabbula’s being small in word (mɛllṯɔ) and his audience’s being large in acuteness of language (lṭišuṯɔ dlɛššɔnɔ) suggests that Rabbula was not entirely comfortable speaking Greek and that he probably would have preferred to deliver his homily in Syriac. A little later in the homily, he goes on to apologize more explicitly for his facility in Greek, since he was a “man of the countryside (quryɔyɔ) and living among country-folk (quryɔye) (where) it is Syriac that we mostly speak” (Overbeck 1865: 241.11–12). 58 While these statements likely involve some rhetorical modesty, 59 they do still seem to suggest that Rabbula was linguistically dominant in Syriac. Thus, when Rabbula was speaking Greek in Constantinople, it would have been a situation of imposition, since he had linguistic dominance in the source language, Syriac. This would have resulted in the transfer of Syriac phonology and syntax into his Greek, changes of which Rabbula himself seems to have been all too well 57.  The heading of this text reads: ‫ܬܘܪܓ�ܡܐ ܕܡܠܠ ܡܪܝ ܪܒܘ�ܠܐ ܐܦܣܩܘܦܐ ܕܐܘܪܗܝ ܒܥܕܬܐ‬ ‫‘ ܕܩܘܣܛܢܛܝܢܦܘܠܝܣ ܩܕܡ ܥ�ܡܐ ܟܠܗ‬The discourse which Mar Rabbula, the bishop of Edessa, spoke in the church of Constantinople before all of the people’. Millar (2013a: 128) understands the word turgɔmɔ here to mean ‘translation’ (note also his incorrect transliteration of the Syriac as trgmʾ without the mater lectionis w). This understanding of the word is, however, unlikely: what would it mean to say, ‘A/The translation which Mar Rabbula, the bishop of Edessa, spoke . . .’? Thus, even though the text preserved in Syriac is a translation from the original Greek of Rabbula’s homily, it is not explicitly described as such. 58.  Note that this statement by Rabbula fits well with a situation often posited in the secondary literature in which local elites in the cities would have had knowledge of Greek, whereas the rural population would have spoken only Syriac. In a now classic article on “Provincial Languages in the Roman Empire,” for instance, MacMullen contrasts the use of Greek in urban areas for trade, tribute, commerce, and administration with the use of Syriac in the “remoter countryside” and “the more backward districts” (MacMullen 1966: 13 and 5, respectively). 59.  So already Brock 1967: 155. In The Orator’s Education (4.1.9; ed. Russell 2001), for instance, Quintilian notes that a standard rhetorical ‘trick’ (simulatio) of the προοίμιον is to feign to be inexperienced or incompetent. As a Syriac comparison, many of the memre by Yaʿqub of Serugh (d. 521) begin with a προοίμιον in which he declares his inadequacy for expressing his subject matter (see Blum 1983: 308–9). For Greek rhetorical training in the Syriac milieu, see Watt 1985; 1986; 1987; 1989; 1990; 1993a; 1993b; 1994a; 1994b; 1995; 1998; 1999; 2005; 2009.

36

Chapter 3

aware. 60 When Rabbula was speaking Syriac, however, the situation would have been one of borrowing, since he had linguistic dominance in the recipient language. Thus, Rabbula seems to have had linguistic dominance in Syriac and so could borrow from Greek into Syriac; he also spoke Greek, though not as his linguistically dominant language, in which case he would have imposed Syriac features onto Greek. Rabbula fell far along on the continuum of knowledge of Greek among native Syriac speakers. He was not, however, at the end of this continuum. There were native speakers of Syriac who wrote exclusively in Greek and seem to have been more a part of the Greco-Roman world than the Syriac-speaking world. One such person is Eusebius of Emesa (died before 359). 61 Eusebius was born in Edessa around 300, and so his native language would have been Syriac. In addition, Eusebius spoke Greek fluently and wrote, it seems, entirely in that language. Unfortunately, little else can be said definitively about Eusebius’s language use. 62 A clearer picture of language use, however, can be ascertained with Theodoret of Cyrrhus (393–466). 63 Theodoret was born in Antioch to wealthy parents, and he seems to have received a thoroughly Greek education. He wrote a number of works in Greek, including biblical commentaries, dogmatic works, an ecclesiastical history, a hagiography of monks from Syria, as well as the Cure for Hellenic Maladies, which engages with pagan Greek thought and philosophy. His written Greek is of a very high literary character. 64 Notwithstanding his facility in Greek, it seems that Theodoret also spoke Syriac. This is mentioned in a number of passages in his History of the Monks of Syria, in which he converses with monks in Syriac. 65 In one instance, Theodoret even understands a demon who speaks to him in Syriac (21.15–16). Brock uses these passages as evidence that Theodoret “normally spoke Syriac” (Brock 1994: 154 n. 27). Similarly, Millar states that these passage “show that he could understand the spoken language” (Millar 2013a: 118; see also p. 36)—that is, Syriac. More caution is needed, however, in the use of such anecdotal stories for establishing Theodoret’s language

60.  As a comparison, the “Syrian rhetor” Lucian of Samosata mentions his foreign accent when speaking Greek (The Double Indictment; ed. Harmon 1921: 136–37; see also Becker 2006: 11). 61.  For Eusebius of Emesa, see ter Haar Romeny 1997: 7–12; Petit, Van Rompay, and Weitenberg 2011: xxiii–xxix; Van Rompay, in GEDSH, 155; Winn 2011. 62.  For discussion, see ter Haar Romeny 1997: 9–10; along with Brock 1998: 715 with n. 15. 63.  For Theodoret, see Urbainczyk 2002. For a discussion of his language use, see Millar 2013a: 35–38. 64.  Photius (d. ca. 893) praises it in his Bibliotheca, 31 (Henry 1959–91: 1.17–18). 65. For discussion, see Urbainczyk 2000. The text is edited in Canivet and LeroyMolinghen 1977–79; with an English translation in Price 1985.

The Sociohistorical Setting

37

use. 66 Nevertheless, there are other indications that Theodoret not only spoke Syriac, but that Syriac seems to have been his native language. The clearest support for this derives from a passage in the Cure for Hellenic Maladies, in which Theodoret states: καὶ ταῦτα λέγω οὐ τὴν Ἑλλάδα σμικρύνων φωνήν ἧς ἁμηγέπη μετέλαχον οὐδὲ ἐναντία γε αὐτῇ ἐκτίνων τροφεῖα . . . I say these things not to belittle the Greek language, in which I have obtained a share to some extent, nor to not make a return to it for bringing me up. (Canivet 1958: 5.75) 67

In its most straightforward interpretation, this passage implies that Greek was not Theodoret’s native language. 68 Thus, Theodoret’s native language seems to have been Syriac. In addition, it is clear that Theodoret had a very high knowledge of Greek. Given these points, Theodoret would represent the very far end of the continuum of Syriac speakers who learned Greek. In Van Coet­ sem’s typology, he might even be approaching neutralization, in which an individual is equally dominant in two languages. To summarize up to this point, there was a continuum of knowledge of Greek among people whose native language was Syriac in Late Antique Syria and Mesopotamia. At one end of the continuum, there was someone like Ouranius, who had little or no knowledge of Greek. At the other end of the continuum, there was someone like Theodoret, whose native language seems to have been Syriac but who wrote extensively in a very high register of Attic Greek and who was fully at home in the Greco-Roman world. Between these two poles, there were a number of native Syriac speakers who had some knowledge of Greek, from Philoxenos and his limited facility in the language to Rabbula and his ability to write and speak fluently. Thus, 66.  See similarly Johnson 2015a: xvi. 67.  It remains unclear what exactly Theodoret intends with ἀμηγεπη, which typically means ‘in one way or another’ (LSJ 82). The translation above follows Urbainczyk (2002: 17) and Millar (2007: 117) in rendering it ‘to some extent’. Canivet translates ‘qui est bien un peu la mienne’ (Canivet 1958: 250). 68.  So also Bardenhewer 1924: 221; Bardy 1948: 19; Brock 1998: 714; Canivet 1977: 38–39 with n. 11. See also MacMullen, who states that Theodoret was “reared in Syriac” (MacMullen 1966: 4). More recently, however, this interpretation has been questioned. Urbainczyk (2002: 16–17), for instance, accepts the straightforward interpretation of the sentence but proposes that it is to be understood ironically, since Theodoret is after all writing in Greek. Similarly, Millar (2007: 117) argues that this is “a conventional expression of modesty” on the part of Theodoret. It should be noted, however, that both Urbainczyk and Millar have an ulterior motive for rejecting the straightforward interpretation of the passage: neither thinks that Syriac was in fact Theodoret’s first language. Their arguments for this are insufficient, however, being built around the logic that Theodoret is a major author of literary Greek, and therefore he must have been a native speaker of Greek. This argument does not hold up, since literary ability and native language are not directly correlated.

38

Chapter 3

there is ample—albeit mostly anecdotal 69—evidence that a number of native Syriac speakers learned Greek to varying degrees but remained linguistically dominant in Syriac. Shifting now to the segment of the Late Antique population whose native language was Greek, we see an interesting difference emerge. There are no attested cases in which a native Greek speaker is known to have learned Syriac. This is of course an argument from silence, but it is striking nonetheless. 70 The lack of evidence for Greek speakers learning Syriac provides an important contrast for the situation described above for Syriac speakers learning Greek. One particularly remarkable foil is Severus (d. 538), who was patriarch of Antioch between 512 and 518 and who continued to serve as the leader of the anti-Chalcedonians until his death. 71 Severus was born to a pagan family in Sozopolis in Pisidia, a region in southwestern Anatolia. As a native of Pisidia, his native language would have been Greek. He was educated in Alexandria and then in Berytus. While in Berytus, he converted to Christianity and eventually became a monk. He was elected patriarch of Antioch in 512, but in 518 with the ascension of the pro-Chalcedonian Justin I, he was forced to flee to Egypt, where he spent the remainder of his life. Of particular interest to the current discussion is Severus’s time in Antioch as Patriarch. While Antioch had a large Greek-speaking population, many members of Severus’s patriarchate would have been Syriac speaking, especially farther to the east. Thus, he would have had good reasons to learn Syriac. There is, however, no indication that he ever did this. All of Severus’s oeuvre was written in Greek and then translated into Syriac during his lifetime. In addition, and more importantly to the point being made here, there is no evidence that he had the ability to use Syriac in any capacity. This is particularly interesting since Syriac speakers were extremely receptive to Severus, who became one of the most popular and influential leaders of the anti-Chalcedonians. 72 Thus, Severus provides an instructive contrast to Syriac speakers learning Greek. He was a native speaker of Greek who had various reasons to learn Syriac, but there is no indication that he actually did so. 73 Severus is not an isolated example; he seems to be 69.  For problems with such anecdotal evidence, see Johnson 2015a: xvi. 70.  A possible exception that proves the rule is Eusebius of Caesarea (d. 339/340). In his Ecclesiastical History (1.13; ed. Lake 1926), Eusebius mentions translating Syriac texts from the archives of Edessa into Greek. Since he was probably born in Caesarea, his native language was probably Greek (or possibly a dialect of Aramaic), but it was certainly not Syriac. Thus, Eusebius would have learned Syriac as a foreign language—that is, if his account of translating Syriac documents into Greek is to be trusted (Millar [2011b: 105] is skeptical). 71.  For Severus, see Brock, in GEDSH, 368–69; and (with more detail) Allen and Hayward 2004. 72.  See the illuminating new study by Moss (forthcoming). 73.  It should be noted that Severus may well have learned Latin since he studied law in Berytus.

The Sociohistorical Setting

39

representative of Greek speakers in Late Antique Syria and Mesopotamia in that he did not learn Syriac. The dichotomy between at least some Syriac speakers learning Greek and few, if any, Greek speakers learning Syriac was motivated by social factors. As the language of empire, Greek was a language of prestige in Late Antique Syria and Mesopotamia. Thus, Greek speakers would have had little impetus to learn Syriac. Even when a Greek speaker such as Severus had a reason to learn Syriac, he still might not have done so, given the relative prestige of the two languages. In contrast, Syriac speakers would have had very good reasons to learn Greek. This is especially true for local Syriac-speaking elites who wished to participate in the broader Greco-Roman world. With these cases of individual language use laid out, I now turn to how contact-induced changes in Syriac are to be analyzed within Van Coetsem’s typology. There are no known examples in which a native Greek speaker learned Syriac. Thus, there is no example of imposition by native Greek speakers. 74 With native Syriac speakers, there was a continuum of knowledge of Greek. In the majority of the known cases, if not all of them, Syriac remained the dominant language for these speakers. 75 Thus, they would have borrowed from Greek into Syriac when using Syriac and imposed from Syriac onto Greek if using Greek. Contact-induced changes in Syriac due to Greek should, then, be analyzed as borrowing, in which the agents of change were dominant speakers of the recipient language. The linguistic data corroborate this analysis of borrowing. Syriac contains a large number of Greek loanwords. There are in fact more than 800 Greek loanwords attested in pre-eighth-century Syriac texts that were not translated from Greek. 76 As discussed above, the transfer of lexemes is the expected effect of borrowing. 77 In contrast, the types of changes that are associated with imposition are not found in Syriac: there is no evidence of the systematic 74.  It is of course likely that there were at least a few native Greek speakers who learned Syriac (see n. 70 above). Their numbers would have been very small, however, and thus it is unlikely that any changes in their Syriac due to imposition would have spread throughout the Syriac-speaking population. For the distinction between contact-induced change on the individual level and the diffusion of those changes to the broader community, see Van Coet­ sem 2000: 40, 57, 281. 75.  At the very far end of the continuum of knowledge of Greek among native Syriac speakers, there may have been a small segment of the population who had equal linguistic dominance in Greek and Syriac, such as perhaps Theodoret, or who even had a switch of linguistic dominance from Syriac to Greek. In these limited cases, there would have been neutralization or imposition, respectively. It should be noted, however, that the number of such individuals would again have been so small that it is unlikely that any changes in their Syriac would have spread throughout the Syriac-speaking population as a whole. 76.  These are analyzed in detail in chaps. 4–6. 77. See Smits 1996: 32–33, 38; Thomason and Kaufman 1988: 74–109; Van Coetsem 1988: 10–11; 1995: 77–80; 1997: 358–59; 2000: 53, 67–73, 137–66.

40

Chapter 3

transfer of phonological, morphological, or syntactic features from Greek to Syriac. The only phonological features transferred are associated with loanword integration, such as the “emphatic” p (see §5.2.6). The only Greek morphological features in Syriac are secondary developments due to analogy, such as the Berufsname suffix -ɔrɔ (see §6.6). The syntactic features transferred are cases of grammatical replication in which speakers of Syriac created a new grammatical structure on the model of a structure in Greek (see §§7–9). These cases of grammatical replication, however, are isolated, nonsystematic, and of limited scope in contrast to the transfer typically witnessed in imposition. 78 Thus, the linguistic evidence also suggests that the contact-induced changes in Syriac due to Greek should be analyzed as borrowing and not as imposition or neutralization. 3.5. Conclusion Aramaic speakers were in contact with the Greek language beginning already in the middle of the first millennium b.c.e. Alexander the Great’s defeats of Darius III in the 330s b.c.e. ultimately led to the establishment of Seleucid control over Syria and Mesopotamia. With the Seleucid Empire came the foundation of Hellenistic cities and the use of Greek as the official language of empire in the region. Contact between Aramaic and Greek became even more pronounced with the Roman conquest of the area in the first centuries of the Common Era. Thus, the Syriac-speaking culture that comes into view in the first centuries of the Common Era was one that had been in contact with the Greco-Roman world and its Greek language for centuries. Given that Greek was the language of the (eastern) Roman Empire, it is no surprise that many native Syriac speakers learned it to one degree or another. In contrast, there are no indications that Greek speakers ever learned Syriac. This suggests that the contact-induced changes in Syriac due to Greek should be analyzed as the result of borrowing within the typology of Van Coetsem. That is, these contactinduced changes were the result of dominant speakers of Syriac transferring features from Greek into their own language. This analysis is corroborated by the linguistic data, since primarily loanwords (which are more common in borrowing) were transferred into Syriac from Greek and not phonology or syntax (which are more common in imposition). 78.  For a discussion of how grammatical replication (or the transfer of structure more generally) fits within a situation of borrowing, see §10.2.

Loanwords ✧ ✧ ✧

Chapter 4

Greek Loanwords in Syriac: The Methodological Framework No language is entirely free from borrowed words, because no nation has ever been completely isolated. (Jespersen 1922: 208 n. 1)

4.1. Overview It is widely recognized that one of the most basic effects of language contact is the transfer of lexemes from one language to another. In his Language, for instance, Sapir notes that “[t]he simplest kind of influence that one language may exert on another is the ‘borrowing’ of words” (Sapir 1921: 193). Given the prolonged period of contact between Syriac and Greek described in chap. 3, it comes as no surprise that Syriac contains numerous Greek loanwords. There are in fact more than 800 Greek loanwords attested in pre-eighth-century Syriac texts that were not translated from Greek. The passage in (4.1) below provides several examples of Greek loanwords in Syriac: (4.1) Memrɔ on Elijah and His Ascension into Heaven, by Yaʿqub of Serugh (d. 521; ed. Bedjan 1905–10: 4.226–59; see also Kaufman 2009: 349–427) ̈ ‫ܐܓܘܢܐ ܘܩܐ̈ܪܣܐ ܦܓܥܘ ܒܗ ܒܥܠ�ܡܐ ܒܝܫܐ‬ ‫ܟ�ܡܐ‬ ‫ܘܗܝܕܝܢ ܥܒܪܗ �ܠܐܬܪܐ ܕܡܘܬܐ ܘܐܓܗܝ ܡܢܗ‬ ̈ ‫ܠܟ�ܡܐ‬ ‫ܦܚܐ ܒܐܘܪܚܗ ܕܥܠ�ܡܐ ܚܙܐ ܕܛܡܝܪܝܢ‬ ‫ܘܫܘܪ ܐܢܘܢ ܒܥܡ�ܠܐ ܪܒܐ ܘܟܢ ܐܬܢܨܚ‬ ̈ ܿ ‫ܢܗܙܘܗ �ܠܐܠܦܗ ܒܝ�ܡܐ ܡܝܬܐ‬ ‫ܟܝܡܘܢܐ‬ ‫ܟ�ܡܐ‬ ̈ ‫ܘܗܝܕܝܢ ܡܛܝ ܠܗܢܐ ܠ�ܡܐܢܐ ܕ�ܠܐ ܡܝܘܬܐ‬ ‫ܟ�ܡܐ ܐܬܟܬܫ ܥܡ ܫܠܝܛܐ ܕܢܛܪ ܐܐܪ‬ kmɔ ʾgẅnʾ wqʾr̈sʾ how contest-m.pl.emp and+battle-m.pl.emp bʿɔlmɔ in+world-m.sg.emp

bišɔ evil-m.sg.emp

43

pḡaʿ(w) beh encounter-suf.3.m.pl in+him

44

Chapter 4

whayden and+then

ʿaḇreh cross-suf.3.m.sg+him dmawtɔ nml+death-m.sg.emp

laḵmɔ to+how

lʾaṯrɔ to+place-m.sg.emp

wʾaḡhi and+escape-suf.3.m.sg

paḥḥe bʾurḥeh trap-m.pl.emp in+way-f.sg.con+him

mɛnneh from+him

dʿɔlmɔ nml+world-m.sg.emp

ḥzɔ daṭmirin see-suf.3.m.sg nml+be.hidden-part.m.pl.abs wašwar and+leap-suf.3.m.sg

ʾennon bʿamlɔ them-m in+work-m.sg.emp

rabbɔ great-m.sg.emp

ʾɛṯnaṣṣaḥ wḵen and+then succeed-suf.3.m.sg kmɔ how

̈ wnʾ kym storm-m.pl.emp

nahzuh shake-suf.3.m.pl+her

byammɔ in+sea-m.sg.emp whayden and+then

lʾɛllp̄eh to+boat-f.sg.con+his

miṯɔ dead-m.sg.emp

maṭṭi lhɔnɔ arrive-suf.3.m.sg to+this-m

lmʾnʾ dlɔ harbor-m.sg.emp nml+neg

mɔyote dead-m.pl.emp kmɔ ʾɛṯkattaš ʿam how struggle-suf.3.m.sg with

šalliṭɔ ruler-m.sg.emp

dnɔṭar ʾʾr nml+guard-part.m.sg.abs air

‘How many contests and battles encountered him in this evil world until he crossed over the place of death and escaped it? How many hidden traps did he see in the path of the world until he jumped over them with great effort and so succeed? How many storms shook his boat in the mortal sea until he arrived at the harbor of the immortals? How much did he struggle with the ruler who guards the air . . . ?’  (233.11–17) This seven-line excerpt derives from a memrɔ, or metrical homily, written in Syriac by the influential West Syriac poet Yaʿqub of Serugh (d. 521). The author was a native Syriac speaker who, according to Brock (1994: 157), probably had no knowledge of Greek. In all likelihood, this homily was delivered to a Syriac-speaking congregation located somewhere near the Euphrates, perhaps in either Ḥawra or Baṭnan da-Serugh. Five of the 16 substantives in the excerpt have a Greek origin:

Greek Loanwords in Syriac: The Methodological Framework (4.2) a. b. c. d. e.

45

‫ ܐܐܪ‬ʾʾr ‘air’ < ἀήρ (LSJ 30) ‫ ܐܓܘܢܐ‬ʾgwnʾ ‘contest’ < ἀγών (PGL 25; LSJ 18–19) ‫ ܟܝܡܘܢܐ‬kymwnʾ ‘storm’ < χειμών (LSJ 1983) ‫ ܠ�ܡܐܢܐ‬lmʾnʾ ‘harbor’ < λιμήν (LSJ 1050) ‫ ܩܐܪܣܐ‬qʾrsʾ ‘battle’ < καιρός (LSJ 859–860)

These words illustrate the topic of this chapter: Greek loanwords in Syriac. 4.2.  History of Research The Greek loanwords in Syriac have been an object of study for more than a millennium. Already in the ninth century, the well-known translator Ḥunayn b. Isḥāq (d. 873) wrote several treatises on Syriac lexicography that probably touched on Greek loanwords in Syriac. In addition to his work on homographs entitled the Book of Similar Words (Kṯɔḇɔ dašmɔhe dɔmyɔye), 1 he wrote a Compendious Lexicon (lhksyqwn bp̄ɔsiqɔṯɔ), which unfortunately does not survive, though it in all likelihood included lemmata for Greek loanwords in Syriac. In addition, Ḥunayn authored a work entitled Explanation of Greek Words with (or: in?) Syriac (puššɔq šmɔhe yawnɔye bsuryɔyɔ). Though again this work does not survive, it may well have been an early treatment dedicated solely to Greek loanwords in Syriac. 2 Ḥunayn’s lexicographical work was incorporated into a number of later lexica. This includes the Lexicon of his student Ishoʿ bar ʿAli, who lived in the second half of the ninth century. 3 In the introduction to his Lexicon, Bar ʿAli states that he employed the Lexicon of Ḥunayn as well as that of another ninth-century lexicographer, Ishoʿ of Merv, when compiling his own Lexicon. 4 Bar ʿAli’s Lexicon includes a number of Greek loanwords that are explained in Syriac and/or in Arabic. In the mid-tenth century, another lexicographer, Ḥasan bar Bahlul, composed a large Lexicon, 5 which relied on Ḥunayn as well as other sources. Bar Bahlul’s Lexicon, like Bar ʿAli’s, contains a considerable number of Greek loanwords with Syriac and/or Arabic definitions. The lexica of Bar ʿAli and Bar Bahlul represent extensive treatments of Greek loanwords in Syriac within the Syriac tradition itself. The lexica of Bar ʿAli and Bar Bahlul were incorporated into the two large Syriac lexica that were published at the end of the nineteenth century: the 1.  Edited in Hoffmann 1880a: 2–49; along with Gottheil 1887: *61–*67; 1889. 2. So Taylor, in GEDSH, 392. 3. The Lexicon is edited in Hoffmann 1874; Gottheil 1910–28. There has been a good deal of confusion in the secondary literature concerning the biography and identity of the lexicographer Bar ʿAli; for which, now see Butts, in GEDSH, 53–54; and (with more detail) Butts 2009a. 4.  Ishoʿ of Merv is probably to be distinguished from Zekarya of Merv, who is often cited in the Lexicon of Ḥasan bar Bahlul (mid-tenth century). So also Baumstark 1922: 241– 42; Butts, in GEDSH, 216–17; against Duval 1907: 297. 5.  Edited in Duval 1888–1901.

46

Chapter 4

Thesaurus Syriacus by Robert Payne Smith (1879–1901), which appeared in an English abridgment as A Compendious Syriac Dictionary edited by his daughter Jessie (1903), and, to a lesser extent, the Lexicon Syriacum by Carl Brockel­mann (1895 [1st ed.]; 1928 [2nd ed.]), which was recently translated into English, with substantial updates and corrections, as A Syriac Lexicon by Michael Sokoloff (2009). These two large Latin lexica, along with their English versions, include lemmata for most of the Greek loanwords that are found in Syriac texts. Outside the standard Syriac lexica, the only monographic study of Greek loanwords in Syriac is A. Schall’s Studien über griechische Fremdwörter im Syrischen (1960). 6 This book is divided into two parts. The first provides an inventory of Greek loanwords found in Syriac literature not translated from Greek up to Ephrem in the middle of the fourth century. The second lists Greek loanwords related to religion, cult, and myth that are found throughout Syriac literature, (unfortunately) disregarding diachronic considerations. While the first part is relatively comprehensive, the second is not only limited in scope but also lacks a number of words and references. 7 Since Schall’s monograph, a number of studies have appeared that analyze Greek loanwords in individual corpora or authors. 8 The greatest bulk of this work has been carried out by Brock. 9 Despite this ever-growing body of literature, a contact linguistic analysis of Greek loanwords in Syriac continues to be a desideratum. 10 The next several chapters below aim to help fill this void in the literature. 4.3. Definition In this book, a loanword is defined as a lexeme that has been transferred from the source language into the recipient language.  11 Loanwords always involve the transfer of phonetic material. That is, they are instances of global copying, as opposed to selective copying, in the Code-Copying Model developed by Johanson (see, e.g., 2002a); and matter replication, as opposed to pattern replication, in the framework of Matras and Sakel (2007b; 2007c). If phonetic material is not transferred, then it is not a case of lexical transfer but of lexical calque, grammatical replication (see §§7– 9), or other kinds of change. 12 6.  A valuable Greek-Syriac index for this work is provided in Voigt 1998a. 7.  So already Brock 1967: 389 with n. 5. 8.  A useful bibliography is available in Voigt 1999–2000. 9.  See, e.g., Brock 1967; 1975; 1982; 1994; 1996; 1999–2000; 2004; 2005; 2010. 10.  So also Brock 1967: 389, 426; 1996: 251–53; 2004: 39; Taylor 2002: 327 n. 61. 11.  A similar definition is found in Haspelmath 2008: 46. See also Haugen 1950b: 213– 14. It should be noted that occasionally the input involves more than one lexeme. This is the case, for instance, with ‫ ܕܝܛܣܪܘܢ‬dyṭsrwn ‘Diatessaron’, from the Greek phrase διὰ τεσσάρων, which literally means ‘through (the) four (Gospels)’. 12.  For some of these, see §7.6 below.

Greek Loanwords in Syriac: The Methodological Framework

47

In the scholarly literature, the terms loanword and lexical borrowing have often been employed interchangeably. 13 This is unfortunate since the term borrowing has been used in so many (contradictory) ways throughout the contact linguistic literature (see §2.2). In this book, borrowing refers to a type—in the sense of typology—of contact-induced change in which the agents of change are dominant speakers of the recipient language (see §§2.3–2.6). Since the transfer of lexemes is attested not only in situations of borrowing but also in situations of imposition and of neutralization, I avoid the use of the term lexical borrowing. Thus, the lexeme that is transferred from the source language to the recipient language is termed a loanword (never a lexical borrowing), and the process is termed lexical transfer (never lexical borrowing). 4.4. Corpus The next several chapters of this book are based on a corpus of more than 800 Greek loanwords and their derivatives found in pre-eighth-century Syriac texts that were not translated from Greek. 14 This corpus has been populated from several sources: concordances to text; 15 indexes to text editions that list Greek loanwords in Syriac, especially those published in the Corpus Scriptorum Christianorum Orientalium (CSCO); my own readings; as well as a systematic exploitation of Michael Sokoloff’s work A Syriac Lexicon (2009), which is a translation (with correction, expansion, and update) of the Lexicon Syriacum by Carl Brockelmann (1895 [1st ed.]; 1928 [2nd ed.]). Some lemmata in the corpus contain only a few references (or sometimes only one), whereas others contain more than 100. It has also been possible to search for additional occurrences of loanwords in three large “databases”: (1) the Comprehensive Aramaic Lexicon (CAL); 16 (2) the Oxford-BYU Syriac Corpus; 17 and (3) Prof. Sebastian Brock’s more than 2,000 card files listing Greek loanwords in Syriac. 18 These three “databases” have been consulted on numerous occasions (though not systematically) and have proven especially useful for establishing the first occurrence of a loanword in Syriac. 13.  See, e.g., Haspelmath 2009: 36. 14.  I hope to publish a glossary of Greek loanwords in Syriac based on this corpus at some point in the future. 15.  New Testament (Kiraz 1993); Book of the Laws of the Countries (Lund 2007); Book of Steps (Kmosko 1926); Demonstrations by Aphrahaṭ (Parisot 1894–1907). 16.  Accessible online at http://cal.huc.edu/. I am grateful to Stephen Kaufman (Professor Emeritus of Bible and Cognate Literature at HUC-JIR/Cincinnati), not only for helping to develop and for curating this important resource, but also for responding to inquiries on various occasions. 17.  I am grateful to Kristian Heal (Brigham Young University), who generously provided me with a Beta-version of the Oxford-BYU Syriac Corpus. 18.  I am grateful to Sebastian Brock (Emeritus Reader in Syriac Studies at Oxford University), who allowed me to digitize his card files over several weeks in August of 2011.

48

Chapter 4

Throughout this book, citations of Greek loanwords in Syriac are systematically provided with references to LS2 and SL (only rarely is a loanword not found in these lexica). The English translations in this book derive from SL. At times, secondary literature relevant to the particular point being made is cited; these citations are not intended to be exhaustive histories of scholarship of the loanword in question. In several instances, it has been important to establish the earliest occurrence of a Greek loanword in Syriac. When this is the case, the earliest text attesting the loanword that is known to me is cited with a heading in bold giving the century of composition. Consider, for instance, the following loanword: ἔθος (LSJ 480) > ‫ ܗܬܘܣ‬htws ‘custom’ (6th cent. Eliya, Life of Yuḥanon of Tella, 84.26 [ed. Brooks 1907: 29–95]; LS2 184; SL 356). To my knowledge, this word is not found in Syriac until the sixth century, when it occurs in the Life of Yuḥanon of Tella by Eliya, which was edited by Brooks. 4.5.  Lehn- oder Fremdwörter? A distinction is usually made in the scholarly literature between Lehnwörter and Fremdwörter. 19 The former are said to have been integrated, to one degree or another, into the recipient language, whereas the latter remain foreign words in the recipient language. Though scholars have at times considered this to be a binary opposition, it is more likely that Lehnwörter and Fremdwörter represent a continuum. Within the context of Syriac-Greek language contact, it is often difficult to determine where a given word falls on the continuum between Lehnwörter and Fremdwörter. 20 There are, however, occasional clues. One such clue is the degree of integration, especially on the morphosyntactic level. 21 Some Greek loanwords in Syriac, for instance, do not regularly occur with the synthetic suffixed genitive pronouns but, rather, prefer analytic independent possessive pronouns based on dil-, as in the following examples from the Lives of the Eastern Saints by Yuḥanon of Ephesus (ca. 589; ed. Brooks 1923–25): (4.3) a. b. c.

‫ ܐܝܣܘܢ ܕܝܠܗܝܢ‬ʾyswn dilhen ‘a copy of them’ (143.7–8) ܿ ‫ܕܝܠܗ‬ ‫ ܦܠܛܝܢ‬plṭyn dilɔh ‘her palace’ (430.7) ܿ ‫ܕܝܠܗ‬ ‫ ܣܩܠܪܐ‬sqlrʾ dilɔh ‘her treasurer’ (420.9–10)

The use of dil- in these examples suggests that ‫ ܐܝܣܘܢ‬ʾyswn ( ‫ ܐܩܛܣܛܣܝܐ‬ʾqṭsṭsyʾ ‘disorder’ (75.7; 80.5; LS2 44; SL 92) b. Latin ducatus (OLD 576; LD 615) > δουκᾶτον (PGL 384) > ‫ ܕܘܩܛܘܢ‬dwqṭwn ‘military command’ (87.2; LS2 163; SL 287) c. ἔθος (LSJ 480) > ‫ ܗܬܘܣ‬htws ‘custom’ (84.26; LS2 184; SL 356) Given the nature of the composition—that is, hagiography—it is unlikely that the author, Eliya, intended to restrict his work to an exclusively bilingual audience, excluding monolingual Syriac speakers. 34 Thus, these are in all likelihood loanwords and not single-word code-switches. The fact that these words are not otherwise attested in Syriac seems, then, to be only an accident of survival. Cases such as this have important implications for the use of relative frequency as a criterion to distinguish single-word codeswitches from loanwords in a corpus such as Syriac. In this book, I choose to err on the side of loanwords. Thus, I consider the default position to be that a Greek word in a Syriac text is a loanword (and not a code-switch) and that the burden of proof lies on establishing that a particular Greek word is a code-switch. 4.7.  Immediate Source and Ultimate Source In this book, a loanword is defined as a lexeme that has been transferred from the source language into the recipient language. It is important to clarify what exactly is meant by source language in this context. In particular, it is necessary to distinguish between immediate source and ultimate source. 35 Immediate source refers to the language from which a lexeme was transferred to the recipient language, whereas ultimate source is a reflection of a word’s etymology. In many instances, the immediate source and the ultimate source are the same. This is the case, for instance, with Syriac ‫ ܗܪܛܝܩܐ‬hrṭyqʾ ‘heretical; 32.  Since this word is not a loanword but a single-word code-switch, it cannot serve as evidence for the use of the genitive as an input form for Greek loanwords in Syriac, as is done in Brock 1996: 255. 33.  See also ἀκριβῶς (LSJ 55) > ‫ ܐܩܪܝܒܘܣ‬ʾqrybws ‘exactly’ (91.2; LS2 45; SL 93), which is otherwise only found in the Lexicon of Bar Bahlul (ed. Duval 1888–1901: 278.2). 34.  For monolingualism as a criterion for distinguishing single-word code-switches from loanwords, see Haspelmath 2009: 40; Myers-Scotton 1993: 193. 35.  For this distinction, see Wohlgemuth 2009: 51.

54

Chapter 4

heretic’ (LS2 183; SL 354), which was transferred from Greek αἱρετικός (PGL 51). Greek is the immediate source since the word was transferred from Greek to Syriac, and Greek is the ultimate source since the word is a native Greek formation. There are, however, a number of loanwords in Syriac for which Greek is the immediate source, but it is not the ultimate source. Syriac ‫ ܣܛܪܦܐ‬sṭrpʾ ‘satrap’ (LS2 469; SL 998), for instance, is a loanword from Greek σατράπης (LSJ 1585). The Greek word, however, is itself a loanword from Old Iranian *xšaθra‑pā. 36 Thus, Greek is the immediate source of Syriac ‫ܣܛܪܦܐ‬ sṭrpʾ, but Old Iranian is the ultimate source. 37 The largest group of words for which Greek is the immediate source but not the ultimate source is the group of Latin words that are found in Syriac. 38 Conversely, there are loanwords in Syriac for which Greek is the ultimate source, but it is not the immediate source. Greek κλῇθρον (LSJ 957), for instance, is the ultimate source of Syriac ‫ܩܪܩܠ‬ qrql ‘grated cover’ (LS2 700; SL 1416). This Greek word, however, reached Syriac by way of Late Latin cracli, a form attested in the Appendix Probi. 39 Included within the group of words for which Greek is the ultimate source but not the immediate source are the Aramaic inheritances in Syriac that derive ultimately from Greek. 40 4.8.  Latin Loanwords in Syriac More than 100 Latin words are found in Syriac texts not translated from Greek that were written before Yaʿqub of Edessa (d. 708). 41 Most of these Latin words probably reached Syriac via Greek. 42 That is, Greek is usually the immediate source for Latin loanwords in Syriac. 43 In some cases, the phonology of the Syriac form is an indication that the word was transferred via Greek. The nasal n in Syriac ‫ ܦܠܛܝܢ‬plṭyn ‘palace’ (LS2 574; SL 1199), for instance, indicates that the immediate source was Greek παλάτιον (LLGE 85; LSJ 1291; see also Mason 1974: 74) and not Latin palatium (OLD 1284; LD 1291). 44 In addition, a majority of the Latin words found in Syriac are also at36.  Ciancaglini 2008: 28, 220–21. For the Iranian form, see Tavernier 2007: 436. 37.  For additional cases like this, see Ciancaglini 2008: 28. 38.  These are discussed in §4.8 below; and Butts forthcoming a. 39.  Baehrens 1922: 8 (s.v., line 209). Compare earlier Latin clathri (OLD 333; LD 350). 40.  These are discussed in §4.9 and collected in appendix 1. 41.  These are collected and analyzed in detail in Butts forthcoming a. 42.  So already Brock 1967: 424 n. 46; 1975: 90; 1996: 255; 1999–2000: 443; 2005: 23; Ciancaglini 2008: 7; Healey 1995: 83; Rochette 2010: 292; Schall 1960: 243–44; Wasserstein 1995: 134. 43.  For immediate source, see §4.7. A similar situation is attested for other Aramaic dialects; all of the Latin words in Palmyrene Aramaic, for instance, probably arrived by way of Greek (Brock 2005: 23). 44.  There are occasional cases in which the phonology points to Latin as the immediate source. The initial voiced bilabial stop of Syriac ‫ ܒܘܪܓܐ‬bwrgʾ ‘tower’ (LS2 95; SL 130)

Greek Loanwords in Syriac: The Methodological Framework

55

tested in Greek as loanwords. Thus, a possible Greek intermediary is known to have existed. It is interesting to note in this regard that most of the Latin words in Syriac are attested in the Greek papyrological record from Egypt and/or in Byzantine Greek. This suggests that these Latin loanwords were used in the Koinē Greek of the eastern Roman Empire (see §4.10), and it is in this way that many of them entered Syriac. A number of Latin words already appear in the earliest period of Syriac (pre-4th century) as well as in the Syriac New Testament. 45 Several of these words were transferred into Aramaic at an earlier period and then inherited in Syriac (see §4.9). Others reached Syriac (via Greek) due to the Roman presence in Syria and Mesopotamia already in the early centuries of the Common Era. The fact that many of the Latin words in the Syriac of this period are also attested in other languages of the Near East, whether Aramaic or not, indicates that they were widespread throughout the Roman Near East. Many of the Latin words that appear in the earliest period of Syriac (pre-4th century) as well as in the Syriac New Testament were probably a deeply ingrained part of the Syriac language by the time of fourth-century authors, such as Aphrahaṭ and Ephrem. The fourth century saw only a small addition to the number of Latin words in Syriac. A number of these words are rare throughout the history of Syriac, and a couple of them seem to be hapax legomena. The presence of rare words, and especially hapax legomena, suggests that Syriac speakers continued to be in contact with the Greco-Roman world in the fourth century. At the same time, however, the paucity of new Latin words in this period indicates a relatively low degree of contact. Contact between Syriac and the Greco-Roman world seems to have increased in the fifth century, when the number of new Latin words in Syriac rises. The appearance of a number of Latin words that are closer to Fremdwörter than they are to Lehnwörter corroborates such an increase in contact. It is also the fifth century that sees more Latin words in Syriac that are not found in other dialects of Aramaic or other languages of the Roman Near East, pointing to an increase in contact between Syriac and the Greco-Roman world relative to that of other languages and cultures. The sixth century represents the zenith of new Latin words in Syriac, pointing to a climax in contact between Syriac culture and the Greco-Roman world. The sixth century continues two trends from the fifth century: the appearance of a large number of Fremdwörter as well as the rise in the number of Latin words in Syriac that are not found in other languages of the Near East. Both of these trends point to a high degree of contact between Syriac and the Greco-Roman world in the sixth century. This is perhaps best illustrated by Yuḥanon of Ephesus (d. 581), who suggests, for instance, that the immediate source is Latin burgus (OLD 245; LD 255) and not Greek πύργος (LSJ 1556; see also Schall 1960: 50–51). 45.  Brock 1967: 424 n. 46; 1999–2000: 443–44.

56

Chapter 4

is known to have resided for a number of years in Constantinople, which was Greek speaking but whose imperial court was officially Latinate. 46 The Latin loanwords in Yuḥanon’s writings may be due to the particular sociolect of Syriac that was in use in Constantinople by Yuḥanon and his audience, which was more influenced by the imperial language of Latin. 47 The seventh century sees a sharp decline in the number of new Latin words in Syriac. This is at least partly due to the absence of a figure such as Yuḥanon of Ephesus in this period. It may also be due, however, to a decrease in contact compared with the sixth century. When analyzed diachronically, the Latin words in Syriac can serve as a thermometer—to borrow a metaphor of Brock—to gauge the changing degrees of contact between Syriac and the Greco-Roman world in Late Antiquity. 48 4.9.  Greek Loanwords as Inheritances in Syriac Greek had been in contact with the Semitic languages of the Near East for at least half a millennium by the time that Syriac emerged in the first centuries of the Common Era. 49 Thus, it comes as no surprise that Greek loanwords are found in Aramaic dialects prior to Syriac. The earliest Greek loanword in Aramaic is the monetary term στατήρ (LSJ 1634), which is first attested on the Abydos Lion Weight from ca. 500 b.c.e. (KAI 263). 50 This loanword is also found in Achaemenid Aramaic texts (TAD C3.7Ar2:3; 3.7Br1:13, 20) 51 as is an additional Greek loanword: πίναξ > pynk ‘plate’ (TAD D7.57:8). 52 The Aramaic of Daniel attests (at least) three Greek loanwords: 53 (4.9) a. κιθάρα, κίθαρις (LSJ 950) > qytrws (Kethiv), qa ṯros (Qere) ‘zither’ (HALOT 1970); compare Syriac ‫ ܩܝܬܪܐ‬qytrʾ (LS2 705; SL 1366) b. συμφωνία (LSJ 1689) > sumponyɔ (Dan 3:5, 15), sypnyh (Dan 3:10 [Kethiv]), supponyɔ (Dan 3:10 [Qere]) ‘symphony’ (HALOT 1937–38); compare Syriac ‫ ܨܦܘܢܝܐ‬ṣpwnyʾ (LS2 635; SL 1297) c. ψαλτήριον (LSJ 2018) > psanṭerin ‘psaltery’ (HALOT 1958) Other dialects of Middle Aramaic also attest Greek loanwords, including the Aramaic of Targum Onqelos and Jonathan (Dalman 1905: 182–87), Nabatean 46.  For Yuḥanon of Ephesus, see Harvey 1990. For his language use, see the brief remarks in Millar 2013a: 109. 47. The Ecclesiastical History of Pseudo-Zacharias (6th cent.) is similar in this regard (ed. Brooks 1919–24). 48.  Brock 1999–2000. 49.  Brock 1996: 251; 1998: 713. 50.  Brock 1996: 251 n. 2; DNWSI 805. 51.  Brock 1967: 418; 1996: 251; DNWSI 805; Muraoka and Porten 1998: 377. 52.  DNWSI 910; Muraoka and Porten 1998: 377. 53.  Brock 1975: 84; Kutscher 1970: 401–2; Rosenthal 1995: §191; Wasserstein 1995: 135.

Greek Loanwords in Syriac: The Methodological Framework

57

Aramaic (Healey 1995), Palmyrene Aramaic (Cantineau 1935: 155; Brock 2005), and Ḥatran Aramaic (Contini and Pagano 2015). Finally, Greek loanwords are found not only in Syriac but in all of the Late Aramaic dialects. In the current study, it is important, whenever possible, to account for how a particular Greek loanword in Syriac relates to the same Greek loanword in other Aramaic dialects, whether contemporary or earlier. 54 Consider, for instance, the Greek word πίναξ ‘board, plank’ (LSJ 1405). In Aramaic, it first appears as a loanword in an Aramaic text from Egypt that dates to the late third century b.c.e. (TAD D7.57:8). In Syriac, the Greek loanword is found already in the New Testament translations, both Old Syriac and Peshiṭta (Brock 1967: 413–14), as well as in texts not translated from Greek, beginning with the fourth-century authors Aphrahaṭ (Demonstrations, 1.729.3 [citing Matt 23:25; ed. Parisot 1894–1907]) and Ephrem (Maḏrɔše on the Nativity, 104.13 [ed. Beck 1959]; Maḏrɔše on Nisibis, 2.87.12 [ed. Beck 1963]). In addition to Syriac, the Greek word appears in the Late Aramaic dialects of Jewish Babylonian Aramaic (DJBA 901), Jewish Palestinian Aramaic (DJPA 431), Christian Palestinian Aramaic (DCPA 327; LSP 156), and Samaritan Aramaic (DSA 690). So, was this Greek word transferred into each of these dialects independently? Or was it transferred into one early dialect and then inherited in later dialects? Or is some combination of these two options possible? Or is there another explanation altogether? There is evidence suggesting that Greek loanwords were transferred between Aramaic dialects. This, for instance, seems to be the case with the verbal root qṭrg ‘to accuse’, which is found in Syriac (LS2 657, 663; SL 1348, 1358–59) as well as Jewish Palestinian Aramaic (DJPA 489), Christian Palestinian Aramaic (DCPA 370; LSP 178), and Samaritan Aramaic (DSA 775). The Greek source for this root is either the noun κατήγορος (LSJ 927) or the infinitive κατηγορεῖν (LSJ 926–27). The Greek source, regardless of whether it was a noun or infinitive, has the voiced velar stop γ followed by the alveolar trill ρ. Each of the Aramaic dialects, however, attests the reverse order, with the alveolar trill preceding the voiced velar stop. 55 There is no regular sound change in Aramaic to account for this development, and so it is necessary to posit an ad hoc change. Given that it is such an irregular change, it is unlikely that this root metathesis would have occurred independently in each of the four Late Aramaic dialects that attest the word; this would after all be an extreme example of drift. It is more likely that the Greek word was transferred into one 54.  For similar considerations involving Iranian loanwords in Syriac, see Ciancaglini 2008: 25–28. 55. It should be noted that an unmetathesized form is occasionally found in Syriac with the noun κατήγορος (LSJ 927) + -ɔnɔ > ‫ ܩܛܓܪܢܐ‬qṭgrnʾ (LS2 657; SL 1350), alongside ‫ ܩܛܪܓܢܐ‬qṭrgnʾ (LS2 663; SL 1359).

58

Chapter 4

dialect of Aramaic, then the (irregular) root metathesis occurred, and only then the word was transferred to other dialects of Aramaic. The example of qṭrg establishes that in at least some cases Greek loanwords were transferred among Aramaic dialects. This leads to a new series of questions: Are these cases of transfer inheritance from mother language to daughter language? Or are they contact-induced transfers among Aramaic dialects? As established by Boyarin (1981), the Late Aramaic dialects cannot be divided into traditional sibling-type relationships with a mother in the Middle Aramaic period. That is, the late West Aramaic dialects of Samaritan Aramaic, Christian Palestinian Aramaic, and Jewish Palestinian Aramaic do not share a common genetic source that is attested in the previous period of Middle Aramaic. This has important implications for the current series of questions, since it renders it impossible for a Greek word to have been transferred into a hypothetical proto-Late Aramaic, or even proto-Late West Aramaic, and then be inherited by each of the daughter languages. Rather, a Greek loanword would have had to be transferred into an Aramaic dialect, then transferred from there to other dialects of Aramaic, and only from these other dialects inherited into the Late Aramaic dialects attested in the historic record. This scenario was likely facilitated by the existence of a high literary dialect such as Standard Literary Aramaic. 56 This supradialect could have served as a repository of Greek loanwords, which would then have been transferred into other dialects, such as the dialect that would later have become Syriac. To illustrate this process, it is worth returning to the example of πίναξ ‘board, plank’. Given the history of Aramaic, one possible scenario would involve the transfer of this word from Greek into the Aramaic dialect attested in TAD D7.57. From this dialect, the word would then have been transferred into other Aramaic dialects, including potentially the precursors of Syriac, Jewish Babylonian Aramaic, Jewish Palestinian Aramaic, Christian Palestinian Aramaic, and Samaritan Aramaic. From these proto-languages, the Greek word would have been inherited by the dialects of Late Aramaic that preserve the word. This scenario could of course be complicated by interdialectical transfer at various stages, including in earlier times, with Standard Literary Aramaic serving as a conduit as well as in Late Aramaic times. This would not, however, significantly affect the outcome. In the case of πίναξ, then, it is incorrect to suppose that Syriac and the other Late Aramaic dialects inherited the loanword directly from the dialect of Aramaic attested in TAD D7.57, where the word is first found. This cannot be the case, since Syriac is not a later form of the Aramaic dialect attested in TAD D7.57. At the same time, however, this dialect could have served as the source for the word in Syriac. In this scenario, Syriac would have inherited the word 56.  For Standard Literary Aramaic, see Greenfield 1974.

Greek Loanwords in Syriac: The Methodological Framework

59

from proto-Syriac, which received the word from the Aramaic dialect attested in TAD D7.57, possibly via Standard Literary Aramaic. Thus, Greek would not be the immediate source of the word; rather, it would have been inherited from earlier Aramaic by Syriac. It can be concluded, then, that Syriac probably contains Greek loanwords that were inherited from an earlier stage of Aramaic as well as Greek loanwords for which Greek was the immediate source. 57 The question is how to identify the inherited words. One potential criterion is the attestation of Greek loanwords in other dialects of Aramaic. More than 60 Greek loanwords in Syriac are also attested in Aramaic dialects prior to Late Aramaic—that is, Achaemenid Aramaic (ca. 583–ca. 333 b.c.e.) and Middle Aramaic (ca. 333 b.c.e.–ca. 200 c.e.). 58 The majority of these are attested in Syriac by at least the fourth century. This is the case, for instance, with the previously discussed example of πίναξ ‘board, plank’ (LSJ 1405) > ‫ ܦܝܢܟܐ‬pynkʾ ‘dish, writing tablet’ (LS2 579; SL 1188). Consequently, if a Greek loanword is attested both in an Aramaic dialect from the Middle Aramaic period or earlier and in Syriac by the fourth century, it seems likely that it was transferred into Aramaic at an earlier period and inherited by Syriac. A list of all the words fulfilling these criteria is given in appendix 1 at the end of this book. There are a few Greek loanwords that are attested in Aramaic dialects prior to Late Aramaic but are not attested in Syriac by the fourth century: ̈ (4.10) a. βασιλική (LSJ 309–310) > ‫ܒܣܝܠܝܩܘܣ‬ bsÿlyqws ‘colonnade, portico’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 274.4 [ed. Brooks 1935]; LS2 80, 940; SL 164–65), already in Palmyrene bslqʾ (PAT 63; 71; see also Brock 2005: 13) b. δόγμα (PGL 377–78; LSJ 441) > ‫ ܕܘܓ�ܡܐ‬dwgmʾ ‘doctrine’ (6th cent. Yuḥanon of Ephesus, Lives of the Eastern Saints, 583.6 [ed. Brooks 1923–25]; LS2 141; SL 277–78), already in Palmyrene dgm ‘decree, decision’ (PAT 355; see also Brock 2005: 15); see also Jewish Palestinian Aramaic dygmʾ, dwgmʾ ‘illustration, model, example’ (DJPA 145, 830) c. ὁμολογία (PGL 957–58; LSJ 1226) > ‫ ܐܡܘܠܓܝܘܣ‬ʾmwlgyws, ‫ ܐܡܠܘܓܝܣ‬ʾmlwgys ‘confession of faith’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 117.26; 131.20 [ed. Brooks 1935]; LS2 25; SL 53), already in Palmyrene ʾmlgyʾ ‘contract [context damaged, sense uncertain]’ (PAT 339–40; see also Brock 2005: 19) 57.  For immediate source, see §4.7. Syriac could of course also contain Greek words from different immediate sources, such as Latin or another Late Aramaic dialect. 58. The dates of these periods are those of Folmer (1995) and differ from those of Fitzmyer (1979).

60

Chapter 4 d. ὕπαρχος (LSJ 1853) > ‫ ܗܘܦܪܟܐ‬hwprkʾ, ‫ ܐܘܦܪܟܐ‬ʾwprkʾ ‘prefect’ (5th cent. Teaching of Addai, 38.21 [ed. Howard 1981]; LS2 43, 182; SL 19; 338), already in Nabatean hprkʾ (DNWSI 292; see also Healey 1993: 108–9; this word has, however, also been connected to ἔπαρχος); see also Late Jewish Literary Aramaic hprkʾ (Tg. Esth. I 10:1; Jastrow 363)

Given that these words are not attested in Syriac until a later period, it is less certain that Syriac inherited them from earlier Aramaic. In fact, these may well be instances in which a word was independently transferred from Greek into different dialects of Aramaic. This is almost certainly the case for some of the words, such as δόγμα, since the loanword in Syriac differs in meaning from the other Aramaic dialect. In the Late Aramaic period, it becomes more difficult to use comparative Aramaic evidence to determine whether or not a Greek loanword in Syriac is an inheritance. This is due to the fact that each of the Late Aramaic dialects is known to have had contact with Greek, though to varying degrees. 59 Given this contact, it is impossible to exclude that a given loanword underwent cases of independent transfer from Greek into multiple dialects of Late Aramaic. Independent cases of transfer in fact seem likely in a number of cases based on the late date of the first occurrence of a loanword in Syriac. Consider, for instance, Greek ταξιώτης, ταξεώτης ‘imperial bodyguard’ (LSJ 1756), which occurs in Jewish Palestinian Aramaic (DJPA 230) and in Syriac (LS2 275; SL 529). Theoretically, this word could have been inherited from earlier Aramaic or transferred into each of the two dialects independently. 60 The latter is, however, by far the more likely scenario in this case since the word in question is not attested in Syriac until the sixth century when it appears in the Ecclesiastical History by Yuḥanon of Ephesus (part 3, 9.18; 158.17 [ed. Brooks 1935]). Thus, given their individual histories of contact with Greek, the Late Aramaic dialects do not provide reliable evidence for determining whether or not a Greek loanword in Syriac has been inherited from earlier Aramaic. In the end, comparative Aramaic evidence provides a criterion for identifying some of the Greek loanwords in Syriac that were inherited from earlier Aramaic. It is not, however, possible to identify all of them. Many of the Greek loanwords that are attested in the earliest layer of Syriac could well have been inherited from earlier Aramaic, and so they would not be the result of language contact between Syriac—at least as it is traditionally understood—and Greek. Nevertheless, pending the discovery of extensive documentation of the Aramaic ancestor of Syriac, it is unlikely that it will ever be possible to identify these inheritances with any degree of certainty. 59.  See the discussion on pp. 206–210 below. 60.  It is also possible that the word was transferred from one of the dialects to another.

Greek Loanwords in Syriac: The Methodological Framework

61

4.10.  The Greek Source The Greek language with which Syriac speakers were in contact was not the Attic of the classical period but, rather, Koinē and then early Byzantine Greek. Koinē Greek developed from Attic in the Hellenistic period and quickly spread over the classical world as well as over much of the ancient Near East. 61 Koinē Greek eventually gave way to the Greek of the Byzantine Era. 62 The best sources for the Greek with which Syriac speakers were in contact are the inscriptions and documents that were written in Late Antique Syria and Mesopotamia. 63 In a majority of cases, Greek loanwords in Syriac reflect Attic Greek. This should not be too surprising since Attic continued to exert significant influence on the orthography of Koinē Greek. That is, the majority of Greek texts, especially literary texts but even documents from Late Antiquity continue to reflect, or at least strive for the orthography of Attic Greek, regardless of how well it reflects the actual pronunciation of the language. Occasionally, however, Greek loanwords in Syriac reflect non-Attic forms that also appear in the inscriptions and documents from Late Antique Syria and Mesopotamia. Greek documents from Syria and Mesopotamia, for instance, attest an assimilation of κ to γ before a voiced stop. 64 This assimilation of [+ voice] is also reflected in the g in ‫ ܐܩܠܝܣܓܕܝܩܘܣ‬ʾqlysgdyqws ‘expert in church law’ (LS2 44; SL 92). This suggests that the Syriac word was transferred from a Koinē form of Attic ἐκκλησιέκδικος (PGL 433). Or, to take a different example, μ assimilates to ν before a labial in the Koinē of Syria and Mesopotamia. 65 This assimilation accounts for the first n in Syriac ‫ ܣܘܢܦܢܘܣ‬swnpnws ‘supervisor of the trades people of Constantinople on behalf of the eparch of the city’ (LS2 485; SL 984), which can be contrasted with the μ in the Attic form σύμπονος (LSJ 1685; PGL 1289). These as well as other examples will be discussed in 61.  Browning 1983: 19–52; Horrocks 2010: 79–188. 62.  Browning 1983: 53–68; Horrocks 2010: 189–369. 63.  Publication information for these Greek texts is discussed above on pp. 29–30. 64.  This is reflected in writings such as ἐγδικίας for ἐκδικίας (P. Euphrates 2.13 [mid3rd]); διεγδικήσειν for διεκδικήσειν (P. Euphrates 9.22–23 [252]); ἐγβένω for ἐκβαίνω (P. Euphrates 17.22 [mid-3rd]); ἐγ διακληρώσεως for ἐκ διακληρώσεως (P. Dura 19.6 [88–89]). This assimilation of [+ voice] is also found in the Koinē Greek of Egypt (Gignac 1976–81: 1.6–80; Mayser 1970: 143–44). 65.  This assimilation is reflected in writings such as διαπενψαμένου for διαπεμψαμένου (P. Euphrates 2.20 [mid-3rd]); ἐνποιηθῇ for ἐμποιηθῇ (P. Euphrates 8.24 [251]); ἐνποιούμενον for ἐμποιούμενον (P. Euphrates 9.23 [252]); ἔνπροσθεν for ἔμπροσθεν (P. Euphrates 16.A.2 [after 239]); ἐ̣νφράξι for ἐμφράξει (P. Euphrates 13.16 [243]); συνβά[ν] for συμβάν (P. Euphrates 2.5 [mid-3rd]); συνπαρόντος for συμπαρόντος (P. Euphrates 6.9 [249]; 9.14 [252]). This assimilation is also attested in the Koinē of Egypt (Gignac 1976–81: 1.167–69; Mayser 1970: 203–7).

62

Chapter 4

more detail in §5.2.3 below. For now, however, it is important to note that occasionally the Greek loanwords in Syriac reflect the Koinē Greek of Late Antique Syria and Mesopotamia. The Greek loanwords in Syriac, thus, serve as a witness to the Greek of Late Antique Syria and Mesopotamia (as so-called Nebenüberlieferungen). This is important because the number of Greek documents from Syria and Mesopotamia is quite limited compared with the extensive material found in Egypt. One of the questions that the abundance of the Egyptian material and the paucity of other material often raise is whether or not the Greek documents from Egypt are representative—in language, history, economics, etc.—of the broader Late Antique Near East. The Greek loanwords in Syriac, as Nebenüberlieferungen for the Greek of Syria and Mesopotamia, suggest that the Egyptian papyri are in some respects representative of a Koinē Greek spread across the Roman Near East. In Greek documents from Egypt, for instance, π is commonly deleted in the cluster μπτ. 66 This deletion is also attested in Syriac ‫ ܩܡܛܪܝܢ‬qmṭryn ‘small chest’ (LS2 672; SL 1377), which can be compared with the Attic form κάμπτριον (LSJ 873). Syriac ‫ ܩܡܛܪܝܢ‬qmṭryn thus suggests that κάμτριον was the Koinē form in Late Antique Syria and Mesopotamia. The fact that the Greek loanword in Syriac reflects a sound change attested in the Greek papyri from Egypt implies that this sound change was not restricted to Egypt but, rather, it extended across the Roman Near East. Similarly, Greek γ is occasionally written instead of κ in word-initial position in the Greek papyri from Egypt, as in γυβερνήτης (P. Grenf. 1.49.21 [220/221]) for Attic κυβερνήτης (LSJ 1004). 67 The voiced velar stop, as opposed to the voiceless velar stop, is also found for this same word in Syriac ‫ ܓܘܒܪܢܝܛܐ‬gwbrnyṭʾ ‘helmsman, pilot’ (LS2 103; SL 210), which is attested once in Aphrahaṭ’s Demonstrations (1.612.2; ed. Parisot 1894–1907), against the much more common spelling ‫ ܩܘܒܪܢܝܛܐ‬qwbrnyṭʾ (with orthographic variants; LS2 645; SL 1323). Again, the agreement between the Greek loanword in Syriac and the writing in the Greek papyri from Egypt suggests that γυβερνήτης with its initial voiced velar stop was a common Koinē form across the Roman Near East. 68 The correspondence between the Greek papyri from Egypt and the Greek loanwords in Syriac is not restricted to phonology but extends also to morphology and lexicon. In the Greek papyri from Egypt, for instance, the ending -ιον is often realized as -ιν. 69 Thus, the frequent use of Syriac -yn to represent this ending almost certainly reflects a Koinē form -ιν and not the Attic form -ιον. 66.  Gignac 1976–81: 1.64; Mayser 1970: 152. 67.  Gignac 1976–81: 1.77; Mayser 1970: 143–44. 68.  The voiced velar stop is also reflected in Latin gubernare (LD 831), which is a loanword from Greek. 69.  Gignac 1981: 2.25–29.

Greek Loanwords in Syriac: The Methodological Framework

63

This is, for instance, the case with γυμνάσιον (LSJ 362) > ‫ ܓܡܢܣܝܢ‬gmnsyn ‘gymnasia’ (LS2 121; SL 242) and Latin palatium (OLD 1284; LD 1291) > παλάτιον (LLGE 85; LSJ 1291) > ‫ ܦܠܛܝܢ‬plṭyn ‘palace’ (LS2 574; SL 1199). Thus, Syriac loanwords establish that, like the Greek of Egypt, the Greek of Syria and Mesopotamia had -ιν for Attic -ιον. 70 With regard to lexicon, it is well known that Latin had a significant influence on Koinē Greek. 71 This is probably nowhere more evident than in the large number of Latin loanwords that occur in Greek papyrological texts from Egypt. 72 Interestingly, the majority of the Latin words in Syriac are also attested in the Greek papyri from Egypt (see §4.8). This suggests that these Latin words were part of the broader Koinē of the eastern Roman Empire. To summarize, most Greek loanwords in Syriac reflect the orthography of Attic Greek of the classical period. Occasionally, however, the Greek loanwords in Syriac reflect forms found in the inscriptions and documents that were written in Late Antique Syria and Mesopotamia. In addition, the Greek loanwords in Syriac at times attest a form that is also found in the Greek papyri from Egypt. In these cases, it is possible to posit a common Koinē form that was spread through the eastern Roman Empire. Given this situation, throughout this book, Greek forms are cited not only from Attic Greek but also at times from the Greek documents from Syria and Mesopotamia as well as from Egypt. 4.11. Conclusion With the methodological framework now established, chaps.  5–6 below provide a contact linguistic analysis of the Greek loanwords in Syriac. Chapter 5 analyzes the phonological integration of Greek loanwords in Syriac, while chap. 6 focuses on morphosyntactic integration. 70.  This is confirmed by the Greek documents from Syria and Mesopotamia, which also have -ιν for Attic -ιον (Welles, Fink, and Gilliam 1959: 48)—for example, δελματ̣ίκιν for δελματ̣ίκιον (P. Dura 30.17 [232]); σεισ̣ ύ̣ρι̣ ν̣ ̣ for σεισύριον (P. Dura 33.13 [240–250]). 71. See Browning 1983: 40–42, 67–68; Rochette 2010: 291–292. 72.  These are collected in Daris 1991.

Chapter 5

The Phonological Integration of Greek Loanwords in Syriac Likewise, other languages have certain letters that the rest of languages are unable to pronounce. As for Syriac speakers, by which I mean the speech of Edessa, it is not their language that prevents them [from this], but it is their writing system because of its incompleteness and its lack of vowel signs. (Yaʿqub of Edessa, The Correctness of Speech) 1

5.1. Overview Although a fair amount of literature has been devoted to Greek loanwords in Syriac, very little attention has thus far been paid to their phonological integration. In the standard grammar of Syriac, Nöldeke (1904) discusses this topic in only a handful of paragraphs. 2 Other grammars, such as those by Brockel­mann (1981) and Muraoka (2005), offer even fewer remarks. In the only monographic study of Greek loanwords in Syriac, Schall (1960) makes a number of passing references to phonological integration, 3 but he never provides a systematic treatment. More recently, Brock (1996: 254–57) and Voigt (1998b) have provided additional insights; neither, however, offers a comprehensive description. In the present chapter, I aim to provide an updated analysis of the phonological integration of Greek loanwords in Syriac. I begin with the integration of consonants (§5.2) and then turn to vowels (§5.3). In contrast to changes in syntax or lexicon, diachronic changes in orthography present a special challenge since even meticulous Syriac scribes, who 1.  The Syriac text is edited in Wright 1871b: 2*.a.5–12. An English translation of the same quotation can be found in Kiraz 2012: 59, where it is mistakenly said to come from Yaʿqub’s Letter on Syriac Orthography (ed. Phillips 1869). 2.  See, e.g., Nöldeke 1904: §4B, 25, 39, 40H, 46, 51. 3.  See, e.g., Schall 1960: 37, 42–44, 50–51, 61–62, 80, 93, 99, 104, 107, 108, 109, 110, 111, 120, 121, 135–36, 148–50; 174, 217, 220, 232, 245.

64

The Phonological Integration of Greek Loanwords in Syriac

65

were generally hesitant to make drastic changes at the word level, were prone to update the orthography of the manuscript before them. 4 This is known anecdotally through Yaʿqub of Edessa (d. 708), who in his Letter on Syriac Orthography implores later scribes not to change his chosen orthography for various words, including Greek loanwords (Phillips 1869: 6.1–8). In addition, many cases of textual transmission betray scribal updates of orthography. This is especially clear in the works of Ephrem, where the fifth- and sixth-century manuscripts from Dayr al-Suryān often preserve an older orthography compared with the later liturgical manuscripts. The Greek loanword ἀγών (PGL 25; LSJ 18–19), for instance, appears in the earlier spelling of ‫ ܐܝܓܘܢܐ‬ʾygwnʾ in ms Brit. Libr. Add. 14,627 (sixth century) but in the standardized spelling of ‫ ܐܓܘܢܐ‬ʾgwnʾ in the later liturgical ms Brit. Libr. Add. 14,506 (9th–10th century; ed. Beck 1964b: 10.14). 5 Given that scribes are known to have updated the orthography of Greek loanwords, it has been necessary in a few instances in this chapter to account not only for the date of composition of a work but also for the date of the manuscript that contains the work. 6 5.2. Consonants 5.2.1.  The Consonantal Inventories of Koinē Greek and Syriac The consonantal inventory of Koinē Greek contained 16 phonemes, which are summarized in table 5.1. 7 The consonantal inventory of Koinē Greek differs only slightly from that of Attic Greek. Attic Greek was characterized by a symmetrical system of 9 stops, with 3 manners of articulation (voiceless unaspirated [κ, π, τ], voiceless aspirated [χ, φ, θ], and voiced [γ, β, δ]) and 3 places of articulation (bilabial [β, π, φ], dental [δ, τ, θ], and velar [γ, κ, χ]). By the Koinē Greek of the Roman period, the voiceless aspirated stops had become voiceless fricatives—that is, *ph > f, *th > θ, and *kh > x. Similarly, the voiced stops eventually became fricatives, as in Modern Greek—that is, 4.  This is not to say that scribes did not also make changes on the word level. To give but one example, ms New Haven, Yale Syriac 5 (1888) preserves the same recension of the Syriac History of Cyriacus and His Mother Julitta as the earlier manuscript in London, Library of the Royal Asiatic Society (1569); it, however, attests extensive syntactical and lexical variants that are best explained as scribal intervention. A different recension (probably translation) of this text is edited in Terpelyuk 2009. 5.  For the dates of these manuscripts, see Wright 1870–72: 2.415, 1.247–49, respectively. 6.  Ideally, future studies of Greek loanwords in Syriac—or for that matter, Syriac grammatical studies more generally—will be better able to account for both date of composition and date of copying. A good model is provided by Hittotology, where it has become increasingly common to refer to the date of the original composition as well as to the date of script, following the conventions of The Chicago Hittite Dictionary—for example, OH/OS is an Old Hittite composition preserved in Old Script whereas OH/NS is an Old Hittite composition preserved in New Script. 7.  In general, see Allen 1987; Gignac 1976–81: 1.63–179; Mayser 1970: 141–217.

66

Chapter 5

Table 5.1.  Consonantal Inventory of Koinē Greek

Liquid Fricative

Stop

Bilabial Dental Alveolar Velar

Nasal

voiceless unaspirated p (π)

t (τ)

k (κ)

voiced

b (β)

d (δ)

g (γ)

voiceless

ɸ (φ)

θ (θ)

s (σ)

voiced

z (ζ)

trill

r (ρ)

lateral approximant

l (λ) m (μ)

x (χ)

Glottal

h (spiritus asper)

n (ν)

*b > β, *g > ɣ, and *d > ð, though it is difficult to date this change precisely. 8 In addition to the stops, there were 4 sonorants in Attic Greek as well as in Koinē Greek. Two of these were liquids, one being an alveolar lateral approximant (λ) and the other being a voiced alveolar trill (ρ). The remaining 2 liquids were nasals, one being bilabial (μ) and the other being alveolar (ν). Alongside these 2 nasal phonemes, there was a velar nasal, which was an allophone of the alveolar nasal and the voiced velar stop. In addition to the stops and sonorants, there were 2 voiceless fricatives in Attic Greek, one alveolar (σ) and the other glottal (spiritus asper). The latter was lost sometime in the Late Antique period. 9 Attic and Koinē Greek also possess several monographs: ξ represents the voiceless unaspirated velar stop κ plus the voiceless alveolar fricative σ; ψ represents the voiceless unaspirated bilabial stop π plus the voiceless alveolar fricative σ. In Attic Greek, ζ was a monograph for /zd/; however, it had developed into a voiced alveolar fricative /z/ by the Koinē Greek of the Roman and Byzantine periods. The consonantal inventory of Classical Syriac included 28 phonemes, which are summarized in table 5.2. 10 Syriac was characterized by several sets of consonantal triads consisting of a voiceless, voiced, and emphatic member. The emphatic member, which is traditionally represented with an under-dot in Semitic studies, was likely glottalic/ejective in earlier stages of Semitic; however, it was probably realized as pharyngeal in Syriac, as in Arabic. 11 Triads 8.  See the discussions in Allen 1987: 29–32; Browning 1983: 26–28; Gignac 1976–81: 1.68–76. 9.  In general see, Harviainen 1976 as well as §§5.2.4, 5.2.5 below. 10.  In general, see Daniels 1997; Muraoka 2005: §3; Nöldeke 1904: §2. 11. For the emphatic consonsants in Semitc, see Kogan 2011: 59–61 with further references.

The Phonological Integration of Greek Loanwords in Syriac

67

p̄ (‫)ܦ‬ ܼ

ṯ (‫)ܬ‬ ܼ

s (‫)ܣ‬

voiced

ḇ (‫)ܒ‬ ܼ

ḏ (‫)ܕ‬ ܼ

z (‫)ܙ‬

Stop Fricative

ܿ k (‫)ܟ‬ ܿ g (‫)ܓ‬

ṣ (‫)ܨ‬

trill

r (‫)ܪ‬

lateral approximant

l (‫)ܠ‬ m (‫)ܡ‬

Glide

w (‫)ܘ‬

ʾ (‫)ܐ‬

q (‫)ܩ‬

emphatic

Nasal

Glottal

voiceless

Pharyngeal

ܿ [ṗ (‫])ܦ‬ ṭ (‫)ܛ‬

Velar

emphatic

Palatal

Dental ܿ t (‫)ܬ‬ ܿ d (‫)ܕ‬

Palato-Alveolar

Bilabial voiced

ܿ p (‫)ܦ‬ ܿ b (‫)ܒ‬

voiceless

Liquid

Alveolar

Table 5.2.  Consonantal Inventory of Syriac

š (‫)ܫ‬

ḵ (‫)ܟ‬ ܼ

ḥ (‫)ܚ‬

h (‫)ܗ‬

ḡ (‫)ܓ‬ ʿ (‫)ܥ‬ ܼ

n (‫)ܢ‬ y (‫)ܝ‬

were found for the dental stops (‫ ܬ‬t, ‫ ܕ‬d, ‫ ܛ‬ṭ), the velar stops (‫ ܟ‬k, ‫ ܓ‬g, ‫ ܩ‬q), and the alveolar fricatives (‫ ܣ‬s, ‫ ܙ‬z, ‫ ܨ‬ṣ). In Syriac, the bilabial stops (‫ ܦ‬p, ‫ܒ‬ b) and pharyngeal fricatives (‫ ܚ‬ḥ [= IPA ħ], ‫ ܥ‬ʿ ) lacked an emphatic member and so had only voiced and voiceless members. 12 It should be noted, however, that an emphatic member did exist for the bilabial stop series in Greek loanwords in Syriac. 13 Sometime after the Old Aramaic period, the non-emphatic bilabial, dental, and velar stops developed fricative allophones postvocalically when ungeminated. 14 By the time of Syriac, these fricatives (both voiced and voiceless) seem to have become phonemic, since the conditioning factor of the allophone was in many cases lost due to a regular vowel deletion rule. This led to minimal pairs such as *garbā́ > garbɔ́ ‘leper’ (LS2 130; SL 255) 12.  An emphatic velar fricative *xʾ may well have existed in Proto-Semitic, as argued by Huehnergard (2003) on the basis of correspondences of Akkadian ḫ and West-Semitic ḥ. The existence of an emphatic bilabial stop *pʾ in Proto-Semitic is unlikely (see the discussion, with bibliography, in Kogan 2011: 80–81; Militarev and Kogan 2000: cv–cvicxvi). 13.  This is discussed below in §5.2.6. 14.  In Semitic Studies, these fricatives are traditionally indicated by an underbar or overbar—that is: p̄ = IPA f; ḇ = IPA β; ṯ = IPA θ; ḏ = IPA ð; ḵ = IPA x; ḡ = IPA ɣ.

68

Chapter 5

versus *garibā́ > garḇɔ́ ‘leprosy’ (LS2 130; SL 255) and *qaṭalatíh > qṭalṯeh ‘she killed him’ versus *qaṭaltíh > qṭalteh ‘I killed him’. 15 The innovative bilabial, dental, and velar fricatives (both voiced and voiceless) were not distinguished from their stop counterparts in the consonantal writing system of Syriac, though diacritics were eventually developed to differentiate them. 16 In addition to the stops and fricatives that occur in triads or biads, there were two glottal phonemes, one being a voiceless stop (‫ ܐ‬ʾ ) and the other being a voiceless fricative (‫ ܗ‬h), as well as a palato-alveolar voiceless fricative (‫ ܫ‬š). Alongside the stops and fricatives, there were six sonorants in Classical Syriac. Two of these were liquids, one being an alveolar lateral approximant (‫ ܠ‬l) and the other being an alveolar trill (‫ ܪ‬r). Two of these were nasals, one being bi­labial (‫ ܡ‬m) and the other being alveolar (‫ ܢ‬n). The remaining two were glides, one being bilabial (‫ ܘ‬w) and the other being palatal (‫ ܝ‬y). 5.2.2.  The Representation of Greek Consonants The representation of Greek consonants in Syriac is remarkably regular. 17 In most cases, each Greek consonantal phoneme is represented by a single consonant in Syriac. These correspondences are summarized in table 5.3. Correspondences that deviate from this chart are usually the result of one of two causes. First, a Koinē form of Greek probably served as the source for some of the words that prima facie seem to exhibit irregular correspondences. This is the case, for instance, with the initial consonant of ‫ ܓܘܒܪܢܝܛܐ‬gwbrnyṭʾ ‘helmsman, pilot’ in Aphrahaṭ’s Demonstrations (ed. Parisot 1894–1907: 1.612.2; see also LS2 103; SL 210), which does not derive from Attic Greek κυβερνήτης (LSJ 1004) but from Koinē γυβερνήτης, a form that is attested in Greek documents from Egypt (P. Grenf. 1.49.21 [220/221]). Cases such as this will be discussed in further detail below (§5.2.3). Second, some of the seemingly irregular correspondences are due to secondary developments. This is the case, for instance, with πινακίδιον (LSJ 1405) > ‫ ܦܢܩܝܬܐ‬pnqytʾ ‘writing tablet, treatise; collection; small book, volume’ (LS2 580; SL 1207). This loanword is also attested once in Syriac as ‫ ܦܢܩܝܕܬܐ‬pnqydtʾ, which represents the Greek source more closely. The form ‫ ܦܢܩܝܬܐ‬pnqytʾ is the result of a regressive assimilation of d to t in Syriac. 18 Excluding cases subsumed under these two categories, there are very few irregular correspondences. 15.  This distinction was extended by analogy to other places in the verbal system: ‫ܚܕܝܬ‬ ܼ ܿ ḥḏiṯ ‘I rejoiced’ versus ‫ܚܕܝܬ‬ ḥdit ‘you (m.sg.) rejoiced’. 16.  Kiraz 2012: §§210–16. 17.  For additional details and analysis, see Butts forthcoming d. 18.  So already Wright 1870–72: 2.633. For this sound change, see Nöldeke 1904: §26B ̄ ḥ(d)tʾ. It should be noted, and compare, e.g., *ḥadtā > *ḥadtɔ > ḥattɔ ‘new’, written ‫ܚܕܬܐ‬ however, that in the later vocalization tradition ‫ ܦܢܩܝܬܐ‬pnqytʾ is realized as /pɛnqiṯɔ/ with ṯ (not the expected tt); see, e.g., Luke 1:63 in Pusey and Gwilliam 1901. The fricativization of

The Phonological Integration of Greek Loanwords in Syriac

69

Table 5.3.  Representation of Greek Consonants in Syriac Greek

Syriac

Syriac

Greek

β

‫ܒ‬b

‫ܐ‬ʾ

spiritus asper (initial)

γ

‫ܓ‬g

‫ܒ‬b

β

 γ /ŋ/

‫ܢ‬n

‫ܓ‬g

γ

δ

‫ܕ‬d

‫ܕ‬d

δ

ζ

‫ܙ‬z

‫ܗ‬h

spiritus asper (initial)

θ

‫ܬ‬t

‫ܙ‬z

ζ

κ

‫ܩ‬q

‫ ܛ‬ṭ

τ

λ

‫ܠ‬l

‫ܟ‬k

χ

μ

‫ܡ‬m

  ‫ ܟܣ‬ks ξ

ν

‫ܢ‬n

‫ܠ‬l

λ

ξ

‫ ܟܣ‬ks

‫ܡ‬m

μ

π

‫ܦ‬p

‫ܢ‬n

ν , γ /ŋ/

ρ

‫ܪ‬r

‫ܣ‬s

σ

 ῥ-

‫ ܪܗ‬rh, ‫ ܪ‬r

‫ܦ‬p

π, φ

 ῤῥ

‫ ܪܗ‬rh, ‫ ܪ‬r

  ‫ ܦܣ‬ps ψ

σ

‫ܣ‬s

‫ܩ‬q

κ ρ

τ

‫ ܛ‬ṭ

‫ܪ‬r

φ

‫ܦ‬p

  ‫ ܪܗ‬rh ῥ-, ῤῥ

χ

‫ܟ‬k

‫ܬ‬t

ψ

‫ ܦܣ‬ps

θ

spiritus asper (initial) ‫ ܗ‬h, ‫ ܐ‬ʾ

The majority of correspondences in table 5.3 are unremarkable, since Greek phonemes tend to be represented by similar Syriac phonemes—for example, the Greek bilabial nasal μ by the Syriac bilabial nasal ‫ ܡ‬m, the Greek alveolar trill ρ by the Syriac alveolar trill ‫ ܪ‬r, etc. Several correspondences do, however, require additional comment. These include the representation of Greek spiritus asper word-initially (§5.2.4), the representation of Greek spiritus asper with ῥ and ῤῥ (§5.2.5); the representation of Greek π (§5.2.6); the representation of the Greek voiceless stops π, τ, and κ (§5.2.7); and the representation of the Greek velar nasal ŋ (§5.2.8). These are dealt with in the following sections. ṯ is to be explained as secondary, probably due to an inner Syriac development whereby the Syriac ending ‑iṯɔ was used to represent the Greek ending -ιδιον.

70

Chapter 5

5.2.3.  A Koinē Greek Input Form Before looking at correspondences that require additional comment, I need to say a few words about possible input forms from Koinē Greek. Many words that prima facie seem to exhibit irregular correspondences from those given in table 5.3 may be explained by positing a different Greek input form. As discussed in §4.10, the Greek language with which Syriac speakers were in contact was not the Attic of the classical period but, rather, Koinē and then early Byzantine Greek. Even though the majority of Greek loanwords in Syriac reflect Attic Greek—no doubt due to its significant prestige, as is reflected, for instance, in the orthography of Koinē Greek—some Greek loanwords in Syriac seem to reflect input forms from Koinē and early Byzantine Greek. Consider, for instance, the Greek loanword ‫ ܐܩܠܝܣܓܕܝܩܘܣ‬ʾqlysgdyqws ‘expert in church law’ (LS2 44; SL 92) < ἐκκλησιέκδικος (PGL 433). The usual representation of Greek κ is the emphatic velar stop q in Syriac. The occurrence of Syriac g, which could reflect either a voiced velar stop /g/ or a voiced velar fricative /ɣ/, indicates that the Greek source was ἐκκλησιέγδικος with the assimilation of κ to γ before the voiced stop δ. Interestingly, an assimilation of κ to γ before a voiced stop is attested in Greek documents from Syria and Mesopotamia. This is reflected in writings such as the following: (5.1) a. ἐγδικίας (P. Euphrates 2.13 [mid-3rd]) for ἐκδικίας b. διεγδικήσειν (P. Euphrates 9.22–23 [252]) for διεκδικήσειν c. ἐγβένω (P. Euphrates 17.22 [mid-3rd]) for ἐκβαίνω d. ἐγ διακληρώσεως (P. Dura 19.6 [88–89]) for ἐκ διακληρώσεως This assimilation of [+ voice] is also found in Greek documents from Egypt. 19 Thus, Syriac g in ‫ ܐܩܠܝܣܓܕܝܩܘܣ‬ʾqlysgdyqws is best explained as reflecting a Greek input form of ἐκκλησιέγδικος (resulting from assimilation) and not as an irregular consonant correspondence of Greek κ reflecting an Attic input form of ἐκκλησιέκδικος. To give another example, the first n in Syriac ‫ ܣܘܢܦܢܘܣ‬swnpnws ‘supervisor of the trades people of Constantinople on behalf of the eparch of the city’ (LS2 485; SL 984) would seem to be an irregular correspondence for μ in the Attic form σύμπονος (LSJ 1685; PGL 1289). Greek documents from Syria and Mesopotamia, however, attest a dissimilation of μ to ν before a labial. This is reflected in writings such as the following: (5.2) a. διαπενψαμένου (P. Euphrates 2.20 [mid-3rd]) for διαπεμψαμένου b. ἐνποιηθῇ (P. Euphrates 8.24 [251]) for ἐμποιηθῇ c. ἐνποιούμενον (P. Euphrates 9.23 [252]) for ἐμποιούμενον d. ἔνπροσθεν (P. Euphrates 16.A.2 [after 239]) for ἔμπροσθεν 19.  Gignac 1976–81: 1.6–80; Mayser 1970: 143–44.

The Phonological Integration of Greek Loanwords in Syriac

71

e. ἐ̣νφράξι (P. Euphrates 13.16 [243]) for ἐμφράξει f. συνβά[ν] (P. Euphrates 2.5 [mid-3rd]) for συμβάν g. συνπαρόντος (P. Euphrates 6.9 [249]; 9.14 [252]) for συμπαρόντος This dissimilation is also attested in Greek documents from Egypt. 20 Thus, Syriac ‫ ܣܘܢܦܢܘܣ‬swnpnws probably reflects a Greek input form of σύνπονος, in which μ has dissimilated to ν before the labial π. To give one final example, several Greek loanwords in Syriac attest what would seem to be the representation of Greek τ by Syriac t: (5.3) a. Latin domesticus (OLD 570; LD 607–8) > δομεστικός (LLGE 41; PGL 380) > ‫ ܕܘܡܣܬܝܩܐ‬dwmstyqʾ ‘domesticus, a Byzantine imperial guard soldier’ (LS2 158; SL 283), as well as ‫ܕܘܡܣܛܝܩܐ‬ dwmsṭyqʾ b. προστάς (LSJ 1526) → accusative singular προστάδα > ‫ܦܪܘܣܬܕܐ‬ prwstdʾ ‘doorpost, lintel; vestibule, portico’ (LS2 603; SL 1233) c. πιστικός ‘faithful’ (LSJ 1408) > ‫ ܦܣܬܝܩܐ‬pstyqʾ ‘sailor to whom responsibility for a ship is entrusted’ (LS2 585; SL 1215–16), as well as ‫ ܦܝܣܛܝܩܣ‬pysṭyqs This representation would be irregular, since Greek τ is usually represented by the emphatic dental stop ṭ in Syriac. This, however, can again be explained by the Greek input form. Greek documents from Syria and Mesopotamia attest an assimilation of τ to θ after σ—for example, κατεσθάθην (P. Dura 46.rev.5 [early 3rd]) for κατεστάθην and ἀφείσθασθαι (P. Dura 31.int.7, ext.33 [204]) for ἀφίστασθαι. This assimilation is also attested in Greek documents from Egypt. 21 Thus, the examples in (5.3) seem to reflect Greek input forms in which τ has assimilated to θ after σ. In each of these three cases, Greek loanwords in Syriac attest what looks like an irregular representation of the Greek source. Upon closer examination, however, this seemingly irregular representation can be accounted for by positing a Koinē input form in which assimilation or dissimilation has occurred. Such an analysis is strengthened by the fact that these changes that are found in the Greek loanwords in Syriac are also attested in Greek documents from Syria and Mesopotamia (especially in P. Dura and P. Euphrates). Finally, it should be pointed out that each of these changes is also attested in Greek documents from Egypt. The previous examples all involved sound changes that are already known from Greek texts from Syria and Mesopotamia. Greek loanwords in Syriac, however, could—at least theoretically—witnesss changes that are not (yet) 20.  Gignac 1976–81: 1.167–69; Mayser 1970: 203–7. 21.  Gignac 1976–81: 1.87; Mayser 1970: 154.

72

Chapter 5

attested in Greek texts from the area. Consider, for instance, Syriac ‫ܩܡܛܪܝܢ‬ qmṭryn ‘small chest’ (LS2 672; SL 1377), which is a loanword from Greek κάμπτριον (LSJ 873). Greek π is not represented in the Syriac word. The deletion of π in the cluster μπτ is, however, well attested in Greek documents from Egypt. 22 Thus, the Syriac form ‫ ܩܡܛܪܝܢ‬qmṭryn (instead of the expected **qmpṭrn) suggests that the deletion of π in the cluster μπτ also occurred in the Greek of Syria and Mesopotamia, even if it is not (yet) attested in Greek documents from this area. 23 The Greek input form for Syriac ‫ ܩܡܛܪܝܢ‬qmṭryn would, then, not be Attic Greek κάμπτριον but, rather, Koinē κάμτριον. A similar case is encountered with what seems to be a representation of Greek σ by Syriac z in the following words: (5.4) a. προθεσμία (LSJ 1481) > ‫ ܦܪܬܙܡܝܐ‬prtzmyʾ ‘fixed time period’ (LS2 610; SL 1256), with an alternative spelling of ‫ܦܪܘܬܣܡܝܐ‬ prwtsmyʾ (LS2 610; SL 1235). b. σμάραγδος (LSJ 1619) > ‫ ܙܡܪܓܕܐ‬zmrgdʾ ‘emerald’ (LS2 200; SL 387); also in Samaritan Aramaic zmrgdy (DSA 234); Christian Palestinian Aramaic zmrgd (DCPA 111; LSP 56); Jewish Palestinian Aramaic zmrgd (DJPA 179); but note also ‫ ܣܡܪܓܕܐ‬smrgdʾ (LS2 482; SL 1021) c. σμίλη (LSJ 1619) > ‫ ܙܡܠܝܐ‬zmlyʾ ‘small knife, scalpel’ (LS2 199; SL 385), also in Tg. Jonathan ʾuzmil (Jer 36:23); Jewish Palestinian Aramaic ʾwzmyl (DJPA 38); Late Jewish Literary Aramaic ʾzml (Tg. Job 16:9; Jastrow 46) 24 This representation would be irregular, since the usual representation of Greek σ is the voiceless alveolar fricative s. The voiced dental fricative z can, however, be explained by an assimilation of σ to ζ before μ. Though this change it is not (yet) attested in Greek documents from Syria and Mesopotamia, it is well attested in Greek documents from Egypt. 25 The previously discussed changes all had clear conditioning factors, and thus a strong case can be made that the Greek loanwords in Syriac reflect changes that had occurred in the Greek source. There are, however, also several cases in which the conditioning factors are less clear or even nonexistent. In such cases, it is not as certain that the Greek input form can be 22.  Gignac 1976–81: 1.64; Mayser 1970: 152. 23.  It should be noted that the representation without p is also found in JBA qamṭrɔ, qumṭrɔ (DJBA 1024) and Tg. Jonathan qəmaṭrəyɔ (2 Kg 10:22), as well as in JPA qmṭryy ‘chest maker’ (DJPA 496). 24.  The same assimilation is found in Armenian զմելին /zmelin/ (NDAE 187; see also Brockelmann 1893: 300) < σμιλίον (LSJ 1619). 25.  Gignac 1976–81: 1.120–21; Mayser 1970: 177.

The Phonological Integration of Greek Loanwords in Syriac

73

used to explain seemingly irregular representations of Greek consonants in Syriac. Nevertheless, I will still outline these examples here, since they may provide additional cases of Koinē Greek input forms, even if they are less certain. Greek β is typically represented in Syriac by b, which was realized either as a voiced bilabial stop /b/ or a voiced bilabial fricative /β/. There is, however, a case in which it is represented in Syriac by p, which was realized either as a voiceless bilabial stop /p/ or a voiceless bilabial fricative /ɸ/: 26 (5.5) κύβος (LSJ 1005) > ‫ ܩܘܦܣܐ‬qwpsʾ ‘cube; piece on a draft board; tessera, mosaic tile; mosaic work; hard stone, flint’ (LS2 685; SL 1340) This seemingly irregular representation may be due to an interchange of β and π in the Greek source, a change that is sporadically attested in Greek documents from Egypt. 27 Alternatively, the presence of p in ‫ ܩܘܦܣܐ‬qwpsʾ ( qup̄sɔ. As already mentioned above, Greek κ is typically represented in Syriac by the emphatic velar stop q. In a few isolated cases, however, it is represented by Syriac g, which was either a voiced velar stop /g/ or a voiced velar fricative /ɣ/: 28 (5.6) a. καλλίας (LSJ 867) > ‫ ܓܠܣ‬gls ‘ape, monkey’ (LS2 118; SL 238) b. κυβερνήτης (LSJ 1004) > ‫ ܓܘܒܪܢܝܛܐ‬gwbrnyṭʾ ‘helmsman, pilot’ (LS2 103; SL 210; see also Schall 1960: 107; only in Aphrahaṭ, Demonstrations, 1.612.2 [ed. Parisot 1894–1907]), though usually ‫ ܩܘܒܪܢܝܛܐ‬qwbrnyṭʾ (with orthographic variants) (LS2 645; SL 1323) This representation may reflect an interchange of γ for κ in the Greek source, which is encountered in Greek documents from Egypt. 29 As already noted, the Greek form with γ is actually attested in a document from Egypt for one of the words in (5.6): γυβερνήτης (P. Grenf. 1.49.21 [220/221]). 30 26.  This representation is also attested in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §154). 27.  Gignac 1976–81: 1.83; Mayser 1970: 145. 28.  This representation is also found in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §9). See also Latin scutum (OLD 1714; LD 1651) > σκοῦτα (LSJ 1616) > ‫ ܣܓܘܛܘܣ‬sgwṭws ‘shield’ (LS2 459; SL 967). 29.  Gignac 1976–81: 6–80; Mayser 1970: 143–44. 30.  The change from voiceless to voiced velar stop is also reflected in Latin gubernare (LD 831), which is a loanword from Greek.

74

Chapter 5

Finally, Greek χ is represented by the emphatic velar stop q in the following words: 31 (5.7) a. χαλκηδών (LSJ 1973) > ‫ ܩܪܟܕܢܐ‬qrkdnʾ ‘chalcedony’ (LS2 696; SL 1411). This would not be the only irregular consonant correspondence in the word: λ = Syriac r, but usually l; κ = Syriac k, but usually q. The phonology is a better fit for καρχηδόνιος ‘Carthaginian’ (LSJ 881) or the like. b. χαράκωμα (LSJ 1977) > ‫ ܩܠܩܘ�ܡܐ‬qlqwmʾ ‘siege engines, entrenchments’ (LS2 670; SL 1375); compare the expected correspondence in Tg. Jonathan krqwmʾ (1 Sam 26:7; Jastrow 669) and Late Jewish Literary Aramaic krqwmʾ (Tg. Job 20:24; Jastrow 669), as well as Jewish Palestinian Aramaic krkwm (DJPA 270), where however the correspondence of κ is irregular c. χάρτης (LSJ 1980); see also Latin charta (OLD 309; LD 325) > ‫ ܩܪܛܝܣܐ‬qrṭysʾ ‘sheet of paper; papyrus’ (LS2 695; SL 1405–6), with the same correspondence in Jewish Palestinian Aramaic qrṭys (DJPA 269), though ‫ ܟܪܛܝܣܐ‬krṭysʾ (LS2 344; SL 650) also occurs in Syriac, with the same correspondence in Jewish Palestinian Aramaic krṭys (DJPA 269) and Christian Palestinian Aramaic krṭys (DCPA 185) Typically, Greek χ is represented by Syriac k, which was realized as a voiceless velar stop or a voiceless velar fricative. The seemingly irregular correspondences in (5.7) may be due to a change of χ to κ before a liquid in the Greek source, a change that is attested in Greek documents from Egypt. 32 While most Greek loanwords in Syriac seem to derive from Attic input forms, the Greek loanwords discussed in this section probably reflect a Koinē or early Byzantine Greek input form. Positing a non-Attic input form can account for what prima facie seem to be a number of irregular representations of Greek consonants in Syriac. This, accordingly, further strengthens the argument that the representation of Greek consonants in Syriac is remarkably regular. In addition, it illustrates how Greek loanwords in Syriac can serve as a witness to the Greek of Late Antique Syria and Mesopotamia (as so-called Nebenüberlieferungen). 33

31.  Schall 1960: 232. This representation is also attested in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §163). 32.  Gignac 1976–81: 1.86–96, esp. top of p. 91; Mayser 1970: 144–45. 33.  See §4.10 above.

The Phonological Integration of Greek Loanwords in Syriac

75

5.2.4. Greek spiritus asper: Word Initial Greek words with initial spiritus asper were realized with an initial voiceless glottal fricative /h/ in Attic Greek 34—for example, ὅρος /horos/ ‘boundary’ (LSJ 1255–56). During the Late Antique period, spiritus asper in word-initial position ceased to be pronounced. 35 Greek spiritus asper is usually represented with h in Syriac 36—for example, ἡνιόχος (LSJ 775) > ‫ ܗܢܝܘܟܐ‬hnywkʾ ‘charioteer’ (LS2 179; SL 348; see also Harviainen 1976: 59) and ὅμηρος (LSJ 1221) > ‫ ܗܡܝܪܐ‬hmyrʾ ‘hostage, pledge’ (LS2 178; SL 345; see also Harviainen 1976: 59). The representation of Greek spiritus asper with Syriac h reflects the earlier Attic pronunciation. 37 In some manuscripts, the initial h is marked with a sublinear dot to indicate that it represents spiritus asper 38 and perhaps that it should not be pronounced. Greek spiritus asper is not represented with h in the following words, however: 39 (5.8) a. ἕνωσις (PGL 486–89; LSJ 579) > ‫ ܐܢܘܣܝܣ‬ʾnwsys ‘combining into one, union’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 27.23 [ed. Brooks 1935]; LS2 30; SL 60; see also Harviainen 1976: 61) b. ἑορταστικός (PGL 504; LSJ 601) > ‫ ܐܘ̈ܪܛـܐܣܛܝܩܐܣ‬ʾwr̈ṭʾsṭyqʾs (pl) ‘festal’ (7th cent. Yaʿqub of Edessa, Letter 13, to Yuḥanon the Stylite of Litarba on Eighteen Biblical Questions, 8.15 [ed. Wright 1867: *1–*24]; only here; LS2 48; SL 9; see also Harviainen 1976: 61) 40 c. ἱερατεῖον (LSJ 820) > ‫ ܐܝܪܛܝܘܢ‬ʾyrṭywn ‘sacristy’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 12.16 [ed. Brooks 1935]; LS2 16; SL 38; see also Harviainen 1976: 62) d. ἱππικός (LSJ 834) > ‫ ܐܦܝܩܘܣ‬ʾpyqws ‘horse’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 114.26 [ed. Brooks 1935]; LS2 42; SL 87; see also Harviainen 1976: 62) 34.  Allen 1987: 52–56. It has also been reconstructed as a voiceless laryngeal fricative /ħ/ (see, e.g., Harviainen 1976: 1 with n. 2). 35.  Harviainen 1976. 36.  Brock 1996: 256; Harviainen 1976: 59–61; Wasserstein 1993: 204. This is also the most common representation in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §78). 37.  In his Letter on Syriac Orthography, Yaʿqub of Edessa (d. 708) refers to representations of spiritus asper with h in Syriac as ‘according to ancient custom’ (mɛṭṭol mʿayyḏuṯɔ ʿattiqṯɔ; ed. Phillips 1869: 6.6–10). 38.  Kiraz 2012: §203; Segal 1953: 26. 39.  Brock 1996: 256; Harviainen 1976: 61–63. Such spellings are also attested in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §79). 40.  The spelling ‫ ܐܘ̈ܪܛـܐܣܛܝܘܩܐܣ‬ʾwr̈ṭʾsṭywqʾs, which is given in SL 9, is not found in the edition of Wright 1867.

76

Chapter 5 e. ἱππόδρομος (LSJ 834) → accusative singular ἱππόδρομον > ‫ ܐܝܦܕܪܡܘܢ‬ʾypdrmwn ‘hippodrome’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 151.7 [ed. Brooks 1935]; LS2 40; SL 81; see also Harviainen 1976: 62) f. ὁλοσηρικόν (LSJ 1218) > ‫ ܐܠܝܣܪܝܩܘܢ‬ʾlysryqwn, ‫ܐܠܘܣܪܝܩܘܢ‬ ʾlwsryqwn, ‫ ܐܠܣܝܪܝܩܘܢ‬ʾlsyryqwn ‘garment entirely of silk’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 139.27 [ed. Brooks 1935]; Lives of the Eastern Saints, 538.10; 540.10 [ed. Brooks 1923–26]; LS2 22; SL 49; see also Harviainen 1976: 62) g. ὁμολογία (PGL 957–58; LSJ 1226) > ‫ ܐܡܘܠܓܝܘܣ‬ʾmwlgyws ‘confession of faith’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 117.26; 131.20 [ed. Brooks 1935]; LS2 25; SL 53; see also Harviainen 1976: 62), note already Palmyrene ʾmlgyʾ (PAT 339–40; see also Brock 2005: 19) h. ὁρίζων (LSJ 1251) > ‫ ܐܘܪܝܙܘܢ‬ʾwryzwn ‘horizon’ (7th cent. Yaʿqub of Edessa, Scholia, 4.150.43 [ed. Benedictus 1732– 46]; Hexaemeron, 172.2.23 [ed. Chabot 1953]; Severos Sebokht, Treatise on the Astrolabe, 84.11 [pl ‫ܐܘܪܘܝܙܘܢܛܘܣ‬ ʾwrwyzwnṭws] [ed. Nau 1899]; LS2 47; SL 22; see also Harviainen 1976: 62)

In all of these cases, the word in question is not attested until the sixth or seventh century. The representation of Greek spiritus asper with the Syriac voiceless glottal stop ʾ reflects the later Koinē pronunciation. In addition, occasionally, two forms of a word are attested in Syriac, one with initial h and another with initial ʾ: 41 (5.9) a. αἵρεσις (PGL 51; LSJ 41) > ‫ ܗܪܣܝܣ‬hrsys ‘difference, opinion; heresies’ (pre-4th cent. Book of the Laws of the Countries, 41.  Brock 1996: 256; Wasserstein 1993: 203–4. For additional examples involving translation literature, compare also ἁπλῶς (LSJ 191) > ‫ ܗܦܠܘܣ‬hplws ‘simply, merely; in vain’ (6th cent. Qiyore of Edessa, Cause of the Liturgical Feasts, 184.11, 185.11 [ed. Macomber 1974]; LS2 181; SL 352–353; see also Harviainen 1976: 59) vs. ‫ ܐܦܠܘܣ‬ʾplws (6th cent. [translation] Theodosius of Alexandria, Theological Discourse, 164.129 [ed. van Roey and Allen 1994]; LS2 40; SL 87; see also Harviainen 1976: 61); ὕπατος (LSJ 1854) > ‫ܗܘܦܛܘܣ‬ hwpṭws (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 72.3; 73.22 [ed. Brooks 1935]; LS2 179; SL 337; see also Harviainen 1976: 59) vs. pl ‫ ܐܦܛܘ‬ʾpṭw (sic without syɔme) ‘consul’ (7th cent. [translation] 1 Esd 3:14; LS2 40; SL 19; see also Harviainen 1976: 62); ὑποδιάκονος (PGL 1448; LSJ 1879) > ‫ ܗܦܘܕܝܩܢܐ‬hpwdyqnʾ ‘subdeacon, member of the minor clergy’ (7th cent. Denḥa, Life of Marutha, 81.13 [ed. Nau 1905a: 52–96]; LS2 ̈ 179; SL 336; see also Harviainen 1976: 60) vs. pl ‫ܐܦܘܕܝܩܢܘ‬ ʾpwdyq̈nw ‘subdeacon, member of the minor clergy’ (5th–6th cent. [translation] Didascalia Apostolorum, 10.9; 30.22; 111.16 [ed. Vööbus 1979]; LS2 179; SL 336; see also Harviainen 1976: 62).

The Phonological Integration of Greek Loanwords in Syriac

77

28.14; 36.17, 36.20 [ed. Drijvers 1965]; LS2 90, 184; SL 180, 355; see also Harviainen 1976: 59) vs. ‫ ܐܪܣܝܣ‬ʾrsys (6th cent. Life of John bar Aphtonia, 23.1 [ed. Nau 1902]; LS2 51; SL 103; see also Harviainen 1976: 61) b. αἱρεσιώτης (PGL 51) > ‫ ܗܪܣܝܘܛـܐ‬hrsywṭʾ ‘heretical; schismatical’ (4th cent. Ephrem, Prose Refutations, Discourse 1, 47.16 [ed. Overbeck 1865: 21–58]; LS2 184; SL 355; see also Harviainen 1976: 59) vs. ‫ ܐܪܣܝܘܛـܐ‬ʾrsywṭʾ (6th cent. Philoxenos, Commentary on Matthew and Luke, 28.26 [ed. Watt 1978]; LS2 51; SL 103; see also Harviainen 1976: 61) c. αἱρετικός (PGL 51) > ‫ ܗܪܛܝܩܐ‬hrṭyqʾ ‘heretical, schismatic’ (5th cent. Julian Romance, 125.20 [ed. Hoffmann 1880b]; Narsai, Memre on Biblical Themes, 17.459; 18.490, 495, 506 [ed. Frishman 1992]; Life of Rabbula, 171.9; 193.11; 194.10, 20 [ed. Overbeck 1865: 157–248]; LS2 183; SL 354; see also Harviainen 1976: 59) vs. ‫ ܐܪܛܝܩܐ‬ʾrṭyqʾ (6th cent. Yuḥanon of Ephesus, Lives of the Eastern Saints, 601.10 [ed. Brooks 1923–25]; LS2 48; SL 98; see also Harviainen 1976: 61) d. ἡγεμών (LSJ 763) > ‫ ܗܓܡܘܢܐ‬hgmwnʾ ‘prefect’ (4th cent. Aphrahaṭ, Demonstrations, 1.973.6 [ed. Parisot 1894–1907]; Book of Steps, 645.20; 648.3 [ed. Kmosko 1926]; LS2 171; SL 340; see also Harviainen 1976: 59) vs. ‫ ܐܝܓܡܘܢܐ‬ʾygmwnʾ ‘prefect’ (4th cent. Book of Steps, 648.15 [ed. Kmosko 1926]; LS2 4; SL 31; see also Harviainen 1976: 62) e. ὑπηρέτης (LSJ 1872) > ‫ ܗܘܦܪܝܛܐ‬hwpryṭʾ ‘slave, servant’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 64.2 [ed. Brooks 1935]; LS2 182; SL 338; see also Harviainen 1976: 59) vs. ‫ ܐܘܦܪܝܛܐ‬ʾwpryṭʾ ‘slave, servant’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 61.23; 64.20 [ed. Brooks 1935]; LS2 43; SL 89; see also Harviainen 1976: 62) Again, there is a clear diachronic tendency: it is only after the 5th century that spiritus asper ceases to be consistently represented with h. The one exception to this tendency is ἡγεμών > ‫ ܐܝܓܡܘܢܐ‬ʾygmwnʾ ‘prefect’, attested in the Book of Steps. This spelling in the Book of Steps, however, probably reflects the date of the manuscript (ca. 12th cent.) and not the supposed date of composition (ca. 400). 42 This seems especially likely since the earlier spelling ‫ܗܓܡܘܢܐ‬ hgmwnʾ also occurs in this text and even within the very same passage (648.3; 42.  See the comments in n. 6 above. For arguments against the traditional dating of this text to the fourth century, see Smith 2014.

78

Chapter 5

see also 645.20). Moreover, in the seventh- or eighth-century ms Jerusalem Syr. 180, which was not used in the edition but which the editor was later able to collate, the spelling with initial h is found instead of initial ʾ at 648.15, again suggesting that the later spelling is due to transmission history. 43 In addition, it should be noted that the existence of two forms for the loanwords in (5.9), one with initial ʾ and the other with initial h, suggests either that the orthography of these words was updated over time or that these words were transferred from Greek to Syriac on more than one occasion. Harviainen (1976: 25–29, 31) has proposed that Greek spiritus asper was lost in the Greek of Syria and Mesopotamia by the mid-fourth century. Greek spiritus asper is not commonly represented by the Syriac voiceless glottal stop, however, until the sixth century in Greek loanwords in Syriac texts. This points to the conservative nature of Greek loanwords in Syriac, which often reflect a more Attic form and not necessarily the spoken Koinē form (see §4.10 and §5.2.3). Greek words with spiritus lenis are realized as vowel initial—for example, ὄρος /óros / ‘mountain’ (LSJ 1255). Word-initial vowels are not, however, tolerated in Syriac. Thus, Greek words with spiritus lenis are usually realized with an initial voiceless glottal stop in Syriac 44—for example, εἰκῇ (LSJ 484) > ‫ ܐܝܩܐ‬ʾyqʾ ‘in vain’ (LS2 16; SL 37–38) and ἐξορία (LSJ 598) > ‫ܐܟܣܘܪܝܐ‬ ʾkswryʾ (LS2 19; SL 43) ‘exile’. In a few cases, however, Greek spiritus lenis is represented in Syriac with an initial h: 45 ̈ hr̈wmʾ ‘sweet spice, fragrant herb’ (5.10) a. ἄρωμα (LSJ 254) > pl ‫ܗܖܘ�ܡܐ‬ (4th cent. Ephrem, Maḏrɔše on Faith, 96.8; 180.4; 199.14 [ed. Beck 1955]; Maḏrɔše on the Church, 80.22 [ed. Beck 1960]; Prose Refutations, Discourse 2–5, 1.52.10 [ed. Mitchell 1912–21]; Maḏrɔše on Paradise, 7.8; 20.27; 49.12, 21 [ed. Beck 1957b]; Maḏrɔše on the Nativity, 42.1; 114.9; 128.12 [ed. Beck 1959]; Maḏrɔše on Nisibis, 127.1 [ed. Beck 1963]; already in Mark 16:1 [SP]; Luke 23:56 [SCP]; 24:1 [P]; LS2 184; SL 354; see also Brock 1967: 394; Harviainen 1976: 63) b. ἔθος (LSJ 480) > ‫ ܗܬܘܣ‬htws ‘custom’ (6th cent. Eliya, Life of Yuḥanon of Tella, 84.26 [ed. Brooks 1907: 29–95]; only here; LS2 184; SL 356; see also Harviainen 1976: 63)

43. See Kmosko 1926: ccciv (s.v. 648.15). 44.  Kiraz 2012: §§603–4. This is also the most common representation in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §77). 45.  Brock 1996: 256; Harviainen 1976: 63–64. This is also attested in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §77). For an additional example involving translation literature, compare also ἐπαρχία (LSJ 611) > ‫‘ ܗܦܪܟܝܘܣ‬province’ (4th–5th cent. [translation] Eusebius of Caesarea, Ecclesiastical History, 76.17 [ed. Wright and McLean 1898]; LS2 182; SL 353; see also Harviainen 1976: 63) alongside the more common ‫( ܐܦܪܟܝܐ‬LS2 43; SL 89; see also Harviainen 1976: 64).

The Phonological Integration of Greek Loanwords in Syriac

79

c. ἐποχή (LSJ 677) > ‫ ܗܦܘܟܝ‬hpwky ‘position with reference to celestial latitude and longitude’ (7th cent. Severos Sebokht, Treatise on the Astrolabe, 244.10 [ed. Nau 1899]; LS2 181; SL 348; see also Harviainen 1976: 63) d. ἰδιώτης (LSJ 819) > ‫ ܗܕܝܘܛـܐ‬hdywṭʾ ‘unskilled, simple, ordinary; stupid’ (4th cent. Aphrahaṭ, Demonstrations, 1.404.26; 1.516.7; 1.693.19; 1.728.2; 1.817.7; 1.920.2 [ed. Parisot 1894–1907]; Book of Steps, 777.7, 10, 11, 12 [ed. Kmosko 1926]; Ephrem, Maḏrɔše on Faith, 149.12; 150.15; 153.13, 17; 163.27; 166.8; 176.8; 242.21 [ed. Beck 1955]; passim; LS2 171; SL 331; see also Harviainen 1976: 26 with n. 5, p. 64; Wasserstein 1993: 204) e. οἰκονόμος (LSJ 1204) > ‫ ܗܘܩܘܢ�ܡܐ‬hwqwnmʾ ‘steward’ (5th cent. Life of Sheʿmon the Stylite, 4.535.3 [ed. Bedjan 1890– 97]; LS2 174; SL 339; see also Harviainen 1976: 64), but note also pl ‫ ܐܩܝܢܡܘ‬ʾqynmw (sic without syɔme) (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 141.28 [ed. Brooks 1935]; LS2 45; SL 20; see also Harviainen 1976: 64) Initial h is the usual representation of spiritus asper in pre-sixth-century Syriac texts, as discussed above, not of spiritus lenis. 46 In some cases, the irregular correspondences in (5.10) are due to so-called Vulgäraspiration—that is, Koinē Greek has aspiration in cases where Attic Greek does not. 47 The word ἰδιώτης, for instance, probably had spiritus asper in Koinē Greek, as reflected in Syriac ‫ ܗܕܝܘܛـܐ‬hdywṭʾ as well as in Coptic hēdiōtēs (Förster 2002: 344). 48 In other cases, however, examples of initial h for spiritus lenis may represent hypercorrections in which Syriac speakers introduced h (mistakenly) supposing that the Greek source had once had spiritus asper though it was no longer pronounced. 49 This hypercorrection in language contact can be compared to English speakers’ pronunciation of French coup de grâce as /kuː də graː/, in which the final sibilant of French grâce /gʁas/ has been deleted by hypercorrection on the basis of the many French loanwords in English in which a final consonant is not pronounced—for example, foie gras /fwaː graː/, faux pas /foː paː/, coup d’état /kuː də taː/, etc. The cases of hypercorrection involving 46.  It should be noted that similar cases of Greek spiritus lenis being represented by initial h are found in Greek loanwords in Coptic (Brock 1996: 256; Harviainen 1976: 37, 75). 47.  Gignac 1976–81: 1.133–38; Mayser 1970: 174–76. 48.  See also Harviainen 1976: 26 with n. 5. 49.  Wasserstein (1993: 204) prefers to see the h in these cases as a representation of Greek ε, η, or αι. This is quite unlikely, however, since h does not represent these vowels until well into the seventh century.

80

Chapter 5

Greek spiritus lenis provide additional support for the loss of spiritus asper in the Greek of Syria and Mesopotamia. 5.2.5. Greek spiritus asper: ῥ and ῤῥ When word-initial, Greek ρ occurred with spiritus asper and was realized as a voiceless alveolar trill /r̥/ in Attic Greek. 50 During the Roman period, this allophonic realization was lost, and initial Greek ρ was a voiced alveolar trill /r/. 51 Greek ρ can be represented with either rh or r in Syriac. In loanwords that are first attested in Syriac by the fifth century, initial Greek ῥ with spiritus asper tends to be represented in Syriac with rh: 52 (5.11) a. ῥητίνη (LSJ 1569) > ‫ ܪܗܛܢܐ‬rhṭnʾ, ‫ ܐܪܗܛܢܐ‬ʾrhṭnʾ ‘resin’ (5th cent. Julian Romance, 51.12 [ed. Hoffmann 1880b], already in Gen 37:25; 43:11; LS2 727; SL 1460) 53 b. ῥήτωρ (LSJ 1570) > ‫ ܪܗܝܛܪܐ‬rhyṭrʾ, ‫ ܪܗܛܪܐ‬rhṭrʾ ‘orator, rhetorician’ (4th cent. Ephrem, Prose Refutations, Discourse 1, 58.21 [ed. Overbeck 1865: 21–58]; LS2 727; SL 1442) This representation reflects the older pronunciation. 54 In loanwords that are not attested until after the fifth century, however, initial Greek ῥ with spiritus asper tends to be represented simply as r in Syriac 55—for example, ῥογα (PGL 1217) > ‫ ܪܘܓܐ‬rwgʾ ‘pay, wages; paying of wages’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 129.26; 270.26 [ed. Brooks 1935]; LS2 711; SL 1443; see also Harviainen 1976: 66). The representation as r (without h) reflects the loss of the allophonic realization of word-initial ρ in the Greek source. As in the case of word-initial ρ, geminated Greek ρρ (written ῤῥ in Byzantine orthography) was realized as a voiceless alveolar trill /r̥/ in Attic Greek. 56 The allophonic realization was lost in the Roman period (Harviainen 1976: 29–32). Medial ῤῥ is represented in Syriac by the alveolar trill r with a following voiceless glottal fricative h in ἀῤῥαβών (LSJ 146) > ‫ ܪܗܒܘܢܐ‬rhbwnʾ ‘pledge, deposit’ (LS2 716; SL 1439; see also Harviainen 1976: 66), which is 50.  Allen 1987: 41–42. 51.  Harviainen 1976. 52.  Brock 1996: 256; Harviainen 1976: 66; Nöldeke 1904: §39; Schall 1960: 99. See also ῥητορεία (LSJ 1569) → accusative singular ῥητορείαν > ‫* *ܪܗܛܪܝܢ‬rhṭryn ‘oratory, set speech’ (5th cent. Julian Romance, 99.4 [ms ‫ ܗܛܪܝܢ‬hṭryn] [ed. Hoffmann 1880b]; only here; LS2 727; SL 1441; see also Harviainen 1976: 66). 53.  The spelling ‫ ܪܛܝܢܐ‬rṭynʾ also occurs in later literature. 54.  In his Letter on Syriac Orthography, Yaʿqub of Edessa (d. 708) refers to writings of ῥ with rh as ‘according to ancient custom’ (mɛṭṭol mʿayyḏuṯɔ ʿattiqṯɔ; ed. Phillips 1869: 6.7–10). 55.  Brock 1996: 256; Harviainen 1976: 66. 56.  Allen 1987: 44–45.

The Phonological Integration of Greek Loanwords in Syriac

81

common from the fourth century onward in Syriac. This spelling reflects the older Attic pronunciation. 57 Medial ῤῥ is represented by the Syriac alveolar trill r without h in the following words: (5.12) a. Latin birrus (LD 239) > βίῤῥος (LSJ 316) > ‫ ܒܝܪܘܢܐ‬byrwnʾ, ‫ܒܪܘܢܐ‬ brwnʾ ‘toga, cloak, patriarch’s chlamys’ (5th cent. Life of Rabbula, 184.26 [ed. Overbeck 1865: 157–248]; LS2 69, 97; SL 143, 187; see also Harviainen 1976: 66) b. καταῤῥάκτης (LSJ 908–9) > ‫ ܩܛܪܩܛܐ‬qṭrqṭʾ ‘sluice, floodgate; step of stairs’ (Bible 1 Kgs 6:8; LS2 664; SL 1359; see also Harviainen 1976: 66) The representation of ῤῥ by Syriac r (without h) reflects the later Koinē pronunciation after the allophonic realization was lost. It should be noted that the representation of Greek ῤῥ with Syriac r (without h) in ‫ ܩܛܪܩܛܐ‬qṭrqṭʾ from the Old Testament Peshiṭta does not necessarily reflect the date of composition (ca. 200) but, rather, may be due to a scribal update in the manuscripts, the earliest of which stem from the sixth century. Various representations of ῤῥ are attested for παῤῥησία (LSJ 1344). Table 5.4 provides a diachronic synopsis of these according to date of composition.  58 In addition to rh and r (without h), Greek ῤῥ is represented by Syriac rr in this word. This is an unusual representation of Greek gemination in Syriac (§5.2.9); thus, it is probably a reflection of the earlier realization of ῤῥ. 5.2.6.  The Syriac Emphatic Bilabial Stop Greek π was a voiceless unaspirated bilabial stop /p/ in Attic Greek as well as in the Koinē Greek of the Roman and Byzantine periods. 59 It is typically represented in Syriac by p, 60 which was realized either as a voiced bilabial stop /p/ or a voiceless bilabial fricative /p̄/ (= IPA /ɸ/) in native Syriac words. There is evidence, however, suggesting that Syriac p, when representing Greek π, was an emphatic bilabial stop. The clearest support for this from the period that is of interest to this study derives from cases of the assimilation of the feature [+ emphatic] due to the presence of this “emphatic” p ( ‫ ܦܛܓܪܐ‬pṭgrʾ ‘gout’ (LS2 563; SL 1180), as well as the expected representation ‫ ܦܘܕܓܪܐ‬pwdgrʾ b. πρόσωπον (LSJ 1533) > ‫ ܦܪܨܘܦܐ‬prṣwpʾ ‘face, countenance; person, party’ (LS2 605; SL 1249–50), for expected ‫*ܦܪܣܘܦܐ‬ *prswpʾ c. πύργος (LSJ 1556) > ‫ ܦܘܪܩܣܐ‬pwrqsʾ ‘tower’ (LS2 607; SL 1173), for expected ‫* *ܦܘܪܓܣܐ‬pwrgsʾ d. συμφωνία (LSJ 1996: 1689) > ‫ ܨܦܘܢܝܐ‬ṣpwnyʾ ‘bagpipe’ (LS2 635; SL 1297), for expected ‫* *ܣܦܘܢܝܐ‬spwnyʾ; see also Aramaic of Daniel sumponyɔ (Dan 3:5, 15), sypnyh (Dan 3:10 [Kethiv]), supponyɔ (Dan 3:10 [Qere]) (HALOT 1937–38) without emphatic ṣ

The Phonological Integration of Greek Loanwords in Syriac

83

In each of these cases, the presence of an “emphatic” p (< Greek π) led to the assimilation of a stop (whether voiced or voiceless) to its emphatic counterpart. An additional assimilation of [+ emphatic] occurs systematically in the ct-stem of πεῖσαι (LSJ 1353–54) > rt. ‫ ܦܝܣ‬pys c ‘to persuade, to convince; to demand, seek, beseech’ (LS2 567; SL 1188), which is often written ‫ܐܬܛܦܝܣ‬ ʾtṭpys. 62 In the later Syriac vocalization traditions, the “emphatic” p is marked with the same diacritic point that is used to distinguish the nonemphatic voiceless stops from their fricative counterparts. 63 In his grammatical works, Bar ʿEbroyo (d. 1286) speaks on several occasions about the “Greek pe” (‫̄ܦܐ‬ ‫ ܝܘܢܝܬܐ‬pe yawnɔytɔ) that is found in ‫ ܦܪܨܘܦܐ‬prṣwpʾ ‘face, countenance; person, party’ (LS2 605; SL 1249–50) < πρόσωπον (LSJ 1533) and other Greek loanwords. 64 An “emphatic” p also occurs with Greek loanwords in Christian Palestinian Aramaic 65—for example, ἀσπίς (LSJ 259) > ʾsṗys ‘snake’ (DCPA 23; LSP 15). In the Christian Palestinian Aramaic script, this “emphatic” p can be written either with the sign for the voiceless bilabial stop (‫ )ܦ‬or with a reversed form of this sign (‫)ܧ‬, which is often transcribed as ṗ. The “emphatic” p in Syriac and Christian Palestinian Aramaic can be compared with the Classical Ethiopic (Gəʿəz) characters that are usually transcribed as p (ፐ) and ṗ (ጰ). 66 These characters occur almost exclusively in loanwords of various origins, including from Greek 67—for example, πνεῦμα ‘spirit’ (LSJ 1424) > penəmu ‘Satan’ (CDG 413) and πόλις ‘city’ (LSJ 1433–34) > ṗolis ‘capital city’ (CDG 414). Christian Palestinian Aramaic, Syriac, and Classical Ethiopic each underwent a prolonged period of contact with Greek that resulted in, inter alia, the presence of a number of Greek loanwords in these languages. Each of these languages dealt in similar but distinct ways with Greek π: Classical Ethiopic went the furthest in innovating two characters to represent the sound; Christian Palestinian Aramaic used an existing character, both in its normal form and in an inverted form; and Syriac used an existing character, which in the later tradition was marked with a diacritic.

62.  See already Schall 1960: 80. 63.  Kiraz 2012: §214; see also §§63, 68; Nöldeke 1904: §15; Segal 1983: 488. In the later West Syriac tradition, this point is placed in the middle of p thereby differentiating it from both the voiceless bilabial stop and the voiceless bilabial fricative (Kiraz 2012: §214). 64. See Voigt 1998b: 532–36 with additional references. 65.  Müller-Kessler 1991: §2.1.2.4. 66.  It should be noted that the Classical Ethiopic reflex of the Proto-Semitic voiceless bilabial stop *p is /f/. 67.  Gragg 2004: 435; Tropper 2002: §31.1.

84

Chapter 5

Since there is no independent sign for the “emphatic” p in the Syriac script and since Syriac p also represents Greek φ, a few homographs result: 68 (5.14) a. ‫ ܐܣܦܝܪ‬ʾspyr, ‫ ܣܦܝܪ‬spyr ‘troop, cohort’ (LS2 36, 493; SL 76, 1031) < σπεῖρα (LSJ 1625) vs. ‫ ܐܣܦܝܪܐ‬ʾspyrʾ, ‫ ܣܦܝܪܐ‬spyrʾ, ‫ ܐܣܦܝܪ‬ʾspyr ‘sphere; circle; ball; pine cone; cake’ (LS2 76, 1031; SL 76, 1031) < σφαῖρα (LSJ 1738) b. ‫ ܛܪܘܦܐ‬ṭrwpʾ ‘solstice’ (LS2 291; SL 550) < τροπή (LSJ 1826) vs. ‫ ܛܪܘܦܐ‬ṭrwpʾ ‘nourishment, support’ (LS2 291; SL 550) psanṭerin ‘psaltery’ (HALOT 1958). 71.  In general, see Marrassini 1990: 39–41; 1999: 329–30; as well as Kutscher 1965: 33–34. 72.  In this regard, Kutscher himself remarks, “It is impossible to say anything definite about this” (Kutscher 1965: 32). 73.  There is, for instance, no mention of aspiration in the recent detailed treatment of Semitic phonetics and phonology by Kogan (2011). 74.  Since Syriac has no independent sign for the “emphatic” p, there is ambiguity in the representation of Greek π and φ, as illustrated by the homographs discussed at the end of §5.2.6.

86

Chapter 5

5.2.8.  The Greek Velar Nasal ŋ in γκ, γγ, γχ, and γμ In the sequences γκ, γγ, γχ, and γμ, Greek γ represents the velar nasal ŋ, 75 which serves as an allophone of the dental nasal ν and the voiced velar stop γ. In the majority of cases, the Greek velar nasal is represented in Syriac with the dental nasal n, as in the following representative examples: (5.15) a. ἀνάγκη (LSJ 101) > ‫ ܐܢܢܩܐ‬ʾnnqʾ (with alternative orthographies) ‘necessity’ (LS2 29; SL 63) b. κόγχη (PGL 759) > ‫ ܩܢܟܐ‬qnkʾ ‘the part of the church in which the holy service is preformed and where the altar stands’ (LS2 677; SL 1385) c. μάγγανον (LSJ 1070) > ‫ ܡܢܓܢܘܢ‬mngnwn ‘instrument of torture’ (LS2 394; SL 780) In the following cases, however, the Greek velar nasal ŋ is not represented in Syriac: (5.16) a. Latin uncinus (OLD 2090; LD 1929) > ὄγκινος (LSJ 1196) > ‫ ܐܘܩܝܢܐ‬ʾwqynʾ ‘hook; anchor; sailors’ sounding line’ (LS2 9; SL 20) b. ἀγκών (LSJ 10) > ‫ ܐܩܘܢܐ‬ʾqwnʾ ‘hollow of the arm or knee’ (LS2 45; SL 92) c. λόγχη (LSJ 1059) > ‫ ܠܘܟܝܬܐ‬lwkytʾ ‘spear’ (LS2 361; SL 679) 76 d. σπόγγος (LSJ 1628) > ‫ ܐܣܦܘܓܐ‬ʾspwgʾ, ‫ ܣܦܘܓܐ‬spwgʾ ‘sponge’ (LS2 35; SL 75) The lack of representation of the Greek velar nasal ŋ in these examples is due to its assimilation to a following consonant. This can be compared with the rare cases in which Greek ν is not represented in Syriac: 77 75.  Allen 1987: 33–39; Woodard 2004b: 616. The pronunciation of γ as the velar nasal ŋ is sometimes reflected in spellings in Greek documents, including those from Syria and Mesopotamia: for γγ, see ἀντισύνγραφα for ἀντισύγγραφα (P. Euphrates 6.29–30 [249]; 7.23 [249]); στρονγυλοπρόσωπον for στρογγυλοπρόσωπον (P. Euphrates 8.13 [251]; 9.12 [252]); συνγραφήν for συγγραφήν (P. Euphrates 8.17 [251]); for γκ, see ἐνκαλ̣έσῃ for ἐγκαλέσῃ (P. Dura 31.int.16 [204]; P. Euphrates 8.25 [251]); ἐν̣κ̣αλῖν for ἐγκαλεῖν (P. Euphrates 14.17 [241]); ἐνκαλλέσειν for ἐγκαλέσειν (P. Dura 31.int.13 [204]); ἐνκαλοῦμε for ἐγκαλοῦμαι (P. Euphrates 3.12 [252–56]; 4.12; [252–56]); ἐνκλήματα for ἐγκλήματα (P. Euphrates 3.11 [252–56]; 4.12 [252–56]); πάνκαλα for πάγκαλα (P. Euphrates 17.9–10 [mid-3rd cent.]); πάνκαλον for πάγκαλον (P. Euphrates 17.2 [mid-3rd]); συνκωμῆται for συγκωμῆται (P. Euphrates 1.10–11 [245]); συνκωμήτης for συγκωμήτης (P. Euphrates 4.6 [252–56]); for γχ, see τυνχάνομεν for τυγχάνομεν (P. Euphrates 1.11 [245]). ̈ lw¨nkdyʾ ‘small spears’ (LS2 368; 76.  Compare, however, λογχίδιον (LSJ 1059) > ‫ܠܘܢܟܕܝܐ‬ SL 679). 77.  See also Latin mansio (OLD 1074; LD 1109) > ‫ ܡܣܝܘܢܐ‬msywnʾ ‘journey of ten parasangs’ (LS2 396; SL 790). It should be noted that Latin ns is normally realized simply as σ in Latin loanwords in Greek (Gignac 1976–81: 1.117–18).

The Phonological Integration of Greek Loanwords in Syriac

87

(5.17) a. ἀνδριάς (LSJ 128) → accusative singular ἀνδριάντα > ‫ܐܕܪܝܢܛܐ‬ ʾdrynṭʾ, ‫ ܐܕܪܝܛܐ‬ʾdryṭʾ ‘statue’ (LS2 6; SL 11), with an additional spelling of ‫ ܐܢܕܪܝܢܛܐ‬ʾndrynṭʾ b. πανδοκεῖον (LSJ 1296–97) > ‫ ܦܘܬܩܐ‬pwtqʾ, ‫ ܦܘܛܩܐ‬pwṭqʾ ‘inn’ (LS2 565, 618; SL 1162, 1177) c. σάνδαλον (LSJ 1582) > ‫ ܣܕ�ܠܐ‬sdlʾ ‘sandal’ (LS2 460, 484; SL 971, 1022), with an additional spelling of ‫ ܣܢܕ�ܠܐ‬sndlʾ d. σινδών (LSJ 1600) > ‫ ܣܕܘܢܐ‬sdwnʾ ‘fine linen cloth’ (LS2 460; SL 970) This lack of representation of Greek ν is either due to an assimilation of ν to a following dental in the Greek source, a change that is attested in Greek documents from Egypt, 78 or to an inner Syriac development whereby n assimilates to a following consonant. Given the regularity of the latter, it seems more likely. 5.2.9.  Greek Gemination With the exception of γγ and ρρ, 79 Greek gemination, which is written with two consonants, was realized as a lengthened sound. 80 In the majority of cases, Greek gemination is represented by a single consonant in Syriac 81—for example, κόσσος (LSJ 985; PGL 772) > ‫ ܩܣܘܣ‬qsws ‘blow on the ear’ (LS2 680; SL 1386) and τύραννος (LSJ 1836) > ‫ ܛܪܘܢܐ‬ṭrwnʾ ‘tyrant’ (LS2 289; SL 549). The Syriac consonantal script does not indicate gemination, and thus it cannot be determined whether or not gemination is represented in these cases without recourse to the later vocalization traditions. Occasionally, Greek gemination is represented by two consonants in Syriac, as in μᾶλλον (LSJ 1076) > ‫ �ܡܐܠܠܘܢ‬mʾllwn, ‫ ܡܠܠܘܢ‬mllwn ‘rather, more’ (LS2 392; SL 766), as well as ‫ ܡܠܘܢ‬mlwn. The spelling of this word with two l’s in Syriac does not appear until the sixth century. In contrast, the spelling with one l is already attested in the fourth century in Ephrem’s Commentary on the Diatessaron (ed. Leloir 1990: 30.19). This suggests that the representation of Greek gemination with two consonants in Syriac is a later phenomenon. 82 This aligns with a trend that the Greek source tends to be represented more closely 78.  Gignac 1976–81: 1.116; Palmer 1945: 2. This change in Greek may additionally involve nasalization of the vowel. 79.  These were discussed in §5.2.8 and §5.2.5, respectively. 80.  Allen 1987: 12–13. 81.  Kiraz 2012: §§603–4. For the representation of Greek gemination in Postbiblical Hebrew and in various dialects of Jewish Aramaic, see Krauss 1898: §41. 82.  It should be noted that, among the many spellings of παῤῥησία (LSJ 1344), ‫ܦܪܪܝܣܝܐ‬ prrysyʾ and ‫ ܦܪܪܣܝܐ‬prrsyʾ are found already in the 3rd-century Acts of Thomas (ed. Wright 1871a: 212.12 and 192:10, respectively; see table 5.4). This unusual spelling for consonantal gemination is, however, probably due to the voiceless pronunciation of ῤῥ (see §5.2.5).

88

Chapter 5

Table 5.5.  Vowel Phonemes of Koinē Greek in the Roman Period Front High Mid Low

Central

i (ι, η, ει) / y (υ, οι)

Back u (ου)

e (ε, αι)

o (ο, ω) a (α)

in Syriac over time, which is even clearer with the representation of Greek vowels in Syriac, to which we now turn. 5.3. Vowels 5.3.1.  The Vocalic Inventories of Koinē Greek and Syriac The vocalic inventory of Koinē Greek in the Roman period consisted of six phonemes, which are summarized in table 5.5. 83 The Koinē vocalic system is the result of a number of developments from the much more complicated system of Attic Greek, which had five short and seven long vowels, plus five short diphthongs and five long diphthongs. 84 In the Koinē Greek of the Roman period, there were two high front vowels /i/ and /y/, which are distinguished by the presence or absence of rounding. The high, front, unrounded /i/ in the Koinē Greek of the Roman period is written ι, which was a high, front, unrounded vowel, either short /i/ or long /iː/, in Attic Greek, as well as with η, which was a long, open-mid, front /ɛ:/ in Attic Greek, and with ει, which was a long, close-mid, front /e:/ in Attic Greek. 85 The high, front, rounded /y/ in the Koinē Greek of the Roman period is written with υ, which was a high, front, rounded vowel, either short /y/ or long /yː/, in Attic Greek, as well as with οι, which was the diphthong /oi/ in Attic Greek. By the middle of the Byzantine period, /y/ lost its rounding and so merged with /i/. Mid front /e/ in the Koinē Greek of the Roman and Byzantine periods is written with ε, which was a mid front short /e/ in Attic Greek, as well as with αι, which was the diphthong /ai/ in Attic Greek. Low central /a/ in the Koinē Greek of the Roman and Byzantine periods is written α, which was a low, central vowel, either short /a/ or long /aː/, in Attic Greek. Mid back /o / in the Koinē Greek of the Roman and Byzantine periods is written with ο, which was a short, mid, back /o / in Attic 83. See Gignac 1976–81: 1.183–333; Horrocks 2010: 160–63; Mayser 1970: 33–141. 84.  For the more complicated vowel inventory of Attic Greek, see Allen 1987: 62–95; Woodard 2004b: 617. 85.  It should be noted that some Koinē dialects preserved η as an open mid-front /ε/ into the Roman period (Allen 1987: 74–75; Butts 2015b: 99–100; Gignac 1976–81: 1.235–42; Mayser 1970: 46–54; Palmer 1934: 170; 1945: 1).

89

The Phonological Integration of Greek Loanwords in Syriac

Table 5.6.  Reconstructed Vowel Phonemes of 4th-Century Syriac Front High Close-mid Mid Open-mid Low

Central

i

Back u

e

o e̞ ɛ

ɔ a

Greek, as well as with ω, which was a long, open-mid, back /ɔ:/ in Attic Greek. High back /u/ in the Koinē Greek of the Roman and Byzantine periods is written ου, which was a long, high, back /uː/ in Attic Greek. In addition to these six vowel phonemes, the Koinē Greek of the Roman and Byzantine periods has two diphthongs: αυ /au/ and ευ /eu/. The vocalic inventory of fourth-century Classical Syriac can be reconstructed with eight phonemes, which are summarized in table 5.6.  86 High back /u/ is the reflex of earlier Aramaic *ū as well as *u in unaccented syllables. Close-mid back /o / is the reflex of earlier Aramaic *u in accented syllables as well as earlier Aramaic *aw in closed syllables. In West Syriac, close-mid back /o / merged with high back /u/. Open-mid back /ɔ/ is the reflex of earlier Aramaic *ā. In West Syriac, this vowel was raised to closemid back /o /. Low central /a/ is the reflex of earlier Aramaic *a. Open-mid front /ɛ/ is the reflex of earlier Aramaic *i. High front /i/ is the reflex of earlier Aramaic *ī. Close-mid front /e/ results from several different contractions, including non-final *-aʾ and word-final *‑áyu and *-íyu. 87 In West Syriac, close-mid front /e/ merged with /ɛ/. Mid front e̞ results from other contractions, 88 including non-final *-iʾ, but it merges with /i/ in West Syriac instead of /ɛ/. The vowel system described in the previous paragraph and summarized in table 5.6 must be reconstructed. This is because the written Syriac vocalization traditions were not developed until after the period that is of interest 86.  In general, see Daniels 1997; Muraoka 2005: §4; Nöldeke 1904: §§8–10. 87.  These are discussed in Blau 1969. 88.  There is no IPA symbol that represents the mid-front unrounded vowel between close-mid e and open-mid ɛ. This is, however, often represented as e̞—i.e., greater tongue lowering of closemid e, or less commonly as ɛ̝—i.e., increased tongue height of open-mid ɛ (Roca and Johnson 1999: 127).

90

Chapter 5

to this study. 89 These vocalization traditions involve the layering of vowel signs—either in the form of diacritic points (East Syriac) or adapted Greek vowels (West Syriac)—onto an inherited consonantal skeleton. 90 Since all data for these vocalization traditions derive from well after the time period that is of interest to this study, this section does not analyze the use of Syriac vowel signs to represent vowels in Greek loanwords. The primary evidence for Syriac vowels prior to the late seventh century is the use of so-called matres lectionis. In scholarship on Northwest Semitic languages, the term mater lectionis (pl matres lectionis), literally, ‘mother of reading’, refers to the use of certain consonants to mark vowels in a consonantal script. In the word ‫ ܓܘܫ�ܡܐ‬gwšmʾ /gušmɔ/ ‘body’ (LS2 136; SL 222–23), for instance, the sign usually used for the bilabial glide w indicates instead the vowel /u/, and the sign for the voiceless glottal stop ʾ indicates the final /ɔ/. In native Syriac words, the sign for the bilabial glide w serves as a mater lectionis for almost all cases of high back /u/ and close-mid back /o /; 91 the sign for the palatal glide y serves as a mater lectionis for all cases of high front /i/ as well as for some cases of close-mid front /e/ and mid front /e̞/; and the sign for the voiceless glottal stop ʾ serves as a mater lectionis for all cases of open-mid back /ɔ/ in final position as well as for many cases of close-mid front /e/ and mid front /e̞/. 92 These same consonants also serve as matres lectionis in Greek loanwords in Syriac. In addition, toward the end of the seventh century, the sign for the voiceless glottal fricative h came to be used as a mater lectionis in Greek loanwords. 5.3.2.  The Representation of Greek Vowels In contrast to the representation of Greek consonants in Syriac, which is remarkably stable and regular (see §5.2), there is a great deal of variation in the representation of Greek vowels in the Syriac script. 93 The various possibilities are summarized in table 5.7. The variety in the representation is due to at least two factors. First, the vowel system of Greek was far from stable, experiencing significant changes from Attic to the Koinē Greek of the Roman period and then more changes into the Byzantine period. These changes in the Greek vowel system can account for a number of the variations in the Syriac representation, such as, for instance, the various representations of Greek υ: the use of y in Syriac as a mater lectionis reflects the later pronunciation /i/ 89.  The use of diacritics for specific vowel phonemes does not appear until the 8th and 9th centuries (Kiraz 2012: §34). Traces of the five-vowel Greek system are also found at this time, though it is not systematically in use until the 10th century (Coakley 2011; Kiraz 2012: §44). 90.  For these systems, see Kiraz 2012: §§138–57, 174–83; Segal 1953: 24–47. 91.  The only regular exceptions are kol ‘all’ and mɛṭṭol ‘because of’, where a mater lectionis is optional (Kiraz 2012: 101A). 92.  For the system of matres lectionis in Syriac, see Kiraz 2012: §§23–26, 33, 131–37. 93.  See already Brock 1996: 256; 2004: 31 n. 5.

The Phonological Integration of Greek Loanwords in Syriac

91

Table 5.7.  Representation of Greek Vowels in Syriac Greek

Syriac common

α

∅

Syriac rare

‫ ܐ‬ʾ, ‫ ܘ‬w, ‫ ܝ‬y

ε

∅

‫ ܝ‬y, ‫ ܐ‬ʾ, ‫ ܗ‬h, ‫ ܘ‬w

η

‫ ܝ‬y, ∅

‫ ܐ‬ʾ, ‫ ܗ‬h, ‫ ܘ‬w

ι

‫ ܝ‬y, ∅

‫ ܐ‬ʾ, ‫ ܘ‬w

ο

‫ ܘ‬w, ∅

‫ ܝ‬y, ‫ ܐ‬ʾ

υ

‫ ܘ‬w, ∅

‫ܝ‬y

ω

‫ ܘ‬w, ∅

αι

∅

αυ

‫ܘ‬w

ει

‫ ܝ‬y, ∅

ευ

‫ܘ‬w

οι

‫ܘ‬w

ου

‫ ܘ‬w, ∅

Greek common

∅

rare

α, ε, η, ι, ο, υ, ω, αι, ει, ου

‫ܐ‬ʾ

α, ε, η, ι, ο, αι

‫ܗ‬h

ε, η

‫ܘ‬w

ο, υ, ω, αυ, ευ, οι, ου

α, ε, η, ι, αι

‫ܝ‬y

η, ι, ει

α, ε, ο, υ, οι

‫ ܝ‬y, ‫ ܐ‬ʾ

‫ܝ‬y

(unrounded), whereas the use of w in Syriac reflects the earlier pronunciation /y/ (rounded), which was probably reinforced by the written orthography of Greek. Or, to take another example, the use of ʾ as a mater lectionis for final η reflects the pronunciation of an open-mid front /ε/, whereas the use of y reflects its later merger with the high, front, unrounded /i/ (Butts 2015b: 99–100). The second and greater source of variation in the representation of Greek vowels in Syriac stems from the optional use of matres lectionis for each of the Greek vowels (excluding diphthongs). Certainly, there are tendencies for particular vowels—for example, usually no mater lectionis with α and ε. In addition, many words have a stable orthography—for example, πρόσωπον (LSJ 1533) > ‫ ܦܪܨܘܦܐ‬prṣwpʾ ‘face, countenance; person, party’ (LS2 605; SL 1249– 50). 94 Nevertheless, a mater lectionis is entirely optional for the representation of many vowels in many Greek loanwords. This is the case, for instance, with the ο’s in Greek ὄργανον (LSJ 1245) > ‫ ܐܘܪܓܢܘܢ‬ʾwrgnwn, ‫ ܐܪܓܢܘܢ‬ʾrgnwn, ‫ ܐܪܓܢܢ‬ʾrgnn ‘instrument, tool’ (LS2 46; SL 21). The optional use of matres lectionis in Greek loanwords in Syriac diverges starkly from their use in native 94.  It should be noted, however, that even with Syriac words that usually have a very stable orthography variation can occur. The spelling ‫ ܦܪܘܨܘܦܐ‬prwṣwpʾ, for instance, is found in a Sinai ms probably from the 9th century (Binder 2014: 17).

92

Chapter 5

7th Century

6th Century

5th Century

4th Century

Pre-4th Century

Table 5.8.  Diachronic Synopsis of Spellings of διαθήκη in Syriac Peshiṭta Old Testament

‫ ܕܝܬܩܐ‬dytqʾ (1 Chr 15:25, 26, 28, 29)

Demonstrations by Aphrahaṭ (fl. 337–45) (ed. Parisot 1894–1907)

‫ ܕܝܬܩܐ‬dytqʾ (1.52.19; passim), ‫ ܕܝܬܩܝ‬dytqy (1.533.11)

Book of Steps (ca. 400) (ed. Kmosko 1926)

‫ ܕܝܬܩܐ‬dytqʾ (40.7; passim), ‫ ܕܝܬܩܝ‬dytqy (201.3; passim)

Maḏrɔše against Heresies by Ephrem (d. 373) (ed. Beck 1957a)

‫ ܕܝܬܩܐ‬dytqʾ (103.18; passim)

Teaching of Addai (ca. 420) (ed. Howard 1981)

‫ ܕܝܬܩܐ‬dytqʾ (35.9; 36.17)

Julian Romance (5th cent.) (ed. Hoffmann 1880b)

‫ ܕܝܬܩܐ‬dytqʾ (75.18)

Memre by Narsai (d. ca. 500) (ed. Frishman 1992)

‫ ܕܝܬܩܐ‬dytqʾ (73.60)

Life of Rabbula (ca. 450) (ed. Overbeck 1865: 159–209)

‫ ܕܝܬܩܐ‬dytqʾ (172.18)

Cause of the Liturgical Feasts by Qiyore of Edessa (6th cent.) (ed. Macomber 1974)

‫ ܕܝܬܩܐ‬dytqʾ (17.16)

Commentary on Matthew and Luke by Philoxenos (d. 523) (ed. Watt 1978)

‫ ܕܝܐܬܝܩܝ‬dyʾtyqy (69.22)

Part 2 by Isḥaq of Nineveh (late 7th cent.) (ed. Brock 1995)

‫ ܕܝܬܩܐ‬dytqʾ (46.19, 20; passim)

Letters by Ishoʿyahb III of Adiabene (d. 659) ‫ ܕܝܬܩܐ‬dytqʾ (31.14; passim) (ed. Duval 1904–5) Letter 13 by Yaʿqub of Edessa (d. 708) (ed. Wright 1867: *1–*24)

‫ ܕܝܐܬܝܩܝ‬dyʾtyqy (19.12)

Syriac works, where the orthography is extremely stable. 95 Or consider Greek τήγανον (LSJ 1786), which appears in Syriac as ‫ ܛܓܢܐ‬ṭgnʾ, ‫ ܛـܐܓܢܐ‬ṭʾgnʾ, and ‫ ܛܝܓܢܐ‬ṭygnʾ ‘frying pan’ (LS2 268; SL 513). Greek η in this example is represented in three different ways in Syriac: with the mater lectionis y, with the mater lectionis ʾ, and without a mater lectionis. This type of variation is quite simply unattested for native Syriac words, but it is not atypical of many Greek loanwords in Syriac. 95.  Noting the exceptions of kol ‘all’ and mɛṭṭol ‘because of’, mentioned in n. 91 above, where a mater lectionis is optional for o .

The Phonological Integration of Greek Loanwords in Syriac

93

Table 5.9.  Yaʿqub of Edessa’s Preferred Orthography for Greek Loanwords Greek source

Yaʿqub’s orthography

σύνοδος

‫ ܣܘܢܘܕܘܣ‬swnwdws (6.8)

παῤῥησία

‫ ܦܐܪܪܝܣܝܐ‬pʾrrysyʾ (6.9)

κατάστασις

‫ ܩܐܛـܐܣܛܐܣܝܣ‬qʾṭʾsṭʾsys (7.3)

φαντασία

‫ ܦܐܢܛܐܣܝܐ‬pʾnṭʾsyʾ (7.3)

θεολογία

‫ ܬܐܘܠܘܓܝܐ‬tʾwlwgyʾ (7.4)

πληροφορία

‫ ܦܠܝܪܘܦܘܪܝܐ‬plyrwpwryʾ (7.4)

φιλοσοφία

‫ ܦܝܠܘܣܘܦܝܐ‬pylwswpyʾ (7.4) ̈ ‫ܐܘܐܢܓܐܠܝܐ‬ ʾwʾn̈gʾlyʾ (7.6)

εὐαγγέλια διαθήκη

‫ ܕܝܐܬܝܩܝ‬dyʾtyqy (7.7)

εὐαγγελιστής

‫ ܐܘܐܢܓܠܝܣܛܐ‬ʾwʾnglysṭʾ (7.7)

In some cases, the instability of the orthography of a Greek loanword may indicate that the word in question is closer to a Fremdwort than a Lehnwort (see §4.5). In other cases, however, the changing orthography of Greek loanwords in Syriac shows that Syriac speakers continued to interact with the Greek source by updating a loanword. This is the case, for instance, with διαθήκη. Table 5.8 provides a diachronic synopsis of various spellings of Greek διαθήκη in Syriac. In the fourth century, there is variation in the representation of the final η with either ʾ or y in Syriac: the former reflects a mid front realization of Greek η, whereas the latter reflects a high front realization (Butts 2015b: 99–100). The orthography then stabilizes as ‫ ܕܝܬܩܐ‬dytqʾ. Beginning in the sixth century, a new orthography ‫ ܕܝܐܬܝܩܝ‬dyʾtyqy is found in the West Syriac tradition. 96 This new orthography provides a fuller representation of the vowel hiatus ια and also reflects the high front realization of final Greek η. The previous example of διαθήκη not only shows that the orthography of Greek loanwords in Syriac changed diachronically, but it also points to a more specific trend: over time, vowels in Greek loanwords tend to be represented more fully in Syriac. This trend can be exemplified by the representation of vowels in Greek loanwords in the Letter on Syriac Orthography by Yaʿqub of Edessa (d. 708). 97 Table 5.9 illustrates the orthography preferred by Yaʿqub, in which every Greek vowel is represented by a Syriac mater lectionis. This is the extreme end of the spectrum in the representation of Greek vowels in Syriac. 96.  Brock 1996: 257. 97.  The text is edited in Phillips 1869.

94

Chapter 5

It should be noted, however, that this is only Yaʿqub’s ideal, which was never fully realized in Syriac. 5.3.3.  Greek Accent In Attic Greek, accent was related to pitch (melody) and was fixed on one of the last three syllables of a word. 98 By at least the end of the fourth century c.e., Greek accent changed from melodic to stress, as it is in modern Greek. 99 In Syriac, accent is related to word stress and falls almost always on the last syllable. 100 There is almost no evidence for how Greek accent was accommodated in Syriac. 101 Accent could be used to explain the rare representation of Greek ε by Syriac y in a word such as ἐπιθέτης (LSJ 634) > ‫ ܐܦܝܬܝܛܐ‬ʾpytyṭʾ ‘imposter’ (LS2 43; SL 87). Counterexamples are numerous, however—for example, μέταλλον (LSJ 1114) > ‫ ܡܛܠܘܢ‬mṭlwn ‘metal; mine, quarry’ (LS2 382; SL 747) and συντέλεια (LSJ 1725–26) > ‫ ܣܘܢܛܠܝܐ‬swnṭlyʾ ‘poll tax’ (LS2 485; SL 983), where there are no matres lectionis for stressed Greek ε. This distribution could theoretically be explained by the fact that the former word preserves the Greek accent (by this time, word stress), possibly as a Fremdwort, and the latter adopt Syriac stress as fully accommodated Lehnwörter. 102 This hypothesis, however, is difficult if not impossible to prove. 103 A place where accent does seem to play a clearer role is in the apocopation of final Greek vowels in Syriac. There are a few Greek loanwords in Syriac in which a final vowel is apocopated: 104 (5.18) a. βῆμα (PGL 295–96; LSJ 314) > ‫ ܒܝܡ‬bym ‘tribunal, raised platform, bema of a church’ (LS2 68; SL 141), singular also attested as ‫ ܒܝ�ܡܐ‬bymʾ, ‫ ܒܐ�ܡܐ‬bʾmʾ b. σπεῖρα (LSJ 1625) > ‫ ܐܣܦܝܪ‬ʾspyr, ‫ ܣܦܝܪ‬spyr ‘troop, cohort’ (LS2 36, 1031; SL 76, 1031) c. τάχα (LSJ 1762) > ‫ ܛܟ‬ṭk ‘perhaps’ (LS2 274; SL 528)

98.  Allen 1987: 116–39; Woodard 2004b: 619. For the historical development, see Pro­ bert 2006. 99.  Allen 1987: 130–31; Gignac 1976–81: 1.325–27. 100.  There are only a few exceptions, such as the imperatives—for example, qṭólayn(y) ‘kill (m.sg) me!’ 101.  For the accommodation of accent in Postbiblical Hebrew and in various dialects of Jewish Aramaic, see Krauss 1898: §36. 102.  For the distinction between Fremdwörter and Lehnwörter, see §4.5. 103.  In fact, ‫ ܐܦܝܬܝܛܐ‬ʾpytyṭʾ can undergo secondary nominal derivations with suffixes (§6.5.3), suggesting that it is a Lehnwort and not a Fremdwort. See ἐπιθέτης (LSJ 634) > ‫ ܐܦܝܬܝܛܐ‬ʾpytyṭʾ ‘imposter’ (LS2 43; SL 87) + -uṯɔ → ‫‘ ܐܦܝܬܝܛܘܬܐ‬imposture’ (LS2 43; SL 87). 104.  See also παράλια (LSJ 1316) > ‫ ܦܪܗܠܝ‬prhly ‘seashore’ (Luke 6:17 [S]; LS2 598; SL 1229; see also Brock 1967: 411).

The Phonological Integration of Greek Loanwords in Syriac

95

In all of these cases, the final α is unaccented and thus could potentially be apocopated. It should be noted that this was only a potentiality, since there are numerous examples in which final unaccented α is not apocopated—for example, εἶτα (LSJ 498) > ‫ ܐܝܛܐ‬ʾyṭʾ ‘then’ (LS2 14; SL 33), where the preservation of final α cannot be accounted for by Syriac morphology. In contrast to unaccented vowels, which have the potential to be apocopated, there are no examples in which a final accented Greek vowel is apocopated in Syriac. 5.3.4.  Greek Vowel Hiatus In Attic Greek as well as in the Koinē Greek of the Roman and Byzantine periods, vowel-initial syllables can occur within words—for example, δι | α | θή | κη. This results in hiatus (also called diaeresis). Syriac, in contrast, does not tolerate vowel-initial syllables in any context, including within words. The accommodation of Greek vowel hiatus in Syriac is accomplished in two ways. 105 First, the vowel hiatus can be resolved by epenthesis of a voiceless glottal stop ʾ or a palatal glide y. This is the case, for instance, in πατριάρχης (PGL 1051–52) > ‫ ܦܛܪܝܪܟܐ‬pṭryrkʾ ‘patriarch’ (LS2 565; SL 1184), where the consonant y resolves the hiatus in Greek ια. Second, Greek vowel hiatus can be contracted in Syriac into a monosyllable. This is the case, for instance, with Latin quaestor (OLD 1534–35; LD 1502–3) > κυαίστωρ (LLGE 63; PGL 784) > ‫ ܩܣܛܘܪ‬qsṭwr ‘quaestor, Byzantine head of judiciary’ (LS2 655, 679; SL 1322), where neither a glide nor the voiceless glottal stop appears in the first syllable of the Syriac. 5.4. Conclusion In the scholarly literature, very little attention has been paid to the phonological integration of Greek loanwords in Syriac. In this chapter, I have begun to remedy this situation by providing a description and analysis of the phonological integration of Greek loanwords in Syriac. I began with the integration of Greek consonants. I showed that the representation of Greek consonants in Syriac is remarkably regular and that almost all seeming deviations can be explained either by positing a Koinē Greek source that differs from Attic Greek or by appealing to secondary developments in Syriac. The majority of consonant correspondences are unremarkable, since Greek phonemes tend to be represented by very similar Syriac phonemes. One exception to this is the series of Greek voiceless stops (π, τ, and κ), which are not represented by the expected Syriac voiceless stops (‫ ܿܦ‬p, ‫ ܿܬ‬t, and ‫ ܿܟ‬k) but by the Syriac emphatic stops (‫ ܿܦ‬ṗ, ‫  ܛ‬ṭ, and ‫ ܩ‬q). In contrast to the consonants, the representation of Greek vowels, including Greek vowel hiatus, in Syriac is much less regular. While some Greek 105.  Nöldeke 1904: §40H. For the accommodation of Greek vowel hiatus in Postbiblical Hebrew and in various dialects of Jewish Aramaic, see Krauss 1898: §§138–51.

96

Chapter 5

loanwords in Syriac exhibit a stable orthography, the representation of Greek vowels with Syriac matres lectionis varies significantly in a large number of words. In some cases, this variation suggests that a word is closer to a Fremd­ wort than a Lehnwort. In other cases, however, the orthography of Greek loanwords in Syriac was updated over time. Often, this update resulted in an orthography that more closely represents the vowels in the Greek source, in line with the diachronic trend that Greek vowels tend to be represented more fully over time in Syriac. This contrasts with a number of contact situations cross-linguistically in which loanwords tend to become more integrated over time. Thus, Syriac writers can be seen updating the orthography of Greek loanwords, even well-established words, as the mechanisms for phonological integration shifted. Phonological integration—and, by extension, lexical transfer more broadly—was, then, not a one-point-in-time event for Syriac speakers. Rather, over time, they continued to interact with the Greek source not only by transferring new loanwords into their language but also by updating the loanwords that were already in their language. The dynamic nature of Greek loanwords in Syriac will continue to be explored in the next chapter on the morphosyntactic integration of Greek loanwords in Syriac (chap. 6).

Chapter 6

The Morphosyntactic Integration of Greek Loanwords in Syriac If loanwords are to be incorporated into the utterances of a new language, they must be fitted into its grammatical structure. (Haugen 1950b: 217)

6.1. Overview The previous chapter analyzed the phonological integration of Greek loanwords in Syriac. The current chapter turns to their morphosyntactic integration. In the scholarly literature, morphosyntactic integration has garnered the least attention of all the topics related to Greek loanwords in Syriac. Nöldeke (1904) devotes only a few sections to this subject throughout his grammar. 1 Schall (1960) fails to provide more than a couple passing remarks. More recently, Brock (1996: 254–56) has added several important pages to the discussion. 2 In this chapter, I aim to provide an overview of the morphosyntactic integration of Greek loanwords in Syriac. This chapter is organized according to parts of speech: nouns (§6.2), verbs (§6.3), and finally, particles (§6.4). 6.2. Nouns 6.2.1. Overview Greek nouns are marked for case, gender, and number. Five different grammatical cases are distinguished: vocative, nominative, genitive, dative, and accusative. Three genders are distinguished: masculine, feminine, and neuter. Two numbers are distinguished: singular and plural. 3 Syriac nouns are marked for gender, number, and state. Two genders are distinguished: masculine and feminine. Two numbers are distinguished: singular and plural. Three states are distinguished: status absolutus, status emphaticus, and status constructus. State is a morphosyntactic category in Aramaic. 1.  See, e.g., Nöldeke 1904: §§88–89, 202L. 2.  See earlier Brock 1967: 392–93. 3.  An earlier dual is preserved in a few remnants.

97

98

Chapter 6

The status constructus marks a noun that is dependent on a following noun, as in ‫ ܡܠܟܘܬ‬malkuṯ in the following example: (6.1) Peshiṭta Gospels (ca. 400; ed. Kiraz 1996) ‫ܙܥܘܪܐ ܕܝܢ ܒܡܠܟܘܬ ܫܡܝܐ ܪܒ ̄ܗܘ ܡܢܗ‬ zʿurɔ

den

small-m.sg.emp but

bmalkuṯ

šmayyɔ

in+kingdom-f.sg.con

heaven-m.pl.emp

raḇ

(h)u

mɛnneh

great-m.sg.abs

he

from+him

‘But, the least in the kingdom of heaven is greater than he’   (Matt 11:11) In earlier Aramaic, the status absolutus was the unmarked form of the noun. In Syriac, however, it only occurs in a limited number of syntactic uses, including in distributive repetition, after the quantifier kol ‘all’ and cardinal numerals, with negatives, in idiomatic expressions introduced by a preposition, with predicate adjectives, and in adverbial forms. 4 In earlier Aramaic, the status emphaticus was the definite form of a noun. In Syriac, however, it is the unmarked form of the noun. Four different categories are of primary interest in the morphosyntactic integration of Greek nouns in Syriac: Greek case endings (§6.2.3), gender (§6.2.4), number (§6.2.5), and Aramaic state (§6.2.6). 6.2.2.  Input Forms Before turning to the morphosyntactic integration of Greek nouns in Syriac, we need to look briefly at the various input forms that are attested for such words. The most common is the nominative singular—for example, κελλαρίτης (LSJ 937; PGL 741) > ‫ ܩܠܪܛܝܣ‬qlrṭys (with alternative orthographies) ‘steward’ (LS2 671; SL 1376) and θρόνος (LSJ 807) > ‫ ܬܪܘܢܘܣ‬trwnws (with alternative orthographies) ‘throne’ (LS2 836; SL 1665). The nominative singular is the citation form in Greek and so the most unmarked form. This fits well with the cross-linguistic tendency for the unmarked form to serve as the input form. 5 In addition to the nominative, the accusative is the only other case that commonly serves as an input form 6—for example, κλείς (LSJ 957) → accusative singular κλεῖδα > ‫ ܩܠܝܕܐ‬qlydʾ, ‫ ܐܩܠܝܕܐ‬ʾqlydʾ ‘key; clasp, buckle’ (LS2 667; SL 1370). The accusative singular also serves as an input form in other dialects 4.  Muraoka 2005: §72; Nöldeke 1904: §§205–10. 5.  The nominative singular is also the most common input form for Greek loanwords in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §§87–93). 6.  Brock 1967: 393; 1996: 254–55. For a possible example of the genitive as an input form for Greek nouns in Syriac, see n. 32 in chap. 4 above (p. 53) with the discussion in the main text.

The Morphosyntactic Integration of Greek Loanwords in Syriac

99

of Aramaic—for example, ἀνδριάς (LSJ 128) → accusative singular ἀνδριάντα > Palmyrene ʾdryṭ ‘statue’ (PAT 335; see also Brock 2005: 12, 25). 7 In addition to the singular, the plural serves as an input form for some Greek loanwords in Syriac. The Greek nominative plural ending ‑οι, for instance, is attested as an input form for Greek second-declension nouns in -ος: 8 (6.2) a. κληρικός (PGL 756) → nominative plural κληρικοί > pl ‫ܩܠܝ̈ܪܝܩܘ‬ qlyr̈yqw ‘clerics’ (LS2 671; SL 1371), singular attested as ‫ ܩܠܝܪܝܩܐ‬qlyryqʾ, ‫ ܩܠܝܪܝܩܘܣ‬qlyryqws, with additional plurals of ‫ ܩܠܝ̈ܪܝܩܐ‬qlyr̈yqʾ, ‫ ܩܠܝ̈ܪܝܩܘܣ‬qlyr̈qws b. ὀρθόδοξος (PGL 971–72; LSJ 1248) → nominative plural ὀρθόδοξοι > pl ‫ ܐܘ̈ܪܬܘܕܘܟܣܘ‬ʾwr̈twdwksw, ‫ ܐ̈ܪܬܘܕܘܟܣܘ‬ʾr̈twdwksw ‘orthodox (pl)’ (LS2 52; SL 105), singular attested as ‫ܐܪܬܕܘܟܣܐ‬ ʾrtdwksʾ, with additional plurals of ‫ ܐܘ̈ܪܬܘܕܘܟܣܐ‬ʾwr̈twdwksʾ, ‫ ܐ̈ܪܬܕܘܟܣܐ‬ʾr̈tdwksʾ It should be noted that singular forms exist alongside plural forms in both of these examples. This enabled the analogical creation of a new plural ending -w in Syriac (see §6.2.5). Other nominative plural forms may occasionally serve as an input form. The Greek nominative plural third-declension ending -ες could, for instance, be attested as an input form in σειρήν (LSJ 1588) → nominative plural σειρῆνες > pl ‫ ܣܝ̈ܪܝܢܣ‬syr̈yns ‘Sirens, name of an animal’ (LS2 499; SL 1007), with an additional plural of ‫ ܣܝ̈ܪܝܢܘܣ‬syr̈ynws. Alternatively, however, ‫ ܣܝ̈ܪܝܢܣ‬syr̈yns could be analyzed as a case of the analogically created plural ending ‑(ʾ)s or -(w)s. 9 To the preceding nominative plural input forms, Nöldeke (1904: §89) proposes that the nominative plural -αι occurs in cases such as διαθήκη (PGL 348; ̈ LSJ 394–95) → nominative plural διαθήκαι > ‫ܕܝܬܩܐ‬ dytq̈ʾ ‘covenant’ (LS2 152; SL 301), singular attested as ‫ ܕܝܬܩܐ‬dytqʾ (with alternative orthographies), ̈ ̈ dÿtqs. 10 There is, however, with additional plurals of ‫ܕܝܬܩܘܣ‬ dytqẅs and ‫ܕܝܬܩܣ‬ no clear evidence to substantiate Nöldeke’s claim, especially since the plural ̈ ‫ܕܝܬܩܐ‬ dytq̈ʾ could be analyzed as a Syriac plural formation with the masculineplural status emphaticus ending -e. The accusative plural also serves as an input form for some Greek loanwords in Syriac. This is the case, for instance, with Greek first-declension 7.  The accusative is also an input form for Greek loanwords in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §97). 8.  Nöldeke 1904: §89. The nominative plural is also attested as an input form for Greek loanwords in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §94). 9.  For the development of these endings, see §6.6.2. ̈ 10.  It should be noted that the plural ‫ܕܝܬܩܐ‬ dytq̈ʾ is very rare. The form is mentioned in ̈ the lexicon of Bar Bahlul (Duval 1888–1901: 1.574). The absolute form ‫ܕܝܬܩܝܢ‬ dyt¨qyn is, however, found in Ephrem, for example, Maḏrɔše against Julian the Apostate, 73.20 (ed. ̈ Beck 1957b), suggesting that the plural ‫ܕܝܬܩܐ‬ dytq̈ʾ also existed at this time.

100

Chapter 6

nouns that end in -ας in the accusative plural 11—for example, οὐσία (LSJ ̈ ̈ 1274–75) → accusative plural οὐσίας > pl ‫ܐܘܣܝܣ‬ ʾwsÿs, ‫ܐܘܣܝܐܣ‬ ʾwsÿʾs ‘essence, substance; wealth’ (LS2 9; SL 18), singular attested as ‫ ܐܘܣܝܐ‬ʾwsyʾ, ̈ with an additional plural of ‫ܐܘܣܝܘܣ‬ ʾwsÿws. It should be noted that a singular is attested for this word, which enabled the analogical creation of a new plural ending -(ʾ)s in Syriac. 12 The accusative plural -ους serves as an input form in some cases for Greek second-declension masculine and feminine nouns—for example, κληρικός (PGL 756) → accusative plural κληρικόυς > pl ‫ ܩܠܝ̈ܪܝܩܘܣ‬qlyr̈yqws ‘cleric’ (LS2 671; SL 1371), singular attested as ‫ ܩܠܝܪܝܩܐ‬qlyryqʾ, ‫ ܩܠܝܪܝܩܘܣ‬qlyryqws, with additional plurals of ‫ ܩܠܝ̈ܪܝܩܐ‬qlyr̈yqʾ , ‫ ܩܠܝ̈ܪܝܩܘ‬qlyr̈yqw. The existence of a nominative singular form alongside the accusative plural enabled the analogical creation of a new plural ending -ws in Syriac. 13 The Greek nominative plural ending -ες or the accusative plural ending -ας serves as an input form for the following third declension noun: πολύπους (LSJ 1441–42) → nominative plural πολύποδες, accusative plural πολύποδας > ̈ pl ‫ܦܘܠܘܦܕܣ‬ pwlwpd̈s ‘polyp’ (LS2 576; SL 1163), singular attested as ‫ܦܐܠܘܦܣ‬ pʾlwps, ‫ ܦܝܠܝܦܘܣ‬pylypws. The Greek plural ending -ματα serves as an input form for some Greek third-declension neuter nouns with stems in τ: (6.3) a. δόγμα (PGL 377–78; LSJ 441) → nominative/accusative plural ̈ dẅgmṭʾ, ‫ܕܘܓܡܐܛـܐ‬ ̈ δόγματα > pl ‫ܕܘܓܡܛܐ‬ dwgm̈ʾṭʾ ‘doctrine’ (LS2 141; SL 277–78), singular attested as ‫ ܕܘܓ�ܡܐ‬dwgmʾ, ̈ ̈ with additional plurals of ‫ܕܘܓܡܘ‬ dwgm̈w, ‫ܕܘܓܡܐ‬ dwgm̈ʾ b. φλέγμα (LSJ 1943) → nominative/accusative plural φλέγματα ̈ > pl ‫ܦܠܓܡܛܐ‬ plgm̈ṭʾ ‘phlegm’ (LS2 571; SL 1195), singular attested as ‫ ܦܠܓ�ܡܐ‬plgmʾ The input form in each of these cases could be analyzed as either nominative or accusative. The Greek plural ending -εις probably serves as an input form for some Greek third-declension nouns with stems in ι: 14 (6.4) a. αἵρεσις (PGL 51; LSJ 41) → nominative/accusative plural αἵρεσεις > pl ‫ ܗ̈ܪܣܝܣ‬hr̈sys, ‫ ܐ̈ܪܣܝܣ‬ʾr̈sys ‘difference, opinion; heresies’ (LS2 51, 184; SL 103, 180, 355), singular attested as ‫ ܗܪܣܝܣ‬hrsys, ‫ ܐܪܣܝܣ‬ʾrsys b. τάξις (LSJ 1756) → nominative/accusative plural τάξεις > pl ̈ ‫ܛܟܣܝܣ‬ ṭksÿs ‘order; rank’ (LS2 90, 274; SL 181, 529), singular 11.  Nöldeke 1904: §89. 12.  For this development, see §6.2.5 and §6.6.2 below. 13.  For this development, see §6.2.5 and §6.6.2 below. 14.  Nöldeke 1904: §89.

The Morphosyntactic Integration of Greek Loanwords in Syriac

101

attested as ‫ ܛܟܣܐ‬ṭksʾ, ‫ ܛܟܣܝܣ‬ṭksys, with an additional plural of ̈ ṭk̈sʾ ‫ܛܟܣܐ‬ The input form in each of these cases could be analyzed as either nominative or accusative. In addition, each of these cases could be alternatively analyzed as an instance in which the singular and plural have the same form (see §6.2.5). If so, the input form is the nominative singular. To the preceding nominative/accusative plural input forms, Nöldeke (1904: §89) proposes that the nominative/accusative plural -α occurs in cases such as εὐαγγέλιον (PGL 555–59; LSJ 705) → nominative/accusative plural εὐαγγέλια ̈ > pl ‫ܐܘܢܓܠܝܐ‬ ʾwnglÿʾ ‘gospel’ (LS2 8; SL 17–18), singular attested as ‫ܐܘܢܓܠܝܘܢ‬ ̈ ʾwnglywn. In this case, however, the plural ‫ܐܘܢܓܠܝܐ‬ ʾwnglÿʾ could be analyzed as a Syriac plural formation with the masculine plural status emphaticus ending -e (see §6.2.5). 15 Thus, it is impossible to determine whether or not the nominative/accusative plural -α also serves as an input form. Setting aside inflection, it should be noted that the diminutive serves as an input form for a number of Greek loanwords in Syriac: (6.5) a. ζώνη (LSJ 759) → ζωνάριον (LSJ 759) > ‫ ܙܘܢܪܐ‬zwnrʾ ‘belt’ (LS2 201; SL 373–74) b. θρόνος (LSJ 807) → θρονίον (LSJ 807) > ‫ ܬܪܘܢܝܘܢ‬trwnywn ‘seat, chair’ (LS2 836; SL 1665) It is interesting to note in this regard that diminutive forms are more common in Koinē Greek than earlier dialects. 16 This probably explains the relatively high number of diminutives that serve as input forms for Greek loanwords in Syriac. To summarize, the most common input form for Greek nouns in Syriac is the nominative singular. This fits well with the cross-linguistic tendency that the most unmarked form usually serves as the input form. The accusative is the only other case that is reliably attested as an input form. In addition to singular input forms, a number of Greek loanwords entered Syriac as plurals. In most (if not all) of these instances, the plural is attested as an input form only when the singular is also found. This suggests that there were multiple transfers of the same lexeme in (at least) two different forms. This is a reflection of the dynamic nature of lexical transfer in Greek-Syriac language contact. Over time, Syriac speakers continued to manipulate the Greek loanwords in their language on the basis of the Greek source language. In the case of input forms, they did this by transferring Greek plural forms into Syriac for Greek loanwords that already existed in their language in the singular. These Greek plural forms came 15.  It should be noted, however, that in his Letter on Syriac Orthography Yaʿqub of Edessa (d. 708) vocalizes this plural as if the source is εὐαγγέλια (ed. Phillips 1869: 7.6). 16.  Gignac 1976–81: 2.28; and especially Palmer 1945: 84–90.

102

Chapter 6

to be used as plurals for the words in question (§6.2.5) and provided the basis for the analogical creation of new plural markers in Syriac (§6.6.2). 6.2.3.  Greek Case Endings A Greek case ending can be accommodated in four possible ways in Syriac. First, it can be removed with the addition of a native Syriac ending—for example, ἰδιώτης (LSJ 819) > ‫ ܗܕܝܘܛـܐ‬hdywṭʾ ‘unskilled, simple, ordinary; stupid’ (LS2 171; SL 331). Second, it can be removed without the addition of a native Syriac ending—for example, βῆμα (PGL 295–296; LSJ 314) > ‫ ܒܝܡ‬bym ‘tribunal, raised platform, bema of a church’ (LS2 68; SL 141), alongside ‫ܒܝ�ܡܐ‬ bymʾ and ‫ ܒܐ�ܡܐ‬bʾmʾ. Third, it can be kept with the addition of a native Syriac ending—for example, νόμος (LSJ 1180) > ‫ ܢܡܘܣܐ‬nmwsʾ ‘law’ (LS2 431; SL 921–22). In most of these cases, the Greek case ending was retained in order to create a triliteral root in Syriac. Fourth, it can be kept without the addition of a native Syriac ending—for example, φύσις (LSJ 1964–65) > ‫ ܦܘܣܝܣ‬pwsys ‘nature’ (LS2 582; SL 1167). The two most common accommodation strategies for Greek case endings in Syriac involve either the removal of the Greek case ending with the addition of a native Syriac ending or the retention of the Greek case ending without the addition of a native Syriac ending. That is, both result in an ending from only one of the languages, whether fully Greek or fully Syriac. In most cases, there is a clear tendency to associate one of these four strategies with a particular noun class. Third-declension nouns in -ις, for instance, tend to retain the Greek ending without the addition of a Syriac ending, whereas first-declension nouns in -ης tend to replace the Greek ending with a Syriac ending. The motivating factors for this distribution, however, remain unclear. Finally, it should be noted that it is not rare for the same Greek loanword to be accommodated according to different strategies—for example, τάξις (LSJ 1756) > ‫ ܛܟܣܐ‬ṭksʾ ‘order; rank’ and ‫ ܛܟܣܝܣ‬ṭksys (LS2 90, 274; SL 181, 529). This suggests either that the same Greek loanword was transferred into Syriac on multiple occasions or that Syriac speakers reaccommodated a Greek loanword on the basis of the Greek source. 6.2.4. Gender Greek has three genders (masculine, feminine, and neuter), whereas Syriac has only two (masculine and feminine). Greek masculine nouns are usually realized as masculine in Syriac 17—for example, masc. γένος (LSJ 344) > masc. ‫ ܓܢܣܐ‬gnsʾ ‘kind, species; family; race, nation’ (LS2 125, 841; SL 179, 249) and masc. τύπος (LSJ 1835) > masc. ‫ ܛܘܦܣܐ‬ṭwpsʾ ‘example, copy; shape, form; symbol; edict’ (LS2 286; SL 520, 1464). Rarely, however, Greek 17.  Nöldeke 1904: §88.

The Morphosyntactic Integration of Greek Loanwords in Syriac

103

masculine nouns are realized as feminine in Syriac. 18 Several of these cases may be due to secondary developments in Syriac—for example, masc. δρόμων (LSJ 450) > fem. ‫ ܕܪܡܘܢ‬drmwn ‘ship, boat’ (LS2 167; SL 324) and masc. κέρκουρος (LSJ 943) > fem. ‫ ܩܪܩܘܪܐ‬qrqwrʾ ‘light boat’ (LS2 701; SL 1416; see also Nöldeke 1904: §88) due to the feminine gender of ‫ ܐܠܦܐ‬ʾɛllp̄ɔ ‘boat’ (LS2 22, 921; SL 50–51) < Akkadian elippu (CAD E 90–95; see also Kaufman 1974: 48). In other cases, the feminine gender may have been phonologically motivated—for example, masc. μαργαρίτης (LSJ 1080) > fem. ‫ ܡܪܓܢܝܬܐ‬mrgnytʾ ‘pearl; Eucharistic wafer’ (LS2 402; SL 826), since the Syriac form has the feminine ending -tɔ presumably (though irregularly) corresponding to Greek τ (see §5.2.2). In other cases, there are not any clear motivating factors—for example, masc. θρόνος (LSJ 807) > fem. ‫ ܬܪܘܢܘܣ‬trwnws (with alternative orthographies) ‘throne’ (LS2 836; SL 1665). Greek feminine nouns are usually realized as feminine in Syriac 19—for example, fem. πολιτεία (LSJ 1434) > fem. ‫ ܦܘܠܝܛܝܐ‬pwlyṭyʾ ‘republic, state’ (LS2 574, 848; SL 1164) and fem. ὕλη (LSJ 1847–48) > fem. ‫ ܗܘ�ܠܐ‬hwlʾ (with alternative orthographies) ‘woods, forest; matter, material; firewood’ (LS2 173, 175; SL 335, 341). Occasionally, however, Greek feminine nouns are realized as masculine in Syriac. 20 Most of the cases are probably to be explained by the accommodation of final Greek ‑α by Syriac -ʾ, which was probably understood as the ending -ɔ of masculine singular nouns in the status emphaticus—for example, fem. ἀπουσία (LSJ 225) > masc. ‫ ܐܦܘܣܝܐ‬ʾpwsyʾ ‘waste, excrement; latrine’ (LS2 41; SL 83) and fem. στοά (LSJ 1647) > masc. ‫ ܐܣܛܘܐ‬ʾsṭwʾ ‘portico’ (LS2 32; SL 68). In other cases, there are not any clear motivating factors—for example, fem. τάξις (LSJ 1756) > masc. ‫ ܛܟܣܐ‬ṭksʾ, ‫ ܛܟܣܝܣ‬ṭksys ‘order; rank’ (LS2 90, 274; SL 181, 529). Syriac has no neuter gender, and so Greek neuter nouns must be accommodated in Syriac either as masculine and/or feminine. Greek neuter nouns are usually realized as masculine in Syriac 21—for example, neut. πρόσωπον (LSJ 1533) > masc. ‫ ܦܪܨܘܦܐ‬prṣwpʾ ‘face, countenance; person, party’ (LS2 605; SL 1249–50). Greek neuters are, however, also realized as feminine in Syriac 22— for example, neut. θέατρον (LSJ 787) > fem. ‫ ܬܐܛܪܘܢ‬tʾṭrwn ‘theater; spectacle’ (LS2 813; SL 1618). The realization of Greek neuter nouns as masculine is more common than feminine by approximately three to two. There are no discernible motivations for the accommodation of a particular Greek neuter noun as either masculine or feminine. In a few rare cases, a Greek neuter noun is found with both genders in Syriac—for example, neut. ξενοδοχεῖον (LSJ 18.  Brock 1996: 256; Nöldeke 1904: §88. 19.  Nöldeke 1904: §88. 20.  Nöldeke 1904: §88; Brock 1996: 256. 21.  Brock 1996: 256; Nöldeke 1904: §88. 22.  Brock 1996: 256; Nöldeke 1904: §88.

104

Chapter 6

1189) > masc./fem. ‫ ܐܟܣܢܘܕܘܟܐ‬ʾksnwdwkʾ, ‫ ܐܟܣܢܕܘܟܝܘܢ‬ʾksndwkywn, ‫ܟܣܢܕܟܝܢ‬ ksndkyn, ‫ ܐܟܣܢܕܟܝܢ‬ʾksndkyn, ‫ ܟܣܢܘܕܘܟܝܢ‬ksnwdwkyn ‘hospital’ (LS2 338, 846; SL 44, 640) and neut. τάγμα (LSJ 1752) > masc./fem. ‫ ܬܓ�ܡܐ‬tgmʾ, ‫ܛܓ�ܡܐ‬ ṭgmʾ ‘order, class; command, precept; troop, cohort’ (LS2 92, 268, 816; SL 185, 512, 1623). In addition to the cases discussed above, a Greek loanword is rarely found with both genders in Syriac. This occurs for different reasons. In some cases, it is due to the Greek source, which itself attests multiple genders—for example, masc./fem. ἀήρ (LSJ 30) > mostly fem., occasionally masc. ‫ ܐܐܪ‬ʾʾr ‘air’ (LS2 1; SL 1). 23 In other cases, however, a Greek loanword in Syriac takes both masculine and feminine agreement due to an inner Syriac development. This is most common with feminine Greek nouns that end in final -ʾ in Syriac—for example, fem. σειρά (LSJ 1588) > masc./fem. ‫ ܣܝܪܐ‬syrʾ ‘thread; chain’ (LS2 472; SL 1007) and fem. χώρα (LSJ 2015) > masc./fem. ‫ ܟܘܪܐ‬kwrʾ ‘land, province’ (LS2 323; SL 612). In both of these cases, the transfer of the feminine Greek word to masculine gender in Syriac is to be explained by an inner Syriac development based on the form of the word—that is, most Syriac nouns ending in ­-ʾ are masculine (as opposed to feminines in -tʾ). Finally, there are cases in which it is unclear why a Greek loanword is attested with multiple genders in Syriac—for example, Latin masc. uncinus (OLD 2090; LD 1929) > masc. ὄγκινος (LSJ 1196) > masc./fem. ‫ ܐܘܩܝܢܐ‬ʾwqynʾ ‘hook; anchor; sailors’ sounding line’ (LS2 9; SL 20). Few conclusions can be drawn from the accommodation of the gender of Greek loanwords in Syriac. The fact that the majority of Greek loanwords in Syriac retain the gender of the Greek source does, however, suggest a relatively high degree of bilingualism for at least part of the Syriac-speaking population. 24 In contrast, the change of Greek feminine nouns ending in -α to masculine nouns in Syriac due to the fact that they end in -ʾ shows that some Syriac speakers lost sight of the Greek source. 6.2.5. Number Greek loanwords in Syriac are declined for number either according to Syriac morphology or according to Greek morphology. Many Greek loanwords in Syriac attest multiple plural formations. Most Greek loanwords in Syriac are declined for number according to Syriac morphology. 25 The Syriac masculine plural is illustrated in the following representative examples: 23.  A similar phenomenon is found with masc. χάρτης (LSJ 1980) > masc./fem. ‫ܟܪܛܝܣܐ‬ krṭysʾ, ‫ ܩܪܛܝܣܐ‬qrṭysʾ ‘sheet of paper; papyrus’ (LS2 344, 695; SL 650, 1405–6), where the feminine gender is to be explained by the feminine Latin charta (OLD 309; LD 325). 24.  For a similar argument involving French loanwords in Brussels Flemish, see Winford 2003: 49–50. 25.  This is also the case for Greek loanwords in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §§315–25).

The Morphosyntactic Integration of Greek Loanwords in Syriac

105

(6.6) a. ἀγών (PGL 25; LSJ 18–19) > ‫ ܐܓܘܢܐ‬ʾgwnʾ ‘struggle’ (LS2 4; ̈ SL 6) → pl ‫ܐܓܘܢܐ‬ ʾgẅnʾ b. χειμών (LSJ 1983) > ‫ ܟܝܡܘܢܐ‬kymwnʾ (LS2 332; SL 619) → pl ̈ ‫ܟܝܡܘܢܐ‬ kymẅnʾ ‘storm’ The Syriac feminine plural is illustrated in the following representative examples: (6.7) a. λόγχη (LSJ 1059) > ‫ ܠܘܟܝܬܐ‬lwkytʾ ‘spear’ (LS2 361; SL 679) → ̈ lẅkytʾ pl ‫ܠܘܟܝܬܐ‬ b. μηχανή (LSJ 1131) > ‫ �ܡܐܟܢܐ‬mʾknʾ, ‫ ܡܝܟܢܐ‬myknʾ ‘machine, siege engine; irrigated land’ (LS2 385, 846; SL 701) → pl ̈ ̈ ‫�ܡܐܟܢܘܬܐ‬ mʾk̈nwtʾ, with additional plurals of ‫�ܡܐܟܢܣ‬ mʾk̈ns and ̈ ‫ �ܡܐܟܢܘܣ‬mʾk̈nws The Syriac masculine plural ending is significantly more common than the feminine plural. More rarely, Greek loanwords in Syriac can be declined for number according to Greek morphology. Alongside the singular, the plural also serves as an input form for some Greek loanwords in Syriac. 26 This is the case, for instance, with Greek second-declension nouns with nominative singular -ος ~ nominative plural -οι. The plural αἱρετικοί, for instance, was transferred into Syriac as ‫ ܗ̈ܪܛܝܩܘ‬hr̈ṭyqw along with the singular αἱρετικός (PGL 51) > ‫ ܗܪܛܝܩܐ‬hrṭyqʾ ‘heretical, schismatic’ (LS2 183; SL 354). The ending -w in ‫ ܗ̈ܪܛܝܩܘ‬hr̈ṭyqw marks plurality. This ending ‑w functions as a plural marker for many other Greek loanwords in Syriac that have a corresponding Greek plural in ‑οι. The ending ‑w is also found as a plural marker with Greek loanwords that do not have a corresponding plural in ‑οι in the source language. This is the case, for instance, with δόγμα (PGL 377–78; LSJ 441) > ‫ ܕܘܓ�ܡܐ‬dwgmʾ ‘doctrine’ (LS2 ̈ 141; SL 277–78), one of the plurals of which is ‫ܕܘܓܡܘ‬ dwgm̈w. The plural ̈ ending -w in ‫ ܕܘܓܡܘ‬dwgm̈w is due to an inner Syriac analogy: ̈ (6.8) ‫ ܗܪܛܝܩܐ‬hrṭyqʾ : ‫ ܗ̈ܪܛܝܩܘ‬hr̈ṭyqw :: ‫ ܕܘܓ�ܡܐ‬dwgmʾ : X = ‫ܕܘܓܡܘ‬ dwgm̈w

This analogy led to the creation of a new plural ending -w that is used with Greek loanwords in Syriac that do not have a Greek plural in -οι such as ̈ ‫ܕܘܓܡܘ‬ dwgm̈w (the expected Greek nominative plural is δόγματα). The plural ending -w is not used with native Syriac words in contrast to the plural endings ‑(w)s and -(ʾ)s. 27 In addition to the case above that involves the Greek nominative singular and plural, the nominative singular and accusative plural serve as input forms 26.  For the Greek input forms, see §6.2.2. 27.  This is discussed immediately below as well as in §6.6.2.

106

Chapter 6

for some Greek loanwords in Syriac. This is the case, for instance, with Greek second-declension nouns with nominative singular -ος ~ accusative plural -ους. The accusative plural σύγκλητους, for instance, was transferred into Syriac ̈ sẅnqlyṭws, along with the nominative singular σύγκλητος (LSJ as ‫ܣܘܢܩܠܝܛܘܣ‬ 1665) > ‫ ܣܘܢܩܠܝܛܘܣ‬swnqlyṭws ‘senate; senator’ (LS2 486; SL 984–85). As in the case of -w discussed above, a new plural ending -ws was created by analogy in Syriac. This new plural ending -ws is found with Greek loanwords in Syriac that do not have a corresponding Greek accusative plural in -ους 28—for example, ἀήρ (LSJ 30) > ‫ ܐܐܪ‬ʾʾr ‘air’ (LS2 1; SL 1) → pl ‫ ܐܐ̈ܪܘܣ‬ʾʾr̈ws (the expected Greek plurals are nom ἀέρες and acc ἀέρας). The new Syriac plural ending -ws that is illustrated in this example is also rarely found with native Syriac words. 29 The Greek plural also serves as an input form with Greek first-declension nouns with nominative-singular -η (or -α) ~ accusative plural -ας. The ̈ accusative plural ἀνάγκας, for instance, was transferred into Syriac as ‫ܐܢܢܩܣ‬ ʾnnq̈s, along with the nominative-singular ἀνάγκη (LSJ 101) > ‫ ܐܢܢܩܐ‬ʾnnqʾ (with alternative orthographies) ‘necessity’ (LS2 29; SL 63). The ending -(ʾ)s is one of the regular plural formations for Greek loanwords in -η (or -α) in Syriac. As in the case with -w and -ws, a new plural ending -(ʾ)s was created by analogy in Syriac—for example, Latin velum (OLD 2024; LD 1965–66) > βῆλον (PGL 295) > ‫ ܘ�ܠܐ‬wlʾ, ‫ ܘܐ�ܠܐ‬wʾlʾ ‘veil, curtain’ (LS2 185; SL 358) → pl ‫̈ܘܠܣ‬ ẅls. Since the plural ending -ws can be written defectively as -s, it cannot be ruled out that this example represents the plural ending -ws. The existence of the plural ending -(ʾ)s can, however, be definitively established by the writing of -ʾs, which occurs rarely with native Syriac words. 30 The Greek plural ending -ματα serves as a plural marker for some Greek third-declension neuter nouns with stems in τ—for example, δόγμα (PGL 377– ̈ dẅgmṭʾ, 78; LSJ 441) → nominative/accusative plural δόγματα > pl ‫ܕܘܓܡܛܐ‬ ̈ ‫ ܕܘܓܡܐܛـܐ‬dwgm̈ʾṭʾ ‘doctrine’ (LS2 141; SL 277–78), singular attested as ̈ ̈ ‫ ܕܘܓ�ܡܐ‬dwgmʾ, with additional plurals of ‫ܕܘܓܡܘ‬ dwgm̈ʾ, ‫ܕܘܓܡܐ‬ dwgm̈ʾ. In this case, the existence of singular and plural forms did not lead to the analogical creation of a new plural ending. In a few, rare cases, the singular and the plural are exactly the same for a Greek loanword in Syriac—for example, μάγγανον (LSJ 1070) > ‫ܡܢܓܢܘܢ‬ mngnwn, pl ‫ ̈ܡܢܓܢܘܢ‬m̈ngnwn ‘instrument of torture’ (LS2 394; SL 780). Most Greek loanwords in Syriac are declined for number either according to Syriac morphology or according to Greek morphology. Prima facie, one might expect this distinction to correlate roughly with Lehnwörter and Fremdwörter, 28.  See already Schall 1960: 99. 29.  For this development, see §6.6.2. 30.  See §6.6.2 below.

The Morphosyntactic Integration of Greek Loanwords in Syriac

107

respectively. 31 That is, Greek loanwords that decline for number according to Greek morphology might be expected to be closer to Fremdwörter than Lehnwörter. Interestingly, however, many of the words that take Greek plural morphology are among the most commonly attested Greek words in Syriac. In addition, they seem to be accommodated fully in Syriac in all other regards. This is the case, for instance, with Greek ἀνάγκη > Syriac ‫ ܐܢܢܩܐ‬ʾnnqʾ (with alternative orthographies) ‘necessity’, which has Greek morphology plurals of ̈ ̈ ‫ܐܢܢܩܘܣ‬ ʾnnq̈ws and ‫ܐܢܢܩܣ‬ ʾnnq̈s. Where does a word such as this fall on the continuum of Fremdwörter versus Lehnwörter? Its plural morphology suggests Fremdwort, since Greek plural morphology in Syriac is predominantly linked to words of Greek origin. 32 In all other regards, however, the word is a fully incorporated Lehnwort. It is, for instance, attested in Syriac already in the early third-century Book of the Laws of the Countries (6.17, 60.12; ed. Drijvers 1965), and it occurs frequently in Syriac texts of all genres from all time periods. In addition, setting aside its plural morphology, the word is fully accommodated in Syriac. This situation is not restricted to Syriac ‫ ܐܢܢܩܐ‬ʾnnqʾ but exists for a number of the Greek words in Syriac that attest Greek plural morphology. The fact that a Syriac word such as ‫ ܐܢܢܩܐ‬ʾnnqʾ ‘necessity’ occurs with Greek plural morphology suggests at the very least that Syriac speakers categorized it with a number of other words that had marked plural morphology (known by the contact linguist to be of Greek origin ultimately). This could indicate that the word is not entirely on the Lehnwörter side of the continuum but that it appears a little to the Fremdwörter side. Another interpretation is also possible, however, and is in fact more likely, given what is known about the Syriac-Greek contact situation more generally. The Greek plural morphology of Syriac ‫ ܐܢܢܩܐ‬ʾnnqʾ ‘necessity’ is probably a reflection of the dynamic nature of Greek lexical transfer in Syriac. That is, even though Greek ἀνάγκη was transferred into Syriac by at least the second century, some Syriac speakers never entirely disconnected the Syriac word from its Greek source, since they continued to be in contact with Greek. This connection is what provided the basis for the word to continue to take a Greek type of plural morphology. It is interesting in this regard that Syriac ‫ ܐܢܢܩܐ‬ʾnnqʾ not only attests a plural ̈ of ‫ܐܢܢܩܣ‬ ʾnnq̈s, which accurately reflects the Greek plural ἀνάγκας, but also a ̈ plural of ‫ܐܢܢܩܘܣ‬ ʾnnq̈ws, the ending of which reflects a different Greek plural ̈ of -ους. The plural ‫ܐܢܢܩܘܣ‬ ʾnnq̈ws rules out the interpretation of the word as a single-word code-switch. It also indicates that Syriac ‫ ܐܢܢܩܐ‬ʾnnqʾ takes a Greek-looking plural morphology that does not necessarily reflect the Greek source accurately. Thus, Syriac speakers categorized a word such as ‫ܐܢܢܩܐ‬ 31.  For this continuum, see the discussion in §4.5. 32.  For the few exceptions that result in new plural endings in Syriac, see §6.6.2.

108

Chapter 6

ʾnnqʾ as taking a special type of plural marking, and it is likely that this categorization was based on an active knowledge that the word was from Greek. The dynamic nature of Greek lexical transfer in Syriac is also evidenced by the fact that many Greek loanwords in Syriac attest multiple plural formations. The plural of ‫ ܕܘܓ�ܡܐ‬dwgmʾ ‘doctrine’ (< δόγμα [PGL 377–78; LSJ 441]), for instance, is attested with three different plural endings: (1) the native Syriac ̈ dẅgmṭʾ ( ‫ ܐ̈ܪܟܣ‬ʾr̈ks + ‑ɔyɔ → ‫ ܐܪܟܘܣܝܐ‬ʾrkwsyʾ ‘angelic’ (LS2 49; SL 100; see also Brock 1967: 393 n. 12; 1996: 260–61 with n. 32) b. εὐαγγέλιον (PGL 555–59; LSJ 705) > ‫ ܐܘܢܓܠܝܘܢ‬ʾwnglywn ‘gospel’ (LS2 8; SL 17–18) + -ɔʾiṯ → ‫ ܐܘܢܓ�ܠܐܝܬ‬ʾwnglʾyt ‘according to the gospel’ (7th cent. Sahdona, Works, 3.112.23 [ed. de Halleux 1960–65]; see also Brock 1996: 260) c. φύσις (LSJ 1964–65) > ‫ ܦܘܣܝܣ‬pwsys ‘nature’ (LS2 582; SL 1167) + -ɔyɔ → ‫ ܦܘܣܝܣܝܐ‬pwsysyʾ ‘of natural philosophy’ (LS2 582; SL 1167; see also Brock 1996: 260–61 with n. 32) d. ὠκεανός (LSJ 2031) > ‫ ܐܘܩܝܢܘܣ‬ʾwqynws ‘ocean’ (LS2 9; SL 20) + -ɔʾiṯ → ‫ ܐܘܩܝܢܐܝܬ‬ʾwqynʾyt ‘like an ocean’ (LS2 9; SL 20) Thus, even though these words are not integrated into the system of Syriac state, they still fall closer to the Lehnwörter side of the continuum, since they undergo secondary nominal derivations. 6.2.7. Summary There is a broad range in the degree of integration of Greek nouns in Syriac. Some nouns have been fully integrated into Syriac and behave just as native vocabulary by declining according to the normal rules of Syriac gender, number, and state. Other nouns, in contrast, express number through Greek morphology and/or do not follow the normal morphosyntactic rules for state in Syriac. In addition, it should be noted that some Greek loanwords encounter particular problems in their integration into Syriac. A good example of this is found with Greek first-declension nouns ending in -η. The final η in these words is phonologically similar to the Syriac masculine plural status emphaticus ending ‑e. This occasionally leads to singular forms being written with Syriac syɔme, which typically marks the morphological category of plurality. Consider, for instance, the word ‫ ܛ̈ܪܘܦܐ‬ṭr̈wpʾ ‘nourishment’ (< τροφή) in the following clause:

35.  Brock 1996: 260 n. 32. Secondary derivations involving Greek loanwords in Syriac are analyzed in §6.5.3; and (with more detail) in Butts 2014.

The Morphosyntactic Integration of Greek Loanwords in Syriac

111

(6.13) Babai the Great, Commentary on the ‘Gnostic Chapters’ by Evagrius of Pontus (ed. Frankenberg 1912). ܿ ‫ܐܝܬܝܗ ܕܟܝܘܬܐ ܘܚܘܪܪܐ ܕܡܢ ̈ܚܫܐ‬ ‫ܛ̈ܪܘܦܐ ܕܗܝ‬ tr̈wpʾ

dhi

ʾiṯeh

daḵyuṯɔ

nourishment

nml+she

ex+her

purity-f.sg.emp

wḥurrɔrɔ

dmɛn

and+freedom‑m.sg.emp nml+from

ḥašše suffering-m.pl.emp

‘nourishment, which is purity and freedom from suffering’ (468.14–15) One expects a form such as ‫ ܛ̈ܪܘܦܐ‬ṭr̈wpʾ ‘nourishment’ with syɔme to be plural. The agreement markers in the clause, however, indicate that it is singular (see the singular pronoun hi and the singular possessive pronominal suffix -eh in ʾiṯeh). Thus, the syɔme in this word is due to the phonological similarity between the final mid front vowel of the Greek source and the Syriac masculine plural status emphaticus ending -e. 36 Occasionally, this similarity leads to the use of singular words with pronominal suffixes as if they were plural, as in the following example: (6.14) Maḏrɔše on Virginity by Ephrem (d. 373; ed. Beck 1962) ̈ ‫ܕܝܬܩܘܗܝ ܕܡܘܫܐ ܠܣܒܪܬܗ ܣܟܝܬ‬ dÿtqwhy

dmuše

convenant-f.sg.con+his

nml+pn to+good.news-f.sg.con+his

lasḇarṯeh

sakkyaṯ wait-suf.3.f.sg

‘The covenant of Moses awaited his good news’ (32.1) ̈ dÿtqwhy (< διαθήκη) must be analyzed as a singular In this example, ‫ܕܝܬܩܘܗܝ‬ noun, given the verbal agreement with sakkyaṯ (3.f.sg), but it takes the pronominal suffixes of a plural noun—that is, ‑aw(hy) instead of -eh. Thus, in this case, the final η of the Greek input form has led to the loanword’s adopting plural morphology. While these examples represent extreme cases, they do serve to illustrate some of the problems that Greek nouns can encounter when they are integrated into Syriac. 6.3. Verbs Syriac contains a number of verbs that are ultimately of Greek origin. A majority of the verbal roots in Syriac that are Greek in origin are denominative 36.  For further analysis with additional examples, see Butts 2015b.

112

Chapter 6

formations. 37 The Syriac verbal root ṭgn d ‘to fry, roast; to torture’, for instance, is derived from the noun ‫ ܛܓܢܐ‬ṭgnʾ ‘frying pan’, which was transferred from Greek τήγανον ‘frying pan’ (LSJ 1786). As this example illustrates, most transitive denominative roots from Greek loanwords that are triliteral occur in the d-stem in Syriac; the c-stem and g-stem also occur, though less commonly. Passives of these denominatives are formed with the respective t-stems. This follows the typical pattern for denominative verbs in Syriac—for example, ‫ ܐܠܗܐ‬ʾalɔhɔ ‘god’ (LS2 21; SL 47) → ‫ ܐܠܗ‬ʾlh d ‘to deify’, dt ‘to be deified’ (LS2 21; SL 47). Denominative verbs from Greek loanwords that involve more than three root consonants follow the typical pattern for these roots in Syriac. 38 These denominative verbs represent the majority of Syriac verbal roots that are of Greek origin. It should be noted, however, that they are not loanwords in the strict sense, but they are, rather, secondary formations from Greek loanwords. In addition to denominative verbs, Syriac contains a small number of verbs that are accommodated according to a strategy known as light verb in Wohlge­ muth’s typological study of the accommodation of loanverbs in the world’s languages (2009). In Wohlgemuth’s typology, a light verb is an accommodation strategy in which a loanverb is employed in combination with a light verb such as ‘to do’, ‘to make’, or ‘to be’ from the recipient language, which bears the inflection and/or grammatical information. 39 Light verb accommodation is illustrated in example (6.15). (6.15) Bohairic Coptic naferdokimazin

mmof

pe

do-past.imperfect.3.m.sg+tempt

dom+him

he

‘he was tempting him’ (ed. Van Rompay apud Datema 1978:  275.28) In this example, dokimazin (< δοκιμάζειν ‘to tempt’ [LSJ 442]) is used in combination with the native Coptic verb er ‘to do’ (Crum 1929–39: 83–84), which bears the grammatical information. The verb er functions almost as an auxiliary with the semantic information contained in the loanverb. Use of a light verb is the second most common accommodation strategy for loanverbs cross-linguistically.

37.  Brock 1967: 401; 1975: 87–88; 1996: 257; 2004: 31–32, 35; Ciancaglini 2008: 8–9. Denominative verbs involving Greek loanwords are also common in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §§280–82). 38. See Nöldeke 1904: §§180–82. 39.  Wohlgemuth 2009: 102–17; Wichmann and Wohlgemuth 2008: 93–96. This is roughly equivalent to “bilingual compound verbs” in Muysken 2000: 193–206. For the term “light verb,” see Wichmann and Wohlgemuth 2008: 91, with reference to Jespersen 1954: 117–18.

The Morphosyntactic Integration of Greek Loanwords in Syriac

113

Light verb strategy is occasionally found in Syriac texts not translated from Greek prior to Yaʿqub of Edessa (d. 708). 40 In the active voice, the Syriac verbal root ‫ ܥܒܕ‬ʿbd ‘to do, make’ (LS2 504, 848; SL 1054–56) is used with a transferred Greek aorist active infinitive: (6.16) Scholia by Yaʿqub of Edessa (d. 708; ed. Phillips 1864) ‫ܿܨܒܐ ܗܘܐ ܕܦܠܝܪܘܦܘܪܝܣܐ ܢܥܒܕܝܘܗܝ‬ ṣɔḇe

(h)wɔ

dplyrwpwrysʾ nɛʿbḏiw(hy)

want-part.m.sg.abs

be-suf.3.m.sg

nml+inform

do-pre.3.m.sg+him

‘he wanted to inform him’ (3.17) In this example, a conjugated form of the verbal root ‫ ܥܒܕ‬ʿbd occurs with ‫ ܦܠܝܪܘܦܘܪܝܣܐ‬plyrwpwrysʾ, which derives from the Greek aorist active infinitive πληροφορῆσαι (LSJ 1419). It should be noted that in this strategy Greek verbs are not accommodated to the root-and-pattern morphology of Syriac. The following Greek aorist infinitives occur with ‫ ܥܒܕ‬ʿbd in Syriac texts not translated from Greek up to Yaʿqub of Edessa (d. 708): (6.17) a. κοσσίσαι, see κόσσος ‘box on the ear, cuff’ (LSJ 985) > ‫ ܩܘܣܝܣܐ‬qwsysʾ with ‫ ܥܒܕ‬ʿbd ‘to give a blow on the ear’ (6th cent. Yuḥanon of Ephesus, Lives of the Eastern Saints, 516.10 [ed. Brooks 1923–25]; LS2 680; SL 1059) b. πληροφορῆσαι (LSJ 1419) > ‫ ܦܠܝܪܘܦܘܪܝܣܐ‬plyrwpwrysʾ with ‫ ܥܒܕ‬ʿbd ‘to exercise (office); to inform; to satisfy’ (7th cent. Yaʿqub of Edessa, Scholia, 3.17 [ed. Phillips 1864]; LS2 577; SL 1058–59) c. προσελθῆναι (with passive morphology), for the expected aoristinfinitive προσελθεῖν (LSJ 1511) > ‫ ܦܪܣܠܬܝܢܐ‬prsltynʾ with ‫ܥܒܕ‬ ʿbd ‘to present a petition to the emperor’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 140.18 [ed. Brooks 1935]; LS2 602, 848; SL 1059) d. συγκροτῆσαι (LSJ 1667) > ‫ ܣܘܢܩܪܛܝܣܐ‬swnqrṭysʾ with ‫ ܥܒܕ‬ʿbd ‘to unite closely’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 37.17 [ed. Brooks 1935]; LS2 486; SL 1058) e. τιτλῶσαι (LSJ 1799) > ‫ ܛܝܛܠܘܣܐ‬ṭyṭlwsʾ with ‫ ܥܒܕ‬ʿbd ‘to make a public case of’ (6th cent. Eliya, Life of Yuḥanon of Tella, 77.7 [ed. Brooks 1907: 29–95]; LS2 273; SL 526, 1057) 41 40.  Brock 1996: 257–58; 1975: 88; 2004: 37; Ciancaglini 2006; 2008: 10, 48–52; Schall 1960: 248. 41.  Alternatively, this word could derive from τίτλος (LSJ 1799). This would, however, be an unmotivated case of the retention of Greek -ς, and thus it seems more likely that the Greek source is the aorist infinitive τιτλῶσαι (so Ghanem 1970: 142 n. 268; LS2 273; SL 1057).

114

Chapter 6

In the passive voice, the Syriac verbal root ‫ ܗܘܝ‬hwy ‘to be(come)’ (LS2 173; SL 333–34) is used with the Greek aorist passive infinitive: 42 (6.18) Works by Rabbula of Edessa (d. 435/436; ed. Overbeck 1865: 210–48, 362–81) ‫ܘܡܢ ܦܪܕܝܣܐ ܿܗܘ ܥܕܢܝܐ ܐܟܣܘ̈ܪܣܬܝܢܐ ܗܘܝܢܢ‬ wmɛn

pardaysɔ

and+from

paradise-m.sg.emp that-m of.Eden-m.sg.emp

haw

ʿḏɛnɔyɔ

ʾkswr̈stynʾ to.be.exiled

hɔwenan be-part.m.pl.abs+we

‘and we were exiled from that paradise of Eden’ (365.14) In this example, a conjugated form of the verbal root ‫ ܗܘܝ‬hwy occurs with ‫ ܐܟܣܘ̈ܪܣܬܝܢܐ‬ʾkswr̈stynʾ, which derives from the Greek aorist passive infinitive ἐξορισθῆναι (LSJ 598). The following Greek aorist passive infinitives occur with ‫ ܗܘܝ‬hwy in Syriac texts not translated from Greek up to Yaʿqub of Edessa (d. 708): (6.19) a. ἐξορισθῆναι (LSJ 598) > ‫ ܐܟܣܘܪܣܬܝܢܐ‬ʾkswrstynʾ with ‫ ܗܘܝ‬hwy ‘to be exiled’ (5th cent. Rabbula of Edessa, Works, 365.14 [ed. Overbeck 1865: 210–48, 362–81]; LS2 19; SL 334) b. χειροτονηθῆναι (PGL 1522–23) > ‫ ܟܝܪܘܛܘܢܝܬܝܢܐ‬kyrwṭwnytynʾ with ‫ ܗܘܝ‬hwy ‘to be ordained’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 173.4 [ed. Brooks 1935]; LS2 344; SL 334) c. σφυρισθῆναι (LSJ Supplement 289) > ‫ ܐܣܦܪܣܬܝܢܐ‬ʾsprstynʾ with ‫ ܗܘܝ‬hwy ‘to be struck with hammers, beat’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 16.3 [ed. Brooks 1935]; LS2 37; SL 1057) 43 42.  For a possible case of the Greek aorist passive infinitive used with Syriac ʿbd, see n. 43 below. 43.  The verb occurs in the following passage: ‫ܘܦܩܕ ܠܗܘܢ‬ ‫ܫܕܪ ܩܠܝ̈ܪܝܩܘ ܘܣܩܘܒܝܛܘ̈ܪܣ‬ ܼ ‫ܕܐܣܦܪܣܬܝܢܐ ܢܥܒܕܘܢܝܗܝ ܥܕ�ܡܐ ܕܓܣܐ ܕ�ܡܐ ܐ�ܠܐ ܡܬܛܦܝܣ ܠܫܘܬܦܘܬܗܘܢ ܘܗܟܢܐ ܡܢ ܬܪܥܣ̈ܪ‬ ‫‘ ܐܣܩܘܒܝܛܘ̈ܪܣ ܐܣܦ̈ܪܣܬܝܢܐ ܗܘܐ ܗܘܐ ܥܕ�ܡܐ ܕܢܦܠ ܒܝܢܬܗܘܢ ܘܐܫܬܬܩ‬and he sent clerics and excubitors commanding them to have him beaten until blood spilt forth unless he be convinced to join them. Thus, he was beaten by twelve excubitors until he fell among them and was silent’ (15.27–16.4). In his Lexicon, Brockelmann (LS2 37), followed by Sokoloff (SL 1057), cites the first occurrence of ‫ ܐܣܦܪܣܬܝܢܐ‬ʾsprstynʾ and gives an active meaning. He does not, however, explain the incongruence of the passive Greek form used with Syriac ʿbd, which usually occurs with the active voice. In addition, he neglects to mention that the same form occurs several lines later with a passive meaning and with Syriac hwy. Given the context, it is clear that the second occurrence of ‫ ܐܣܦܪܣܬܝܢܐ‬ʾsprstynʾ in 16.3 has a passive meaning, which conforms to the general pattern of passive infinitive with hwy. This is the example cited above. As for the first occurrence of ‫ ܐܣܦܪܣܬܝܢܐ‬ʾsprstynʾ, the sense seems to

The Morphosyntactic Integration of Greek Loanwords in Syriac

115

Brock (1996: 257–58 n. 25; 2004: 38) has pointed to the late fifth- to early sixth-century translations of the Didascalia (ed. Vööbus 1979) and Athanasius’s Life of Antony (ed. Draguet 1980) as the earliest texts attesting the light verb strategy in Syriac. The example of ‫ ܐܟܣܘܪܣܬܝܢܐ‬ʾkswrstynʾ in example (6.19a) would represent an earlier, if not the earliest case of light verb strategy in Syriac, if the work is in fact by Rabbula of Edessa (d. 456/6). 44 Regardless, the light verb strategy is extremely rare in the fifth century and only becomes more frequent in the sixth and seventh centuries. 45 Light verb strategy is, however, never common in Syriac texts not translated from Greek, though it does occur more frequently in texts translated from Greek. 46 The development of the light verb accommodation strategy in Syriac has been linked to various external factors. Brock, for instance, has argued that the use of the light verb strategy in Syriac is due to contact with (non-Sahidic) Coptic, where a similar construction exists consisting of the native Coptic verb er ‘to do’ and a Greek infinitive. 47 The Coptic construction was illustrated in example (6.15). The Coptic construction is indeed structurally similar to the Syriac active voice construction, and so it potentially could have provided the model for this. 48 Coptic could not, however, have served as the model for the passive voice construction in Syriac, since no comparable construction exists in Coptic. Given that contact with Coptic cannot account for the entire Syriac construction (active and passive), it seems more likely that the use of the light verb accommodation strategy in Syriac is an internal development. This is especially the case since Syriac follows a well-established pattern of using a light verb meaning ‘to do’ with the active voice and a light verb meaning ‘to be(come)’ with the passive voice. 49 be ‘that they make him to be beaten’. For a similar pattern, compare Biblical Aramaic uḇʿo dɔniyye(ʾ)l wəḥaḇrohi ləhiṯqəṭɔlɔ ‘and they sought Daniel and his companions to be put to death’ (Dan 2:13) [and+seek-suf.3.m.pl pn and companion-m.pl.con+his to+to.be.killedinf]. In this case, then, ‫ ܐܣܦܪܣܬܝܢܐ‬ʾsprstynʾ would still be passive but used with Syriac ʿbd. It is unclear, however, if this should be interpreted as a light verb strategy or if Syriac ʿbd is a full finite verb, and ‫ ܐܣܦܪܣܬܝܢܐ‬ʾsprstynʾ is a simple loanverb or a code-switch. 44.  Ciancaglini (2006: 175; 2008: 50) claims that the light verb strategy is already attested in Ephrem (d. 373), citing πληροφορῆσαι (LSJ 1419) > ‫ ܦܠܪܘܦܘܪܝܣܐ‬plrwpwrysʾ with ‫ ܥܒܕ‬ʿbd ‘to inform’ in Benedictus 1732–46: 4.157.44 (correct the citation of line 43 in SL 1059). The commentary edited by Benedictus (1732–46: 4.116–93), however, is not genuine Ephrem but the Scholia of Yaʿqub of Edessa (the title of the work even mentions Yaʿqub of Edessa!). 45.  Brock 1996: 257–58; 2004: 37. 46.  For examples, see Ciancaglini 2008: 49; SL 334; 1056–60. 47.  Brock 1975: 88; see also 2004: 37 n. 13. Citing Brock, Van Rompay similarly notes that the use of the light verb strategy in Syriac “parallels, and may be derived from, a similar structure in all Coptic dialects except Sahidic” (in GEDSH, 106). 48.  For criticisms, however, see Ciancaglini 2006; 2008: 50. 49.  Wohlgemuth 2009: 109, 253.

116

Chapter 6

Ciancaglini has also argued that language contact played a role in the introduction of the light verb strategy in Syriac, but in her case it is contact with Iranian, not with Coptic. 50 Her proposal is even more unlikely, however, since most if not all of the examples of the light verb strategy in Syriac are found in authors (and translators) who wrote within the Roman Empire, such as Yuḥanon of Ephesus and Yaʿqub of Edessa. If contact with Iranian had played a role in the development of the light verb strategy in Syriac, then one would expect the strategy to feature in texts from Iranian-speaking areas. This is not the case, however. In response to arguments that external factors played a role in the development of the light verb strategy in Syriac, I should also stress that the light verb strategy is common cross-linguistically—the second most common in fact— occurring in over 104 languages in Wohlgemuth’s cross-linguistic sample. Thus, even though two languages with which Syriac was in contact have light verb strategies ([non-Sahidic] Coptic and Iranian) and even though cases of the transfer of accommodation strategies are attested cross-linguistically, 51 there is not sufficient evidence to suggest that external factors played a role in the development of the light verb strategy in Syriac. Rather, it seems to have been an internal Syriac development. Finally, it should be noted that in Syriac the verbal root ‫ ܥܒܕ‬ʿbd is found in several constructions that are superficially similar to the light verb strategy, as in the following example: (6.20) Acts of Thomas (ca. 200–250; ed. Wright 1871a) ‫ܿܗܘ ܕܥܕ�ܡܐ ܠܣܘܦܐ ܐܓܘܢܐ ܥܒܕ ܥܡ ܟܝܢܟ‬ haw

daʿḏammɔ

lsawpɔ

ʾgwnʾ

that-m.sg

nml+until

to+end-m.sg.emp

struggle-m.sg.emp

ʿḇaḏ

ʿam

kyɔnɔḵ

do+suf.3.m.sg

with

nature-m.sg.con+your-m.sg

‘who struggled against your nature until the end’ (200.1–2) In this example, a conjugated form of ‫ ܥܒܕ‬ʿbd is used with the noun ‫ܐܓܘܢܐ‬ ʾgwnʾ < ἀγών (PGL 25; LSJ 18–19). Cases such as this are not to be analyzed as the light verb accommodation strategy, however, since the Greek source ἀγών is a noun. 52 A similar case is found with ἄθλησις (PGL 46; LSJ 32) > ‫ ܐܬܠܝܣܐ‬ʾtlysʾ with ‫ ܥܒܕ‬ʿbd ‘fight, struggle’ (LS2 55; SL 112). 53 50.  Ciancaglini 2006; 2008: 48–52. 51.  Wichmann and Wohlgemuth 2008: 105–6, 108 with reference to Bakker 1997. 52. See Wichmann and Wohlgemuth 2008: 91. This contrasts with Ciancaglini (2006; 2008: 48–52), who argues that these are the same phenomenon. 53. Alternatively, Sokoloff (SL 112) gives the input form as the aorist infinitive ἀθλῆσαι. Brock’s proposed input of ἄθλησις seems more likely, however, since the word is more often used as the substantive ‘fight, struggle’.

The Morphosyntactic Integration of Greek Loanwords in Syriac

117

In general, the number of Greek loanverbs in Syriac is relatively small. While there are more than 800 Greek nouns and even 15 Greek particles in Syriac, 54 there are only a limited number of Greek loanverbs in Syriac (setting aside the denominative formations). In addition, these Greek loanverbs in Syriac use the light verb accommodation strategy and thus are not accommodated to Syriac root-and-pattern morphology. The relatively small number of Greek loanverbs in Syriac is likely due to the complex morphological structure of the Syriac verb. 55 6.4. Particles Approximately 15 Greek particles are attested in Syriac texts not translated from Greek before Yaʿqub of Edessa (d. 708). 56 Several Greek particles are already found in pre-fourth-century Syriac: 57 (6.21) a. μέν (LSJ 1101–2) > ‫ ܿܡܢ‬mn ‘indeed’ (pre-4th cent. Odes Sol. 18:7 [ed. Charlesworth 1973; for this interpretation, see Butts 2013]; 4th cent. Ephrem, Prose Refutations, 33.21–27 [ed. Overbeck 1865]; though not common until the sixth and seventh centuries; also in NT; LS2 393; SL 778) b. εἰκῇ (LSJ 484) > ‫ ܐܝܩܐ‬ʾyqʾ ‘in vain’ (pre-4th cent. Acts Thom. 220:10 [ed. Wright 1871a]; 4th cent. Aphrahaṭ, Demonstrations, 1.568.8, 9 [ed. Parisot 1894–1907]; Book of Steps, 288.20; 508.8 [ed. Kmosko 1926]; Ephrem, Prose Refutations, 44.4; 53.24 [ed. Overbeck 1865], Maḏrɔše against Julian the Apostate, 87.28 [ed. Beck 1957b], Maḏrɔše on Nisibis, 53.1; 122.7; 124.10 [ed. Beck 1963]; also in OT and NT; occurs throughout Classical Syriac; LS2 16; SL 37–38; see also Brock 1967: 398; 1975: 89; 1996: 259) c. τάχα (LSJ 1762) > ‫ ܛܟ‬ṭk ‘perhaps’ (pre-4th cent. Exod 32:30; Num 23:3; 4th cent. Aphrahaṭ, Demonstrations, 1.632.9; 1.696.14; 1.753.20; 2.133.18 [ed. Parisot 1894–1907], Ephrem, Prose Refutations, 34.6 [ed. Overbeck 1865], 2.24.46 [ed. Mitchell 1912–21]; Memrɔ on Our Lord, 31.9 [ed. Beck 1966]; Maḏrɔše on Nisibis, 22.9 [ed. Beck 1961a], 90.9, 15 [ed. Beck 1963]; Maḏrɔše against Heresies, 9.4; 44.3; 142.25 [ed. Beck 1957a]; Letter to Publius, 285.14; 293.18 [ed. Brock 1976]; remains common throughout Classical Syriac; LS2 274; SL 528; see also Brock 1967: 421; 1975: 89; 1996: 260) 54.  Greek particles in Syriac are analyzed immediately below in §6.4. 55.  For structure playing a role in lexical transfer, see Winford 2003: 52. 56.  Greek particles are also found in Postbiblical Hebrew and in various dialects of Jewish Aramaic (Krauss 1898: §§113–15). 57.  Brock 1975: 89; Butts 2013.

118

Chapter 6

Several additional Greek particles are first attested in fourth-century Syriac: 58 (6.22) a. γοῦν (LSJ 358) > ‫ ܓܘܢ‬gwn in ‫ ܒܕܓܘܢ‬bdgwn ‘at any rate’ (4th cent. Ephrem, Maḏrɔše on Nisibis, 13.15; 16.8; 37.3; 25.8, 9; 37.3; 42.13; 44.12 [ed. Beck 1961a]; 4.23; 20.9; 31.8 [ed.  Beck 1963]; Maḏrɔše on the Church, 2.2; 56.20 [ed. Beck 1960]; Maḏrɔše on the Fast, 27.5; 35.17 (appendix) [ed. Beck 1964b]; Maḏrɔše on Pascha, 22.3; 28.15, 20; 41.11 [ed. Beck 1964a]; Memre on Faith, 17.13; 18.24; 37.22 [ed. Beck 1961b]; remains common throughout Classical Syriac; LS2 59; SL 118; see also Brock 1996: 258; 1999–2000: 440) b. κἄν (LSJ 873) > ‫ ܩܢ‬qn ‘and if’ (4th cent. Book of Steps, 317.25 [ed. Kmosko 1926]; extremely rare; LS2 673; SL 1379; see also Brock 1996: 259) 59 c. οὖν (LSJ 1271–72) > ‫ ܐܘܢ‬ʾwn ‘really’ (4th cent. Aphrahaṭ, Demonstrations, 328.6 [ed. Parisot 1894–1907]; Ephrem, Commentary on the Diatessaron, 52.7 (quotation); 62.21; 68.21; 70.7; 98.21 (quotation); 108.2; 116.18; 160.9; 170.12 [ed. Leloir 1963]; 5th cent. Julian Romance, 120.19 [ed. Hoffmann 1880b]; not common; see also Brock 1975: 89; 1996: 259) d. μᾶλλον (LSJ 1076) > ‫ ܡܠܘܢ‬mlwn, ‫ �ܡܐܠܠܘܢ‬mʾllwn, ‫ܡܠܠܘܢ‬ mllwn ‘rather, more’ (4th cent. Ephrem, Commentary on the Diatessaron, 30.19 [ed. Leloir 1963; 1990]; LS2 392; SL 766) After the fourth century, Greek particles (including adverbs) continue to be added to Syriac: 60 (6.23) a. ἀκριβῶς (LSJ 55) > ‫ ܐܩܪܝܒܘܣ‬ʾqrybws ‘exactly’ (6th cent. Eliya, Life of Yuḥanon of Tella, 91.2 [ed. Brooks 1907: 29–95]; very rare; LS2 45; SL 93; see also Brock 1975: 89; 1996: 259) b. ἁπλῶς (LSJ 191) > ‫ ܗܦܠܘܣ‬hplws, ‫ ܐܦܠܘܣ‬ʾplws ‘simply, merely; in vain’ (6th cent. Qiyore of Edessa, Cause of the Liturgical Feasts, 184.11; 185.11 [ed. Macomber 1974]; LS2 40, 181; SL 87, 352–53; see also Brock 1975: 89 with n. 56; 1996: 259) c. ἄρα (LSJ 232–33) > ‫ ܐܪܐ‬ʾrʾ ‘therefore, then’ (5th cent. Narsai, Memre on Biblical Themes, 20.551 [ed. Frishman 1992]; LS2 45; SL 9; see also Brock 1996: 259; 1999–2000: 440) 61 58.  Brock 1975: 89. 59.  For the date of composition of this text, see n. 42 in chap. 5 above (p. 77). 60.  Brock 1975: 89. Perhaps also ὤ (LSJ 2029), εὖ (LSJ 704) > ‫ ܐܝܘ‬ʾyw ‘oh, woe!’ (5th cent. Balai, Memre on Joseph, 26.7 [ed. Bedjan 1891]; Narsai, Memre, 1.138.3 [ed. Mingana 1905]; LS2 14; SL 32). ̈ 61.  The Narsai example reads: ‫ܐܠܗܝܐ‬ ‫ ܐܪܐ ܐܢܫܐ ܚܕܘ ܕܗܘܝܬܘܢ‬ʾrʾ (ʾ)nɔšɔ ḥḏaw dahwayton ʾalɔhɔye ‘Therefore, rejoice, people, for you have become divine beings’ [therefore

The Morphosyntactic Integration of Greek Loanwords in Syriac

119

d. ἆρα (LSJ 232–33) > ‫ ܐܪܐ‬ʾrʾ ‘(introducing a question)’ (5th cent. Narsai, Memre on Biblical Themes, 20.549 [ed. Frishman 1992]; LS2 45; SL 9; see also Brock 1996: 259; 1999–2000: 440) 62 e. εἶτα (LSJ 498) > ‫ ܐܝܛܐ‬ʾyṭʾ ‘then’ (6th cent. Yuḥanon of Ephesus, Lives of the Eastern Saints, 581.9; 605.5 [ed. Brooks 1923–25]; LS2 14; SL 33; see also Brock 1996: 258) f. μάλιστα (LSJ 1076) > ‫ ܡܠܝܣܛܐ‬mlysṭʾ ‘most especially, exceedingly’ (6th cent. Ishay, Cause of the Martyrs, 18.7, 39.10 [ed. Scher 1909]; LS2 393; SL 771; see also Brock 1975: 89; 1996: 259) g. πάντως (LSJ 1301) > ‫ ܦܢܛܘܣ‬pnṭws ‘certainly, absolutely’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 151.30 [ed. Brooks 1935]; LS2 579; SL 1204–5; see also Brock 1996: 259) h. πότε (LSJ 1454) > ‫ ܦܘܛـܐ‬pwṭʾ ‘ever’ (6th cent. Yuḥanon of Ephesus, Lives of the Eastern Saints, 434.5; 435.5, 9 [ed. Brooks 1923–25]; LS2 559; SL 1162) i. τέως (LSJ 1786) > ‫ ܛـܐܘܣ‬ṭʾws ‘awhile, at least’ (6th cent. Qiyore of Edessa, Cause of the Liturgical Feasts, 116.9; 122.16 [ed. Macomber 1974]; LS2 265; SL 507; see also Brock 1996: 260 [“very probably a seventh-century introduction”]) A majority of these particles entered Syriac at the height of Syriac-Greek contact in the sixth century. Morphologically, the Greek particles in Syriac do not require accommodation. Syntactically, many of them preserve features of their Greek source. The particle ‫ ܿܡܢ‬mn ‘indeed’ (< μέν [LSJ 1101–2]), for instance, occurs in second position in Syriac, just as its Greek source does. In addition to these Greek loanwords in Syriac, there are two frequently occurring Syriac particles that are connected with Greek: ‫ ܕܝܢ‬dyn ‘then, but’ (LS2 151; SL 296–97) and ‫ ܓܝܪ‬gyr ‘indeed’ (LS2 114; SL 230). These two particles function in the same was as Greek δέ ‘but’ (LSJ 371–72) and γάρ ‘for’ (LSJ 338–39), respectively, even to the point of occurring in second position. These Syriac particles, however, are not loanwords from Greek; rather, they represent the adaptation of native Semitic material on the model of Greek. Their development is discussed in detail in chap. 9.

rejoice-imp.m.pl nml+be-suf.2.m.pl divine-m.pl.emp]. For the use of ἄρα (LSJ 232–33) in first position in Koinē Greek, see Luke 11:48. ̈ ‫ ܐܪܐ‬ʾrʾ saḵle lmɔn kɔryɔ 62.  The Narsai example reads: ‫ܣܟ�ܠܐ ܠܡܢ ܟܪܝܐ ܠܟܘܢ ܥܠ ܪܘܡܪܡܢ‬ lḵon ʿal rumrɔman ‘Why, foolish ones, are you pained by our exaltation?’ [int foolish-m. pl.emp to+what to.grieve-part.f.sg.abs to+you-m.pl concerning exaltation-m.s.con+our].

120

Chapter 6

It has often been pointed out that the transfer of particles is rarer than the transfer of nouns, adjectives, and verbs cross-linguistically. 63 Thus, the transfer of these Greek particles into Syriac points to a high degree of contact between Syriac and Greek. Interestingly, three Greek particles were transferred into Syriac already in its earliest history, with another four being added by the fourth century. This suggests that there was significant contact between the two languages already at an early period. 64 In addition, a number of Greek particles were transferred into Syriac in the sixth century, pointing to an increase in contact at this period. 6.5.  Secondary Nominal Derivations Involving Greek Loanwords in Syriac 6.5.1. Overview The previous sections dealt with the morphosyntactic integration of Greek loanwords in Syriac (§§6.2–6.4). Integration is not the end of the story for Greek loanwords in Syriac, however. Integrated loanwords can undergo the same derivational processes as native Syriac words. As Winford notes, “Once incorporated, [loanwords] become fair game for both derivational and inflectional processes internal to the recipient language” (Winford 2003: 59). This section analyzes various types of secondary nominal derivations involving Greek loanwords in Syriac. These derivations are divided into two categories: (1) nominal derivations involving root-and-pattern morphology (§6.5.2); (2) nominal derivations involving suffixes (§6.5.3). 6.5.2.  Root-and-Pattern Morphology As discussed in §6.3, Syriac literature before Yaʿqub of Edessa (d. 708) contains more than 20 verbal roots that are ultimately of Greek origin. From these verbal roots, various nouns can be derived according to standard Syriac nominal patterns. 65 An active and passive participle can, for instance, be theoretically derived for all of the verbal roots—for example, ‫ ܡܙܕܘܓܐ‬mɛzdawwḡɔ ‘the married one’ [part.m.sg.emp] ← rt. ‫ ܙܘܓ‬zwg d ‘to yoke; to join’; dt ‘to be married’ (LS2 191; SL 369) ← ‫ ܙܘܓܐ‬zwgʾ ‘yoke, pair; chariot’ (LS2 90, 191; SL 180, 369–70) < ζεῦγος (LSJ 754), ζυγόν (LSJ 757). In addition, many roots attest a nomen agentis form, e.g, ‫ ܡܝܩܢܢܐ‬myaqqnɔnɔ ‘characteristic’ (LS2 307; SL 754) ← rt. ‫ ܝܩܢ‬yqn d ‘to delineate’ (LS2 307; SL 582) ← ‫ ܐܝܩܘܢܐ‬ʾyqwnʾ ‘image, representation’ (LS2 45, 307; SL 38, 569) < εἰκών (LSJ 485). These 63.  See, e.g., Muysken 1980; Poplack, Sankoff, and Miller 1988: 62–65; Winford 2003 2003: 51. 64.  This is discussed in more detail in §§10.2–10.3 below. 65.  For Syriac nominal patterns, see Nöldeke 1904: §§92–140; as well as Fox 2003, with comparative Semitic evidence.

The Morphosyntactic Integration of Greek Loanwords in Syriac

121

nomen agentis forms with the suffix ‑ɔn can occur with a number of additional suffixes, including -ɔyɔ, -ɔʾiṯ, -ɔyuṯɔ, and ‑uṯɔ. 66 Other than the participle and nomen agentis, the only nominal pattern that is widely attested with verbal roots that are ultimately of Greek origin is CuCCɔC- (< *CuCCāC-), which is the usual pattern for deriving abstract substantives from d-stem verbal roots and, by extension, quadriliteral roots in Syriac. 67 The following nouns are formed according to this pattern from verbal roots that are ultimately of Greek origin: 68 (6.24) a. ζεῦγος (LSJ 754), ζυγόν (LSJ 757) > ‫ ܙܘܓܐ‬zwgʾ ‘yoke, pair; chariot’ (LS2 90, 191; SL 180, 369–70) → rt. ‫ ܙܘܓ‬zwg d ‘to yoke; to join’; dt ‘to be married’ (LS2 191; SL 369) → ‫ ܙܘܘܓܐ‬zuwwɔḡɔ ‘marriage’ (LS2 371; SL 371; see also Brock 1967: 400 n. 21; 2004: 37) b. καλῶς (LSJ 871) > rt. ‫ ܩܠܣ‬qls d ‘to praise’ (LS2 669; SL 1373) → ‫ ܩܘܠܣܐ‬qullɔsɔ ‘praise, elegy’ (LS2 669; SL 1329) c. κατήγορος (LSJ 927) or κατηγορεῖν (LSJ 926–27) > rt. ‫ܩܛܪܓ‬ qṭrg ‘to accuse; to apply’ (LS2 657, 663; SL 1348, 1358–59) → ‫ ܩܘܛܪܓܐ‬quṭrɔḡɔ ‘accusation’ (LS2 663; SL 1327) d. κύβος (LSJ 1005) > ‫ ܩܘܦܣܐ‬qwpsʾ ‘cube; piece on a draft board; tessera, mosaic tile; mosaic work; hard stone, flint’ (LS2 685; SL 1340) → rt. ‫ ܩܦܣ‬qps dt ‘to be provided with mosaics’ (LS2 685; SL 1394–95) → ‫ ܩܘܦܣܐ‬quppɔsɔ ‘provision of mosaics’ (inscription 1.8, 2.9 [dated to 509, 595] [ed. Krebernik 1991]; for this interpretation, see Brock 2004: 37, against the editor) e. ναυαγός (LSJ 1161) > rt. ‫ ܢܘܓ‬nwg d ‘to wreck a ship’; dt ‘to suffer shipwreck’ (LS2 418; SL 895) → ‫ ܢܘܘܓܐ‬nuwwɔḡɔ ‘shipwreck’ (LS2 418; SL 896) f. παραγγέλλειν (LSJ 1306) > rt. ‫ ܦܪܓܠ‬prgl ‘to admonish, warn; to send a declaration, warning; to excite, urge on; to forbid, prohibit; to hold back, restrain; to impede, hinder’ (LS2 592; SL 1226–27) → ‫ ܦܘܪܓ�ܠܐ‬purgɔlɔ ‘order; precept; confinement; threats’ (LS2 592; SL 1169; see also Brock 2004: 37) g. παῤῥησία (LSJ 1344) > ‫ ܦܪܗܣܝܐ‬prhsyʾ (with alternative orthographies) ‘freedom of speech; permission; liberty; familiarity, openness’ (LS2 602; SL 1245–46) → rt. ‫ ܦܪܣܝ‬prsy ‘to lay bear, reveal, uncover; to put to shame, expose’ (LS2 601; SL 1245) → ‫ ܦܘܪܣܝܐ‬pursɔyɔ ‘revealing, laying bare; 66. See Butts 2014: 230–33. 67.  For this pattern, see Nöldeke 1904: §§117, 123. 68.  Brock 2004: 37.

122

Chapter 6 uncovering, shame; male genital area, pudenda’ (LS2 601; SL 1171) h. προνοῆσαι (LSJ 1490–91) > rt. ‫ ܦܪܢܣ‬prns ‘to divide, distribute; to provide for, supply; to manage, administer’ (LS2 599; SL 1243; see Butts forthcoming c) → ‫ܦܘܪܢܣܐ‬ purnɔsɔ ‘nourishment, food; help; divine providence; administration; diocese’ (LS2 599; SL 1170–1171; see also Brock 1996: 261) i. τάξις (LSJ 1756) > ‫ ܛܟܣܐ‬ṭksʾ, ‫ ܛܟܣܝܣ‬ṭksys ‘order; rank’ (LS2 90, 274–75; SL 181, 529) → rt. ‫ ܛܟܣ‬ṭks d ‘to order’, dt ‘to be set in order, arranged’ (LS2 275; SL 529) → ‫ ܛܘܟܣܐ‬ṭukkɔsɔ ‘arrangement, rule’ (LS2 275; SL 529) j. τέχνη (LSJ 1785) > ‫ ܛܟܢܐ‬ṭknʾ ‘guile’ (LS2 274; SL 528–29) → rt. ‫ ܛܟܢ‬ṭkn d ‘to bestow care upon’; dt ‘to give attention, be busy with; to devise, contrive; to beguile, deceive’ (LS2 274, 844; SL 28) → ‫ ܛܘܟܢܐ‬ṭukkɔnɔ ‘guile’ (LS2 274; SL 517) k. τήγανον (LSJ 1786) > ‫ ܛܓܢܐ‬ṭgnʾ (with alternative orthographies) ‘frying pan’ (LS2 268; SL 513) → rt. ‫ ܛܓܢ‬ṭgn d ‘to fry, torture’ (LS2 268; SL 512–13) → ‫ ܛܘܓܢܐ‬ṭuggɔnɔ ‘torment, torture’ (LS2 268; SL 515)

CuCCɔC- is the only nominal pattern besides the participle and nomen agentis that is widely attested with verbal roots that are ultimately of Greek origin. The nominal pattern C1C2ɔC3- is attested with two verbal roots that do not occur in the d-stem: 69 (6.25) a. πεῖσαι (LSJ 1353–54) > rt. ‫ ܦܝܣ‬pys c ‘to persuade, to convince; to demand, seek, beseech’; ct ‘to be persuaded; to obey’ (LS2 567; SL 1188) → ‫ ܦܝܣܐ‬pyɔsɔ ‘persuasion, conviction’ (LS2 567; SL 1188) b. μελέτη (LSJ 1096) or μελετᾶν (LSJ 1096) > rt. ‫ ܡܠܛ‬mlṭ g ‘give attention to; attend to’ (LS2 391; SL 768) → ‫ ܡܠܛܐ‬mlɔṭɔ ‘care, attention; zeal’ (LS2 3991; SL 768) The use of the pattern C1C2ɔC3- in these cases is to be explained by the fact that these triliteral roots are not associated with the d-stem and so could not be used with the pattern CuCCɔC-. In addition, note that other nouns can be derived from these nouns—for example, pyɔs- + ‑(a)t → ‫ ܦܝܣܬܐ‬pyɔsṯɔ ‘persuasion’ (LS2 567; SL 1189). Additional nominal patterns are only sparsely attested with verbal roots of ultimate Greek origin. The Berufsname pattern CaCCɔC- is attested, for 69.  For this nominal pattern in Syriac, see Nöldeke 1904: §109; Fox 1996: 185–86, 226–27, 235.

The Morphosyntactic Integration of Greek Loanwords in Syriac

123

instance, in ‫ ܦܪܢܣܐ‬parnɔsɔ ‘steward, administrator’ (LS2 599; SL 1243–44) ← rt. ‫ ܦܪܢܣ‬prns ‘to divide, distribute; to provide for, supply; to manage, administer’ (LS2 599; SL 1243; see Butts forthcoming c) < προνοῆσαι (LSJ 1490–91). 70 In general, then, derivations involving root-and-pattern morphology with Greek loanwords are restricted in two ways. First, root-and-pattern morphology is only attested with Greek loanwords that have an independent verbal root. This shows that the root plays an essential role with internal nominal derivations in Syriac, reflecting the broader Semitic situation more generally. 71 There are no examples in Syriac in which a Greek loanword that does not have an independent verbal root produces internal nominal derivations involving root-and-pattern morphology. That is, a noun **nummɔsɔ ‘legality’ is never derived from ‫ ܢܡܘܣܐ‬nmwsʾ ‘law’ (LS2 431; SL 921–22) < νόμος (LSJ 1180) according to the abstract pattern CuCCɔC-, since no independent verbal root **nms exists in the language. Second, even when an independent verbal root does exist, only a limited number of nominal derivations are possible. Participles, nomina agentis, and abstracts of the pattern CuCCɔC- can be derived for most roots, and the abstract pattern C1C2ɔC3- and the Berufsname pattern CaCCɔC- are also found in isolated cases. No other internal nominal patterns are attested, however, including common substantival patterns such as *C1aC2C3-, *C1iC2C3-, and *C1uC2C3- and common adjectival patterns such as C1aC2C2iC3- (< * C1aC2C2īC3-). This suggests that internal nominal derivations were not fully productive in creating new nominal derivations synchronically in Syriac. Some patterns could indeed be used, but many patterns were simply not productive. Derivations involving root-and-pattern morphology with Greek loanwords, then, are restricted in that: (1) they can only occur if there is an independent verbal root; (2) they are only found with a limited set of nominal patterns, including participles, nomina agentis, and abstracts of the pattern CuCCɔC and are not attested with most of the internal nominal patterns in Syriac. 72 In both of these ways, nominal derivations involving root-and-pattern morphology differ from derivations with Syriac suffixes, which is the subject of the next section (§6.5.3). 70.  For the Berufsname pattern CaCCɔC- in Syriac, see Fox 2003: 260–61; Nöldeke 1904: §115. 71.  For the broader Semitic context, see Fox 2003: 44–45. 72.  It is likely that both of these restrictions are not limited to Greek loanwords in Syriac but apply more broadly to all lexemes in the language. That is, even for native Syriac lexemes, it is unlikely that nouns can be derived via root-and-pattern morphology unless a verbal root exists. In addition, it is likely that, synchronically, Syriac only has a limited number of productive internal nominal patterns. That is, a Syriac speaker could not freely form a *C1aC2C3- noun from any verbal root in the language but, rather, speakers learned a set of lexemes that were *C1aC2C3- nouns.

124

Chapter 6

6.5.3. Suffixes In addition to the internal derivations discussed in the previous section (§6.5.2), Greek loanwords in Syriac can undergo derivation with Syriac suffixes. The noun ‫ ܛܪܘܢܘܬܐ‬ṭrwnwtʾ ‘tyranny’ (LS2 289; SL 1184), for instance, is derived from ‫ ܛܪܘܢܐ‬ṭrwnʾ ‘tyrant’ (< τύραννος [LSJ 1836]) with the addition of the suffix ‑uṯɔ, which forms abstract substantives. The use of the abstract suffix -uṯɔ in ‫ ܛܪܘܢܘܬܐ‬ṭrwnwtʾ does not differ from its use with native Syriac words—for example, malkɔ ‘king’ (LS2 391; SL 772) + -uṯɔ → malkuṯɔ ‘kingdom’ (LS2 392; SL 772–73). Since an exhaustive study of the use of Syriac derivational suffixes with Greek loanwords can be found elsewhere, 73 it is not necessary to rehearse the details here. It is useful, however, to summarize some of the conclusions that can be drawn from the use of Syriac derivational suffixes with Greek loanwords. To begin, suffixes can be used with any incorporated Greek loanword, whether or not it has an independent verbal root. Thus, ‫ ܢܡܘܣܐܝܬ‬nmwsʾyt ‘according to the law’ can be derived from ‫ ܢܡܘܣܐ‬nmwsʾ ‘law’ (LS2 431; SL 921– 22) < νόμος (LSJ 1180), despite the fact that an independent verbal root **nms does not exist in the language. In addition, the set of Syriac suffixes used with Greek loanwords is the same as that used with non-Greek words in Syriac. This indicates that, in contrast to secondary nominal derivations involving root-andpattern morphology, secondary nominal derivations involving Syriac suffixes are fully productive in Syriac with Greek loanwords. The Syriac suffixes used with Greek loanwords include both simple suffixes and complex suffixes. The simple suffixes are the abstract suffix -uṯɔ, the adverbial suffix -ɔʾiṯ, the adjectival suffix -ɔyɔ (so-called nisba), and the nomen agentis suffix ‑ɔnɔ. The abstract suffix ‑uṯɔ is already attested with several Greek loanwords in pre-fourth-century Syriac. 74 The fifth and sixth centuries saw a large expansion in the use of the abstract suffix -uṯɔ with Greek loanwords, with a number of forms also being introduced in the fourth century. Greek loanwords with the adverbial suffix -ɔʾiṯ are rare in the earliest period of Syriac. This suffix came to be used more frequently with Greek loanwords over time, however, reaching an apex in the seventh century. This trajectory fits well with a more general increase in the frequency of the adverbial suffix -ɔʾiṯ throughout the history of Syriac. 75 73. See Butts 2014, building on the foundational work in Brock 1996: 260–61; 1999– 2000: 440–42; 2004. 74.  It is interesting to note that two of these types occur already in the Old Syriac documents but are very rare in later Syriac: ἄρχων (LSJ 254) > ‫ ܐܪܟܘܢܐ‬ʾrkwnʾ ‘ruler, archon; leader, chief’ (LS2 49; SL 100) + -uṯɔ → ‫ ܐܪܟܘܢܘܬܐ‬ʾrkwnwtʾ ‘rulership’ (P. Euphrates 20 3.5 [ed. Drijvers and Healey 1999: 231–48]) and στρατηγός (LSJ 1652) > ‫ ܣܛܪܛܓܐ‬sṭrṭgʾ strategos’ (LS2 34, 469; SL 71, 998) + -uṯɔ → ‫ ܣܛܪܛܓܘܬܐ‬sṭrṭgwtʾ ‘strategos-ship’ (P. Dura 28 1.5). 75. See Nöldeke 1875: 200 n. 3.

The Morphosyntactic Integration of Greek Loanwords in Syriac

125

The so-called nisba suffix -ɔyɔ (< *­-āy-) occurs only rarely with Greek loanwords before the sixth century. 76 There are, for instance, no examples in Ephrem, 77 and the only type attested in Narsai is ‫ ܢܡܘܣܝܐ‬nmwsyʾ ‘legal’ (Narsai, Memre, 1.32.2; 2.74.8, 2.311.17, passim [ed. Mingana 1905]; LS2 431; SL 922; Brock 1996: 260 n. 33; 1999–2000: 442 [no less than 27 times in Narsai]; 2004: 32 [pre-6th cent.]; 2010: 13–14) ← ‫ ܢܡܘܣܐ‬nmwsʾ ‘law’ (LS2 431; SL 921–22) + -ɔyɔ < νόμος (LSJ 1180). 78 After the fifth century, Greek loanwords with the adjectival suffix -ɔyɔ become much more common. The large number of examples of Greek loanwords with the nisba suffix -ɔyɔ from the sixth and seventh centuries (at least 19 types) shows that this suffix became more common after the fifth century. 79 There is a small group of Greek loanwords in Syriac that are only attested with the adjectival suffix ‑ɔyɔ: (6.26) a. Latin praetor (OLD 1448; LD 1436) > πραίτωρ (LLGE 92; PGL 1126; LSJ 1458) + -ɔyɔ > ‫ ܦܪܛܘܪܝܐ‬prṭwryʾ ‘praetor’ (6th cent. Yuḥanon of Ephesus, Ecclesiastical History, part 3, 309.28 [ed. Brooks 1935]; Lives of the Eastern Saints, 159.3 [ed. Brooks 1923–25]; LS2 596; SL 1237) b. βάρβαρος (LSJ 306) + -ɔyɔ > ‫ ܒܪܒܪܝܐ‬brbryʾ ‘barbarian’ (4th cent. Ephrem, Maḏrɔše on the Church, 89.16 [ed. Beck 1960]; already in Acts 28:2, 4, 6; Rom 1:14; 1 Cor 14:11; Col 3:11; LS2 95; SL 186) c. βυρσεύς (LSJ 333) + -ɔyɔ > ‫ ܒܘܪܣܝܐ‬bwrsyʾ ‘tanner’ (6th cent. Philoxenos, Discourses, 142.9 [ed. Budge 1894]; already in Acts 9:43; 10:6, 32; LS2 97; SL 131; see also Brock 1996: 260 n. 33; 2004: 32 [pre-6th cent.], 34) d. γερδιός, γέρδιος (LSJ 345), see also Latin gerdius (OLD 761; LD 811) + -ɔyɔ > ‫ ܓܪܕܝܝܐ‬grdyyʾ ‘weaver’ (pre-4th cent. 1 Sam 17:7; 1 Chr 11:23; 20:5; also Jdt 16:14; LS2 132; SL 258) e. λῃστής (LSJ 1046) + -ɔyɔ > ‫ ܠܣܛܝܐ‬lsṭyʾ ‘bandit’ (4th cent. Aphrahaṭ, Demonstrations, 1.337.25 [ed. Parisot 1894–1907]; Book of Steps, 165.6, 9 [ed. Kmosko 1926]; also in Matt 21:13 [CP]; 27:38 [P]; Mark 11:17 [P]; 15:27 [P]; Luke 10:30 [P]; 19:46 [P]; 22:52 [P]; LS2 368; SL 692–93; see also Brock 1967: 406–7; 2004: 32 [pre-6th cent.], 35) f. ξένος (LSJ 1189) + -ɔyɔ > ‫ ܐܟܣܢܝܐ‬ʾksnyʾ ‘strange, foreign; stranger’ (pre-4th cent. Acts Thom., 175:5, 7; 183:12; 231:3; 76.  Brock 2004: 32. 77.  Brock 1999-2000: 442. 78.  It should be noted that ‫ ܢܡܘܣܝܐ‬nmwsʾ is found already in early translations from Greek: Tob 2:13; Titus of Bostra, Against the Manichaeans, 176.28 (ed. de Lagarde 1859); Eusebius, Theophany, 3.3 (ed. Lee 1842 [cited by section numbers]). 79.  For further support of this argument, see §7.3 below.

126

Chapter 6 242:11 [ed. Wright 1871: 2.171–333]; Odes Sol. 17:6 [ed. Charlesworth 1973]; also in Isa 58:7; Job 31:32; Ps 69:9; Ruth 2:10; Matt 25:35 [SCP], 38 [SCP], 43 [CP], 44 [SCP]; 27:7 [SP]; passim; LS2 338; SL 44), already in Palmyrene Aramaic ʾksny (PAT 337–38; see also Brock 2005: 19); see also Jewish Babylonian Aramaic ʾaḵsənɔyɔ (DJBA 131); Jewish Palestinian Aramaic ʾksnyy (DJPA 58); Christian Palestinian Aramaic ʾksnʾy (DCPA 15; LSP 8)

Several of these words are found early in the history of Syriac. It should be stressed, however, that the use of Syriac suffixes with these particular words is not related to secondary nominal derivation but is due to the integration of the loanwords. Unlike the abstract suffix -uṯɔ, the adverbial suffix -ɔʾiṯ, and the adjectival suffix -ɔyɔ (so-called nisba), which have been discussed so far, the nomen agentis suffix -ɔnɔ is not productive with Greek loanwords in Syriac. Rather, it is basically limited to derived-stem nomina agentis from verbal roots that are Greek in origin 80—for example, ‫ ܡܛܟܣܢܐ‬mṭakksɔnɔ ‘someone who puts in order’ (LS2 275; SL 747; see also Brock 1996: 261; 2004: 36 [first attested in 6th cent.]) ← rt. ‫ ܛܟܣ‬ṭks d ‘to order’, dt ‘to be set in order, arranged’ (LS2 275; SL 529) ← ‫ ܛܟܣܐ‬ṭksʾ, ‫ ܛܟܣܝܣ‬ṭksys ‘order; rank’ (LS2 90, 275–76; SL 181, 529) < τάξις (LSJ 1756). This is a reflection of the limited use of this suffix in Syriac more broadly. There is a small group of Greek loanwords in Syriac that are only attested with the adjectival suffix -ɔnɔ: (6.27) a. κατήγορος (LSJ 927) > ‫ ܩܛܪܓܢܐ‬qṭrgnʾ, ‫ ܩܛܓܪܢܐ‬/qṭeḡrɔnɔ/ (with alternative orthographies) ‘accuser’ (LS2 657, 663; SL 1350, 1359) b. ταραχή (LSJ 1758) > ‫ ܛܪܟܢܐ‬ṭrknʾ ‘whisperer, tale-bearer; shrew, sagacious’ (LS2 289, 845; SL 553–54) The use of -ɔnɔ in these words can be compared with the Greek loanwords in Syriac that only occur with the adjectival suffix -ɔyɔ, which are listed in example (6.26). The complex suffixes in Syriac include the adjectival suffix -tɔnɔ, the adverbial suffix ‑tɔnɔʾiṯ, the abstract suffix -tɔnɔyuṯɔ, the abstract suffix -ɔyuṯɔ, the adjectival suffix ‑ɔnɔyɔ, the adverbial suffix -ɔnɔʾiṯ, and the abstract suffix -ɔnɔyuṯɔ. These suffixes are rare in the period of investigation of this study. Some notable forms do occur in early Syriac texts, however, such as the following two forms from the same root but involving different suffixes:

80.  Brock 2004: 36. For such roots, see §6.3.

The Morphosyntactic Integration of Greek Loanwords in Syriac

127

(6.28) a. σχῆμα (LSJ 1745) > ‫ ܐܣܟ�ܡܐ‬ʾskmʾ (with alternative orthographies) ‘form’ (LS2 35, 89; SL 74, 178) + -tɔnɔʾiṯ → ‫ ܐܣܟܡܬܢܐܝܬ‬ʾskmtnʾyt ‘cleverly; in pretense, feignedly’ (pre-4th cent. Book of the Laws of the Countries, 6.10 [ed. Drijvers 1965]; already in Pseudo-Clementine Recognitions, 90.11 [ed. de Lagarde 1861]; Titus of Bostra, Against the Manichaeans, 13.26 [ed. de Lagarde 1859]; LS2 35; SL 73; see also Brock 1996: 261; 1999–2000: 442; 2004: 35) b. σχῆμα (LSJ 1745) > ‫ ܐܣܟ�ܡܐ‬ʾskmʾ (with alternative orthographies) ‘form’ (LS2 35, 89; SL 74, 178) + -tɔnɔ → ‫ ܐܣܟܡܬܢܐ‬ʾskmtnʾ ‘modest; just; hypocritical’ (4th cent. Aphrahaṭ, Demonstrations, 1.604.25 [ed. Parisot 1894–1907]; already in Eusebius, Theophany, 1.37 [ed. Lee 1842 (cited by section numbers)]; LS2 35; SL 73; see also Brock 1996: 261; 1999–2000: 442; 2004: 35 [first attested in 4th cent.]) With the notable exception of these two words, most of the complex suffixes are not attested with Greek loanwords until the fifth century, and even then they remain rare during the period of Syriac with which this study is concerned (up to ca. 700). The use of a Syriac suffix with a Greek word provides an important clue for establishing a terminus ante quem for its incorporated into Syriac. 81 The occurrence of ‫ ܐܣܟܡܬܢܐܝܬ‬ʾskmtnʾyt ‘cleverly; in pretense, feignedly’ already in the second-century Book of the Laws of the Countries (6.10; ed. Drijvers 1965), for instance, shows that ‫ ܐܣܟ�ܡܐ‬ʾskmʾ ‘form’ was incorporated into Syriac by that time. This is important, since it is often difficult to determine where a given word falls on the continuum between Lehnwort and Fremdwort. 82 One might suppose, for instance, that a word such as ‫ ܐܘܩܝܢܘܣ‬ʾwqynws ‘ocean’ (LS2 9; SL 20) < ὠκεανός (LSJ 2031) is closer to the Fremdwort side of the continuum, since it does not follow the morphosyntactic rules of Syriac state. 83 This may or may not be true for its early occurrences in the Acts of Thomas (198:8; ed. Wright 1871: 2.171–333), the Book of the Laws of the Countries (50.12; ed. Drijvers 1965), 84 and Aphrahaṭ’s Demonstrations (1.660.8; ed. Parisot 1894–1907). By the seventh century, however, the word is certainly closer to the Lehnwort side of the continuum (even though it still does not follow the morphosyntactic rules of Syriac state), since it occurs with the Syriac derivational suffix -ɔʾiṯ as ‫ ܐܘܩܝܢܐܝܬ‬ʾwqynʾyt ‘like an ocean’ in Babai the Great’s Commentary on the ‘Gnostic Chapters’ by Evagrius of Pontus 81.  Correct terminus post quem in Butts 2014. 82.  See §4.5 above. 83.  See §6.2.6 above. 84.  This should be added to Schall 1960.

128

Chapter 6

(14.32; ed. Frankenberg 1912). Thus, the occurrence of a Greek loanword with a Syriac suffix can help to establish a terminus ante quem for a Greek word’s being on the Lehnwort side of the Lehnwort–Fremdwort continuum. Secondary nominal derivations involving suffixes are extremely rare in Aramaic outside Syriac. The only suffix attested with Greek loanwords in other Aramaic dialects is the abstract suffix *-ūt-. In Palmyrene Aramaic, for instance, the following words occur with the abstract suffix *-ūt-: 85 (6.29) a. ἐπιμελητής (LSJ 645–46) > ʾpmlṭ ‘curator’ (PAT 342) + *-ūt → ʾpmlṭw ‘curatorship’ (PAT 342; see also Brock 2005: 16) b. πρόεδρος (LSJ 1476) > *plhdr ‘president’ + *-ūt → plhdrw ‘presidency’ (PAT 400; see also Brock 2005: 20) c. στρατηγός (LSJ 1652) > ʾsṭrṭg ‘general’ (PAT 341) + *-ūt → ʾsṭrṭgw ‘term as general, command, campaign, expedition’ (PAT 341; see also Brock 2004: 32 n. 7; 2005: 16), also already in Achaemenid Aramaic ʾsṭrṭg[w] (DNWSI 87–88) The abstract suffix *-ūt is also used in Christian Palestinian Aramaic—for example, ἀθλητής (PGL 46; LSJ 32) > ʾtlyṭ (DCPA 35; LSP 20) + *-ūt → ʾtlyṭw (DCPA 35). These examples from Palmyrene Aramaic and Christian Palestinian Aramaic show that secondary nominal derivations involving the use of native Aramaic suffixes with Greek loanwords are not limited to Syriac but are also found in other dialects of Aramaic, albeit in a limited number.  86 It is interesting to note in this regard that Palmyrene Aramaic and Christian Palestinian Aramaic are, after Syriac, among the dialects of Aramaic that had the most intense contact with Greek. Thus, the use of secondary nominal derivations involving native suffixes with Greek loanwords seems to correlate with degrees of contact. In this book, I am focusing primarily on texts that were not translated from Greek. In the introduction (chap. 1), however, I pointed out that one of the two main loci for contact between Greek and Syriac was the Syriac translations of Greek texts. The use of Syriac suffixes with Greek loanwords helps to illustrate the role that Syriac translations from Greek must have played in Syriac-Greek language contact. In many cases, the first attestation of a Greek loanword with a Syriac suffix is in a text translated from Greek. 87 Translation literature thus facilitated the creation of new combinations of Syriac derivational endings with Greek loanwords. Over time, some of these 85.  Brock 2005: 25. The ending -w in the following examples is the status absolutus of the abstract suffix *-ūt. 86. Compare Brock, who states, “Syriac is the only Late Aramaic dialect which develops this potential” (Brock 2004: 32). 87.  This was noted already by Brock 1999–2000: 441. See also, with further detail, Butts 2014.

The Morphosyntactic Integration of Greek Loanwords in Syriac

129

neologisms came to be used in Syriac texts not translated from Greek, and a few even became quite common there. This corroborates the observation that Syriac translations from Greek facilitated changes to the Syriac language, as attested in native literature. 6.5.4. Summary In this section, I analyzed nominal derivations involving Greek loanwords in Syriac. These were divided into two categories: those involving root-andpattern morphology (internal derivation) and those involving suffixes (external derivation). These two categories have a number of differences. To begin, the only Greek loanwords that undergo nominal derivations involving root-andpattern morphology are those for which an independent verbal root is also attested in Syriac. That is, a noun **nummɔsɔ ‘legality’ cannot be derived from ‫ ܢܡܘܣܐ‬nmwsʾ ‘law’ (LS2 431; SL 921–22) < νόμος (LSJ 1180) according to the pattern CuCCɔC-, since no independent verbal root exists. This illustrates the essential role that roots play in internal nominal derivations in Syriac. This contrasts with nominal derivations involving suffixes in which no such restriction exists, as is illustrated by ‫ ܢܡܘܣܐܝܬ‬nmwsʾyt ‘according to the law’. A second difference between the two categories of nominal derivation relates to the issue of productivity. Roots of Greek origin are fully productive in Syriac as verbal forms. In addition, participles, nomina agentis, and abstracts of the pattern CuCCɔC- can be derived for most if not all roots. Beyond this, however, root-and-pattern morphology is severely restricted in creating new nouns from verbal roots of ultimate Greek origin. In contrast, nominal derivations from Greek loanwords involving suffixes do not seem to be limited. In fact, suffixes can be productively applied to Greek loanwords already in the earliest period of Syriac. Over time, the use of suffixes with Greek loanwords continues to increase as these suffixes become used more frequently and as Greek loanwords become more integrated into Syriac. 6.6.  Structural Consequences of Loanwords 6.6.1. Overview It is well known that the incorporation of loanwords can have structural consequences in the phonology and the morphology of the recipient language. 88 In English, for instance, there are a number of loanwords from Latin in which both the singular and plural were transferred: (6.30) a. alumnus ~ alumni b. fungus ~ fungi 88.  See, e.g., Bloomfield 1933: 452–54; King 2000: 46–47; Sapir 1921: 201–2; Smits 1996: 39; Van Coetsem 2000: 90–91; Weinreich 1953: 31; Winford 2003: 53–58; 2005: 386–88; Winter 1973: 145–46.

130

Chapter 6

Based on pairs such as these, English speakers developed a new plural suffix -i for singular nouns ending in -us. This plural suffix -i is found with Latin loanwords such as status and apparatus, for which the plural is occasionally cited as stati and apparati instead of the Latinate plurals statūs and apparatūs (both fifth declension, not second) or the now-common English plural statuses and apparatuses. The plural suffix -i also occurs with English nouns that are not of Latin origin, such as the Greek loanword octopus, where the plural octopi frequently appears instead of the Greek plural octopodes. The plurals stati, apparati, and octopi are the result of analogy within English: (6.31) alumnus : alumni :: fungus : fungi :: status : X = stati  :: apparatus : X = apparati :: octopus : X = octopi The plural ending -i does not, then, represent the transfer of a morpheme from Latin to English but instead is the result of analogy in English. This process is no different from analogy involving native words. Thus, the plural ending ‑i in English is contact-induced only in the sense that the words on which the analogy is based are the result of language contact; the ending does not represent the direct transfer of a morpheme from Latin to English, however. Given the substantial number of Greek loanwords in Syriac, it is not surprising that these words served as the basis for secondary analogical developments in Syriac. In this section, I discuss two cases such as this: (1) the development of the Syriac plural suffixes -(w)s and -(ʾ)s (§6.6.2); and (2) the development of the Syriac Berufsname suffix -ɔrɔ (§6.6.3). 6.6.2.  The Syriac Plural Suffixes -(w)s and -(ʾ)s The first instance to be discussed in which Greek loanwords provide the basis for analogical developments in Syriac has already been introduced in the analysis of Greek plural morphology with Greek loanwords. 89 At times, the Greek plural serves as an input form alongside the singular. The accusative plural κληρικόυς, for instance, was transferred into Syriac as ‫ ܩܠܝ̈ܪܝܩܘܣ‬qlyr̈yqws, along with the nominative singular κληρικός (PGL 756) > ‫ ܩܠܝܪܝܩܐ‬qlyryqʾ ‘cleric’ (LS2 671; SL 1371). In this case, then, the suffix -ws functions as a plural ending. This plural ending -­ ws also occurs with Greek loanwords that do ̈ not have a corresponding Greek plural in -ους—for example, ‫ܐܢܢܩܘܣ‬ ʾnnq̈ws, which is one of the plural forms of ‫ ܐܢܢܩܐ‬ʾnnqʾ (with alternative orthographies) ‘necessity’ (LS2 29; SL 63) < ἀνάγκη (LSJ 101). As already noted, this use of -ws as a plural marker is the result of analogy: ̈ (6.32) ‫ ܩܠܝܪܝܩܐ‬qlyryqʾ : ‫ ܩܠܝ̈ܪܝܩܘܣ‬qlyr̈yqws :: ‫ ܐܢܢܩܐ‬ʾnnqʾ : X = ‫ܐܢܢܩܘܣ‬ ʾnnq̈ws

89.  See §6.2.5 above.

The Morphosyntactic Integration of Greek Loanwords in Syriac

131

The plural suffix -ws is found primarily with Greek loanwords in Syriac, as ̈ in the examples of ‫ ܩܠܝ̈ܪܝܩܘܣ‬qlyr̈yqws and ‫ܐܢܢܩܘܣ‬ ʾnnq̈ws. It does, however, also occur with at least one native Syriac word. One of the attested plurals of Syriac ‫ ܩܪܝܬܐ‬qriṯɔ ‘village, town’ is ‫ ܩܘ̈ܪܝܘܣ‬qwr̈yws (5th cent. Life of Rabbula, 161.6 [ed. Overbeck 1865: 157–248]). The form ‫ ܩܘ̈ܪܝܘܣ‬qwr̈yws derives from the native Syriac plural ‫ ܩܘ̈ܪܝܐ‬qur̈yɔ with the analogically created plural suffix ‑ws. The basis for this analogy is difficult to determine, but it may stem from the fact that Syriac ‫ ܩܪܝܬܐ‬qriṯɔ has an irregular—that is, marked—native Syriac plural in ‫ ܩܘ̈ܪܝܐ‬qur̈yɔ. 90 Thus, the plural suffix -ws is not a direct transfer from Greek but represents an analogical development in Syriac based on Greek loanwords. This development led to the use of the plural suffix -ws (ultimately from Greek -ους) with many Greek loanwords, including those that do not have a plural in -ους, and with at least one native Syriac word. A similar development led to the creation of a plural suffix -(ʾ)s in Syriac. The Greek plural served as an input form with Greek first-declension nouns with nominative singular -η (or -α) ~ accusative plural -ας. The accusative ̈ plural ἀνάγκας, for instance, was transferred into Syriac as ‫ܐܢܢܩܣ‬ ʾnnq̈s, along with the nominative singular ἀνάγκη (LSJ 101) > ‫ ܐܢܢܩܐ‬ʾnnqʾ (with alternative orthographies) ‘necessity’ (LS2 29; SL 63). Based on a pair such as this, Syriac speakers analogically created a new plural suffix -(ʾ)s. This analogically created plural suffix -(ʾ)s is occasionally attested with native Syriac words. Alongside ‫ ܩܘ̈ܪܝܘܣ‬qwr̈yws, which was discussed above, one of the attested plurals of Syriac ‫ ܩܪܝܬܐ‬qriṯɔ ‘village, town’ is ‫ ܩܘ̈ܪܝܐܣ‬qwryʾs (7th cent. Apocalypse of Pseudo-Methodius, 8.12 [ed. Reinink 1993]). The form ‫ ܩܘ̈ܪܝܐܣ‬qwryʾs derives from the native Syriac plural ‫ ܩܘ̈ܪܝܐ‬qur̈yɔ with the analogically created plural suffix -(ʾ)s. Though outside the time period that is of immediate interest to ̈ this book, the plural suffix -(ʾ)s also occurs in ‫ܓܢܣ‬ gn̈s (8th cent. Chronicle of Zuqnin, 1.131.14 [ed. Chabot 1927–49]), which is one of the plurals of the native Syriac word ‫ ܓܢܬܐ‬gannṯɔ ‘garden’ (LS2 122; SL 250). 91 While the analogical basis for the extension of the plural -(ʾ)s to ‫ ܩܘ̈ܪܝܐܣ‬qwryʾs could again be explained by the existence of an irregular plural in Syriac, it remains 90.  This is one of the few words in Syriac reflecting the so-called broken plurals that are common in Arabic, Ethiopian Semitic (especially Gəʿəz), Old South Arabian, and Modern South Arabian and that are probably to be reconstructed to Proto-Semitic. For an analysis of the broken plurals in the Semitic languages, see Ratcliffe 1998a; 1998b. For their reconstruction to Proto-Semitic, see, inter alia, Goldenberg 1977: 473–75 (= 1998: 298–300); Greenberg 1955; Hetzron 1976: 102; Huehnergard 1987; 2005: 159; Huehnergard and Rubin 2011: 272–73. 91.  An alternative spelling of ‫ ܓܐܢܐܣ‬gʾnʾs (without syɔme) occurs in a section from the history of Dionysios of Tel Maḥre (d. 845), which is quoted in the 12th-century Chronicle by Michael Rabo (Chabot 1899–1910: 4.448c.12). It should be noted that the irregular orthography of the first syllable of ‫ ܓܐܢܐܣ‬gʾnʾs is recorded neither in Nöldeke 1904: §89 nor in SL2 122.

132

Chapter 6

much less clear what the analogical basis for the extension of the plural -(ʾ)s ̈ gn̈s could have been. to ‫ܓܢܣ‬ To summarize, the plural suffixes -(w)s and -(ʾ)s derive ultimately from the Greek accusative plural endings -ους and -ας, respectively. These are not direct transfers from Greek, however. Rather, they represent analogical developments based on a number of Greek loanwords in Syriac that appear both in a singular and plural form. Both plural suffixes occur commonly with Greek loanwords of various types, but they are extremely rare with native Syriac words, being restricted to only a handful of examples. 6.6.3.  The Syriac Berufsname Suffix -ɔrɔ The second instance to be discussed in which Greek loanwords provide the basis for analogical developments in Syriac involves the Syriac Berufsname suffix ‑ɔrɔ. The most common nominal formation for Berufsnamen in Syriac is C1aC2C2ɔC3- 92—for example, ‫ ܓܢܒܐ‬gannɔḇɔ ‘thief’ (LS2 124; SL 244) ← gnb ‘to steal’ (LS2 124; SL 243–44). In addition to this nominal pattern, Berufs­namen are also occasionally formed with the suffix -ɔrɔ in Syriac: 93 (6.33) a. ‫ ܐܠܦܪܐ‬ʾɛllp̄ɔrɔ ‘sailor’ (LS2 22; SL 51) 94← ‫ ܐܠܦܐ‬ʾɛllp̄ɔ ‘boat’ (LS2 22, 921; SL 50–51) < Akkadian elippu (CAD E 90–95; see also Kaufman 1974: 48) b. ‫ ܐܣܛܘܢܪܐ‬ʾɛsṭonɔrɔ ‘stylite’ (LS2 33; SL 69) ← ‫ ܐܣܛܘܢܐ‬ʾɛsṭonɔ ‘pillar’ (LS2 33; SL 68) < Iranian (Ciancaglini 2008: 7, 110); see, for instance, Pahlavi stun(ag) (CPD 78) c. ‫ ܐܣܛܣܝܪܐ‬ʾɛsṭasyɔrɔ ‘quarrelsome, factious’ (LS2 33; SL 70) ← ‫ ܐܣܛܣܝܣ‬ʾsṭsys ‘uproar, disturbance’ (LS2 33; SL 69–70) < στάσις (LSJ 1634) d. ‫ ܒܙܝܩܪܐ‬bɔziqɔrɔ ‘falconer’ (LS2 65; SL 133) ← ‫ ܒܙܝܩܐ‬bɔziqɔ ‘falcon’ (LS2 65; SL 133) < Iranian (Ciancaglini 2008: 125); see, for instance, Pahlavi bāz (CPD 18) 95 The Berufsname suffix -ɔrɔ in these words ultimately reflects later Greek -άριος, which itself is from Latin -arius. 96 The suffix -ɔrɔ was not, however, a 92.  Fox 2003: 260–61; Nöldeke 1904: §115. 93.  Brockelmann 1908: §223b1; Ciancaglini 2008: 7; Nöldeke 1904: §140. Though these words are loanwords, it is improbable that Syriac speakers would have analyzed all of them as such, especially ‫ ܐܦ�ܠܐ‬ʾɛllp̄ɔ ‘boat’. 94. This Syriac noun is probably the source of Christian Palestinian Aramaic ʾlprʾ ‘sailor’ (DCPA 18–19; so also Müller-Kessler 1991: §4.2.1.10.2). 95.  Compare, however, Jewish Babylonian Aramaic bɔzyɔr ‘falconer’ (DJBA 182–83), which derives not from an Iranian noun with the Greek suffix -άριος but from an Iranian noun with an Iranian suffix—e.g., Modern Persian bāzyār (Steingass 1892: 146). 96. For the relationship between the Latin and Greek suffixes, see Mason 1974: 3; Palmer 1945: 48–49.

The Morphosyntactic Integration of Greek Loanwords in Syriac

133

direct morphological transfer from Greek into Syriac. Rather, the development of the suffix -ɔrɔ in Syriac is due to inner-Syriac analogy. In Syriac, there are a number of Greek loanwords that contain the -άριος suffix, as is illustrated in the following examples: (6.34) a. ἀποκρισιάριος (LSJ 204) > ‫ ܐܦܘܩܪܝܣܪܐ‬ʾpwqrysrʾ, ‫ܐܦܩܪܝܣܪܐ‬ ʾpqrysrʾ ‘legate’ (LS2 43; SL 89) b. Latin veredarius (OLD 2035; LD 1973) > βερεδάριος, βερηδάριος, οὐερεδάριος (GLBRP 306; LLGE 34, 79) > ‫ܒܝܠܕܪܐ‬ byldrʾ (with alternative orthographies) ‘letter carrier’ (LS2 75; SL 141) c. Latin vestiarius (OLD 2048; LD 1981) > βεστιάριος (GLBRP 306; LLGE 34; see also Mason 1974: 12) > ‫ ܒܣܛܝܪܐ‬bsṭyrʾ ‘person in charge of wardrobe’ (LS2 80; SL 163) d. Latin galearius (LD 800) > γα(λ)λιάριος (LSJ 337; LLGE 38) > ‫ ܓܠܝܪܐ‬glyrʾ ‘galearius, military servant’ (LS2 118; SL 237–38) e. δρομωνάριος (PGL 388) > ‫ ܕܪܘܡܢܪܐ‬drwmnrʾ ‘sailor (LS2 167; SL 324) f. Latin cancellarius (LD 276) > καγκελλάριος (LSJ 848; LSJ Sup. 162; GLBRP 610; LLGE 48; PGL 681; see also Mason 1974: 4, 58) > ‫ ܩܢܩܠܪܐ‬qnqlrʾ ‘notary’ (LS2 679; SL 1386) g. Latin cubicularis (OLD 463; LD 486) > κουβικουλάριος (PGL 779) > ‫ ܩܘܒܩܠܪܐ‬qwbqlrʾ ‘chamberlain’ (LS2 644; SL 1309) h. Latin quaestionarius (OLD 1535; LD 1502) > κυαιστιωνάριος (GLBRP 694; LLGE 63) 97 > ‫ ܩܣܛܘܢܪܐ‬qsṭwnrʾ ‘torturer’ (LS2 679; SL 1387) i. Latin lecticarius (OLD 1012; LD 1045–46) > λεκτικάριος (GLBRP 709; LLGE 66; LSJ Sup. 194) > ‫ ܠܩܛܝܩܪܐ‬lqṭyqrʾ ‘priest who carry funeral biers’ (LS2 370; SL 697) j. Latin notarius (OLD 1192; LD 1217) > νοτάριος (GLBRP 786; LLGE 74–75; PGL 922–23; see also Mason 1974: 69–70) > ‫ ܢܘܛܪܐ‬nwṭrʾ (with alternative orthographies) ‘notarius, a Byzantine official’ (LS2 427; SL 898, 911) k. παραμονάριος (PGL 1022) > ‫ ܦܪܡܘܢܪܐ‬prmwnrʾ ‘verger, church keeper’ (LS2 598; SL 1242) l. σακκελάριος (PGL 1221) > ‫ ܣܩܠܪܐ‬sqlrʾ ‘treasurer’ (LS2 494; SL 1040) m. Latin silentarius (LD 1698) > σιλεντιάριος (LSJ 1598) > ‫ ܣܠܢܛܝܪܐ‬slnṭyrʾ ‘silentary’ (LS2 477; SL 1013) 97.  Brockelmann (LS2 679), followed by Sokoloff (SL 1387), gives the Greek intermediary as κουεστιονάριος, which is not, however, found in the Greek lexica.

134

Chapter 6 n. σπαθάριος (LSJ 1623) > ‫ ܐܣܦܬܪܐ‬ʾsptrʾ ‘guardsman’ (LS2 37; SL 78) o. σχολάριος (LSJ 1747) > ‫ ܐܣܟܠܪܐ‬ʾsklrʾ ‘palace guard’ (LS2 34; SL 74) p. Latin tabellarius (OLD 1897–98; LD 1831) > ταβελλάριος (GLBRP 1067; LLGE 109; LSJ 1752; see also Mason 1974: 4, 6, 90–91) > ‫ ܛܒܠܪܐ‬ṭblrʾ ‘keeper of records’ (LS2 266; SL 510–11) q. Latin tabularius (OLD 1899; LD 1832) > ταβουλάριος (GLBRP 1067; LLGE 110; PGL 1370) > ‫ ܛܒܘܠܪܐ‬ṭbwlrʾ ‘keeper of records’ (LS2 267; SL 509) r. Latin chartularius (LD 326) > χαρτου(λ)λάριος (LSJ 1980) > ‫ ܟܪܛܘܠܪܐ‬krṭwlrʾ ‘archivist’ (LS2 344; SL 650)

Given the relatively large number of Greek loanwords that have the suffix -άριος in Syriac and given their consistent semantics, Syriac speakers would have been able to deduce the meaning of the suffix -ɔrɔ < -άριος. In addition, several of these Greek loanwords in -άριος were also transferred in a form without the suffix: (6.35) a. δρόμων (LSJ 450) > ‫ ܕܪܡܘܢ‬drmwn ‘ship, boat’ (LS2 167; SL 324) ~ δρομωνάριος (PGL 388) > ‫ ܕܪܘܡܢܪܐ‬drwmnrʾ ‘sailor’ (LS2 167; SL 324) b. Latin lectica (OLD 1012; LD 1045) > λεκτίκιον (GLBRP 709; LLGE 66; LSJ 1037) > ‫ ܠܩܛܩܝܢ‬lqṭqyn, ‫ ܠܩܛܝܩܝܢ‬lqṭyqyn ‘small litter’ (LS2 370; SL 697) ~ Latin lecticarius (OLD 1012; LD 1045–46) > λεκτικάριος (GLBRP 709; LLGE 66; LSJ Sup. 194) > ‫ܠܩܛܝܩܪܐ‬ lqṭyqyrʾ ‘priest who carries funeral biers’ (LS2 370; SL 697) c. σχολή (LSJ 1747–48) > ‫ ܐܣܟܘ�ܠܐ‬ʾskwlʾ (with alternative orthographies) ‘lecture hall’ (LS2 34, 474; SL 73, 1008) ~ σχολάριος (LSJ 1747) > ‫ ܐܣܟܠܪܐ‬ʾsklrʾ ‘palace guard’ (LS2 34, 474; SL 74, 1008) Based on pairs such as these, Syriac speakers created a new Berufsname suffix ‑ɔrɔ through analogy: (6.36) ‫ ܐܣܟܘ�ܠܐ‬ʾskwlʾ ‘lecture hall’ : ‫ ܐܣܟܠܪܐ‬ʾsklrʾ ‘palace guard’ :: ‫ ܕܪܡܘܢ‬drmwn ‘ship, boat’ : ‫ ܕܪܘܡܢܪܐ‬drwmnrʾ ‘sailor’ :: ‫ ܐܠܦܐ‬ʾɛllp̄ɔ ‘boat’ : X = ‫ ܐܠܦܪܐ‬ʾɛllp̄ɔrɔ ‘sailor’ The Syriac Berufsname suffix ‑ɔrɔ does not, then, represent the direct transfer of Greek ‑άριος (or Latin -arius), but rather, it results from an analogical development in Syriac based on Greek loanwords. This new Berufsname suffix ‑ɔrɔ is only rarely attested with non-Greek words in Syriac, and all of these non-Greek words are nonnative Aramaic. In cases such as ‫ ܐܣܛܘܢܪܐ‬ʾɛsṭonɔrɔ

The Morphosyntactic Integration of Greek Loanwords in Syriac

135

‘stylite’, the motivation for the analogical extension of -ɔrɔ may have been that ‫ ܐܣܛܘܢܐ‬ʾɛsṭonɔ ‘pillar’ is obviously not Aramaic. In other cases, however, such as ‫ ܐܠܦܪܐ‬ʾɛllp̄ɔrɔ ‘sailor’, it seems less likely that Syriac speakers would have been cognizant of the foreign origin (in this case, Akkadian) of ‫ܐܠܦܐ‬ ʾɛllp̄ɔ ‘boat’. 6.6.4. Summary Prima facie, the Syriac Berufsname suffix ‑ɔrɔ could represent the direct transfer of the Greek derivational suffix -άριος (or Latin -arius). Similarly, the Syriac plural suffixes -(w)s and -(ʾ)s could represent the direct transfer of the Greek inflectional endings -ους and -ας, respectively. Upon closer examination, however, we find that a different explanation is more likely. These three suffixes in Syriac are not cases of the direct transfer of morphology from Greek to Syriac. Rather, they are instances in which Syriac speakers analogically created new morphological suffixes on the basis of Greek loanwords in their language. The analogical creation of the plural suffixes -(w)s and -(ʾ)s was made possible by the fact that different input forms exist for the same Greek loanword in Syriac. Similarly, the existence of Greek loanwords with the suffix ‑άριος as well as without it enabled the analogical creation of the Berufsname suffix ‑ɔrɔ. These changes, then, illustrate the ramifications of the influx of a large number of Greek loanwords into Syriac. While the changes discussed in this section do not represent the transfer of morphology from Greek into Syriac, they do show, in an extended way, the effects that contact with Greek had on Syriac. In these particular cases, this contact resulted in changes that reached all the way to the morphology of Syriac. It should be noted, however, that these new suffixes were never productive in Syriac; rather, they are attested with only a limited subset of words, most of which are not native to Syriac. 6.7. Conclusion The morphosyntactic integration of Greek loanwords in Syriac varies significantly by parts of speech. Particles require basically no integration. Verbs, in contrast, require more integration but only slightly more, since the majority of Greek loanverbs in Syriac are accommodated by a strategy termed light verb, in which the native Syriac verbal roots ‫ ܥܒܕ‬ʿbd ‘to do, make’ and ‫ܗܘܝ‬ hwy ‘to be(come)’ occur in conjunction with Greek infinitives. This strategy does not require the Greek verb to be integrated into Syriac root-and-pattern morphology, but instead the Greek verb is left basically unintegrated with a native Syriac light verb containing all of the inflectional information. Thus, for most Greek loanverbs in Syriac, morphosyntactic integration is minimal. In contrast to verbs and particles, Greek nouns underwent a more involved process of morphosyntactic integration in Syriac. In fact, Greek nouns, which are marked for one of five cases (vocative, nominative, genitive, dative, and

136

Chapter 6

accusative), one of three genders (masculine, feminine, and neuter), and one of two numbers (singular and plural), are often integrated as fully inflectional Syriac nouns, which are marked for one of two genders (masculine and feminine), one of two numbers (singular and plural), and one of three states (status absolutus, status emphaticus, and status constructus). This requires significant accommodation on the morphological level. It should be noted, however, that Greek nouns are not always—or even usually—accommodated to Syriac derivational structure. Thus, a loanword such as τόμος (LSJ 1804) > ‫ ܛܘܡܣܐ‬ṭwmsʾ ‘tome’ (LS2 280; SL 518), where ‫ ܛܘܡܣܐ‬ṭwmsʾ could reflect a native Syriac nominal derivation of the pattern C1uC2C3ɔ, represents an exception rather than the rule. Many Greek nouns are integrated into the inflectional structure of Syriac to one degree or another, but many fewer are integrated into the derivational structure of Syriac. In contrast to the particles and verbs, Greek nouns in Syriac show a significant degree of variation in their accommodation. In fact, in a number of cases, a single Greek noun can attest multiple accommodations for the Greek case endings, multiple accommodations of gender, and/or multiple plural formations in Syriac. This diversity suggests that some Greek nouns were transferred into Syriac on more than one occasion and accommodated differently by different speakers. In addition, in some cases, Syriac speakers seem to have maintained a connection between the loanword in Syriac and its Greek source, enabling them to adjust the accommodation of the loanword over time. This type of variety in the morphosyntactic accommodation of loanwords is only possible in a contact situation that stretches over an extended period of time. The Greek particles and verbs in Syriac, in turn, establish a high level of contact at various times. The existence of Greek particles in the earliest period of Syriac literature suggests a significant degree of contact already in the first centuries of the Common Era. The addition of a number of Greek particles and verbs into Syriac in the sixth century reflects the peak of contact at this time.

Grammatical Replication ✧ ✧ ✧

Chapter 7

Grammatical Replication: The Methodological Framework Knowledge of syntax is important in any language, and in Aramaic where the syntax in particular reflects the history of language most faithfully, it is of crucial significance. (Rosenthal 1995: 1)

7.1. Overview Almost all previous scholarly literature discussing contact-induced changes in Syriac due to Greek has been restricted to loanwords. Loanwords, however, are only one of the many different categories of contact-induced change. In chaps 7–9, I discuss a different category of contact-induced change, in which speakers of Syriac adapted inherited Aramaic material by replicating it on a Greek model. This category of change will be termed grammatical replication. The current chapter discusses the methodological framework for grammatical replication. Chapters 8–9 contain case studies of grammatical replication in Syriac due to Greek: the Syriac copula ʾiṯaw(hy) replicated on Greek ἐστίν (chap. 8) and the Syriac conjunctive particle den replicated on Greek δέ (chap. 9). 7.2. Definition This chapter discusses a category of contact-induced change that is termed grammatical replication, following the work of Heine and Kuteva. 1 Grammatical replication is defined in this book as a contact-induced change in which speakers of the recipient language create a new grammatical structure on the model of a structure of the source language. 2 Unlike loanwords, which involve the transfer of phonetic material, grammatical replication involves the transfer 1.  Heine and Kuteva 2003; 2005; 2006: 48–96; 2008; 2010. 2. Following Weinreich (1953: 30–31), Heine and Kuteva use replica language and model language in lieu of what is here termed recipient language and source language, respectively. I prefer the latter terms, because they can be used with other types of contactinduced change, such as loanwords (see §2.2).

139

140

Chapter 7

Graph 7.1.  Diachronic Frequency of Nisba Adjectives of semantic-conceptual material from the source language to the recipient language (Heine and Kuteva 2006: 68). 7.3.  Change in Frequency of a Pattern The most basic change in grammatical replication involves an increase in the frequency of a pattern. 3 In such cases, a pattern of low frequency in the recipient language becomes more frequent because it corresponds to a pattern in the source language. This represents the raising of a minor-use pattern to a major-use pattern. 4 This change often involves the selection and favoring of one pattern in the recipient language at the expense of another pattern. 5 This aspect of grammatical replication is analogous to indirect transfer in the work of SilvaCorvalán (1994), which she defines as “the higher frequency of use of a form in language S . . . in contexts where a partially corresponding form in language F is used either categorically or preferentially” (1994: 4). It also corresponds to Aikhenvald’s enhancement, “whereby certain marginal constructions come to be used with more frequency if they have an established correspondent in the source language” (Aikhenvald 2002: 238). Finally, it is similar to Mougeon and Beniak’s (1991: 10–12, passim) covert interference, in which “a minoritylanguage feature may undergo a gradual decline and eventual loss because it lacks an interlingual counterpart in the majority language . . . [and which] is accompanied by a concomitant rise in the use of the feature taking over the function vacated by the disappearing feature” (1991: 11). 3.  Heine and Kuteva 2003: 547; 2006: 50–57; 2010: 89; see also Thomason 2003: 711 n. 6. 4.  Heine and Kuteva 2006: 52; see also Poplack and Levey 2010: 393 with bibliography referred to therein. 5.  King 2000: 89; Poplack 1996: 289.

141

Grammatical Replication: The Methodological Framework

Table 7.1.  Frequency of Nisba Adjectives (Excluding Gentilics and Ordinal Numbers) Total Types

Nisba Types

%

740

3

1241

9

0.73

Selection of Aphrahaṭ (fl. 337–345)

996

4

0.40

Ephrem (d. 373), Prose Refutations, Discourse 1

992

7

0.71

Teaching of Addai (ca. 420)

922

7

0.76

Life of Rabbula (ca. 450)

1512

18

1.19

Selection of Philoxenos (d. 523)

1006

12

1.12

Book of the Laws of the Countries (ca. 220) Acts of Thomas (ca. 200–250), Acts 1–7

0.43

579

3

0.52

Eliya (mid-6th cent.)

1493

27

1.81

Selection of Yuḥanon of Ephesus (d. ca. 589)

1087

20

1.84

Denḥa (d. 649)

1082

21

1.94

Selection of Yaʿqub of Edessa (d. 708)

1109

22

1.98

Shemʿun of Beth Arsham (d. before 548)

The increase in the frequency of a pattern through grammatical replication can be illustrated with an example from Syriac-Greek language contact involving Syriac adjectives derived with the so-called nisba ending ‑ɔy (< *-āy). 6 In Syriac, as in other dialects of Aramaic, the nisba ending ‑ɔy forms gentilics, as in (7.1a), ordinal numbers, as in (7.1b), and other types of adjectives, as in (7.1c): (7.1) a. ‫ ܐܪܡܝܐ‬ʾɔrɔmɔyɔ ‘Aramean’ (LS2 50; SL 101) < *ʾarām ‘Aram’ + *-āyb. ‫ ܬܠܝܬܝܐ‬tliṯɔyɔ ‘third’ (LS2 826; SL 1649) < *θalīθ- ‘(passive participle of rt. θlθ “to be three”)’ + *-āyc. ‫ ܐܠܗܝܐ‬ʾalɔhɔyɔ ‘divine’ (LS2 21, 921; SL 47) < *ʾalāh- ‘God’ + *-āyIt is the last type—in other words, excluding gentilics and ordinal numbers— that is of concern here. Throughout Classical Syriac, this type of nisba adjective increased in frequency (Brock 2010). This increase is illustrated in table 7.1, which charts the percentage of nisba types per total lexeme types (verb, noun, and particle) across a corpus of more than 125,000 tokens from a selection of 12 prose texts. 7 The same data are charted in graph 7.1. 6.  For the nisba ending in Semitic, see Butts 2010: 81–82. 7.  In this book, type refers to a pattern, whereas token refers to actual instances of this pattern. Thus, in this case, type tracks whether a lexeme occurs in a given corpus (i.e., it is

142

Chapter 7

As can be seen in the table, the percentage of nisba types per total types increases from 0.43% in the Book of the Laws of the Countries (ca. 220) to 1.98% in the selected Letters of Yaʿqub of Edessa (d. 708). While this may initially seem to be a small increase, it must be remembered that this represents the increase of types of the nisba adjective relative to all of the types in the corpus. In other words, the Book of the Laws of the Countries contains 740 different words (or types), and only 3 of these are nisba adjectives (0.43%). In contrast, the selected Letters of Yaʿqub of Edessa contain 22 different nisba adjectives (types) in a corpus of 1,109 different words (1.98%). This means that the frequency of nisba adjectives in the selected Letters of Yaʿqub of Edessa is more than four times as much as that in the Book of the Laws of the Countries (an increase of 460% to be exact). The diachronic increase in the data set is further illustrated by the trend line in graph 7.1. The relatively high R2 value of 0.7257 suggests that the trendline is an accurate representation of the data. 8 Both table 7.1 and graph 7.1, then, illustrate that nisba adjectives, excluding gentilics and ordinal numbers, became increasingly more common throughout the history of Syriac. 9 The increase in nisba adjectives over the history of Syriac is due to contact with Greek. Syriac, at least in the early period, contained many fewer adjectives than Greek, often using other constructions, such as the so-called adjectival genitive 10—for example, daḵyɔnɔ ‘(lit.) of nature’, 11 where Greek would use an adjective—for example, φυσικός ‘natural’ (LSJ 1964). Over time, however, Syriac innovated new adjectival formations to replace the adjectival genitives. 12 Many of these new adjectives in Syriac were formed with the nisba ending -ɔy—for example, kyɔnɔyɔ ‘(lit.) pertaining to nature’ (LS2 321; SL 620; see also Brock 2010: 113). It is the creation of new adjectival formations such as kyɔnɔyɔ to replicate Greek adjectives that led to the diachronic increase in the nisba adjective discussed in the previous paragraph.

binary), whereas token tracks how many times a lexeme occurs in a given corpus. The selection of texts is the same as that used below for the verbless clause (§8.5). 8.  For readers less familiar with statistics, the coefficient of determination, or R2, ranges from 0 to 1.0 and reflects how well the regression line fits the data, with 1.0 indicating that the line perfectly fits the data. 9.  It should be noted that there is one statistical outlier in the data: Shemʿun of Beth Arsham (d. before 548), who has a relatively low percentage of nisba types per total types. This might be explained by the fact that Shemʿun was of Persian origin and had less exposure to Greek, the language of the Eastern Roman Empire. If Shemʿun of Beth Arsham is excluded from the dataset, the R2 value jumps to 0.8633. 10.  For the term adjectival genitive, see Waltke and O’Connor 1990: 148–154. ̈ 11.  See, e.g., ‫ܐܢܝܢ‬ ‫ ܡܛܠ ܕܕܟܝܢܐ‬mɛṭṭol daḏḵyɔnɔ ʾɛnnen ‘because they are natural’ [because nml+nml+nature-m.sg.emp they-f] (Book of the Laws of the Countries, 22.7–8 [ed. Drijvers 1965]). 12.  This trend was noted already in Brock 1990: 322; 2010; Becker 2006: 136.

Grammatical Replication: The Methodological Framework

143

One final piece to this puzzle is that it must be established that Syriac speakers actually equated Syriac adjectives, especially those derived with the nisba ending ‑ɔy, with Greek adjectives. 13 They could, of course, have chosen to identify Greek adjectives, such as φυσικός ‘natural’, with Syriac adjectival genitive constructions, such as daḵyɔnɔ ‘(lit.) of nature’. This did not in fact happen, however. Rather, it was Syriac adjectives that Syriac speakers identified with Greek adjectives. This identification can be established from the typology of translation technique, as is illustrated in the following example: (7.2) Second Epistle to Succensus by Cyril of Alexandria (ed. Schwartz 1927: 1.1.6.157–62) ἵνα

ᾖ

τὸ

πάθος

that

be-pres.act.subj.3.sg

art-nom.sg.neut

incident-nom.sg.neut

ἑκούσιον voluntary-nom.sg.neut

‘so that suffering would be voluntary’ (161.7) (7.3) Earlier Syriac Translation (ed. Ebied and Wickham 1975: 47–53 [Syr.], 39–43 [ET]) ‫ܕܢܗܘܐ ܚܫܐ ܕܨܒܝܢܐ‬ dnɛhwe

ḥaššɔ

dṣɛḇyɔnɔ

nml+be-pre.3.m.sg

suffering-m.sg.emp

nml+will-m.sg.emp

‘so that suffering would be of the will’ (51.27–28 [Syr.], 42.32–33 [ET]) (7.4) Later Syriac Translation (Brit. Libr. Add. 12,154, f. 188r., cited according to King 2008: 216) ‫ܡܛܠ ܕܨܒܝܢܝܐ ܢܗܘܐ ܚܫܐ‬ mɛṭṭol

dṣɛḇyɔnɔyɔ

nɛhwe

because

nml+voluntary-m.sg.emp

be-pre.3.m.sg suffering-m.sg.emp

ḥaššɔ

‘so that suffering would be voluntary’ In the earlier translation, cited in (7.3), the Greek adjective ἑκούσιον ‘voluntary’ is rendered by the Syriac adjectival genitive dṣɛḇyɔnɔ ‘of will’. A different translation of this same word, however, is found in the later translation in (7.4), in which the adjectival genitive was replaced by the Syriac nisba adjective ṣɛḇyɔnɔyɔ ‘(lit.) pertaining to will’. Thus, in the later translation, the Syriac nisba adjective ṣɛḇyɔnɔyɔ replicates the Greek adjective ἑκούσιον in contrast 13.  This is what Weinreich calls “interlingual identification” (Weinreich 1953: 7). See also Heine and Kuteva 2003: 531; Van Coetsem 1988: 21.

144

Chapter 7

to the earlier translation with the adjectival genitive. According to the wellestablished typology of Syriac translation technique, later translations, such as that in (7.4), tend to provide a more formal equivalence in comparison with earlier translations, such as in (7.3), often to the point that the lexical and morphological material of Syriac is mapped onto the semantic and grammatical categories of Greek. 14 This example thus shows that Syriac speakers equated Greek adjectives with Syriac adjectives, including those derived with the nisba ending ‑ɔy, rather than, for instance, with adjectival genitives. Throughout the history of Syriac, then, the nisba ending -ɔy became more frequent as Syriac speakers attempted to replicate Greek adjectives. This example of grammatical replication did not result in a new function for the nisba ending -ɔy since it already formed adjectives in early Syriac (and earlier in Aramaic for that matter). Rather, contact with Greek resulted in an increase in the frequency of the ending. That is, the nisba ending -ɔy was raised from a minor-use pattern to a major-use pattern. This example is illustrative of one of the more basic changes in grammatical replication, in which there is a diachronic increase in the frequency of a pattern in the recipient language due to its identification with a pattern in the source language. Additional examples involving a diachronic increase in the frequency of a pattern are illustrated in the case studies following this methodological introduction. 7.4.  Creation of a New Structure In addition to causing a change in frequency of a pattern in the recipient language, grammatical replication can result in the creation of new structures in the recipient language. This occurs when a structure in the recipient language comes to be used in new contexts on the model of the source language. 15 This, then, represents an extension in the function of a structure in the recipient language due to the function of the corresponding structure in the source language. 16 The creation of a new function due to grammatical replication can again be illustrated with an example from Syriac-Greek language contact, this time involving the use of Syriac lwɔṯ ‘toward; at, with’ with the verbal root ʾmr ‘to say’ on the model of the use of Greek πρός ‘on the side of, in the direction of’ with a verb of speech. The uses of Syriac lwɔṯ overlap significantly with those of Greek πρός in that both can express spatial relations, whether directional or locative. 17 Consider, for instance, the following example: 14.  On the translation technique in this particular passage, see King 2008: 216, 266–68. For the broader typology, see Brock 1979a; 2007a: 937–42; King 2008: 175–276. 15.  Heine and Kuteva 2006: 52. 16.  Weinreich 1953: 30–31. For extension as a mechanism of syntactic change, see Harris and Campbell 1995: 97–119. 17.  For Syriac lwɔṯ, see LS2 362; SL 682. For Greek πρός, see Humbert 1960: §§544–47; LSJ 1496–99.

Grammatical Replication: The Methodological Framework

145

(7.5) Hebrew Vorlage wəhinne ʾiššɔ and+behold woman-f.sg

šoḵɛḇɛṯ lay.part.f.sg

margəloṯɔ(y)w from+feet-f.pl.con+his

‘and behold a woman was lying at his feet’ (Ruth 3:8) (7.6) Greek Septuagint καὶ and

ἰδοὺ behold

γυνὴ κοιμᾶται woman-nom-f.sg sleep-pres.ind.mid.3.sg

ποδῶν feet-gen.m.pl

πρὸς πρὸς

αὐτοῦ he-gen.m.sg

‘and behold a woman was sleeping at his feet’ (Ruth 3:8) (7.7) Old Testament Peshiṭta (latter half of 2nd cent.) ‫ܐܢܬܬܐ ܕܕܡܟܐ ܠܘܬ ̈ܪܓܠܘܗܝ‬ ʾa(n)ttṯɔ dḏɔmḵɔ woman-f.sg.emp nml+sleep-part.f.sg.abs

lwɔṯ lwɔṯ

rɛḡlaw(hy) feet-f.pl.con+his

‘a woman, who was sleeping at his feet’ (Ruth 3:8) The Greek in (7.6) and the Syriac in (7.7) are independent translations of the Hebrew Vorlage in (7.5). Thus, it is noteworthy that Syriac lwɔṯ and Greek πρός are used in the exact same context: both are used for a locative relation. In addition, and particularly important for the point being argued here, Syriac lwɔṯ and Greek πρός overlap in a large number of verb-preposition combinations (i.e., particle verbs). Both, for instance, are used with verbs of returning, as is illustrated in the following example: (7.8) Hebrew Vorlage wənɔšuḇɔ and+return-pre.1.pl

ʾăleḵɛm toward+you-m.pl

‘and we will return to you’ (Gen 22:5) (7.9) Greek Septuagint ἀναστρέψωμεν πρὸς return-fut.act.ind.1.pl toward

ὑμᾶς you-acc.m.pl

‘and we will return to you’ (Gen 22:5) (7.10) Old Testament Peshiṭta (latter half of 2nd cent.) ‫ܘܢܗܦܘܟ ܠܘܬܟܘܢ‬ wnɛhpoḵ and+return-pre.1.pl

lwɔṯḵon toward+you-m.pl

‘and we will return to you’ (Gen 22:5)

146

Chapter 7

This example shows that Greek uses πρός, in (7.9), and Syriac uses lwɔṯ, in (7.10), in verb-preposition combinations that express ‘return to’, in this case (independently) translating the Hebrew Vorlage in (7.8). There are a number of other verb-preposition combinations in which Greek uses πρός and Syriac uses lwɔṯ, including to ‘bring to’ (Gen 2:19; 43:23); ‘to go to’ (Gen 15:15); ‘to turn aside to’ (Gen 19:3); ‘to be gathered to /at’ (Gen 25:8); ‘to draw near to’ (Gen 27:22; 37:18; 43:19; 45:4); ‘to send to’ (Gen 32:4); ‘to go up to’ (Gen 44:17, 24; 45:9); ‘to go down to’ (Gen 45:9); and even ‘to cry out to’ (Gen 4:10). The uses of Syriac lwɔṯ, then, overlap in a number of places with those of Greek πρός; they do not overlap entirely, however. One such use where they do not—or at least, did not initially—overlap is with certain verbs of speech. In early Syriac texts, the verbal root ʾmr ‘to say’ governs a dative object marked with the preposition l- ‘to, for’, as in the following example: (7.11) Book of the Laws of the Countries (ca. 220; ed. Drijvers 1965) ܿ ‫ܢ‬ ‫ܐܡܪ ܠܗ ܒܪ ܕܝܨ‬ ʾɔmar

leh

bar dayṣɔn

say-part-m.sg.abs

to+him

PN

‘Bardaiṣan said to him’ (4.14) In contrast, Greek πρός marks the dative object of various verbs of speech, such as λέγειν and εἰπεῖν: (7.12) Luke εἶπεν

δὲ

πρὸς αὐτὸν

say-aor.act.ind.3.sg but to

he-acc.m.sg

ὁ

ἄγγελος

art-nom.m.sg

angel-nom.m.sg

‘and the angel said to him’ (Luke 1:13) This, then, represents a difference between the use of Greek πρός and the original use of Syriac lwɔṯ. By the sixth century, however, the dative object of the verbal root ʾmr ‘to say’ in Syriac could also be marked with the preposition lwɔṯ ‘toward’ (Brock 2008: 4), as in the following example: (7.13) Apocalypse of Pseudo-Methodius (692; ed. Reinink 1993) ‫ܐܡܪ ܠܘܬ ܫܡܥܘܢ‬ ܼ ʾɛmar

lwɔṯ

say-suf.3.m.sg

toward PN

šɛmʿon

‘he said to Simon’ (20.9)

Grammatical Replication: The Methodological Framework

147

This represents an extension in the use of lwɔṯ on the model of Greek πρός. This extension resulted in a new function for Syriac lwɔṯ as it came to be used in a new context with the verb ʾmr ‘to say’. 18 The extension was facilitated by the fact that Syriac lwɔṯ and Greek πρός already overlapped in a number of uses, especially for spatial relations (whether directional or locative). The extension merely added one more use to Syriac lwɔṯ. Though admittedly of limited scope, this example is illustrative of a more dramatic type of change in grammatical replication in which a form in the recipient language becomes used in a new context on the model of the source language. Additional examples involving the creation of a new grammatical function are illustrated in the case studies following this methodological introduction. 7.5.  Grammatical Replication within Historical Linguistics As this last example illustrated, grammatical replication is not itself a mechanism of change but involves various mechanisms of change, such as reanalysis and extension. In the case of Syriac lwɔṯ, for instance, the major mechanism of change was extension. In addition to reanalysis and extension, cases of grammatical replication may also involve grammaticalization. 19 In these cases, a structure in the source language is replicated in the recipient language by following a common path of grammaticalization. It should be stressed, however, that there are cases in which grammatical replication does not involve grammaticalization. 20 These are called restructuring in the terminology of Heine and Kuteva. 21 Grammatical replication, then, is a broader category, which sometimes encompasses grammaticalization. 22 Thus, grammatical replication is not itself a mechanism of change but can involve various mechanisms of change, such as reanalysis, extension, and/or grammaticalization. 23

18.  Similar cases involving contact-induced changes in verb-preposition combinations in the French of Prince Edward Island are analyzed in detail in King 2000. 19.  Heine and Kuteva 2003; 2005: 79–122; 2006: 57–68; 2008; 2010: 87. Literature on grammaticalization theory is vast; for introductions, see Heine et al. 1991; Hopper and Traugott 1993; Heine and Kuteva 2002. For grammaticalization in Semitic, see Huehnergard 2006; and especially Rubin 2005. 20.  This differs from Sakel, who states that cases of pattern replication, which is roughly equivalent to grammatical replication in this study (see below, p. 148), “inherently involve a process of grammaticalization” (Sakel 2007: 17). 21.  Heine and Kuteva 2006: 65; see also the diagrams in 2006: 95 (fig. 2.1); 2008: 59 (fig. 1); 2010: 87 (fig. 4.1). 22.  Heine and Kuteva 2006: 64–65; 2010: 86–87. 23.  See similarly Aikhenvald 2003: esp. p.  3. This differs from Harris and Campbell (1995), who argue that borrowing, which is roughly equivalent to grammatical replication in this book (see below, p. 149), is itself a mechanism of change alongside extension and reanalysis.

148

Chapter 7

In the contact linguistic literature, it has become increasingly clear that contact-induced change and internally motivated change are not mutually exclusive. 24 Thus, in this study I do not adopt a binary framework according to which a change is either contact induced or internally motivated. Rather, a change can be contact induced, internally motivated, or both. Nevertheless, it is still important to establish whether or not language contact played a role in a given change. A good deal of scholarly literature has, in fact, been devoted to this question. 25 Establishing that language contact is a factor is especially important in cases involving grammatical replication, since these are often the most difficult to prove. 26 In the case studies of grammatical replication that follow, I have attempted to trace systematically the contact-induced changes in question with the support of historical data. This allows for a convincing case to be made for language contact playing a role in the described changes. As a final control for proving contact, the sister dialects of Syriac have also proven useful. 7.6.  Alternative Designations for Grammatical Replication I have already noted that the field of contact linguistics lacks a uniform terminology (§2.2). This is particularly the case for changes that are termed grammatical replication in this study, which it seems that almost every contact linguist calls by a different name. Thus, it is useful to conclude this methodological introduction with a survey of various alternative designations that have been applied to similar types of contact-induced change in the contact linguistic literature. 27 Grammatical replication is similar to the replication of linguistic patterns in Matras and Sakel’s typological project on Language Convergence and Linguistic Areas. 28 In this framework, the replication of linguistic patterns “pertains to the semantic and grammatical meanings and the distribution of a construction or structure” in contrast to the replication of linguistic matter, which involves “actual phonological segments” (Matras and Sakel 2007b: 7; see also 2007c). Thus, their “replication of linguistic patterns” is an exact synonym for grammatical replication as used in this book. 24.  See, e.g., Dorian 1993; Heine and Kuteva 2003: 531–32; 2006: 7, 73–79; Hickey 2002; 2010b: 15–16, 21; Jones and Esch 2002 [with many additional references]; Poplack 1996: 290; Thomason 1986: 278–79; 2001: 86, 88; 2010: 32, 34. 25.  See, e.g., Harris and Campbell 1995: 407–8; Haugen 1950b: 226–28; Heine and Kuteva 2006: 73–79; Kutscher 1954: 240–43; Mønnesland 1999: 327–36; Poplack 1996: 290; Poplack and Levey 2010; Thomason 2001: 86, 88, 91–95; 2003: 708–10; 2004: 8–9; 2010: 34–35. 26. See Thomason 2003: 709; and especially Poplack and Levey 2010. 27.  See also the surveys in Kuteva 2005: 6–13; Ross 2006: 96–97; 2007: 132–35. 28.  Matras and Sakel 2007a; 2007c; Sakel 2007. See also http://www.llc.manchester .ac.uk/research/projects/lcla/.

Grammatical Replication: The Methodological Framework

149

Grammatical replication also encompasses what Harris and Campbell term borrowing, which they define as “a change in which a foreign syntactic pattern (either a duplication of the foreign pattern or at least a formally quite similar construction) is incorporated into the borrowing language through the influence of a donor pattern found in a contact language” (Harris and Campbell 1995: 122). 29 A number of other linguists have also termed types of contactinduced change similar to grammatical replication as “structural borrowing,” 30 “syntactic borrowing,” 31 or “grammatical borrowing.” 32 Grammatical replication also shares similarities with metatypy, a type of contact-induced change that has been described in a series of studies by Ross and also employed by others. 33 Ross defines metatypy as “a diachronic process whereby the morphosyntactic constructions of one of the languages of a bilingual speech community are restructured on the model of the constructions of the speakers’ other language” (Ross 2007: 116). In his various publications, Ross vacillates over whether this restructuring affects morphosyntactic constructions, as in this definition, or is restricted to “syntax” (Ross 2006: 95) or is extended to “semantic and morphosyntactic structure” more generally (Ross 1996). In his work before 2006, Ross included varying degrees of restructuring within the category of metatypy. Since 2006, however, Ross has narrowed his definition of metatypy to include only such restructuring as results in a change in type, with type to be understood in the sense of typology—for example, a change from SOV to SVO word order. Ross now refers to similar kinds of contact-induced change that do not result in a change in type as calquing or more specifically grammatical calquing. Thus, in Ross’s morerecent work, grammatical calquing and metatypy result in similar changes but differ in degree (occasional vs. systemic). Grammatical replication as defined in this study, then, is a broader category of contact-induced change, which encompasses both Ross’s metatypy and grammatical calquing. A number of linguists in addition to Ross have labeled contact-induced changes similar to grammatical replication as calques. 34 Some prefer to qualify the term calque further, such as “lexicon-syntactic calques” (Silva-Corvalán 1994: 174–84) or Ross’s “grammatical calques,” mentioned above. Grammatical replication is similar to selective copying in the Code-Copying Model developed by Johanson (see, e.g., 2002a) and subsequently employed 29. For a slightly different definition that incorporates “replication,” see Harris and Campbell 1995: 51. 30.  See, e.g, Aikhenvald 2002; 2003; Emeneau 1962; Nadkarni 1975; Thomason and Kaufman 1988: 67; Winford 2003: 12. 31.  See, e.g, Appel and Muysken 1987: 158–62; Mønnesland 1999. 32.  See, e.g., King 2000; Matras and Sakel 2007a: 1; Wohlgemuth 2009: 224, 272. 33.  See, e.g., Ross 1996; 2001; 2003; 2006; 2007; 2008; as well as Bowden 2005. See also Noonan 2010: 56. 34.  See, e.g., Heath 1984: 367; Hetzron 1976: 99.

150

Chapter 7

by others. 35 The Code-Copying Model describes contact-induced change in terms of “elements of a foreign code being copied into the code of the recipient language” (Johanson 2002a: 8–9). This copying can be either global or selective. In global copying, a unit of a foreign code is copied into the basic code in its entirety—that is, as “a block of material, combinational, semantic and frequential structural properties” (2002a: 9). The most common examples of global copying are what are called loanwords in this book. In selective copying, in contrast, only selected properties are copied. Johanson’s selective copying is a broader category than grammatical replication in that it can also include, inter alia, copies of phonology and semantics; nevertheless, grammatical replication, as employed in this book, is similar to the selective copying of (morpho‑)syntax in Johanson’s Code-Copying Model. 36 Grammatical replication overlaps with what a number of scholars term convergence. Though the term convergence was employed in earlier contact linguistic literature (e.g., Weinreich 1953: 113), its more recent use seems to be based primarily on the influential study of language contact in the Kupwar village (India) by Gumperz and Wilson (1971). In their study, Gumperz and Wilson use convergence to refer to a series of contact-induced changes that led Marathi, Hindi, and Kannada to develop the same surface syntactic structure resulting in the intertranslatability of the three languages. This use of convergence has been adopted by a number of linguists. Silva-Corvalán, for instance, defines convergence as “the achievement of greater structural similarity in a given aspect of the grammar of two or more languages, assumed to be different at the onset of contact” (Silva-Corvalán 1994: 4–5; 1995: 8). Similarly, Thomason (2007: 187; see also 2003: 700) uses convergence to refer to a type of contact-induced change that usually occurs in situations of long-term bilingualism in which structures common to both languages are favored, often resulting in a change of frequency of existing patterns and not in the addition of new patterns. In addition, Aikhenvald employs convergence in the sense of “structural isomorphism, whereby the grammar and semantics of one language are almost fully replicated in another” (Aikhenvald 2002: 6). Matras (2010) has also used convergence in the sense of his pattern replication, which was mentioned above (p. 148). Convergence is used in similar senses by a number of other scholars. 37 In many of these cases, convergence involves systemic changes. 38 Thus, grammatical replication would be a broader category, including convergence. 35.  See, e.g., Csató 2001; 2002; Hayasi 2000; Kıral 2000; Menz 2000. 36.  For the relationship between grammatical replication and Johanson’s Code-Copying Model, see Heine and Kuteva 2005: 6–7. 37.  See, e.g., Myers-Scotton 2006: 271; Poplack 1996: 286; Poplack and Levey 2010: 399; Pray 1980; Sridhar 1978; Winford 2009: 281–82. 38.  See also Matras 2010: 68.

Grammatical Replication: The Methodological Framework

151

Grammatical replication is similar to indirect diffusion in the work of Aikhen­vald (2002; see also Heath 1978). In Aikhenvald’s framework, indirect diffusion refers to the transfer “of categories, or of terms within a category” (2002: 4). Indirect diffusion can involve a number of different changes, including (1) “the emergence of new categories and new paradigms . . . through reanalysis of existing grammatical patterns and through shared grammaticalization processes” (2002: 237); (2) “the creation of new categories—by what can be called ‘loan translation’ . . . or by introducing new morphemes” (2002: 237); and (3) “changes known as ‘enhancement’—whereby certain marginal constructions come to be used with more frequency if they have an established correspondent in the source language” (2002: 238). In each of these, Aikhenvald’s indirect diffusion is similar to grammatical replication, as used in this book. These represent only some of the many different terms by which contactinduced changes similar to grammatical replication are known in the contact linguistic literature. Others include “modeling,” 39 “covert interference,” 40 “pattern transfer,” 41 “indirect transfer,” 42 “loanshift,” 43 “congruence,” 44 “interference,” 45 “resyntactization,” 46 “loan translation,” 47 and so forth. Out of these various terms, grammatical replication has been adopted in this book because it is broad enough to include various kinds of change that affect the structural material of language, especially (morpho-)syntax. In addition, the theory of grammatical replication as developed by Heine and Kuteva can be applied equally to situations of borrowing, imposition, and neutralization. 48 7.7. Conclusion This chapter has established the methodological framework for a kind of contact-induced changed termed grammatical replication. Grammatical replication was defined as a contact-induced change in which speakers of the recipient language create a new grammatical structure on the model of a structure of the source language. Grammatical replication can result in various kinds of change in the recipient language. Two in particular were discussed and illustrated: (1) an increase in the frequency of a pattern, and (2) the creation of new structures. Grammatical replication, as defined in this book, is similar to a 39.  Silva-Corvalán 1994. 40.  Mougeon and Beniak 1991: 10–11, and passim. 41.  Heath 1984: 367. 42.  Silva-Corvalán 1994. 43.  Haugen 1950a: 289; 1950b: 215, 219–20. 44.  Corne 1999: 8, 9, and passim; Mufwene 2001: 23, and passim. 45.  Weinreich 1953: 30–31. 46.  Appel and Muysken 1987: 158–59. 47.  Türker 1999. 48.  For this typology, see §§2.3–2.6.

152

Chapter 7

number of other changes discussed in the contact linguistic literature, including borrowing, metatypy, calque, and convergence. The next two chapters provide extended case studies of grammatical replication in Syriac due to Greek. Chapter 8 argues that the development of the Syriac copula ʾiṯaw(hy) ‘he is’ is due, at least partly, to its replication on the Greek verbal copula ἐστίν ‘he is’. Chapter 9 discusses the replication of the Syriac conjunctive particle den ‘then, but’ on the model of Greek δέ ‘but’.

Chapter 8

The Syriac Copula ʾiṯaw(hy) Replicated on Greek ἐστίν No doubt even the best original writings in Syriac give evidence of the strong influence of Greek Syntax. . . . The Greek idiom exercised its influence with all the greater force and effect, precisely at those points where Syriac itself exhibited analogous phenomena. (Nöldeke 1904: ix–x)

8.1. Overview The past several decades have witnessed a number of syntactic studies on Syriac. While Nöldeke’s Compendious Syriac Grammar (1904)—with an occasional clarification from Duval (1881) and Brockelmann (1981)—remains unsurpassed in its description of the phonology and morphology of Classical Syriac, 1 studies of word order, 2 cleft sentences, 3 and the particle d- 4 (to name only a few) have not so much refined Nöldeke’s description as entirely replaced it. 5 Within this resurgence of syntactic research on Syriac, the most significant progress has been made in the analysis of the verbless clause. Stemming from the watershed study of Goldenberg (1983) with important additions by others, 6 the Syriac verbless clause has become increasingly well understood. That being said, however, its description is far from complete. In particular, studies of the Syriac verbless clause—like Syriac grammatical studies in general—have been limited by a lack of diachronic perspective. In addition, there continues 1.  So also Goshen-Gottstein 1989: 236–37; Van Rompay 2001. 2.  Avinery 1975; 1976; 1984; Joosten 1993; Muraoka 1972. 3.  Goldenberg 1971; 1990; Wertheimer 2001a; 2001c. 4.  Wertheimer 2001b. 5.  It is for this reason that I am currently preparing a new syntax of Classical Syriac to be published with Ugarit-Verlag in the series Lehrbücher orientalischer Sprachen. 6.  See, e.g., Goldenberg 1991; 2006; Joosten 1989; 1992; 1996: 77–96; 2006; Muraoka 1975; 1977; 1997: §§102–9; 2006; Pat-El 2006; van Peursen 2006a; 2006b; Van Rompay 1991; Wertheimer 2002.

153

154

Chapter 8

to be no agreement on the possible role that contact with Greek played in changes in verbless clause formation in Syriac, with some arguing that contact with Greek was a factor, 7 while others maintain that it was not. 8 The current chapter explores the role that contact with Greek played in the creation of a fully functioning copula in Syriac from the existential particle ʾiṯ ‘there is’ plus a pronominal suffix. Among the many attested changes in this development, I argue that two are specifically the result of contact with Greek: (1) the extension of the copulaic use of ʾiṯ to verbless clauses with substantival predicates (§8.3); (2) the raising of copulaic ʾiṯ from a minor-use pattern to a major-use pattern throughout the history of Syriac (§8.5). Throughout this chapter, I pay particular attention to establishing that language contact did in fact play a role in the described changes. This is important for the field of Syriac studies since, as has already been mentioned, this remains an open question. In addition, this represents a valuable contribution to the field of contact linguistics. Contact linguistic scholarship continues to dispute whether or not structure can be transferred in situations of borrowing. In the words of Poplack, “The transfer of grammatical structure in a situation of language contact has had a contentious history in linguistic thought, and no consensus has yet been reached regarding its nature, extent, or even its existence” (Poplack 1996: 285). At least part of this disagreement stems from the fact that many of the purported cases of structural transfer in situations of borrowing are based on insufficient data and lack adequate analysis. 9 In this chapter, therefore, I aim to provide an example of the transfer of structure in a situation of borrowing. 10 8.2.  Verbless Clause Formation in Syriac It is necessary to begin with an overview of verbless clause formation in Syriac. In Syriac, verbless clauses can be constructed in two basic ways (Wertheimer 2002), which will be termed Pattern A and Pattern B. 11 Pattern A consists of the word order predicate-subject with the subject restricted to an enclitic personal pronoun, as in (8.1): 12 7.  See, e.g., Jenner 2003: 307; Joosten 1996: 107; 1999: 213–14; Muraoka 1985: 77; 2006: 131–34; Wertheimer 2002: 12–13. 8.  See most recently Pat-El 2006: 342–44. 9.  See recently Poplack and Levey 2010. 10.  Arguments in favor of analyzing contact-induced changes in Syriac due to Greek as borrowing are presented in §3.4 above. 11.  There is a third (marginal) pattern for the verbless clause in Syriac, which consists of the simple juxtaposition of subject and predicate (Nöldeke 1904: §§310, 312b; Muraoka 1987: §103; Joosten 1992: 586–87; 1996: 91–93; Butts 2006: 58–61). For my use of verbless clause instead of “nominal clause” or “non-verbal clause,” see Butts 2006: 56 n. 13. 12.  For this pattern, see Goldenberg 1983 with additions in Goldenberg 1991; Joosten 1989; 1992; 1996: 77–96; Muraoka 1975; 1977; 1997: §§102–9; Pat-El 2006; Van Rompay 1991; Wertheimer 2002.

The Syriac Copula ʾiṯaw(hy) Replicated on Greek ἐστίν

155

(8.1) Syriac Acts of Thomas (3rd cent.; ed. Wright 1871a) ‫ܘܓܒܪܐ ܐܢܐ ܥܒܪܝܐ‬ wgaḇrɔ

(ʾ)nɔ

and+man-m.sg.emp I

ʿɛḇrɔyɔ Hebrew-m.sg.emp

‘I am a Hebrew man’ (172:13) In this sentence, the nexus between the subject (ʾ)nɔ ‘I’ and the predicate gaḇrɔ ‘man’ is expressed by the syntactic juxtaposition of the two terms. 13 The subject in this type of verbless clause is restricted to an enclitic personal pronoun in Syr­ iac. 14 In example (8.1), the enclitic status of the pronoun (ʾ)nɔ ‘I’ is indicated by its phonologically reduced form—the independent form is ʾɛnɔ—as well as by the fact that it interrupts the noun-adjective phrase gaḇrɔ ʿɛḇrɔyɔ ‘Hebrew man’. When a subject other than a personal pronoun is to be expressed with a Pattern A verbless clause, the logical subject is extraposed either to the front or to the rear of the predicate-subject nucleus, with the personal pronoun resuming the extraposed logical subject: 15 (8.2) Syriac Acts of Thomas (3rd cent.; ed. Wright 1871a) ‫ܡܛܠ ܕܛܝܒܘܬܝ ܥܡܟ ܗܝ‬ mɛṭṭol

dṭaybuṯ(y)

ʿammɔḵ

(h)i

because

nml+grace-f.sg.con+my

with+you-m.sg she

‘because my grace is with you’ (172:16) In this example, the predicate is ʿammɔḵ ‘with you’, and the subject is (h)i ‘she’, which refers to the extraposed logical subject ṭaybuṯ(y) ‘my grace’. This type of extraposition in verbless clauses is not limited to Syriac but occurs in other dialects of Aramaic, such as Achaemenid Aramaic (Muraoka and Porten 1998: 294–96) and Jewish Palestinian Aramaic (Butts 2006: 61–64). Verbless clauses belonging to Pattern B are constructed with the existential particle ʾiṯ ‘there is’ plus a possessive pronominal suffix: 16 13.  Goldenberg 1983: 111–12; 1987–88: 113–15. 14.  There are rare instances in which the pronoun does not seem to be enclitic: ʿam hɔy ger dʾalɔhɔ ʾa(n)tton [with that-f.sg for nml-god-m.sg.emp you-m.pl] ‘For, you are with that of God’ (Philoxenos, Letter to the Monks of Beth Gawgal; ed. Vaschalde 1902: 158.16). It should be noted that the second-person pronouns, including ʾa(n)tton, do not have a marked, unattached enclitic form of the pronoun (in contrast to the first- and third-person pronouns); it is still noteworthy, however, that ʾa(n)tton is not in the enclitic word position in this example. It should also be noted that the pronoun is not enclitic in some other dialects of Aramaic (see n. 23 below). 15.  Zewi 1996: 41–55; and especially Goldenberg 1998: 165–67. 16.  For the various uses of ʾiṯ in Syriac, including the copulaic use, see Jenner 2003; Joosten 1996: 97–107; Goldenberg 1983: 117–131; Muraoka 1977; 2006; Nöldeke 1904: §§301–8; Wertheimer 2002: 4–5.

156

Chapter 8

(8.3) Syriac Acts of Thomas (3rd cent.; ed. Wright 1871a) ‫ܒܪ ܥܣܪܝܢ ܓܝܪ ܘܚܕܐ ܫܢܝܢ ܐܝܬܝ ܝܘܡܢܐ‬ bar

ʿɛsrin

ger

son-m.sg.con twenty-m.pl.abs for ʾiṯay

yawmɔnɔ

ex+my

today

waḥḏɔ

šnin

and+one-f.sg.abs

years-m.pl.abs

‘for I am twenty-one years old today’ (317:19–20) In this example, ʾiṯ serves as the nexus between the subject ‘I’, which is expressed by a possessive pronominal suffix, and the predicate bar ʿɛsrin . . . ‘son of twenty . . .’. In Syriac, there are also rare examples in which an enclitic personal pronoun occurs instead of a possessive pronominal suffix: 17 (8.4) Letter 47 by Timotheos I (d. 823; ed. Braun 1901) ܿ ‫ܐܢܝܢ ܕܝܢ ܠܓܡܪ‬ ̈ ‫ܒܗܘ �ܡܐ ܕܐܚܝܕܝܢܢ‬ ‫ܠܝܬ‬ layt

ʾɛnnen

den

laḡmar

bhaw

neg+ex

they-f

but

completely

in+that-m.sg

mɔ

dʾaḥiḏinan

what

nml+be.held-part.m.pl+we

‘But, they are not at all in that (book) that we possess’ (306.10–11) 18 (8.5) Syriac Translation of Eusebius’s Ecclesiastical History (before 420; 19 ed. Wright and McLean 1898) ̈ ‫ܠܝܬ ܐܢܘܢ‬ ‫ܢܒܝܐ‬ layt

ʾɛnnon

nbiye

neg+ex

they-m

prophet-m.pl.emp

‘they are not prophets’ (297.13–14) 20 17.  Nöldeke 1904: §302; Goldenberg 1983: 117; Van Rompay 1994: 82–83; Joosten 1996: 107. 18. Contrast this with a similar construction, but with a suffix, several lines later: wlaytaw(hy) hɔnɔ pɛṯḡɔmɔ bšaḇʿin [and+neg+ex+his this-m sentence-m.sg.emp in+seventym.pl.abs] ‘and this sentence is not in the Septuagint’ (Letter 47 by Timotheos I; ed. Braun 1901: 306.14). Similarly see Braun 1901: 304.21–22. 19.  This translation is preserved in one of the earliest dated Syriac manuscripts, St. Petersburg, Public Library, Cod. Syr. 1 (461/462). The translation must have predated this manuscript by at least half a century, however, since the Syriac version was the basis of an Armenian translation from the first decades of the 5th century (Van Rompay 1994: 73 n. 15; see also Merx, apud Wright and McLean 1898: xiii–xvii). 20. Translating Greek οὐκ εἰσὶ προφῆται [neg be-pres.act.ind.3.pl prophet-m.pl.nom] ‘they are not prophets’.

The Syriac Copula ʾiṯaw(hy) Replicated on Greek ἐστίν

157

As the latter example illustrates, most cases of ʾiṯ, or the negative layt ( *ʾḏáyn > *ʾiḏáyn > **ʾɛḏén. 26 Note, however, that there are isolated cases in which an initial glottal stop is lost in Syriac—for example, *ʾunāsaʾ > nɔšɔ ‘man, humanity’ (but written ⟨ʾnšʾ⟩) (LS2 31; SL 65). This occurs more commonly before ḥ 27—for example, *ʾaxātaʾ > ḥɔṯɔ ‘sister’ (LS2 10; SL 503) and *ʾaxarataʾ > ḥarṯɔ ‘end’ (LS2 13; SL 497). These cases of aphaeresis of the initial glottal stop are irregular phonological developments, however. This suggests that the aphaeresis of the initial glottal stop in Syriac den (< *ʾiðayn) is due to its replication on Greek δέ. One possibility is that the aphaeresis of the initial glottal stop is a result of the phonological erosion that often occurs in grammaticalization. 28 This phonological erosion would not be surprising given that many cases of grammatical replication also involve grammaticalization. 29 Another possibility is that the initial glottal stop was deleted in an effort to make Syriac den resemble Greek δέ more closely. This would then be a case in which Syriac den was replicated phonologically on Greek δέ, which would be significant, since grammatical replication involves the transfer of semantic-conceptual material, but it is usually not thought to involve the transfer of phonological material. Given the paucity of comparable cases in the literature, it is difficult to choose between these two options (or perhaps it is not an either/or choice). Regardless, it is clear that the aphaeresis of the initial glottal stop in Syriac den (< *ʾiðayn) is due to its replication on Greek δέ and that this development led to the former resembling the latter phonologically. These developments in the syntax, semantics, and phonology of Syriac den can be contrasted with Syriac hɔyden ‘then’ (LS2 174; SL 340), which derives from *ʾiðayn with a prefixed *hā‑. 30 Syriac hɔyden more closely resembles earlier Aramaic reflexes of *ʾiðayn than Syriac den, as is illustrated in the following example:

‫ ܡܪܟ ܒܟܘܠܠܟ‬mɔn den ḥawwi lan ʾɛn ʾɛṣṭḇi mɔrɔk bḵullɔlɔḵ [who den show-suf.3.m.sg to+us if be.pleased‑suf.3.m.sg lord-m.sg.con+your-m.sg in+coronation-m.sg.con+your-m. sg] ‘would that we be shown if our Lord is pleased with your coronation!’ (History of ʿAbdɔ damšiḥɔ; ed. Bedjan 1890–97: 1.190.10–11; editio princeps in Corluy 1886; I am currently preparing a new edition of this text with translation in collaboration with Simcha Gross). 26.  For the reduction of the pretonic short vowel followed by the secondary epenthesis of *i (or more rarely *a), compare *ʾamára > *ʾamár > *ʾmár > *ʾimár > ʾɛmár ‘he said’. 27. See Huehnergard 1998: 269 n. 19 with bibliographical references there. 28.  For phonological erosion in grammaticalization, see, inter alia, Hopper and Traugott 2003: 154–59; Rubin 2005: 4–5. 29.  Heine and Kuteva 2003; 2005: 79–122; 2006: 57–68; 2008. 30.  For the broader Semitic context of *hā-, see Hasselbach 2007: 21, passim.

The Syriac Conjunctive Particle den Replicated on Greek δέ

185

(9.17) Demonstrations by Aphrahaṭ (336/7; ed. Parisot 1894–1907) ̈ ‫ܘܡܘܫܐ ܗܘܐ ܒܡܕܝܢ ܬܠܬܝܢ‬ ‫ܫܢܝܢ ܗܝܕܝܢ ܟܕ ܣܓܝ ܥܠܝܗܘܢ ܐܘܠܨܢܐ‬ ‫ܐܦܩ ܐܢܘܢ ܡܢ ܡܨܪܝܢ‬ wmuše

hwɔ

bmɛḏyan tlɔṯin

šnin

and+pn

become-suf.3.m.sg

in+gn

year-f.pl.abs

thirty-m.pl.abs

hɔyden

kaḏ

sḡi

ʿlayhon

ʾulṣɔnɔ

hɔyden

when

great-m.sg.abs

against+them-m

hardship-m.sg.emp

ʾappɛq

ʾɛnnon

mɛn

mɛṣren

bring.out-suf.3.m.sg

them-m

from

gn

‘Moses was in Midian for thirty years. Then, when the suffering was great against them, he led them out of Egypt’ (65.2–4) In this example, Syriac hɔyden occurs in clause-initial position and functions as a temporal adverb meaning ‘then, at that time’, just like earlier Aramaic reflexes of *ʾiðayn. 31 In addition, hɔyden even preserves a trace of the initial syllable of Aramaic *ʾiðayn in the palatal glide y: *hā + *ʾiðayn > *hāʾidayn > *hāyiḏayn > hɔyden. 32 The syntactic, semantic, and phonological differences between Syriac hɔyden and den highlight the degree to which the latter has been replicated on Greek δέ. Syriac den is already attested in the Peshiṭta version of the Pentateuch, which was translated (from Hebrew) by the middle of the second century. 33 Thus, these developments in syntax, semantics, and phonology had already occurred in Syriac by at least that time. Nevertheless, den is rare in the Peshiṭta Pentateuch occurring only 48 times in over 115,000 total tokens. 34 This is less than once every 2,400 tokens. In texts from the third and fourth centuries, den is encountered much more frequently, occurring once every 190 tokens in the Book of the Laws of the Countries (ca. 220; ed. Drijvers 1965), once every 207 tokens in Acts 1–7 of the Acts of Thomas (ca. 200–250 c.e.; ed. Wright 1871a: 2.171–251 [Syr.]), once every 327 tokens in Demonstrations 1–3 by Aphrahaṭ (fl. 337–45; ed. Parisot 1894–1907), and once every 80 tokens in Discourse 1 of the Prose Refutations by Ephrem (d. 373; ed. Overbeck 1865: 21–58). This is summarized in table 9.1. Thus, by the third century, Syriac den was occurring much more frequently than it had in the second century.

31.  In fact, in his Letter on Syriac Orthography, Yaʿqub of Edessa (d. 708) specifically states that hɔyden is a ba(r)t qɔlɔ zaḇnɔyṯɔ ‘word of time’ (ed. Phillips 1869: 6.12–13). 32.  All of these changes are regular, except for the loss of the fricativization of *ḏ, which can be explained by analogy with Syriac den. 33.  For the date, see Weitzman 1999: 248–58. 34.  These numbers are based on CAL and differ slightly from Taylor 2002.

186

Chapter 9

Table 9.1.  Frequency of Syriac den in Early Syriac Prose Tokens of den

Total Tokens

Frequency

Peshiṭta Pentateuch (ca. 150)

48

115,523

1 : 2,407

Book of the Laws of the Countries (ca. 220)

39

7,420

1 : 190

Acts of Thomas (ca. 200–250 c.e.), Acts 1–7

77

15,721

1 : 204

36

11,772

1 : 327

118

9,322

1 : 79

Aphrahaṭ (fl. 337–45), Demonstrations, 1–3 Ephrem (d. 373), Prose Refutations, Discourse 1

The dramatic increase in the frequency of Syriac den is due to its replication on Greek δέ. Greek δέ occurs at a much higher frequency than once every 2,400, the rate of occurrence of den in the Peshiṭta Pentateuch. The comparatively greater frequency of Greek δέ can be illustrated by the following legal text from Dura-Europos, in which δέ is used to introduce most new sentences: (9.18) P. Dura 12 (225–50) 35  τῶν δὲ τελευ̣[τη]σ̣ άντω[ν τ]ὰς κληρονομείας ἀποδίδοσ̣ [θ]ε τοῖς [ἄγ]χιστα γένους, ἀγχιστε͂ς δὲ οἵδε· ἐὰν μὴ [τέ]κ̣ να λείπῃ ἢ υἱοποιήσητε κατὰ τὸν νόμον πατὴρ ἢ μήτηρ, μὴ ἄλλῳ ἀνδρὶ συνοικοῦσα· ἐὰν δὲ μηθεὶς τούτων ᾖ ἀδελφοὶ ὁμ[οπ]άτριοι· ἐὰν δὲ μηδὲ οὗτοι ὦσιν ἀδελφὲ ὁμοπάτριοι· ἐὰν δὲ μηθὶς τούτων ᾖ, πατρὸς δὲ πατὴρ ἢ πατρὸς μήτηρ ἢ ἀνεψιὸς ἀπὸ πατρὸς γεγεννημένος, τούτων ἡ κληρονομία ἔστω. ἐὰν δὲ μηθὶς τούτων ὑπάρχῃ βασιλικὴ ἡ οὐσία ἔστω. κατὰ δὲ ταῦτα ἔστωσαν καὶ αἱ ἀγχιστίαι   ‘(δέ) The inheritances of those who have died are to be rendered to the next of kin of the family; (δέ) the next of kin are these: If (the deceased) does not leave children or has not legally adopted a son, the father or the mother who has not been married to another man (receives the inheritance). (δέ) If neither of these is alive, brothers of the same father (receive the inheritance). (δέ) If none of these is alive, sisters of the same father (receive the inheritance). (δέ) If none of these is alive, (δέ) the inheritance belongs to the father’s father, the father’s mother, or a male cousin on the father’s side. 35.  The text is reproduced as on the actual document; note the following differences from standard Koinē orthography: κληρονομείας for κληρονομίας; ἀποδίδοσ̣ [θ]ε for ἀποδίδοσθαι; ἀγχιστε͂ς for ἀγχιστεῖς; υἱοποιήσητε for υἱοποιήσηται; ἀδελφὲ for ἀδελφαὶ; μηθὶς for μηθεὶς (2×); ἀγχιστίαι for ἀγχιστείαι.

The Syriac Conjunctive Particle den Replicated on Greek δέ

187

(δέ) If none of these is alive, the property is the king’s. (δέ) The rights of kin should also be according to these things’.   (P. Dura 12.3–18) In this text, δέ occurs 8 times, or once every 9.75 tokens. While this greater frequency of occurrence is not representative of all Greek texts, it does clearly illustrate that δέ occurs at a greater rate of frequency in Greek than den did in the phase of Syriac represented in the Peshiṭta Pentateuch. It is this comparatively greater frequency of Greek δέ that led to an increase in the frequency of den in the early history of Syriac. Thus, by the third century, Syriac den had not only been replicated on Greek δέ in its syntax, semantics, and phonology, but it had also become more frequent due to contact with Greek. In addition to being replicated on Greek δέ in its syntax, semantics, and phonology, Syriac den underwent a further development. In Greek, δέ can be used in conjunction with the second-position particle μέν to form a construction glossed ‘on the one hand . . . on the other hand . . .’ (Smyth 1956: §2904): (9.19) P. Dura 31 (204) 36 Ναβουσάμαος

μὲν

τῇ

Ἀκόζζει

συνοικεῖν

PN

men

art-dat.f.sg

PN

cohabit-inf.pres.act

ἑτέρῳ

ἀνδρὶ

ὃ

ἂν

αὐτὴ

another-m.sg.dat

man-m.sg.dat

rel-acc.sg

cnd

she-nom

αἱρῆται

Ἄκοζζις

δὲ

τῷ

Ναβουσαμάῳ

take-pres.act.subj.3.sg

PN

de

art-dat.m.sg

PN

γαμεῖν

ἄλλην

marry-inf.pres.act another-acc.f.sg

γυναῖκαν

ὃ

woman-acc.f.sg rel-acc.sg

ἂν

αὐτὸς

βούληται

cnd

he-nom

want-pres.act.subj.3.sg

‘Nabusamaos, on the one hand (μὲν), (gives) to Akozzis to cohabitate with another man whom she chooses; Akozzis, on the other hand (δὲ), (gives) to Nabusamaos to marry another woman whom he wants’ (P. Dura 31.9–12) In Syriac, a similar construction is formed with man (LS2 393; SL 778), a loanword from Greek μέν (LSJ 1101–2), and den: 36.  The text is reproduced as on the actual document; note the following difference from standard Koinē: γυναῖκαν for γυναῖκα.

188

Chapter 9

(9.20) Letter 13 by Yaʿqub of Edessa (d. 708; ed. Wright 1867: *1–*24) ‫ �ܠܐܒܪܗܡ ܕܝܢ ܢܝܚܐܝܬ ܘܗܢܝܐܝܬ‬.‫ܠܩ̈ܪܩܣܐ ܿܡܢ ܛܪܕ ܡܢ ܚܩܠܗ ܕܐܒܪܗܡ‬ ‫ܐܡܪ܆ ܐܒܪܗܡ ܐܒܪܗܡ‬ man

lqarqɔse

ṭraḏ

to+vulture-m.pl.emp man

mɛn

ḥaqleh

expel-suf.3.m.sg from

field.m.sg.con+his

dʾaḇrɔhɔm

lʾaḇrɔhɔm

den

niḥɔʾiṯ

whanyɔʾiṯ

nml+pn

to+pn

den

gentle-adv

and+pleasant-adv

ʾɛmar

ʾaḇrɔhɔm

ʾaḇrɔhɔm

. . .

say-suf.3.m.sg

pn

pn

. . .

‘On the one hand (man), he (= God) expelled the vultures from the field of Abraham. On the other hand (den), he called out gently and pleasantly to Abraham, “Abraham, Abraham, . . .”’ (5*.10–11) The Syriac man . . . den . . . construction in (9.20) exactly replicates the Greek μέν . . . δέ . . . construction in (9.19). This construction is already attested in the fourth-century Syriac of Ephrem: (9.21) Prose Refutations, Discourse 1 by Ephrem (d. 373; ed. Overbeck 1865) ܿ ‫ ܕܐܢ ܐܢܫ ܢܚܘܣ ܥܠ ܙܪܥܐ ܟܢܝܫܐ ܕ�ܠܐ‬.‫ܗܦܟܬܗ ܕܗܕܐ‬ ‫ܫܡܥ ܬܘܒ‬ ܿ ܿ ܿ ܿ .‫ ܒܗܝ ܕܚܣ ܕ�ܠܐ ܢܒܕܪ‬.‫ܡܣܬܒܪ ܕܥܒܕ‬ ‫ܢ‬ ‫ܡ‬ ‫ܚܟܝ�ܡܐܝܬ‬ . . ‫ܢܒܕܪܝܘܗ ܼܝ‬ ܼ ܿ ‫ܕܢܚܙܝܗ ܠܝܙܦܬܗ ܡܒܕܪܬܐ ܕܐܟܪܐ ܟܕ ܒܩܪܢܐ ܘܪܒܝܬܐ‬ ‫ܐܡܬܝ ܕܝܢ‬ ܿ ‫ܡܟܢܫܐ ܘܦܪܥܐ ܠܗ ܐܪܥܐ ܿܗܝ ܕܝܢ ܡܬܚܙܝܐ ܿܗܝ ܦܪܘܫܘܬܐ ܕܚܣܬ‬ ܿ ܿ ‫ܕ�ܠܐ ܬܒܕܪ ܕܥܘܝܪܘܬܐ ܗܝ‬ šmaʿ

tuḇ

hp̄ɔḵtɔh

dhɔḏe

d(ʾ)en

hear-imp.m.sg

then

opposite-f.sg.con+her

of+this-f.sg

nml+if

ʾɛnɔš

nḥus

man-m.sg.abs

spare-pre.3.m.sg on

ʿal

knišɔ

dlɔ

be-gathered-part.m.sg.emp nml+neg

zarʿɔ seed-m.sg.emp nbaddriw(hy)

ḥakkimɔʾiṯ

scatter-pre.3.m.sg+him

wise-adv

man

mɛstḇar

man

be.considered-part.m.sg.abs nml+do-part.m.sg.abs in+that-f.sg

dḥɔs

dʿɔḇɛḏ dlɔ

nḇaddar

nml+spare-suf.3.m.sg nml+neg scatter-pre.3.m.sg dnɛḥzeh

lizɛp̄ṯeh

nml+see-suf.1.pl+her

to+investment-f.sg.con+his

mḇaddarṯɔ

dʾakkɔrɔ

be.scattered-part.f.sg.emp nml+farmer-m.sg.emp

bhɔy ʾɛmaṯ(y)

den

when

den

kaḏ when

The Syriac Conjunctive Particle den Replicated on Greek δέ

189

bqarnɔ

wrɛbbiṯɔ

mḵannšɔ

in+principal-f.sg.emp

and+interest-f.sg.emp

gather-part.f.sg.abs

wp̄ɔrʿɔ

leh

ʾarʿɔ

and+recompense-part.f.sg.abs to+him land-f.sg.emp mɛtḥazyɔ

hɔy

pɔrošuṯɔ

hɔy den37 then dḥɔsaṯ

be.seen-part.f.sg.abs that-f.sg discernment-f.sg.emp nml+spare-suf.3.f.sg dlɔ

tḇaddar

nml+neg

scatter-pre.3.f.sg nml+blindness-f.sg.emp

daʿwiruṯɔ

(h)y

. . .

she

. . .

‘hear then the opposite of this. If a man spares the gathered seed so as not to scatter it, on the one hand (man), it is thought that he acted wisely in sparing (it) so as not to scatter it; on the other hand (den), when we see the scattered investment of the farmer being collected in capital and interest as well as the earth rewarding him, that discernment which spared (the seed) so as not to scatter it (now) appears to be blindness’ (33.21–27)  37 This example establishes that the man . . . den . . . construction is already attested in Syriac by the fourth century. 38 The construction is rare in this period, however, and does not become common until the fifth century. 9.4.  Late Aramaic Comparanda Most dialects of Late Aramaic do not exhibit a development with *ʾiðayn similar to that witnessed in Syriac den. In Samaritan Aramaic, for example, the reflex of *ʾiðayn, which is written ʾdyn, functions as a clause-initial temporal adverb meaning ‘then’ (DSA 8), as illustrated in the following example: (9.22) ms C of the Samaritan Targum (Late Aramaic; ed. Tal 1980–83) ʾdyn

šry

then

begin-suf.3.m.sg to+call-inf in+name-m.sg.con

lmqry

bšm

yhwh pn

‘Then, (the people) began to call upon the name of the Lord’   (Gen 4:26) 39 37.  The temporal adverb ‫ ܗܝܕܝܢ‬hɔyden is written here as two words (for this, see Payne Smith 1879–1901: 1002). In his Letter on Syriac Orthography, Yaʿqub of Edessa (d. 708) seems to imply that, when written separately, hɔy den is not marked for time (ed. Phillips 1869: 6.12–15). 38.  It also shows that man (< Greek μέν) is attested at least by this time in Syriac; for additional details, see Butts 2013. 39.  The Hebrew Vorlage reads ʾɔz huḥal liqro(ʾ) bšem Yhwh [then begin-suf.3.m.sg for+call‑inf on+name-m.sg.con pn] ‘Then, (people) began to call on the name of the Lord’, with Samaritan Aramaic ʾdyn translating its Hebrew cognate ʾɔz ‘then’. For this verse more broadly, see Fraade 1984, with comments on the Samaritan version at p. 29.

190

Chapter 9

As this example illustrates, the use of ʾdyn in Samaritan Aramaic is similar to that of *ʾiðayn in earlier dialects of Aramaic. Besides Syriac, the only Late Aramaic dialect in which *ʾiðayn may have been replicated on Greek δέ is Christian Palestinian Aramaic. Like Syriac den and Greek δέ, Christian Palestinian Aramaic dy is a conjunctive particle that is restricted to second position: 40 (9.23) Christian Palestinian Aramaic (Late Aramaic; ed. Müller-Kessler and Sokoloff 1998a) ʾzlw

dy

go-suf.3.m.pl then

‘then they went’ (Mark 11:4) 41 Unlike Syriac den, Christian Palestinian Aramaic dy does not have a final nasal. Thus, it is impossible to determine whether dy is a loanword from Greek δέ or another example of the grammatical replication of Aramaic *ʾiðayn on the model of Greek δέ. 42 If the latter is the case, then Christian Palestinian Aramaic dy represents a further step of phonological erosion as compared with Syriac den. 9.5. Conclusion Both Syriac den and Greek δέ are conjunctive particles that occur in second position and mark a change in topic from the first clause to the second clause. Despite the obvious semantic, syntactic, and phonological similarities between the two, Syriac den is not a loanword from Greek but represents an inheritance from Aramaic *ʾiðayn that has been replicated on Greek δέ. This grammatical replication resulted in changes in the syntax, semantics, and phonology of Syriac den. These changes occurred already by the time of the translation of the Old Testament Peshiṭta in the mid-second century. In addition, this grammatical replication resulted in an increase in the frequency of Syriac den from the second century to the third century. Finally, by the fourth century, Syriac den (< *ʾiðayn) occurs with man, a loanword from Greek μέν, in a construction that exactly replicates Greek μέν . . . δέ . . . ‘on the one hand . . . , on the other hand . . .’. This case of grammatical replication is particularly interesting since the identification of Aramaic *ʾiðayn with Greek δέ seems to have been at least partly based on phonology. In addition, grammatical replication 40.  DCPA 85; LSP 44; Müller-Kessler 1991: 148; numerous additional attestations can be found in Müller-Kessler and Sokoloff 1996; 1997; 1998a; 1998b; 1999. 41.  The Greek Vorlage reads καὶ απῆλθον [and go-aor.act.ind.3.pl] ‘and they went’, with καὶ ‘and’ instead of δέ. 42.  For the former interpretation, see Müller-Kessler 1991: 148; DCPA 85. For the latter interpretation, see LSP 44. The latter is also implied in Brock 1996: 258.

The Syriac Conjunctive Particle den Replicated on Greek δέ

191

led to a closer phonological similarity between the two function words, either through grammaticalization or through the transfer of phonology. Thus, this case establishes that, while grammatical replication is primarily related to the semantic-conceptual and results in the transfer of (morpho-)syntactic material, it is not entirely removed from phonology. Rather, in this case, phonology played a key role: it facilitated the grammatical replication and may have even been transferred in the replication process. 9.6.  Excursus: Syriac ger and Greek γάρ It is impossible to discuss the replication of Syriac den on the model of Greek δέ without mentioning Syriac ger ‘truly, indeed’ and Greek γάρ ‘for’. Both of these are conjunctive particles that occur in second position and introduce a reason or a cause or more generally strengthen a proposition. 43 This is illustrated in the following example: (9.24) Hebrew Bible kol

ʾăšɛr- to(ʾ)məri

ʾɛʿĕśɛ-

lɔḵ

all-m.sg

rel

do-pre.1.sg

to+you-f.sg

say-pre.2.f.sg

ki

yoḏeaʿ

for

know-part.m.sg all-m.sg.con

kɔl-

šaʿar

ʿammi

ki

ʾešɛṯ

ḥayil

ʾɔt

that

woman-f.sg.con

strength-m.sg

you-f.sg

gate-m.sg people-m.sg.con+my

‘whatever you say, I will do for you, for all of the assembly of my people know that you are a woman of strength’ (Ruth 3:11) (9.25) Greek Septuagint πάντα

ὅσα

ἐὰν

εἴπῃς

all-acc.n.pl

as.great.as-acc.n.pl

if

say-aor.act.subj.2.sg γὰρ

ποιήσω

σοι

do-fut.act.ind.1.sg

you-dat.sg know-perf.act.ind.3.sg

πᾶσα

φυλὴ

οἶδεν λαοῦ

μου

gar ὅτι

all-nom.f.sg tribe-nom.f.sg people-gen.m.sg my-gen

that

γυνὴ

σύ

δυνάμεως

εἶ

woman-nom.f.sg power-gen.f.sg be-pres.act.ind.2.sg you-nom.sg

‘whatever you say, I will do for you, for all of the assembly of my people know that you are a woman of strength’ (Ruth 3:11) 43.  For Syriac ger, see LS2 114; SL 230. For Greek γάρ, see Denniston 1996: 56–114; Humbert 1960: §§689–96; LSJ 338–39.

192

Chapter 9

(9.26) Old Testament Peshiṭta (latter half of 2nd cent.) ܿ ‫ܟܠܗ ܫܪܒܬܐ ܕܥܡܢ ܕܐܢܬܬܐ‬ ‫ܟܠ ܕܬܐܡܪܝܢ ܠܝ ܐܥܒܕ ܠܟܝ ܝܕܥܐ ܓܝܪ‬ ‫ܐܢܬܝ ܕܥܘܫܢܐ‬ kol

dṯe(ʾ)mrin

all-m.sg.abs

nml+say-pre.2.f.sg to+me

li

ʾɛʿbɛḏ do-pre.1.sg

ger kollɔh

lɛḵ

yɔḏʿɔ

to+you-f.sg

know-part.f.sg.abs ger all-m.sg.con+her

šarbṯɔ

dʿamman

dʾa(n)ttṯɔ

tribe-f.sg.emp

nml+people-m.sg.con+our

nml+woman-f.sg.emp

ʾa(n)tt(y)

dʿušnɔ

you-f.sg

nml+strength-m.sg.emp

‘whatever you say to me, I will do for you, for all of the tribe of our people know that you are a woman of strength’ (Ruth 3:11) The Greek in (9.25) and the Syriac in (9.26) are independent translations of the Hebrew passage in (9.24). Thus, it is noteworthy that Syriac ger and Greek γάρ both occur in second position and that both introduce the second clause, which gives the reason for the first clause. Thus, like Syriac den and Greek δέ, Syriac ger and Greek γάρ share phonological, syntactic, and semantic similarities. The question here is whether Syriac ger is a loanword from Greek γάρ or whether Syriac ger is an inheritance from earlier Aramaic that has been replicated on Greek γάρ, just as Syriac den was replicated on Greek δέ. Both opinions are found in the literature. 44 Unfortunately, the wealth of data that are available for tracking the development of *ʾiðayn to Syriac den is lacking for Syriac ger. There is, in fact, no evidence for its pre-Syriac history in Aramaic. This prima facie makes the interpretation of Syriac ger as a loanword appealing. Nevertheless, there are two arguments in favor of analyzing Syriac ger as the replication of earlier Aramaic material on Greek γάρ. First, the representation of Greek α by Syriac y would be quite unusual. 45 It should be noted, however, that two alternative orthographies occur: 44.  For the former, see, e.g., Brock 1967: 423; 1975: 89; for the latter, see, e.g., Brockel­ mann 1981: §53; Brock 1996: 258; Ciancaglini 2008: 6. 45.  The only possible comparandum for this representation is ἀγών (PGL 25; LSJ 18– 19) > ‫ ܐܝܓܘܢܐ‬ʾygwnʾ ‘struggle’ (LS2 4; SL 6). According to Brockelmann (1908: §94r), this is due to the dissimilation of the vowel in the initial syllable to e before the back vowel. It should be noted that the expected spelling ‫ ܐܓܘܢܐ‬ʾgwnʾ occurs much more commonly, especially later.

The Syriac Conjunctive Particle den Replicated on Greek δέ

193

(9.27) a. ‫ ܓܪ‬gr in Old Syriac Gospels, Luke 18:25 (Sinaiticus) (ed. Kiraz 1996; see also Brock 1996: 258 n. 28) b. ‫ ܓܐܪ‬gʾr in Aphrahaṭ’s Demonstrations, variant at 368.17 (ed. Parisot 1894–1907) Nevertheless, these (early) alternative orthographies are exceedingly rare. Thus, the standard orthography of Syriac ger with medial y provides a strong argument against the loanword hypothesis. Second, though earlier Aramaic evidence for Syriac ger is lacking, there is a potential cognate in Arabic jayri, rarely jayra. 46 This particle occurs in clause-initial position and can be glossed as ‘verily, truly; yes’, as in the following example: (9.28) Classical Arabic jayri

lā

ʾafʿalu

truly

neg

do-pre.1.sg this-m.sg

ðālika

‘Truly, I will not do this’ (Lane 1863–93: 493) The phonological correspondence between Arabic jairi and Syriac ger is entirely regular, 47 and thus they could well be cognate. Together, these two pieces of evidence suggest that Syriac ger is not a loanword from Greek γάρ but that it is an inheritance from earlier Aramaic that has been replicated on Greek γάρ, just as Syriac den was replicated on Greek δέ. In this case, the interlingual identification of Syriac ger with Greek γάρ would have been due to the fact that both are function words with an overlapping use of strengthening a proposition. In addition, their phonological similarity would have facilitated their identification, again as in the case of Syriac den and Greek δέ. After having been identified with one another, Syriac ger was replicated on Greek γάρ leading to the movement of Syriac ger to second position as well as to the new use of Syriac ger to introduce a reason or a cause. The only Late Aramaic dialect that attests a cognate to Syriac ger is Christian Palestinian Aramaic. 48 Like Syriac ger and Greek γάρ, Christian Palestinian Aramaic g(y)r is a second-position particle that introduces a reason or a cause or more generally strengthens a proposition, 49 as is illustrated in the following example: 46.  BK 361; Lane 1863–93: 493; Wright 1896–98: 286B. 47.  For the monophthongization of the diphthong in Syriac, compare *bayt > beṯ ‘house of’. 48.  The form gyr ‘for’ occurs in Tg. Proverbs 29:19 (Jastrow 241). While this text is written in Late Jewish Literary Aramaic, it is a translation from Syriac. Thus, gyr here is to be explained as a loanword from the Syriac Vorlage. 49.  DCPA 76; LSP 44; Müller-Kessler 1991: 148; numerous additional attestations can be found in Müller-Kessler and Sokoloff 1996; 1997; 1998a; 1998b; 1999.

194

Chapter 9

(9.29) Christian Palestinian Aramaic (Late Aramaic; ed. Müller-Kessler and Sokoloff 1997) nṭyr

gr

hwʾ lmrʾ

be.kept-part.m.sg.abs

gr

he

to+lord-m.sg.emp

‘For, it was kept for the Lord’ (Exod 12:42) 50 Christian Palestinian Aramaic g(y)r is probably an inheritance from earlier Aramaic that has been replicated on Greek γάρ, since it is otherwise difficult to explain the orthography with medial y. 51 Thus, both Christian Palestinian Aramaic g(y)r and Syriac ger, while cognate with Arabic jairi, occur in second position and can introduce a reason or a cause due to their replication on Greek γάρ. The case of Syriac ger and Christian Palestinian Aramaic g(y)r thus adds an additional example in which phonological similarity led to interlingual identification and ultimately to grammatical replication. 50. The Greek Vorlage reads νυκτὸς προφυλακή ἐστιν τῷ κυρίῳ [night-gen.f.sg vigil‑nom.f.sg be-pres.act.ind.3.sg art-dat.m.sg lord-dat.m.sg] ‘it was a vigil of the night for the Lord’, without γάρ. 51. In this context, it should be noted that Brock (1996: 258) attributes the spelling without y to a secondary adaptation of the Christian Palestinian Aramaic form to the Greek spelling.

Chapter 10

Conclusion . . . le grand problème, le problème éternel, celui de la symbiose et de l’interpénétration de la civilisation occidentale, c’est-à-dire grecque, et de la culture orientale dans le Proche Orient, un processus qui commença après la conquête de l’Orient par Alexandre, et qui continue toujours. . . . (Rostovtzeff 1943: 44–45)

10.1. Overview In the introduction (chap. 1), I noted that this book is located at the intersection of the field of contact linguistics and the study of ancient languages. I also suggested that each of these fields can and should inform the study of the other. In the first two sections of this conclusion, I aim to substantiate this claim. Section 10.2 illustrates how Syriac data can add to ongoing debates in the field of contact linguistics concerning the transfer of structure in situations of borrowing. Section 10.3 shows that analyzing Syriac data within a contact linguistic framework can help to answer questions about when Syriac speakers first had intense contact with the Greek language. After these two sections, I conclude with a discussion of the Greco-Roman context of Syriac (§10.4). 10.2.  The Transfer of Structure in Situations of Borrowing The transfer of structure has a long and contested history in the field of contact linguistics. 1 In this section, I focus on one particular aspect of this question that has been raised several times in recent scholarship: can structure be borrowed? Note that borrowed here refers to the technical sense established in chap.  2—that is, transferred in situations in which the agents of change are linguistically dominant in the recipient language. Most contact linguists agree that structure can be transferred in cases of imposition (source language 1.  For a recent overview of what can be transferred in language contact, see Curnow 2001.

195

196

Chapter 10

agentivity), in which the agents of change are linguistically dominant in the source language, or in Thomason and Kaufman’s terms, in language shift. It remains an open question, however, whether or not structure can be transferred in borrowing, in which the agents of change are linguistically dominant in the recipient language. On the affirmative side of this question are Thomason and Kaufman, who in their borrowing scale have categories ranging from slight structural borrowing to heavy structural borrowing. 2 Similarly, Van Coetsem allows for an extended mode of borrowing in which phonological and grammatical material can be transferred alongside lexical material. 3 There are, however, a number of linguists who restrict what can be transferred in situations of borrowing. In a study of contact-induced changes in Prince Edward Island French due to English, for instance, King (2000) argues that the seeming cases of grammatical borrowing were not due to the direct transfer of grammatical structure but were the result of the transfer of lexical items. 4 Based on this, she calls into question whether grammatical structure can be transferred in situations of borrowing. 5 King concludes her study by stating that “[i]t is expected that in other case studies of language contact in which structural borrowing seems superficially to have occurred, it will also be discovered that the actual path of change has instead involved core lexical borrowing followed by reanalysis” (King 2000: 176). This is but one example in which structural borrowing has been questioned in the literature. A discussion of the issue, with similar conclusions, can be found in Winford’s textbook on contact linguistics (Winford 2003: 61–100). 6 At least part of the disagreement over whether or not structure can be transferred in situations of borrowing stems from the fact that many of the purported cases of structural borrowing in the contact linguistic literature are based on insufficient data and lack adequate analysis. Poplack, in particular, has drawn attention to this, noting that, “[i]n theory, the view that anything can be borrowed under the right circumstances seems uncontroversial. But in practice, when an apparent case of convergence is pursued scientifically, it often disappears” 2.  Thomason and Kaufman 1988: 74–109. See also Thomason 2001: 69–71. 3.  Van Coetsem 2000: 215–36; 2003: 86–88. 4.  A summary article is available in King 2005. 5.  King does not explicitly make a distinction between situations of borrowing (recipient language agentivity) versus imposition (source language agentivity). Nevertheless, she does note that three speakers in her corpus are “not fluent in French” and often transfer “elements or structures” from English into French (King 2000: 89, 175–76). This prompts her to restrict her general conclusion that there is no evidence for the transfer of grammatical structure to “fluent speakers of French,” excluding the three “non-fluent speakers” from this conclusion (King 2000: 175–76). King thus implicitly adopts a distinction similar to that of borrowing (recipient language agentivity) versus imposition (source language agentivity), and it is for this reason that her thesis has been restricted to borrowing here. 6.  For others, see, e.g., Hickey 2010b; Louden 2000: 96; Silva-Corvalán 1995b.

Conclusion

197

(Poplack 1996: 304). The question of structural borrowing is then inextricably tied up with the question of establishing that a given change is in fact contact induced. 7 Poplack and Levey conclude a recent study that stresses this point by stating, “Contact-induced change is not an inevitable, nor possibly even a common, outcome of language contact. Only more accountable analyses of more contact situations will tell. In the interim, the burden of proof is on those who claim that it has occurred” (Poplack and Levey 2010: 412). It is here that an ancient language such as Syriac can be of assistance. The extensive written record of Syriac, which spans more than two millennia, combined with the considerable body of comparative data available for earlier and contemporaneous dialects of Aramaic enable the historical linguist to trace changes, including contact-induced changes, step by step from their prehistory through their completion as well as to establish in many cases whether or not contact played a role in these changes. 8 In this book, I have presented several examples in which structure was transferred in a situation of borrowing (chaps. 7–9). 9 In §7.3, I argued that the Syriac adjectival ending -ɔy became more frequent throughout the history of Syriac due to its identification with the more frequently occurring Greek adjectives. This resulted in its raising from a minor-use pattern to a major-use pattern. In §7.4, I showed that Syriac lwɔṯ ‘toward’ came to be used with the verb ʾmr ‘to say’ due to its identification with Greek πρός ‘toward’, which could be used with various verbs of speech. Thus, Syriac lwɔṯ acquired a new function due to its replication on Greek πρός. Chapter 8 presented a more detailed case in which the Syriac copula ʾiṯaw(hy) ‘he is’ was replicated on the model of Greek ἐστίν ‘he is’. This resulted in the extension of the copulaic use of ʾiṯ to verbless clauses with substantival predicates and in the raising of copulaic ʾiṯ from a minor-use pattern to a major-use pattern throughout the history of Syriac. Finally, chap.  9 showed how earlier Aramaic *ʾiðayn ‘then, at that time’ was replicated on Greek δέ ‘then, but’ to produce Syriac den ‘then, but’. This replication resulted in changes in the syntax, semantics, and phonology of Syriac den as well as in an increase in its frequency. In each of these examples, an attempt was made to trace systematically the contact-induced changes in 7.  See also King 2000: 46–47 as well as the discussion above at p. 148. 8.  The Syriac situation, thus, contrasts with that usually encountered by the historical linguist. Dorian, for instance, notes that “there will seldom be the ideal breadth and depth of material on which to base an assessment of change in terms of external or internal motivation” (Dorian 1993: 152). Similarly, Poplack and Levey state that “[t]he first step in establishing the existence of change is comparison over time. This may not be simple or straightforward, given the often fragmentary nature of surviving diachronic evidence” (Poplack and Levey 2010: 394). 9.  It was established in §3.4 that contact-induced changes in Syriac due to Greek are to be analyzed as borrowing, in which speakers linguistically dominant in the recipient language, Syriac, transferred features from the source language, Greek.

198

Chapter 10

question with the support of historical data so that a convincing case could be made that language contact played a role in the described changes. In this book, then, I have presented several cases in which structure was transferred in a situation of borrowing. Thus, in answer to the question “Can structure be borrowed?” we can respond “Yes.” This “yes” comes with an immediate caveat, however: the structure transferred in each of the cases investigated is restricted. Several of the cases discussed here do not involve the creation of a new structure but, rather, a change in the distribution of an existing structure. This is the case, for instance, with the increase in the frequency of the adjectival ending -ɔy (§7.3), the increase in the frequency of the copula ʾiṯaw(hy) (§8.5), and the increase in frequency of Syriac den (§9.3). Several other changes discussed did involve the creation of a new grammatical function but only as an extension of an existing grammatical structure. This is the case, for instance, with the extension of Syriac lwɔṯ to verbs of saying (§7.4) and the extension of the copula ʾiṯaw(hy) to substantival predicates (§8.3). In general, then, all of the examples of grammatical replication analyzed in this book differ from the transfer typically witnessed in imposition, in that they are isolated, nonsystematic, and of limited scope. The cases presented in this study are similar to some of the changes in Los Angeles Spanish investigated by Silva-Corvalán (1994). The extension of the Syriac copula to verbless clauses with substantival predicates, for instance, is comparable to a case described by Silva-Corvalán in which Spanish cómo acquired an additional meaning due to its replication on English ‘how’ (1994: 176–77). In the case of both Spanish cómo and Syriac ʾiṯaw(hy), there are no radical changes to the grammatical system of the recipient language but only extensions of an already existing variant. Once more cases such as these are identified and analyzed adequately in the literature, it will be possible to formulate parameters for the way that structure is transferred in situations of borrowing. 10 For now, however, this book has provided several cases in which structure was transferred in a situation of borrowing, though the structure transferred in each case was restricted. 10.3.  The Beginning of Syriac-Greek Language Contact Having illustrated how an ancient language such as Syriac can contribute to the field of contact linguistics (§10.2), I now turn to the way that contact linguistics can inform the study of Syriac. One of the more contested questions in the study of Syriac-Greek language contact revolves around when intense contact between Syriac and Greek began. While it is widely accepted that by the fifth century Syriac authors were influenced by Greek, there is no consensus in the scholarly literature concerning how much earlier this intense contact ex10. For another potential case, see Smits 1999, with the comments of Van Coetsem 2003: 86–87.

Conclusion

199

tended. According to the traditional view, fourth-century Syriac authors such as Aphrahaṭ (fl. 337–45) and Ephrem (d. 373) lived in a purely Semitic (or Aramaic) linguistic and cultural context that had not yet been influenced by Greek. 11 In his classic study of Greek loanwords in Syriac, for instance, Schall states that “Afrahaṭ (schrieb zwischen 337 und 345) war wesentlich frei vom Einfluss des griechischen Geistes” (Schall 1960: 3). Similarly, in an encyclopedia article on Ephrem from the late 1960s, Murray claims that “Ephrem knew no Greek, shows no debt to Greek philosophy, and expresses contempt for Greek thought,” and a little later he adds, “Ephrem is heir to a Judaeo-Christian tradition which developed largely in isolation from the Greek-speaking world” (Murray 1967: 221–22). 12 This traditional view was predominant primarily in the twentieth century, 13 but it continues to be held by some scholars today. In a relatively recent article, for instance, Pat-El argues that the development of a productive copula in Syriac, which was discussed in detail above (chap. 8), was not due to contact with Greek, because this verbless clause pattern is already found in the Syriac of Ephrem who, according to her line of thought, was “among writers who have no knowledge of Greek” (Pat-El 2006: 343). Among the most vocal opponents of this traditional view is H. J. W. Drijvers. 14 Throughout his work, Drijvers maintains that Edessa and the surrounding areas were “thoroughly hellenized” by the turn of the Common Era with widespread Syriac-Greek bilingualism. Other scholars have adopted this position, especially in more recent years. In her recent monograph on Iranian loanwords in Syriac, for instance, Ciancaglini writes, “The area of western Syria and northern Mesopotamia was once part of the Seleucid empire; linguistic and archaeological evidence shows that the area was thoroughly Hellenized from the beginning of the Christian era” (Ciancaglini 2008: 6). Between these two poles of the spectrum, there are a number of intermediate positions. In a recent article, for instance, Healey discusses “The Edessan Milieu and the Birth of Syriac” (Healey 2007). He points out that Syriac began as the local Aramaic dialect of the region around Edessa, being witnessed in the Old Syriac inscriptions and legal documents. The Edessan dialect of 11.  For a synopsis of this traditional view with citations of many representatives, see Shepardson 2008: 65–66 with n. 191; and especially Possekel 1999: 1–7. See also Healey 2007: §5. 12.  In his later work, Murray steps back from this position, conceding that early Syriac authors were influenced by Greek (Murray 1982: 9–10). In the new introduction to the reprint of his Symbols of Church and Kingdom, Murray clarifies that “the homelands of the authors studied in this book would have been mainly Syriac-speaking, though with varying knowledge of Greek” (Murray 2004: 3). 13.  It in fact goes back much earlier. Already in the fifth century, the church historian Sozomen states that Ephrem was ‘ignorant of Greek learning’ (Ἑλληνικῆς παιδείας ἄμοιρος) (Ecclesiastical History, 3.16; ed. Bidez et al. 1983–96: 2.152). 14.  See, e.g., Drijvers 1970; 1980; 1998.

200

Chapter 10

Aramaic, according to Healey, was eventually transformed into a prestigious literary language due to several factors, including its use as an administrative language, as a royal language, and above all, as a religious language. According to Healey, however, one factor that did not play a role in this transformation is Syriac’s interaction with Greek. Healey argues that Greek linguistic influence in Edessa is “mostly connected with Romanization in the third century a.d.” (Healey 2007: 121). Thus, a figure such as Bardaiṣan (154–222) represents only a narrow circle associated with the royal court and is not indicative of more widespread Greco-Roman contact at the time (Healey 2007: 120). In Healey’s words: though Bardaisan may form a prominent peak of Hellenism, it is not clear that he is the tip of an iceberg of any great significance. That there was Greek culture in Edessa is clear, but much more clear is the underlying dominance of native religious and linguistic tradition. (Healey 2007:124)

Thus, Healey emphasizes Syriac’s continuity with earlier Aramaic. Healey concludes his discussion of “The Edessan Milieu and the Birth of Syriac” by stating, “[I]n the formative period the Edessan milieu was not hellenized to any significant extent, while Syriac’s ancestry is to be sought in the local Aramaic dialects of northern Mesopotamia, gradually transformed into a prestige language of religious literature” (Healey 2007: 125). According to Healey, then, Syriac only began to undergo significant Greek influence in the third century, when the Roman Empire expanded into Syria and Mesopotamia. Or, to take another example, a more nuanced adaptation of the traditional view can be found in the more recent works by Brock, where he proposes that Syriac authors from the fourth century had relatively limited contact with Greek but that a major transition occurred in the fifth century, when contact became increasingly more intense: “The earliest major writers, Aphrahaṭ (active 337–345) and Ephrem (d. 373), although far from untouched by the influence of Greek language and culture, are nevertheless comparatively unhellenized in their style and language” (Brock 1996: 253). Thus, for Brock, fourth-century Syriac authors were influenced by Greek—just relatively less so than later authors. 15 15.  This marks a departure from Brock’s earlier work, in which he adopts the traditional view that Syriac authors from the fourth century were basically devoid of Greek influence. In a study from 1975, he states, “Aphrahaṭ was a writer who was virtually untouched by Greek culture, and one can safely assume that he knew little, if any, Greek” (Brock 1975: 81). Similarly, in another study from slightly later, he considers Aphrahaṭ and Ephrem to be “representatives of a Syriac culture that is still essentially semitic in its outlook and thought patterns” (Brock 1982: 17). In a later study, however, he states: “[B]y the fourth century ad, Greek and Semitic cultures had already been interacting in the Middle East for over half a millennium: no Syriac writer of Ephrem’s time is going to be purely Semitic in character or totally unhellenized” (Brock 1992: 143). In a more recent study, Brock concludes that “the fact that Ephrem was evidently heir to a Syriac lexical stock that had already been

Conclusion

201

It is clear, then, that there is no agreement in the scholarly literature over when the period of intense contact between Syriac and Greek began. In this book, however, I have introduced new evidence pertaining to this question from grammatical replication. The previous chapters have presented several cases in which grammatical replication occurred in Syriac already by the second century. Chapter §8 established that by at least the early second century the Syriac copula ʾiṯaw(hy) ‘he is’ had been extended to verbless clauses with substantival predicates on the model of Greek ἐστίν ‘he is’. In addition, I showed in chap.  9 that, already by the time of the translation of the Old Testament Peshiṭta in the mid-second century, Aramaic *ʾiðayn ‘then, at that time’ had been replicated on Greek δέ ‘but, then’ in its syntax, semantics, and phonology to produce Syriac den ‘but, then’. This grammatical replication also resulted in an increase in the frequency of Syriac den from the second century to the third century. Finally, in the excursus in §9.6, I argued that, by the second century, Syriac ger ‘indeed, for’ had been replicated in its syntax and semantics on Greek γάρ ‘indeed, for’. These cases of grammatical replication have significant implications for establishing a terminus ante quem for extensive contact between Syriac speakers and the Greek language. It is well established that for contact-induced changes such as grammatical replication to take place in a situation of borrowing there must be a high degree of bilingualism that extends over a considerable period of time. 16 This is due to the nature of the change itself. In grammatical replication, speakers of the recipient language equate a grammatical structure in their own language (the recipient language) with a grammatical structure in the source language. This necessitates that speakers of the recipient language have a high enough proficiency in the source language to make such structural equations. In the words of Thomason: “You can’t borrow what you don’t know” (Thomason 2010: 41). 17 In addition, this bilingualism must extend for considerably enriched by borrowings from Greek gives support to the view that he was living in a milieu that was already considerably hellenized” (Brock 1999–2000: 449). He adds the caveat, however, that a diachronic perspective still allows one to “characterize the writings of fourth-century Syriac authors as being comparatively unhellenized” (Brock 1999–2000: 449 n. 45; italics mine). For the progression in Brock’s thought on this issue, see Possekel 1999: 5–7. 16.  See, e.g., Heine and Kuteva 2003: 531; 2005: 13; Johanson 2002a: 50; Poplack 1996: 285; Thomason 2010: 37. It should be stressed that this applies only to contact situations of borrowing (recipient language agentivity). In situations of imposition (source language agentivity), these kinds of change can occur as quickly as a generation. Again, it should be noted that that contact-induced changes in Syriac due to Greek are to be analyzed as borrowing, in which speakers linguistically dominant in the recipient language, Syriac, transferred features from the source language, Greek, as argued in §3.4 above. 17.  In less colorful terms, Hickey states, “[I]t is probably true that the borrowing of ‘systemic’ material—inflections, grammatical forms, sentence structures—can only occur via bilinguals” (Hickey 2010b: 8).

202

Chapter 10

at least several generations. In fact, Heine and Kuteva note that in many cases bilingualism lasts for as many as three to five centuries before grammatical replication occurs. 18 Even adopting a more conservative estimate, the cases of grammatical replication that had occurred in Syriac by the second century indicate that there must have been significant Syriac-Greek bilingualism among at least some speakers by at least the turn of the Common Era. Returning then to the question of when intense contact between Syriac and Greek began, the traditional view that rejects intense Syriac-Greek contact before the fifth century is in need of revision. In their language, fourth-century Syriac authors such as Aphrahaṭ and Ephrem must have been heirs to an Aramaic culture that had long been in contact with the Greco-Roman world and its Greek language. In fact, this contact must extend back to the turn of the Common Era. 10.4.  Syriac in Its Greco-Roman Context Recent years have seen growing interest in the Greco-Roman context of Syriac Christianity. This has, for instance, been the primary focus of a number of studies by Millar (see especially 2011a; 2011b; 2012; 2013a; 2013b). Millar’s work has in turn evoked several responses, including recently by Johnson (2015b) and ter Haar Romeny (2015). Each of these scholars, in his own way, has sought to investigate interactions between the Greek- and Syriac-speaking worlds of Late Antiquity. Surprisingly, however, none of these scholars interacts with changes to the Syriac language itself. That is, while these scholars provide ample discussion of the use of Syriac, including especially material remains such as inscriptions and manuscripts, they rarely, if ever, discuss contact-induced changes in Syriac due to Greek. 19 In this book, I have adopted a different approach. By focusing on the Syriac language, I have sought to introduce new—and arguably more-reliable—evidence for locating Syriac Christianity within its Greco-Roman context. Aramaic was in contact with Greek already from the mid-first millennium b.c.e., when the Greek monetary term στατήρ appears in the Abydos Lion Weight (KAI 263). Contact between Aramaic and Greek increased with Alexander’s defeat of Darius III in the 330s b.c.e., which brought Syria and Mesopotamia under the control of the Seleucid Empire for the next two centuries. Thus, by the time that Edessa became a Roman colonia at the beginning of the third century, the Aramaic-speaking inhabitants of Syria and Mesopotamia had already been in contact with the Greco-Roman world and 18.  Heine and Kuteva 2003: 531. Similarly, Poplack concludes that, “All cases of borrowing involving extensive structural change in the borrowing language have a history of several hundred years of contact” (Poplack 1996: 305–6). 19.  Johnson even goes so far as to suggest that contact linguistics does not provide a viable method for investigating a case of ancient language contact such as Syriac and Greek (see n. 27 in chap. 1 above).

Conclusion

203

its Greek language for more than half a millennium. The effects of this contact are witnessed in the more than 60 Greek loanwords that were transferred into Aramaic prior to Syriac and then inherited in Syriac. 20 These words include typical Greco-Roman cultural terms such as στρατηγός (LSJ 1652) > ‫ ܣܛܪܛܓܐ‬sṭrṭgʾ ‘strategos’ (LS2 34, 469; SL 71, 998) and ἐπίτροπος (LSJ 669) > ‫ ܐܦܛܪܘܦܐ‬ʾpṭrwpʾ ‘prefect; manager’ (LS2 40; SL 86). There are also nouns belonging to more-abstract semantic groups, however, such as γένος (LSJ 344) > ‫ ܓܢܣܐ‬gnsʾ ‘kind, species; family; race, nation’ (LS2 125, 179, 841; SL 179, 249), κίνδυνος (LSJ 952) > ‫ ܩܝܢܕܘܢܘܣ‬qyndwnws ‘danger’ (LS2 676; SL 1363–64), and χρῶμα (LSJ 2012) > ‫ ܟܪܘ�ܡܐ‬krwmʾ ‘color; nature’ (LS2 347; SL 648). In addition, several Greek verbs were inherited in Syriac from earlier Aramaic. The Greek loanwords that were inherited in Syriac from earlier Aramaic point to more than casual contact between Greek and pre-Syriac Aramaic already before the Roman period. In the first centuries c.e., the Roman Empire expanded eastward with the region of Osrhoene and the important Syriac-speaking center of Edessa coming under greater Roman influence in the mid-second century. The earliest Syriac texts, such as the Old Testament Peshiṭta (translated from Hebrew), the Odes of Solomon (ca. 2nd cent.), and the Book of the Laws of the Countries (ca. 220), stem from this period. 21 These texts already show signs of significant contact with the Greco-Roman world and its Greek language. The Book of the Laws of the Countries, for instance, contains 25 different Greek loanwords that occur a total of 114 times. 22 This means that 5.26% of the noun types and 5.35% of the noun tokens in this text are Greek loanwords. This can be compared with 4.68% of the noun types and 2.84% of the noun tokens that are Greek loanwords in Discourse 1 of the Prose Refutations by Ephrem (d. 373). 23 Thus, the early third-century Book of the Laws of the Countries contains a higher concentration of Greek loanwords than the equally philosophical Prose Refutations by Ephrem, which stems from the latter half of the fourth century. This illustrates the degree of contact between Greek and Syriac already by the second century. 20.  These are collected in appendix 2 and are discussed more generally in §4.9. 21.  The Old Testament Peshiṭta is in the process of being reedited under the auspices of the Leiden Peshiṭta Institute (for this text, see also the important monograph by Weitzman 1999). The Odes of Solomon are edited with English translation in Charlesworth 1973; a facsimile edition of the witnesses is also available in Charlesworth 1981. It should be noted that both the date (first–third centuries) and the original language (Syriac or Greek) of the Odes continues to be disputed. The Book of the Laws of the Countries is edited with English translation in Drijvers 1965. 22.  The total token count of this text is 7,420. For an earlier treatment of the Greek loanwords in this text, see Schall 1960: 71–80. 23.  The text is edited in Overbeck 1865: 21–58. The total token count of this text is 9,322.

204

Chapter 10

The effects of language contact by at least the first centuries of the Common Era are not limited to loanwords but also extend to changes such as grammatical replication. Already by the time of the Peshiṭta Pentateuch (ca. 150), for instance, Aramaic *ʾiðayn ‘then, at that time’ had been replicated on Greek δέ ‘but, then’ in its syntax, semantics, and phonology to produce Syriac den ‘but, then’. By the time of the Book of the Laws of the Countries, Syriac den had also become more frequent due to its replication on Greek δέ. 24 Or, to take a different example, the Syriac copula ʾiṯaw(hy) ‘he is’ is attested with a substantival predicate already in the Odes of Solomon (20:1). This is the result of an extension that occurred on the model of Greek ἐστίν ‘he is’. 25 As argued in §10.3, these cases of grammatical replication indicate that there must have been significant Syriac-Greek bilingualism by at least the turn of the Common Era. During the first centuries of the Common Era, then, the Aramaic dialect of Edessa was changing due to contact with Greek. In particular, Greek loanwords were augmenting the native Aramaic vocabulary, and native Aramaic material was being adapted to replicate constructions in Greek. Thus, fourthcentury Syriac authors such as Aphrahaṭ (fl. 337–45) and Ephrem (d.  373) were heirs to an Aramaic language that had already been significantly changed by the Greek language. It has often been noted that all, or almost all, of the Greek loanwords in Aphrahaṭ are also found in the Syriac Bible. 26 It is usually concluded from this that Aphrahaṭ, who lived in the Sassanian Empire, adopted these words from the Bible. A different conclusion is possible, however, in light of the scenario being proposed here. The fact that words of ultimate Greek origin occur in the Syriac Bible and in early Syriac literature could well suggest that these words were already part of the Syriac language by Aphrahaṭ’s time. This seems to be the case with Greek loanwords in the Syriac Old Testament Peshiṭta, which was translated from Hebrew not from Greek. 27 This may also be the case with the Old Syriac Gospels, which are much less tied to their Greek Vorlage than even the fourth-century Peshiṭta translation. 28 Similarly, 24.  These changes involving Syriac den (< *ʾiðayn) were analyzed in chap. 9 above. 25.  This was discussed in detail in §8.3. 26.  See, e.g., Brock 1967: 390; 1975: 81; Haefeli 1932: 190; Schall 1960: 87. 27.  There is not yet an exhaustive treatment of the Greek loanwords in the Syriac Old Testament Peshiṭta. See, however, the initial remarks in Joosten 1998. 28.  In the preface to the 2nd edition of his grammar, Nöldeke states, The Syriac Bible has been more largely drawn upon than in the former edition, particularly as regards the Gospels, and especially the Synoptic Gospels. These last exhibit almost invariably an exceedingly flowing, idiomatic style of Syriac, which upon the whole reads better than the Semitic Greek of the original. This feature comes into still stronger relief in the more ancient form of the text—as contained in C. (Curetonianus) and S. (Sinaiticus)— than in our usual text P. (Peshitā). (Nöldeke 1904: xiii)

For the Greek loanwords in the Old Syriac and Peshiṭta Gospels, see Brock 1967.

Conclusion

205

each of the Greek loanwords in Aphrahaṭ could already have been part of Syriac by at least the fourth century. 29 Contact between Greek and Syriac was not restricted to the early centuries of the Common Era but continued and even increased throughout the period of Classical Syriac. By comparing loanwords in Ephrem (d. 373) and Narsai (d. ca. 500), Brock has convincingly shown that Greek-Syriac contact became more intense in the century after the death of Ephrem. Similar conclusions can be drawn from the Greek loanwords found in other texts. In the first seven acts of the Acts of Thomas (ca. 200–250), for instance, 5.03% of the noun types and 2.52% of the noun tokens are of Greek origin. 30 This can be compared with the Life of Rabbula (ca. 450), where Greek loanwords account for 6.59% of the noun types and 3.37% of the noun tokens. 31 This demonstrates an increase in Greek loanwords from the third to the fifth century. Moving even later in time, 10.47% of the noun types and 6.00% of the noun tokens are of Greek origin in the mid-sixth-century Life of Yuḥanon of Tella by Eliya. 32 This demonstrates the increasingly intense contact between Syriac and Greek throughout the period of Classical Syriac. Similar conclusions can be reached based on the number of Greek particles and Greek verbs that entered Syriac during the sixth century. 33 The picture provided by loanwords can be corroborated by two of the cases of grammatical replication presented in this study. In §7.3, I showed that throughout the history of Syriac the adjectival ending -ɔy became increasingly more frequent as Syriac speakers attempted to replicate Greek adjectives. There was, for instance, a 460% increase in the frequency of nisba adjectives from the Book of the Laws of the Countries (ca. 220) to the selected Letters of Yaʿqub of Edessa (d. 708). This demonstrates an increase in Syriac-Greek 29.  Brock hints at a similar conclusion when he states that “the vast majority of Greek words to be found in the Old Syriac and Peshiṭta Gospels became well established in the literary language, and it is very likely that many of them were already so” (Brock 1967: 426; italics mine). He goes on to state, however, that “there is hardly any surviving evidence for this” (1967: 426). 30.  The Syriac text of the Acts of Thomas is edited in Wright 1871: 2.171–333 (Syr.). The total token count of this sample is 15,721. The Acts of Thomas were in all likelihood composed in Syriac (Attridge 1990). The date of composition is most likely the first half of the third century (Bremmer 2001b: 73–77). The Syriac original was translated into Greek at an early date (the Greek text is edited in Bonnet 1903: 99–291). The content of the Syriac text that is now extant shows signs of revision, often bringing it more in line with the emerging orthodoxy. The language of the Syriac text, however, contains a number of early forms (Wright 1871a: 2.xiv–xv) that indicate that the language belongs to the earliest period of Syriac. 31.  The text is edited in Overbeck 1865: 159–209. The total token count of this text is 12,000. 32.  The text is edited in Brooks 1907: 29–95. The total token count of this text is 15,815. 33.  See §6.4 and §6.3, respectively.

206

Chapter 10

contact from the early third century to the beginning of the eighth century. The distribution of verbless clauses points to a similar conclusion. I argued in §8.5 that Syriac verbless clauses with a copula of ʾiṯ plus pronominal suffix became increasingly more common throughout the history of Syriac due to their identification with Greek verbless clauses with the verbal copula ἐστίν. In the selections from the fourth-century authors Aphrahaṭ and Ephrem, for instance, less than 20% of the verbless clauses with substantival predicates are formed with ʾiṯaw(hy). In contrast, ʾiṯaw(hy) occurs in just fewer than 40% of the verbless clauses with substantival predicates in the selections from the sixth-century authors Philoxenos (d. 523) and Shemʿun of Beth Arsham (d. before 548). Finally, almost all (98%) of the verbless clauses with a substantival predicate are formed with ʾiṯaw(hy) in the selection from Yaʿqub of Edessa (d. 708). This again illustrates that contact between Syriac and Greek continued until at least the beginning of the eighth century. Syriac was, then, in contact with Greek for centuries. One of the many interesting aspects of this continuity of contact is that it enabled contact-induced changes to extend over generations of speakers. This is perhaps most obvious in the cases of grammatical replication involving the increase in the adjectival ending ‑ɔy (§7.3) and the copula ʾiṯaw(hy) (§8.5), which were mentioned in the previous paragraph. The dynamic nature of contact between Syriac and Greek is, however, also witnessed in the Greek loanwords in Syriac. Since a number of Syriac speakers knew Greek to one degree or another, some Greek loanwords in Syriac never became entirely disassociated from their Greek source. This can be seen, for instance, in the diachronic changes to the orthography of Greek loanwords in Syriac. In contrast to what is generally witnessed cross-linguistically, Greek loanwords did not always become more integrated in Syriac over time. Rather, in a number of cases, the opposite in fact occurred, and Greek loanwords in Syriac came to represent the Greek source more closely over time. In these cases, some Syriac speakers never lost sight of the Greek origin of certain loanwords and were thus able to reshape them based on the source language. This trend reaches its apex with the bilingual Yaʿqub of Edessa (d. 708), who in his Letter on Syriac Orthography, uses a mater lectionis in Syriac to represent every vowel in Greek loanwords. Contact between Greek and Syriac can be analyzed not only diachronically but also comparatively. That is, locating Syriac among its sister dialects of Jewish Palestinian Aramaic, Christian Palestinian Aramaic, and Samaritan Aramaic in the Levant and Jewish Babylonian Aramaic and Mandaic in Mesopotamia provides another vantage point by which to evaluate the GrecoRoman context of Syriac. Comparing the concentration of Greek loanwords in each of the Late Aramaic dialects, for instance, shows that Syriac underwent particularly intense contact with Greek. Greek loanwords in Jewish Babylo-

Conclusion

207

nian Aramaic and Mandaic are much rarer than they are in Syriac. 34 In fact, one can read page after page of text in both of these dialects without encountering a single Greek loanword, a feat that is simply impossible for Syriac. In addition, the few Greek loanwords that do occur in these two dialects are often inheritances from early stages of Aramaic. 35 The scarcity of Greek loanwords in Jewish Babylonian Aramaic and Mandaic, especially when compared with Syriac, is not surprising since these two dialects are located much farther east in Mesopotamia, where they would have had significantly less contact with the Greco-Roman world. In contrast to Jewish Babylonian Aramaic and Mandaic in Mesopotamia, the Late Aramaic dialects located farther to the west contain numerous Greek loanwords. Among the Late Aramaic dialects located in the Levant, Samaritan Aramaic contains the lowest concentration of Greek loanwords. In a sample of prose texts from the Tibat Marqe, for instance, only 0.28% of the noun types are of Greek origin. 36 Similar low numbers are found in poetry written in Samaritan Aramaic: in selected Piyyuṭim by Marqe, for instance, only 0.37% of the noun types are of Greek origin. 37 These numbers from Samaritan Aramaic can be compared with a prose and poetry selection from Ephrem. In Discourse 1 of Ephrem’s Prose Refutations, 4.68% of the noun types are of Greek origin. 38 Fewer Greek loanwords are found in Ephrem’s poetry, but still 2.74% of the noun types are of Greek origin in a selection from two collections of Ephrem’s maḏrɔše. 39 Thus, Greek loanwords are significantly more common in Ephrem than in Samaritan Aramaic from roughly the same time period. A higher concentration of Greek loanwords can be found in Jewish Palestinian Aramaic. In a selection of Piyyuṭim, for instance, 4.94% of the noun types and 3.87% of the noun tokens are of Greek origin. 40 These numbers are 34.  For Jewish Babylonian Aramaic, see Sokoloff 2011: 669; for Mandaic, see Burtea 2011: 682; Müller-Kessler 2004: 57; Voigt 2007: 150. 35.  For this phenomenon, see §4.9 and appendix 1. 36.  This sample consists of §§1–40 of Book 1 of the Tibat Marqe (ed. Ben-Ḥayyim 1988) and contains 5,023 total tokens. Hebrew material was excluded from the analysis. 37.  This sample consists of Marqe 1, incipit ʾdyq ʿlynn mrn, ed. Ben-Ḥayyim 1967: 133–46; Marqe 2, incipit ʿbwdh dʿlmh, ed. Ben-Ḥayyim 1967: 146–53; Marqe 3, incipit ʾthw ʾlhynw, ed. Ben-Ḥayyim 1967: 154–59; Marqe 4, incipit ʾthw ʾlhnn, ed. Ben-Ḥayyim 1967: 159–65; Marqe 14, incipit hʾ sbyʿn npšhtn, ed. Ben-Ḥayyim 1967: 214–23. The total token count of this sample is 2,245. 38.  The text is edited in Overbeck 1865: 21–58. The total token count of this sample is 9,322. 39.  This sample consists of Against Julian the Apostate (ed. Beck 1957b: 71–91) and Hymns on Nisibis, 13–21 (ed. Beck 1961a: 34–59). The total token count of this sample is 5,477. 40.  This sample consists of Poems 1–2, 4–5, 14–15, 34–40 in Yahalom and Sokoloff 1999. The total token count of this sample is 2,119.

208

Chapter 10

higher than those found in the selection of Ephrem’s poetry mentioned above, in which 2.74% of the noun types and 1.95% of the noun tokens are of Greek origin. They are in fact very similar to the concentration of Greek loanwords in the poetry of Narsai: 5.52% of the noun types and 3.48% of the noun tokens are of Greek origin in a sample of two memre by Narsai. 41 It should be pointed out, however, that the Jewish Palestinian Aramaic Piyyuṭim were written after Narsai’s time, which suggests that Syriac may well have undergone more intense contact with Greek than Jewish Palestinian Aramaic. An even higher concentration of Greek loanwords can be found in at least some prose texts written in Jewish Palestinian Aramaic. In a sample from Genesis Rabbah, for instance, 13.70% of the noun types and 8.82% of the noun tokens are of Greek origin. 42 This concentration of Greek loanwords is higher than that of most Syriac texts; in the Life of Yuḥanon of Tella mentioned above, for instance, 10.47% of the noun types and 6.00% of the noun tokens are of Greek origin. The Greek language, however, is highly thematized in Genesis Rabbah. Consider the following passage, which makes a Greek word play on the place-name Cappadocia: A man came to R. Yose b. Ḥalafta and said to him, “It was said to that man (i.e., me) in a dream: Go, take the property of your father from Cappadocia!” [R. Yose] said to him, “Did your father ever (lit., in his days) go to Cappadocia?” He said, “No.” [R. Yose] said to him, “Go, count twenty beams on the ceiling of your house, and you will find it.” He said to him, “There are not twenty (beams) on it.” [R. Yose] said to him, “If there are not twenty (beams) on it, count from their beginning to their end and from their end back to their beginning. Where you find twenty, you will find it.” He went and did (this), and he found (it). From where had R. Yose b. Ḥalafta learned this? From Cappadocia (qpdwqyh = κάππα δόκια ‘20 beams’). (Gen. Rab. 68:12; Theodor and Albeck 1912–36: 784; translated from the text edited by Kutscher apud Rosenthal 1967: 1/1.63) 43

Such a thematization of the Greek language suggests that many of the Greek words in Genesis Rabbah may not be established Lehnwörter but, rather, Fremdwörter employed for particular rhetorical and literary purposes. Regard41.  The two memre are On the Translation of Enoch and Elijah (ed. Frishman 1992: 3–20) and On the Tower of Babel and the Division of Tongues (Frishman 1992: 69–86). The total token count of this sample is 9,221. 42.  This sample consists of most of the longer Aramaic passages of Genesis Rabbah, according to ms Vat. Ebr. 30, as edited in Kutscher’s contribution to Rosenthal 1967: 1/1.59– 67. The total token count of this sample is 2,436, not including the Hebrew passages that are occasionally provided by Kutscher. 43.  The same story, though with minor differences, occurs in y. Maʿaś. Šen. 4.6, 55b as well as in Lam. Rab. 1:1 (ed. Buber 1899: 54.10–55.1). For a comparison of the various Palestinian versions, see Mandel 2000: 84–89. The story is also found in b. Ber. 56b, where the etymology is presented as part Aramaic (qwpʾ ‘beam’) and part Greek (δέκα ‘ten’; a corruption according to Alexander 1995: 240–41). For further discussion, see Hasan-Rokem 2000: 88–107; Smelik 2013: 19.

Conclusion

209

less, Jewish Palestinian Aramaic clearly contains a large number of Greek loanwords, reflecting its setting in the multilingual world of ancient Palestine, where Hebrew, Aramaic, and Greek (all in various dialects) were an integral part of the linguistic landscape. 44 The Greek loanwords in Syriac can finally be compared with those in Christian Palestinian Aramaic. While it is widely acknowledged that Christian Palestinian Aramaic was significantly influenced by Greek, this case of language contact has received only passing reference in the secondary literature. One of the difficulties in studying contact-induced changes in Christian Palestinian Aramaic due to Greek is that all of the extant Christian Palestinian Aramaic texts, with the exception of the inscriptions and (some of) the amulets, were translated from Greek. Thus, it is often difficult, if not impossible, to establish that a particular lexeme or grammatical feature is due to a change in the language itself and not a superficial result of translation. Notwithstanding this limitation, a large number of Greek words occur in Christian Palestinian Aramaic. In a sample of Christian Palestinian Aramaic prose texts, for instance, 13.15% of the noun types and 8.37% of the noun tokens are of Greek origin. 45 These numbers are very similar to those found in the sample from the Jewish Palestinian Aramaic Genesis Rabbah mentioned above and surpass those for the Syriac Life of Yuḥanon of Tella. It must again be stressed, however, that unlike the Syriac texts treated in this book, these Christian Palestinian Aramaic texts are translations from Greek, which may at least partially explain the high concentration of Greek loanwords in them. To summarize these comparisons with Late Aramaic, Syriac contains a much higher concentration of Greek loanwords than Jewish Babylonian Aramaic and Mandaic in Mesopotamia as well as Samaritan Aramaic in the Levant. In contrast, a similarly high—and even higher—concentration of Greek loanwords can be found in Jewish Palestinian Aramaic and Christian Palestinian Aramaic. Some caveats are necessary, however. Genesis Rabbah indeed contains a high concentration of Greek loanwords, higher even than a Syriac text such as the Life of Yuḥanon of Tella, which itself has a high concentration of Greek loanwords by Syriac standards. The Greek language is highly thematized in Genesis Rabbah, however, and this may explain the large and diverse number of Greek loanwords found in it. The Greek words in the Jewish Palestinian Aramaic Piyyuṭim are more likely to be true Lehnwörter, and thus they may provide a more accurate indication of the degree of contact between 44.  There is a significant body of scholarly literature on the presence of Greek in this time and place; see, for instance, the classic studies of Lieberman 1942 and 1950; or the more recent remarks in Fraade 2010; 2014; Hezser 2001: 227–50; Sperber 2006. 45.  This sample consists of the early texts edited in Müller-Kessler and Sokoloff 1996, which includes The Forty Martyrs of the Sinai Desert, Eulogios the Stone-Cutter, and Ana­ stasia. The total token count of this sample is 7,399.

210

Chapter 10

this dialect of Aramaic and Greek. 46 The concentration of Greek loanwords in these poems is comparable with the concentration found in Narsai, even though the Jewish Palestinian Aramaic Piyyuṭim stem from a later date. This suggests that Syriac may well have had a more intense period of contact with Greek than Jewish Palestinian Aramaic, even if the Greek language is highly thematized in some Jewish Palestinian Aramaic literature. Greek loanwords also occur frequently in Christian Palestinian Aramaic. These texts are translations from Greek, however, which may account for the large number of Greek loanwords found in them. Nevertheless, it is clear that this dialect of Aramaic underwent an intense period of contact with Greek similar to that experienced by Syriac. This analysis receives further support from the cases of grammatical replication presented in this book. In §§8.3–8.4, I argued that the extension of the copulaic use of *ʾīθay in Syriac to verbless clauses with substantival predicates was due to contact with Greek. A similar extension is also found in Christian Palestinian Aramaic. None of the other dialects of Late Aramaic, however, attests this development. Similarly, in chap. 9, I illustrated how earlier Aramaic *ʾiðayn was replicated on Greek δέ to produce Syriac den. This change did not occur in Mandaic and Jewish Babylonian Aramaic in Mesopotamia or in Samaritan Aramaic and Jewish Palestinian Aramaic in the Levant. The only dialect that may attest a similar development is again Christian Palestinian Aramaic. Finally, the same distribution among the Late Aramaic dialects is found with the replication of Aramaic ger ‘truly, indeed’ on Greek γάρ: this change is found in Syriac and Christian Palestinian Aramaic but not in the other Late Aramaic dialects (§9.6). The cases of grammatical replication presented in this book, then, show that Syriac and Christian Palestinian Aramaic diverge from the other Late Aramaic dialects in a number of ways due to contact with Greek. By analyzing contact-induced changes in Syriac due to Greek, I have sought to shed new light on the Greco-Roman context of the Syriac language and, more broadly, of Syriac Christianity. As the Aramaic dialect of Edessa was transformed into the literary language of Syriac during the first centuries of the Common Era, a number of contact-induced changes due to Greek were taking place. The vocabulary of Syriac was augmented with numerous Greek loanwords, exceeding those found in the other dialects of Late Aramaic, with the exception of Christian Palestinian Aramaic and possibly also of Jewish Palestinian Aramaic. In addition, inherited Aramaic material was adapted to replicate Greek constructions, thereby departing not only from the earlier Aramaic dialects but also from the other Late Aramaic dialects. Again, only 46.  Note the following comment by Brock on the usefulness of poetry for studying loanwords: “Poetry is especially suitable, since most of the Greek words used will be genuine loanwords, rather than temporary, or learned, borrowings” (Brock 1999–2000: 439).

Conclusion

211

Christian Palestinian Aramaic shares some of these developments with Syriac. The differences between Syriac and the other Late Aramaic dialects—excluding Christian Palestinian Aramaic—were only further accentuated as Syriac continued to be in contact with Greek for centuries. The end result of this long period of intense contact with Greek was a dialect of Aramaic called Syriac that differed in a number of ways from the other Late Aramaic dialects of Jewish Babylonian Aramaic and Mandaic in Mesopotamia and of Samaritan Aramaic and Jewish Palestinian Aramaic in the Levant. The Syriac language itself, thus, provides clear evidence for the pervasive impact that the GrecoRoman world had on Syriac Christianity beginning already in the early centuries of the Common Era and extending throughout Late Antiquity.

Appendix 1

Greek Loanwords Inherited in Syriac The following words are attested both in an Aramaic dialect prior to the second century c.e. (Middle Aramaic or earlier) and in Syriac by the fourth century. Thus, based on the arguments presented in §4.9, I propose that it is likely that they were transferred into Aramaic at an earlier period and then inherited in Syriac.

a. ἀήρ (LSJ 30) > ‫ ܐܐܪ‬ʾʾr ‘air’ (pre-4th cent. Odes Sol. 5:5 [ed. Charlesworth 1973]; also in OT and NT; LS2 1; SL 1), already in Tg. Onq. ʾawwer ‘air’ (Cook 2008: 5); see also Christian Palestinian Aramaic ʾʾr (DCPA 1; LSP 1); Jewish Palestinian Aramaic ʾawwer, ʾbyr ‘air, space’ (DJPA 38); Samaritan Aramaic ʾwyr ‘open space’ (DSA 13); Jewish Babylonian Aramaic ʾăwerɔ (DJBA 87–88); Mandaic aiar ‘upper atmosphere, air, ether, wind’ (MD 14) b. ἀνδριάς (LSJ 128) → accusative singular ἀνδριάντα > ‫ܐܢܕܪܝܢܛܐ‬ ʾndrynṭʾ, ‫ ܐܕܪܝܛܐ‬ʾdryṭʾ ‘statue’ (pre-4th cent. Old Syriac Inscriptions As 1.5 [ed. Drijvers and Healey 1999]; also in 2 Chr 14:2; LS2 6; SL 11), already in Palmyrene ʾdrṭ (PAT 335; see also Brock 2005: 12); see also Late Jewish Literary Aramaic ʾndrṭʾ (Tg. Esth. 1 3:2; Jastrow 81); Jewish Palestinian Aramaic ʾndrṭ (DJPA 64); Jewish Babylonian Aramaic ʾandrɔṭɔ (DJBA 144) ̈ c. ἀρχαί (LSJ 252) > ‫ ܐܪܟܐ‬ʾrkʾ, in the phrase ‫ܐܖܟܐ‬ ‫ ܒܝܬ‬byt ʾr̈kʾ ‘archive’ (4th cent. Ephrem, Sermons I, 43.11 [ed. Beck 1970a]; LS2 49; SL 100, 145), already in Palmyrene ʾrkʾ (PAT 347; see also Brock 2005: 12) d. ἄρχων (LSJ 254) > ‫ ܐܪܟܘܢܐ‬ʾrkwnʾ ‘ruler, archon; leader, chief’ (pre-4th cent. P. Euphrates 6.36, 43; 7.34, 38 [ed. Feissel, Gascou, and Teixidor 1997], also in NT; LS2 49; SL 100), already in Palmyrene ʾrkwn (PAT 343; see also Brock 2005: 12); see also Late Jewish Literary Aramaic ʾrkwn (Tg. Job 21:28; Jastrow 121); Christian Palestinian Aramaic ʾrkwn (DCPA 29; LSP 18); Jewish Palestinian Aramaic ʾrkwn (DJPA 75); Samaritan Aramaic ʾrkwn (DSA 63); Jewish Babylonian 212

Greek Loanwords Inherited in Syriac

213

Aramaic ʿrkn (DJBA 881–882); Mandaic arkun (MD 37–38); Judean Aramaic ʾrkwn (DNWSI 109) e. Latin assarium (OLD 186) > ἀσσάριον (LLGE 31; LSJ 260) > ‫ ܐܣܪܐ‬ʾsrʾ ‘assarius, small copper coin’ (Bible Matt 10:29 [SP]; Luke 12:6 [SP]; LS2 38; SL 80; see also Brock 1967: 394), also in Ḥatran ʾs (Contini and Pagano 2015: 129); Palmyrene ʾsr (PAT 341; see also Brock 2005: 12–13); see also Christian Palestinian Aramaic ʾsr (DCPA 24; LSP 16); Jewish Babylonian Aramaic ʾissɔrɔ (DJBA 123) f. αὐτοκράτωρ (LSJ 280–81) > ‫ ܐܘܛܩܪܛܘܪ‬ʾwṭqrtwr ‘emperor’ (pre-4th cent. Old Syriac Parchments 1.1; 2.2; 3.1 [ed. Drijvers and Healey 1999: 231–48]; LS2 8; SL 14), already in Palmyrene [ʾw]ṭqrṭwr (PAT 335; see also Brock 2005: 13) g. βάσις (LSJ 310) > ‫ ܒܣܣ‬bss ‘base’ (pre-4th cent. Exod 25:31; 35:16; 37:17; 38:5; 39:39; etc.; LS2 81; SL 166), already in Nabatean bss (DNWSI 179; Healey 1993: 69–70, 255; 1995: 77); Tg. Jonathan bəsis (1 Kgs 7:27, 30; Jastrow 179); see also Late Jewish Literary Aramaic bsys (Tg. Ps.-J. Exod 25:31); Christian Palestinian Aramaic bsys (DCPA 53); Jewish Palestinian Aramaic bsys (DJPA 106); Samaritan Aramaic bsys (DSA 105) h. βουλευτής (LSJ 324–25) > ‫ ܒܘܠܘܛـܐ‬bwlwṭʾ ‘counselor, senator’ (NT Mark 15:43 [SP]; Luke 23:50 [SCP]; LS2 62; SL 127; see also Brock 1967: 396), already in Palmyrene b(y)lwṭ (PAT 346; see also Brock 2005: 13); see also Christian Palestinian Aramaic bwlwṭys (DCPA 42; LSP 23); Jewish Palestinian Aramaic bwlwwṭys (DJPA 87) i. βωμός (LSJ 334) > ‫ ܒܘܡܣܐ‬bwmsʾ ‘altar’ (pre-4th cent. 2 Chr 14:2; 31:1; LS2 78; SL 127), already in Nabatean bms (DNWSI 168; or βῆμα); see also Christian Palestinian Aramaic bwms (DCPA 42; LSP 23); Jewish Babylonian Aramaic bimosɔ (DJBA 201) j. γένος (LSJ 344) > ‫ ܓܢܣܐ‬gnsʾ ‘kind, species; family; race, nation’ (pre-4th cent. Acts Thom. 201:17; 244:16; 245:5 [ed. Wright 1871a: 2.171–333]; Odes Sol. 41:8 [ed. Charlesworth 1973]; also in OT and NT; LS2 90, 125, 841; SL 179, 249), already in Ḥatran gns (Contini and Pagano 2015: 132); Palmyrene gns (PAT 354; see also Brock 2005: 13); Tg. Jonathan p l gnysn (1 Sam 17:4); see also Late Jewish Literary Aramaic gynsʾ (Tg. Ps.-J. Exod 12:47; Jastrow 260); Christian Palestinian Aramaic gns (DCPA 75; LSP 39); Samaritan Aramaic gnws (DSA 154); Jewish Palestinian

214

Appendix 1 Aramaic gynws (DJPA 128); Jewish Babylonian Aramaic ginsɔ (DJBA 297); Mandaic ginsa (MD 91) k. Latin denarius (OLD 514; LD 545) > δηνάριον (LLGE 40; LSJ 388) > ‫ ܕܝܢܪܐ‬dynrʾ ‘gold denar’ (pre-4th cent. Old Syriac Parchments 1.ii [abbreviation], 9; 2.ix, 16, 17, 18, 22 [ed. Drijvers and Healey 1999: 231–48]; P. Euphrates 7.29; 10.22 [ed. Feissel, Gascou, and Teixidor 1997]; also in OT and NT; LS2 160; SL 297), already in Ḥatran dnr (DNWSI 256; Contini and Pagano 2015: 133); Judean Aramaic dynr (DNWSI 256); Palmyrene dnrʾ, dynr (PAT 356; see also Brock 2005: 14); Tg. Jonathan pl denɔrin (2 Kgs 5:5); see also Late Jewish Literary Aramaic dynrʾ (Tg. Ps.-J. Exod 30:13); Christian Palestinian Aramaic dynr (DCPA 86; LSP 45); Jewish Palestinian Aramaic dynr (DJPA 147; Jastrow 302) Jewish Babylonian Aramaic denɔrɔ (DJBA 334); Mandaic dinara (MD 108) l. διάταγμα (LSJ 414) > ‫ ܕܝܛܓ�ܡܐ‬dyṭgmʾ ‘order, charge’ (pre-4th cent. Ezra 4:18; 8:36; LS2 150; SL 294), already in Palmyrene dyṭgmʾ (PAT 356; see also Brock 2005: 14); see also Late Jewish Literary Aramaic dyṭgmʾ (Tg. Esth. 2 3:15; Jastrow 294); Jewish Palestinian Aramaic dyyṭgmh (DJPA 145) m. ἐξέδρα (LSJ 589) > ‫ ܐܟܣܕܪܐ‬ʾksdrʾ ‘exedra’ (pre-4th cent. 1 Kgs 7:4; Ezek 40:38, 45, 46; 41:10; 42:1, 4, 5, 7; 44:19; LS2 18; SL 43), already in Palmyrene ʾksdrʾ, ʾkšdrʾ (PAT 337; see also Brock 2005: 15; Blau 1970: 58 n. 17 [on the spelling with š]); Tg. Jonathan ʾaḵsaḏrɔ (Judg 3:23; Jastrow 64); see also Late Jewish Literary Aramaic ʾksdrwy (Tg. Ps. 104:3); Jewish Babylonian Aramaic ʾaḵsaḏrɔ (DJPA 131) n. ἐπαρχία (LSJ 611) > ‫ ܗܦܪܟܝܘܣ‬hprkyws ‘province’ (4th cent. [translation] Eusebius of Caesarea, Ecclesiastical History, 76.17 [ed. Wright and McLean 1898]; LS2 43, 182; SL 89, 353), already in Judean Aramaic hprkyh (DJA 44); Nabatean hprkyʾ (DNWSI 292); see also Late Jewish Literary Aramaic ʾprkyʾ (Tg. Lam. 1:1; Jastrow 59); Christian Palestinian Aramaic ʾprkyʾ, hprkyʾ (DCPA 26, 104); Jewish Palestinian Aramaic ʾyprkyyh (DJPA 53) o. ἐπίτροπος (LSJ 669) > ‫ ܐܦܛܪܘܦܐ‬ʾpṭrwpʾ ‘prefect; manager’ (4th cent. Book of Steps, 464.7, 8, 12, 17, 18, 22; 465.1, 3, 6 [ed. Kmosko 1926]; also in NT; LS2 40; SL 86), already in Judean Aramaic ʾpṭrp (DNWSI 94); Palmyrene ʾpṭrp (PAT 342; see also Brock 2005: 16); see also Late Jewish Literary Aramaic ʾpwṭrwpws (Tg. Ps.-J. Gen 39:4; Jastrow 102); Christian

Greek Loanwords Inherited in Syriac

215

Palestinian Aramaic ʾṗyṭrwpws, hṗyṭrwpʾ (DCPA 25; LSP 16); Jewish Palestinian Aramaic ʾpyṭrwpws (DJPA 69–70); Jewish Babylonian Aramaic ʾappiṭroppɔ (DJBA 155) p. ζεῦγος (LSJ 754), ζυγόν (LSJ 757) > ‫ ܙܘܓܐ‬zwgʾ ‘yoke, pair; chariot’ (pre-4th cent. Acts Thom. 215:6; 231:9; 238:5; 242:3, 9 [ed. Wright 1871a: 2.171–333], also in OT and NT; LS2 90, 188; SL 180, 369–70), already in Tg. Jonathan zoḡ (2 Kgs 9:25); see also Late Jewish Literary Aramaic zwgʾ (Tg. Ps.-J. Lev 15:9); Christian Palestinian Aramaic zwg (DCPA 108; LSP 54); Jewish Palestinian Aramaic zoḡ (DJPA 400); Samaritan Aramaic zwg (DSA 223–24); Jewish Babylonian Aramaic zoḡɔ ‘pair’ (DJBA 400–401), zygʾ (DJBA 406) q. ἡγεμών (LSJ 763) > ‫ ܗܓܡܘܢܐ‬hgmwnʾ, ‫ ܐܝܓܡܘܢܐ‬ʾygmwnʾ ‘prefect’ (4th cent. Aphrahaṭ, Demonstrations, 1.973.6 [ed. Parisot 1894–1907]; Book of Steps, 645.20; 648.3; 648.15 [ed. Kmosko 1926]; also in OT and NT; LS2 4, 171; SL 31, 340), already in Nabatean hgmwn (DNWSI 270); Palmyrene hgmwn, hygmwn (PAT 359; Brock 2005: 16); see also Late Jewish Literary Aramaic hgmwnʾ (Tg. Esth. 2 8:7; Jastrow 331); Christian Palestinian Aramaic hyg(y)mwn (DCPA 98; LSP 50); Jewish Babylonian Aramaic haḡmonɔ (DJBA 360) r. Latin caesar (OLD 254; LD 265) > καῖσαρ (LSJ 860) > ‫ܩܣܪ‬ qsr ‘Caesar, emperor’ (pre-4th cent. Old Syriac Inscriptions As 49.7 [mostly restored; ed. Drijvers and Healey 1999]; Old Syriac Parchments 1.1; 2.2; 3.1 [ed. Drijvers and Healey 1999: 231–48]; also in NT; LS2 680; SL 1388), already in Judean Aramaic qysr (DJA 77–78); Nabatean qysr (DNWSI 1018–19; Healey 1995: 81); Palmyrene qysr (PAT 406; see also Brock 2005: 17); see also Christian Palestinian Aramaic qysr (DCPA 372; LSP 179); Jewish Palestinian Aramaic qysr (DJPA 372); Jewish Babylonian Aramaic qesɔr (DJBA 1014–15) s. Latin centurio (OLD 300; LD 316) > κεντυρίων (LLGE 53; PGL 744) > ‫ ܩܢܛܪܝܘܢ‬qnṭrywn, ‫ ܩܢܛܪܘܢܐ‬qnṭrwnʾ ‘centurion’ (4th cent. Aphrahaṭ, Demonstrations, 1.92.6 [ed. Parisot 1894– 1907]; Ephrem, Maḏrɔše on Faith, 35.3, 15; 196.4 [ed. Beck 1955]; Maḏrɔše on the Nativity, 18.4 [ed. Beck 1959]; also in NT; LS2 677; SL 1382–83), already in Nabatean qnṭryn (DNWSI 1015; Healey 1993: 209, 264; 1995: 77); Palmyrene qṭrywn (PAT 405–6; see also Brock 2005: 17); see also Christian Palestinian Aramaic qnṭrywnʾ (DCPA 377; LSP 181) t. κιθάρα, κίθαρις (LSJ 950) > ‫ ܩܝܬܪܐ‬qytrʾ ‘cithern, lyre’ (pre-4th cent. Odes Sol. 6:1; 7:17; 14:8; 26:3 [ed. Charlesworth 1973];

216

Appendix 1 also in OT and NT; LS2 705; SL 1366), already in Daniel qytrws (Kethiv), qaṯros (HALOT 1970); Tg. Jonathan qtrws (Isa 5:12; Jastrow 1434); see also Christian Palestinian Aramaic qytr (DCPA 373; LSP 186) u. κίνδυνος (LSJ 952) > ‫ ܩܝܢܕܘܢܘܣ‬qyndwnws ‘danger’ (pre-4th cent. Odes Sol. 38:5; 39:8 [ed. Charlesworth 1973]; also in NT; LS2 676; SL 1363–64), already in Palmyrene qdns (PAT 404; see also Brock 2005: 17); see also Christian Palestinian Aramaic qyndynws (DCPA 371; LSP 179) v. Latin collarium, collare (OLD 350; LD 365) > κολλάριον (LLGE 56; LSJ 972) > ‫ ܩܘܠܪܐ‬qwlrʾ ‘iron collar’ (pre-4th cent. 1 Chr 20:3; 2 Sam 12:31; LS2 671; SL 1330), already in Tg. Jonathan qolɔr (Ezek 19:9; Jastrow 1330); see also Jewish Palestinian Aramaic qwlr (DJPA 479) w. Latin colonia (OLD 355; LD 370) > κολωνία (LLGE 56; PGL 766; LSJ 974) > ‫ ܩܘܠܘܢܝܐ‬qwlwnyʾ ‘colony’ (pre-4th cent. Old Syriac Parchments 1.4; 3.4 [ed. Drijvers and Healey 1999: 231– 48]; also in NT; LS2 669; SL 1329), already in Palmyrene qlnyʾ (PAT 406; see also Brock 2005: 18); see also Jewish Babylonian Aramaic qɔlɔnyɔ (DJBA 1021) x. Latin legio (OLD 1013–14; LD 1047) > λεγιών, λεγεών (LLGE 65; PGL 794) > ‫ ܠܓܝܘܢܐ‬lgywnʾ ‘legion’ (4th cent. Book of Steps, 153.9 [ed. Kmosko 1926]; common in Ephrem; also in OT and NT; LS2 358; SL 673), already in Palmyrene lgywn (PAT 376; see also Brock 2005: 18); Tg. Jonathan pl ligyonin (Ezek 30:9); see also Late Jewish Literary Aramaic lgywnʾ (Tg. Job 15:24; Jastrow 692); Christian Palestinian Aramaic l(y)gywn (DCPA 199; LSP 101); Jewish Palestinian Aramaic liḡyon (DJPA 281) y. λεκάνη (LSJ 1037) > ‫ ܠܩܢܐ‬lqnʾ ‘platter, basin’ (pre-4th cent. Acts Thom. 221:14 [ed. Wright 1871a: 2.171–333]; also in OT and NT; LS2 370; SL 697), already in Tg. Jonathan ləqinəʾ, variant liqnɔ (Judg 6:38; Jastrow 719); see also Late Jewish Literary Aramaic lqynʾ (Tg. Job 32:19; Jastrow 719); Jewish Babylonian Aramaic liqnɔ (DJBA 633) z. λιμήν (LSJ 1050) > ‫ ܠ�ܡܐܢܐ‬lmʾnʾ ‘harbor’ (pre-4th cent. Acts Thom. 206:19 [ed. Wright 1871a: 2.171–333]; Odes Sol. 38:3 [ed. Charlesworth 1973]; also in OT and NT; LS2 367; SL 691– 92), already in Palmyrene lmn ‘emporium’ (PAT 377; see also Brock 2005: 18); see also Christian Palestinian Aramaic lmyn (DCPA 201); Jewish Palestinian Aramaic lmyn (DJPA 284) aa. μαγίς (LSJ 1071) > ‫ ܡܓܣܐ‬mgsʾ ‘jar, dish’ (pre-4th cent. Exod 37:16; Num 4:7; LS2 374; SL 710), already in Tg. Onq. məgisəṯɔ

Greek Loanwords Inherited in Syriac

217

(Num 4:7; Cook 2008: 144); see also Late Jewish Literary Aramaic mgysʾ (Tg. Ps. 123:2; Jastrow 728); Jewish Babylonian Aramaic mḡisɔ (DJBA 640) ab. μηλωτή (LSJ 1127) > ‫ ܡܝܠܬܐ‬myltʾ ‘carpet; covering; pillow’ (4th cent. Ephrem, Maḏrɔše on the Nativity, 46.6 [ed. Beck 1959]; LS2 383; SL 752), already in Palmyrene mlṭ (PAT 381– 82; to be added to Brock 2005); Tg. Onq. melɔ ‘fine wool’ (Gen 49:11; Cook 2008: 151); see also Late Jewish Literary Aramaic myltʾ (Tg. Esth. 2 1:6; Jastrow 775); Samaritan Aramaic mylt (DSA 464); Jewish Babylonian Aramaic melṯɔ (DJBA 669–70) ac. Latin mille (OLD 1109; LD 1144) > μίλιον (LSJ 1134) > ‫ܡܝ�ܠܐ‬ mylʾ ‘one-thousand paces; mile-stone’ (pre-4th cent. Acts Thom. 195:17; 196:10; 238:11 [ed. Wright 1871a: 2.171–333]; also in NT; LS2 383; SL 752), already in Palmyrene m(yl) (PAT 380; see also Brock 2005: 18); see also Late Jewish Literary Aramaic pl mylyn (Tg. Ps.-J. Exod 12:37); Christian Palestinian Aramaic myl (DCPA 220; LSP 109); Jewish Palestinian Aramaic myl (DJPA 304–5); Jewish Babylonian Aramaic milɔ (DJBA 667) ad. Latin modium (OLD 1123; LD 1155) > μόδιος (LLGE 73; LSJ 1140) > ‫ ܡܘܕܝܐ‬mwdyʾ ‘corn measure, peck; container’ (NT Matt 5:15 [C]; LS2 375; SL 721–22), already in Achaemenid Aramaic mdy (CAL); Palmyrene mdʾ (PAT 378; see also Brock 2005: 18–19); see also Christian Palestinian Aramaic mwdyʾ (DCPA 211; LSP 107); Jewish Palestinian Aramaic mwdyy (DJPA 294); Jewish Babylonian Aramaic moḏyɔ (DJBA 645) ae. μοχλός (LSJ 1149) > ‫ ܡܘܟ�ܠܐ‬mwklʾ ‘bolt for fastening door’ (pre-4th cent. Odes Sol. 17:10 [ed. Charlesworth 1973]; also in OT; LS2 385; SL 724), already in Ḥatran mklʾ (DNWSI 624; Contini and Pagano 2015: 136); Tg. Onq. pl mugləsayyɔ (1 Kgs 7:50; Jastrow 738) af. ναός (LSJ 1160) > ‫ ܢܘܣܐ‬nwsʾ ‘temple; fortress, citadel’ (pre4th cent. Acts Thom. 174:8; 181:7; 185:12 [ed. Wright 1871a: 2.171–333]; also in OT and NT; LS2 421, 922; SL 901), already in Nabatean nws (DNWSI 723); Palmyrene nws (PAT 231; see also Brock 2005: 19); see also Christian Palestinian Aramaic nws, nʾws (DCPA 254; LSP 121); Jewish Palestinian Aramaic nwws (DJPA 344); Mandaic nausa (MD 282) ag. νόμος (LSJ 1180) > ‫ ܢܡܘܣܐ‬nmwsʾ ‘law’ (pre-4th cent. Acts Thom. 225:12, 16; 226:6, 8; 229:8 [ed. Wright 1871a: 2.171–333]; Book of the Laws of the Countries, 48× [see Lund 2007: 180–81] [ed. Drijvers 1965]; Old Syriac Parchments 1.16 [ed. Drijvers and Healey 1999: 231–48]; also in OT and NT;

218

Appendix 1 LS2 431; SL 921–22), already in Palmyrene nmws (PAT 389; see also Brock 2005: 19); Tg. Jonathan nimosɔ (1 Sam 2:12); see also Late Jewish Literary Aramaic nmwsʾ (Tg. Ps. 1:2; Jastrow 905); Christian Palestinian Aramaic nymws (DCPA 263–64; LSP 123); Jewish Palestinian Aramaic nimos, nwmws (DJPA 349, 839); Samaritan Aramaic nymws (DSA 523); Mandaic nimusa (MD 298) ah. ξένος (LSJ 1189) + adjectival ending -ɔyɔ > ‫ ܐܟܣܢܝܐ‬ʾksnyʾ ‘strange, foreign; stranger’ (pre-4th cent. Acts Thom. 175:5, 7; 183:12; 231:3; 242:11 [ed. Wright 1871a: 2.171–333]; Odes Sol. 17:6 [ed. Charlesworth 1973]; also in OT and NT; LS2 338; SL 44), already in Palmyrene ʾksny (PAT 337–38; see also Brock 2005: 19); see also Late Jewish Literary Aramaic ʾksnyʾ (Tg. Ps.-J. Deut 27:18); Christian Palestinian Aramaic ʾksnʾy (DCPA 15; LSP 8); Jewish Palestinian Aramaic ʾksnyy (DJPA 58); Samaritan Aramaic ʾksnʾy (DSA 29); Jewish Babylonian Aramaic ʾaḵsənɔyɔ (DJBA 131) ai. Latin sextarius (OLD 1751; LD 1688) > ξέστης (LLGE 76–77) > ‫ ܩܣܛܐ‬qsṭʾ ‘vase, urn; measure’ (4th cent. Ephrem, Commentary on Genesis and Part of Exodus, 146.21, 22 [ed. Tonneau 1955]; also in OT and NT; LS2 679; SL 1387), already in Palmyrene qsṭwn (PAT 406; see also Brock 2005: 19); see also Late Jewish Literary Aramaic qsṭʾ (Tg. Ps. Exod 30:24); Christian Palestinian Aramaic qysṭ (DCPA 372; LSP 181); Jewish Palestinian Aramaic qsyṭ (DJPA 498); Jewish Babylonian Aramaic qisṭɔ, qystʾ (DJBA 1014) aj. πεῖσαι (LSJ 1353–54) > rt. ‫ ܦܝܣ‬pys c ‘to persuade, to convince; to demand, seek, beseech’, ct ‘to be persuaded; to obey’ (pre4th cent. Acts Thom. 172:17; 180:15; 181:19; 182:6; 221:3, 5; 240:6; 241:3 [ed. Wright 1871a: 2.171–333]; Book of the Laws of the Countries, 14× [see Lund 2007: 200–201] [ed. Drijvers 1965]; Odes Sol. 8:17; 39:8 [ed. Charlesworth 1973]; LS2 567; SL 1188), already in Tg. Onq. pys (Cook 2008: 108 [s.v. ṭps]; see Butts 2012: 158); see also Late Jewish Literary Aramaic pys (Tg. Ps.-J. Gen 19:3); Christian Palestinian Aramaic ṗys (DCPA 327–28; LSP 156); Jewish Palestinian Aramaic pys (DJPA 430–31); Samaritan Aramaic pys (DSA 676); Jewish Babylonian Aramaic pys (DJBA 899–900) ak. πίναξ (LSJ 1405) > ‫ ܦܝܢܟܐ‬pynkʾ ‘dish, writing tablet’ (4th cent. Aphrahaṭ, Demonstrations, 1.729.3 [citing Matt 23:25] [ed. Parisot 1894–1907]; Ephrem, Maḏrɔše on the Nativity, 104.13 [ed. Beck 1959]; Maḏrɔše on Nisibis, 2.87.12 [Beck 1963]; also

Greek Loanwords Inherited in Syriac

219

in NT; LS2 579; SL 1188), already in Achaemenid Aramaic pynk (DNWSI 910); see also Christian Palestinian Aramaic ṗynks (DCPA 327; LSP 156); Jewish Palestinian Aramaic pynk (DJPA 431); Samaritan Aramaic pnk (DSA 690); Jewish Babylonian Aramaic pinḵɔ (DJBA 901) al. πλατεῖα (LSJ 1413–14); see also Latin platea (OLD 1388; LD 1385) > ‫ ܦܠܛܝܐ‬plṭyʾ ‘open space, square’ (pre-4th cent. Jer 5:1; 9:20; Song 3:2; LS2 574; SL 1199), already in Palmyrene plṭyʾ (PAT 400–401; see also Brock 2005: 20); see also Late Jewish Literary Aramaic plṭyh (Tg. Job 29:7; Jastrow 1179); Jewish Palestinian Aramaic plṭyh (DJPA 435); Samaritan Aramaic plṭyh (see also DJPA 435) am. πολιτεία (LSJ 1434) > ‫ ܦܘܠܝܛܝܐ‬pwlyṭyʾ ‘republic, state’ (4th cent. [translation] Eusebius of Caesarea, Theophany, 56 [ed. Lee 1842]; LS2 574, 848; SL 1164), already in Palmyrene plṭyʾ (PAT 400; see also Brock 2005: 20); see also Christian Palestinian Aramaic ṗwlyṭyʾ (DCPA 325; LSP 158) an. πόρπη (LSJ 1451) > ‫ ܦܪܦܐ‬prpʾ ‘clasp, buckle, ring’ (pre-4th cent. Exod 35:11; LS2 604; SL 1248), already in Tg. Onq. p l purpin (Exod 26:6; Cook 2008: 229); see also Late Jewish Literary Aramaic pl pwrpyyʾ (Tg. Ps.-J. Exod 26:6); Jewish Palestinian Aramaic prp (DJPA 450) ao. πραγματευτής (LSJ 1458) > ‫ ܦܪܓܡܛܘܛـܐ‬prgmṭwṭʾ ‘agent, merchant’ (pre-4th cent. P. Euphrates 6.36, 7.34 [ed. Feissel, Gascou, and Teixidor 1997]; LS2 592; SL 1227), already in Palmyrene prgmṭt (PAT 401; see also Brock 2005: 20); see also Jewish Palestinian Aramaic prgmṭwwṭ (DJPA 444) ap. προνοῆσαι (LSJ 1490–91) > rt. ‫ ܦܪܢܣ‬prns ‘to divide, distribute; to provide for, supply; to manage, administer’ (4th cent. Book of Steps, 4.19; 60.13, 14; 76.19; 381.14 [ed. Kmosko 1926]; also in OT; LS2 599; SL 1243; see Butts forthcoming c), attested already in Palmyrene prns (PAT 401; DNWSI 940); Tg. Jonathan prns (Ezek 34:8 [2×]; Isa 57:8; Jastrow 1231); see also Late Jewish Literary Aramaic prns (Ps. J. Gen 30:30, Lev 25:35; Jastrow 1231); Christian Palestinian Aramaic prns (DCPA 341; LSP 163); Jewish Palestinian Aramaic prns (DJPA 448, 842); Samaritan Aramaic prns (DSA 704–5); Jewish Babylonian Aramaic prns (DJBA 935) aq. σάνδαλον (LSJ 1582) > ‫ ܣܕ�ܠܐ‬sdlʾ ‘sandal’ (NT Mark 6:9 [S]; LS2 460, 484; SL 971, 1022), already in Tg. Jonathan sandəlin (Isa 11:15; Jastrow 1004–5); see also Late Jewish Literary Aramaic sndl (Tg. Ps. 108:10; Jastrow 1004–5); Jewish

220

Appendix 1 Palestinian Aramaic sndl (DJPA 383); Jewish Babylonian Aramaic sandlɔ (DJBA 821); Mandaic sandla (MD 313) ar. σημεῖον (LSJ 1593) > ‫ ܣܡܝܘܢ‬smywn ‘sign; zenith; example’ (4th cent. Ephrem, Prose Refutations, Discourse 2–5, 108.28 [ed. Mitchell 1912–21]; LS2 480; SL 1017–18), already in Ḥatran smʾ (DNWSI 790; Contini and Pagano 2015: 136); Palmyrene smy (PAT 392; see also Brock 2005: 20); see also Late Jewish Literary Aramaic symywnʾ ‘bond, shackle’ (Tg. Job 13:27; Jastrow 1000); Christian Palestinian Aramaic sym(y)wn (DCPA 285; LSP 135); Jewish Palestinian Aramaic symn (DJPA 375); Samaritan Aramaic symn (DSA 584) as. σμίλη (LSJ 1619) > ‫ ܙܡܠܝܐ‬zmlyʾ ‘small knife, scalpel’ (pre-4th cent. Jer 36:23; LS2 199; SL 385), already in Tg. Jonathan ʾuzmil (Jer 36:23); see also Late Jewish Literary Aramaic ʾzml (Tg. Job 16:9; Jastrow 46); Jewish Palestinian Aramaic ʾwzmyl (DJPA 38) at. στατήρ (LSJ 1634) [> Pahlavi stēr (CPD 77) (?)] > ‫ܐܣܬܝܪܐ‬ ʾstyrʾ, ‫ ܐܣܬܪܐ‬ʾstrʾ ‘stater, coin, weight’ (4th cent. Ephrem, Maḏrɔše against Julian the Apostate, 75.3 [ed. Beck 1957b]; Ephrem, Maḏrɔše on Nisibis, 2.55.4 [Beck 1963]; also in OT and NT; LS2 38; SL 80), already in Achaemenid Aramaic str, pl sttrn (DNWSI 805); Ḥatran ʾstr (DNWSI 92; Contini and Pagano 2015: 130); see also Christian Palestinian Aramaic ʾsṭṭyr (DCPA 22; LSP 15); Jewish Palestinian Aramaic sṭw (DJPA 372); Jewish Babylonian Aramaic ʾisterɔ (DJBA 123); Mandaic astira (MD 30) au. στοά (LSJ 1647) > ‫ ܐܣܛܘܐ‬ʾsṭwʾ ‘portico’ (pre-4th cent. 1 Kgs 6:3; passim; also in NT; LS2 32; SL 68), already in Ḥatran ʾsṭwʾ (DNWSI 87; Contini and Pagano 2015: 130); Judean Aramaic sṭwh (DNWSI 783); Palmyrene ʾsṭwʾ (PAT 341; see also Brock 2005: 21); see also Christian Palestinian Aramaic ʾsṭwʾ (DCPA 22; LSP 15); Jewish Palestinian Aramaic sṭw (DJPA 372) av. στρατηγός (LSJ 1652) > ‫ ܣܛܪܛܓܐ‬sṭrṭgʾ ‘strategos’ (pre4th cent. Old Syriac Parchments 1.v3 [Drijvers and Healey 1999: 231–48]; also in NT; LS2 34, 469; SL 71, 998), already in Nabatean ʾsrtgʾ (DNWSI 92; see also Healey 1993: 108; 1995: 77); Palmyrene ʾsṭrṭg (PAT 341; see also Brock 2005: 21); Tg. Jonathan ʾisṭarṭigay (1 Sam 10:5); see also Late Jewish Literary Aramaic ʾsṭrṭyg (Tg. Chr. 1 11:16); Christian Palestinian Aramaic ʾsṭrṭygws (DCPA 22; LSP 15); Jewish Palestinian Aramaic ʾysṭrṭyg (DJPA 52); Jewish Babylonian Aramaic ʾisṭrɔṭeḡɔ (DJBA 122)

Greek Loanwords Inherited in Syriac

221

aw. συμφωνία (LSJ 1689) > ‫ ܨܦܘܢܝܐ‬ṣpwnyʾ ‘bagpipe’ (pre-4th cent. Acts Thom. 174:14 [ed. Wright 1871a: 2.171–333]; also in OT and NT; LS2 635; SL 1297), already in Daniel sumponyɔ (Dan 3:5, 15), sypnyh (Dan 3:10 [Kethiv]), supponyɔ (Dan 3:10 [Qere]) (HALOT 1937–38) ax. σῶμα (LSJ 1749) → nominative/accusative plural σώματα > pl ‫ ܣܘܡܛܐ‬swmṭʾ (sic; without syɔme) ‘bodies’ (4th cent. Ephrem, Prose Refutations, Discourse 2–5, 2.6.45 [ed. Mitchell 1912–21]; LS2 479; SL 981), already in Palmyrene swm (PAT 391; see also Brock 2005: 22); see also Jewish Palestinian Aramaic smh (DJPA 381) [possibly a code-switch] ay. τάγμα (LSJ 1752) > ‫ ܬܓ�ܡܐ‬tgmʾ ‘order, class; command, precept; troop, cohort’ (pre-4th cent. Book of the Laws of the Countries, 28.23 [ed. Drijvers 1965]; Odes Sol. 35:4 [ed. Charlesworth 1973]; also in OT; LS2 92, 268, 816; SL 185, 512, 1623), already in Palmyrene tgmʾ ‘association’ (PAT 418; see also Brock 2005: 22); Jewish Babylonian Aramaic tgmmyh (DJBA 1194) az. Latin talaria (OLD 1901; LD 1835) > ταλάριον (LLGE 110) > pl ‫ ܛܠ̈ܪܐ‬ṭlr̈ʾ ‘sandals’ (NT Mark 6:9 [P]; Acts 12:8; LS2 278; SL 535), already in Tg. Jonathan ṭallɔriṯɔ (1 Kgs 2:5; Jastrow 538) ba. τάξις (LSJ 1756) > ‫ ܛܟܣܐ‬ṭksʾ ‘order; rank’ (pre-4th cent. Acts Thom. 240:2 [ed. Wright 1871a: 2.171–333]; Book of the Laws of the Countries, 32.12, 32.14, 62.9 [ed. Drijvers 1965]; also in NT; LS2 90, 274–75; SL 181, 529), already in Palmyrene ṭksys ‘row’ (PAT 368; see also Brock 2005: 22); see also Late Jewish Literary Aramaic pl ṭyksyn (Tg. Ps.-J. Lev 24:10); Christian Palestinian Aramaic ṭks, ṭksys (DCPA 148; LSP 74); Jewish Palestinian Aramaic ṭqs ‘banner’ (DJPA 224), ṭqsys ‘regiment of troops’ (DJPA 230); Mandaic ṭaksa (MD 174) bb. ταῶς (LSJ 1763) > ‫ ܛܘܣܐ‬ṭwsʾ ‘peacock’ (4th cent. Ephrem, Commentary on Genesis and Part of Exodus, 59.22 [ed. Tonneau 1955]; Maḏrɔše against Heresies, 170.16 [ed. Beck 1957a]; also in OT; LS2 271; SL 519), already in Tg. Jonathan pl ṭwɔsin (1 Kgs 10:22); see also Late Jewish Literary Aramaic ṭwwsʾ (Tg. Esth. 2 1:2; Jastrow 522); Jewish Palestinian Aramaic ṭwws (DJPA 221); Samaritan Aramaic ṭʾws (DSA 307); Jewish Babylonian Aramaic ṭwwsʾ (DJBA 496); Mandaic ṭausa (MD 173) bc. ὑπατεία (LSJ 1854) > ‫ ܗܦܛܝܐ‬hpṭyʾ ‘consulship; gift of a consul’ (pre-4th cent. Old Syriac Parchments 1.2; 2.4; 3.2 [ed. Drijvers

222

Appendix 1 and Healey: 231–48]; LS2 179, 842; SL 337), already in Judean Aramaic hpṭyh (DJA 44) bd. Latin fascia (OLD 677; LD 726) > φασκία (LLGE 114) > ‫ ܦܣܩܝܬܐ‬psqytʾ ‘bandage used to wrap a corpse’ (NT John 11:44 [SP]; LS2 585, 923; SL 1215), already in Tg. Jonathan pəsiqayyɔ (Isa 3:24); see also Late Jewish Literary Aramaic psyqyyhwn (Tg. Ps.-J. Num 25:1); Christian Palestinian Aramaic psqyʾ (DCPA 336; LSP 160); Jewish Palestinian Aramaic pysqy (DJPA 432) be. χαράκωμα (LSJ 1977) > ‫ ܩܠܩܘ�ܡܐ‬qlqwmʾ ‘seige engines, entrenchments’ (pre-4th cent. Deut 20:20; passim; LS2 670; SL 1375), already in Tg. Jonathan karqomɔ (1 Sam 26:7; Jastrow 669); see also Late Jewish Literary Aramaic krqwmʾ (Tg. Job 20:24; Jastrow 669); Jewish Palestinian Aramaic krkwm (DJPA 270) bf. χιλίαρχος (LSJ 1992) > ‫ ܟܠܝܪܟܐ‬klyrkʾ, ‫ ܟܝܠܝܪܟܐ‬kylyrkʾ ‘chiliarch’ (NT Matt 8:5 [S], 8 [S], 13 [S]; Mark 6:21 [SP]; John 18:12 [SP]; passim; LS2 329; SL 618), already in Nabatean klyrkʾ (DNWSI 512; see also Healey 1995: 77) bg. χρῶμα (LSJ 2012) > ‫ ܟܪܘ�ܡܐ‬krwmʾ ‘color; nature’ (4th cent. Ephrem, Maḏrɔše on the Church, 28.21 [ed. Beck 1960]; Commentary on Genesis and Part of Exodus, 127.21; 151.22, 24, 25 [ed. Tonneau 1955]; Maḏrɔše against Heresies, 32.1; 46.4; 145.18 [ed. Beck 1957a]; also in OT; LS2 347; SL 648), already in Tg. Onq. kərum (Exod 28:20; 39:13; Cook 2008: 131); see also Late Jewish Literary Aramaic krwn (Tg. Ps.-J. Exod 28:20; Jastrow 665); Jewish Palestinian Aramaic krwm (DJPA 268); Samaritan Aramaic krwm ‘gem’ (DSA 408); Jewish Babylonian Aramaic krwm (DJPA 268) bh. ψῆφος (LSJ 2022–23) > ‫ ܦܣܦܣܐ‬pspsʾ ‘small pebble; game with dice’ (4th cent. Ephrem, Maḏrɔše against Heresies, 35.26 [ed. Beck 1957a]; LS2 582; SL 1212), already in Judean Aramaic psyps ‘mosaic’ (DNWSI 922); see also Jewish Palestinian Aramaic psyps ‘mosaic stone, mosaic floor’ (DJPA 440)

Appendix 2

Citations for Verbless Clauses This appendix contains references for the data cited regarding the diachronic increase in the frequency of Pattern B verbless clauses (§8.5). Verbless Clauses with Substantival Predicates Book of the Laws of the Countries (ca. 220; ed. Drijvers 1965)

• Pattern A 25× (4.9, 15, 15–16; 6.5; 10.11; 12.8 [2×]; 14.22, 25; 18.23; 20.14, 15, 16; 22.5, 11, 22.14–15, 16, 17–18; 28.6–7, 25; 32.12; 36.7–9; 50.1 54.6; 60.23) • Pattern B 1× (12.3–4).

Acts 1–7, Acts of Thomas (ca. 200–250; ed. Wright 1871a: 2.171–251 [Syr.]) • Pattern A 34× (172.13; 178.15–16; 179.17–18, 19; 181.1 [negative], 2, 9; 183.8; 185.8; 186.17; 188.3, 5 [2×]; 195.10; 198.2, 3, 5, 6, 8, 9; 199.4–5, 11; 202.10; 213.13; 214.21–215.1; 216.21–217.1; 219.6; 220.18; 223.14; 227.6–7; 236.18; 237.2–3; 240.7–8; 248.17) • Pattern B 5× (217.2; 240.12, 13–14, 15–16; 249.3)

Demonstrations 1–3 by Aphrahaṭ (fl. 337–45; ed. Parisot 1894–1907)

• Pattern A 31× (8.5, 9, 12; 9.4 [biblical quotation], 8 [biblical quotation]; 12.5; 13.5, 16–17, 21–22 [biblical quotation], 24; 16.6, 14; 17.26; 21.15 [biblical quotation], 17; 24.14; 24.23–25.1; 25.2 [biblical quotation]; 29.8; 33.1; 45.4; 52.12, 21–22, 25; 57.5, 8; 60.24–25 [biblical quotation]; 96.10; 97.2–3; 101.19; 132.14–15 [biblical quotation]). • Pattern B 6× (8.5–6; 20.12; 24.8; 52.19; 97.14; 136.7)

Prose Refutations, Discourse 1 by Ephrem (d. 373; ed. Overbeck 1865: 21–58) • Pattern A 35× (21.12, 14; 22.6, 21; 24.26–27; 30.6, 7–8; 31.2, 4–5; 33.4, 33.27; 34.17–18; 36.4; 37.20; 38.16, 19, 26; 40.12; 41.2, 25; 44.12 [fully transformed cleft]; 44.17 [negative], 18; 47.20, 23–24 [negative], 24; 48.2, 3; 49.15–16 [negative]; 52.7; 55.4 [2×], 6; 57.24; 58.14) • Pattern B 4× (23.2–3, 6; 35.11; 58.21–22)

Teaching of Addai (ca. 420; ed. Howard 1981)

• Pattern A 31× (3*.25; 4*.1; 6*.12–13; 9*.18–19; 10*.8; 13*.12; 15*.6; 17*.21, 21–22; 18*.25; 19*.17; 20*.8–9; 24*.24, 25; 25*.5; 13–14, 21, 26*.18–19; 27*.5, 21–22; 28*.11–12, 13, 23; 29*.1 [2×], 1–2; 33*.12–13; 34*.18–19; 36*.2; 42*.4; 44*.1) • Pattern B 4× (19*.4–5, 6; 27*.1, 4)

Life of Rabbula (ca. 450; ed. Overbeck 1865: 159–209)

• Pattern A 1× (197.26) • Pattern B 15× (162.9, 27; 163.1, 3; 163.8–9 [negative], 9, 10, 12; 173.6, 18; 177.4; 183.20–21; 184.18, 20; 208.14–15)

223

224

Appendix 2

Four Letters by Philoxenos (d. 523; ed. Frothingham 1886: 28–48; Vaschalde 1902: 127–73) • Pattern A 57×: 30.12; 34.24; 34.24–36.1; 36.19–20; 42.22; 130.18–131.1 [biblical quotation]; 136.18 [2×]; 137.1 [2×], 2, 17; 138.1, 1–2, 12, 13; 139.8, 11, 14; 142.10; 143.16; 145.7; 147.2–3, 14–15; 148.8; 149.3, 17; 150.20; 151.4; 152.14–15, 18, 21; 153.1 [2×], 9, 18–19, 19, 19–20; 154.15–16; 155.3, 6, 8, 9; 156.19–20; 157.5–6, 7; 159.6; 164.6; 166.15; 167.13; 168.15–16, 17; 169.2–3, 18; 170.5, 11; 171.16 • Pattern B 34×: 28.12; 30.5; 30.18; 32.22; 34.9; 40.17; 46.18–20, 22; 129.3– 4; 130.9; 131.6–7; 132.4; 133.16–17; 134.6–7; 140.12; 143.16–17; 146.16; 147.7; 148.1–2; 149.6; 151.6; 154.5; 155.1, 11–12; 156.19; 163.6–7, 8, 9; 165.5, 7; 168.2–3, 5–6; 172.2, 10–11

Letter on Ḥimyarite Persecution by Shemʿun of Beth Arsham (d. before 548; ed. Guidi 1881) • Pattern A 11× (3*.12, 12–13, 18 [2×], 22, 22–23; 4*.20; 6*.17; 11*.7; 12*.26; 15*.1) • Pattern B 7× (1*.4–5; 2*.7–8; 3*.10, 11–12, 25–26; 9*.7, 26)

Life of Yuḥanon of Tella by Eliya (mid-6th cent.; ed. Brooks 1907: 29–95)

• Pattern A 16× (44.10–11; 55.24–25; 61.4; 64.13–14 [biblical citation]); 68.17; 71.16 [2×]; 71.25–26; 72.2; 73.24; 81.15; 84.4–5; 85.3; 86.1–2; 89.13; 91.9) • Pattern B 13× (31.8–9; 34.8–9; 45.6; 53.14; 53.18–19; 71.21; 72.9; 73.19; 78.6–7; 82.13; 84.25–26; 92.11; 94.4–5)

Lives of Eastern Saints 10, 24, 36 by Yuḥanon of Ephesus (d. ca. 589; Brooks 1923–25: 1.137–58, 2.513–26, 2.624–41) • Pattern A 3× (147.2–3, 3–4; 150.8; 154.10) • Pattern B 12× (142.4; 145.3; 145.6; 146.11–12; 147.13; 150.5; 151.2; 311.11–312.1; 314.13–315.1; 422.8–9; 423.8–9; 429.9–10)

Life of Marutha by Denḥa (d. 649; ed. Nau 1905a: 52–96)

• Pattern A 2× (72.8–9 [quotation of Gregory the Theologian]; 72.13) • Pattern B 7× (63.10; 72.5; 79.6; 86.10–11; 86.14–87.1; 91.13; 94.4)

Letter 13 and 18 by Yaʿqub of Edessa (d. 708; ed. Wright 1867: *1–*24; Rignell 1979) • Pattern A 1× (13.8*.1) • Pattern B 41× (13.2.22; 13.3.30–13.4.1; 13.5.3; 13.6.11; 13.7.3–4; 13.9.6, 8, 12; 13.12.12–13, 15, 21–23; 13.13.6–7 [biblical quotation]; 13.14.27, 29; 13.15.8, 27; 13.15.28–13.16.1; 13.16.13, 14, 29; 13.17.2– 4, 10–11, 15–16, 17–18; 13.18.19; 13.19.1–2, 5, 6; 13.20.3, 6, 10, 13, 20–21; 13.21.8–9; 13.22.5–6; 13.23.15; 18.52.9–10; 18.56.7; 18.62.2; 18.64.12–13; 18.66.13)

Verbless Clauses with Prepositional Phrase Predicates Book of the Laws of the Countries (ca. 220; ed. Drijvers 1965) • Pattern A 3× (16.6–7, 17; 58.2) • Pattern B 2× (18.1–2; 22.8)

Citations for Verbless Clauses

225

Acts 1–7, Acts of Thomas (ca. 200–250; ed. Wright 1871a: 2.171–251 [Syr.]) • Pattern A 8× (172.16; 182.18; 198.1–2; 204.17; 206.7; 209.20; 237.1–2; 244.16) • Pattern B 4× (198.8; 204.1; 205.4; 206.8)

Demonstrations 1–3 by Aphrahaṭ (fl. 337–345; ed. Parisot 1894–1907)

• Pattern A 7× (9.10 [biblical quotation], 10–11 [biblical quotation]; 49.19–20 [negative]; 61.21–22; 64.5–6; 88.14–15; 132.13–14 [biblical quotation]) • Pattern B 2× (72.18–19, 23–24; see also 61.13–14, but the predicate is probably the adverb hɔkannɔ ‘thus’)

Prose Refutations, Discourse 1 by Ephrem (d. 373; ed. Overbeck 1865: 21–58) • Pattern A 7× (39.10 [negative]; 40.17; 43.6–7, 23, 24–25; 55.23–24; 56.21–22) • Pattern B 3× (34.25 [negative]; 44.11; 46.27)

Teaching of Addai (ca. 420; ed. Howard 1981) • Pattern A 1× (47*.5–6) • Pattern B 1× (21*.19)

Life of Rabbula (ca. 450; ed. Overbeck 1865: 159–209) • Pattern A 1× (192.6 [biblical quotation]) • Pattern B 1× (195.19)

Four Letters by Philoxenos (d. 523; ed. Frothingham 1886: 28–48; Vaschalde 1902: 127–73) • Pattern A 2× (148.11–12 [biblical quotation]; 158.16) • Pattern B 21× (34.23 [biblical quotation]; 44.4, 8; 130.12–13; 131.8–9; 133.12; 135.5; 138.7; 140.10; 141.13; 148.2; 149.18; 150.3–4; 151.4; 158.10; 161.14; 163.10, 13, 14; 165.17; 171.22–172.1)

Letter on Ḥimyarite Persecution by Shemʿun of Beth Arsham (d. before 548; ed. Guidi 1881) • Pattern A 0× • Pattern B 2× (10*.18–19, 26)

Life of Yuḥanon of Tella by Eliya (mid-6th cent.; ed. Brooks 1907: 29–95)

• Pattern A 3× (77.9–10; 83.24 [biblical quotation]; 91.11–12) • Pattern B 17× (31.4; 32.24–25; 34.17–18; 42.6; 43.22; 61.20–21; 69.3, 7–8; 70.14; 72.10, 10–11, 11 [2×]; 76.13; 82.18–19; 95.6, 12)

Lives of Eastern Saints 10, 24, 36 by Yuḥanon of Ephesus (d. ca. 589; Brooks 1923–25: 1.137–58, 2.513–26, 2.624–41) • Pattern A 3× (145.6, 9; 150.4) • Pattern B 7× (142.5; 146.8–9; 147.14; 150.6; 317.8; 318.9; 424.1)

Life of Marutha by Denḥa (d. 649; ed. Nau 1905a: 52–96) • Pattern A 0× • Pattern B 2× (71.9; 83.6)

Letter 13 and 18 by Yaʿqub of Edessa (d. 708; ed. Wright 1867: *1–*24; Rignell 1979) • Pattern A 2× (13.13.5–6 [biblical quotation]; 13.16.16) • Pattern B 10× (13.2.24; 13.3.1–2; 13.8.19; 13.14.22–23; 13.15.22–23; 13.16.3, 16–17; 13.21.30, 13.22.23; 13.23.26)

Bibliography Aartun, K. 1974 Die Partikeln des Ugaritischen. Kevelaer. Abū ʿAssāf, ʿAlī 1972 Kitābāt suryāniyya jadīda fī al-matḥaf al-waṭanī bidimašq. Les annales archeologiques arabes syriennes 22: 135–44. Adams, J. N. 2003 Bilingualism and the Latin Language. Cambridge. Aggoula, B. 1992 Studia Aramaica III. Syria 59: 391–422. Aikhenvald, A. Y. 1996 Areal Diffusion in North-West Amazonia: The Case of Tariana. Anthropological Linguistics 38: 73–116. 2002 Language Contact in Amazonia. Oxford. 2003 Mechanisms of Change in Areal Diffusion: New Morphology and Language Contact. Journal of Linguistics 39: 1–29. Aikhenvald, Alexandra Y., and R. M. W. Dixon 2001a Areal Diffusion and Genetic Inheritance: Problems in Comparative Linguistics. Oxford. 2001b Introduction. Pp. 1–26 in Aikhenvald and Dixon 2001a. 2006 Grammars in Contact. Oxford. Alexander, P. S. 1995 Bavli Berakhot 55a–57a: The Talmudic Dreambook in Context. JJS 46: 230–48. Al-Jallad, A. forthcoming  Graeco-Arabica I: The Southern Levant. In Le context de naissance de l’écriture arabe: Écrit et écritures araméennes et arabes au 1er millénaire après J.-C., ed. F. Briquel-Chatonnet, M. Debié, and L. Nehmé. OLA. Louvain. Al-Jallad, A., R. Daniel, and O. al-Ghul 2013 The Arabic Toponyms and Oikonyms in 17. Pp. 23–48 in The Petra Papyri, vol. 2, ed. L. Koenen et al. Amman. Allen, P., and C. T. R. Hayward 2004 Severus of Antioch. London. Allen, W. S. 1987 Vox Graeca: A Guide to the Pronunciation of Classical Greek. 3rd ed. Cambridge.

226

Bibliography

227

Alwan, K. 1989 Jacques de Saroug: Quatre homélies métriques sur la Création. CSCO 508–9. Louvain. Amar, J. P. 2011 The Syriac Vita Tradition of Ephrem the Syrian. CSCO 629–30. Louvain. Appel, R., and P. Muysken 1987 Language Contact and Bilingualism. London. Assemani, G. S. 1719–28  Bibliotheca Orientalis Clementino-Vaticana. Rome. Attridge, H. W. 1990 The Original Language of the Acts of Thomas. Pp. 241–50 in Of Scribes and Scrolls: Studies on the Hebrew Bible, Intertestamental Judaism, and Christian Origins Presented to John Strugnell on the Occasion of His Sixtieth Birthday, ed. H. W. Attridge, J. J. Collins, and T. H. Tobin. Lanham, MD. Avinery, I. 1975 The Position of the Demonstrative Pronoun in Syriac. JNES 34: 123–27. 1976 The Position of Declined kl in Syriac. Afroasiatic Linguistics 3: 108–9. 1984 The Position of Declined kl in Syriac. JAOS 104: 333. Baarda, T. 1975 The Gospel Quotations of Aphrahat, the Persian Sage: Aphrahat’s Text of the Fourth Gospel. Amsterdam. Baasten, M. F. J., and W. T. van Peursen 2003 Hamlet on a Hill: Semitic and Greek Studies Presented to Professor T. Muraoka on the Occasion of His Sixty-Fifth Birthday. OLA 118. Louvain. Baehrens, W. A. 1922 Sprachlicher Kommentar zur vulgärlateinischen Appendix Probi. Halle. Bakker, E. J. 1993 Boundaries, Topics, and the Structure of Discourse. An Investigation of the Ancient Greek Particle δέ. Studies in Language 17: 275–311. Bakker, P. 1994 Michif, the Cree-French Mixed Languages of the Métis Buffalo Hunters in Canada. Pp. 13–33 in Bakker and Mous 1994. 1997 A Language of Our Own: The Genesis of Michif, the Mixed Cree-French Language of the Canadian Métis. Rev. ed. Oxford Studies in Anthropological Linguistics 10. New York. 2003 Mixed Languages as Autonomous Systems. Pp. 107–50 in Matras and Bakker 2003. Bakker, P., and M. Mous 1994 Mixed Languages: 15 Case Studies in Language Intertwining. Amsterdam. Bakker, P., and R. A. Papen 1997 Michif: A Mixed Language Based on Cree and French. Pp. 295–363 in Thomason 1997. Bardenhewer, O. 1924 Geschichte der altkirchlichen Litteratur, vol. 4. Freiburg im Breisgau. Bardy, G. 1948 La question des langues dans l’église ancienne. Paris.

228

Bibliography

Barsoum, I. A. 2003 The Scattered Pearls: A History of Syriac Literature and Sciences, trans. M. Moosa. 2nd ed. Piscataway, NJ. Bátori, I. 1979 Russen und Finnougrier: Zweisprachigkeit und sprachliche Interferenz. Pp. 1–26 in Standardsprache und Dialekte in mehrsprachigen Gebieten Europas: Akten des 2. Symposions über Sprachkontakt in Europa, Mannheim 1978, ed. P. S. Ureland. Tübingen. Bauer, H., and P. Leander 1927 Grammatik des biblisch-Aramäischen. Halle. Baumstark, A. 1922 Geschichte der syrischen Literatur. Bonn. Beck, E. 1955 Des heiligen Ephraem des Syrers Hymnen de Fide. CSCO 154–55. Louvain. 1957a Des heiligen Ephraem des Syrers Hymnen contra Haereses. CSCO 169–70. Louvain. 1957b Des heiligen Ephraem des Syrers Hymnen de Paradiso und Contra Julianum. CSCO 174–75. Louvain. 1959 Des heiligen Ephraem des Syrers Hymnen de Nativitate (Epiphania). CSCO 186–87. Louvain. 1960 Des heiligen Ephraem des Syrers Hymnen de Ecclesia. CSCO 198–99. Louvain. 1961a Des heiligen Ephraem des Syrers Carmina Nisibena, vol. 1. CSCO 218–19. Louvain. 1961b Des heiligen Ephraem des Syrers Sermones de Fide. CSCO 212–13. Louvain. 1962 Des heiligen Ephraem des Syrers Hymnen de Virginitate. CSCO 223–24. Louvain. 1963 Des heiligen Ephraem des Syrers Carmina Nisibena, vol. 2. CSCO 240–41. Louvain. 1964a Des heiligen Ephraem des Syrers Paschahymnen (de Azymis, de Crucifixione, de Resurrectione). CSCO 248–49. Louvain. 1964b Des heiligen Ephraem des Syrers Hymnen de Ieiunio. CSCO 246–47. Louvain. 1966 Des heiligen Ephraem des Syrers Sermo de Domino Nostro. CSCO 270–71. Louvain. 1970a Des heiligen Ephraem des Syrers Sermones, vol. 1. CSCO 305–6. Louvain. 1970b Des heiligen Ephraem des Syrers Sermones, vol. 2. CSCO 311–12. Louvain. 1979 Ephraem Syrus: Sermones in Hebdomadam Sanctam. CSCO 411–12. Louvain. Becker, A. H. 2006 Fear of God and the Beginning of Wisdom: The School of Nisibis and Christian Scholastic Culture in Late Antique Mesopotamia. Philadelphia. Bedjan, P. 1890–97  Acta Martyrum et Sanctorum. 7 vols. Paris. 1891 Histoire complète de Joseph par Saint Ephrem. 2nd ed. Paris.

Bibliography

229

1895 Histoire de Mar-Jabalaha, de trois autres patriarches, d’un prêtre et de deux laïques, nestoriens. Paris. 1903 Homiliae S. Isaaci, vol. 1. Leipzig. 1905–10  Homiliae Selectae Mar-Jacobi Sarugensis. 4 vols. Paris. 1909 Mar Isaacus Ninivita: De perfectione religiosa. Paris. Beeston, A. F. L. 1984 Sabaic Grammar. Manchester. Beeston, A. F. L., M. A. Ghul, W. W. Müller, and J. Ryckmans 1982 Sabaic Dictionary (English-French-Arabic). Louvain-la-Neuve and Beirut. Bell, G. 1982 The Churches and Monasteries of the Tur ʿAbdin. London. Bellinger, A. R., and C. B. Welles 1935 A Third-Century Contract of Sale from Edessa in Osrhoene. Yale Classical Studies 5: 95–154 with pls. 1–3. Ben-Ḥayyim, Z. 1967 The Literary and Oral Tradition of Hebrew and Aramaic amongst the Samaritans, vol. 3/2: The Recitation of Prayers and Hymns. Jerusalem. [Hebrew] 1988 ‫( תיבת מרקה‬Tibåt Mårqe): A Collection of Samaritan Midrashim. Jerusalem. [Hebrew] Benedictus, P. 1732–46  Sancti patris nostri Ephraem Syri Opera omnia quae exstant Graece, Syriace, Latine. Rome. Beyer, K. 1966 Der reichsaramäische Einschlag in der ältesten syrischen Literatur. ZDMG 116: 242–54. 1984 Die aramäischen Texte vom Toten Meer. Göttingen. 1986 The Aramaic Language: Its Distribution and Subdivisions, trans. J. Healey. Göttingen. Biberstein-Kazimirski, A. de 1860 Dictionnaire arabe-français. Paris. Bickell, G. 1873–77  Sancti Isaaci Antiochi doctoris Syrorum opera omnia, syriace, arabice­que primus edidit, latine vertit. Giessen. Bidez, J. 1983–96  Sozomène: Histoire ecclésiastique. SC 306, 418, 495, 516. Paris. Biesen, K. den 2011 Annotated Bibliography of Ephrem the Syrian. 2nd ed. Published by the author. Binder, Mathias 2014 ‘Your Death, O God’: Christological and Other Adaptations of Šubḥalmaran’s End Time Book in the Syriac Manuscript M20N from Mount Sinai. JEastCS 66: 1–35. Björkman, E. 1902 Scandinavian Loan-Words in Middle English, Part 2. Halle a. S.

230

Bibliography

Blau, J. 1969 The Origins of Open and Closed e in Proto-Syriac. BSOAS 32: 1–9. [= Blau 1998: 299–307] 1970 On Pseudo-Corrections in Some Semitic Languages. Jerusalem. 1972 Marginalia Semitica II. IOS 2: 58–82. [= Blau 1998: 221–46] 1998 Topics in Hebrew and Semitic Linguistics. Jerusalem. Bloomfield, L. 1933 Language. New York. Blum, G. G. 1969 Rabbula von Edessa: Der Christ, der Bischof, der Theologe. CSCO 300. Louvain. 1983 Zum Bau von Abschnitten in Memre von Jacob von Sarug. Pp. 307–21 in IIIo Symposium Syriacum 1980: Les contacts du monde syriaque avec les autres cultures, ed. R. Lavenant, S.J. OCA 221. Rome. Bonnet, M. 1903 Acta Philippi et Acta Thomae accedunt Acta Barnabae. Leipzig. Bordreuil, P., and D. Pardee 2009 A Manual of Ugaritic. LSAWS 3. Winona Lake, IN. Bou Mansour, T. 2003 Une clé pour la distinction des écrits des Isaac d’Antioche. ETL 79: 365–402. Bowden, John 2005 Language Contact and Metatypic Restructuring in the Directional System of North Maluku Malay. Concentric: Studies in Linguistics 31: 133–58. Bowersock, G. W. 1990 Hellenism in Late Antiquity. Ann Arbor, MI. 2000 The Syriac Life of Rabbula and Syrian Hellenism. Pp. 255–71 in Greek Biography and Panegyric in Late Antiquity, ed. T. Hägg and P. Rousseau. Berkeley, CA. Boyarin, D. 1978 On the History of the Babylonian Jewish Aramaic Reading Traditions: The Reflexes of *a and *ā. JNES 37: 141–60. 1981 An Inquiry into the Formation of the Middle Aramaic Dialects. Pp. 613–49 in Bono homini donum: Essays in Historical Linguistics in Memory of J. Alexander Kerns, vol. 2, ed. Y. L. Arbeitman and A. R. Bomhard. Amsterdam. Braun, O. 1901 Ein Brief des Katholikos Timotheos I über biblische Studien des 9. Jahrhunderts. OC 1: 299–313. Bremmer, J. N. 2001a The Apocryphal Acts of Thomas. Louvain. 2001b The Acts of Thomas: Place, Date and Women. Pp. 74–90 in Bremmer 2001a. Brendemoen, B., E. Lanza, and E. Ryen 1999 Language Encounters across Time and Space: Studies in Language Contact. Oslo. Briant, P. 1978 Colonisation hellénistique et populations indigènes: La phase d’installation. Klio 60: 57–92.

Bibliography

231

1979 Des Achéménides aux rois hellénistiques: Continuités et ruptures. Annali della Scuola Normale Superiore di Pisa 9: 1375–1414. 1999 Alexandre et l’héritage achéménide: Quelques réflexions et perspectives. Pp. 209–17 in International Congress, Alexander the Great: From Macedonia to the Oikoumene, Veroia 27–31/5/1998. Veroia. 2002 From Cyrus to Alexander: A History of the Persian Empire. Winona Lake, IN. Briquel Chatonnet, F., and A. Desreumaux 2011a Syriac Inscriptions in Syriac. Hugoye 14: 27–44. 2011b Oldest Syriac Christian Inscription Discovered in North-Syria. Hugoye 14: 45–61. Brock, S. P. 1967 Greek Words in the Syriac Gospels (vet and pe). Le Muséon 80: 389–426. 1975 Some Aspects of Greek Words in Syriac. Pp. 80–108 in Synkretismus im syrisch-persischen Kulturgebiet, ed. A. Dietrich. Göttingen. [= Brock 1984: iv] 1976 Ephrem’s Letter to Publius. Le Muséon 89: 261–305. 1979a Aspects of Translation Technique in Antiquity. GRBS 20: 69–87. [= Brock 1984: iii] 1979b Jacob of Edessa’s Discourse on the Myron. OC 63: 20–36. 1982a From Antagonism to Assimilation: Syriac Attitudes to Greek Learning. Pp. 17–34 in Garsoïan, Mathews, and Thomson 1982. [= Brock 1984: v] 1982b The Homily by Marutha of Tagrit on the Blessing of the Waters at Epiphany. OC 66: 51–74. 1984 Syriac Perspectives on Late Antiquity. London. 1989a Some Observations on the Use of Classical Syriac in the Late Twentieth Century. JSS 34: 363–75. 1989b Three Thousand Years of Aramaic Literature. ARAM 1: 11–23. 1990 Diachronic Aspects of Syriac Word Formation: An Aid for Dating Anonymous Texts. Pp. 321–33 in V Symposium Syriacum 1988, ed. R. Lavenant. Rome. 1991 Some New Syriac Documents from the Third Century ad. ARAM 3: 259–67. 1992 The Luminous Eye: The Spiritual World Vision of Saint Ephrem. 2nd ed. Kalamazoo, MI. 1994 Greek and Syriac in Late Antique Syriac. Pp. 149–60 in Literacy and Power in the Ancient World, ed. A. K. Bowman and G. Woolf. Cambridge. [= Brock 1999a: i] 1995 Isaac of Nineveh (Isaac the Syrian): ‘The Second Part’, Chapters IV–XLI. CSCO 554–55. Louvain. 1996 Greek Words in Syriac: Some General Features. Scripta classica Israelica 15: 251–62. [= Brock 1999a: xv] 1997 A Brief Outline of Syriac Literature. Bakerhill and Kottayam. 1998 Syriac and Greek Culture. Pp.  708–19 in Cambridge Ancient History, vol. 13: The Late Empire, a.d. 337–425, ed. A. Cameron and P. Garnsey. Cambridge. 1999a From Ephrem to Romanos: Interactions between Syriac and Greek in Late Antiquity. Aldershot.

232

Bibliography

1999b Fragments of Ps. John Chrysostom, Homily on the Prodigal Son, in Christian Palestinian Aramaic. Le Muséon 112: 333–62. 1999–2000  Greek Words in Ephrem and Narsai: A Comparative Sampling. ARAM 11–12: 439–49. 2003 Some Diachronic Features of Classical Syriac. Pp. 95–111 in Baasten and van Peursen 2003. 2004 Secondary Formations from Greek Loanwords in Syriac. Pp. 31–39 in Verbum et calamus: Semitic and Related Studies in Honour of the Sixtieth Birthday of Professor Tapani Harviainen, ed. H. Juusola, J. Laulainen, and H. Palva. Helsinki. 2005 Greek and Latin Words in Palmyrene Inscriptions: A Comparison with Syriac. Pp. 11–25 in A Journey to Palmyra: Collected Essays to Remember Delbert R. Hillers, ed. E. Cussini. Leiden. 2007a Translation, Greek and Syriac. Pp. 935–46 and 957–59 in A History of Ancient Greek: From the Beginnings to Late Antiquity, ed. A.-F. Christidis. Cambridge. 2007b A Brief Guide to the Main Editions and Translations of the Works of Saint Ephrem. Pp. 281–338 in Saint Éphrem: Un poète pour notre temps. Antélias. 2008 The History of the Holy Mar Maʿin with a Guide to the Persian Martyr Acts. Persian Martyr Acts in Syriac: Text and Translation 1. Piscataway, NJ. 2010 A Criterion for Dating Undated Syriac Texts: The Evidence from Adjectival Forms in -aya. Parole de l’Orient 35: 111–24. 2011 A Soghitha on the Daughter of Jephtha, by Isaac. Hugoye 14: 3–25. Brock, S. P., A. M. Butts, G. A. Kiraz, and L. Van Rompay 2011 Gorgias Encyclopedic Dictionary of the Syriac Heritage. Piscataway, NJ. Brockelmann, C. 1893 Die griechischen Fremdwörter im Armenischen. ZDMG 47: 1–42. 1895 Lexicon Syriacum. Berlin. 1908 Grundriß der vergleichenden Grammatik der semitischen Sprachen, vol. 1: Laut- und Formenlehre. Berlin. 1928 Lexicon Syriacum. 2nd ed. Halis Saxonum. [trans. Sokoloff 2009] 1981 Syrische Grammatik. 13th ed. Leipzig. Brooks, E. W. 1907 Vitae virorum apud Monophysitas celeberrimorum. CSCO 7–8. Louvain. 1919–24  Historia ecclesiastica Zachariae Rhetori vulgo adscripta, vols. 1–2. CSCO 83–84, 87–88. Louvain. 1923–25  John of Ephesus: Lives of the Eastern Saints, vols. 1–3. PO 17/1; 18/4; 19/2. Paris. 1935 Iohannis Ephesini Historiae Ecclesiasticae, Pars Tertia. CSCO 105. Louvain. Brown, P. 1971 The World of Late Antiquity, ad 150–750. New York. Browning, R. 1983 Medieval and Modern Greek. 2nd ed. Cambridge. Bruns, P. 1991 Aphrahat: Unterweisungen, vols. 1–2. Fontes Christiani 5/1–2. Freiburg.

Bibliography

233

Buber, S. 1899 Midrash Ekhah rabbah. Vilna. Buccini, A. F. 1992 The Development of Umlaut and the Dialectal Position of Dutch in Germanic. Ph.D. diss., Cornell University. Budge, E. A. W. 1894 The Discourses of Philoxenus, Bishop of Mabbôgh, a.d. 485–519. London. Burtea, B. 2004 Šarh ḏ-paruanaiia: A Mandaean Ritual Explanatory Commentary. ARAM 16: 85–93. 2011 Mandaic. Pp. 670–85 in Weninger 2011. Burkitt, F. C. 1913 Euphemia and the Goth, with the Acts of Martyrdom of the Confessors of Edessa. London. Buth, R. 1990 ʾĕdayin/tote: Anatomy of a Semitism in Jewish Greek. MAARAV 5–6: 33–48. Butts, A. M. 2006 Observations on the Verbless Clause in the Language of Neophyti I. Aramaic Studies 4: 53–67. 2009a The Biography of the Lexicographer Ishoʿ bar ʿAli (ʿĪsā b. ʿAlī). OC 93: 60–71. 2009b The Afflictions of Exile: A Syriac Memrā by David Puniqāyā. Le Muséon 122: 53–80. 2010 The Etymology and Derivation of the Syriac Adverbial Ending -ɔʾiθ. JNES 69: 79–86. 2012 Review of Cook 2008. JNES 71: 61–62. 2013 Greek μέν in Early Syriac. Hugoye 16: 211–23. 2014 The Use of Syriac Derivational Suffixes with Greek Loanwords. Or 83: 207–237. 2015a Semitic Languages in Contact. Studies in Semitic Languages and Linguistics 82. Leiden. 2015b The Use of syāmē as a Phonological Marker in Syriac. Hugoye 18: 93–109. 2015c Between Aramaic *ʾiðayn and Greek δέ: The Linguistic History of Syriac den. Pp. 13–30 in Syriac in Its Multi-Cultural Context, ed. H. Teule et al. Eastern Christian Studies. Louvain. forthcoming a  Latin Words in Classical Syriac. Hugoye. forthcoming b  Old Syriac. In Brill Encyclopedia of Early Christianity, ed. Paul J. J. van Geest and Bert Jan Lietaert Peerbolte. Leiden. forthcoming c  The Etymology of Aramaic (and Hebrew) √prns ‘to distribute, supply’. JAOS. forthcoming d  The Integration of Consonants in Greek Loanwords in Syriac. Cameron, A. 2012 The Mediterranean World in Late Antiquity, 395–700 a d. 2nd ed. Abingdon. Campbell, L. 1993 On Proposed Universals of Grammatical Borrowing. Pp. 91–109 in Historical Linguistics 1989, ed. H. Aertsen and R. J. Jeffers. Amsterdam.

234

Bibliography

Canali de Rossi, F. 2004 Iscrizioni dello estremo oriente greco: Un Repertorio. Bonn. Canivet, P. 1958 Théodoret de Cyr: Thérapeutique des maladies helléniques. SC 57. Paris. 1977 Le monachisme syrien selon Théodoret de Cyr. TH 42. Paris. Canivet, P., and A. Leroy-Molinghen 1977–79  Théodoret de Cyr: Histoire des moines de Syrie. SC 234, 257. Paris. Cantineau, J. 1935 Grammaire du palmyrénien épigraphique. Cairo. 1951 Le consonantisme du sémitique. Sem 4: 79–94. Carlson, T. 2011 A Light from ‘The Dark Centuries’: Isḥaq Shbadnaya’s Life and Works. Hugoye 14: 191–214. Chabot, J.-B. 1899–1910  Chronique de Michel le Syrien, patriarche jacobite d’Antioche (1166– 1199). 4 vols. Paris. 1916 Chronicon ad annum Christi 1234 pertinens, vol. 1. CSCO 81. Louvain. 1927–49  Chronicon anonymum Pseudo-Dionysianum vulgo dictum. CSCO 91, 104, 121. Louvain. 1953 Iacobi Edesseni Hexaemeron. CSCO 92, 97. Louvain. Charlesworth, J. H. 1973 The Odes of Solomon. Oxford. Chialà, S. 2011 Isacco di Ninive: Terza collezione. CSCO 637–38. Louvain. Ciancaglini, C. A. 2006 L’origine delle locuzioni verbali con ʿbd in siriaco. Pp. 173–84 in Loquentes linguis: Studi linguistici e orientali in onore di Fabrizio A. Pennacchietti, ed. P. G. Borbone, A. Mengozzi, and M. Tosco. Wiesbaden. 2008 Iranian Loanwords in Syriac. Wiesbaden. Clyne, M. G. 1967 Transference and Triggering: Observations on the Language Assimilation of Postwar German-Speaking Migrants in Australia. The Hague. Coakley, J. F. 2011 When Were the Five Greek Vowel-Signs Introduced into Syriac Writing? JSS 56: 307–25. Contini, R., and P. Pagano 2015 Notes on Foreign Words in Hatran Aramaic. Pp. 126–57 in Butts 2015a. Cook, E. M. 2008 A Glossary of Targum Onkelos: According to Alexander Sperber’s Edition. Leiden. Corluy, J. 1886 Acta sancti Mar Abduʾl Masich, aramaice et latine, edidit nunc primum ex cod. Londinensi (Addit. mss. 12174). AB 5: 5–52. Corne, C. 1999 From French to Creole: The Development of New Vernaculars in the French Colonial World. Westminster Creolistics Series 5. London.

Bibliography

235

Cotton, H. M., W. E. H. Cockle, and F. G. B. Millar 1995 The Papyrology of the Roman Near East: A Survey. The Journal of Roman Studies 85: 214–35. Cross, F. M., Jr., and D. N. Freedman 1952 Early Hebrew Orthography: A Study of the Epigraphic Evidence. New Haven, CT. Crum, W. E. 1929–39  A Coptic Dictionary. Oxford. Csató, É. Á. 2001 Syntactic Code-Copying in Karaim. Pp. 271–83 in The Circum-Baltic Languages: Typology and Contact, ed. Ö. Dahl and M. Koptjevskaja-Tamm. Amsterdam. 2002 Karaim: A High-Copying Language. Pp. 315–27 in Jones and Esch 2002. Cureton, W. 1864 Ancient Syriac Documents. London. Curnow, T. J. 2001 What Language Features Can Be ‘Borrowed’? Pp. 412–36 in Aikhenvald and Dixon 2001a. Dalman, G. 1905 Die Grammatik des jüdisch-palästinischen Aramäisch. 2nd ed. Leipzig. Daniels, P. T. 1993 Classical Syriac Phonology. Pp. 127–40 in Kaye 1997: vol. 1. Daris, Sergio 1991 Il lessico Latino nel Greco d’Egitto. 2nd ed. Barcelona. Datema, C. 1978 Amphilochii Iconiensis Opera: Orationes, pluraque alia quae supersunt, nonnulla etiam spuria. CCSG 3. Turnhout. Degen, R. 1969 Altaramäische Grammatik. Wiesbaden. Denniston, J. D. 1996 The Greek Particles. 2nd ed. London. Díez Macho, A. 1968–79  Neophyti 1 Targum Palestinense ms de la Biblioteca Vaticana. Madrid. Díez Merino, L. 1992 The Adverb in Qumran Aramaic. Pp. 22–47 in Studies in Qumran Aramaic, ed. T. Muraoka. Louvain. Dillon, M. 1945 Linguistic Borrowing and Historical Evidence. Language 21: 12–17. Dillmann, A. 1907 Ethiopic Grammar, ed. C. Bezold, trans. J. A. Crichton. 2nd ed. London. Dion, P-.E. 1974 La Langue de Yaʾudi. Waterloo. Donner, H., and W. Röllig 1969–73  Kanaanäische und aramäische Inschriften mit einem Beitrag von O.  Röss­ler. Wiesbaden.

236

Bibliography

Doran, R. 2006 Stewards of the Poor: The Man of God, Rabbula, and Hiba in Fifth-Century Edessa. CSS 208. Kalamazoo, MI. Dorian, N. 1993 Internally and Externally Motivated Change in Language Contact Settings: Doubts about Dichotomy. Pp. 131–155 in Historical Linguistics: Problems and Perspectives, ed. C. Jones. London. Draguet, R. 1972 Le commentaire du livre d’Abba Isaïe (logoi I–XV) par Dadišoʿ Qaṭraya (VIIe s.). CSCO 326–27. Louvain. 1980 La vie primitive de S. Antoine. CSCO 417–18. Louvain. Drijvers, H. J. W. 1965 The Book of the Laws of the Countries. Assen. 1966 Bardaiṣan of Edessa. Assen. 1970 Bardaiṣan of Edessa and the Hermetica: The Aramaic Philosopher and the Philosophy of His Time. JEOL 21: 190–210. 1977 Hatra, Palmyra und Edessa. ANRW II.8: 799–906. 1980 Cults and Beliefs at Edessa. Leiden. 1992 Syrian Christianity and Judaism. Pp. 124–46 in The Jews among Pagans and Christians, ed. J. Lieu, J. North, and T. Rajak. London. 1998 Syriac Culture in Late Antiquity: Hellenism and Local Traditions. Mediterraneo Antico 1: 95–113. Drijvers, H. J. W., and J. F. Healey 1999 The Old Syriac Inscriptions of Edessa and Osrhoene. Leiden. Drijvers, J. W. 1999 Rabbula, Bishop of Edessa: Spiritual Authority and Secular Power. Pp. 139– 154 in Drijvers and Watt 1999. Drijvers, J. W., and J. W. Watt 1999 Portraits of Spiritual Authority: Religious Power in Early Christianity, Byzantium and the Christian Orient. Religions in the Graeco-Roman World 137. Boston. Drower, E. S., and R. Macuch 1963 A Mandaic Dictionary. Oxford. Duval, R. 1888–1901  Lexicon Syriacum auctore Hassano bar Bahlule, vols. 1–3. Paris. 1904–5  Išōʿyahb Patriarchae III: Liber Epistularum. CSCO 11–12. Louvain. 1907 La littérature syriaque. 3rd ed. Bibliothèque de l’enseignement de l’histoire ecclésiastique: Anciennes littératures chrétiennes 2. Paris. Ebied, R. Y., and L. R. Wickham 1975 A Collection of Unpublished Syriac Letters of Cyril of Alexandria. CSCO 359–60. Louvain. Eck, W. 2000 Latein als Sprache politischer Kommunikation in Städten der östlichen Provinzen. Chiron 30: 641–60.

Bibliography

237

Emeneau, M. B. 1962 Bilingualism and Structural Borrowing. Proceedings of the American Philosophical Society 106: 430–42. Faber, A. 1981 Phonetic Reconstruction. Glossa 15: 233–62. 1985 Akkadian Evidence for Proto-Semitic Affricates. JCS 37: 101–7. Feissel, D., and J. Gascou 1989 Documents d’archives romains inédits du Moyen-Euphrate (IIIe siècle après J.-C.). CRAIBL 1989: 535–61. 1995 Documents d’archives romains inédits du moyen Euphrate (IIIe s. après J.-C.), part 1: Les pétitions (P. Euphr. 1 à 5). Journal des Savants 1995: 65–119. 2000 Documents d’archives romains inédits du moyen Euphrate (IIIe s. après J.-C.), part 3: Actes diverses et lettres (P. Euphr. 11 à 17). Journal des Savants 2000: 157–208. Feissel, D., J. Gascou, and J. Teixidor 1997 Documents d’archives romains inédits du moyen Euphrate (IIIe s. après J.-C.), part 2: Les actes de vente-achat (P. Euphr. 6 à 10). Journal des Savants 1997: 3–57. Ferguson, C. 1959 Diglossia. Word 15: 325–40. Fiey, J.-M. 1968 Notule de littérature syriaque: La Démonstration XIV d’Aphraate. Le Muséon 81: 449–54. Fischer, W. 2002 A Grammar of Classical Arabic. New Haven, CT. Fitzmyer, J. A. 1979a Wandering Aramean: Collected Aramaic Essays. Missoula, MT. 1979b The Phases of the Aramaic Language. Pp. 57–84 in Fitzmyer 1979a. 2004 The Genesis Apocryphon. 3rd ed. Rome. Fitzmyer, J. A., and D. J. Harrington 1978 A Manual of Palestinian Aramaic Texts (Second Century b.c.–Second Century a.d.). Rome. Flom, G. T. 1905 Review of Björkman 1902. The Journal of English and Germanic Philology 3: 422–26. Foley, W. A. 1986 The Papuan Languages of New Guinea. Cambridge. Folmer, M. 1995 The Aramaic Language in the Achaemenid Period: A Study in Linguistic Variation. OLA 68. Louvain. Förster, H. 2002 Wörterbuch der griechischen Wörter in den koptischen dokumentarischen Texten. Texte und Untersuchungen zur Geschichte der altchristlichen Literatur 148. Berlin. Fox, J. 2003 Semitic Noun Patterns. HSS 52. Winona Lake, IN.

238

Bibliography

Fraade, S. 1984 Enosh and His Generation: Pre-Israelite Hero and History in Postbiblical Interpretation. Chico, CA. 2012 Language Mix and Multilingualism in Ancient Palestine: Literary and Inscriptional Evidence. Jewish Studies 48: 1*–40*. 2014 The Rehov Inscriptions and Rabbinic Literature: Matters of Language. Pp. 225–38 in Talmuda de-Eretz Israel: Archaeology and the Rabbis in Late Antique Palestine, ed. Steven Fine and Aaron Koller. Studia Judaica 73. Boston. Fränkel, S. 1903 Beiträge zum syrischen Wörterbuche. ZA 17: 85–90. Frankenberg, W. 1912 Evagrius Ponticus. Berlin. Frishman, J. 1992 The Ways and Means of the Divine Economy: An Edition, Translation and Study of Six Biblical Homilies by Narsai. Ph.D. diss., Leiden University. Frothingham, A. L. 1886 Stephen bar Sudaili: The Syrian Mystic and the Book of Hierotheos. Leiden. Garr, W. R. 2004 A Dialect-Geography of Syria–Palestine, 1000 to 586 b .c.e. Winona Lake, IN. Garsoïan, N. G., T. F. Mathews, and R. W. Thomson 1982 East of Byzantium: Syria and Armenia in the Formative Period. Washington, DC. Gascou, J. 2009 The Papyrology of the Near East. Pp. 473–94 in The Oxford Handbook of Papyrology, ed. R. S. Bagnall. New York. 2012 The Diversity of Languages in Dura-Europos. Pp. 74–96 in Edge of Empires: Pagans, Jews, and Christians at Roman Dura-Europos, ed. Jennifer Y. Chi and Sebastian Heath. Princeton. Gelb, I. J., et al. 1956–2006  The Assyrian Dictionary of the Oriental Institute of the University of Chicago. Chicago. Gensler, O. D. 2000 Why Semitic Adverbializers (Akkadian -iš, Syriac ‑āʾīṯ) Should Not Be Derived from Existential *ʾīṯ. JSS 45: 233–65. Ghanem, J. R. 1970 The Biography of John of Tella (d. a.d. 537) by Elias. Ph.D. diss., University of Wisconsin, Madison. Gibson, J. C. L. 1975 Textbook of Syrian Semitic Inscriptions, vol. 2: Aramaic Inscriptions, Including Inscriptions in the Dialect of Zenjirli. Oxford. Gignac, F. T. 1976–81  A Grammar of the Greek Papyri of the Roman and Byzantine Periods. 2 vols. Milan.

Bibliography

239

Glare, P. G. W. 1982 Oxford Latin Dictionary. Oxford. Gołąb, Z. 1959 Some Arumanian-Macedonian Isogrammatisms and the Social Background of Their Development. Word 15: 415–35. Goldenberg, G. 1971 Tautological Infinitive. IOS 1: 36–85. [= Goldenberg 1998: 66–115] 1977 The Semitic Languages of Ethiopia and Their Classification. BSOAS 40: 461–507. [= Goldenberg 1998: 286–332] 1983 On Syriac Sentence Structure. Pp. 97–140 in Arameans, Aramaic, and the Aramaic Literary Tradition, ed. Michael Sokoloff. Ramat-Gan. [= Goldenberg 1998: 525–68] 1987–88  The Contribution of Semitic Languages to Linguistic Thinking. Jaar­ bericht van het Vooraziatisch-Egyptisch Genootschap ‘Ex Oriente Lux’ 30: 107–15. [= Goldenberg 1998: 1–9] 1990 On Some Niceties of Syriac Syntax. Pp. 335–44 in V Symposium Syriacum 1988, ed. R. Lavenant, S.J. OCA 236. Rome. [= Goldenberg 1998: 579–90] 1991 On Predicative Adjectives and Syriac Syntax. BO 48: 716–26. [= Goldenberg 1998: 556–78] 1998 Studies in Semitic Linguistics: Selected Writings. Jerusalem. 2006 Comments on ‘Three Approaches to the Tripartite Nominal Clause in Syriac’ by Wido van Peursen. Pp. 175–84 in van Keulen and van Peursen 2006. Goldstein, J. A. 1966 The Syriac Deed of Sale from Dura-Europos. JNES 25: 1–16. Golovko, E. V. 1994 Copper Island Aleut. Pp. 113–21 in Bakker and Mous 1994. 1996 A Case of Nongenetic Development in the Arctic Area: The Contribution of Aleut and Russian to the Formation of Copper Island Aleut. Pp. 63–77 in Language Contact in the Arctic: Northern Pidgins and Contact Languages, ed. E. H. Jahr and I. Broch. Trends in Linguistics: Studies and Monograph 88. Berlin. Golovko, E. V., and N. B. Vakhtin 1990 Aleut in Contact: The CIA Enigma. Acta Linguistica Hafniensia 22: 97–125. Goshen-Gottstein, M. H. 1989 Exercises in Semitic Linguistics 1: Classical Syriac. Jerusalem Studies in Arabic and Islam 12: 233–42. Gottheil, R. J. H. 1887 A Treatise on Syriac Grammar by Mâr(i) Eliâ of Ṣôbhâ. Berlin. 1910–28  Bar ʿAli (Ishoʿ). The Syriac-Arabic Glosses, vols. 1–2. Atti della R. Accademia dei Lincei: Classe di Scienzi morali, storiche e filologiche Ser. 5, vol. 13. Rome. Gragg, G. 2004 Geʾez (Aksum). Pp. 427–53 in Woodard 2004a. Grainger, J. D. 1990 The Cities of Seleukid Syria. Oxford.

240

Bibliography

Greenberg, J. 1955 Internal a-Plurals in Afroasiatic (Hamito-Semitic). Pp. 198–204 in Afrikanistische Studien, ed. J. Lukas. Berlin. Greenfield, J. C. 1974 Standard Literary Aramaic. Pp. 280–89 in Actes du premier congrès international de linguistique sémitique et chamito-sémitique: Paris 16–19 juillet 1969, ed. A. Caquot and D. Cohen. The Hague and Paris. Gröber, G. 1904–6  Grundriss der Romanischen Philologie, vol. 1. 2 vols. Strassburg. Guidi, I. 1881 La lettera di Simeone vescovo di Bêth-Aršâm sopra I martiri omeriti. Reale Accademia dei Lincei Ser. 3a: Memorie della classe di scienze morali, storiche e filologiche, vol. 7. Rome. Gumperz, J. J. 1971 Language in Social Groups: Essays by J. Gumperz, selected and introduced by A. S. Dil. Stanford, CA. Gumperz, J. J., and R. Wilson 1971 Convergence and Creolization: A Case from the Indo-Aryan/Dravidian Border in India. Pp. 151–67 in Pidginization and Creolization of Languages, ed. D. Hymes. Cambridge. [= Gumperz 1971: 251–73] Gutas, D. 1998 Greek Thought, Arabic Culture: The Graeco-Arabic Translation Movement in Baghdad and Early ʿAbbāsid Society (2nd–4th / 8th–10th Centuries). London. Guy, G. R. 1990 The Sociolinguistic Types of Language Change. Diachronica 7: 47–67. Gzella, H. 2015 A Cultural History of Aramaic: From the Beginnings to the Advent of Islam. Leiden. Haar Romeny, B. ter 1997 A Syrian in Greek Dress: The Use of Greek, Hebrew, and Syriac Biblical Texts in Eusebius of Emesa’s Commentary on Genesis. TEG 6. Louvain. 2015 How Greek Was Syriac Christianity? Plenary Address at VIIth North American Syriac Symposium, The Catholic University of America, June 21–24, Washington, DC. Haefeli, L. 1932 Stilmittel bei Aphrahat dem persischen Weisen. Leipzig. Hafez, O. 1996 Phonological and Morphological Integration of Loanwords into Egyptian Arabic. Égypte / Monde arabe 27–28: 383–410. Halleux, A. de 1960–65  Martyrius (Sahdona): Oeuvres spirituelles, vols. 1–3: Livre de la Perfection; vol. 4: Lettres à des amis solitaires: Maximes sapientiales. CSCO 200–201, 214–15, 252–55. Louvain. 1963 Philoxène de Mabbog: Sa vie, ses écrits, sa théologie. Louvain.

Bibliography

241

1977 Commentaire du prologue johannique (Ms. Br. Mus. Add. 14,534). CSCO 380–81. Louvain. Harmon, A. M. 1921 Lucian, of Samosata, vol. 3. LCL 130. London. Harrak, A. 1992 The Ancient Name of Edessa. JNES 51: 209–14. Harris, A. C., and L. Campbell 1995 Historical Syntax in Cross-Linguistic Perspective. Cambridge. Harvey, S. A. 1990 Asceticism and Society in Crisis: John of Ephesus and the Lives of the Eastern Saints. The Transformation of Classical Heritage 18. Berkeley, CA. Harviainen, T. 1976 On the Loss of the Greek /h/ and the So-Called Aspirated Rhō. StOr 45: 1–88. Hasan-Rokem, G. 2000 Web of Life: Folklore and Midrash in Rabbinic Literature. Stanford, CA. Haspelmath, M. 2008 Loanword Typology: Steps toward a Systematic Cross-Linguistic Study of Lexical Borrowability. Pp. 43–62 in Stolz, Bakker, and Salas Palomo 2008. 2009 Lexical Borrowing: Concepts and Issues. Pp. 35–54 in Haspelmath and Tadmor 2009a. Haspelmath, M., and U. Tadmor 2009a Loanwords in the World’s Languages: A Comparative Handbook. Berlin. 2009b The Loanword Typology Project and the World Loanword Database. Pp. 1–34 in Haspelmath and Tadmor 2009a. Hasselbach, R. 2007 Demonstratives in Semitic. JAOS 127: 1–27. Haugen, E. 1950a Problems of Bilingualism. Lingua 2: 271–90. [= Haugen 1972: 59–78] 1950b The Analysis of Linguistic Borrowing. Language 26: 210–31. [= Haugen 1972: 79–109] 1972 The Ecology of Language: Essays by Einar Haugen, selected and introduced by A. S. Dil. Stanford, CA. Hausherr, I. 1937 Gregorii Monachi Cyprii De theoria sancta quae syriace interpretata dicitur visio divina. Rome. Hayasi, T. 2000 Lexical Copying in Turkic: The Case of Eynu. Pp. 433–39 in Studies on Turkish and Turkic Languages, ed. A. Göksel and C. Kerslake. Turcologica 46. Wiesbaden. Healey, J. F. 1993 The Nabataean Tomb Inscriptions of Madaʾin Salih. JSS Supplement 1. Oxford. 1995 Lexical Loans in Early Syriac: A Comparison with Nabataean Aramaic. Studi epigrafici e linguistici 12: 75–84. 2006 A New Syriac Mosaic Inscription. JSS 51: 313–27.

242

Bibliography

2007 The Edessan Milieu and the Birth of Syriac. Hugoye 10: 115–27. 2008 Some Lexical and Legal Notes on a Syriac Loan Transfer of 240 ce. Pp. 211–26 in Kiraz 2008. 2009 Aramaic Inscriptions and Documents of the Roman Period. Textbook of Syrian Semitic Inscriptions 4. Oxford. Heath, J. 1978 Linguistic Diffusion in Arnhem Land. Australian Aboriginal Studies: Research and Regional Studies 13. Canberra. 1984 Language Contact and Language Change. Annual Review of Anthropology 13: 367–84. Heine, B., and T. Kuteva 2002 World Lexicon of Grammaticalization. Cambridge. 2003 On Contact-Induced Grammaticalization. Studies in Language 27: 529–72. 2005 Language Contact and Grammatical Change. Cambridge. 2006 The Changing Languages of Europe. Oxford. 2008 Constraints on Contact-Induced Linguistic Change. Journal of Language Contact, THEMA 2: 57–90. 2010 Contact and Grammaticalization. Pp. 86–105 in Hickey 2010a. Heine, H., U. Claudi, and F. Hünnemeyer 1991 Grammaticalization: A Conceptual Framework. Chicago. Henry, R. 1959–91  Photius: Bibliothèque. Paris. Hetzron, R. 1976 Two Principles of Genetic Reconstruction. Lingua 38: 89–108. Hezser, C. 2001 Jewish Literacy in Roman Palestine. TSAJ 81. Tübingen. Hickey, R. 2002 Internal and External Forces Again: Changes in Word Order in Old English and Old Irish. Language Sciences 24: 261–83. 2010a The Handbook of Language Contact. Chichester, West Sussex. 2010b Language Contact: Reconsideration and Reassessment. Pp. 1–28 in Hickey 2010a. Hillers, D. R., and E. Cussini 1996 Palmyrene Aramaic Texts. Baltimore, MD. Hoffmann, G. 1874 Syrisch-Arabische Glossen: Autographie einer gothaischen Handschrift ent­haltend Bar Ali’s Lexikon von Alaf bis Mim. Kiel. 1880a Opuscula nestoriana. Kiliae. 1880b Iulianos der Abtruennige: Syrische Erzaehlungen. Leiden. Hoftijzer, J., and K. Jongeling 1995 Dictionary of the North-West Semitic Inscriptions. Leiden. Hopper, P. J., and E. C. Traugott 1993 Grammaticalization. Cambridge, MA. Horner, G. W. 1911–24  The Coptic Version of the New Testament in the Southern Dialect, Otherwise Called Sahidic and Thebaic. Oxford.

Bibliography

243

Horrocks, G. C. 2010 Greek: A History of the Language and Its Speakers. 2nd ed. Oxford. Howard, G. 1981 The Teaching of Addai. Chico. Hoyland, R. G. 2015 In God’s Path: The Arab Conquest and the Creation of an Islamic Empire. Oxford. Huehnergard, J. 1987 Three Notes on Akkadian Morphology. Pp. 181–93 in “Working with No Data”: Semitic and Egyptian Studies Presented to Thomas O. Lambdin, ed. D. M. Golomb. Winona Lake, IN. 1995 What Is Aramaic? ARAM 7: 261–82. 2002 Introduction to the Comparative Study of the Semitic Languages. Unpublished manuscript. Cambridge, MA. 2003 Akkadian ḫ and West Semitic *ḥ. Pp. 102–19 in Studia Semitica, ed. L. Kogan. Moscow. 2005 Features of Central Semitic. Pp. 155–203 in Biblical and Oriental Essays in Memory of William L. Moran, ed. A. Gianto. BibOr 48. Rome. 2006 On the Etymology of the Hebrew Relative šɛ-. Pp. 103–25 in Biblical Hebrew in Its Northwest Semitic Setting: Typological and Historical Perspectives, ed. S. E. Fassberg and A. Hurvitz. Winona Lake, IN, and Jerusalem. Huehnergard, J., and A. D. Rubin 2011 Phyla and Waves: Models of Classification of the Semitic Languages. Pp. 259–78 in Weninger 2011. Hug, V. 1993 Altaramäische Grammatik der Texte des 7. und 6. Jh.s v. Chr. Heidelberg. Humbert, J. 1960 Syntaxe grecque. 3rd ed. Paris. Janda, I. H. 1976 English Hungarian and Hungarian English Language Interference Phenomena in Chicago. Pp. 590–95 in The Second Lacus Forum 1975, ed. P. A. Reich. Columbia, SC. Jansma, T. 1965 Die Christologie Jakobs von Serugh und ihre Abhängigkeit von der alexandrinischen Theologie und der Frömmigkeit Ephraems des Syrers. Le Muséon 78: 5–46. 1969 Natuur, lot en vrijheid: Bardesanes, de filosoof der Arameeërs en zijn images. Wageningen. Jastrow, M. 1886–1903  Dictionary of the Targumim, the Talmud Babli and Yerushalmi, and the Midrashic Literature. New York. Jenner, K. D. 2003 The Use of the Particle ‫ ܐܝܬ‬in the Syro-Hexaplaric Psalter and the Peshitta. Pp. 287–308 in Baasten and van Peursen 2003. Jespersen, O. 1922 Language: Its Nature, Development and Origin. London.

244

Bibliography

1948 Growth and the Structure of the English Language. 9th ed. Oxford. 1954 A Modern English Grammar on Historical Principles, vol. 6. London. Johanson, L. 1992 Strukturelle Faktoren in türkischen Sprachkontakten. Stuttgart. 2002a Structural Factors in Turkic Language Contacts. Richmond. [ET of rev. version of Johanson 1992] 2002b Contact-Induced Change in a Code-Copying Framework. Pp. 285–313 in Jones and Esch 2002. Johnson, S. F. 2015a Languages and Cultures of Eastern Christianity: Greek. Surrey. 2015b Introduction: The Social Presence of Greek in Eastern Christianity, 200–1200 ce. Pp. 1–122 in Johnson 2015a. Jones, M. C., and E. Esch 2002 Language Change: The Interplay of Internal, External and Extra-linguistic Factors. Contributions to the Sociology of Language 86. Berlin. Joosten, J. 1988 The Use of Some Particles in the Old Testament Peshitta. Textus 14: 175–83. 1989 The Predicative Adjective in the Status Emphaticus in Syriac. BO 46: 18–24. 1992 The Negation of the Non-verbal Clause in Early Syriac. JAOS 112: 584–88. 1993 On Ante-position of the Attributive Adjective in Classical Syriac and Biblical Hebrew. ZAH 6: 149–92. 1996 The Syriac Language of the Peshitta and Old Syriac Versions of Matthew. Leiden. 1998 Greek and Latin Words in the Peshitta Pentateuch: First Soundings. Pp. 37–47 in Symposium Syriacum VII: 1996, ed. R. Lavenant, S.J. OCA 256. Rome. 1999 Materials for a Linguistic Approach to the Old Testament Peshitta. Journal of the Aramaic Bible 1: 203–18. 2006 Comments on ‘Three Approaches to the Tripartite Nominal Clause in Syriac’ by Wido van Peursen. Pp. 185–88 in van Keulen and van Peursen 2006. Joseph, B. D., and R. D. Janda 2003 The Handbook of Historical Linguistics. Oxford. Kaufman, S. A. 1974 Akkadian Influences on Aramaic. Chicago. 1992 Aramaic. Pp. 173–78 in vol. 4 of ABD. 1997 Aramaic. Pp. 114–30 in The Semitic Languages, ed. R. Hetzron. London. 2009 Jacob of Sarug’s Homilies on Elijah. Piscataway, NJ. Kaye, A. S. 1997 Phonologies of Asia and Africa (including the Caucasus). 2 Vols. Winona Lake, IN. Kayser, C. 1886 Die Canones Jacob’s von Edessa. Leipzig. Kennedy, H., and J. H. W. G. Liebeschuetz 1988 Antioch and the Villages of Northern Syria in the Fifth and Sixth Centuries a.d.: Trends and Problems. Nottingham Medieval Studies 32: 65–90.

Bibliography

245

Keulen, P. S. F. van, and W. T. van Peursen 2006 Corpus Linguistics and Textual History: A Computer-Assisted Interdisciplinary Approach to the Peshiṭta. Assen. Khan, G. 2002 The Neo-Aramaic Dialect of Qaraqosh. Studies in Semitic Languages and Linguistics 36. Leiden. King, D. 2007 Paul of Callinicum and His Place in Syriac Literature. Le Muséon 120: 327–49. 2008 The Syriac Version of the Writings of Cyril of Alexandria: A Study in Translation Technique. CSCO 626. Louvain. King, R. 2000 The Lexical Basis of Grammatical Borrowing: A Prince Edward Island French Case Study. Amsterdam Studies in the Theory and History of Linguistic Science, Series IV. Current Issues in Linguistic Theory 209. Amsterdam. 2005 Crossing Grammatical Borders: Tracing the Path of Contact-Induced Linguistic Change. Pp. 233–51 in Dialects across Borders: Selected Papers from the 11th International Conference on Methods in Dialectology (Methods XI), Joensuu, August 2002, ed. M. Filppula et al. Amsterdam Studies in the Theory and History of Linguistic Science, Series IV. Current Issues in Linguistic Theory 273. Amsterdam. Kıral, F. 2000 Das gesprochene Aserbaidschanisch von Iran: Eine Studie zu den syntaktischen Einflüssen des Persischen. Turcologica 43. Wiesbaden. Kiraz, G. A. 1993 A Computer-Generated Concordance to the Syriac New Testament. Leiden. 1996 Comparative Edition of the Syriac Gospels. Leiden. 2007 Kthobonoyo Syriac: Some Observations and Remarks. Hugoye 10: 129–42. 2008 Malphono w-Rabo d-Malphone: Studies in Honor of Sebastian P. Brock. Piscataway, NJ. 2012 Tūrrāṣ Mamllā: A Grammar of the Syriac Language, vol. 1. Piscataway, NJ. Kleyn, H. G. 1882 Het leven van Johannes van Tella door Elias. Leiden. Kmosko, M. 1907 S. Simeon bar Sabbaʿe. PS 2/3. Paris. 1926 Liber Graduum. PS 3. Paris. Koehler, L., and W. Baumgartner 1994–2000  The Hebrew and Aramaic Lexicon of the Old Testament, rev. ed. W. Baumgartner and J. J. Stamm. 4 vols. Leiden. Kogan, L. 2011 Proto-Semitic Phonetics and Phonology. Pp. 54–119 in Weninger 2011. Krauss, S. 1898 Griechische und lateinische Lehnwörter im Talmud, Midrasch und Targum, vol. 1. Berlin. 1899 Griechische und lateinische Lehnwörter im Talmud, Midrasch und Targum, vol. 2. Berlin.

246

Bibliography

Krebernik, M. 1991 Schriftfund aus Tall Biʿa 1990. MDOG 123: 41–70. Kutscher, E. Y. 1954 New Aramaic Texts. JAOS 74: 233–48. 1965 Contemporary Studies in North-Western Semitic. JSS 10: 21–51. 1970 Aramaic. Pp. 347–412 in Sebeok 1970. Lado, R. 1957 Linguistics across Cultures: Applied Linguistics for Language Teachers. Ann Arbor, MI. Lagarde, P. A. de 1859 Titi Bostreni contra manichaeos libri quatuor syriace. Bern. 1861 Clementis Romani recognitiones syriace. Osnabrück. Lake, K. 1926 Eusebius: The Ecclesiastical History, vol. 1. LCL 153. Cambridge, MA. Lampe, G. W. H. 1961 A Patristic Greek Lexicon. Oxford. Lane, E. W. 1863–93  An Arabic-English Lexicon. 8 vols. London. Lattke, M. 1979–98  Die Oden Salomos in ihrer Bedeutung für Neues Testament und Gnosis, vols. 1–4. OBO 25/1–4. Freiburg. 1993a Dating the Odes of Solomon. Antichthon 27: 45–59. [= Lattke 1979–98: 4.113–32) 1993b Die griechischen Wörter im syrischen Text der Oden Salomos. ARAM 5: 285–302. [= Lattke 1979–98: 4.133–50] 1995 Oden Salomos. FC 19. Freiburg. 2009 Odes of Solomon: A Commentary. Hermeneia. Minneapolis. Layton, B. 2004 A Coptic Grammar with Chrestomathy and Glossary: Sahidic Dialect. 2nd ed. PLO n.s. 20. Wiesbaden. Lebon, J. 1930 Textes inédits de Philoxène de Mabboug. Le Muséon 43: 17–84, 149–220. Lee, S. 1842 Eusebius, Bishop of Caesarea, on the Theophania or Divine Manifestation of Our Lord and Saviour Jesus Christ. London. Lehiste, I. 1988 Lectures on Language Contact. Cambridge. Leloir, L. 1963 Saint Éphrem: Commentaire de l’Évangile concordant. Texte syriaque (Manuscrit Chester Beatty 709). CBM 8. Dublin. 1990 Saint Éphrem: Commentaire de l’Évangile concordant. Texte syriaque (Ma­ nus­crit Chester Beatty 709), folios additionels. CBM 8. Louvain. Leslau, W. 1987 Comparative Dictionary of Geʿez (Classical Ethiopic). Wiesbaden. Lewis, C. T., and C. Short 1969 A Latin Dictionary. Oxford.

Bibliography

247

Liddell, H., and R. Scott 1996 A Greek-English Lexicon, rev. H. Stuart Jones and R. McKenzie. Oxford. Lieberman, S. 1942 Greek in Jewish Paestine. New York. 1950 Hellenism in Jewish Palestine. New York. Littmann, E., and M. Höfner 1956 Wörterbuch der Tigrē-Sprache: Tigrē-Deutsch-Englisch. Wiesbaden. Look, A. E. 1929 The History of Abba Marcus of Mount Tharmaka. Oxford. Louden, M. L. 2000 Contact-Induced Phonological Changes in Yiddish: Another Look at Weinreich’s Riddles. Diachronica 17: 85–110. Loveday, L. J. 1996 Language Contact in Japan: A Socio-linguistic History. Oxford. Lund, J. A. 2007 The Book of the Laws of Countries. A Key-Word-in-Context Concordance. Piscataway, NJ. 2013 Gender Documentation of Nouns in Syriac Lexicography: Remarks on the Renovated Lexicon Syriacum. Hugoye 16: 3–14. Machiela, D. A. 2009 The Dead Sea Genesis Apocryphon: A New Text and Translation with Introduction and Special Treatment of Columns 13–17. STDJ 79. Leiden. MacKenzie, D. N. 1971 A Concise Pahlavi Dicitonary. London. MacMullen, R. 1966 Provincial Languages in the Roman Empire. AJP 87: 1–16. 1982 The Epigraphic Habit in the Roman Empire. AJP 103: 233–46. Macomber, W. F. 1974 Six Explanations of the Liturgical Feasts by Cyrus of Edessa. 2 vols. CSCO 355–56. Louvain. Macuch, R. 1965 Handbook of Classical and Modern Mandaic. Berlin. 1976 Geschichte der spät- und neusyrischen Literatur. Berlin. Mandel, P. 2000 Between Byzantium and Islam: The Transmission of a Jewish Book in the Byzantine and Early Islamic Periods. Pp. 74–106 in Transmitting Jewish Traditions: Orality, Textuality, and Cultural Diffusion, ed. Y. Elman and I. Gershoni. New Haven, CT. Mango, M. M. 1982 The Continuity of the Classical Tradition in the Art and Architecture of Northern Mesopotamia. Pp. 115–134 in Garsoïan, Mathews, and Thomson 1982. Mankowski, P. V. 2000 Akkadian Loanwords in Biblical Hebrew. HSS 47. Winona Lake, IN.

248

Bibliography

Markschies, C. 2012 Hellenisierung des Christentums: Sinn und Unsinn einer historischen Deutungskategorie. Leipzig. Marrassini, P. 1990 Some Considerations on the Problem of the ‘Syriac Influences’ on Aksumite Ethiopia. Journal of Ethiopian Studies 23: 35–46. Mason, H. J. 1974 Greek Terms for Roman Institutions. Toronto. Matras, Y. 2009 Language Contact. Cambridge. 2010 Contact, Convergence, and Typology. Pp. 66–85 in Hickey 2010a. Matras, Y., and P. Bakker 2003 The Mixed Language Debate: Theoretical and Empirical Advances. Trends in Linguistics: Studies and Monographs 145. Berlin. Matras, Y., and J. Sakel 2007a Grammatical Borrowing in Cross-Linguistic Perspective. Berlin. 2007b Introduction. Pp. 1–13 in Matras and Sakel 2007a. 2007c Investigating the Mechanisms of Pattern Replication in Language Convergence. Studies in Language 31: 829–65. Mayser, M. 1970 Grammatik der griechischen Papyri aus der Ptolemäerzeit, vol. 1: Laut- und Wortlehre, part 1: Einleitung und Lautlehre. 2nd ed. Berlin. Menz, A. 2000 Gagausische Syntax: Eine Studie zum kontaktinduzierten Sprachwandel. Turcologica 41. Wiesbaden. Merkelbach, M., and J. Stauber 2005 Jenseits des Euphrat: Griechische Inschriften. Munich. Meyer, E. A. 1990 Explaining the Epigraphic Habit in the Roman Empire: The Evidence of Epitaphs. JRS 80: 74–96. Michelson, D. 2014 The Practical Christology of Philoxenos of Mabbug. Oxford. Militarev, A., and L. Kogan 2000 Semitic Etymological Dictionary, vol. 1: Anatomy of Man and Animals. Münster. Millar, F. 1987 The Problem of Hellenistic Syria. Pp. 110–33 in Hellenism in the East, ed. A. Kuhrt and S. Sherwin-White. London. [= Millar 2006: 3–31] 1993 The Roman Near East 31 b.c.–a.d. 337. Cambridge, MA. 2006 Rome, the Greek World, and the East, vol. 3: The Greek World, the Jews, and the East. Chapel Hill, NC. 2007 Theodoret of Cyrrhus: A Syrian in Greek Dress. Pp. 105–25 in From Rome to Constantinople: Studies in Honour of Averil Cameron, ed. H. Amirav and B. ter Haar Romeny. Louvain. 2011a Greek and Syriac in Edessa: From Ephrem to Rabbula (ce 363–435). Semitica et Classica 4: 99–113.

Bibliography

249

2011b Greek and Syriac in Edessa and Osrhoene, ce 213–363. Scripta Classica Israelica 30: 93–111. 2012 Greek and Syriac in Fifth-Century Edessa: The Case of Bishop Hibas. Semitica et Classica 5: 151–65. 2013a Religion, Language and Community in the Roman Near East. Oxford. 2013b The Evolution of the Syrian Orthodox Church in the Pre-Islamic Period: From Greek to Syriac? JECS 21: 43–92. Mingana, A. 1905 Narsai doctoris Syri homiliae et carmina. 2 vols. Mosul. 1907 Sources syriaques. Leipzig. Mitchell, C. W. 1912–21  S. Ephraim’s Prose Refutations of Mani, Marcion, and Bardaisan. 2 vols. Oxford. Moller, G. I. 1988 Towards a New Typology of the Syriac Manuscript Alphabet. JNSL 14: 153–97. Mønnesland, S. 1999 Syntactic Borrowing or Internal Development? Location vs Motion in Serbian and Albanian. Pp. 321–39 in Brendemoen, Lanza, and Ryen 1999. Moss, C. 1929 Isaac of Antioch: Homily on the Royal City. ZS 7: 295–306. 1932 Isaac of Antioch: Homily on the Royal City. ZS 8: 61–72. 1935 Jacob of Serugh’s Homilies on the Spectacles and the Theater. Le Muséon 48: 87–112. Moss, Y. forthcoming  Incorruptible Bodies: Christology, Society, and Authority in Late Antiquity. Christianity in Late Antiquity 1. Berkeley, CA. Mougeon, R., and E. Beniak 1991 Linguistic Consequences of Language Contact and Restriction: The Case of French in Ontario, Canada. Oxford. Mous, M. 1994 Maʾa or Mbugu. Pp. 175–200 in Bakker and Mous 1994. 2001 Maʾa as an Ethno-Register of Mbugu. SUGIA: Sprache und Geschichte in Afrika 16–17: 293–320. 2003 The Linguistic Properties of Lexical Manipulation and Its Relevance for Maʾá. Pp. 209–35 in Matras and Bakker 2003. Mufwene, S. S. 2001 The Ecology of Language Evolution. Cambridge. Müller-Kessler, C. 1991 Grammatik des Christlich-Palästinisch-Aramäischen. Hildesheim. 2004 The Mandaeans and the Question of Their Origin. ARAM 16: 47–60. Müller-Kessler, C., and M. Sokoloff 1996 The Forty Martyrs of the Sinai Desert, Eulogios, the Stone-Cutter, and Ana­ stasia. A Corpus of Christian Palestinian Aramaic 3. Groningen.

250

Bibliography

1997 The Christian Palestinian Aramaic Old Testament and Apocryphal Version from the Early Period. A Corpus of Christian Palestinian Aramaic 1. Groningen. 1998a The Christian Palestinian Aramaic New Testament Version from the Early Period: Gospels. A Corpus of Christian Palestinian Aramaic 2A. Groningen. 1998b The Christian Palestinian Aramaic New Testament Version from the Early Period: Acts of the Apostles and Epistles. A Corpus of Christian Palestinian Aramaic 2A. Groningen. 1999 The Catechism of Cyril of Jerusalem in the Christian Palestinian Aramaic Version. A Corpus of Christian Palestinian Aramaic 4. Groningen. Muraoka, T. 1972 Remarks on the Syntax of Some Types of Noun Modifier in Syriac. JNES 31: 192–94. 1975 On the Nominal Clause in the Old Syriac Gospels. JSS 20: 28–37. 1977 On the Syriac Particle it. BO 34: 21–22. 1987 Classical Syriac for Hebraists. Wiesbaden. 2005 Classical Syriac: A Basic Grammar with a Chrestomathy. 2nd ed. PLO n.s. 19. Wiesbaden. 2006 Further Remarks on ‫ ܐܝܬ‬Clauses in Classical Syriac. Pp. 129–34 in Text, Translation, and Tradition: Studies on the Peshitta and Its Use in the Syriac Tradition Presented to Konrad D. Jenner on the Occasion of His Sixty-Fifth Birthday, ed. W. T. van Peursen and R. B. ter Haar Romeny. Leiden. Muraoka, T., and B. Porten 1998 A Grammar of Egyptian Aramaic. Leiden. Murray, R. 1967 Ephrem Syrus. Pp. 220–23 in A Catholic Dictionary of Theology. London. 1982 The Characteristics of the Earliest Syriac Christianity. Pp. 3–16 in Garsoïan, Mathews, and Thomson 1982. 2004 Symbols of Church and Kingdom: A Study in Early Syriac Tradition. Rev. ed. London. Muysken, P. 1981 Halfway between Quechua and Spanish: The Case for Relexification. Pp. 52–78 in Historicity and Variation in Creole Studies by A. Highfield and A. Valdman. Ann Arbor, MI. 1994 Media Lengua. Pp. 201–5 in Bakker and Mous 1994. 1997 Media Lengua. Pp. 365–426 in Thomason 1997a. 2000 Bilingual Speech: A Typology of Code-Mixing. Cambridge. Myers-Scotton, Carol 1993 Duelling Languages: Grammatical Structure in Code-Switching. Oxford. 2002 Contact Linguistics: Bilingual Encounters and Grammatical Outcomes. Oxford. 2006 Multiple Voices: An Introduction to Bilingualism. Malden, MA. Nadkarni, M. V. 1975 Bilingualism and Syntactic Change in Konkani. Language 51: 672–83. Nau, F. 1899 Le traité sur l’astrolabe plan de Sévère Sébokt. JA 9/13: 56–101, 238–303.

Bibliography

251

1902 Vie de Jean bar Aphtonia: Texte syriaque. Bibliothèque hagiographique orientale 2. Paris. 1905a Histoires d’Ahoudemmeh et de Marouta. PO 3/1. Paris. 1905b Traduction des lettres XII et XIII de Jacques d’Édesse (exégèse biblique). ROC 10: 197–208, 258–82. 1907 Bardesanes: Liber legum regionum. PS 1/2. Paris. 1913 La seconde partie de l’histoire de Barhadbesabba ʿArbaïa. PO 9/5. Paris. 1932 La première partie de l’histoire de Barhadbesabba ʿArbaïa. PO 23/2. Paris. Nöldeke, T. 1868 Beiträge zur Kenntniss der aramäischen Dialecte, II: Ueber den christlichpalästinischen Dialect. ZDMG 22: 443–527. 1875 Mandäische Grammatik. Halle. 1904 Compendious Syriac Grammar: Translated from the Second and Improved German Edition by James A. Crichton. Leipzig. [Repr. Winona Lake, IN, 2001] Noonan, M. 2010 Genetic Classification and Language Contact. Pp. 48–65 in Hickey 2010a. Olinder, G. 1937 Iacobi Sarugensis Epistulae quotquot supersunt. CSCO 110. Paris. Olmo Lete, G. del, and J. Sanmartín 2003 A Dictionary of the Ugaritic Language in the Alphabetic Tradition. Leiden. Ortiz de Urbina, I. 1958 Patrologia Syriaca. 2nd ed. Rome. Overbeck, J. J. 1865 S. Ephraemi Syri Rabulae episcopi Edesseni Balaei aliorumque Opera selecta. Oxford. Owens, R. J., Jr. 1983 The Genesis and Exodus Citations of Aphrahat the Persian Sage. MPIL 3. Leiden. Palmer, A. 2015 Sumatar Revisited: The Long Inscription of 165 ce. JSS 60: 63–92. Palmer, L. R. 1934 Prolegomena to a Grammar of the Post-Ptolemaic Papyri. JTS 35: 170–75. 1945 A Grammar of the Post-Ptolemaic Papyri. London. Pardee, D. 2000 Les textes rituels. Ras Shamra-Ougarit 12. Paris. 2003–4  Review of Tropper 2000 (online version). AfO 50. 2004 Ugaritic. Pp. 288–318 in Woodard 2004a. Parisot, I. 1894–1907  Aphraatis Sapientis Persae Demonstrationes. PS 1/1–2. Paris. Pat-El, N. 2006 Syntactical Aspects of Negation in Syriac. JSS 51: 329–48. Payne Smith, R. 1879–1901  Thesaurus Syriacus. Oxford. 1903 A Compendious Syriac Dictionary, ed. J. Payne Smith. Oxford. [Repr. Winona Lake, IN, 1998]

252

Bibliography

Petermann, H. 1867 Thesaurus s. Liber magnus vulgo “Liber Adami” appellatus opus Mandaeorum summi ponderis. Leipzig. Petit, F., L. Van Rompay, and J. J. S. Weitenberg 2011 Eusèbe d’Émèse: Commentaire de la Genèse. TEG 15. Louvain. Peursen, W. P. van 2006a Three Approaches to the Tripartite Nominal Clause in Syriac. Pp. 157–73 in van Keulen and van Peursen 2006. 2006b Response to the Responses. Pp. 197–204 in van Keulen and van Peursen 2006. Peursen, W. P. van, and T. C. Falla 2009 The Particles ‫ ܶܓܝܪ‬and ‫ ܷܕܝܢ‬in Classical Syriac: Syntactic and Semantic Aspects. Pp. 63–99 in Foundations for Syriac Lexicography, vol. 2: Colloquia of the International Syriac Language Project, ed. P. J. Williams. Perspectives on Syriac Linguistics 3. Piscataway, NJ. Phillips, G. 1864 Scholia on Passages of the Old Testament by Mar Jacob. London. 1869 A Letter by Mār Jacob, Bishop of Edessa on Syriac Orthography. London. Pierre, M. J. 1988–89  Aphraate le sage persan, vols. 1–2. SC 349, 359. Paris. Poplack, S. 1993 Variation Theory and Language Contact. Pp. 251–86 in American Dialect Research, ed. D. Preston. Amsterdam. 1996 The Sociolinguistic Dynamics of Apparent Convergence. Pp. 285–309 in Towards a Social Science of Language: Papers in Honour of William Labov, vol. 2. Social Interaction and Discourse Structures, ed. G. Guy et al. Amsterdam Studies in the Theory and History of Linguistic Science, Series IV. Current Issus in Linguistic Theory 128. Philadelphia. Poplack, S., and S. Levey 2010 Contact-Induced Grammatical Change: A Cautionary Tale. Pp. 391–419 in Language and Space: An International Handbook of Linguistic Variation, vol. 1: Theories and Methods, ed. P. Auer and J. E. Schmidt. Berlin. Poplack, S., and M. Meechan 1995 Patterns of Language Mixture: Nominal Structure in Wolof-French and Fongbe-French Bilingual Discourse. Pp. 199–232 in One Speaker, Two Languages, ed. P. Muysken and L. Milroy. Cambridge. Poplack, S., and D. Sankoff 1984 Borrowing: The Synchrony of Integration. Linguistics 22: 99–135. Poplack, S., D. Sankoff, and C. Miller 1988 The Social Correlates and Linguistic Processes of Lexical Borrowing and Assimilation. Linguistics 26: 47–104. Porten, B., and J. A. Lund 2002 Aramaic Documents from Egypt: A Key-Word-in-Context Concordance. Winona Lake, IN.

Bibliography

253

Porten, B., and A. Yardeni 1986–93  Textbook of Aramaic Documents from Ancient Egypt. 4 vols. Jerusalem and Winona Lake, IN. Possekel, U. 1999 Evidence of Greek Philosophical Concepts in the Writings of Ephrem the Syriac. CSCO 380. Louvain. 2004 Bardaisan of Edessa on the Resurrection: Early Syriac Eschatology in Its Religious-Historical Context. OC 88: 1–28. 2006 Bardaisan of Edessa: Philosopher or Theologian? ZAC 10: 442–61. 2008 Orpheus among the Animals: A New Dated Mosaic from Osrhoene. OC 92: 1–35. 2009 Die Schöpfungstheologie des Bardaisan von Edessa. Pp. 219–29 in Edessa in hellenistisch-römischer Zeit: Religion, Kultur und Politik zwischen Ost und West, L. Grei­siger, C. Rammelt, and J. Tubach. Beiruter Texte und Studien 116. Beirut. 2012 Bardaisan and Origen on Fate and the Power of the Stars. JECS 20: 515–41. Pray, B. 1980 Evidence of Grammatical Convergence in Dakhini Urdu and Telegu. Berkeley Linguistics Society 6: 90–99. Price, R. M. 1985 A History of the Monks of Syria by Theodoret of Cyrrhus. CSS 88. Kala­ mazoo, MI. Probert, P. 2006 Ancient Greek Accentuation: Synchronic Patterns, Frequency Effects, and Prehistory. Oxford. Pusey, P. E., and G. H. Gwilliam 1901 Tetraeuangelium sanctum; juxta simplicem Syrorum versionem ad fidem codicum Massorae, editionum denuo recognitum. Oxford. Ramelli, I. 2009 Bardaisan of Edessa: A Reassessment of the Evidence and a New Interpretation. Piscataway, NJ. Ratcliffe, R. R. 1998a The “Broken” Plural Problem in Arabic and Comparative Semitic: Allomorphy and Analogy in Non-concatenative Morphology. Amsterdam. 1998b Defining Morphological Isoglosses: The ‘Broken’ Plural and Semitic Subclassification. JNES 57: 81–123. Rayfield, J. R. 1970 The Languages of a Bilingual Community. The Hague. Reinink, G. J. 1983 Das syrische Alexanderlied: Die drei Rezensionen. CSCO 454–55. Louvain. 1993 Die syrische Apokalypse des Pseudo-Methodius. CSCO 540–41. Louvain. Ri, Su-Min 1987 La Caverne des trésors: Les deux recensions syriaques. CSCO 486–87. Louvain.

254

Bibliography

Ridolfini, F. P. 2006 Le “Dimostrazioni” del “Sapiente Persiano.” Verba Seniorum n.s. 14. Rome. Rignell, K.-E. 1979 A Letter from Jacob of Edessa to John the Stylite of Litarab concerning Ecclesiastical Canons. Lund. Rignell, L. G. 1941 Briefe von Johannes dem Einsiedler. Lund. Roca, I., and W. Johnson 1999 A Course in Phonology. Oxford. Rochette, B. 2010 Greek and Latin Bilingualism. Pp. 281–93 in A Companion to the Ancient Greek Language, ed. E. J. Bakker. Malden, MA. Roey, A. van, and P. Allen 1994 Monophysite Texts of the Sixth Century. OLA 56. Louvain. Rosenthal, F. 1967 An Aramaic Handbook. PLO 10. Wiesbaden. 1995 A Grammar of Biblical Aramaic. PLO n.s. 5. Wiesbaden. Ross, M. D. 1985 Current Use and Expansion of Tok Pisin: Effects of Tok Pisin on Some Vernacular Languages. Pp. 539–556 in Handbook of Tok Pisin (New Guinea Pidgin), ed. S. A. Wurm and P. Mühlhäusler. Canberra. 1987 A Contact-Induced Morphosyntactic Change in the Bel Languages of Papua New Guinea. Pp. 583–601 in A World of Language: Papers Presented to Professor S. A. Wurm on His 65th Birthday, D. C. Laycock and W. Winter. Canberra. 1991 Refining Guy’s Sociolinguistic Types of Language Change. Diachronica 8: 119–29. 1996 Contact-Induced Change and the Comparative Method: Cases from Papua New Guinea. Pp. 180–217 in The Comparative Method Reviewed: Regularity and Irregularity in Language Change, ed. M. Durie and M. Ross. New York. 2001 Contact-Induced Change in Oceanic Languages in North-West Melanesia. Pp. 134–66 in Aikhenvald and Dixon 2001a. 2003 Diagnosing Prehistoric Language Contact. Pp. 174–98 in Motives for Language Change, ed. R. Hickey. Cambridge. 2006 Metatypy. Pp. 95–99 in vol. 8 of Encyclopedia of Language and Linguistics, ed. K. Brown. 2nd ed. Oxford. 2007 Calquing and Metatypy. Journal of Language Contact, THEMA 1: 116–43. 2008 A History of Metatypy in the Bel Languages. Journal of Language Contact, THEMA 2: 149–64. Ross, S. K. 2001 Roman Edessa: Politics and Culture on the Eastern Fringes of the Roman Empire, 114–242 c.e. London. Rostovtzeff, M. 1943 L’Orient et la civilisation grecque: Doura-Europos sur l’Euphrate. Renaissance 1: 43–59.

Bibliography

255

Rubin, A. D. 2005 Studies in Semitic Grammaticalization. HSS 57. Winona Lake, IN. Russell, D. A. 2001 Quintilian: The Orator’s Education. Books 1–2. LCL 124. Cambridge, MA. Sachau, E. 1870 Inedita Syriaca: Eine Sammlung syrischer Übersetzungen von Schriften griechischer Profanliteratur. Vienna. Sakel, J. 2007 Types of Loan: Matter and Pattern. Pp. 15–29 in Matras and Sakel 2007a. Sakel, J., and Y. Matras 2008 Modelling Contact-Induced Change in Grammar. Pp. 63–87 in Stolz, Bakker, and Salas Palomo 2008. Salmons, J. 1990 Bilingual Discourse-Marking: Code-Switching, Borrowing and Convergence in Some German-American Dialects. Linguistics 28: 453–80. Sandfeld, M. K. 1938 Problèmes d’interférences linguistiques. Pp. 59–61 in Actes du quatrième congrès international de linguistes. Copenhagen. Sankoff, D., S. Poplack, and S. Vanniarajan 1990 The Case of the Nonce Loan in Tamil. Language Variation and Change 2: 71–101. Sankoff, G. 2002 Linguistic Outcomes of Language Contact. Pp. 638–38 in The Handbook of Language Variation and Change, ed. J. K. Chambers, P. Trudgill, and N. Schilling-Estes. Oxford. Sapir, E. 1921 Language: An Introduction to the Study of Speech. New York. Sartre, M. 2001 D’Alexandre à Zénobie: Histoire du Levant antique, IVe siècle avant J.-C.– IIIe siècle après J.-C. Paris. 2005 The Middle East under Rome. Cambridge, MA. Schall, A. 1960 Studien über griechische Fremdwörter im Syrischen. Darmstadt. Scher, A. 1907 Mar Barḥadbšabba ʿArbaya évêque de Ḥalwan (VIe siècle): Cause de la fondation des écoles. PO 4/4. Paris. 1909 Traités d’Isaï le Docteur et de Hnana d’Adiabène sur les martyrs, le vendredi d’or et les rogations. PO 7/1. Paris. Schlesinger, M. 1928 Satzlehre der aramäischen Sprache des babylonischen Talmuds. Leipzig. Schmitt, R. 1983 Die Sprachverhältnisse in den östlichen Provinzen des Römischen Reiches. ANRW II.29.2: 554–86. Schulthess, F. 1903 Lexicon Syropalaestinum. Berolini.

256

Bibliography

Schwartz, E. 1927 Acta conciliorum oecumenicorum, vol. 1/6: Concilium universale Ephesenum. Berlin. Sebeok, T. 1970 Current Trends in Linguistics, vol. 6: Linguistics in South West Asia and North Africa. The Hague. Segal, J. B. 1953 The Diacritical Point and the Accents in Syriac. London Oriental Series 2. London. 1970 Edessa ‘The Blessed City’. London. Shepardson, C. 2008 Anti-Judaism and Christian Orthodoxy: Ephrem’s Hymns in Fourth-Century Syria. Washington, DC. Sihler, A. L. 1995 New Comparative Grammar of Greek and Latin. Oxford. Silva-Corvalán, C. 1994 Language Contact and Change: Spanish in Los Angeles. Oxford. 1995a Spanish in Four Continents: Studies in Language Contact and Bilingualism. Washington, DC. 1995b The Study of Language Contact: An Overview of the Issues. Pp. 3–14 in Silva-Corvalán 1995b. 1998 On Borrowing as a Mechanism of Syntactic Change. Pp. 225–46 in Romance Linguistics: Theoretical Perspectives, ed. A. Schwegler, B. Tranel, and M. Uribe-Etxebarria. Amsterdam. 2008 The Limits of Convergence in Language Contact. Journal of Language Contact, THEMA 2: 213–24. Siman, E. P. 1984 Narsaï: Cinq homélies sur les paraboles évangéliques. Paris. Smelik, W. F. 2013 Rabbis, Language and Translation in Late Antiquity. Cambridge. Smith, K. 2014 A Last Disciple of the Apostles: The “Editor’s” Preface, Rabbula’s Rules, and the Date of the Book of Steps. Pp. 72–96 in Breaking the Mind: New Studies in the Syriac “Book of Steps,” ed. K. S. Heal and R. A. Kitchen. Washington, DC. Smits, C. 1996 Disintegration of Inflection: The Case of Iowa Dutch. The Hague. 1998 Two Models for the Study of Language Contact: A Psycho-Linguistic Perspective versus a Socio-cultural Perspective. Pp. 377–90 in Historical Linguistics 1997: Selected Papers from the 13th International Conference on Historical Linguistics, Düsseldorf, 10–17 August 1997, ed. M. S. Schmid, J. R. Austin, and D. Stein. Amsterdam. 1999 Working Class Maastricht versus Middle Class Maastricht: In Search for a Theoretical Framework. Leuvense Bijdragen 88: 453–76. Smyth, H. W. 1956 Greek Grammar. Cambridge.

Bibliography

257

Sokoloff, M. 2002a. A Dictionary of Jewish Babylonian Aramaic of the Talmudic and Geonic Periods. Ramat-Gan. 2002b A Dictionary of Jewish Palestinian Aramaic of the Byzantine Period. 2nd ed. Ramat-Gan. 2003 A Dictionary of Judean Aramaic. Ramat-Gan. 2009 A Syriac Lexicon: A Translation from the Latin, Correction, Expansion, and Update of C. Brockelmann’s Lexicon Syriacum. Winona Lake, IN. [2nd, corrected printing, Winona Lake, IN, 2012] 2011 Jewish Babylonian Aramaic. Pp. 660–70 in Weninger 2011. 2014 A Dictionary of Christian Palestinian Aramaic. OLA 234. Louvain. Sperber, D. 2006 Rabbinic Knowledge of Greek. Pp. 627–40 in The Literature of the Sages, vol. 2: Midrash and Targum, Liturgy, Poetry, Mysticism, Contracts, Inscriptions, Ancient Science and the Languages of Rabbinic Literature, ed. S. Safrai, Z. Safrai, J. Schwartz, and P. J. Tomson. Assen. Sridhar, S. N. 1978 Linguistic Convergence: Indo-Aryanization of Dravidian Languages. Studies in the Linguistic Sciences 8: 197–215. Stein, P. 2003 Untersuchungen zur Phonologie und Morphologie des Sabäischen. Rahden. Steiner, R. C. 1977 The Case for Fricative-Laterals in Proto-Semitic. New Haven, CT. 1982 Affricated Ṣade in the Semitic Languages. New York. Steingass, F. J. 1930 A Comprehensive Persian-English Dictionary. London. Stolz, T. 2008 Romancisation World-Wide. Pp. 1–24 in Stolz, Bakker, and Salas Palomo 2008. Stolz, T., D. Bakker, and R. Salas Palomo 2008 Aspects of Language Contact: New Theoretical, Methodological and Empirical Findings with Special Focus on Romancisation Processes. Berlin. Strothmann, W. 1972 Johannes von Apamea. PTS 11. Berlin. 1988 Kohelet-Kommentar des Johannes von Apamea. GOFS 30. Wiesbaden. Tal, A. 1980–83  The Samaritan Targum of the Pentateuch: A Critical Edition. Tel-Aviv. [Hebrew] 2000 A Dictionary of Samaritan Aramaic. Leiden. Tavernier, J. 2007 Iranica in the Achaemenid Period (ca. 550–330 b.c.): Lexicon of Old Iranian Proper Names and Loanwords, Attested in Non-Iranian Texts. OLA 158. Louvain. Taylor, D. G. K. 2002 Bilingualism and Diglossia in Late Antique Syria and Mesopotamia. Pp. 298–331 in Bilingualism in Ancient Society, ed. J. N. Adams, M. Janse, and S. Swain. Oxford.

258

Bibliography

Teixidor, J. 1989 Les derniers rois d’Édesse d’après deux nouveaux documents syriaques. ZPE 76: 219–22. 1990 Deux documents syriaques du IIIe siècle après J.-C., provenant du Moyen Euphrate. CRAIBL 1990: 144–66. 1991–92  Un document syriaque de fermage de 242 après J.-C. Semitica 41–42: 195–208. 1992 Bardesane d’Edesse: La première philosophie syriaque. Paris. Terpelyuk, A. A. 2009 Martyrdom of Mār Qūryāqūs and Yōlīṭī (Cyriacus and Julitta). Moscow. [Russian] Teule, H., and C. F. Tauwinkl, with B. ter Haar Romeny and J. van Ginkel 2010 The Syriac Renaissance. Louvain. Theodor, J., and C. Albeck 1912–36  Bereschit Rabba mit kritischem Apparat und Kommentar. Berlin. Thomason, S. G. 1986 Contact-Induced Language Change: Possibilities and Probabilities. Pp. 261–83 in Beiträge zum 2. Essener Kolloquium über “Kreolsprachen und Sprachkontakte,” ed. N. Boretzky, W. Enninger, and T. Stolz. Bochum. 1997a Contact Languages: A Wider Perspective. Amsterdam. 1997b On Mechanisms of Interference. Pp. 181–207 in Language and Its Ecology: Essays in Memory of Einar Haugen, ed. S. Eliasson and E. H. Jahr. Trends in Linguistics: Studies and Monographs 100. Berlin. 1997c A Typology of Contact Languages. Pp. 71–88 in The Structure and Status of Pidgins and Creoles, ed. A. K. Spears and D. Winford. Amsterdam. 1997d Mednyj Aleut. Pp. 449–68 in Thomason 1997a. 1997e Ma’a (Mbugu). Pp. 469–87 in Thomason 1997a. 2001 Language Contact: An Introduction. Washington, DC. 2003 Contact as a Source of Language Change. Pp. 687–712 in Joseph and Janda 2003. 2004 Determining Language Contact Effects in Ancient Contact Situations. Pp. 1–14 in Lenguas en contacto: El testimonio escrito, ed. P. Bádenas de la Peña et al. Madrid. 2005 Contact-Induced Changes: Classification and Processes. Diachronica 22: 373–427. 2007 Language Contact and Deliberate Change. Journal of Language Contact, THEMA 1: 41–62. 2008 Social and Linguistic Factors as Predictors of Contact-Induced Change. Journal of Language Contact, THEMA 2: 42–56. 2009 On the Unity of Contact Phenomena: The Case of Borrowing. Pp. 279–305 in Multidisciplinary Perspectives on Code-Switching, ed. K. De Bot, L. Isurin, and D. Winford. Amsterdam. 2010 Contact Explanations in Linguistics. Pp. 31–47 in Hickey 2010a. Thomason, S. G., and T. Kaufman 1988 Language Contact, Creolization, and Genetic Linguistics. Berkeley.

Bibliography

259

Tonneau, R.-M. 1955 Sancti Ephraem Syri in Genesim et in Exodum Commentarii. CSCO 152–53. Louvain. Torrey, C. C. 1935 A Syriac Parchment from Edessa of the Year 243 a.d. ZS 10: 32–45. Torczyner, H. 1916 Die Entstehung des semitischen Sprachtypus. Vienna. Trask, R. L. 2000 The Dictionary of Historical and Comparative Linguistics. Edinburgh. Tropper, J. 1993 Die Inschriften von Zincirli. Münster. 1997 Lexikographische Untersuchungen zum Biblisch-Aramäischen. JNSL 23: 105–28. 2000 Ugaritische Grammatik. Münster. 2002 Altäthiopisch: Grammatik des Geʿez mit Übungstexten und Glossar. ELO 2. Münster. 2003 Das ugaritische Briefformular. Pp. 63–74 in Bote und Brief, A. Wagner. Frankfurt am Main. Tubach, J. 2011 Zur Interpretation des Perlenliedes: Exegetische Prämissen und ihre Schluss­ folgerungen. Pp. 231–58 in Syrien im 1.–7. Jahrhundert nach Christus: Akten des 1. Tübinger Tagung zum Christlichen Orient (15.–16. Juni 2007), ed. D. Bumazhnov and H. R. Seeliger. Tübingen. Türker, E. 1999 Codeswitching and ‘Loan Translations’. Pp. 111–23 in Brendemoen, Lanza, and Ryen 1999. Unnik, W. C. van 1937 Nestorian questions on the administration of the eucharist, by Ishoʿyabh IV. Haarlem. Urbainczyk, T. 2002 Theodoret of Cyrrhus: The Bishop and the Holy Man. Ann Arbor, MI. Van Coetsem, F. 1988 Loan Phonology and the Two Transfer Types in Language Contact. Publications in Language Sciences 27. Dordrecht. 1990 Review of Thomason and Kaufman 1988, Lehiste 1988, and Wardhaugh 1987. Language in Society 19: 260–68. 1995 Outlining a Model of the Transmission Phenomenon in Language Contact. Leuvense Bijdragen 84: 63–85. 1997 Language Contact: Neutralization as the Missing Link in Language Transmission. Leuvense Bijdragen 86: 357–71. 2000 A General and Unified Theory of the Transmission Process in Language Contact. Heidelberg. 2003 Topics in Contact Linguistics. Leuvense Bijdragen 92: 27–100. Van Rompay, L. 1991 Some Reflections on the Use of Post-Predicative hwā in Classical Syriac. Pp. 210–19 in Studies in Hebrew and Aramaic Syntax Presented to Professor

260

Bibliography

J. Hoftijzer, ed. K. Jongeling, H. L. Murre-Van den Berg, and L. Van Rompay. Leiden. 1994 Some Preliminary Remarks on the Origins of Classical Syriac as a Standard Language. Pp. 70–89 in Semitic and Cushitic Studies, ed. G. Goldenberg and S. Raz. Wiesbaden. 1996 The Christian Syriac Tradition of Interpretation. Pp. 612–41 in Hebrew Bible / Old Testament. The History of Its Interpretation, vol. 1: From the Beginnings to the Middle Ages (until 1300), part 1: Antiquity, ed. M. Sæbø. Göttingen. 2000 Past and Present Perceptions of Syriac Literary Tradition. Hugoye 3: 71–103. 2001 Review of Nöldeke 1904. Hugoye 4: 280–84. 2007a Syriac Studies: The Challenges of the Coming Decade. Hugoye 10: 23–35. 2007b Oh That I Had Wings like a Dove! Some Remarks on Exclamatory Clauses in Syriac. Pp. 91–105 in Studies in Semitic and General Linguistics in Honor of Gideon Goldenberg, ed. T. Bar and E. Cohen. Münster. 2010 Humanity’s Sin in Paradise: Ephrem, Jacob of Sarug, and Narsai in Conversation. Pp. 199–217 in Jacob of Serugh and His Times, G. A. Kiraz. Pis­ca­ta­way,  NJ. Vaschalde, A. A. 1902 Three Letters of Philoxenus, Bishop of Mabbôgh (485–519): Being the Letter to the Monks, the First Letter to the Monks of Beth-Gaugal, and the Letter to Emperor Zeno. Rome. 1915 Babai Magnai. Liber de Unione. CSCO 79. Louvain. Vergote, J. 1938 Grec biblique. Pp. 1319–69 in Dictionnaire de la Bible: Supplément, vol. 3. Expiation-Herméneutique, ed. L. Pirot. Paris. Vildomec, V. 1963 Multlingualism. Leiden. Voigt, R. 1979 Die Laterale im Semitischen. WO 10: 93–114. 1992 Die Lateralreihe /ś ṣ́ ź/ im Semitischen. ZDMG 142: 37–52. 1998a Griechischer Wortindex zu Anton Schalls ‘Studien über griechische Frendwörter im Syrischen’. Pp. 539–43 in Symposium Syriacum VII (Uppsala 1996), ed. R. Lavenant. OCA 256. Rome. 1998b Das emphatische p des Syrischen. Pp. 527–37 in Symposium Syriacum VII (Uppsala 1996), ed. R. Lavenant. OCA 256. Rome. 1999–2000  Griechische Fremdwörter im Syrischen: Eine Bibliographie. GraecoArabica 7/8: 555–70. 2007 Mandaic. Pp. 149–66 in Kaye 2007: vol. 1. Vööbus, A. 1960 Syriac and Arabic Documents Regarding Legislation Relative to Syrian Asceticism. Stockholm. 1970a Syrische Kanonessammlungen: Ein Beitrag zur Quellenkunde, vol. 1: Westsyrische Originalurkunden 1, A. CSCO 307. Louvain. 1970b Syrische Kanonessammlungen: Ein Beitrag zur Quellenkunde, vol. 1. Westsyrische Originalurkunden 1, B. CSCO 307. Louvain. 1979 The Didascalia Apostolorum in Syriac. CSCO 401–2, 407–8. Louvain.

Bibliography

261

Walters, J. 2015 Re-considering the Compositional Unity of Aphrahat’s Demonstrations. Paper presented at VIIth North American Syriac Symposium, The Catholic University of America, June 21–24, Washington, DC. Waltke, B. K., and M. O’Connor 1990 An Introduction to Biblical Hebrew Syntax. Winona Lake, IN. Wardhaugh, R. 1987 Languages in Competition, Dominance, Diversity, and Decline. Oxford. Wasserstein, A. 1993 A Note on the Phonetic and Graphic Representation of Greek Vowels and of the Spiritus Asper in the Aramaic Transcription of Greek Loanwords. Scripta Classica Israelica 12: 200–208. 1995 Non-Hellenized Jews in the Semi-Hellenized East. Scripta Classica Israelica 14: 111–37. Watt, J. W. 1978 Philoxenus of Mabbug: Fragments of the Commentary on Matthew and Luke. CSCO 392–93. Louvain. 1985 Antony of Tagrit as a Student of Syriac Poetry. Le Muséon 98: 261–79. 1986 The Fifth Book of the Rhetoric of Antony of Tagrit. CSCO 480–81. Louvain. 1987 Antony of Tagrit on Rhetorical Figures. Pp. 317–25 in IV Symposium Syriacum 1984: Literary Genres in Syriac Literature, ed. H. J. W. Drijvers, R. Lavenant, S.J., C. Molenberg, and G. J. Reinink. OCA 229. Rome. 1989 Syriac Panegyric in Theory and Practice. Le Muséon 102: 271–98. 1990 The Rhetorical Structure of the Memra of Eli of Qartamin on Philoxenus of Mabbug. Pp. 299–306 in V Symposium Syriacum 1988, ed. R. Lavenant, S.J. OCA 236. Rome. 1993a Grammar, Rhetoric, and the Enkyklios Paideia in Syriac. ZDMG 143: 45–71. 1993b The Syriac Reception of Platonic and Aristotelian Rhetoric. ARAM 5: 579–601. 1994a Syriac Rhetorical Theory and the Syriac Tradition of Aristotle’s Rhetoric. Pp. 243–60 in Peripatetic Rhetoric after Aristotle, ed. W. W. Fortenbaugh and D. C. Mirhady. New Brunswick, NJ. 1994b The Philosopher-King in the ‘Rhetoric’ of Antony of Tagrit. Pp. 245–58 in VI Symposium Syriacum 1992, ed. R. Lavenant, S.J. OCA 247. Rome. 1995 From Themistius to al-Farabi: Platonic Political Philosophy and Aristotle’s Rhetoric in the East. Rhetorica 13: 17–41. 1998 From Synesius to al-Farabi: Philosophy, Religion, and Rhetoric in the Christian Orient. Pp. 265–77 in Symposium Syriacum VII: 1996, ed. R. Lavenant, S.J. OCA 256. Rome. 1999 A Portrait of John bar Aphtonia, Founder of the Monastery of Qenneshre. Pp. 155–69 in Drijvers and Watt 1999. 2005 Aristotelian Rhetoric in Syriac: Bar Hebraeus. Butyrum Sapientiae. Book of Rhetoric. Aristoteles Semitico-Latinus 18. Leiden. 2009 Literary and Philosophical Rhetoric in Syriac. Pp. 141–54 in Literary and Philosophical Rhetoric in the Greek, Roman, Syriac, and Arabic Worlds, ed. F. Woerther. Hildesheim.

262

Bibliography

Weinreich, U. 1953 Languages in Contact: Findings and Problems. New York. Weitzman, M. P. 1999 The Syriac Version of the Old Testament. Cambridge. Welles, C. B., R. O. Fink, and J. F. Gilliam 1959 The Excavations at Dura-Europus. Final Report V, Part 1: The Parchments and Papyri. New Haven, CT. Weninger, S. 2011 The Semitic Languages: An International Handbook. Berlin. Wertheimer, A. 2001a Special Types of Cleft Sentences in Syriac. JSS 46: 221–41. 2001b The Functions of the Syriac Particle d-. Le Muséon 114: 259–89. 2001c More Thoughts about Cleft Sentences. Hebrew Linguistics 49: 21–34. [Hebrew] 2002 Syriac Nominal Sentences. JSS 47: 1–21. Whitney, W. D. 1881 On Mixture in Language. Transactions of the American Philological Association (1869–1896) 12: 5–26. [Repr. pp. 170–91 in Whitney on Language: Selected Writings of William Dwight Whitney, ed. M. Silverstein. Cambridge, MA, 1971] Wichmann, S., and J. Wohlgemuth 2008 Loan Verbs in a Typological Perspective. Pp. 89–121 in Stolz, Bakker, and Salas Palomo 2008. Windisch, E. 1897 Zur Theorie der Mischsprachen und Lehnwörter. Berichte über die Verhandlungen der Königlich Sächsischen Gesellschaft der Wissenschaften zu Leipzig: Philologisch-historische Classe 49: 101–26. Winford, D. 2003 An Introduction to Contact Linguistics. Oxford. 2005 Contact-Induced Changes: Classification and Processes. Diachronica 22: 373–427. 2007 Some Issues in the Study of Language Contact. Journal of Language Contact. THEMA 1: 22–39. 2009 On the Unity of Contact Phenomena and Their Underlying Mechanisms: The Case of Borrowing. Pp. 279–305 in Multidisciplinary Approaches to Code Switching, L. Isurin, D. Winford, K. De Bot. Studies in Bilingualism 41. Amsterdam. Winn, R. E. 2011 Eusebius of Emesa: Church and Theology in the Mid-Fourth Century. Washington, DC. Winter, W. 1973 Areal Linguistics: Some General Considerations. Pp. 135–47 in Current Trends in Linguistics, vol. 11: Diachronic, Areal, and Typological Linguistics, ed. T. A. Sebeok. The Hague. Wohlgemuth, J. 2009 A Typology of Verbal Borrowings. Berlin.

Bibliography

263

Woodard, R. D. 2004a The Cambridge Encyclopedia of the World’s Ancient Languages. Cambridge. 2004b Attic Greek. Pp. 614–49 in Woodard 2004a. Wright, W. 1867 Two Epistles of Mār Jacob, Bishop of Edessa. Journal of Sacred Literature and Biblical Record 10: 430–60. 1869 The Homilies of Aphraates, the Persian Sage. London. 1870–72  Catalogue of Syriac Manuscripts in the British Museum Acquired since the Year 1838. London. 1871a Apocryphal Acts of the Apostles. London. 1871b Fragments of the ‫ ܬܘܪܨ ܡܡܠ�ܠܐ ܢܗܪܝܐ‬or Syriac Grammar of Jacob of Edessa. Clerkenwell. 1894 A Short History of Syriac Literature. London. 1896–98  A Grammar of the Arabic Language. 3rd ed. Cambridge Wright, W., and N. McLean 1898 The Ecclesiastical History of Eusebius Pamphili, 265–339, Bishop of Caesarea. London. Yahalom, J., and M. Sokoloff 1999 Jewish Palestinian Aramaic Poetry from Late Antiquity. Jerusalem. [Hebrew] Yakubovich, I. S. 2010 Sociolinguistics of the Luvian Language. Leiden. Yallop, C. 1977 Alyawarra: An Aboriginal Language of Central Australia. Canberra. Zewi, T. 1996 The Definition of the Copula and the Role of 3rd Independent Personal Pronouns in Nominal Sentences of Semitic Languages. Folia Linguistica Historica 17: 41–55.

Index of Authors Aartun, K.  176 Adams, J. N.  6 Aggoula, B.  29 Aikhenvald, A. Y.  11, 52, 140, 147, 149, 150, 151 Albeck, C.  208 Alexander, P. S.  208 Al-Ghul, O.  85 Al-Jallad, A.  85 Allen, P.  38 Allen, W. S.  65, 66, 75, 80, 81, 86, 87, 88, 94 Appel, R.  149, 151 Assemani, G. S.  2 Attridge, H. W.  205 Avinery, I.  153 Baarda, T.  31 Baehrens, W. A.  54 Bakker, P.  5, 19, 116, 181 Bar-Asher, E.  181 Bardenhewer, O.  37 Bardy, G.  37 Barsoum, I. A.  2 Bátori, I.  15 Bauer, H.  158 Baumstark, A.  2, 45 Beck, E.  57, 65, 78, 79, 82, 92, 99, 111, 117, 118, 125, 207, 212, 215, 217, 218, 220, 221, 222 Becker, A. H.  5, 31, 32, 34, 36, 142 Bedjan, P.  79, 118, 184 Beeston, A. F. L.  176 Bell, G.  4, 31 Bellinger, A. R.  29 Benedictus, P.  76, 115 Ben-Ḥayyim, Z.  165, 166, 167, 207 Beniak, E.  140, 151

Beyer, K.  1, 6, 179 Bidez, J.  199 Biesen, K. den  31 Binder, M.  91 Blau, J.  89, 158, 214 Bloomfield, L.  11, 129 Blum, G. G.  34, 35 Bonnet, M.  205 Bordreuil, P.  176 Bowden, J.  149 Bowersock, G. W.  28, 29 Boyarin, D.  58 Braun, O.  156 Bremmer, J. N.  205 Briant, P.  25, 26 Brock, S. P.  1, 2, 3, 4, 5, 7, 28, 29, 31, 32, 34, 35, 36, 37, 38, 44, 46, 47, 48, 49, 52, 53, 54, 55, 56, 57, 59, 64, 75, 76, 78, 79, 80, 81, 84, 90, 93, 94, 97, 98, 99, 103, 109, 110, 112, 113, 115, 116, 117, 118, 119, 121, 122, 124, 125, 126, 127, 128, 141, 142, 144, 146, 175, 190, 192, 193, 194, 200, 201, 204, 205, 210, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221 Brockelmann, C.  46, 47, 64, 72, 114, 132, 133, 153, 175, 176, 177, 192 Brooks, E. W.  33, 48, 51, 56, 59, 60, 75, 76, 77, 78, 79, 80, 82, 109, 113, 114, 118, 119, 125, 205, 224, 225 Brown, P.  26 Browning, R.  61, 63, 66 Bruns, P.  31 Buber, S.  208 Buccini, A. F.  19 Budge, E. A. W.  125 Burtea, B.  207

264

Index of Authors Buth, R.  179 Butts, A. M.  3, 45, 54, 68, 88, 91, 93, 110, 111, 117, 121, 122, 123, 124, 127, 128, 141, 154, 155, 157, 168, 175, 189, 218, 219 Cameron, A.  26 Campbell, L.  11, 18, 144, 147, 148, 149 Canali de Rossi, F.  4, 29, 31 Canivet, P.  36, 37 Cantineau, J.  57 Carlson, T.  3 Chabot, J.-B.  3, 76, 131 Charlesworth, J. H.  117, 126, 159, 203, 212, 213, 215, 216, 217, 218, 221 Chialà, S.  82 Ciancaglini, C. A.  2, 12, 48, 49, 54, 57, 112, 113, 115, 116, 132, 175, 192, 199 Clyne, M. G.  13 Coakley, J. F.  90 Cockle, W. E. H.  29 Contini, R.  57, 213, 214, 217, 220 Cook, E. M.  212, 217, 218, 219, 222 Corluy, J.  184 Corne, C.  151 Cotton, H. M.  29 Cross, F. M., Jr.  158 Crum, W. E.  112 Csató, É. Á. 150 Curnow, T. J.  195 Dalman, G.  56 Daniel, A. R.  85 Daniels, P. T.  66, 89 Daris, S.  63 Datema, C.  112 Degen, R.  158 Denniston, J. D.  181, 191 Díez Macho, A.  168 Díez Merino, L.  179 Dillmann, A.  176 Dillon, M.  14 Dion, P-.E.  176 Dixon, R. M. W.  11

265

Doran, R.  34 Dorian, N.  148, 197 Draguet, R.  115 Drijvers, H. J. W.  1, 7, 28, 29, 52, 77, 107, 108, 124, 127, 142, 146, 157, 160, 181, 185, 199, 203, 212, 213, 214, 215, 216, 217, 218, 220, 221, 223, 224 Drijvers, J. W.  34 Drower, E. S.  165 Duval, R.  45, 53, 92, 99, 153 Ebied, R. Y.  143 Emeneau, M. B.  149 Esch, E.  148 Falla, T. C.  174, 181 Feissel, D.  30, 50, 212, 214, 219 Fiey, J.-M.  31 Fink, R. O.  27, 29, 63 Fischer, W.  176, 177 Fitzmyer, J. A.  1, 59, 179, 180 Flom, G. T.  14 Foley, W. A.  13 Folmer, M.  59 Fox, J.  120, 122, 123, 132 Fraade, S.  29, 189, 209 Frankenberg, W.  111, 128 Freedman, D. N.  158 Frishman, J.  77, 92, 118, 119, 208 Frothingham, A. L.  224, 225 Garr, W. R.  158 Gascou, J.  27, 29, 30, 50, 212, 214, 219 Gensler, O. D.  158 Ghanem, J. R.  33, 113 Gibson, J. C. L.  158 Gignac, F. T.  61, 62, 65, 66, 70, 71, 72, 73, 74, 79, 81, 86, 87, 88, 94, 101 Gilliam, J. F.  27, 29, 63 Gołąb, Z.  13 Goldenberg, G.  131, 153, 154, 155, 156 Goldstein, J. A.  29 Golovko, E. V.  19

266

Index of Authors

Goshen-Gottstein, M. H.  4, 153 Gottheil, R. J. H.  45 Gragg, G.  83 Grainger, J. D.  26 Greenberg, J.  131 Greenfield, J. C.  58 Gröber, G.  14 Gross, S.  184 Guidi, I.  224, 225 Gumperz, J. J.  13, 150 Gutas, D.  3 Guy, G. R.  16, 21, 22 Gwilliam, G. H.  68 Haar Romeny, B. ter  36, 202 Haefeli, L.  204 Hafez, O.  11, 52 Halleux, A. de  32, 110 Harmon, A. M.  36 Harrak, A.  26 Harrington, D. J.  180 Harris, A. C.  144, 147, 148, 149 Harvey, S. A.  56 Harviainen, G.  66, 75, 76, 77, 78, 79, 80, 81 Hasan-Rokem, G.  208 Haspelmath, M.  11, 16, 46, 47, 48, 52, 53 Hasselbach, R.  176, 184 Haugen, E.  11, 13, 46, 97, 148, 151 Hayasi, T.  150 Hayward, C. T. R.  38 Heal, K.  47 Healey, J. F.  1, 3, 28, 29, 52, 54, 57, 60, 124, 160, 199, 200, 212, 213, 214, 215, 216, 217, 220, 222 Heath, J.  13, 149, 151 Heine, B.  11, 12, 13, 52, 139, 140, 143, 144, 147, 148, 150, 151, 175, 176, 180, 184, 201, 202 Heine, H.  147 Henry, R.  36 Hetzron, R.  11, 131, 149 Hezser, C.  209 Hickey, R.  7, 12, 16, 148, 196, 201

Hoffmann, G.  45, 77, 80, 82, 92, 109, 118 Höfner, M.  177 Hopper, P. J.  147, 184 Horrocks, G. C.  61, 88 Howard, G.  60, 82, 92, 223, 225 Hoyland, R. G.  3 Huehnergard, J.  67, 131, 147, 184 Hug, V.  177 Humbert, J.  144, 181, 191 Janda, I. H.  12, 15 Jansma, T.  28, 31 Jastrow, M.  60, 72, 74, 193, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222 Jenner, K. D.  154, 155, 162 Jespersen, O.  14, 43, 112 Johanson, L.  13, 46, 149, 150, 201 Johnson, S. F.  6, 37, 38, 89, 202 Jones, M. C.  148 Joosten, J.  48, 153, 154, 155, 156, 157, 161, 163, 175, 183, 204 Kaufman, S. A.  1, 2, 7, 11, 12, 13, 14, 15, 16, 18, 19, 20, 21, 22, 23, 39, 43, 47, 103, 132, 149, 196 Kennedy, H.  29 Khan, G.  158 King, D.  5, 34, 52, 143, 144 King, R.  12, 129, 140, 147, 149, 196, 197 Kıral, F.  150 Kiraz, G. A.  2, 47, 64, 68, 75, 78, 81, 82, 83, 87, 90, 98, 161, 193 Kmosko, M.  47, 77, 78, 79, 92, 117, 118, 125, 214, 215, 216, 219 Kogan, L.  66, 67, 85 Krauss, S.  73, 74, 75, 78, 81, 87, 94, 95, 98, 99, 104, 112, 117 Krebernik, M.  121 Kuteva, T.  139, 140, 143, 144, 147, 148, 150, 151, 175, 176, 180, 184, 201, 202 Kutscher, E. Y.  56, 85, 148, 168, 208

Index of Authors Lado, R.  15 Lagarde, P. A. de  125, 127 Lake, K.  38 Lane, E. W.  176, 193 Lattke, M.  159, 175 Leander, P.  158 Lebon, J.  32 Lee, S.  125, 127, 219 Leloir, L.  87, 118 Leroy-Molinghen, A.  36 Levey, S.  7, 12, 140, 148, 150, 154, 197 Lieberman, S.  209 Liebeschuetz, J. H. W. G.  29 Littmann, E.  177 Louden, M. L.  21, 23, 196 Lund, J. A.  47, 177, 217, 218 Machiela, D. A.  179, 180 MacMullen, R.  29, 35, 37 Macomber, W. F.  76, 92, 118, 119 Macuch, R.  2, 165 Mandel, P.  208 Mango, M. M.  27 Mankowski, P. V.  48 Markschies, C.  8 Marrassini, P.  85 Mason, H. J.  54, 132, 133, 134 Matras, Y.  18, 19, 46, 52, 148, 149, 150 Mayser, M.  61, 62, 65, 70, 71, 72, 73, 74, 79, 88 McLean, N.  78, 156, 214 Meechan, M.  52 Menz, A.  150 Merkelbach, M.  29 Merx, A.  156 Meyer, E. A.  29 Michelson, D.  32 Militarev, A.  67 Millar, F.  25, 26, 27, 28, 29, 30, 31, 32, 34, 35, 36, 37, 38, 56, 202 Miller, C.  52, 120 Mingana, A.  118, 125 Mitchell, C. W.  78, 117, 220, 221 Moller, G. I.  29

267

Mønnesland, S.  148, 149 Moss, Y.  38 Mougeon, R.  140, 151 Mous, M.  19 Mufwene, S. S.  21, 151 Müller-Kessler, C.  83, 132, 169, 190, 193, 194, 207, 209 Muraoka, T.  2, 56, 64, 66, 89, 98, 153, 154, 155, 157, 158, 172 Murray, R.  199 Muysken, P.  19, 50, 112, 120, 149, 151 Myers-Scotton, C.  50, 52, 53, 150 Nadkarni, M. V.  149 Nau, F.  28, 49, 76, 77, 79, 82, 224, 225 Nöldeke, T.  2, 64, 66, 68, 80, 81, 83, 89, 95, 97, 98, 99, 100, 101, 102, 103, 109, 112, 120, 121, 122, 123, 124, 131, 132, 153, 154, 155, 156, 165, 169, 175, 181, 183, 204 Noonan, M.  149 O’Connor, M.  142, 182 Olinder, G.  31 Ortiz de Urbina, I.  2 Overbeck, J. J.  34, 35, 77, 80, 81, 82, 92, 114, 117, 131, 185, 188, 203, 205, 207, 223, 225 Owens, R. J., Jr.  31 Pagano, P.  57, 213, 214, 217, 220 Palmer, A.  87, 88, 101, 132 Papen, R. A.  19 Pardee, D.  176 Parisot, I.  31, 47, 57, 62, 68, 73, 77, 79, 82, 92, 117, 118, 125, 127, 185, 193, 215, 218, 223, 225 Pat-El, N.  4, 31, 153, 154, 163, 199 Payne Smith, R.  46, 189 Petermann, H.  165 Petit, F.  36 Peursen, W. T. van  153, 174, 181 Phillips, G.  64, 65, 75, 80, 81, 82, 93, 101, 113, 185, 189 Pierre, M. J.  31

268

Index of Authors

Poplack, S.  7, 12, 52, 120, 140, 148, 150, 154, 196, 197, 201, 202 Porten, B.  56, 155, 158, 177 Possekel, U.  27, 28, 29, 199, 201 Pray, B.  150 Price, R. M.  36 Probert, P.  94 Pusey, P. E.  68

Sokoloff, M.  46, 47, 114, 116, 133, 168, 169, 190, 193, 194, 207, 209 Sperber, D.  209 Sridhar, S. N.  150 Stauber, J.  29 Stein, P.  176 Steingass, F. J.  132 Stolz, T.  13

Ramelli, I.  28 Ratcliffe, R. R.  131 Rayfield, J. R.  12, 15 Reinink, G. J.  131, 146 Ridolfini, F. P.  31 Rignell, K.-E.  224, 225 Roca, I.  89 Rochette, B.  26, 54, 63 Roey, A. van  76 Rosenthal, F.  56, 109, 139, 158, 168, 208 Ross, M. D.  12, 13, 16, 21, 22, 26, 28, 29, 148, 149 Rostovtzeff, M.  195 Rubin, A. D.  131, 147, 184

Tal, A.  189 Tauwinkl, C. F.  2 Tavernier, J.  54 Taylor, D. G. K.  3, 6, 29, 45, 46, 185 Teixidor, J.  28, 29, 30, 50, 212, 214, 219 Terpelyuk, A. A.  65 Teule, H.  2 Theodor, J.  208 Thomason, S. G.  7, 11, 12, 13, 14, 15, 16, 18, 19, 20, 21, 22, 23, 24, 39, 52, 140, 148, 149, 150, 196, 201 Tonneau, R.-M.  218, 221, 222 Torczyner, H.  177 Torrey, C. C.  29 Trask, R. L.  11, 12 Traugott, E. C.  147, 184 Tropper, J.  83, 158, 176, 177 Türker, E.  151

Sakel, J.  46, 52, 147, 148, 149 Sandfeld, M. K.  13 Sankoff, D.  52, 120 Sapir, E.  1, 43, 129 Sartre, M.  25, 26 Schall, A.  46, 48, 54, 55, 64, 73, 74, 80, 81, 83, 97, 106, 113, 127, 199, 203, 204 Scher, A.  5, 119 Schlesinger, M.  164 Schwartz, E.  143 Segal, J. B.  26, 27, 28, 75, 83, 90 Shepardson, C.  199 Sihler, A. L.  11 Silva-Corvalán, C.  12, 13, 140, 149, 150, 151, 196, 198 Smelik, W. F.  6, 208 Smith, K.  46, 77, 189 Smits, C.  12, 13, 16, 17, 18, 22, 23, 39, 129, 198 Smyth, H. W.  162, 187

Urbainczyk, T.  36, 37 Vakhtin, N. B.  19 Van Coetsem, F.  11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30, 37, 39, 40, 129, 143, 196, 198 Van Rompay, L.  2, 6, 31, 34, 36, 112, 115, 153, 154, 156, 174, 175, 183 Vaschalde, A. A.  155, 224, 225 Vergote, J.  14, 15 Vildomec, V.  13, 15 Voigt, R.  46, 64, 81, 83, 84, 207 Vööbus, A.  34, 76, 115 Walters, J.  31 Waltke, B. K.  142, 182 Wasserstein, A.  54, 56, 75, 76, 79

Index of Authors Watt, J. W.  35, 77, 92 Weinreich, U.  12, 13, 17, 18, 25, 129, 139, 143, 144, 150, 151 Weitenberg, J. J. S.  36 Weitzman, M. P.  185, 203 Welles, C. B.  27, 29, 63 Wertheimer, A.  153, 154, 155, 157 Whitney, W. D.  18 Wichmann, S.  7, 13, 112, 116 Wickham, L. R.  143 Wilson, R.  13, 150 Windisch, E.  13, 14, 15 Winford, D.  12, 16, 23, 50, 52, 104, 117, 120, 129, 149, 150, 196 Winn, R. E.  36

269

Winter, W.  13, 15, 129 Wohlgemuth, J.  7, 11, 13, 18, 52, 53, 112, 115, 116, 149 Woodard, R. D.  81, 86, 88, 94 Wright, W.  2, 28, 31, 49, 64, 65, 68, 75, 78, 82, 87, 92, 116, 117, 126, 127, 155, 156, 176, 185, 188, 193, 205, 213, 214, 215, 216, 217, 218, 221, 223, 224, 225 Yahalom, J.  168, 207 Yakubovich, I. S.  6, 50 Zewi, T.  155

Index of Biblical Sources Hebrew Bible

Genesis 2:19 146 4:10 146 4:26 189 15:15 146 19:3  146, 218 22:5 145 25:8 146 27:22 146 30:30 219 32:4 146 37:18 146 37:25 80 39:4 214 43:11 80 43:19 146 43:23 146 44:17 146 44:19 214 44:24 146 44:34 168 45:4 146 45:9 146 49:11 217 Exodus 12:37 217 12:42 194 12:47 213 25:31 213 26:6 219 28:20 222 30:13 214 30:24 218 32:30 117 35:11 219

Exodus (cont.) 35:16 213 37:16 216 37:17 213 38:5 213 39:13 222 39:39 213 Leviticus 15:9 215 24:10 221 25:35 219 Numbers 4:7  216, 217 23:3 117 25:1 222 Deuteronomy 20:20 222 27:18 218 Judges 3:23 214 6:38 216 1 Samuel 1:15 169 2:12 218 10:5 220 17:4 213 17:7 125 26:7  74, 222 2 Samuel 12:31 216

270

Isaiah 3:24 222 5:12 216 11:15 219 57:8 219 58:7 126 1 Kings 2:5 221 6:3 220 6:8 81 7:4 214 7:27 213 7:50 217 10:22 221 2 Kings 5:5 214 9:25 215 10:22 72 Jeremiah 5:1 219 9:20 219 36:23  72, 220 Ezekiel 19:9 216 30:9 216 34:8 219 40:38 214 40:45 214 40:46 214 41:10 214 42:1 214 42:4 214

271

Index of Biblical Sources Ezekiel (cont.) 42:5 214 42:7 214 Psalms 69:9 126 Song of Songs 3:2 219 Job 31:32 126 Daniel 2:11 159 2:13 115 2:14 179 2:15 179 2:17 179 2:19 179 2:25 179 2:26 159 2:35 179 2:36 159 2:46 179 2:48 179 3:3 179 3:5  56, 82, 221 3:10  56, 82, 221 3:13 179 3:14 159 3:15 159 3:18 159

Deuterocanonical Books 1 Esdras 3:14 76

New Testament Matthew 5:15 217 8:5 222 10:29 213 11:11 98 12:8 161 21:13 125

Daniel (cont.) 3:19 179 3:24 179 3:26 179 3:30 179 4:4 179 4:16 179 5:6 179 5:8 179 5:9 179 5:13 179 5:17 179 5:24 179 5:29 179 6:4 179 6:5 179 6:6 179 6:7 179 6:12 179 6:13 179 6:14 179 6:15 179 6:16 179 6:17 179 6:19 179 6:20 179 6:22 179 6:24 179 6:26 179 7:1 179 7:11 179 7:19 179 20:1 159

Ruth 1:14  174, 175, 181, 182 1:18  182, 183 2:10 126 3:8 145 3:11  191, 192 Daniel 5:3 179 Ezra 4:9 179 4:18 214 4:23 179 4:24 179 5:2 179 5:4 179 5:9 179 5:16 179 6:1 179 6:13 179 8:36 214 1 Chronicles 11:23 125 20:3 216 20:5 125 2 Chronicles 14:2  212, 213 31:1 213

Judith 16:14 125

Tobit 2:13 125

Matthew (cont.) 23:25  57, 218 25:35 126 25:38 126 25:43 126 25:44 126 27:7 126

Matthew (cont.) 27:38 125 Mark 6:9  219, 221 6:21 222 11:4 190

272

Index of Biblical Sources

Mark (cont.) 11:17 125 15:27 125 15:43 213 16:1 78

Luke (cont.) 19:46 125 22:52 125 23:50 213 23:56 78 24:1 78

Luke 1:13 146 1:63 68 6:17 94 10:30 125 11:48 119 12:6 213 18:25 193

Acts (cont.) 9:43 125 10:6 125 12:8 221 28:2 125 Romans 1:14 125

John 11:44 222 11:54 82 18:12 222

1 Corinthians 14:11 125

Acts 1–7  185, 223, 225

Colossians 3:11 125

‫‪Index of Syriac Words‬‬

‫‪78, 117‬‬ ‫‪120‬‬ ‫‪75‬‬ ‫‪153–73, 197–98,‬‬ ‫‪201, 204, 206, 210‬‬ ‫‪214‬‬ ‫‪78‬‬ ‫‪114–15‬‬ ‫‪104‬‬ ‫‪104‬‬ ‫‪104‬‬ ‫‪125–26, 218‬‬ ‫‪112‬‬ ‫‪112, 141‬‬ ‫‪141‬‬ ‫‪76‬‬ ‫‪76‬‬ ‫‪76‬‬ ‫‪103, 132, 134–35‬‬ ‫‪132, 134–35‬‬ ‫‪59, 76‬‬ ‫‪59, 76‬‬ ‫‪144–47, 197–98‬‬ ‫‪87, 212‬‬ ‫‪75‬‬ ‫‪86, 106–8, 130–31‬‬ ‫‪107–8, 130–31‬‬ ‫‪106–8, 131‬‬ ‫‪103, 220‬‬ ‫‪132, 134–34‬‬ ‫‪132, 134–34‬‬ ‫‪132‬‬ ‫‪132‬‬ ‫‪134‬‬ ‫‪134‬‬ ‫‪127‬‬ ‫‪127‬‬

‫)‪ (cont.‬ܐ‬ ‫ ܐܝܩܐ‬ ‫ ܐܝܩܘܢܐ‬ ‫ ܐܝܪܛܝܘܢ‬ ‫ ܐܝܬ‬ ‫ ܐܟܣܕܪܐ‬ ‫ ܐܟܣܘܪܝܐ‬ ‫ ܐܟܣܘܪܣܬܝܢܐ‬ ‫ ܐܟܣܢܕܘܟܝܘܢ‬ ‫ ܐܟܣܢܕܟܝܢ‬ ‫ ܐܟܣܢܘܕܘܟܐ‬ ‫ ܐܟܣܢܝܐ‬ ‫ ܐܠܗ‬ ‫ ܐܠܗܐ‬ ‫ ܐܠܗܝܐ‬ ‫ ܐܠܘܣܪܝܩܘܢ‬ ‫ ܐܠܝܣܪܝܩܘܢ‬ ‫ ܐܠܣܝܪܝܩܘܢ‬ ‫ ܐܠܦܐ‬ ‫ ܐܠܦܪܐ‬ ‫ ܐܡܘܠܓܝܘܣ‬ ‫ ܐܡܠܘܓܝܣ‬ ‫ ܐܡܪ‬ ‫ ܐܢܕܪܝܢܛܐ‬ ‫ ܐܢܘܣܝܣ‬ ‫ ܐܢܢܩܐ‬ ‫̈‬ ‫ ܐܢܢܩܘܣ‬ ‫̈‬ ‫ ܐܢܢܩܣ‬ ‫ ܐܣܛܘܐ‬ ‫ ܐܣܛܘܢܐ‬ ‫ ܐܣܛܘܢܪܐ‬ ‫ ܐܣܛܣܝܣ‬ ‫ ܐܣܛܣܝܪܐ‬ ‫ ܐܣܟܘ�ܠܐ‬ ‫ ܐܣܟܠܪܐ‬ ‫ ܐܣܟ�ܡܐ‬ ‫ ܐܣܟܡܬܢܐ‬

‫‪273‬‬

‫‪45, 104, 106, 212‬‬ ‫‪106‬‬ ‫‪45, 65, 105, 116,‬‬ ‫‪192 n. 45‬‬ ‫‪105‬‬ ‫‪87, 212‬‬ ‫‪87‬‬ ‫)‪93 (table 5.9‬‬ ‫)‪93 (table 5.9‬‬ ‫‪213‬‬ ‫‪118‬‬ ‫‪110‬‬ ‫‪101‬‬ ‫‪101, 110‬‬ ‫‪100‬‬ ‫‪100‬‬ ‫‪100‬‬ ‫‪100‬‬ ‫‪77‬‬ ‫‪60‬‬ ‫‪86, 104‬‬ ‫‪110, 127–28‬‬ ‫‪110, 127–28‬‬ ‫‪91‬‬ ‫‪76‬‬ ‫‪75 n. 40‬‬ ‫‪75‬‬ ‫‪76‬‬ ‫‪99‬‬ ‫‪99‬‬ ‫‪65, 77-78, 192‬‬ ‫‪n. 45‬‬ ‫‪77–78, 215‬‬ ‫‪118 n. 60‬‬ ‫‪95, 119‬‬ ‫‪48-49‬‬ ‫‪76‬‬

‫ܐ‬ ‫ ܐܐܪ‬ ‫ ܐܐ̈ܪܘܣ‬ ‫ ܐܓܘܢܐ‬ ‫̈‬ ‫ ܐܓܘܢܐ‬ ‫ ܐܕܪܝܛܐ‬ ‫ ܐܕܪܝܢܛܐ‬ ‫̈‬ ‫ ܐܘܐܢܓܐܠܝܐ‬ ‫ ܐܘܐܢܓܠܝܣܛܐ‬ ‫ ܐܘܛܩܪܛܘܪ‬ ‫ ܐܘܢ‬ ‫ ܐܘܢܓ�ܠܐܝܬ‬ ‫̈‬ ‫ ܐܘܢܓܠܝܐ‬ ‫ ܐܘܢܓܠܝܘܢ‬ ‫ ܐܘܣܝܐ‬ ‫̈‬ ‫ ܐܘܣܝܐܣ‬ ‫̈‬ ‫ ܐܘܣܝܘܣ‬ ‫̈‬ ‫ ܐܘܣܝܣ‬ ‫ ܐܘܦܪܝܛܐ‬ ‫ ܐܘܦܪܟܐ‬ ‫ ܐܘܩܝܢܐ‬ ‫ ܐܘܩܝܢܐܝܬ‬ ‫ ܐܘܩܝܢܘܣ‬ ‫ ܐܘܪܓܢܘܢ‬ ‫ ܐܘܪܘܝܙܘܢܛܘܣ‬ ‫ ܐܘ̈ܪܛـܐܣܛܝܘܩܐܣ‬ ‫ ܐܘ̈ܪܛـܐܣܛܝܩܐܣ‬ ‫ ܐܘܪܝܙܘܢ‬ ‫ ܐܘ̈ܪܬܘܕܘܟܣܐ‬ ‫ ܐܘ̈ܪܬܘܕܘܟܣܘ‬ ‫ ܐܝܓܘܢܐ‬ ‫ ܐܝܓܡܘܢܐ‬ ‫ ܐܝܘ‬ ‫ ܐܝܛܐ‬ ‫ ܐܝܣܘܢ‬ ‫ ܐܝܦܕܪܡܘܢ‬

‫‪Index of Syriac Words‬‬

‫‪99‬‬ ‫‪116‬‬

‫)‪ (cont.‬ܐ‬ ‫ ܐ̈ܪܬܘܕܘܟܣܘ‬ ‫ ܐܬܠܝܣܐ‬

‫‪94, 102‬‬ ‫‪118‬‬ ‫‪213‬‬ ‫‪213‬‬ ‫‪54-55 n. 44‬‬ ‫‪125‬‬ ‫‪132‬‬ ‫‪132‬‬ ‫‪133‬‬ ‫‪94, 102‬‬ ‫‪94, 102‬‬ ‫‪81‬‬ ‫‪133‬‬ ‫‪59‬‬ ‫‪213‬‬ ‫‪125‬‬ ‫‪81‬‬

‫ܒ‬ ‫ ܒܐ�ܡܐ‬ ‫ ܒܕܓܘܢ‬ ‫ ܒܘܠܘܛـܐ‬ ‫ ܒܘܡܣܐ‬ ‫ ܒܘܪܓܐ‬ ‫ ܒܘܪܣܝܐ‬ ‫ ܒܙܝܩܐ‬ ‫ ܒܙܝܩܪܐ‬ ‫ ܒܝܠܕܪܐ‬ ‫ ܒܝܡ‬ ‫ ܒܝ�ܡܐ‬ ‫ ܒܝܪܘܢܐ‬ ‫ ܒܣܛܝܪܐ‬ ‫̈‬ ‫ ܒܣܝܠܝܩܘܣ‬ ‫ ܒܣܣ‬ ‫ ܒܪܒܪܝܐ‬ ‫ ܒܪܘܢܐ‬

‫‪131 n. 91‬‬ ‫‪193‬‬ ‫‪62, 68, 73‬‬ ‫ܒܕܓܘܢ ‪see‬‬ ‫‪90‬‬ ‫‪119, 181, 191–94,‬‬ ‫‪201, 210‬‬ ‫‪133‬‬ ‫‪73‬‬ ‫‪63‬‬ ‫‪132‬‬ ‫‪131–32‬‬ ‫‪102, 203, 213–14‬‬ ‫‪131–32‬‬ ‫‪193‬‬ ‫‪67-68‬‬ ‫‪67-68‬‬ ‫‪125‬‬ ‫‪59-60, 100, 105–6,‬‬ ‫‪108‬‬

‫ܓ‬ ‫ ܓܐܢܐܣ‬ ‫ ܓܐܪ‬ ‫ ܓܘܒܪܢܝܛܐ‬ ‫ ܓܘܢ‬ ‫ ܓܘܫ�ܡܐ‬ ‫ ܓܝܪ‬ ‫ ܓܠܝܪܐ‬ ‫ ܓܠܣ‬ ‫ ܓܡܢܣܝܢ‬ ‫ ܓܢܒܐ‬ ‫̈‬ ‫ ܓܢܣ‬ ‫ ܓܢܣܐ‬ ‫ ܓܢܬܐ‬ ‫ ܓܪ‬ ‫ ‪ /garbɔ/‬ܓܪܒܐ‬ ‫ ‪ /garḇɔ/‬ܓܪܒܐ‬ ‫ ܓܪܕܝܝܐ‬ ‫ܕ‬ ‫ ܕܘܓ�ܡܐ‬

‫‪274‬‬

‫)‪ (cont.‬ܐ‬ ‫ ܐܣܟܡܬܢܐܝܬ‬ ‫‪127‬‬ ‫ ܐܣܦܘܓܐ‬ ‫‪86‬‬ ‫‪ (< σπεῖρα) 84, 94‬ܐܣܦܝܪ‬ ‫‪ (< σφαῖρα) 84‬ܐܣܦܝܪ‬ ‫‪ (< σπεῖρα) 84, 94‬ܐܣܦܝܪܐ‬ ‫‪ (< σφαῖρα) 84‬ܐܣܦܝܪܐ‬ ‫ ܐܣܦܪܣܬܝܢܐ‬ ‫‪114–15 with‬‬ ‫‪n. 43‬‬ ‫ ܐܣܦܬܪܐ‬ ‫‪134‬‬ ‫ ܐܣܪܐ‬ ‫‪213‬‬ ‫ ܐܣܬܝܪܐ‬ ‫‪220‬‬ ‫ ܐܣܬܪܐ‬ ‫‪220‬‬ ‫̈‬ ‫ ܐܦܘܕܝܩܢܘ‬ ‫‪76 n. 41‬‬ ‫ ܐܦܘܣܝܐ‬ ‫‪103‬‬ ‫ ܐܦܘܩܪܝܣܪܐ‬ ‫‪133‬‬ ‫ ܐܦܛܘ‬ ‫‪76 n. 41‬‬ ‫ ܐܦܛܪܘܦܐ‬ ‫‪203, 214–15‬‬ ‫ ܐܦܝܩܘܣ‬ ‫‪75‬‬ ‫ ܐܦܝܬܝܛܐ‬ ‫‪94‬‬ ‫ ܐܦܝܬܝܛܘܬܐ‬ ‫‪94 n. 103‬‬ ‫ ܐܦܠܘܣ‬ ‫‪76 n. 41, 118‬‬ ‫ ܐܦܩܪܝܣܪܐ‬ ‫‪133‬‬ ‫ ܐܦܪܟܝܐ‬ ‫‪78 n. 45‬‬ ‫ ܐܩܘܢܐ‬ ‫‪86‬‬ ‫ ܐܩܛܣܛܣܝܐ‬ ‫‪53‬‬ ‫ ܐܩܝܢܡܘ‬ ‫‪79‬‬ ‫ ܐܩܠܝܕܐ‬ ‫‪98‬‬ ‫ ܐܩܠܝܣܓܕܝܩܘܣ‬ ‫‪61, 70‬‬ ‫ ܐܩܪܝܒܘܣ‬ ‫‪53 n. 33, 118‬‬ ‫ )‪ (< ἄρα‬ܐܪܐ‬ ‫‪118–19‬‬ ‫ )‪ (< ἆρα‬ܐܪܐ‬ ‫‪119‬‬ ‫ ܐܪܓܢܘܢ‬ ‫‪91‬‬ ‫ ܐܪܓܢܢ‬ ‫‪91‬‬ ‫ ܐܪܗܛܢܐ‬ ‫‪80‬‬ ‫ ܐܪܛܝܩܐ‬ ‫‪77‬‬ ‫ ܐ̈ܪܟܐ‬ ‫‪212‬‬ ‫ ܐܪܟܘܢܐ‬ ‫‪124 n. 74, 212–13‬‬ ‫ ܐܪܟܘܢܘܬܐ‬ ‫‪124 n. 74‬‬ ‫ ܐܪܟܘܣܝܐ‬ ‫‪110‬‬ ‫ ܐ̈ܪܟܣ‬ ‫‪110‬‬ ‫ ܐܪܡܝܐ‬ ‫‪141‬‬ ‫ ܐܪܣܝܘܛـܐ‬ ‫‪77‬‬ ‫ ܐܪܣܝܣ‬ ‫‪76-77, 100‬‬ ‫ ܐ̈ܪܣܝܣ‬ ‫‪100‬‬ ‫ ܐ̈ܪܬܕܘܟܣܐ‬ ‫‪99‬‬

‫‪275‬‬

‫‪Index of Syriac Words‬‬

‫‪79‬‬ ‫‪221–22‬‬ ‫‪76 n. 41, 118‬‬ ‫‪78 n. 45, 214‬‬ ‫‪53–54, 77, 105‬‬ ‫‪105‬‬ ‫‪77‬‬ ‫‪76-77, 100‬‬ ‫‪100‬‬ ‫‪48, 53, 78‬‬

‫)‪ (cont.‬ܗ‬ ‫ ܗܦܘܟܝ‬ ‫ ܗܦܛܝܐ‬ ‫ ܗܦܠܘܣ‬ ‫ ܗܦܪܟܝܘܣ‬ ‫ ܗܪܛܝܩܐ‬ ‫ ܗ̈ܪܛܝܩܘ‬ ‫ ܗܪܣܝܘܛـܐ‬ ‫ ܗܪܣܝܣ‬ ‫ ܗ̈ܪܣܝܣ‬ ‫ ܗܬܘܣ‬

‫‪106‬‬ ‫‪106‬‬ ‫‪106‬‬

‫ܘ‬ ‫ ܘܐ�ܠܐ‬ ‫ ܘ�ܠܐ‬ ‫̈‬ ‫ ܘܠܣ‬

‫‪120–21, 215‬‬ ‫‪120–21‬‬ ‫‪121‬‬ ‫‪101‬‬ ‫‪72, 220‬‬ ‫‪72‬‬

‫ܙ‬ ‫ ܙܘܓ‬ ‫ ܙܘܓܐ‬ ‫ ܙܘܘܓܐ‬ ‫ ܙܘܢܪܐ‬ ‫ ܙܡܠܝܐ‬ ‫ ܙܡܪܓܕܐ‬

‫‪68-69 n. 18‬‬

‫ܚ‬ ‫̄‬ ‫ ܚܕܬܐ‬

‫‪92‬‬ ‫‪119‬‬ ‫‪134‬‬ ‫‪134‬‬ ‫‪104‬‬ ‫‪112, 122‬‬ ‫‪92, 112, 122‬‬ ‫‪122‬‬ ‫‪122‬‬ ‫‪122‬‬ ‫‪136‬‬ ‫‪221‬‬ ‫‪102‬‬ ‫‪92‬‬ ‫‪113‬‬ ‫‪94, 117‬‬ ‫‪122‬‬

‫ܛ‬ ‫ ܛـܐܓܢܐ‬ ‫ ܛـܐܘܣ‬ ‫ ܛܒܘܠܪܐ‬ ‫ ܛܒܠܪܐ‬ ‫ ܛܓ�ܡܐ‬ ‫ ܛܓܢ‬ ‫ ܛܓܢܐ‬ ‫ ܛܘܓܢܐ‬ ‫ ܛܘܟܢܐ‬ ‫ ܛܘܟܣܐ‬ ‫ ܛܘܡܣܐ‬ ‫ ܛܘܣܐ‬ ‫ ܛܘܦܣܐ‬ ‫ ܛܝܓܢܐ‬ ‫ ܛܝܛܠܘܣܐ‬ ‫ ܛܟ‬ ‫ ܛܟܢ‬

‫‪100, 106, 108‬‬ ‫‪100, 106, 108‬‬ ‫‪100, 105–6, 108‬‬ ‫‪100, 106, 108‬‬ ‫‪33, 109‬‬ ‫‪71‬‬ ‫‪71‬‬ ‫‪53‬‬ ‫‪92-93 (with tables‬‬ ‫)‪5.8, 5.9‬‬ ‫‪214‬‬ ‫‪46 n. 11‬‬ ‫‪48, 109‬‬ ‫‪119, 174–94,‬‬ ‫‪197–98, 201, 204,‬‬ ‫‪210‬‬ ‫‪214‬‬ ‫‪92–93 (with table‬‬ ‫‪5.8), 99, 111‬‬ ‫‪99‬‬ ‫‪99‬‬ ‫‪92-93 (with table‬‬ ‫)‪5.8‬‬ ‫‪99 n. 10‬‬ ‫‪99‬‬ ‫‪133–34‬‬ ‫‪103, 134‬‬ ‫‪77-78, 215‬‬ ‫‪79, 102‬‬ ‫‪78‬‬ ‫‪114, 135, 163 n. 30,‬‬ ‫‪171 nn. 42-43‬‬ ‫‪103‬‬ ‫‪76 n. 41‬‬ ‫‪77‬‬ ‫‪60‬‬ ‫‪79‬‬ ‫‪80 n. 52‬‬ ‫‪184–85, 189 n. 37‬‬ ‫‪225‬‬ ‫‪75‬‬ ‫‪75‬‬ ‫‪76 n. 41‬‬

‫)‪ (cont.‬ܕ‬ ‫̈‬ ‫ ܕܘܓܡܐ‬ ‫̈‬ ‫ ܕܘܓܡܐܛـܐ‬ ‫̈‬ ‫ ܕܘܓܡܘ‬ ‫̈‬ ‫ ܕܘܓܡܛܐ‬ ‫ ܕܘܟܣ‬ ‫ ܕܘܡܣܛܝܩܐ‬ ‫ ܕܘܡܣܬܝܩܐ‬ ‫ ܕܘܩܛܘܢ‬ ‫ ܕܝܐܬܝܩܝ‬ ‫ ܕܝܛܓ�ܡܐ‬ ‫ ܕܝܛܣܪܘܢ‬ ‫ ܕܝܠ‬ ‫ ܕܝܢ‬

‫ ܕܝܢܪܐ‬ ‫ ܕܝܬܩܐ‬ ‫̈‬ ‫ ܕܝܬܩܐ‬ ‫̈‬ ‫ ܕܝܬܩܘܣ‬ ‫ ܕܝܬܩܝ‬ ‫̈‬ ‫ ܕܝܬܩܝܢ‬ ‫̈‬ ‫ ܕܝܬܩܣ‬ ‫ ܕܪܘܡܢܪܐ‬ ‫ ܕܪܡܘܢ‬ ‫ܗ‬ ‫ ܗܓܡܘܢܐ‬ ‫ ܗܕܝܘܛـܐ‬ ‫̈‬ ‫ ܗܖܘ�ܡܐ‬ ‫ ܗܘܝ‬ ‫ ܗܘ�ܠܐ‬ ‫ ܗܘܦܛܘܣ‬ ‫ ܗܘܦܪܝܛܐ‬ ‫ ܗܘܦܪܟܐ‬ ‫ ܗܘܩܘܢ�ܡܐ‬ ‫ ܗܛܪܝܢ‬ ‫ ܗܝܕܝܢ‬ ‫ ܗܟܢܐ‬ ‫ ܗܡܝܪܐ‬ ‫ ܗܢܝܘܟܐ‬ ‫ ܗܦܘܕܝܩܢܐ‬

‫‪276‬‬

‫‪Index of Syriac Words‬‬ ‫‪181‬‬ ‫‪45, 216‬‬ ‫‪125‬‬ ‫‪134‬‬ ‫‪133–34‬‬ ‫‪134‬‬ ‫‪216‬‬

‫ ܠܡ‬ ‫ ܠ�ܡܐܢܐ‬ ‫ ܠܣܛܝܐ‬ ‫ ܠܩܛܝܩܝܢ‬ ‫ ܠܩܛܝܩܪܐ‬ ‫ ܠܩܛܩܝܢ‬ ‫ ܠܩܢܐ‬

‫‪105‬‬ ‫‪105‬‬ ‫‪105‬‬ ‫‪105‬‬ ‫‪87-88, 118‬‬ ‫‪216–17‬‬ ‫‪217‬‬ ‫‪217‬‬ ‫‪120‬‬ ‫‪90 n. 91, 92 n. 95‬‬ ‫‪126‬‬ ‫‪90 n. 91, 92 n. 95‬‬ ‫‪94‬‬ ‫‪105‬‬ ‫‪217‬‬ ‫‪217‬‬ ‫‪120‬‬ ‫ ‪87-88, 118‬‬ ‫‪122‬‬ ‫‪122‬‬ ‫‪119‬‬ ‫‪124‬‬ ‫‪124‬‬ ‫‪87-88, 118‬‬ ‫‪52‬‬ ‫‪117, 119, 187–89‬‬ ‫‪183–84 n. 25‬‬ ‫‪106‬‬ ‫‪86, 106‬‬ ‫‪86 n. 77‬‬ ‫‪103‬‬

‫ܡ‬ ‫ �ܡܐܟܢܐ‬ ‫̈‬ ‫ �ܡܐܟܢܘܣ‬ ‫̈‬ ‫ �ܡܐܟܢܘܬܐ‬ ‫̈‬ ‫ �ܡܐܟܢܣ‬ ‫ �ܡܐܠܠܘܢ‬ ‫ ܡܓܣܐ‬ ‫ ܡܘܕܝܐ‬ ‫ ܡܘܟ�ܠܐ‬ ‫ ܡܙܕܘܓܐ‬ ‫ ܡܛܘܠ‬ ‫ ܡܛܟܣܢܐ‬ ‫ ܡܛܠ‬ ‫ ܡܛܠܘܢ‬ ‫ ܡܝܟܢܐ‬ ‫ ܡܝ�ܠܐ‬ ‫ ܡܝܠܬܐ‬ ‫ ܡܝܩܢܢܐ‬ ‫ ܡܠܘܢ‬ ‫ ܡܠܛ‬ ‫ ܡܠܛܐ‬ ‫ ܡܠܝܣܛܐ‬ ‫ ܡܠܟܐ‬ ‫ ܡܠܟܘܬܐ‬ ‫ ܡܠܠܘܢ‬ ‫ ‪ /mɛn/‬ܡܢ‬ ‫ )‪ܿ /man/ (< μέν‬ܡܢ‬ ‫ ‪ܿ /mɔn/‬ܡܢ‬ ‫ ܡܢܓܢܘܢ‬ ‫̈‬ ‫ ܡܢܓܢܘܢ‬ ‫ ܡܣܝܘܢܐ‬ ‫ ܡܪܓܢܝܬܐ‬

‫‪121‬‬ ‫‪121‬‬ ‫‪133‬‬ ‫‪217‬‬

‫ܢܢ‬ ‫ ܢܘܓ‬ ‫ ܢܘܘܓܐ‬ ‫ ܢܘܛܪܐ‬ ‫ ܢܘܣܐ‬

‫)‪ (cont.‬ܛ‬ ‫ ܛܟܢܐ‬ ‫ ܛܟܣ‬ ‫ ܛܟܣܐ‬

‫‪122‬‬ ‫‪122, 126‬‬ ‫‪100-103, 122, 126,‬‬ ‫‪221‬‬ ‫‪100-101‬‬ ‫‪60‬‬ ‫‪100-103, 122,‬‬ ‫‪126‬‬ ‫‪100-101‬‬ ‫‪221‬‬ ‫‪87, 124‬‬ ‫‪124‬‬ ‫‪84‬‬ ‫‪84, 110–11‬‬ ‫‪126‬‬

‫̈‬ ‫ ܛܟܣܝܣ‬ ‫ ܛܠ̈ܪܐ‬ ‫ ܛܪܘܢܐ‬ ‫ ܛܪܘܢܘܬܐ‬ ‫ )‪ (< τροπή‬ܛܪܘܦܐ‬ ‫ )‪ (< τροφή‬ܛܪܘܦܐ‬ ‫ ܛܪܟܢܐ‬

‫‪120‬‬

‫ܝ‬ ‫ ܝܩܢ‬

‫‪90 n. 91, 92 n. 95,‬‬ ‫‪98‬‬ ‫‪104‬‬ ‫‪181‬‬ ‫‪222‬‬ ‫‪45, 105‬‬ ‫‪105‬‬ ‫‪142‬‬ ‫‪114‬‬ ‫‪90 n. 91, 92 n. 95,‬‬ ‫‪98‬‬ ‫‪222‬‬ ‫‪104‬‬ ‫‪104‬‬ ‫‪203, 222‬‬ ‫‪134‬‬ ‫‪74, 104 n. 23‬‬ ‫‪216‬‬ ‫‪86, 105‬‬ ‫‪105‬‬ ‫‪86 n. 76‬‬ ‫–‪144–47, 197‬‬ ‫‪98‬‬

‫̈‬ ‫ ܛܟܣܐ‬ ‫ ܛܟܣܝܘܛـܐ‬ ‫ ܛܟܣܝܣ‬

‫ܟ‬ ‫ ܟܘܠ‬ ‫ ܟܘܪܐ‬ ‫ ܟܝ‬ ‫ ܟܝܠܝܪܟܐ‬ ‫ ܟܝܡܘܢܐ‬ ‫̈‬ ‫ ܟܝܡܘܢܐ‬ ‫ ܟܝܢܝܐ‬ ‫ ܟܝܪܘܛܘܢܝܬܝܢܐ‬ ‫ ܟܠ‬ ‫ ܟܠܝܪܟܐ‬ ‫ ܟܣܢܕܟܝܢ‬ ‫ ܟܣܢܘܕܘܟܝܢ‬ ‫ ܟܪܘ�ܡܐ‬ ‫ ܟܪܛܘܠܪܐ‬ ‫ ܟܪܛܝܣܐ‬ ‫ܠ‬ ‫ ܠܓܝܘܢܐ‬ ‫ ܠܘܟܝܬܐ‬ ‫̈‬ ‫ ܠܘܟܝܬܐ‬ ‫̈‬ ‫ ܠܘܢܟܕܝܐ‬ ‫ ܠܘܬ‬

‫‪277‬‬

‫‪119‬‬ ‫‪87‬‬ ‫‪100‬‬ ‫‪33 n. 48, 103, 219‬‬ ‫‪102, 110‬‬ ‫‪110‬‬ ‫‪121‬‬ ‫‪122‬‬ ‫‪121–22‬‬ ‫‪82‬‬ ‫‪87‬‬ ‫‪82‬‬ ‫‪95‬‬ ‫)‪93 (table 5.9‬‬ ‫‪52-53‬‬ ‫‪100‬‬ ‫‪57–59, 218–19‬‬ ‫‪83, 122, 218‬‬ ‫‪122‬‬ ‫‪71‬‬ ‫‪122‬‬ ‫‪100‬‬ ‫‪100‬‬ ‫‪219‬‬ ‫‪48-49, 54, 63‬‬ ‫)‪93 (table 5.9‬‬ ‫‪113, 115 n. 44‬‬ ‫‪119‬‬ ‫‪68‬‬ ‫‪68‬‬ ‫‪222‬‬ ‫‪222‬‬ ‫‪71‬‬ ‫‪121‬‬ ‫‪219‬‬ ‫‪94 n. 104‬‬ ‫‪82 (table 5.4), 121‬‬ ‫‪71‬‬ ‫‪91 n. 94‬‬ ‫‪72‬‬ ‫‪125‬‬ ‫‪33, 109‬‬ ‫‪133‬‬ ‫‪122–23, 219‬‬ ‫‪123‬‬

‫‪Index of Syriac Words‬‬ ‫)‪ (cont.‬ܦ‬ ‫ ܦܘܛـܐ‬ ‫ ܦܘܛܩܐ‬ ‫̈‬ ‫ ܦܘܠܘܦܕܣ‬ ‫ ܦܘܠܝܛܝܐ‬ ‫ ܦܘܣܝܣ‬ ‫ ܦܘܣܝܣܝܐ‬ ‫ ܦܘܪܓ�ܠܐ‬ ‫ ܦܘܪܢܣܐ‬ ‫ ܦܘܪܣܝܐ‬ ‫ ܦܘܪܩܣܐ‬ ‫ ܦܘܬܩܐ‬ ‫ ܦܛܓܪܐ‬ ‫ ܦܛܪܝܪܟܐ‬ ‫ ܦܝܠܘܣܘܦܝܐ‬ ‫ ܦܝܠܝܣ‬ ‫ ܦܝܠܝܦܘܣ‬ ‫ ܦܝܢܟܐ‬ ‫ ܦܝܣ‬ ‫ ܦܝܣܐ‬ ‫ ܦܝܣܛܝܩܣ‬ ‫ ܦܝܣܬܐ‬ ‫ ܦܠܓ�ܡܐ‬ ‫̈‬ ‫ ܦܠܓܡܛܐ‬ ‫ ܦܠܛܝܐ‬ ‫ ܦܠܛܝܢ‬ ‫ ܦܠܝܪܘܦܘܪܝܐ‬ ‫ ܦܠܝܪܘܦܘܪܝܣܐ‬ ‫ ܦܢܛܘܣ‬ ‫ ܦܢܩܝܕܬܐ‬ ‫ ܦܢܩܝܬܐ‬ ‫ ܦܣܦܣܐ‬ ‫ ܦܣܩܝܬܐ‬ ‫ ܦܣܬܝܩܐ‬ ‫ ܦܪܓܠ‬ ‫ ܦܪܓܡܛܘܛـܐ‬ ‫ ܦܪܗܠܝ‬ ‫ ܦܪܗܣܝܐ‬ ‫ ܦܪܘܣܬܕܐ‬ ‫ ܦܪܘܨܘܦܐ‬ ‫ ܦܪܘܬܣܡܝܐ‬ ‫ ܦܪܛܘܪܝܐ‬ ‫ ܦܪܛܘܪܝܢ‬ ‫ ܦܪܡܘܢܪܐ‬ ‫ ܦܪܢܣ‬ ‫ ܦܪܢܣܐ‬

‫–‪102, 108–9, 123‬‬ ‫‪25, 129, 217–18‬‬ ‫‪124, 129‬‬ ‫‪125‬‬

‫)‪ (cont.‬ܢܢ‬ ‫ ܢܡܘܣܐ‬ ‫ ܢܡܘܣܐܝܬ‬ ‫ ܢܡܘܣܝܐ‬ ‫ܣ‬ ‫ ܣܓܘܛܘܣ‬ ‫ ܣܕܘܢܐ‬ ‫ ܣܕ�ܠܐ‬ ‫ ܣܘܡܛܐ‬ ‫ ܣܘܢܘܕܘܣ‬ ‫ ܣܘܢܛܠܝܐ‬ ‫ ܣܘܢܦܢܘܣ‬ ‫ ܣܘܢܩܠܝܛܘܣ‬ ‫̈‬ ‫ ܣܘܢܩܠܝܛܘܣ‬ ‫ ܣܘܢܩܪܛܝܣܐ‬ ‫ ܣܛܪܛܓܐ‬

‫‪73 n. 28‬‬ ‫‪87‬‬ ‫‪87, 219–20‬‬ ‫‪221‬‬ ‫)‪93 (table 5.9‬‬ ‫‪94‬‬ ‫‪61, 70-71‬‬ ‫‪106‬‬ ‫‪106‬‬ ‫‪113‬‬ ‫‪124 n. 74, 203,‬‬ ‫‪220‬‬ ‫‪124 n. 74‬‬ ‫‪54‬‬ ‫‪104‬‬ ‫‪99‬‬ ‫‪99‬‬ ‫‪133‬‬ ‫‪220‬‬ ‫‪72‬‬ ‫‪87‬‬ ‫‪86‬‬ ‫‪84, 94‬‬ ‫‪84‬‬ ‫‪94‬‬ ‫‪33-34‬‬ ‫‪48-49, 133‬‬

‫ ܣܛܪܛܓܘܬܐ‬ ‫ ܣܛܪܦܐ‬ ‫ ܣܝܪܐ‬ ‫ ܣܝ̈ܪܝܢܘܣ‬ ‫ ܣܝ̈ܪܝܢܣ‬ ‫ ܣܠܢܛܝܪܐ‬ ‫ ܣܡܝܘܢ‬ ‫ ܣܡܪܓܕܐ‬ ‫ ܣܢܕ�ܠܐ‬ ‫ ܣܦܘܓܐ‬ ‫ )‪ (< σπεῖρα‬ܣܦܝܪ‬ ‫ )‪ (< σφαῖρα‬ܣܦܝܪ‬ ‫ )‪ (< σπεῖρα‬ܣܦܝܪܐ‬ ‫ ܣܦܪܐ‬ ‫ ܣܩܠܪܐ‬

‫‪113, 116, 135‬‬

‫ܥ‬ ‫ ܥܒܕ‬

‫‪100‬‬ ‫)‪93 (table 5.9‬‬ ‫)‪82 (table 5.4‬‬ ‫‪82 (table 5.4),‬‬ ‫)‪93 (table 5.9‬‬ ‫‪33‬‬ ‫‪82‬‬

‫ܦ‬ ‫ ܦܐܠܘܦܣ‬ ‫ ܦܐܢܛܐܣܝܐ‬ ‫ ܦܐܪܣܝܐ‬ ‫ ܦܐܪܪܝܣܝܐ‬ ‫ ܦܕܓܘܓܐ‬ ‫ ܦܘܕܓܪܐ‬

‫‪Index of Syriac Words‬‬

‫‪99-100, 130–31‬‬ ‫‪121‬‬ ‫‪74, 222‬‬ ‫‪98‬‬ ‫‪62, 72‬‬ ‫‪118‬‬ ‫‪215‬‬ ‫‪215‬‬ ‫‪86‬‬ ‫‪133‬‬ ‫‪87‬‬ ‫‪218‬‬ ‫‪133‬‬ ‫‪95‬‬ ‫‪215‬‬ ‫‪121‬‬ ‫‪74, 104 n. 23‬‬ ‫‪131‬‬ ‫‪74‬‬ ‫‪103‬‬ ‫‪54‬‬

‫)‪ (cont.‬ܩ‬ ‫ ܩܠܝ̈ܪܝܩܘܣ‬ ‫ ܩܠܣ‬ ‫ ܩܠܩܘ�ܡܐ‬ ‫ ܩܠܪܛܝܣ‬ ‫ ܩܡܛܪܝܢ‬ ‫ ܩܢ‬ ‫ ܩܢܛܪܘܢܐ‬ ‫ ܩܢܛܪܝܘܢ‬ ‫ ܩܢܟܐ‬ ‫ ܩܢܩܠܪܐ‬ ‫ ܩܣܘܣ‬ ‫ ܩܣܛܐ‬ ‫ ܩܣܛܘܢܪܐ‬ ‫ ܩܣܛܘܪ‬ ‫ ܩܣܪ‬ ‫ ܩܦܣ‬ ‫ ܩܪܛܝܣܐ‬ ‫ ܩܪܝܬܐ‬ ‫ ܩܪܟܕܢܐ‬ ‫ ܩܪܩܘܪܐ‬ ‫ ܩܪܩܠ‬

‫‪80-81‬‬ ‫‪80‬‬ ‫‪80‬‬ ‫‪80 n. 52‬‬ ‫‪80‬‬ ‫‪80‬‬ ‫‪80 n. 53‬‬

‫ܪ‬ ‫ ܪܗܒܘܢܐ‬ ‫ ܪܗܛܢܐ‬ ‫ ܪܗܛܪܐ‬ ‫ *ܪܗܛܪܝܢ‬ ‫ ܪܗܝܛܪܐ‬ ‫ ܪܘܓܐ‬ ‫ ܪܛܝܢܐ‬

‫)‪93 (table 5.9‬‬ ‫‪103‬‬ ‫‪104, 221‬‬ ‫‪35 n. 57‬‬ ‫‪141‬‬ ‫‪98, 103‬‬ ‫‪101‬‬

‫ܬ‬ ‫ ܬܐܘܠܘܓܝܐ‬ ‫ ܬܐܛܪܘܢ‬ ‫ ܬܓ�ܡܐ‬ ‫ ܬܘܪܓ�ܡܐ‬ ‫ ܬܠܝܬܝܐ‬ ‫ ܬܪܘܢܘܣ‬ ‫ ܬܪܘܢܝܘܢ‬

‫‪278‬‬

‫‪121‬‬ ‫‪113‬‬ ‫‪219‬‬ ‫‪82-83, 91, 103‬‬ ‫)‪82 (table 5.4‬‬ ‫)‪82 (table 5.4‬‬ ‫‪72‬‬

‫)‪ (cont.‬ܦ‬ ‫ ܦܪܣܝ‬ ‫ ܦܪܣܠܬܝܢܐ‬ ‫ ܦܪܦܐ‬ ‫ ܦܪܨܘܦܐ‬ ‫ ܦܪܪܝܣܝܐ‬ ‫ ܦܪܪܣܝܐ‬ ‫ ܦܪܬܙܡܝܐ‬

‫‪143–44‬‬ ‫‪56, 82, 221‬‬

‫ܨ‬ ‫ ܨܒܝܢܝܐ‬ ‫ ܨܦܘܢܝܐ‬

‫)‪93 (table 5.9‬‬ ‫‪45‬‬ ‫‪133‬‬ ‫‪62, 73‬‬ ‫‪121‬‬ ‫‪216‬‬ ‫‪121‬‬ ‫‪216‬‬ ‫‪113‬‬ ‫‪73, 121‬‬ ‫‪121‬‬ ‫‪131‬‬ ‫‪131‬‬ ‫‪131‬‬ ‫‪51‬‬ ‫‪57 n. 55, 126‬‬ ‫‪57-58, 121‬‬ ‫‪57 n. 55, 126‬‬ ‫‪81‬‬ ‫‪49-50‬‬ ‫‪203, 216‬‬ ‫‪56, 215–16‬‬ ‫‪98‬‬ ‫‪99-100, 130–31‬‬ ‫‪99-100‬‬ ‫‪99-100‬‬ ‫‪99-100‬‬

‫ܩ‬ ‫ ܩܐܛـܐܣܛܐܣܝܣ‬ ‫ ܩܐܪܣܐ‬ ‫ ܩܘܒܩܠܪܐ‬ ‫ ܩܘܒܪܢܝܛܐ‬ ‫ ܩܘܛܪܓܐ‬ ‫ ܩܘܠܘܢܝܐ‬ ‫ ܩܘܠܣܐ‬ ‫ ܩܘܠܪܐ‬ ‫ ܩܘܣܝܣܐ‬ ‫ )‪ (< κύβος‬ܩܘܦܣܐ‬ ‫ ‪ /quppɔsɔ/‬ܩܘܦܣܐ‬ ‫ ܩܘ̈ܪܝܐ‬ ‫ ܩܘ̈ܪܝܐܣ‬ ‫ ܩܘ̈ܪܝܘܣ‬ ‫ ܩܘܪܝܠܝܣܘܢ‬ ‫ ܩܛܓܪܢܐ‬ ‫ ܩܛܪܓ‬ ‫ ܩܛܪܓܢܐ‬ ‫ ܩܛܪܩܛܐ‬ ‫ ܩܝܛܘܣ‬ ‫ ܩܝܢܕܘܢܘܣ‬ ‫ ܩܝܬܪܐ‬ ‫ ܩܠܝܕܐ‬ ‫ ܩܠܝܪܝܩܐ‬ ‫ ܩܠܝ̈ܪܝܩܐ‬ ‫ ܩܠܝ̈ܪܝܩܘ‬ ‫ ܩܠܝܪܝܩܘܣ‬

Index of Greek Words α

ἀγκών 86 ἀγών  45, 65, 105, 116, 192 n. 45 ἀήρ  45, 104, 106, 212 ἀθλῆσαι  116 n. 53 ἄθλησις 116 ἀθλητής 128 αἵρεσις  76–77, 100 αἱρεσιώτης 77 αἱρετικός  54, 77, 105 ἀκαταστασία 53 ἀκριβῶς  53 n. 33, 118 ἁμηγέπη  37 n. 67 ἀνάγκη  86, 106–8, 130 ἀνδριάς  87, 98–99, 212 ἀντισύγγραφα  86 n. 75 ἀντισύνγραφα  86 n. 75 ἁπλῶς  76 n. 41, 118 ἀπό 52 ἀποκρισιάριος 133 ἀπουσία 103 ἄρα 118–19 ἆρα 119 -αριος 132–35 ἀῤῥαβών 80–81 ἀρχαί 212 ἀρχή 110 ἄρχων  124 n. 74, 212–13 ἄρωμα 78 ἀσπίς 83 ἀσσάριον 213 αὐτοκράτωρ 213 ἀφείσθασθαι 71 ἀφίστασθαι 71

β

βάρβαρος 125 βασιλική 59 βάσις 213 βερεδάριος 133 βερηδάριος 133 βεστιάριος 133 βῆλον 106 βῆμα  94, 102, 213 βίῤῥος 81 βουλευτής 213 βυρσεύς 125 βωμός 213 γ γαλ(λ)ιάριος 133 γάρ  119, 191–94, 201, 210 γένεσις 32 γέννησις 32 γένος  102, 203, 213– 14 γερδιός 125 γέρδιος 125 γοῦν 118 γυβερνήτης  62, 68, 73 γυμνάσιον 63 δ δέ  119, 174–94, 197–98, 201, 204, 210 δελματ̣ίκιν  63 n. 70 δελματίκιον  63 n. 70 δηνάριον 214 διὰ τεσσάρων  46 n. 11 διαθήκη  92 (table 5.8), 93 (table 5.9), 99, 111

279

δ (cont.) διαπεμψαμένου  61 n. 65, 70 διαπενψαμένου  61 n. 65, 70 διάταγμα 214 διεγδικήσειν  61 n. 64, 70 διεκδικήσειν  61 n. 64, 70 δόγμα  59–60, 100, 105–6, 108 δοκιμάζειν 112 δομεστικός 71 δουκᾶτον 53 δούξ  33, 109 δρόμων  103, 134 δρομωνάριος 133–34 ε ἐγ  61 n. 64, 70 ἐγβένω  61 n. 64, 70 ἐγδικίας  61 n. 64, 70 ἐγκαλεῖν  86 n. 75 ἐγκαλέσειν  86 n. 75 ἐγκαλέσῃ  86 n. 75 ἐγκαλοῦμαι  86 n. 75 ἐγκλήματα  86 n. 75 ἔθος  48, 53, 78 εἰκῇ  78, 117 εἰκών 120 εἶτα  95, 119 ἐκ  61 n. 64, 70 ἐκβαίνω  61 n. 64, 70 ἐκδικίας  61 n. 64, 70 ἐκκλησιέγδικος  61, 70 ἐκκλησιέκδικος  61, 70

280 ε (cont.) ἑκούσιον 143–44 ἐμποιηθῇ  61 n. 65, 70 ἐμποιούμενον  61 n. 65, 70 ἔμπροσθεν  61 n. 65, 70 ἐμφράξει  61 n. 65, 71 ἐνκαλ̣έσῃ  86 n. 75 ἐν̣κ̣αλῖν  86 n. 75 ἐνκαλλέσειν  86 n. 75 ἐνκαλοῦμε  86 n. 75 ἐνκλήματα  86 n. 75 ἐνποιηθῇ  61 n. 65, 70 ἐνποιούμενον  61 n. 65, 70 ἔνπροσθεν  61 n. 65, 70 ἐ̣νφράξι  61 n. 65, 71 ἕνωσις 75 ἐξέδρα 214 ἐξορία 78 ἐξορισθῆναι 114 ἑορταστικός 75 ἐπαρχία  78 n. 45, 214 ἔπαρχος 60 ἐπιθέτης 94 ἐπιμελητής 128 ἐπίτροπος  203, 214–15 ἐποχή 79 ἐστίν  153–73, 197–98, 201, 204, 206, 210 εὖ  118 n. 60 εὐαγγέλια  93 (table 5.9), 101 εὐαγγέλιον  101, 110 see also εὐαγγέλια εὐαγγελιστής  93 (table 5.9) ζ ζεῦγος  120–21, 215 ζυγόν  120–21, 215 ζωνάριον 101 ζώνη 101 η ἡγεμών  77–78, 215

Index of Greek Words η (cont.) ἡνιόχος 75 θ θέατρον 103 θεολογία  93 (table 5.9) θρονίον 101 θρόνος  98, 101, 103 ι -ιδιον  68–69 n. 18 ἰδιώτης  79, 102 ἱερατεῖον 75 -ιν 62–63 -ιον 62–63 ἱππικός 75 ἱππόδρομος 76 ἴσον 48–49 κ καγκελλάριος 133 καιρός 45 καῖσαρ 215 καλλίας 73 καλῶς 121 κάμπτριον  62, 72 κάμτριον  62, 72 κἄν 118 κάππα δόκια  208 καρχηδόνιος 74 καταῤῥάκτης 81 κατάστασις  93 (table 5.9) κατεσθάθην 71 κατεστάθην 71 κατηγορεῖν  57–58, 121 κατήγορος  57–58, 121, 126 κελλαρίτης 98 κεντυρίων 215 κέρκουρος 103 κῆτος 49–50 κιθάρα  56, 215–16 κίθαρις  56, 215–16 κίνδυνος  203, 216 κλείς 98 κλῇθρον 54

κ (cont.) κληρικός 99–100, 130–31 κόγχη 86 κολλάριον 216 κολωνία 216 κοσσίσαι 113 κόσσος  87, 113 κουβικουλάριος 133 κουεστιονάριος 133 n. 97 κυαιστιωνάριος 133 κυαίστωρ 95 κυβερνήτης  62, 68, 73 κύβος  73, 121 κύριε ἐλέησον  51 λ λεγεών 216 λεγιών 216 λεκάνη 216 λεκτικάριος 133–34 λεκτίκιον 134 λῃστής 125 λιμήν  45, 216 λόγχη  86, 105 λογχίδιον  86 n. 76 μ μάγγανον  86, 106 μαγίς 216–17 μάλιστα 119 μᾶλλον  87–88, 118 μαργαρίτης 103 μελετᾶν 122  μελέτη 122 μέν  117, 119, 187–90 μέταλλον 94 μηλωτή 217 μηχανή 105 μίλιον 217 μόδιος 217 μοχλός 217 ν ναός 217 ναυαγός 121

ν (cont.) νόμος  102, 108–9, 123–25, 129, 217–18 νοτάριος 133 ξ ξενοδοχεῖον 103–4 ξένος  125–26, 218 ξέστης 218 ο ὄγκινος  86, 104 οἰκονόμος 79 ὁλοσηρικόν 76 ὅμηρος 75 ὁμολογία  59, 76 ὄργανον 91 ὀρθόδοξος 99 ὁρίζων 76 ὅρος 75 ὄρος 78 οὐερεδάριος 133 οὖν 118 οὐσία 99–100 π πάγκαλα  86 n. 75 πάγκαλον  86 n. 75 παιδαγωγός 33 παλάτιον  48–49, 54, 63 πανδοκεῖον 87 πάνκαλα  86 n. 75 πάνκαλον  86 n. 75 πάντως 119 παραγγέλλειν 121 παράλια  94 n. 104 παραμονάριος 133 παῤῥησία  81–82, 87 n. 82, 93 (table 5.9), 121–22 πατριάρχης 95 πεῖσαι  83, 122, 218 πινακίδιον 68 πίναξ  56–59, 218–19 πιστικός 71 πλατεῖα 219

Index of Greek Words

281

π (cont.) πληροφορῆσαι  113, 115 n. 44 πληροφορία  93 (table 5.9) πνεῦμα 83 ποδάγρα 82 πόλις 83 πολιτεία  33 n. 48, 103, 219 πολύπους 100 πόρπη 219 πότε 119 πραγματευτής 219 πραίτωρ 125 πραιτώριον  33, 109 πρόεδρος 128 προθεσμία 72 προνοῆσαι  122–23, 219 προοίμιον  35 n. 59 πρός  144–47, 197–98 προσελθεῖν 113 προσελθῆναι 113 προστάς 71 πρόσωπον  82–83, 91, 103 πύργος  54–55 n. 44, 82

σ (cont.) σμάραγδος 72 σμίλη  72, 220 σμιλίον  72 n. 24 σπαθάριος 134 σπεῖρα  84, 94 σπόγγος 86 στάσις 132 στατήρ  56, 84–85 n. 70, 202, 220 στοά  103, 220 στρατηγός  124 n. 74, 128, 203, 220 στρογγυλοπρόσωπον 86 n. 75 στρονγυλοπρόσωπον 86 n. 75 συγγραφήν  86 n. 75 σύγκλητος 106 συγκροτῆσαι 113 συγκωμῆται   86 n. 75 συγκωμήτης  86 n. 75 συμβάν  61 n. 65, 71 συμπαρόντος  61 n. 65, 71 σύμπονος  61, 70–71 συμφωνία  56, 82, 221 συνβά[ν]  61 n. 65, 71 συνγραφήν  86 n. 75 συνκωμῆται   86 n. 75 συνκωμήτης  86 n. 75 σύνοδος  93 (table 5.9) συνπαρόντος  61 n. 65, 71 σύνπονος  61, 70–71 συντέλια 94 σφαῖρα 84 σφυρισθῆναι 114 σχῆμα 127 σχολάριος 134 σχολή 134 σῶμα 221 

ρ ῥητίνη 80 ῥητορεία  80 n. 52 ῥήτωρ 80 ῥογα 80 σ σακκελάριος 48–49, 133 σάνδαλον  87, 219–20 σατράπης 54 σειρά 104 σειρήν 99 σεισ̣ ύ̣ρι̣ ν̣ ̣  63 n. 70 σεισύριον  63 n. 70 σημεῖον 220 σιλεντιάριος 133 σινδών 87 σκοῦτα  73 n. 28

τ ταβελλάριος 134 ταβουλάριος 134

282 τ (cont.) τάγμα  104, 221 ταλάριον 221 ταξεώτης 60 τάξις  100–103, 122, 126, 221 ταξιώτης 60 ταραχή 126 τάχα  94, 117 ταῶς 221 τέχνη 122 τέως 119 τήγανον  92, 112, 122 τίτλος  113 n. 41 τιτλῶσαι 113 τόμος 136 τροπή 84 τροφή  84, 110–11 τυγχάνομεν  86 n. 75 τυνχάνομεν  86 n. 75

Index of Greek Words τ (cont.) τύπος 102 τύραννος  87, 124 υ ὕλη 103 ὕπαρχος 60 ὑπατεία 221–22 ὕπατος  76 n. 41 ὑπηρέτης 77 ὑποδιάκονος  76 n. 41 φ φαντασία  93 (table 5.9) φασκία 222 φιλοσοφία  93 (table 5.9) φλέγμα 100 φυλή 52–53 φυσικός 142–43 φύσις  102, 110

χ χαλκηδών 74 χαράκωμα  74, 222 χάρτης  74, 104 n. 23 χαρτου(λ)λάριος 134 χειμών  45, 105 χειροτονηθῆναι 114 χιλίαρχος 222 χρῶμα  203, 222 χώρα 104 ψ ψαλτήριον  56, 84–85 n. 70 ψῆφος 222 ω ὤ  118 n. 60 ὠκεανός  110, 127–28

Index of Subjects Abgar VIII ​26, 28 Abgarids ​26 Abydos Lion Weight ​see Achaemenid under Aramaic accent ​15, 24 n. 41, 36 n. 60, 89, 94– 95 Achaemenid Aramaic ​see under Aramaic Achaemenid Empire ​1, 25 see also Achaemenid under Aramaic Acts of Thomas ​81 (table 5.4), 87 n. 82, 116, 125–27, 140–41 (graph 7.1, table 7.1), 155–56, 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 185–86 (with table 9.1), 205, 213, 215–18, 221, 223, 225 adjectival genitive ​142–44 Adme ​26 see also Edessa Against the Manichaeans ​ see Titus of Bostra Akkadian ​1–2, 67 n. 12, 84–85, 103, 132, 135 Alexander the Great ​25, 40, 202 Alexandria ​38 Amharic ​1 Amid ​29, 50 analogy ​40, 68 n. 15, 99–102, 105–8, 129–35, 185 n. 32 Anastasia ​209 see also Christian Palestinian Aramaic under Aramaic Antioch ​26, 36, 38 Apamea ​26 aphaeresis ​184 Aphrahaṭ ​31, 55, 199–202, 204–6 Demonstrations ​31, 47 n. 15, 57, 62, 68, 73, 77, 79, 82 (table 5.4),

Aphrahaṭ, Demonstrations (cont.) 92 (table 5.8), 117–18, 125, 127, 140–41 (graph 7.1, table 7.1), 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 185–86 (with table 9.1), 193, 215, 218, 223, 225 Apocalypse of Pseudo-Methodius ​5, 131, 146–47 Appadana ​29 Appendix Probi ​54 Arab conquests ​1–3, 26 Arabic ​1–3, 45, 66, 84–85, 131 n. 90, 176–77, 193–94 Aramaic ​1–7, 26, 29, 38 n. 70, 50 n. 25, 56–60, 73 nn. 26 and 28, 74 n. 31, 75 nn. 36 and 39, 78 nn. 44 and 45, 81 n. 60, 87 n. 81, 94 n. 101, 95 n. 105, 98 n. 5, 99 nn. 7 and 8, 104 n. 25, 109 n. 34, 112 n. 37, 114–15, 117 n. 56, 128, 148, 212–22 Achaemenid ​56–59, 84–85 n. 70, 128, 155, 158, 177–78, 202, 217, 219, 220 see also Aramaic of Ezra under Aramaic Aramaic of Daniel ​56, 82, 84–85 n. 70, 114–15 n. 43, 158–59, 164–65, 169, 179, 216–17, 221 Aramaic of Ezra  ​158 n. 26, 179 Aššur ostracon (KAI 233) ​177 Christian Palestinian Aramaic ​3–4, 57–58, 72, 74, 83–84, 126, 128, 132 n. 94, 163–64, 169–70, 190, 193–94, 206, 209–22 Ḥatran ​56–57, 180 n. 19, 213–14, 217, 220 Jewish Babylonian Aramaic ​3, 57–58, 72 n. 23, 126, 132 n. 95, 164–65, 169–70, 206–7, 209–22

283

284

Index of Subjects

Aramaic (cont.) Jewish Palestinian Aramaic ​3, 57– 60, 72, 74, 126, 155, 157 n. 23, 167–70, 206–22 Judean ​213–15, 220, 222 Late Aramaic ​3–4, 6–8, 57–60, 84, 128 n. 86, 148, 157, 163–70, 189–90, 193, 206–11 see also Christian Palestinian Aramaic, Jewish Babylonian Aramaic, Jewish Palestinian Aramaic, Late Jewish Literary Aramaic, Mandaic, Samaritan under Aramaic Late Jewish Literary Aramaic ​60, 72, 74, 193 n. 48, 212–22 Mandaic ​3, 164–65, 169–70, 206–7, 209–14, 217–18, 220–21 Middle Aramaic ​56–60, 84, 158–60, 163, 180 n. 19, 212–22 see also Aramaic of Daniel, Ḥatran, Judean, Nabataean, Palmyrene, Qumran, Targum Onkelos under Aramaic Nabataean ​3 n. 13, 56–57, 60, 213–15, 217, 220, 222 Neo-Aramaic ​2, 158 Old Aramaic ​157–58 Palmyrene ​3–4 n. 13, 54 n. 43, 56– 57, 59, 76, 99, 126, 128, 212–21 Qumran ​179–80 Samʾalian ​176–77 Samaritan ​3, 57–58, 72, 165–70, 189–90, 206–7, 209–13, 215, 217–22 Standard Literary Aramaic ​58–59 Tg. Onqelos, Tg. Jonathan ​56–57, 72, 74, 212–22 Arbela ​25 Aristotle ​5 Armenian ​72 n. 24, 156 n. 19 aspect ​ see TAM assimilation ​61, 68, 70–73, 81–83, 86– 87, 156 n. 19 Aššur ostracon  ​see under Aramaic Athanasius Life of Antony ​115

Austronesian ​21–23 Babai Commentary on the “Gnostic Chapters” by Evagrius of Pontus ​ 111, 127–28 Babylonian Talmud ​164, 208 n. 43 see also Jewish Babylonian Aramaic under Aramaic Balai Memre on Joseph ​118 n. 60 Bar ʿAli, Ishoʿ Lexicon ​45 Bar Bahlul, Haṣan Lexicon ​45, 53 n. 33, 99 n. 10 Bardaiṣan ​5, 28, 200 see also Book of the Laws of the Countries, Dialogue on Fate Bar ʿEbroyo ​83 Basil of Caesarea ​5 Baṭnan da-Serugh ​44 Bel  ​21–23 Berufsname ​ see -ɔrɔ, CaCCɔC- under morphemes, Syriac Berytus ​38 Beth Garmai ​32 Bibliotheca ​ see Photius Book of Similar Words ​ see Ḥunayn b. Isḥāq Book of Steps ​47 n. 15, 77–79, 92 (table 5.8), 117–18, 125, 214–16, 219 Book of the Laws of the Countries ​28, 47 n. 15, 76–77, 107–8, 127, 140–42 (graph 7.1, table 7.1), 146, 157, 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 181, 185–86 (with table 9.1), 203–6, 217–18, 221, 223–24 see also Bardaiṣan, Dialogue on Fate borrowing (general) ​11–12, 20–24, 43, 47, 147 n. 23, 149, 152 borrowing (recipient language agentivity) 16, 17–19, 24, 30, 36, 39–40, 47, 151, 154, 173, 195–98, 201–2

Index of Subjects borrowing, grammatical ​see grammatical borrowing borrowing scale ​20 broken plural ​131 n. 90 Byzantine Greek ​55, 61–63, 65–66, 70–74, 80, 88–89 calque ​11–12, 46, 149, 152 see also grammatical calque calque, grammatical ​see grammatical calque Canons ​ see Rabbula Carrhae ​ see Ḥarran case (Greek) ​102 Cause of the Foundation of the Schools ​ 5 n. 20 Cause of the Liturgical Feasts ​ see Qiyore of Edessa Cause of the Martyrs ​ see Ishay Christian Palestinian Aramaic  ​see under Aramaic Chronicle ​ see Michael Rabo Chronicle of 1234 ​3 Chronicle of Zuqnin ​131 Clement ​ see Pseudo-Clementine, Recognitions and Homilies Code-Copying Model ​46, 149–50 code-switching ​50–53, 107, 114–15 n. 43, 221 Cohors Vicesima Palmyrenorum ​27 Commentary on Genesis and Part of Exodus ​ see Ephrem ​ Commentary on John ​ see Philoxenos Commentary on Matthew and Luke ​ see Philoxenos Commentary on the Diatessaron ​ see Ephrem Compendious Lexicon ​ see Ḥunayn b. Isḥāq congruence ​151 Constantinople ​34–36, 55–56 convergence ​150–52, 196–97 converted language ​5 Copper Island Aleut ​see Mednyj Aleut Coptic ​14, 79, 112, 115–16 copula ​153–73, 197–99, 201, 204, 206, 210

285

Correctness of Speech, The ​ see Yaʿqub of Edessa covert interference ​140, 151 creole ​21 Cure for Hellenic Maladies ​ see Theodoret of Cyrrhus Cyril of Alexandria ​5 On Orthodox Faith ​34 Second Epistle to Succensus ​143–44 Daniel  ​see under Aramaic Darius III ​25, 40, 202 Dawid Puniqoyo ​3 Dayr al-Suryān ​65 Demonstrations ​ see Aphrahaṭ Denḥa Life of Marutha ​76 n. 41, 82 (table 5.4), 140–41 (graph 7.1, table 7.1), 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 224–25 denominative ​111–12, 117 diacritic point ​68, 83, 90 diaeresis ​ see hiatus Dialogue on Fate ​4–5 see also Bardaiṣan, Book of the Laws of the Countries Didascalia Apostolorum ​76 n. 41, 115 diffusion ​ see indirect diffusion Dionysios of Tel Maḥre ​131 n. 91 Discourses ​ see Philoxenos dissimilation ​70–71, 192 n. 45 Diyarbakır ​ see Amid Double Indictment ​ see Lucian of Samosata drift ​57 dual ​ see number Dura-Europos ​27, 29, 186 see also P. Dura Ecclesiastical History ​ see Eusebius of Caesarea, Sozomen, Yuḥanon of Ephesus Edessa ​1, 26–32, 34, 36, 38 n. 70, 52, 64, 199–200, 202–4, 210 see also Adme Edessa, School of ​see School of the Persians

286

Index of Subjects

Egypt ​14, 38, 55, 57, 61–63, 68, 70–74, 84 n. 70, 87 Egyptian ​14 see also Coptic Eliya Life of Yuḥanon of Tella ​33, 48, 53, 78, 82 (table 5.4), 109, 113, 118, 140–41 (graph 7.1, table 7.1), 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 205, 208–9, 224–25 emphatic ​66–67, 70–71, 73–74, 84–85, 95 see also emphatic p emphatic p ​40, 81–84 enclitic ​154–57, 166–67, 169, 177 English ​16–17, 24 n. 41, 79–80, 129–30, 196, 198 enhancement ​140, 151 Ephrem ​4–5, 31, 55, 65, 115 n. 44, 125, 199–202, 204–8, 215–16 Commentary on Genesis and Part of Exodus ​218, 221–22 Commentary on the Diatessaron ​87, 118 Letter to Publius ​117 Maḏrɔše against Heresies ​92 (table 5.8), 117, 221, 222 Maḏrɔše against Julian the Apostate ​​ 99 n. 10, 117, 207 n. 39, 220 Maḏrɔše on Faith ​78–79, 118, 215 Maḏrɔše on Nisibis ​57, 78, 117–18, 207 n. 39, 218, 220 Maḏrɔše on Paradise ​78 Maḏrɔše on Pascha ​118 Maḏrɔše on the Church ​78, 118, 125, 222 Maḏrɔše on the Fast ​118 Maḏrɔše on the Nativity ​57, 78, 215, 217–18 Maḏrɔše on Virginity ​111 Memrɔ on Faith ​118, 215 Memrɔ on Our Lord ​82 (table 5.4), 117 Prose Refutations ​77–78, 80, 117, 140–41 (graph 7.1, table 7.1), 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 185–86 (with table 9.1),

Ephrem, Prose Ref. (cont.) 188–89, 203, 207, 220–21, 223, 225 Sermons ​212 epigraphic habit ​29 Erbil ​ see Arbela Ethiopic, Classical ​83, 85, 176–77 Eulogios the Stone-Cutter ​209 see also Christian Palestinian Aramaic under Aramaic Euphrates ​27, 29, 44, 50, 52 Eusebius of Caesarea ​5, 38 n. 70 Ecclesiastical History ​38 n. 70, 78 n. 45, 156, 214 History of the Martyrs in Palestine ​ 28 Panegyric on the Christian Martyrs ​ 28 Theophany ​28, 125 n. 78, 127, 219 Eusebius of Emesa ​36 Evagrius of Pontus ​5, 111, 127–28 exclamatory clause ​183–84 n. 25 existential particle ​154–70, 173–74 Explanation of Greek Words with (or: in?) Syriac  see Ḥunayn b. Isḥāq extension ​131–32, 135, 144–47, 154, 158–64, 167–70, 204, 197–98, 204, 210 extraposition ​155, 161 Ezra ​ see under Aramaic feminine ​ see gender Feriale Duranum ​27 Flemish ​104 n. 24 Forty Martyrs of the Sinai Desert ​169, 209 see also Christian Palestinian Aramaic under Aramaic Fremdwort/Fremdwörter ​48–50, 55, 93–96, 106–7, 127–28, 208–9 French ​13–14, 16–17, 79, 104 n. 24, 147, 196 Galen ​5 Gaugamela ​25 Gəʿəz ​ see Ethiopic, Classical gemination ​67, 80–82, 85, 87–88

Index of Subjects gender ​102–4 Genesis Rabbah ​168, 208–10 see also Jewish Palestinian Aramaic under Aramaic gentilic ​ see -ɔyɔ under morphemes, Syriac German ​13–14 Ginza Rba ​165 see also Mandaic under Aramaic global copying ​46, 149–50 grammatical borrowing ​149, 196 grammatical calque ​149 grammatical replication ​2–3, 40, 46, 137–94, 198, 201–2, 204–6, 210 grammaticalization ​147, 151, 176, 184, 191 Greco-Arabic translations ​3 Gregory of Nazianzus ​5 Gregory of Nyssa ​5 Ḥarran ​26 Ḥasan bar Bahlul ​see Bar Bahlul, Ḥasan Ḥatran  ​see under Aramaic Ḥawra ​44 Hebrew ​1–2, 6 n. 23, 50 n. 25, 73 nn. 26 and 28, 74 n. 31, 75 nn. 36 and 39, 78 nn. 44 and 45, 81 n. 60, 87 n. 81, 94 n. 101, 95 n. 105, 98 n. 5, 99 nn. 7 and 8, 104 n. 25, 112 n. 37, 117 n. 56, 145–46, 174–77, 181–83, 185, 189 n. 39, 191–92, 203–4, 209 hiatus ​93, 95 hierarchy of borrowing ​18 Hindi ​150 History of ʿAbdɔ damšiḥɔ ​183–84 n. 25 History of the Martyrs in Palestine ​ see Eusebius of Caesarea History of the Monks of Syria ​ see Theodoret of Cyrrhus Hittite ​50 n. 25, 65 n. 6 Ḥunayn b. Isḥāq ​5 Book of Similar Words ​45 Compendious Lexicon ​45 Explanation of Greek Words with (or: in?) Syriac ​45

287

immediate source ​53–55, 59 imperfect learning ​14, 20–21, 23–24 imposition (source language agentivity) ​ 14, 16–17, 18–19, 22–24, 30, 35, 39–40, 47, 151, 195–96, 198, 201 n. 16 indirect diffusion ​151 indirect transfer ​140, 151 input form ​52 n. 31, 53 n. 32, 70–74, 98–102, 105–6, 111, 116 n. 53, 130–31, 135 interference (general) ​12, 151 see also covert interference, interference through shift interference through shift ​12, 14, 20–23, 24 n. 41 interlingual identification ​143 n. 13 Iranian ​1–2, 29, 48 n. 20, 49 n. 23, 54, 57 n. 54, 116, 132, 199 see also Middle Persian, Pahlavi Ishay Cause of the Martyrs ​119 Isḥaq of Nineveh ​5 Part 3 ​82 (table 5.4), 92 (table 5.8) Isḥaq Shbadnaya ​3 Ishoʿ bar ʿAli ​see Bar ʿAli, Ishoʿ Ishoʿ of Merv ​45 Isocrates ​5 Italian ​24 n. 41 Jacob ​ see Yaʿqub Jewish Babylonian Aramaic  ​see under Aramaic Jewish Palestinian Aramaic  ​see under Aramaic John ​ see Yuḥanon Jonathan, Targum ​see under Aramaic Judean Aramaic  ​see under Aramaic Julian Romance ​77, 80, 82 (table 5.4), 92 (table 5.8), 109, 118 Justin I ​38 Kallinikos ​33 see also Pawlos of Kallinikos Kannada ​150 Koinē Greek ​55, 61–63, 65–66 (with table 5.1), 68, 70–74, 76, 78–79,

288

Index of Subjects

Koinē Greek (cont.) 81, 88–90 (with table 5.5), 95, 101, 118–19 n. 61, 186 n. 35, 187 n. 36 Kupwar ​150 language attrition ​18, 21–23 language maintenance ​20–23 language shift ​20–23, 196 Late Aramaic  ​see under Aramaic Late Jewish Literary Aramaic  ​see under Aramaic Latin ​6 n. 23, 26–27, 29, 33 n. 47, 38 n. 73, 48, 53–56, 59 n. 57, 62–63, 71, 73–74, 81, 86, 95, 104, 106, 109, 125, 129–30, 132–35, 213–19, 221–22 Lehnwort/Lehnwörter ​48–50, 55, 93–96, 106–7, 110, 127–28, 208–9 Letter 13 ​see Yaʿqub of Edessa Letter 18 ​see Yaʿqub of Edessa ​ Letter 47 ​see Timotheos I Letter on Ḥimyarite Persecution ​ see Shemʿun of Beth Arsham Letter on Syriac Orthography ​ see Yaʿqub of Edessa Letter to Maron of Anazarbus ​ see Philoxenos Letter to Publius  ​see Ephrem Letter to the Monks of Beth Gawgal ​ see Philoxenos Letters ​ see Rabbula lexicon-syntactic calque ​149 Life of Antony ​ see Athanasius Life of John bar Aphtonia ​77 Life of Marutha ​ see Denḥa Life of Rabbula ​34, 77, 81–82 (with table 5.4), 92 (table 5.8), 131, 140–41 (graph 7.1, table 7.1), 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 205, 223, 225 Life of Sheʿmon the Stylite ​79 Life of Yuḥanon of Tella ​ see Eliya light verb ​112, 115, 135 light verb strategy ​112–17, 135 linguistic dominance ​15 n. 15, 17–18, 20–24, 35, 36, 39 n. 75

Lives of the Eastern Saints ​ see Yuḥanon of Ephesus loan translation ​151 loanshift ​151 loanverb ​7 n. 30, 111–17, 135 loanword ​2–3, 11, 13–15, 20, 24, 26 n. 6, 30, 33 n. 47, 39–40, 42–136, 139, 150, 187, 190, 192–93, 199– 210, 212–22 see also loanverb Lucian ​5 Lucian of Samosata ​36 n. 60 Luwian ​6 n. 23, 50 n. 25 Maʾa ​19 Maḏɔše . . . ​see Ephrem Mandaic  ​see under Aramaic manuscripts, Syriac Jerusalem, Syr. 180 ​78 London, Brit. Libr. Add. 12,150 ​28 London, Brit. Libr. Add. 12,154 ​143 London, Brit. Libr. Add. 14,506 ​65 London, Brit. Libr. Add. 14,627 ​65 London, Library of the Royal Asiatic Society ​65 n. 4 New Haven, Yale Syriac 5 ​65 n. 4 St. Petersburg, Cod. Syr. 1 ​156 n. 19 Marathi ​150 Maron of Anazarbus ​32 masculine ​ see gender mater lectionis ​35 n. 57, 90–94, 96, 158 n. 25, 177 n. 14, 206 matter replication ​46, 148 Media Lengua ​19 Mednyj Aleut ​19 ​ Memrɔ . . . ​see Balai, Ephrem, Narsai, Yaʿqub of Serugh metatypy ​149, 152 Michael Rabo Chronicle ​131 n. 91 Michif ​19 Middle Aramaic  ​see under Aramaic Middle Persian ​2 see also Pahlavi mixed language ​5, 19, 21 model language ​139 n. 2

Index of Subjects modeling ​151 monophthongization ​177 n. 14, 193 n. 47 mood ​ see TAM morphemes, Syriac -ɔʾiṯ ​110, 121, 124, 126–27 -ɔnɔ ​57 n. 55, 124, 126 -ɔnɔʾiṯ ​126 -ɔnɔyɔ ​126 -ɔnɔyuṯɔ ​126 -ɔrɔ ​40, 132–35 -ɔyɔ ​110, 121, 124–26, 140–44 (with graph 7.1, table 7.1), 197– 98, 205–6, 218 -ɔyuṯɔ ​121, 126 -(ʾ)s ​99–100, 105–8, 130–32, 135 -tɔnɔ ​126–27 -tɔnɔʾiṯ ​126–27 -tɔnɔyuṯɔ ​126 -uṯɔ ​94 n. 103, 121, 124, 126, 128 -w ​99, 105–8 -(w)s ​99–100, 105–8, 130–32, 135 *C1aC2C3- ​123 CaCCɔC- ​122–23, 132 C1aC2C2iC3- ​123 C1C2ɔC3- ​122–23 *C1iC2C3- ​123 CuCCɔC- ​121–23, 129 *C1uC2C3- ​123, 136 morphosyntax  ​7, 11–12, 48, 52, 63, 96, 97–136, 149, 151, 158 n. 26, 172, 191 see also syntax mosaic ​27–28 Nabatean  ​see under Aramaic Narsai ​205, 210 Memre ​77, 92 (table 5.8), 118–19, 125, 208 native language ​4, 12, 15 n. 15, 17–18, 20–24, 31–32, 34, 36–38 Neapolis ​ see Appadana Neo-Aramaic  ​see under Aramaic Neo-Assyrian Empire ​1 Neo-Babylonian Empire ​1 Neofiti, Targum  ​ see Targum Neofiti

289

neuter ​ see gender neutralization ​19, 20, 24, 30, 37, 39–40, 47, 151 nexus ​155–56 nisba ​ see -ɔyɔ under morphemes, Syriac Nisibis ​4, 26, 29, 31 nomen agentis ​120–24, 126, 129 see also -ɔnɔ under morphemes, Syriac nominal clause ​see verbless clause number ​104–8 nunation ​177 Odes of Solomon ​117, 126, 159–60, 203–4, 212–13, 215–18, 221 Old Aramaic  ​see under Aramaic Old Syriac Documents ​124 n. 74, 199– 200, 213–17, 220–22 see also P. Dura 28, P. Euphrates 19, P. Euphrates 20 Old Syriac Gospels ​see Index of Biblical Sources Old Syriac Inscriptions ​1, 3 n. 13, 28, 160 n. 28, 199–200, 212, 215 On Orthodox Faith ​ see Cyril of Alexandria Onqelos, Targum ​ see under Aramaic Orator’s Education, The ​ see Quintilian Orpheus ​28 Osrhoene ​26, 29, 31, 203 Ouranius ​31–32, 37 P. Dura ​27, 71 12 ​162, 186–87 19 ​61 n. 64, 70 28 ​29, 33 n. 46, 52–53, 124 n. 74, 213–17, 220–22 see also Old Syriac Documents 30 ​63 n. 70 31 ​71, 86 n. 75, 187 33 ​63 n. 70 46 ​71 P. Euphrates ​29–30, 50, 71 1 ​86 n. 75 2 ​61 nn. 64–65, 70–71

290

Index of Subjects

P. Euphrates (cont.) 3 ​86 n. 75 4 ​86 n. 75 6 ​30, 50–51 (with Fig. 4.1), 61 n. 65, 71, 86 n. 75, 163, 212, 219 7 ​30, 50, 86 n. 75, 212, 214, 219 8 ​61 n. 65, 70, 86 n. 75 9 ​61 nn. 64–65, 70–71, 86 n. 75 10 ​214 13 ​61 n. 65, 71 14 ​86 n. 75 16 ​61 n. 65, 70 17 ​61 n. 64, 70, 86 n. 75 19 ​29–30, 61 n. 64, 70, 213–15, 221–22 see also Old Syriac Documents 20 ​29–30, 124 n. 74, 213, 215–16, 221–22 see also Old Syriac Documents P. Grenf. 1 ​62, 68, 73 Pahlavi ​132, 220 see also Middle Persian Palmyrene  ​see under Aramaic Panegyric on the Christian Martyrs ​ see Eusebius of Caesarea Papuan ​21–23 Part 3 ​ see Isḥaq of Nineveh Parthian Empire ​1, 26 participle ​120–23, 129, 141, 158–59, 163–64 particles ​117–20, 135–36, 139, 145, 153, 174–94, 205 see also existential particle pattern replication ​46, 147–48 pattern transfer ​151 Pawlos of Kallinikos ​34 Persian, Modern ​132 n. 95 Peshiṭta ​ see Index of Biblical Sources pharyngealization ​66–67, 84 Philoxenian Version ​32 Philoxenos ​32–33, 37, 140–41 (graph 7.1, table 7.1), 170–73 (tables 8.2– 8.3, graphs 8.1–8.2), 206, 224–25 Commentary on John ​32 Commentary on Matthew and Luke ​ 77, 92 (table 5.8)

Philoxenos (cont.) Discourses ​125 Letter to Maron of Anazarbus ​32 Letter to the Monks of Beth Gawgal ​ 155 n. 14 Phoenician ​84–85 Phoenix ​28 Photius ​ Bibliotheca ​36 n. 64 pidgin ​21 Pisidia ​38 Piyyuṭim (Jewish Palestinian Aramaic) ​ 168, 207–10 see also Jewish Palestinian Aramaic under Aramaic Piyyuṭim (Samaritan Aramaic) ​166–67, 207 see also Samaritan under Aramaic Plato ​5 n. 20, 28 plural ​ see number Plutarch ​5 Polykarpos (translator) ​32 Porphyry ​5 Prince Edward Island ​see French Prose Refutations ​ see Ephrem Pseudo-Clementine, Recognitions and Homilies ​28, 127 Qenneshrin ​34 Qiyore of Edessa Cause of the Liturgical Feasts ​76 n. 41, 92 (table 5.8), 118–19 Quintilian ​35 n. 59 Qumran Aramaic  ​see under Aramaic R2 value ​142 n. 8 Rabbula ​34–37, 114–15 see also Life of Rabbula Canons ​34 Letters ​34 Sermon ​35 Works ​114 reanalysis ​147, 151, 160, 178, 196 recipient language agentivity ​see borrowing

Index of Subjects recipient language ​12, 13, 15–17, 19–24, 30, 36, 39, 46–48, 53, 112, 120, 129, 139–40, 144, 147, 150–51, 172–73, 195–98, 201 see also borrowing (recipient language agentivity) replica language ​139 n. 2 replication, grammatical ​see grammatical replication Reshʿayna ​26 see also Sergios of Reshʿayna restructuring ​147, 149 resyntactization ​151 Roman Empire ​1, 6, 25–30, 32, 40, 55, 63, 116, 142 n. 9, 200, 203 root-and-pattern morphology ​113, 117, 120–24, 129, 135 Sabaic ​176 Sahdona Works ​110 Samʾalian  ​see under Aramaic Samaritan  ​see under Aramaic Samaritan Targum ​189–90 see also Samaritan under Aramaic Sassanian Empire ​1, 6, 26, 29, 31, 33– 34, 204 Scholia ​ see Yaʿqub of Edessa School of the Persians ​32 Second Epistle to Succensus ​ see Cyril of Alexandria secondary derivations ​120–29 selective copying ​46, 149–50 Seleucia-Ctesiphon ​3, 26 Seleucid Empire ​1, 25–26, 40, 199, 202 Seleucus I Nicator ​26 Sergios of Reshʿayna ​5 n. 20, 34 Sermons ​ see Ephrem Severos Sebokht Treatise on the Astrolabe ​76, 79 Severus of Antioch ​5, 38–39 Shemʿun of Beth Arsham ​ Letter on Ḥimyarite Persecution ​ 140–41 (graph 7.1, table 7.1),

291

Shemʿun of Beth Arsham, Letter (cont.) 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 206, 224–25 shift-induced interference  ​see interference through shift Singara ​26 singular ​ see number source language agentivity ​see imposition source language ​12, 15–19, 21, 23–24, 30, 35, 46–47, 53, 101, 105, 139–40, 144, 147, 151, 196–97, 201, 206 see also imposition Sozomen Ecclesiastical History ​199 n. 13 Sozopolis ​38 Spanish ​198 spiritus asper ​66, 69, 75–82 spiritus lenis ​78–80 stability gradient of language ​15, 18– 20 Standard Literary Aramaic  ​see under Aramaic state (Aramaic) ​97–98, 108–10, 136 see also status absolutus, status constructus, status emphaticus status absolutus ​98, 128 n. 85 status constructus ​98 status emphaticus ​98–99, 101, 103, 109–11 structural borrowing ​20, 149, 196–97 see also borrowing (general) Sumerian ​2 ​ syɔme ​76 n. 41, 79, 110–11, 131 n. 91, 221 syntactic borrowing ​149 syntax ​2, 7, 13–15, 20–24, 30, 35, 40, 64, 109, 139–94 see also grammatical replication, morphosyntax TAM ​163 n. 30, 171 nn. 42–43 Targum, Samaritan ​see Samaritan Targum Targum Jonathan ​ see under Aramaic

292

Index of Subjects

Targum Neofiti ​168 see also Jewish Palestinian Aramaic under Aramaic Targum Onqelos ​ see under Aramaic Teaching of Addai ​60, 82 (table 5.4), 92 (table 5.8), 140–41 (graph 7.1, table 7.1), 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 223, 225 Tella ​29, 33 see also Yuḥanon of Tella tense ​ see TAM Themistius ​5 Theodore of Mopsuestia ​5 Theodoret of Cyrrhus ​36–37, 39 n. 75 Cure for Hellenic Maladies ​36–37 History of the Monks of Syria ​36–37 Theodosius of Alexandria Theological Discourse ​76 n. 41 Theological Discourse ​ see Theodosius of Alexandria Theophany ​ see Eusebius of Caesarea Tibat Marqe ​165–67, 207 see also Samaritan under Aramaic Tigre ​177 Timotheos I Letter 47 ​156 Titus of Bostra ​ Against the Manichaeans ​28, 125 n. 78, 127 token ​141–42, 161, 171, 185–87 (with table 9.1), 203, 205, 207–9 translation technique ​5, 143–44, 160–62 Treatise on the Astrolabe ​ see Severos Sebokht type ​124 n. 74, 141–42, 203, 205, 207–9 Ugaritic ​50 n. 25, 176 ultimate source ​53–54 Urfa  ​see Edessa

ʾUrhɔy ​ see Edessa verb ​111–17 verbless clause ​153–73, 197–99, 201, 206, 210, 223–25 Vita ​ see Life Vulgäraspiration ​79 Windisch’s Law ​13 Works ​ see Rabbula or Sahdona Yaʿqub of Edessa ​3, 116 Correctness of Speech, The ​64 Letter 13 ​49–50, 75, 140–41 (graph 7.1, table 7.1), 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 188, 205–6, 224–25 Letter 18 ​140–41 (graph 7.1, table 7.1), 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 205–6, 224–25 Letter on Syriac Orthography ​64 n. 1, 65, 75 n. 37, 80 n. 54, 82 (table 5.4), 92 (table 5.8), 93–94 (with table 5.9), 101 n. 15, 185 n. 31, 189 n. 37 Scholia ​76, 113, 115 n. 44 Yaʿqub of Serugh ​31, 35 n. 59, 43–44 Memrɔ on Elijah and His Ascension into Heaven ​43–44 Yuḥanon of Ephesus ​55–56, 116 Ecclesiastical History ​59–60, 75–77, 79–80, 113–15, 119, 125 Lives of the Eastern Saints ​33 n. 45, 48, 50–51, 59, 76–77, 82 (table 5.4), 113, 119, 140–41 (graph 7.1, table 7.1), 170–73 (tables 8.2–8.3, graphs 8.1–8.2), 224–25 Yuḥanon of Tella ​33–34 see also Life of Yuḥanon of Tella Zekarya of Merv ​45 n. 4