Translating Religion; Linguistic Analysis of Judeo-Arabic Sacred Texts From Egypt - Hary, Benjamin H

Translating Religion

Études sur le Judaïsme Médiéval Fondées par

Georges Vajda Dirigées par

Paul B. Fenton

TOME XXXVIII

Translating Religion Linguistic Analysis of Judeo-Arabic Sacred Texts from Egypt

by

Benjamin H. Hary

LEIDEN • BOSTON 2009

This book is printed on acid-free paper. Library of Congress Cataloging-in-Publication Data Hary, Benjamin H. Translating religion : linguistic analysis of Judeo-Arabic sacred texts from Egypt / by Benjamin H. Hary. p. cm. — (Etudes sur le judaïsme médiéval ; v. 38) Includes bibliographical references and index. ISBN 978-90-04-17382-8 (hardback : alk. paper) 1. Judeo-Arabic language— Dialects—Egypt. 2. Judeo-Arabic literature—Egypt—History and criticism. 3. Jews— Egypt—Languages. I. Title. II. Series. PJ5079.5.E49H37 2009 492.7’70962—dc22

2009003328

ISSN 1568-5004 ISBN 978 90 04 17382 8 © Copyright 2009 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Hotei Publishing, IDC Publishers, Martinus Nijhoff Publishers and VSP. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. printed in the netherlands

È·‡ ¯ÎÊÏ ®≤∞∞≥—±π≤¥© È¯‰ ¯È‡Ó ®±π∏∑—±π∞≥© È¯‰ ÏÁ¯ È˙·ÒÂ ‰¢·ˆ˙ ÌÈÈÁÏ Ï„·ÈÈ ÈÁ‡ „Â·ÎÏÂ È¯‰ ÈÏ‡

In memory of my father Meir Hary (1924–2003) In recognition of my grandmother Rachel Hary (1903–1987) and in honor of my brother Eli Harry (1950– )

CONTENTS List of Figures and Tables....................................................................xi Preface................................................................................................xiii Technical Notes..................................................................................xix Introduction......................................................................................xxiii PART I JUDEO-ARABIC: THE LANGUAGE OF ARABIC-SPEAKING JEWS Chapter One – The Jewish Linguistic Spectrum.....................................5 The Sociolinguistics of Jewish Varieties.....................................5 Issues of Terminology ..............................................................8 The Emergence and Development of Jewish Varieties..............13 Crossing Religious Boundaries................................................16 The Characteristics of Jewish Varieties ...................................19 A List of Jewish Varieties........................................................25 Chapter Two – Judeo-Arabic within the Jewish Linguistic Spectrum.......................................................................................29 An Overview of Judeo-Arabic................................................29 The History of Judeo-Arabic...................................................32 The Structure of Judeo-Arabic.................................................37 The Judeo-Arabic Continuum .................................................39 The Diachronic Development of Judeo-Arabic Continuglossia ..................................................................41 The Current State of Judeo-Arabic.........................................44

CONTENTS

Chapter Three – The Translation of Sacred Texts into Judeo-Arabic (the ¡ar˙) ................................................................51 Translation and Issues of Sacredness.......................................52 The Translator’s Dilemma.......................................................57 The Development of the ¡ar˙ ..................................................60 The Cairo Collection................................................................63 Previous Studies on the ¡ar˙....................................................65 The Framework for the Linguistic Analysis of the ¡ar˙............68 Head-to-Toe Scanning.............................................................74 The Translation Continuum....................................................83 The Work of the ¡ar˙an...........................................................85 Chapter Four – Spoken Egyptian Judeo-Arabic: The Evidence from the ¡ar˙ Texts.......................................................................91 Spoken Egyptian Judeo-Arabic — An Introduction................91 Methodology..........................................................................93 1. Phonetics and Phonology...................................................100 2. Morphology .....................................................................112 3. Syntax ...............................................................................125 4. Lexicon ..............................................................................131 5. Summary of Colloquial Features of Spoken Egyptian Judeo-Arabic...................................................................134 Chapter Five – Additional Linguistic Issues of the ¡ar˙ Tradition......137 Issue One: Pseudocorrections ...............................................137 Issue Two: The Theoretical Background of the Use of Hebrew and Aramaic Components in Judeo-Arabic.........144 Cases of Interference in Direction A...........................148 Cases of Interference in Direction B...........................153 Issue Three: The Hebrew and Aramaic Lexicon in the ¡ur¥˙...156 Summary...............................................................................159

viii

CONTENTS

PART II A LINGUISTIC MODEL OF THE JUDEO-ARABIC TRANSLATIONS OF SACRED TEXTS Chapter Six – Applying the Model......................................................163 Introductory Notes ...............................................................163 Methodological Notes...........................................................165 Dynamic Literal/Interpretive Linguistic Tension: Complex Cases................................................................166 Calque Translations...............................................................178 The Organization of the Examples.........................................182 Chapter Seven – The Phrase and the Word Levels...............................183 Word-for-Word Translation..................................................183 Word Order: Syntactic Adaptation........................................188 Word Order: Adverbs............................................................192 Word Order: Numerals..........................................................198 Lexicon: Semantic Considerations.........................................200 Lexicon: Sound/Appearance Considerations.........................205 Chapter Eight – The Morphosyntactic Level.....................................213 Negation: Nominal ................................................................213 Negation: Verbal....................................................................215 Prepositions .........................................................................219 Coordinating Particles and Conjunctions...............................231 Conditional Particles.............................................................234 Independent Personal Pronouns............................................237 Pronominal Suffixes..............................................................238 Relative Pronouns.................................................................240 Demonstrative Pronouns ......................................................244 Interrogative Pronouns and Particles.....................................248 Verb Conjugation: Infinitives.................................................250 Verb Conjugation: Finite Verbs..............................................256

ix

CONTENTS

Cases: Accusative .................................................................257 Cases: Directional .................................................................264 Definiteness..........................................................................266 Agreement: Number ..............................................................274 Agreement: Gender...............................................................280 Tense and Aspect..................................................................287 Mood ....................................................................................293 Voice: Passive .......................................................................295 Numerals with Counted Nouns .............................................299 Chapter Nine – The Segment Level ....................................................303 Assimilation..........................................................................304 Emphatization and Deemphatization....................................305 Elision ..................................................................................306 Orthographic Marking of the Glides ......................................307 Diacritic Marks.....................................................................310 Hebrew-Influenced Orthography..........................................311 Summary...............................................................................327 Bibliography.....................................................................................329 Index .................................................................................................347

x

LIST OF FIGURES AND TABLES Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7-1 Figure 7-2 Figure 7-3 Figure 7-4 Figure 8-1 Figure 8-2 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13

Periodization of Judeo-Arabic...........................................34 The Judeo-Arabic Continuum............................................38 The Continuglossic Nature of Judeo-Arabic.......................42 The Deviation/Literal Continuum of Definiteness...............68 The Interpretive/Literal T-M-A Continuum........................70 The Interpretive/Literal Continuum...................................70 The Phrase Continuum......................................................71 The Lexical Continuum I....................................................71 The Lexical Continuum II...................................................72 The Morphosyntactic Continuum.......................................72 The Word Level Continuum...............................................75 The Segmental Continuum.................................................75 The Less Literal/More Literal Continuum..........................85 What Happens in the Process of Translation?....................86 The ¡ar˙an’s Work............................................................87 The Development of the ≠im in Urban Egyptian Arabic.......96 Direction A: Hebrew and Aramaic Are the Recipient Languages .................................................145 Figure 14 Direction B: Arabic Is the Recipient Language..................146 Table 1 Table 2 Table 3 Table 4

The Translation of Hebrew ÏÂ˜ in Genesis and Exodus........72 The Translation of Hebrew ÈÏÂ˜· ÚÓ˘ in Genesis and Exodus..................................................................73 The Example from the Passover Haggadah......................76 The Linguistic Model of the Analysis of the ¡ur¥˙ ..........81–82

PREFACE This book is about Arabic-speaking Jews, what and how they write and speak, and how they composed and used their liturgical or sacred writings in the past. How did I come to write such a book? Scholars at times explore issues that are deeply meaningful to them, and at other times they investigate topics that are as far removed from them as possible. When I wrote about Arabic-speaking Jews previously, I did not consider it to be a topic I was personally involved with. However, when I sat down to write the preface to this book, I realized that this topic is, after all, very personal to me. Despite my best efforts to be objective in my work, I know that biases probably remain of which I am unaware. Thus, I begin by situating myself in relationship to the material about which I am writing by establishing my background and making explicit my motives for writing this book. Growing up in Haifa, Israel, in the 1960s was not an easy task for a boy who was searching for his own identities, caught between the conflicting worlds of Jews and Arabs, Ashkenazim and Sephardim, secular and religious communities, and the like. My mother, Miriam Rebensaft, and her family had fled Nazi Germany in 1936, saving their lives, but not their Berlin middle-class status. The family settled in Bat-Galim, a small community in Haifa on the Mediterranean coast that absorbed many German Jews. The family remained proud of its German roots. My maternal grandmother Alice never bothered to learn Hebrew. She considered herself German and struggled to get along with the locals—Jews and Arabs—whom she referred to as diese barbarischen asiatischen Leute (these barbaric people from Asia). I was continually amazed by her bluntness.

PREFACE My grandmother also objected to my mother’s marriage to my father, Meir Hary. His father Haim Hary, a staunch Zionist, had been raised in Brody, a city in present-day Ukraine, around the turn of the twentieth century. With a degree in architecture from the University of Vienna, Haim Hary had come to then-Palestine to “build” Haifa. But he was not a “total” Ashkenazi, or Jew of Central and Eastern European descent.1 Although Haim came from Brody, it was well known that his family’s origins could be traced back to sixteenth-century Safad, at that time the center of Jewish intellectual life in Ottoman Palestine. Furthermore, Haim’s third wife, Rachel, my paternal grandmother, who had been born in Haifa, had a grandfather from Morocco, which also gave my father Sephardi origins. It was hard enough for my grandmother Alice to live in “the Orient,” as she called it, but to let her daughter marry a non-Ashkenazi was beyond her understanding.2 The marriage took place, however, and my grandmother learned to accept her son-in-law. Our household, though, was very German. When I first went to Berlin in 1977, following my mother’s early death, I was invited to breakfast by her surviving cousin, who served weich gekochtes Ei (soft-boiled egg) with the typical small silver spoon. I responded, “This is just like Israel,” not realizing that there was nothing Israeli about the soft-boiled eggs or the spoon, and that in fact I had been raised with a strong German cultural influence. I grew up in a family that always aimed to provide the children with a better education in order to improve the family’s socioeconomic situation amid the troubled economy of Israel at that time. I was sent to one of the best semiprivate schools in Israel, The Hebrew Reali School in Haifa. The atmosphere in the school, which thrived on

1

2

The term Ashkenazi is itself problematic and worthy of analysis; compare my discussion of the terms Mizrahi, Sephardi, and “Arab Jews” on pp. 30–32, below. I am fortunate to have found my mother’s diary after her death in 1976, in which she described, with much agony, the difficult situation in which she found herself when considering marriage.

xiv

PREFACE snobbish Ashkenazi elite culture, did not look favorably upon frenkim, the derogatory Hebrew term for non-Ashkenazim in the 1960s, or upon Christian and Muslim Arabs. Considering that the non-Ashkenazi origin of our family was ignored and the Moroccan background practically denied, there was no surprise that I completely identified with my German heritage and saw myself as such. Nevertheless, I was attracted to the Arabic language. Trying to understand the “other” intrigued me from a young age. I heard at home and in school that Arabs were “bad” and that “we could not trust them,” yet I sometimes heard my uncles speaking in Arabic with my grandmother Rachel, and from time to time saw Arab friends from Haifa coming and going in our apartment. I studied Arabic seriously in high school and majored in Arabic and Islamic studies. I used Arabic during my military service and then studied it at the Hebrew University of Jerusalem and the University of California, Berkeley, where I earned my doctorate. Choosing Judeo-Arabic as my primary area of research closed a circle for me, because I finally got to explore the religiolect which was used by my grandmother Rachel’s relatives in Morocco, as well as by many of my fellow Israelis. Although many scholars of Judeo-Arabic are native speakers of the religiolect or at least had heard it at home, growing up I only heard a little bit of German in addition to Hebrew. By studying Judeo-Arabic, I am in a sense reclaiming my Moroccan roots, and this is one of the reasons that motivated me to write this book. It is in this context and these circumstances that I situate myself and my research on Judeo-Arabic. In acknowledgment of my heritage, I dedicate this book to the memory of my father Meir Hary Ï¢Ê, in recognition of my paternal grandmother Rachel Hary Ï¢Ê, and in honor of my loving brother Eli. The work on this project began in 1994, when I was first introduced to the Cairo Collection. As explained in chapter 3 (pp. 63–65), the collection consists of more than one hundred photocopied manuscripts and is housed at the Institute of Microfilmed Hebrew Manuscripts in the Jewish National and University Library in Jerusalem. The manuxv

PREFACE scripts contain mainly Jewish liturgical texts written in Hebrew, Aramaic, and Judeo-Arabic. Many of them are ¡ur¥˙, which are verbatim translations of sacred religious and liturgical Hebrew/Aramaic texts into Judeo-Arabic, and in this case the Egyptian (mostly Cairene) variety of Judeo-Arabic. I started to publish articles about the nature of these translations and soon it became apparent that the writing of a volume or two was warranted. This book is the first, dealing with theoretical issues concerning these texts. The next volume, Sacred Texts: The Tradition of the ¡ar˙ in Egyptian Judeo-Arabic, with Critical Editions and Translations of the Book of Genesis, the Book of Esther, and the Passover Haggadah, due to appear in 2009, includes critical editions of three sample ¡ur¥˙ along with their translations into English and a linguistic introduction and commentary. I could not have completed such a project without the enormous help that I received from family and friends. I wish to thank my closest friend, Martin J. Wein, who read parts of the book and helped me shape some of my ideas about Christian, Jewish, and Muslim religiolects. Working with him on a separate project on these religiolects enriched me in ways beyond description. Martin has always been a source of love, energy, and intellectual challenge. I have also enjoyed many conversations about Judeo-Arabic with friends and colleagues: David Blumenthal, Piero Capelli, Rkia Cornell, Vincent Cornell, Shoshana Felman, Ruby Lal, Jeff Lesser, Frank Lewis, Deborah Lipstadt, Roxani Margariti, Gordon Newby, Yaron Peleg, Marina Rustow, Jacob Wright, and Ofra Yeglin. I thank Nick Fabian, Illan Gonen, Gene McGarry, and Michael Gugenheimer, who not only edited and proofread different chapters of the book, but also offered valuable comments and ideas. I would also like to thank the following colleagues, friends, and students, who helped me with different sections of the book at various stages: Angelika Bammer, Elitzur Bar-Asher, Moshe Bar-Asher, Sarah Benor, Michael Berger, Joshua Blau, Shmuel Bolozky, Alan Cienki, Abraham David, David Engel, María A. Gallego, Ophira Gamliel, Elena xvi

PREFACE Glaznov-Corrigan, Noa David, Sander Gilman, Galia Hatav, Nate Hoffer, Geoffrey Horowitz, Shlomo Izre’el, George Jochnowitz, Joshua Keller, Geoffrey Khan, Rina Kreitman, Damon Lynch, Aharon Maman, Meira Polliack, Emily Pollokoff, Gabi Rosenbaum, Ora Schwartzwald, Elana Shohamy, Doron Shultziner, Robert Smith, Debra Spitulnik, Devin Stewart, Sasson Somekh, Norman Stillman, Joseph Tedghi, Yosef Tobi, Don Tuten, Ofra Tirosh-Becker, and Sarah Willen. Friends and family stood by me during the long process of writing this book and always bestowed on me emotional support and love: Ursula Blumenthal, Shuki Cohen, the five Harys in Los Angeles (Chris, David, Eliane, Goni, and Tania), Kimberly Katz, Raphy Marom, Wendy Newby, Arnon Rolnick, Mark Tanner, Tsipi Wagner, and especially my father, Meir Hary Ï¢Ê, who always kept asking when the book would appear; also my brother Eli and my sister-in-law Rachel, my nephews, Yuval and Tomer, and their partners Anat and Rony, my great-nephew Yonatan, my niece Mika, and my cousin Yael. Many thanks go to Paul Fenton, the editor of the series, to Michiel Klein Swormink, Publishing Manager, and his team at the Brill office in Boston, Michael Mozina and Jennifer Pavelko, who were cooperative and engaging throughout the process of bringing the book to print. I am also grateful to the units at Emory University of which I am a member: the Department of Middle Eastern and South Asian Studies, the Tam Institute for Jewish Studies, and the Program in Linguistics. All have been consistently helpful and supportive of my research and teaching. Additionally, the Social Science Research Council funded the beginning of this study in 1995; the Emory University Research Committee also contributed funds to the project in 1995; the Institute of Critical International Studies at Emory College of Arts and Sciences funded several international trips connected to this study; the Jewish Studies Enrichment Fund and the Woodruff Fund both helped with the editing; and the Emory College of Arts and Sciences and Emory Graduate School helped finance the completion of this book. xvii

PREFACE Thanks also go to Sulaiman Jubran (chair) and colleagues in the Department of Arabic at Tel Aviv University, where I taught in 2001. I would like to acknowledge the helpful and dedicated staff at the Oxford Centre for Hebrew and Jewish Studies in Yarnton Manor at Oxford University, where I spent the spring and summer of 2005 as a Visiting Skirball Fellow. The staff at the Bodleian Library in Oxford was also very helpful. I am also grateful to the staff at the Center for Advanced Judaic Studies at the University of Pennsylvania in Philadelphia, who were helpful with assisting me with ms. HB 15. Finally, the Israeli Academic Center in Cairo and especially its administrative and financial director, Amr Zakarya, as well as its general director, Gabi Rosenbaum, gave me an emotional and intellectual home during my visits to Egypt: ‰Î¯·‰ ÏÚ ÌÏÂÎ Â‡Â·È. People who know me are familiar with my love of a good cup of cappuccino. I cannot complete my acknowledgments without expressing my gratitude to the staff in several cafés on three continents, where I spent countless hours writing this book: Midcity Cafe, Midtown Starbucks, and Octane Coffee Bar and Lounge in Atlanta; Caffè Nero on Broad Street in Oxford; and Cafeneto and Saquella Café in central Tel Aviv. Atlanta, December 2008 Benjamin H. Hary

xviii

TECHNICAL NOTES (i) In this volume I use a broad phonemic conventional transcription, which is employed regularly in Middle Eastern and Islamic scholarship, for all the Judeo-Arabic citations. The citations include the JudeoArabic spelling in Hebrew characters, followed by a phonemic transcription when deemed necessary. Then follows the citation from the manuscripts (in bold) and the folio and the line numbers (all in parentheses), succeeded by the translation into English in single quotation marks. For example, ‰Ï‚Ú˙Ò‡ /ista>galna/ (15 29a,8) ‘we hastened’ is cited from ms. 151 (Genesis), folio 29a, line 8; and ¯˘‡„Á /˙idaa¡ar/ (74 23,17) ‘eleven’ is cited from ms. 74 (Haggadah), folio 23, line 17. In some chapters (such as 4 and 9), a more precise transcription is needed, so the allophonic (narrower) transcription is displayed, as is customary, between square brackets [ ], while the (broader) phonemic transcription appears between slashes / /. (ii) The transcription normally does not indicate initial glottal stops for the following reasons: first, it is not customary in works of Middle Eastern and Islamic Studies; second, the phoneme is not always attested in the spoken variety; and finally, Wehr’s dictionary of standard Arabic (1976, 1994) uses the practice adopted here. Thus, in the following examples the initial glottal stop is not marked in the transcription: ‰Á‡ /i˙na/ (15 10a,16) ‘we’ and Â˙‡ /intu/ (74 14,1) ‘you (masc. pl.).’ The same occurs with nouns that follow the definite article: /al-um¥r/ ‘the things,’ rather than /al-
For some peculiarities with the numbering of the folios in ms. 15 (especially folios 0 and 23-1), see Hary 2009, the critical edition of Genesis.

TECHNICAL NOTES Since the transcription is often based on Spoken Egyptian JudeoArabic, if the initial glottal stop comes from an Old Arabic qåf, it is indicated in the transcription for historical reasons, as is the custom in Hinds and Badawi’s dictionary of Egyptian Arabic (1986): ÂÓÂ˜ /<ømo/ (3 4,7) ‘his people’ and √‡ˆÂ˜ /ala/ ‘on’ and /fi/ ‘in’ are usually denoted without the final long /å/ or long /•/; however, ‰Ï‡ /ila/ ‘to,’ also the definite direct object marker, may be transcribed as /ilå/ when deemed necessary. The relative pronoun È≤Ï‡ is transcribed /allaƒ•/ with long /•/ because it reflects Classical Arabic in the texts of the ¡ur¥˙. (vi) The transcription for Judeo-Arabic is as follows: For consonants: √ ø, /
xx

◊ ÿ Ÿ ⁄ · ‚ „

/†/ /Ω/ />/ /\/ /f/ /q/ /k/

‰ Â Ê Á … Ë Í

/l/ /m/ /n/ /h/ /t/ or /h/ /w/ /y/

TECHNICAL NOTES For long vowels: Í /•/, /∑/ Ë /¥/, /ø/

«

For short vowels: /i/, /e/ /u/, /o/

/a/

/å/

(vii) The transcription for Hebrew follows the guidelines of The Jewish Quarterly Review. Note that final ‰ is not marked in the transcription except for established terms such as Torah and Haggadah. (viii) The following abbreviations are used in the volume: fem. feminine p. page l. line pl. plural masc. masculine pp. pages ms. manuscript sg. singular mss. manuscripts vol. volume n. note

xxi

INTRODUCTION Translations of Hebrew and Aramaic sacred texts into Jewish languages, religiolects,1 and varieties have been historically widespread throughout the Jewish world. Among Judeo-Arabic speakers, the tradition of such translations is known as the ¡ar˙2 (pl. ¡ur¥˙). The present study analyzes the intricacies of this genre of translating Hebrew and Aramaic sacred texts into Egyptian Judeo-Arabic by examining specific eighteenth- and nineteenth-century manuscripts of the ¡ur¥˙. Haim Blanc, who has written extensively on Egyptian Judeo-Arabic, wrote in 1981 (n. 5) that “A Judeo-Arabic literary tradition peculiar to Egypt must have existed, but has not been investigated in detail.” My publication of megillat p¥r•m il-mißriyy•n ‘the Cairene Purim Scroll’ in 1992, and now the present study are attempts to fill this lacuna described by Blanc almost thirty years ago. This study, however, does not attempt to reconstruct the original ¡ur¥˙ of the Jews of Egypt, whose use of the genre has a long history that is still unclear. Rather, the study aims to provide a plausible representation of what this translation tradition might have been like in Egypt in the period under study. The volume also attempts to shed light on the linguistic peculiarities of the genre. It develops a linguistic model of the process of translation of these sacred texts. Rather than viewing these texts as merely literal or “verbatim” translations, as has been the generally accepted approach,

1 2

For the definition of the term religiolect, see pp. 12–13. The literal meaning of the term is ‘interpretation,’ ‘explanation,’ or ‘commentary.’ See Wehr 1994:541.

INTRODUCTION this study traces in great detail the literal/interpretive linguistic tension with which the translators/interpreters/composers of these texts, known as ¡ar˙anim, actually struggled in their work. In addition to the desire to provide a verbatim translation of a sacred text, these translators also had to consider the linguistic parameters of the target religiolect and make decisions that affected their readership’s ability to read and use the translation. The book is divided into two parts. Part 1 investigates Judeo-Arabic, the language of Arabic-speaking Jews in general. Part 2 develops a linguistic model of the Judeo-Arabic translations of sacred texts. Chapter 1 sets the stage for the analysis by summarizing the spectrum of Jewish linguistic usage in historical and sociolinguistic terms. It challenges definitions of generally accepted terminology and establishes new terms such as Jewish-defined language, majority language, (language) variety, linguistic intelligibility, religiolect, castelect, and more. The chapter also tracks the emergence and development of Jewish languages, followed by a discussion of instances where Christians and Muslims have participated in the Jewish linguistic spectrum and adopted some of its usages. Furthermore, the chapter maps the prototype of a Jewish language and lists the various Jewish languages mentioned in the literature. I must emphasize that my analysis employs a measure of social construction, for lack of a better term, in grouping together a variety of languages/dialects/varieties under the rubric of “Jewish languages.” The notion of “Jewish languages,” “Christian languages,” or “Muslim languages” provides a useful analytic tool, but ultimately does not constitute an “element in the world.” Rather, it is an artificial grouping of an array of language formations that we can productively analyze because of the threads that link them religiously, linguistically, historically, and socioculturally. This is how I would like to clarify the distinction between commonplace notions of “Jewishness” and the kind of analytic grouping at the heart of my sociolinguistic analysis. In her effort to be a careful and responsible historian, Stern (2008) xxiv

INTRODUCTION has adopted a similar approach, attempting to destabilize archaeologists’ prior assumptions about what counts as “Jewish” or what makes something “Jewish.” She has claimed that “[r]eliance on essentialist or syncretistic models of cultural dynamics has limited past evaluations of ancient Jewish populations.” Using the methods of historical linguistics, among other tools, she has reexamined data on North African Jews and demonstrated “how direct comparison of Jewish material evidence with that of its neighbors allows for a reassessment of what the category of ‘Jewish’ might have meant in different North African locations and periods.” According to Stern, this examination “allows for a more informed and complex understanding of Jewish cultural distinctiveness.”3 Chapter 2 explores Judeo-Arabic within the general framework of Jewish religiolects. It reviews the history of Judeo-Arabic and analyzes its structure, while discussing the language continuum used by JudeoArabic writers and speakers and tracing its diachronic development. This chapter also tackles some additional terminological issues, especially with respect to the denotation of Arabic-speaking Jews. Finally, the chapter offers some new insights into the status of JudeoArabic today. Chapter 3 examines the literary genre of the translation of sacred Hebrew and Aramaic texts into Jewish religiolects. It considers the perceived sanctity of the translated texts and demonstrates how translators dealt with the constant linguistic tension between their desire to provide a verbatim translation of the sacred text and their need to adapt the translation to the linguistic parameters of the target religiolect so that readers could comprehend them. The chapter analyzes the reasons why such translations were made and traces the evolution of the ¡ar˙ in Judeo-Arabic, especially in North Africa and Egypt beginning in the fifteenth century, while also taking into account Saadia’s

3

I thank Sarah Willen, who called my attention to Stern’s work and provided me with the quoted text. xxv

INTRODUCTION earlier translation of the Bible in the tenth century. This is followed by a review of previous scholarship on the ¡ar˙ and a discussion of the Cairo Collection, from which several manuscripts relevant to this book are taken. This chapter then offers a linguistic method for analyzing the ¡ar˙ that is based on scanning the text on several linguistic levels, and then establishing a continuum of least-to-most-literary translations. Examples are provided from various categories and linguistic features. The chapter concludes with a description of various mechanisms that the ¡ar˙anim in Judeo-Arabic used when translating sacred texts. Chapter 4 departs to some degree from the previous analysis and attempts to demonstrate how a careful and thorough linguistic analysis of the ¡ur¥˙ can contribute to our understanding of the spoken variety among Egyptian Jews in the eighteenth- and nineteenth-century. In this chapter I examine selected features of spoken Egyptian JudeoArabic in phonetics and phonology, morphology, syntax, and the lexicon. The chapter summarizes the characteristics of spoken Egyptian Judeo-Arabic that are reflected in the texts of the ¡ur¥˙, and then explores how these characteristics set this variety apart from the spoken Egyptian Arabic varieties used by Christians and Muslims. Finally, the chapter addresses methodological issues connected to the reconstruction of the spoken variety used by Egyptian Jews, as it may be extracted from the texts of the ¡ur¥˙, and demonstrates the connection between the orthography and phonetics/phonology as well as its limitations. It also points out similar orthographical trends in today’s publications in modern Egyptian dialect. Thirty years ago Haim Blanc argued that the dialect spoken by Cairene Jews was not distinct from the dialect spoken by Christians and Muslims, as opposed to the situation in other regional varieties of spoken Judeo-Arabic. For example, Blanc argued that Baghdadi JudeoArabic was indeed distinct in many ways from Christian and Muslim Baghdadi dialects,4 which I now call religiolects. Since the publication 4

See Blanc 1964 and Rabin et al. 1979:49–52.

xxvi

INTRODUCTION of Blanc’s work, additional texts have become available for the examination of Late and Modern Egyptian Judeo-Arabic. In 1992 I published the beginning of my investigation of Judeo-Arabic; in the present study I continue that work while searching for colloquial elements that are present in the written texts of the religiolect. Furthermore, Gabriel Rosenbaum conducted extensive recordings of modern Egyptian Judeo-Arabic and reported on his findings in 2002 (see bibliography). All of these studies affirm that spoken Egyptian Judeo-Arabic is distinct from the varieties used by Christians and Muslims in Egypt. Chapter 5, too, is an excursus and explores the use of pseudocorrections in Judeo-Arabic in general and in the Egyptian ¡ur¥˙ in particular. It demonstrates how both hypercorrections and hypocorrections, two different types of pseudocorrections, are used in the texts, while also discussing the implications of the standardization of pseudocorrections in Egyptian Judeo-Arabic for both the literary and the spoken varieties. This chapter develops a broad theoretical model for the use of Hebrew and Aramaic components in Jewish religiolects, focusing in particular on Judeo-Arabic. Using Uriel Weinreich’s work on languages in contact, this model demonstrates how components are transferred into Judeo-Arabic from two different directions, explaining the reasons for each direction and giving examples of each direction from the various Egyptian ¡ur¥˙. Finally, the chapter provides a discussion of Hebrew and Aramaic lexical items employed in the Egyptian ¡ur¥˙. Part 2 of this book offers an in-depth analysis of the linguistic model of the Egyptian Judeo-Arabic ¡ur¥˙ (see table 4, pp. 81–82), as introduced more generally in chapter 3 through hundreds of examples extracted from the texts of the ¡ur¥˙. These examples, presented according to several levels, categories, and features, demonstrate the constant literal/interpretive tension in which the ¡ar˙anim found themselves. Most of the examples provided in this volume are based upon seven manuscripts, four of which are taken from the Cairo Collection: xxvii

INTRODUCTION • Ms. HB 15 (=CAJS Rare ms. 255 5 ), located at the Center for Advanced Judaic Studies at the University of Pennsylvania in Philadelphia – a partial ¡ar˙ of the book of Genesis • Ms. 1302, Jerusalem Ben Zvi Institute – the book of Esther • Mss. 3, 74, 91, and 93 from the Cairo Collection – Egyptian Judeo-Arabic Passover Haggadah (located in Jerusalem) The critical editions, translations, and analyses of these manuscripts will appear in Hary, Sacred Texts: The Tradition of the ¡ar˙ in Egyptian Judeo-Arabic, with Critical Editions and Translations of the Book of Genesis, the Book of Esther, and the Passover Haggadah, forthcoming in 2009. Chapter 6 explores the methodological considerations related to the organization of these examples and applies the linguistic model to complex examples from the ¡ur¥˙ in order to demonstrate how the ¡ar˙anim translated their sacred texts. The chapter also analyzes the ultimate verbatim translations, known as calque translations, and demonstrates how they were incorporated into the ¡ur¥˙. Chapters 7 through 9 arrange the examples, taken from the texts of the ¡ur¥˙, in descending units of grammatical structure, from the phrase level down through the lexical, morphosyntactic, and segment levels in thirteen categories, focusing on various linguistic features. Chapter 7 lists examples at the phrase and word levels, chapter 8 treats the morphosyntactic level, and chapter 9 deals with the segment level. The book concludes with an extensive bibliography and an index which primarily covers part 1. Because part 2 offers hundreds of examples, the index covers the general themes of the linguistic categories and features (which are also found in table 4 on pp. 81–82). The index also covers references from chapters 1 through 6 to Saadia Gaon’s 5

In fact, ms. 255 is the new number assigned to the manuscript. It was originally catalogued by Glatzer as ms. 317. This manuscript is actually a ¡ar˙ of the Torah, with some parts missing. The manuscript also includes haf†arot, which are chapters from the Prophets read in synagogues after the weekly portion of the Torah.

xxviii

INTRODUCTION translation (tafs•r), as well as the Protestant Arabic translation of the Hebrew Bible; however, these references from chapters 7 through 9 are not indexed, because they are cited as alternative translations to the ¡ur¥˙. Today, Judeo-Arabic is an endangered religiolect, perhaps on the verge of extinction. Although the SIL International Ethnologue project maintains that as of the mid-1990s there were close to 500,000 JudeoArabic speakers, that number has declined today to just under 400,000 speakers, and it is estimated that the last native speaker of the variety will die this century. Therefore, I view the research on Judeo-Arabic language, culture, and history as a “salvage operation” to record and preserve one of the most fascinating phenomena in Jewish, Arab, and Middle Eastern cultures.

xxix

PART I JUDEO-ARABIC: THE LANGUAGE OF ARABIC-SPEAKING JEWS

CHAPTER ONE THE JEWISH LINGUISTIC SPECTRUM This chapter investigates the spectrum of Jewish linguistic usage in historical and sociolinguistic terms. It opens with an examination of how sociolinguistics and history have inquired into when, how, and why Jews have written and spoken differently from their neighbors. It then tracks the emergence and development of “Jewish languages.”1 This is followed by a discussion of instances where Christians and Muslims have adopted Jewish linguistic usages, leading to a proposal for some modifications in the accepted terminology. The chapter goes on to map the prototype of a Jewish language and lists the various Jewish languages mentioned in the literature. Finally, definitions of several terms are given, including Jewish-defined language, majority language, (language) variety, linguistic intelligibility, religiolect, castelect, and more. The Sociolinguistics of Jewish Varieties Sociolinguistic studies attempt to analyze language use according to variables such as place of birth and language acquisition, place of domicile, age, gender, sexual orientation, socioeconomic factors, occupation, education, and so forth. An important variable which is often overlooked is religious affiliation and identity. Indeed, Jews almost everywhere tend to speak and write somewhat differently

1

I use the term “Jewish language/s” in quotation marks here because I have doubts about its accuracy. This is how it is meant to be used throughout the book. Later in the chapter I coin the term Jewish-defined languages.

CHAPTER ONE from their (non-Jewish)2 neighbors, in the same way that young people tend to speak and write differently from the elderly. Although this study emphasizes the connection between religion and language, this does not mean that Jewish varieties need to be examined only through the prism of religion. Jewish languages are also migration languages, and they are also varieties of their own majority languages. Thus Judeo-Italian, for example, in addition to being a Jewish language, is also a migration language and a variety of Italian. The distinct features of a language used by Jews can range from as little as a few Hebrew or Aramaic words, to thoroughgoing linguistic innovations in all areas of the language, resulting in a language form 2

6

The term “non-Jews” may have developed in the United States as a politically correct euphemism for the traditional Jewish terms goyim and shiksas as a result of their negative overtones. While the intention was positive, the category of “non-Jews” and its derivatives are still problematic. They continue an awkward dichotomy and play down differences among various “non-Jews,” so that all people who are not Jewish are lumped into one group. Clearly, it is not useful to group all Christians, Muslims, Hindus, and adherents of other faiths into a single category of “non-Jews.” In addition, the term is a definition ex negativo, lacking any positive descriptive content. People should preferably be defined by what they are, rather than by what they are not. “Non-Jews” generally do not use the term to describe themselves, except in Jewish-defined contexts, e.g., in Israel. For example, a woman in China (with no connection to Jews) would not call herself a “non-Jew” when describing her identity. The term also effects a division of humankind into two groups, along the lines of traditional rabbinical theological thought; “non-Jews” is thus really a nonacademic religious term. Since the academic world should be multi-religious and/or secular, with a strong critical tradition, the use of specific religious terminology (such as “nonbelievers,” “sinners,” “the righteous,” “the enlightened,” or “non-Jews”) would be inappropriate. In addition, the term leaves no room for interreligious syncretism, or for people who affiliate with several religions at the same time–for example, people who can be considered to be both Jewish and Christian (Wein 2005, 2008). Finally, this term implies that Jews can be grouped easily into a single category. However, Moroccan Jews, for example, can as easily be grouped with Moroccan Muslims (in terms of some aspects of food, dress, language, for example) as with Ashkenazim, or even seen as a separate group. Thus, his dichotomy flattens any in-group Jewish differences. Consequently, I have avoided this term whenever possible, and I only use it for convenience to avoid cumbersome circumlocutions. I thank Martin J. Wein, personal communication.

THE JEWISH LINGUISTIC SPECTRUM that is largely unintelligible3 to others. The spectrum of Jewish linguistic practice thus runs the gamut from some Yiddish words in the lexicon of secular Jews in America today to languages like Judeo-Arabic, Judeo-Spanish, and Yiddish.4 Consequently, the field of Jewish linguistics needs to explore questions such as the following: • What constitutes a Jewish language? • Do Jews speak different languages, dialects, or “religiolects”?5 • What are the similarities and the differences between Jewish languages in various times and places? • How and why do Christians and Muslims use elements of Jewish languages? • How do Jewish languages differ from their related co-territorial counterparts,6 and in what ways are the former imbued with Jewish “culture”? • What role does language play in the emergence of a collective identity and the creation of community boundaries? • How do non-Jews view language use among Jews?

3 4

5 6

See below, pp. 10–11, on the issue of mutual intelligibility. Birnbaum (1979) and Weinreich (1980) were among the first to discuss this phenomenon. Many scholars followed: Paper 1978; Rabin et al. 1979; Bunis 1981; Fishman 1981, 1985a, 1985b; Gold 1981, 1989; Rabin 1981; Wexler 1981, 2006; Hary 2004; Myhill 2004; Spolsky and Benor 2006; Hary and Wein 2008. Benor (2008) has critiqued the literature and offered important insights. See below, pp. 12–13, for a definition of the term religiolect. The term co-territorial language has been used to denote the dominant language used alongside “Jewish languages” in a specific area. The term is problematic because the dichotomy it creates gives rise to the mistaken impression that one of the languages is more “real” while the other is “co-.” In other words, it leads to the perception of one language (the “co-”) as marked and the other as unmarked. The situation on the ground, in fact, is more complex, since Jews have been known to use several languages simultaneously, and have often even mixed their uses of the various languages. It is therefore best to employ the term majority language, which does not have a judgmental connotation.

7

CHAPTER ONE Issues of Terminology Scholars have so far investigated about twenty7 Jewish languages. These range from Judeo-Arabic, Judeo-Berber, and Judeo-Persian, used by Jews in the Middle East, to Judeo-Italian, Judeo-Provençal, and Yiddish, which originated in Europe.8 In view of the sociolinguistic considerations mentioned above, the Jewish linguistic spectrum should also include any other distinctive modes of speech and writing used by Jews, in addition to the more than twenty or so Jewish languages. Jews develop such distinctive language forms out of a wish to distinguish themselves from their neighbors, or due to outside encouragement or pressure, depending on majority-minority dynamics. The distinctive features in question may involve aspects of Jewish ritual practice, cuisine, and fashion, or also specific characteristics of speech and writing. To take one example, Jewish language use in the Netherlands has received only scant scholarly attention to date, but further research into the matter, including a compilation of a corpus of written and spoken “Judeo-Dutch,” would no doubt reveal some interesting features of this language form. The same is true of many other language zones with a Jewish population, such as “Judeo-German” (Matras 1991), “Judeo-Polish” (Brzezina 1986), or “Judeo-Russian” (Verschik 2007), to name a few.9 It is probably impossible to define the concept of Jewish languages 7 8

9

8

A list of sixteen languages is provided in Rabin et al. 1979:58–66. See below, pp. 25–27, for a list of Jewish languages. The languages on this list are all part of the Jewish linguistic spectrum; other varieties may be added to the spectrum. When I was a visiting Skirball Fellow at the Oxford Centre for Hebrew and Jewish Studies in the spring of 2005, I gave two separate talks about the Jewish linguistic spectrum in Oxford and in Manchester. On both occasions people from Amsterdam approached me after the lecture, telling me that they had always noticed that the Dutch they spoke possessed some special features which, they assumed, were to be attributed to the fact that they were Jewish. See also Jacobs (2005:303–306) who has mentioned “Jewish Dutch” and other post-Yiddish “ethnolects.”

THE JEWISH LINGUISTIC SPECTRUM solely on the basis of linguistic considerations,1 0 as it is difficult to find linguistic criteria that are common to Judeo-Arabic, Judeo-Spanish, Judeo-Tat, and Yiddish, for example. In other words, it is difficult to conduct a comparative study of Jewish languages based on genetic1 1 or typological1 2 classifications: not all Jewish languages are genetically related (as are the Semitic languages, for example), nor do all possess common typological characteristics (such as the Subject-Object-Verb word order in Japanese and Turkish, for example, a feature which is usually also associated with other grammatical characteristics such as postpositions, adjective-noun order, and more). Jewish languages therefore need to be examined within a different framework where sociohistorical and sociolinguistic factors are also taken into account, since these languages share cultural commonalities.1 3 Norman Stillman (1991) may have been the first to make a serious study of this connection by showing parallels between Jewish and Muslim languages, although Haim Rabin had alluded earlier to Muslim and Christian languages (Rabin et al. 1979:42–43). Stillman demonstrated that the common bonds between Jewish and Muslim languages are the Hebrew script and the Hebrew/Aramaic vocabulary, and the Arabic script and Arabic vocabulary, respectively. 10

Thus, I adopt the same approach taken by S. Morag (Rabin et al. 1979:53) and M. Zand (ibid., 55). 11 By genetic classification I mean language families that are postulated to have originated from protolanguages, such as the Indo-European or Afro-Asiatic languages. 12 By typological classification I mean languages that share similar structural features. For example, if the usual word order in a language is Object-Verb, it will also tend to have post- rather than prepositions. These features are called implied universals. 13 Paper uses the term Kulturbund (1978:vii). I have reservations about this term since there is very little evidence for a strong Jewish Kulturbund in most periods in history. For example, marriages between Ashkenazi and Sephardi Jews were banned in many places (e.g., Amsterdam), being considered mixed marriages. Furthermore, the term itself has associations with the Central European Romantic nationalism of the nineteenth century that I would like to avoid.

9

CHAPTER ONE In light of this discussion, a sociolinguistic approach would appear to be more appropriate for Jewish languages than a purely linguistic approach. Furthermore, it is not clear whether the term language is in fact suitable for the entire Jewish linguistic spectrum. Certainly in some instances the term language would seem to be fully justified—for example, in the case of Yiddish, especially after its official standardization at the Czernovitz conference in Austria-Hungary in 1908. What constitutes a language that separates it from a dialect? What are the prerequisites of a speech form that turn it from a dialect into a language? The famous Yiddish phrase, “A language is a dialect with an army and a navy,” attributed to Max Weinreich,1 4 clearly does not answer these questions. Linguists in general prefer to use the term variety when they do not wish to commit themselves to either of the terms language or dialect. In fact, Joshua Fishman (1985) has convincingly demonstrated that there are no clear linguistic or social criteria that can be used to distinguish between a language and a dialect. The term variety is fuzzy and vague, and as such can describe ambiguous situations. Language has to be defined along linguistic, geographical, historical, political, religious, and sociological lines; the definition of a language as a collection of mutually intelligible dialects is inadequate for a number of reasons. First, mutual intelligibility is relative and needs to be examined along a continuum, not in absolute terms. For example, rather than saying that English and German are not mutually intelligible, it would be more fitting to argue that English and German are less mutually intelligible than German and Dutch. Second, mutual intelligibility is sometimes connected to issues of political power. In fact, we know of African tribes who claim that they do not understand a neighboring tribe, whereas the latter perfectly understands the former.

14

The actual quote is taken from Weinreich 1945:13, ÔÃ‡ ËÈÓ Ë˜ÚÏÈÈ„ Ã‡ ÊÈ‡ ÍÃ‡¯t˘ Ã‡ ËÀ‡ÏÙ ÔÂ‡ ÈÈÓ¯Ã‡. Weinreich writes that he heard it from someone else; George Jochnowitz thinks that the latter may be Joshua Fishman.

10

THE JEWISH LINGUISTIC SPECTRUM Upon closer examination, it turns out that the tribe that understands its neighbors’ speech possesses more political power. The “defending” tribe, in an effort to maintain its independence, argues that its members do not comprehend the language of the powerful neighboring tribe (Chambers and Trudgill 1998:3–5). Third, there are different degrees of understanding, depending on directionality. Indeed, it is claimed that Danes understand Norwegians better than Norwegians understand Danes. Finally, there are languages, e.g., in Scandinavia, that are mutually more intelligible than are some of the dialects of what is considered the same language, e.g., Arabic or Chinese. Thus, the definition of a language as a collection of mutually intelligible dialects is untenable. In sum, there is no sharp dividing line between a language and a dialect; the distinction between the two is fluid and depends on numerous variables, which may change over time. For example, the dialects spoken in southern Sweden today were deemed Danish until 1658, when this region of Denmark became part of Sweden. After forty years the same dialects were already termed Swedish. The new appellation was quite likely a result of the Swedish conquest, and not due to any significant changes in people’s speech habits. Thus, geographical, historical, and political, rather than linguistic factors caused the dialects to be called Swedish instead of Danish (ibid., 9–11). It is in fact doubtful whether in many cases it is even possible or advisable to determine if a certain speech variety is separate or not. For example, Benor shows that Orthodox Jewish English, also called “Yeshivish” or “Yinglish,” is variously considered to be a separate language, a dialect, or a jargon, while some do not notice anything that distinguishes it from other varieties of English. Linguists describe the various varieties and possible ideologies associated with them, irrespective of whether a given language variety can be considered “separate” from another.1 5

15

See Benor, Jewish Languages Listserv, March 4, 2005. I have been much influenced by Sarah Benor’s work in the field of the Jewish linguistic spectrum.

11

CHAPTER ONE Overall, it is better to use the vague term variety or language variety rather than language.1 6 In the past I have also used a different term, ethnolect,1 7 in the context of Jewish varieties. However, the term “ethnic” is very problematic and has undergone several changes in meaning. In popular usage its meaning is close to “racial,” but the academic usage is very different. Thus, for example, “ethnicity”1 8 has been defined as a “named human population with a myth [emphasis mine] of common ancestry, shared memories and cultural elements, a link with a historic territory or homeland and a measure of solidarity” (Smith 1993:49).1 9 A better and more suitable term is religiolect, which I mentioned briefly in Hary 1992:xviii n. 1. The term religiolect avoids the messiness of “ethnicity” and relates directly to the religious backgrounds of the people who use this language variety. A religiolect is thus a language variety with its own history and development, which is used by a religious community.2 0 A Jewish 16

Gold and Prager have used the term lect: “In order to see the objects of our inquiry ranged in a continuum, we choose to speak of lects, which we do not arrange in any rank of preference … We collect data as we find it in any Jewish lect, even those whose distinguishing marks appear to be few” (Prager 1986:225). 17 See Hary 1992, 1995:74, 1996c:727–28, 1997a:35–36, 1997b:220, n. 2, 1999:67–68, and elsewhere. 18 The problems associated with the terms “ethnic,” “ethnicity,” and the like are the reason for placing them in quotation marks throughout this study. 19 This is one of the classic definitions of ethnicity. A. D. Smith is a theorist of nationalism, and so his definition needs to be understood in this context. 20 A special case may be that of Yiddish-speaking secular Jews of the first half of the twentieth century, for example, who might have been uncomfortable with the term religiolect in reference to their variety. However, the term religiolect describes not a personal identity of its speakers, but characteristics of a variety that had often been embedded before the rise of secularism. Furthermore, even secular Jews may resort to religious self-definition in group construction, e.g., Bundists, Folkists, or Zionists. The term religiolect, however, may not fit all religious communities. In India the term castelect is more suitable. Thus, Jewish Malayalam is one of many castelects of Kerala. The dichotomy, then, in Kerala is not necessarily between the various religious communities, but rather between /ambalakkår/ ‘those who go to temples,’ and /pal6l6ikkår/ ‘those who go to /pal6l6i /, prayer shrines, i.e., churches/mosques/synagogues.’

12

THE JEWISH LINGUISTIC SPECTRUM religiolect, then, is a spoken and/or written variety employed by the Jewish population of a specific area, although it later may extend also to other communities and areas (see below). Our knowledge of the Jewish religiolects of the past is inadequate, since in many cases scholars began to study them when it was too late and only a handful of speakers were still using them—or, worse still, they had already become extinct. New Jewish religiolects have been created in modern times, due in part to migration patterns, conversion, and an increase in Jewish identity. Some of these modern varieties (see below, p. 27) have only been investigated in part, if at all. The Emergence and Development of Jewish Varieties2 1 Jewish varieties, we assume, developed in parts of the Diaspora from preexisting languages and were used in both written and spoken forms by Jews within their communities. Some may have developed as a result of the migration and dispersion of the Jews throughout Africa, Asia, and Europe during the early centuries of the Common Era (Birnbaum 1971:68). These varieties initially came into being out of a desire for integration into the non-Jewish environment, but later came to be a hallmark of “continuing Jewish consciousness and identity” (Ben-Sasson 1971:771). Put another way, the Jews’ initial adoption of a preexisting language in the Diaspora may have been an attempt to fit into their new environment, but later the language established itself as a Jewish variety, with Hebrew script, Hebrew and Aramaic linguistic elements, and other distinctive characteristics, thus becoming a symbol of Jewish identity and an actual obstacle to integration. It is also possible that Jewish varieties have developed as a result of conversions to Judaism, and not just as a result of Jewish migration. In other words, non-Jews who converted to Judaism or non-Jewish

21

I thank Martin J. Wein for some of the ideas expressed in this section and for our long conversations and debates about these issues. See also the discussion in Wexler 2006:xv–xix.

13

CHAPTER ONE members of Jewish or mixed households—e.g., manumitted slaves, servants, nannies—may also have contributed to the development of Jewish varieties. This latter point has often been ignored in Jewish studies. It should be reemphasized, therefore, that the strict differentiation between Jews and other communities and religious groups only emerged gradually. The historian H. H. Ben Sasson (1994:277) has the following to say about what the boundaries between Jews and non-Jews looked like in the period that produced such important Jewish varieties as JudeoAramaic and ancient Judeo-Greek (Yevanic): In the Second Temple era, the Jewish faith expanded as it had never before and has never since. Throughout the Roman Empire and beyond it, people adopted the Jewish faith or at least part of the Jewish way of life. Large sections of the Jewish people made it their concern to convert the heathen to Jewish monotheism and took pride in the fact that Jewish customs were to be found everywhere.

The subsequent rise of Christianity did not immediately halt the expansion of emerging rabbinic Judaism. Apart from the conversion of two kingdoms to nonrabbinic forms of Judaism—Himyar in southern Arabia in the fifth century C.E. and Khazaria in southern Russia in the first half of the eighth century C.E.—there are numerous indications of nonrabbinic conversions to Judaism of Berber tribes in North Africa in the pre-Islamic period. In addition, a systematic rabbinic conversion effort seems to have been directed at slaves owned by Jews. This was also practical, since rabbinic law banned non-Jews from handling food in Jewish households. Rabbinic law encouraged circumcision and ritual immersion of slaves, i.e., formal conversion and liberation,2 2 and strongly discouraged their resale. When freed, these people became full-fledged Jews (Rosenbloom 1978:80). There are further indications that sexual 22

According to rabbinic law, if a non-Jewish slave is converted, s/he becomes free, because Jews are not allowed to hold Jewish slaves.

14

THE JEWISH LINGUISTIC SPECTRUM relations between slaveholders and slaves were not uncommon, and that these were sometimes resolved through full conversion and liberation of the slave, so that the resulting offspring were considered Jews.2 3 In the view of Ben Zion Wacholder (1956:106), there is sufficient documentation to establish that many if not all of the Judaized slaves were finally emancipated and absorbed by the Jewish population. A reasonable guess might be that between the seventh and the eleventh centuries Middle Eastern and North African Jewry doubled as a result of the proselyting of slaves.

In the light of combined slave absorption and mass conversion, the rise of Judeo-Arabic on three continents as a giant among Jewish varieties is hardly surprising, with many speakers of Judeo-Arabic being of converted background. On a smaller scale, also through possible conversion, Judeo-Berber emerged in North Africa in the same period and earlier. Elsewhere, Jewish memorial books from the Crusade massacres in early medieval Germany, the cradle of Yiddish, contain significant numbers of converts to Judaism in their lists of victims, although in this case no documentary evidence has so far been found for mass conversion or systematic proselytizing. While legal restrictions and bans on conversion to Judaism had been imposed from the outside much earlier, the path to conversion in the Old World was effectively blocked only after Christian and Muslim control over vast populations of the recently converted solidified in the Middle Ages and the Renaissance. Jewish religious law acknowledged this development through a more restrictive interpretation of proselytism, such as in the Shul˙an >Arukh (1565). By then, major Jewish varieties with written records had already become established.

23

As convincingly stated in Rabinowitz 1971:1187, “Sometimes Jews became over-intimate with women slaves and had them undergo ritual immersion for the purpose of proselytism; their children were regarded as full-fledged proselytes. The best known of these cases concerns the Exilarch Bustanai b. Óaninai.”

15

CHAPTER ONE Considering the permeable or even blurred boundaries between Jews and non-Jews in many periods including the present, it would be rather unrealistic not to expect any linguistic input from proselytes. Benor (2004) has also pointed out in her research on newly-Orthodox Jews in the United States that the linguistic impact of the newcomers on Jewish English may be above average due to their possible tendency to “hyperaccommodate.” In fact, it is quite safe to assume that many Jewish religiolects emerged through a combination of two historical processes: on the one hand, a considerable amount of migration and subsequent integration of Jews, including adoption of preexisting languages outside core Jewish areas; and, on the other hand, a massive influx into many Jewish communities of non-Jews, who brought with them their own languages and enriched them with their newly-won religious and educational heritage.2 4 This interplay of acculturation and reculturation of migrants and locals, of mutual integration and isolation of Jewish and “nonJewish” communities, and of transference and replacement of language elements appears to be mirrored in the complex structure and eclectic nature of many Jewish religiolects. Crossing Religious Boundaries In recent years there has been a surge in interest in Jewish linguistics in general and in the definition of Jewish languages in particular. Benor has recently proposed that Jewish languages be considered not necessarily in terms of distinct systems, but rather as consisting of a “distinctively Jewish repertoire” from which Jews choose when they use their variety (Benor 2009). In other words, what has so far been considered the field of Jewish linguistics is really only the tip of the iceberg. First of all, the concept of religion-based varieties needs to be 24

For a present-day example, see Benor (2004). She describes many factors in the development of “Orthodox” Jewish English, one of which is the contribution of newly-Orthodox Jews. In other words, not just FFB (“Frum from birth”) Jews helped in the creation of Jewish English, but also Jews who became Orthodox.

16

THE JEWISH LINGUISTIC SPECTRUM extended so as to encompass varieties used by adherents of other faiths as well. For example, just as we can identify Jewish-defined languages, we could also recognize Christian- or Muslim-defined languages.2 5 Second, the term language should be broadened to include varieties that are not clearly distinguished from the majority languages.2 6 Finally, the participation of non-Jews in the Jewish linguistic spectrum, and religious crossover in general, should also be taken into account. This last point is especially relevant to the discussion here. There are indeed cases where Christians or Muslims have entered into the Jewish linguistic spectrum. For example, American non-Jews may use some Jewish English elements, mostly in the lexicon, especially if they live in a city with a sizeable Jewish population, like New York or Los Angeles. Thus, it is not unusual to encounter a Catholic Italian American asking a porter to schlep ‘carry’ her suitcase. This phenomenon is by no means new, as the following examples indicate: • As far back as the Middle Ages, Hebrew and Aramaic lexical items entered some Christian-German dialects in the Rhine Valley via Yiddish dialects and has survive until today. For instance, in Hessonian dialects Schmiere stehen ‘to keep a lookout’ comes from Hebrew ‰¯ÈÓ˘ /¡mira/ ‘guard’; schäckern ‘to flirt’ comes from Hebrew ¯ÂÎÈ˘ /¡ikor/ ‘drunken’; and Ganove ‘a thief’ comes from Hebrew ·‚ /ganav/, with the same meaning, spread beyond local dialects and even into standard spoken German. • Christian and Muslim craftsmen borrowed professional terminology from their Jewish colleagues in their respective trade jargons and even argot. For example, Primo Levi reports the use of Judeo-Italian elements in Northern Italy among Christian furriers. In Cairo and in Alexandria, Christian and Muslim goldsmiths still use an argot they think of as “Hebrew” or “Jewish,” which contains Hebrew and 25 26

See Norman Stillman’s work (1991) on Jewish and Muslim languages. See the reference above to Judeo-Dutch and Benor’s proposal for a “distinctively Jewish repertoire.”

17

CHAPTER ONE Aramaic elements. For example, the use of the word [¡a>>ål] ‘a thief’ seems to be derived from Aramaic [¡qal] ‘take.’ • Judeo-Persian used in Isfahan is easily distinguished from the Persian used by Muslims in that city (Rabin et al. 1979:53). However, Muslim Iranians in some Iranian villages, such as Sede and others, use the same Judeo-Persian employed in Isfahan only by Jews (ibid., 56). • According to reports from early modern Saloniki, non-Jews, especially those who worked in the city’s harbor, employed JudeoSpanish as their daily speech. The historical background of this linguistic phenomenon is the demographic prevalence of Jews in the Saloniki port, which was so pronounced that there were periods when it was closed on Shabbat. • In Ruthenia (now western Ukraine) it was common for Orthodox Christian nannies to learn Yiddish and use it to communicate with the Jewish families for whom they worked. They would also teach Jewish children the Hebrew prayers, while Hebrew blessings were widespread among the general Orthodox Christian population.2 7 • The greatest challenge to traditional definitions of Jewish languages is the case of Modern Hebrew as used in Israel today. The majority of Israel’s non-Jews, over a fifth of the country’s citizens, are to varying degrees bilingual, usually Hebrew-Arabic and sometimes Hebrew-Russian. In spite of popular misperceptions, the linguistic community of Hebrew in Israel is no longer defined by religion, but, for the most part, by citizenship or residence. Since, as we have seen, Christians or Muslims may enter into the Jewish linguistic spectrum, 2 8 or they might use some of what started out as distinctively Jewish features, it would certainly be preferable to use the term “Jewish-defined languages,” i.e., languages that were 27

See Erez 1959:231–44, 249–52; Sole 1959:149; and Wein 2007, the section on Christians and Jews in Ruthenia. 28 This is in contrast to M. Zand who claims that “a Jewish language … serves … only the Jewish population of that area” (Rabin et al. 1979:55).

18

THE JEWISH LINGUISTIC SPECTRUM defined by Jews, often standardized by Jewish Bible translations or the Hebrew Bible itself, but not all of whose elements were used exclusively by them. Of course, there is also a need for a discourse on Christian- as well as Muslim-defined languages.2 9 The Characteristics of Jewish Varieties It is generally acknowledged that Jewish varieties share a number of traits. The prototypical Jewish religiolect possesses certain features, ranging from script and grammatical structure to a specific tradition of translating sacred texts. Of course, not every Jewish religiolect needs to have all the features in order to qualify, as the existence of the Jewish linguistic spectrum makes clear. The first and most conspicuous of these features is the consistent use of Hebrew characters in writing.3 0 Jews almost invariably adopted the spelling conventions of Talmudic orthography, including the use of word-final letter forms and the occasional marking of vowel sounds, using available (consonantal) letters and/or other signs. Thus, the Hebrew script in this case symbolizes the Jewishness of the community. It is, in fact, not uncommon to find the script used as a mark of the religious affiliation of the users of a language, as with the Arabic script which is used for writing Aljamiado,3 1 (Muslim) Chinese, Jawi (Malay), M åppil¢l¢a-Malayalam,3 2 Persian, Olttoman Turkish, and Urdu, for

29

For a treatment of the issue, see Hary-Wein 2008. Of course there are some exceptions. For example, Schwarzwald notes that most of the Judeo-Spanish literature in the Ottoman Empire in the last few hundred years was written in Hebrew characters (2001:82); however, Ottoman Judeo-Spanish was written in Hebrew characters only until the 1920s, when Attaturk legislated the use of Latin characters for Turkish. 31 López-Morillas has writen that the Arabic script in Aljamiado (a Spanish variety used by the Moriscos in the sixteenth century) and in other Muslim languages, as well as the Hebrew script in Jewish languages, became “an explicit emblem for the religious and cultural cohesion of the linguistic group” (1994:17). 32 This is a castelect of Muslims in North Malabar. It is written in Arabic script with some orthographic adjustments to the phonetic system of Malayalam. 30

19

CHAPTER ONE example—all languages used by predominantly Muslim language communities. Similarly, the Cyrillic script of Serbian symbolizes the importance of the Eastern Orthodox Church in that language community, whereas Croatian, although practically identical to Serbian, at least until the recent political developments, is written with the Latin script, in line with the Roman Catholic background of most of its users.3 3 The second trait to be found among many Jewish varieties is the use, sometimes simultaneously, of different traditions of orthography. Such competition among various orthographic systems is typical of a situation in which the choice among linguistic systems transmits implicit political, cultural, and religious messages. This can be seen, for example, in the Soviet spelling reform of Standard Yiddish. In contrast to the decision to use the traditional spelling of Hebrew/Aramaic-derived words taken at the Czernovitz conference, in the USSR there was an attempt to dissociate Yiddish from its religious roots (among others) by abolishing this orthographic tradition. Thus, the name of the Yiddish Communist newspaper and publishing house Emes ‘The Truth’ (a literal translation of Russian /pravda/) was spelled phonetically ÒÚÓÚ (/emes/) rather than traditionally ˙Ó‡ (/emet/), like the Hebrew word from which it is derived.3 4 An example from Judeo-Arabic is the historical competition among the Phonetic, the Arabicized, and the Hebraized orthographies (Hary 1996c). The tension between the latter two types of orthography from the fifteenth century onwards reflected the changing dynamics of interreligious relations, including the changing proximity between Jews and Muslims. In other words, the emergence of a Hebraized orthography in Late Judeo-Arabic was driven, among 33

I do not wish to imply here that people actually “choose” a script to fit their need for religious identification. While this may sound democratic, it is probably quite unhistorical. However, whenever rulers imposed a religion on a country, the religious authorities would usually be given responsibility for education, and they would of course impose their standards, including the script, on that country. 34 This does not mean that /emet/ was ever pronounced by Yiddish speakers. See Hary 1992:112–13 for further examples.

20

THE JEWISH LINGUISTIC SPECTRUM other things, by an increasing fragmentation of society along religious lines, evident also in Christian-dominated countries in the same period (Israel 1989). Similarly, the switch from the Phonetic to the Arabicized orthography in the tenth century and the subsequent disappearance of the former (Blau and Hopkins 1984:13–15; Hary 1996c:737) may hint at increased literacy. It is clear that the Phonetic orthography reflected a culture that was centered on oral rather than written transmission. This kind of competition between various systems is sometimes characteristic of varieties lacking standard forms. A third typical feature of Jewish religiolects is the incorporation of Hebrew and Aramaic elements. These are found not only in the religious and cultural sphere, but in the entire lexicon, as well as in the phonology, morphology, and syntax. For example, in Later Judeo-Arabic some authors use /ilå/ ‘to’ as a marker for the definite direct object, in imitation of the Hebrew accusative marker ˙‡ /et/ in Hebrew (Hary 1991b), something that is not found in non-Jewish Arabic varieties. Furthermore, in the Judeo-Arabic dialect of Peq•>•n the Hebrew root <-t-t ‘signal’ is used in an Arabic verbal pattern, /bi<áttit/ ‘(he) sends signals’ (Geva-Kleinberger 2005:50). Similarly, in Jewish English, Hebrew words such as ‰ÎÏ‰ ‘Jewish law,’ ¯˘Î ‘kosher,’ and ˘¯„ ‘drash’ (a biblical interpretation) take the English morphemes -ic, -ally, -ed, and -ing to create the following respectively: halakhically ‘as far as Jewish law is concerned,’ non-hekhshered ‘(food) without a rabbinic seal of kashrut,’ kashering ‘rendering (vessels and kitchen surfaces) kosher,’ and drashing ‘presenting a (biblical) interpretation’ (Benor 2004; 2009). In addition, Judeo-Italian speakers and writers insert Hebrew roots into Italian paradigms, such as /paxad/ ‘be afraid,’ /paxadoso/ ‘timid,’ and /impaxadito/ ‘scared,’ using the Hebrew root p-˙-d for ‘be afraid.’ Another example from Judeo-Italian is the verb /gannavyare/ ‘to steal,’ based on the Hebrew root g-n-v ‘steal,’ as in the sentence guarda che non gannavi ‘watch that she does not steal

21

CHAPTER ONE (from you).’ In Jewish Malayalam3 5 as well, speakers use Hebrew lexemes with Malayalam forms: /sår2a ppe*††u/ ‘suffered, got into trouble’ consists of /sår2a/ ‘trouble,’ taken from Hebrew ‰¯ˆ and followed by /pe*††u/ (past of /pe*†-/); /s!ålomåyi/ ‘died’ includes /s!ålom/ ‘peace,’ taken from Hebrew ÌÂÏ˘ and followed by /åyi/ ‘to be’ (past of /åk-/). Finally, in Judeo-Spanish, the Hebrew roots ¡-˙-d ‘bribe,’ k-f-r ‘deny, be heretic,’ and d-r-¡ ‘interpret, expound’ take Spanish patterns to form the following Judeo-Spanish verbs: ¯‡„ÁÂ˘ /¡ohadear/ ‘to bribe,’ ¯‡¯Ù‡˜ /kafrar/ ‘to deny the existence of God’, and ¯‡Ò¯‡„Ø¯‡˘¯‡„ /dar¡ar/ or /darsar/ ‘to interpret’ (Hary 1999:74 n. 17). The fourth trait is that some Jewish varieties have developed a distinct spoken form, one which is mostly “unintelligible” 3 6 to people outside the community (written Jewish languages are obviously unreadable to most non-Jews, if only because of the use of the Hebrew script). Baghdadi Judeo-Arabic (Blanc 1964) is one example. The fifth trait typical of most Jewish varieties is that their literature is written by Jews for a Jewish readership and usually deals with Jewish topics. However, it has happened that non-Jewish epics as well as other works have been translated or adapted in several Jewish varieties. Furthermore, this trait is not as exclusive as had been assumed in the scholarly literature (Blau 1999:49), considering the issue of crossing religious boundaries discussed above (pp. 16–19). Sixth, a migrated or displaced dialectalism has developed in many Jewish varieties. In other words, Jewish varieties in a certain region sometimes feature dialectal characteristics that are uncommon in that region. This is usually due to Jewish migration and dispersion. For example, in Cairene Judeo-Arabic one can encounter the forms /niktibniktíbu/ for the first person singular – first person plural imperfect, otherwise typically found in “western” Arabic dialects (Fischer and

35 36

Some examples from Jewish Malayalam are taken from Gamliel 2008. See above, p. 10–11, on the issue of intelligibility.

22

THE JEWISH LINGUISTIC SPECTRUM Jastrow 1980).3 7 One would not expect to find these forms in Cairo; their appearance among Cairene Jews is probably due to Jewish migration from Morocco or Alexandria to Cairo.3 8 Another example of migrated or displaced dialectalism can be found in Judeo-Italian. In the southern Italian dialects (Gyoto-Italian) one finds the form /li donni/ ‘the women’ (however rare) instead of the standard /le donne/. In addition, a typical characteristic of central Italian dialects is a system of seven vowels. The combination of these two regional features can only be found in Judeo-Italian, suggesting a synthesis of dialectal elements from different regions due to migration among the Jewish communities in Italy. A seventh feature of many Jewish-defined languages is that they preserve archaic forms which have become extinct in the respective majority languages. For example, in Egyptian Judeo-Arabic the verbal pattern /fu>ul/ (chapter 4, p. 117, 2.2.1.1) has survived, as opposed to /fi>il/, which has replaced it in the modern Egyptian dialect. In addition, the interrogative pronouns /∑¡/ ‘what,’ /l∑¡/ ‘why,’ and /k∑f/ ‘how’ of Cairene Judeo-Arabic have survived in the sentence-initial position (pp. 114–16, 2.1.6), in contrast to the situation in the standard dialect, where other pronouns, /∑h/ ‘what,’ /l∑h/ ‘why,’ and /ezzåy/ ‘how’ appear at the end of the sentence. Furthermore, the demonstrative pronoun /de/ ‘this (masc.),’ an older Cairene form, survived among Jews through the twentieth century.3 9 Jewish Malayalam also possesses many archaic forms, the most striking of which is the dative ending 37

See Hary 1992:278 and the references there. See also chapter 4, p. 118, 2.2.2, for the clarification of the term “western” dialects. 38 On the other hand, as this form exists not only among Cairene Jews, but also in the west Delta (Behnstedt 1978:69), it is perhaps not of Maghrebi origin, but may have developed independently in Egypt. However, the existence of /niktib-niktíbu/ in the west Delta does not preclude the idea of migrated or displaced dialectalism as advanced above. See other examples of migrated dialectalism in Judeo-Arabic in chapter 4, p. 103, 1.1.6; p. 105, 1.6; p. 110, 1.13.4; and p. 114, 2.1.4.3. 39 Blanc 1974:216; Rosenbaum 2002c:126; see also pp. 113–14, 2.1.4.1.

23

CHAPTER ONE /-ikku*/, instead of /-u*/, for nouns and pronouns ending in /-an/, e.g., /j•vanikku*/ ‘for life’ (instead of /j•vanu*/) and /avanikku*/ ‘for him (third person singular with dative ending)’ (instead of /avanu*/).4 0 Preserving archaic forms is also typical of migration languages. For example, French Quebec preserves archaic forms that are not used in Parisian French. Similarly, “western” Arabic dialects preserve forms that are common in the Koran but are not found in “eastern” dialects, for example, the word /˙¥t/ ‘fish,’ also found in the Koran, versus its “eastern” equivalent /samak/.4 1 In this respect it is only fitting that Judeo-Spanish and Yiddish, two Jewish religiolects with a rich history of migration, use many archaisms. The former preserves the archaic Old Spanish phonemes /¡/ and /d≈/, as opposed to /x/ for both in modern Spanish. Yiddish has kept the archaic word hait, which has disappeared as an independent word from German, surviving only as a suffix, e.g., Kindheit (Birnbaum 1979:10). Eighth, Jewish speakers have usually considered their varieties as separate from the majority languages and have given them special names, such as /il-lu\a dyalna/ ‘our language’ or /il->arabiyya dyalna/ ‘our Arabic’ in Moroccan Judeo-Arabic, whereas Muslim Moroccan Arabic is termed /il->arabiyya dilmsilm•n/ ‘the Arabic of the Muslims’ (Bar-Asher 1988; Stillman 1988). In Jewish Malayalam, speakers call their variety /malayalam ¡elanu/ ‘our Malayalam’ and distinguish it from other varieties of Malayalam, /ze lo malayalam ¡elanu/ ‘this is not our Malayalam.’ 4 2 Furthermore, Kerala Jews also refer to their language variety as malbarit or cochinit.

40

According to Ophira Gamliel, this archaic form is represented even today in the speech of Kerala Jews in Israel. Ayyar states that (1993:27–28) the dative /-ukku*/ after /-an/ was alternating with /u*/ in the earliest Malayalam inscriptions (note that the phonemes /u/ and /i/ alternate in Malayalam). There is only one text that has this archaic ending, the Råmacaritam from the thirteenth century. 41 For more examples see Shin>ar in Rabin et al. 1979:56–57. 42 They say it in Hebrew. In fact, “our Malayalam” is probably something that other speakers of castelects in Kerala might use to refer to their variety.

24

THE JEWISH LINGUISTIC SPECTRUM Ninth, the “spirit” of Jewish-defined languages, their reservoir of images, formulations, concepts, and icons, is derived from Jewish sources in Hebrew and Aramaic, usually sacred texts. Tenth, many Jewish religiolects share a unique literary genre, the verbatim translation of sacred religious and liturgical Hebrew/Aramaic texts into the various Jewish religiolects (¡ar˙, pl. ¡ur¥˙, in JudeoArabic;4 3 ¡ar> or ¡ar˙ in Judeo-Neo-Aramaic; tavsili in Judeo-Georgian; tefila in Judeo-Italian; tamsir in Jewish Malayalam; ladino in JudeoSpanish; taytsh in Yiddish; etc.). The translations include the Bible, Midrashic literature, Pirkei Avot (“Ethics of the Patriarchs,” a tractate of moral and religious teachings from the Second Temple period and the first centuries of the Common Era), the Passover Haggadah, the Siddur or prayer book, the Talmud, and more. However, the existence of these typical features of a Jewish variety does not mean that in order to qualify as such a variety needs to possess them all. Whenever a language variety used by Jews differs, even if only slightly, from the majority language, it deserves to be considered as part of the Jewish linguistic spectrum. Jewish varieties are thus best placed on a continuum stretching from those with a high concentration of the most prominent characteristics (Yiddish, for example) to those with only few and marginal traits (varieties of secular Jewish English, for example). Other forms of Jewish linguistic practice are located somewhere between these two poles. A List of Jewish Varieties Jewish varieties are numerous and, as the following list shows, reflect Jewish history and geography. 4 4 Beside Hebrew, the primary Jewishdefined language (although see Ornan 1985), Jewish forms of Aramaic 43 44

See chapter 3 for a detailed analysis of the genre of the ¡ar˙. See the Jewish Languages Research Website (http://www.jewish-languages. org/), edited and designed by Sarah Benor and Tsuguya Sasaki/Tsvi Sadan. This website is an important endeavor toward a new understanding of the Jewish linguistic spectrum.

25

CHAPTER ONE began to develop even before the beginning of the Common Era. Before the end of the Second Temple period, Hellenistic Jews began to employ the Greek Koiné in its Jewish form, Yevanic, which many centuries later in the Balkans came to be known as Judeo-Greek. Judeo-Arabic began to develop in the seventh century C.E., with the spread of Islam in the Middle East and North Africa; Jews from Spain to Iraq adopted forms of Arabic and created Judeo-Arabic varieties. In North Africa, Judeo-Berber (Berberic) emerged, and in Iran, Judeo-Persian (Parsic). In Christian Europe, Latin eventually gave rise to at least six different Jewish religiolects: Judeo-Italian (Italkian) in Italy, JudeoProvençal (Shuadit) in southern France and Judeo-French (Zarphatic) in the north, Judeo-Catalan (Catalanic) in the eastern part of the Iberian Peninsula, Judeo-Portuguese (Portugesic) in the western part, and Judeo-Spanish (Ladino, Jidyó, Judezmo) in between. After the expulsion of the Jews from the Iberian Peninsula toward the end of the fifteenth century, Judeo-Spanish spread east through the Balkans to Turkey and Palestine and south to Morocco and over to some parts of North Africa, in the form of Haketiya (Moroccan JudeoSpanish). Yiddish originated in the tenth century among central European Jews, probably in Southwest Germany, and spread throughout Central and Eastern Europe and into Italy, and centuries later also to the Americas, Australia, Palestine, and South Africa. Before the Holocaust, three-quarters of world Jewry spoke Yiddish. Furthermore, Canaanic (Knaanic, also known as Judeo-Czech and JudeoSlavic) emerged in Slavic-speaking areas, and Judeo-Alsatian (YédischDaïtsch) in Alsace. In the east, Kurdish Jews use Judeo-Neo-Aramaic, Judeo-Arabic dialects, and also Judeo-Kurdish with mixed Hebrew, Turkish, and Arabic elements. In Central Asia, Judeo-Tajik (Bukharic; some speakers call it Farsi) is employed; Judeo-Tat (Judeo-Tatic, or Juhuri, of the Iranian family) is used by Jews in Daghestan in the eastern and northern Caucasus, and Judeo-Georgian (Gurjic) is used by Jews in Georgia in the southern Caucasus. Judeo-Crimchak (of the Turkic family) is employed by Crimean Jews, both Rabbinic and 26

THE JEWISH LINGUISTIC SPECTRUM Karaite. Further to the east, Jewish Malayalam developed, especially among the Jews of Kerala in southern India. Most speakers of these religiolects have emigrated to Israel, France, North America, and elsewhere. Consequently, Jewish religiolects have declined, some have become endangered, and others are now extinct. On the other hand, these varieties are being replaced in recent decades by new Jewish religiolects. In Australia, Canada, South Africa, the United Kingdom, the United States, and elsewhere, some forms of Jewish English have emerged, especially in the last century and particularly among Orthodox Jewish communities. In France a form of Judeo-French has emerged, and in Argentina, Mexico, and other Spanishspeaking places, new varieties of Latin American Judeo-Spanish are being formed. The same holds true for Judeo-Dutch in the Netherlands, for Judeo-Russian in Israel and Russia, and so forth (see p. 8). Three Jewish religiolects hold a special place in Jewish culture, because they have been used both over a wide geographical area and for a long period of time: Judeo-Arabic, Judeo-Spanish, and Yiddish. Among these Judeo-Arabic holds pride of place: it has had the longest recorded history, from the pre-Islamic period to the present; it spans the widest continuous geographical area, from Spain to Yemen and Iraq; and “it was the medium of expression for one of the foremost periods of Jewish cultural and intellectual creativity” (Stillman 1988:3–4).

27

CHAPTER TWO JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM This chapter explores Judeo-Arabic within the framework of other Jewish religiolects. It reviews the history of this religiolect and analyzes its structure in some detail. It then discusses the language continuum employed by users of Judeo-Arabic and traces its diachronic evolution. The chapter also tackles some terminological issues, especially with respect to the denotation of Arabic-speaking Jews. Finally, the chapter discusses the state of Judeo-Arabic today.1 An Overview of Judeo-Arabic Judeo-Arabic is a religiolect (see pp. 12–13) that has been spoken and written in various forms by Jews throughout the Arabic-speaking world. Judeo-Arabic literature deals for the most part with Jewish topics, and is written by Jews for a Jewish readership. Several important features distinguish it from other varieties of Arabic. These include a mixture of elements of Classical and post-Classical Arabic, dialectal components, pseudocorrections, and pseudocorrections that have become standardized. In other words, it is a typical mixed variety. Judeo-Arabic also possesses a number of specific additional sociolinguistic and sociocultural features that set it apart: the use of Hebrew rather than Arabic characters, various traditions of Judeo-Arabic orthography, elements of Hebrew and Aramaic vocabulary and grammar,

1

Some of the material in this chapter was published in different form in Hary 1992 and 2003.

CHAPTER TWO and the style of the ¡ar˙ in Judeo-Arabic texts.2 Judeo-Arabic speakers have been a topic of discussion academically and politically for many years in Israeli society. Many designations for speakers of Judeo-Arabic exist, including Mizrahim (or Á¯ÊÓ‰ ˙Â„Ú, lit. “the communities of the East”), Sephardim (lit. “Spaniards”), and “Arab Jews.” Actually, the term Mizrahim, lit. “Easterners” (translated as “oriental Jews”) is of course a misnomer, since Moroccan Jews, for example, hardly count as being from the east, if the point of reference is Israel.3 The term Sephardim has its own problems. Strictly speaking, it refers to Jews whose ancestors had been expelled from the Iberian Peninsula, up to and especially in 1492, and who then settled in the Ottoman Empire and other countries. Although many Jews of the Ottoman Empire, especially in Arabic-speaking communities, adopted the religious ways and liturgical customs of the expellees from the Iberian Peninsula, pre-Sephardi traditions also survived in many areas, including North Africa. Finally, the term “Arab Jews,” attested historically in various documents but now used only sporadically, may be misleading because the word “Arab” could be perceived as an “ethnic” marker. This leads to three unresolved issues: (i) The word “Arab” as an “ethnic” marker in the current Israeli sense did not exist historically or sociologically before the creation of modern Israel, so Arabic-speaking Jews in the past were conceptualizing something entirely different when designating themselves as “Arab Jews.” (ii) The concept of “ethnicity” itself remains unclear in most contexts, 2

3

The ¡ar˙ is a genre composed of literal translations of Jewish religious sacred texts from Hebrew into Judeo-Arabic. The reference here is to the style of this genre, characterized by Hebrew and/or Aramaic interference. Another term for this style is “Hebraism.” An imaginary line drawn diagonally across the Mediterranean, from the Strait of Gibraltar to the Black Sea, has historically distinguished the Jewish “west” (in fact, north) from the Jewish “east” (in fact, south). This raises a number of questions, such as: Who set this imaginary line? Who used it? For what purposes?

30

JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM the Israeli case included; it is therefore best avoided in academic discourse, unlike the concepts of language or religion, which can be measured and marked more easily. (iii) The term “Arab Jews” bears controversial political connotations in Israel. For example, it may suggest a connection between “Arab Jews” and “Arab Israelis,” whose identity constructions seem similar on the surface, but in fact differ profoundly on various levels. For example, Arab Israelis in general feel less connected to the State of Israel than “Arab Jews.” In addition, many “Arab Jews” object to the term, sometimes strongly, because of the current Arab-Israeli conflict, among other reasons. Although some Israeli intellectuals today refer to themselves as “Arab Jews,” 4 they are probably quite aware that their use of the term with its current connotations is rather remote from the way it may have been used by Jews in premodern Egypt, for example, where the political context was significantly different. Although in the past I have used the term “Jews of Arab lands,” this designation in retrospect may not be the most appropriate. The expression associates “lands” with a nationality, since the term “Arab” may be used to refer to a specific (pan-)nationalism.5 Thus, the use of this term would seem to establish a link, in the Romantic sense, 4

5

For example, members of the “Mizrahi Democratic Rainbow.” Professor Sasson Somekh of Tel Aviv University, an expert on the writings of the Egyptian author Nag•b Ma˙f¥z, spoke out at an “Iraqis conference” held at Tel Aviv University in May 2008 against this term, which he claimed was being used for political ends and/or in order to follow current trends. He defined an “Arab Jew” as a person born in an Arabic-speaking Jewish home, who lived in an Arabic-speaking Jewish community in an Arab Muslim environment, and was competent in literary Arabic, the basis of Arab culture. Indeed, I heard Professor Somekh identify himself as an “Arab Jew” at one of the Middle Eastern Studies Association meetings a few years ago. The term “Arab” need not necessarily be identified with nationalism. A Syrian, for example, may ask herself, “Am I a Syrian or an Arab?” However, there have been attempts to demonstrate that the term “Arab” in the context of pan-nationalism encompasses all Arabic-speaking nations.

31

CHAPTER TWO between just one population group and a specific territory. Such a link is factually inaccurate, since many minorities—Jews, Kurds, Berbers, and others—who live in the “Arab lands” have their own national movements. Control of a given territory by a certain population is thus a historical and not a geographical fact; i.e., there is no “natural link” between human population groups and specific territories. Nationalism in the Middle East developed mainly in the twentieth century. Consequently, the terms “Arab Jews” or “Arabic-speaking Jews,” as historical and cultural designations, are best avoided in reference to any time before the end of the nineteenth century. Afterwards the terms become ambiguous, especially “Arab Jews,” unless one specifically stipulates that the word “Arab” is not being employed in the more recent sense of nationality. Today, such “Arab Jews” are in reality almost exclusively multilingual Israeli, French, or North American nationals who for the most part do not hold any “Arab” citizenship (except for some Moroccan Jews). When referring to the time period from the beginning of the twentieth century to the present, the term “Jews of Arabic-speaking backgrounds” is thus more suitable. For premodern times the term “Arabic-speaking Jews” is fitting as well, and is therefore used extensively in this volume. The two latter terms would probably also be acceptable to more people than the term “Arab Jews.”6 The History of Judeo-Arabic At two points in its history, Judeo-Arabic underwent dramatic changes in its structure and use. The first change occurred during the fifteenth century, when the Jewish world reduced its contact with its Arab 6

This topic deserves further investigation. The following questions could be posed to subjects and then analyzed: What does the term “Arab Jews” mean to you? What does it evoke? What does the Arab part evoke and what does the Jews part evoke? Provide ten associations when you hear the term “Arab Jews,” etc. For a recent discussion of some of these issues, see Gottreich 2008 and Levy 2008.

32

JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM counterpart. Although a great number of Jews settled in the Ottoman Empire after the expulsion of the Jews from Spain in 1492, and in some ways experienced even more intense contact with the Muslim world, many curtailed their contacts with Arabs, their language, and their culture. Jews felt the need for more separation from their Muslim (and Christian) neighbors and began to congregate in exclusively Jewish neighborhoods (sometimes with active encouragement by the authorities) such as ˙art il-yah¥d (in Egypt) or mEllå˙ (in North Africa). This change was especially marked in some areas like North Africa, but less so in others like Yemen, where close contacts between Jews and non-Jews persisted for some time. As a result, Judeo-Arabic did not develop along the same lines everywhere. Because of the change in contact between the cultures in the fifteenth century, not only did the structure of Literary Written Judeo-Arabic (Hary 1992:79) come to incorporate more dialectal elements, but also more works were written in Hebrew. In fact, Hebrew, Arabic, and Judeo-Arabic were sometimes assigned different usage functions (Drory 1992, 2000). In the twentieth century this religiolect again experienced a dramatic change with the rise of Jewish and Arab national movements, the outbreak of the Arab-Israeli conflict, and the consequent emigration of Jews from Arabic-speaking areas. These changes brought about the loss or near loss of the religiolect. The changes of both the fifteenth and the twentieth centuries brought about an increased use of dialectal elements in Judeo-Arabic texts. However, the changes in the fifteenth century were unique because they featured the development of the Hebraized orthography (Hary 1996c), characterized by, among other things, greater Hebrew/Aramaic influence on Judeo-Arabic spelling. Both changes were intimately connected to the decreased contact between Jews and their Arab neighbors, which led to a somewhat more insular Jewish existence in the Arabic-speaking areas.

33

CHAPTER TWO For purposes of discussion and analysis it is convenient to divide the history of the religiolect into the following periods:7 Pre-Islamic Judeo-Arabic, Early Judeo-Arabic (eighth/ninth to tenth centuries), Classical Judeo-Arabic (tenth to fifteenth centuries), Later Judeo-Arabic (fifteenth to nineteenth centuries), and Contemporary Judeo-Arabic (twentieth century). This periodization, however, should not draw attention away from the major changes that occurred in the fifteenth and then in the twentieth century, as represented in figure 1: Judeo-Arabic

Medieval

Late

Modern

Pre-Islamic > Early > Classical > Later > Contemporary (Change I) (Change II) Figure 1. Periodization of Judeo-Arabic

There is a linear connection between medieval, late, and modern Judeo-Arabic.8 In other words, although the religiolect experienced two dramatic changes in its development, one in the fifteenth and the other in the twentieth century, it can still be divided into successive periods, each of which was influenced by its predecessor. The following paragraphs shed some light on each of the periods. There is some evidence that the Jews in the Arabian Peninsula during the pre-Islamic period used a type of Arabic Jewish dialect 7

8

I have offered a periodization of Judeo-Arabic elsewhere: Hary 1992:78, 1995:74–77, 1996c:730, 1997b:200–203, 2003:52–53; Elqayam and Hary 1997: 111–12; and more. In the periodization here, unlike in the previous instances, I am taking into account the two dramatic changes that occurred in this religiolect. The dialectologist Haim Blanc was the first to point me in this direction. In his works on spoken Egyptian Judeo-Arabic (1974, 1981, 1985) he demonstrated a linear connection between the different periods of Judeo-Arabic in Egypt.

34

JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM called al-Yah¥diyya (Newby 1971, 1988:21–23; Gil 1984:206). This dialect was similar to the Arabic dialect used by the majority, but included some Hebrew and Aramaic lexemes, especially in the domains of religion and culture. Some of these Hebrew and Aramaic words passed into the speech and writing of the Arabs. This explains why words of Hebrew and Aramaic origin appear in the Quran. There is no evidence, however, that Pre-Islamic Judeo-Arabic ever served as the vehicle of a distinct literature. For example, the poetry of the Jewish poet as-SamawÅdiyå< did not differ from that of his Arab contemporaries and, in fact, it constitutes part of the canon of Arabic literature, and not of Jewish literature. Were it not for Arab sources which report that he was Jewish, this fact would probably have remained unknown. In other words, as-SamawÅdiyå< was an Arab poet who happened to be Jewish.9 Yet there may also have existed alYah¥diyya writings in Hebrew characters (Newby 1971:220).1 0 After the conquests of early Islam, the Jews in the newly-conquered lands adopted the conquerors’ language. They began to incorporate Arabic into their writing and gradually developed their own religiolect. The second period of Judeo-Arabic began during the ninth century, and in Egypt already in the eighth century. This was the main period in which the Judeo-Arabic Phonetic orthography was used, though alongside the Arabicized orthography. Since the Phonetic orthography was phonetically based (Blau and Hopkins 1987:124–25; Hary 1996c) it did not imitate the orthography of Classical Arabic. Therefore, in Early Judeo-Arabic only scribes who were educated in Classical Arabic and wrote for readers versed in it used the Arabicized orthography, 9

See Snir 2005:488–91. He has claimed that as-Samaw
35

CHAPTER TWO which was based on the mechanical transfer of Classical Arabic spelling into Hebrew characters. The appearance of Saadia ibn Yosef al-Fayy¥m•’s (882–942 C.E.) translation of the Pentateuch into Judeo-Arabic marks the beginning of the third period, Classical Judeo-Arabic. Although the written form of this language contained dialectal features as well as pseudocorrections, it tended to follow the model of Classical Arabic to a large extent (Blau 1980, 1981). The works written in this period covered the entire spectrum of literary composition: theology, philosophy, biblical exegesis, philology, grammar and lexicography, law, ritual, and literature, in addition to commercial and private correspondence. Furthermore, the number of such works in this period exceeded the number of Judeo-Arabic works of any other single period. The fourth period, Later Judeo-Arabic, reflects the beginning of the first dramatic change in the history of Judeo-Arabic, as stated above. This period lasted from the fifteenth to the nineteenth century. The shift from Classical to Later Judeo-Arabic was accompanied by “the increased social isolation of the Jews of the Arab world at the end of the Middle Ages within restrictive quarters, such as the mEllå˙ and ˙årat il-yah¥d” (Stillman 1988:5). During this period many more dialectal elements penetrated into the written language, and the tradition of the ¡ar˙—that is, the literal translation of Hebrew and Aramaic religious sacred texts into Judeo-Arabic—developed. Historical, halakhic, liturgical and other texts were written in this period, many of them aimed at the general public rather than the erudite elite. Toward the end of this period, and even more so in the following period, an extensive folk literature also came into being. This period, too, witnessed the continued use of the Hebraized orthography (Hary 1996c), i.e., Judeo-Arabic written with spelling conventions that were relatively heavily influenced by Hebrew and Aramaic. It was also at the beginning of this period that Jewish scholars began to write in Hebrew; by the end of the period Hebrew had become the preferred written language. Yemen was an exception in this development, because its Jewish 36

JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM community was more isolated. The literary language of the third period, Classical Judeo-Arabic, continued to be used there well past the fifteenth century. The emergence of the literary language of the fifth period marks the beginning of the second dramatic change in the history of the religiolect. Contemporary Judeo-Arabic of the twentieth century is characterized by greater production of ßur¥˙, folktales, and other types of popular literature.1 1 In this period the texts are characterized by more dialectal components than in previous periods and exhibit local elements taken from the spoken variety. However, North Africans had already begun to use their local dialect in writing during earlier periods. As a result, Jewish readers from other Arabic-speaking areas found Maghrebi texts difficult if not impossible to understand. Furthermore, beginning in the previous period and continuing into this period, several dialectal centers developed and flourished among Arabic-speaking Jews. Thus, there arose Maghrebi Judeo-Arabic, Egyptian Judeo-Arabic, Syrian Judeo-Arabic, Iraqi Judeo-Arabic, and Yemenite Judeo-Arabic, each with its own local flavor. The Structure of Judeo-Arabic Because it is the meeting point of Classical Arabic, Arabic dialects, Hebrew, and Aramaic, Judeo-Arabic exists in numerous mixed forms. As a result, one feature of Literary Written Judeo-Arabic is that it contains, among other elements, many colloquial characteristics. Figure 2 illustrates the continuum in Judeo-Arabic. Note that the dramatic changes in Judeo-Arabic that occurred during the fifteenth and the twentieth centuries resulted in a shift in the nature of the continuglossia,1 2 so that more and more dialectal elements penetrated 11

For a useful list of works in contemporary Judeo-Arabic, see Corré 1989. A list of Iraqi Judeo-Arabic folk literature, including poetry, can be found in Avishur 1979. 12 The term continuuglossia was introduced in Hary 2003 with the use of two u’s. Here I use continuglossia with one u to reflect the Latin origin more

37

CHAPTER TWO the writings composed in this religiolect. This had the effect of reducing the gap between the left and the right poles of the continuum. JUDEO-ARABIC Hebrew/Aramaic

Hebrew/Aramaic

Literary Written Judeo-Arabic (Varieties Bn )

Dialectal Spoken JA (Varieties C)

Standard Arabic (Variety A)

Arabic Dialects

Figure 2. The Judeo-Arabic continuum

At the right end of the Judeo-Arabic continuum one finds Dialectal Spoken Judeo-Arabic (Hary 1992:79). The left side of the Arabic continuum containing standard Arabic (the acrolect) is not found in a fully-developed form in Literary Written Judeo-Arabic; however, it is a source of style-shifting which many authors attempted to use, with mixed success. In other words, the language of Judeo-Arabic authors only approached standard Arabic. Had they written in a language that was too much like standard Arabic, their writings would have lost their distinctive identity, and would not have been considered JudeoArabic. On the other hand, standard Arabic is still the anchor for the properly (I thank Michiel Klein Swormink, personal communication). It is largely meant to replace the term diglossia (Ferguson 1959) by emphasizing that a continuum describes the situation better than a dichotomy. In the case of Arabic, rather than stressing a contrast between standard and colloquial Arabic, the proposed term refers to a continuum on which the Arabic varieties are located.

38

JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM left side of the Judeo-Arabic continuum. Thus, it is clear that the Jews, a minority language community, defined themselves linguistically according to the values of the Arabs, the dominant majority. The Judeo-Arabic Continuum Many of the linguistic characteristics of the various Judeo-Arabic dialects (Varieties C) throughout their history can be identified by means of a careful analysis of Judeo-Arabic texts (Varieties Bn ).1 3 By identifying and setting aside the elements of Classical Arabic, pseudocorrections, and the style of the ¡ar˙ in a written text, the student of Judeo-Arabic can isolate dialectal elements derived from colloquial Judeo-Arabic. Such an analysis should be done by comparison to the modern dialects.1 4 The influence of standard Arabic is particularly evident in the area of pseudocorrections. The reason for this is that Judeo-Arabic authors at times attempted to write in the more prestigious standard Arabic, with varying degrees of success and with occasional pseudocorrections. A number of Judeo-Arabic authors did master standard Arabic and wrote in it. Their writings in standard Arabic, however, cannot be considered Judeo-Arabic, and thus lie outside the scope of this religiolect. Maimonides (1135–1204) can serve as a good example of Classical Judeo-Arabic. He was certainly also capable of writing in standard Arabic (Variety A1 5), and indeed did so; without a doubt, he was able to switch between the different varieties of the language, adapting his writing to his readership. As a result, some of his works, such as his medical writings, which were aimed at Christian and Muslim readers, are in standard Arabic (Variety A), not in Classical Judeo-Arabic. In other works, such as his letters to his coreligionists, he used Literary Written Classical Judeo-Arabic (Varieties Bn ). 13

See Hary 1992:11ff and 1996a for detailed explanations of Varieties A, B n, and C. 14 See chapter 4, pp. 93ff, for a further treatment of this methodology. 15 On Variety A and its placement on the Arabic continuum, see Hary 1995:77– 80, 1996a:71–75.

39

CHAPTER TWO Judeo-Arabic writers’ and speakers’ attitudes are important for understanding the religiolect. Since they did not have the same ideal of al->arabiyya ‘the [pure] Arabic’ as their Muslim neighbors, they allowed themselves to rely more on colloquial elements when writing. But on the other hand, they aspired to write in the prestigious standard Arabic (Variety A), which they did not always master. This, in turn, resulted in pseudocorrections, a typical component of Judeo-Arabic (Hary 1992:62–67, 2003). To conclude, Judeo-Arabic, standard Arabic, Arabic dialects, Hebrew, and Aramaic were all part of the linguistic inventory of Jewish society in areas where Arabic was spoken. But standard Arabic, Arabic dialects, Hebrew, and Aramaic were not part of the Judeo-Arabic continuglossia, although they were in close contact with the religiolect and influenced its structure. The main difference between the continuglossic situation of Arabic and that of Judeo-Arabic lies in the functions of the varieties of B n . Whereas in Arabic, Bn —or, as it is sometimes termed, al- lu\a al-wuß†å ‘the intermediate language’—is used both orally (most often in the media) and in writing (mainly in private letters and personal communication, but also in modern prose, dramatic dialogues, and occasionally also in modern poetry), it is not employed as the main variety for literary compositions, as is the case with the varieties of Bn of Judeo-Arabic. Generally speaking, Judeo-Arabic literary texts have been composed in Bn , whereas Arabic literary texts have been composed, for the most part, in Variety A. This last observation should be accepted with some reservation, since in recent years there has been a significant increase in the publication of written colloquial Arabic in Egypt.1 6 However, it is still the case that the majority of literary texts in the rest of the Arab world are composed in Variety A. In our description of the language community of Jews in Arabicspeaking areas, we see how continuglossia is intimately tied to the use 16

See chapter 4, p. 97, and Rosenbaum 2004.

40

JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM of other languages, Hebrew and Aramaic in the present case. This situation is not unique to Judeo-Arabic. It is in fact compatible with what Ferguson has said about Tamil and the effect on it of Sanskrit and English, and about Arabic in some parts of the Arab world where French, English, Syriac, or Coptic also plays a role.1 7 In fact, continuglossia occurs in many other speech communities.1 8 Faroese, for example, is a West Scandinavian language spoken by between forty and fifty thousand people on the Faroe Islands. For centuries Danish was the language of administration, religion, education, and culture on the islands, and Faroese was the spoken vernacular. But from the nineteenth century onward, there has been a systematic attempt to replace Danish with a “pure” form of Faroese, a written version of the language free from the “corrupted” vernacular with its foreign influences. In fact, a conscious effort has been made to create an intralingual continuglossic situation with two opposing poles: a purist written (and later oral) variety and a colloquial variety, where Danish is still in the background.1 9 Faroese is thus not just another example of an intralingual continuglossic situation, but also an interesting case of planned continuglossia. The Diachronic Development of Judeo-Arabic Continuglossia The continuglossic state of Judeo-Arabic has evolved throughout its history.2 0 The situation in Classical Judeo-Arabic was not the same as that in Later Judeo-Arabic. The structure of the latter underwent 17

See the discussion in Ferguson 1959:337. Ferguson alludes to this in ibid., 326. 19 I thank John Thomeson who sent me his manuscript. See Thomeson n.d. 20 I do not, of course, claim that my diachronic analysis of Judeo-Arabic is the only way to understand its history or the history of Arabic in general. Much of the data on this religiolect has been lost, and it is therefore possible to hypothesize other “histories” of Judeo-Arabic that are consistent with the available information. I have attempted here to sketch one possible history of Judeo-Arabic that may be useful for gaining a better understanding of how the history of Arabic has developed. 18

41

CHAPTER TWO several changes due to the fact that its literary varieties exhibited a stronger dialectal base than the literary varieties of previous periods. The relative position of a typical text of Later Judeo-Arabic (Varieties Bn ) would therefore fall more toward the colloquial end or right side of the continuum than would a typical text of Classical Judeo-Arabic. In Contemporary Judeo-Arabic the dialectal base of Bn is again stronger than it was in Later Judeo-Arabic; the relative position of an average text of this variety is shifted even further toward the colloquial end of the continuum. In other words, the “linguistic distance” (Ferguson 1996:57–58) between Bn and C has changed over the centuries: in Classical Judeo-Arabic the distance between the two is larger than in Later Judeo-Arabic, and larger still than in Contemporary Judeo-Arabic. Figure 3 shows the relative position of a typical text in Bn in different periods of Judeo-Arabic:

JUDEO-ARABIC

Varieties Bn 1

Varieties C 2

3

4

1 = Literary Written Classical Judeo-Arabic 2 = Literary Written Later Judeo-Arabic 3 = Literary Written Contemporary Judeo-Arabic 4 = Spoken Dialectal Judeo-Arabic Figure 3. The continuglossic nature of Judeo-Arabic

Figure 3 is a schematic representation of the history of continuglossia in Judeo-Arabic. The exact position of Literary Written Judeo-Arabic (Bn ) can never be fixed, even within a specific period, not only because of the countless lectal possibilities involved, but also because it may 42

JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM shift along the continuum, depending on the nature of the texts involved, their writers and readers, and other variables. The diagram does, however, attempt to show the relative positions of average or typical texts of Bn in the different periods of Judeo-Arabic. Thus, it is clear from the diagram that the number of dialectal elements in Literary Written Contemporary Judeo-Arabic, for example, is greater than in Literary Written Later Judeo-Arabic, because the contemporary variety is generally located closer to the continuum’s colloquial or right end. The same holds true for the relation of Literary Written Later to Classical Judeo-Arabic, in that the dialectal elements in the Later period are more extensive than those of the Classical period. There are several possible explanations for why the dialectal components in Judeo-Arabic became more conspicuous with the passage of time. First, as mentioned above, the two major changes in the history of Judeo-Arabic occurred during the fifteenth century, at the end of Classical Judeo-Arabic and the beginning of Later Judeo-Arabic; and during the twentieth century, at the end of Later Judeo-Arabic and the beginning of Contemporary Judeo-Arabic. During both of these transition periods, more dialectal elements began to appear in the texts. This linguistic situation was coupled with a conscious desire on the part of Jews to distance themselves from Arabic culture and its written expressions. Because of this separation, or perceived separation, Jews may have cared even less for preserving Classical Arabic, and thus may have allowed more dialectal components to enter their writings. In addition, they started to write more in Hebrew. Moreover, in Later Judeo-Arabic a Hebraized orthography began to develop, heavily influenced by Hebrew/Aramaic. In other words, rather than mechanically transferring Arabic letters into Hebrew characters or imitating Classical Arabic spelling as represented in the Arabicized orthography devised in the tenth century, Jews from the fifteenth century onward developed a Hebraized orthography and allowed much greater Hebrew/Aramaic influence on the spelling of Judeo-Arabic. Finally, the increased dialectal components in Later and Contemporary Judeo-Arabic may represent 43

CHAPTER TWO a decline in the level of education in the Muslim world in general and in the Arabic-speaking Jewish world in particular, which started at the end of the Middle Ages, in the fifteenth century. The diachronic development of Judeo-Arabic is comparable in some ways to the history of Maltese. Like Judeo-Arabic, Maltese is used by “non-Muslims,” Christians in this case. More importantly, Maltese, like Judeo-Arabic, is also not written in Arabic characters; it uses the Latin alphabet (chapter 1, pp. 19–20). Maltese, unlike Judeo-Arabic, however, has been more isolated from the main body of Arabic speakers, and thus the vernacular has moved even further away from the center. Consequently, many features of Maltese are further removed from standard Arabic than are those in Judeo-Arabic. The phonemes /∞/ and />/ both disappeared, although both are still reflected in the conservative orthography. In addition, /x/ merged with /˙/ (/˙obz/ ‘bread’; /a˙na/ ‘we’). Third, emphatic phonemes became nonemphatic, although the vowels sometimes indicate where an emphatic phoneme existed earlier. Finally, numerous Italian and Sicilian loanwords have been fully integrated into Maltese, resulting in a major morphological restructuring of the language (Versteegh 1997:209–11). The investigation of marginal, minority religiolects or language varieties such as Judeo-Arabic and Maltese makes it easier to understand the diachronic development of Arabic in general. In fact, such investigations open a small window onto Arabic continuglossia in general and can explain some of its historical developments, as well as the development of Arabic dialects throughout history, since the periphery (in these cases, Judeo-Arabic or Maltese) so often points to the center (in this case, Arabic in general). The Current State of Judeo-Arabic As mentioned in the previous chapter, Judeo-Arabic is one of the more significant Jewish religiolects. However, Yiddish and JudeoSpanish enjoy greater recognition and prestige in both Jewish and “non-Jewish” circles. There are several reasons for this. The dominance 44

JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM of Ashkenazi Jewry throughout the twentieth century in two influential Jewish societies, in the United States and in Palestine/Israel, has advanced the prestige of Yiddish over other Jewish religiolects and varieties. In the United States, a special organization, YIVO (˘È„ÈÈ ËÂËÈËÒÈ‡ ¯ÚÎÈÏË»Ù‡ ‹ ˘ÒÈÂÂ ‘The Institute for Jewish Studies’), was reestablished to support the teaching and study of Yiddish. Despite competition from Hebrew, especially in twentieth-century Palestine, Yiddish continues to enjoy greater prestige than any other Jewish religiolect (except Hebrew). The tragedy of the Holocaust, coupled with Stalin’s crackdown on Yiddish and the consequent loss of a large number of Yiddish speakers and a fair number of Judeo-Spanish speakers, also helped to increase nostalgic interest in these two religiolects during the twentieth century. In 1996 the Knesset, the Israeli legislature, adopted two laws, the Law of the National Authority for Yiddish Culture (1996) and the Law of the National Authority for Ladino Culture (1996), that established national agencies for the study of Yiddish and Ladino, respectively. Moreover, the Film Industry Regulations of 2001 state specifically that “a film is considered Israeli if the main language in the original copy of the film is either Hebrew, Arabic, Yiddish, or Ladino or some combination of them.” Although the regulations recognize Arabic, as it is one of the two official languages used in the State of Israel, they do not recognize Judeo-Arabic as such. The omission of Judeo-Arabic from the regulations may have adverse consequences because the designation of a film as Israeli entitles its producers to receive grants from the Ministry of Culture. Of course, a movie filmed in Judeo-Arabic could well be recognized as Israeli, since Judeo-Arabic is a variety of Arabic. But symbolically, the fact that Judeo-Arabic is not mentioned in the official regulations of 2001 is very telling. In yet another example, in December 2001 the Israeli Postal Service (È‡ÏÂ·‰ ˙Â¯˘‰) issued stamps recognizing the Yiddish and the Ladino legacies. All these measures constitute clear symbolic signs of the relative importance of Yiddish and Judeo-Spanish in Israeli society. However, 45

CHAPTER TWO in none of the above examples is recognition given to the Judeo-Arabic linguistic heritage: the Knesset has not adopted a law establishing a national authority for the study of Judeo-Arabic; Judeo-Arabic is not regarded as an officially recognized language in the definition of an Israeli film; and the Israeli postal service has not issued commemorative stamps recognizing Judeo-Arabic culture.2 1 The Israeli public has at most a limited acquaintance with the term “Judeo-Arabic.” An average high-school or university graduate in Israel would likely recognize the words “Yiddish” or “Ladino,” but would be puzzled if confronted with the term “Judeo-Arabic.” Even within the Judeo-Arabic speech community in Israel there is little awareness of the linear link between Medieval, Late, and Modern Judeo-Arabic, or of the connection between the various varieties of Arabic. For example, the famous Israeli soccer player Haim Revivo, who played several years in the Spanish professional soccer league and was very popular there, was asked in an interview about his extensive knowledge of languages. In addition to Hebrew and Spanish, he was asked if he spoke Arabic as well. “No, no, I don’t speak any Arabic,” answered Revivo. “But I thought that you spoke Arabic with your grandmother who came from North Africa,” continued the interviewer. “Oh, that’s very different,” answered Revivo: “I only spoke Moroccan with her.” There are probably several reasons for the Israeli public’s lack of familiarity with the term “Judeo-Arabic,” as exemplified in Revivo’s failure to see the connection between his grandmother’s native tongue, Moroccan Judeo-Arabic, and other varieties of Arabic. One reason is that Israeli Jews may wish to avoid the term “Arabic” because of its connotations in the context of the Arab-Israeli conflict; another reason may be the fact that the various Judeo-Arabic geographical varieties are markedly different from one another, and thus differ also from the 21

I thank my student, Gidon Tikotski, who complained about this in early 2002 to Mr. Yitshaq Granot, Director of Stamp Production and Issuance at the Israeli Postal Company; however, nothing has been done as of yet to rectify the situation.

46

JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM familiar local Palestinian dialect (Myhill 2004:122). Avoidance of the word “Iraq” in the term by which the Iraqi Jewish community in Israel designates itself may be another possible example. The community is called Ï·· ˙Â„‰È Babylonian Jewry. However, this term has a long history in the Babylonian Talmud, in Benjamin of Tudela’s writings, in the responsa literature, and in the contemporary Jewish community in Mumbai, India,2 2 which may explain the avoidance of the word “Iraq” in the community’s name.2 3 In the Israeli academic community, however, a number of scholars of Medieval and Later Judeo-Arabic have gained prominence, among them many of Arab descent (Avishur, Bar-Asher, Chetrit, Toby, and more). But even in the academic world the situation is embarrassing, since Israeli universities have not created new positions for the teaching and study of Judeo-Arabic in more than a decade. The current “politically correct” attitude towards “Sephardi/Mizrahi” culture in Israel and the United States may have dictated greater recognition of Judeo-Spanish, but so far has not reached Judeo-Arabic. For example, in recent years a “Sephardi/Mizrahi” caucus has been established through the commendable efforts of Aviva Ben-Ur of the University of Massachusetts at Amherst and Norman Stillman of the University of Oklahoma. This caucus meets annually as part of the annual conference of the Association for Jewish Studies. At a recent meeting, one of the panels discussed the incorporation of “Sephardi/Mizrahi” elements into the Jewish Studies curriculum. One of the participants proudly outlined new proposals that had been initiated at his institution to include what he termed “Sephardi and Mizrahi” components of the curriculum. While some Sephardi materials were presented, nothing about Judeo-Arabic or the culture of Arabic-speaking Jews was even 22 23

I thank Shalom Goldman for his remarks on this matter. In a markedly different mode, in May 2008 Tel Aviv University organized an academic congress entitled “The Iraqi Conference,” on the acculturation of Iraqi Jews into Israeli society. The organizers specifically used the term “Iraq” and avoided the traditional term “Babylonian Jewry.”

47

CHAPTER TWO mentioned. Furthermore, Jewish scholarship on Bible translation, although recognizing Saadia Gaon’s tenth-century Judeo-Arabic translation of the Bible, often ignores the huge range of Judeo-Arabic biblical translations. Frederick Greenspahn has quoted Joseph Hertz, the British Chief Rabbi of the first half of the twentieth century, saying that “the history of Jewish Bible translations would summarize the history of the Jews,”2 4 adding that “it is particularly striking to note those languages in which there are several Jewish translations. These include Greek, Aramaic, Yiddish/German, and English, which constitute the major centers of diaspora Jewish life, further illustrating the intimate connection between the history of Jewish Bible translation and of the Jews” (2006:181). It is disappointing to see Greenspahn ignore the plethora of Judeo-Arabic biblical translations as well as the Arabic-speaking Jewish diaspora that for many centuries consisted of more than half of the Jewish population in the world. One very positive development deserves to be mentioned, however. The publishing house Brill, thanks to the great efforts on the part of an editorial team headed by Norman Stillman of the University of Oklahoma, is in the last phases of publishing an Encyclopedia of Jews in the Islamic World, in which Judeo-Arabic has a prominent place. The Judeo-Arabic religiolect today is endangered and close to becoming extinct. The extensive emigration of Arabic-speaking Jews from the late 1940s through the 1960s is the main reason for this situation. Most of these Arabic-speaking Jews came to Israel (although some also immigrated to France, North America, and other places), where they were under great pressure to drop Judeo-Arabic and adopt Hebrew. Today there are still sizeable Jewish communities in Tunisia and in Morocco.2 5 In Morocco, though, most of the Jewish speech community uses French rather than Moroccan Judeo-Arabic. There 24 25

Hertz 1936, 2:74, quoted in Greenspahn 2006:181. According to www.jewishvirtuallibrary.org, and based on American Jewish Year Book, as of 2006 there were 1,100 Jews in Tunisia and 3,000 in Morocco.

48

JUDEO-ARABIC WITHIN THE JEWISH LINGUISTIC SPECTRUM are still speakers of Judeo-Arabic in Israel (and elsewhere) and a show in Moroccan Judeo-Arabic has been broadcast weekly on Israeli radio. According to the SIL International Ethnologue project, as of the mid1990s there were close to 500,000 speakers of Judeo-Arabic, and I assume that the number has declined today to just under 400,000 speakers (see also Spolsky and Shoahamy 1999:3). This population, however, is aging, so that Judeo-Arabic’s use as a native religiolect will likely disappear in the near future. Consequently, there is an urgent need to encourage research on Judeo-Arabic.

49

CHAPTER THREE THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC (THE ¡ar˙) This chapter begins with a general inquiry into the translation of sacred religious texts, mostly liturgical in nature, into Jewish religiolects. It considers the perceived sanctity of the translated texts and demonstrates how translators dealt with the constant linguistic tension between their desire to provide as literal a translation of the original sacred text as possible, and the need to make this translation from Hebrew or Aramaic fit the linguistic parameters of the target religiolect so that the reader could comprehend the texts. The chapter also analyzes the reasons why such translations were made and traces the evolution of this genre, called ¡ar˙ (pl. ¡ur¥˙) in Judeo-Arabic, especially in North Africa, including Egypt, beginning in the fifteenth century, while also taking into account Saadia’s earlier translation of the Bible in the tenth century. This is followed by a review of previous scholarship on the ¡ar˙ and a discussion of the Cairo Collection, from which several manuscripts relevant to this book are taken. The chapter then offers a linguistic model for the analysis of the translations of sacred texts, based on scanning the text in descending units of grammatical structure, from the phrase level down through the lexical, morphosyntactic, and segment levels, and employing a continuum of least-to-most-literal translations. Examples are provided of various categories and linguistic features. The chapter concludes with a description of two mechanisms which translators/interpreters, called ¡ar˙anim in Judeo-Arabic, used when performing translations of sacred texts.

CHAPTER THREE Translation and Issues of Sacredness1 Jewish sacred texts are written primarily in Hebrew and Aramaic and are used, among other things, for liturgy and for study. As mentioned in chapter 1 (p. 25), these texts include the Hebrew Bible, the Talmud, the Siddur or prayer book, the Passover Haggadah, Midrashic literature, and Pirkei Avot, “Ethics of the Patriarchs,” a tractate of moral and religious teachings dating from Second Temple times and the period following the Second Temple’s destruction. The first translations of sacred texts into Jewish religiolects date back to the Gaonic period in Babylonia in the early Middle Ages. In Late Judeo-Arabic the genre is known as ¡ar˙ (pl. ¡ur¥˙), in JudeoNeo-Aramaic it is termed ¡ar> or ¡ar˙, tavsili in Judeo-Georgian, tefila in Judeo-Italian, tamsir in Jewish Malayalam, ladino in Judeo-Spanish, and taytsh in Yiddish; the genre is also documented in Judeo-Provençal, Judeo-Persian, Judeo-Berber, and other Jewish religiolects. Ladino religious literature had its beginnings in pre-expulsion Spain. However, it ripened and flourished only after the expulsion. Three main texts were continuously translated, especially in Constantinople and Salonika, but also in other Sephardi Jewish centers such as Amsterdam, Livorno, Venice, Vienna, and elsewhere. These texts were the Hebrew Bible, the Passover Haggadah, and the tractate of Pirkei Avot, which Sephardi Jews read on Saturdays between Passover and Pentecost (Schwarzwald 1992:12). Sephardi Jews put these texts to both liturgical and pedagogical use. They were taught to students in religious schools and were read in the synagogue and at home (haf†arot, Bible, Pirkei Avot). Some were read on specific Jewish holidays (biblical megillot, Passover Haggadah). In general, a Jewish religiolect draws from and is influenced by both Hebrew and Aramaic. However, the texts of the ¡ar˙, the Ladino religious literature, tefila, and other translations of sacred texts into Jewish varieties are extreme forms of their respective religiolects, since 1

Some of the issues in this section have also been discussed in Hary 2004.

52

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC they not only draw from Hebrew and Aramaic, but are in fact based on and dependent on them. This makes the genre unique; its investigation clearly reveals the connection between language, religion, and culture. In other words, the intricacies of the ¡ar˙ and other translations of Jewish religious literature demonstrate how a Jewish religiolect operates in a minority Jewish society living under specific linguistic tensions, as will be shown later in the chapter. The translation of sacred texts into different Jewish languages, religiolects, and varieties has been widespread throughout the Jewish world. As we have seen in chapter 1, the occurrence of this genre is a common feature of many Jewish languages. Most Jews learn some Hebrew and Aramaic, but their competence in these languages can vary greatly. When they consult a sacred text, many Jews thus rely not only on the original Hebrew or Aramaic version, but also on a translation in their local variety. Significantly, although numerous Jewish authorities have come out against Bible translations, 2 “Jewish tradition has not merely tolerated [these] translations, but on occasion accorded them with a degree of authority approaching that of the Hebrew” (Greenspahn 2006:181). Thus, the British Chief Rabbi of the first half of the twentieth century, Joseph Hertz, has claimed that “translations of the Bible share in the sacredness of the Original” (1936, 2:71), and German Jewish theologian and philosopher Franz Rosenzweig (1886– 1929) has argued that the “Bible must surely be the first book to be translated and then held equal to the original translation” (1971:366).3 It has been claimed that the ¡ur¥˙ translations were composed primarily for the use of women and children, whose Hebrew was not up to par (Bar-Asher 1988:29–30). However, the translations were not meant to replace the Hebrew Bible, just to complement it. In fact, 2

3

See Greenspahn 2002:61 n. 104: ‰Î¯ˆ ÏÎ Ì‚¯˙‰Ï ‰ÏÂÎÈ ‰¯Â˙‰ ‰˙È‰ ‡Ï (Soferim 1:7) ‘The Torah could not be adequately translated’ and ˘„Â˜‰ È¯ÙÒ ·Â˙ÎÏ Â¯È˙‰ ‡Ï ˘„Â˜‰ ÔÂ˘Ï· ‡Ï‡ (Nahmanides) ‘It was only permitted to write the sacred books in the holy language.’ Both of these quotations appear in Greenspahn 2006:181, including n. 15.

53

CHAPTER THREE Jews continued to read the Bible in Hebrew all over the world, no matter what their linguistic limitations were, and therefore “the Talmud mandates that the Bible be translated at the time that it is read” (Greenspahn 2006:187). Moreover, the ¡ur¥˙ translations still required some knowledge of Hebrew and Aramaic, and thus the claim that the translations were done because of a lack of Hebrew competence is not as accurate as it may seem at first sight. In fact, the translations were written mostly in Hebrew characters, as was customary in Jewish varieties. At times they also incorporated Hebrew and Aramaic words and elements, frequently translating the Hebrew etymologically “so that Hebrew connotations will be more apparent” (Greenspahn 2002:44). Finally, these translations occasionally closely followed the syntactic structures of the Hebrew or Aramaic original, rather than those of the target language, i.e., the Jewish religiolect. The target language in these cases was thus augmented in an effort to translate the original text as literally as possible (Hary 1995). Greenspahn was therefore correct in asserting that the claim that “these translations are evidence of and adjustment to Jewish assimilation” is inaccurate (Greenspahn 2000:6). According to him, “Jewish versions of the Bible are not simply accommodations to linguistic necessity, but also an expression of communal identity and an assertion of ownership of the Bible” (idem, 2006:195). In fact, these translations “reflect the communities which produce them” (ibid., 194). The sacredness in which these translations are held is not uniform. Indeed, the degree of sanctity of holy texts within the Jewish tradition varies, depending on the text, the place, and the time. For example, all of the Hebrew Bible is considered sacred, but its holiest part is the Torah, or the Five Books of Moses. Further, the Ten Commandments are more sacred than other parts of the Torah. Similarly, Genesis, the first book of the Torah, is more sacred than the Song of Songs, but the Song of Songs is still part of the Hebrew Bible, and so derives its sanctity from its inclusion in the sacred canon. In postbiblical sacred texts, the question of the degree of sanctity arises as well. For example, 54

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC different Midrashim4 may possess different degrees of sanctity. Thus Halakhic Midrashim are considered more sacred, because they are more closely linked to the Torah than Aggadic Midrashim are.5 The notion of sacredness is therefore best understood as occupying a continuum. On one end there is the Hebrew Bible, usually considered the ultimate sacred text in Jewish tradition. Other texts and phenomena then can be located along the continuum, at varying distances from the Hebrew Bible, each in accordance with its specific degree of sacredness. Texts such as translations of the Hebrew Bible or Halakhic Midrashim that are strongly connected to the Hebrew Bible will be found closer to the sacred end of the continuum than texts that are not connected directly to the Hebrew Bible.6 But not only texts reside along this continuum, which can accommodate other cultural elements as well—orthography, for example. As shown above in chapter 1 (pp. 19–21), writing systems often serve as religious symbols. The Arabic alphabet is a marker for Islam in languages such as Persian, Urdu, and Osmanli. The Cyrillic alphabet used in Serbian is a marker for the Eastern Orthodox Church, just as the Latin alphabet, in which Croatian is written, is a marker for Catholicism. Jewish languages are most often written in Hebrew characters; consequently, the Hebrew alphabet, as the marker of Hebrew/Jewish religious culture, may itself be considered sacred. Texts written in Jewish religiolects using Hebrew characters may thus be sacred, although

4

5

6

The terms Midrash (sg.) and Midrashim (pl.) refer to a specific rabbinic literature of homilies, interpretations, and biblical exegesis. Midrashim offer commentaries on some books of the Hebrew Bible. The issue in rabbinic literature concerns the status of Aggadic Midrashim, which were the subject of some controversy in the Middle Ages. In other words, the rabbis were not sure how literally to take them. See R. Abraham ben Ha-Rambam’s lengthy essay (reprinted in most editions of >Ein Ya>aqov), where he discusses the many categories of the Aggadic Midrashim. I thank Michael Berger, personal communication. I thank Gordon Newby for several of the ideas expressed here, due to a number of extended conversations we had on this topic.

55

CHAPTER THREE the degree of sanctity is determined primarily by the sacredness of the text. For example, translations of the Bible into Jewish varieties are particularly sacred because, in addition to using Hebrew characters, they are closely associated with the most sacred end of the continuum. I call this relationship “sanctity by association.”7 Thus, translations of sacred texts such as the Bible into Jewish varieties are more sacred than other writings in Jewish varieties, whose sanctity derives merely from their use of the Hebrew alphabet—a marker of Jewish religion and culture; the translations are not sacred to the same degree as the Hebrew original itself. Saadia’s tenth-century translation of the Bible (tafs•r) raises an interesting question with regard to the concept of “sanctity by association,” particularly in relation to the ¡ur¥˙ translations of the fifteenth century and beyond (see below, pp. 60–63). Saadia translated the Bible into Classical Judeo-Arabic, with few colloquial elements, using an idiomatic, nonliteral style of translation; but the ¡ur¥˙ translations were composed in Later Judeo-Arabic, frequently in a verbatim style. Which of these translations was considered more sacred? According to the concept of “sanctity by association,” the ¡ur¥˙ translations and the tafs•r would both be expected to be located near the sacred end of the continuum because they are translations of the Hebrew Bible. But the ¡ur¥˙ translations are indeed considered more sacred than the tafs•r because of their verbatim style. Saadia’s translation was widely read throughout the Arabic-speaking Jewish world and was certainly considered “sacred.” However, as mentioned below (pp. 61–62), Rabbi Issachar ben Susan criticized Saadia in the introduction to his sixteenth-century ¡ar˙ to the Bible for having written in a language that was difficult to understand, adding that therefore the tafs•r was ignored and neglected. This is a clear indication that Issachar 7

In Hary 2004:234 I called this “guilt by association.” I now prefer to call it “sanctity by association.” For example, since a translation of the book of Genesis is associated with the original sacred Hebrew text, this association makes the translation sacred as well. I thank Nick Fabian for his suggestion.

56

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC ben Susan may have considered Saadia’s tafs•r to be less sacred than scholars had previously thought. For this reason Issachar ben Susan thought that a new translation was needed.8 The Translator’s Dilemma The texts of the ¡ar˙, the Ladino religious literature, tefila, and translations of sacred texts into other Jewish varieties exhibit a constant linguistic tension between the translator’s desire to retain the original sacred text word for word and the need to produce a translation that readers could understand. Greenspahn was correct in arguing that “translations that preserve the sound, syntax, and etymological relationships of the original … are unlikely to read smoothly in their target languages. It is, therefore, not surprising that numerous Jewish renderings have been criticized for being overly literal and wooden” (2002:51). This is one of the reasons why many translators sought to balance their literal translations with interpretations. In the ¡ar˙, for example, we find Judeo-Arabic verbatim translations that result in “un-Arabic” structures which imitate the Hebrew source and deviate from standard Judeo-Arabic. On the other hand, in order to produce an easily comprehensible translation, the text must be interpreted from time to time; this is done through the use of word substitution, paraphrase, and the addition of flavor from the local dialect. At times the translation is uncompromisingly literal: every Hebrew word is equivalent to exactly one word in Judeo-Arabic, in order to preserve the Hebrew syntactic structure. In these cases the Judeo-Arabic translation seems strange to native speakers, because the Arabic words become subject to the grammatical rules that govern their Hebrew equivalents and the translator risks creating structures that are unacceptable in Arabic. For example, in the Egyptian Judeo-Arabic Passover Haggadah the following example appears: ÈÎÁÏ ‰ÈÏÚ ‰ÈÈˆÂ ‘we are

8

I thank Geoffrey Khan for helping me pose these questions.

57

CHAPTER THREE duty-bound to tell.’9 This is a verbatim translation of the Hebrew ‰ÂˆÓ ¯ÙÒÏ ÂÈÏÚ. The reason given in the literature for this method of translation is pedagogical. In teaching Jewish sacred texts, one or several words were read aloud in Hebrew, and these were immediately followed by the Judeo-Arabic equivalent from the ¡ar˙. Therefore, the latter had to maintain a word order which exactly followed the Hebrew text. Students were taught to recite after the teacher, first in Hebrew and then in Judeo-Arabic, as follows: ‰ÂˆÓ - ‰ÈÈˆÂ; ÂÈÏÚ - ‰ÈÏÚ; ¯ÙÒÏ - ÈÎÁÏ.1 0 Through such rote repetition they were also indirectly taught Hebrew grammar. For example, in Arabic there is no equivalent to the Hebrew particle ˙‡ /et/ which marks the definite direct object; in Egyptian ¡ur¥˙ it was usually rendered with the Judeo-Arabic word ‰Ï‡ /ila/, so that the latter became the marker for the definite direct object:1 1 Ï·‚ ‰Ï‡ ÂÈ˘ÚÏ ˙ËÚÂ ÂÈ˘Ú ‰Ï‡Â ·Â˜ÚÈ ‰Ï‡ ˜ÁˆÈÏ ˙ËÚÂ ˜ÁˆÈ ‰Ï‡ ÂÏ ˙ËÚÂ ¯ÈÚ˘ ‘and I gave him Isaac, and I gave unto Isaac Jacob and Esau, and I gave unto Esau Mount Seir.’1 2 This is the translation of the Hebrew, ¯ÈÚ˘ ¯‰ ˙‡ ÂÈ˘ÚÏ Ô˙‡Â ÂÈ˘Ú ˙‡Â ·˜ÚÈ ˙‡ ˜ÁˆÈÏ Ô˙‡Â ˜ÁˆÈ ˙‡ ÂÏ Ô˙‡Â. By learning to correlate the Hebrew /et/ with the Judeo-Arabic equivalent /ila/, students would eventually understand the syntactic function of /et/. It was for such didactic purposes that a Judeo-Arabic word order which exactly followed the Hebrew text was considered necessary. The requisites of teaching, however, were not the only reason for word-for-word translation. A tradition of literal biblical translation had prevailed for centuries before the development of the ¡ar˙. This method of translation, already used in the Targums, no doubt played a key role in shaping the ¡ar˙. Indeed, the desire for literal translation 9

This sentence is taken from an Egyptian Judeo-Arabic Haggadah, ms. 3 3,6. See the critical edition in Hary 2009. 10 Read from left to right: /mitzva/ - /wißiyya/ (in Egyptian Judeo-Arabic), etc. 11 See Hary 1991, 1992:300–303 and chapter 8 of this volume, pp. 257–64. 12 Taken from an Egyptian Judeo-Arabic Haggadah, ms. 3 6,7–9. See the critical edition in Hary 2009. I underlined /ila/ in the Judeo-Arabic and /et/ in the Hebrew.

58

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC was so compelling for Judeo-Arabic authors that they were willing to violate rules of Arabic linguistic structure, to the point of assigning new functions to prepositions (/ila/ ‘to,’ for example) in order to produce a text that imitated the sacred Hebrew original as literally as possible. This willingness to violate Arabic grammar may have been connected to the role that Judeo-Arabic played in maintaining Jewish identity in Diaspora minority communities. The demand for verbatim translations of Hebrew sacred texts into Judeo-Arabic may well reflect a strong desire to connect to Jewish heritage in a foreign environment (Muslim in this case). Because of their tight connection to the Hebrew sacred texts, the ¡ur¥˙ gained acceptance as holy texts themselves. As such, they were not updated and eventually became unintelligible as the distance between the language of an old ¡ar˙ and contemporary users became greater. Nevertheless, the ¡ar˙ exhibits numerous examples of nonliteral translations or the use of local dialectal elements in the texts. Thus, the ¡ar˙ demonstrates the constant tension between literal translation and the need to interpret the text and adapt it to the standard Judeo-Arabic style. The same type of linguistic tension is also found in the Ladino religious literature. The translations in this literature are also typically quite literal, but here and there one finds examples of interpretation in them. The Hebrew phrase ¯ÓÂÁÂ Ï˜ ‘an inference from minor to major (a minori ad majus)’ is literally rendered in all the Saloniki versions of the Ladino translations of Pirkei Avot as Ò‡„‡‚ÊÈÙ Â‡ Ò‡‡ÈÈ»·ÈÏ livianas o pesgadas ‘lightness or heaviness’ (Schwarzwald 1989:7). The root ‰‡¯ ‘see,’ although often translated literally as ‰¯È»· vera ‘see,’ is also interpreted in other places as temera ‘fear’ (ibid., 13). This tension may be exasperated because the paradigms of ‰‡¯ and ‡¯È may overlap (Exod 32:25), causing a possible ambiguity. Furthermore, as a manifestation of the linguistic tension, it is common to find stylistic variations in the literature. For example, the phrase ı¯‡ Í¯„ ‘good (or basic) manners’ may be translated in three different ways: ‰¯ÈÈË È„ ÂÊÂ‡ uso de tierra, ‰¯ÈÈË È„ È¯·ÓÂËÒÂ˜ costumbre de 59

CHAPTER THREE tierra, ‰¯ÈÈË È„ ‰Ò‡ÊÂ‡ usança de tierra (Schwarzwald 1989:15). Stylistic variation may also arise in cases where writers wished to elevate the language to what they perceived to be a more respectable literary level. Furthermore, Ladino religious literature is also characterized by archaisms, 1 3 few Hebrew words, and homophony. For example, archaic participial forms are common: ÔÈÈÊÈ„ dizien ‘say’ and Ô‡Ó‡ aman ‘love’; also, the verbal forms ‘I will come, he will come’ are commonly translated with the archaic forms ‰¯È»· ¨È¯È»· verné, verná instead of vendré, vendrá. In addition, the few Hebrew words that do appear in the Ladino literature are drawn from a limited cultural and religious vocabulary. Other Hebrew words such as ÂÏÈÙ‡ ‘although’ and Ï‰˜ ‘public’ are regularly used in Judeo-Spanish, and are therefore not perceived by speakers as Hebrew words. Finally, the choice of words in Ladino reflects homophony, an attempt to adhere as much as possible to the sound of the Hebrew original: the word Ò¯Ù ‘wage’ is rendered by the similar-sounding ÂÈÈÒÈ¯Ù precio rather than by salario; ˘ÙÁÏ and ˘˜·Ï, both meaning ‘seek,’ are rendered by ¯‡˜˘Â· buscar, and ÈÚ and ÔÎÒÓ are both translated as ÂÈ˜ÒÈÓ ‘poor’ (ibid., 10–12). The linguistic tension discussed here has been mentioned by other scholars as well. Greenspahn, for example, has mentioned that the “attachment to Hebrew [in Jewish translations of the Bible] would seem to contradict the very enterprise of translation, while incorporating Jewish tradition can jeopardize the literalistic approach to the Bible long associated with Jews. And indeed, such tensions are manifest in the very nature of these renderings” (2002:61). The Development of the ¡ar˙ In the long Judeo-Arabic tradition of translating sacred texts, two historical “breaks” from previous traditions took place. In the tenth century, Saadia Gaon, who was more committed to the Arabic text 13

The use of archaisms is common to many Jewish religiolects. See chapter 1, pp. 23–24.

60

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC than to the Hebrew, departed from the tradition of verbatim translation that had been the norm since the days of the Septuagint and the Targum. His Judeo-Arabic translation of the Bible closely followed the model of post-Classical Arabic; it became a popular text that was widely used and read throughout the Arabic-speaking Jewish world. Furthermore, Saadia’s translation marks the beginning of the period of Classical Judeo-Arabic, and its orthography, imitating Classical Arabic, served as the basis for Classical Judeo-Arabic or Arabicized orthography.1 4 The second break came in the fifteenth century, when the literary genre of the ¡ar˙, or the literal translation of Hebrew and Aramaic religious sacred texts into Judeo-Arabic, began to develop and flourish locally in different communities (Bar-Asher 1988:29–30). The ¡ar˙ was meant to replace Saadia Gaon’s work in the spirit of a previous tradition, by reviving the literal translation pattern of Onqelos. According to Bar-Asher, the ¡ur¥˙ were composed to provide basic education to young students and to the general public, whose knowledge of Hebrew and Aramaic was considered inadequate. This was also the case in other Jewish religiolects. For example, Judeo-Italian tefilot latini or tefilot vulgar were written for the use of women, as reflected in the Judeo-Italian translations of prayer books. In these translations, adjectives and nouns referring to those who pray are in the feminine gender. This may suggest that women were less likely than men to know Hebrew and Aramaic (Jochnowitz 2001). However, as demonstrated above, the motivation behind these translations may have been more complex. In fact, there were several reasons why new translations of religious sacred texts were needed. In the sixteenth century Rabbi Issachar ben Susan wrote a ¡ar˙ to the Bible in whose Hebrew introduction he wrote that “Saadia wrote [his translation] in Classical Arabic … and [his] language is difficult for anyone

14

See Blau and Hopkins 1984; Hary 1992:82–85, 1996c; see also chapter 2, pp. 35–36.

61

CHAPTER THREE unaccustomed to it, even if he is a native speaker” (Sasson 1932:64, my translation). In addition, Ben Susan noted that, because Saadia translated the Bible in proper Arabic style, students, and even some teachers, found that their knowledge of Arabic, some six hundred years later, was not adequate for understanding his translation. This is the reason, so Ben Susan claimed, that Saadia’s translation had been neglected, ignored, and sometimes even criticized. Ben Susan in fact reported overhearing an important old rabbi saying that he had “no pleasure from our Rabbi Saadia’s translation because we do not understand what he says” (ibid., my translation). In addition, Ben Susan specifically indicated that Saadia’s translation required more interpretation. Furthermore, women, children, and uneducated people were in need of comprehensible texts for their liturgical and educational use (ibid., 65, 67). All of these considerations led Ben Susan to his conclusion that a new translation of the Bible was needed, which he termed ¡ar˙. He then composed the ¡ar˙ in the Arabic of his time (the sixteenth century) and place (the Maghreb). Avishur (1988:45) has summarized Ben Susan’s arguments and concluded that Saadia’s translation was not suitable for teaching purposes. Because Saadia’s translation was not verbatim, it did not suit the teaching method used by Jews, in which students learned by reciting Hebrew and Aramaic texts together with their Judeo-Arabic equivalent. Piamenta (1988:76) has added that the vocabulary Saadia used in his translation was not understood by the average educated speaker of later periods, and therefore the need for new ¡ur¥˙ arose. Moreover, there may be an additional reason for the development of the ¡ar˙. As I have mentioned elsewhere, 1 5 in the fifteenth century the Jewish world began to sever its contacts with Arab Muslim culture. Jewish authors and translators found a way to reconnect to their Jewish identity via the translation of sacred texts. As more elements from the original Hebrew and Aramaic texts were embedded into the 15

Hary 1995:75 and chapter 2 of this volume, pp. 33–34, 36.

62

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC Judeo-Arabic translations (and not just Hebrew or Aramaic words per se), the reader could feel closer to the original texts. Thus, the ¡ar˙, with its literal adherence to the Hebrew and Aramaic original, opened a small window that allowed Jews to reconnect to sacred Jewish texts even in a language other than Hebrew. Saadia’s translation, which was not literal, did not make this possible (Hary 1994b:25–26). Saadia’s tenth-century translation thus eventually led to a number of distinct results. First, it motivated the genesis and development of the ¡ar˙ as the old translation became less accessible. Second, the fact that Saadia’s translation was not literal forced the ¡ar˙anim to compose verbatim translations that could be used for teaching according to the customary method of repetition. Finally, as Saadia’s tafs•r was no longer understood five hundred years after it was written, the ¡ar˙anim realized that sometimes they had to break away from literal translation and interpret the Hebrew text to some extent. In short, for linguistic and pedagogical reasons as well as for purposes of identity, the ¡ar˙ began to develop in the fifteenth century and eventually replaced Saadia’s translation of the Bible. As of now the various ¡ur¥˙ produced from the fifteenth century to the present are still being collected from three sources: manuscripts, printed versions, and recordings (Avishur 1988:40, 1991:141). In Egypt, unlike most of North Africa, very few ¡ur¥˙ exist in print; there are even fewer recordings, and not many manuscripts. For these reasons, the ¡ur¥˙ from the Cairo Collection, some of which are analyzed in this volume and in Hary 2009, in addition to other manuscripts, stand out as an especially rare and valuable source of information on the development of the Egyptian Judeo-Arabic ¡ar˙. The Cairo Collection The Cairo Collection consists of more than one hundred photocopied manuscripts, mostly from Egypt, dating from the eighteenth century through the twentieth. In the 1980s this collection was brought from a synagogue in Cairo to the Institute of Microfilmed Hebrew Manuscripts 63

CHAPTER THREE in the Jewish National and University Library in Jerusalem.1 6 The manuscripts contain mainly Jewish liturgical texts written in Hebrew, Aramaic, and Judeo-Arabic. The large number of noteworthy documents in the collection has made it possible to reconstruct many features of Egyptian Judeo-Arabic of the eighteenth century and later, and to give us a better understanding of Jewish life in premodern and modern Egypt. The collection is organized in ten boxes. Most of the manuscripts in Hebrew and Aramaic contain piyyu†im; others consist of Shabbat laws, seli˙ot or penitential prayers, ritual slaughter laws, and divorce laws. A number of Hebrew manuscripts contain commentaries on several books of the Bible, and documents addressing liturgical issues such as prayers for the New Year and prayers for Shavuot evening. Most of the manuscripts in Judeo-Arabic are ¡ur¥˙: translations of Passover Haggadot and of the books of Isaiah, Jeremiah (including the haf†ara for the ninth of Av), Ezekiel, the twelve Minor Prophets, Psalms, Job, 1 7 Ecclesiastes, and Ruth. Additional Judeo-Arabic manuscripts include isråa¡ar ˙axam•m ‘The Story of the Ten Rabbis.’ The local flavor of the Egyptian dialect comes through in many of the Judeo-Arabic manuscripts.1 9 One whole manuscript, 16

In consultation with the staff, I named this collection The Cairo Collection. In 2005 my graduate student, Ms. Noa David, completed her master’s thesis, consisting of a critical edition and a linguistic analysis of ten chapters of the ¡ar˙ manuscript of the book of Job in Egyptian Judeo-Arabic from the Cairo Collection. See David 2005. 18 I am currently preparing this text for publication. 19 See chapter 4 for a treatment of the Egyptian Judeo-Arabic dialect. Even in the titles mentioned above, one feature of this dialect stands out: a preference for the vowel /u/. In standard Egyptian Arabic the word /
64

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC interestingly enough, consists of letters and different personal lists, all in Yiddish. In addition, several manuscripts are written in more than one language: some are in Hebrew and Aramaic (such as a commentary on Maimonides’ ‰¯Â˙ ‰˘Ó, ritual slaughter laws, and midrashim); others are in Hebrew and Judeo-Arabic (such as bilingual editions of Passover Haggadot). One manuscript, written in 1906 by the Ashkenazi Rabbi of Cairo, Ô‰ÂÎ‰ ÔÂ¯‰‡ Ï„ÚÓ ÔÂ¯‰‡, comprises testimonies, agreements, and requests for divorce agreements in four languages (Hebrew, Arabic, Yiddish, and French). In sum, this collection warrants a great deal of scholarly interest. The corpus for the present study (see preface) includes four manuscripts from the Cairo Collection (mss. 3, 74, 91 and 93—all variants of the Egyptian Judeo-Arabic Passover Haggadah). Previous Studies on the ¡ar˙ In recent years there has been a growing interest in Jewish varieties in general, and in the phenomenon of literal translations of sacred texts from Hebrew and Aramaic into the various Jewish varieties and religiolects in particular. Such translations have been studied in Avishur 2001; Avrahami 1994; Bar-Asher 2001, 2002; Ben-Oren 2000; Chetrit 2007; Cuomo 2000; Doron 1979; Hary 2000a; Kaplan and Mulugetta 2000; Kasher 2000; Maman 2000; Sabar 2000; Schwarzwald 1989; Sephiha 1988; Tedghi 1994; Timm 2007; Tirosh-Becker 2006; Toby 1996; Turniansky 2007; Yerushalmi 2000 and Zafrani 1988.2 0 Although many of the above-mentioned publications specifically address JudeoArabic ¡ur¥˙, there have been a number of other recent studies on translations of the Bible into Arabic in general. The most notable study of translations composed in a minority society is by Polliack (1997), which analyzes the Karaite tradition of biblical translations into Arabic in the tenth and eleventh centuries C.E.

20

This is by no means exhaustive: it is just a sample of recent studies on the translation of sacred texts into Jewish religiolects and varieties.

65

CHAPTER THREE These studies on the ¡ur¥˙ have created a general framework for understanding the nature of the ¡ar˙ and the verbatim character of such translations. Bar-Asher, for example, has characterized Maghrebi ¡ar˙ as basically verbatim with some deviations. Citing Job 1:6 ÌÈ‰Ï‡‰ È· Â‡Â·ÈÂ ÌÂÈ‰ È‰ÈÂ ÌÎÂ˙· ÔË˘‰ Ì‚ ‡Â·ÈÂ ’‰ ÏÚ ·ˆÈ˙‰Ï ‘one day the divine beings came before God, and the devil came along with them,’ he demonstrated how the verse was literally translated in a Moroccan ¡ar˙ as ÂÈ‡µÂ ¯‡‰ Ô‡∑Â ÌÂ‰ËÒÂÙ ·‚‚ÒÓÏ ‡˙Á ‡µÂ ‰‡Ï‡ ˙Ï·‡˜Ó ÂÙ˜ÂÂ˙È ‡∑ÈÈ‡ÏÓÏ „‡ÏÂ (Bar-Asher 1988:10–11). According to Bar-Asher this verbatim translation imposes a “syntactic strangeness” or a “syntactic anomaly,” which sounds discordant to speakers of Judeo-Arabic. A simple verse such as ¯·„ÈÂ ¯Ó‡Ï ‰˘Ó Ï‡ ’‰ (Exod 6:10) ‘And God spoke unto Moses, saying’ is translated as ®ÏÂ˜ÈÏ Â‡© ÔÏÈ‡˜ ‰˘Ó ‡ÏÈ‡ ‰‡Ï‡ ÌÏÏ∑˙Â. Bar-Asher has claimed that instead of ‡ÏÈ‡ ‰‡Ï‡ ÌÏÏ∑˙Â, we would have expected ‡ÚÓ ‰‡Ï‡ ÌÏÏ∑˙Â (the preposition with rather than unto) and instead of ÔÏÈ‡˜ (or ÏÂ˜ÈÏ), we would have expected ÂÏ Ï‡˜Â ‘and told him’ (Bar-Asher 1988:11). Tedghi has provided other examples of the verbatim method of translation in a Moroccan Siddur (1994:93–94), two of which, both from the Amidah prayer, are especially worth noting: (i) ¯ÙÚ ÈÈ˘ÈÏ Â˙ÂÓ‡ ÌÈÈ˜ÓÂ ‘And keeps His faith to those who sleep in the dust,’ itself influenced by Dan 12:2, is translated ˙·‡˙ÓÂ ·‡¯˙ ÔÈÒÚ‡Ï Â˙‡Ó‡ with the strange and unclear phrase /l-na>s•n t-tråb/. (ii) ‰˙‡ Â˙Ï‰˙ ÈÎ ‰Ú˘ÂÂ ÂÚÈ˘Â‰ ‡Ù¯Â ’‰ Â‡Ù¯ ‘heal us, O Sovereign, and we will heal; save us and we will be saved, for You are our glory,’ which forms the eighth blessing of the Amidah, is translated word for word: ÔÈ˙ ‡‡¯ÎÒ ÔÈ‡ Â˙‡‚Â ‡˙È‚ ÂÂ‡ÂÂ‡„Â ‰‡Ï‡ ‡È ‡ÈÂÂ‡„ ‘heal us O God; deliver us and we will be delivered, for You are our gratitude.’

66

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC The studies also point to deviations from verbatim translation in the ¡ar˙. For example, Bar-Asher has shown this with respect to uses of the definite article. The sentence ‰È· È¯Ó‡ ÔÈ·‰Ï ¯ÒÂÓÂ ‰ÓÎÁ ˙Ú„Ï (Prov 1:2) ‘for learning wisdom and discipline; for understanding words of insight’ is translated in the Tafilalt tradition thus: ·‡„‡ÏÂ ‡Ò‡ÈÈ∑Ï Û¯ÚÈ ‡Ó‡‰ÙÏ Ï‡ÂÂ˜ Ì‰‰ÙÈ (1988:12). In other words, although the definite article is not used in the Hebrew source, it is used in the ¡ar˙, reflecting also dialectal use. Tedghi has examined differences in gender, number, tense, and definiteness, as well as examples of free translation. For example, in the sentence ÌÈ¯ÂÒ‡ ¯È˙Ó ’‰ ÌÈ·Ú¯Ï ÌÁÏ Ô˙Â ÌÈ˜Â˘ÚÏ ËÙ˘Ó ‰˘ÂÚ ÌÈ¯‚ ¯ÓÂ˘ ’‰ ÌÈ˜È„ˆ ·‰‡ ’‰ ÌÈÙÂÙÎ Û˜ÂÊ ’‰ ÌÈ¯ÂÚ Á˜ÂÙ ’‰ (Ps 146:7–9) ‘who conducts justice for the oppressed, gives food to the hungry; the Sovereign who sets prisoners free, the Sovereign who restores light to the blind; the Creator who straightens up the bent, loves the righteous, and protects the strangers,’ which is translated as ÔÈÓÂÏ„ÓÏ Ú¯Ò ÏÓÚÈ ÔÈÈÁÓÏ Ê‰È ‰‡Ï‡ ÔÈÈÓÚÓÏ ÏÁÈ ‰‡Ï‡ ÔÈËÂ·¯ÓÏ ˜ÒÙÈ ‰‡Ï‡ ÔÈ‡ÚÈÊÏ Ì‡ÚË ÈËÚÈ ÌÈ¯È‚Ï È„ÁÈ Â‡Ï‡ ÔÈÁÏ‡Ò ·ÁÈ ‰‡Ï‡, the ¡ar˙an translated all the Hebrew active participles using the Judeo-Arabic imperfect, thus deviating from a verbatim translation; for example, the Hebrew participle ‰˘ÂÚ is translated by the Judeo-Arabic imperfect ÏÓÚÈ (Tedghi 1994:101).2 1 In sum, as Bar-Asher has put it so clearly, “In general, the ¡ar˙ is a peshat [i.e., straightforward and literal] verbatim translation, but with specific deviations from it toward the syntax of the spoken dialect and even a few referrals to the derash.”2 2 Tedghi has noted deviations in several lexical and morphosyntactic categories, as mentioned above (1994:94–104), and he concluded that although the ¡ar˙an “chose the verbatim translation, … he does not reconstruct the Hebrew text blindly”

21

Of course, the ¡ar˙an may have thought, with some reason, that the Hebrew active participle and the Arabic imperfect serve the same function. If so, the translation is verbatim functionally but not grammatically. 22 Bar-Asher 1988:15, my translation. Bar-Asher has cited examples such as Onkelos or Saadia’s translations (1988, paragraph 17b).

67

CHAPTER THREE (ibid., 104, my translation). In the example from the end of the Amidah, ÌÈÓÁ¯Â ‰˜„ˆ „ÒÁÂ ÔÁ ÌÈÈÁ ‰Î¯·Â ‰·ÂË ÌÂÏ˘ ÌÈ˘ ‘Lay peace, goodness, blessing, life, grace, mercy, righteousness, and compassion,’ the translation in the Moroccan Siddur is ‡ÁÓÏ ÒÈÚÏÂ ‡Î¯‡·ÓÏÂ ‡ÁÈÏÓÏ ‡ÈÈÙ‡ÚÓÏ ÏÚÊ ˙‡ÓÁ¯Â ‰˜„ˆ Ï„ÙÏÂ (Tedghi 1994:104). The sentence is translated literally as is easily seen in the verbatim word order, but Tedghi has characterized it as a deviation in definiteness, because most of the Judeo-Arabic nouns receive the definite article, which is not the case in the Hebrew original. Both Bar-Asher and Tedghi have made it clear that literal translation and deviation from it are in opposition; they can therefore be presented as the two ends of a scale. Although Tedghi did not use the concept of a scale, it is clear that on a continuum of the definiteness category, his example mentioned above would approach the deviation side:

Deviation Literal ˙‡ÓÁ¯Â ‰˜„ˆ Ï„ÙÏÂ ‡ÁÓÏ ÒÈÚÏÂ ‡Î¯‡·ÓÏÂ ‡ÁÈÏÓÏ ‡ÈÈÙ‡ÚÓÏ ÏÚÊ Figure 4. The deviation/literal continuum of definiteness

The Framework for the Linguistic Analysis of the ¡ar˙ In a number of previous publications I took the approach discussed above and extended it somewhat from a linguistic point of view. I attempted to address the phenomenon of the ¡ar˙ through the eyes of the ¡ar˙anim and their work. The translators’ struggle between literal translation and interpretation can be plotted on a literal/ interpretive scale or continuum in each of the different categories listed below. On the one hand, ¡ar˙anim felt impelled to follow the long tradition of verbatim biblical translation, as found in the Septuagint, Onqelos, and the like. They were also committed to deliver a text that would fit the pedagogical needs of word-for-word translation. Furthermore, as mentioned above, literal translation helped both the ¡ar˙an and the text’s readers/users reconnect to and strengthen their Jewish identity. 68

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC This method of translation, however, created many “un-Arabic” sentences that may have been nearly incomprehensible to average native speakers. Speakers and readers of the religiolect may have found the resulting Judeo-Arabic structure strange, and consequently the ¡ar˙an ran the risk of inserting grammatical structures into the translations that were not usual in Arabic. However, the ¡ar˙anim of the fifteenth century and later also felt the need to interpret certain points in the text, and so they did not blindly follow the model of literal translation. Instead, they would substitute words, formulate paraphrases, and add flavor from their local dialect. They wanted to ensure that their translations would be understood, and not become merely a flat reflection of the Hebrew/ Aramaic text. In sum, the ¡ar˙anim were dealing with a constant literal/interpretive linguistic tension (Hary 1995:84). I have demonstrated how this tension operates in each of the following nine linguistic categories: word order, paraphrase and changes in word order, the definite direct object, prepositions and particles, tense-mood-aspect (T-M-A), the definite article, negation, gender and number, and Hebrew elements. For example, in the T-M-A category, the ¡ar˙an translates the Hebrew participle in ÔÈÏÎÂ‡ Â‡ ‘we eat’ (from the Passover Haggadah) literally as ‰Á‡ ÔÈÏÎ‡Â (ms. 3 2,12), using the Judeo-Arabic participle as well. In other manuscripts, though, he translates the same phrase as ÂÏÎ‡ ‰Á‡ (ms. 74 1,7) and as ÂÏÎ‡ ÔÁ (ms. 93 13,3–4), using the Judeo-Arabic imperfect form, as in colloquial Cairene Judeo-Arabic;2 3 this use shows a tendency toward interpretive translation which backs away from the literal mode (Hary 1995:86–92). In figure 5 the above-mentioned examples are shown on a scale along the interpretive/literal continuum in the T-M-A category:

23

Note that the Cairene Judeo-Arabic dialect possesses the “western” feature of /niktib-niktíbu/ in the verbal conjugation. See chapter 4 of this volume, pp. 118–19, 2.2.2.2, and Hary 1992:278, 2.2.2, as well as the references there.

69

CHAPTER THREE

Interpretive ÂÏÎ‡ ‰Á‡

Literal ÔÈÏÎ‡Â ‰Á‡

Figure 5. The interpretive/literal T-M-A continuum

But the analysis is not always so straightforward. For instance, Exod 18:19, ÍˆÚÈ‡ ÈÏÂ˜· ÚÓ˘ ‰˙Ú ‘now listen to me, I will give you counsel,’ is translated by ¬¯Â˘‡ ÈÏÂ˜ È¥ ÚÓÒ‡ È˙˜ÂÂÏ„ (ms. 15 57a,19) ‘now hear my words, I will give you advice.’ At first glance this translation appears to be rather literal: each word in Hebrew has a Judeo-Arabic equivalent in the same order. Figure 6 shows the three middle constituents of the translation along the continuum:

Interpretive

Literal ÈÏÂ˜ È¥ ÚÓÒ‡

Figure 6. The interpretive/literal continuum

A closer analysis of this example, however, reveals traces of both verbatim and interpretive tendencies in the translation. Figure 6 does not reflect the complexity of the interpretive/literal tension in this case, because it does not show how elements in the phrase pull it in opposite directions on our continuum. The translation ÈÏÂ˜ È¥ ÚÓÒ‡ follows the word-for-word technique, pointing to a verbatim translation (figure 7-1); however, this translation also has an interpretive character (figure 7-2), since instead of the exact meaning of the Hebrew phrase ÈÏÂ˜· ÚÓ˘ ‘listen to me,’ it is rendered as ‘hear my words,’ adding an interpretive mode to the translation. This tendency is also evident in the Protestant Arabic translation of the Hebrew Bible: wðu?? ?B" l?L? ?Ý« ‘hear my voice’ or w"u?? ?I" l?L? ?Ý« ‘hear my words.’ In Saadia’s tafs•r, however, the literal tendency predominates, as Saadia tried to capture the original Hebrew meaning in his translation ÈÓ Ï·˜‡ ‘take it from me’ (see table 2). Furthermore, 70

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC if the word ÏÂ˜ ‘voice’ is carefully observed, the verbatim nature of the translation can be seen, for the ¡ar˙an copied the Hebrew word into the Judeo-Arabic translation. This word exists in Arabic as well, but with a different meaning, ‘word, speech.’ The ¡ar˙an preferred to use a Hebrew word in Arabic dress in order to preserve the original form of the Hebrew text by using a similar-sounding word (figure 7-3). As mentioned in chapter 7 (feature 3-2), this is common in Judeo-Arabic ¡ur¥˙ as well as in many other Jewish-defined religiolects (Greenspahn 2002:46). In other places (see tables 1 and 2 below), however, the ¡ar˙an did not use a similar sounding word, but rather ÒÁ or ˙Âˆ ‘voice.’ Moreover, the Hebrew preposition · ‘in’ is translated into Judeo-Arabic È¥, in keeping with common Arabic use. Whenever the ¡ar˙an refrained from using the Arabic preposition bi, which is similar to its Hebrew cognate, he moved toward the interpretive end of the continuum (figure 7-4). In sum, the complexities due to the literal/interpretive tension are best unraveled by not only examining the phrase as a whole, but also studying its parts. Figures 7-1 through 7-4 demonstrate such an analysis:

Interpretive (word-for-word translation)

Literal ÈÏÂ˜ È¥ ÚÓÒ‡

Figure 7-1. The phrase continuum

Interpretive (the meaning of the phrase) ÈÏÂ˜ È¥ ÚÓÒ‡

Literal

Figure 7-2. The lexical continuum I

71

CHAPTER THREE

Interpretive (choice of word)

Literal ÏÂ˜

Figure 7-3. The lexical continuum II

Interpretive (choice of preposition)

Literal È¥

Figure 7-4. The morphosyntactic continuum

Table 1 below shows a number of occurrences of the Hebrew ÏÂ˜ in various Arabic translations of Genesis and Exodus, while table 2 illustrates various translations of the example cited above, ÈÏÂ˜· ÚÓ˘: Hebrew Original

Citation

Egyptian Saadia ¡ar˙

Protestant Translation

·˜ÚÈ ÏÂ˜ ÏÂ˜‰

Gen 27:22

ÒÁ

˙Âˆ

u%

¯ÙÂ˘‰ ÏÂ˜ È‰ÈÂ

Exod 19:19

ÒÁ

˙Âˆ

¯ÙÂ˘‰ ÏÂ˜ ˙‡Â

Exod 20:14-15

ÒÁ

˙Âˆ

„Á‡ ÏÂ˜ ÌÚ‰ ÏÎ ÔÚÈÂ

Exod 24:3

ÏÂ˜

˙Âˆ

u%

u%

u%

… ‰ÓÁÏÓ ÏÂ˜ … ÌÚ‰ ÏÂ˜ Exod 32:17-18 … ˙ÂÚ ÏÂ˜ ÔÈ‡ ˙ÂÚ ÏÂ˜ … ÏÂ˜ ÔÈ‡Â

ÒÁ; ÏÂ˜

˙Âˆ

u%

‰ÁÓ· ÏÂ˜ Â¯È·ÚÈÂ

ÒÁ

˙Âˆ

u%

Exod. 36:6

24

Table 1. The translation of Hebrew ÏÂ˜ in Genesis and Exodus

24

Hebrew ÏÂ˜ in Exod 32:17–18 is translated into Judeo-Arabic ÒÁ ‘voice.’ However, ‰ÓÁÏÓ ÏÂ˜ ‘the voice of war’ is translated intoJudeo-Arabic ÏÂ˜ in ms. 15 73b,1, but in the margin the ¡ar˙an changed it back to ÒÁ.

72

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC Hebrew Original

Citation Egyptian ¡ar˙

Saadia

Protestant translation w"uI" ÆÆÆ lLÝ

ÈÏÂ˜· Ì‰¯·‡ ÚÓ˘ÈÂ Gen 26:5

È˙Âˆ·

ÈÏÂ˜ ÆÆÆ Ï·˜

ÈÏÂ˜· ÚÓ˘

Gen 26:5

È˙Âˆ È¥ ÚÓÒ‡

ÈÓ Ï·˜‡

ÈÏÂ˜· ÚÓ˘

Gen 27:13

È˙Âˆ È¥ ÚÓÒ‡

ÈÓ Ï·˜‡

ÈÏÂ˜· ÚÓ˘

Gen 30:6

È˙Âˆ È¥ ÚÓÒ

È˙Âˆ ÚÓÒ

ÈÏÂ˜· ÂÚÓ˘È ‡Ï

Exod 4:1

ÈÏÂ˜ È¥ ÂÚÓÒÈ

ÈÓ ÔÂÏ·˜È

ÈÏÂ˜· ÚÓ˘

Exod 18:19

ÈÏÂ˜ È¥ ÚÓÒ‡

ÈÓ Ï·˜‡

ÈÏÂ˜· ÂÚÓ˘˙

Exod 19:5

ÈÏÂ˜ È¥ ÂÚÓÒ˙

ÈÈ¯Ó‡ Ì˙Ï·˜ wðuB" r²FLÝ

w"uI" lLÝ«

w"uI" lLÝ«

wðuB" lLÝ«

w"uI" ÊuFL&¹

wðuB" lLÝ«

Table 2. The translation of Hebrew ÈÏÂ˜· ÚÓ˘ in Genesis and Exodus

Our analysis of ÈÏÂ˜ È¥ ÚÓÒ‡ reveals the complexity of the literal/ interpretive tension in the ¡ar˙. Clearly the framework used in the past is insufficient. Using previous methods of analysis, the arrow in figure 6 obviously approaches the literal end. In figure 7, on the other hand, using the new approach, the arrow changes its placing on the continuum from 7-1 through 7-4. In other words, if the phrase ÚÓÒ‡ ÈÏÂ˜ È¥ is treated only as a whole, as was done in previous analyses, we will lose the linguistic traces left by the ¡ar˙anim when they were translating the text and coping with the dilemma posed by the contrast between the literal and the interpretive tendencies. In the past, scholars who wrote on this issue have provided different examples in each category (tense, definiteness, number, etc.); however, never has an analysis of different categories in the same example been published. In the framework adopted for this volume and exemplified in much detail in part 2, I take each example and show how the above-mentioned linguistic tension is evident simultaneously at different linguistic levels. For instance, at the phrase level,2 5 ÈÏÂ˜ È¥ ÚÓÒ‡ can be placed both closer to the literal end (figure 7-1) and closer to the 25

Here the term phrase refers to elements of sentence structure above the word level, including clauses and syntactic phrases, in accordance with the general meaning of phrase.

73

CHAPTER THREE interpretive end (figure 7-2), depending on which linguistic level and component (phrase level and syntactic structure or lexical level and meaning, respectively) we examine. Further down on the lexical level, ÈÏÂ˜ is closer to the literal end (figure 7-3), while on the morphosyntactic level, the preposition È¥ is closer to the interpretive end (figure 7-4). Head-to-Toe Scanning Building on my previous work, I show in this study how each word and phrase in the ¡ar˙ must be scanned from head to toe at four basic linguistic levels: the phrase level, the word level, the morphosyntactic level, and the segment level. The purpose of the scan is to reveal traces of the ¡ar˙an’s work, as these are the data at our disposal. From such traces one can make inferences about the linguistic dilemmas which had to be resolved during the process of translation. For example, the Hebrew word ‰¯Â˙ ‘Torah’ in the Haggadah is variously translated by three different words: ‰¯Â˙ (for example in ms. 74 22,13), ‰¯ÂË (for example in ms. 93 90,12), and ‰ÚÈ¯˘ (for example in ms. 3 16,22). The Judeo-Arabic word ‰ÚÈ¯˘ in the translation is a trace which points to a decision concerning the choice of a word from the lexicon, as the ¡ar˙an could have chosen one of the other words (‰¯ÂË ¨‰¯Â˙), which can be found elsewhere in the ¡ar˙.2 6 The choice of ‰ÚÈ¯˘ then bears witness to an interpretive tendency (figure 8-1). A different choice of word is Judeo-Arabic ‰¯Â˙, which is an exact copy of the Hebrew ‰¯Â˙, indicating a literal tendency (figure 8-1). Similarly, the word ‰¯ÂË also points to the literal tendency at the word level (figure 8-1). At the segment level, the word ‰¯Â˙ is also on the literal side of the scale, because it uses the sounds of the Hebrew equivalent ‰¯Â˙ (figure 8-2). In comparison, in the translation ‰¯ÂË, the letter Ë constitutes a trace which may refer to regressive emphatization (/t/ > [†] preceding the [r¢]). This by itself is an indication of an interpretive tendency, since the ¡ar˙an here used a phonological variant to change the original 26

Table 4 below should be consulted in order to follow this example.

74

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC sound of the Hebrew ˙ into Judeo-Arabic /†/ (figure 8-2).

Interpretive (word level/ choice of word)

Literal ‰ÚÈ¯˘

‰¯Â˙ ‰¯ÂË

Figure 8-1. The word level continuum

Interpretive (segment level/ regressive assimilation)

Literal ‰¯ÂË

‰¯Â˙

Figure 8-2. The segmental continuum

In the same way, other linguistic features can be found and analyzed at the appropriate linguistic levels. The reader may think that such an analysis would have to assume that different translations were created by the same ¡ar˙an, in order to demonstrate the complexities of the literal/interpretive linguistic tension. Different ¡ur¥˙ were indeed composed by the same ¡ar˙an; however, it also happened, of course, that various ¡ur¥˙ were written by different ¡ar˙anim. But even if the translations were composed by different ¡ar˙anim, the literal/ interpretive tension is still there to be analyzed, because the ¡ar˙anim appear to have all belonged to one “school of translation,” even if there was no established formal institution. The ¡ar˙anim did not work in isolation, but rather they were part of a group of people, some very learned and others less so, who composed translations of sacred texts and worked within the same modes and principles. 2 7 The following example from the Passover Haggadah illustrates the framework for this linguistic analysis of the ¡ar˙. Table 3 represents a 27

I thank Gunvor Mejdell for alerting me to this point. See also Bar-Asher 1988:8–10 and below in this chapter, pp. 89–90 and chapter 6, p. 165.

75

CHAPTER THREE sentence from the Haggadah in four different manuscripts and thus can be used as a good example for comparison. The example also contains several features from all the linguistic levels and many categories. Part 2 of this volume is an expansion of the analysis of the following sentence. It includes examples of all levels, categories, and features based on table 4.

Â‡¯˜ ‡Ï ÍÓ˘· ¯˘‡ ˙ÂÎÏÓÓ ÏÚÂ ÍÂÚ„È ‡Ï ¯˘‡

ÌÈÂ‚‰ Ï‡

28

¬ÓÒ‡ È¥ Â‰„ ÌÏ È≤Ï‡ ‰ËÏˆ Ï‡ ‰ÏÚÂ ¬Â¥¯Ú ÌÏ È≤Ï‡ ·ÂÚ˘ Ï‡ ‰ÏÚ

29

Â‰„ ÌÏ ¬ÓÒ‡· È≤Ï‡ ‰ËÏˆ Ï‡ ‰ÏÚÂ ¬Â¥¯Ú ÌÏ È≤Ï‡ ·ÂÚ˘ Ï‡ ‰ÏÚ

30

Â‰„ ÌÏ ¬ÓÒ‡· È≤Ï‡

ÔÈË‡Ïˆ ‰ÏÚÂ ¬ÂÙ¯ÚÈ ÌÏ È≤Ï‡ ·ÂÚ˘ Ï‡ ‰ÏÚ

31

Â‰„ ÌÏ ¬ÓÒ‡· È≤Ï‡ ÔÈË‡Ïˆ Ï‡ ‰ÏÚÂ ¬Â¥¯ÚÈ ÌÏ È≤Ï‡ ·ÂÚ˘ Ï‡ ‡ÏÚ

Í˙ÓÁ ÍÂÙ˘ ‰ „ ‚ ‰ ¬˙˜‡ÓÁ ·ÂÎÒ‡ ± ¬˙‡ÓÁ ·ÂÎÒ‡ ≤ ¬˙ÈÈÓÁ ·ÎÒ‡ ≥ ¬˙˜‡ÓÁ ·ÂÎÒ‡ ¥

Translation: ‘Pour out your wrath upon the nations that did not know you and upon the kingdoms that did not call your name.’ Table 3. The example from the Passover Haggadah

The analysis of the linguistic features is performed as follows, using the model provided in table 4: (1) ÍÂÙ˘ > ·ÎÒ‡Ø·ÂÎÒ‡: (i) Level: morphosyntactic; Category: TMA; Feature: tense/ aspect (11-1). Judeo-Arabic ·ÎÒ‡Ø·ÂÎÒ‡ ‘pour’ translates the Hebrew imperative ÍÂÙ˘ literally, using an imperative form in Judeo-Arabic as well. (ii) Level: word; Category: lexicon; Feature: root choice (considerations of sound/appearance) (3-2). The decision to use the JudeoArabic root s-k-b was interpretive, since another choice was available to the ¡ar˙an, one which would have been closer to 28

Ms. Ms. 30 Ms. 31 Ms. 29

76

3, folio 23,4. 74, folio 13,1. 91, folio 10b,4. 93, folio 63,10.

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC the sound or appearance of the Hebrew original (Õ · ”). (2) Í˙ÓÁ > ¬˙˜‡ÓÁ/¬˙‡ÓÁ/¬˙ÈÈÓÁ (i) Level: word; Category: lexicon; Feature: root choice (considerations of sound/appearance) (3-2). The decision to use the JudeoArabic roots ˙-m-q and ˙-m-y was literal, since the ¡ar˙an chose Judeo-Arabic roots with sounds close to those of the Hebrew: W)ULŠ ‘anger’ (‚ Â Õ) and WOLŠ ‘rage’ (Ë Â Õ). (3) Ï‡ > ‰ÏÚ/‡ÏÚ (i) Level: morphosyntactic; Category: prepositions/particles; Feature: prepositions (5-1). The choice of the Judeo-Arabic preposition >ala represents the interpretive tendency, since the resulting translation is not literal. In fact, in a verbatim translation we would expect the Judeo-Arabic preposition ila. (ii) Level: segment; Category: orthography/phonology; Feature: Hebrew-influenced orthography (13-6). The spelling ‰ÏÚ with a word-final he to mark the vowel /a/ is probably in imitation of Hebrew orthography, and the final alef in ‡ÏÚ probably represents the influence of the orthography of the Babylonian Talmud, as part of the Hebraized orthography tradition of Judeo-Arabic,3 2 and should therefore be considered literal. (4) ¯˘‡ > È≤Ï‡ (i) Level: morphosyntactic; Category: pronouns; Feature: relative pronouns (6-3). The Judeo-Arabic relative pronoun È≤Ï‡ is commonly used throughout the texts of the ¡ur¥˙ to modify all nouns, regardless of their gender and number. This is an indication of literal translation, since it accords with the rules of Hebrew syntax. (5) ÍÂÚ„È > ¬Â¥¯Ú/¬ÂÙ¯ÚÈ/¬Â¥¯ÚÈ (i) Level: morphosyntactic; Category: TMA; Feature: tense/aspect (9-1). Judeo-Arabic ¬ÂÙ¯ÚÈ/¬Â¥¯ÚÈ ‘know you’ translates Hebrew ÍÂÚ„È in a literal way, since the same tense/aspect (imperfect) is 32

See Hary 1996c:732, 1999a:77–79, and 2009.

77

CHAPTER THREE used in both. However, the translation ¬Â¥¯Ú is interpretive, since it makes use of the perfect tense/aspect, whereas Hebrew uses the imperfect form. The latter also represents a possible hypocorrection that has been standardized in Later Egyptian Judeo-Arabic.3 3 (ii) Level: segment; Category: orthography/phonology; Feature: diacritic marks (13-5). On the one hand, the spelling of Ù without the supralinear dot in ¬ÂÙ¯ÚÈ may represent a literal tendency, as it could be an imitation of the Hebrew letter fe (Ù). On the other hand, the spelling of ¥ with a supralinear dot in ¬Â¥¯Ú/¬Â¥¯ÚÈ represents the fricative pronunciation of [f] rather than the stop [p], so it could denote the interpretive tendency, as this is part of the orthographic tradition of standard Arabic, indicating the få< with a supralinear dot (·) (see chapter 9, p. 311, 13-5.3). (6) ˙ÂÎÏÓÓ ÏÚÂ > ÔÈË‡Ïˆ Ï‡ ‰ÏÚÂ/ÔÈË‡Ïˆ ‰ÏÚÂ (i) Level: morphosyntactic; Category: definiteness; Features: adding the definite article when needed (9-1), and deleting the definite article when not needed (9-4). In ÔÈË‡Ïˆ Ï‡ ‰ÏÚÂ ‘upon the kingdoms,’ the ¡ar˙an added the definite article in the JudeoArabic text in order to conform to Arabic structure, even though the definite article is lacking in the Hebrew text. This represents the interpretive side of the scale (9-1). On the other hand, in ÔÈË‡Ïˆ ‰ÏÚÂ the translation is verbatim, as the definite article is absent both in the Hebrew and the Judeo-Arabic text, although it is required by Arabic grammar (9-4). The ¡ar˙an’s translation in this case follows the Hebrew text slavishly. (7) ˙ÂÎÏÓÓ > ‰ËÏˆ/ÔÈË‡Ïˆ (i) Level: word; Category: lexicon; Feature: word choice (semantic considerations) (3-1). The choice of ÔÈË‡Ïˆ ‘sultans’ represents the interpretive tendency, since Hebrew ˙ÂÎÏÓÓ ‘kingdoms’ does 33

See chapter 4, pp. 126–27, 3.3.3; chapter 5, pp. 141–43; and chapter 8, pp. 215–17, 4-2.1; as well as Hary 1992:294–95 and 314.

78

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC not mean the rulers themselves. On the other hand, the choice of ‰ËÏˆ ‘sultanate, kingdom’ represents a literal translation. (ii) Level: morphosyntactic; Category: number and gender; Feature: number: plural (10-2). The choice of the singular ‰ËÏˆ ‘sultanate, kingdom’ represents the interpretive tendency, since the Hebrew ˙ÂÎÏÓÓ ‘kingdoms’ is in the plural form. (iii)Level: segment; Category: orthography/phonology; Feature: emphatization and deemphatization (13-2). The regressive partial assimilation (emphatization) /s/ > [ß] in ‰ËÏˆ/ÔÈË‡Ïˆ, triggered by /†/, represents an interpretive tendency. (8) ÍÓ˘· > ¬ÓÒ‡·/¬ÓÒ‡ È¥ (i) Level: morphosyntactic; Category: prepositions/particles; Feature: prepositions (5-1). The choice of the Judeo-Arabic preposition bi represents the literal tendency, since it slavishly copies the Hebrew preposition ·, but the choice of the JudeoArabic preposition f• is interpretive wherever it better fits Arabic prepositional use. (9) Â‡¯˜ > Â‰„ (i) Level: word; Category: lexicon; Feature: root choice (semantic considerations) (3-1). The choice of the Judeo-Arabic root ‰„ is taken from colloquial Egyptian Arabic, pointing to an interpretive mode. (10) Â‡¯˜ ‡Ï ÍÓ˘· > ¬ÓÒ‡ È¥ Â‰„ ÌÏ/Â‰„ ÌÏ ¬ÓÒ‡· (i) Level: phrase; Feature: word-for-word translation (1-1). This sentence represents another good example of the literal/ interpretive linguistic tension. The sentence Â‰„ ÌÏ ¬ÓÒ‡· ‘they did not call your name’ is a verbatim translation of Â‡¯˜ ‡Ï ÍÓ˘·, and the word order is the same in both languages. But in the translation ¬ÓÒ‡ È¥ Â‰„ ÌÏ the Judeo-Arabic word order is different and was chosen to accommodate its structure: this translation thus leans toward the interpretive side of the scale.

79

CHAPTER THREE To conclude, the sentence analyzed above from head to toe reveals a complex literal/interpretive linguistic tension, in which the components move back and forth along the continuum in a multidimensional manner according to various linguistic levels, categories, and features. Table 4 below illustrates these levels, categories, and features, which form the basis of a linguistic model for analyzing the literal/interpretive linguistic tension, as demonstrated in great detail in part 2 of this study, using examples from various Egyptian Judeo-Arabic ¡ur¥˙.

80

192–98 199–200

2-2 Adverbs 2-3 Numerals

Negation

Prepositions/Particles

Pronouns

Verb Conjugation

4

5

6

7

Morphosyntactic

Lexicon

3

Word

200–205

244–48 248–50

6-4 Demonstrative pronouns 6-5 Interrogative pronouns and particles 7-2 Finite verbs

256

250–56

240–44

6-3 Relative pronouns

7-1 Infinitives

238–40

6-2 Pronominal suffixes

234–37

5-3 Conditional particles

237–38

231–34

5-2 Coordinating particles and conjunctions 6-1 Independent personal pronouns

219–31

215–19

4-2 Verbal 5-1 Prepositions

213–15

4-1 Nominal

3-2 Word (or root) choice: considerations of sound/appearance 205–12

3-1 Word (or root) choice: semantic considerations

188–92

2-1 Syntactic adaptation

Word Order

2

183–88

1-1 Word-for-word translation

1

Analysis on pp.

Phrase

Feature

Category

Level

THE LINGUISTIC MODEL OF THE ANALYSIS OF THE ¡UR¥˙

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC

81

82

Numerals

12

13

TMA

11

305–6 306 307–10 310–11 311–27

13-3 Elision 13-4 Orthographic marking of the glides 13-5 Diacritic marks 13-6 Hebrew-influenced orthography

304

13-2 Emphatization and deemphatization

Orthography/Phonology 13-1 Assimilation

299–302

295–98

11-3 Voice: passive 12-1 With counted nouns

293–95

11-2 Mood

280–87

10-3 Gender

287–93

274–80

10-2 Number: plural 11-1 Tense/Aspect

274

10-1 Number: dual

272–73

9-4 Deleting the definite article where not needed

Agreement

271–72

9-3 Deleting the definite article where needed

10

270–71

9-2 Adding the definite article where not needed

Definiteness

9

266–70

264–65

8-2 Directional 9-1 Adding the definite article where needed

257–64

8-1 Accusative

Cases

8

Analysis on pp.

Feature

Category

Table 4. The linguistic model of the analysis of the ¡ur¥˙

Segment

Level

CHAPTER THREE

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC The Translation Continuum A translation as a whole can also be placed on a theoretical literal/ interpretive continuum (see figure 9). In a given aspect, a certain translation may be more literal or more interpretive than another. For example, the translation of names within a literal tradition can be placed on a continuum from more to less literal:3 4 (i) There are biblical names that are copied into the Judeo-Arabic ¡ur¥˙ unchanged in a clear literal translation: Ï‡ÏÏ‰Ó (15 0-1,1) ‘Mahalalel’; „¯È (15 0-1,2) ‘Jared’; ÍÂÁ (15 0-1,3) ‘Enoch’; ÁÏ˘Â˙Ó (15 0-1,6) ‘Metushelah’; ÍÓÏ (15 0-1,10) ‘Lemech’; ÁÂ (15 0-1,10) ‘Noah’; Ì˘ (15 0-1,17) ‘Shem’; ÌÁ (15 0-1,18) ‘Ham’; ˙ÙÈ (15 0-1,18) ‘Jephet’; ¯ÂÁ (15 2b,8) ‘Nahor’; ‰˜·¯ (15 2b,15) ‘Rebecca’; Ï‡Â˙· (15 2b,16) ‘Bethuel’; ‰ÎÏÓ (15 2b,16) ‘Milcah’; Ì‰¯·‡ (15 2b,16) ‘Abraham’; ‰¯ÂË˜ (15 4b,11) ‘Keturah’; ‡·˘ (15 4b,13) ‘Sheba’; ¯¥Ú (15 4b,15) ‘Epher’; Ú„È·‡ (15 4b,15) ‘Abida’; ‰Ú„Ï‡ (15 4b,15) ‘Eldaah.’ These names appear in the translation in their original form, although the ¡ar˙an could have changed some of the spellings to reflect the Judeo-Arabic phonetics, as was done elsewhere in a more interpretive translation (see below). (ii) Further along the continuum toward the interpretive side is the translation of the Hebrew name of (possibly) the Hittites ˙ÕÁ (Gen 23:16). It is translated into Judeo-Arabic ˙ÈÁ (15 2a,4), with a spelling that is not identical. Here the ¡ar˙an made sure that the short /e/ vowel in the Hebrew name ˙ÕÁ was expressed in the Judeo-Arabic translation by adding the yod.3 5 A similar example is ÌÈÈ¯Â‰ Ì‡¯‡ (15 2b,7) ‘Aram-Naharaim,’ where the Judeo-Arabic

34

For a more exhaustive treatment of this issue, see chapter 9, pp. 320–27, 13-6.8. 35 Although the representation of short /i/ or /e/ in Later Egyptian Judeo-Arabic is not as common as the representation of short /u/ (Hary 1992:249, 2.1.2), it is quite prevalent in later Egyptian ¡ur¥˙ (David 2005:73, 2.1.3) as well as in the orthography of eighteenth- and nineteenth-century Judeo-Arabic letters from the Geniza (Wagner 2007:68).

83

CHAPTER THREE spelling is not identical to the Hebrew, probably due to phonetic issues, such as a possible long /å/ pronounced in Ì‡¯‡ and a possible /u/ in ÌÈÈ¯Â‰. Vowel lengthening /a/ > /å/ may also explain the alef in Ô‡¯ÓÊ (15 4b,12) ‘Zimran,’ Ô‡˘˜È (15 4b,12) ‘Yokshan,’ and more.3 6 The consonant /w/ is expressed in the personal name ÁÂÂ˘ (15 4b,13) /¡uwa˙/ ‘Shua˙’ by writing two vavs. (iii)Moving still further along the continuum, the following examples are found: ı¥¯Î /karfaß/ (74 21,2) ‘Karpas’ (greens for the Passover Seder), ¯∂Ú ÈÏ‡ /eΩer/ (93, 15,9) ‘Eliezer,’ and ‰È¯∂Ú />aΩarya/ (93 15,9) ‘Azarya.’ Here a Hebrew common noun and two Hebrew personal names were copied into the Judeo-Arabic translation, but with a phonetic modification. The spellings suggest Judeo-Arabic emphatization of /s/ > [ß] and /z/ > [Ω]3 7 respectively, an indication of a slightly more interpretive translation. (iv)The translation of the name of the cave ‰ÏÙÎÓ‰ ‘the Machpela’ into Judeo-Arabic ‰È˙Ó Ï‡ (15 2a,9), literally ‘the doubled,’ although clearly a case where the ¡ar˙an chose a literal translation, is still less literal than the previous example. Indeed, the ¡ar˙an translated the Hebrew root k-f-l ‘double’ verbatim into Judeo-Arabic t-n-y ‘double,’ and also copied the Hebrew locative initial mem into Judeo-Arabic locative m•m, even though he did not use the same root in the Judeo-Arabic translation. (v) Much closer toward the less literal side of the continuum, one finds the translation of the biblical Hebrew place name ‡¯ÓÓ ‘Mamre’ (Gen 23:17). In the ¡ar˙ its translation is ‚¯Ó (15 2ba,6) ‘meadow.’ This is the ¡ar˙an’s attempt to describe a place with a field and vegetation. The various examples, organized in the above groups, are reflected in figure 9:

36 37

See chapter 4, p. 102, 1.1.5, and chapter 5, pp. 148–49, 1.1. See also chapter 5, p. 149, 1.4.2 and p. 150, 1.4.3.

84

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC

Least Literal

(v)

(iv)

(iii)

(ii)

(i) Most Literal

Figure 9. The less literal/more literal continuum

The usefulness of the continuum can be observed in the following examples: (i) Whereas ˙¯ƒÁ È¥ (15 2a,4) ‘in the presence of,’ which translates ÈÊ‡· ‘in the hearing of’ (Gen 23:16), is a clear interpretation of the Hebrew, ˙¯ƒÁÏ (15 2a,7) ‘in the presence of’ is less obvious. The latter translates ÈÈÚÏ ‘in the eyes of’ (Gen 23:18), and is still interpretive, but less so than the previous example, as it is closer to the Hebrew meaning (see chapter 7, pp. 203–4, 3-1.12). (ii) Despite the fact that ı‡¯ /r¢åß/ (91 2b,10) ‘head’ and ıÂ¯Á‡ /u˙r¢uß/ (15 2a,19) ‘beware’ are literal translations of Hebrew ˘‡¯ and ¯Ó˘‰ respectively, the Judeo-Arabic spellings, with the emphatic [ß], are examples of Judeo-Arabic assimilation in the environment of emphatic [r¢], and so indicative of a more interpretive translation. The Work of the ¡ar˙an The detailed theoretical framework laid out here provides for the analysis of the linguistic features in the ¡ar˙. In the course of defining this framework, I attempt to understand how the ¡ar˙an undertook the translations of the Hebrew or Aramaic text. I assume that he was in possession of the original Hebrew or Aramaic text, or at least that he was competent in the text. The ¡ar˙anim of that period memorized the Bible and other sacred texts, so that even if they did not have the actual text in front of them, they remembered it as if it were. The ¡ar˙an quite likely intended to make a literal translation. In other words, he would attempt to find a Judeo-Arabic equivalent for every Hebrew or Aramaic word. Although throughout the ¡ar˙ both literal and interpretive tendencies are seen at each level of the translation, the

85

CHAPTER THREE literal tendency seems to have dominated, since it is more frequently encountered than the interpretive tendency. Furthermore, a literal translation gave rise to “un-Arabic” sentences, which could not have existed in the ¡ar˙ if the guiding principle had been interpretive. In Saadia’s tafs•r, the guiding principle was interpretive, and indeed such “un-Arabic” sentences are not to be found there. In fact Saadia’s tafs•r obeyed, for the most part, the rules of Classical Arabic. In other words, the ¡ar˙an intended to translate the text verbatim, for the various reasons enumerated above (pp. 53–54, 58–59, 61–63). However, the ¡ar˙ includes nonverbatim traces as well. How did these traces find their way into the ¡ar˙ if they were in conflict with the guiding principle of verbatim translation? Was there a separate mechanism that enabled these interpretive traces to find their way into the final product, the ¡ar˙? What happened on the way from the original Hebrew/Aramaic input to the moment the ¡ar˙an wrote down the Judeo-Arabic output of the ¡ar˙? Figure 10 illustrates these questions:

input (original text)

?!

output (the ¡ar˙)

Figure 10. What happens in the process of translation?

Two kinds of mechanisms are assumed to operate inside the rectangle (figure 11). Mechanism A represents the guiding principle of verbatim translation, according to which a ¡ar˙an put a Judeo-Arabic equivalent in place of each Hebrew or Aramaic component. This is the way the ¡ar˙an intended to render the text. In other words, Mechanism A is a deliberate process. Its output, an “intermediate product” (IP), is not 86

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC attested regularly; however, it can be assumed that it would be a rather complete literal translation with many “un-Arabic” sentences and structures that would make comprehension difficult, if not impossible. Mechanism B then takes the IP and allows the interpretive tendencies to find their way in. Mechanism B manipulates the text by way of change, addition, and deletion,3 8 in order to facilitate comprehension. The end result is a complex output, the ¡ar˙, which contains a mixture of both literal and interpretive traces. Figure 11 illustrates these mechanisms:

* IP

Mechanism B

Mechanism A

input (original text)

output (the ¡ar˙)

Figure 11. The ¡ar˙an’s work

We may safely assume that a ¡ar˙an was more aware of using Mechanism A than Mechanism B. Furthermore, it could even be supposed that he may have been unaware of the existence of Mechanism B, and that he was in fact convinced that he did indeed translate the text verbatim. We may also assume that a ¡ar˙an would have used a different linguistic competence for each of the mechanisms. For Mechanism A he would find an equivalent for the Hebrew component in his store of Judeo-Arabic linguistic knowledge. For Mechanism B he would use his knowledge of other linguistic traditions, such as standard Arabic, colloquial Arabic, or the language of previous translations, such as those of Saadia (Classical Judeo-Arabic) or Onqelos

38

Tedghi mentions “broadening,” “narrowing,” and “free translation” (1994: 96–100).

87

CHAPTER THREE (Aramaic). Moreover, it may well have been the case that translators/interpreters had not just the Hebrew text in front of them, but also Saadia’s translation, either in a physical copy or in their mind, since that text was so authoritative. But translators may also have consciously discarded Saadia’s translation, and yet unconsciously consulted it occasionally in order to resolve translation difficulties. Some, of course, may have lost their knowledge of Saadia’s translation altogether. Furthermore, many translators may have realized that a totally verbatim translation was impossible, and therefore considered the use of calculated compromises in the interests of intelligibility and readability. As mentioned above, the IP is not attested regularly, since the end product usually includes Mechanism B. However, a good illustration of an IP and the process of the ¡ar˙an’s work can be shown in two different translations of a clause from the Haggadah, Ú„ÂÈ ÂÈ‡˘ „Á‡Â ÏÂ‡˘Ï ‘and the one who does not know how to ask.’ In ms. 93 16,7 the ¡ar˙an translated this sentence as Ï‡ÒÈÏ ¥¯ÚÈ ÂÒÈÏ È≤Ï‡ „Á‡ÂÂ. This is a verbatim translation which would have sounded strange to native speakers of Arabic for a number of reasons. First, the translation’s word order strictly follows the Hebrew original. Second, ÂÒÈÏ is a hybrid form of the negative particle /lays/ and the third pronominal suffix /-o/, in imitation of the Hebrew negative ÔÈ‡ and the third pronominal suffix Â≠. Finally, Ï‡ÒÈÏ translates the Hebrew infinitive construct ÏÂ‡˘Ï with the Arabic particle /li/, which is phonetically equivalent to the Hebrew preposition Ï, although in Arabic the particle /an/ would have been expected. It seems that the source text ÂÈ‡˘ „Á‡Â ÏÂ‡˘Ï Ú„ÂÈ went through Mechanism A (figure 11) to result in a possible IP of Ï‡ÒÈÏ ¥¯ÚÈ ÂÒÈÏ È≤Ï‡ „Á‡ÂÂ. This IP then went through Mechanism B, which did not produce any change. Had Mechanism B caused any changes, interpretive traces would have been detected in the output. Thus, we assume that in this case the IP was the same as the final output (the ¡ar˙).

88

THE TRANSLATION OF SACRED TEXTS INTO JUDEO-ARABIC Ms. 3 4,10, however, offers a different translation of the same Hebrew clause: Ï‡ÒÈ ˘¥¯ÚÈ ‡Ó „Á‡ÂÂ ‘and the one who does not know how to ask.’ Clearly, this translation leans more toward the interpretive side of the scale and is formulated in standard Egyptian Judeo-Arabic. As in the first translation, the Hebrew clause ÏÂ‡˘Ï Ú„ÂÈ ÂÈ‡˘ „Á‡Â went through Mechanism A to produce an IP, perhaps ¥¯ÚÈ ÂÒÈÏ È≤Ï‡ „Á‡ÂÂ Ï‡ÒÈÏ as in the previous example. In the example in ms. 3, however, Mechanism B did cause the ¡ar˙an to employ his linguistic knowledge of colloquial Egyptian Arabic to arrive at the output ˘¥¯ÚÈ ‡Ó „Á‡ÂÂ Ï‡ÒÈ, which indeed includes several features of colloquial Egyptian Arabic: the negative /ma -¡/, as in [majI?rafS] ‘does not know,’ and the asyndetic clauses [w¶wa:©Idma] ‘the one who does not’ and [jI?rafSjIsal] ‘does not know how to ask.’ To conclude, both translations of ÏÂ‡˘Ï Ú„ÂÈ ÂÈ‡˘ „Á‡Â would appear to result in the same IP after going through Mechanism A. We assume that the difference between the two translations is the result of Mechanism B. In the first example, Ï‡ÒÈÏ ¥¯ÚÈ ÂÒÈÏ È≤Ï‡ „Á‡ÂÂ, Mechanism B does not operate and therefore the IP and the output are the same. In the second example, Ï‡ÒÈ ˘¥¯ÚÈ ‡Ó „Á‡ÂÂ, however, Mechanism B does operate, and therefore the output is different than the IP. The main manipulations that Mechanism B performs on the IP in the second example are the following: omission of the relative pronoun È≤Ï‡, adaptation of ¥¯ÚÈ ÂÒÈÏ to the Judeo-Arabic negation, and use of the asyndetic clause Ï‡ÒÈ ˘¥¯ÚÈ. Since a number of different ¡ar˙anim each composed ¡ur¥˙, many different manuscripts of the same text may exist. When two different versions of a translation from two different manuscripts are compared and analyzed, as is the case here with the versions from mss. 3 and 93, we do not necessarily assume that one and the same ¡ar˙an composed them, although we do consider it likely that they are copies of one earlier prototype ¡ar˙. However, it certainly may also be the case that two manuscript versions of a text were written by two different ¡ar˙anim. How, then, can the same analysis be assigned to 89

CHAPTER THREE both? In fact, even if the ¡ur¥˙ in question were composed by two or more different ¡ar˙anim, they can still be analyzed as texts that share a tradition of translation. It is probable that the tradition of the ¡ar˙ was unified, at least in Egypt, if not elsewhere. Consequently, the same analysis, using the two mechanisms A and B, can be assigned to examples from different manuscripts. Throughout this study, I assume the existence of such a tradition of translation and analyze examples from various Egyptian Judeo-Arabic ¡ur¥˙ accordingly.3 9 To summarize, Mechanism A and Mechanism B scan the text differently and use different working methods. Mechanism A scans the text horizontally, i.e., word by word, and for each Hebrew word substitutes a Judeo-Arabic equivalent to produce the IP. Mechanism B then scans the latter from head to toe all the way from the phrase level through the word level and the morphosyntactic level, down to the segment level, while performing interpretive manipulations where needed. Part 2 of this book illustrates these mechanisms in detail.

39

See p. 75 above, including n. 27, and chapter 6, p. 165.

90

CHAPTER FOUR SPOKEN EGYPTIAN JUDEO-ARABIC: THE EVIDENCE FROM THE ¡arh¢ TEXTS This chapter introduces the methodological considerations for reconstructing the spoken Judeo-Arabic variety used by Egyptian Jews, as extracted from the texts of the ¡ur¥˙. It examines the various linguistic levels in the texts, emphasizes the connection between the orthography and phonetics/ phonology, including its limitations, and points out similar orthographical trends in today’s published modern Egyptian dialect. It then analyzes selected characteristics of dialectal features of spoken Egyptian Judeo-Arabic in phonetics and phonology, morphology, syntax, and the lexicon. The chapter concludes with a summary which highlights the characteristics of spoken Egyptian JudeoArabic, as reflected in the texts of the ¡ur¥˙, that set it apart from spoken Egyptian Arabic used by Christians and Muslims.1 Spoken Egyptian Judeo-Arabic — An Introduction Nada Tomiche was the first to document Egyptian Jewish speech. In 1968 she highlighted several features in Egyptian Judeo-Arabic speech, distinguishing them from the Christian and Muslims dialects in phonetics, morphology, and lexicon. Haim Blanc followed suit, laying the foundation for research on Egyptian Judeo-Arabic in three important 1

It is likely that most of the data presented here refer to the religiolect spoken in Cairo by Jews in the eighteenth and nineteenth centuries. However, it is also probable that Jews in Alexandria and elsewhere had similar linguistic traits (see Tomiche 1968:1179–80; Rosenbaum 2002c:118). I use “Egyptian” and “Cairene” Judeo-Arabic interchangeably to refer to these features.

CHAPTER FOUR articles (1974, 1981, 1985). In 1979, despite the limited data collected on the speech of Egyptian Jews, Blanc argued that the dialect spoken by Cairene Jews was not distinct from the dialect spoken by Christian and Muslims, as opposed to spoken Baghdadi Judeo-Arabic, which was distinct in many ways from Christian and Muslims Baghdadi dialect (Rabin et al. 1979:49–52). 2 With the recordings of contemporary Egyptian Jews conducted by Gabriel Rosenbaum and the collection and publication of Egyptian Judeo-Arabic material, including ¡ur¥˙, done by me (Hary 1992, 2009), there are now ample data to trace the development of spoken Egyptian Judeo-Arabic from the sixteenth century until today. These data have confirmed several of Tomiche’s discoveries and all of Blanc’s findings about the dialect, which in 1974 he termed “non-standard Cairene.” Moreover, Rosenbaum (2002b, 2002c)3 has identified new dialectal features of Egyptian Judeo-Arabic, a process that is continued in this chapter. Rosenbaum has rightly asserted that had Blanc seen the new materials, he would have been convinced that the term “Egyptian Judeo-Arabic” was justified (2002c:118).

2

3

Likewise, S. Morag (Rabin et al. 1979:53) has claimed that modern Egyptian Judeo-Arabic does not include any independent linguistic features. Rosenbaum (2002c:118) has claimed that in Egypt the Christians (Copts) and the Muslims speak the same variety. Recent research on religiolects (chapter 1 of this volume; Benor 2008; Hary and Wein 2008) has not supported this claim. There are simply not enough data available. Had there been systematic recordings of Christians and Muslims, using corpus linguistics methodology, and had there been a comprehensive analysis of Egyptian Christian and Muslim texts from the premodern eras, both of these varieties would have been identified as distinct to some degree. Moreover, Tomiche (1968:1180) has alluded to this issue by noting that “[d]ans sa phonétique, il se caractérise par une absence de vélarisation qui contraste avec les dialectes des Musulmans, mais qui le rapproche du language des Coptes.”

92

SPOKEN EGYPTIAN JUDEO-ARABIC Methodology The texts of the ¡ur¥˙ contain several levels and elements, including those of colloquial speech. Through a careful examination of the texts, these elements can be extracted in order to reconstruct, at least in part, the dialect used by Egyptian Jews during the eighteenth and nineteenth centuries, the period of the composition of the ¡ur¥˙. The texts of Judeo-Arabic ¡ur¥˙ include a mixture of several layers:4 (i) Classical and post-Classical Arabic; (ii) Pseudocorrections; (iii) Standardized pseudocorrections; (iv) Verbatim or direct translations from Hebrew and Aramaic into Judeo-Arabic; (v) Traces from earlier translations of sacred texts, especially that of Saadia Gaon; and (vi) Dialectal components. The tracing of the dialect, then, can be conducted by isolating elements (i) through (v), thereby allowing elements of the spoken dialect (vi) to surface. The findings should then be compared with the modern dialects,5 as well as with documentation from premodern Cairene Arabic, to confirm the evidence. This comparison, of course, has to be done carefully and meticulously to avoid the many complications that may arise. For example, standardized pseudocorrection phenomena (element [iii] above) may pose a special difficulty. It is not easy to discern whether standardized pseudocorrections have become part of the dialect or have just been standardized in the written texts and become productive in them. In order to exemplify this methodology, consider the following sentence: ÔÈÚ Ï‡ ˙ÏÊÂ ‡‰¥¯Ú ÌÏ Ï‚‡¯Â ‰È¯Î· ÈÂÂ˜ ¯∂Ó Ï‡ ˙ÒÂÁ ‰È·ˆ Ï‡Â ˙ÚÏËÂ ‡‰˙¯‚ ˙ÏÓÂ (15 2b,17–18) ‘and the girl (was) very good looking, a virgin, and no man had known her. She went down to the spring, 4 5

See Blau 1991 for the different layers in Classical Judeo-Arabic. See Blau in Rabin et al. 1979:47–48 for a similar discussion.

93

CHAPTER FOUR filled her jar, and came up,’ which is the translation of ‰‡¯Ó ˙·›Ë ‰¯Ú‰Â ÏÚ˙Â ‰„Î ‡ÏÓ˙Â ‰ÈÚ‰ „¯˙Â ‰Ú„È ‡Ï ˘È‡Â ‰ÏÂ˙· „›‡Ó (Gen 24:16). In this translation Classical and post-Classical components (element [i] above) are clear: the i∂åfa \ayr ˙aq•qiyya in /˙usnat al-manΩar/ ‘good looking’ and the use of Judeo-Arabic perfect verbs. The latter does not constitute a verbatim translation, since the Hebrew uses the imperfect followed by the vav consecutive. The verbatim translation elements (element [iv] above), on the other hand, are also evident: lack of /wa-kånat/ ‘and (she) was’ at the beginning of the sentence in the ¡ar˙, in order to mirror the Hebrew; lack of vav conjunctive before ‰È¯Î· ‘virgin,’ again, to meticulously copy the Hebrew; the use of the Judeo-Arabic noun ÔÈÚ ‘spring’ to imitate the Hebrew ÔÈÚ, as other Arabic nouns could have been chosen; and more. It is also possible that the choice of ÔÈÚ ‘spring’ may have been indirectly influenced by Saadia’s translation (element [v] above), which was so prevalent among Arabic-speaking Jews. Furthermore, the choice of ¯∂Ó Ï‡ ˙ÒÂÁ ‘good looking’ by the ¡ar˙an may also be connected to the influence of Saadia’s work. The use of /lam/ before the perfect verb in ‡‰¥¯Ú ÌÏ ‘(he) did not know her’ may reveal a standardized pseudocorrection (element [iii] above), which became part of Egyptian Judeo-Arabic dialect. As a pseudocorrection becomes prevalent in a variety—spoken or written—it reaches a point where it ceases to be a pseudocorrection and becomes an accepted form of the variety. It thus becomes standardized and productive. In Later Egyptian Judeo-Arabic in general, and in the texts of the ¡ur¥˙ in particular, there are countless examples where the negative particle /lam/ is followed by the perfect: ÏÒ¯ ÌÏ (15 27a,16) ‘(he) did not send’; ˙˙‡‚ ÌÏ (1302 2a,3) ‘(she) did not come’ (p. 120, 2.2.5.1.2); ‰·¯˜ ÌÏ (3 16,20) ‘(he) did not bring us near’ (chapter 8, pp. 215–17, 4-2.1, pp. 293–95, 11-2). These and other examples may stem from hypocorrected forms6 that have been 6

For a detailed discussion of hypocorrections, hypercorrections, and other phenomena of pseudocorrections, see Blau 1970:12–15; Hary 2007.

94

SPOKEN EGYPTIAN JUDEO-ARABIC standardized in the variety and become productive. In the dialect, the regularly used particle to negate the past is /ma/. The writer does not choose the latter construction because it is an unmarked form, dominant in the dialect, and thus not prestigious. Instead, he chooses the prestigious marked Classical Arabic negation particle /lam/. The writer, however, “corrects” the construction only halfway: although he changes the negative particle /ma/ to /lam/, he does not change the perfect form following it to the jussive, as required by Classical Arabic. This example follows the criteria for hypocorrections (Hary 2007:277–78). However, because these forms have been regularly used in Later Egyptian JudeoArabic, they have been standardized in writing, as evident in various manuscripts from that period, and have become productive. The question remains, however, whether the use of /lam/ followed by the perfect has become part of the spoken dialect. Rosenbaum has argued (2002a) that it has, producing evidence for its existence in the dialect. He has noted that “the negative particle lam in colloquial Arabic texts derives from the negative particle lam in common use in standard Arabic” and commented that it “is also possible that they are the result of pseudo-corrections which became productive” (ibid., 591), but he has not supplied the above-mentioned analysis in detail. Wagner, who has traced the development of the language used in eleventh- to nineteenth-century Judeo-Arabic letters from the Cairo Geniza, disagreed, claiming that the use of /lam/ in late Egyptian Judeo-Arabic dialect “is a very unlikely scenario” (2007:177). Arnold, on the other hand, has reported on the use of /lam/ in the spoken Arabic of the Jews of Iskenderoun in the northeastern Mediterranean and speculates that it comes from the combination of /lå/ and /må/ (2006–2007:11), and not necessarily from Classical Arabic /lam/. To conclude, the standardized and productive use of /lam/ as a negative particle in Judeo-Arabic ¡ur¥˙ may reflect living usage and may be part of the dialect of the Jews in Egypt, although this has not been fully proven (see also chapter 5, pp. 141–43).

95

CHAPTER FOUR To return to the example presented on pp. 93–94, after the isolation of the above elements, several characteristics that are part of the spoken dialect remain: the Egyptian adverb /
/g/, /g’/, /≠/ th

6 –7 cent.

th

th

8 –11 cent.

/≠/ th

th

12 –17 cent.

/≠/, /g/ th

th

/g/ th

th

17 –18 cent. 19 –20 cent.

Figure 12. The development of the ≠•m in urban Egyptian Arabic

The following notes, however, demonstrate the intricate and complex relationship between orthography and phonetics/phonology: (i) There is clear evidence that the uvular stop /q/ has shifted to /
96

SPOKEN EGYPTIAN JUDEO-ARABIC and probably earlier.7 The orthography in the texts of the ¡ur¥˙, for the most part, does not reflect this shift and uses qof (˜) in such cases. Only infrequently does the orthography in the ¡ur¥˙ mark the reflex of old Arabic /q/ as alef (‡) (see below, p. 106, 1.7.1). Nonetheless, based on other evidence, the shift /q/ > /aßf¥r/ Bird’s Milk by Y¥suf al-Qa>•d was published in 1994 and —uÞ Áœ p?¹b!« /id-d•k da †ør/ This Rooster Is an Ox by °mån Bakr• as recently as 2007.8 In such books the glottal stop that clearly shifted from the uvular stop /q/ is spelled with an Arabic qåf, reflecting standard Arabic spelling and not dialectal pronunciation, along the lines of a similar orthographic practice used in Egyptian Judeo-Arabic. 7 8

See Grotzfeld 1967; Hary 1992:263. Bird’s Milk is the literal translation; the idiomatic translation is That’s Impossible. See also Rosenbaum (2008:391): “One of the significant results of this activity is the publication of several prose texts written completely in the colloquial, thus eliminating the traditional stylistic distinction between narration and dialogue.” Furthermore, in n. 3 he has provided a list of additional novels. Since then, in addition to Bakr•’s novel, the following were also published in Cairo: ‰“UÐ /båzil/ Puzzle by Óusayn >Abd al->Al•m in 2005, and “u??&« …e?¹U??Ž />åyza atgawwez/ I’d Like to Get Married by |åda >Abd al->Ål in 2007.

97

CHAPTER FOUR (ii) The spelling of ˘‡ (3 2,8) ‘what,’ which, surprisingly enough, appears more than twenty times in ms. 3 of the Passover Haggadah from the Cairo Collection and several times in ms. 91,9 may indicate that this interrogative pronoun was pronounced with a short /e/ (/e¡/), another variant of ˘È‡ /∑¡/. There is, however, no confirmation for this pronunciation in any other source.1 0 The same orthographical analysis applies to ˘Ï (3 23,10) and ˘ÈÏ /l∑¡/ (15 3a,17; 5b,4; and more) ‘why.’ Lacking any other supporting evidence, it is safe to assume that the vowel in these interrogative pronouns is indeed long /∑¡/ and /l∑¡/ and that the spelling is simply defectiva.1 1 (iii)Although there is ample evidence in the orthography of Late JudeoArabic in general and Late Egyptian Judeo-Arabic in particular for the appearance of an alef where a short /a/ is expected, it is not a solid proof for the lengthening of /a/ into [å]. This spelling usually occurs in one-syllable words:1 2 È‡˘1 3 /¡ay/ (93 1,2) ‘thing’; Ô‡Ú />an/ (93 2,11; 11,18a) ‘about’; Ï‡¡ /xall/ (74 21,2) ‘vinegar’; Ë‡˘ /¡a††/ (91 3a,1) ‘shore’; È‡Ê /zayy/ (91 2b,8) ‘like.’1 4 Interestingly, Khan has considered this alef to mark the long vowel /å/: “This spelling reflects, indeed, the fact that in these words the vowel is lengthened in the spoken language” (1991:226, my translation). However, it has not been proven that the vowel had shifted to

9

In mss. 15, 1302, 74, and 93 the spelling is always with a yod, ˘È‡ /∑¡/, indicating a long vowel. See the reference to these manuscripts in the introduction to this volume. 10 Tomiche 1968:1180; Blanc 1974:216; Rosenbaum 2002c:126. See also pp. 114–15, 2.1.6.1. 11 See the scriptio defectiva described in Wagner 2007:76 for ‰Ù /f•h/ ‘in it.’ 12 See also Hary 1992:249, 2.1.3; 1994b:377; Khan 1991:226, although Wagner (2007:70) has cited several examples of multisyllabic words: ‰¯‡Ó ‘time,’ ‡‰˙·‡˙Î ‘I wrote to her.’ 13 For a similar spelling in Tunisian Judeo-Arabic, see Doron 1995:134. 14 Ì‡Î /kåm/ (91 7a,17; 8a,5; 93 31,14; 37,14) ‘how much’ is probably long, as is the case in standard Egyptian Arabic, although there is also a variant with a short vowel /kam/.

98

SPOKEN EGYPTIAN JUDEO-ARABIC become long, because there is no supporting evidence for this phenomenon elsewhere. This issue then demonstrates that the orthography cannot be used as the only tool to trace the dialect. It is more probable that the alef was added to “create” a longer word, as a word with only two letters may have seemed too short to the ¡ar˙ånim. In sum, these three cases demonstrate the need to consider the orthography cautiously and not as a solid proof for phonetic or phonological structure. Therefore, in this chapter, if the orthography is not supported by additional evidence, then it is not taken as proof of a phonetic or phonological feature. Below are several notes on the descriptive linguistic analysis of spoken Egyptian Judeo-Arabic, as extracted from the texts of the ¡ur¥˙, that use the methods outlined above. The examples cited in this chapter are only a small selection from the texts of the ¡ur¥˙; many are not unique to Egyptian Judeo-Arabic dialect, but are also found in standard Egyptian Arabic. Each example appears in the JudeoArabic orthography, followed by a conjectured transcription when deemed necessary and plausible. Then appears a reference to the manuscript(s) from which it is cited, followed by a translation into English.

99

CHAPTER FOUR 1. Phonetics and Phonology 1.1 Vowel shifts 1.1.1 /å/ > [a]: È˙¥ÈÈË /†ayfeti/ (15 3b,9) ‘my family’; ¬˙ÈÈ¯‚ /garyetak/ (74 14,13; 93 67,13) ‘your (masc.) maid.’ 1.1.2 /a/ > [i] or [e]/[E], sometimes as part of imåla process: Ô‡ÚÈ‚ /gi>ån/ (3 2,4; 74 1,1; 93 12,5) ‘hungry’; ÏÈÁÓ /ma˙Ell/ (93 1,8) ‘place’; Â˙È¯¥Âˆ /ßufrito/ (93 2,14) ‘his table’; ˙ÂˆÓ ˙È˙‡Ï˙ /talåtit maßøt/ (93 11,12) ‘three matzot (unleavened bread).’ 1.1.3 /a/ > [u]: ˙˜ÂÂ ÏÂ„ /dulwaåla/ (91 2b,15; 93 16,4) ‘exalt’; ·ÚÂ˘ /¡u>b/ (93 22,9; 22,10) ‘people’; ¯Â¡‡ /åxur/ (15 6b,16 and more; 3 11,12 and more) ‘another’; ÔÈ¯¡Â‡ /uxr•n/ (91 9b,14; 93 45,5) ‘other (pl.).’1 6 1.1.4 /i/ > [u]: Ì‡ÊÂ¡ /xuzåm/ (15 3a,5) ‘nose ring’; 1 7 ‰‡È¯ÂÚ />uryåna/ (3 8,15) ‘naked’;1 8 ‰ƒÂ¯ /ru∂a/ (3 21,10) ‘contentment, acceptance.’1 9 Furthermore, note the /u/ in /a¡ar ˙axam•m/ ‘The Story of Ten Rabbis.’ 15

It is possible that the shift comes from standard Egyptian /dilwa /u/); however, the shift /håƒa/ > /da/ > /du/ is presented here. 16 Note that in standard colloquial Egyptian Arabic these examples do not undergo the shift, as they do in Egyptian Judeo-Arabic: /dilwaåla/ (ibid., 598); /¡a>b/ (ibid., 466); /åxar/ and /åxar•n/ (ibid., 11); although /råxar/ and its plural form /ruxr•n/ also exist. 17 In standard colloquial Egyptian Arabic, both /xuzåm/ and /xizåm/ exist (ibid., 249); however, in Egyptian Judeo-Arabic the preference is for /xuzåm/. 18 In standard Egyptian Arabic this shift does not occur: />iryåna/ (ibid., 574). 19 Hinds and Badawi have not recorded the shift in standard colloquial Egyptian Arabic; they wrote /ri∂a/ (1986:340). However, my Egyptian consultants have told me that /ru∂a/ exists in standard Egyptian as well. 20 In standard Cairene /
100

SPOKEN EGYPTIAN JUDEO-ARABIC The following examples exhibit the same shift of /i/ > [u] in standard colloquial Egyptian Arabic that also occurs in Egyptian Judeo-Arabic dialect: Ï˜Â˙ /tu /u/ from standard /mråt-/ ‘wife of’ (Hinds and Badawi 1986:815). Even in the less likely case that the word has come from Hebrew, it still has the vowel /u/, reflecting the preference in Egyptian Judeo-Arabic.2 2 1.1.4.1 In sum, there is a clear preference in spoken Egyptian JudeoArabic for the vowel /u/.2 3 Sometimes standard Cairene dialect features both /u/ and /i/ for certain forms; however, Egyptian Judeo-Arabic demonstrates a consistent preference for /u/. For example, the verbal pattern /fu>ul/ (see below, p. 117, 2.2.1.1) occurs both in standard Egyptian dialect and in Egyptian Judeo-Arabic dialect; however, Muslims prefer /fi>il/ and Jews prefer /fu>ul/. Furthermore, Jews opt for /xuzåm/ ‘nose ring’ and /¡ubbåk/ ‘window,’ whereas /xizåm/ and /¡ebbåk/ respectively are more common in the standard spoken variety in Cairo. In addition, Jews use the vowel /u/ even in cases where the standard variety calls for /e/ or /i/ or even /a/: />uryåna/ (3 8,15) ‘naked,’

21

See also Khan 1991:226. See 1.1.4.1 below and the examples from Hebrew, /kupp¥r/ ‘Yom Kippur’ and /kutubbå/ ‘marriage contract.’ 23 Already in 1991 Khan (p. 226 n. 10) had noticed this phenomenon, but he did not have the same extensive data that exist today. 22

101

CHAPTER FOUR /†u¡t/ ‘washtub,’ and /muxadda/ ‘pillow’2 4 appear in Egyptian JudeoArabic as opposed to />iryåna/, /†i¡t/, and /mexadda/ in the standard variety. Along the same lines, /dulwaåla/ (91 2b,15)2 6 ‘exalt’ appear in Egyptian Judeo-Arabic vs. /tabårak/ and /ta>åla/ in standard Egyptian. In the Egyptian Judeo-Arabic Haggadah the negation particle ˘ÂÓ /mu¡/ (3 1,4) appears, whereas standard Cairene employs both /mi¡/ and /mu¡/. Rosenbaum (2002b:37) has reported that words taken from Hebrew undergo a similar shift: /kupp¥r/ vs. Hebrew /kipp¥r/ ‘Yom Kippur’ and /kutubbå/ vs. Hebrew /ketubbå/ ‘marriage contract.’ 1.1.5 /a/ > [å] This shift occurs in Hebrew words borrowed into the ¡ar˙ (chapter 5, 1.1): ‰‡ÏË· /ba†alå/ (93 1,6) ‘in vain’; Ô‡ÁÏÂ˘ /¡ul˙ån/ (74 21,17; 91 10a,11) ‘table,’ and is consistent with Rosenbaum’s finding: /koh∑n/ ‘priest’ and /kohenå/ ‘daughter or wife of a priest’ (2002b:36). This shift also occurs in personal names adopted from the Hebrew: Ô‡¯ÓÊ /zimrån/ (15 4b,12) ‘Zimran’; Ô‡˘˜È /yok¡ån/ (15 4b,12) ‘Yokshan’; Ô‡„Ó /mEdån/ (15 4b,12) ‘Medan’; Ô‡È„Ó /mEdyån/ (15 4b,12) ‘Midian’; ˜‡·˘È /yi¡båq/ (15 4b,13) ‘Yishbak’; Ô‡„„ /dEdån/ (15 4b,14) ‘Dedan.’ The same process occurs in place names: Ô‡˙Â„ /dotån/ (15 22a,8) ‘Dotan’; ‰˙‡Ó˙ /timnåta/ (15 23a, 15 and 16) ‘to Timnah’ (chapter 7, 3-2.5 and chapter 8, 8-2.1). 1.1.6 /å/ > /a/ preceding />/, /˙/, or /h/: ˙‡Ú¯„ /dira>åt/ (15 37a,9) ‘arms’; Ú˙· /bEta>/ (91 7a,11) and ˙ÈÚ˙· /bEta>et/ (91 8b,2; 10a,12) ‘genitive marker, of (masc., fem.)’; ˙‡‰Ï‡ /ilahåt/ (93 19,2) ‘gods’; ‰ÏÈ‡ /ilah/ (91 9b,7) ‘God’ and ‡‰Ï‡ /ilahna/ (91 9b,10) ‘our God.’ 24

The last two examples are taken from Rosenbaum 2002b:37. Behnstedt and Woidich (1985, 2:178–80) have recognized about forty variants for /dilwaA∂åyma. 26 See above, p. 100, 1.1.3, for a complete reference to these examples. 25

102

SPOKEN EGYPTIAN JUDEO-ARABIC This shift occurs regularly in Maghrebi dialects2 7 but is unusual in standard Egyptian dialect.2 8 This feature, then, is an example of migrated or displaced dialectalism, discussed in chapter 1 (pp. 22–23). Sometimes this vowel shortening occurs elsewhere:2 9 ‰ÈÈËˆÂ /waß†a-niyya/ (74 21,3; 91 11,12) ‘middle’; ˙‡ÒÎ /kasåt/ (74 21,6; 91 11,18) ‘cups’; ÔÈÎÒÓ /masak•n/ (93 12,4) ‘poor (pl.).’ 1.1.7 /i/ or /e/ > [•] or [∑] in words borrowed from the Hebrew: ıÈÓÁ /˙am∑ß/ (93 1,3; 1,6 and more) ‘unleavened’ (chapter 5, p. 149, 1.2). 1.2 The diphthongs The diphthongs contracted to become monophthongs in Egyptian Judeo-Arabic dialect, certainly by the nineteenth century (Blanc 1981:195). 1.2.1 /aw/ > /ø/: ÂÓÂ˜ /<ømo/ (3 4,7) ‘his people’; ‰ËÂˆ /ßø†na/ (3 9,17) ‘our voice.’ Sometimes, however, the contraction has not happened: „Â‚ÂÂÓ /mawg¥d/ (15 3b,12 and many other occurrences) ‘present, in attendance (masc. sg.)’; ‰„ÂÂ‰ /hawda/ (15 3b,12 and many other occurrences) ‘indeed.’ These forms may reflect classicisms, but in some instances the monophthongization has yet to be firmly established. 1.2.2 /ay/ > /∑/: ËÈµ /∞∑†/ (15 2a,5) ‘field’; ‡‰ÈÏÚ />al∑ha/ (15 3b,19 and more) ‘on her’; ÔÈ˙‡ /itn∑n/ (15 5b,6) ‘two’; ˙ÈÏÈÏ /l∑lit/ (93 1,1) ‘the night of’; ÒÈÁ /˙∑s/ (93 1,3) ‘since’; ËÈËÚ‡ /a>†∑†/ (91 3a,2) ‘I gave.’ 1.3 The disappearance of the hamza 1.3.1 As is common in Arabic dialects in general, the hamza frequently 27

See Khan 1991:228 (for example, ‰ÚÓ‚ /gama>a/ ‘group’) and the reference there (n. 19). The same occurs in Classical Maghrebi Judeo-Arabic: ÚÓ˙˘‡Ï‡ /al-i¡tima>/ ‘the meeting’ (Blau 1995:23–24). 28 The shortening of this vowel in standard Egyptian can occur when /bEtå>/ appears before an element that begins with a consonant, such as in /bta>kullo/ ‘know-all,’ although the variant /btå>kullo/, with the long vowel /å/, also exists. 29 There is, of course, the possibility that this is only a spelling variation, not indicating a change in the phonetics.

103

CHAPTER FOUR disappears: ÔÏÈ‡˜ /qåylan/ (15 0-1,13 and more) ‘saying’; ‰¯Ó /mara/ (15 2a,17) ‘wife’; ¯È· /b•r/ (15 22a,14) ‘well’; Ô‡„Ó /madån/ (1302 2a,1) ‘nations’ (madå /madån/).3 0 This shift is clearer in ˙ËÈÁ ¬Á·„Ó /˙∑†at madba˙ak/ (3 21,10) ‘the wall of your altar,’ where the shift /å /∑/, including the disappearance of the hamza, is evident. 1.3.2 The disappearance of the hamza is seen in the orthography of the alif mamd¥da and vowel shortening, /å /a/, either by spelling with a he: ‰ÒÂ¯ /ruasa/ (1302 1b,16) ‘officials, leaders’; ‰Ó‰Â¥ /fuhama/ (3 3,5) ‘wise people’; ‰„˙·‡ /ibtida/ (3 5,17) ‘beginning’; ‰ÓÒ /sama/ (3 24,6) ‘heavens’; ‰Ë·ÂÚ />uba†a/ (3 24,16) ‘imbecile (pl.)’; or with an alef: ‡¯ÊÂ /wuzara/ (1302 1b,5) ‘ministers, nobles’; ‡Ò‡Â¯ /ruasa/ (1302 1b,5 and many other occurrences) ‘officials, leaders’; ‡µ /\ena/ (1302 1b,6) ‘wealth’; ‡ÏÎÂ /wukala/ (1302 2a,13) ‘deputies.’ This spelling is also part of the Hebraized orthography (Hary 1999:79). 1.4 The shift of the interdentals This shift is common in urban Arabic dialects; therefore, it is not surprising to see it in colloquial Egyptian Judeo-Arabic as well. The interdentals frequently become stops. In some cases, however, where the affiliation with Classical Arabic is more desired on the part of the ¡ar˙anim, as a language elevation technique, the interdentals become fricatives (Hary 1992:258–60, 3.6; 1994b:378), because fricatives are considered “closer” to interdentals in speakers’ phonemic inventory. 1.4.1 /®/ > /t/:3 1 Ï˜Â˙ /tu /s/: ÒÈÁ /˙∑s/ (93 1,3) ‘since’; ˙Ò¯ÂÂ‡Â /wi-
It is possible that a scribal error has occurred and the text should read ÔÈ‡„Ó, but even so, it could be /madåyin/ along with the disappearance of the hamza. 31 Although the orthography in ˙¯ÂÂÈÏ /li-yuwarris/ (3 6,9) ‘to inherit’ features a tav, the shift in the spoken variety is probably to the fricative /s/. See also ˙Ò¯ÂÂ‡Â in 1.4.2.

104

SPOKEN EGYPTIAN JUDEO-ARABIC 1.4.3 /ƒ/ > /d/: ·‰„ /dahab/ (15 3a,6) ‘gold’; „È· /nEb•d/ (3 1,1) ‘wine’; ‰¯ÂÎ„ /dEk¥ra/ (3 7,9) ‘males’; Â„Â /widno/ (3 24,12) ‘his ear’; ÁÈ‡·„ /dabåyi˙/ (3 25,6) ‘sacrifices.’ 1.4.4 /ƒ/ > /z/: ˙·ÊÚ˙‡ /it>azzebit/ (15 19a,20) ‘(she) was punished’; ÈÊÏ‡ /il-lazi/ (1302 2a,11) ‘that’; ÊÂ¡‡È /yåxuz/ (93 11,8) ‘(he) takes’; Â˙‡Ê /zåto/ (3 4,19) ‘himself’; ÊÈÊÏ /laz•z/ (3 27,10) ‘delightful.’3 2 1.5 The realization of Classical Arabic /≠•m/ For the most part, the realization of Classical Arabic /≠•m/ is the velar stop /g/: ¯‚˘ /¡agar/ (15 2a,6) ‘trees’; ÒÂÏ‚ /gul¥s/ (1302 1b,3) ‘sitting’; Ô‡ÚÈ‚ /gi>ån/ (3 2,4) ‘hungry.’ This is in line with Blanc 1981:189–93; Davies 2005:xxxv;3 3 and Hary 1996b (see above, p. 96). 1.6 The realization of Classical Arabic />ayn/ There is a weakening of the voiced pharyngeal fricative />/ in Later Egyptian Judeo-Arabic dialect,3 4 as reflected in ¯˘‡˙‡ /itnaa¡ar/ (15 19b,9; 1302 2b,9) ‘twelve’; ¯˘‡„Á /˙idaa¡ar/ (74 23,17) ‘eleven’ (see below 2.2.7.2);3 5 „‰‡ /ahd/ (91 3a,8) ‘pact.’ This phenomenon is not uncommon in Semitic languages, occurring in other Judeo-Arabic dialects as well, and may be a feature of migrated dialectalism.3 6

32

Despite the fact that in an equivalent manuscript „È„Ï (74 20,1) ‘delightful’ appears, there is no supporting evidence for the pronunciation of /lad•d/. Thus, the orthography of „È„Ï may not reflect the shift /ƒ/ >/d/. 33 Davies has claimed that the phoneme was pronounced as /g/ by Y¥suf al-⁄irb•n• in seventeenth-century Egypt. 34 The orthography is a good indication of this weakening; however, it does not supply us with solid proof in this case. Support for this weakening comes from both modern Egyptian dialects and other Judeo-Arabic dialects. For example, in the modern Judeo-Arabic dialect of Peq•>•n, the “articulation of > is very weak in all positions” (Geva-Kleinberger 2005:47)—for example, /bsa(>)d¥na/ ‘they help us.’ 35 The weakening of the voiced pharyngeal fricative />/ appears in words for numbers in many Arabic dialects, and can also be considered consonant deletion. 36 For example, in the Judeo-Arabic dialects of Haifa (Geva-Kleinberger 2004:43), of Peq•>•n (idem 2005:47), and of Tiberias (idem 2008:9).

105

CHAPTER FOUR 1.7 The realization of Classical Arabic /qåf/ 1.7.1 As is the case in the modern dialect of Cairo, the shift /q/ > / / /∞/ also occurs: Â¯ƒµ /∞i∂ru/ (3 19,16) ‘were able.’ 1.8 The glides The glides occur frequently in the texts, denoted usually by two vavs and two yods. 1.8.1 /w/: ÈÎ¯ÂÂ /wirki/ (15 2a,14) ‘my thigh’; ÔÈ„Â‚ÂÂÓ /mawg¥d•n/ (1302 1b,7) ‘present, in attendance (pl.)’; ‰ÂÂ¯ˆÓ /maßarwa/ (3 8,17) ‘Egyptians.’ 1.8.2 In words borrowed from the Hebrew, the voiced labiodental fricative /v/, which is common in Hebrew but does not exist in Arabic, becomes the glide /w/ to adapt to the Arabic phonetic structure (chapter 5, p. 150, 1.6): Â‡ÂÂˆÂÂ /wi-ßiwwånu/ (93 1,14) ‘and He commanded us.’ As reported in Rosenbaum (2002c:123), there is no indication in the ¡ur¥˙ of the use of /v/ in Egyptian Judeo-Arabic. This explains the pronunciation of the letter · in Hebrew words borrowed into Judeo-Arabic as a voiced bilabial stop /b/ (chapter 5, p. 150, 1.6.1). 1.8.3 /y/: ¯ÈÈ‡„ /dåyir/ (15 2a,7) ‘around’; ˙ÈÈÓ /miyyEt/ (1302 1b,6) ‘one hundred’; ‰ÈÈ¯· /barriyya/ (3 16,14) ‘desert.’

37

See also chapter 5, p. 150, 1.5, and Rosenbaum 2002c:123.

106

SPOKEN EGYPTIAN JUDEO-ARABIC 1.9 The phoneme /p/ This phoneme appears in Egyptian Judeo-Arabic dialect, borrowed from the Hebrew: ÌÈ˘µÏÙ Ï‡ (15 4b,17) ‘the concubines’; ˘„ÏÙ (15 1a,9) ‘Pildash’; ı¯‡Ù (15 23-1a,5) ‘Paretz.’ Rosenbaum (2002c:123) has reported that in modern Egyptian Judeo-Arabic dialect the phoneme /p/ can alternate with its voiced counterpart /b/: /pur•m/ alongside /bur•m/ ‘Purim’ (see also chapter 5, p. 154, 3). 1.10 Assimilation 1.10.1 Partial assimilation is seen in the sound change /d/ > /t/ (devoicing) in ÏÂ¡˙˙ /titxul/ (15 36b,7) ‘enter,’ clearly indicating speech use.3 8 1.10.2 A voicing assimilation is the shift /s/ > /z/ in the environment of voiced segments: ‡‰ÂÊÓµÈÂ /wi-yi\miz¥ha/ (3 22,8) ‘and (they) dip it’; ÌÂ‰ÂÊÓµÈÂ /wi-yi\miz¥hum/ (3 22,10) ‘and (they) dip them.’3 9 1.11 Emphatization (tafx•m) It is common in spoken Egyptian Judeo-Arabic to find cases of emphatization (tafx•m) or velarization as a partial assimilation that occurs in the environment of other emphatic phonemes. 1.11.1 /t/ > [†]: In the environment of emphatic /ß/: ÂËÂˆ /ßø†o/ (15 8b,15) ‘his voice’; ‰ËÂˆ /ßø†na/ (3 9,17) ‘our voice’; and ‰ËˆÏ¡ /xallaß†Ena/ (93 78,6) ‘you saved us.’ In the environment of emphatic [r¢]: ¯·ËÚÂ˙ /tu>†abar¢/ (93 11,5) ‘is considered’; ·¯ÂË /†ur¢ab/ (91 9a,14) ‘graves’; and ÔÈË¯Ë¥ /fi†Er¢†∑n/ (91 10a,4) ‘two matzas.’ In the environment of emphatic /†/: ËÈËÚ‡ /a>†∑†/ (91 3a,2) ‘I gave’; ÈÂË‡ÂÂÁË‡ /i†˙wa†¥ni/ (3 24,13) ‘(they) encompassed me’; and ÈË‡Ë·Â¯ /ruba†å†i/ (74 14,14; 91 11b,6) ‘my bondage’ (Hary 1992:255, 3.3). 1.11.2 /d/ > [∂]: In the environment of emphatic [r¢]: Èƒ¯ /r¢a∂i/ (15 0-2,6) ‘bad’; ¯ÿƒ· /ba∂∂ar¢/ (15 Additional Folio-b,1) ‘(he) came early’; 38

See chapter 9, p. 304, 13-1.1. See also ÍÁ˙È˙· /bEtit˙ak/ ‘you are laughing’ (TS AS 209.274/7, quoted in Khan 1992:231 and Wagner 2007:36). 39 In Egyptian Arabic /\amas/-/yi\mis/ means ‘dip,’ whereas /\ammis//yi\ammis/ means ‘eat while dipping the food’ (Hinds and Badawi 1986:630).

107

CHAPTER FOUR Â¯ƒÁ„ /da˙∂ar¢u/ (15 10a,13) ‘they caused to roll down’; ¯ƒ˜ /ni<∂ar¢/ (15 10a,20); ¯ƒ˜È /yi<∂ar¢/ (91 2b,10) ‘able’; ‡¯ƒÂ˜ / [ß]: In the environment of the realization of /q/: ‰ÈÈˆ‡˜ / [ß]: ı¥¯Î /kar¢faß/ (74 21,2) ‘Karpas, greens for the Passover seder,’ although it alternates with nonemphatic /s/: Ò¥¯Î /karfas/ (74 21,2). It may also occur in /z/ > /Ω/: ¯∂Ú ÈÏ‡ /eli>eΩer¢/ (93 15,9) ‘Eliezer’; ‰È¯∂Ú />aΩar¢ya/ (93 15,9 and more) ‘Azarya’; and ¯∂Ú Ï‡ /el>aΩar¢/ (93 15-b,10) ‘Elazar’ (see chapter 5, p. 149, 1.4.). 1.11.5 In words borrowed from Hebrew into Egyptian Judeo-Arabic, there is evidence that speakers used the emphatic /ß/ for the Hebrew /ßadi/ (ˆ): ıÈÓÁ /˙am∑ß/ (93 1,3) ‘unleavened food’ (chapter 5, 1.3). 1.12 Loss of Emphatization (tarq•q) This phenomenon, which usually occurs in the environment of nonemphatic segments, is more common in Later Egyptian Judeo-Arabic dialect than in its Muslim counterpart.4 1 40 41

See Doron 1995:133 for the emphatization in Tunisian Judeo-Arabic. See Blanc 1974 n. 6. In standard Cairene, deemphatization is more common

108

SPOKEN EGYPTIAN JUDEO-ARABIC 1.12.1 /ß/ > /s/: ÚÒÈÏ /li-yisna>/ (91 3a,17) ‘to do’; ˛Â¸ÒÂ˜¯È /yir<úsu/ (74 11,17) ‘(they) dance’; ÒÂ /nuss/ (91 10a,12) ‘half’ (see also 1.1.4 and 1.14); ¯Ò˜ / /d/: ¯„¡‡ /axdar/ (1302 1b,8) ‘green’; ‰˜È„ (74 17,16) and ‡˜È„ /di /t/: È˙¥ÈÈ˙ /tayféti/ (15 3b,7) ‘my family,’ although È˙¥ÈÈË /†ayféti/ (15 3b,9; 3b,10) also occurs; ¬‡È¯˙ /tar• [\], which reflects a voicing process, may occur: Ì‡ÊÂµ Ï‡ /il-\uzåm/ (15 3a,14; 4a,2) ‘the nose ring,’ although Ì‡ÊÂ¡ /xuzåm/ (15 3a,5) and ‰ÓÊ¡‡ Ï‡ /il-axzima/ (15 18b,19) occur as well. among women. See Mitchel 1962:24; Tomiche 1964:17, 98. Norlin (1987), in his phonetic study of Egyptian Arabic, does not report deemphatization, but he records only (Cairene) male speakers. Tomiche (1968:1178) has reported the deemphatization in Egyptian Judeo-Arabic dialect and also among the Christians. Compare also with the Judeo-Arabic dialect of Peq•>•n: /sadi
109

CHAPTER FOUR 1.13.2 The opposite shift, reflecting devoicing, occurs as well: /\/ > [x]: ‰ÈÏ‚¯ ÏÒ¡ÈÏ /li-yixsil/ (15 3a,19) ‘to wash his feet’; ÂÏÒ¡‡Â (15 18b,15) ‘and they washed,’ although /\/ occurs in other cases of /\asal/ ‘wash’: ÌÂ‰ÈÏ‚¯ ÂÏÒµÂ (15 29b,14) ‘and they washed their feet’; Â‰‚Â ÏÒµÂ (15 30a,3) ‘and he washed his face’; „È· Ï‡ È¥ ÏÒµ (15 36b,15) ‘wash in wine.’ 1.13.3 The merger of /∂/ and /Ω/, which is common in Arabic dialects, is also seen in spoken Egyptian Judeo-Arabic. The merger produces /∂/: ‰ÏÈÏÂƒ /∂ull∑la/ (15 17a,13) ‘shaded area’; √Á /˙a∂∂/ (1302 5a,20) ‘pleasure’; Ïƒ /∂ill/ (91 10b,15; 10b,16) ‘shadow’; ¯Âƒ‡ /an∂ur/ (74 15,8; 91 12a,3; 93 70,3) ‘I see.’ 1.13.4 The shift /¡/ > /s/ occurs in Egyptian Judeo-Arabic but is not common: ÒÂ„˜ /-sab>/ (15 6b,20; 7a,16) ‘Beer-Sheba.’4 6 1.14 Deletion of final consonants 1.14.1 In ÒÂ /nuss/ (91 10a,12) ‘half,’ the final consonant /f/ is deleted from the consonant cluster /sf/ (cf. 1.1.4 and 1.12.1). This type of deletion may occur in many dialects; for example, /abßar/ > /baßar/ > /baßß/ ‘see.’ 1.15 The definite article The definite article in the ¡ur¥˙ is written for the most part morphophonemically with the separate morpheme Ï‡: √¯‡ Ï‡ (15 0-1,14) ‘the land’; Ì‡ÈÈ‡ Ï‡ (1302 1b,3) ‘the days’; and ˙È· Ï‡ (93 1,1) ‘the house.’ The exception ·ÂÚ˘‡ /i¡-¡u>¥b/ (74 11,12) ‘the nations’ points to the spoken pronunciation of the definite article with the regular full assimilation that usually occurs with coronal phonemes. This phonetic spelling is part of the Hebraized orthography (Hary 1996c:732), although it is not wide-spread. It is possible to assume that, in the 46

See Hary 1992:260, 3.7, and chapter 5, p. 151, 1.7. This feature is common in Moroccan Judeo-Arabic (Stillman 1988:31, 3) and may be part of migrated dialectalism (chapter 1, p. 22–23).

110

SPOKEN EGYPTIAN JUDEO-ARABIC Egyptian Judeo-Arabic dialect, the assimilation rules apply also to definite articles that precede nouns taken from Hebrew (Rosenbaum 2002c:124–25); however, it is not seen in the texts of the ¡ur¥˙ because of the common morphophonemic spelling: ÍÂÙ˘ Ï‡ (3 colophon) ‘the (prayer) Pour (Your Wrath)’ (see chapter 5, p. 151, 2.1). 1.16 Complex phonological processes 1.16.1 The /†/ in the noun ¯ÂË4 7 /†or¢/ (93 85,16) ‘ox’ has undergone two phonological processes: first, the interdental became a stop (/®/ > /t/) as is common in urban dialects (see 1.4), and then it underwent emphatization (/t/ > [†]) in the environment of emphatic [r¢] (see above, p. 107, 1.11.1).

47

Note that in contemporary written Egyptian dialect, the spelling is similar, as in, for example, the title of the novel —u?Þ Áœ p¹b?!« This Rooster Is an Ox, published in 2007 in Cairo. See also above, p. 97.

111

CHAPTER FOUR 2. Morphology 2.1 The Pronouns 2.1.1 Independent pronouns ‰Á‡ /i˙na/ (15 10a,16; 3 2,12; 3,3; 74 1,6; 10,9; 91 8b,1) ‘we’; Â˙‡ /intu/ (74 14,1; 91 10b,19; 93 65,11) ‘you (masc. pl.).’ The third plural form is a distinct feature of Cairene Judeo-Arabic dialect: ÔÓÂ‰ /humman/ (15 18a,10; 18a,15 and many more examples; 3 13,16; 17,20; 93 78,12) ‘they,’ 4 8 as opposed to standard Cairene /humma/, although the former may appear in regional Egyptian dialects as well as among lower socioeconomic speakers. 2.1.2 Possessive pronouns 2.1.2.1 First person singular: The usual /-i/ occurs regularly: È„ÈÒ /s•di/ (15 2a,1) ‘my master’; ÈÈ· /b˙ni/ (15 2a,2) ‘between me’; È·‡ /ibni/ (15 17b,10) ‘my son.’ After a long vowel the pronoun /-ya/ appears: ‰È‡¯ÂÂ /waråya/ (15 2a,18) ‘after me’; ‡ÈÂ·‡ (15 2b,1 and more) and ‰ÈÂ·‡ /ab¥ya/ (15 7b,15) ‘my father’; ‰È‡ÚÓ /ma>åya/ (15 10a,5 and more) ‘with me’; ‰ÈÏ /liyya/ (15 2b,1) ‘to me.’ 2.1.2.2 Second person masculine singular: The colloquial pronoun /-ak/ occurs regularly: ¬˙ÈÈÓ /mayyetak/ (15 2a,2) ‘your dead’; ¬˙ËÏÒ /sal†antak/ (1302 3a,15) ‘your kingdom’; ¬„È·Ú />ab•dak/ (3 7,19) ‘your slaves.’ 2.1.2.3 Second person feminine singular: The colloquial pronoun /-ik/ or /-ek/ appears: ¬Ë· /ba†nek/ (15 5b,6) ‘your belly’; ¬˙Á‡È¯Ï /liriyå˙tek/ (3 25,17) ‘to your rest’; however, after a long vowel the pronoun is /-ki/: ÈÎ‡˘Á‡ /a˙¡åki/ (15 5b,7) ‘your womb’; ÈÎÂ·‡ /ab¥ki/ (15 3a,7; 1302 4a,8) ‘your father.’ 2.1.2.4 Third person masculine /-o/ (or /-u/) appears after a consonant. Following prepositions and particles: ÂÈ· /b˙no/ (15 12b,13) ‘between him’; ÂÚÓ /ma>o/ (15 4a,12; 1302 4b,18) ‘with him’; Â„Ú />ando/ (15 48

There is only one example of ÌÂ‰ in Genesis (15 5a,11), but I suspect it reflects classical use. See also Rosenbaum 2002b:38 and 2002c:126.

112

SPOKEN EGYPTIAN JUDEO-ARABIC 7a,13) ‘with/by him’; ÂÓ /minno/ (15 36a,15; 1302 4a,1) ‘from him’; ÂÓ‡„˜ /alo/ (15 13a,15) ‘(he) let him’; Â·Á /˙abbo/ (15 31a,3) ‘(he) loved him’; Â¥„‡ˆÈÂ /wi-yißådfo/ (15 31a,14) ‘(he) met with it’; Â·È‚‡ /ag•bo/ (15 31a,20) ‘I bring him.’ The objective pronoun is /-h/ after long vowels: ‰Â„¡‡Â /wi-axad¥h/ (15 22a,18) ‘and they took him’; ‰‡¯˙˘‡Â /wi-i¡taråh/ (15 23-1a,6) ‘and (he) bought him’; ‰ÂÏÊ /nazzil¥h/ (15 23-1a,8) ‘(they) brought him down.’ 2.1.4 Demonstrative pronouns It is not uncommon to find in the ¡ur¥˙ the regular colloquial Cairene demonstrative pronouns /da/ for the masculine and /di/ for the feminine; however, specific Egyptian Judeo-Arabic variants exist in the texts as well: /de/ for masculine singular ‘this’ and /døli/ or /hadøli/ for plural ‘these,’ as opposed to standard Egyptian /døl/. 2.1.4.1 /da/: ¯¡˙¥ÂÓ ‰„ /da muftaxar/ (3 3,7–8) ‘this is praiseworthy’; ‰„ ··Ò· /bisabab da/ (93 17, 5–6) ‘for this’; ‰„ ¯ÈË¥ /fi†•r da/ (3 18, 10) ‘this unleavened bread.’ A variant among Cairene Jews is /de/:4 9 Ï‡ ‰È„ ·‡· /de il-båb/ (91 12a,15) ‘this is the gate’; ‰È„ ¯ÈË¥ /fi†•r de/ (91 8b,10) ‘this unleavened bread’; ‰È„ ¯¯ÂÓ /mor∑r de/ (91 8b,8) ‘this bitter herb.’ Furthermore, the pronoun È„ in the sentence ‡ÏÚÂ È„ ˘È‡ 49

This form is archaic and survived among the Jews in Cairo, a typical feature of Jewish-defined languages. See Blanc 1974:216 and Rosenbaum 2002c:126. See also chapter 1, pp. 23–24.

113

CHAPTER FOUR È„ ˘È‡ (1302 3b,17) ‘what is this and about what is that?’ may represent Egyptian Judeo-Arabic masculine /de/, since it translates the Hebrew masculine form ‰Ê ‘this’ (Esth 4:5). 2.1.4.2 /di/: È„ ··Ò· /bi-sabab di/ (91 2b, 3–4) ‘for this’; ‰È‡ˆÚ Ï‡ È„ /di l->aßåya/ (3 13,2) ‘this is the rod’; È„ ˙‡Î /kånit di/ (3 26,19) ‘this was.’ 2.1.4.3 The Egyptian Haggadah exhibits Cairene Judeo-Arabic /døli/: „‡ÏÂÂ‡ Ï‡ ÈÏÂ„ (3 10,17) ‘these are the sons’; ˙‡·¯ƒ ¯˘Ú ÈÏÂ„ (3 13,14) ‘these are the ten plagues’; ÈÏÂ„ ˙‡ÓÏÎ ˙‡Ï˙ (3 17,18–19) ‘these three words’; ÔÓÂ‰ ÈÏÂ„Â (3 17,20) ‘and these are.’ Alongside this form, another variant is found in Genesis and Esther, /hadøli/: ÈÏÂ„‰ Ì‡ÏÎ Ï‡Î (15 3a,13) ‘like these words’; ÈÏÂ„‰ Ì‡ÏÎ Ï‡ (15 24a,5) ‘these words’; ÏÂÎ ‰¯ÂË˜ „‡ÏÂÂ‡ ÈÏÂ„‰ (15 4b,15–16) ‘all of these are the children of Ketura’; ÈÏÂ„‰ Èƒ‡¯‡ Ï‡ (15 6a,11) ‘these lands’; ÈÏÂ„‰ ‰È‡ˆÚ Ï‡ (15 23b,18–19) ‘these rods’; ÈÏÂ„‰ Ì‡ÈÈ‡ Ï‡ (1302 1b,7) ‘these days.’ This variant /hadøli/ is a feature of migrated dialectalism (chapter 1, pp. 22–23) as it exists in Christian (/hadøli/) and Judeo-Arabic (/haƒøli/) Baghdadi dialects.5 0 2.1.5 Relative pronouns /illi/ is employed, reflecting dialectal use: ‰È‡ËˆÂ Ï‡ ‰¯ÈË¥ Ï‡ Â„¡‡ÈÂ ÌÂ‰È¥ ÈÏ‡ (3 1,9–10) ‘and (they) take the middle matza which is inside’; ‰¯˙˘‡ ÈÏ‡ È„‚ Ï‡ ÏÎ‡Â (74 24,14–15 and more) ‘and (he) ate the kid that (he) bought’; ˘¥¯ÚÈ ‡Ó ÈÏ‡Â (91 2b,8) ‘and he who does not know’; ‡˙‡‰·‡Ï ˙È¥˜ÂÂ ÈÏ‡ ‡ÈÈ‰ (91 3a,13) ‘it is that (promise), which has stood by our fathers.’ 2.1.6 Interrogative pronouns 2.1.6.1 ˘È‡ /∑¡/ (15 2a,2; 6a,20; and more; 1302 2b,8; 4b,17; and more; 74 2,13; 10,6; and more; 93 17,3; 39,4; and more) and ˘‡ (3 1,13; 2,8; and many other occurrences; 91 8b,8; 10b,2; and more) ‘what’ (see

50

See Blanc 1964:138. Muslim Baghdadi Arabic exhibits /h(a)ƒøla/; however, Erwin (2004:290) reports occasional /haƒøli/.

114

SPOKEN EGYPTIAN JUDEO-ARABIC above, p. 98, ii) are common in Cairene Judeo-Arabic dialect, reflecting the preservation of an older form in standard Cairene.5 1 Furthermore, this pronoun appears at the beginning of the sentence, as opposed to standard Cairene /∑h/, which is usually postposed (/ismak ∑h/ ‘what is your name?’): ˙Ï‚Ú˙Ò‡ ‰„ ˘È‡ (15 8a,6) ‘what is this that you hurried?’; ¬˙·ÏË ˘È‡Â (1302 4a,17; 4b,1; and more) ‘and what is your request?’; ‰„‚‰ Ï‡ ‡È‰ ˘‡ (3 1,13) ‘what is the Haggadah?’ From the data in the ¡ur¥˙, it seems that in the eighteenth and nineteenth centuries Jews in Cairo used the interrogative pronoun /∑¡/ exclusively. Rosenbaum (2002c:126) has shown that in twentiethcentury Cairene, Jews used /∑¡/ alongside standard Cairene /∑h/. In the texts of the ¡ur¥˙ there are two instances of ‰È‡ /∑h/ (93 67,2; 70, 2) ‘what,’ indicating the possible beginning of the linguistic change. However, the pronoun is still preposed: ‰ÏÏ‡Ï ·ÈÂÂ‡‚‡ ‰È‡ (93 67,2) ‘how shall I repay to God?’; Ô‡Ò‡ Ï‡ ÈÏ ÚˆÈ ‰È‡ (93 70,2) ‘what will the man do to me?’ In contrast, the equivalent interrogative pronoun ‡Ó /ma/ in 93 13,1, for example, could be a borrowing from the literary variety or an imitation of the Hebrew ‰Ó and is not a reflection of the dialect. 2.1.6.2 ˘È‡ „˜ /åyil 51

Note that in standard colloquial Egyptian, /∑¡/ may appear in the proverbs and frozen phrases: /∑¡>arrafk/ ‘How did you find out?’; /∑¡˙ål law/ ‘What would happen if … ?’; and more (Hinds and Badawi 1986:46). In standard Egyptian /i¡mi>na/ ‘what does it mean?’ the vowel of the interrogative is a short /i/.

115

CHAPTER FOUR †ayyibåt/ (91 7a,17) ‘how many good deeds?’; Â·¯ƒ‡ Ì‡Î /kåm in∂arabu/ (93 31,14) ‘how many times were they smitten?’ 2.1.6.3 ˘ÈÏ /l∑¡/ (15 5b,20; 9a,6; and many more occurrences; 91 10b,8) and ˘Ï /le¡/ (3 23,10; 74, 13,4) ‘why’ (see above, p. 98, [ii]) are typical in Cairene Judeo-Arabic dialect, preserving an older form of standard Cairene and appearing preposed in the sentence: ÔÓ ¥˜Â˙ ˘ÈÏ ‰¯· (15 3a,17) ‘why do you stand outside?’; ÂÏÂ˜È ˘Ï (3 23,10) ‘why do they say?’ This is in contrast to postposed standard Cairene /l∑h/: /ru˙t l∑h/ ‘why did you go?’ 2.1.6.4 ¥ÈÎ /k∑f/ (15 30b,2; 1302 6a,2) ‘how’ in Cairene Judeo-Arabic also preserves an older form and is preposed as well: ÚÏË‡ ¥ÈÎ (15 31,b) ‘how (can) I go forth?’ This is in contrast to standard modern Cairene where /izzåy/ is employed as the postposed interrogative pronoun for ‘how.’ 2.1.6.5 The following are other common interrogative pronouns that are typical to the dialect: ÔÈÓ /m•n/ (15 3a,6) ‘who?’; ÔÈ¥ /f∑n/ (15 22a,6) ‘where?’; ÔÈÓ /min∑n/ (15 10a,16; 3 14,18) ‘from where?’ 2.1.7 Genitive marker The texts of the ¡ur¥˙ reflect at times the Cairene genitive marker /bitå>/ or /bEta>/ (masc. sg.), with some attestation of /bitå>a/ or /bEta>a/ (fem. sg.) but no attestation of /bit¥>/ (pl.). 5 2 The following examples from the Haggadah keep the gender and number agreement: ¥ˆ Â„Â¡‡È ÔÈÓÂ˜È¥‡ Ú‡˙· È·¡Ó Ô‡Î È≤Ï‡ ‰¯ËÈÈ¥ Ï‡ (93 47,10) ‘(the participants) hold half of the hidden matza of the afikoman’;5 3 ‰¯ÈË¥ Ï‡ ¥ˆÂ Â„¡‡È ÔÈÓÂ˜È¥‡ ˙Ú‡˙· ‰ÈÈ·‡¡Ó Ï‡ (74 22,1) ‘(the participants) take half of the hidden matza of the afikoman’; ÔÈÓÂ˜¥ Ï‡ ˙ÈÚ˙· ‰¯ÈË¥ Ï‡ ÒÂ Â„¡È 52

Note that in Classical Egyptian Judeo-Arabic there is evidence for both the feminine and the plural forms. See Blau 1980:159. 53 In the days of the temple, the meal began and ended with the lamb meat of the Passover sacrifice; in the period since the destruction of the temple, the meal begins and ends with the eating of matza; the last bit, whether of the Passover meat or of the matza, is called the afikoman, meaning the “last bit” (David Blumenthal, personal communication).

116

SPOKEN EGYPTIAN JUDEO-ARABIC (91 10a,12) ‘(the participants) take half of the matza of the afikoman.’ However, in the following examples the marker /bitå>/ (masc. sg.) modifies feminine nouns, which may indicate the frozen use of the marker in Cairene Judeo-Arabic dialect: Ú‡˙· ˙‡Î … ‰·¯ƒÂ ‰·¯ƒ ÏÎ ˙‡·¯ƒ Ú·¯ (3 14, 18–19 through 15,1–2) ‘each plague … was of four plagues’; ˙‡·¯ƒ ÒÓ¡ Ú‡˙· ˙‡Î … ‰·¯ƒÂ ‰·¯ƒ ÏÎ (3 15, 9–10 through 15,11) ‘each plague … was of five plagues’; Ú‡˙· ‡ˆÓ Ï‡ Â„Â¡‡È ÔÈÓÂ˜È¥‡ (3 22,13) ‘(the participants) hold the matza of the afikoman’ (see also chapter 5, pp. 151–52, 2.3). In Á·Âˆ Ï‡ Ú‡˙· ÚÓ˘ ˙ÈÈ¯˜ (3 3,15-16) ‘the reading of the morning Shema’ it is possible that /bitå>/ modifies ÚÓ˘ (masc.) and not ˙ÈÈ¯˜ (fem.), but in the following example, the masculine singular marker /bitå>/ modifies two (inanimate) nouns: ÔÈÁ ¯ÂÎ· Ï‡¯˘È Ú‡˙· ÍÂ¥˘ Ï‡Â È·¯Ú Ï‡· ‰„‚‰ Ï‡ ·‡˙Î ‡≤‡‰ (3 colophon) ‘This is the book of the Haggadah in Arabic and the section pour, both of them of Israel Be˙or Óanin.’ On the other hand, this nonagreement may be connected to the nature of the verbatim translation (see chapter 8, 10-3.16) and does not necessarily reflect spoken use. 2.2 The verb 2.2.1 The perfect 2.2.1.1 The pattern /fu>ul/, typical in Later Egyptian Judeo-Arabic (Hary 1992:280–85), is common in the spoken dialect as well: ˙¯˙ÂÎ /kutret/ (15 0-2,5) ‘(it) grew’; ıÂÏÂ¡ /xuluß/ (3 5,2) ‘was redeemed’; ÏÂ˜Â˙ /tu
uΩmu/ (3 8,9) ‘(they) grew mighty’; È˙¯Â·ÂÎ /kuburti/ (91 4b,15) ‘you (fem.) grew’; ˙¯Â˜Â¥ /fuil/ also occurs in Later Egyptian Judeo-Arabic and is reflected in the spoken variety: È¯‚ /giri/ (15 2b,18) ‘(he) ran’;

117

CHAPTER FOUR ˙ÈÈ¯‚ /giryit/ (15 3a,12) ‘(she) ran’; ÂÈË¡ /xi†yu/ (15 24a,5) ‘(they) sinned’; ÂÈÈÓ /nimyu/ (93 22,10) ‘(they) grew, increased’; ÈÂÂÈ˜ /araft/ (15 12a,15) ‘you (masc.) knew.’ 2.2.1.3.2 /ti/ for second person feminine singular: È˙¯Â˙ÂÎ /kuturti/ (91 4b,15) ‘you (fem.) increased.’ 2.2.1.3.3 /tu/ for second person plural: Â˙È‚ /g∑tu/ (15 27b,2; 27b,5; 27b,7; and more) ‘you (pl.) came’; Â˙ÚÓÒ ÌÏÂ /wi-lam simi>tu/ (15 28a,3) ‘and you (pl.) did not hear’; Â˙Ó„Ú (15 28b,7) ‘you (pl.) lost’ or ‘you (pl.) were deprived.’ 2.2.2 The imperfect Haim Blanc (1974) has divided Arabic dialects into A-dialects and N-dialects. The former are typically “eastern” with /akteb–nekteb/ for the first person singular—first person plural imperfect forms, whereas the latter are “western” with /nekteb–nektebu/ for the same forms. The division into A- and N-dialects is preferable to the geographical distinction, because the boundaries between east and west are unclear. In fact, thanks to Blanc, we know that in Egypt both the A-dialects and the N-dialects coexist. One of the most salient features of Egyptian Judeo-Arabic dialect is the use of the N-dialect /nekteb–nektebu/ forms, alongside the A-forms /akteb–nekteb/. 2.2.2.1 The first person singular imperfect in Later Egyptian JudeoArabic dialect may be of the /nekteb/ type as is the case in the N-dialects: ¥¯Ú ‰‡ /ana ni>raf/ (93 88,6 and more) ‘I know.’ 2.2.2.2 The first person plural imperfect in Later Egyptian Judeo-Arabic can be of the /niktébu/ type as is the case in N-dialects. It is especially surprising in Cairo:5 4 ÂËÚ /ni>†u/ (15 18a,3) ‘we give’; Â„¡‡ /naxdu/ (15 54

Behnstedt and Woidich (1985 2:210–12) have identified /niktib–niktíbu/ in

118

SPOKEN EGYPTIAN JUDEO-ARABIC 18a,3; 18a,12) ‘we take’; Â„ÂÚ˜ /ni<>udu/ (15 18a,4) ‘we sit’; ÂÏÂ˜ /ni<¥lu/ (15 22a,11) ‘we say’; ÂÂÎ /nik¥nu/ (3 2,7) ‘we will’; ÂÒÓµ /ni\misu/ (93 13,2) ‘we dip’; ÂÏÎ‡ /naklu/ (91 8b,1; 93 13,4) ‘we eat’; Â·¯˘ /ni¡rábu/ (93 13,7) ‘we drink’; ¬Â¯ÂÎ˘È /ni¡kuruk/5 5 (93 75,16) ‘we thank you.’ The paucity of examples of the /nekteb/ type suggests that a shift may be occurring in nineteenth-century Cairene Judeo-Arabic dialect toward a possible /akteb–niktébu/ pattern, but there are not enough data to establish this change. 2.2.2.3 The imperfect with /bi/ exists in Egyptian Judeo-Arabic in its regular mood: ÂÚ¯È· /biyir>u/ (15 25a,6) ‘(they) graze’; ÂÎÁÈ· Â‡ÎÂ /wi-kånu biyi˙ku/ (3 3,13) ‘and (they) were telling’; ‰ÓÏÚÈ· /bi>allimna/ (91 4b,2) ‘(they) teach us’; ¯∂‡· /banΩur/ (1302 4b,10) ‘I see.’ 2.2.3 The participle The participle ÔÈÏÎ‡Â /wakl•n/ (3 2,12) ‘have eaten (pl.)’ uses the colloquial root w-k-l. 2.2.4 The passive 2.2.4.1 The passive in the dialect employs the inform, among other forms: Ô¥„‡ /indafan/ (15 5a,4) ‘(he) was buried’; ¬Ï‰‡ /anhalik/ (15 18b,10) ‘I will be annihilated’; ÌÎÁ‡ /in˙akam/ (1302 2a,12) ‘(it) was decreed’; ‡ËÚÈ /yin>a†a/ (1302 2a,15) ‘(it) is given’; ÚÓÒ‡ /insama>/ (1302 2b,3) ‘(it) was heard’; ˙≤¡‡‡ /inaraf/ (1302 3a,1) ‘(it) was known’; «˙Á·‡ /inba˙as/ (1302 3a,2) ‘(it) was investigated’; Ì˙¡‡ /inxatam/ (1302 3b,3) ‘(it) was sealed’; ÏÚ‚‡ /inga>al/ (1302 3b,20) ‘(it) was issued’; Úˆ˙ /tinßini>/ (1302 4b,1) ‘(it) will be done’; ‡ÈÚ·‡ /inba>∑na/ (1302 5b,1) ‘we were sold’; Ï‡˜˙ /tin<ål/ (74 2,3) ‘(it) is recited’; ¬¥È‡ /infakk/ (91 2b,5) ‘(it) was redeemed’; Â·¯ƒ‡ /in∂arabu/ (3 14,6) ‘(they) were Egypt, but not in Cairo. It seems that this phenomenon is restricted in Cairo only to Judeo-Arabic. See also Hary 1992:278, 2.2.2, and 1994b:379. 55 Note that there is orthographical evidence here for /i-u/: /ni¡kuru/, whereas Mitchell (1978:262) reports /u-u/: /nu¡kur/ in standard Egyptian.

119

CHAPTER FOUR smitten’; ¯ÂÂ„‡ /indawwar/ (3 20,15) ‘(it) was turned.’ 2.2.4.2 Infrequently in the data from the ¡ur¥˙ the itfa>al form is used for passive constructions: ¯·¡˙‡ /itxabar/ (15 23a,15) ‘(it) was told’; ÂÏ‰·˙‡ /itbahalu/ (15 31b,9) ‘(they) were scared’; ˙˙˘˙‡ /it¡attit/ (1302 2a,18) ‘(it) was scattered’; ‡ÏÓ˙‡ /itmala/ (1302 3a,10; 4b,5) ‘(he) was filled’; Â‰„˙‡ /itnadahu/ (1302 3b,1) ‘(they) were called.’ 2.2.5 Verbal forms 2.2.5.1 Form I 2.2.5.1.1 It is common in the dialects as well as in spoken Egyptian Judeo-Arabic to observe the shift from form IV to form I. Thus, several verbs appear in the texts in the verbal form I rather than IV: ‰ËÚ />a†a/ (15 24b,20; 74 2,7) ‘(he) gave’; ÂÏÒ¯ /rasalu/ (15 4a,18) ‘(they) sent.’ 2.2.5.1.2 The verb ‘come’ The texts reveal some colloquial characteristics of this verb which also appear in standard Cairene: ˙È‚ /g∑t/ (15 3b,11) ‘I came’; ‰È‚ /g∑na/ (15 15b,4) ‘we came’; Â˙È‚/ÌÂ˙È‚ /g∑tu(m)/ (15 7a,6; 27b,2; and more) ‘you (pl.) came.’ The third person form in the ¡ur¥˙ is both ‰È‚ /geh/ (15 8b,1 24a,14; and more) and ‰‚ /gah/ (15 18b,13; 19a,2; and more) ‘he came’ as in standard Cairene. Furthermore, the texts feature specific Egyptian Judeo-Arabic traits: ˙˙‡‚ /gåtit/ (15 28a,2 and more) ‘she came’ (/gat/ in the standard variety) and Â‚ /gu/ (3 3,14) ‘(they) came’ (/gum/ in the standard variety). Egyptian Judeo-Arabic has also /ega/ ‘he came’ and /egu/ ‘they came,’ attested in earlier periods.5 6 2.2.5.1.3 The verbs ‘eat’ and ‘take’ Whereas the short forms /kal/ ‘he ate’ and /xad/ ‘he took’ are common in standard Egyptian dialect, spoken Egyptian Judeo-Arabic does not employ them in the texts of the ¡ur¥˙. Of course it is possible that the long forms appearing in the ¡ur¥˙ represent standard literary 56

See also Blanc 1974:215; Hary 1992:285 and the references there; Rosenbaum 2002c:126.

120

SPOKEN EGYPTIAN JUDEO-ARABIC Arabic and not the colloquial; however, it is telling that the standard Cairene short forms /kal/ and /xad/ are missing:5 7 ˙ÏÎ‡ /akalt/ (15 8b,5) ‘I ate’; ÏÎ‡ /akal/ (15 6a,3; 13,8; and more; 93 83,9) ‘he ate’; ˙ÈÏÎ‡ /akalit/ (15 22a,12; 93 84,8; 84,16; and more) ‘she ate’; ‰ÏÎ‡ /akalna/ (93 59,19; 62,12) ‘we ate’; Â˙ÏÎ‡ /akaltu/ (15 22b,3) ‘you (pl.) ate’; ÂÏÎ‡ /akalu/ (15 14b,20; 15a,13; and more; 3 2,2; 93 62,15) ‘they ate’; ˙„¡‡ /axadt/ (15 24b,1; 36a,20) ‘I took’; „¡‡ /axad/ (15 15b,17; 16a,12; 1302 2b,3) ‘he took’; ˙„¡‡ /axadet/ (15 11b,12; 14a,18; and more) ‘she took’; Â„¡‡ /axadu/ (15 18a,19; 18b,2; and more; 1302 1b,8) ‘they took.’ 2.2.5.1.4 The verb ‘go’ The verb /r¢å˙/ is usual in standard Egyptian dialect, occurring also in Egyptian Judeo-Arabic dialect: ˙ÁÂ¯ /ru˙t/ (15 18b,3) ‘I went’; ˙ÁÂ¯ /ru˙t/ (15 14a,11) ‘you (masc. sg.) went’; Á‡¯ /rå˙/ (15 2b,7; 4b,2; and more) ‘he went’; ˙Á‡¯ /rå˙et/ (15 4b,1; 4b,2; and more) ‘she went’; ÂÁ‡¯ /rå˙u/ (15 28a,10; 38a,16; and more) ‘they went.’ However, Cairene Judeo-Arabic dialect has also the unusual variant of /ar¢å˙/: Á‡¯‡ (15 6b,5; 9b,1; 9b,6) ‘he went’ (Blanc 1974:215; Rosenbaum 2002c:126). 2.2.5.1.5 Shift of final-hamza verbs It is common in the dialect for final-hamza verbs to shift to “defective” verbs (R3 =/ R3 =/y/). There is evidence for this shift in the various Egyptian ¡ur¥˙: ÂÏÓÈ /yimlu/ (15 2b,12; 3 1,1) ‘(they) fill’; ˙ÏÓ˙‡ /itmalet/ (15 0-2,12) ‘was filled.’ 2.2.5.2 Form V Sometimes it is difficult to distinguish in the dialect between verbal form V preceded by a prothetic alif and itfa>al: ‰ÓÏÚ˙‡ /it>allEmna/ (3 5,11) or /it>alEmna/ ‘we learned.’

57

See confirmation for the lack of the short forms in Blanc 1974:215 and Hary 1992:285–86.

121

CHAPTER FOUR 2.2.5.3 itfa>>al and itfa>al This is a common pattern in spoken Egyptian Judeo-Arabic as well as in standard Egyptian dialect, which can be used also for the passive (see above, p. 120, 2.2.4.2) in Later Egyptian Judeo-Arabic (Hary 1992:287): ‰˘Ó˙‡ /itma¡¡a/ (15 0-1,6) ‘(he) walked’; ·ÂÂ‡˙‡ /itu/ (15 10a,13) ‘(they) were gathered’; ÂË·¡˙‡ /itxabba†u/ (15 5b,4) ‘(they) fumbled about’; ˜ÓÁ˙‡Â /w-it˙amaet/ (93 2,3) ‘four hours’; Ï‡ ¥ÈÁˆÓ ˙ÈÒÓ¡ ‰¯ÂË /xamset/ (93 891,3 and more) ‘five books of the Torah’; ˙È˙Ò ‰˘Ó Ï‡ ÌÈ¯„Ò /sittet/ (93 90,3–4 and more) ‘the six Mishnah sections’; Ì‡ÈÈ‡ ˙ÈÚ·Ò /sab>et/ (93 90,12 and more) ‘seven days’; ¯Â‰˘‡ ˙ÈÚÒ˙ /tis>et/ (93 92,7) ‘nine months’; ‰¯ÂË Ï‡ ˙‡ÓÏÎ ˙È¯˘Ú />a¡áret/ (93 92,6–7 and more) ‘the ten commandments (words) of the Torah’; È≤Ï‡ ˙ÈÚ‡Ò /så>et/ (93 18,7) ‘the time in which’; ‡˘ÂÓ ˙ÈÚÈ¯˘ /¡ari>et/ (93 62,20) ‘the law of Moses’; Â‚‡ÂÂÓ‡ ˙È˘Âµ‡Î /a\u¡et/ (93 77,14) ‘like the multitude of its waves’; Ï‡¯÷È ˙ÈÚÓ‚ /gamá>et/ (93 79,8) ‘the house of Israel’; and possibly ‰¯˜Â¥ Ï‡ ˙ÈÈ¡¯Ò /sarxiyet/ (93 79,3) and ÔÈÎÒÓ Ï‡ ˙ÈÈ¡¯Ò (93 79,3–4) ‘the cry of the poor’; ‰ÈÈÏ‡„ Ï‡ ˙ÈÈ‰Î‡¥

122

SPOKEN EGYPTIAN JUDEO-ARABIC /fakhiyet/ (93 82,16) ‘the fruit of the vine.’5 8 2.3.2 The dual and the sound plural 2.3.2.1 The n¥n of the dual and of the masculine sound plural is preserved with pronominal suffixes and in the first term of an i∂åfa in Late Egyptian Judeo-Arabic dialect as well as in standard Egyptian: Â˙¡Â‡ ÔÈ„‡ /id∑n uxto/ (15 3a,15) ‘his sister’s hands’; ÌÂ‰È˙‡ /itn∑nhum/ (15 24a,11) ‘both of them’; ‰Ï‡‚¯ Ï‡ ÔÈÏ‚¯Â /wi-rigl∑n ir-riggåla/ (15 3a,20) ‘the men’s feet’; ÂÈÂÂ„Ú />aduww•no/ (15 4b,1) ‘his enemies’; ‡‰È¯∂‡ ÏÎ ÔÈÈÚ />en∑n kullE naΩr•nha/ (1302 2b,16) ‘the eyes of all who saw her’; ¬ÏÓ Ï‡ ÔÈÈÚ È¥ /fi >en∑n il-malik/ (1302 2a,15–16) ‘in the king’s eyes’; ‡È‰¯‡Î5 9 (93 23,9) /karh•nna/ ‘the people who hate us’; ‰ÈÈ·‚‡ ‰„‡·Ú ÔÈ„·‡Ú />abid•n >ibåda agnabiyya/ (3 5,17–18) ‘worshippers of foreign idols.’ 2.4 The adjective 2.4.1 The Egyptian colloquial adjective pattern /fu>ayyal/ appears in the ¡ar˙: ÔÈÚ‡È¥Â¯ /rufayya>•n/ (15 25a,7; 25a,9; 25a,12; 25a,13; 25b,12; 25b,13; 25b,14; 25b,19; 25b,20; 26a,4) ‘lean, thin’; ÔÈÏ˛È¸ÈÏÂ‡ /ulayyil•n/ (15 25a,7) ‘few.’ 2.5 The numerals The texts of the ¡ur¥˙ exhibit many references to the forms of the numerals in Late Egyptian Judeo-Arabic dialect. 2.5.1 1–10: ÔÈ˙‡ /itn∑n/ (15 0-1,2) ‘two’ with the shift of the interdental /®/ to the stop /t/. 2.5.2 11–19: ¯˘‡„Á /˙idaa¡ar/ (74 23,17) ‘eleven’ with the /ar/ ending typical of the Egyptian dialect and with a possible weakening of the 58

The orthography in the last three examples probably indicates /i/, as is mentioned in the section. Had the pronunciation been /a/ (/sarxiyat/ and /fakhiyat/), it is possible that the ¡ar˙an would have written one yod only. 59 This phenomenon is less obvious in the spelling in another manuscript, ‰È‰¯Î (3 9,5), where the two n¥ns collapse into one.

123

CHAPTER FOUR pharyngeal fricative />/, or simply consonant deletion (see also above, p. 104, 1.6); ¯˘Ú ÔÈ˙‡ /itn∑n>a¡ar/ (15 5a,13) ‘twelve’ with the shift of the interdental to a stop.6 0 More typical Egyptian forms are ¯˘‡˙‡ /itnaa¡ar/ (15 19b,9; 1302 2b,9) ‘twelve’; and ¯˘Ú ˙Ú·¯‡ /arba>ata>¡ar/ (93 1,1) ‘fourteen.’ 2.5.3 20–90: ÔÈ˙Ï˙6 1 /talat•n/ (15 0-1,1) ‘thirty’ with the shifts of the long /å/ to short /a/ and the interdental /®/ to the stop /t/; ÔÈÓ˙6 2 /taman•n/ (15 0-1,9) ‘eighty’ with the same shifts as the previous example, ÔÈ˙Ï˙. 2.5.4 100–900: ÔÈ˙ÈÓ /mit∑n/ (3 15,17) ‘two hundred’; ‰È‡Ó ÒÓÂ¡ /xumsumiyya/ (15 0-1,17) ‘five hundred’; ‰È‡ÓÓ˙ /tamanmiyya/ (15 0-1,1) ‘eight hundred’ with short vowels and the shift of the interdental /®/ to the stop /t/. According to the ¡ur¥˙, Cairene Jews still used older forms in the numerals in the eighteenth and nineteenth centuries: ‰Ò ˙ÈÈ‡ÓØ˙È‡Ó ÚÂ·¯Â‡ /urbu>miyya/ (3 6,17; 91 3a,10) ‘four hundred’ 6 3 as opposed to the standard Cairene /rub>umiyya/.6 4 2.5.5 1,000 and upwards: ¥‡Ï‡ ˙¯˘Ú />a¡rattalåf/ (1302 3a,17) ‘ten thousand.’

60

In another example the supralinear dot above the tav indicates a literary form: ¯˘Ú È«˙‡ (15 27b,8) ‘twelve.’ 61 In ÔÈ«˙‡Ï«˙ (15 5a,14) ‘thirty,’ the orthography represents the literary variety. 62 The form ÔÈÈÓ˙ /tmany•n/ (15 0-1,11) ‘eighty’ with the glide /y/ is unusual. 63 The spelling of ˙È‡ÓÂÚ·¯‡ (15 2a,4) does not demonstrate standard Cairene pronunciation of /rub>umiyya/. 64 See also Blanc 1974:217. In fact, today Cairene Jews still use this nonstandard form, as Rosenbaum (2002c:126) has reported.

124

SPOKEN EGYPTIAN JUDEO-ARABIC 3. Syntax 3.1 Demonstrative pronouns In the ¡ar˙, the demonstrative pronoun usually follows the noun it refers to, as is usual in colloquial Egyptian Judeo-Arabic: ‰„ ‰Ïµ Ï‡ (15 26a,11) ‘this famine’; ‰„ ¯ÈË¥ (3 18,10) ‘this matza’; ‰„ ¯¯Ó (3 18,19) ‘this bitter herb’; ÈÏÂ„‰ Ì‡ÏÎ Ï‡ (15 3a,13; 10b,10–11) ‘these words’; ÈÏÂ„‰ Èƒ‡¯‡ Ï‡ ÏÂÎ (15 6a,8–9; 6a,11) ‘all these lands’; Ï‡Â ÈÏÂ„‰ ‰Ï‡‚Â¯ (15 18a,10) ‘and these men’; ÈÏÂ„‰ Ì‡ÈÈ‡ Ï‡ (1302 1b,5) ‘these days’; ÈÏÂ„ ˙‡ÓÏÎ ˙‡Ï˙ (3 17,18–19) ‘these three words’; Ï‡ ÏÎ ÈÏÂ„‰ ·È‡‚Ú (3 19,17–18) ‘all these miracles.’ Even when using standard literary demonstrative pronouns, the colloquial word order is preserved: ‰„‡‰ ÌÂÈ Ï‡ (15 7a,16) ‘this day’; ‰„‡‰ ˙È· Ï‡ (15 24b,8) ‘this house’; È„‡‰ ‰˜Èƒ Ï‡ (15 28a,2) ‘this anguish’; È„‡‰ ‰È„‰ Ï‡ (15 29a,17) ‘this gift’; È„‡‰ √¯‡ Ï‡ (15 6a,7; 35b,7) ‘this land’; È„‡‰ ‰ÏÈÏ Ï‡ (3 2,9) ‘this night’; È„‡‰ ‰ÈÂ„ Ï‡ (3 4,3) ‘this world’; È„‡‰ È„‡·Ú Ï‡ (3 4,17) ‘this worship.’ This word order follows regular colloquial Egyptian use, for example, /ill∑la di/ ‘this night.’ At the same time, and perhaps as part of the verbatim translation, this word order also translates literally the Hebrew demonstrative pronoun word order (‰Ê‰ ‰ÏÈÏ‰ ‘this night,’ for example; see chapter 8, pp. 244–48, 6-4). 3.2 The definite article Sometimes the definite article may be dropped from the noun in a noun-adjective phrase, as in ‰„Â„ÓÓ Ï‡ Ú‡¯„ È¥Â ‰„È„˘ Ï‡ „È È¥ (3 2,20–3,1)‘with a mighty hand and an outstretched arm’ (See also chapter 8, pp. 268–69, 9-1.7). This may reflect dialectal behavior (Hary 1992: 30–33). 3.3 Negation 3.3.1 The use of /ma … ¡/ to negate the imperfect is regular in spoken Egyptian Judeo-Arabic: ˘¥¯ÚÈ ‡Ó /mayi>raf¡/ (3 4,10) ‘(he) does not know’; ˘Â˜¯¥È ‡Ó /mayifrE<¥¡/ (74 2,11) ‘(they) do not separate’; ‡Ó

125

CHAPTER FOUR ˘ÏÂ˜È /may
See above, pp. 94–95; see also chapter 5, pp. 141–43, and chapter 8, pp. 215–17, 4-2.1, and the references there for further explanations. 66 See chapter 8, 4-1.3–4-1.5, and the references there for further explanations.

126

SPOKEN EGYPTIAN JUDEO-ARABIC ‘and not through a seraph’; and in isolation: ¬„È·ÚÂ È„ÈÒ ÌÏ ‰ÈÏ‡ ÂÏ‡˜Â ÏÎ‡ Â¯˙˘ÈÏ Â‚ (15 27b,5) ‘And they told him, “No, sir, your servants have come to purchase food” ’; Â˙È‚ √¯‡ Ï‡ ˙¯ÂÚ Ô‡ ÌÏ ÌÂ‰ÈÏ‡ Ï‡˜Â Â¯Â∂˙Ï (15 27b,7) ‘And he said to them, “No, you have come to see the land in its nakedness” ’; ÌÂ˙‡ ÔÈÒÒ‚ Ô‡ ÔÂÚ¯¥ ˙‡ÈÁÂ ÌÏ Ô‡Î‡Â (15 27b,14) ‘And if not, by Pharaoh’s life, you are indeed spies.’ 3.4 Agreement: Verbs 3.4.1 The verb preceding its subject agrees with its number, which is not uncommon in Arabic dialects in general. In other words, a plural verb may come before a plural subject: Ï‡Ó‚ Ï‡ ÂÓ˙ ®È≤Ï‡Î© Ô‡ÎÂ (15 3a,4) ‘and when the camels had finished drinking’; Á¯ËÓ Ï‡ Ï‰‡ ÂÏ˜ÒÂ (15 6a,13) ‘and the men of the place asked’; ‰Â·‡ „È·Ú Â¯¥Á (15 6b,7) ‘his father’s servants dug’; ¯‡¯‚ Ô‡ÈÚÂ¯ ÂÓˆ¡˙‡Â (15 6b,14) ‘and the herdsmen of Gerar fought’; ‰˙‡‰·‡ ÂÏÎ‡ È≤Ï‡ (3 2,1–2) ‘which our fathers ate.’ This is also the case before inanimate subjects, as in Â‡ÎÂ ÁÏ˘Â˙Ó Ì‡È‡ ÏÂÎ (15 0-1,11) ‘all the days of Metushelah were …’;6 7 Ì‡È‡ Ï‡ ÂÏ ÂÏÂÂË (15 6a,16) ‘the days had passed for him.’ 3.4.2 The gender of plural verbs (including participles) is usually masculine, even when it refers to feminine human beings, as is usual in Arabic dialects: ÔÈÈ‡ÏÓ Ï‡ /il-mallay•n/ (15 2b,9) ‘the women who draw water’ and ‰ÈÓ ÂÏÓÈÏ ÔÈ‚¯‡¡ „Ï· Ï‡ Ï‰‡ ˙‡·Â /wi-banåt ahl al-balad xarg•n li-yimlu mayya/ (15 2b,11-12) ‘and the townsmen’s daughters come out (masc. pl.) to draw (masc. pl.) water.’ The same is seen in Esther in the gender of the verb in ÂÏ„‰·ÈÏ ‡Ò Ï‡ (1302 2a,2) ‘the women will treat (masc. pl.) … contemptuously.’ 3.5 Agreement: Nouns 3.5.1 For the most part, masculine plural adjectives modify nonhuman plural nouns even if they are feminine: Ï‡Â ¯∂Ó Ï‡ ÔÈÒÂÁ ˙‡¯˜· Ï‡ Ô‡ÓÂÒ (15 24a,9–10) ‘good-looking and sturdy cows’; ÔÈ¯¡Â‡ ˙‡¯˜· 67

In this example /kull/ ‘all’ cannot be considered the subject; it must be /ayyåm/ ‘days,’ because otherwise the plural verb /kånu/ could not be explained.

127

CHAPTER FOUR (15 25a,7) ‘other cows’; ÔÈ·ÈÈËÂ Ô‡ÓÂÒ … ˙‡Ï·Ò (15 25a,11) ‘ears of grain … solid and healthy’; ÔÈ¯ÈÈµÓ È‡Â‡ (1302 1b,11) ‘various vessels.’ Sometimes, however, a feminine plural adjective modifies the nonhuman feminine plural noun, as in ÌÁÏ Ï‡ Ô‡ÓÂÒÂ ¯∂Ó Ï‡ ˙‡ÒÂÁ ˙‡¯˜· (15 25a,5–6) ‘good-looking and sturdy cows’; however, this may not necessarily represent colloquial use, but rather a verbatim translation (chapter 8, pp. 284–85, 10-3.13). 3.6 Agreement: Pronouns 3.6.1 The dual agreement in the spoken variety is lost, becoming plural, as is expected in the dialects: ÔÂ¯‰Â ‰ÒÂÓ ÌÂ‰ È≤Ï‡ (93 88,11) ‘who they are Moses and Aaron.’ 3.6.2 As seen above, the gender of the plural suffixed pronouns is usually masculine, even when referring to feminine nouns, as is common in the dialect: Ì‰ÏÂ˜· Ì‰Ó‡„˜ Ì‰Ê‡Â‚‡ ÂÏ„‰·ÈÏ ‡Ò Ï‡ /in-niså li-yibahdilu agwåzhom uddåmhom bi-<ølhom/ (1302 2a,2) ‘the women will treat their (masc. pl.) husbands contemptuously in their (masc. pl.) presence, saying (masc. pl.) …’ Similarly, in Ì‰˜‡ÂÊ ‡ËÚÈÂ ‡Ò Ï‡ ∂¥‡Á /˙åfiΩ in-niså wi-yin>a†a zawå
SPOKEN EGYPTIAN JUDEO-ARABIC your eyes see’; ¯ˆÓ ÔÈÈ‡‚ Ï‡ Ï‡¯˘È „‡ÏÂÂ‡ ÈÓ‡Ò‡ ÏÂ„Â (15 32b,20–33a,1) ‘these are the names of the children of Israel who have come to Egypt’; ‰ÈÏ‡ ÔÈ·È¯˜ Ï‡Â (1302 1b,19) ‘his closest (advisers)’; ÔÈÒÏ‡‚ Ï‡ ‰ËÏÒ Ï‡ È¥ ‡ÏÂÂ‡ (1302 1b,21) ‘who occupy the first (place) in the kingdom’; Ô‡ËÏÒ Ï‡ ÔÈÓ„‡¡ (1302 2a,12) ‘the king’s servants’; ‰Á‡ ÔÈÒÓµÓ (3 2,10–11) ‘we dip’; ÔÈ·¯‡˘Â ÔÈÏÎ‡Â ‰Á‡ (3 2,12) ‘we have (just) eaten and drunk’; ÔÈ¥¯‡Ú ‰ÏÂÎ (3 3,5) ‘all of us know.’ 3.7.2 Some colloquial adverbs take the “frozen” accusative case in Egyptian Judeo-Arabic dialect, as is common in many other Arabic dialects: ÔÏÈ‡˜ (3 5,10) ‘saying’; Ô„‚ (1302 1b,18) ‘very.’6 8 3.7.3 Note the use of a yod as a marker for the “defective” /-in/ that becomes /-i/ in the dialect: È‡Â‡ /awåni/ (1302 1b,10) ‘vessels’ (at a¡xåß/ (74 2,8) ‘four sons’ is a typical example of colloquial use of the numerals followed by the counted noun. In Ï‡Ó‚ ‰¯˘Ú />a¡ára gimål/ (15 2b,6) ‘ten camels,’ however, the i∂åfa structure of numeral is missing. On the other hand, the orthography in the latter example may not be used as solid evidence for the pronun68

Note that the “frozen” accusative case in the Hebraized orthography is usually marked phonetically with a nun. See Hary 1992:297–98, 4.1.4; 1996c:732, ii; Khan 1991:230, b; Wagner 2008:59. 69 The term /ma˙b¥b/ means ‘gold piece’ and in the Ottoman period ‘sequin’ (see Wehr). However, it also means ‘gold pound’ or ‘pound note’ in Egyptian Arabic (see Hinds and Badawi).

129

CHAPTER FOUR ciation of the example, and the i∂åfa is probably used in this example in the spoken variety: />a¡árat gimål/. 3.8.3 The counted noun following the numbers may be in the plural: ÌÂ‰˙‡¯·ÂÎ ¯˘Ú ÔÈ˙‡ (15 5a,13) ‘twelve of their nobles’; Â„‡ÏÂ ¯˘Ú „Á‡Â (15 16a,13) ‘and his eleven children’; ¬„È·Ú ¯˘Ú È«˙‡ (15 27b,8) ‘twelve of your servants’; ÔÈÒ ÔÈ˙Ï˙Â ‰È‡ÓÓ˙ (15 0-1,1) ‘eight hundred and thirty years.’ In other examples, however, the counted noun following the numbers is in the singular, following Classical Arabic structure: ‰ƒ¥ ÔÈ¯˘Ú· (15 22b,6–7) ‘for twenty pieces of silver’; ‰Ò ÔÈ˙Ò (15 5b,12) ‘sixty years’; ‰Ò ˙È‡ÓÓ˙ (15 0-1,3–4) ‘eight hundred years’; ‰ƒ¥ Ï‡˜˙Ó ¥‡Ï‡ ˙¯˘ÚÂ (1302 3a,17) ‘and ten thousand miskals [a type of weight] of silver’; ‰È„Ó ÔÈ¯˘ÚÂ ˙Ú·ÒÂ ˙È‡Ó (1302 1b,3) ‘a hundred and twenty-seven provinces’; ÌÂÈ ÔÈ‡Ó˙Â ˙ÈÈÓ (1302 1b,6) ‘a hundred and eighty days’; ‰Ò ÔÈÚ·Ò (3 3,19) ‘seventy years’ (see also chapter 8, pp. 299–301, 12-1.4–12-1.7). 3.9 Interrogative pronouns As mentioned above (pp. 114-16, 2.1.6), the word order of the interrogative pronouns in Cairene Judeo-Arabic dialect is preposed: È„ ‡ÏÚ È¡„¯ÓÏ ‰ÈÈ¯·ÎÂ ¯‡˜Â‡ Úˆ‡ ˘È‡ (1302 4b,17) ‘what honor and respect has been conferred on Mordecai for this?’; ‰È¯Î· ‡ÈÏ ‡„ ˘ÈÏÂ (15 5b,20–6a,1) ‘of what (use) is my birthright to me?’; ¯„˜‡ ¥ÈÎÂ ¯∂‡ (1302 6a,2–3) ‘and how can I see?’ This is in contrast to standard Cairene, where the word order is postposed: /ismak ∑h/ ‘what is your name?’ 3.10 Asyndetic embedded clauses It is common in Egyptian Judeo-Arabic dialect as well as in standard Egyptian dialects to have asyndetic embedded clauses: Ï‡ÒÈ ˘¥¯ÚÈ ‡Ó /mayi>raf¡ yisal/ or /mayi>raf¡ yisa¡¡u/ (3 1,4) ‘(they) are not going to eat.’

130

SPOKEN EGYPTIAN JUDEO-ARABIC 4. Lexicon 4.1 Prepositions: Several prepositions employed in the ¡ar˙ are reflections of Egyptian Judeo-Arabic colloquial use: /•d• dialect with the same meaning: /˙ada/ ‘with’; /˙idåya/ ‘with me’; /˙adåna/ ‘with us’ (Behnstadt and Woidich 1994, 4:81) and probably comes from Classical Arabic /˙idå
70

See also Blanc 1974:217 and Rosenbaum 2002b:38.

131

CHAPTER FOUR 4.3 Verbs 4.3.1 Some typical colloquial verbs are: /ga/ ‘come’ with its Cairene Judeo-Arabic forms (see p. 120, 2.2.5.1.2); /akal/ ‘eat,’ including its participle ÔÈÏÎ‡Â /wakl•n/ (3 2,12) ‘have eaten (pl.)’ (p. 119, 2.2.3); /axad/ ‘take’ (pp. 120–21, 2.2.5.1.3); /rå˙/ and /arå˙/ ‘go’ (p. 121, 2.2.5.1.4). 4.3.2 Furthermore, the employment of the verb ¥‡˘ /¡åf/ (15 3a,14) ‘(he) saw’ is typical of dialectal use. 4.3.3 The following typical colloquial Egyptian verbs occur in the ¡ar˙, as evidence of Later Egyptian Judeo-Arabic dialect: Â¯ƒÁ„ /da˙∂aru/ (15 10a,13) ‘(they) caused to roll down’ (Hinds and Badawi 1986:279); È˘¯Î /kara¡ni/ (3 26,10) ‘(he) drove me away’; ˙ÈÎÊ /zak∑t/ (3 3,19) ‘I gained’ (see chapter 5, p. 153, 2.7). 4.3.4 Some verbs appearing in the ¡ar˙ demonstrate a different meaning than standard Egyptian, which may carry over to the dialect: ÂÓ˘Ú˙‡ /it>a¡¡imu/ (3 23,17) ‘trust in.’7 1 4.4 Nouns 4.4.1 The following typical colloquial Egyptian nouns occur in the ¡ar˙, as evidence of Later Egyptian Judeo-Arabic dialect: Ì‡ÊÂ¡ /xuzåm/ (15 3a,5) ‘nose ring’ (standard Arabic /xizåm/); Ï˜Â˙ /tu∑¡/ (15 6a,2; 8a,2; 10a,6; and more; 93 1,4) ‘bread’; ËÒ ‡Ï· /bala sEnt/ (15 10b,13) ‘for nothing; for free’; ÈÊÂ‚ /gøzi/ (15 11a,16; 11a,18; 11b,20; and more) ‘my husband’; Â˙‡¯Ó /mråto/ (15 21a,13; 23-1b,2; 23-1b,16; and more) ‘his wife’; Ô·Ï /laban/ (15 36b,17) ‘milk’; ‰‚‡Á /˙åga/ (3 4,15) ‘nothing’; Â„Â /widno/ (3 24,12) ‘his ear’; ‰·‡· /båba/ (74 25,14 and more) ‘daddy’; ‰ËÂ¥ /f¥†a/ (74 21,4) ‘towel’; ·Â·ÁÓ /ma˙b¥b/ (74 24,14 and more; 93 83,4 and more) ‘pound note.’ 71

Hinds and Badawi (1986:580) have defined /it>a¡¡im/ as ‘be hopeful, be made hopeful.’

132

SPOKEN EGYPTIAN JUDEO-ARABIC 4.4.2 Some nouns show a different meaning in the ¡ar˙, which may be carried over from the dialect: Â„¥ /fadw/ (3 21,12) ‘redemption, liberation’;7 2 ‰Ú¥Â„ /duf>a/ (3 23,18) ‘stumbling.’

72

Hinds and Badawi (1986:645) have defined /fadw/ ‘food or money given to the poor in compensation for the omission of certain religious duties’ as part of Islamic law; however, /fidå
133

CHAPTER FOUR 5. Summary of Colloquial Features of Spoken Egyptian Judeo-Arabic Previous scholarship investigated several characteristics of Late7 3 and Modern7 4 Egyptian Judeo-Arabic dialect. As mentioned at the beginning of the chapter (pp. 91–92), the greatest obstacle in the past has been the paucity of material that can reveal Egyptian Judeo-Arabic dialectal features. The texts of the Egyptian ¡ur¥˙, analyzed in this volume as well as in Hary 2009, confirm and contribute to the study of these characteristics. In fact, these texts can be of tremendous help in establishing major linguistic changes over the last three centuries in Egypt. 7 5 As is common in other Jewish religiolects (chapter 1, pp. 21–24), many of the dialectal characteristics of Egyptian Judeo-Arabic incorporate Hebrew and Aramaic elements and are archaic forms of the majority dialect; some of them reflect migrated dialectalism. The following is a summary of selected various features:7 6 (i) (ii) (iii) (iv) (v) (vi) 73

The preference for the vowel /u/ (pp. 101–2, 1.1.4.1): ·ÚÂ˘ /¡u>b/ ‘people’; ˙˜ÂÂ ÏÂ„ /dulwa /å/ in words originating in Hebrew (p. 102, 1.1.5): ‰‡ÏË· /ba†alå/ ‘in vain’; Ô‡ÁÏÂ˘ /¡ul˙ån/ ‘table.’ The shift /å/ > /a/ preceding />/, /˙/, or /h/ (1.1.6): Ú˙· /bita>/ ‘genitive marker, of (masc.)’; ‡‰Ï‡ /ilahna/ ‘our God.’ The weakening of the voiced pharyngeal fricative />/ (p. 105, 1.6): ¯˘‡„Á /˙idaa¡ar/ ‘eleven’; „‰‡ /ahd/ ‘pact.’ The existence of the voiceless bilabial stop /p/ (p. 107, 1.9): Ï‡ ÌÈ˘µÏÙ ‘the concubines’; ˘„ÏÙ ‘Pildash.’ Emphatization in words originating in Hebrew (p. 108, 1.11.4):

See, for example, Tomiche 1968:1179–81; Blanc 1974, 1981, 1985; and Hary 1992, 1994b, 1996c, 1997a. 74 See, for example, Rosenbaum 2002b, 2002c, 2002d. 75 I am planning on writing in the near future a study about tracking these linguistic changes. 76 The selected examples do not include manuscript citations for ease of reading; however, the reader is referred back to the paragraphs in the chapter where they are properly cited.

134

SPOKEN EGYPTIAN JUDEO-ARABIC ı¥¯Î /kar¢faß/ ‘greens for the Passover seder’; ‰È¯∂Ú />aΩar¢ya/ ‘Azarya.’ (vii) Deemphatization (pp. 108–9, 1.12): ÒÂ /nuss/ ‘half’; ¯„¡‡ /axdar/ ‘green.’ (viii) The occasional shift /¡/ > /s/ (p. 110, 1.13.4): ÒÂ„˜ // (pp. 116–17, 2.1.7). (xiv) The verbal pattern /fu>ul/ (p. 117, 2.2.1.1): ıÂÏÂ¡ /xuluß/ ‘was redeemed’; È˙¯Â˙ÂÎ /kuturti/ ‘you (fem.) increased.’ (xv) The use of /nekteb–nektébu/ imperfect verbs (pp. 118–19, 2.2.2): ¥¯Ú ‰‡ /ana ni>raf/ ‘I know’; Â„¡‡ /naxdu/ ‘we take’; Â„ÂÚ˜ /ni<>udu/ ‘we sit.’ (xvi) Unique forms of the verb ‘come’ (p. 120, 2.2.5.1.2): ˙˙‡‚ /gåtit/ ‘she came’; Â‚ /gu/ ‘they came.’ (xvii) The appearance of the “long” forms of ‘eat’ and ‘take’ (pp. 120–21, 2.2.5.1.3): Â˙ÏÎ‡ /akaltu/ ‘you (pl.) ate’; Â„¡‡ /axadu/ ‘they took.’ (xviii) A special form of the verb ‘go’ (p. 121, 2.2.5.1.4): Á‡¯‡ /ar¢å˙/. 135

CHAPTER FOUR (xix) Roots originating in Hebrew may receive Arabic morphological patterns, creating mixed forms (chapter 5, p. 153, 2.7): ˙ÈÎÊ /zak∑t/ ‘I gained’ and ‰ÎÊ˙‡ /itzaka/ ‘(he) gained’; ¯„‡„Ò /sadådir/ ‘sections of the Mishna’ (chapter 5, p. 153, n. 24). (xx) The use of archaic numeral forms (p. 124, 2.5.4): ˙È‡Ó ÚÂ·¯Â‡ /urbu>miyya/ ‘four hundred.’ (xxi) The use of an alternative feminine ending morpheme in the first term of an i∂åfa, /-et/ or /-it/ (2.3.1): Ì‡ÈÈ‡ ˙ÈÚ·Ò /sab>etayyåm/ ‘seven days’; ‡˘ÂÓ ˙ÈÚÈ¯˘ /¡ari>et m¥sa/ ‘the law of Moses.’ (xxii) The possible extensive use of the negation particle /lam/ (3.3.3): with perfect verbs: ˙ÏÂ˜ ÌÏ ‘I did not say’; Ô‡Î ÌÏ ‘(he) was not’; with imperfect verbs: ¯·¡˙ ÌÏ ‘(she) will not inform’; ÌÏ È˘˙¡ ‘we will not be abashed’; with nouns: ÌÂ˙‡ ÔÈÒÒ‚ ÌÏ Ô‡ ¥¯Ú‡Â ‘and I will know that you are not spies’; with pronouns: ÂÏ ÌÏÂ ÈÏ ‘to me and not to him’; with prepositions: Â„ÁÂÏ ¬ÏÓ Ï‡ ‡ÏÚ ÌÏ ‘not only against the king’; in isolation: È„ÈÒ ÌÏ ‰ÈÏ‡ ÂÏ‡˜Â ‘And they told him, “No, sir.” ’ (xxiii) The use of the adverb /kamåna/ ‘also.’ (xxiv) The use of words originating in Hebrew: ÌÈ˘µÏÙ /pila\¡•m/ ‘concubines’; ¯¥Ò /sefer/ ‘scroll’; Ï¯‡Ú />aril/, plural ÔÈÈÏ¯‡Ú />ariliy•n/ ‘Christian man/men’; ˙‡·˘ /¡abbåt/ ‘Saturday, Shabbat’; ‰ÙÂÁ /˙uppa/ ‘marriage’; ‰¯Â˙ /tøra/ and ‰¯ÂË /†øra/ ‘Torah’; ·È¯ÈÚ />∑r∑b/ ‘evening’; ÔÈ¥‚ /gefen/ ‘vine’; ÒÂ„˜ // ‘the prayer of the Shema’; ıÈÓÁ /˙am∑ß/ ‘unleavened food’; ˙ÈˆÂ¯Á /˙arøß∑t/ ‘Óaroset’; ‰ˆÓ /maßå/ ‘matza’; ı¥¯Î /kar¢faß/ ‘Karpas, greens for the Passover seder’; ¯¯ÂÓ /morer/ ‘bitter herbs’; ÔÈÓÂ˜È¥‡ /afikum•n/ ‘afikoman.’7 7 77

For more examples, see chapter 5, especially pp. 156–59.

136

CHAPTER FIVE ADDITIONAL LINGUISTIC ISSUES OF THE ¡arh¢ TRADITION This chapter examines the use of pseudocorrections in Judeo-Arabic in general and in the Egyptian ¡ur¥˙ in particular. It demonstrates how both hypercorrections and hypocorrections, which are two different types of pseudocorrections, are used in the texts, and it discusses the implications of the standardization of pseudocorrections in Egyptian Judeo-Arabic for both the literary and the spoken varieties. Furthermore, the chapter develops a theoretical model for the use of Hebrew and Aramaic components in Jewish religiolects in general and in Judeo-Arabic in particular. Using Uriel Weinreich’s work on languages in contact, this model demonstrates how components are transferred to Judeo-Arabic from two different directions, supplying the reasons for each direction and giving examples for each direction from the various Egyptian ¡ur¥˙. The chapter ends with a discussion of Hebrew and Aramaic lexical items employed in the Egyptian ¡ur¥˙.1 Issue One: Pseudocorrections Pseudocorrections are forms that “result from speakers’ and writers’ desire to speak and write a more prestigious variety and to avoid stigmatized forms” (Hary 2007:277). The prevalence of these forms in Judeo-Arabic texts results from a traceable social process. At times Judeo-Arabic authors want to embellish their writings with prestigious Classical Arabic forms, but their competence in Classical Arabic is 1

Some of the material in this chapter was published in a different form in Hary 1999 and 2007.

CHAPTER FIVE insufficient. So, they “correct” certain forms although these corrections are not necessary according to Classical Arabic norms (Blau 1970). Ultimately, these “corrected” forms create mechanisms for language change.2 I described in detail (Hary 2007:275) the process of the creation of pseudocorrections, also termed “overcorrectness”: Whenever a variety of a language with social, religious, economic, or other prestige comes into contact with a variety without such prestige, speakers and writers of the latter will, at times, try to use forms of the former even if the forms are unnecessary in that linguistic environment. They will change or “correct” forms of the prestigeless variety and utilize the forms of the language with the prestige. However, sometimes these speakers do not have sufficient knowledge of the prestigious variety. Thus, they change or “correct” forms that do not need to be changed even according to the prestigious variety, and may arrive at forms that are “too corrected” (hence, hypercorrections) or “halfway corrected” (or “not corrected enough”; hence, hypocorrections). Furthermore, speakers and writers may sometimes mix forms. Such pseudocorrections stem from the desire of the speakers to “decorate” themselves with forms of the prestigious variety and to avoid the dominant usage of the prestigeless variety, often out of overself-denial in order to gain prestige in their speech or writings. At times, such corrections can become standardized in the prestigeless variety, thus contributing to language change.

The differences between hypercorrection and hypocorrection are explained in detail in Hary 2007:277–78. John Ohala (1993) has discussed both forms in regard to sound change. A famous example for this kind of sound change is the emergence of the umlaut in the history of Old English (Yu 2006:527). 2

See Labov 1972:178–80. Examples of a linguistic change caused by a pseudocorrection are the use of hamza in Arabic (Hary 2007:279), or the use of /lam/ in Egyptian Judeo-Arabic (chapter 4, pp. 94–95, and chapter 8, pp. 215–17, 4-2.1), although there are scholars who interpret this phenomenon differently.

138

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION As mentioned in chapter 4, pp. 94–95, pseudocorrections are one of several components in the texts of Judeo-Arabic ¡ur¥˙. Pseudocorrections, however, are particularly important because it is only possible to identify dialectal features in the texts of the ¡ur¥˙ if the other components, including pseudocorrections, are isolated. Identifying pseudocorrections also helps to reconstruct the history of Arabic dialects. For example, there are many texts of Literary Written Middle Arabic (Hary 1992:55ff) that exhibit a mixture of Classical Arabic elements, vernacular features, pseudocorrected forms, and standardized examples of such forms. The characteristics of Middle Arabic dialects can be extracted only by isolating Classical Arabic elements, on the one hand, and pseudocorrections, on the other hand, leaving the dialectal features of the texts as the remainder. Only a careful examination of Middle Arabic texts can reveal dialectal or pseudocorrected features and distinguish between them.3 There are a number of examples of hypercorrections in the ¡ur¥˙ that might be analyzed. For example, the shift of the uvular stop /q/ to the glottal stop /
See the methodology section in chapter 4, pp. 93ff. See Grotzfeld 1967; Hary 1992:263; Rosenbaum 2002c:123. See also chapter 4, pp. 96–97 and p. 106, 1.7.1.

139

CHAPTER FIVE asked’; ¯ÈÂÂ‡Ò˜ / /ß/ in the environment of other emphatic phonemes is not uncommon (chapter 4, p. 108, 1.11.3). The scribe, who probably pronounced the word /waßa†/ as a result of emphatization, understood that in the orthography only one of the two emphatic consonants was originally emphatic, but he did not know which one. Thus, he mixed

140

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION the two, resulting in the hypercorrection spelling of ˙ˆÂ.6 In addition to hypercorrections, hypocorrections can also be detected in the texts of the ¡ur¥˙. For example, there is ample use of the negative particle /lam/ in Late and Modern Egyptian Judeo-Arabic texts. This particle can come before verbs in both the perfect and imperfect forms, and it can negate verbless clauses (nouns, pronouns, and prepositions).7 Arnold has made a case for the appearance of this particle in the spoken dialect of the Jews of Iskenderoun in the northeastern Mediterranean, demonstrating that it is the combination of /lå/ and /må/, and thus accounting for its emphatic character (2006– 2007:11–12). Rosenbaum (2002a) has attempted to show that the particle was part of colloquial Egyptian Arabic speech, although Wagner has disagreed (2007:175–77). Rosenbaum also reviewed other scholars’ positions and concluded that none of them has really considered /lam/ to be part of colloquial Egyptian speech. He has claimed that Blau (1970:95), Hary (1987:280, 1992:314), Hopkins (1984:153), and Khan (1991: 231–32, 1992:236 n. 38) all considered the particle /lam/ to be a pseudocorrection, a pseudoclassical form, or a hypocorrection. Avishur (1992:166) and Woidich (1968:108) have considered /lam/ to be an element of standard Arabic, and Davies, while considering it as part of the colloquial, has still regarded it as a “[violation] of the classical rules” (1981:304). Moreover, Spitta-Bey did not consider /lam/ to be part of the colloquial (1880:169 n. 1). Furthermore, Goiten also considered /lam/ to be a pseudoclassical form (1972:258, quoted in Hary 1992:294 n. 154). Rosenbaum has admitted that the particle derived from standard Arabic and may be used in the colloquial as “the result of either pseudocorrection or a desire to elevate the linguistic style” (2002a:595). With the publication of more texts in Late Egyptian Judeo-Arabic in Hary 2009, there are now many texts in both Late 6

7

See Blau 1995:35‚ and Hary 2009, the critical edition of the Haggadah, note to ms. 3 25,8. See Hary 1992:294–95 and also chapter 4, pp. 94–95 and pp. 126–27, 3.3.3, and also chapter 8, 4-1.3–4-1.5; 4-2.1–4-2.3; 4-2.7.

141

CHAPTER FIVE and Modern Egyptian Judeo-Arabic that show the use of /lam/ in the Egyptian dialect. For Late Egyptian Judeo-Arabic: (i) Fragments of Egyptian Judeo-Arabic folktales from seventeenthcentury Cairo (Lebedev 1965, 1977) (ii) Il-maßri wir-r•fi ‘Townsmen and Fellah,’ a seventeenth-century Geniza text (Goitein 1972) (iii) Darkhei no>am ‘Ways of Pleasantness,’ in which verbatim testimonies of seventeenth-century Egyptian Judeo-Arabic are quoted (Blanc 1981, 1985) (iv) Megillat p¥r•m il-mißriyy•n ‘The Cairene Purim Scroll,’ a sixteenthcentury Egyptian Judeo-Arabic historical and liturgical text (Hary 1992) (v) ⁄ur¥˙ of Genesis, Esther, and Passover Haggadah (Hary 2009) Examining the data from Egyptian Judeo-Arabic written and oral texts (see above i–v), in addition to Rosenbaum’s analysis (2002a) and the materials on standard Egyptian dialect published by Davies (1981) and >Awwåd (1968), as well as Modern Judeo-Arabic folktales about R. Abraham ibn Ezra (Avishur 1992), and Rosenbaum’s extensive Modern Judeo-Arabic recordings (2002b, 2002c), it is possible to trace the history of the negative particle /lam/ in Egyptian Arabic. The plethora of examples provided in Late and Modern Egyptian JudeoArabic for the use of /lam/ preceding both perfect and imperfect verbs, as well as before verbless clauses, leaves no doubt as to the extensive use of the particle in the texts (the literary variety) as early as the sixteenth century. It also raises the possibility of the employment of the particle in colloquial Egyptian Arabic. Furthermore, Jewish religiolects in general (chapter 1, 23–24) tend to maintain archaic forms. Thus, it is not surprising that the phenomenon has extended all the way to twentieth-century Egyptian Judeo-Arabic variety. There are two questions that are still answered only partially. The first question is whether /lam/ was indeed an integral part of the spoken variety, as promoted by Rosenbaum (2002a, see discussion 142

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION above). The second question is what the origin of this phenomenon might be; however, it is safe to assume that the desire to elevate speech style, so common in the creation of pseudocorrections in JudeoArabic, has prompted the process of hypocorrection in the case of /lam/ preceding perfect verbs.8 Once this mechanism was in process, /lam/ preceding perfect verbs became standardized in Egyptian JudeoArabic9 and was extended to other syntactic positions: before imperfect verbs as well as before verbless clauses preceding nouns, pronouns, and prepositions. Rosenbaum’s contribution, then, lies in insisting that the feature has to be part of the colloquial (although we can not be sure of it), whereas my analysis attempts to present the process of its extensive use, at least in the written variety. Chapter 8 of this volume provides many examples of the particle before verbless clauses (p. 214, 4-1.3–4-1.5), before perfect verbs (pp. 215–17, 4-2.1), and before imperfect verbs (pp. 217–19, 4-2.3; 4-2.7).

8

9

On this process see Hary 1987:280; 1992:294, 314; 2007:278–79. See also chapter 4, 94–95, and chapter 8, pp. 215–17, 4-2.1. See Hary 1992:67, 2007:278–79.

143

CHAPTER FIVE Issue Two: The Theoretical Background of the Use of Hebrew and Aramaic Components in Judeo-Arabic Jewish religiolects in general share the common feature of including and incorporating Hebrew and Aramaic components in them (chapter 1, 21–22). These components need to be studied within the framework of languages or dialects in contact and they can be found along most of the Jewish linguistic spectrum. Hebrew and Aramaic elements should be considered together because the Hebrew found throughout the Jewish linguistic spectrum before the twentieth century is postbiblical and naturally mixes with Aramaic (Blau 1981:133 n. 1). Furthermore, speakers and writers of the religiolects did not distinguish between Hebrew and Aramaic elements. From the twentieth century onward, however, influenced by the Zionist movement and the State of Israel, Jewish religiolects include elements of modern Israeli Hebrew (Benor 2008). Judeo-Arabic, as explained in chapter 2 (p. 37) is the meeting point of Classical Arabic, Arabic dialects, and Hebrew and Aramaic. The interaction of these varieties is diagrammed in figure 2 (p. 38). Indeed, Hebrew and Aramaic influence Judeo-Arabic—as well as other Jewish religiolects—in all areas of the variety, but especially in the lexicon. Blau has claimed that the extent to which Hebrew and Aramaic elements are used in Jewish religiolects in general, and in Judeo-Arabic in particular, is dependent upon the personal style of the author, the literary genre, and the Hebrew and Aramaic knowledge of the readers (Blau 1981:44–45). The use of Hebrew and Aramaic components, however, positioned within the larger framework of varieties in contact, also appears to depend upon stimulus or resistance factors of interference (Hary 1999:68). Interference consists of deviations from the norms of either variety that occur as a result of influence from another variety (Lehiste 1988:1–27). In other words, interference between Hebrew and Aramaic on the one hand and Judeo-Arabic on the other hand can be viewed from two directions: (i) Direction A: Hebrew and Aramaic are the recipient languages and 144

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION Arabic is the source or primary language. As Hebrew and Aramaic experience interference from Arabic, they may accept and encourage this interference or resist it. As a result of this interference, the Hebrew and Aramaic features, along with other elements, take on the Arabic structure and grammar to shape Judeo-Arabic. For example, when a Hebrew noun is used in Judeo-Arabic and experiences interference from the Arabic structure, it takes an Arabic plural morpheme (however infrequently) in Direction A. Thus, the Hebrew word Ï¯Ú ‘uncircumcised man’ is transferred “as is” into Judeo-Arabic Ï¯Ú where it assumes the additional meaning of a Christian man. Its plural, however, is ÔÈÈÏ¯‡Ú, which uses the Arabic plural morpheme /•n-/ in Direction A (see below, 2.6, pp. 152–53). Thus, the plural noun />ariliy•n/ is incorporated into the Arabic structure. Figure 13 illustrates this direction: Arabic

Hebrew/Aramaic

Judeo-Arabic

Arabic Figure 13. Direction A: Hebrew and Aramaic are the recipient languages

(ii) Direction B: Arabic is the recipient language and Hebrew and Aramaic are the source languages. In this direction Arabic experiences interference from Hebrew and Aramaic. Arabic then may accept and encourage or resist this interference, which, along with other elements, helps shape Judeo-Arabic. For example, Hebrew plural nouns are frequently transferred into Judeo-Arabic “as is” with their Hebrew plural morpheme /-im/ (masculine) or /-ot/ (feminine). In such cases

145

CHAPTER FIVE the Arabic experiences interference from Hebrew, and this interference leaves these nouns with their Hebrew plural morphemes untouched along the lines of Direction B. Thus, Hebrew nouns such as ÌÈ˘µÏÙ ‘concubines’ (with the masculine plural form /-im/) and ˙ÂÂˆÓ ‘commandments’ (with the feminine plural form /-ot/) appear in JudeoArabic with their Hebrew plural morphemes. Figure 14 illustrates Direction B: Hebrew/Aramaic

Arabic

Judeo-Arabic

Hebrew/Aramaic Figure 14. Direction B: Arabic is the recipient language

Uriel Weinreich, in his pioneering study of languages in contact, has argued that varieties in contact are subject to many factors that stimulate or resist interference (1979:64–65). In Jewish religiolects in general and in Judeo-Arabic in particular, the level of interference is related to the existence of a native speaker community, which speaks the source or primary language. In general, Direction A is common to Jewish religiolects, primarily because Hebrew and Aramaic, which were used for liturgy, rabbinic literature, and rabbinic correspondence, functioned largely as literary languages (ibid., 89). Without the backing of a community of native speakers, Hebrew and Aramaic are prone to heavy interference. When Hebrew and Aramaic elements are found in Jewish religiolects, these elements are, for the most part, integrated grammatically and lexically into the Jewish religiolect (in Direction A),

146

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION because the source, primary, or dominant languages are used by communities of native speakers, while Hebrew and Aramaic are not. Thus, although both words in the Yiddish phrase /ye¡íva bóxer/ ‘a student of a religious academy’ come from Hebrew (/ye¡ivá/ > /ye¡íva/; /baxúr/ > /bóxer/), the word order is Germanic rather than Semitic, because the Hebrew components are fully integrated into the Germanic structure of Yiddish. Similarly, the Judeo-Italian verbal form gannavi ‘she steals’ has an Italian grammatical structure, although its root comes from the Hebrew g-n-v ‘steal.’ Likewise Hebrew and Aramaic elements in Judeo-Arabic texts tend to experience interference in Direction A. They are incorporated fully into the Arabic structure, primarily because the linguistic community of Arabic-speaking Jews lacks the competence of Hebrew and Aramaic native speakers. Thus, the phrase ˙ÂÏ‡˘Ï‡ ‰≤‰ ‘these questions’1 0 (Blau 1981:137), which contains a Hebrew word, still follows Arabic grammatical structure. Hebrew ˙ÂÏ‡˘ ‘questions’ is treated as an inanimate plural; therefore, the preceding demonstrative pronoun ‰≤‰ ‘this’ is in the singular feminine form, following standard Arabic practice. Moreover, ˙ÂÏ‡˘ ‘questions’ takes the Arabic definite article /al-/ as opposed to the Hebrew /ha-/, thereby fully incorporating the Hebrew noun into the Arabic grammatical structure. In addition to the encouraged interference that occurs in Direction A, as demonstrated in the previous examples, there are also examples of resistance to interference. In these cases the Hebrew and Aramaic components are not fully incorporated. For example, the Hebrew phrase in the following sentence is not fully integrated into the Arabic: ‰˜·ÈÂ ‰‡ÏË·Ï ‰Î¯· ¬˘ È¥ (93 1,6) ‘there is a doubt whether the blessing (was recited) in vain.’1 1 The Hebrew phrase ‰‡ÏË·Ï ‰ Î¯· ‘the blessing (was 10

The Hebrew words appear in a different font (Gilboa) and they are italicized in the translation. Sometimes it is difficult to decide if the word is written in Hebrew or is the Judeo-Arabic spelling of the word, copied from the Hebrew. 11 The word ‰Î¯· ‘blessing’ is in Hebrew as indicated by the khaf; ‰‡ÏË·Ï has an additional alef, reflecting the pronunciation of the long /å/ (see below, 1.1).

147

CHAPTER FIVE recited) in vain’ is transferred into the Judeo-Arabic text “as is” without any modification, except, of course, for the lengthening of the vowel (see below, pp. 148–49, 1.1) and the supralinear dot above the kaf, despite the fact that certain adjustments would better incorporate it into the Arabic structure. In this case the resistance to interference stems from the literary and social prestige associated with the Hebrew phrase ‰‡ÏË·Ï ‰Î¯· as it is used commonly in Jewish liturgical life.1 2 In Judeo-Arabic in general and in Egyptian Judeo-Arabic in particular, the incorporation of Hebrew and Aramaic elements tends for the most part to follow Direction A.1 3 Indeed, the Hebrew and Aramaic elements in the Egyptian ¡ur¥˙ often experience interference in Direction A and are incorporated fully into the Arabic structure, a result that follows from the fact that the ¡ar˙anim have Egyptian Arabic native capability. The texts of the Egyptian ¡ur¥˙, however, prove to be more complicated. They not only incorporate elements from Hebrew and Aramaic, but at times they are also dependent on primary texts in these languages. Thus, there are also many instances in the texts of the ¡ur¥˙ where Direction B operates and the Arabic in fact undergoes interference from Hebrew and Aramaic. The following analysis includes examples of both Direction A and Direction B interference in the Egyptian ¡ur¥˙. Cases of Interference in Direction A 1. When Hebrew and Aramaic words are incorporated into Judeo-Arabic, they undergo phonetic or phonological interference from Arabic, thus changing to “adapt” to the Arabic phonetic structure. 1.1 The vowel /a/ becomes lengthened [å]: Ú‡Ó˘1 4 /¡Emå>/ (74 2,1) ‘the prayer of the Shema’; Ô‡ÒÈ /nisån/ (93 1,1) ‘(month of) Nisan’; ‰‡¡¯· /bEraxå/ (93 1,12) ‘blessing’; Â‡ÂÂˆÂÂ /wi-ßiwwånu/ (93 1,14) ‘and He 12

For more examples of this resistance, see Hary 1999:71–72. For details see ibid., 72–75. 14 This corrects the spelling I provided wrongly (ÚÓ˘) in Hary 1999:73. 13

148

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION commanded us’;1 5 ˙‡·˘ /¡abbåt/ (93 2,12) ‘Saturday, Shabbat’; ‰‡¯˜ÈÓ /miqrå/ (74 20,18) ‘occasion’; Ì‡„ÚÂÓ /mo>adåm/ (74 20,9) ‘their time.’ The lengthening occurs also in names: Ô‡ÚÎ /kEna>ån/ (15 21b,1) ‘Canaan’; „‡ÚÏ‚ /gil>åd/ (15 22b,1) ‘Gil>ad.’1 6 1.2 The vowel /i/ or /e/ becomes lengthened [•]/[∑]: ıÈÓÁ /˙am∑ß/ (93 1,3; 1,6 and more) ‘unleavened food’; ·È¯ÈÚ />∑r∑b/1 7 (93 2,9) ‘evening’; ÔÈ¥‚ /gef∑n/ or /gefen/ (74 20,10) ‘vine’; ˙ÈˆÂ¯Á /˙arøß∑t/ (91 10a,7) ‘Óaroset’ (chapter 4, p. 103, 1.1.7). 1.3 In Hebrew words with /ßadi/ (ˆ) that are borrowed into Egyptian Judeo-Arabic, there is evidence that speakers used the emphatic /ß/ to adapt to the Arabic phonetics: ‰ˆÓ /maßå/1 8 (3 1,11; 74 21,4) ‘matza’ and ˙ÂˆÓ /maßøt/ (3 1,8; 74 21,3) ‘matzas’; ‰ˆÓ ÈˆÂÓ /møßi maßå/ (74 21,4; 91 10a,2) ‘bring forth matza’; ıÈÓÁ /˙am∑ß/ (93 1,3) ‘unleavened food’; Â‡ÂÂˆÂÂ /wi-ßiwwånu/ (93 1,14) ‘and He commanded us’ (chapter 4, p. 108, 1.11.5). 1.4 Emphatization occurs in Hebrew words borrowed into JudeoArabic, thus employing a regular Arabic phonological process in line with Direction A (chapter 4, p. 108, 1.11.4). 1.4.1 /t/ > /†/: ËÚÓ ÈËÓ· /bi-mE†ei mE>a†/ (3 8,1) ‘few in number,’ although it alternates with È˙Ó· (93 22,1). 1.4.2 /z/ > /Ω/: ¯∂Ú ÈÏ‡1 9 /eli>eΩer¢/ (93 15,9) ‘Eliezer’; ‰È¯∂Ú />aΩar¢ya/ (93 15,9 and more) ‘Azarya’; ¯∂Ú Ï‡ /el>aΩar¢/ (93 15-b,10) ‘Elazar,’ although these names alternate with nonemphatic /z/: ¯ÊÚÈÏ‡ /eli>ezer¢/ 15

See also Wagner 2007:83, 144. For more examples, see chapter 4, p. 102, 1.1.5, and chapter 7, pp. 207–8, 3-2.5. Cf. Rosenbaum 2002b:36. 17 For different words such as />urubbå/, />erubbå/, and />arubbå/ in Egyptian Judeo-Arabic, see Rosenbaum 2002c:134, 4.5.3.2. 18 Rosenbaum has provided evidence that in words borrowed from Hebrew, Egyptian Jews kept the original Hebrew stress on the last syllable, unlike standard Egyptian dialect (2002c:124, 135). 19 Notice that the /eli/ in ¯∂Ú ÈÏ‡ /eli>eΩer¢/ and /el/ in ¯∂Ú Ï‡ /el>aΩar¢/ are separate, as if /eli/ functions as a separate word and /el/ as the definite article. 16

149

CHAPTER FIVE (3 3,10) ‘Eliezer’; ‰È¯ÊÚ />azar¢ya/ (3 3,11) ‘Azarya’; ¯ÊÚÏ‡ /el>azar¢/ (3 3,11) ‘Elazar.’ 1.4.3 /s/ > /ß/: ı¥¯Î /kar¢faß/ (74 21,2) ‘Karpas, greens for the Passover seder,’ although it alternates with nonemphatic /s/, where the emphatization does not occur: Ò¥¯Î /karfas/ (74 21,2); ˙ˆÂ¯Á /˙arøßet/ (74 21,11) and ˙ÈˆÂ¯Á /˙arøß∑t/ (91 10a,7) ‘Óaroset,’ also alternating with ˙ÒÂ¯Á /˙arøset/ (3 22,11); ÔÈ·ÂˆÓ /mEßub•n/ (93 48,13) ‘seated.’2 0 1.5 The shift /q/ > / /w/ occurs in words borrowed from the Hebrew, in line with Direction A and following the desire to adapt to the Arabic phonetic structure: Â‡ÂÂˆÂÂ /wi-ßiwwånu/ (93 1,14) ‘and He commanded us.’ Rosenbaum reports the use of the voiced labiodental fricative /v/ in Egyptian Judeo-Arabic, as in />avøn/ ‘sin’ (2002c:123), but there is no reference to this use in the texts of the Egyptian ¡ur¥˙ (chapter 4, p. 106, 1.8.2). 1.6.1 The letter · in Hebrew words borrowed into Judeo-Arabic is pronounced as a voiced bilabial stop /b/, even if in Hebrew it is an unvoiced labiodental fricative /v/. This is in order to follow the Arabic phonetic inventory: ·È¯ÈÚ />∑r∑b/ (93 2,9) ‘evening.’

20 21

See chapter 4, p. 108, 1.11.4, and Hary 1999:73 for both 1.4.2 and 1.4.3. See chapter 4, pp. 96–97 and p. 106, 1.7.1.

150

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION 1.7 The shift /¡/ > /s/ occurs in the ¡ur¥˙ in some words borrowed from the Hebrew, but the shift is not common: ÒÂ„˜ /sab>/ (15 6b,20; 7a,16) ‘Beer-Sheba.’ In ‰·È÷ (15 4b,20) ‘old age’ the shift /¡/ > /s/ could have happened, but it can also mark the use of a Hebrew word, as is the case in ‡÷Ó (15 5a,10) ‘Masa.’2 2 2. Direction A occurs also through morphosyntactic processes. 2.1 The definite article attached to Hebrew words is the Arabic morpheme /al-/ in a morphophonemic spelling (chapter 4, 1.15), thus incorporating the Hebrew words into the Judeo-Arabic structure: Ï‡ ÌÈ˘µÏÙ (15 4b,17) ‘the concubines’; ‰„‚‰ Ï‡ (3 colophon) ‘the Haggadah’; ıÈÓÁ Ï‡ (93 1,3) ‘the unleavened’; ‰ Î¯· Ï‡ (93 1,12) ‘the blessing’; ˙ÒÂ¯Á Ï‡ (3 22,8) ‘the Óaroset’; ÍÂÙ˘ Ï‡ (3 colophon) ‘the (prayer) Pour (Your Wrath)’; ˙·˘ Ï‡ (93 2,10) ‘the Shabbat’; Ï‡ ÁÈ˘Ó (3 4,4) ‘the messiah.’ 2.2 As mentioned frequently in this volume, the ¡ur¥˙ are verbatim translations of sacred texts. At times, when the Hebrew original does not have the definite article, the ¡ar˙an making the translation inserts the Arabic definite article in Direction A to adapt the translation to regular Arabic style. Thus, although in Gen 49:12 there are no definite articles preceding the nouns in the Hebrew text, ÌÈ˘ Ô·ÏÂ ÔÈÈÓ ÌÈÈÚ ÈÏÈÏÎÁ ·ÏÁÓ ‘(his) eyes are darker than wine and (his) teeth are whiter than milk,’ the ¡ar˙an adds them in order to be in line with Arabic structure: Ô·Ï Ï‡ ÔÓ Ô‡Ò Ï‡ √ÈÈ·‡Â „È· Ï‡ ÔÓ ÔÂÈÚ Ï‡ ¯ÓÁ‡ (15 36b,16–17) ‘(his) eyes are more red than wine and (his) teeth are whiter than milk’ (chapter 8, pp. 267–68, 9-1.4–9-1.6). 2.3 The Arabic genitive marker /bEtå>/ is added to Hebrew words and phrases that are incorporated into the text in Direction A: ˙‡È¯˜ ˙˜Â Á·Âˆ Ï‡ Ú‡˙· ÚÓ˘ (93 15,3) or, with the addition of the definite article, Á·Âˆ Ï‡ Ú‡˙· Ú‡Ó˘ ˙‡È¯˜ Ï‡ ˙˜Â (74 15,3) ‘the time of reciting the 22

See Hary 1992:260, 3.7, and chapter 4, p. 110, 1.13.4.

151

CHAPTER FIVE morning Shema’; ÔÈÓÂ˜È¥‡ Ú‡˙· ‡ˆÓ Ï‡ (3 22,13) ‘the matza of the afikoman’ (see chapter 4, pp. 116–17, 2.1.7). 2.4 Arabic prepositions may be added to Hebrew words, incorporating them into the Judeo-Arabic text in Direction A. In the title of the ¡ar˙ of Esther, the preposition /bi/ is added after the Hebrew: ¯˙Ò‡ ˙Ï"‚Ó È·¯Ú Á¯˘· (1302 1b,1) ‘the Scroll of Esther in Arabic ¡ar˙.’ In the following example from the Haggadah, „Ú· Ô‡ÒÈ· ¯˘Ú ˙Ú·¯‡ ˙ÈÏÈÏ ˙È·¯Ú (93 1,1) ‘the night of the fourteenth of the month of Nisan after the evening prayer,’ the Arabic prepositions /bi/ ‘in’ and /ba>d/ ‘after’ flank the Hebrew word Ô‡ÒÈ ‘(the month of Nisan),’ although the latter could be considered Judeo-Arabic; /ba>d/ comes before Hebrew ˙È·¯Ú. Both the prepositions /bi/ ‘in’ and />ala/ ‘on’ are used in the following example flanking Hebrew words to incorporate them into the text: ‰Áˆ ‡ÏÚ ‰ Î¯· Ï‡ ˛˙¸ÂˆÓ· ‰ÎÊ˙‡ (93 1,11–12) ‘he gains (the merit of performing) the commandment of (reciting) the blessing in the right way.’ It seems that in the following example the preposition /bi/ is in Arabic: ˙·˘· ÁÒÙ ·È¯ÈÚ ÌÂÎÁÂÈ ‡≤‡ (93 2,9) ‘if the evening before Passover falls on Friday night …’ However, there is the possibility that the preposition /bi/ is indeed in Hebrew (/be-/) and that in this case the Hebrew underwent resistance and was transferred into the Arabic “as is.” In the following sentence, though, the Arabic preposition /fi/ is used to incorporate the Hebrew words into the Judeo-Arabic text: ˘„˜ÓÏ ¯ ÎÊ ÂÏÂ˜ÈÂ ˙ÒÂ¯Á Ï‡ È¥ ÌÂ‰ÂÊÓµÈÂ (3 22,10–11) ‘and dip them in the Óaroset, and recite the blessing, in memory of the Temple.’ 2.5 Arabic demonstrative pronouns are used with Hebrew words to integrate them into the Arabic text: ‰‡¡¯· Ï‡ ‰„‰ (93 1,12) ‘this blessing’; ‰„ ¯¯ÂÓ (74 10,14) ‘this bitter herb.’ 2.6 Hebrew nouns in Egyptian Judeo-Arabic seldom receive Arabic plural morphemes in Direction A: ÔÈÈÏ¯‡Ú2 3 ‘uncircumcised (Christian) men.’ Rosenbaum (2002c:125) has reported /mamzer•n/ ‘bastards,’

23

See Hary 1992:155 n. 119.

152

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION but the form alternates with /mamzer•m/ (see below, Direction B, 4). 2.7 As is common in Jewish religiolects, Hebrew roots can receive the majority language patterns, creating mixed forms (chapter 1, pp. 21–22). Thus, the Hebrew root s-d-r of ‰˘Ó È¯„Ò ‘the sections of the Mishna’ receives interference from the Arabic plural form /fa>å>il/, resulting in Judeo-Arabic ¯„‡„Ò (74 22,16) ‘sections (of the Mishna).’2 4 Along similar lines, the Hebrew root z-k-y ‘gain’ receives interference from the Arabic and may take either a first verbal pattern, resulting in ˙ÈÎÊ /zak∑t/ (3 3,19) ‘I gained,’ or an /itfa>al/ form, resulting in ‰ÎÊ˙‡ /itzaka/ (93 1,11) ‘(he) gained’ (chapter 4, p. 132, 4.3.3). 2.8 Hebrew terms and lexical items are incorporated syntactically into the text in Direction A to be part of the Judeo-Arabic text: Ï‡ „‡ÏÂÂ‡ÏÂ ˙‡È„‰ Ì‰¯·‡ ‰ËÚ Ì‰¯·‡Ï È≤Ï‡ ÌÈ˘µÏÙ (15 4b,17) ‘and to Abraham’s sons by concubines, Abraham gave gifts’; ÍÂÙ˘ Ï‡Â È·¯Ú Ï‡· ‰„‚‰ Ï‡ (3 colophon) ‘The Haggadah in Arabic and (the prayer) Pour (Your Wrath)’; ‰¯ÈÓÁ ÏÎ ÏÂ˜È ‡Ó ÔÂ„· (93 2,11) ‘without reciting (the prayer in Aramaic) all kinds of leavened food’; ˙·˘ Ï‡ È¥ ÔÂÎÈ Ô‡Î‡ (93 8,15) ‘if (the Seder) falls on Shabbat’; ‰‰ ÔÓ ˘Â„˜ Ï‡ Â„·È ÏÂÁ ÔÂÎÈ Ô‡Î‡ (93 9,4) ‘If (the seder) falls on a regular day, (the participants) start the Sanctification from here.’ In the following example, however, the Arabic interference of the Hebrew is resisted and the Hebrew term ÍÂ¯· ˘Â„˜‰ ‡Â‰ ‘the holy One, blessed be He’ is inserted “as is,” including the Hebrew definite article /ha-/: Á·„Â ‡Â‰ ÍÂ¯· ˘Â„˜‰ ‰È‚Â (93 87,14) ‘and the holy One, blessed be He, came and slew.’ Cases of Interference in Direction B Unlike other Judeo-Arabic texts, the ¡ur¥˙ contains many examples of interference in Direction B where, rather than Hebrew and Aramaic components being fully incorporated into the Arabic structure, the Arabic structure instead experiences heavy interference from Hebrew 24

It is possible that in another context ¯„‡„Ò can also indicate the plural of siddur ‘prayer book.’ I thank Yona Sabar, personal communication.

153

CHAPTER FIVE and Aramaic elements. Direction B occurs in the ¡ur¥˙ when important Hebrew and Aramaic texts are translated into Judeo-Arabic. In other words, the role of the community of Arabic native speakers is less significant in the texts of the ¡ur¥˙ that are dependent on these Hebrew and Aramaic texts. In these cases, the language “spirit” of the ¡ur¥˙ is that of a Hebrew or Aramaic sacred Jewish text. Although the ¡ar˙anim were not native Hebrew or Aramaic speakers, they nevertheless based certain elements of the Judeo-Arabic translations quite closely on Hebrew and Aramaic, the original languages of these sacred texts. In these cases, the Arabic is prone to heavy interference in Direction B. 3. When Hebrew and Aramaic components are found in Judeo-Arabic, in most cases they undergo phonetic and phonological change in Direction A to adapt to the Arabic phonemic inventory. However, the occurrence of the phoneme /p/ in Egyptian Judeo-Arabic (chapter 4, p. 107, 1.9) is an example of Direction B. In this case, the Arabic undergoes interference from the Hebrew, does not resist the phoneme, which is not found in its phonemic inventory, and accepts it “as is.” This phoneme appears in words borrowed from the Hebrew: Ï‡ ÌÈ˘µÏÙ (15 4b,17) ‘the concubines’; Ô„Ù (15 5b,1 and more) ‘Padan’; ÌÈ˙˘ÏÙ (15 6a,5 and more) ‘Philistines’; Ï‡ÈÙ (15 16b,3) ‘Peniel’; Ï‡ ÈÊ¯Ù (15 18b,9) ‘the Perizzites’; ÈËÂÙ (15 26b,12) ‘Poti’; ÁÒÙ (3 4,14 and many other places) ‘Passover’; ‰ÙÂÁ (93 90,12 and many other places) ‘marriage’; ˜ÂÒÙ (3 3,21 and many other places) ‘verse’; ‰Ú¯Ù (3 7,8 and many other places) ‘Pharaoh.’ However, the name ¯ÙÈËÂÙ ‘Potiphar’ is transferred into Judeo-Arabic in Direction A, and thus /p/ is adapted to the Arabic phonetic structure, becoming the labiodental fricative /f/: ¯¥ÈËÂ¥ /fo†ifar/ (15 23-1a,6) ‘Potiphar.’ 4. A Hebrew noun used in Judeo-Arabic does not usually receive an Arabic plural morpheme in Direction A (see above, p. 145), but is transferred into the Judeo-Arabic “as is” in Direction B with its Hebrew plural morpheme /-im/ (masculine) or /-ot/ (feminine): ÌÈ˘µÏÙ (15 4b,17) 154

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION ‘concubines’; ÌÈ˙˘ÏÙ (15 6a,5 and more) ‘Philistines’; ÌÈÈ¯ˆÓ Ï‡ (91 4b,17) ‘The Egyptians,’ although it may alternate with ‰ÂÂ¯ˆÓ Ï‡ /il-mEßarwa/ (3 8,17); ÁÒÙ ÌÈÈ„ (3 4,14) ‘the laws of Passover’; ˙ÂÂˆÓ (93 1,11) ‘commandments’; ˙Â¯ÂÓ˘ ˙ÂˆÓ (3 1,8) ‘guarded matzas.’ 2 5 5. The Judeo-Arabic accusative marker ‰Ï‡ /ilå/ undergoes interference from Hebrew in Direction B. Consequently, in Later Egyptian JudeoArabic ¡ur¥˙, the translation of the Hebrew definite direct object marker ˙‡ as ‰Ï‡ /ilå/ is almost obligatory. See chapter 8, pp. 257–64, 8.1 for further details. 6. Many additional morphological and syntactical processes where the interference is in Direction B are reported in Hary 1999:82–88. Furthermore, part 2 of this volume presents countless examples of literal or verbatim translations from the Hebrew and Aramaic texts into Judeo-Arabic. In these verbatim translations Direction B is evident in innumerable instances.

25

For more examples, see Hary 1999:74 and Rosenbaum 2002c:125, 3.2 (/mamzer•m/ ‘bastards, illegitimate children’).

155

CHAPTER FIVE Issue Three: The Hebrew and Aramaic Lexicon in the ¡ur¥˙ There are many Hebrew and Aramaic words that found their way into Egyptian Judeo-Arabic not only in the religious and liturgical spheres, but also in other areas (Rosenbaum 2002c:129–43). In the texts of the ¡ur¥˙ there are not many Hebrew and Aramaic words, as the ¡ar˙anim made it their point to attempt to translate every Hebrew and Aramaic word into Judeo-Arabic. This is common not only in Judeo-Arabic ¡ur¥˙, but also in many other translations of sacred texts into various Jewish religiolects. Nonetheless, there are some instances where the Hebrew and Aramaic words are copied and inserted into the JudeoArabic ¡ur¥˙, especially proper and place names. The following paragraphs present a selection of these words. 6. There are many Hebrew and Aramaic proper names in the ¡ur¥˙. Some are adapted to varying extents to the Arabic structure through Direction A, while others are copied “as is” into the Judeo-Arabic texts in Direction B. Figure 9 in chapter 3 (p. 85) illustrates the continuum of translation from least literal to most literal, where “least literal” includes names that have been adapted to the Arabic structure in Direction A, and “most literal” includes names that have been copied “as is” into the Judeo-Arabic ¡ur¥˙ in Direction B. 2 6 6.1 Proper names transferred in Direction A have been adapted to the Arabic phonetic structure. This adaptation ranges from the spelling of the labiodental fricative /f/ with ¥ (including the supralinear dot), to the spelling of the emphatic /ß/ with ˆ, to the marking of the long vowel /å/ with an alef, and more: ÔÂ¯¥Ú (15 5a,2 and many other places) ‘Ephron’; ¯ÁÂˆ (15 5a,3) ‘Ío˙ar’; ¯‚‡‰ (15 5a,7) ‘Hagar’; ¯‡„˜ (15 5a,10) ‘Qedar’; ÔÂÚ¯¥ (15 31b,7 and many other places) ‘Pharaoh,’ alternating with ‰Ú¯Ù (see below, 6.2); ¥ÒÂÈ (15 31b,4 and many other places) ‘Joseph’; ÔÈ·Â‡¯ (15 33a,2 and more) ‘Reuben’; Â‡ÒÚ (15 5b,10 and more) ‘Esau,’ alternating

26

For more examples and further discussion, see chapter 3, pp. 83–85.

156

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION with ÂÈ˘Ú (see below, 6.2); ‡Â·¯Á (1302 1b,14) ‘Óarbona’; Ô¡ÂÓÓ (1302 1b,22 and more) ‘Memukhan’; È¡„¯Ó (1302 2a,17 and more) ‘Mordecai’; ¯∂Ú ÈÏ‡ /eli>eΩer¢/ (93 15,9) ‘Eliezer,’ alternating with ¯ÊÚÈÏ‡ /eli>ezer¢/ (see below, 6.2); ‰È¯∂Ú />aΩar¢ya/ (93 15,9 and more) ‘Azarya,’ alternating with ‰È¯ÊÚ />azarya/ (see below, 6.2); ¯∂Ú Ï‡ /el>aΩar¢/ (93 15-b,10) ‘Elazar,’ alternating with ¯ÊÚÏ‡ /el>azar¢/ (see below, 6.2). 6.2 Proper names transferred in Direction B are copied “as is” into the Judeo-Arabic texts: ˜ÁˆÈ (15 5a,1 and many other places) ‘Isaac’; Ï‡ÚÓ˘È (15 5a,2 and many other places) ‘Ishmael’; ‰¯˘ (15 5a,4 and many other places) ‘Sarah’; ˙Á (15 5a,4) ‘Óet’ (Hittite); ˙ÂÈ· (15 5a,9) ‘Nebayot’; Ï‡·„‡ (15 5a,10) ‘Adbeel’; ÚÓ˘Ó (15 5a,10) ‘Mishma’; ‰ÓÂ„ (15 5a,10) ‘Duma’; „„Á (15 5a,11) ‘Óadad’; ÌÈ˙˘ÏÙ (15 6a,5 and more) ‘Philistines’; ÔÈÓÈ· (15 32a,19 and more; 1302 2a,18) ‘Benjamin’; Ì‰¯·‡ (15 2a,1 and more; 3 6,2) ‘Abraham’; ¯ÂÁ (15 2a,8 and more; 3 6,2) ‘Na˙or’; ÔÓÂ‰Ó (1302 1b,14) ‘Mehuman’; ‡˙Ê· (1302 1b,14) ‘Bizzta’; ‡˙‚· (1302 1b,14) ‘Bigta’; ¯˙Ê (1302 1b,15) ‘Zetar’; ¯È‡È (1302 2a,17) ‘Yair’; ÈÚÓ˘ (1302 2a,17) ‘Shimi’; ˘È˜ (1302 2a,17) ‘Qish’; ¯˙Ò‡ (1302 2b,2 and more) ‘Esther’; ÔÓ‰ (1302 3a,4 and more) ‘Haman’; ‰Ú¯Ù (3 7,8 and more) ‘Pharaoh,’ alternating with ÔÂÚ¯¥ (see above, 6.1); ¯ÊÚÈÏ‡ /eli>ezer/ (3 3,10) ‘Eliezer,’ alternating with ¯∂Ú ÈÏ‡ /eli>eΩer¢¢/ (see above, 6.1); ‰È¯ÊÚ />azarya/ (3 3,11) ‘Azarya,’ alternating with ‰È¯∂Ú />aΩar¢ya/ (see above, 6.1); ¯ÊÚÏ‡ /el>azar/ (3 3,11) ‘Elazar,’ alternating with ¯∂Ú Ï‡ /el>aΩar¢¢/ (see above, 6.1); Ú˘Â‰È (3 5,20 and more) ‘Joshua’; Á¯˙ (3 6,2) ‘Tera˙’; ÂÈ˘Ú (3 6,8 and more) and Â÷Ú (93 19,9 and more) ‘Esau,’ alternating with Â‡ÒÚ (see above, 6.1). 6.3 Place names transferred in Direction A have been adapted to Arabic phonetics, either by lengthening the vowel /a/ to /å/ and marking it with an alef, or by marking the labiodental fricative /f/ with ¥, using the supralinear dot, as well as by other means:

157

CHAPTER FIVE ¯‡¯‚ (15 6b,10 and other places) ‘Gerar’; ‰Ú·˘ (15 7a,15) ‘Shib>a’; Ú·˘ ¯È· (15 7a,16) ‘Bir Sheba>’; ‡¯ÓÓ (15 5a,3) ‘Mamre’; Ô‡ÚÎ (15 34a,1 and more) ‘Canaan,’ alternating with ÔÚÎ (see below, 6.4); ˙¯¥‡ (15 35b,14 and more) ‘Ephrat’; ˜¯· ¯„‡· (3 3,12–13) ‘Bnei Braq.’ 6.4 Place names transferred in Direction B are copied “as is” into the Judeo-Arabic texts: ‡¯ÓÓ (15 5a,3) ‘Mamre’; ¯Â˘ (15 5a,15) ‘Shur’; Ì¯‡ Ô„Ù (15 5b,1) ‘Padan-Aram’; ˜÷Ú (15 6b,15) ‘Eseq’; ‰Ë÷ (15 6b,17) ‘Si†na’; Ï‡¯˘È (15 16a,20 and more; 3 2,5 and more) ‘Israel’; Ô˘Â˘ (1302 1b,4 and more) ‘Shushan’; ÌÈÏ˘Â¯È (1302 2a,18) ‘Jerusalem’; ‰„Â‰È (1302 2b,1) ‘Judah’; Ï·· (1302 2b,1) ‘Babylon’; Ô˘Â‚ (3 7,19 and more) ‘Goshen,’ although it is also translated ‰ÊÈ‚ (15 31b,20 and more) ‘Giza’; ÔÚÎ (3 7,18 and more) ‘Canaan,’ alternating with Ô‡ÚÎ (see above 6.3); ¯ÈÚ˘ (3 6,9) and ¯ÈÚ÷ (93 19,10) ‘Se>ir’; ÈÈÒ (3 16,20 and more) ‘Sinai.’ 7. Liturgical and religious terms: ¯¥Ò (15 title) ‘book’ (religious);2 7 ÔÒÈ (1302 3a,12) and Ô‡ÒÈ (93 1,1) ‘(month of) Nisan’; ¯„‡ (1302 3a,14; 3b,6 and more) ‘(month of) Adar’; ÔÂÈÒ (1302 6a,7) ‘(month of) Sivan’; ¯¯ÂÓ (3 2,15) ‘bitter herbs’; Ú‡Ó˘ (74 2,1) alternating with ÚÓ˘ (3 3,15) ‘the prayer of the Shema’; ÁÈ˘Ó (3 4,4) ‘messiah’; ÁÒÙ ÌÈÈ„ (3 4,14) ‘the Laws of Passover’; ÁÒÙ Ô·¯Â˜ (3 4,15) ‘the sacrifice of Passover’; ÔÈÓÂ˜È¥‡ (3 22,13 and more) ‘afikoman’; ˙Â¯ÂÓ˘ ˙ÂˆÓ (3 1,8) ‘guarded matzas’; ˜ÂÒÙ (3 3,21 and many other places) ‘verse’; ı¥¯Î (74 21,2) /karfaß/ ‘Karpas’ (greens for the Passover Seder), alternating with Ò¥¯Î /karfas/ (74 21,2); ˙ˆÂ¯Á /˙arøßet/ (74 21,11) ‘Óaroset,’ alternating with ˙ÈˆÂ¯Á /˙arøß∑t/ (91 10a,7); ‰ÏÈÓ (93 91,5) ‘circumcision’; ‰ÙÂÁ (93 90,12) ‘marriage’; Ï‡¯÷È Ë·˘ (93 93,11) ‘Israelite tribe’ or ‘tribe of Israel’; ‡Â‰ ÍÂ¯· ˘Â„˜‰ (93 87,14) ‘the holy One, blessed be He’; Ï‡

27

See also Rosenbaum 2002c:134.

158

ADDITIONAL LINGUISTIC ISSUES OF THE ¡ar˙ TRADITION ˙·˘ (93 8,15) and ˙‡·˘ /¡abbåt/ (93 2,12) ‘(the) Shabbat, (the) Saturday’; ÏÂÁ (93 9,4) ‘a regular day’; ÒÂ„˜ /
Summary This chapter highlights additional linguistic issues relevant to the analysis of Egyptian ¡ur¥˙: the use of pseudocorrections in the texts and their standardization, the Hebrew and Aramaic components in the ¡ur¥˙ and the ways they were incorporated into the texts, and a brief look at the Hebrew and Aramaic lexicon used in the Judeo-Arabic texts.

159

PART II A LINGUISTIC MODEL OF THE JUDEO-ARABIC T RANSLATIONS OF SACRED TEXTS

CHAPTER SIX APPLYING THE MODEL This chapter, as an introduction to part 2, applies the linguistic model of the ¡ur¥˙, introduced in chapter 3 of part 1. In other words, it provides complex examples of how the ¡ar˙anim translated their sacred texts using Mechanisms A and B (pp. 85–90). The chapter discusses methodological considerations concerning the organization of the examples and demonstrates the dynamic literal/interpretive linguistic tension that characterizes the translations. Furthermore, it discusses how calque translations, which represent the ultimate in verbatim translation, were incorporated into the ¡ur¥˙. Introductory Notes In the following paragraphs the ¡ar˙anim’s work is analyzed in great detail, showing the literal/interpretive linguistic tension, which is the outcome of translation Mechanisms A and B. I bring forth examples from the various texts of the ¡ur¥˙ analyzed in this study: the books of Genesis and Esther as well as the Passover Haggadah. These examples are organized according to table 4, displayed on pp. 81–82.1 This organization cannot be done in neat linguistic categories, since in several cases examples may appear under different features. Good illustrations of this kind of classification are examples (i) and (ii) below. Note that each example begins with the Judeo-Arabic text, 1

The following examples are taken only from the actual texts of the ¡ur¥˙. For the most part, they do not include the non-¡ur¥˙ parts of the texts which accompany the Haggadah. In other words, in general they do not contain any grammatical phenomena from the Haggadah which are instructional or descriptive.

CHAPTER SIX followed, when deemed necessary, by a broad transcription, then accompanied by a citation of the manuscript (in bold) along with numbers of the folio(s) and the line(s), and then the translation into English. The reader can find in Hary 2009 the actual critical editions of the ¡ur¥˙ along with the folio and line numbering as well as their translation into English. (i) ‰ÏÈÏ Ï‡ ¬Ï‡≤ (74 1,17) ‘that night.’ This example involves demonstrative pronouns and gender agreement. In one analysis, this example may fall under Level: Morphosyntactic; Category: Pronouns; Feature 6-4: Demonstrative pronouns, as it exhibits a certain demonstrative pronoun word order (chapter 8, 6-4.11). However, this example, following a different analysis, may also be classified under Level: Morphosyntactic; Category: Agreement; Feature 10-3: Gender, since it demonstrates an irregular gender agreement (ibid., 10-3.15), resulting from a literal translation of the Hebrew ‰ÏÈÏ‰ Â˙Â‡. (ii) ¯˙ÎÈÏ Ô‡Ò‡ Ï‡ ‰„˙·‡ Ô‡ Ô‡ÎÂ (15 0-1,18) ‘when men began to increase.’ This example includes features of infinitives, prepositions, and finite verbs. The translation of the Hebrew infinitive construct ·]Ï in ·]Ï Ì„‡‰ ÏÁ‰ ÈÎ È‰ÈÂ ‘and it came to pass when man began to increase’ (Gen 6:1) is Judeo-Arabic ¯˙ÎÈÏ ‘to increase’ with the preposition /li-/ followed by the finite verb /yiktir/. In one analysis this example can be found in Level: Morphosyntactic; Category: Verb Conjugation; Feature 7-1: Infinitives. This is an example of a verbatim translation of the Hebrew preposition ≠Ï, associated with the infinitive construct, into Judeo-Arabic /li-/, followed by a finite verb, even if in regular Arabic style we would have expected the particle /an/. This kind of example may also fall under Level: Morphosyntactic; Category: Prepositions/Particles; Feature 5-1: Prepositions, as it involves the preposition /li-/, or even under Level: Morphosyntactic; Category: Verb Conjugation; Feature 7-2: Finite verbs, as it also involves the employment of a Judeo-Arabic finite verb rather than the Hebrew infinitive construct.

164

APPLYING THE MODEL Methodological Notes As described in chapter 3 (pp. 75 and 89–90), various ¡ar˙anim must have composed the ¡ur¥˙, since many manuscripts of the same ¡ar˙ exist. When two separate examples of the same text from two different manuscripts are compared and assigned the same analysis, it is actually assumed that both examples were composed by the same ¡ar˙an. Consider, for example, the phrase ÈÂÂ˜ ÈÂÂ˜ Ï‡· (93 22,10–11) ‘very very,’ which translates rather literally the Hebrew „‡Ó „‡Ó·. However, in other manuscripts of the Egyptian Haggadah, the Hebrew phrase is translated more interpretively and in accordance with regular colloquial use: ÈÂÂ˜ ÈÂÂ˜ (3 8,9) and ÈÂÂ‡˜ ÈÂÂ‡˜ (91 4b,13). In other words, when these two examples are compared and analyzed in chapter 8, p. 230, 5-1.37, it is assumed that mss. 3, 91, and 93 are copies of an earlier prototype ¡ar˙. It is quite possible, though, that these manuscripts have been composed by two or more different ¡ar˙anim. However, the ¡ur¥˙ can be viewed within one translation tradition. It is probable that the tradition of the ¡ar˙ was unified either by a “chief ¡ar˙an,” who directed the work of the translation, or by “translation schools” that included rabbis and disciples who engaged in the work of the ¡ar˙. Consequently, the same analysis can be assigned to examples from different manuscripts. The detailed analysis presented in this chapter highlights the understanding that the ¡ar˙ created its own Judeo-Arabic grammar and structure. For example, the use of /ila/ to mark the definite direct object is obligatory in Egyptian Judeo-Arabic ¡ur¥˙: ‰Ï‡ ¥ÒÂÈ ·‡‚Â ÔÂÚ¯¥ ‰Ï‡ ·Â˜ÚÈ ¬¯‡·Â ÔÂÚ¯¥ Ì‡„Â˜Ï Â¥˜ÂÂ ‰Â·‡ ·Â˜ÚÈ (15 34a,12–13; underlining is mine) ‘And Joseph brought Jacob, his father, and presented him to Pharaoh. Jacob then blessed Pharaoh.’ As is customary in the texts of the ¡ur¥˙, /ila/ marks the definite direct objects ·Â˜ÚÈ ‘Jacob’ and ÔÂÚ¯¥ ‘Pharaoh,’ precisely emulating the Hebrew ˙‡ in Genesis.

165

CHAPTER SIX The kind of overlapping in the features discussed above, coupled with the need to search for Mechanism B in the ¡ur¥˙, as mentioned in chapter 3, highlights the need to scan the texts “from head to toe.” This method of scanning yields a better understanding of the ¡ar˙an’s work. Therefore, the following section provides examples that illustrate this complex analysis, using the method of scanning “from head to toe” and presenting several examples that can be analyzed on different levels (phrase, word, morphosyntax, and segment) and thereby produce different analytical results. Some examples illustrate the literal nature of the translation and some the interpretive nature. Furthermore, numerous examples demonstrate simultaneous literal/interpretive translation of the different features listed in table 4 (pp. 81–82).. Dynamic Literal/Interpretive Linguistic Tension: Complex Cases The following examples from the ¡ur¥˙ of Genesis and the Passover Haggadah can be scanned “from head to toe” all the way from the phrase through the word and the morphosyntactic levels, down to the segment level. All of the examples demonstrate the literal/intepretive linguistic tension interwoven in the work of the ¡ar˙anim. (1) ÌÏÎ˙ÈÏ Ì˙È Ï·˜ ‡Â‰ Ô‡ÎÂ (15 2b,15) ‘He had scarcely finished speaking’ is a strange word sequence, which looks on the surface like an exact translation of the Hebrew ¯·„Ï ‰ÏÎ Ì¯Ë ‡Â‰ È‰ÈÂ (Gen 24:15) in Level: Phrase; Feature 1-1: Word-for-word translation. Indeed, when this sentence is further scanned, similar verbatim translations occur, but not always. On Level: Morphosyntactic; Category: Prepositions/Particles; Feature 5-1: Prepositions, the preposition Ï·˜ ‘before’ followed by the verb Ì˙È ‘finished’ is a clear literal translation of ‰ÏÎ Ì¯Ë, whereas in regular Arabic structure we would have expected the particle /an/ in between (/qabla an yatimm/). In fact, in Saadia’s translation the particle /an/ appears in this sentence, ‚¯ÙÈ Ô‡ Ï·˜ Â‰Ù. Further, on the morphosyntactic level, either on Level: Morphosyntactic; Category: Prepositions/ Particles; Feature 5-2: Coordinative particles and conjunctions, or on Level:

166

APPLYING THE MODEL ÌÏÎ˙ÈÏ /li-yitkallim/ ‘to speak’ translates verbatim Hebrew ¯·„Ï with the equivalent Judeo-Arabic particle /li-/. Regular Arabic style, though, requires the particle /an/. Moreover, the independent pronoun ‡Â‰ ‘he’ is also translated literally, as evidenced by the spelling, which may be an imitation of the Hebrew ‡Â‰; however, the pronunciation /huwwa/ complicates the analysis by pointing toward the interpretive direction (Level: Morphosyntactic; Category: Pronouns; Feature 6-1: Independent personal pronouns). Another indication of the interpretive direction can be seen on Level: Morphosyntactic; Category: TMA; Feature 11-1: Tense/Aspect . Here, the verb Ô‡ÎÂ is in the perfect, although in the Hebrew È‰ÈÂ is in the imperfect followed by the vav consecutive. Conversely, the second verb Ì˙È is in the imperfect, following regular Arabic structure, unlike the Hebrew perfect ‰ÏÎ, also in an interpretive mode. On the segmental level, Level: Segment; Category: Orthography/Phonology; Feature 13-6: Hebrewinfluenced orthography, the literal mode is further observed in the spelling of the independent pronoun ‡Â‰ with a final alef as an imitation of the Hebrew spelling and as part of the Hebraized orthography. Therefore, the above example, when scanned from “head to toe” on different levels, can produce both literal and interpretive translations. Morphosyntactic; Category: Verb Conjugation; Feature 7-2: Finite verbs,

(2) ˙‡ÓÂ ‰Ò ÔÈ˙ÒÂ ÔÈ˙‡Â ‰È‡Ó ÚÒ˙ „¯È Ì‡È‡ ÏÂÎ Â‡ÎÂ (15 0-1,4-5) ‘And all the days of Jared came to 962 years, and then he died’ translates the Hebrew ˙ÂÓÈÂ ‰˘ ˙Â‡Ó Ú˘˙Â ‰˘ ÌÈ˘˘Â ÌÈ˙˘ „¯È ÈÓÈ ÏÎ ÂÈ‰ÈÂ (Gen 5:20). When this sentence is scanned in various ways, the outcome may differ depending on the level. On the phrase level in Level: Phrase; Feature 1-1: Word-for-word translation, the Judeo-Arabic sentence looks like a verbatim translation, but upon a closer look, interpretive translation is also revealed. Remaining on the phrase level, in Level: Phrase; Category: Word order; Feature 2-3: Numerals, the word order of the numerals clearly follows regular Judeo-Arabic structure, not imitating the Hebrew original, in a typical interpretive translation. On the word level, the literal translation is observed. Both ÏÂÎ /kull/ ‘all’ and „¯È

167

CHAPTER SIX /yered/ ‘Jared’ are similar to their Hebrew equivalents in Level: Word; Category: Lexicon; Feature 3-2: Word choice (considerations of sound/appearance). Finally, on the morphosyntactic level we see another example of interpretive translation. Both Â‡ÎÂ /wa-kånu/ ‘(they) were’ and ˙‡ÓÂ /wa-måt/ ‘and (he) died’ are in the perfect, whereas Hebrew ÂÈ‰ÈÂ and ˙ÂÓÈÂ are in the imperfect, succeeding vav consecutive. The ¡ar˙an then interprets these grammatical forms in the ¡ar˙ under Level: Morphosyntactic; Category: TMA; Phenomenon 11-1: Tense/Aspect. Conversely, Â‡ÎÂ /wa-kånu/ ‘and (they) were’ shows a clear verbatim translation in Level: Morphosyntactic; Category: Agreement; Feature 10-2: Number: plural, as the verb is in the plural preceding a singular subject (ÏÂÎ /kull/ ‘all’) or an inanimate plural subject (Ì‡È‡ /ayyåm/ ‘days’). This phenomenon is unusual in Arabic but it imitates the Hebrew plural ÂÈ‰ÈÂ in a literal translation. A comparable literal translation is evident in ms. 15 0-1,7, where the singular verb Ô‡ÎÂ appears in a similar sentence Ì‡È‡ ÏÂÎ Ô‡ÎÂ ‘And all the days were … ,’ but this time the Hebrew is also in the singular (È‰ÈÂ, Gen 5:23). (3) ¯‚‡˙Ï ˙ÂÂ¥Ó (15 2a,4), which translates ¯Á›ÒÏ ¯·›Ú ‘according to the going merchants’ weights’ (Gen 23:16), means literally ‘passed to a merchant.’ It is a clear verbatim translation, even when the sentence is scanned in different ways. First, on the phrase level in Level: Phrase; Feature 1-1: Word-for-word translation, it is a verbatim translation of the Hebrew. Second, on the word level in Level: Word; Category: Lexicon; Feature 3-1: Word Choice (semantic considerations), the ¡ar˙an chose carefully the Judeo-Arabic words, which translate exactly the biblical Hebrew words. In other words, ˙ÂÂ¥Ó and ¯‚‡˙ are literal translations of ¯·›Ú and ¯Á›Ò respectively. Third, on the morphosyntactic level in Level: Morphosyntactic; Category: TMA; Feature 11-1: Tense/Aspect, the participle in Judeo-Arabic ˙ÂÂ¥Ó is the exact equivalent of the participle in Hebrew ¯·›Ú. Similarly, under Level: Morphosyntactic; Category: Prepositions/Particles; Feature 5-1: Prepositions, the Judeo-Arabic preposition /li-/ in Judeo-Arabic ¯‚‡˙Ï is an exact copy of the Hebrew preposition ≠Ï ‘to’ in ¯Á›qÃÏ.

168

APPLYING THE MODEL Finally, the translation ¯‚‡˙Ï /li-tågir/ ‘to a merchant’ poses a problem. Why did the ¡ar˙an subtract the definite article in the Judeo-Arabic translation, when he actually needed it both in the verbatim translation, as the Hebrew includes it in ¯Á›qÃÏ, and in the interpretive translation, as the definite article is called for by regular Judeo-Arabic structure under Level: Morphosyntactic; Category: Definiteness; Feature 9-4: Deleting the definite article where not needed? It is possible, that the reading of ¯‚‡˙Ï is actually /li-t-tågir/ ‘to the merchant’ in a phonetic spelling, where both the alif and the låm of the Arabic definite article are not pronounced. The alif is waßla and the låm of the definite article assimilates fully to the tå<. This sort of spelling is not surprising in Late Egyptian JudeoArabic spelling, as part of the Hebraized orthography which also includes phonetic spelling (Hary 1996c). (4) È≤Ï‡ ‰È·ˆ Ï‡ ˙‡ÎÂ (15 2b,12) ‘and the young woman who...’ translates ¯˘‡ ‰¯Ú‰ ‰È‰Â (Gen 24:14). When the sentence is scanned in different categories, both literal and interpretive tendencies are revealed. On the surface, when the relative pronoun in this sentence is examined on Level: Morphosyntactic; Category: Pronouns; Feature 6-3: Relative pronouns , its role as a literal translation is apparent. As seen in chapter 8 (pp. 240–41, 6-3.1 through 6-3.3), the relative pronoun È≤Ï‡ /allaƒ•/ is a “frozen” masculine form of Later Egyptian Literary Judeo-Arabic that is used to modify any noun it follows, regardless of its gender or number. Consequently, /allaƒ•/ is the verbatim equivalent of the Hebrew relativizer ¯˘‡. Furthermore, /allaƒ•/ does not follow the regular Arabic gender paradigm on Level: Morphosyntactic; Category: Agreement; Feature 10-3: Gender, and thus is once more adhering to the verbatim translation. However, the verb ˙‡ÎÂ ‘and (she) was’ is in the feminine, although it translates the masculine Hebrew ‰È‰Â. The same appears in the translation of ‰ÓÏÚ‰ ‰È‰Â (Gen 24:43) ‘let the young woman’ into ‰È·ˆ Ï‡ ˙‡ÎÂ (15 3b,14) ‘and the young woman was.’ Both translations are indeed interpretive under Level: Morphosyntactic; Category: Agreement; Feature 10-3: Gender and Level: Morphosyntactic; Category: TMA; Feature 11-1: Tense/Aspect .

169

CHAPTER SIX In fact, the ¡ar˙an did not understand the modal meaning of the Hebrew ‰È‰Â, and opted to use the feminine verb ˙‡ÎÂ in order to agree with the feminine noun in Judeo-Arabic, not following the Hebrew masculine verb ‰È‰. (5) ‰¯ÓÓ ‚¯Ó (15 19b,14) translates the Hebrew place name ‡¯ÓÓ (Gen 35:27) and can be used as a good example of the complexities of the literal/interpretive tension. On the word level (Level: Word; Category: Lexicon; Feature 3-2: Word choice (considerations of sound/appearance)), the choice of ‚¯Ó ‘meadow’ is a calque translation, since the ¡ar˙an tried to express the meaning of the name. He did so at other points in the manuscript as well (15 2a,6; 2a,9). The addition of the word ‰¯ÓÓ to the translation of the name intensifies the verbatim translation, because it is almost a copy of the Hebrew original (‡¯ÓÓ). On the segmental level (Level: Segment; Category: Orthography/Phonology; Feature 13-6: Hebrewinfluenced orthography), however, the word ‰¯ÓÓ is spelled with a final he, which may indicate a literal translation, since it imitates the Hebraized orthography, where frequently the ¡ar˙an has spelled the final alif with a he. However, the appearance of the Judeo-Arabic ‰¯ÓÓ is not exactly like the Hebrew ‡¯ÓÓ, and thus ‰¯ÓÓ is not a precise copy. This may point in the direction of interpretive translation. In another place in Genesis (49:30), the Hebrew name ‡¯ÓÓ is translated simply as ‰¯ÓÓ (15 37b,1) ‘Mamre.’ This also may indicate both verbatim translation, as the same name appears in the ¡ar˙ and the ¡ar˙an has used the Hebraized orthography with the final he, as well as interpretive translation, since the Judeo-Arabic ‰¯ÓÓ is not an exact copy of the Hebrew ‡¯ÓÓ (see chapter 9, p. 322, 13-6.8). (6) ··Ò Ï‡ ˘‡ (3 2,8) ‘what is the reason?’ translates the Hebrew ‰Ó ‘what’ in a nonverbatim manner for emphasis and clarification (Level: Phrase; Feature 1-1: Word-for-word translation). At the same time, the interrogative pronoun itself, ˘‡ (3 2,8) ‘what,’ is an Egyptian colloquial choice (chapter 4, pp. 114–15, 2.1.6.1) away from literal tendencies and more in the direction of an interpretive mode (Level: Word; Category: 170

APPLYING THE MODEL Lexicon; Feature 3-2: Word choice (considerations of sound/appearance).

On the other hand, ‡Ó ‘what’ in another manuscript of the Haggadah (93 13,1) translates the Hebrew ‰Ó in a literal manner because it is a similar word with a similar sound. (7) ÔÈÒÓµÓ ‰Á‡ ÒÈÏ (3 2,10-11) ‘we do not dip.’ This sentence translates verbatim the Hebrew ÔÈÏÈ·ËÓ Â‡ ÔÈ‡ in Level: Phrase; Feature 1-1: Word-forword translation. Furthermore, the negation phrase ‰Á‡ ÒÈÏ follows verbatim the Hebrew Â‡ ÔÈ‡ ‘we do not,’ where the ¡ar˙an created an exact Judeo-Arabic equivalent to the Hebrew under Level: Morphosyntactic; Category: Negation; Feature 4-1: Nominal and did not use the more synthetic regular Arabic /lasna/. Similarly, ÔÈÒÓµÓ, as a participle, is equivalent to the Hebrew/Aramaic participle ÔÈÏÈ·ËÓ in a literal translation in Level: Morphosyntactic; Category: TMA; Feature 11-1: Tense/Aspect. However, in another manuscript, ÂÒÓµ (93 13,2) ‘we dip’ in the imperfect is more in line with regular Egyptian Judeo-Arabic use (chapter 4, pp. 118–19, 2.2.2.2) and thus approaches the interpretive mode. (8) ¯ˆÓ ‚Â¯¡ È¥ ÈÎÁÏ ‰ÈÏÚ ‰ÈÈˆÂ (3 3,5) ‘it is our duty to tell the story of the exodus from Egypt.’ This sentence is a clear word-for-word translation of the Hebrew ÌÈ¯ˆÓ ˙‡ÈˆÈ· ¯ÙÒÏ ÂÈÏÚ ‰ÂˆÓ under Level: Phrase; Feature 1-1: Word-for-word translation, and is strange in regular Arabic style. However, in order to make the translation even more verbatim, the ¡ar˙an could have chosen the Judeo-Arabic preposition /bi-/, in imitation of the Hebrew /bE-/, in Level: Morphosyntactic; Category: Prepositions/particles; Feature 5-1: Prepositions. He did not do so, and in a slightly more interpretive translation he preferred the prevalent JudeoArabic /fi/ in order to facilitate better comprehension. Furthermore, ÈÎÁÏ ‘for us to tell’ also presents competing translation tendencies: on the one hand, under Level: Morphosyntactic; Category: Verb Conjugation; Feature 7-1: Infinitives, the translation is verbatim, because the ¡ar˙an imitated the Hebrew preposition ≠Ï in ¯ÙÒÏ with the Judeo-Arabic /li-/, although he could have chosen regular Arabic /an/. On the other hand, under Level: Morphosyntactic; Category: Verb Conjugation; Feature 7-2: 171

CHAPTER SIX Finite verbs, the translation approximates the more interpretive tendency,

because the ¡ar˙an used a finite verb, unlike the Hebrew. (9) The independent pronoun ‰Á‡ (3 2,12) ‘we’ (chapter 4, p. 112, 2.1.1) is translated from the Hebrew into colloquial Egyptian JudeoArabic interpretively under Level: Morphosyntactic; Category: Pronouns; Feature 6-1: Independent personal pronouns, but if the same pronoun is scanned on the segmental level, there is a more verbatim outcome: in Level: Segment; Category: Orthography/ Phonology; Feature 13-6: Hebrewinfluenced orthography,

the pronoun, spelled with a final he instead of alef, is in line with the Hebraized orthography, thus reflecting a literal mode.

(10) ‰‚‡Á Ò‚ ÁÒÙ Ô·¯Â˜ „Ú· ÔÈ·¯˜Ó ÒÈÏ ÁÒÙ ÌÈÈ„Î ÂÏ ÏÂ˜ ‰˙‡ Ô‡ÓÎ (3 4,14) ‘you too must spell out to him according to the laws of Passover, that we do not offer anything at all after the Passover sacrifice’ translates the Hebrew ÔÓÂ˜ÈÙ‡ ÁÒÙ‰ ¯Á‡ ÔÈ¯ÈËÙÓ ÔÈ‡ ¨ÁÒÙ‰ ˙ÂÎÏ‰Î ÂÏ ¯ÂÓ‡ ‰˙‡ Û‡Â. On the surface, under Level: Phrase; Feature 1-1: Word-for-word translation, this complex example looks like a verbatim translation of the Hebrew original, as an Arabic reader would find this sentence strange. However, upon a closer look, using the scanning model on different levels, other issues are revealed. On both the phrase and the word levels, the possessive compound ÁÒÙ ÌÈÈ„ ‘the laws of Passover’ is translated literally without any modifications to the first term of the i∂åfa and uses a Hebrew word (ÌÈÈ„), although it is different from the Hebrew original (˙ÂÎÏ‰). The idea of using a Hebrew word is in line with a more verbatim translation under Level: Word; Category: Lexicon; Feature 3-2: Word choice (considerations of sound/appearance), but the choice of a different Hebrew word is more interpretive. In other manuscripts, though, the ¡ar˙an used ÔÈÈ„ (74 2,11; 93 16,13) ‘laws’ more interpretively in an attempt to achieve an Arabic plural suffix /-•n/ rather than the Hebrew plural suffix ÌÈ-, although he may have imitated the Mishnaic Hebrew or the Aramaic suffix. However, even in mss. 74 and 93, the ¡ar˙an still used a 172

APPLYING THE MODEL Hebrew word in a literal mode on the word level, and did not modify the first term of the i∂åfa, leaving the /n¥n/ untouched, thus reflecting again a more verbatim translation on the phrase level. In yet another feature on the phrase level, the ¡ar˙an succeeded in adapting syntactically ÁÒÙ ÌÈÈ„ ‘the laws of Passover’ into the phrase ÌÈÈ„Î ÁÒÙ by using the Judeo-Arabic preposition /ka-/ in a more interpretive translation, although the kaf may indicate the Hebrew /kE-/ in ˙ÂÎÏ‰Î ÁÒÙ‰. Continuing the scanning, there are more examples on the word level. The ¡ar˙an used the Hebrew word Ô·¯Â˜2 ‘sacrifice’ in a literal mode to explain the Hebrew original interpretively, and in another manuscript he used the Arabic Ô‡·¯Â˜ (74 2,11) ‘offering’ interpretively as evidenced by the use of the alef. In yet another part of the sentence, the ¡ar˙an chose the colloquial Arabic ‰‚‡Á Ò‚ ‘nothing’ to interpret the Hebrew original under Level: Word; Category: Lexicon; Feature 3-1: Word choice (semantic considerations). Scanning further down on the morphosyntactic level, Ô‡ÓÎ ‘also,’ which translates the Hebrew Û‡Â, lacks a translation of the coordinating vav under Level: Morphosyntactic; Category: Preposition/Particles; Feature 5-2: Coordinative particles and conjunctions. Moreover, ÂÏ ÏÂ˜ ‘spell out to him’ translates verbatim the Hebrew ÂÏ ¯ÂÓ‡; however, in another of the Haggadah manuscript (74 2,11) the colloquial ÂÏÂ˜ /
2

It is possible that Ô·¯Â˜ reflects the Judeo-Arabic /qurbån/ ‘sacrifice’ where the alef is missing in scriptio defectiva.

173

CHAPTER SIX manuscripts (74 2,11 and 93 16,13) it is translated by the colloquial /ma … ¡/ in Level: Morphosyntactic; Category: Negation; Feature 4-2: Verbal more in the interpretive direction. In addition, the ¡ar˙an did not transfer the definite article from the Hebrew ÁÒÙ‰ ˙ÂÎÏ‰Î into the Judeo-Arabic ÁÒÙ ÌÈÈ„Î ‘according to the laws of Passover,’ although it was needed on Level: Morphosyntactic; Category: Definiteness; Feature 9-4: Deleting the definite article where not needed. Finally, on the segmental level ‰˙‡ is spelled with a final he in imitation of the Hebrew spelling and as part of the Hebraized orthography in a literal translation under Level: Segment; Category: Orthography/Phonology; Feature 13-6: Hebrew-influenced orthography. (11) ‰ÏÈÏ Ï‡ ¬Ï˙ (3 3,13–14) ‘that night’ translates the Hebrew Â˙Â‡ ‰ÏÈÏ‰. On the surface, when scanning the gender of the demonstrative pronoun, the outcome is an interpretive translation: the pronoun is in the feminine agreeing with the feminine Arabic noun ‰ÏÈÏ ‘night,’ but in Hebrew the pronoun Â˙Â‡ is in the masculine. However, in another manuscript of the Haggadh, ‰ÏÈÏ Ï‡ ¬Ï‡≤ (74 1,17) ‘that night’ appears, where the masculine demonstrative pronoun /ƒålik/ modifies a feminine noun /l∑la/ ‘night’ (see chapter 8, pp. 247–48, 6-4.11). This choice reflects a literal translation, so that the Hebrew masculine pronoun Â˙Â‡ is rendered with an equivalent Judeo-Arabic masculine pronoun /ƒålik/. This analysis works on Level: Morphosyntactic; Category: Agreement; Feature 10-3: Gender. The analysis becomes more complicated when this example is scanned under Level: Morphosyntactic; Category: Pronouns; Feature 6-4: Demonstrative pronouns. Under this scanning, the demonstrative pronouns in ‰ÏÈÏ Ï‡ ¬Ï˙ (3 3,13–14), ‰ÏÈÏ Ï‡ ¬Ï‡≤ (74 1,17) ‘that night’ and ÌÂÈ Ï‡ ¬Ï‡≤ (3 5,12) ‘that day’3 are translated interpretively as they follow regular Arabic style, where the demonstrative pronouns precede the modified nouns. Other demonstrative pronouns, on the other hand, are translated literally as they follow the Hebrew word 3

The latter example translates the Hebrew ‡Â‰‰ ÌÂÈ· , where the demonstrative pronoun follows the noun it modifies.

174

APPLYING THE MODEL order. For example, ‰≤‡‰ ¯ÈË¥ Ï‡ „ÈÚ ÌÂÈ (93 60,9) ‘this festival of unleavened bread,’ which follows verbatim the Hebrew ‰Ê‰ ˙ÂˆÓ‰ ‚Á ÌÂÈ; and È„‡‰ ‰ÈÂ„ Ï‡ (3 4,3) ‘this world,’ which follows literally the Hebrew ‰Ê‰ ÌÏÂÚ‰. Furthermore, in the ¡ar˙ some demonstrative pronouns are translated into colloquial Egyptian Judeo-Arabic to go along with the interpretive translation: ¯¡˙¥ÂÓ ‰„ ‰„ÂÂ‰ /hawda da muftaxar/ (3 3,7–8) ‘this (masc.) is praiseworthy,’ translating the Hebrew Á·Â˘Ó ‰Ê È¯‰; ‰È‡ˆÚ Ï‡ È„ /di l->ißåya/ (3 13,2) ‘this (fem.) is the rod,’ translating ‰ËÓ‰ ‰Ê; and ÈÏÂ„ ˙‡ÓÏÎ ˙‡Ï˙ /talåt kalimåt døli/ (3 17,18–19) ‘these three words,’ translating ÂÏ‡ ÌÈ¯·„ ‰˘ÂÏ˘. On the other hand, these last three examples also present verbatim translations since the demonstrative pronoun placement in them is similar to that of the Hebrew, following its word order. (12) „Â„Á Ï‡ ÔÈ· „‰Ú Ï‡ È¥ (3 6,14–15) ‘in the pact between the edges’ translates the Hebrew ÌÈ¯˙·‰ ÔÈ· ˙È¯Ÿ·œa. Scanning the definite article in Level: Morphosyntactic; Category: Definiteness; Feature 9-1: Adding the definite

yields an interpretive translation, as the ¡ar˙an added the definite article to the noun />ahd/ ‘pact’ to follow regular Arabic style, despite the fact that in the Hebrew original the definite article is lacking (˙È¯Ÿ·œa). In other words, the ¡ar˙an chose in this case not to slavishly copy the Hebrew text. Furthermore, scanning the preposition /fi/ under Level: Morphosyntactic; Category: Prepositions/Particles; Feature 5-1: Prepositions yields similar results. There, the phrase Ï‡ È¥ „Â„Á Ï‡ ÔÈ· „‰Ú (3 6,14–15) ‘in the pact between the edges’ also follows the interpretive translation, as the ¡ar˙an chose the more common Judeo-Arabic preposition /fi/ to translate the Hebrew ≠·. In two other manuscripts (ËÂË˘ Ï‡ ÔÈ· „‰‡· in 91 3a,8 and Ï‡ ÔÈ· „‰Ú· ËÂË˘ in 93 20,3), though, the ¡ar˙an chose the Judeo-Arabic preposition /bi-/ to imitate the Hebrew preposition in a verbatim way. article where needed

(13) ·ÒÎÓ· Â‚Â¯¡È ¬Ï‡≤ „Ú·Â ‡‡ Ú¯‡˘‡ ÂÓ„¡‡ È≤Ï‡ ·Ú˘ Ï‡ ‰Ï‡ ‰‡ÓÎÂ ÌÈ∂Ú (3 6,18–19) ‘And also the nation, whom they served, will I judge, and afterward they will leave with great possessions’ translates 175

CHAPTER SIX the Hebrew ÏÂ„‚ ˘‹Î¯· Â‡ˆÈ ÔÎ È¯Á‡Â ¨ÈÎ‡ Ô„ Â„›·ÚÈ ¯˘‡ ÈÂ‚‰ ˙‡ Ì‚Â. Although this sentence is taken from the Passover Haggadah, it is also a quotation from Gen 15:14 and is clearly a verbatim translation of the Hebrew, especially under Level: Phrase; Feature 1-1: Word-for-word translation. Scanning this sentence on the word level under Level: Word; Category: Lexicon; Feature 3-1: Word choice (semantic considerations) indicates the literal/interpretive tension. The ¡ar˙an translates the Hebrew word Ô„ ‘judge’ into Judeo-Arabic Ú¯‡˘‡ /u¡åri>/ ‘I enact a law’ with a slightly different meaning, placing the translation closer to the interpretive side, as opposed to the translation ÌÂÎÁ‡ /a˙kum/ ‘I judge,’ appearing in other manuscripts of the Haggadah (91 3a,11; 93 20,6), which is closer to the literal side. Similarly, scanning under Level: Word; Category: Lexicon; Feature 3-2: Word choice (considerations of sound/appearance) in Â‚Â¯¡È ÌÈ∂Ú ·ÒÎÓ· (3 6,19) ‘they will leave with great possessions,’ the ¡ar˙an chose the Judeo-Arabic preposition /bi-/ to translate the Hebrew ≠·. This is an exact rendering of the Hebrew, despite the fact that the ¡ar˙an could have chosen the more regular Arabic preposition /fi/. Down on the morphosyntactic level, the verbatim translation is seen in Level: Morphosyntactic; Category: Cases; Feature 8-1: Accusative , where the ¡ar˙an copied the Hebrew direct definite object marker ˙‡ /et/ into Judeo-Arabic ‰Ï‡ /ila/ as is so usual, and in fact almost obligatory, in Egyptian ¡ur¥˙: ·Ú˘ Ï‡ ‰Ï‡ ‰‡ÓÎÂ ‘and also the nation’ translating the Hebrew ÈÂ‚‰ ˙‡ Ì‚Â. Similarly, under Level: Morphosyntactic; Category: TMA; Feature 11-1: Tense/Aspect, the verbs ÂÓ„¡È (91 3a,11) and ÂÓ„¡˙ÒÈ (93 20,6) ‘(they) serve’ are both in the imperfect, following the Hebrew Â„›·ÚÈ ‘they serve’ literally; however, in another manuscript the ¡ar˙an employed the Judeo-Arabic perfect ÂÓ„¡‡ (3 6,18) ‘(they) served’ in a more interpretive direction. Further down on the segmental level, under Level: Segment; Category: Orthography/Phonology; Feature 13-6: Hebrew-influenced orthography, the influence of the Hebraized orthography is seen in the direction of literal translation. Both the preposition ‰Ï‡ that marks the definite direct object and the independent pronoun ‰‡ (91 3a,11) ‘I’ clearly follow 176

APPLYING THE MODEL the literal direction, as they are directly influenced by Hebrew orthography, with its use of Judeo-Arabic final he to denote both the alif maqß¥ra bi-ßurat il-yå< and final alif respectively. In two other manuscripts, on the other hand, the spelling of ‡‡ (3 6,19; 93 20,6) ‘I’ is more interpretive as it follows the Arabicized orthography. (14) The phrase È„‡‰ ‰ÈÂ„ Ï‡ (3 4,3) ‘this world,’ which translates ‰Ê‰ ÌÏÂÚ‰, reveals opposing directions of translation. Scanning the morphosyntactic level under Level: Morphosyntactic; Category: Pronouns; Feature 6-4: Demonstrative pronouns, the demonstrative pronoun word order follows the Hebrew word order in a literal translation, although a different word order in Literary Arabic could have been expected. The ¡ar˙ translation, though, follows the same word order as colloquial Arabic, /id-dunya di/. On the other hand, scanning on Level: Morphosyntactic; Category: Agreement; Feature 10-3: Gender, the gender of the demonstrative pronoun is translated interpretively, in accord with the feminine gender of the Arabic noun it modifies and not the gender of the Hebrew masculine noun ÌÏÂÚ ‘world.’ Down on the segmental level under Level: Segment; Category: Orthography/Phonology; Feature 13-6: Hebrew-influenced

‰ÈÂ„ exhibits the literal/interpretive linguistic tension as the word ends with a he in imitation of the Hebrew orthography in a literal mode, but in a different manuscript the spelling ‡ÈÈÂ„ (93 13,2) reveals the interpretive mode, since the final alef, representing the Arabicized orthography, follows the Arabic spelling with its final alif. orthography,

(15) The sentence Ï‡¯˘È ÂÓÂ˜Ï ‰ÚÈ¯˘ Ï‡ ‰ËÚ È≤Ï‡ ¬Â¯·Ó (74 2,7) ‘Blessed is He who has given the Torah to His people, Israel,’ which translates the Hebrew Ï‡¯˘È ÂÓÚÏ ‰¯Â˙ Ô˙˘ ÍÂ¯·, can be analyzed differently in various categories. Scanning under Level: Morphosyntactic; Category: Cases; Feature 8-1: Accusative results in a verbatim translation: the Judeo-Arabic sentence lacks the definite object marker /ila/ just as the Hebrew does not employ /et/. This occurs despite the fact that the ¡ar˙an added the definite article in the ¡ar˙ (see chapter 8, p. 259, 8-1.7). On the other hand, in another category (Level: Morphosyntactic; Category: 177

CHAPTER SIX Definiteness; Feature 9-1: Adding the definite article where needed)

the definite article appears in the ¡ar˙, although it does not in the Hebrew original. This phenomenon is analyzed interpretively, as the presence of the definite article deviates from its absence in the Hebrew original.

Calque Translations Borrowing, or the introduction of linguistic features from one language to another, is common in the texts of the ¡ur¥˙. Borrowing is a gradual process: first a linguistic feature is introduced into the host language through a bilingual community and is not yet adapted phonologically or morpho-logically. Once it is integrated into the host language and adopted by the monolingual community, it becomes part of the host language (Mahootian 2006:513). One type of borrowing is the loan, a “linguistic unit (usually a lexical item) which has come to be used in a language or dialect other than the one where it originated” (Crystal 2003:275). According to Crystal, there are several kinds of loan processes: (i) Loan words, where both form and meaning are adopted by a host language with some adaptation in the phonological system. For example, some Palestinians use the morpheme /p/ in their Arabic in words borrowed from the Hebrew: /kupatxolim/ ‘health maintenance organization, HMO’; Japanese takshi from English taxi. (ii) Loan blends, where the meaning is borrowed along with only partial borrowing of the form. For example, when some Palestinians borrow Hebrew words and insert them into their dialect, they only change some phonological features of the borrowed word. Thus, in /kubatxolim/, the Hebrew voiceless bilabial stop /p/ is changed to the voiced equivalent /b/, because the former does not exist in the Arabic phonemic inventory; however, the voiceless velar fricative /x/ is not changed to the native Arabic voiceless pharyngeal fricative /˙/. (iii) Loan shifts, where the meaning is borrowed, but the form follows 178

APPLYING THE MODEL the phonological behavior of the host language. In Hebrew /xumus/ ‘hummus’ is taken from Arabic /˙ummuß/;4 Japanese maketo ‘indoor market’ is borrowed from English market. (iv) Loan translations, where the morphemes of the borrowed word are translated verbatim. For example, accomplished fact (from French fait accompli), superman (from German Übermensch), Hebrew /gored¡xakim/ (from English skyscraper), Hebrew /zemargi¡tov/ (from English it feels good). Loan translations are also known as calque translations, a term used in historical linguistics, comparative linguistics, and sociolinguistics “to refer to a type of borrowing, where the morphemic constituents of the borrowed word or phrase are translated item by item into equivalent morphemes in the new language” (Crystal 2003:61). The term comes from French calquer ‘to trace, to copy.’ Bussmann cites four processes for the creation of calques (1996:62): (i) Through meaning that is borrowed through change and expansion of the meaning in the new language. For example, write, originally ‘to scratch,’ influenced by Latin scribere. (ii) Through a neologism, which is loosely based on a foreign concept. (iii) By way of a verbatim loan translation of the several morphemes. For example, crispbread (from German Knäckebrot), power politics (from German Machtpolitik). (iv) Through a loose loan translation. For instance, brotherhood from Latin fraternitas. Because of the ¡ar˙an’s desire to translate the sacred text verbatim, and sometimes because of his poor understanding of the text, many calque translations found their way into the ¡ar˙. In calque translations the ¡ar˙an has translated the original text word for word, reflecting each word with an equivalent in the target language. Furthermore, the 4

Thus, in the target language, in this case Hebrew, /˙/ becomes /x/, /ß/ becomes /s/, and the geminate is abolished. Egyptians always say that they can identify Israeli tourists in Cairo by their typical (Naot) sandals and their frequent request to eat /xumus/, not /˙ummuß/.

179

CHAPTER SIX ¡ar˙an has also translated into the host language (Judeo-Arabic) the morphemes of the guest language (Hebrew or Aramaic). The end result usually cannot be understood in the host language if knowledge of the original text is missing or inadequate. The following is a representation of several examples from each of the ¡ur¥˙ discussed in this book: •

•

•

•

Genesis: Ú„Â ‡ÏÂ (Gen 41:21) ‘and it was not known’ is rendered ¥¯Ú ÌÏÂ (15 25b,15–16) ‘and we did not know.’ In this calque translation the ¡ar˙an has “copied” the Hebrew nif>al verbal prefix as a JudeoArabic imperfect first person plural verbal prefix /n-/, thus changing the meaning slightly. This is done in order to achieve a verbatim translation (chapter 8, p. 217, 4-2.2). ‰ÏÙÎÓ‰ ˙¯ÚÓ (Gen 25:9) ‘the cave of Machpela’ is translated by a Judeo-Arabic calque: ‰È˙Ó Ï‡ ˙¯‡µÓ (15 5a,2) ‘the cave of the double.’ In this case the Hebrew ‰ÏÙÎÓ is translated verbatim into Judeo-Arabic ‰È˙Ó to indicate the meaning of the name in a calque translation (chapter 9, pp. 322–23). È‡] ÈÁÏ ¯‡· (Gen 25:11) ‘Be>er La˙ai Ro>i’ is translated word for word by the ¡ar˙an into È‡¯∂ ÈÈÁ Ï‡ ¯È· (15 5a,6) ‘a well of the living, seeing.’ It is clear that this phrase does not make sense on its own; it is nonetheless a calque translation of the Hebrew place name (chapter 9, p. 323). ¯˘‡Î (Gen 24:22 and other places) ‘when, where, as’ is analyzed by the ¡ar˙an as consisting of two morphemes: ≠Î and ¯˘‡; thus the conjunction is translated verbatim into Judeo-Arabic, also with the two attached morphemes, /ka-/ and /allaƒ•/, resulting in È≤Ï‡Î (15 3a,4 and other places) ‘when.’ In Esther the conjunction is also rendered as a calque translation, but in a more analytic structure: È≤Ï‡ ‡ÓÎ (1302 2b,22 and more) ‘when’ (chapter 8, p. 232, 5-2.3).

Esther: • ‰Ò„‰ (Esth 2:7) ‘Hadassah’ is rendered ‡ÈÒ¯Ó (1302 2b,1) ‘Mars•na’ 180

APPLYING THE MODEL

•

(myrtle). The ¡ar˙an actually transferred the meaning of the Hebrew name ‘Hadassah’ (myrtle) into the Judeo-Arabic calque translation (see chapter 9, p. 323, 13-6.8). The Hebrew name ‡˙„Ó‰ ‘Hammedatha’ (Esth 9:10) is rendered Ï‡ ‡˙„Ó (1302 7a,1) ‘the Medatha.’ The ¡ar˙an has mistakenly interpreted the Hebrew name as having two morphemes: the definite article /ha-/ and the name ‡˙„Ó, and thus translated it accordingly (see chapter 7, p. 191, n. 3).

Haggadah: • „‚Î ‘as against’ is reflected verbatim in „‡ˆÂ˜Î (3 4,8), clearly a calque translation (see chapter 8, p. 224, 5-1.15). • Hebrew „ÂÚ·Ó ‘whilst’ has the Judeo-Arabic equivalent ÔÈÁ· ÔÓ (3 5,13), composed of three parts (see chapter 8, pp. 224–25, 5-1.16). • ‰ÓÎÂ ‰ÓÎ ˙Á‡ ÏÚ ‘how much more so?’ is reflected literally in the ¡ar˙: ˘È‡ „˜Â ˘È‡ „˜ ‰„Á‡Â ‰ÏÚ (3 17,5), where each of the Hebrew words has a Judeo-Arabic equivalent. • ˙È·‰ ˙¯˜Ú ‘the mistress of the house’ is reflected verbatim in ˙ÓÈ˜Ú ˙È· Ï‡ (3 20,10), where the ¡ar˙an has translated Hebrew ˙¯˜Ú into ‰ÓÈ˜Ú ‘sterile,’ using its root ‘sterile, barren.’ This is indeed the intended meaning in Ps 113; however, in modern Hebrew the meaning has changed into ‘the mistress of the house.’ • „‡Ó È˙ÈÚ È‡ ‘I am greatly afflicted’ is translated literally as ‡‡ ÈÂÂ˜ ˙·ÂÂ‡‚ (3 24,20) ‘I greatly answered.’ The Hebrew verb È˙ÈÚ stems from the root >-n-h, which can mean both ‘poor’ and ‘answer.’ The ¡ar˙an has chosen the second meaning and arrived at the bizarre calque translation (chapter 7, pp. 201–2, 3-1.5).

181

CHAPTER SIX The Organization of the Examples The organization of the examples according to the categories presented in table 4 facilitates the examination of the literal/interpretive linguistic tension as well as the translation mechanisms (A and B). In the following chapters of part 2, I cite examples of literal translations (Lit.) as well as interpretive translation (Int.) for each feature. Following the literal and the interpretive examples, I present cases, when possible, that illustrate both literal and interpretive elements at the same time (L/I). Seldom do I use the same example to illustrate different features. Obviously, for reasons of space and redundancy, not all the examples that can be extracted from the texts of the ¡ur¥˙ can be presented, but usually at least three examples from each text for every feature are offered. The examples are taken from Genesis, Esther, and the Passover Haggadah and presented in that order. The organization of the examples employs a numbering system that follows table 4 (pp. 81–82). Each feature is divided into three parts: literal translation (Lit.), interpretive translation (Int.) and literal/ interpretive translation (L/I). The examples are marked in consecutive numbers; thus the examples of Level: Morphosyntactic; Category: Verb Conjugation; Feature 7-1: Infinitives, for instance, are numbered from 7-1.1 through 7-1.15, where 7-1.1 through 7-1.8 are examples of literal translation, 7-1.9 through 7-1.12 are examples of interpretive translations, and 7-1.13 through 7-1.15 demonstrate the literal/interpretive translation with its linguistic tension (chapter 8, pp. 250–56). In order to facilitate reading the text and finding the various examples, the headers of the odd-numbered pages in chapters 7 through 9 change in order to indicate the various features that are discussed within each chapter. Thus, the reader can use the headers as a guide to the contents of each chapter, reducing the constant need to refer to table 4. For purposes of comparison, when deemed necessary, both Saadia Gaon’s translation (Derenbourg 1893) and the Protestant Arabic translation of the Bible (Al-Kitåb al-Muqaddas 2004) are provided.

182

CHAPTER SEVEN THE PHRASE AND THE WORD LEVELS This chapter analyzes the ¡ur¥˙ of Genesis, Esther, and the Passover Haggadah, scanning selected examples through the phrase and word levels. Based on the theoretical model of the analysis of the ¡ar˙, including table 4, presented in chapter 3 (pp. 81–82), many examples of the different categories and features appear below. Note the explanation of the organization of the examples at the end of chapter 6 (p. 182). Note also that the headers on the odd-numbered pages of this chapter include, in addition to the name of the chapter as in the rest of the volume, the various features discussed on the relevant pages for easier referencing. Word-for-Word Translation Level: Phrase; Category 1 Feature 1-1 — Word-for-word translation Lit.: The following examples are word-for-word translations, which are typical of the ¡ar˙: (1-1.1) In Genesis: The translation of Ï˜˘ ˙›‡Ó Ú·¯‡ ı¯‡ ÈÚÓ˘ ÈI‡ ¯›·˜ Í˙Ó ˙‡Â ‡È‰ ‰Ó ÍÈ·Â ÈÈ· ÛÒÎ (Gen 23:15) into Judeo-Arabic ‰Ï‡Â ‡È‰ ˘È‡ ÍÈ·Â ÈÈ· ‰ƒ¥ Ï‡˜˙Ó ˙È‡Ó Ú·¯‡ √¯‡ ‰ÚÓÒ‡ È„ÈÒ Ô¥„‡ ¬˙ÈÈÓ (15 2a,1-2) ‘My master, hear us! Land worth four hundred silver (coins), what is it between me and you? (Go and) bury your dead’ is a clear word-for-word translation with the exception of the first plural pronominal suffix /-na/ in ‰ÚÓÒ‡, rather than the singular pronoun /-ni/ in the Hebrew ÈÚÓ˘. Both Saadia and the Protestant translation of the Bible

CHAPTER SEVEN offer more interpretive translations for this verse. Saadia in his translation, »‰ƒÙ Ï‡˜«˙Ó »‰È‡Ó Ú·¯‡ ÈÂ‡Ò˙ »‰ƒÂ¯ ÈÓ ÚÓÒ‡ È„ÈÒ ‡È Í˙ÈÓ ‡‰ÈÙ ÔÙ„‡ ÍÈ·Â ÈÈ· È‰ ‡Ó, added a few words after /isman• ar∂un bi-arba>i mi
THE PHRASE AND THE WORD LEVELS: WORD-FOR-WORD TRANSLATION Abimelech king of the Philistines looked down through the window and saw, behold! Isaac was jesting with Rebecca his wife.’ (1-1.2) In Esther: The sentence Â˙ËÏÒ ¯‡˜Â‡ ‡µ ‡Ï‡ Â¯Â∂ È¥ ¯‡˙Î Ì‡ÈÈ‡ Â˙ÈÈ¯·Î ˙ÈÈ¯¡¥ ¯‡˜Â‡ ‡Ï‡Â (1302 1b,5–6) ‘with his display of the richness of his glorious kingdom and the honor of his splendorous majesty for many days’ is clearly a word-forword translation of the Hebrew ¯˜È ˙‡Â Â˙ÂÎÏÓ „Â·Î ¯˘Ú ˙‡ Â˙›‡¯‰· ÌÈ·¯ ÌÈÓÈ Â˙ÏÂ„‚ ˙¯‡Ù˙ (Esth 1:4), which actually yields some sort of an “un-Arabic” sentence. The same is true for Ï‡ ¯Ó‡ ‡„Î Ô‡ ·ˆµÓ ÒÈÏ ÔÂ‡˜ Ï‡Î ·¯˘ Ï‡Â Ï‚¯Â Ï‚¯ ‡ƒ¯Î ÚˆÈÏ Â˙È· ¯È·Î ÏÎ ‡ÏÚ ¬ÏÓ (1302 1b,5–6) ‘and the drinking was according to the law, there was no coercion, for the king had ordered all the officers of his house to do according to every man’s will,’ which translates verbatim ÔÈ‡ ˙„Î ‰È˙˘‰Â ˘È‡Â ˘È‡ ÔÂˆ¯Î ˙Â◊ÚÏ Â˙È· ·¯ ÏÎ ÏÚ ÍÏÓ‰ „ÒÈ ÔÎ≠ÈÎ Ò›‡ (Esth 1:8). Similarly, the sentence ‡Ï‡ ¬ÏÓ Ï‡ ÔÈ‡„Ó ÏÎ ‡Ï‡ ·È˙‡ÎÓ ÏÒ¯Â ‡‰˙·‡˙ÎÎ ‰È„ÓÂ ‰È„Ó (1302 2a,9) ‘and he sent letters to all the kingdom’s provinces, to each province in its own script’ is a verbatim translation of the Hebrew ÍÏÓ‰ ˙ÂÈ„Ó ÏÎ Ï‡ ÌÈ¯ÙÒ ÁÏ˘ÈÂ ‰·˙ÎÎ ‰È„ÓÂ ‰È„Ó Ï‡ (Esth 1:22). In the Protestant translation, which is less verbatim and more in line with Arabic structure, we see /wa-arsala kutuban ilå kulli buldåni l-maliki ilå kulli bilådin ˙asaba kitåbatihå/. (1-1.3) In the Haggadah: The Judeo-Arabic translation ‰ÈˆÂ ¯ˆÓ ‚Â¯¡ È¥ ÈÎÁÏ ‰ÈÏÚ (3 3,6) ‘It is incumbent upon us to tell the exodus from Egypt’ is a word-for-word translation of the Hebrew original ÌÈ¯ˆÓ ˙‡ÈˆÈ· ¯ÙÒÏ ÂÈÏÚ ‰ÂˆÓ, although, as in example 5-1.34 in chapter 8 (pp. 228–29), the preposition /bi-/ could have been used in an even more verbatim translation. Similarly in the Haggadah, È‚Â¯¡ È¥ ÈÏ ‰ÏÏ‡ Úˆ ‰„ ··Ò È¥ ÂÏ ÏÂ˜Â ÂÏ ÌÏÂ ÈÏ ¯ˆÓ ÔÓ (3 4,20–5,1) ‘and say to him, it is because of that which God did for me when I went out from Egypt, for me 185

CHAPTER SEVEN and not for him’ is a verbatim translation of the Hebrew original ÂÏ ‡ÏÂ ÈÏ ÆÌÈ¯ˆÓÓ È˙‡ˆ· ÈÏ ÈÈ ‰˘Ú ‰Ê ¯Â·Ú· ÂÏ ¯ÂÓ‡Â. Here too, the preposition /bi-/ could have been used in an even more verbatim translation. Likewise, the Judeo-Arabic translation ËÈµ Ï‡ ˙‡·Î ‰ÂÂ·¯ ¬˙ÏÚ‚ (3 8,12–13) ‘I have made you multiply ten thousand like the vegetation of the field’ is a word-for-word translation that reproduces the word order of the Hebrew original ÁÓˆÎ ‰··¯ ÍÈ˙˙ ‰„˘‰. Sometimes, the word-for-word translation is exact, disregarding any medieval or other interpretation of the biblical text: (1-1.4) The sentence ‰¯˘Ú ÂÂ‡ ‰ÚÓ ‰È·ˆ Ï‡ „Ú˜˙ (15 3a,14) ‘Let the girl stay with us two days or ten’ is an almost exact translation of the Hebrew ¯Â◊Ú Â‡ ÌÈÓÈ Â˙‡ ‰¯Ú‰ ·˘˙ (Gen 24:55), except for the translation of the Hebrew ÌÈÓÈ ‘days’ into Judeo-Arabic ÔÈÓÂÈ ‘two days.’ This translation does not take into account the medieval interpretations of ‘a year or ten months.’ The Protestant translation is similar, but Saadia follows the medieval interpretation. Similarly, the Judeo-Arabic Â‰ÓÂ‡ „Ú· ˜ÁˆÈ ‰ÏÒ˙‡Â (15 4b,10–11) ‘and Isaac was comforted after his mother’s (death)’ is a literal translation of the Hebrew ÂÓ‡ È¯Á‡ ˜ÁˆÈ ÌÁÈÂ (Gen 24:67). Both Saadia and the Protestant translation used /ba>da mawt ummihi/ ‘after his mother’s death’ interpretively. Int.: On the other hand, at times the ¡ar˙an has added to the translation interpretive words lacking in the Hebrew original. (1-1.5) In Genesis: The sentence Â˙˘‡ ‰˜·¯ ¯‰˙Â ß‰ ÂÏ ¯˙ÚÈÂ (Gen 25:21) in the ¡ar˙ is an interpretive translation. In his translation Â˙‡¯Ó ˙„ÏÂÂ ˙Ï·ÁÂ ‰ÏÏ‡ Â˙‡Ïˆ Ï·˜Â (15 5b,3) ‘God accepted his prayer, and his (Isaac’s) wife became pregnant and gave birth,’ the ¡ar˙an “explained” Hebrew ¯˙ÀÚÈÂ ‘allowed Himself to be entreated’ as Â˙‡Ïˆ Ï·˜Â ‘accepted his prayer’; he did not mention 186

THE PHRASE AND THE WORD LEVELS: WORD-FOR-WORD TRANSLATION Rebecca by her name although it was mentioned in the Hebrew text, and he added ˙„ÏÂÂ ‘and she gave birth’ for clarification. In ‰ÏÏ‡ „Ú ÔÓ ¯Ó‡ ·ÂÏË˙Ï ˙Á‡¯Â (15 5b,5) ‘she went to ask God (about) the issue,’ the ¡ar˙an added the noun ¯Ó‡ ‘issue’ in order to clarify the Hebrew equivalent, ß‰ ˙‡ ˘¯„Ï ÍÏ˙Â (Gen 25:22). Similarly, the Judeo-Arabic sentence ÔÓÊ ¯‰˘ ‰È‡ÚÓ „ÂÚ˜Â˙Â (15 10b,11–12) ‘and stay with me a month’s time’ is a clear interpretation of the Hebrew equivalent ÌÈÓÈ ˘„›Á ÂÓÚ ·˘ÈÂ (Gen 29:14) ‘and he stayed with him for a month.’ First, in the Judeo-Arabic translation the person is changed, from the third (·˘ÈÂ) to the second („ÂÚ˜Â˙) and in another place from the third (ÂÓÚ) to the first (‰È‡ÚÓ), using direct speech; and second, the translation of ÌÈÓÈ ‘days’ is expressed as ÔÓÊ ‘time’ in the JudeoArabic ¡ar˙. (1-1.6) In Esther: The translation Ï‡ ÔÈÓ„‡¡ Ô‡È·ˆ Ï‡ ÂÏ‡˜Â Ô‡ Â‰Ï Ô‡ËÏÒ (1302 2a,12) ‘The courtiers who served the ruler said …’ does not follow the Hebrew ÂÈ˙¯˘Ó ÍÏÓ‰ È¯Ú Â¯Ó‡ÈÂ (Esth 2:2) word for word, in a probable attempt to clarify it. Similarly, in ¬ÏÓ Ï‡ ˙È· ‡Ï‡ Ô‡ÓÎ È‚˙ ÌÏ (1302 2b,13) ‘she would not come again to the king’s house,’ translating the Hebrew ÍÏÓ‰ Ï‡ „ÂÚ ‡Â·˙ ‡Ï (Esth 2:14), the ¡ar˙an added the word /b∑t/ ‘house’ in an interpretive manner. Likewise, the phrase ÈÈÓÈ Ï‚¯ (1302 2a,17–18) may seem at first glance to be a literal translation of ÈÈÓÈ ˘È‡ ‘a Benjaminite’ (Esth 2:5), but the addition of ÔÈÓÈ· Ë·Ò ÔÓ (1302 2a,18) ‘from the tribe of Benjamin’ is clearly interpretive, so that readers would fully understand the phrase. (1-1.7) In the Haggadah: „‡ÏÂ‡ ÂÂÎ Ï‡¯˘È √¯‡ È¥ ‰È‡‚ Ï‡ ‰Ò Ï‡ ÔÈ˜Â˙ÚÓ (3 2,6–7) ‘next year we will be free children in the Land of Israel.’ Clearly this sentence does not follow the Hebrew verbatim, ÔÈ¯ÂÁ È· ‰‡·‰ ‰˘Ï. The ¡ar˙an “explained” the context by adding the phrase ÂÂÎ Ï‡¯˘È √¯‡ È¥ ‘in the Land of Israel we 187

CHAPTER SEVEN will be …’ for clarification. The deviation from the literal translation may reflect the need to adjust the translation to a different readership: the biblical text is composed for the people in the Land of Israel, whereas the Judeo-Arabic Haggadah translation is for the people in the Diaspora, hence the need to add the phrase ÂÂÎ Ï‡¯˘È √¯‡ È¥ ‘in the Land of Israel we will be …’ It is common to find that the Hebrew ¯Ó‡˘ and ¯Ó‡˘ ‰ÓÎ are translated into the Judeo-Arabic ¡ar˙ as ˜ÂÒÙ Ï‡ Ï‡˜ ‡ÓÎ (3 3,21 and many other places) ‘as the (biblical) verse reads’ in an interpretive translation by adding the word ˜ÂÒÙ Ï‡ ‘the verse’ (see also below, p. 211, 3-2.21). Furthermore, ··Ò Ï‡ ˘‡ (3 2,8) ‘what is the reason?’ This phrase translates the Hebrew ‰Ó ‘what’ in a nonverbatim manner for emphasis and clarification. Moreover, in an interpretation of the Hebrew ¯Â·„‰ ÈÙ≠ÏÚ ÒÂ‡ ‘compelled by the word,’ the ¡ar˙an wrote in Judeo-Arabic ˜Ï‡¡ Ï‡ ÏÂ˜ ‰ÏÚ ·ÂˆµÓ (3 7,14) ‘compelled by the creator’s words.’ L/I: (1-1.8) The Hebrew phrase ÁÈ˘Ó‰ ˙ÂÓÈ ‘the days of the Messiah’ is translated literally in ms. 93 15,16 ÁÈ˘Ó Ï‡ Ì‡ÈÈ‡; however, it is interpreted in ms. 3 4,4 ÁÈ˘Ó Ï‡ ˙˜Â ‘the time of the Messiah.’ In ms. 74 2,6 both options are used in an interpretive mode: ÁÈ˘Ó Ï‡ Ì‡ÈÈ‡ ˙˜Â ‘the time of the days of the Messiah.’ Word Order: Syntactic Adaptation Level: Phrase; Category 2: Word Order Feature 2-1 — Syntactic adaptation Lit.: It is usual in the ¡ar˙ to find the verb preceding its subject in typical VSO Arabic word order when this is also the case in the Hebrew original. For example, both /wa-˙alifu/ ‘and they swore’ and /wa-rasalhum/ ‘and (he) sent them’ precede their respective subjects /rågil/ ‘man’ and ‘Isaac’ in ÌÂ‰ÏÒ¯Â ‰Â¡‡Ï Ï‚‡¯ Â¥ÈÏÁÂ

188

THE PHRASE AND THE WORD LEVELS: SYNTACTIC ADAPTATION ˜ÁˆÈ (15 7a,12) ‘and they swore to one another and then Isaac sent them away,’ following similar word order in the Hebrew verse ˜ÁˆÈ ÌÁÏ˘ÈÂ ÂÈÁ‡Ï ˘È‡ ÂÚ·˘ÈÂ (Gen 26:31). However, the morphosyntactic adaptation of the Hebrew text does not extend to the number of the verb, as the verb Â¥ÈÏÁÂ /wa-˙alifu/ ‘and they swore’ is in the plural, following literally its Hebrew equivalent ÂÚ·˘ÈÂ. Moreover, if the Hebrew original exhibits an SVO word order, where the verb follows the subject, the ¡ar˙an has not adapted it syntactically to the Arabic but has translated this word order verbatim in an SVO word order, as the following examples show. (2-1.1) In Genesis: In the translation of Gen 24:35 ˙‡ Í¯· ß‰Â „‡Ó È„‡, a similar SVO word order is revealed in the ¡ar˙: ‰ÏÏ‡Â ÈÂÂ˜ È„ÈÒ ‰Ï‡ ¬¯‡· (15 3b,2) ‘and God greatly blessed my master.’ In the same way the SVO word order occurs in ‰Ï‡ ¥Ï¡ Ô‡˘˜ÈÂ ‡·˘ (15 4b,13) ‘and Yokshan begot Sheba’ in order to slavishly follow the Hebrew ‡·˘ ˙‡ „ÏÈ Ô˘˜ÈÂ (Gen 25:3). The same applies to ·Â˜ÚÈ ‰Ï‡ ˙·Á ‰˜·¯Â (15 5b,15) ‘and Rebecca loved Jacob,’ which follows the same SVO word order of the Hebrew ‰˜·¯Â ·SÚÈ ˙‡ ˙·‰›‡ (Gen 25:28). (2-1.2) In Esther: ¯ˆ˜ Ï‡ Ô˘Â˘ È¥ Ô‡Î È„Â‰È Ï‚¯ (1302 2a,17) ‘There was a Jewish man in Shushan, the capital’ contains the same SVO word order of the Hebrew in Esth 2:5, ‰È‰ È„Â‰È ˘È‡ ‰¯È·‰ Ô˘Â˘·. Likewise, the Judeo-Arabic text keeps the Hebrew SVO word order in ¬ÏÓ Ï‡ ¯Ó‡· ÔÈÚÂÙ„Ó Â‚¯¡ ÔÈÈ‡¯‚ Ï‡ (1302 3b,8) ‘The runners went out, rushed by the king’s command’ in a literal translation. Similar is the SVO word order in both Ï‡Â ‡ËÚ‡ ¯Ó‡ (1302 3b,8) ‘the command was given’ and in ¬ÏÓ Ï‡ ÂÙÈÈÎ˙È ÂÒÏ‚ ÔÓ‰Â (1302 3b,8–9) ‘the king and Haman were sitting and enjoying themselves,’ following the original Hebrew texts ‰˙ ˙„‰Â and ˙Â˙˘Ï Â·˘È ÔÓ‰Â ÍÏÓ‰Â (Esth 3:15) respectively. (2-1.3) In the Haggadah: ¯ˆÓ È¥ ÔÂÚ¯¥Ï ‰ÂÎ „È·Ú (3 2,19) ‘we were Pharaoh’s slaves in Egypt’ exhibits SVO word order just 189

CHAPTER SEVEN like the Hebrew original, ÌÈ¯ˆÓ· ‰Ú¯ÙÏ ÂÈÈ‰ ÌÈ„·Ú. Similarly, in Ï‡Â ÂÏÂ˜È ‰ÓÏÂÚ (3 4,2–3) ‘and the sages say,’ the verb follows its subject to imitate Hebrew word order: ÌÈ¯ÓÂ‡ ÌÈÓÎÁÂ. In the same way, the verb in ¯ˆÓ ÂÏÊ Â„‡ÏÂÂ‡Â (91 3a,5) ‘and his children went down into Egypt’ follows the subject in an SVO word order, just like the Hebrew original, ÌÈ¯ˆÓ Â„¯È ÂÈ·Â. This type of verbatim translation and avoidance of syntactic adaptation in the Judeo-Arabic structure may even occur when the Hebrew original uses an unusual word order for emphasis or stylistic reasons, as in the following examples with OV word order. (2-1.4) In Genesis: The Hebrew word order OV in Ô˙ ˙$„‚ÓÂ ‰Ó‡ÏÂ ‰ÈÁ‡Ï ‘and he gave gifts to her brother and her mother’ (Gen 24:53) is retained in the ¡ar˙ in a verbatim translation, ‡‰ÓÂ‡ÏÂ ‡‰Â·‡Ï ÏÚ‚ ˙‡È„‰Â (15 4a,10–11) ‘and he gave gifts to her father 1 and her mother.’ The same applies to Ô‡È„‚ „ÂÏ‚ ‰Ï‡Â ‰È„È‡ ‰ÏÚ ˙Ò·Ï ÊÚ‡Ó Ï‡ (15 7b,20–8a,1) ‘with the skins of the goat-kids she dressed his arms,’ where the OV word order follows the same word order as in ÂÈ„È ÏÚ ‰˘È·Ï‰ ÌÈÊÚ‰ ÈÈ„‚ ˙]›Ú ˙‡Â (Gen 27:16). Similarly, the OV word order in ¬Ï Â˙ÏÚ‚ „ÈÒ (15 8b,12) ‘a master have I installed over you’ follows the Hebrew OV word order in ÍÏ ÂÈ˙Ó˘ ¯È·‚ (Gen 27:37). (2-1.5) In Esther: The ¡ar˙an has employed OV word order in ‡‰˙˜È¥¯Ï ¬ÏÓ Ï‡ ÈËÚÈ ‡‰˙ËÏÒÂ (1302 2a,6) ‘and the queenship will the king give to her friend,’ in order to translate the Hebrew OV word order in ‰˙ÂÚ¯Ï ÍÏÓ‰ Ô˙È ‰˙ÂÎÏÓÂ (Esth 1:19). In ˙¯˘ÚÂ ‰Úˆ Ï‡ È≤‡‰ ÔÈÚ‡ˆ „È ‡ÏÚ ÔÊÂ‡ ‰ƒ¥ Ï‡˜˙Ó ¥‡Ï‡ (1302 3a,17–18) 1

The word /li-ab¥ha/ ‘to her father’ in ‡‰ÓÂ‡ÏÂ ‡‰Â·‡Ï ÏÚ‚ ˙‡È„‰Â (15 4a,10–11) ‘and he gave gifts to her father and her mother’ mistranslates the Hebrew ‰ÈÁ‡Ï ‘to her brother’ (Gen 24:53). Later in the chapter, this mistake does not occur in Ï‡˜Â ‡‰ÓÂ‡Â ‡‰Â¡‡ (15 4a,13–14) ‘her brother and mother said,’ translating the Hebrew ‰Óœ‡Â ‰ÈÁ‡ ¯Ó‡ÈÂ (Gen 24:55). See Hary 2009, chapter 3, “Mistakes and Omissions in the Texts.”

190

THE PHRASE AND THE WORD LEVELS: SYNTACTIC ADAPTATION ‘I will weigh out ten thousand miskals2 of silver into the hands of those who are engaged in this business,’ the ¡ar˙an has used OV word order to imitate the Hebrew original in ¯ÎÎ ÌÈÙÏ‡ ˙¯˘ÚÂ ‰Î‡ÏÓ‰ È◊›Ú È„È ÏÚ ÏÂ˜˘‡ ÛÒÎ (Esth 3:9). A similar OV word order is employed in ÂÏ˙˜ „Â‰È Ï‡ Â„Ú ‡˙„Ó Ï‡ Ô·‡ ÔÓ‰ „‡ÏÂ‡ ˙¯˘Ú (1302 7a,1) ‘The ten sons of Haman, son of Madatha, enemy of the Jews, they killed,’ following the Hebrew in Ô· ÔÓ‰ È· ˙¯◊Ú Â‚¯‰ ÌÈ„Â‰È‰ ¯¯›ˆ ‡˙„Ó‰ (Esth 9:10).3 (2-1.6) In the Haggadah: The Hebrew word order OVS in ¯‚ ÍÚ¯Ê ‰È‰È ‘your seed will be a stranger’ (although ¯‚ is technically a predicate) is translated verbatim in the ¡ar˙: ¬ÏÒ ÔÂÎÈ ·È¯µ (3 6,16) ‘your seed will be a stranger,’ in a similar unusual OVS word order. Similarly, the Hebrew OV word order in ÁÓˆÎ ‰··¯ ÍÈ˙˙ ‰„˘‰ is transferred intact into the Judeo-Arabic with the same OV word order ¬˙ÏÚ‚ ËÈµ Ï‡ ˙‡·Î ‰ÂÂ·¯ (3 8,12–13) ‘I have made you multiply ten thousand like the vegetation of the field.’ In the same way, the word order in the Hebrew Â˜Ï ÌÈ‰ ÏÚÂ ˙ÂÎÓ ÌÈ˘œÓÁ, where the locative precedes the verb for emphasis, is strictly imitated in the Judeo-Arabic ¡ar˙ Â·¯ƒ‡ ¯Á· Ï‡ ‰ÏÚÂ ‰·¯ƒ ÔÈÒÓ¡ (3 14,7–8) ‘and over the sea they (the Egyptians) were smitten with fifty plagues.’ The avoidance of syntactic adaptation of the Judeo-Arabic structure may sometimes cause slight misunderstandings of the text, as in the following example: (2-1.7) The phrase ÂÏ ˙„ÏÂ ‘bore to him’ in ¯ÂÁ Ô·‡ Ï‡Â˙· ˙· ‰ÎÏÓ ÂÏ ˙„ÏÂ È≤Ï‡ (15 4a,1) ‘the daughter of Bethuel, the son of 2

3

A miskal is a measure of weight; in Egypt one miskal is equivalent to 4.68 grams. ‡˙„Ó Ï‡ can be considered a calque translation, as it translates the Hebrew name ‡˙„Ó‰ ‘Hammedatha.’ The ¡ar˙an has mistakenly interpreted the Hebrew name as having two morphemes: the definite article /ha-/ and the name. See chapter 6, p. 181, and Hary 2009, chapter 3, “Mistakes and Omissions in the Texts.”

191

CHAPTER SEVEN Nahor, whom Milcah (his wife) bore to him’ is an exact rendition of the Hebrew ÂÏ ‰„ÏÈ (Gen 24:47) without any syntactic adaptation. Saadia (‰Ï ‰˙„ÏÂ) and the Protestant translation (t?? ?! t?? ðb?? ?!Ë) employed syntactic adaptation by adding the ‘returning’ pronoun (/al->å
THE PHRASE AND THE WORD LEVELS: ADVERBS exhibiting a strong literal translation tendency. (2-2.1) In many instances the place of the adverb Ô‡ÓÎ /kamån/ ‘also’ in the Judeo-Arabic phrase follows verbatim the place of Ì‚ in the biblical Hebrew phrase. For example, Ô‡ÓÎ ˙Ï‡˜Â È˜Ò‡ ¬Ï‡Ó‚Ï (15 3a,1) ‘And she said, “I will also water your camels” ’ translates verbatim ·‡˘‡ ÍÈÏÓ‚Ï Ì‚ ¯Ó‡˙Â (Gen 24:19); ˙‡·Ï Á¯ËÓ Ô‡ÓÎÂ ‰ÚÓ ¯È˙Î ¥ÏÚ Ô‡ÓÎÂ Ô·˙ Ô‡ÓÎ ‰ÈÏ‡ ˙Ï‡˜Â (15 3a,8–9) ‘and she said to him, “There is a lot of straw as well as forage in our place, and also a place to spend the night” ’ translates verbatim ÔÂÏÏ ÌÂ˜Ó Ì‚ ÂÓÚ ·¯ ‡ÂÙÒÓ Ì‚ Ô·˙ Ì‚ ÂÈÏ‡ ¯Ó‡˙Â (Gen 24:25); and ‰ÈÏÚ Ô‡ÓÎ ÂÓˆ‡¡˙‡Â (15 6b,16) ‘and they fought also over it’ translates verbatim ‰ÈÏÚ Ì‚ Â·È¯ÈÂ (Gen 26:21). The same occurs in the ¡ar˙ of the book of Esther. In È‚˙ ÌÏ ¬ÏÓ Ï‡ ˙È· ‡Ï‡ Ô‡ÓÎ (1302 2b,13) ‘she would not go again to the house of the king’ and in Ô‡ÓÎ ¬˙·ÏË ˘È‡Â (1302 7a,4) ‘and what is your further request?,’ the place of /kamån/ in the phrase is the same as the place of „ÂÚ in the original Hebrew in Esth 2:14 and 9:12 respectively. Furthermore, the Haggadah displays the same feature. In the following examples, the place of Ô‡ÓÎ /kamån/ or ‰‡ÓÎ /kamåna/ follows verbatim the place of the Hebrew word Û‡ or Ì‚: Ô‡ÓÎ Â‡Ò ‰Ï‡ Ò¯„ ‰˙‡ (3 4,20) ‘you also, extinguish his teeth’; ÂÓ„¡‡ È≤Ï‡ ·Ú˘ Ï‡ ‡Ï‡ ‰‡ÓÎÂ (3 6,18) ‘and also that nation whom they worshiped’; and ‰È‰¯Î ‰ÏÚ ‡Â‰ Ô‡ÓÎ „‡ÊÈÂ (3 9,4–5) ‘may also join our enemies.’ (2-2.2) Similar is the word order of ÔƒÈÈ‡ /ay∂an/ ‘also’ in the various ¡ur¥˙. In Classical Arabic /ay∂an/ is generally located after the phrase to which it refers,4 but in the Egyptian ¡ur¥˙ it usually follows verbatim the place of its biblical equivalent. Thus, in the book of Esther the place of /ay∂an/ in Ï‡ È˙˘Â ÔƒÈÈ‡

4

See Reckendorf 1895–98:110–11. In colloquial Egyptian Arabic the use of the adverb /ay∂an/ is limited (Hinds and Badawi 1986:46).

193

CHAPTER SEVEN ‰ÓÂÊÚ ˙Úˆ ‰ÎÏÓ (1302 9b,13) ‘Queen Vashti also threw a banquet’ is similar to the place of its Hebrew equivalent Ì‚ in È˙˘Â Ì‚ ‰˙˘Ó ‰˙˘Ú ‰ÎÏÓ‰ (Esth 1:9) before and not after È˙˘Â, which it modifies. Furthermore, in „Â‰È ÏÏ ¯Î‡· ÔƒÈÈ‡ ‡ËÚÈ (1302 7a,5) ‘let tomorrow also be given to the Jews,’ /ay∂an/ precedes /båkir/ ‘tomorrow’ as is the case in ÌÈ„Â‰ÈÏ ¯ÁÓ Ì‚ Ô˙È (Esth 9:13). The same occurs in È¥ ÔƒÈÈ‡ Ô˘Â˘ È¥ È≤Ï‡ „Â‰È Ï‡ Â˜ÂÂ‚˙‡Â ¯„‡ ¯‰˘Ï ±¥ Ï‡ ÌÂÈ (1302 7a,8) ‘Then the Jews in Sushan were assembled on the fourteenth day of Adar as well,’ following strictly the placement of Ì‚ in Esth 9:15. (2-2.3) The adverb Ë˜¥ /faqa†/ ‘only’ in regular Arabic use usually comes after the phrase to which it relates; however, in ‰Ï‡ Ë˜¥ Ú‚¯˙ ÌÏ È·‡ (15 2b,4) ‘Only, do not take my son back’ the adverb /faqa†/ comes before the phrase it modifies in a clear word-for-word translation of ·˘˙ ‡Ï È· ˙‡ ˜¯ (Gen 24:8). The same holds true for the placement of /faqa†/ in ¬ÚÓ ‰Úˆ È≤Ï‡ÎÂ ·ÈÈË Ë˜¥ (15 7a,9–10) ‘just as we have done only good with you,’ following directly its position in the Hebrew ÂÈ˘Ú ¯˘‡ÎÂ ·ÂË ˜¯ ÍÓÚ (Gen 26:29). In the same way, the ¡ar˙an has placed /faqa†/ in ¬Ó Ì∂Ú‡ ÈÒ¯ÂÎ Ï‡ Ë˜¥ (15 26b,4) ‘only in the throne shall I be above you’ before the phrase it modifies, as in the Hebrew ÍÓÓ Ï„‚‡ ‡ÒÎ‰ ˜¯ (Gen 41:40). Furthermore, /faqa†/ more frequently translates in Genesis Hebrew Í‡ ‘only,’ and in this case too it appears before the phrase it modifies in a verbatim translation of the Hebrew. Thus, the place of /faqa†/ in Ï‡˜Â ‡È‰ È˙¡Â‡ ˙ÏÂ˜ Ë˜¥Â ‡È‰ ¬˙‡¯Ó ‰„ÂÂ‰ Ë˜¥ (15 6a,18–19) ‘and he said, “Indeed, she is your wife!” How (lit. only) can you say, “She is my sister?” ’ follows verbatim its place in the Hebrew biblical verse ‡È‰ È˙ÂÁ‡ ˙¯Ó‡ ÍÈ‡Â ‡È‰ Í˙˘‡ ‰‰ Í‡ ¯Ó‡ÈÂ (Gen 26:9). 5

5

The second /faqa†/ in this example actually translates Hebrew ÍÈ‡ ‘how’; apparently the scribe confused ÍÈ‡ and Í‡. See Hary 2009, chapter 3, “Mistakes and Omissions in the Texts.”

194

THE PHRASE AND THE WORD LEVELS: ADVERBS Similarly, in È˙Âˆ È¥ ÚÓÒ‡ Ë˜¥ È·‡ ‡È (15 7b,16–17) ‘My son, only heed my voice,’ the ¡ar˙an has placed /faqa†/ before the phrase it modifies, following the Hebrew ÈÏÂ˜· ÚÓ˘ Í‡ È· (Gen 27:13). Along the same lines /faqa†/ appears in ‚¯¡ ‚Â¯¡ Ë˜¥ Ô‡ÎÂ ˜ÁˆÈ ‰È‚Â „Ú ÔÓ ·Â˜ÚÈ (15 7b,20–8a,1) ‘and it was when Jacob had scarcely left from the presence of Isaac,’ translating verbatim ˜ÁˆÈ ÈÙ ˙‡Ó ·˜ÚÈ ‡ˆÈ ‡›ˆÈ Í‡ È‰ÈÂ (Gen 27:30). (2-2.4) Along the same lines, the ¡ar˙an translates the Hebrew modal interjection (or vocative) ‡ using the Judeo-Arabic adverb Ô‡Ï‡ /al<ån/ ‘now,’ placing it in the same position in the sentence that ‡ occupies in the Hebrew original. Thus, in Genesis, /al<ån/ follows the imperative verb in ÈÎ¯ÂÂ ˙Á˙ ¬„È Ô‡Ï‡ ÏÚ‚‡ (15 2a,14) ‘place now your hand under my thigh,’ just as ‡ does in the Hebrew ÈÎ¯È ˙Á˙ Í„È ‡ ÌÈ˘ (Gen 24:2). The same applies to ¥„‡ˆ ÌÂÈ Ï‡ ÈÓ‡„Â˜Ï Ô‡Ï‡ (15 2b,10) ‘(you) make it happen now in front of me today,’ following ÌÂÈ‰ ÈÙÏ ‡ ‰¯˜‰ (Gen 24:12), and to ¬˙¯‚ ÔÓ ‰ÈÓ ÏÈÏ˜ ÔÓ Ô‡Ï‡ ÈÈË¥˘ (15 2b,19) ‘let me drink now a little water from your jug,’ following verbatim ËÚÓ ‡ ÈÈ‡ÈÓ‚‰ Í„ÎÓ ÌÈÓ (Gen 24:17). The ¡ar˙an was so strict about translating every Hebrew ‡ with Judeo-Arabic /al<ån/ that he mistakenly parsed the verb ‰ËÁ‡ ‘I would bear the loss’ (Gen 31:39), translating it as Ô‡Ï‡ ˙ÈË¡‡ (15 14b,7–8). In Esther the Hebrew interjection ‡ does not appear, but in the Haggadah, occurrences similar to those in Genesis appear. The position of Ô‡ Ï‡ /al<ån/ ‘now’ in √¯‡ È¥ ¬„È·Ú Ô‡ Ï‡ Â„ÂÚ˜È Ô˘Â‚ (3 7,20) ‘let your servants dwell now in the land of Goshen’ follows verbatim the position in its Hebrew equivalent ‡ Â·˘È Ô˘Â‚ ı¯‡· ÍÈ„·Ú. In the same way, the position of /al<ån/ in Ï‡ ÔÈ¥ ÌÂ‰‰‡Ï‡ Ô‡ (3 23,11) ‘where is their God now?’ follows verbatim the position of its Hebrew equivalent in Ì‰È‰¿‡ ‡ ‰È‡. Likewise in ÂÓÂ˜ ÏÂÎÏ Ô‡ Ï‡ „‡ˆÂ˜ ‰¥ÂÂ‡ ‰ÏÏ‡Ï È¯Â„ (3 25,3–4) ‘My vows to God I will pay now in the presence of all his people,’ following literally the Hebrew ÂÓÚ ÏÎÏ ‡ ‰„‚ ÌÏ˘‡ ß‰Ï È¯„. 195

CHAPTER SEVEN (2-2.5) Whereas the Judeo-Arabic adverb /al<ån/ is reserved by the ¡ar˙anim for the translation of Hebrew ‡ in both Genesis and the Passover Haggadah, Egyptian Judeo-Arabic /dilw
6

/dilwa
196

THE PHRASE AND THE WORD LEVELS: ADVERBS (2-2.6) The same applies to the sentence position of the adverb ÈÂÂ˜ ‘very’ in the ¡ar˙, as it follows verbatim the position of „›‡Ó in the Hebrew text. In Genesis, for example, Ï‡ ˙ÒÂÁ ‰È·ˆ Ï‡Â ÈÂÂ˜ ¯∂Ó (15 2b,17) ‘and the girl is very good looking’ translates verbatim „›‡Ó ‰‡¯Ó ˙·›Ë ‰¯ÚÓ‰Â (Gen 24:16); È„ÈÒ ‰Ï‡ ¬¯‡· ‰ÏÏ‡Â ÈÂÂ˜ (15 3b,2) ‘and God has greatly blessed my master’ translates literally „›‡Ó ÈI‡ ˙‡ Í¯· ß‰Â (Gen 24:35); and ÈÂÂ˜ ¯Â·ÂÎ Ô‡ „Ú (15 6b,5) ‘until he became very great’ translates verbatim Ï„‚ ÈÎ „Ú „›‡Ó (Gen 26:13). The same occurs in Esther, where Hebrew „›‡Ó is translated by Egyptian Judeo-Arabic /giddan/. In both cases where the adverb appears, its position in the sentence follows the Hebrew equivalent: Ô„‚ ¬ÏÓ Ï‡ Ë¡ÒÂ (1302 1b,17–18) ‘then the king became very angry,’ translating „›‡Ó ÍÏÓ‰ Ûˆ˜ÈÂ (Esth 1:12) and Ô„‚ ‰ÎÏÓ Ï‡ ˙˘Ú˙¯‡Â (1302 3b,14–15) ‘and the queen was greatly distressed,’ translating „›‡Ó ‰ÎÏÓ‰ ÏÁÏÁ˙˙Â (Esth 4:4). In the Haggadah ÈÂÂ˜ ÈÂÂ˜ Ï‡· ÂÓ∂ÂÚÂ (93 22,10–11) ‘and they increased a lot’ follows literally the Hebrew „‡Ó „‡Ó· ÂÓˆÚÈÂ (see p. 230, 5-1.37). Similarly, in ÈÂÂ˜ ÏÈ˜˙ ‰¥ (3 12,7) ‘very heavy destruction’ the ¡ar˙an has placed /
CHAPTER SEVEN In Esther: ¥˜‡Â ÔÓ‰ ‡„ÂÂ‰ (1302 4b,20) ‘Here Haman is standing’ translates „Ó›Ú ÔÓ‰ ‰‰ (Esth 6:5); ‰·˘¡ Ï‡ ‡„ÂÂ‰ ÔƒÈÈ‡ (1302 5b,10) ‘indeed also the tree’ translates ıÚ‰ ‰‰ Ì‚ (Esth 7:9); and ˙ÈËÚ ÔÓ‰ ˙È· ‡„ÂÂ‰ (1302 6a,4) ‘I gave indeed Haman’s house’ translates È˙˙ ÔÓ‰ ˙È· ‰‰ (Esth 8:7). In the Haggadah: in both of the following examples /hawda/ appears at the same position as its Hebrew equivalent. ‰„ ‰„ÂÂ‰ ¯¡˙¥ÂÓ (3 3,7–8) ‘Indeed, this is praiseworthy’ translates ‰Ê È¯‰ Á·%˘Ó; and ‰Ò ÔÈÚ·Ò Ô·‡Î ‰‡ ‰„ÂÂ‰ (3 3,18–19) ‘Indeed, I am about seventy years old’ translates ‰˘ ÌÈÚ·˘ Ô·Î È‡ È¯‰. Int.: (2-2.8) The adverb and vocative ‰„ÂÂ‰ ‘behold, indeed, here’ can, at times, be positioned at a different place in the sentence than its Hebrew equivalent in an interpretive translation: ‰˜·¯Â ‰‚¯‡¡ ‰„ÂÂ‰ (15 2b,15) ‘and indeed Rebecca is leaving’ translates ˙‡ˆ$È ‰˜·¯ ‰‰Â (Gen 24:15). L/I: (2-2.9) In the Haggadah the adverb Ë˜¥ /faqa†/ translates Hebrew „·Ï· ‘only,’ following verbatim the same word order as its Hebrew equivalent, but at the same time it also follows the item it modifies in an interpretive mode, unlike the examples above in 2-2.3, where /faqa†/ follows verbatim the Hebrew, but does not succeed the item it modifies as required by the Arabic structure: ‰Â¥ÈÏ ‰ÈÏÚ ¥˜Â Ë˜¥ „Á‡Â ÌÏ Ô‡ (3 7,2–3) ‘for not one only has risen up to annihilate us’ translates ÂÈ˙ÂÏÎÏ ÂÈÏÚ „ÓÚ „·Ï· „Á‡ ‡Ï˘; and ‡Â‰ ¬¯‡·Â˙ ÒÂ„Â˜ Ï‡ ¬¥ Ë˜¥ ‰˙‡‰·‡Ï ÌÏ È≤Ï‡ (3 19,9–10) ‘it was not only our ancestors that the Holy One, blessed be He, redeemed’ translates verbatim the Hebrew ÂÈ˙Â·‡ ˙‡ ‡Ï˘ ‡Â‰ ÍÂ¯· ˘Â„˜‰ Ï‡‚ „·Ï·. Word Order: Numerals Feature 2-3 — Numerals Int.: It is quite common in the ¡ar˙ to find the regular Judeo-Arabic word order for the numerals (hundreds-units-tens), which is not 198

THE PHRASE AND THE WORD LEVELS: NUMERALS in line with Hebrew word order. This is a typical interpretive translation that favors Judeo-Arabic structure over a literal translation (the following examples are just a small sample): (2-3.1) In Genesis: The phrase ÔÈÒ ÔÈ˙Ï˙Â ‰È‡ÓÓ˙ (15 0-1,1) ‘eight hundred and thirty years’ follows regular Judeo-Arabic word order, and thus translates interpretively the different word order of the Hebrew ‰˘ ˙Â‡Ó ‰Ó˘Â ‰˘ ÌÈ˘Ï˘ ‘thirty years and eight hundred years’ (Gen 5:16). Likewise, ‰ÒÓ¡Â ‰È‡ÓÓ˙ ‰Ò ÔÈÚÒ˙Â (15 0-1,2) ‘eight hundred and ninety-five years’ has regular Judeo-Arabic word order, despite the different Hebrew word order: ‰˘ ˙Â‡Ó ‰Ó˘Â ‰˘ ÌÈÚ˘˙Â ˘ÓÁ (Gen 5:17). In the same way, the word order of the numerals in ‰Ò ÔÈÓ˙Â ÔÈ˙‡Â ‰È‡Ó (15 0-1,12–13) ‘one hundred and eighty-two years’ is regular, and not the same as the Hebrew ‰˘ ˙‡ÓÂ ‰˘ ÌÈÓ˘Â ÌÈ˙˘ (Gen 5:28), indicating an interpretive translation. (2-3.2) The same occurs in the ¡ar˙ of Esther: ˙Ú·ÒÂ ˙È‡Ó ‰È„Ó ÔÈ¯˘ÚÂ (1302 1b,3; 6a,9) ‘one hundred and twenty-seven provinces’ observes regular Judeo-Arabic word order, not following the Hebrew word order of ‰È„Ó ‰‡ÓÂ ÌÈ¯˘ÚÂ Ú·˘ (Esth 1:1; 8:9) in an interpretive translation. Furthermore, the ¡ar˙an in Esther has used Arabic numerals in his translation: ¯‰˘ Ï‡ È¥ ‰È¥ ≤≥ È¥ ÔÂÈÒ ¯‰˘ ‡ÂÂ‰ ˙Ï‡˙ Ï‡ (1302 6a,7) ‘in the third month, i.e., the month of Sivan, on the twenty-third.’ (2-3.3) In the Haggadah as well, the ¡ar˙an has translated the Hebrew word order of the numerals interpretively: ÔÈÒÓ¡Â ÔÈ˙ÈÓ ‰·¯ƒ (3 15,17) ‘two hundred and fifty plagues’ does not follow the Hebrew word order ˙ÂÎÓ ÌÈ˙‡ÓÂ ÌÈyÓÁ, but is an interpretive translation. L/I.: As seen above, the ¡ar˙an has usually retained regular JudeoArabic word order in the numerals, but in the following examples, the numerals’ word order is the same as in the Hebrew; it may appear to be a literal translation, but it actually follows regular

199

CHAPTER SEVEN Judeo-Arabic word order at the same time. (2-3.4) In Genesis: ‰Ò ÔÈ˙ÒÂ ‰ÒÓ¡ (15 0-1,5) ‘sixty-five years’ is in regular Judeo-Arabic word order, which also follows the Hebrew word order: ‰˘ ÌÈ˘˘Â ˘ÓÁ (Gen 5:21). Similarly, ˙Ï˙ ‰Ò ‰È‡Ó (15 0-1,7) ‘three hundred years’ follows the Hebrew word order of ‰˘ ˙Â‡Ó ˘Ï˘ (Gen 5:22), but is also in regular Arabic structure. Lexicon: Semantic Considerations Level: Word; Category 3: Lexicon Feature 3-1 — Word or root choice (semantic considerations) Lit.: (3-1.1) There are a few cases where the ¡ar˙an has translated the meanings of Hebrew place and personal names into JudeoArabic, and not just phonetically transcribed the name. Thus, the translation of the name of the cave ‰ÏÙÎÓ‰ ‘the Machpelah’ into Judeo-Arabic ‰È˙Ó Ï‡ (15 2a,9), literally ‘the doubled,’ is a clear literal choice by the ¡ar˙an. He translated verbatim the Hebrew root k-f-l ‘double’ into Judeo-Arabic t-n-y ‘double’ and also copied the Hebrew locative initial mem into Judeo-Arabic locative m•m, as well as the pattern ma-R1 R2 R3 -a.7 Of course, had the ¡ar˙an copied the Hebrew Machpelah into Judeo-Arabic with the actual phonetics of the Hebrew name, as he did in many other place names as well as personal names (see below, 3-2.1), we would have had an even more literal translation. Similarly, the Hebrew place name È‡] ÈÁÏ ¯‡· (Gen 24:62) is translated verbatim semantically into ¯∂‡ ÈÈÁ Ï‡ ¯È· (15 4b,3) ‘the well of the living looker.’8 The same occurs in Esther with the translation of the personal Hebrew name ‰Ò„‰ (Esth 2:7) ‘myrtle’ into Judeo-Arabic ‡ÈÒ¯Ó (1302 2b,1) ‘myrtle.’ Along

7 8

See also the section on calque translation in chapter 6, pp. 178–81. Although notice that the ¡ar˙an translated ÈÁÃÏ as ÈÈÁ Ï‡ and not the more literal ÈÈÁ ÏÏ.

200

THE PHRASE AND THE WORD LEVELS: SEMANTIC CONSIDERATIONS the same lines, in the Haggadah the ¡ar˙an translated ˜¯· È· into ˜¯· ¯„‡· (3 3,12–13) ‘the (large) towns (districts) of Braq.’9 Other translations of words according to their literal meaning are: (3-1.2) The ¡ar˙an translated literally the verb Ú„È ‘know’ in the biblical (sexual) sense into Judeo-Arabic />araf/: ÌÏ Ï‚‡¯Â ‡‰¥¯Ú (15 2b,17) ‘and no man had known her.’ (3-1.3) It seems that when the ¡ar˙an translated the Hebrew ·ÂËÏ … ‰&p˘ÈÂ ‘treated her … with kindness’1 0 (Esth 2:9) into ‡‰¯ÈÈµ (1302 2b,6) ‘change her,’ he preferred a verbatim translation, rather than interpretive. (3-1.4) The translation of the Hebrew Â˙›‡¯‰· ‘when he displayed’ (Esth 1:4) into Â¯Â∂ È¥ (1302 1b,5) has probably gone through a few stages. First, the ¡ar˙an wished to translate verbatim Â˙›‡¯‰·, using the Judeo-Arabic root n-Ω-r, equivalent to the Hebrew root ‰‡¯. However, since the Hebrew verb is in the hif>il form, the ¡ar˙an needed to use the IV form in Judeo-Arabic. Since verbal form IV has quite frequently become form I in the colloquial and in Later Judeo-Arabic,1 1 the ¡ar˙an used verbal form I here as well, with the meaning of verbal form IV ‘display’ and not the meaning of form I ‘see.’ It is also possible, of course, that the ¡ar˙an simply mistranslated Hebrew ‰‡¯‰ ‘display’ into Judeo-Arabic /naΩar/ ‘see’ in form I. (3-1.5) Sometimes, the ¡ar˙an was so eager to translate the Hebrew text verbatim that he even ignored the actual meaning of the lexeme. Consequently, the text contains many calque 9

The ¡ar˙an took the Hebrew È· to mean ‘large towns’ and connected it to the root b-n-y ‘build.’ Cf. the use of ‰È˙$· to mean ‘its villages’ in Num 21:25. 10 This is one possible translation (Tanakh 1988); there are other possibilities. 11 For Later Judeo-Arabic, see Hary 1992:280, 2.5.2; and Davies 1981:117–18. For the colloquial, see Mitchell 1978:72–74. Even in Classical Judeo-Arabic we find this phenomenon, although it is limited (Blau 1980:73–74).

201

CHAPTER SEVEN translations (chapter 6, pp. 178–81), which are typical of translations of sacred texts in various Jewish varieties. Thus, in the Haggadah, the ¡ar˙an translated Hebrew „‡Ó È˙ÈÚ È‡ ‘I am greatly afflicted’ literally as ÈÂÂ˜ ˙·ÂÂ‡‚ ‡‡ (3 24,20) ‘I answered greatly.’ The Hebrew verb È˙ÈÚ stems from the root >-n-h, which can mean both ‘poor’ and ‘answer.’ The ¡ar˙an chose the second meaning and arrived at the bizarre and unclear calque translation in the Judeo-Arabic text. Int.: (3-1.6) Sometimes the ¡ar˙an translated names of places interpretively, either by translating them according to meaning or by using actual Arabic place names. Thus, the biblical Hebrew place name ‡¯ÓÓ ‘Mamre’ (Gen 23:17) was translated interpretively into Judeo-Arabic ‚¯Ó (15 2ba,6) ‘meadow,’ probably the ¡ar˙an’s attempt to describe the place with the field and the vegetation surrounding it. Furthermore, the ¡ar˙an translated the place names Ì˙Ù and ÒÒÓÚ¯ in the Haggadah into the Egyptian place names ÌÂÈ¥ Ï‡ and ‰ÒÓ‰· Ï‡ (3 9,9-10) respectively. (3-1.7) The word È˙ÈÈ¯„Â‚ (15 2b,1) ‘my home (root) country’ is an interpretive choice of the ¡ar˙an translating Hebrew È˙„ÏÂÓ (Gen 24:7) ‘my native land.’ The ¡ar˙an used the Judeo-Arabic root g-z-r (‘root, source’) and not the root w-l-d employed by both Saadia (È„ÏÂÓ, Derenbourgh 1893:34) and the Protestant translation (/m•låd•/, Al-Kitåb al-Muqaddas 2004:29). (3-1.8) The choice of the verb È‚˙ (15 2b,3) ‘come’ is a possible interpretation of the Hebrew ˙ÎÏÏ … ‰·‡˙ ‘want’ (Gen 24:8). In other words, the ¡ar˙an collapsed the two verbs into one, È‚˙, although later in the sentence he used ÁÂ¯˙Ï ‘to go.’ Saadia and the Protestant translation did not do that; the former used in his translation È∏ÓÏ‡ … ˜Ù‡Â˙ ‘agree to come,’ while the Protestant translation employed ¡UAð ‘want.’1 2 12

It is also possible that the ¡ar˙an misunderstood the Hebrew verb ‰·‡˙ ‘consent’ and confused it with the root ‡Â· ‘come.’ See Hary 2009, “General

202

THE PHRASE AND THE WORD LEVELS: SEMANTIC CONSIDERATIONS (3-1.9) The ¡ar˙an translated ‰ÓÎ˘ ‘her shoulder’ (Gen 24:15) into Judeo-Arabic ‡‰Ò‡¯ (15 2b,16) ‘her head’ interpretively. This translation can be seen as “cultural,” since it is common to see women in Egypt carring their jars of water over their heads, rather than on their shoulders. Interestingly enough, both Saadia’s translation and the Protestant translation exhibit the more literal translation /katf/ ‘shoulder.’ (3-1.10) In the sentence ‰ÚÈ‡Ë ¯˙Ò‡ È¡„¯Ó Ì‡ÏÎ ‡Ï‡Â (1302 2b,22) ‘and Esther obeys Mordechai’s words,’ the ¡ar˙an clearly interpreted Hebrew ‰◊›Ú ‘do’ in ‰◊›Ú ¯˙Ò‡ ÈÎ„¯Ó ¯Ó‡Ó ˙‡Â (Esth 2:20) as ‰ÚÈ‡Ë ‘obey.’ (3-1.11) The ¡ar˙an interpreted Hebrew „È ‘hand’ metaphorically as ‰¯„˜Ø‰¯„Â˜ ‘might.’ Thus, ¬ÏÓ Ï‡ ˙¯„˜Î (1302 1b,11) ‘with the king’s might’ appears in Esther, translating the Hebrew original ÍÏÓ‰ „ÈÎ ‘in accordance with the king’s wealth; in abundance’ (Esth 1:7). In the Haggadah as well, we find Ï‡ ‰ÓÈ∂Ú Ï‡ ‰¯„Â˜ (3 14,11) ‘the great strength,’ translating the Hebrew ‰ÏÂ„‚‰ „È‰. Furthermore, Judeo-Arabic ·¯ Ï‡ ˙¯„Â˜ (3 14,9) ‘God’s might’ translates ÌÈ‰¿‡ Ú·ˆ‡; and ‰ÏÏ‡ ˙¯„Â˜ (3 26,13) ‘God’s might’ translates ÈÈ ÔÈÓÈ ‘God’s right hand.’1 3 (3-1.12) ˙¯ƒÁ È¥ ‘in the presence of’ is used in the texts of the ¡ur¥˙ in some sort of interpretive mode. The idiom is employed in the translation of several Hebrew phrases. For example, È¥ ˙¯ƒÁ (15 2a,4; 30b,19; 37b,13) translates ÈÊ‡· ‘in the ears of’ (Gen 23:16; 44:18; 50:4) and is clearly a metaphorical interpretation of the Hebrew. The same applies to ¬ÏÓ Ï‡ ˙¯ƒÁ È¥ (1302 2a,4; 3a,17; 5a,20; 5b,19; 7a,5), which translates ÍÏÓ‰ ÏÚ (Esth 1:19; 3:9; 5:4, 8; 9:13). Similarly, the use of ˙¯ƒÁÏ (15 2a,7) ‘in the presence of,’

13

Comments on the Critical Editions.” Note that the last two examples do not refer to ‘hand,’ but to ‘finger’ and ‘right hand’ respectively.

203

CHAPTER SEVEN which translates ÈÈÚÏ ‘in the eyes of’ (Gen 23:18); ¬˙¯ƒÁÏ (15 34b,15) ‘in your presence,’ translating ÍÈÈÚÏ ‘in your eyes’ (Gen 47:19); ÌÂ‰˙¯ƒÁÏ (15 28a,7) ‘in their presence,’ which translates Ì‰ÈÈÚÏ ‘in their eyes’ (Gen 42:24); ÌÂÎ˙¯ƒÁÏ (3 12,18) ‘in your presence,’ translating ÍÈÈÚÏ ‘in your eyes’;1 4 ¬˙¯ƒÁ ‰ÏÚ (15 32b,14) ‘in your presence,’ rendering ÍÈÈÚ ÏÚ ‘in your eyes’ (Gen 46:4); and the common use of ˙¯ƒÁ È¥ (over twenty-five occurrences in Genesis, ten occurrences in Esther, and at least one occurrence in the Haggadah) to translate ÈÈÚ·1 5 are also interpretive, but less so than the examples in the preceding paragraph. The same applies to ¬˙¯ƒÁ ‰ÏÚ (15 32b,14) ‘in your presence,’ which translates ÍÈÈÚ ÏÚ ‘on your eyes’ (Gen 46:4). Moreover, ˙¯ƒÁ È¥ (1302 4b,13; 5b,20), translating ÈÙÏ (Esth 5:14; 8:5) and Â˙¯ƒÁ È¥ (1302 5b,20), translating ÂÈÙÏ (Esth 8:5), are even less interpretive.1 6 (3-1.13) The choice of /¡a††/ ‘shore’ in ¯‰ Ï‡ Ë˘ È¥ (3 6,1) ‘at the shore of the river’ to translate the Hebrew ¯‰‰ ¯·Ú· ‘beyond the river’ is an interpretive choice. In another place, when the Hebrew text appears with the preposition ‘from,’ as in ¯·ÚÓ ¯‰‰, the ¡ar˙ exhibits the preposition /min/: ¯‰ Ï‡ Ë˘ ÔÓ (3 6,5–6). Thus, the ¡ar˙an consistently translates ¯·Ú ‘beyond’ with Ë˘ ‘shore’ and not with any lexeme that expresses the notion of ‘beyond.’1 7 L/I: (3-1.14) The Hebrew word ‰¯Â˙ ‘Torah’ is translated in three different ways in the Haggadah: literally as ‰¯Â˙ (for example,

14

Notice the inconsistency in the translation of the pronominal suffix. This translation also occurs in combination with various pronominal suffixes. 16 For a further treatment of “more” or “less” interpretive (or literal) translation, see pp. 83–85 above, especially p. 85. 17 See BDB:719, ¯‰‰ ¯·Ú· ‘beyond the river,’ but also ¯·ÚÓ ‘to the other side of.’ 15

204

THE PHRASE AND THE WORD LEVELS: APPEARANCE CONSIDERATIONS in 74 22,13) and ‰¯ÂË (for example in 93 90,12), obviously taken from the Hebrew original. It is, however, also translated interpretively by ‰ÚÈ¯˘ (for example in 3 3,6), avoiding the phonetically similar words /tøra/ and /†øra/. (3-1.15) The choice of the word ¯‡˙ÂÎ /kutår/ (3 21,6) ‘many’ in the Haggadah to translate Hebrew ÌÈ¯Á‡ ‘other’ is an interpretive choice of the ¡ar˙an; however, in other manuscripts, the translation is verbatim: ÔÈ¯¡‡ (74 12,5) and ÔÈ¯¡Â‡ (91 9b,14; 93 45,5) ‘other (pl.).’ (3-1.16) In the Haggadah the ¡ar˙an chose to translate ÛÂÒ ÌÈ ‘the Red Sea’ as ÔÂ≤¯Â‡ Ï‡/ÔÂ„¯Â‡ Ï‡ ¯Á· (93 73,13; 73,15; 91 113a,2; 13a,4) ‘the Jordan River,’ thus providing his own interpretation, while in 74 16,17, the ¡ar˙an kept the original meaning: ÌÂÊÏÂ˜ Ï‡ ¯Á· ‘the Red Sea’ (Wehr 1994:920). Lexicon: Sound/Appearance Considerations Feature 3-2 — Word (or root) choice: considerations of sound/ appearance Lit.: This feature is a common phenomenon in Jewish translations of the Hebrew Bible.1 8 Saadia makes a special effort to use Arabic words that sound like the Hebrew originals.1 9 The same occurs in the Septuagint, which uses /drepanon/ (1 Sam 13:21) ‘sickle’ for the Hebrew Ô·¯„ /darban/ ‘goad,’ and /trof∑/ (Ps 111:5) ‘food’ for the Hebrew Û¯Ë /†eref/ ‘prey’ (Greenspahn 2002:46). (3-2.1) In several places in the ¡ur¥˙, one may find Hebrew biblical personal names copied unchanged into the Judeo-Arabic in clear literal mode. The following examples are just a sample. In Genesis: Ï‡ÏÏ‰Ó (15 0-1,1) ‘Mahalalel’; „¯È (15 0-1,2) ‘Jared’; ÍÂÁ (15 0-1,3) ‘Enoch’; ÁÏ˘Â˙Ó (15 0-1,6) ‘Metushelah’; ÍÓÏ

18 19

See above, pp. 83–85, and Greenspahn 2002:46. See Malter 1921:145 and Galliner 1903:11, quoted by Greenspahn 2002:46 n.16.

205

CHAPTER SEVEN (15 0-1,10) ‘Lemech’; ÁÂ (15 0-1,10) ‘Noah’; Ì˘ (15 0-1,17) ‘Shem’; ÌÁ (15 0-1,18) ‘Ham’; ˙ÙÈ (15 0-1,18) ‘Jephet’; ¯ÂÁ (15 2b,8; also in the Haggadah, 3 6,2) ‘Nahor’; ‰˜·¯ (15 2b,15) ‘Rebecca’; Ï‡Â˙· (15 2b,16) ‘Bethuel’; ‰ÎÏÓ (15 2b,16) ‘Milcah’; Ì‰¯·‡ (15 2b,16) ‘Abraham’; ‰¯ÂË˜ (15 4b,11) ‘Keturah’; ‡·˘ (15 4b,13) ‘Sheba’; ¯¥Ú (15 4b,15) ‘Epher’; Ú„È·‡ (15 4b,15) ‘Abida’; ‰Ú„Ï‡ (15 4b,15) ‘Eldaah’; and, of course, the more common names Ì‰¯·‡ (15 2a,1; also in the Haggadah, 3 6,2); ˜ÁˆÈ (15 2a,17; also in the Haggadah, 3 6,8); and ·Â˜ÚÈ (15 5b,11; also in the Haggadah, 3 6,8).2 0 In the book of Esther: È˙˘Â (1302 1b,13) ‘Vashti’; ÔÓÂ‰Ó (1302 1b,14) ‘Mehuman’; ‡˙Ê· (1302 1b,14) ‘Bizzeta’; ‡Â·¯Á (1302 1b,14) ‘Óarbona’; ¯˙Ê (1302 1b,15) ‘Zetar’; ÒÎ¯¡ (1302 1b,15) ‘Carcas’; È¡„¯Ó (1302 2a,17) ‘Mordechai’; ¯È‡È (1302 2a,17) ‘Yair’; ÈÚÓ˘ (1302 2a,17) ‘Shimei’; ¯˙Ò‡ (1302 2b,2) ‘Esther’; and more. In the Haggadah: ¯ÊÚÈÏ‡ (3 3,10) ‘Eliezer’; ÚÂ˘Â‰È (3 3,10) ‘Joshua’; ‡·È˜Ú (93 14,15) and ‰·È˜Ú (3 3,12) ‘Akiva’; ÔÂ¥¯Ë (3 3,12) ‘Tarfon’; Á¯˙ (3 6,2) ‘Terah’; ‰„Â‰È (3 14,2) ‘Judah’; ÈÒÂÈ ÈÏÈÏ‚ Ï‡ (3 14,5) ‘Jose, the Galilean’; Ï‡ÈÏÓ‚ (3 17,18) ‘Gamliel.’ (3-2.2) Similarly, place names are also sometimes copied into the ¡ar˙ verbatim. In Genesis: Ì¡˘ (15 22a,1) ‘Shechem’; ÔÂ¯·Á (15 22a,4) ‘Hebron’; and ·ÈÊ¡ (15 23a,5) ‘Chezib.’ In Esther: Ô˘Â˘ (1302 1b,4) ‘Shushan’; ÌÈÏ˘Â¯È (1302 2a,18) ‘Jerusalem’; and Ï·· (1302 2b,1) ‘Babylon.’ In the Haggadah: ¯ÈÚ˘ (3 6,9) ‘Seir’; ÔÚÎ (3 7,18) ‘Canaan’; Ô˘Â‚ (3 7,19) ‘Goshen’; and ÈÈÒ (3 16,20) ‘Sinai.’ In the following examples, although the ¡ar˙an has chosen personal and place names that are phonetically similar to the 20

This spelling (·Â˜ÚÈ) may also indicate the Judeo-Arabic pronunciation /ya>q¥b/. ·˜ÚÈ appears once in 15 10b,4.

206

THE PHRASE AND THE WORD LEVELS: APPEARANCE CONSIDERATIONS Hebrew names they translate, the Judeo-Arabic spelling indicates more clearly the Judeo-Arabic pronunciation. This is a slight shift along the continuum—away, to a small degree, from the literal translation end, but still far from the interpretive side. Thus, the addition of matres lectionis (vowel letters) may indicate the quality of the vowel or its quantity. (3-2.3) The marking of /e/: The Hebrew name ˙ÕÁ ‘Óet’ (possibly Hittites) (Gen 23:16) is translated literally into Judeo-Arabic ˙ÈÁ (15 2a,4). Here the ¡ar˙an has made sure that the short /e/ vowel in the Hebrew name ˙ÕÁ is expressed in the Judeo-Arabic translation by the yod and thus has not copied the Hebrew name blindly. (3-2.4) The marking of /u/: In the same way, the addition of the vav in ÍÂÁ (15 4b,15) ‘Óanoch,’ as opposed to the Hebrew Í$Á (Gen 25:4), emphasizes the /o/ (or, maybe /ø/) pronunciation in /˙anox/ (or /˙anøx/). (3-2.5) /a/ > /å/: The spelling of ÌÈÈ¯Â‰ Ì‡¯‡ (15 2b,7) ‘AramNaharaim’ may reflect a possible long /å/ in Ì‡¯‡ and a possible /o/ or /u/ in ÌÈÈ¯Â‰. Furthermore, although the name Ô·Ï (15 3a,13; 3 7,7) ‘Laban’ appears to be a copy of the Hebrew Ô·Ï, in the same verse in Genesis, Ô‡·Ï (15 3a,13) /labån/ appears, possibly indicating the lengthening of the second short vowel /a/ > [a:]. The same holds true for the following personal and place names with a possible Judeo-Arabic /å/ pronunciation: Ô‡¯ÓÊ (15 4b,12) ‘Zimran’ for Hebrew Ô¯ÓÊ (Gen 25:2); Ô‡˘˜È (15 4b,12) ‘Yokshan’ for Hebrew Ô˘˜È (Gen 25:2); Ô‡„Ó (15 4b,12) ‘Medan’ for Hebrew Ô„Ó (Gen 25:2); Ô‡È„Ó (15 4b,12) ‘Midian’ for Hebrew ÔÈ„Ó (Gen 25:2); ˜‡·˘È (15 4b,13) ‘Yishbak’ for Hebrew ˜·˘È (Gen 25:2); Ô‡„„ (15 4b,14) ‘Dedan’ for Hebrew Ô„„ (Gen 25:3); Ô‡ÚÎ (15 21b,1) ‘Canaan’ for Hebrew ÔÚÎ (Gen 37:1); Ô‡˙Â„ (15 22a,8) ‘Dotan’ for Hebrew Ô˙I (Gen 37:17);

207

CHAPTER SEVEN and „‡ÚÏ‚ (15 22b,1) ‘Gil>ad’ for Hebrew „ÚÏ‚ (Gen 37:25).2 1 (3-2.6) In the same way, the glide /w/ appears in the personal name ÁÂÂ˘ (15 4b,13) ‘Shuah,’ where the glide in /¡uwa˙/ is indicated by the two vavs. (3-2.7) Emphatization (tafx•m) also can be shown in this way. In the Haggadah the personal names ¯∂Ú Ï‡ /el>aΩar/ (93 15-b,10) ‘Elazar’ and ‰È¯∂Ú />aΩarya/ (93 15,9 and more) ‘Azarya’ demonstrate emphatization (see chapter 4, p. 108, 1.11.4 and chapter 5, pp. 149–50, 1.4.2). The following examples demonstrate the ¡ar˙an’s choice of Judeo-Arabic words that are phonetically similar to their Hebrew equivalen, indicating a clear verbatim choice of word or root: (3-2.8) In Genesis: The choice of the verb ¬ÿ¯·Â (15 2b,8) ‘caused to kneel down’ is preferred because of the phonetic proximity to the Hebrew verb ÍVŸ·'ÈÂ (Gen 24:11); the ¡ar˙an could have chosen a different verb, for example, ŒU?½« as in Saadia or the Protestant translation. Similarly, the ¡ar˙an chose the JudeoArabic ÔÈÚ /<∑n/ (15 2b,11 and 18) ‘spring’ to translate the Hebrew ÔÈÚ (Gen 24:13 and 24:16), probably because of sound proximity in a literal fashion. Furthermore, Ô·˙ /tibn/ (15 3a,3) ‘straw’ is phonetically close to the Hebrew Ô·˙ (Gen 24:25). Along the same lines, the ¡ar˙an chose the verbal noun ÏÎ‡Ó (15 24b,12) ‘eating’ in order to follow verbatim the Hebrew ÏÎ‡Ó (Gen 40:17), where he could have chosen the more ordinary /
21

See also chapter 4, p. 102, 1.1.5 and chapter 5, pp. 148–49, 1.1.

208

THE PHRASE AND THE WORD LEVELS: APPEARANCE CONSIDERATIONS The use of Hebrew words in the ¡ar˙ is not common, but it does occur as part of the literal translation: (3-2.10) Although the verb ÈÈÁ /˙ayy/ (15 4b,18) ‘live’ exists in regular Arabic use, its choice in this example is clearly a verbatim translation of the Hebrew ÈÁ (Gen 25:7). Both Saadia and the Protestant translation used the more common />å¡/ ‘live.’ (3-2.11) Judeo-Arabic lexemes may change their meanings in order to imitate their phonetically similar Hebrew counterparts. Thus, Judeo-Arabic ‰È„Ó /mad•na/ (1302 1b,3) changes its meaning in the ¡ar˙ from ‘city’ to ‘province,’ as it is a copy of the Hebrew ‰È„Ó ‘state, province’ (Esth 1:1) (Hary 1994a:32). Int.: Hebrew place names are sometimes translated interpretively by their Judeo-Arabic counterparts in the various ¡ur¥˙. (3-2.12) In Genesis the place name ·‚‰ ı¯‡ ‘the region of the Negev’ (Gen 24:62) is translated interpretively as ÈÏ·˜ Ï‡ √¯‡ ‘the land of the south’ (15 4b,3). (3-2.13) In Esther the Hebrew place name ˘ÂÎ ‘Cush,’ identified as Ethiopia (Esth 1:1), is interpreted in the ¡ar˙ as ˘·Á Ï‡ (1302 1b,2) ‘Ethiopia.’ Both Saadia and the Protestant translation have used the Hebrew word /k¥¡/. Similarly, the Hebrew place name Â„Â‰ ‘India’ (Esth 1:1) is translated in the ¡ar˙ as „‰ Ï‡ (1302 1b,2). Both Saadia and the Protestant translation also used /al-hind/. Likewise, the Hebrew place name Ò¯Ù (Esth 1:3, 14) is translated interpretively by either „‰ Ï‡ (1302 1b,5) ‘India’ or Ì‚Ú Ï‡ (1302 1b,20) ‘Persia.’ Both Saadia and the Protestant translation translated the place name as /fåris/, a more literal tendency. (3-2.14) In the Haggadah regular Judeo-Arabic interpretive translations appear in ¯ˆÓ (3 2,3 and elsewhere) ‘Egypt’; ÏÈ Ï‡ (3 10,19) ‘the Nile’; ÔÂ„¯Â‡ Ï‡ (3 20,17) ‘the Jordan’; and more. L/I: (3-2.15) The personal name Â◊Ú ‘Esau’ is translated with an interpretive orthography in Genesis as Â‡ÒÚ (15 5b,10); however,

209

CHAPTER SEVEN in the Haggadah the name is copied from Hebrew in a more literal translation: ÂÈ˘Ú (3 6,8) and Â˘Ú (91 3a,4). Similarly, the personal name ‰Ú¯Ù ‘Pharaoh’ is translated interpretively in Genesis and in the Haggadah into ÔÂÚ¯¥ (15 22b,18; 3 2,19); however, ‰Ú¯Ù (3 7,8), copied from the Hebrew, also occurs, but rarely. This is good evidence for a verbatim imitation of the Hebrew letters in Judeo-Arabic, as the letter fe without the supralinear dot hardly exists in the Egyptian ¡ur¥˙. In the same way, the personal name ˘Â¯Â˘Á‡ ‘Ahasuerus’ in the book of Esther is usually translated interpretively as È¯È˘„Ê‡ Ï‡; however, in 1302 1b,15 ˘Â¯Â˘Á‡ appears in the ¡ar˙ in a verbatim translation. Saadia used ˘¯Â˘Á‡ and the Protestant translation employed /
THE PHRASE AND THE WORD LEVELS: APPEARANCE CONSIDERATIONS girls’ (25:6) and in Esther (2:14) ¯‡Â‚ /≠uwårin/ ‘slave girls,’ whereas the Protestant translation was more consistent with /sarår•y/ ‘concubines.’ (3-2.19) ‡Ó (93 13,1) ‘what’ translates the Hebrew ‰Ó in a literal manner, whereas the choice of ˘‡ in another manuscript (3 2,8) is an Egyptian Judeo-Arabic colloquial interrogative pronoun more in the direction of the interpretive mode. (3-2.20) Judeo-Arabic ÏÂÎ (3 5,20) ‘every, all’ always translates Hebrew ÏÎ, maybe also because of sound and lexical proximity. However, when the Hebrew ÏÎ relates to time, as in ‰ÏÈÏ‰ Â˙Â‡ ÏÎ ‘all that night,’ the ¡ar˙an translated it in an interpretive mode, using the Judeo-Arabic /†¥l/ ‰ÏÈÏ Ï‡ ¬Ï˙ ÏÂË (3 3,13–14) ‘all that night.’ (3-2.21) ˜ÂÒÙ (3 3,21 and many other places) ‘verse’ is a Hebrew word used in the ¡ar˙ in a verbatim way. 2 2 However, this word does not appear in the Hebrew text (we find there ¯Ó‡˘ and ‰ÓÎ ¯Ó‡˘), and it is used in the ¡ar˙ in an interpretive translation (see p. 188, 1-1.7, above). (3-2.22) The translation ¯‰˘ Ï‡ Ò‡¯ (3 5,11) ‘the beginning of the month’ is the typical verbatim translation of the Hebrew ˘„›Á ˘‡¯; however, in another manuscript (93 14,4–5), Ï‡ ÏÂ‡ ¯‰˘ appears as a regular Arabic phrase in an interpretive translation. Furthermore, the ¡ar˙an has added the definite article in the Judeo-Arabic translation (¯‰˘ Ï‡) as needed, even when the Hebrew did not have it, in yet another indication of the interpretive mode (see other examples in chapter 8, p. 268, 9-1.6). (3-2.23) Whereas the Judeo-Arabic Egyptian dialect form Ï‡ ‰ÂÂ¯ˆÓ /il-maßarwa/ (3 8,17) ‘the Egyptians’ is employed in an interpretive translation, in another manuscript ÌÈÈ¯ˆÓ Ï‡ (91

22

Note the letter fe without the supralinear dot, which hardly appears in Egyptian ¡ur¥˙, another indication of the literal nature of the translation.

211

CHAPTER SEVEN 4b,17) ‘the Egyptians’ appears as a transcribed Hebrew word (ÌÈÈ¯ˆÓ) in a verbatim translation. (3-2.24) Furthermore, in the Haggadah, in one manuscript the Hebrew phrase is incorporated into the Judeo-Arabic verbatim as follows: Á·„Â ‡Â‰ ÍÂ¯· ˘Â„˜‰ ‰È‚Â (93 87,14) ‘and the Holy One, blessed be He, slew.’ In another manuscript, however, the Hebrew phrase is translated into Judeo-Arabic: ‡Â‰ ¬¯‡·Â˙ Ò„˜ÂÓ Ï‡ ‰È‚Â Á·„Â (74 25,14) ‘and the Holy One, blessed be He, slew.’ With regard to the literal/interpretive linguistic tension, some additional interesting examples appear in the choice of Hebrew words in the ¡ar˙ that are written and pronounced with Arabic phonological influence. (3-2.25) ˙È¯˜ (3 3,15) ‘reading’ is a copy of the Hebrew word ˙‡È¯˜, but with hamza deletion in the dialect, the word is written without the alif and its pronunciation probably approximates /qriat/. On the one hand, ˙È¯˜ is a literal translation because the ¡ar˙an simply copied the Hebrew word, but on the other hand, it is interpretive, because he used Judeo-Arabic orthography and phonology. In other manuscripts of the Haggadah the ¡ar˙an copied the Hebrew word letter for letter in a more verbatim manner: ˙‡È¯˜ (74 2,1; 93 15,3). (3-2.26) Similarly, the Hebrew prayer ÚÓ˘ ‘Hear!’ was copied into the ¡ar˙ in mss. 3 3,15 and 93 15,3 verbatim. However, in ms. 74 2,1 it is spelled under the influence of Judeo-Arabic phonology, where the vowel /a/ is lengthened: Ú‡Ó˘ /¡Emå>/.

212

CHAPTER EIGHT THE MORPHOSYNTACTIC LEVEL Like the previous and the following chapters, this chapter analyzes the ¡ur¥˙ of Genesis, Esther, and the Passover Haggadah, scanning, this time, selected examples through the morphosyntactic level. The analysis of the data is based upon the theoretical model advanced in chapter 3, including the use of table 4 (pp. 81–82). Negation: Nominal Level: Morphosyntactic; Category 4: Negation Feature 4-1 — Nominal Lit.: (4-1.1) The negation of the first plural independent pronoun /i˙na/ takes an interesting twist in ÔÈÒÓµÓ ‰Á‡ ÒÈÏ (3 2,10–11) ‘we do not dip.’ This sentence translates word for word the Hebrew equivalent ÔÈÏÈ·ËÓ Â‡ ÔÈ‡. In Arabic the phrase ‘we are not’ consists of the more synthetic /lasna/; however, in order to follow the more analytic Hebrew Â‡ ÔÈ‡, the ¡ar˙an created an exact Judeo-Arabic equivalent to the Hebrew. (4-1.2) Similarly, the negation of ÂÒÈÏ (15 0-1,8) ‘he is not’ is a clear literal translation, as the ¡ar˙an combined the pronominal suffix /-o/ with /lays/ in the same way the Hebrew has it, namely ÂÈ‡, whereas the result /layso/ does not exist in regular Arabic use.1 In Genesis: ˘ÂÂ˜Ó ÂÒÈÏ È≤Ï‡ (15 12b,7) ‘who is not speckled’ translates „S ÂÈ‡ ¯˘‡ (Gen 30:33); ÒÓ‡ Ï‡ ÒÓ‡Î ‡È‡ÚÓ ÂÒÈÏ 1

This feature can also be analyzed under 6-2.1, pronominal suffixes (p. 238).

CHAPTER EIGHT (15 13a,12) ‘who is not with me2 as in earlier days’ translates ÌÂ˘Ï˘ ÏÂÓ˙Î ÂÓÚ ÂÈ‡ (Gen 31:2); and ÂÒÈÏ „ÏÂ Ï‡Â (15 22b,9) ‘and the boy is not (here)’ renders ÂÈ‡ „ÏÈ‰ (Gen 37:30). In the Haggadah: Ï‡ÒÈ ¥¯ÚÈ ÂÒÈÏ È≤Ï‡Â (3 5,8) ‘and the one who does not know how to ask’ translates the Hebrew ÂÈ‡˘Â ÏÂ‡˘Ï Ú„ÂÈ. /lam/ can be used in the ¡ur¥˙ to negate nouns, pronouns, and prepositions in “un-Arabic” constructions, although there is a possibility of colloquial use here (see pp. 126–27, 3.3.3).3 (4-1.3) In Genesis, ˙˜Â ÌÏ (15 10a,19) ‘not yet time’ translates ˙Ú ‡Ï (Gen 29:7); Ô‡ÓÎ Ï‡˜È ·Â˜ÚÈ ÌÏ (15 16a,19–20)4 ‘no longer will (your name) be called Jacob’ translates ¯ÓÀ‡È" ·SÚÈ ‡Ï „ÂÚ (Gen 32:29); Ô‡Ï‡ ÌÏ (15 17a,1) ‘not now’ translates ‡ Ï‡ (Gen 33:10); ÏÒ Ï‡ ÔÂÎÈ ÂÏ ÌÏ (15 23a,8–9) ‘the seed would not count as his’ translates Ú¯Ê‰ ‰È‰È ÂÏ ‡Ï (Gen 38:9). (4-1.4) In the following examples in Esther, the /lam/ precedes the prepositions />ala/ and /ka-/ respectively: ¬ÏÓ Ï‡ ‡ÏÚ ÌÏ Â„ÁÂÏ (1302 1b,23) ‘not only against the king’ translates ÏÚ ‡Ï Â„·Ï ÍÏÓ‰ (Esth 1:16), and ÔÂ‡˜ Ï‡Î ÌÏ È≤Ï‡ ¬ÏÓ Ï‡ Ì‡„˜Ï È‚‡ (1302 4a,11) ‘I will go before the king, not according to the law’ renders ˙„Î ‡Ï ¯˘‡ ÍÏÓ‰ Ï‡ ‡Â·‡ (Esth 4:16). (4-1.5) In the Haggadah as well, /lam/ negates nouns, pronouns, and prepositions: ¬ÈÏÓ ÌÏÂ ‰‡ (91 5b,16) ‘I, and not an angel’ renders Í‡ÏÓ ‡ÏÂ È‡; ÂÏ ÌÏÂ ÌÂÎÏ (3 4,18) ‘to you (pl.) and not to him’ translates ÂÏ ‡ÏÂ ÌÎÏ; ÌÂ‰Ï ÌÏ √¯‡ È¥ (3 6,16) ‘in a land that is not theirs’ translates Ì‰Ï ‡Ï ı¯‡·; Ë˜¥ „Á‡Â ÌÏ Ô‡ (3 7,2–3) ‘for not one only’ renders „·Ï· „Á‡ ‡Ï˘; and ¬‡ÏÓ „È ‰ÏÚ ÌÏ (3 11,10–11) ‘not by an angel’ translates Í‡ÏÓ È„È ÏÚ ‡Ï. 2 3 4

Note that Hebrew ÂÓÚ ‘with him’ is rendered ‡È‡ÚÓ ‘with me’ in Judeo-Arabic. On the use of /lam/ in Egyptian Arabic, see Rosenbuam 2002a:588–95. Semantically the /lam/ here negates the verb Ï‡˜È, but it directly precedes the personal name ·Â˜ÚÈ.

214

THE MORPHOSYNTACTIC LEVEL: NEGATION – VERBAL L/I.: (4-1.6) The negation particle /lays/ is used in the ¡ur¥˙ in regular Arabic style, but at the same time it also serves as a literal translation, thus reflecting the literal/interpretive tension. In Genesis: ‰ÚÓ Ï‚‡¯ ÒÈÏ (15 15a,5) ‘there is no man with us’ translates ÂÓÚ ˘È‡ ÔÈ‡ (Gen 31:50); ‰ÈÓ ‰È¥ ÒÈÏ Èƒ‡¥ ¯È· Ï‡Â (15 22a,19) ‘and the pit is empty with no water in it’ translates ÌÈÓ Â· ÔÈ‡ ˜V ¯Â·‰Â (Gen 37:24); and ¯È· Ï‡ È¥ ¥ÒÂÈ ÒÈÏ (15 22b,8) ‘Joseph is not in the pit’ translates ¯Â·· ÛÒÂÈ ÔÈ‡ (Gen 37:29). In the book of Esther: ·ˆµÓ ÒÈÏ ÔÂ‡˜ Ï‡Î ·¯˘ Ï‡Â (1302 1b,11–12) ‘Drinking was according to the law, where no one was forced’ translates Ò›‡ ÔÈ‡ ˙„Î ‰È˙˘‰Â (Esth 1:8); and ¯˙Ò‡ Ì‡Â ·‡ ‡‰Ï ÒÈÏ Ô‡ ÂÓÚ ˙· (1302 2b,2) ‘Esther, his uncle’s daughter, as she does not have a father and a mother’ translates Ì‡Â ·‡ ‰Ï ÔÈ‡ ÈÎ Â„I ˙· ¯˙Ò‡ (Esth 2:7). In the Haggadah: Ìµ ÏÏ ‰Ú¯Ó ÒÈÏ (3 7,17) ‘there is no pasture for the flocks’ translates Ô‡ˆÃÏ ‰Ú¯Ó ÔÈ‡; and Ô‡ËÏÂˆ ‰Ï ÒÈÏ (93 75,9) ‘we do not have a king’ translates ÍÏÓ ÂÏ ÔÈ‡. Negation: Verbal Feature 4-2 — Verbal Lit.: (4-2.1) The use of the particle /lam/ to negate the perfect in the ¡ar˙ is a verbatim translation that imitates the Hebrew original in its use of the negation particle and the following tense/aspect (see also pp. 293–94, 11-2.1–11-2.3). In Genesis: ‡‰¥¯Ú ÌÏ Ï‚‡¯Â (15 2b,17) ‘and no man had known her’ translates ‰Ú„È ‡Ï ˘È‡Â (Gen 24:16); ˙¥¯Ú ÌÏ ‰‡Â (15 9b,19) ‘and I did not know’ translates È˙Ú„È ‡Ï ÈÎ‡Â (Gen 28:16); and ·Â˜ÚÈÏ ˙„ÏÂÂ ÌÏ (15 11b,1–2) ‘(she) had not borne to Jacob’ renders ·SÚÈÏ ‰„ÏÈ ‡Ï (Gen 30:1). In Esther: ˙ÏÓÚ ÌÏ (1302 1b,22) ‘(she) did not do’ translates ‰˙◊Ú ‡Ï (Esth 1:15); ÈÈ˘ ˙·ÏË ÌÏ (1302 2b,15) ‘(she) did not ask for anything’ translates ¯·„ ‰˘˜· ‡Ï (Esth 2:15); and ·‰ ÏÏÂ Ì‰ÿ„È Âÿ„Ó ÌÏ (1302 7a,11) ‘and they did not lay their hands on 215

CHAPTER EIGHT the spoils’ translates Ì„È ˙‡ ÂÁÏ˘ ‡Ï ‰Ê··Â (Esth 9:16). In the Haggadah: ÌÎÁ ÌÏ (3 7,9) ‘did not issue an edict’ translates ¯Ê‚ ‡Ï; ÏÊ ÌÏ (3 7,15) ‘did not go’ translates „¯È ‡Ï; and Úˆ ÌÏ (3 16,2) ‘did not execute’ translates ‰◊Ú ‡Ï. In the above examples the ¡ar˙an could have used /ma … ¡/ instead, in an interpretive colloquial translation, as he indeed did in some other places (see below, p. 218, 4-2.5). Note that in Later Egyptian Judeo-Arabic in general, and in the texts of the ¡ur¥˙ in particular, the negation particle /lam/ appears, not infrequently, preceded by a perfect verb.5 This may be an outcome of the appearance of a common hypocorrection, occurring frequently, which consequently resulted in the standardization of the phenomenon in Later Egyptian JudeoArabic. In other words, in the Judeo-Arabic dialect, the regular particle for negating the past tense was /må/. The writer did not choose it because it was the unmarked form, dominant in his dialect, and thus not prestigious enough to suit his taste. Instead he chose the prestigious marked Classical Arabic negation particle /lam/. However, he “corrected” only halfway: he only changed /må/ to /lam/ and did not replace the perfect form following it with the jussive, as required by standard Classical Arabic.6 When a pseudocorrection becomes prevalent in the variety, at some point it ceases to be a pseudocorrection and becomes an accepted form of the variety. Here, because these forms were regularly used in Later Egyptian Judeo-Arabic, 5

6

See Blau 1981:223; Hary 1992:294–95, 314; 2007:276–77; Rosenbuam 2002:588–95. See also the extensive discussion of the phenomenon in this volume in chapter 4, pp. 94–95, 126–27; and in chapter 5, pp. 141–43. This example follows the criteria for hypocorrections as outlined in Hary 1992:63–67 and 2007. The underlying form differs from the form in the prestigious variety; the resulting form contains a vernacular feature (the use of the perfect form, not the jussive); it does not go far enough (i.e., it does not change to the jussive); and the form /lam/ followed by the perfect does not exist in the prestigious variety, nor in the dialect.

216

THE MORPHOSYNTACTIC LEVEL: NEGATION – VERBAL they became standardized at one point at least in writing. This is a case where a pseudocorrection was created, used quite frequently, and then became standardized as part of the dialect. I agree with Rosenbaum that there is a possibility, albeit unproven, that /lam/ was used in Egyptian Arabic the dialect, 7 but I maintain that my original analysis of this phenomenon, first articulated in my doctoral thesis in 1987 and later in 1992 (pp. 67, 294, 314), is valid. In other words, synchronically we may say that /lam/ is part of colloquial Egyptian Arabic, but there are good reasons to assume that diachronically entered to colloquial Egyptian from the literal variety, through hypocorrections that later became standardized. (4-2.2) The example ¥¯Ú ÌÏÂ (15 25b,15–16) ‘and we did not know’ is a calque translation of the Hebrew Ú„Â ‡ÏÂ (Gen 41:21) ‘and it was not known.’ In this case the ¡ar˙an transferred the Hebrew nif>al verbal prefix into a Judeo-Arabic imperfect first person plural verbal prefix /n-/, thus changing the meaning slightly. This was done in order to reflect the verbatim translation.8 (4-2.3) The use of /lam/ followed by the imperfect reflects a literal translation, but it may also reflect spoken Cairene JudeoArabic.9 In Genesis: Ú¥¯È ÌÏ (15 26b,10) ‘will not raise’ translates ‡Ï ÌÈ¯È (Gen 41:44); Â˙ÂÓ˙ ÌÏÂ (15 27b,19) ‘and you may not die’ renders Â˙ÂÓ˙ ‡ÏÂ (Gen 42:20); and ¬ÈÏ‡ Â·È‚‡ ÌÏ (15 28b,9) ‘I will not bring him to you’ translates ÍÈÏ‡ Â‡È·‡ ‡Ï (Gen 42:37). In Esther: ¯ÈÈµ˙È ÌÏÂ (1302 2a,5) ‘and it will not change’ translates ¯Â·ÚÈ ‡ÏÂ (Esth 1:19); ¯·¡˙ ÌÏ (1302 2b,7) ‘(she) will not inform’ renders „È‚˙ ‡Ï (Esth 2:10); and ¯„˜˙ ÌÏ (1302

7 8 9

Rosenbuam 2002a:595. See chapter 6, pp. 178–81, “Calque Translation.” See chapter 4, p. 126, 3.3.3, and later in this chapter, p, 294, 11-2.4–11-2.6.

217

CHAPTER EIGHT 5a,14) ‘you will not be able’ translates ÏÎÂ˙ ‡Ï (Esth 6:13). In the Haggadah: È˘˙¡ ÌÏ (93 59,9–10) ‘we will not be embarrassed’ translates ÌÏÎ$ ‡Ï; ÏÂ˜È ÌÏ (3 17,18) ‘is not saying’ translates ¯Ó‡ ‡Ï; ‚Â¯¡È ÌÏ (3 17,19) ‘is not performing’ renders ‡ˆÈ ‡Ï; and ¯Ò˜ ÌÏ (93 58,2) ‘is not lacking’ translates ¯ÒÁ ‡Ï. (4-2.4) The negation particle /lays/ can be used, although rarely, to negate verbs in a verbatim way, creating a “strange” Arabic construction. In the book of Esther the ¡ar˙an used /lays/ to translate verbatim the Hebrew ÔÈ‡ in ·È˘‰Ï ÔÈ‡ (Esth 8:8); however, he used the particle with the passive verb /yurga>/, Ú‚¯È ÒÈÏ (1302 6a,6) ‘(it) will not be revoked.’ Note that in Hebrew the phrase ·È˘‰Ï ÔÈ‡ can mean both ‘(it) will not be revoked’ and ‘(it) must not be revoked,’ while in Judeo-Arabic only the first meaning is possible. Int.: (4-2.5) The spoken use of /ma … ¡/ is used infrequently to negate the imperfect in an interpretive translation, using spoken Egyptian Judeo-Arabic: ˘¥¯ÚÈ ‡Ó /mayi>raf¡/ (3 4,10) ‘(he) does not know.’ This same particle can be used to negate the perfect, also in an interpretive translation: ˘È·ÁÓ /ma˙abb∑¡/ (15 23-1a,19) ‘(he) did not want’ (see pp. 125–26, 3.3.1). (4-2.6) Similarly infrequent is the spoken use of /mu¡/ to negate the participle, which is also an interpretive translation, using regular spoken Egyptian Judeo-Arabic: ÔÈÁÈ‡¯ ˘ÂÓ /mu¡ ray˙•n/ (3 1,4) ‘(they) are not going’ (chapter 4, p. 126, 3.3.2). L/I: (4-2.7) The negation of the imperfect with /lam/ to indicate the negated imperative is usual in the ¡ur¥˙ in a literal translation, as /lam/ is the verbatim equivalent of the Hebrew ÏÃ‡: In Genesis: ¥‡¡˙ ÌÏ (15 7a,1) ‘do not fear’ translates ‡¯È˙ Ï‡ (Gen 26:24); „Â¡‡˙ ÌÏ (15 9a,10) ‘do not take’; and ÈÏ ÈËÚ˙ ÌÏ ÈÈ˘ (15 12b,2) ‘do not give me a thing.’ In Esther: Â·¯˘˙ ÌÏÂ ÂÏÎ‡˙ ÌÏÂ (1302 4a,10) ‘and do not eat and do not drink’ translates Â˙˘˙ Ï‡Â ÂÏÎ‡˙ Ï‡Â (Esth 4:16) and 218

THE MORPHOSYNTACTIC LEVEL: PREPOSITIONS ÈÈ˘ ı˜˙ ÌÏ (1302 5a,9) ‘do not leave out a thing’ renders Ï‡ ¯·„ ÏÙ˙ (Esth 6:11). In the Haggadah: ‰‚ÈÂÁ˙ ÌÏÂ (93 59,6) ‘and do not make us need’ translates ÂÎÈ¯ˆ˙ Ï‡ ‡Â; ‰‡Ò˙ ÌÏ (93 60,1) ‘do not forget us’ renders ÂÁÎ˘˙ Ï‡; and ‰ÎÂ¯˙˙ ÌÏÂ (93 60,2) ‘do not leave us’ translates ÂÁÊ˙ Ï‡Â. However, Esther includes an example where the negation particle /lå/ is used in a more interpretive mode: È¥ ÈÓÓ¡˙ ‡Ï ¬Ò¥ (1302 4a,6) ‘do not fool yourself,’ thus demonstrating the literal/interpretive linguistic tension. Prepositions Level: Morphosyntactic; Category 5: Prepositions/Particles Feature 5-1 — Prepositions Lit.: (5-1.1) The Judeo-Arabic preposition /ka-/ ‘like, about’ is regularly selected to translate verbatim the Hebrew ≠Ÿk, probably because of phonetic and semantic proximity, using the same word order as the Hebrew: ‰Ò ÔÈÚ·Ò Ô·‡Î ‰‡ (3 3,18-19) ‘I am about seventy years old’ translates ‰˘ ÌÈÚ·˘ Ô·Î È‡; ·Î‡ÂÎÎ ‰ÓÒ Ï‡ (3 8,4) ‘like the stars of the heaven’ renders È·ÎÂÎÎ ÌÈÓ˘‰; and Ô‡¥¯¡ Ï‡Î ˛Â¸ÒÂ˜¯È Ï‡·‚ Ï‡ (74 11,17-12,1) ‘the mountains danced like young sheep’ renders ÌÈÏÈ‡Î Â„˜¯˙ ÌÈ¯‰‰. This preposition is used in the translation even when regular Arabic use does not call for it, just to keep the biblical “spirit” in the ¡ar˙ in a typical literal translation. 1 0 This is the way that the Judeo-Arabic /ka-/ behaves in Â„ÈÒ ÂÚÈÓÒÎ Ô‡ÎÂ (15 23-1b,15–16) ‘when his master heard,’ translating literally È‰ÈÂ ÂÈI‡ Ú›Ó˘Î (Gen 39:19); and in ¬ÏÓ Ï‡ ÒÂÏ‚Î Ì‡ÈÈ‡ Ï‡ ¬Ï‡≤ È¥ (1302 1b:3) ‘in those days as the king sat,’ which translates ÍÏÓ‰ ˙·˘Î Ì‰‰ ÌÈÓÈ· (Esth 1:2). Similarly, the Judeo-Arabic 10

Note that in Hebrew ≠Ÿk is also employed as an aspectual morpheme for a continuous action, much like English as.

219

CHAPTER EIGHT preposition /bi-/ ‘in’ renders verbatim Hebrew ≠· in ˙Ï‡ÓÎ·Â ÈÏÂ„‰ Ì‡ÈÈ‡ Ï‡ (1302 1b,7) ‘and when these days were fulfilled,’ which translates ‰Ï‡‰ ÌÈÓÈ‰ ˙Â‡ÏÓ·Â (Esth 1:5). (5-1.2) In the Egyptian ¡ar˙ it is a common practice to translate Hebrew „Ú ‘until’ as „Ú, probably because of phonetic proximit y, as is the case in the previous section, which indicates a verbatim translation. In Genesis this preposition appears in this way regularly in 15 3a,1; 3b,1; 5a,15; 6b,5; 7,16; and more. Similarly, in the ¡ar˙ of the book of Esther, Judeo-Arabic „Ú (1302 3b,11; 4a,17; 4b,1; and more) is also used regularly to translate Hebrew „Ú. In the Passover Haggadah, though, the ¡ar˙an used the Judeo-Arabic preposition „ÚÏ (3 3,14; 3,20; 20,4) ‘until’ to translate Hebrew „Ú in a more common colloquial use. In one case the ¡ar˙an used „ÚÂ (3 20,4) ‘and until’ to translate literally the Hebrew „ÚÂ with the coordinating vav, but in another manuscript of this translation (93 43,5) he still employed the more colloquial form „ÚÏ. This literal translation is seen even more clearly in the translation of the Hebrew „›‡Ó „Ú ‘to a very (big) extent’ (Gen 27:33, 34) into Judeo-Arabic ÈÂÂ˜ „Ú (15 8b,5; 8b,7). Likewise, ¯˘‡ „Ú ‘until that’ (Gen 27:44; 28:15; 29:8) is also literally translated into È≤Ï‡ „Ú (15 9a,4; 9b,16; 10a,20). In the same verbatim way, Hebrew Ì‡ „Ú ‘until if, whether’ (Gen 24:19, 33) is translated into Judeo-Arabic Ô‡Î‡ „Ú (15 3a,1; 3b,1). Finally, the same occurs in the sentence ÍÏÓ ¬ÏÓ È·‡ ¥¯˙˘‡Â ¬‡·Â˘ Ï‡ „Ú È¥ ÌÈ˙˘ÏÙ (15 6a,16-17) ‘and Abimelech, king of the Philistines, gazed down through the window,’ translating ÔÂÏÁ‰ „Ú· ÌÈ˙˘ÏÙ ÍÏÓ ÍÏÓÈ·‡ Û˜˘ÈÂ (Gen 26:8). In this example the ¡ar˙an literally matched the Hebrew „Ú· ‘through’ (Gen 26:8) with the two equivalent Judeo-Arabic prepositions /fi/ and />and/ to produce „Ú È¥. In the book of Esther a similar phenomenon occurs in the sentence ¬ÏÓ Ï‡ ·‡· Ì‡„˜Ï „Ú ‡‚Â (1302 3b,11) ‘and he came to the front of the king’s gate.’ In this 220

THE MORPHOSYNTACTIC LEVEL: PREPOSITIONS instance too, the ¡ar˙an matched the Hebrew ÈÙÏ „Ú (Esth 4:2) with the equivalent Judeo-Arabic prepositions Ì‡„˜Ï „Ú. (5-1.3) The use of „Ú is also applied elsewhere. Thus „Ú ÔÓ (15 2a,11; 5a,4; 8a,20; and more; 1302 5a,7) ‘from’ translates literally the Hebrew ˙‡Ó (Gen 23:20; 25:10; 27:30; and more; Esth 7:7). The same occurs in È„Ú ÔÓ (15 31a,13) ‘from me,’ which translates È˙œ‡Ó (Gen 44:28); Â„Ú ÔÓ (15 7a,13) ‘from him’ rendering Â˙‡Ó (Gen 26:31); ÌÂÎ„Ú ÔÓ (15 7a,6) ‘from you (masc. pl.)’ for ÌÎ˙‡Ó (Gen 26:27); and ÌÂ‰„Ú ÔÓ (15 28a,6) ‘from them (masc.)’ for Ì˙‡Ó (Gen 42:24). Furthermore, „Ú ÔÓ (15 26a,13; 31a,14; 31a,19; and more; 1302 5a,7) ‘from’ also translates ÌÚÓ (Gen 41:33; 44:29, 32; and more); È„Ú ÔÓ (15 14a,13) ‘from me’ for ÈÓÚÓ (Gen 31:31); and ‰„Ú ÔÓ (15 6b,9) ‘from us’ for ÂÓÚÓ (Gen 26:16). In one place in Genesis (24:27), though, ÌÚÓ is translated verbatim as ‰ÚÓ ÔÓ (15 3a,11) ‘from.’ (5-1.4) The Judeo-Arabic preposition /li-/ in „Ï· ÏÏ ‰¯· (15 2a,11) ‘outside the city’ translates verbatim the Hebrew ıÂÁÓ ¯ÈÚÏ (Gen 24:11), whereas regular Arabic use calls for a zero preposition, /barra l-balad/. (5-1.5) The combination of the Judeo-Arabic prepositions /li-/ and /min/ is used to translate literally the equivalent Hebrew ≠ÓÏ: ÔË˜ „ÚÂ ÏÂ„‚ÓÏ ‘big and small alike’ (Esth 1:5) becomes ÔÓÏ ¯Èµˆ „ÚÂ ¯È·Î (1302 1b,7–8). Some Judeo-Arabic verbs take the same prepositions as their Hebrew equivalent just to satisfy the verbatim translation, and are thus different from regular Arabic use: (5-1.6) The following sentences exemplify the nonuse of prepositions or the employment of a zero preposition: ¯„˜ ÌÏ ·ÈÈË ÂÂ‡ Èƒ¯ ¬ÈÏ‡ ÌÏÎ˙ (15 4a,7–8) ‘we cannot speak to you bad or good’ follows verbatim the Hebrew Â‡ Ú¯ ÍÈÏ‡ ¯·„ ÏÎÂ ‡Ï ·ÂË (Gen 24:50), with a zero preposition preceding ‘bad or good.’ In Saadia as well as in the Protestant translation, the

221

CHAPTER EIGHT preposition /bi-/ is used in an interpretive translation, for example /bi-¡arr/ ‘in bad.’ In the same way, the Hebrew sentence ˘„Á ¯È˘ ÍÏ ‰„ÂÂ ‘and we shall thank you with a new song’ is translated verbatim into ‰„È„‚ ‰ÁÈ·Ò˙ ¬Ï ¯ÂÎ˘Â (3 21,11), using a zero preposition for the concept of “with,” following the Hebrew original. (5-1.7) The Judeo-Arabic preposition /ma>a/ ‘with’ in ÚˆÂ È„ÈÒ ÚÓ Ïƒ¥Ó (15 2b,10) ‘and deal graciously with my master’ is a verbatim translation of the Hebrew ÌÚ in È„‡ ÌÚ „ÒÁ ‰˘ÚÂ (Gen 24:12). Rather than /ma>a/, the preposition /ilå/ is expected in this sentence, as is the case in Saadia and the Protestant translation. (5-1.8) In the sentence ˙¯‡µÓ ‰Ï‡ Â˙‡¯Ó ‰¯˘ ‰Ï‡ Ì‰¯·‡ Ô¥„ ‰È˙Ó Ï‡ ËÈµ (15 2a,8–9) ‘Abraham buried his wife Sarah in the cave of the field of the Machpelah,’ the second preposition ‰Ï‡ ‘to’ follows slavishly Hebrew Ï‡ in Gen 23:19, despite the fact that it means ‘in.’ Here, the ¡ar˙an could have translated interpretively using Judeo-Arabic /fi/, but chose to be “loyal” and literal to the Hebrew original. (5-1.9) In the phrase ÔÈÚ Ï‡ ‰ÏÚ Ï‡Ó‚ Ï‡ ‰ÏÚ ¥˜‡Â ‰„ÂÂ‰Â (15 3a,15–16) ‘and behold, he stood by the camels at the spring,’ the first />ala/ translates verbatim Hebrew ÏÚ, meaning in this context ‘by, beside.’ The ¡ar˙an could have chosen another Judeo-Arabic preposition to reflect the meaning more accurately, but he opted to preserve the Hebrew text by using ‰ÏÚ. In the Protestant translation, the translator chose the preposition bMŽ ‘by’ to better reflect the accurate meaning; however, Saadia also selected ÈÏÚ. (5-1.10) The Judeo-Arabic preposition /fi/, which means ‘for’ in ¬˙· ÏÁ¯ È¥ ÔÈÒ Ú·Ò ¬Ó„¡‡ (15 10b,16) ‘I will serve you seven years for your daughter Rachel,’ is a literal translation of the Hebrew Í˙· ÏÁ¯· ÌÈ˘ Ú·˘ Í„·Ú‡ (Gen 29:18) and is different from regular Arabic use. 222

THE MORPHOSYNTACTIC LEVEL: PREPOSITIONS (5-1.11) ‰È¥ ¬¯‡Ú˙ÈÂ (3 9,5) ‘and he will fight us’ is a literal translation of the Hebrew Â· ÌÁÏÂ, as /it>årak/ usually takes a zero preposition. The translation could have been even more literal had the ¡ar˙an chosen the preposition /bi-/.1 1 (5-1.12) In order to keep the translation as literal as possible, the ¡ar˙an at times used the preposition /li-/ ‘to,’ even if the Arabic verb required a zero preposition. In Genesis: ÏÁ¯Ï ·Â˜ÚÈ Ò‡·Â (15 10b,6) ‘Then Jacob kissed Rachel’ translates ÏÁ¯Ï ·˜ÚÈ ˜˘ÈÂ (Gen 29:11); ÏÁ¯Ï ·Â˜ÚÈ ¯·¡Â (15 10b,7) ‘and Jacob told Rachel’ renders ÏÁ¯Ï ·˜ÚÈ „‚ÈÂ (Gen 29:12); ‡‰Â·‡Ï ˙¯·¡Â (15 10b,8) ‘and she told her father’ translates ‰È·‡Ï „‚˙Â (Gen 29:12); and Â‰Ï Ò‡·Â Â‰Ï ˜ÚÂ (15 10b,9–10) ‘and he embraced him and kissed him’ renders ˜·ÁÈÂ ÂÏ ˜˘ÈÂ ÂÏ (Gen 29:13). Similarly, the preposition /li-/ in ¬·‡Ï ¯·¡˙Â (3 5,9) ‘and you shall tell your son’ from the Haggadah constitutes a literal translation of the Hebrew ≠Ï in Í·Ï ˙„‚‰Â; the Arabic verb /axbar/ does not take this preposition, but rather a zero preposition. The same applies to ‡‰ÓÂ‡ ˙È·Ï ˙¯·¡Â (15 3a,12) ‘and (she) told (this) to her mother’s household.’ (5-1.13) The preposition /li-/ becomes subject to the Hebrew original, even when it creates “un-Arabic” sentences: Ô‡ÓÎ È˜Ò‡ ¬Ï‡Ó‚Ï (15 3a,1; 3b,16) ‘I will also water your camels.’ Here, the Judeo-Arabic /li-/ is an imitation of the Hebrew ≠Ï in ·‡˘‡ ÍÈÏÓ‚Ï Ì‚ (Gen 24:19, 44). Furthermore, in Gen 24:46 the Hebrew ≠Ï does not appear, ‰˜˘‡ ÍÈÏÓ‚ Ì‚, but the ¡ar˙ still uses it: È˜Ò‡ ¬Ï‡Ó‚Ï Ô‡ÓÎÂ (15 3b,19). This supports the argument that the ¡ar˙ created its own unique Judeo-Arabic structure. In another place, however, when the Hebrew does not have the preposition ≠Ï, in ‰˙˜˘‰ ÌÈÏÓ‚‰ Ì‚ (Gen 24:46), neither does the ¡ar˙, ˙˜Ò Ï‡Ó‚ Ï‡ Ô‡ÓÎÂ (15 3b,20) ‘and (she) 11

See Chapter 3, pp. 83–85, “The Translation Continuum.”

223

CHAPTER EIGHT also watered the camels.’ Both Saadia and the Protestant translation used a zero preposition in all cases with the verb /saqå/ ‘water, give drink,’ as is customary in Arabic structure. (5-1.14) The preposition />ala/ in ‰˜‡ÓÁ È¡„¯Ó ‡ÏÚ ÔÓ‰ ‡ÏÓ˙‡Â (1302 4b,5) ‘Haman was filled with wrath at Mordecai’ follows faithfully the Hebrew ÏÚ in ‰ÓÁ ÈÎ„¯Ó ÏÚ ÔÓ‰ ‡ÏÓÈÂ (Esth 5:9). Both Saadia and the Protestant translation employed />ala/ as well. Several prepositions, used in combination with other prepositions, nouns, or even verbs, are employed in the ¡ar˙ to translate literally their Hebrew equivalent. In other words, these Judeo-Arabic prepositions are clear literal translations of the Hebrew originals, using calque translations.1 2 (5-1.15) „‡ˆÂ˜Î (3 4,8) ‘as against’ clearly translates verbatim the Hebrew „‚Î, using a combination of literal translation and the local dialect in order to be understood. Hebrew „‚ is translated with the colloquial [
12

See also chapter 6, pp. 178–81, “Calque Translation.”

224

THE MORPHOSYNTACTIC LEVEL: PREPOSITIONS /˙•n/ to arrive at /min ˙•n/. Similarly, the Judeo-Arabic ÔÈÁ· ÔÓ (3 5,13) ‘whilst’ translates literally the Hebrew „ÂÚ·Ó, this time with its three morphemes: Hebrew ≠Ó is translated by /min/, ≠· by /bi-/, and „ÂÚ by /˙•n/ to arrive in the innovative /min bi-˙•n/. (5-1.17) Similarly, combinations of Judeo-Arabic />ala/ with other nouns and prepositions are also representations of a verbatim translation. Thus, ÏÂ˜ ‰ÏÚ (3 7,14) ‘in accordance with’ translates literally the Hebrew ÈÙ ÏÚ (where ÏÂ˜ ‘word’ renders Hebrew ‰Ù ‘mouth’), and „È ‰ÏÚ (3 11,10) ‘by means of, through’ translates verbatim the Hebrew È„È ÏÚ. Likewise, ··Ò ‰ÏÚ (3 18,3; 18,4; 18,10) ‘for a reason, because’ is a literal translation of Hebrew ÌÂ˘ ÏÚ, with />ala/ translating Hebrew ÏÚ ‘on’ and /sabab/ translating Hebrew ÌÂ˘ ‘reason.’ In another manuscript Ô‡˘ ‡ÏÚ (91 8a,18; 8b,1) is used to translate ÌÂ˘ ÏÚ. In the same manner, ‰‚Â ‰ÏÚ (15 2a,9) ‘on the face of, facing’ is a verbatim translation of the Hebrew ÈÙ ÏÚ, with />ala/ translating Hebrew ÏÚ ‘on,’ as in the previous example, and /wagh/ translating Hebrew ÈÙ ‘face of.’ Still with the preposition />ala/, ‰ÏÚ ÔÓ (15 4b,6; 15 24b,13) ‘from above’ occurs in Genesis, translating verbatim the Hebrew ÏÚÓ (Gen 24:64; 40:17). The Protestant translation, on the other hand, chose />an/ in this example in a more interpretive translation, and Saadia used both />an/ and /fawqa/ respectively. In the same way, in ˜ÁˆÈ ‰ÏÚ ÔÓ ÌÂ‰ÏÒ¯Â (15 4b,18) ‘then (he) sent them away from Isaac,’ the ¡ar˙an used ‰ÏÚ ÔÓ to translate verbatim the Hebrew ÏÚÓ in ˜ÁˆÈ ÏÚÓ ÌÁÏ˘ÈÂ (Gen 25:6), whereas both Saadia and the Protestant translation used />an/. (5-1.18) The Judeo-Arabic preposition /li-/ is used both in ··ÒÏ (3 4,18) ‘in accordance with’ and in Ï‚‡Ï (74 2,13) in verbatim translation of the Hebrew ≠Ï in ÈÙÏ. (5-1.19) Similarly, both the Judeo-Arabic prepositions /fi/ and /bi-/ in ··Ò È¥ (3 4,20) ‘for this’ and in ··Ò· (91 2b,3) translate literally the Hebrew ≠· in ¯Â·Ú· ‘for.’ 225

CHAPTER EIGHT (5-1.20) Judeo-Arabic ÂÈ·‡Ó È¥ (15 2a,16) /fi mab∑no/ ‘among it’ translates verbatim the Hebrew Â·¯˜· (Gen 24:3). (5-1.21) Judeo-Arabic ˙˜ÂÂÏ (15 2b,9) /li-wa
226

THE MORPHOSYNTACTIC LEVEL: PREPOSITIONS (5-1.26) In ·¯ Ï‡ ‰ÚÓ ÍÂÁ ‰˘Ó˙‡Â (15 0-1,6) ‘and Enoch walked with God,’ which renders ÌÈ‰¿‡‰ ˙‡ ÍÂÁ ÍÕÏ‰˙ÈÂ (Gen 5:22), the ¡ar˙an translated Hebrew ˙‡ into Judeo-Arabic /ma>a/ ‘with’ in a clear interpretive manner. In other words, the ¡ar˙an did not translate Hebrew ˙‡ blindly into Judeo-Arabic /ilå/ in this instance. Realizing that Hebrew ˙‡ had indeed a different meaning, the ¡ar˙an expressed it in his translation (see also pp. 261–62, 8-1.14). (5-1.27) In the same way, the ¡ar˙an interpretively translated Hebrew ÈÙ ÏÚ ‘on the surface’ using Judeo-Arabic />ala/ in ‰ÏÚ √¯‡ Ï‡ (15 0-1,18) ‘on earth’ (‰Ó„‡‰ ÈÙ ÏÚ, Gen 6:1). (5-1.28) The use of the preposition /min/ in ÏÈÏ˜ ÔÓ Ô‡Ï‡ ÈÈË¥˘ ‰ÈÓ (15 2b,19) ‘let me sip a little water,’ despite the fact that it is lacking in the Hebrew ÌÈÓ ËÚÓ ‡ ÈÈ‡ÈÓ‚‰ (Gen 24:17), points to an interpretive translation, reflecting regular Arabic style. (5-1.29) Similarly, the ¡ar˙an employed /minhum/ ‘from them’ to translate Ì˙›‡ ‘them’ in a clear interpretive translation in ÌÂ‰Ó √¯‡ Ï‡ ˙Ï˙Ó‡Â (3 8,9–10) ‘and the land was filled with them,’ rendering Ì˙›‡ ı¯‡‰ ‡ÏÓ˙Â. At times the ¡ar˙an used spoken Judeo-Arabic prepositions in a probable attempt to translate interpretively: (5-1.30) The Egyptian Judeo-Arabic preposition /
227

CHAPTER EIGHT since the ¡ar˙an chose an Egyptian Judeo-Arabic dialectal preposition for his translation. At other times the ¡ar˙an selected a nonequivalent JudeoArabic preposition to translate the Hebrew original, also in an interpretive attempt. In other words, the ¡ar˙an may have translated Hebrew ≠· ‘in, at’ into Judeo-Arabic /min/ ‘from’ or Hebrew Ï‡ ‘to’ into Judeo-Arabic /fi/ ‘in’ for clarification. (5-1.32) Thus, although the Hebrew text uses the preposition Ãa ‘in the’ in ıÂÁÃa „ÓÚ˙ ‘(you [masc. sg.]) stand outside’ (Gen 24:31), the ¡ar˙an employed the Judeo-Arabic preposition ÔÓ /min/ ‘from’ in an interpretive translation: ‰¯· ÔÓ ¥˜Â˙ (15 3a,17–18). (5-1.33) Similarly, ÌÈ˘‰ ˙È· Ï‡ appears in Esth 2:3, but the Judeo-Arabic translation employs /fi/ in ‡Ò Ï‡ ˙È· È¥ (1302 2a,14) ‘in the women’s quarters’ in place of Hebrew Ï‡. L/I: (5-1.34) The Hebrew preposition ≠· ‘in, at’ may be translated literally into Judeo-Arabic with the phonetically similar JudeoArabic equivalent /bi-/, but it may also be translated interpretively with Judeo-Arabic /fi/. Thus, the use of the Judeo-Arabic preposition /bi-/ in ÌÈ∂Ú ·ÒÎÓ· Â‚Â¯¡È (3 6,19) ‘they come out with great wealth’ is literal, as it is phonetically similar to the Hebrew ≠· in ÏÂ„‚ ˘ÂÎ¯· Â‡ˆ"È . On the other hand, in ‚Â¯¡ È¥ ÈÎÁÏ ¯ˆÓ (3 3,6) ‘to tell the story of the exodus from Egypt’ the ¡ar˙an chose the Judeo-Arabic preposition /fi/, and not /bi-/, in translating ÌÈ¯ˆÓ ˙‡ÈˆÈ· ¯ÙÒÏ, in order to be better understood by employing the common Judeo-Arabic preposition. This choice of the preposition /fi/ in the ¡ar˙ to translate Hebrew ≠· in the interpretive mode is extremely common and occurs frequently in the texts of the ¡ur¥˙. Similarly, verbatim translation appears in ÏÈÏ‡˜ Ò‡· (91 4b,7) and in ÏÈÏ˜ Ò‡· (93 22,6) ‘with a few people,’ where the Judeo-Arabic preposition /bi-/ is used to render the Hebrew ≠· in ËÚÓ È˙Ó·. In 3 228

THE MORPHOSYNTACTIC LEVEL: PREPOSITIONS 8,2, however, the interpretive translation ÏÈÏ˜ Ò‡ È¥ occurs, with the more common use of the preposition /fi/, thus reflecting the literal/interpretive linguistic tension. In the same way /bi-/ in Ò¥ ÔÈÚ·Ò· (91 4b,7; 93 22,6) ‘with seventy people’ reflects a literal translation, compared with the more interpretive use of /fi/ in Ò¥ ÔÈÚ·Ò È¥ (3 8,2), rendering the Hebrew ˘Ù ÌÈÚ·˘·. (5-1.35) Prepositions preceding Hebrew „È /yad/ in prepositional phrases can also be translated in literal/interpretive manner. Thus, following the previous examples with /bi-/ and /fi/, the same phenomenon appears in the translation of the Hebrew „È· ‘in the hand of.’ In Genesis it is translated interpretively È¥ „È (15 8a,2; 12b,12; 36b,8; 24a,1), whereas in Esther, Hebrew „È· is treated both literally „È· (1302 3b,4; 6a,11) and interpretively „È ÔÚ (1302 1,12; 1,15). In the Haggadah, it is translated literally by „È· or „ÈÈ· in mss. 74, 91, and 93 and less literally by „È È¥ in ms. 3 (2,20; 11,6; 12,4). Hebrew „ÈÓ (Gen 32:12; 33:20; 38:20; 48:22) is translated literally as „È ÔÓ (15 15b,13; 17a,16; 23b,9; 36a,20) ‘from,’ and so is Hebrew „È ˙Á˙ (Gen 41:35), which is translated verbatim as „È ˙Á˙ (15 26a,17) ‘under the authority of.’ Such is also the case with Hebrew ÈØ„È ÏÚ (Gen 41:42; Esth 3:9, 6:9) ‘into the hand/s of,’ which is translated verbatim by „È ‰ÏÚ (15 26b,6) and „È ‡ÏÚ (1302 3a,17–18; 5a,5) respectively. On the other hand, Hebrew „È Ï‡ is translated interpretively in Esther (2:3, 8, 14) by „È ‡ÏÚ (1302 2a,14; 2b,4; 2b,12) ‘in the charge of.’ (5-1.36) The verb /daxal/ ‘cohabit’ in its sexual sense takes the preposition /ilå/ in a literal translation in ‡‰ÈÏ‡ Ï¡„Â (15 11a,1) ‘and he cohabited with her,’ following the Hebrew preposition Ï‡ in ‰ÈÏ‡ ‡›·ÈÂ (Gen 29:23). The same occurs in ‰Ï‡ Ô‡ÓÎ Ï¡„Â ÏÁ¯ (15 11a,11) ‘and he cohabited with Rachel also,’ where the ¡ar˙an followed verbatim the same Hebrew preposition in ‡›·ÈÂ ÏÁ¯ Ï‡ Ì‚ (Gen 29:30). On the other hand, in an interpretive mode, the ¡ar˙an translated ‰ÈÏ‡ ‰‡Â·‡Â (Gen 29:21) as ÏÂ¡„‡Â 229

CHAPTER EIGHT ‡‰ÈÏÚ (15 11a,1) ‘and I may cohabit with her,’ using the JudeoArabic preposition />ala/ to translate Hebrew Ï‡. (5-1.37) The Hebrew adverb „‡Ó „‡Ó· ‘very very’ is translated literally as ÈÂÂ˜ ÈÂÂ˜ Ï‡· (93 22,10–11) ‘very very’ 1 3 to account for the preposition ≠· ‘in’ at the beginning of the adverb (chapter 7, p. 197, 2-2.6). However, in other manuscripts, the phrase is translated more interpretively and in accordance with regular colloquial use: ÈÂÂ˜ ÈÂÂ˜ (3 8,9) and ÈÂÂ‡˜ ÈÂÂ‡˜ /al∑na/ in ÔÈÈ¯ˆÓ Ï‡ ‰ÈÏÚ ÂÒ‡Â (91 5a,1) ‘and the Egyptians were harsh toward us,’ which translates ÌÈ¯ˆÓ‰ Â˙Â‡ ÂÚ¯ÈÂ, is used interpretively as it follows regular Arabic use, whereas in ‰ÂÂ¯ˆÓ Ï‡ ‰‡ÈÈ‡ ÂÒ‡˜Â (3 9,2), the object pronoun /iyyåna/ is employed to imitate the Hebrew Â˙Â‡ in a verbatim translation. (5-1.41) The literal/interpretive linguistic tension is vividly seen in Esth 2:17. The ¡ar˙ reads, ‡‰Ò‡¯ ‡ÏÚ È¥ ‰ËÏÒ ‚‡˙ ÏÚ‚Â 13

Note, however, that the phrase ÈÂÂ˜ ÈÂÂ˜ Ï‡· employs the definite article, which is not used in the Hebrew original.

230

THE MORPHOSYNTACTIC LEVEL: COORDINATING PARTICLES (1302 2b,18–19) ‘then (he) set the royal crown upon her head,’ with the preposition /fi/ erased in the manuscript. It seems that the author wished to translate the Hebrew ˙ÂÎÏÓ ¯˙Î Ì˘ÈÂ ‰˘‡¯· literally and did so by using /fi/ to translate Hebrew ≠·, but finally regretted this for fear of forming an odd Arabic sentence. Accordingly, he erased the preposition /fi/ and wrote />ala/ instead.1 4 Coordinating Particles and Conjunctions Feature 5-2 — Coordinating particles and conjunctions Lit.: (5-2.1) Judeo-Arabic particle Ô‡ (/an/ or /in/) is used to translate several Hebrew particles and pronouns.1 5 For example, the ¡ar˙an used Ô‡ ‘that’ to translate verbatim the Hebrew ÈÎ ‘that,’ usually at the beginning of a sentence. In Genesis: ‰„˙·‡ Ô‡ Ô‡ÎÂ (15 0-1,18) ‘When men began’ translates ÏÁ‰ ÈÎ È‰ÈÂ (Gen 6:1); ÔÓÂ‰ ÔÈ·ÈÈË Ô‡ … Â¯∂Â (15 0-1,19–20) ‘and they saw … that they were good’ translates ˙›·›Ë ÈÎ ÆÆÆ Â‡¯ÈÂ ‰‰ (Gen 6:2) and È˙¥ÈÈË ‰Ï‡ È‚˙ Ô‡ (15 3b,10) ‘then you come to my people’ translates È˙ÁÙ˘Ó Ï‡ ‡Â·˙ ÈÎ (Gen 24:41). In Esther: È„Â‰È ‡ÂÂ‰ È≤Ï‡ Ì‰Ï ¯·¡ Ô‡ (1302 3a,9) ‘that he told them that he is Jewish’ translates verbatim ‡Â‰ ¯˘‡ Ì‰Ï „È‚‰ ÈÎ È„Â‰È (Esth 3:4); ÈÓÂ˜Â ‡‡ ‡ÈÚ·‡ Ô‡ (1302 5b,1) ‘that we have been sold, I and my people’ renders ÈÓÚÂ È‡ Â¯ÎÓ ÈÎ (Esth 7:4); and ‰ÈÈ„¯ Ï‡ ‰ÈÏ‡ ˙ˆÏ¡ Ô‡ (1302 5b,6) ‘that evil reached him’ renders ‰Ú¯‰ ÂÈÏ‡ ‰˙ÏÎ ÈÎ (Esth 7:7). In the Haggadah: ·¯‰˙ Ô‡ ¯Á· ‡È ¬Ï ˘‡ (3 20,17) ‘what happened to you, O sea, that you withdraw?’ translates ÍÏ ‰Ó ÒÂ˙ ÈÎ ÌÈ‰; ÂÏƒ¥ Ô‡ÓÈ„ Ô‡ (93 58,1) ‘for his grace is forever’ 14

See also Hary 2009, the edition of ms. 1302. In another ¡ar˙ of the book of Esther, printed in Livorno in 1868 and used by Jews from Cairo, Alexandria, Damascus, and Aleppo, the translator used /fi/ in this place. 15 For the use of Ô‡ as the conditional particle and the relative pronoun, see 5-3 and 6-3.9 respectively in this chapter.

231

CHAPTER EIGHT renders Â„ÒÁ ÌÏÂÚÏ ÈÎ; ·ÈÈË Ô‡ ‰‡Ï‡Ï Â¯ÂÎ˘‡ (93 63,2) ‘praise God, for He is good’ translates ·ÂË ÈÎ ß‰Ï Â„Â‰; and ‰Ï‡ È¥È ·Á Ô‡ ·Â˜ÚÈ (3 23,6) ‘for he wished to annihilate Jacob’ renders ÏÎ‡ ÈÎ ·SÚÈ ˙‡. (5-2.2) At times, the Judeo-Arabic particle Ô‡ translates Hebrew ≠˘ with the meaning of ‘for, that’ in the Haggadah: ÚÈÓ‚ È¥ Ô‡ ÈÏ‡ÈÏ Ï‡ (3 2,10) ‘that on all other nights’ renders ˙ÂÏÈÏ‰ ÏÎ·˘; ‰Â¥ÈÏ ‰ÈÏÚ ¥˜Â Ë˜¥ „Á‡Â ÌÏ Ô‡ (3 7,2–3) ‘For it was not one man only who rose against us to destroy us’ translates „Á‡ ‡Ï˘ Â˙ÂÏÎÏ ÂÈÏÚ „ÓÚ „·Ï·; and ÌÎÁ ÌÏ ‰Ú¯Ù Ô‡ (3 7,8) ‘that Pharaoh did not decree’ translates ¯Ê‚ ‡Ï ‰Ú¯Ù˘. Furthermore, biblical Hebrew ÈÎ with the meaning of ‘for, because’ may be also translated into Judeo-Arabic Ô‡. For example, ˙˜Â Ï‡ ‡≤‡‰ È¥ È˙ÎÒ˙ ‡≤‡ Ô‡ (1302 4a,7) ‘for if you remain silent at this time’ translates ˙‡Ê‰ ˙Ú· È˘È¯Á˙ ˘¯Á‰ Ì‡ ÈÎ (Esth 4:14), and Ìµ ÏÏ ‰Ú¯Ó ÒÈÏ Ô‡ (3 7,17) ‘because there is no pasture for the flocks’ translates Ô‡ˆÏ ‰Ú¯Ó ÔÈ‡ ÈÎ. (5-2.3) A more literal rendering, reflecting a calque translation (chapter 6, pp. 178–81), can be seen in È≤Ï‡Î (15 3a,4; 4a,9; 7a,9; and many other places) ‘when.’ This conjunction is a verbatim translation of the Hebrew ¯˘‡Î ‘when, where, as’ (Gen 24:22 and other places), where the two morphemes of the conjunction are translated literally: the Hebrew preposition ≠Î ‘as’ is translated by the Judeo-Arabic preposition /ka-/, and the Hebrew relative pronoun ¯˘‡ is translated by Judeo-Arabic /allaƒ•/ and is attached to the preposition, resulting in È≤Ï‡Î /ka-allaƒ•/. A similar example appears in Esther, where ‡ÓÎ È≤Ï‡ (1302 2b,22; 5a,8) also translates ¯˘‡Î (Esth 2:20; 6:10), with the same analysis, but in a more analytic structure. (5-2.4) The ¡ar˙an chose Judeo-Arabic ÔÈÁ ‘then’ in È¯·˙ ÔÈÁ È‡¥ÏÁ ÔÓ (15 3b,10) ‘then (you) will be absolved from my oath’ to translate verbatim the Hebrew Ê‡ in È˙Ï‡Ó ‰˜!z Ê‡ (Gen 24:41) without paying any attention to the form of the verb 232

THE MORPHOSYNTACTIC LEVEL: COORDINATING PARTICLES that follows the particle. The same applies to ˙˙¯¥ ÔÈÁ (15 36b,5) ‘(you) have desecrated,’ which translates ˙ÏÏœÁ Ê‡ (Gen 49:4). Int.: (5-2.5) The Hebrew particle ≠˘ /¡E/ in ÌÈÁ‹Ó ¯Â¯ÓÂ ‰ˆÓ ˘È˘ ‰Ú˘· ÍÈÙÏ is omitted in the translation in favor of an interpretive and regular Judeo-Arabic style, ¯¯ÂÓ Ï‡Â ¯ÈË¥ Ï‡ ˙Ú‡Ò È¥ ¬Ó‡„Â˜Ï ÔÈËÂËÁÓ (3 5,14–15) ‘at the time when the unleavened bread and the bitter herbs are placed before you.’1 6 L/I: (5-2.6) The Judeo-Arabic coordinating conjunction /wa-/ is missing before the adjective />a`•m/ ¯È·ÎÂ ÌÈ∂Ú ¯È·Î ·Ú˘Ï (3 7,11–12) ‘a great, mighty, and populous nation’ in a literal translation, as it is lacking also in the Hebrew ·¯Â ÌÂˆÚ ÏÂ„‚ ÈÂ‚Ï. This is as opposed to regular Judeo-Arabic style, where the adjectives are all connected with the coordinating conjunction /wa-/. However, in a different place in the Haggadah, the coordinating conjunction is nonetheless added in the ¡ar˙, despite the fact that it is lacking in the Hebrew text in an interpretive translation: ÌÈ∂ÚÂ ¯È·Î ·Ú˘Ï (3 8,7–8) ‘a great and mighty nation,’ which translates ÌÂˆÚ ÏÂ„‚ ÈÂ‚Ï.1 7 (5-2.7) The sentence ÂÏ ÏÈ‡Á˙ ‰Ï‡Ú˙ (3 9,3) ‘Come, let us deal wisely with him’ follows verbatim the Hebrew ÂÏ ‰ÓÎÁ˙ ‰·‰ as an asyndetic sentence. In other manuscripts, though, the JudeoArabic version is syndetic, yielding a more interpretive translation, ÂÏ ÏÈÈ‡Á˙È ‰˙Á ‰Ï‡Ú˙ (91 5a,1–2). (5-2.8) As in 5-2.2 the ¡ar˙an translated verbatim Hebrew ≠˘ ‘that, for’ in ‡Â‰ ÍÂ¯· ˘Â„˜‰˘ into Judeo-Arabic Ô‡ in ÒÂ„Â˜ Ï‡ Ô‡ ‡Â‰ ¬¯‡·Â˙ (93 20,1–2) ‘For the Holy One, blessed be He.’ In another manuscript, the ¡ar˙an did not do so, and the sentence lacks a translation of the Hebrew ≠˘: ‡Â‰ ¬¯‡·Â˙ ÒÂ„Â˜ Ï‡ (3 16 17

See example 6-3.12 below for a different analysis. Note that there are variant readings of this sentence in the Haggadah, following biblical ÌÂˆÚ ÏÂ„‚ ÈÂ‚Ï (Deut 26:5).

233

CHAPTER EIGHT 6,13) ‘The Holy One, blessed be He,’ which is more interpretive. Conditional Particles Feature 5-3 — Conditional particles Int.: The Hebrew conditional particle Ì‡ is frequently translated by regular Judeo-Arabic /in/ and usually is followed by the perfect /kån/ of the conditional verbal structure. (5-3.1) In Genesis: Èƒ¯ ‰‡ÚÓ Úˆ˙ Ô‡Î Ô‡ (15 7a,9–10) ‘if you do harm to us’ translates ‰Ú¯ ÂÓÚ ‰˘Ú˙ Ì‡ (Gen 26:29), and Ô‡Â ‰‡ ‰˙ÈÈÓ ÒÈÏ Ô‡Î (15 11b,3) ‘and if not, let me die’ renders Ì‡Â ÈÎ‡ ‰˙Ó ÔÈ‡ (Gen 30:1). (5-3.2) In Esther: ¬ÏÓ Ï‡ ˙¯ƒÁ È¥ ÔÒÁÈ Ô‡Î Ô‡ (1302 2a,4) ‘if it pleases the king’ translates ·ÂË ÍÏÓ‰ ÏÚ Ì‡ (Esth 1:19); Ô‡Î Ô‡ ‰ËÏÒ ÏÏ È˙ÏˆÁ ‡≤‰Î ˙˜ÂÏ (1302 4a,8) ‘if you have become a queen for a time such as this’ renders ˙ÂÎÏÓÏ ˙Ú‚‰ ˙‡ÊÎ ˙ÚÏ Ì‡ (Esth 4:14); and È¡„¯Ó „Â‰È Ï‡ ÏÒ ÔÓ Ô‡Î Ô‡ (1302 5a,13) ‘If Mordecai is of Jewish descent’ renders ÈÎ„¯Ó ÌÈ„Â‰È‰ Ú¯ÊÓ Ì‡ (Esth 6:13). (5-3.3) In the Haggadah: /in kån/ in the conditional meaning appears only once, and not in the language of the translation itself but rather in the directions assigned for conducting the seder: ˙·˘ Ï‡ ‚Â¯¡ È¥ „ÈÚ Ï‡ ÔÂÎÈ Ô‡Î Ô‡Â (93 10,4) ‘and if the holiday falls on Shabbat.’ Judeo-Arabic /in/ and /kån/ appear together regularly, and they are often combined into one word orthographically, Ô‡Î‡. (5-3.4) In Genesis: ÁÂ¯˙Ï ‰¯Ó Ï‡ È‚˙ ÌÏ Ô‡Î‡Â (15 2b,3) ‘if the girl does not wish to go’ translates ˙ÎÏÏ ‰˘‡‰ ‰·‡˙ ‡Ï Ì‡Â (Gen 24:8); È‡¥ÏÁ ÔÓ È¯·Ó ÔÂÎ˙Â ¬Ï ÂËÚÈ ÌÏ Ô‡Î‡Â (15 3b,11) ‘and if they do not give you [the girl], you will be released from my adjuration’ translates È˙Ï‡Ó È˜ ˙ÈÈ‰Â ÍÏ Â˙È ‡Ï Ì‡Â (Gen 24:41); and ‰ÈÏ Â¯·¡ È„ÈÒ ‰ÚÓ ˜ÁÂ Ïƒ¥ ÔÈÚ‡ˆ ÌÂ˙‡ Ô‡Î‡ (15 4a,4–5) ‘if you do what is kind and right to my master, inform me’ translates

234

THE MORPHOSYNTACTIC LEVEL: CONDITIONAL PARTICLES ÈÏ Â„È‚‰ ß‰ ˙‡ ˙Ó‡Â „ÒÁ ÌÈ˘›Ú ÌÎ˘È Ì‡ (Gen 24:49). (5-3.5) In Esther: ¬ÏÓ Ï‡ ‡‰È· Èƒ‡¯ Ô‡Î‡ Ô‡ (1302 2b,13–14) ‘unless the king desires her’ renders ÍÏÓ‰ ‰· ıÙÁ Ì‡ ÈÎ (Esth 2:14); È‚‰ Ï‡˜ È≤Ï‡ Ô‡Î‡ Ô‡ (1302 2b,15) ‘but rather what Hegai said’ translates È‚‰ ¯Ó‡È ¯˘‡ ˙‡ Ì‡ ÈÎ (Esth 2:15); and ¬ÏÓ Ï‡ ˙¯ƒÁ È¥ ÔÒÁÈ Ô‡Î‡Â (1302 4b,2) ‘and if it pleases the king’ translates ÍÏÓ‰ ÈÈÚ· ÔÁ È˙‡ˆÓ Ì‡ (Esth 5:8). (5-3.6) As seen above in 5-3.3, in the Haggadah Ô‡Î‡ /in kån/ does not appear in the language of the translation itself, but rather it appears several times in the directions assigned for conducting the seder: ‰‰ ÔÓ ˘Â„˜ØÒÂ„˜ Ï‡ Â„·È ÏÂÁ ÔÂÎÈ Ô‡Î‡ (93 8,12; 9,4) ‘if (Passover) falls on a weekday, begin the kiddush from here,’ and ‰‰ ÔÓ ÒÂ„˜ Ï‡ Â„·È ˙·˘ Ï‡ È¥ ÔÂÎÈ Ô‡Î‡ (93 8,15) ‘if (Passover) falls on Shabbat, begin the kiddush from here.’ The literary conditional particle ‡≤‡ is quite rare in the texts of the ¡ur¥˙: (5-3.7) In Esther: ˙˜Â Ï‡ ‡≤‡‰ È¥ È˙ÎÒ˙ ‡≤‡ Ô‡ (1302 4a,7) ‘if you keep silent at this time’ translates ˙Ú· È˘È¯Á˙ ˘¯Á‰ Ì‡ ÈÎ ˙‡Ê‰ (Esth 4:14). (5-3.8) In the Haggadah ‡≤‡ appears in directions given for administering the seder, as in 5-3.3 and 5-3.6: ·È¯ÈÚ ÌÂÎÁÂÈ ‡≤‡ ˙·˘· ÁÒÙ (93 2,9) ‘if the evening before Passover occurs on Shabbat.’ The Hebrew conditional particle ÂÏ‡ or ÂÏ, and ‡ÏÂÏ in the negative, can be translated with either ®Ô‡Î© ÂÏ or Ô‡Î Ô‡, and ‰ÏÂÂÏ/‡ÏÂÂÏ in the negative, as in the following examples. (5-3.9) In Genesis: in ¥ÒÂÈ ‰Ï È·¡È ÂÏ ÂÏ‡˜Â (15 38a,11) ‘And they said, “What if Joseph still bears a grudge against us,” ’ ÂÏ /law/ translates Hebrew ÂÏ in Gen 50:15. Furthermore, in both ‰ÈÏ Ô‡Î ˜ÁˆÈ ÚÊ¥Â Ì‰¯·‡ ‰‡Ï‡ ‡ÈÂ·‡ ‰‡Ï‡ ‡ÏÂÂÏ (15 14b,12–13) ‘Had not the God of my father, the God of Abraham and the 235

CHAPTER EIGHT fear of Isaac, been with me ...’ (translating Gen 31:42) and Ô‡ ‰Ï‚Ú˙Ò‡ ‰ÏÂÂÏ (15 29a,8) ‘if we had not hastened’ (rendering Gen 43:10),1 8 ‰ÏÂÂÏ/‡ÏÂÂÏ /lawla/ translates the Hebrew negative conditional particles ‡ÏÂÏ and ÈÏÂÏ respectively. (5.3-10) In Esther: in ‡ÈÚ·‡ ¯‡Â‚ÏÂ „È·ÚÏ Ô‡Î Ô‡Â (1302 5b,1–2) ‘had we (only) been sold as slaves and mistresses’ Ô‡Î Ô‡Â translates Hebrew ÂÏ‡Â in Esth 7:4. (5.3-11) In the Haggadah: in ‰‚¯¡‡ ÌÏ Ô‡Î ÂÏÂ (3 3,1–2) ‘and if he had not brought us out’ Ô‡Î ÂÏÂ translates Hebrew ÂÏ‡Â. Similarly, /wa-lawkån/ in ¬‡‰ Ô‡Î Ô‡ÎÂÏÂ (3 5,2) ‘and if he had been there’ renders Hebrew ÂÏ‡Â. In the same way, the section of dayyenu is inundated with Ô‡Î ÂÏ translating Hebrew ÂÏ‡, such as in ÌÂ‰˙‡‰‡Ï‡ È¥ Úˆ Ô‡Î ÂÏ (3 16,3) ‘if he had executed judgment upon their gods.’ (5.3-12) The following example features a translation of Hebrew ÂÏ, not in its conditional use but in its “wish” meaning: Ï‡˜Â ¬Ó‡ÏÎÎ ÔÂÎÈ ˙È¯‡È Ô‡·Ï (15 12b,9) ‘And Laban said, “May it be according to your words,” ’ translating Í¯·„Î È‰È ÂÏ Ô‰ Ô·Ï ¯Ó‡ÈÂ (Gen 30:34). Judeo-Arabic Ô‡Î Ô‡ also translates Hebrew Ì‡ and other particles in their nonconditional meanings. (5.3-13) For example, in Genesis, Ô‡Î Ô‡ translates the emphatic meaning of Hebrew Ì‡ in ‰‡ ‰„ ˘ÈÏ Ô‡Î Ô‡ (15 5b,4–5) ‘why is this, then, happening to me?,’ whereas Ô‡Î Ô‡ translates ÔÎ Ì‡ ‘then’ in ÈÎ&‡ ‰Ê ‰ÓÏ ÔÎ Ì‡ (Gen 25:22). 1 9 Furthermore, Ô‡Î Ô‡ renders the comparative meaning of Ì‡ in Ô‡ Â‡ÒÚ È·‡ ‰„ ‰˙‡ ÌÏ Ô‡Î (15 8a,8) ‘Are you my son Esau or not?,’ translating Hebrew Ì‡ ‘or’ in ‡Ï Ì‡ Â˘Ú È· ‰Ê ‰˙‡‰ (Gen 27:21). In another 18

Notice that the verb used in the Hebrew text is Â‰Ó‰Ó˙‰ ‘we dawdled,’ the opposite of /ista>galna/ ‘we hastened.’ 19 In this example it is also possible that the ¡ar˙an translated Ì‡ into Ô‡ and ÔÎ into Ô‡Î, following the orthographic and phonetic similarity.

236

THE MORPHOSYNTACTIC LEVEL: INDEPENDENT PRESONAL PROUNOUNS case, Ô‡Î Ô‡ in ¬ÚÓ ‰ÏÏ‡ Ô‡Î Ô‡ ‰¯∂ ¯È∂ (15 7a,7) ‘we have indeed seen that God was with you’ simply translates Hebrew ‰È‰ ÈÎ in ÍÓÚ ß‰ ‰È‰ ÈÎ ÂÈ‡¯ Â‡¯ (Gen 26:28). Yet in another case, Ô‡Î Ô‡ is employed to render ‘so that’ or ‘until’ in Ô‡ÓÎ ˙Ï‡˜Â Â·¯˘ÈÏ ÂÓ˙ Ô‡Î‡ „Ú È˜Ò‡ ¬Ï‡Ó‚Ï (15 3a,1) ‘and she said, “Let me draw water for your camels, so that they finish drinking,” ’ translating Ì‡ „Ú ‘so that’ in ˙Â˙˘Ï ÂÏÎ Ì‡ „Ú ·‡˘‡ ÍÈÏÓ‚Ï Ì‚ ¯Ó‡˙Â (Gen 24:19).2 0 Independent Personal Pronouns Level: Morphosyntactic; Category 6: Pronouns Feature 6-1 — Independent personal pronouns Int.: It is common to find in the Egyptian ¡ur¥˙ regular Judeo-Arabic translations of the independent personal pronouns, reflecting interpretive translations.2 1 (6-1.1) Both ‰˙‡ (15 3b,16; 3 4,14; 74 2,11; and many more occurrences) and ‡˙‡ (1302 3a,8; 91 2b,3; 93 16,12) ‘you (masc. sg.)’ reflect Judeo-Arabic pronunciation of /inta/ in an interpretive translation (6-1.2) Both È˙‡ (15 3a,6; 4a,1; 4a,20; 1302 4a,6; 3 8,15; 91 4b,16; 93 23,1) and ˙‡ (1302 4a,8) ‘you (fem. sg.)’ reflect Judeo-Arabic pronunciation of /inti/, demonstrating interpretive translation. (6-1.3) The independent pronouns ‰Á‡ (15 10a,16; 3 2,12; 3,3; 74 1,6; 10,9; 91 8b,1) and ÔÁ (93 13,2; 14,4) ‘we’ are interpretive translations into colloquial and literary JudeoArabic respectively. (6-1.4) As in the previous example, standard Arabic ÌÂ˙‡ (15 4a,4 and many more occurrences; 3 24,5; 93 62,15) and Ì˙‡ 20

For a different use of /inkån/ see the section on interrogative pronouns, p. 248, example 6-5.2. 21 For an in-depth analysis of the orthography of independent personal pronouns, see chapter 9, pp. 318–19, 13-6.7.

237

CHAPTER EIGHT (1302 6a,5) as well as the colloquial Â˙‡ (74 14,1; 91 10b,19; 93 65,11) ‘you (masc. pl.)’ are typical interpretive translations. (6-1.5) ÌÂ‰ (only in 15 5a,11) and ÔÓÂ‰ (15 18a,10; 18a,15; and more) ‘they’ are also interpretive translations, reflecting standard and colloquial Egyptian Judeo-Arabic use respectively. Pronominal Suffixes Feature 6-2 — Pronominal suffixes Lit.: (6-2.1) The pronominal suffix /-o/ in ÂÒÈÏ (15 0-1,8 and other places; 3 5,8) ‘he is not’ is a clear verbatim translation of the Hebrew suffix Â≠ in ÂÈ‡. The uninflected standard /laysa/ would have been sufficient according to Arabic usage, but in order to render literally the Hebrew, which includes the pronominal suffix, the ¡ar˙an added the equivalent pronominal suffix to the Judeo-Arabic /laysa/, resulting in /layso/ (see pp. 213–14). (6-2.2) The ¡ar˙an has frequently used the uncommon JudeoArabic /iyyå-/ followed by the appropriate pronominal suffixes in order to render verbatim the Hebrew definite direct object ˙‡ /et/ followed by its pronoun suffixes, although he could have employed the pronominal suffixes following the verbs in a more synthetic and more regular Judeo-Arabic style. 2 2 In Genesis the ¡ar˙an translated verbatim È˙›‡ Â¯Á‡˙ Ï‡ (Gen 24:56) into ‰È‡È‡ Â¯¡‡˙ ÌÏ (15 4a,15) ‘do not delay me,’ where ‰È‡È‡ /iyyåya/ renders È˙›‡. Similarly, ‰È‡È‡ ÌÂ˙‰¯Î ÌÂ˙‡Â (15 7a,6) ‘you hate me’ translates verbatim È˙›‡ Ì˙‡˘ Ì˙‡Â (Gen 26:27), where ‰È‡È‡ /iyyåya/ renders È˙›‡; and ‰ÈÏ‡ ‡‰‡È‡ ·‡‚Â (15 11a,3) ‘and he brought her to him’ translates verbatim ‡·ÈÂ ÂÈÏ‡ ‰˙›‡ (Gen 26:27), where ‡‰‡È‡ /iyyåha/ renders ‰˙›‡. The same occurs in the Haggadah. In the example ‰‡ÈÈ‡ ˙È˘ÓÂ (93 19,7–8) ‘and I led him,’ the ¡ar˙an used Judeo-Arabic 22

An even rarer translation of ˙‡ plus a pronominal suffix is the preposition /ilå/ plus a pronominal suffix. See below in this chapter, p. 261, 8-1.12.

238

THE MORPHOSYNTACTIC LEVEL: PRONOMINAL SUFFIXES ‰‡ÈÈ‡ /iyyåh/ ‘him’ in order to translate verbatim the Hebrew Â˙Â‡ ÍÈÏÂ‡Â, although he could have employed the pronominal suffix /-o/ in /wa-ma¡¡eto/ in a more regular Judeo-Arabic style. Similarly, the ¡ar˙an’s choice of ‰‡ÈÈ‡ /iyyåh/ ‘him’ in ˙¯ÂÂÈÏ ‰‡ÈÈ‡ (3 6,9) ‘to inherit it’ reflects his intention to produce a literal translation. In the same way, the example ÔÈ˜È‡ƒÓ ÌÂ‰‡ÈÈ‡ (3 11,3) ‘oppress them’ demonstrates how the ¡ar˙an preferred to give a literal translation of the Hebrew original Ì˙Â‡ ÌÈˆÁÂÏ, avoiding the more common Judeo-Arabic pronominal suffix /-hum/. (6-2.3) On the other hand, when the Hebrew uses the third plural pronominal suffix (rather than /et/ followed by the pronoun suffix) as in ÌÂ„·ÚÂ, the ¡ar˙an imitated it in yet another verbatim translation of the Hebrew: ÌÂ‰ÂÓ„¡˙ÒÈÂ (91 3a,9) or ¿¨™¨º®í∂¨ (3 6,17) ‘and they served them.’ In this case, the ¡ar˙an’s choice does coincide with regular Judeo-Arabic use. Int.: (6-2.4) In order to achieve a standard Arabic style, the ¡ar˙an did not render the Hebrew pronoun suffix Í≠ in the Hebrew phrase ‡ Í˘È Ì‡ ‘if indeed you …’ (Gen 24:42), resulting in the more interpretive Ô‡Ï‡ „Â‚ÂÂÓ Ô‡Î‡ (15 3b,12) ‘if available now,’ without the pronoun /-ak/ attached to /mawg¥d/. Likewise, the ¡ar˙an did not translate the Hebrew pronoun suffix ÌÎ≠ in Ì‡ ÌÈ◊›Ú ÌÎ˘È ‘if you do’ (Gen 24:49), and in fact, he did not translate ÌÎ˘È at all, but rather interpreted it to be the independent pronoun ÌÂ˙‡ ‘you (masc. pl.)’ in ÔÈÚ‡ˆ ÌÂ˙‡ Ô‡Î‡ (15 4a,4–5) ‘if you do.’ (6-2.5) Although the feminine singular pronominal suffix does not occur in the Hebrew ÌÈÈ„˘ ‘breasts,’ it does appear in the ¡ar˙ in an interpretive translation, ¬Ê‡Ê· (3 8,14) ‘your breasts.’ (6-2.6) In an instance of self-identification with the protagonists in the text, the ¡ar˙an translated ÌÈÒœÓ È¯◊ ÂÈÏÚ ÂÓÈ◊ÈÂ ‘and they set taskmasters on him (the Israelites)’ into ˙‡ÂÂ˘· ‰ÈÏÚ ÂÏÚ‚Â

239

CHAPTER EIGHT ÒÎÓ (3 9,7–8) ‘and they set taskmasters over us.’ Consequently, the Hebrew third masculine singular pronominal suffix is not translated verbatim, but rather by the first plural /-na/. This occurs in all the various manuscripts of the Haggadah; however, interestingly enough, it does not take place in the Judeo-Arabic ¡ar˙ of Exodus, from which this verse is taken, and the translation there is more literal: ‰ÈÒ‡˜ ˙‡Â˘· ‰ÈÏÚ ÂÏÚ‚Â (15 39b,11–12) ‘and they appointed cruel masters on him (the Israelites),’ following ÌÈÒœÓ È¯◊ ÂÈÏÚ ÂÓÈ◊ÈÂ in Exod 1:11. (6-2.7) A similar, yet not identical, situation occurs in the translation of Gen 26:7. The Hebrew text uses the third person verbal pronoun in È˙˘‡ ¯Ó‡Ï ‡¯È ÈÎ ‡È‰ È˙›Á‡ ¯Ó‡ÈÂ ‘and he said, “She is my sister,” as he was afraid to say, “my wife” ’ (emphasis is mine). The Judeo-Arabic ¡ar˙ uses the first person verbal pronoun /xuft aq¥l/ in È˙‡¯Ó ÏÂ˜‡ ˙¥Â¡ Ô‡ ‡È‰ È˙¡Â‡ Ï‡˜Â (15 6a,14) ‘and he said, “She is my sister,” as I was afraid to say, “my wife” ’ (emphasis is mine). L/I: (6-2.8) The ¡ar˙an was consistent in both the literal translation of Hebrew ˙‡ /et/ followed by pronominal suffixes into /iyyå-/ with pronominal suffixes, as seen in 6-2.2, and the literal translation of pronominal suffixes following verbs, as seen in 6-2.3. Thus, indeed, the Hebrew Ì˙›‡ ÂÚÂ is translated verbatim as ÌÂ‰‡ÈÈ‡ Â¥ÚÈÂ (3 6,17; 93 20,5) ‘and they tortured them,’ following the examples in 6-2.2. However, as an exception, in another manuscript, the ¡ar˙an used the pronominal suffix /-hum/ in an interpretive translation: ÌÂ‰Â¥ÚÈÂ (91 3a,10) ‘and they oppressed them,’ as is usually the case in Arabic structure. Relative Pronouns Feature 6-3 — Relative pronouns Lit.: The relative pronoun, usually È≤Ï‡ /allaƒ•/ (masc. sg.), is used throughout the ¡ur¥˙ to modify all nouns, regardless of their

240

THE MORPHOSYNTACTIC LEVEL: RELATIVE PROUNOUNS number and gender. This is done to follow the Hebrew ¯˘‡, which does not change according to number or gender, unlike its Classical Arabic equivalent. Furthermore, È≤Ï‡ is a “frozen” form of the Later Egyptian Literary Judeo-Arabic relative pronoun.2 3 This is evident in many instances in the ¡ur¥˙. (6-3.1) In Genesis: ÌÏÎ˙‡ È≤Ï‡ ‰ƒ¥ Ï‡ (15 2a,3–4) ‘the money about which he had spoken,’ where /allaƒ•/ modifies the feminine noun /fa∂∂a/. In both Â‰ÚÓ È≤Ï‡ ‰Ï‡‚¯ Ï‡ ÔÈÏ‚¯Â (15 3a,20) ‘and the feet of the men who are with him’ and ÌÈ˘µÏÙ Ï‡ „‡ÏÂÂ‡ÏÂ Ì‰¯·‡Ï È≤Ï‡ (15 4b,17) ‘and to the sons of Abraham’s concubines,’ /allaƒ•/ modifies plural nouns: /rigåla/ and /pila\¡im/. (6-3.2) In Esther: In ¬ÏÓ Ï‡ Ô‡„Ó ÏÎ È¥ È≤Ï‡ ÌÓÂ‡ Ï‡ ÏÎ ‡ÏÚÂ (1302 2a,1) ‘against all the nations which are in all the provinces of the king,’ È≤Ï‡ modifies an inanimate plural. In ‰ÈÈ·ˆ Ï‡Â ¬ÏÓ Ï‡ ÔÈÈÚ È¥ ÔÒÁ˙ È≤Ï‡ (1302 2a,15–16) ‘and the girl who pleases the king,’ /allaƒ•/ modifies a feminine noun. Similarly, /allaƒ•/ in ‰È„ÓÂ ‰È„Ó ÏÎ È¥ È≤Ï‡ ˙‡Â˘· ÏÏÂ (1302 3b,2) ‘and to the Pashas who are in every town’ modifies the plural /ba¡awåt/ ‘Pashas.’ (6-3.3) In the Haggadah as well, “frozen” È≤Ï‡ /allaƒ•/ modifies feminine nouns, as in ÔÈ˜È‡ƒÓ ¯ˆÓ È≤Ï‡ ‰˜Èƒ Ï‡ ‰Ï‡ ˙¯∂Â ÌÂ‰‡ÈÈ‡ (3 11,3–4) ‘I have seen the oppression wherewith the Egyptians oppress them’; or plural inanimate nouns, as in ˘‡ È≤Ï‡ Ì‡ÎÁ‡ Ï‡Â ÌÂÒ¯ Ï‡Â „‡‰Â˘ Ï‡ (3 4,7–8) ‘what are the testimonies, statutes, and judgments which …’; and ¯˘Ú ÈÏÂ„ ÒÂ„Â˜ Ï‡ ·‡‚ È≤Ï‡ ˙‡·¯ƒ (3 13,14) ‘these are the ten plagues which the Holy One has brought.’ Sometimes, the position of the relative pronoun is innovative and does not follow regular Judeo-Arabic structure; however, it strictly follows the position of its Hebrew equivalent. 23

See Hary 1992:308–9. In Classical Judeo-Arabic /allaƒ•/ is not always the preferred relative pronoun. See Blau 1980:235–37.

241

CHAPTER EIGHT (6-3.4) The position of /allaƒ•/ in È≤Ï‡ ‰‚¯‡¡ ‰„ÂÂ‰ ‰˜·¯Â ˙„ÏÂÂ˙‡ (15 2b,15) ‘and then Rebecca, who was born … , came out’ is irregular according to Arabic word order, but it follows verbatim the Hebrew original ‰„Ï'È ¯˘‡ ˙‡ˆ&È ‰˜·¯ ‰‰Â (Gen 24:15),2 4 following the literal translation method. È≤Ï‡ /allaƒ•/ (masc.) is also used to translate Hebrew ≠˘ and ¯˘‡ even in their complementizer meaning. Consequently, /allaƒ•/ gains the complementizer function in the texts of the ¡ur¥˙ in Later Egyptian Judeo-Arabic, in addition to its function of relative pronoun. (6-3.5) In Genesis: ˙˜‡ÓÁ Ú‚¯˙ È≤Ï‡ „Ú ÏÈÏ˜ Ì‡È‡ ÂÚÓ „ÂÚ˜˙Â ¬Â¡‡ (15 9a,4) ‘and (you) stay with him a few days until your brother’s anger has subsided,’ in which /allaƒ•/ translates the complementizer Hebrew ¯˘‡ (Gen 27:44). 2 5 Similarly, in ÌÏ ˙Úˆ Ô‡Î Ô‡ È≤Ï‡ „Ú ¬ÎÂ¯˙‡ (15 9b,17) ‘I will not leave you until I do … ,’ Judeo-Arabic È≤Ï‡ /allaƒ•/ translates the Hebrew complementizer ¯˘‡ in Gen 28:15. Although in the Hebrew text neither ≠˘ or ¯˘‡ appears in ÚÈ„Â‰ È¯Á‡ ÛÒÂÈ Ï‡ ‰Ú¯Ù ¯Ó‡ÈÂ Í˙Â‡ ÌÈ‰ÂÏ‡ (Gen 41:39), the complementizer meaning exists in ¬‡È‡ ·¯ Ï‡ ¥¯Ú È≤Ï‡ „Ú· ¥ÒÂÈ ‰Ï‡ ÔÂÚ¯¥ Ï‡˜Â (15 26b,1–2) ‘Then Pharaoh said to Joseph, “After God has informed you.” ’ Indeed, the Protestant translation rendered the sentence with the complementizer /må/ in tÒK"« pLKŽ« U# bFÐ ‘after God has informed you.’ (6-3.6) In the book of Esther: in ˙ÏÓÚ ÌÏ È≤Ï‡ ‡ÏÚ (1302 1b,22) ‘that (she) did not do’ Judeo-Arabic È≤Ï‡ translates the Hebrew complementizer ¯˘‡ in ‰˙˘È ‡Ï ¯˘‡ ÏÚ (Esth 1:15). Similarly, in È˙˘Â È‚˙ ÌÏ È≤Ï‡ ¯ÈÈµ˙È ÌÏÂ (1302 2a,5) ‘(it) will 24

Although notice here that the ¡ar˙an did not follow verbatim the word order of ‰˜·¯ ‰‰Â, as he translated ‰„ÂÂ‰ ‰˜·¯Â. 25 See below, p. 281 n. 51 to section 10-3.2, a reference to the irregular agreement in ÏÈÏ˜ Ì‡È‡ ‘a few days.’

242

THE MORPHOSYNTACTIC LEVEL: RELATIVE PROUNOUNS not be changed [revoked], stating that Vashti may not come,’ Judeo-Arabic È≤Ï‡ is employed for the complementizer ¯˘‡ in Esth 1:19. Likewise, in È„Â‰È ‡ÂÂ‰ È≤Ï‡ Ì‰Ï ¯·¡ Ô‡ (1302 3a,9) ‘that he informed them that he was Jewish,’ /allaƒ•/ is used to translate the complementizer ¯˘‡ in Esth 3:4. The Protestant translation used the complementizer /anna/ in its translation ÍœuN¹ tÒ½UÐ r¼d?³š« tÒ½ô ‘because he informed them that he is Jewish.’ (6-3.7) In the Haggadah as well, Judeo-Arabic /allaƒ•/ translates the Hebrew complementizer ≠˘ in the following examples: Hebrew „¯È ‡Ï˘ „ÓÏÓ ‘teaching that he did not go down’ is translated as ÏÊ ÌÏ È≤Ï‡ ÌÏÚ˙‡ (3 7,15). Similarly, Hebrew Ï‡¯˘È ÂÈ‰˘ „ÓÏÓ ‘teaching that Israel was’ is translated as ÌÏÚ˙‡ Ï‡¯˘È Â‡Î È≤Ï‡ (3 8,6–7). Also Judeo-Arabic È≤Ï‡ ÏÈ‡˜ ‰˙‡ ÔÈÈ¯ˆÓ Ï‡ Â·¯ƒ‡ (3 14,5–6) ‘you say that the Egyptians were smitten’ translates Hebrew ÌÈÈ¯ˆÓ‰ Â˜Ï˘ ¯ÓÂ‡ ‰˙‡. In some cases, È≤Ï‡ /allaƒ•/ (masc. sg.) translates verbatim the original Hebrew ¯˘‡ followed by the preposition ≠Ï indicating possession, and sometimes it is translated verbatim followed by the preposition /li-/: (6-3.8) The phrase ¯ˆÓ ¬ÏÓÏ È≤Ï‡ Ê‡·¡ Ï‡Â (15 24a,12–13) ‘and the baker of the king of Egypt’ follows literally the Hebrew in Gen 40:5 ÌÈ¯ˆÓ ÍÏÓÏ ¯˘‡ ‰Ù›‡‰Â. Similarly, È≤Ï‡ È¥ (15 23-1b,2) ‘since’ translates verbatim the Hebrew ¯˘‡· (Gen 39:9). Infrequently, Judeo-Arabic Ô‡ rather than È≤Ï‡ is used for the relative pronoun: (6-3.9) This is the case in two instances in Esther, where the ¡ar˙an translated Hebrew ¯˘‡ as Ô‡: ÂÒ‡¯ È¥ ‰ËÏÒ ‚‡˙ ÏÚ‚È Ô‡ (1302 5a,4–5) ‘whose royal crown is put on his head,’ rendering the relative pronoun ¯˘‡ in Â˘‡¯· ˙ÂÎÏÓ ¯˙Î Ô(z$ ¯˘‡Â (Esth 6:8); and ÂÓÎÁÈ „Â‰È Ï‡ Ô‡ (1302 6b,2)‘that the Jews will reign,’ also translating ¯˘‡ in ÌÈ„Â‰È‰ ÂËÏ˘È ¯˘‡ (Esth 9:1). 243

CHAPTER EIGHT Int.: (6-3.10) In an interpretive translation, the colloquial relative pronoun /illi/ also occurs in the texts of the ¡ur¥˙, translating Hebrew ≠˘ and ¯˘‡, as a manifestation of the dialect: ÈÏ‡ ‡ÈÈ‰ ‡˙‡‰·‡Ï ˙È¥˜ÂÂ (91 3a,13) ‘it is that (promise) which has stood by our fathers’ translates ÂÈ˙Â·‡Ï ‰„ÓÚ˘ ‡È‰. L/I: (6-3.11) The ¡ar˙an translated the Hebrew relative pronoun ≠‰2 6 literally as the Judeo-Arabic definite article Ï‡, even when the latter is not used as a relative pronoun in regular Arabic usage. He did so in order to achieve an exact verbatim translation: for example, ÈÎÁÈÏ ¯˙ÎÓ Ï‡ ÏÎÂ (3 3,6–7) ‘and all who tell at length,’ which translates ¯ÙÒÏ ‰·¯Ó‰ ÏÎÂ. In other manuscripts (74 1,14; 93 14,7), however, the interpretive mode prevails and the regular Arabic relative pronoun /allaƒ•/ appears: ÏÎÂ ÈÎÁÈÏ ¯˙ÎÈ È≤Ï‡ ‘and all who tell at length.’ (6-3.12) Sometimes the ¡ar˙an translated Hebrew ≠˘ using the Judeo-Arabic relative pronoun È≤Ï‡ just for the sake of a verbatim translation. For example, ÍÈÙÏ ÌÈÁ‹Ó ¯Â¯ÓÂ ‰ˆÓ˘ ‰Ú˘· was translated into ¬Ó‡„Â˜ ÔÈËÂËÁÓ ¯È¯ÂÓ Ï‡Â ¯ÈË¥ Ï‡ È≤Ï‡ ˙Ú‡Ò È¥ (93 18,7–8) ‘at the time when unleavened bread and bitter herbs are placed before you.’ In other manuscripts, however, regular Arabic use was preferred instead of a literal translation: ¯ÈË¥ Ï‡ ˙Ú‡Ò È¥ ¬Ó‡„Â˜Ï ÔÈËÂËÁÓ ¯¯ÂÓ Ï‡Â (3 5,14–15).2 7 Demonstrative Pronouns Feature 6-4 — Demonstrative pronouns In the ¡ar˙ the demonstrative pronouns /håda/ ‘this’ (masc.), /hådi/ ‘this’ (fem.), and /hadøli/ ‘these’ are usually translated literally in accordance with the Hebrew word order, where the demonstrative

26

The use of ≠‰ as a relative pronoun is much less frequent in Hebrew than its use as the definite article. 27 See above, p. 233, example 5-2.5, for a different analysis.

244

THE MORPHOSYNTACTIC LEVEL: DEMONSTRATIVE PROUNOUNS pronouns follow the noun they modify.2 8 This behavior is not consistent with literary Arabic use, where demonstrative pronouns precede the nouns they modify. However, this is in line with colloquial Arabic use, where demonstrative pronouns follow the nouns they modify, as is the case in the ¡ar˙. On the other hand, this is not the case with /ƒalika/ ‘that’ (masc.) and /tilka/ ‘that’ (fem.). In the latter cases, the translation is interpretive, because these pronouns precede the nouns they modify, as is the case in literary Arabic use, but not according to Hebrew word order. Lit.: The demonstrative pronoun /håda/ ‘this’ (masculine): (6-4.1) In Genesis: The placement of the demonstrative pronoun in ‰„‡‰ ÌÂÈ Ï‡ (15 35a,9) ‘this day’ follows verbatim the Hebrew demonstrative pronoun word order in ‰Ê‰ ÌÂÈ‰ (Gen 47:26), and not literary Arabic usage /håƒå l-yawm/. However, this translation does follow colloquial Egyptian Arabic word order, where the demonstrative pronoun follows the noun it modifies: /il-yøm da/. The same occurs in the translation of ‰Ê‰ ÌÂ˜Ó‰ (Gen 28:17) as ‰≤‡‰ Á¯ËÓ Ï‡ (15 9b,20) ‘this place’ and in ‰„‡‰ ˙È· Ï‡ (15 24b,8) ‘this house,’ which is the verbatim translation of ‰Ê‰ ˙È·‰ (Gen 40:14). (6-4.2) In the Haggadah: In the examples ‰≤‡‰ ¯ÈË¥ Ï‡ „ÈÚ ÌÂÈ (93 60,9) ‘this festival of unleavened bread’ and ˙·Ò Ï‡ ÌÂÈ È¥ ‰≤‡‰ (93 82,14) ‘on this Saturday,’ the Judeo-Arabic demonstrative pronoun ‰≤‡‰ follows the noun it modifies, as is the case in Hebrew in ‰Ê‰ ˙ÂˆÓ‰ ‚Á ÌÂÈ and ‰Ê‰ ˙·˘‰ ÌÂÈ· respectively. The demonstrative pronoun /hådi/ ‘this’ (fem.): (6-4.3) In Genesis: È„‡‰ √¯‡ Ï‡ (15 13b,5) ‘this land’ is a literal translation following the Hebrew demonstrative pronoun 28

This seems to work for Genesis and for the Haggadah. In Esther only the plural demonstrative pronoun /hadøli/ follows the noun it modifes, and even this feature may not be a literal translation, since the demonstrative pronoun is a dialectal form, thus following colloquial Judeo-Arabic use. See chapter 4, p. 125, 3.1.

245

CHAPTER EIGHT word order in ˙‡Ê‰ ı¯‡‰ (Gen 31:13). The same applies for Ï‡ È„‡‰ ‰˜Èƒ (15 28a,2) ‘this anguish,’ which translates ˙‡Ê‰ ‰¯ˆ‰ (Gen 42:21), and for È„‡‰ ‰·ËˆÓ Ï‡ (15 15a,9–10) ‘this pillar,’ following the Hebrew ˙‡Ê‰ ‰·ˆÓ‰ (Gen 31:52). (6-4.4) In the Haggadah: The phrase È„‡‰ ‰Ò Ï‡ (3 2,6) ‘this year’ translates verbatim the Aramaic demonstrative pronoun word order in ‡Î‰ ‡˙˘‰, not following regular Arabic use as in /håƒihi s-sana/ ‘this year.’ Other examples include È„‡‰ ‰ÏÈÏ Ï‡ (3 2,9) ‘this night,’ which translates verbatim ‰Ê‰ ‰ÏÈÏ‰; and Ï‡ È„‡‰ ‰ÈÂ„ (3 4,3), which translates literally ‰Ê‰ ÌÏÂÚ‰. The masculine and feminine demonstrative pronouns in Esther follow an interpretive translation (6-4.10), but the plural demonstrative pronoun /hadøli/ ‘these’ follows the Hebrew word order in all of the Egyptian ¡ur¥˙. The reason is that the plural pronoun itself is in colloquial Judeo-Arabic and thus follows colloquial word order, which also coincides with the Hebrew word order. (6-4.5) In Genesis: The examples ÈÏÂ„‰ Ì‡ÏÎ Ï‡Î (15 3a,13) 2 9 ‘like these words,’ translating ‰Ï‡ ÌÈ¯·„Î (Gen 24:28); Ï‡ ÏÂÎ ÈÏÂ„‰ Èƒ‡¯‡ (15 6a,8–9) ‘all these countries,’ which translates Ï‡‰ ˙›ˆ¯‡‰ ÏÎ (Gen 26:3); and ÈÏÂ„‰ ‰Ï‡‚Â¯ Ï‡Â (15 18a,10) ‘and these men,’ translating ‰Ï‡‰ ÌÈ˘‡‰ (Gen 34:21) demonstrate a word order similar to that in the Hebrew original and in colloquial Arabic use. (6-4.6) In Esther: The example ÈÏÂ„‰ Ì‡ÈÈ‡ Ï‡ (1302 1b,7) follows the Hebrew word order in ‰Ï‡‰ ÌÈÓÈ‰ (Esth 1:5), as in Judeo-Arabic colloquial use. (6-4.7) In the Haggadah: The word order of the demonstrative pronoun /(ha)døli/ ‘these’ in ÈÏÂ„ ˙‡ÓÏÎ ˙‡Ï˙ (3 17,18–19) ‘these three words’ and in ÈÏÂ„‰ ·È‡‚Ú Ï‡ ÏÎ (3 19,17–18) ‘all 29

See also pp. 277–78, 10-2.8 for an analysis of number agreement in this example.

246

THE MORPHOSYNTACTIC LEVEL: DEMONSTRATIVE PROUNOUNS these miracles’ follows verbatim the Hebrew ÂÏ‡ ÌÈ¯·„ ‰˘Ï˘ and ÂÏ‡‰ ÌÈÒ‰ ÏÎ respectively, and also follows Judeo-Arabic colloquial word order. Int.: The demonstrative pronoun /ƒalik/ ‘that’ (masc.): (6-4.8) In a marked difference from the examples above, ¬Ï‡≤ ÌÂÈ Ï‡ (3 5,12) ‘that day’ appears in literary Arabic word order. This is an interpretive translation, not following the Hebrew word order of the equivalent demonstrative pronoun ‡Â‰‰ ÌÂÈ·. The demonstrative pronoun /tilka/ ‘that’ (fem.): (6-4.9) The word order of /tilka/ is translated interpretively in the texts of the ¡ur¥˙, not following the equivalent Hebrew demonstrative pronoun word order. For example, Ï‡ ¬Ï˙ È¥ √¯‡ (15 6b,3) ‘in that land’ translating ‡È‰‰ ı¯‡· (Gen 26:12); È¥ ‰Ò Ï‡ ¬Ï˙ (15 6b,4) ‘in that year’ translating ‡È‰‰ ‰˘· (Gen 26:12); and ‰ÏÈÏ Ï‡ ¬Ï˙ (15 6b,20; 3 3,13–14) ‘that night’ are in regular Arabic structure, where the demonstrative pronoun precedes the modified noun, a departure from Hebrew word order. The ¡ar˙ of the book of Esther generally follows regular Arabic structure in an interpretive translation: (6-4.10) ÌÂÈ Ï‡ ‡≤‡‰Â (1302 2a,3) ‘and this day’; ‰Úˆ Ï‡ È≤‡‰ (1302 3a,18) ‘this craft’; ÌÂÈ Ï‡ ¬Ï‡≤ (1302 7a,2) ‘that day’; and ‰ÏÈÏ Ï‡ ¬Ï˙ (1302 4b,14) ‘that night.’ Ì‡ÈÈ‡ Ï‡ ¬Ï‡≤ (1302 1b,3) ‘those days’ is interesting since it follows Arabic structure in terms of word order, but not in terms of agreement. The agreement, though, does not indicate a literal translation, as it does not follow the Hebrew Ì‰‰ ÌÈÓÈ· (Esth 1:2). The plural demonstrative pronoun /hadøli/ ‘these’ is treated above in 6-4.6. L/I: (6-4.11) On the one hand, in the example ‰ÏÈÏ Ï‡ ¬Ï‡≤ (74 1,17) ‘that night,’ the demonstrative pronoun follows regular

247

CHAPTER EIGHT Arabic word order and not Hebrew word order; thus, as in example 6-4.8 above, it is translated interpretively. On the other hand, the demonstrative pronoun /ƒalik/ in this example is in the masculine, whereas it modifies a feminine noun. This is in line with the Hebrew masculine demonstrative pronoun in ‰Ê‰ ‰ÏÈÏ‰;3 0 thus, it represents a verbatim translation.3 1 Interrogative Pronouns and Particles Feature 6-5 — Interrogative pronouns and particles Lit.: (6-5.1) In the example Ô‡·Ï ‰Ï‡ Â¥¯Ú˙ Ô‡Î‡ ÌÂ‰Ï Ï‡˜Â (15 10a,16–17) ‘and he said to them, “Do you know Laban?,” ’ the Hebrew interrogative particle ≠‰ in Ô·Ï ˙‡ Ì˙Ú„È‰ Ì‰Ï ¯Ó‡ÈÂ (Gen 29:5) is translated as /inkån/, which usually renders the Hebrew Ì‡ ‘if.’ This is an attempt to translate quite literally the Hebrew interrogative particle. (6-5.2) Another attempt to translate the Hebrew interrogative particle ≠‰ is found in the Haggadah, where Judeo-Arabic Ô‡ is used: È‚ÈÏ ·¯ Ï‡ ·¯‚ Ô‡Î Ô‡ Â‡ (3 12,14) ‘or did God try to go?’ translating ‡›·Ï ÌÈ‰¿‡ ‰Ò‰. (6-5.3) Yet another effort to translate the interrogative particle ≠‰ results in an even more literal translation in the following example, as the ¡ar˙an interpreted it as a definite article: Ï‡˜Â Â‰Ï Ì‡ÏÒ Ï‡ ÌÂ‰Ï (15 10a,17–18) ‘and he said to them, “Is he well?,” ’ which is the translation of ÂÏ ÌÂÏ˘‰ Ì‰Ï ¯Ó‡ÈÂ (Gen 29:6). Int.: (6-5.4) The Hebrew interrogative particle ≠‰ is sometimes rendered in the ¡ar˙ in an interpretive mode. Thus, ‰ÚÓ ÈÁÂ¯˙ Ï‚‡¯ Ï‡ (15 4a,17) ‘will you go with the man?’ does not reflect the Hebrew ≠‰ in ˘È‡‰ ÌÚ ÈÎÏ˙‰ (Gen 24:58). Similarly, ‰ÏÏ‡ Á‚ Â˜È¯Ë (15 3a,3–4) ‘Did God make his journey prosperous?’ 30 31

See also p. 282, 10-3.4, for example. See also chapter 6, pp. 174–75, example 11.

248

THE MORPHOSYNTACTIC LEVEL: INTERROGATIVE PROUNOUNS/PARTICLES translates implicitly the interrogative particle ≠‰ in ß‰ ÁÈÏˆ‰‰ ÂÎ¯„ (Gen 24:21). L/I: (6-5.5) The Hebrew interrogative pronoun ‰Ó ‘what’ is usually translated interpretively in the texts of the ¡ur¥˙ by colloquial Egyptian Judeo-Arabic ˘È‡ /∑¡/ (15 2a,2; 6a,20; 8a,6; and more; 1302 3b,17; 4b,17; 5a,1; and more; 74 2,10; 2,13; 2,16; and more; 93 16,11; 17,3; 17,10; and more) and ˘‡ (3 2,9; 4,12; 4,17; and more; 91 2b,1; 2b,6; 3a,17; and more) ‘what.’3 2 Rarely are literary Judeo-Arabic ‡≤‡Ó (1302 1b,21) and ‡Ó (93 13,1) ‘what’ used. However, a clearly verbatim translation emerges in one interesting case: The Hebrew particle ‰Ó in ÌÂ˜Ó‰ ‡¯Â ‰Ó ‰Ê‰ (Gen 28:17) means in this context ‘how’ as an adverbial particle and is not the interrogative particle ‘what.’ The ¡ar˙an, in his desire to be verbatim, rendered the Hebrew adverbial exclamatory particle ‰Ó literally as Judeo-Arabic ˘È‡ ‘what’: ‰≤‡‰ Á¯ËÓ Ï‡ È¥ ¥È¡ÂÓ ˘È‡ (15 9b,20) ‘What full of awe is (in) this place.’ (6-5.6) In a similar manner, various Hebrew interrogative adverbs meaning ‘where’ are translated interpretively by colloquial Judeo-Arabic ÔÈ¥ /f∑n/ ‘where.’ For example, ‰‡ is rendered by ÔÈ¥ (15 16a,4; 22b,9); ‰ÙÈ‡ also by ÔÈ¥ (15 22a,6); and the same holds true for ‰È‡ (15 23b,9; 3 23,11; 74 13,5; 91 10b,9;3 3 93 64,8) and È‡ (1302 5b,3). Similar are ÔÈ¥ ÔÓ /min f∑n/ (3 14,5) ‘from where,’ which translates interpretively the Hebrew ÔÈÓ, and ÔÈÓ /min∑n/ (3 14,17; 74 7,4; 7,10; 91 6b,10; 7a,1) ‘from where.’ However, the latter may be considered a verbatim translation because of its orthographic/phonetic proximity to the Hebrew ÔÈÓ. Rarely is literary Judeo-Arabic ÔÈ‡ ÔÓ (93 31,9; 32,7) used. The example È·‡ ‡È Úˆ‡ ˘È‡ ÔÈ¥ ¬ÏÂ (15 32

The interrogative /∑¡/ is a feature of colloquial Egyptian Judeo-Arabic. See chapter 4, pp. 114–15, 2.1.6.1. 33 Note that in this example the letter /fe/ lacks the supralinear dot: ÔÈÙ.

249

CHAPTER EIGHT 8b,13) ‘(where) what shall I do for you, my son?,’ which translates the Hebrew È· ‰˘Ú‡ ‰Ó ‡ÂÙ‡ ‰ÎÏÂ (Gen 27:37) ‘and what then shall I do for you, my son?,’ on the other hand, is a verbatim translation based on a misreading or pseudocorrection. The ¡ar˙an considered Hebrew ‡ÂÙ‡ ‘then’ to be the interrogative adverb ‰ÙÈ‡ ‘where’ because of their orthographic/phonetic resemblance, thus translating it similarly by Judeo-Arabic ÔÈ¥ ‘where.’ Verb Conjugation: Infinitives Level: Morphosyntactic; Category 7: Verb Conjugation Feature 7-1 — Infinitives Lit.: The Hebrew preposition ≠Ï in infinitive constructs is often translated verbatim by Judeo-Arabic /li-/ followed by a finite verb, even if regular Arabic style calls for the particle /an/. However, this feature should not be placed on the extreme literal side of the continuum, since Judeo-Arabic employs a finite verb after /li-/, unlike the uninflected Hebrew verb following ≠Ï. (7-1.1) In Genesis: In ‰È‡¯ÂÂ ÁÂ¯˙Ï ‰¯Ó Ï‡ È‚˙ ÌÏ ‰Ó·Â¯ (15 2a,18) ‘maybe the woman will not come to follow me’ 3 4 the ¡ar˙an followed verbatim the Hebrew infinitive construct ˙ÎÏÏ in Gen 24:5 with the Judeo-Arabic preposition /li-/ in ÁÂ¯˙Ï /li-tr¥˙/ ‘to come,’ although the particle /an/ is expected in regular Arabic use. Similarly, in Ï‡ ÚˆÈÏ È·ˆ Ï‡ ¯¡‡˙‡ ÌÏÂ ‰≤‡‰ Ì‡ÏÎ (15 18a,7) ‘and the lad did not delay doing this thing,’ the ¡ar˙an used /li-/ in ÚˆÈÏ ‘to do’ instead of the expected /an/ in a verbatim translation of ˙Â◊ÚÏ (Gen 34:19). Furthermore, the same use of /li-/ occurs in Â„ÂÚ˜ÈÏ ‘to sit’ in Â„ÂÚ˜ÈÏ ‰Ï‡‚Â¯ Ï‡ ‰Ï ÂÚÂÂ‡Ë (15 18a,14–15) ‘the people comply with us to sit…’ which translates ˙·˘Ï in Gen 34:22. 34

For a possible analysis of ÁÂ¯˙Ï ÆÆÆ È‚˙, see chapter 7, pp. 202–3, 3-1.8.

250

THE MORPHOSYNTACTIC LEVEL: INFINITIVES (7.1-2) In Esther as well, the Hebrew infinitive construct is translated with /li-/ followed by the inflected verb, even if the particle /an/ is required by the Judeo-Arabic structure. Thus, the Judeo-Arabic /li-tigi/ ‘to come (3rd fem. sg.)’ in ‡„ ‡ÓÏÂ ‡Ï‡ È‚˙Ï … ¯˙Ò‡ ˙˜Â (1302 2b,14–15) ‘and when the time came for Esther … to come’ is a verbatim translation of the Hebrew ‡Â·Ï (Esth 2:15). In the same way, the ¡ar˙an employed Judeo-Arabic /li-/ verbatim in Â·È‚ÈÏ ‘to bring’ in … ÔÓÂ‰ÓÏ Ï‡˜ ‰ÎÏÓ Ï‡ È˙˘Â Â·È‚ÈÏ (1302 1b, 14–15) ‘he told Mehuman … to bring Queen Vashti,’ translating verbatim the Hebrew infinitive construct ‡È·‰Ï in Esth 1:11. Similar is the use of /li-/ with the finite verb ·È‚ÈÏ ‘to bring’ in ‰ÎÏÓ Ï‡ È˙˘Â ‡Ï‡ ·È‚ÈÏ ¯Ó‡ (1302 2a, 2-3) ‘he ordered to bring Queen Vashti,’ a verbatim translation of the Hebrew ‡È·‰Ï in Esth 1:17. (7-1.3) In the Haggadah: The Judeo-Arabic preposition /li-/ in ÈÎÁÈÏ ‘to tell’ in ÈÎÁÈÏ ¯˙ÎÓ Ï‡ ÏÎÂ (3 3,6–7) ‘and all who tell at length’ is a clear word-for-word translation of the Hebrew ≠Ï in ¯ÙÒÏ as part of the infinitive construct. Similarly, in ·Ò‡Á ÚˆÈÏ „Á Ï‡ ‰Ï‡ (3 6,13–14) ‘He premeditated the end to perform,’ the ¡ar˙an translated the Hebrew ˙Â˘ÚÏ with the Judeo-Arabic ÚˆÈÏ, using /li-/ in a verbatim way. In the same way, the ¡ar˙an chose once more the Judeo-Arabic preposition /li-/ in ÚˆÈÏ ‘to do’ in ÚˆÈÏ ÈÏÓ¯‡ Ï‡ Ô·Ï ·ÏË ˘‡ (3 7,7–8) ‘What did Lavan the Aramean ask to do’ to imitate the Hebrew ≠Ï in ˙Â˘ÚÏ, whereas he could have chosen the more normative Arabic particle /an/. (7-1.4) As mentioned in the above examples, it is quite regular in the ¡ar˙ to find the Judeo-Arabic preposition /li-/ translating verbatim the Hebrew ≠Ï as part of the infinitive construct, where one would expect the particle /an/. When the expected particle /an/ is used, as in Ï‡˜˙ Ô‡ ˙ÈƒÁ ÌÏÂ (93 15,11) ‘(I) did not understand (why the exodus from Egypt) is recited,’ it actually translates Hebrew ≠˘, as in ¯Ó‡˙˘ È˙ÈÎÊ ‡ÏÂ. In another 251

CHAPTER EIGHT translation of the same place in ms. 3, though, the Judeo-Arabic /li-/ is still preserved: ÏÂ˜‡Ï ˙ÈÎÊ ÌÏÂ (3 3,19). It is quite common in the ¡ur¥˙ to find instances where the ¡ar˙an followed the biblical text verbatim by translating Hebrew infinitive constructs preceded by the prepositions ≠· or ≠Î and sometimes followed by a pronoun suffix as well. On the other hand, as can be seen below in the L/I section (pp. 254–56), especially in the book of Esther, the ¡ar˙an was not consistent in this literal translation and at times he translated these Hebrew infinitive constructs using Judeo-Arabic finite verbs. (7-1.5) In Genesis: The Hebrew infinitive construct Â˙Á˜· in ‰˜·¯ ˙‡ Â˙Á˜· (Gen 26:20) is translated verbatim by Judeo-Arabic maßdar preceded by the preposition /fi-/ and followed by the pronominal suffix /-o/ in order to imitate the Hebrew: Â„¡‡ È¥ ‰˜·¯ ‰Ï‡ (15 5a,20) ‘when taking (he took) Rebecca’; ÛÈËÚ‰·Â Ô‡ˆ‰ (Gen 30:42) is rendered literally as Ìµ Ï‡ ¯È¡‡˙ È¥Â (15 13a,4) ‘and when the flock is late’; ˙Â‡¯Î (Gen 33:10) is translated verbatim as ¯∂ÓÎ in ¬‡ÏÓ Ï‡ ‰‚Â ¯∂ÓÎ (15 17a,2–3) ‘like seeing the face of the angel.’ Likewise is the verbatim translation of the infinitive construct ÂÁ¯·· in ÂÈÁ‡ ÈÙÓ ÂÁ¯·· (Gen 35:7) as Â·Â¯‰ È¥ in ‰Â¡‡ ‰‚Â ÔÓ Â·Â¯‰ È¥ (15 19a,5) ‘during his escape from his brother.’ (7-1.6) In Esther there are cases of verbatim translations of this feature, such as ÒÂÏ‚Î (1302 1b,3) ‘while sitting,’ which is the literal translation of the Hebrew ˙·)˘Î (Esth 1:2). The situation, however, is more complex as is seen in the L/I section below (pp. 254–56). Furthermore, the Hebrew infinitive absolute ˙Â˜˘‰Â (Esth 1:7), which is not preceded by a preposition, is translated verbatim by the verbal noun ·¯˘ Ï‡Â (1302 1b,10) ‘and drinking.’ (7-1.7) In the Haggadah more literal translations of the infinitive construct preceded by ≠· or ≠Î appear. Thus, ÌÈ¯ˆÓÓ È˙‡ˆ· is translated into ¯ˆÓ ÔÓ È‚Â¯¡ È¥ (3 5,1) ‘during my leaving Egypt’; 252

THE MORPHOSYNTACTIC LEVEL: INFINITIVES ÌÈ¯ˆÓ ˙‡ ÂÙ‚· is literally translated into ¯ˆÓ ‰Ï‡ Â·¯ƒ È¥ (3 18,7) ‘when striking Egypt’; and ÌÈ¯ˆÓÓ Ï‡¯˘È ˙‡ˆ· is rendered into È¥ ¯ˆÓ ÔÓ Ï‡¯˘È ‚Â¯¡ (3 20,13) ‘upon Israel’s leaving Egypt.’ (7-1.8) As for the biblical Hebrew infinitive absolute followed by an imperfect, the ¡ar˙an translated it literally. In Ú‚¯‡ ÚÂ‚¯ ¬·‡ ‰Ï‡ (15 2a,18) ‘shall I indeed take your son back?’ the ¡ar˙an used the Arabic absolute object (Wright 1974, 2:54ff) to strictly translate the Hebrew Í· ˙‡ ·È˘‡ ·˘‰ (Gen 24:5). Similarly, in the Haggadah the ¡ar˙an translated Ú„˙ Ú„È (Gen 15:13) ‘you will surely know’ literally, also using the Arabic absolute object ¥¯Ú˙ ¥È¯Ú (3 6,16) ‘you will surely know.’ In both examples the infinitive precedes the finite verb in order to follow the Hebrew word order and deviates from regular Arabic structure. Int.: (7-1.9) The regular use of the preposition /li-/ in the verbatim translation of biblical Hebrew infinitive constructs does not appear in Gen 24:33. The verse Ï›Î‡Ï ÂÈÙÏ Ì◊ÂÈÂ ‘and there was set (food) in front of him to eat’ is translated interpretively with the use of the noun /akl/ ‘food,’ as in ÏÎ‡ ÂÓ‡„Â˜ ÏÚ‚Â (15 3a,20) ‘and he set food before him,’ although the definite article is expected to precede the noun, or at least the tanw•n indefinite accusative in a more literal style. (7-1.10) Similarly, sometimes the Hebrew infinitive construct is not translated literally with /li-/ followed by the finite verb, but is interpreted for clarity. For example, the infinitive ˙Â◊ÚÏ ‘to do’ in È˙˘Â ‰ÎÏÓ· ˙Â˘ÚÏ ‰Ó ˙„Î (Esth 1:15) is translated by a Judeo-Arabic passive for clarification in ‡≤‡Ó ÔÂ‡˜ Ï‡ ·ÒÁ È˙˘Â ‰ÎÏÓ Ï‡ È¥ ÏÚ¥ÂÈ (1302 1b,21–22) ‘by law, what should be done to Queen Vashti.’ (7-1.11) In three cases in Esther the infinitive construct is translated into regular Arabic structure using /an/ followed by the finite verb, as in Ï‚¯ ÏÏ ÚˆÈ Ô‡ ˜È‡Ï ˘È‡ (1302 5a,1) ‘what is appropriate to be done to the man’ for ˘È‡· ˙Â˘ÚÏ ‰Ó (Esth 253

CHAPTER EIGHT 6:6); ‡≤Î ÚˆÈ Ô‡ ¬ÏÓ Ï‡ Ï‡˜Â (1302 7a,7) ‘The king ordered (that this is) to be done like that’ for ÔÎ ˙Â˘ÚÕ‰Ï ÍÏÓ‰ ¯Ó‡ÈÂ (Esth 9:14); and ÚˆÈ ÂÂÎÈ Ô‡ Ì‰ÈÏÚ ˙ÿ·˙ÈÏ (1302 7b,4) ‘charging them to make’ for ÌÈ˘›Ú ˙ÂÈ‰Ï Ì‰ÈÏÚ ÌÈ˜Ï (Esth 9:21). (7-1.12) In the Haggadah an interpretive translation of the infinitive construct ¯›Ó‡Ï appears. Instead of the usual verbatim translation of /li-/ followed by the finite verb, È¥ ¬·‡Ï Â¯·¡˙Â ÔÏÈ‡˜ ÌÂÈ Ï‡ ¬Ï‡≤ (3 5,9–10) ‘and you shall tell your son on that day, saying’ occurs, where ÔÏÈ‡˜ translates interpretively Hebrew ¯›Ó‡Ï. L/I: (7-1.13) The translation of Gen 24:30, which includes two infinitive constructs, ˙›‡¯Î ‘when (he) saw’ and ÂÚÓ˘ÎÂ ‘and when (he) heard,’ illustrates the literal/interpretive linguistic tension. On the one hand, the second is translated literally ÂÚÓÒÎÂ /wa-ka-sam>o/ (15 3a,15), where each Hebrew morpheme has a verbatim Judeo-Arabic equivalent: ≠Â - /wa-/; ≠Î - /ka-/; ÚÓ˘ /sam>/ and the pronominal suffix ≠Â - /-o/, as seen above in 7-1.5. On the other hand, the first infinitive construct in the biblical verse is translated unusually in an interpretive way ¥‡˘ Ô‡ Ô‡ÎÂ (15 3a,14), where the Hebrew infinitive construct is interpreted by a Judeo-Arabic perfect, usually used in the spoken variety. (7-1.14) The linguistic tension concerning the translation of the Hebrew infinitive construct preceded by the prepositions ≠Î /kE-/ and ≠· /bE-/ can be seen best in several places in the ¡ar˙ of the book of Esther. Sometimes the Hebrew infinitives are translated as such by Judeo-Arabic verbal nouns, thus keeping the literal translation: ÒÂÏ‚Î (1302 1b,3) ‘while sitting’ follows the Hebrew ˙·)˘Î (Esth 1:2); Â¯Â∂ È¥ (1302 1b,5) ‘when he displayed’ translates the Hebrew Â˙›‡¯‰· (Esth 1:4); ˙Ï‡ÓÎ·Â (1302 1b,7) ‘when (these days) were fulfilled’ renders the Hebrew ˙‡ÂÏÓ·Â (Esth 1:5); and Ì‰ÏÂ˜· (1302 3a,8) ‘when

254

THE MORPHOSYNTACTIC LEVEL: INFINITIVES they said’ translates the Hebrew Ì¯Ó‡· (Esth 3:4). The Protestant translation used finite verbs in these examples: /lammå jalasa/ ‘when he sat’ for Esth 1:2; /liyuΩhira/ ‘to display’ for Esth 1:4; /wa-lammå nqa∂at/ ‘and when (the days) were completed’ for Esth 1:5; /wa-kån¥ yaq¥l¥na/ ‘and they said’ for Esth 3:4. Other times in the ¡ar˙, the Hebrew infinitives are translated interpretively into finite verbs preceded by the conjunction ‡ÓÏ in order to make comprehension easier: ˙ÈÈ„‰ ‡ÓÏ (1302 2a,11) ‘when (the rage) subsided’ translates the Hebrew Í&÷Î (Esth 2:1); Â˙‡Ó ‡ÓÏÂ (1302 2b,3) ‘when (her father and mother) died’ is the equivalent of the Hebrew ˙ÂÓ·Â (Esth 2:7); ‡ÓÏ ÚÓÒ‡ (1302 2b,3) ‘when (it) was announced’ renders the Hebrew ÚÓ˘‰· (Esth 2:8); ‡„ ‡ÓÏÂ (1302 2b,9) ‘and when (it) arrived’ translates the Hebrew ÚÈ‚‰·Â (Esth 2:12); and ¯∂ ‡ÓÏ (1302 4a,15) ‘when (he) saw’ renders the Hebrew ˙Â‡¯Î (Esth 5:2). Here too the Protestant translation employed finite verbs, for the most part, in its translation: />indamå sakana/ ‘when (the king’s rage) had subsided’ for Esth 2:1; /fa-lammå måta/ ‘when (the parents) had died’ for Esth 2:7; and /fa-lammå sumi>a/ ‘when (the king’s order and his law) were heard’ for Esth 2:8. This is not the case in Genesis and in the Haggadah. There, the ¡ar˙an translated verbatim any Hebrew infinitive constructs preceded by ≠· or ≠Î (see above p. 252–53, 7-1.5 and 7-1.7 respectively), although their occurrence in the Haggadah is rather scarce. A good illustration of the literal-interpretive tension is the translation of ı·˜‰· ‘(they) were brought together’ (Esth 2:8; 2:19). In the first citation the ¡ar˙an translated it verbatim with the Judeo-Arabic verbal noun Ô‡ÈÓÏ· (1302 2b,4) ‘in gathering,’ and in the second he translated it interpretively as a finite verb preceded by /lamma/, ÂÓÏ˙‡ ‡ÓÏÂ (1302 2b,20) ‘and when they gathered.’ (7-1.15) Whereas in the example Ï‡ÒÈÏ (93 16,7) ‘to ask,’ the 255

CHAPTER EIGHT regular Judeo-Arabic preposition /li-/ appears, translating literally the Hebrew infinitive construct ÏÂ‡˘Ï, in another manuscript Ï‡ÒÈ ˘¥¯ÚÈ (3 4,10) ‘does not know how to ask’ occurs, without the Judeo-Arabic /li-/, in accordance with interpretive translation into colloquial Egyptian Judeo-Arabic. Verb Conjugation: Finite Verbs Feature 7-2 — Finite verbs Lit.: (7-2.1) The ¡ar˙an transferred verbatim the pi>el form as well as the root and meaning of the finite verb ˜ÁÃˆÓ in Gen 26:8 into its Judeo-Arabic fa>>al equivalent ¬ÁƒÓ (15 6a,17) ‘(Isaac) was jesting.’ Int.: (7-2.2) It is common in the ¡ar˙ to translate Hebrew finite verbs into regular Arabic style. In Genesis: Â·¯˘ÈÏ ÂÓ˙ (15 3a,1) ‘they have completed to drink’ for ˙Â˙˘Ï ÂÏÎ (Gen 24:19) and È˜Ò‡ ¬Ï‡Ó‚Ï (15 3a,1) ‘I will water your camels’ for ·‡˘‡ ÍÈÏÓ‚Ï (Gen 24:19). In Esther: ‰ÓÂÊÚ ˙Úˆ (1302 1b,13) ‘threw a banquet’ for ‰˙˘Ó ‰˙˘Ú (Esth 1:9) and È‡ËÏÒ ¯Ó‡ ‚¯¡È (1302 2a,4–5) ‘let a royal order go out’ for ˙ÂÎÏÓ ¯·„ ‡ˆÈ (Esth 1:19). In the Haggadah: ˙˜Â ÏˆÁ (3 3,15) ‘time arrived’ for ÔÓÊ ÚÈ‚‰; ¬‡¯Ú Ï‡ ‰¥„‡ˆÈ (3 9,4) ‘if a battle finds us unexpectedly’ for ‰ÓÁÏÓ ‰‡¯˜˙; and ÂÏ ÏÂ˜ ‰˙‡ (3 4,14) ‘you, tell him’ for ‰˙‡ ÂÏ ¯ÂÓ‡. L/I: (7-2.3) The verbs in È˙È‚Â È˙¯Â·ÂÎÂ È˙¯Â˙ÂÎÂ (91 4b,15) ‘(you [fem.]) multiplied, increased and came …’ translate literally the Hebrew È‡·˙Â ÈÏ„‚˙Â È·¯˙Â with the equivalent of the Qal conjugation. However, in È˙È‚Â ¬˙Ó∂ÚÂ ¬˙¯˙ÎÂ (3 8,13–14) ‘(I) multiplied and increased you, and (you [fem.]) came …’; the ¡ar˙an interpreted the Hebrew text and changed the first two verbs into causal forms. Still, the verb È˙È‚Â ‘and (you [fem.]) came’ is conjugated like the Hebrew original finite verb È‡·˙Â.

256

THE MORPHOSYNTACTIC LEVEL: ACCUSATIVE Cases: Accusative Level: Morphosyntactic; Category 8: Cases Feature 8-1 — Accusative This feature is one of the more interesting manifestations of the work of the ¡ar˙anim’. Since ˙‡ /et/ is a clear and consistent marker of the definite direct object in Hebrew, the ¡ar˙anim felt the necessity to transfer it into their ¡ur¥˙. This was done in order to faithfully transfer the Hebrew text into the Judeo-Arabic translation. Indeed, in Later Egyptian Judeo-Arabic ¡ur¥˙, the translation of the definite direct object marker ˙‡ as ‰Ï‡ /ilå/ is almost obligatory. 3 5 In other words, the definite direct object is almost always marked in the texts of the ¡ur¥˙. There are various reasons for preferring /ilå/ over /li-/. First, /ilå/ is a free morpheme, which can serve as an appropriate equivalent to the Hebrew ˙‡—also a free morpheme, whereas /li-/ is a clitic. Second, /ilå/ is missing from the spoken dialect of Later Egyptian Judeo-Arabic and therefore can be assigned a new function. Finally, languages typically need a marker for the unexpected. The subject is evident in a sentence, whereas the direct object is less so and, as a result, it needs a marker. The more prominent the marker is, the better choice it is. Indeed, /ilå/ is more prominent than /li-/. Lit.: (8-1.1) In Genesis /ilå/ translates word for word Hebrew ˙‡ in the following examples: ˜ÁˆÈ ‰Ï‡ ¥Ï¡ Ì‰¯·‡ (15 5a,19–20) 3 6 ‘Abraham begot Isaac’ translates ˜ÁˆÈ ˙‡ „ÈÏÂ‰ Ì‰¯·‡ (Gen 25:19); ‰ÈÓ Ï‡ ¯‡È·‡ ‰Ï‡ ¯¥ÁÂ (15 6b,10–11) ‘and he dug the wells of water’ translates ÌÈÓ‰ ˙]‡· ˙‡ ¯›ÙÁÈÂ (Gen 26:18); and „‚ÂÂ ·¯ Ï‡ ¬„È·Ú ·≤ ‰Ï‡ (15 30b,12–13) ‘God has found the sin of your servants’ translates ÍÈ„·Ú ÔÂÚ ˙‡ ‡ˆÓ ÌÈ‰Ï‡‰ (Gen 44:16). (8-1.2) In Esther /ilå/ behaves similarly: ¯‡˜Â‡ ‡µ ‡Ï‡ Â¯Â∂ È¥ 35

See Hary 1991b for the development of the use of /ilå/ from Later to Modern Judeo-Arabic and the translation of ˙‡ in other Judeo-Arabic dialects and other Jewish religiolects. 36 The underlining of /ilå/ and ˙‡ is mine, for emphasis only.

257

CHAPTER EIGHT Â˙ËÏÒ (1302 1b,5–6) ‘displaying the great glory of his kingdom’ translates Â˙ÂÎÏÓ „Â·Î ¯˘›Ú ˙‡ Â˙›‡¯‰· (Esth 1:4); Ï‡ È˙˘Â ‡Ï‡ ·È‚ÈÏ ‰ÎÏÓ (1302 2a,3) ‘to bring Queen Vashti’ translates ˙‡ ‡È·‰Ï ‰ÎÏÓ‰ È˙˘Â (Esth 1:17); and Hebrew ¯˙Ò‡ ˙‡ ÍÏÓ‰ ·‰‡ÈÂ ‘and the king loved Esther’ (Esth 2:17) is translated verbatim as Ï‡ ·ÁÂ ¯˙Ò‡ ‡Ï‡ ¬ÏÓ (1302 2b,17), using /ilå/. (8-1.3) Finally, in the Haggadah the same phenomenon appears: ‰˙‡‰·‡ ‰Ï‡ … ‰‚¯¡‡ ÌÏ (3 3,1–2) ‘(he) did not deliver us, our forefathers, out of …’ translates ÂÈ˙Â·‡ ˙‡ … ‡ÈˆÂ‰ ‡Ï; ‰ÏÂÎ ‰ÚÈ¯˘ Ï‡ ‰Ï‡ ÔÈ¥¯‡Ú (74 1,13) ‘all of us are learned in the Torah’ renders ‰¯Â˙‰ ˙‡ ÌÈÚ„ÂÈ ÂÏ‹k; and ‰Ï‡ ‰Ï ‡ËÚ‡ Ô‡Î ÂÏ ÌÂ‰Ï‡Ó (3 16,7) ‘if he had given us their wealth’ translates ÂÏ‡ ÌÂÓÓ ˙‡ ÂÏ Ô˙. However, Judeo-Arabic /ilå/ is usually employed to mark the definite direct object only when the Hebrew equivalent has the accusative marker ˙‡, and it is usually omitted when it is lacking in the Hebrew text. (8-1.4) In Genesis: Ï‡Ó‚ Ï‡ ¬ÿ¯·Â (15 2b,8) ‘and he made the camels kneel down’ does not contain /ilå/, although it is expected before the definite direct object, because the Hebrew text does not have it: ÌÈÏÓ‚‰ Í¯Ÿ·È* Â (Gen 24:11). Similar examples are: ˙ÏÓÂ ‡‰˙¯‚ (15 2b,18) ‘and she filled her jar,’ which translates ‡ÏÓ˙Â ‰„Î (Gen 24:16), and ˙È· Ï‡ ˙Èƒ¥ (15 3a,18) ‘I vacated the house,’ which renders ˙È·‰ È˙ÈÙ (Gen 24:31). (8-1.5) In Esther: /ilå/ does not appear in the following examples since the Hebrew equivalents do not contain the Hebrew ˙‡: Ì‰Ê‡Â‚‡ ÂÏ„‰·ÈÏ (1302 2a,2) ‘to treat their husbands contemptuously’ translates Ô‰ÈÏÚ· ˙ÂÊ·‰Ï (Esth 1:17); Ì‰Ê‡Â‚‡ Â¯·˙ÚÈ (1302 2a,7) ‘(they) respect their husbands’ renders ¯˜È Â˙È Ô‰ÈÏÚ·Ï (Esth 1:20); and ‰·˘¡ Ï‡ ÚˆÂ (1302 4b,13) ‘and he prepared (made) the gallows (tree)’ translates ıÚ‰ ˘ÚÈÂ (Esth 5:14).

258

THE MORPHOSYNTACTIC LEVEL: ACCUSATIVE (8-1.6) In the Haggadah the Judeo-Arabic accusative marker /ilå/ does not appear before /¡idyat \a∂abo/ in ˙ÈÈ„˘ ÌÂ‰È¥ ÏÒ¯È Â·ƒµ (3 15, 12) ‘He sent forth against them the fierceness of his anger,’ in slavish imitation of the Hebrew original, which lacks the accusative marker ˙‡: ÂÙ‡ ÔÂ¯Á Ì· ÁÏ˘È. The same occurs in ‰˙Ê‡Ú ‰¥Î ÌÏÂ (3 16,14) ‘and he did not meet our needs,’ which translates ÂÈÎ¯ˆ ˜ÙÒ ‡ÏÂ, and in ‰‡„Ú‡ ˜¯µ ÌÏÂ (3 16,12) ‘and he did not sink our enemies,’ which renders ÂÈ¯ˆ Ú˜˘ ‡ÏÂ. This is also the case in the other manuscripts (74, 91, and 93), although in another place in the ¡ar˙, ‰‡„Ú‡ ‰Ï‡ ˜¯µ (3 17,11) appears with the Judeo-Arabic accusative marker /ilå/, showing the productivity of this phenomenon (see below). (8-1.7) In yet another interesting example, the ¡ar˙an translated the Hebrew ‰¯Â˙ Ô˙ ‘gave the Torah’ by adding the definite article to the Judeo-Arabic object, as expected in regular Arabic style: ‰ÚÈ¯˘ Ï‡ ‰ËÚ (74 2,7) ‘gave the Torah’ (see examples below in 9-1.4, 9-1.5, and 9-1.6). Despite the fact that the translated phrase contains the definite direct object, the ¡ar˙an did not add /ilå/ as expected, because the Hebrew original did not have ˙‡ (and did not need it because the direct object there is indefinite); thus the ¡ar˙an preserved the word-for-word translation without the definite direct object marker.3 7 On the other hand, sometimes the Judeo-Arabic accusative marker /ilå/ comes before a definite direct object, even if the Hebrew original lacks it, thus demonstrating how this marker behaves regularly in the ¡ar˙, which creates its own grammar and structure. In this case, a literal translation method has become part of the style of the ¡ar˙, along with the productive use of /ilå/. Examples of this phenomenon are found in Genesis and the Haggadah, but not in Esther. 37

This example can also be analyzed as an interpretive translation because of the lack of the accusative marker /ilå/.

259

CHAPTER EIGHT (8-1.8) In Genesis: The phrase È˜È¯Ë ‰Ï‡ Á‚ÂÓ (15 3b,13) ‘make successful my way’ translates the Hebrew ÈÎ¯„ ÁÈÏˆÓ (Gen 24:42), although the latter does not have the marker /et/ in it. Similar is the example ˙‡·Â „‡ÏÂÂ ‰Ï‡ ¥ÈÏ¡Â (15 0-1,4) ‘(he) begot sons and daughters,’ which renders ˙Â·Â ÌÈ· „ÏÂÈÂ (Gen 5:19). Later in the chapter, though, the Judeo-Arabic text is regular, without /ilå/: ˙‡·Â „‡ÏÂÂ‡ ¥ÈÏ¡Â (15 0-1,7) ‘(he) begot sons and daughters,’ following the Hebrew ˙Â·Â ÌÈ· „ÏÂÈÂ (Gen 5:22). (8-1.9) In the Haggadah: The sentence ‰Ï‡ Ï˙˜ Ô‡Î ÂÏ ÌÂ‰¯‡Î·‡ (3 16,7) ‘if he had killed their firstborn’ translates Hebrew Ì‰È¯ÂÎ· ‚¯‰ ÂÏ‡ without ˙‡. Likewise, in the sentence ÌÂ‰¯‡Î·‡ ‰Ï‡ Ï˙˜ ÌÏÂ (3 16,4) ‘(he) did not kill their firstborn,’ the ¡ar˙an used the definite direct object marker /ilå/ even when ˙‡ was lacking in the Hebrew Ì‰È¯ÂÎ· ‚¯‰ ‡ÏÂ, although in other manuscripts (74, 91, and 93) as well as in another place in the same manuscript (3 17,9), /ilå/ is not used in the translation, following the Hebrew text verbatim. In the same way, the use of the Judeo-Arabic accusative marker /ilå/ takes on a life of its own and is sometimes used in the ¡ar˙ to translate other items with similar accusative meanings: (8-1.10) In the example È˜Ò‡ ¬Ï‡Ó‚ ‰Ï‡Â (15 2b,13) ‘and I will water your camels,’ /ilå/ renders Hebrew Ì‚Â ‘and also’ in Ì‚Â ‰˜˘‡ ÍÈÏÓ‚ (Gen 24:14), but uses /ilå/ as a definite direct object marker.3 8 (8-1.11) In another example from the book of Esther, ÔÓ‰ ‡ˆÂ ÌÂ˜ ‡Ò‡Â¯ ‡Ï‡Â ‰È„ÓÂ ‰È„Ó ÏÎ È¥ È≤Ï‡ ˙‡Â˘· ÏÏÂ ¬ÏÓ Ï‡ ‰¯ÊÂÏ ÌÂ˜Â (1302 3b,1–2) ‘Haman commanded the viziers of the king and the pashas, who were in every province, and the high 38

Note that a few verses below, in Gen 24:19, the ¡ar˙an has translated the Hebrew Ì‚ into regular Egyptian Judeo-Arabic Ô‡ÓÎ ‘also.’

260

THE MORPHOSYNTACTIC LEVEL: ACCUSATIVE officials of every people,’ the two instances of /li-/ and one instance of /ilå/ (underlined above) translate Hebrew Ï‡, which appears to substitute for ˙‡ in this verse (Esth 3:12) as the definite direct object marker. Furthermore, the Judeo-Arabic accusative marker /ilå/ can sometimes translate the Hebrew object pronoun that contains ˙‡, as is seen in the following examples in Genesis:3 9 (8-1.12) The preposition /ilayha/ in ‡‰ÈÏ‡ ˙Ï˜ÒÂ (15 3b,20) ‘and I asked her’ is a verbatim translation of the Hebrew inflected pronoun ‰˙›‡ in ‰˙›‡ Ï‡˘‡Â (Gen 24:47). Likewise, Judeo-Arabic /ilayhum/ in ÌÂ‰ÈÏ‡ ˙¥„‡ˆÓ Ï‡ ÏÂÎ ‰Ï‡ ÂÏ Â¯·¡Â (15 28a,15) ‘and they told him of all that had happened to them’ is a copy of Ì˙›‡ in Ì˙›‡ ˙]S‰ ÏÎ ˙‡ ÂÏ Â„È‚ÈÂ (Gen 42:29). In addition, with regard to the accusative, the ¡ar˙an sometimes translated the Hebrew original with an equivalent accusative—however, without the actual tanw•n, as it is generally lacking in Judeo-Arabic. (8-1.13) The Hebrew ·ÂË Â‡ Ú¯ ÍÈÏ‡ ¯·„ ÏÎÂ ‡Ï ‘(we) will not be able to speak to you bad or good’ (Gen 24:50) is translated verbatim into ·ÈÈË ÂÂ‡ Èƒ¯ ¬ÈÏ‡ ÌÏÎ˙ ¯„˜ ÌÏ with /ra∂i/ and /†ayyib/ in the accusative case, without the actual tanw•n. Int.: Infrequently prepositions are used to translate Hebrew ˙‡ in an interpretive manner: (8-1.14) In several instances in Genesis the preposition /ma>a/ ‘with’ translates Hebrew ˙‡ interpretively, as the ¡ar˙an has distinguished between the homonyms ˙‡ and ÌÚ (BDB, pp. 84–87): ·¯ Ï‡ ‰ÚÓ ÍÂÁ ‰˘Ó˙‡Â (15 0-1,6; 0-1,8) ‘and Enoch walked with God’ translates ÌÈ‰¿‡‰ ˙‡ ÍÂÁ ÍÏ‰˙ÈÂ (Gen 5:22, 24) (see also p. 227, 5-1.26.); È„ÈÒ ‰ÚÓ ˜ÁÂ Ïƒ¥ ÔÈÚ‡ˆ (15 4a,5) 39

For the more usual translation of ˙‡ followed by suffixed pronouns, see above, pp. 238–40, feature 6-2, pronominal suffixes.

261

CHAPTER EIGHT ‘(you) perform kindness and favor with my master’ renders È„‡ ˙‡ ˙Ó‡Â „ÒÁ ÌÈ˘›Ú (Gen 24:49); ¬˙‡¯Ó ‰ÚÓ ÌÂ˜ Ï‡ „Á‡Â „˜¯ (15 6b,1) ‘one of the people might have slept with your wife’ translates Í˙˘‡ ˙‡ ÌÚ‰ „Á‡ ·Î˘ (Gen 26:10). Moreover, the phrase ‰ÏÏ‡ „Ú ÔÓ ¯Ó‡ ·ÂÏË˙Ï (15 5b,5) ‘in order for (you) to ask for instructions from God’ translates ß‰ ˙‡ ˘]„Ï ÍÏ˙Â (Gen 25:22), where ˙‡ is translated interpretively by Judeo-Arabic „Ú ÔÓ. (8-1.15) In Esther, the preposition /ka-/ ‘as’ in ¯Ó‡Î ˙ÏÓÚ ÌÏ ¬ÏÓ Ï‡ (1302 1b,22) ‘(she) did not do what the king had instructed (her)’ renders interpretively the Hebrew ˙‡ in ‡Ï ÍÏÓ‰ ¯Ó‡Ó ˙‡ ‰˙˘Ú (Esth 1:15). (8-1.16) Likewise, in the Haggadah the preposition /min/ translates Hebrew ˙‡ in ÌÂ‰Ó √¯‡ Ï‡ ˙Ï˙Ó‡Â (3 8,10) ‘and the land was filled with them,’ where the Hebrew Ì˙›‡ is rendered by Judeo-Arabic /minhum/, according to the governing verb ‰Ï˙Ó‡. Similarly, the preposition />ala/ translates Hebrew ˙‡ in the sentence ÌÈÈ¯ˆÓ Ï‡ ‰ÈÏÚ ÂÒ‡Â (91 4b,17) ‘and the Egyptians ill-treated us,’ where the Hebrew Â˙Â‡ is translated by JudeoArabic />al∑na/. (8-1.17) Seldom did the ¡ar˙an paraphrase Hebrew ˙‡, as in the translation of the Hebrew Á& ÍÏ‰˙‰ ÌÈ‰¿‡‰ ˙‡ (Gen 6:9): È¥ Á› ‰˘Ó˙‡ ·¯ Ï‡ ˙Ú‡Ë (15 0-2,11) ‘Noah walked with God in obedience.’ (8-1.18) In addition, regarding the accusative, it is uncommon to find the accusative tanw•n used in the text, but it does appear when it is also used in the colloquial as a “frozen” form: Ô„‚ ¬ÏÓ Ï‡ Ë¡ÒÂ (1302 1b,17–18) ‘then the king became very angry.’ (8-1.19) Judeo-Arabic /ilå/ does not only represent the marker for the definite direct object. It may also render Hebrew Ï‡ ‘to’ in regular interpretive translation: ¬ÏÓ Ï‡ ÔÈ‡Ê¡ ‡Ï‡ ·È‚ÈÏ (1302 3a,18), rendering ÍÏÓ‰ ÈÊ‚–Ï‡ ‡È·‰Ï (Esth 3:9). Thus, 262

THE MORPHOSYNTACTIC LEVEL: ACCUSATIVE Egyptian Judeo-Arabic /ilå/ has two functions. L/I: The above-mentioned verbatim translation of ˙‡ as Judeo-Arabic /ilå/ is almost obligatory. The exceptions are few and far between. (8-1.20) In Genesis: In È˙Á Ï‡ È¯‡· ˙· ˙È„Â‰È ‰¯Ó „¡‡Â (15 7a,17) ‘and (he) took as a wife Judith, the daughter of Beeri, the Hittite,’ the Hebrew ˙‡ in È˙Á‰ È¯‡·–˙· ˙È„Â‰È–˙‡ ‰˘‡ Á˜ÈÂ (Gen 26:34) is not translated as Judeo-Arabic /ilå/, indicating an interpretive mode. However, the ¡ar˙an did employ the accusative marker /ilå/ in the continuation of the verse, ‰Ï‡Â È˙Á Ï‡ ÔÂÏ‡ ˙· ˙‡ÓÒ· (15 7a,17–18) ‘and Basemath, the daughter of Elon, the Hittite,’ slavishly translating the Hebrew È˙Á‰ Ô¿È‡–˙· ˙Ó˘·–˙‡Â, which includes the Hebrew ˙‡ in a literal mode. (8-1.21) In Esther: The phrase ‰ÎÏÓ Ï‡ È˙˘Â Â·È‚ÈÏ (1302 1b,15) ‘for (them) to bring Queen Vashti’ does not include Judeo-Arabic /ilå/, although ˙‡ appears in the Hebrew: ‰ÎÏÓ‰ È˙˘Â–˙‡ ‡È·‰Ï (Esth 1:11). Later in the chapter, however, the Judeo-Arabic accusative marker /ilå/ occurs in: ‰ÎÏÓ Ï‡ È˙˘Â ‡Ï‡ ·È‚ÈÏ (1302 2a,3), translating the same Hebrew sentence ‰ÎÏÓ‰ È˙˘Â–˙‡ ‡È·‰Ï in Esth 1:17. Furthermore, the Hebrew sentence ˙‡Â È˙˘Â ˙‡ ¯ÎÊ ‰ÈÏÚ ¯Ê‚ ¯˘‡ ˙‡Â ‰˙˘Ú ¯˘‡ (Esth 2:1) is translated as ‡Ï‡ ¯Î≤ ‡‰ÈÏÚ ÌÎÁ‡ È≤Ï‡Â ˙Úˆ ÈÊÏ‡ ‡Ï‡Â È˙˘Â4 0 (1302 2a,11-12) ‘(he) remembered Vashti, what she had done, and what had been decided against her,’ where the third ˙‡ in the sentence is not rendered by /ilå/, probably because of the translation of the previous ˙‡ as /ilå/.Likewise, in the sentence È‰ ‡‰¯ÈÈµÂ ‡‰˙‡È·ˆÂ (1302 2b,6) ‘and he transferred her and her maids,’ no /ilå/ appears in the translation, although ˙‡ exists in the equivalent ‰È˙Â¯Ú ˙‡Â ‰˘ÈÂ (Esth 2:9). 40

Notice that the fricative /z/ appears in ÈÊÏ‡ (following the second /ilå/), representing the loss of the interdentals in the dialect; however, the classical spelling È≤Ï‡ with the interdental /ƒ/ appears in the continuation of the verse. See chapter 4, p. 105, 1.4.4.

263

CHAPTER EIGHT

(8-1.22) In the Haggadah: The sentence Ï‡ ˙È· ‰Ï ‰· ÌÏÂ Ò„˜Ó (3 17,4) ‘and (He) did not build the holy shrine for us’ does not have the accusative marker /ilå/, despite the fact that ˙‡ exists in the Hebrew ‰¯ÈÁ·‰ ˙È· ˙‡ ÂÏ ‰· ‡ÏÂ; however, in other manuscripts (74 and 91), /ilå/ is included in a verbatim translation. Cases: Directional Feature 8-2 — Directional Lit.: The most striking verbatim translation of the directional morpheme occurs when the Hebrew directional suffix ‰≠ is transferred unchanged into Judeo-Arabic. (8-2.1) In Genesis: In the sentence ‰È‡˙Â„ ÁÂ¯ ÂÏ‡˜ (15 22a,7–8) ‘they said, “let us go to Dotan,” ’ the directional marker /-a/ in the Judeo-Arabic ‰È‡˙Â„ imitates the Hebrew ‰≠ in ‰È˙I (Gen 37:17).4 1 A similar example is ‰˙‡Ó˙ (15 23a,15; 23a,16) ‘to Timnah.’ These translations, of course, could have been simple orthographic transcriptions of the Hebrew words ‰È˙I (Gen 37:17) and ‰˙Ó˙ (Gen 38:12, 13) in the Judeo-Arabic text, but the addition of the alef in the Judeo-Arabic spelling in ‰È‡˙Â„ and ‰˙‡Ó˙ indicates Arabic pronunciation /dotåyna/ and /timnåna/ (chapter 4, p. 102, 1.1.5) respectively, with a Hebrew directional suffix ‰≠. Less striking, but still quite verbatim are the cases where the Hebrew directional ‰≠ is transformed into a zero preposition with implied accusative in the Judeo-Arabic. (8-2.2) In Genesis: In ˙È· Ï‡ Ï‚‡¯ Ï‡ ‰‚Â (15 3a,18) ‘and the man came to the house,’ which translates literally ˘È‡‰ ‡›·ÈÂ ‰˙È·‰ (Gen 24:32), the Hebrew directional ‰≠ in ‰˙È·‰ is 41

Note that the Hebrew directional suffix sometimes includes also an internal base change or the addition of /t/ when the place name ends with an /-a/.

264

THE MORPHOSYNTACTIC LEVEL: DIRECTIONAL transformed into a zero preposition and implied accusative Ï‡ ˙È·. However, this translation also follows colloquial Arabic use, namely the accusative directionality of the verb /ga/. Similarly, the directional ‰ˆ¯‡ ‘to the ground’ (Gen 24:52) is translated literally by √¯‡ in the sentence ‰ÏÏ‡Ï √¯‡ „‚ÒÂ (15 4a,10) ‘and he bowed to the ground before God.’ In Saadia the translation reads />alå l-ar∂/ and in the Protestant translation /ilå l-ar∂/. In the same way, the biblical ÂÓ‡ ‰¯◊ ‰Ï‰›‡‰ ˜ÁˆÈ À‰Œ‡œ·ÈÂ ‘Isaac brought her to the tent of his mother Sarah’ (Gen 24:67) is translated verbatim in the ¡ar˙ into ‰¯˘ ˙ÓÈ¡ ˜ÁˆÈ ‡‰·‡‚Â ÂÓÂ‡ (15 4b,9), with the directional implied in ÂÓÂ‡ ‰¯˘ ˙ÓÈ¡ ‘to the tent of his mother Sarah.’ Furthermore, in ÔÓ ÌÂ‰ÏÒ¯Â È˜¯˘ ÈÈÁ Â˙‡ÒÈÏÂ Â‰·‡ ˜ÁˆÈ ‰ÏÚ (15 4b,18) ‘then he sent them away from his son Isaac eastward, when he was still alive,’ which translates ‰Ó„L ÈÁ Â„ÂÚ· Â· ˜ÁˆÈ ÏÚÓ ÌÁÏ˘ÈÂ (Gen 25:6), the use of È˜¯˘ ‘eastward’ is directional following a zero preposition and translates verbatim the Hebrew ‰Ó„L. Int.: (8-2.3) The biblical Hebrew directional ‰ÈÚ‰ ‘to the spring’ (Gen 24:16) is translated interpretively by the ¡ar˙an as the Judeo-Arabic preposition /ilå/ in ÔÈÚ Ï‡ ‰Ï‡ ˙ÏÊÂ (15 2b,17–18) ‘and (she) went down to the spring.’ Similarly in Genesis, the ¡ar˙an translated interpretively the Hebrew directional ‰ˆÂÁ‰ ‘outward’ as ‰¯· ÔÓ (15 3a,14), with the preposition /min/. L/I: (8-2.4) In the Haggadah biblical Hebrew directional ‰ÓÈ¯ˆÓ ‘to Egypt’ is frequently rendered in the ¡ar˙ as a zero preposition in a verbatim way: ¯ˆÓ ÏÊÂ (3 7,10–11 and many other places) ‘and he went down to Egypt.’ However, it is also translated interpretively in ¯ˆÓÏ ÏÊÂ (93 21,6) ‘and he went down to Egypt’ (see also p. 230, 5-1.39).

265

CHAPTER EIGHT Definiteness Level: Morphosyntactic; Category 9: Definiteness Feature 9-1 — Adding the definite article where needed Lit.: The following examples are just a sample of the many literal translations where the ¡ar˙an added the definite article in the Judeo-Arabic ¡ar˙, where it is needed in the Arabic and where it also follows the Hebrew original.4 2 This is the standard construction and it is in agreement in all the relevant languages and varieties: Aramaic, Arabic dialects, Classical Arabic, Hebrew, and Judeo-Arabic. (9-1.1) In Genesis: In the example ÏÂÎ Ï‡ È¥ (15 2a,12) ‘in all (things),’ the ¡ar˙an added the definite article to the noun /kull/ in regular Judeo-Arabic style, following as well the Hebrew text ÏÎÃa (Gen 24:1). However, both Saadia and the Protestant translation wrote È˘ ÏÎ ÈÙ and ¡w??ý Òq? ?+ w??,4 3 respectively, without the definite article. In the same way, Ï‡ ‰Ï‡ Ô‡ÓÎ ˙¯‚Â ¯È· (15 3a,2) ‘and (she) continued running to the well’ includes the definite article, following the Hebrew original, Ï‡ „ÂÚ ı¯˙Â ¯‡·‰ (Gen 24:20). The same applies to the definite article preceding /rågil/ ‘man’ in ·‰≤ Ì‡ÊÂ¡ Ï‚‡¯ Ï‡ „¡‡Â (15 3a,4–5) ‘the man took a golden nose ring,’ following verbatim the Hebrew original ·‰Ê ÌÊ ˘È‡‰ Á˜ÈÂ (Gen 24:22). (9-1.2) In Esther: In Ú·‡Ò Ï‡ ÌÂÈ Ï‡ È¥ (1302 1b,14) ‘on the seventh day,’ both the noun and its adjective appear with the definite article as is the case in the Hebrew ÈÚÈ·˘‰ ÌÂÈÃa (Esth 1:10). In the same way, the definite article preceding ¯ˆ˜ in ¯ˆ˜ Ï‡ Ô˘Â˘ (1302 2a,14) ‘Susan the capital’ follows the Hebrew ‰¯È·‰ Ô˘Â˘ in Esth 2:3. Saadia as well used the definite article in ˜ÒÂ‚Ï‡ ÔÒÂÒ ÈÏ‡; however, the translator in the Protestant

42 43

This feature exists in many ¡ur¥˙ of the Bible. See Zafrani 1980:38–59. This is in fact an i∂åfa construction where the definiteness is implied. See also below in 9-1.2.

266

THE MORPHOSYNTACTIC LEVEL: DEFINITENESS translation chose to employ sÓýuÓý W?FK-, without the definite article. Likewise, the definite articles in ∂¥‡Á ¬ÏÓ Ï‡ È˘‡ÂË ÔÓ ‡Ò Ï‡ (1302 2a,14–15) ‘one of the king’s eunuchs, the keeper of the women’ follow the Hebrew ÌÈ˘‰ ¯ÓÂ˘ ÍÏÓ‰ ÒÈ¯Ò (Esth 2:3). (9-1.3) In the Haggadah the definite article precedes the nouns in Ì‡ÎÁ‡ Ï‡Â ÌÂÒ¯ Ï‡Â „‰‡Â˘ Ï‡ ˘‡ (3 4, 12–13) ‘what are the testimonies, statutes, and judgments,’ as is the case in the Hebrew original. In the same way the definite articles in Ï‡ ˘‡ ÌÂÎÏ È„‡‰ È„‡·Ú (3 4,17–18) ‘what does this service mean to you?’ follow regular Arabic structure as well as the Hebrew original. Although the phrase ˜‡ÂÂ‡Ê Ï‡ ˜‡ÂÂ‡Ê È¥ (3 8,14) ‘with ornaments’ contains the definite article, whereas it seems to be lacking in the Hebrew ÌÈÈ„Ú È„⁄ÚÃa; it is present, however, preceding È„Ú, in È„⁄ÚÃa /ba->adi/ ‘with the ornament.’ Int.: On the other hand, the following is also just a sample of examples where the ¡ar˙an added the definite article in the ¡ar˙, even when it is lacking in the Hebrew original, in a clear interpretive translation, following regular Arabic use (see also chapter 5, p. 151, 2.2). (9-1.4) In Genesis: In ·¯µÂÓ Ï‡ ˙˜ÂÂÏ (15 2b,9) ‘at evening time,’ the ¡ar˙an added the definite article to the noun, as required in regular Arabic structure in an interpretive translation. The definite article is lacking in the Hebrew ·¯Ú ˙ÚÏ (Gen 24:11), but it is also employed in both Saadia and the Protestant translation. Similarly, in the false i∂åfa ¯∂Ó Ï‡ ˙ÒÂÁ ‰È·ˆ Ï‡Â (15 2b,17) ‘and the young woman with the fine beauty,’ the noun appears with the definite article (¯∂Ó Ï‡), as expected in Arabic, and not following the Hebrew text’s ‰‡¯Ó (Gen 24:16) without the definite article. The same occurs in Ï‡ ‰ÈÏÚ ˙¥˙Ï‡Â Ï‡Ó˘ Ï‡ ‰ÏÚ ÂÂ‡ ÔÈÓÈ (15 4a,6) ‘and I turn, right or left.’ Both ÔÈÓÈ Ï‡ ‘right’ and Ï‡Ó˘ Ï‡ ‘left’ have the definite article,

267

CHAPTER EIGHT although they translate Hebrew ÔÈÓÈ ‘right’ and Ï‡Ó◊ ‘left’ (Gen 24:49), which lack the definite article. Like the Hebrew, Saadia did not use the definite article, but rather the indefinite accusative ‰ ˇ« ¯ÒÈ Â‡ ‰ ˇ« ÓÈ, and the Protestant translation employed the same, /yam•nan
268

THE MORPHOSYNTACTIC LEVEL: DEFINITENESS translates the Hebrew ‰ÈÂË ÚÂ¯Ê·Â ‰˜ÊÁ „È·, which does not exhibit any definite article. Here, we witness an interesting example of the literal/interpretive linguistic tension. On the one hand, in ‰„È„˘ Ï‡ ‘the mighty’ and in ‰„Â„ÓÓ Ï‡ ‘the outstretched,’ the ¡ar˙an added the Judeo-Arabic definite article, as it may be used thus in regular Judeo-Arabic structure. On the other hand, in „È ‘hand’ and in Ú‡¯„ ‘arm,’ he did not do so and remained faithful to the Hebrew text, which omits the definite article.4 4 (9-1.8) Ú‡Ó˘ ˙‡È¯˜ Ï‡ ˙˜Â ÂÏˆÁ (74 1,17–2,1) ‘the time to recite the prayer of the Shema> has come.’ In this example, the phrase Ú‡Ó˘ ˙‡È¯˜ ‘reciting the Shema>’ is a copy of the Hebrew ÚÓ˘ ˙‡È¯˜ and is treated as a compound noun in a verbatim translation. Thus, adding a definite article to it is an interpretive move, since it is absent in the Hebrew. The phrase is not treated as an i∂åfa, since in that case the ¡ar˙an would have probably attached the definite article to the second term of the i∂åfa (Ú‡Ó˘). Other manuscripts (3 3,15; 93 15,3), however, follow the Hebrew original in a literal translation and do not include the definite article: ÚÓ˘ ˙®‡©È¯˜ ˙˜Â ÂØÏˆÁ. (9-1.9)4 5 In „·‡ ÔÓ (3 6,1–2) ‘forever’ the ¡ar˙an did not add the definite article as required by regular Arabic use, so that he could follow verbatim the Hebrew original, ÌÏÂÚÓ without the definite article. In other manuscripts, however, the definite article is added: „·‡ Ï‡ ÔÓ (93 19,1) ‘forever’ in a more interpretive manner. (9-1.10) Similarly, in ¬‡¯Ú ‡¥„‡ˆÈ Ô‡ (93 23,8) ‘that should a war occur’ there is no definite article attached to the noun />iråk/ ‘war,’ in order to imitate the lack of the definite article 44

This phenomenon, however, may be related to a colloquial feature (see chapter 4, p. 125, 3.2). 45 Both examples 9-1.9 and 9-1.10 can also be treated under feature 9-4, as literal translations.

269

CHAPTER EIGHT in the Hebrew ‰ÓÁÏÓ ‰‡¯˜˙ ÈÎ. In another manuscript, however, the definite article is added, as required by regular Arabic style, in an interpretive manner, ¬‡¯Ú Ï‡ ‰¥„‡ˆÈ Ô‡ (3 9,4). Feature 9-2 — Adding the definite article where not needed Lit.: The ¡ur¥˙ do not exhibit this feature in literal translation. Int.: As above, this phenomenon does not occur regularly, even interpretively. (9-2.1) In Genesis, in the sentence Ï‡ ÌÂ¥ ‰ÏÚ ‰¯È·Î Ï‡ ‰¯‚Á Ï‡Â ¯È· (15 10a,12–13) ‘and the big stone is on the mouth of the well,’ the ¡ar˙an added the definite article to the adjective ‰¯È·Î /kab•ra/ ‘big,’ where it was not required in either the Hebrew or the Arabic. The Hebrew text reads ÈÙ ÏÚ ‰ÏI‚ Ô·‡‰Â ¯‡·‰ (Gen 29:2) ‘and the stone on the mouth of the well is big.’ Consequently, the addition of the Judeo-Arabic definite article has changed the meaning of the Hebrew verse. In both Saadia and the Protestant translation, the definite article does not appear before the adjective, and so the latter translations are more in line with the Hebrew. In Saadia ÈÏÚ «ˇ‰ÓÈ∂Ú «ˇ‰¯¡ˆÂ ‡‰ÈÙ ‘and a big stone is (found) on it’ appears and the Protestant translation reads «Îd?O? ³?+ ÊU??+ d?¾? ³"« r, vKŽ d??−?(«Ë ‘and the stone on the mouth of the well was big.’ Similarly, the ¡ar˙an added the definite article to ÌÎÁ ‘custody’ in Ï‡ È¥ ÌÂ‰‡È‡ ÏÚ‚Â ‰ÈÏÚ˘Ó Ï‡ ˙˘‡· ˙È· ÌÎÁ (15 24a,8) ‘and he placed them in custody, in the house of the chief executioner,’ where it was not needed in the Judeo-Arabic. The Hebrew equivalent ¯Ó˘ÓŸa ‘in custody’ (Gen 40:3) contains no definite article and it is omitted by both Saadia and the Protestant translation. (9-2.2) In Esther, in the phrase ¬ÏÓ Ï‡ ‰‚Â ÔÈ¯∂‡ Ï‡ (1302 1b,20–21) ‘those who see the king’s face,’ which translates the Hebrew ÍÏÓ‰ ÈÙ È‡] (Esth 1:14), there is no need for the definite article; however, this translation is interpretive, em-

270

THE MORPHOSYNTACTIC LEVEL: DEFINITENESS ploying the direct object ¬ÏÓ Ï‡ ‰‚Â /wagh il-malik/ that follows ÔÈ¯∂‡ Ï‡ rather than using an i∂åfa, as in the Hebrew. Feature 9-3 — Deleting the definite article where needed Lit.: The following is a sample of examples of literal translations where the ¡ar˙an subtracted the definite article from the following nouns, although it is required by regular Arabic structure, in agreement with the Hebrew original: (9-3.1) In Genesis: The example Â‰Ï È≤Ï‡ ÏÂÎ È¥ (15 2a,13) ‘in all that is his’ shows the lack of the definite article, where it is required in Judeo-Arabic. This translation follows the Hebrew ÂÏ ¯˘‡ ÏÎŸa (Gen 24:2). Similarly, in both Saadia’s ‰Ï ‡Ó ÚÈÓ∏ ÈÏÚ and the Protestant translation’s t" U?# lO??L?ł vKŽ, the definite article does not appear, following the Hebrew original. In the same way, the omission of the definite article from the nouns ‰ƒ¥ /fi∂∂a/ or /fa∂∂a/ ‘silver,’ ·‰≤ /ƒahab/ ‘gold,’ and ÌÂ„‰ /hud¥m/ ‘clothes’ in ÌÂ„‰Â ·‰≤ ÈÚ‡ÂÂ‡Â ‰ƒ¥ ÈÚ‡ÂÂ‡ „·Ú Ï‡ ‚¯¡Â (15 4a,10–11) ‘and the servant brought out objects of silver and gold, and clothes’ is a verbatim translation of ·‰Ê ÈÏÎÂ ÛÒÎ ÈÏÎ „·Ú‰ ‡ˆÂÈÂ ÌÈ„‚·Â (Gen 24:53), where the same Hebrew nouns lack the definite article. (9-3.2) In the book of Esther: the omission of the definite article from the noun in ‰ÓÂÊÚ Úˆ (1302 1b,4) ‘threw a banquet’ is the verbatim equivalent of the Hebrew ‰˙˘Ó ‰˘Ú (Esth 1:3), which also lacks the definite article. The definite article is absent from both Saadia’s ‡ÒÏ‚Ó Úˆ and the Protestant translation’s WÐœQ?? ? # ÂU?? ? -«. Similarly, the definite article is omitted in È‡ËÏÒ ¯Ó‡ (1302 2a,4–5) ‘royal edict’ and in ·È˙‡ÎÓ ÏÒ¯Â (1302 2a,9) ‘and he sent letters,’ literally following the Hebrew ˙ÂÎÏÓ ¯·„ (Esth 1:19) and ÌÈ¯ÙÒ ÁÏ˘ÈÂ (Esth 1:22) respectively. Note that Saadia, unlike the Egyptian ¡ar˙, used the definite article in ÍÏÓÏ‡ ¯Ó‡ to render the first phrase, but not in £˙Ú·Â ·˙Î· for the second phrase. 271

CHAPTER EIGHT (9-3.3) In the Haggadah: in ‰Ó‰Â¥ ‰ÏÂÎ ‰ÓÏÂÚ ‰ÏÂÎ ÂÏÂ ‰˙Á (3 3,4–5) ‘even if all of us were wise, all of us men of knowledge,’ the definite article is lacking from the nouns, as is the case in the Hebrew ÌÈÂ· ÂÏ‹k ÌÈÓÎÁ ÂÏ‹k ÂÏÈÙ‡Â in the Haggadah. Int.: As expected, there are no examples of omitting the definite article in the ¡ar˙ where required that do not follow the Hebrew. Feature 9-4 —Deleting the definite article where not needed Lit.: (9-4.1) In Genesis: In ‡‰¥¯Ú ÌÏ Ï‚‡¯Â (15 2b,17) ‘and no man had known her,’ the ¡ar˙an did not supply the article, despite the fact that it is needed, in order to imitate the Hebrew text ‰Ú„È ‡Ï ˘È‡Â (Gen 24:16), which also lacks the definite article. Similarly, in ‰ÈÓÂ Ï‡Ó‚ ÏÏ ¥ÏÚÂ Ô·˙ ÏÚ‚Â (15 3a,19) ‘and he gave straw and forage to the camels, and water,’ the definite article is not attached to Ô·˙ /tibn/, ¥ÏÚ />alaf/, or ‰ÈÓ /mayya/, although it is expected in regular Arabic use. Both Saadia and the Protestant translation lack the definite articles, which may also be due to the influence of the Hebrew text. The lack of the definite article indicates a verbatim translation, as it is lacking in the Hebrew original as well: ÌÈÓÂ ÌÈÏÓ‚ÃÏ ‡ÂÙÒÓÂ Ô·˙ Ô˙ÈÂ (Gen 24:32). In the same way, the word ˜Á /˙aqq/ ‘truth’ in ˜È¯Ë È¥ ˜Á (15 4a,3–4) ‘the way of truth’ does not have the definite article, in order to follow verbatim the Hebrew ˙Ó‡ Í¯„· (Gen 24:48), although it is required in regular Arabic use, as is seen in Saadia’s translation, ˜ÁÏ‡ ˜È¯Ë È¥. The Protestant translation used a noun-adjective phrase, 5#√ o¹dÞ w, ‘in a true way.’ (9-4.2) In Esther the ¡ar˙an did not add the definite article ¯È·Î ‘big’ and ¯Èµˆ ‘small’ in ¯Èµˆ „ÚÂ ¯È·Î ÔÓÏ (1302 1b,7–8) ‘big and small alike,’ in order to imitate the Hebrew original ÔË˜ „ÚÂ ÏÂ„‚ÓÏ (Esth 1:5), which also lacks the definite article. Likewise, the ¡ar˙an considered /sanat it-tålita/ in Ï‡ ˙Ò È¥ ‰˙Ï‡˙ (1302 1b,4) ‘in the year of the third’ to be an i∂åfa, in order to imitate the Hebrew possessive compound ˘ÂÏ˘ ˙˘· 272

THE MORPHOSYNTACTIC LEVEL: DEFINITENESS (Esth 1:3). This is the reason for subtracting the definite article from ‰Ò /sana/, whereas in standard Arabic /sana/ would require the definite article in a noun-adjective phrase. Indeed, both Saadia and the Protestant translation used a noun-adjective phrase /f• as-sana a®-®åli®a/ ‘in the third year.’ Int.: As expected, there are no examples that do not follow the Hebrew original of subtracting the definite article in the ¡ar˙ where it is unnecessary. L/I: (9-4.3) In the Haggadah there are several cases where the literal/interpretive linguistic tension is evident. In the following example, the ¡ar˙an did not add the definite article to the participle Ò¯‡Á /˙åris/ ‘who keeps,’ although it is expected in regular Arabic use (Wright 1974, 2:63): ÂÓ˘Ú Ò¯‡Á ¬Â¯·Ó (3 6,12) ‘blessed be He who keeps His promise.’ This is done to follow verbatim the Hebrew original Â˙ÁË·‰ ¯ÓÂ˘ ÍÂ¯·, which also lacks the definite article. In another manuscript, though, the ¡ar˙an has added the definite article in an interpretive translation: ÂÓ˘Ú Ò¯‡Á Ï‡ ¬Â¯·Ó (93 20,1). Similarly, in „‰Ú· ËÂË˘ Ï‡ ÔÈ· (93 20,3) ‘in the pact between the banks,’ the ¡ar˙an subtracted the definite article from the noun „‰Ú />ahd/ ‘pact’ where subtraction was not required. As in the previous example, he did so in order to follow the Hebrew original ˙È¯·œa ÌÈ¯˙·‰ ÔÈ·. In another manuscript, however, the ¡ar˙an added the definite article to arrive at a more interpretive translation: „Â„Á Ï‡ ÔÈ· „‰Ú Ï‡ È¥ (3 6,14–15). (9-4.4) In a rare example, √¯‡ È¥ ¯ÂÂ‡‚Ï (3 7,17) ‘to settle in the land,’ the noun √¯‡ /ar∂/ ‘land’ lacks the definite article, although in the Hebrew original it exists in ı¯‡Àa ¯Â‚Ï. In another manuscript, though, √¯‡ ÏÈÙ (91 4b,3) ‘in the land’ appears, conforming with the Hebrew original and regular Arabic structure.

273

CHAPTER EIGHT Agreement: Number Level: Morphosyntactic; Category 10: Agreement Feature 10-1 — Number: dual Lit.: Because the dual form ceased to be productive in biblical Hebrew for the most part, it seems that the ¡ar˙an did not use it in the ¡ar˙, but rather translated the Hebrew literally. (10-1.1) In Genesis: The phrase ¯ÈÂÂ‡Ò‡ ÔÈ˙‡ (15 3a,5) ‘two wristbands’ is a verbatim translation of the Hebrew ÌÈ„ÈÓˆ È˘ (Gen 24:22). In order to be faithful to the original text, the ¡ar˙an avoided using the Arabic dual form. Similarly, ÔÈ˙‡ ˙‡· (15 10b,13) ‘two daughters’ is a verbatim translation of the Hebrew ˙Â· È˙˘ (Gen 29:16). The use of ¯‡ÂÂ‚‡ Ï‡ ÔÈ˙‡ (15 14a,16) ‘the two maidservants’ to translate ˙›‰Ó⁄‡‰ È˙˘ (Gen 31:33) is similar (see below p. 299, 12-1.1–12.1.3). (10-1.2) In Esther: The phrase È˘‡ÂË ÔÈ˙‡ (1302 2b,23) ‘two eunuchs’ translates verbatim the Hebrew ÈÒÈ¯Ò È˘ (Esth 2:21). (10-1.2) In the Haggadah: In ¬Ó‡„Â˜Ï ÔÈËÂËÁÓ ¯¯ÈÓ Ï‡Â ¯ÈË¥ Ï‡ (3 5,14–15) ‘leavened bread and bitter herb are placed before you,’ it is safe to assume that the participle ÔÈËÂËÁÓ /ma˙†¥†•n/ is in the plural and not in the dual, in a verbatim translation following the Hebrew original plural ÌÈÁ‹Ó. Int.: (10-1.3) There are some instances, on the other hand, where the ¡ar˙an used the dual form in regular Arabic style, for example, ÔÈ˙¯Ó (3 2,12) ‘twice.’ Feature 10-2 — Number: plural Lit.: The ¡ar˙ exhibits cases where a plural verb precedes its plural subject in order to follow verbatim the Hebrew text, although there are other explanations for this phenomenon, as seen below. (10-2.1) In Genesis: The verb ÂË·¡˙‡ /itxaba†u/ ‘they fumbled about’ in ‡‰È·Ó È¥ „‡ÏÂÂ‡ Ï‡ ÂË·¡˙‡Â (15 5b,4) ‘and the children fumbled about inside her’ is a plural form in imitation of the Hebrew ‰·¯˜· ÌÈ·‰ ÂˆˆÂ¯˙ÈÂ (Gen 25:22), despite the fact that 274

THE MORPHOSYNTACTIC LEVEL: AGREEMENT – NUMBER Classical Arabic calls for the verb to be singular before its plural subject. However, this phenomenon occurs also in the colloquial and is prevalent in Later Judeo-Arabic (Hary 1992: 290–91) as well as in Classical Judeo-Arabic (Blau 1995: 129–30), so it is a dialectal as well as a regular Judeo-Arabic phenomenon. The same occurs in Ô‡È·Âˆ Ï‡ Â¯·ÎÂ (15 5b,12–13) ‘and the boys grew up,’ with the plural verb preceding its plural subject, following the Hebrew ÌÈ¯Ú‰ ÂÏ„‚ÈÂ (Gen 25:27). Similar is ÔÈÈƒ¯ Ï‡Â ÔÈÚ‡È¥Â¯ Ï‡ ˙‡¯˜· Ï‡ ÂÏÎ‡Â4 6 (15 25b,14) ‘and the lean ugly cows ate,’ where the verb ÂÏÎ‡Â ‘and (they) ate’ is in the plural despite the fact that it comes before its subject ˙‡¯˜· Ï‡ ‘the cows’ in a verbatim translation of ˙ÂÚ¯‰Â ˙Â˜¯‰ ˙Â¯Ù‰ ‰ÏÎ‡˙Â (Gen 41:20), and also in agreement with colloquial use. Even when the subject is an inanimate plural, its preceding verb can be plural to imitate the Hebrew original: Ï‡Ï‰Ó ÈÓÈ ÏÎ ÂÈ‰ÈÂ (Gen 5:17) is translated literally: Â‡ÎÂ Ï‡ÏÏ‰Ó Ì‡È‡ ÏÂÎ (15 0-1,1) ‘all Mahalalel’s days were ...’4 7 (10-2.2) Also in Genesis the verb appears in the plural before its dual subject: ÌÂ‰È˙‡ Ì‡ÏÁ ÂÓÏÁÂ (15 24a,11) ‘and both of them dreamed a dream,’ imitating the Hebrew original ÂÓÏÁÈÂ Ì‰È˘ ÌÂÏÁ (Gen 40:5). In the following example with a dual subject, Ï‡Â˙·Â Ô·Ï ·ÂÂ‡‚Â (15 4a,6) ‘Then Laban and Bethuel responded,’ on the surface it seems as if it is translated interpretively, because a singular verb /wa-gåwab/ precedes Laban and Bethuel. However, upon a closer investigation, one notices the Hebrew verb in the Hebrew equivalent Ï‡Â˙·Â Ô·Ï ÔÚÈÂ (Gen 24:50), which is also in the singular; thus, a verbatim translation is achieved after all. (10-2.3) In Esther: The verb ÂÏ‡˜Â in Ï‡ ÔÈÓ„‡¡ Ô‡È·ˆ Ï‡ ÂÏ‡˜Â

46 47

For the colloquial adjective ÔÈÚ‡È¥Â¯, see chapter 4, p. 123, 2.4.1. However, /kull/ ‘all’ may be considered to be the subject and not /ayyåm/ ‘days.’ Even if this is the case, a singular verb would be expected.

275

CHAPTER EIGHT Ô‡ËÏÒ (1302 2a,12) ‘and the courtiers who served the ruler said’ is in the plural, although it precedes a plural noun. This is a verbatim translation of the plural Hebrew verb Â¯Ó‡ÈÂ (Esth 2:2). Similar is the case in ¬ÏÓ Ï‡ „È·Ú ÂÏ‡˜Â (1302 3a,7) ‘and the king’s courtiers said,’ following the Hebrew plural in Â¯Ó‡ÈÂ ÍÏÓ‰ È„·Ú (Esth 3:3). The plural verb /gu/ behaves similarly in ‡‰È˘‡ÂËÂ ¯˙Ò‡ ˙‡È·ˆ Â‚Â (1302 3b,14) ‘Esther’s maids and her eunuchs came.’ In this case the Judeo-Arabic verb /gu/ is a masculine form preceding the feminine plural subject ˙‡È·ˆ /ßabiyåt/, whereas in the Hebrew original the verb is in the feminine plural form ‰ÈÒÈ¯ÒÂ ¯˙Ò‡ ˙Â¯Ú ‰‡·˙Â (Esth 4:4). Note that in Saadia, the first two verbs mentioned above that precede plural subjects are also in the plural, but the third one in Esth 4:4 is in the feminine singular, following regular Arabic structure. In the Protestant translation, however, all the verbs above that precede plural subjects are in the singular, according to regular Arabic structure. (10-2.4) In the Haggadah: In the following examples the verbs preceding their plural nouns are in the plural as well: Ï‡ Ê·Â¡ ‰˙‡‰·‡ ÂÏÎ‡ È≤Ï‡ ÔÈÎ‡ÒÓ (3 2,1–2) ‘the bread of affliction, which our fathers ate’; ÌÂ‰≤ÈÓ‡Ï˙ Â‚ ‡Ó „ÚÏ (3 3,14) ‘until their students came’; and ÔÈ„·‡Ú ‰˙‡‰·‡ Â‡Î (3 5,17) ‘our fathers were worshippers.’ All of these translate verbatim the Hebrew, where the plural verbs also precede the plural nouns. Quite frequently, when modifying nonhuman plural nouns, the modifiers may be in the plural in order to strictly copy the Hebrew original, although we would have expected the modifiers to be in the feminine singular. However, spoken Arabic may also exhibit a similar phenomenon (Mitchell 1978:24–25), and in this case the translation cannot be considered verbatim, but rather interpretive. (10-2.5) In Genesis: In the example Ô‡Î‡ „Ú È˜Ò‡ ¬Ï‡Ó‚Ï Ô‡ÓÎ Â·¯˘ÈÏ ÂÓ˙ (15 3a,1) ‘I will also water your camels, until they 276

THE MORPHOSYNTACTIC LEVEL: AGREEMENT – NUMBER finish drinking,’ the verbs ÂÓ˙ ‘(they) finished’ and Â·¯˘È ‘(they) drink’ are in the plural, although they modify the nonhuman plural Ï‡Ó‚ ‘camels,’ probably in deference to the Hebrew text, ˙Â˙˘Ï ÂÏÎ Ì‡ „Ú ·‡˘‡ ÍÈÏÓ‚Ï Ì‚ (Gen 24:19), where the verb ÂÏÎ ‘finished’ is also in the plural. The same holds true when the verb precedes the subject: Â·¯˘ÈÏ Ï‡Ó‚ Ï‡ ÂÓ˙ (15 3a,4) ‘the camels finished drinking.’ In the same way, in Ï‡Ó‚ ‰„ÂÂ‰Â ÔÈÈ‡‚ (15 4b,5) ‘Here, the camels are approaching,’ the participle ÔÈÈ‡‚ is in the plural to render verbatim the Hebrew original ÌÈ‡· (Gen 24:63). Similarly, plural adjectives may follow nonhuman plural nouns in a verbatim translation: ¯∂Ó Ï‡ ˙‡ÒÂÁ ˙‡¯˜· ÌÁÏ Ï‡ Ô‡ÓÂÒÂ (15 25a,5–6) ‘good-looking and sturdy cows’ follows verbatim the Hebrew ¯˘· ˙›‡È¯·Â ‰‡¯Ó ˙ÂÙÈ ˙Â¯Ù (Gen 41:2). (10-2.6) In Esther: In ÔÈ¯ƒÁ˙ÒÓ ÂÂÎÈÏ Ì‡Â˜‡ Ï‡ ÏÎÏ (1302 3b,7) ‘to all the peoples, in order for them to be ready,’ both the verb ÂÂÎÈ /yak¥nu/ and the participle ÔÈ¯ƒÁ˙ÒÓ /mustax∂ir•n/ are in the plural, although they modify Ì‡Â˜‡ /aqwåm/ ‘peoples, nations.’ (10-2.7) In the Haggadah: In ¯‡˙ÂÎ Ï‡ „‡ÈÈÚ‡ Ï‡Â „ÈÚ‡ÂÂÓ ÏÏ (3 21,6) ‘to other feasts and festivals,’ the adjective ¯‡˙ÂÎ /kutår/ is in the plural, although it modifies nonhuman plural nouns, in order to imitate the Hebrew plural nouns in ÌÈÏ‚¯ÏÂ ÌÈ„ÚÂÓÏ ÌÈ¯Á‡. The same, however, may occur in the dialect. Similarly, the pronominal suffix /hum/ in ÌÂ‰ÂÓ„ in Ï‡ ÔÓÂ ÁÈ‡·Ê Ï‡ ÔÓ ÌÂ‰ÂÓ„ ÏˆÁÈ È≤Ï‡ ÁÈ‡Ò¥ (3 21,8–10) ‘of the sacrifices and of the holy paschal offerings, whose blood …’ is in the plural, following the Hebrew original ÌÓ„, although it modifies the nonhuman plural nouns ÁÈ‡·Ê ‘sacrifices’ and ÁÈ‡Ò¥ ‘paschal offerings.’ Other issues of number agreement: (10-2.8) The imitation of the Hebrew plural occurs also in the

277

CHAPTER EIGHT demonstrative pronouns, as in ÈÏÂ„‰ Ì‡ÏÎ Ï‡Î4 8 (15 3a,13) ‘like these words’ and ÈÏÂ„‰ Ì‡ÏÎ Ï‡ „Ú· Ô‡ÎÂ (15 24a,5) ‘and it came to pass after these things.’ Here the ¡ar˙an used the spoken Egyptian Judeo-Arabic plural form ÈÏÂ„‰ /hadøli/ ‘these’ (chapter 4, p. 114, 2.1.4.3) in order to imitate the Hebrew plural form in ‰Ï‡‰ ÌÈ¯·„Î (Gen 24:28) and ‰Ï‡‰ ÌÈ¯·„‰ (Gen 40:1) respectively, although the Judeo-Arabic plural demonstrative pronoun modifies the singular noun /kalåm/. Both the Protestant translation and Saadia employ /håƒihi l-um¥r/. (10-2.9) In ÔÈ„Â‚ÂÂÓ Ï‡ ÌÂ˜ Ï‡ ÏÎÏ (1302 1b,7) ‘to all the people who were present,’ the ¡ar˙an preferred to be true to the Hebrew original and change the Judeo-Arabic number from singular to plural in order to imitate the biblical style with its plural adjective, ÌÈ‡ˆÓ‰ ÌÚ‰ ÏÎÏ (Esth 1:5). Consequently, nonagreement and verbatim translation in the ¡ar˙ are favored here over regularity and nonliteral translation. (10-2.10) As in the previous example, even when the plural subject is collective, the verb preceding it and the adjective modifying it are in the plural, in order to imitate the Hebrew: ÔÈ¯Â‰˘Ó Ï‡¯˘È Â‡Î È≤Ï‡ (3 8,6–7) ‘and Israel was renowned.’ This sentence copies the Hebrew ÌÈÈÈÂˆÓ Ï‡¯˘È ÂÈ‰˘ ‘that the children of Israel were distinguished.’ Similar is the example ÌÂ‰‡ÈÈ‡ ÔÈ˜È‡ƒÓ ¯ˆÓ È≤Ï‡ (3 11,3–4) ‘that the Egyptians harass them,’ which renders the Hebrew Ì˙›‡ ÌÈˆÁÂÏ ÌÈ¯ˆÓ ¯˘‡. (10-2.11) In ÔÈÚÂÓ‚Ó Â‡Î È≤Ï‡ ÆÆÆ ¯ÊÚÏ‡ È·¯Â ÚÂ˘Â‰È È·¯Â ¯ÊÚÈÏ‡ È·¯ (3 3,10–12) ‘Rabbi Eliezer, Rabbi Yehoshua and Rabbi Elazar … who were assembled,’ the relative pronoun È≤Ï‡ is in the singular, although it modifies a plural subject, for two reasons. First, the relative pronoun translates the Hebrew ¯˘‡ in a literal way, and second, it is a “frozen” form used in the Judeo-Arabic texts for all numbers and genders (see pp. 240–44, 6-3). 48

See above p. 246, 6-4.5 for an analysis of the demonstrative pronouns.

278

THE MORPHOSYNTACTIC LEVEL: AGREEMENT – NUMBER Int.: (10-2.12) The verse Ï‡Ó‚ Ï‡ ‰ÏÚ ˙·Î¯Â ‡‰‡È·ÂÒÂ ‰˜·¯ ˙Á‡¯Â Ï‚‡¯ Ï‡ ‰¯Â ˙Á‡¯Â (15 4b,1–2) ‘Then Rebecca and her maids went and rode the camels, and went …’ exhibits interpretive translation. The ¡ar˙an employed the initial verb ˙Á‡¯ in the singular, following standard Arabic structure, but also in line with the Hebrew. The continuation of the verse suggests a departure from the Hebrew because the Judeo-Arabic verbs ˙·Î¯ and ˙Á‡¯ are in the singular, as opposed to the Hebrew ‰·Î¯˙ and ‰ÎÏ˙, where they appear in the plural: ‰˜·¯ Ì˜˙Â ‰ÎÏ˙Â ÌÈÏÓ‚‰ ÏÚ ‰·Î¯˙Â ‰È˙]ÚÂ (Gen 24:61). (10-2.13) In ÆÆÆ ‡˙Ó„‡ ¯˙˘ ‰˘¯Î ‰ÈÏ‡ ÔÈ·È¯˜ Ï‡Â (1302 1b,19) ‘Those closest to him—Carshena, Shetar, Admata …’ the ¡ar˙an preferred regular Arabic style over a literal translation of the Hebrew ‡˙Ó„‡ ¯˙˘ ‡˘¯Î ÂÈÏ‡ ·]˜‰Â (Esth 1:14) by changing the Hebrew singular ·]˜ ‘close’ to Judeo-Arabic plural ÔÈ·È¯˜ in order to accommodate the Arabic structure. (10-2.14) In Esther: Ì‡ÏÎ Ï‡ ‡≤‡‰ (1302 2a,11) ‘this discussion’ follows regular Arabic structure, where Ì‡ÏÎ /kalåm/ is a collective noun, although it translates a Hebrew plural noun ‰Ï‡‰ ÌÈ¯·„‰ (Esth 2:1).4 9 L/I: (10-2.15) The sentence ‡‰ÓÂ‡Â ‡‰Â¡‡ Ï‡˜Â (15 4a,13–14) ‘her brother and mother said’ may seem an interpretive translation, as the verb /wa-qål/ ‘said’ preceding its subjects /ax¥ha waummuha/ ‘her brother and her mother’ is in the singular. However, upon a closer observation, this verb is also in the singular in the Hebrew original ‰Óœ‡Â ‰ÈÁ‡ ¯Ó‡ÈÂ (Gen 24:55), thus creating a literal translation as well. Both Saadia and the Protestant translation also employ the singular verb /wa-qåla/. Quite frequently the ¡ar˙an preferred to translate interpretively, observing regular Judeo-Arabic structure when ensuring 49

See also below, p. 280, 10-2.16–10-2.18.

279

CHAPTER EIGHT the agreement of the demonstrative pronouns;5 0 however, these translations also follow the Hebrew original: (10-2.16) In Genesis: The spoken Egyptian Judeo-Arabic plural demonstrative pronoun /hadøli/ is used in ÈÏÂ„‰ Èƒ‡¯‡ Ï‡ (15 6a,11) ‘these lands’ in order to modify the plural noun /arå∂i/ in a colloquial construction. The same occurs in ‰Ï‡‚Â¯ Ï‡Â ÈÏÂ„‰ (15 18a,10) ‘and these men’ and in ÈÏÂ„‰ ‰È‡ˆÚ Ï‡ (15 23b,18–19) ‘these rods.’ These are interpretive translations into the spoken variety, but they also follow the Hebrew use of plural demonstrative pronouns. (10-2.17) In Esther: The plural demonstrative pronoun /hadøli/ is a spoken form that is compatible with colloquial agreement, appearing in ÈÏÂ„‰ Ì‡ÈÈ‡ Ï‡ (1302 1b,7) ‘these days.’ This translation follows also the Hebrew agreement: ‰Ï‡‰ ÌÈÓÈ‰ (Esth 1:5). (10-2.18) In the Haggadah: The ¡ar˙an used the spoken Egyptian Judeo-Arabic plural demonstrative /døli/ to modify the plural nouns „‡ÏÂ‡ /awlåd/ and ˙‡ÓÏÎ /kalimåt/ respectively in „‡ÏÂ‡ Ï‡ ÈÏÂ„ (3 10,17) ‘these are the sons’; ÈÏÂ„ ˙‡ÓÏÎ ˙‡Ï˙ (3 17, 18–19) ‘these three words’; and ÔÓÂ‰ ÈÏÂ„Â (3 17,20) ‘and these are.’ The Hebrew also uses the plural demonstrative pronouns in these cases. Agreement: Gender Feature 10-3 — Gender Lit.: There are instances in the ¡ur¥˙ where the ¡ar˙an matched the gender of a word in the Judeo-Arabic translation with the gender of the corresponding word in the Hebrew text; the translation thus departs from standard Arabic usage.

50

This has not always been the case. For example, note the use of the singular demonstrative pronoun in Ì‡ÈÈ‡ Ï‡ ¬Ï‡≤ (1302 1b,3 and 2b,22) ‘those days,’ which translates Ì‰‰ ÌÈÓÈ· (Esth 1:2; 2:21).

280

THE MORPHOSYNTACTIC LEVEL: AGREEMENT – GENDER (10-3.1) In Genesis: The verb Ô‡ÎÂ in ‰ÏÏ‡ ˙Î¯· Ô‡ÎÂ (15 23-1a,14) ‘the Creator’s blessing was’ is in the masculine, although the noun /baraka/ ‘blessing’ is in the feminine. This is done in order to imitate the masculine Hebrew verb in the biblical verse, ß‰ ˙Î¯· È‰ÈÂ (Gen 39:5). Similarly, although ˜È¯Ë /†ar•q~†ar•
51

On the other hand, the translation ÏÈÏ˜ Ì‡È‡ (15 9a,4) ‘a few days’ is unique as it does not follow verbatim the Hebrew ÌÈ„Á‡ ÌÈÓÈ (Gen 27:44), nor does it follow regular Arabic use.

281

CHAPTER EIGHT although it refers to the masculine noun ¯ˆÓ ‚Â¯¡ /xur¥g mißr/ ‘the exodus from Egypt.’ This is a literal translation designed to imitate the original Hebrew feminine ÌÈ¯ˆÓ ˙‡ÈˆÈ. Sometimes, the Judeo-Arabic demonstrative pronoun follows verbatim the gender of its Hebrew equivalent, thus deviating from regular Arabic structure: (10-3.4) In Genesis: The feminine demonstrative pronoun ¬Ï˙ /tilka/ in ˙˜ÂÂ Ï‡ ¬Ï˙ È¥ (15 22b,19) ‘at that time’ follows the Hebrew feminine pronoun in ‡È‰‰ ˙Ú· (Gen 38:1) and not standard Arabic agreement structure. Likewise, in the following examples the feminine demonstrative pronoun È„‡‰ /hådi/ is used in imitation of the Hebrew feminine equivalent ˙‡Ê: È„‡‰ Úˆ‡ ÔÓ (15 30b,14) ‘for me to act this (way)’ follows the Hebrew original ˙‡Ê ˙Â˘ÚÓ (Gen 44:17), and ÂÚˆ‡ È„‡‰ ˙ÈˆÂ ‰˙‡Â (15 32a,13–14) ‘and you are bidden, do as follows’ renders ‰˙‡Â ˙‡Ê ˙ÈÂ‹ˆ (Gen 45:19). (10-3.5) In Esther: /tilka/ in ˙˜Â Ï‡ ¬Ï˙ (1302 6a,7) ‘that time’ follows the gender of its Hebrew equivalent ‡È‰‰ ˙Ú· (Esth 5:9), and does not follow standard Arabic agreement structure. (10-3.6) In the Haggadah: In ‰È¡˘ Ï‡ ˙ÈÈÏ‚˙ ‰„ (3 12,13) ‘this denotes the appearance of the Divine Presence,’ the ¡ar˙an has used the masculine demonstrative pronoun ‰„ /da/ to slavishly copy the Hebrew equivalent in ‰ÈÎ˘ ÈÂÏÈ‚ ‰Ê, although according to Arabic agreement standards a feminine pronoun is called for. Int.: (10-3.7) In Genesis: The noun ÔÈÈ‡ÏÓ Ï‡ (15 2b,9) ‘those (women) who draw water’ is in the masculine, following regular colloquial Arabic use (chapter 4, p. 128, 3.6.3), in which the feminine form is seldom used, and does not imitate the Hebrew feminine noun ˙›·‡&˘‰ ‘the women who draw water’ (Gen 24:11) in an interpretive translation. Similarly, the participle ÔÈ‚¯‡¡ (15 2b,12) ‘come out’ is in the masculine, reflecting regular 282

THE MORPHOSYNTACTIC LEVEL: AGREEMENT – GENDER colloquial Arabic use, despite the fact that in the Hebrew the feminine plural is employed in ˙›‡ˆ&È (Gen 24:13). Furthermore, the adjective ÔÈ·ÈÈË /†ayyib•n/ in ÔÓÂ‰ ÔÈ·ÈÈË Ô‡ (15 0-1,20) ‘for good-looking they are’ is in the masculine, despite the fact that it refers to a feminine noun, Ô‡Ò‡ Ï‡ ˙‡· (15 0-1,19) ‘the daughters of men.’ This constitutes an interpretive translation according to spoken Egyptian Judeo-Arabic standards and does not imitate the feminine noun in ‰‰ ˙›·›Ë ÈÎ (Gen 6:2). (10-3.8) In Esther: In Ì‰ÏÂ˜· Ì‰Ó‡„˜ Ì‰Ê‡Â‚‡ ÂÏ„‰·ÈÏ ‡Ò Ï‡ (1302 2a,2) ‘the women will treat their husbands contemptuously in their (masc. pl.) presence, saying (masc. pl.) …’ the verb ÂÏ„‰·È and the several pronouns appear in the masculine plural when referring to the feminine plural human nouns Ï‡ ‡Ò ‘the women,’ following colloquial use. Similar is ‡Ò Ï‡ Ì‰Ê‡Â‚‡ Â¯·˙ÚÈ (1302 2a,7) ‘the women respect their husbands,’ where both the verb Â¯·˙ÚÈ and the pronominal suffix /-hum/ in Ì‰Ê‡Â‚‡ are in the masculine plural describing a feminine plural noun, ‡Ò Ï‡ ‘the women.’ The same occurs in ‡Ò Ï‡ ∂¥‡Á Ì‰˜‡ÂÊ ‡ËÚÈÂ (1302 2a,15) ‘the keeper of the women; and let them be provided with their cosmetics,’ where the pronoun in Ì‰˜‡ÂÊ ‘their cosmetics’ is in the masculine plural. (10-3.9) The Hebrew ‰È¯ÚÂ Ì]Ú ˙‡Â ‘and you (fem.) are naked and bare’ in the Haggadah contains Ì]Ú in the masculine. The ¡ar˙an translated it into />uryåna/ in the feminine to match the gender of the independent pronoun /inti/ ‘you (fem.)’ in an interpretive translation, ‰¥Â˘ÎÓÂ ‰‡È¯ÂÚ È˙‡Â (3 8,15) ‘and you (fem.) are naked and bare.’ The ¡ar˙anim quite frequently employed interpretive translation with regard to the gender of the demonstrative pronouns, thus following regular Arabic agreement structure. (10-3.10) In Genesis: The demonstrative pronoun ‰„ /da/ in ‰„ È‡¥ÏÁ (15 2b,4) ‘this oath’ is in the masculine, not following

283

CHAPTER EIGHT the Hebrew ˙‡Ê È˙Ú‹·˘Ó ‘from this (fem.) oath to me,’ as it modifies a masculine Judeo-Arabic noun. In ‰„‰ ÌÂ‰Ï ÚˆÂ (15 28a,9) ‘and he did that to them,’ the ¡ar˙an employed the masculine demonstrative pronoun ‰„‰ /håda/ following regular Arabic use and not imitating ÔÎ Ì‰Ï ˘ÚÈÂ (Gen 42:25). (10-3.11) In Esther: ‰ÏÈÏ Ï‡ ¬Ï˙ È¥ (1302 4a,14) ‘on that night’ the feminine demonstrative pronoun ¬Ï˙ /tilka/ follows regular Arabic agreement structure, although the Hebrew uses the masculine demonstrative noun ‡Â‰‰ ‰ÏÈÏ· (Esth 6:1). This phenomenon occurs because of the different genders of the Hebrew ‰ÏÈÏ (masc.) and the Judeo-Arabic ‰ÏÈÏ (fem.). In the same way, in È„ ˘È‡ ‡ÏÚÂ È„ ˘È‡ (1302 3b,17) ‘what this is and why this is,’ the ¡ar˙an employed the feminine È„ /di/, not following the masculine ‰Ê in ‰Ê ‰Ó ÏÚÂ ‰Ê ‰Ó (Esth 4:5). A similar example is the verse È„ ‡ÏÚ È¡„¯ÓÏ ‰ÈÈ¯·ÎÂ ¯‡˜Â‡ Úˆ‡ ˘È‡ (1302 4b,17) ‘what honor and greatness have been conferred on Mordecai for this?,’ with the colloquial use of the feminine pronoun È„ /di/ translating the Hebrew masculine demonstrative pronoun ‰Ê in Esth 6:3. (10-3.12) In the Haggadah: The masculine demonstrative pronoun ‰„ /da/ in ‰„ ˘‡ (3 5,4) ‘what is this?’ reflects an interpretive translation, as it does not follow the Hebrew original ˙‡Ê ‰Ó, which contains a feminine demonstrative pronoun. Similarly, the demonstrative pronouns È„‡‰ /hådi/ and ¬Ï˙ /tilka/ are common in È„‡‰ ‰ÏÈÏ Ï‡ (3 2,9; 2,11; 2,13; and more) ‘this night’ and in ‰ÏÈÏ Ï‡ ¬Ï˙ (3 3,13–14) ‘that night.’ These pronouns agree with the gender of the Judeo-Arabic ‰ÏÈÏ ‘night’ and not with the Hebrew, indicating an interpretive mode. The same occurs in È„‡‰ ‰ÈÂ„ Ï‡ (3 4,3) ‘this world,’ where the ¡ar˙an has followed Judeo-Arabic agreement. L/I: (10-3.13) In Genesis, in the story of Joseph, the adjectival agreement used in the interpretation of Pharaoh’s dream shows

284

THE MORPHOSYNTACTIC LEVEL: AGREEMENT – GENDER the literal/interpretive linguistic tension. For example, Ú·Ò ÌÁÏ Ï‡ Ô‡ÓÂÒÂ ¯∂Ó Ï‡ ˙‡ÒÂÁ ˙‡¯˜· (15 25a,5–6) ‘seven handsome and sturdy cows,’ on the one hand, contains an example of literal agreement that strictly follows the Hebrew original: ˙‡ÒÂÁ ‘handsome’ is a feminine plural adjective following the Hebrew feminine plural ˙ÂÙÈ (Gen 41:2) as well as modifying the feminine plural noun ˙‡¯˜· ‘cows.’ On the other hand, the masculine plural adjective Ô‡ÓÂÒ ‘sturdy’ appears; here the ¡ar˙an did not imitate the original Hebrew feminine plural adjective ˙›‡È¯· ‘healthy’ but practiced interpretive translation (chapter 4, pp. 127–28, 3.5.1). The same occurs in ÔÈÚÏ‡Ë ÔÈ¯¡Â‡ ˙‡¯˜· (15 25a,7) ‘other cows rising,’ where the ¡ar˙an employed the masculine plural adjective ÔÈ¯¡Â‡ and the masculine participle ÔÈÚÏ‡Ë to modify a feminine plural noun, thereby not following the gender of the Hebrew original. Similarly, later in the chapter, the ¡ar˙an again used the interpretive mode in Ï‡ ÔÈÒÂÁ ˙‡¯˜· Ï‡ Ú·Ò ¯∂Ó (15 25a,9) ‘seven handsome and sturdy cows,’ by employing the masculine plural adjective ÔÈÒÂÁ to modify the feminine plural noun ˙‡¯˜·. This way, he avoided the literal translation and used a Judeo-Arabic Egyptian colloquial pattern. (10-3.14) In Esther: ˙˜Â Ï‡ ¬Ï˙ (1302 6a,7) ‘at that time’ renders ‡È‰‰ ˙Ú· (Esth 8:9), where the feminine demonstrative pronoun /tilka/ modifies the masculine noun ˙˜Â /waqt/. This is a verbatim translation in an attempt to imitate the equivalent Hebrew feminine noun ˙Ú ‘time.’ In other places, though, interpretive translations appear, where ˙˜Â Ï‡ ‡≤‡‰ (1302 4a,7) ‘at this (masc.) time’ and ‡≤‰Î ˙˜ÂÏ (1302 4a,8) ‘for a time such as this (masc.)’ follow regular Arabic agreement rules, not imitating the original Hebrew ˙‡Ê‰ ˙Ú· and ˙‡ÊÎ ˙ÚÏ (Esth 4:14) respectively. (10-3.15) In ‰ÏÈÏ Ï‡ ¬Ï‡≤ (74 1,17) ‘that night,’ the masculine demonstrative pronoun ¬Ï‡≤ /ƒålika/ modifies a feminine noun 285

CHAPTER EIGHT ‰ÏÈÏ /l∑la/ ‘night.’ This is done in order to imitate literally the gender of the equivalent Hebrew masculine pronoun Â˙Â‡ ‘that, the same,’ which modifies the masculine Hebrew noun ‰ÏÈÏ ‘night.’ In another manuscript, however, the demonstrative pronoun is in the feminine, agreeing with its feminine Arabic noun in an interpretive translation: ‰ÏÈÏ Ï‡ ¬Ï˙ (3 3,13–14) ‘that (fem.) night.’ Furthermore, in the example ‰„ ··Ò· /bisabab da/ (93 17, 5–6) ‘for this reason,’ the demonstrative pronoun ‰„ /da/ is masculine, rendering the Hebrew masculine demonstrative ‰Ê ¯Â·Ú·, while agreeing with the masculine JudeoArabic ··Ò /sabab/. However, in another manuscript the translation is not literal, as the ¡ar˙an used the feminine demonstrative pronoun È„ /di/: È„ ··Ò· /bisabab di/ (91 2b, 3–4) ‘for this (fem.) reason.’ Similar is ‰¥ Ï‡ ‰„ (3 12,4) ‘this is the murrain,’ where the demonstrative ‰„ /da/ is in the masculine while in another manuscript the ¡ar˙an has used the feminine ‰¥ Ï‡ ‰È„ (91 6a,1) ‘this is the murrain,’ possibly influenced by the seemingly feminine /fanå/ with the /-a/ ending. (10-3.16) The colloquial genitive marker /bitå>/ (masc. sg.) usually appears in the Haggadah, with only some attestation for /bitå>a/ (fem. sg.). In the following examples the marker /bitå>/ (masc. sg.) modifies feminine nouns, which may indicate verbatim translation of the Hebrew genitive marker Ï˘ ‘of.’ Since in Hebrew, the marker Ï˘ ‘of’ is not declined for gender, the ¡ar˙an may have chosen to do the same in the Judeo-Arabic ¡ar˙, employing the unmarked form /bitå>/, and thus the feminine marker /bitå>a/ is not used here: Ú·¯ Ú‡˙· ˙‡Î … ‰·¯ƒÂ ‰·¯ƒ ÏÎ ˙‡·¯ƒ (3 14,18–19 – 15,1–2) ‘each plague … was of four plagues’; ˙‡·¯ƒ ÒÓ¡ Ú‡˙· ˙‡Î … ‰·¯ƒÂ ‰·¯ƒ ÏÎ (3 15,9–10 – 15,11) ‘each plague … was of five plagues’; ‡ˆÓ Ï‡ Â„Â¡‡È ÔÈÓÂ˜È¥‡ Ú‡˙· (93 48,13) ‘they hold the matza of the afikoman.’ On the other hand, in other manuscripts the ¡ar˙an followed the interpretive method and used the feminine form /bitå>a/ for 286

THE MORPHOSYNTACTIC LEVEL: TENSE AND ASPECT agreement: ÔÈÓÂ˜È¥‡ ˙Ú‡˙· ‰ÈÈ·‡¡Ó Ï‡ ‰¯ÈË¥ Ï‡ ¥ˆÂ Â„¡‡È (74 22,1) ‘(the participants) take half of the hidden matza of the afikoman’; ÔÈÓÂ˜¥ Ï‡ ˙ÈÚ˙· ‰¯ÈË¥ Ï‡ ÒÂ Â„¡È (91 10a,12) ‘(the participants) take half of the matza of the afikoman.’5 2 Tense and Aspect Level: Morphosyntactic; Category 11: TMA Feature 11-1 — Tense/aspect Lit.: It seems that in many instances the ¡ar˙an has meticulously translated the Hebrew participle as an Arabic participle, even if it is strange in regular Arabic style. This is done to produce a verbatim grammatical translation. (11-1.1) In Genesis: ‰¯˙˘ÂÓÏ /li-mu¡tara/ (15 2a,7) ‘to the possession’ translates the Hebrew ‰˜ÓÏ (Gen 23:18), attempting to keep the seemingly participle unchanged in a verbatim translation; however, ‰˜Ó can be considered a noun (BDB, p. 889). Similarly, the ¡ar˙an has preserved the participle in ÔÎ‡Ò /såkin/ (15 2a,7) ‘dwell’ in order to render slavishly the Hebrew participle ·˘ÂÈ (Gen 24:3), although he could have chosen the more regular Arabic imperfect. In the same way, when translating Gen 24:13, the ¡ar˙an preserved the participle forms of both ·ˆ ‘stand’ and ˙›‡ˆ&È ‘come out’ by employing ·ˆ˙Ó /muntaßib/ (15 2b,11) and ÔÈ‚¯‡¡ /xarg•n/ (15 2b,12) respectively in the Judeo-Arabic, whereas he could have used the imperfect in a more regular Arabic style. (11-1.2) In Esther: The participle ÔËÏÒ˙Ó ‘rule’ in Ï‡ ‡Â‰ … ÔËÏÒ˙Ó Ï‡ È¯È˘„Ê‡ (1302 1b,2) ‘he is Xerxes who rules …’ translates verbatim the Hebrew participle ÍÏ›Ó in ˘Â¯Â˘Á‡ ‡Â‰ … ÍÏ›Ó‰(Esth 1:1); the participles ÔÈ¯∂‡ ‘those who see’ and ÔÈÒÏ‡‚ ‘those who rank’ in ‡ÏÂÂ‡ ÔÈÒÏ‡‚ Ï‡ ¬ÏÓ Ï‡ ‰‚Â ÔÈ¯∂‡ Ï‡ 52

For a possible frozen use of the masculine singular form /bitå>/ in Egyptian Judeo-Arabic dialect, see chapter 4, pp. 116–17, 2.1.7.

287

CHAPTER EIGHT ‰ËÏÒ Ï‡ È¥ (1302 1b,20–21) ‘those seeing the face of the king, occupying the first rank in the kingdom’ render verbatim the two participles ÈÕ‡] and ÌÈ·˘&È in Esth 1:14; and the participle È·¯Ó ‘bring up’ in ¯˙Ò‡ ‡ÈÈ‰ ‡ÈÒ¯Ó Ï‡ È·¯Ó Ô‡ÎÂ (1302 2b,1–2) ‘and he (Mordecai) had brought up Myrtle, i.e., Esther’ translates the Hebrew participle ÔÓ›‡ ‘foster’ in Esth 2:7. (11-1.3) In the Haggadah: The Hebrew participle ÔÈ·ÂÒÓ ‘recline’ is preserved in the Judeo-Arabic ¡ar˙ in the participle ÔÈÈÎ˙Ó (3 2,16). Similarly, the Hebrew ÌÈ„·Ú%˘Ó ‘enslaved’ is also rendered in the Judeo-Arabic as a participle, ÔÈÓ„¡˙ÒÂÓ (3 3,4) ‘enslaved.’ The same is true of the translation of ˙‡ ÌÈÚ„ÂÈ ÂÏÎ ‰¯Â˙‰ ‘all of us know the Torah,’ which also uses the participle in ‰ÚÈ¯˘ Ï‡ ÔÈ¥¯‡Ú ‰ÏÂÎ (3 3,5–6). Int.: Not infrequently do we find in the texts of the ¡ur¥˙ interpretive Arabic translations in regard to the aspect of Hebrew verbs, both perfect and imperfect, following vav consecutive. In these cases the ¡ar˙anim indicated clearly their interpretation of the tense/aspect. The ¡ar˙anim used the regular Arabic coordinating conjunction /wa-/ to render Hebrew vav consecutive. (11-1.4) In Genesis: The ¡ar˙an has employed the Judeo-Arabic perfect to translate the Hebrew imperfect5 3 following the vav consecutive: ‰Ò ÔÈ˙ÒÂ ÔÈ˙‡Â ‰È‡Ó „¯È ˘‡ÚÂ (15 0-1,2–3) ‘And Jared had lived 162 years,’ in which ˘‡ÚÂ /wa->å¡/ translates Hebrew imperfect ÈÁÈÂ (Gen 5:18); ÏÚ‚Â (15 2b,5) ‘And (he) put’ exhibits a Judeo-Arabic perfect verb, translating Hebrew imperfect following vav consecutive, Ì◊ÈÂ (Gen 24:9); Ï‡ ÂÏÎ‡Â ˙‡¯˜· (15 25a,8) ‘and the cows ate’ renders ˙Â¯Ù‰ ‰ÏÎ‡˙Â (Gen 41:4). 53

Note that “the Hebrew imperfect” refers only to the grammatical form of the verb. The actual tense/aspect of the verb following vav consecutive is complex in biblical Hebrew. See Hatav 2004 and 2007, where she has convincingly argued that biblical Hebrew verb forms do not encode tenses but only aspect and modality.

288

THE MORPHOSYNTACTIC LEVEL: TENSE AND ASPECT (11-1.5) In Esther: As above, the ¡ar˙an has translated Hebrew imperfect following vav consecutive with the Judeo-Arabic perfect: È¯È˘„Ê‡ Ï‡ Ì‡ÈÈ‡ È¥ Ô‡ÎÂ (1302 1b,2) ‘and it came to pass, in the days of Xerxes’ renders ˘Â¯Â˘Á‡ ÈÓÈ· È‰ÈÂ (Esth 1:1); Ô¡ÂÓÓ Ï‡˜Â (1302 1b,22) ‘and Memuchan said’ translates ¯Ó‡ÈÂ ÔÎ‹ÓÓ (Esth 1:16); and ·È˙‡ÎÓ ÏÒ¯Â (1302 2a,9) ‘and he sent letters’ translates the Hebrew imperfect following vav consecutive ÌÈ¯ÙÒ ÁÏ˘ÈÂ (Esth 1:22). (11-1.6) In the Haggadah as well, the ¡ar˙an has employed the perfect to translate the Hebrew imperfect following vav consecutive: ˜ÁˆÈ ‰Ï‡ ÂÏ ˙ËÚÂ (3 6,7–8) ‘and I gave him Isaac’ renders ˜ÁˆÈ ˙‡ ÂÏ Ô˙‡Â; ¬‡‰ ÔÓ ‡‰‡Ï‡ ‰ÏÏ‡ ‰‚¯¡‡Â (3 2,19–20) ‘and the Eternal One, our God, has taken us out of there’ translates Ì˘Ó ÂÈ‰Ï‡ ß‰ Â‡ÈˆÂÈÂ; and ÔÈ¯¡‡ ˙‡‰‡Ï‡ Â„·ÚÂ (3 6,2-3) ‘and they served other gods’ translates ÌÈ¯Á‡ ÌÈ‰¿‡ Â„·ÚÈÂ. Conversely, the ¡ar˙an has employed the Judeo-Arabic imperfect to translate a Hebrew perfect that follows vav consecutive: (11-1.7) In Genesis: ¬Á¯ËÓ ‰ÏÚ ¬Ú‚¯ÈÂ (15 24b,5) ‘and (he) will restore your place’ translates the Hebrew ÍÎ ÏÚ Í·È˘‰Â (Gen 40:13); Ïƒ¥ ‰È‡ÚÓ Ô‡Ï‡ Úˆ˙Â (15 24b,7) ‘and (you) will do me a favor’ renders the Hebrew „ÒÁ È„ÓÚ ‡ ˙È˘ÚÂ (Gen 40:14); and ‰„‡‰ ˙È· Ï‡ ÔÓ È‚¯¡˙Â (15 24b,8) ‘and (you) will get me out of this house’ translates the Hebrew ‰Ê‰ ˙È·‰ ÔÓ È˙‡ˆÂ‰Â (Gen 40:14). In all these examples the Hebrew is in the perfect following a vav consecutive, and the Judeo-Arabic interpretive translation is in the imperfect. Sometimes the interpretation is made clearer when the ¡ar˙an has used the colloquial imperfect prefix /bi-/: ‚ÈÏ¡ Ï‡ È¥ ÂÚ¯È·Â 5 4 54

The noun ‚ÈÏ¡ ‘bay, canal’ translates the Hebrew ÂÁ‡ ‘reed grass.’ Wehr explains that Z?OK?š was the “[n]ame of Cairo’s ancient city canal which was abandoned and leveled at the end of the 19th century” (1994:293).

289

CHAPTER EIGHT (15 25a,6) ‘and they graze in a grassy area in Cairo.’ This verse translates the Hebrew imperfect following vav consecutive in ÂÁ‡· ‰ÈÚ¯˙Â (Gen 41:2). (11-1.8) In Esther: The following Judeo-Arabic verbs are in the imperfect, rendering Hebrew perfects that succeed vav consecutive: ¬ÏÓ Ï‡ ¯Ó‡ ÚÓÒÈÂ (1302 2a,7) ‘and the king’s order is heard’ translates ÍÏÓ‰ Ì‚˙Ù ÚÓ˘Â (Esth 1:20); ˘ÈÚÈÂ (1302 4a,4) ‘and may (he) live’ renders ‰ÈÁÂ (Esth 4:11); and Â˘È·Ï‰Â ‘dressed up’ (Esth 6:9) is translated as the Judeo-Arabic imperfect ÂÒ·ÏÈÂ (1302 5a,5). (11-1.9) In the Haggadah: ‰ÈÏ‡ ÏÂ˜˙Â (3 5,4) ‘and you say to him’ translates Hebrew ÂÈÏ‡ ˙¯Ó‡Â; ¬·‡Ï ¯·¡˙Â (3 5,9) ‘and you shall tell your son’ translates Í·Ï ˙„‚‰Â; and ÌÂ‰‡ÈÈ‡ Â¥ÚÈÂ (3 6,17) ‘and they tortured them’ renders Ì˙Â‡ ÂÚÂ. At other times, regardless of vav consecutive, the ¡ar˙an has changed the morphological verbal forms of Hebrew aspects and tenses in the Judeo-Arabic in an interpretive manner. For example, a Hebrew /po>el/ participle may be translated into Judeo-Arabic as an imperfect form, /yaf>al/. While the meaning of the verb may be similar in both languages, the morphological form differs and thus the translation is considered interpretive. (11-1.10) In Genesis: The ¡ar˙an has translated the Hebrew participle ÏÎ›‡ ‘eat’ (Gen 40:17) by the Judeo-Arabic imperfect form in an interpretive manner: ÌÂ‰‡È‡ ÏÂÎ‡È ¯ÈË Ï‡Â (15 24b,13) ‘and the birds are eating them.’ Furthermore, in ÌÏÁ ÔÂÚ¯¥Â (15 25a,4) ‘and Pharaoh dreamed,’ the ¡ar˙an has also changed the Hebrew participle ÌÏ›Á ‘dream’ (Gen 41:1) in the Judeo-Arabic translation, but this time into a perfect form according to context. In another instance, the ¡ar˙an translated the biblical Hebrew participle ¯˙ÂÙ ‘interpreter’ as a Judeo-Arabic verbal noun: ÒÈÏÂ ÔÂÚ¯¥Ï ÌÂ‰‡È‡ ¯Èˆ¥˙ (15 25a,16–17) ‘and there is no interpretation of them for Pharaoh.’

290

THE MORPHOSYNTACTIC LEVEL: TENSE AND ASPECT (11-1.11) In Esther: In ‡˘Ó˙È Ô‡Î È¡„¯Ó (1302 2b,8) ‘Mordecai was walking up and down,’ the ¡ar˙an translated the Hebrew participle ÍÏ‰˙Ó ‘walk about’ (Esth 2:11) using the Judeo-Arabic imperfect form, although semantically /kån yaf>al/ is past habitual. In the same way, the ¡ar˙an translated the Hebrew participle ÚÈ‚Ó ‘reach’ (Esth 4:3) using the Judeo-Arabic imperfect form ÏˆÁÈ Ô‡Î (1302 3b,13) ‘reach.’ Furthermore, the verb È‚˙ /tigi/ in ‰ÈÈ·ˆ Ï‡ È‚˙ ‡≤‡‰·Â (1302 2b,11) ‘in this way the girl would go’ is in the imperfect, although its Hebrew equivalent is in the perfect: ‰‡· ‰¯Ú‰ ‰Ê·Â (Esth 2:13). The same holds true for the two imperfect verbs È‚˙ and Ú‚¯˙ in Ú‚¯˙ ‡ÈÈ‰ Á·ˆ Ï‡ È¥Â È‚˙ ‡ÈÈ‰ ·¯µÓ Ï‡¥ (1302 2b,12) ‘She would go in the evening and in the morning she would return,’ which translate the two perfect Hebrew verbs in ‡È‰ ¯˜··Â ‰‡· ‡È‰ ·¯Ú· ‰·˘ (Esth 2:14). In the same way, the colloquial imperfect morphological form /banΩur/ in È„Â‰È Ï‡ È¡„¯Ó ‡Ï‡ ¯∂‡· È≤Ï‡ (1302 4b,10) ‘as long as I see Mordecai, the Jew’ translates the Hebrew participle ‰‡] (Esth 5:13). (11-1.12) In the Haggadah: ÔÏÈ‡˜ /qaylan/ (3 5,10) ‘saying’ is an interpretive translation of the Hebrew ¯Ó‡Ï, since it does not use the Hebrew infinitive construct. Instead, this form is a participle in accordance with regular Judeo-Arabic style. In ‰Â·‡ ÚÈÈƒ ÈÏÓ¯‡ (3 7,10) ‘An Aramean would have destroyed his father,’ the ¡ar˙an translated the participle „·›‡ as the JudeoArabic perfect ÚÈÈƒ. Conversely, in ˙·‡ ¬¯Ú˘Â (3 8,14–15) ‘and your hair is grown,’ the ¡ar˙an translated the perfect ÁÕnˆœ as the Judeo-Arabic participle ˙·‡ in an interpretive mode. L/I: (11-1.13) The translation ÔÈÒÓµÓ (3 2,11) ‘dip’ is equivalent to the Hebrew participle ÔÈÏ·ËÓ in a literal translation; however, in ÂÒÓµ (93 13,2) ‘(we) dip,’ the verb is in the imperfect, more in line with regular Arabic usage and thus in the interpretive mode. Similarly, the participle ÔÈÏÎ‡Â (3 2,12) ‘have eaten’ is

291

CHAPTER EIGHT equivalent to the Hebrew participle ÔÈÏÎÂ‡ in a literal translation.5 5 However, other manuscripts use the colloquial imperfect form ÂÏÎ‡ (91 8b,1; 93 13,4) ‘we eat’ in an interpretive mode. Likewise, the Judeo-Arabic participle ¯˙ÎÓ /mukattar/ in ÏÎÂ ÈÎÁÈÏ ¯˙ÎÓ Ï‡ (3 3,6–7) ‘and all who tell at length’ literally translates the Hebrew participle ‰·¯Ó, whereas in other manuscripts the more common Judeo-Arabic imperfect ¯˙ÎÈ occurs in an interpretive mode: ÈÎÁÈÏ ¯˙ÎÈ È≤Ï‡ ÏÎÂ (74 1,14; 93 14,7). (11-1.14) The translation of the Hebrew participles in the Haggadah further exhibits the literal/interpretive linguistic tension. The Hebrew participle „ÓÏÓ is translated literally by the Judeo-Arabic participle ÌÏÚ˙Ó5 6 (93 21,13; 22,8), but in other manuscripts, it is translated by the perfect ÌÏÚ˙‡ (3 7,15; 8,6) and the colloquial bi-imperfect ‰ÓÏÚÈ· (91 4b,2) and ‡ÓÈÏÚÈ· (91 4b,10). Further in the Haggadah, the perfect ‰ˆÏ¡ ‘delivered us’ appears in ÌÂ‰„È ÔÓ ‰ˆÏ¡ ‡Â‰ (3 7,5) ‘He delivered us from their hands,’ differing from the Hebrew participle ÂÏÈˆÓ in an interpretive mode. However, in 93 20, 13–14 a similar participle in the Judeo-Arabic appears in a literal translation, ÌÂ‰„È ÔÓ ‡ˆÏ¡Ó ‡Â‰. (11-1.15) The literal/interpretive tension is also manifest in perfect and imperfect verbs in the Haggadah. The ¡ar˙an interpretively translated the Hebrew imperfect Â„·ÚÈ by a JudeoArabic perfect: ÂÓ„¡‡ (3 6,18) ‘(they) served.’ However, in other manuscripts he preserved the imperfect in a literal 55

Note that the literal translation is a reflection of morphology only. Semantically, the participle ÔÈÏÎ‡Â /wakl•n/ signifies present perfect: ‘have eaten.’ 56 This verb is unclear to me. It appears at the end of line 13 in 93 21 as ÏÚ˙Ó, so I assumed a final mem at the end, as if it had been cut off at the end of the line by some physical fault in the manuscript. The resulting verb ÌÏÚ˙Ó is strange, with its combination of verbal forms VII and VIII. It is difficult to assume ÏÚ˙Ó ‘cursed,’ as it does not fit the context. In any case, for our purposes here, I treat the verb as a participle. Furthermore, the appearance of ÌÏÚ˙Ó here becomes clearer later in the manuscript (93 22,8).

292

THE MORPHOSYNTACTIC LEVEL: MOOD translation: ÂÓ„¡È (91 3a,11) and ÂÓ„¡˙ÒÈ (93 20,6) ‘(they) shall serve.’ Mood Feature 11-2 — Mood Lit.: It is common to see the perfect succeed the negation particle ÌÏ /lam/ in the ¡ar˙,5 7 whereas the imperfect jussive mood is expected in regular Arabic use. One of the reasons for this phenomenon in the ¡ur¥˙ could be the adherence to the Hebrew perfect verb. On the other hand, one must consider the possibility of colloquial use here.5 8 (11-2.1) In Genesis: ¯¡‡˙‡ ÌÏ (15 18a,7) ‘(he) was not late’ translates the Hebrew ¯ÃÁÕ‡ ‡Ï (Gen 34:19); Â˘¯Î ÌÏ (15 19a,1) ‘(they) did not drive away’ translates the Hebrew ÂÙ„¯ ‡Ï (Gen 35:5); and ˙¯„˜ ÌÏ (15 20a,9) ‘(it) was not able’ translates the Hebrew ‰ÏÎÈ ‡Ï (Gen 36:7). (11-2.2) In Esther: ˙ÏÓÚ ÌÏ (1302 1b,22) ‘(she) did not do’ translates the Hebrew ‰˙˘Ú ‡Ï (Esth 1:15); ˙˙‡‚ ÌÏ (1302 2a,3) ‘(she) did not come’ translates the Hebrew ‰‡· ‡Ï (Esth 1:17); and ˙¯·¡ ÌÏ (1302 2b,10) ‘(she) did not inform’ translates the Hebrew ‰„È‚‰ ‡Ï (Esth 2:10). (11-2.3) In the Haggadah: Ô‡Î ÌÏ (91 2b,5) ‘(he) was not’ translates the Hebrew ‰È‰ ‡Ï; ‰‚¯¡‡ ÌÏ (3 3,1–2) ‘(he) did not deliver us’ translates the Hebrew ‡ÈˆÂ‰ ‡Ï, although in another manuscript the ¡ar˙an has kept the perfect but used the negation particle /la/: ‚¯¡‡ ‡Ï (93 14,2–3) ‘(he) did not deliver’; and ÌÏ ˙ÈÎÊ (3 3,19) ‘(I) did not understand’ translates È˙ÈÎÊ ‡Ï. At the same time, the ¡ar˙an quite infrequently has also employed the imperfect, following ÌÏ /lam/ in adherence to the 57

This is a common phenomenon in Later Egyptian Judeo-Arabic. See chapter 4, pp. 93–95 and p. 126, 3.3.3, and chapter 5, 141–43, as well as Hary 1992:294–95, 2007:276–77. See also pp. 215–17, 4-2.1, earlier in this chapter. 58 See the analysis in chapter 4, pp. 94–95 and p. 126, 3.3.3.

293

CHAPTER EIGHT Hebrew original, although this feature may also reflect colloquial Egyptian use. This is done to indicate either a negated imperfect or a negated imperative (see above pp. 217–18, 4-2.3). (11-2.4) In Genesis: Â·¯ƒ ÌÏ (15 22a,13) ‘(we) will not hit him’ translates the Hebrew ÂÎ ‡Ï (Gen 37:21), and ¥¯ÚÈ ÌÏ (15 26a,10) ‘will not be known’ translates the Hebrew ÚB+eÈ ‡Ï (Gen 41:31). (11-2.5) In Esther: ¯ÈÈµ˙È ÌÏ (1302 2a,5) ‘(it) will not change’ translates the Hebrew ¯Â·ÚÈ ‡Ï (Esth 1:19); È‚˙ ÌÏ (1302 2a,5) ‘(she) will not come’ translates the Hebrew ‡Â·˙ ‡Ï (Esth 1:19); and ¯·¡˙ ÌÏ (1302 2b,7) ‘(she) will not say’ translates the Hebrew „È‚˙ ‡Ï (Esth 2:10). The phrase Â·¯˘˙ ÌÏÂ ÂÏÎ‡˙ ÌÏÂ (1302 4a,10) ‘and do not eat and do not drink’ translates the Hebrew imperative Â˙˘˙ Ï‡Â ÂÏÎ‡˙ Ï‡Â (Esth 4:16). (11-2.6) In the Haggadah: ¥¯ÚÈ ÌÏ (74 2,9) ‘(he) does not know’ translates the Hebrew Ú„ÂÈ ÂÈ‡; ‰Ï ı˜È ÌÏ (93 58,2) ‘(we) will not lack’ translates the Hebrew imperative ¯ÒÁÈ Ï‡ ÂÏ; and ‰‚ÈÂÁ˙ ÌÏ (93 59,6) ‘do not make us need’ translates the Hebrew imperative ÂÎÈ¯ˆ˙ Ï‡. Int.: It is more uncommon for the ¡ar˙an to employ the imperfect after ÌÏ /lam/ as is expected in standard Arabic (although in the jussive mood), even when the Hebrew original is in the perfect. This can also be in line with the possible colloquial Judeo-Arabic use of /lam/ in Egyptian Arabic.5 9 (11-2.7) In Genesis: The ¡ar˙an has used the imperfect in the ¡ar˙, although in the Hebrew original the perfect is used: ÌÏ ¬ÂÒÓÏ (15 7a,9) ‘(we) did not touch you’ translates the Hebrew ÍÂÚ‚ ‡Ï (Gen 26:29);6 0 Â¯„˜È ÌÏ (15 21b,7) ‘(they) were not 59 60

See chapter 4, pp. 94–95, and Rosenbaum 2002a:588–95. It is possible, however, that the ¡ar˙an has mistakenly interpreted the Hebrew ÍÂÚ‚ as if it were in the imperfect, because of the initial nun which can be confused with the imperfect first plural nun prefix.

294

THE MORPHOSYNTACTIC LEVEL: PASSIVE able’ translates the Hebrew ÂÏÎÈ ‡Ï (Gen 37:4); and Â·È‚‡ ÌÏ (15 29a,7) ‘I did not bring him’ translates the Hebrew ÂÈ˙Â‡È·‰ ‡Ï (Gen 43:9). (11-2.8) In the Haggadah: ÏÂ˜È ÌÏ (3 17,18) ‘(he) did not say’ translates the Hebrew ¯Ó‡ ‡Ï, and ‚Â¯¡È ÌÏ (3 17,19) ‘(he) did not come out’ renders the Hebrew ‡ˆÈ ‡Ï with its perfect form. Voice: Passive Feature 11-3 — Voice: passive Lit.: There are many instances where the ¡ar˙an has used the JudeoArabic passive, usually by employing verbal form VII, infa>al (see chapter 4, pp. 118–19, 2.2.4) to render literally the Hebrew passive equivalent. (11-3.1) In Genesis: The following are Judeo-Arabic passive forms: Ô¥„‡ /indafan/ in Ì‰¯·‡ Ô¥„‡ ¬‡‰ (15 5a,4) ‘there Abraham was buried’ translates the Hebrew ¯·R (Gen 25:10); ÂÚÓ‚È /yangami>u/ in Ô‡ÚËÂ˜ Ï‡ ÏÂÎ ÂÚÓ‚È (15 10b,1) ‘all the flocks will be gathered’ translates Hebrew ÂÙÒÀ‡È" (Gen 29:8); and ÚˆÈ /yanßani>/ in ‰Á¯ËÓ È¥ ÚˆÈ ÌÏ (15 11a,6–7) ‘(it) will not be done in our place’ renders ‰˘ÀÚÈ" (Gen 29:26). (11-3.2) In Esther: The passive ·˙ÎÈ /yinkitib/ renders verbatim in È„ÓÂ Ò¯Ù ÚÈ‡¯˘ È¥ ·˙ÎÈÂ (1302 2a,5) ‘let it be written in the laws of Persia and Media’ its Hebrew equivalent in È˙„· ·˙ÀkÈÂ È„ÓÂ Ò¯Ù (Esth 1:19). Similar are the following passive forms: ÌÎÁ‡ /in˙akam/ in ‡‰ÈÏÚ ÌÎÁ‡ È≤Ï‡Â (1302 2a,12) ‘and what had been decreed against her,’ translating ¯Ê‚ (Esth 2:1) and ÚÓÒ‡ /insama>/ in ¬ÏÓ Ï‡ ¯Ó‡ ÚÓÒ‡ ‡ÓÏ (1302 2b,3) ‘When the king’s order was heard,’ rendering ÚÓ,˘‰œ · (Esth 2:8). (11-3.3) In the Haggadah: Ï‡˜˙ /tin<ål/ in ‚Â¯¡ È¥ Ï‡˜˙ È≤Ï‡ ¯ˆÓ (74 2,2–3) ‘that the exodus from Egypt will be told’ and ¬¥È‡ /infakk/ in ¬¥È‡ Ô‡Î ÌÏ (91 2b,5) ‘had not been redeemed’ are both passive forms that render literally the Hebrew passive forms ¯ÓÀ‡z- and Ï‡‚ respectively. 295

CHAPTER EIGHT The ¡ar˙an has also used other forms to translate the Hebrew passive, including the itfa>al form, passive participle, and more: (11-3.4) In Genesis, using the form itfa>al: ‡‰ ÂÚÓ‚˙‡Â (15 10a,13) ‘and (they) were gathered here’ translates ‰Ó˘ ÂÙÒ‡Â (Gen 29:3); ¯‡Ó˙Ï ¯·¡˙‡Â (15 23a,15–16) ‘and Tamar was told’ translates ¯Ó˙Ï „‚'ÈÂ (Gen 38:13); and Â‰‚Â ÔÓ ÂÏ‰·˙‡ (15 31b,9) ‘(they) were scared of his face’ renders ÂÈÙÓ ÂÏ‰· (Gen 45:3). In ·ˆËÓ ÌÏÒ ‰„ÂÂ‰Â (15 9b,9) ‘and here a ladder is set,’ the ¡ar˙an has used a passive participle to translate ÌÏÒ ‰‰Â ·ˆ‹Ó (Gen 28:12). (11-3.5) In Esther, using the form itfa>al: ÌÈÏ˘Â¯È ÔÓ ˙˙˘˙‡ (1302 2a,18) ‘(he) was exiled from Jerusalem’ translates ‰Ï‚À‰ ÌÈÏ˘Â¯ÈÓ (Esth 2:5); ‡ÏÓ˙‡ (1302 3a,10; 4b,5) ‘(he) was filled’ renders ‡ÏÀÓÈ (Esth 3:5; 5:9); and Â‰„˙‡ (1302 3b,1) ‘(they) were called’ translates Â‡¯∆JÈ$ (Esth 3:12). The ¡ar˙an has also used the passive participle: „ÿ¯¥ÓÂ ˙ÿ˙˘Ó (1302 3a,15) ‘scattered and dispersed’ renders „¯ÙÓÂ ¯Ê‹ÙÓ (Esth 3:8). Furthermore, in the following verse it is possible that the Judeo-Arabic internal passive, an unusual feature in the religiolect, is employed: Ì‡ÈÈ‡ Ï‡ ¯‡·¡‡ ·‡˙Î È¥ ·˙Î‡Â ‰·˘¡ Ï‡ ‡ÏÚ ÌÂ‰È˙‡ Â·Ïˆ‡Â (1302 3a,2–3) ‘and the two of them were hanged on a stake and it was recorded in the book of the history of days,’ translating verbatim the two Hebrew passive verbs ÂÏ,zÈÂ and ·˙ÀkÈÂ (Esth 2:23). (11-3.6) In the Haggadah, using the form itfa>al: ˙Ï˙Ó‡ /imtalit/ in ÌÂ‰Ó √¯‡ Ï‡ ˙Ï˙Ó‡Â (3 8,9–10) ‘and the land was filled with them’ translates Hebrew ‡ÏÀnz! . The passive participle ·ÂˆµÓ /ma\ß¥b/ in ˜Ï‡¡ Ï‡ ÏÂ˜ ‰ÏÚ ·ÂˆµÓ (3 7,14) ‘compelled by the word of the creator’ renders literally the Hebrew ÒÂ‡. Int.: On the other hand, at times the ¡ar˙an has chosen not to translate a Hebrew passive by a Judeo-Arabic passive for interpretive or stylistic reasons. The following are some examples.

296

THE MORPHOSYNTACTIC LEVEL: PASSIVE (11-3.7) In Genesis: In ¬Ï ˙·È‚ È≤Ï‡ È˙Î¯· ‰Ï‡ Ô‡Ï‡ „Â¡ (15 17a,3) ‘take now my blessing which (I) brought to you,’ the ¡ar˙an has used the active verb ˙·È‚ /gibt/ ‘(I) brought’ to translate the Hebrew passive in ÍÏ ˙‡À·‹‰ ¯˘‡ È˙Î¯· ˙‡ ‡ Á˜ ‘please accept my gift which has been brought to you’ (Gen 33:11). (11-3.8) In Esther: The Hebrew passive verb ‰‡¯˜ ‘(she) was summoned’ (Esth 2:14) is not rendered as passive in the JudeoArabic ¡ar˙, ÌÒ‡· ‡‰‰„ÈÂ (1302 2b,14) ‘and (he) calls her by name.’ (11-3.9) In the Haggadah: In many places the ¡ar˙an has translated the Hebrew passive ¯Ó‡˘ ‰ÓÎ ‘as it was said’ into Judeo-Arabic active voice, for example in ˜ÂÒÙ Ï‡ Ï‡˜ ‡ÓÎ (3 8,2) ‘as the verse said.’ The ¡ar˙an infrequently has chosen to employ the JudeoArabic passive, although the Hebrew original does not have it: (11-3.10) In Genesis: The ¡ar˙an has employed the form VII verbal pattern infa>al in È¯·˙Â (15 2b,4) ‘(you) are exonerated’ to translate the Hebrew active ˙È˜Â ‘(you) will be clear’ (Gen 24:8). (11-3.11) In Esther: The use of the Judeo-Arabic passive /yuf>al/ ‘will be done’ in È˙˘Â ‰ÎÏÓ Ï‡ È¥ ÏÚ¥ÂÈ ‡≤‡Ó (1302 1b,21–22) ‘what will be done with Queen Vashti’ is interpretive, as it translates a Hebrew infinitive construct in È˙˘Â ‰ÎÏÓ· ˙Â˘ÚÏ ‰Ó (Esth 1:15). (11-3.12) In the Haggadah, the passive verb Â·¯ƒ‡ /in∂arabu/ in ¯ˆÓ È¥ ÔÈÈ¯ˆÓ Ï‡ Â·¯ƒ‡ (3 14,6) ‘The Egyptians were smitten in Egypt’ translates the Hebrew active verb Â˜Ï ‘(they) suffered.’ Similar is ¯ÂÂ„‡ /indawwar/ (3 20,15) ‘(it) was turned,’ translating the Hebrew active verb ·›Ò$È. L/I: (11-3.13) The ¡ar˙an has employed the active verb Â¥ÈÏ¡

297

CHAPTER EIGHT /xallifu/6 1in ÌÂ‰Ï Â¥ÈÏ¡ ˙‡·Â (15 0-1,19) ‘and they begot daughters for themselves’ in an interpretive mode, although the Hebrew contains a passive verb Ì‰Ï Â„Ï'È ˙Â·Â ‘and daughters were born to them’ (Gen 6:1). On the other hand, in later occurrences the ¡ar˙an, in a verbatim translation, has translated the Hebrew passive ‰„Ï'È ‘was born’ (Gen 24:15) with the Judeo-Arabic passive ˙„ÏÂÂ˙‡ (15 2b,15), and he has rendered the Hebrew passive „Ï'È in ÌÈ· È˘ „Ï'È ÛÒÂÈÏÂ (Gen 41:50) with the Judeo-Arabic passive ¥Ï¡˙‡ in „‡ÏÂÂ‡ ÔÈ˙‡ ¥Ï¡˙‡ ¥ÒÂÈÏÂ (15 26b,20) ‘and two children were born to Joseph.’ (11-3.14) The ¡ar˙an has translated the Hebrew passive ¯ÓÀ‡zin ÌÈ¯ˆÓ ˙‡ÈˆÈ ¯ÓÀ‡-z˘ È˙ÈÎÊ ‡ÏÂ into Ï‡˜˙ in Ï‡˜˙ Ô‡ ˙ÈƒÁ ÌÏÂ ¯ˆÓ ‚Â¯¡ (93 15,11) ‘I did not understand why the exodus from Egypt is recited’ in a verbatim manner (see 11-3.3); however, in another manuscript he has employed the active verb ˙ÈÎÊ in an interpretive mode: ¯ˆÓ ‚Â¯¡ È¥ ÏÂ˜‡Ï ˙ÈÎÊ ÌÏÂ (3 3,19) ‘I did not make it to tell about the exodus from Egypt.’ (11-3.15) The Hebrew ‡Â‰ Ì‚ ÛÒÂÂ is translated into passive verbal form VII infa>al to render verbatim the Hebrew nif>al, ‡Â‰ Ô‡ÓÎ „‡ÊÈÂ (3 9,4) ‘will also be joined.’ In other manuscripts, though, the ¡ar˙an has preferred to use an active verb in a more interpretive manner, ‡ÂÂ‰ Ô‡ÓÎ „ÈÊÈÈÂØ„ÈÊÈÂ (91 5a,3; 93 23,9) ‘he will also join.’ (11-3.16) The Hebrew passive participle ·Â˙Î ‘is written’ is translated verbatim by the Judeo-Arabic passive participle ·Â˙ÎÓ (93 79,4) ‘is written,’ but Hebrew ¯Ó‡˘ is rendered by the Judeo-Arabic active voice, ˜ÂÒÙ Ï‡ Ï‡˜ ‡ÓÎ (93 79,1) ‘as the verse said.’

61

Of course there is a slight possibility that Â¥ÈÏ¡Â should be read as the passive /wa-xullifu/; however, the vowel /u/ is usually marked in Later Egyptian Judeo-Arabic, and previous examples in the ¡ar˙ indicate the reading /xallifu/.

298

THE MORPHOSYNTACTIC LEVEL: NUMERALS Numerals with Counted Nouns Level: Morphosyntactic; Category 12: Numerals Feature 12-1 — With counted nouns Lit.: The number ÔÈ˙‡ /itn∑n/ ‘two’ can accompany the counted noun, replacing a dual form and imitating Hebrew structure (see above p. 274, 10-1.1). This, of course, can also reflect spoken Egyptian Judeo-Arabic usage (chapter 4, p. 129, 3.8.1). (12-1.1) In Genesis: ·ÂÚ˘ ÔÈ˙‡ (15 5b,6) ‘two nations’ renders ÌÈÈ&‚ È˘ (Gen 25:23); ÌÓÂ‡Ï ÔÈ˙‡ (15 5b,6) ‘two peoples’ translates ÌÈÓ‹‡Ï È˘ (Gen 25:23); ÊÚ‡Ó Ô‡È„‚ ÔÈ˙‡ (15 7b,10) ‘two kids’ translates ÌÈÊÚ ÈÈ„‚ È˘ (Gen 27:9); and ˙‡ËÁÓ ÔÈ˙‡Ï (15 15b,7) ‘for two camps’ renders ˙ÂÁÓ È˘Ï (Gen 32:8). (12-1.2) In Esther: È˘‡ÂË ÔÈ˙‡ (1302 2b,23; 4b,15) ‘two eunuchs’ translates ÍÏÓ‰ ÈÒÈ¯Ò È˘ (Esth 2:21). (12-1.3) In the Haggadah: ·Â·ÁÓ ÔÈ˙‡· (93 83,4) ‘for two gold pounds’ renders Aramaicized Hebrew ÈÊÂÊ È¯˙·. The phrase ˙Â¡¯· ÔÈ˙‡ (3 22,4) ‘two blessings’ appears in the instructions to the Passover leader. This could indicate a verbatim translation, but it could also reflect colloquial Egyptian Judeo-Arabic use. There are other instances where the counted nouns follow verbatim the Hebrew original and not regular Arabic structure: (12-1.4) ÌÂ‰˙‡¯·ÂÎ ¯˘Ú ÔÈ˙‡ (15 5a, 13) ‘twelve of their chiefs,’ where the counted noun is in the plural following the Hebrew plural Ìœ‡È˘ ‘chieftains’ (Gen 25:16). In regular Arabic structure, a singular counted noun following the number ‘twelve’ is expected. The same applies to Â„‡ÏÂ ¯˘Ú „Á‡Â (15 16a,13) ‘and his eleven children,’ translating ÂÈ„ÏÈ ¯˘Ú „Á‡ ˙‡Â (Gen 32:23) and to ¬„È·Ú ¯˘Ú È«˙‡ (15 27b,8) ‘twelve of your servants,’ rendering ÍÈ„·Ú ¯˘Ú ÌÈ˘ (Gen 42:13) (chapter 4, p. 130, 3.8.3). At times the counted noun is in agreement with the Hebrew original in a literal translation, but at the same time it also adheres to standard Arabic structure: 299

CHAPTER EIGHT (12-1.5) In Genesis the following nouns appear in the singular, following verbatim the Hebrew original, but also adhering to standard Arabic use: ‰ƒ¥ ‘a piece of silver’ in ‰ƒ¥ ÔÈ¯˘Ú· (15 22b,6–7) ‘for twenty pieces of silver’ translates ÛÒÎ ÌÈ¯˘Ú· (Gen 37:28); ‰Ò ‘year’ in ‰Ò ˙È‡ÓÓ˙ (15 0-1,3–4) ‘eight hundred years’ and in ‰Ò ÔÈ˙Ò (15 5b,12) ‘sixty years’ renders ‰˘ ˙Â‡Ó ‰›Ó˘ (Gen 5:19) and ‰˘ ÌÈ˘˘ (Gen 25:26) respectively. (12-1.6) In Esther: In ‰ƒ¥ Ï‡˜˙Ó ¥‡Ï‡ ˙¯˘ÚÂ (1302 3a,17) ‘and ten thousand miskals [a type of weight] of silver,’ which translates Hebrew ÛÒÎ ¯ÎÎ ÌÈÙÏ‡ ˙¯˘ÚÂ ‘and ten thousand talents of silver’ (Esth 3:9), both in Hebrew and in Judeo-Arabic the first counted noun is in the plural and the second in the singular; the counted noun „‡ÏÂ‡ ‘children’ in ÔÓ‰ „‡ÏÂ‡ ˙¯˘Ú (1302 7a,1) ‘the ten sons of Haman’ is plural in both the ¡ar˙ and the Hebrew original ÔÓ‰ È· ˙¯˘Ú (Esth 9:10). Similar are the counted nouns in the following two examples: ˙Ú·ÒÂ ˙È‡Ó ‰È„Ó ÔÈ¯˘ÚÂ (1302 1b,3) ‘one hundred and twenty-seven provinces,’ translating ‰È„Ó ‰‡ÓÂ ÌÈ¯˘ÚÂ Ú·˘ (Esth 1:1), and ˙ÈÈÓ ÌÂÈ ÔÈ‡Ó˙Â (1302 1b,6) ‘one hundred and eighty days,’ rendering ÌÂÈ ˙‡ÓÂ ÌÈÂÓ˘ (Esth 1:4). (12-1.7) In the Haggadah: the counted nouns ˙‡·¯ƒ ‘plagues’ and ˙‡ÓÏÎ ‘words’ are plural in both the ¡ar˙ and the Hebrew: ˙‡·¯ƒ ¯˘Ú (3 13,14; 3 14,7) ‘ten plagues’; ˙‡·¯ƒ ÒÓ¡ (3 15,11) ‘five plagues’; ˙‡ÓÏÎ ˙‡Ï˙ (3 17,18–19) ‘three words’; the counted noun ‰Ò ‘year’ is singular in both the ¡ar˙ and the Hebrew original: ‰Ò ÔÈÚ·Ò (3 3,19) ‘seventy years.’ Rarely does the word order of the counted noun follow the nonstandard Hebrew original in a literal translation, where the counted noun precedes the numeral: (12-1.8) In Genesis: ÔÈ˙È‡Ó ˙‡‚Ú ÔÈ¯˘Ú ‰ÒÈÈ˙Â ÔÈ˙È‡Ó ÊÚ‡Ó ÔÈ¯˘Ú Ô‡¥¯¡Â (15 15b,18–19) ‘two hundred she-goats and twenty he-goats, two hundred ewes and twenty rams’ translates ÌÈÊÚ

300

THE MORPHOSYNTACTIC LEVEL: NUMERALS ÌÈ¯˘Ú ÌÈÏÈ‡Â ÌÈ˙‡Ó ÌÈÏÁ¯ ÌÈ¯˘Ú ÌÈ˘È˙Â ÌÈ˙‡Ó (Gen 32:15); Ô‡¯˙ ‰¯˘Ú ˘ÂÁÂ ÔÈ¯˘Ú ¯È‡ÓÁ ‰¯˘Ú ¯˜·Â ÔÈÚ·¯‡ (15 15b,19–20) ‘forty cows and ten bulls, twenty she-asses and ten he-asses’ renders ‰¯˘Ú ÌXÈÚÂ ÌÈ¯˘Ú ˙&&˙‡ ‰¯˘Ú ÌÈ¯ÙÂ ÌÈÚ·¯‡ ˙Â¯Ù (Gen 32:16); and ‰¯˘Ú ¥ÒÂÈ Ô‡ÂÂ¡‡ (15 27a,15) ‘ten of Joseph’s brothers,’ translates ‰¯˘Ú ÛÒÂÈ ÈÁ‡ (Gen 42:3) (see also below, p. 302, 12-.1.11). At times the structure of the numeral does not follow Arabic structure, but rather imitates the Hebrew original in a verbatim translation: (12-1.9) In Genesis: The numeral in Ï‡Ó‚ ‰¯˘Ú (15 2b,6) ‘ten camels’ follows the Hebrew ÌÈÏÓ‚ ‰¯˘Ú (Gen 24:10), whereas in regular Arabic usage an i∂åfa construction of the numeral with the counted noun is expected, />a¡árat gimål/. The same is true of the phrase ·‰„ ‰¯˘Ú (15 3a,5–6) ‘ten (pieces of) gold,’ which follows verbatim the Hebrew ·‰Ê ‰¯˘Ú (Gen 24:22) in terms of both the absolute form of the number ‘ten,’ as in the previous example, and the number of the counted noun, which is singular in the Judeo-Arabic translation, just like the Hebrew. (12-1.10) In the Haggadah: ı‡¡˘‡ Ú·¯‡ (3 4,8) ‘four sons’ appears, but in two other manuscripts the i∂åfa rule is preserved: ı‡¡˘‡ ˙Ú·¯‡ (74 2,8) and ı¡˘‡ ˙Ú·¯‡ (93 16,5–7). Int.: At times the ¡ar˙an chose to follow regular Arabic structure when using the singular counted noun, instead of imitating the Hebrew original: (12-1.11) In Genesis: ÔÈ¯˘Ú Ô‡¥¯¡Â … ÔÈ¯˘Ú ‰ÒÈÈ˙Â ÔÈ˙È‡Ó ÊÚ‡Ó (15 15b,18–19) ‘two hundred she-goat and twenty he-goats … and twenty rams’ translates the Hebrew ÌÈ¯˘Ú ÌÈ˘È˙Â ÌÈ˙‡Ó ÌÈÊÚ ÌÈ¯˘Ú ÌÈÏÈ‡Â … (Gen 32:15). In this example the Judeo-Arabic counted nouns are in the singular, following regular Arabic structure, whereas the equivalent Hebrew counted nouns are in the plural (for the literal replication of word order, see 12-1.8). (12-1.12) In the Haggadah: ‰·¯ƒ ÔÈÒÓ¡ (3 14,7-8) ‘fifty plagues’; 301

CHAPTER EIGHT ‰·¯ƒ ÔÈÚ·¯‡ (3 15,6) ‘forty plagues’; and ‰·¯ƒ ÔÈ˙È‡Ó (3 15,7) ‘two hundred plagues,’ where the noun ‰·¯ƒ ‘plague’ is singular, while the equivalent Hebrew ˙ÂÎÓ ‘plagues’ is plural. Some counted nouns reflect dialectal use in an interpretive translation: (12-1.13) In Genesis: The counted noun in ÔÈÒ ÔÈ˙Ï˙Â ‰È‡ÓÓ˙ (15 0-1,1) ‘eight hundred and thirty years’ is in the plural, whereas the Hebrew counted noun is in the singular: ‰˘ ÌÈ˘¿˘ ‰˘ ‰›Ó˘Â (Gen 5:16) (see chapter 4, p. 130, 3.8.3). L/I: (12-1.14) As seen above (12-1.11–12-1.12), sometimes the counted noun follows regular Arabic structure and not the Hebrew original, as is the case with the Judeo-Arabic singular ·ÎÂÎ ‘star’ in the verse ‰ÈÏ ÔÈ„‚‡Ò ·ÎÂÎ ¯˘Ú „Á‡ÂÂ (15 21b,16–17) ‘and eleven stars were bowing down to me’ (unlike 12-1.4 on pp. 299–300). The singular Judeo-Arabic ·ÎÂÎ ‘star’ translates the original Hebrew plural ÌÈ·ÎÂÎ ‘stars’ in ÌÈ·ÎÂÎ ¯˘Ú „Á‡Â ÈÏ ÌÈÂÁ˙˘Ó (Gen 37:9), interpretively following Classical Arabic structure. However, the participle ÔÈ„‚‡Ò ‘bowing down’ that follows the counted noun is in the plural, imitating the Hebrew plural participle ÌÈÂÁ˙˘Ó ‘bowing down’ in a verbatim manner.

302

CHAPTER NINE THE SEGMENT LEVEL Following the path of inquiry in the last two chapters, this chapter analyzes the ¡ur¥˙ of Genesis, Esther, and the Passover Haggadah, scanning selected examples through the segmental level and following the model presented in table 4 of chapter 3 (pp. 81–82). This chapter emphasizes Judeo-Arabic orthography and phonetics/phonology and the connection between them.1 As mentioned in chapter 4 (pp. 96–99), Judeo-Arabic orthography ought to be considered cautiously when reconstructing phonetic and phonological structure. When the orthography, however, is supported by additional evidence, phonetic and phonological features can be established with greater certainty. Many of the examples presented in this chapter are connected to the Hebraized orthography (Hary 1991a:124, 1996c:732), where Hebrew influenced Judeo-Arabic orthography in the late periods of the religiolect. The Hebrew influence on the orthography reflects a tendency toward literal translation, because the Judeo-Arabic Hebraized orthography is closely associated with the Hebrew sacred texts. Conversely, the features discussed in this chapter are also connected to spoken Egyptian JudeoArabic elements. By adhering more to the colloquial, the ¡ar˙an has turned to the interpretive mode, and he has done so in order to make the translated sacred texts more comprehensible to his readers.

1

I have discussed the connection between Judeo-Arabic orthography and phonetics/phonology in chapter 4, pp. 96–99, as well as in Hary 1990, 1991a, 1996b, and elsewhere.

CHAPTER NINE Note that throughout the chapter, phonemic transcription appears, as is customary, between slashes / / and allophonic transcription between square brackets [ ] (see above, p. xix). Assimilation Level: Segment; Category 13: Orthography/Phonology Feature 13-1 — Assimilation Int.: The following is evidence in the texts of the ¡ur¥˙ that assimilation processes have occurred. This can be seen as a sign of an interpretive translation process, whereby the ¡ar˙an represented colloquial use in the texts. (13-1.1) In Genesis partial assimilation occurs in ÏÂ¡˙˙ /titxul/ (15 36b,7) ‘enter’ as devoicing /d/ > /t/ (see p. 107, 1.10.1). (13-1.2) In Esther the verb /xammim/ ‘imagine’ may reflect a full assimilation process [n] > [m] in the Arabic verb /xammin/ in ¬Ò¥ È¥ ÈÓÓ¡˙ ‡Ï (1302 4a,6) ‘do not imagine in your soul.’2 L/I: (13-1.3) In Genesis the regressive voicing shift /x/ > [\] may occur at times: Ì‡ÊÂµ Ï‡ /il-\u zåm/ (15 3a,14; 4a,2) ‘the nose ring.’ This can be seen as a voicing assimilation in the environment of the voiced [z], compared to the literary form /xuzåm/, although Ì‡ÊÂ¡ /xuzåm/ (15 3a,5) and ‰ÓÊ¡‡ Ï‡ /il-axzima/ (15 18b,19), without the assimilation, occur in the text as well. (13-1.4) The opposite regressive devoicing shift also appears: /\/ > [x]. This can be seen as a voicing assimilation in the environment of the voiceless [s]: ‰ÈÏ‚¯ ÏÒ¡ÈÏ /li-yixsal rigl∑h/ (15 3a,19) ‘to wash his feet’ and ÂÏÒ¡‡Â (15 18b,15) ‘and they washed,’ although µ /\/ occurs in other examples of the verb: ÌÂ‰ÈÏ‚¯ ÂÏÒµÂ /wi-\aßalu rigl∑hom/ (15 29b,14) ‘and they washed their feet’; Â‰‚Â ÏÒµÂ (15 30a,3) ‘and he washed his face’; È¥ ÏÒµ „È· Ï‡ (15 36b,15) ‘wash in wine.’

2

However, there is also the possibility of a scribal error in the manuscript.

304

THE SEGMENT LEVEL: EMPHATIZATION AND DEEMPHATIZATION Emphatization and Deemphatization Feature 13-2 — Emphatization and Deemphatization Int.: In the following examples, despite the fact that the Hebrew names were copied into the Judeo-Arabic, their spelling illustrates the possible emphatization that might have occurred in the emphatic environment in a more interpretive translation. (13-2.1) In Genesis: /d/ becomes [∂] in Â¯ƒÁ„ /da˙∂ar¢u/ (15 10a,13) ‘they caused to roll down,’ probably because of the environment of the emphatic /r¢/. (13-2.2) In the Haggadah: in ‰ËÂˆ /ßø†na/ (3 9,17) ‘our voice,’ /t/ becomes emphatic [†] in the environment of emphatic /ß/. In another example, ı¥¯Î /kar¢faß/ (74 21,2) ‘Karpas’ (greens for the Passover Seder) translates the Hebrew ÒÙ¯Î, and the emphatization /s/ > [ß] may have occurred in the environment of [r¢] . A similar emphatization occurs in ¯∂Ú Ï‡ /el>aΩar¢/ (93 15-b,10) ‘Elazer’; ¯∂Ú ÈÏ‡ /eli>eΩer¢/ (93 15,9) ‘Eliezer’; and in ‰È¯∂Ú (93 15,9 and more) />aΩar¢ya/ ‘Azariah,’ with the emphatization of /z/ > [Ω] in the environment of [r¢]. More infrequently, deemphatization may occur in the environment of plain obstruents. (13-2.3) In Genesis: Â·ÂÏ˙˙ /til¥bu/ (15 29a,7) ‘you seek,’ where /†/ becomes [t] after plain /t/, compared with the literary form /ta†l¥bu/. (13-2.4) In Esther: ¯„¡‡ /axdar/ (1302 1b,8) ‘green,’ where /∂/ > [d] following plain /x/, compared with the literary form /ax∂ar/. (13-2.5) In the Haggadah: emphatic /ß/ becomes plain /s/ in ÚÒÈÏ /li-yisna>/ (91 3a,17) ‘for him to do,’ compared with the literary form /yaßna>/. L/I: (13-2.6) In Genesis deemphatized [t] occurred in È˙¥ÈÈ˙ /tayfati/ (15 3b,7) ‘my family’ and in ¬‡È¯˙ /tar•
CHAPTER NINE /†ar• [†], influenced by the environment of the emphatic [r¢]. Elision Feature 13-3 — Elision Int.: The disappearance of the hamza is common in spoken Egyptian Judeo-Arabic, and where it is evident in the texts of the ¡ur¥˙, it indicates an interpretive mode of translation: (13-3.1) In Genesis: ÔÏÈ‡˜ /qåylan/ (15 0-1,13 and more) ‘saying,’ compared to literary Arabic /qå /a/ and the hamza is elided (see 13-6.5). (13-3.3) In the Haggadah: both ‰Ó‰Â¥ /fuhama/ (3 3,5) ‘wise people,’ compared to literary Arabic /fuhamå/ in an interpretive mode. 3

4

Note that the vowel /a/ is represented here by the letter ‰, probably influenced by Hebrew script as part of the Hebraized orthography (Hary 1996c:732). This represents a phonetic spelling, which is one part of the Hebraized orthography of Late Judeo-Arabic (ibid.).

306

THE SEGMENT LEVEL: ORTHOGRAPHIC MARKING OF THE GLIDES Orthographic Marking of the Glides Feature 13-4 — Orthographic marking of the glides Lit.: This phenomenon is connected to the Hebraized orthography used in premodern Egyptian ¡ur¥˙ represented in this volume. As mentioned in Hary (1996c:732), marking the glides by two vavs and two yods to denote /w/ and /y/ respectively is part of this orthographic tradition. This marking is in line with a literal translation technique, as it is heavily influenced by Talmudic orthography, where this marking occurs. (13-4.1) In Genesis: ·ÂÂ‡‚È (15 25b,8) ‘answer’; È‡„ÂÂ‰ (15 25b,9) ‘here I am’; ‰ÏÎÂÂ ÏÈÎÂÂÈÂ (15 26a,14–15) ‘appoint agents,’ as well as √ÈÈ·‡ (15 12b,11; 13b,3; 36b,17) ‘white’; È˙ÈÈ„‰Ø˙ÈÈ„‰ (15 17a,2; 30a,9) ‘(my) gift’; and ˙‡·ÈÈËØÔÈ·ÈÈËØ·ÈÈË (15 2b,7; 25b,18; 1b,20; and many other cases) ‘good,’ although the two yods may also refer to a ¡adda; however, ÔÈ·ÈË (pl.) (15 25b,20; 26a,3; 26a,16) ‘good’ with one yod also appears (see below, 13-4.4). (13-4.2) In Esther: ¯‡˜ÂÂ‡ (1302 6a,21) ‘glory’; ÔÈ„Â‚ÂÂÓ (1302 1b,7) ‘present (pl.)’; È˙„ÏÂÂÓ (1302 6a,3) ‘my birth place, my family’ as well as ˙ÈÈ¯¡¥ (1302 1b,6) ‘pride’; Â˙ÈÈ¯·Î (1302 1b,6) ‘his majesty’; √ÈÈ·‡ (1302 1b,8) ‘white’; and ÔƒÈÈ‡ (1302 1b,13) ‘also.’ (13-4.3) In the Haggadah: ‰„‡ÂÂÊ (3 18,16; 74 10,12; 91 8b,7; 93 40,6) ‘provisions, supplies’; ‰ÂÂ¯‡ˆÓ (3 20; 74 10,14; 91 8b,9) ‘Egyptians’; È¯‡ÂÂÈÏØÈ¯ÂÂÈ®Ï© (3 19,6; 74 11,1; 91 8b,11; 93 40,16) ‘to look (upon)’; as well as ˙‡ÈÈÁ (74 10,15; 91 8b,9; 93 40,15) ‘life’; ‰ÈÈÒ‡˜Ø‰ÈÈˆ‡˜Ø‰ÈÈÒ‡ (74 10,16; 91 8b,13) ‘harsh’; and Ô‡ÏÈÈ‡˜ØΩ‡ÏÈ‡˜ (74 11,2; 91 8b,15; 93 41,11) ‘saying.’ Int.: At times, the ¡ar˙an has marked the glides with one vav or one yod, a choice that is less influenced by the Hebraized orthography and thus approaches the interpretive mode. (13-4.4) In Genesis examples of /w/ denoted with one vav are

307

CHAPTER NINE rare: Â‡ (15 30b,2) ‘or,’ although ÂÂ‡ ‘or’ appears more frequently (15 4a,6; 4a,14; 31a,1; and more); and ‰„Â‰ (15 18a,11) ‘here,’ although it is very rare; ‰„ÂÂ‰ (15 2b,11; 2b,15; 3b,13; and many more occurrences) ‘here’ with two vavs is much more common. For /y/ with one yod: Ì‡È‡ (15 12,13) ‘days’; Ω‡ÏÈ‡˜ (15 3b,5) ‘saying’; and ÔÈ·ÈË (15 25b,20; 26a,3; 26a,16) ‘good (pl.).’ (13-4.5) In Esther there are some examples where the /w/ is indicated with one vav: ¯‡˜Â‡ (1302 1b,6; 4b,6; 4b,9; and more) ‘glory,’ although ¯‡˜ÂÂ‡ (1302 6a,21) occurs as well; È‡Â‡ (1302 1b,6; 4b,6; 4b,9; and more) ‘vessels’; È‡ÏÂ‡ (1302 3b,1) ‘first,’ although see ‡ÏÂÂ‡ (1302 1b,21) ‘first’; „‡ÏÂ‡ (1302 4b,7; 6a,12; 7a,7) ‘children,’ although „‡ÏÂÂ‡ (1302 7a,6) ‘children’ does exist. The glide /y/ denoted with one yod is more rare: Ì‡È‡ (1302 2b,10) ‘days,’ although Ì‡ÈÈ‡ (1302 1b,2; 1b,3; 1b,6; and more) is much more prevalent; ÔÈ‡„Ó (1302 2a,9; 6b,6; 6b,8; and more) ‘provinces,’ but see also ÔÈÈ‡„Ó (1302 6b,3) with two yods; È˘‡ÂË ÔÈ˙‡ (1302 2b,23; 4b,15) ‘two eunuchs’ signifies contraction of the diphthong in /itn∑n/. (13-4.6) In the Haggadah the glide /w/ is sometimes spelled with one vav: Â‡ (3 2,12) ‘or,’ although ÂÂ‡ (74 1,7) also occurs; ÂÏÂ (3 2,11) ‘even,’ although ÂÂÏÂ (93 13,2) also appears; ˙¯ÂÈÏ (93 19,10) ‘to inherit,’ although ˙¯ÂÂÈÏ (3 6,9) appears too. Similarly, the /y/ is marked with one yod, either to indicate the contraction of the diphthong, as in ·Â·ÁÓ ÔÈ˙‡· (93 83,4) ‘for two gold pounds,’ or as a spelling device: ÈÏ‡ÈÏ (3 2,10) ‘nights,’ although ÈÏ‡ÈÈÏ (74 1,6; 93 13,6) also appears; ÔÈÈÎ˙Ó (3 2,16) ‘reclined,’ although ÔÈÎ˙Ó (93 13,8) also occurs. L/I: The literal/interpretive tension in marking the glides is evident throughout the different manuscripts of the various Egyptian Judeo-Arabic ¡ur¥˙ examined in this volume and in Hary 2009. Mss. 74, 91, and 93 of the Haggadah frequently mark the glides with two vavs and two yods to denote /w/ and /y/

308

THE SEGMENT LEVEL: ORTHOGRAPHIC MARKING OF THE GLIDES respectively, as influenced by the Talmudic orthography, in an attempt to be closer to the literal mode of translation. Ms. 3 of the Haggadah, however, approaches the interpretive end of the literal/ interpretive continuum and often denotes the glides with one vav and one yod. The issue in ms. 15 (Genesis) and ms. 1302 (Esther) is a bit more complicated. In ms. 15 the ¡ar˙an has marked /w/ with two vavs, approaching the literal end of the continuum, but he has usually marked /y/ with only one yod, representing a more interpretive mode. He sometimes, however, has marked the /y/ with two yods, but usually this marking denotes a gemination (¡adda). The situation is quite the opposite in ms. 1302. There, the ¡ar˙an has marked /y/ with two yods, approaching the literal end of the continuum, but he usually—not always—marked /w/ with only one vav, representing a more interpretive mode. The following are just a few examples of each of the glides: the literal/interpretive tension is exhibited by marking the glides with two vavs and two yods in a literal mode in mss. 15, 74, 91, and 93 and by not marking the glides in this way in a more interpretive practice in ms. 3. Ms. 1302 exhibits here its own particularities. (13-4.7) Marking /w/: „‡ÏÂÂ‡ (74 1,3; 91 4b,12; 93 14,4; 15 0-1,1) ‘children’ contains two vavs marking /w/ in a literal mode, whereas „‡ÏÂ‡ (3 2,7; 1302 6a,125 ) approaches the interpretive mode by using the Arabicized orthography and not marking the glide. The same phenomenon occurs in the following: ÂÂÏÂ (74 2,15; 93 13,2; 15 36b,14) ‘if,’ as opposed to ÂÏÂ (3 2,11); ÂÂ‡ (74 1,7; 93 57,6; 15 4a,6) ‘or,’ as opposed to Â‡ (3 2,12); Ø¯‡˜ÂÂ‡ ¯‡˜ÂÂ (74 11,12; 91 14b,14; 93 81,5; and even in 3 20,6) ‘glory,’ as opposed to ¯‡˜Â‡ (1302 1b,6; 4b,6; 4b,9; and more).

5

The usual spelling in 1302 is indeed „‡ÏÂ‡ with one vav (4b,7; 6a,12; 7a,1; 7a,7); however, there is one place in the manuscript where „‡ÏÂÂ‡ appears with two vavs (1302 7a,6).

309

CHAPTER NINE (13-4.8) Marking /y/: ˙ÈÈ„˘Ø˙ÈÈ„È˘ (74 2,16; 91 2b,6; 93 17,10) ‘might’ includes two yods marking the palatal glide /y/ using a literal mode, whereas ˙È„˘ (3 5,5) employs the Arabicized orthography, marks the glide with only one yod, and approaches the interpretive mode. The same occurs in the following: „ÈÈ· (74 1,10) ‘with a hand,’ as opposed to Â„ÈØ„È È¥ (15 30b,14; 3 2,20); ‰ÈÈÂ„Ø‡ÈÈÂ„ (74 2,5; 93 15,15) ‘world,’ as opposed to ‰ÈÂ„ in 3 4,3; 15 37a,14; ÈÏ‡ÈÈÏ Ï‡ (74 1,5; 93 13,1) ‘the nights,’ as opposed to ÈÏ‡ÈÏ Ï‡ (3 2,10); and √ÈÈ·‡ (15 12b,11; 1302 1b,8) ‘white,’ with two yods in a literal mode. This literal tendency is not always the case in mss. 15 and 1302, although it is more frequent in 1302 (see also above): 6 Ì‡ÈÈ‡ (1302 1b,2 and many other cases) ‘days,’ as opposed to Ì‡È‡ (15 12,13; 1302 2b,10, only in one place) ‘days.’ Another example is ÔÏÈÈ‡˜ (91 2b,9); Ô‡ÏÈÈ‡˜ (93 18,3) ‘saying’ with two yods; and ÔÏÈ‡˜ (3 5,10) and Ω‡ÏÈ‡˜ (15 3b,5; 7b,7; 10a,5; and many other cases) with one yod.7 Diacritic Marks Feature 13-5 — Diacritic marks Lit.: (13-5.1) In rare cases the ¡ar˙an has marked Arabic få< with a pe without a supralinear dot, moving in the direction of a literal translation and shying away from imitating Arabic orthography, where the supralinear dot exists (·):8 In Genesis: ˙ÙÈ (15 0-1,18) ‘Yefet’ and ÈÙ (15 38a,5) ‘in’ and in Esther: 9 ÏÚÙÂÈ (1302 1b,21) ‘be done’; ˜ÂÙ (1302 3a,5) ‘above’; and ÈÙÈÏ (1302 3a,11) ‘to eliminate.’ The pe without a supralinear dot in ÍÂÙ˘ Ï‡ (3 6

7 8 9

In the Ì‡ÈÈ‡ ‘days’ example, the two yods may mark the gemination (ta¡d•d) and not the glide, but this is not the case in ÔƒÈÈ‡ /ay∂an/ (1302 1b,13) ‘also,’ where the two yods clearly mark the glide [y]. In only one place in ms. 15 does Ω‡ÏÈÈ‡˜ appear with two yods: 15 3a,16. Of course, this kind of marking could also be interpreted as a scribal error. This feature is much more widespread in Esther.

310

THE SEGMENT LEVEL: HEBREW-INFLUENCED ORTHOGRAPHY colophon) ‘the (prayer) Pour Your Wrath’ represents the voiceless fricative labiodental /f/, but the word ÍÂÙ˘ is in Hebrew. There are other cases where pe without a supralinear dot appears, but they indicate the voiceless stop bilabial /p/, not the fricative /f/, sometimes using Hebrew lexemes in the JudeoArabic text. In Genesis: ˘„ÏÙ (15 1a,9) ‘Pildash’ and Ô„Ù (15 5b,1) ‘Paddan’; in Esther: Ò¯Ù (1302 2a,3) ‘Persia’; in the Haggadah: ÁÒÙ (93 2,9 and many other occurrences) ‘Passover’; ˜ÂÒÙ Ï‡ (3 3,21 and many other occurrences) ‘the verse’; and ‰Ú¯Ù (3 7,8 and more) ‘Pharaoh.’ Int.: (13-5.2) In an interesting example, the ¡ar˙an has added the madda to the Hebrew alef in ÏÎ° ÌÏ (15 3a,20) ‘I will not eat,’ following standard Arabic orthography in an interpretive mode. L/I: (13-5.3) In ¬ÂÙ¯ÚÈ (91 10b,4) ‘(they will) know you’ the spelling of Ù without the supralinear dot represents a literal tendency, because it can be an imitation of the Hebrew orthography fe (Ù). However, in ¬Â¥¯ÚÈ (93 63,10) the spelling of Judeo-Arabic ¥ with a supralinear dot demonstrates the interpretive tendency, as this is part of the orthographic tradition of standard Arabic, namely, indicating the få< with a supralinear dot (chapter 3, p. 78, ii). Hebrew-Influenced Orthography Feature 13-6 — Hebrew-influenced orthography As is the case in Feature 13-4 above, this phenomenon is intimately connected to the Hebraized orthography used frequently in premodern Egyptian ¡ur¥˙. However, the distribution of this phenomenon in the various manuscripts is not the same as that of Feature 13-4. For example, ms. 3 consistently marks Arabic final alif with a he in clear imitation of Hebrew orthography, as the final /-a/ sound in Hebrew is often marked with a he (Hebrew ‰). Consequently, marking final alif with a he (Judeo-Arabic ‰) is an indication of literal translation, as it 311

CHAPTER NINE is so influenced by Hebrew orthography. On the other hand, in Feature 13-4, ms. 3 exhibits some interpretive modes by frequently not marking the glides with two vavs or two yods. The following examples illustrate this phenomenon of Hebrewinfluenced orthography, where Genesis (ms. 15) as well as mss. 3 and 74 of the Haggadah usually exhibit the Hebrew-influenced phenomena in a literal mode. Esther (ms. 1302), on the other hand, does not demonstrate Hebrew-influenced features, thus revealing a more interpretive mode. Mss. 91 and 93 of the Haggadah are less consistent: they usually do not exhibit the Hebrew-influenced orthography in an interpretive mode, but not always. L/I: (13-6.1) Marking final alif: In Genesis (ms. 15) as well as in the Haggadah (ms. 3) the final alif is marked with a he (‰), for the most part, as an imitation of Hebrew spelling for final /-a/ sound, and as part of the Hebraized orthography in a literal mode. On the other hand, Esther frequently denotes the final alif with an alef (‡) as part of the Arabicized orthography in an interpretive mode. Mss. 74, 91, and 93 of the Haggadah are not always consistent, although ms. 74 tends to mark the final alif literally with a he (‰) while mss. 91 and 93 tend to mark it interpretively with an alef (‡©Æ For example, ‰„ÂÂ‰ (15 2b,11; 2b,15; 3a,16; and many more occurrences; 3 3,7; 3,18; 74 1,14; 2,2; 91 6a,1)1 0 ‘indeed, here,’ as opposed to ‡„ÂÂ‰ (1302 4b,20; 5b,10; 6a,4; 3 12,5; 74 18,15; 91 14a,7; 93 14,8; 15,10; 27,10; 78,12); ‰‰ (15 8b,5; 23b,12; 3 2,5; 2,6; 74 1,2; 1,3; 93 8,12; 9,4; 78,9) ‘here,’ as opposed to ‡‰ (15 10a,13; 21b,7; 27b,11; 29a,9; 93 12,6; 12,7); ‰ÏÂÎ (15 27b,6; 3 3,4; 74 1,12) ‘all of us,’ as opposed to ‡ÏÂÎ (93 13,8) in the Arabicized orthography in an interpretive mode; ‰ÂÎ (15 28a,17; 3 3,3; 74 1,10) ‘(we) were,’ as opposed to ‡ÂÎ (93 14,1); ‰Â·‡ (15 13a,7; 28a,18; 29b,19; 31a,18; and more; 3 10

Exceptions exist in these manuscripts; see the examples below for details.

312

THE SEGMENT LEVEL: HEBREW-INFLUENCED ORTHOGRAPHY 6,14; 7,8; 93 59,5; 61,2) ‘our father,’ as opposed to ‡Â·‡ (91 3a,8; 3a,18; 93 20,3; 21,4); ‰˙‡‰·‡ (15 33b,18; 34a,5; 3 2,3; 3,2; 5,17; 7,2; 9,16; and many other occurrences; 74 1,1; 1,11; 5,4; and more; 91 5a,10; 5a,12; 8a,19; and more; 93 58,8; 58,9; 60,4; and many more occurrences) ‘our fathers,’ as opposed to ‡˙‡‰·‡ (91 2b,14; 3a,13; 8b,16; and more; 93 14,3; 18,14; 20,11; and more); ‰ËÂˆ (3 9,17; 10,8; 74 5,4; 5,9) ‘our voice,’ as opposed to ‡ËÂˆ (91 5a,11; 5a,16; 93 24,7; 25,3). Another prevalent example is ‰≤‡‰ (15 2a,18; 2b,6; 4a,18; and many more occurrences; 3 2,2; 74 20,3; 93 60,9; 82,14; 82,15) ‘this,’ as opposed to ‡≤‡‰ (1302 2a,3; 2a,11; 2b,11; and many other occurrences) and ‡≤‰ (93 12,4; 28,15); the particle ‡ÓÏ (1302 2a,1; 2a,11; 2b,2; and more) ‘when’ follows the usual trend (although ‰ÓÏ occurs once in Esther [1302 1b,14]). In somewhat different manuscript behavior, the relative pronoun ‰Ó (74 1,17) ‘that’ ends with a final he (‰), marking final alif in imitation of Hebrew in the literal mode, whereas in other manuscripts (15 31b,13; 1302 3a,7; 4a,3; 4a,17; 3 3,14; 93 15,2) the pronoun is spelled ‡Ó with a final alef in the interpretive mode. The interrogative pronoun ‡≤‡Ó (1302 1b,21) ‘what’ does follow the usual trend. ‰≤‡ÓÏ (93 64,7) ‘why,’ with a final he, points to the inconsistency of ms. 93 (see also below, p. 319, 13-6.7). (13-6.2) Marking final vowel /-a/: It is often difficult to decide whether he (‰) and alef (‡) mark final alif or final vowel /a/. In Genesis (ms. 15) as well as in the Haggadah (ms. 3), for the most part final vowel /-a/ is marked with a he as an imitation of the Hebrew spelling for a final /-a/ vowel, and as part of the Hebraized orthography in a literal mode. On the other hand, Esther (ms. 1302) marks the final alif with ‡ as part of the Arabicized orthography in an interpretive

313

CHAPTER NINE mode.1 1 Mss. 74, 91, and 93 of the Haggadah are not always consistent, although ms. 74 tends to mark the final alif literally with a he (‰) while mss. 91 and 93 tend to mark it interpretively with an alef (‡). For example, ‰„·‡ (3 5,8) ‘begin’ as opposed to ‡„·‡ (91 2b,8); ‰„Î (15 3a,16; 13a,15; 13a,17; and more) ‘this way’ as opposed to ‡„Î (1302 1b,12);1 2 ‰˙‡ (15 3b,16; 7a,10; 8a.4; and many more occurrences; 3 4,14; 5,8; 14,5; and more; 74 2,14; 12,3; 12,9; 91 2b,8; 6b,10) ‘you’ as opposed to ‡˙‡ (1302 3a,8; 91 2b,3; 9b,10; 93 16,12; 18,2; 31,9; and more). (13-6.3) Marking alif maqß¥ra bi-ß¥rat il-yå<: The trend shown above in 13-6.1 continues here. The ¡ur¥˙ of Genesis (ms. 15) and the Haggadah (mss. 3 and 74) mark, for the most part, alif maqß¥ra bi-ß¥rat il-yå< with a he (‰) in imitation of the Hebrew orthography, in a literal translation mode. The ¡ar˙ of Esther (ms. 1302), on the other hand, marks alif maqß¥ra bi-ß¥rat il-yå< with an alef (‡) in a more interpretive manner. As above, mss. 91 and 93 are less consistent, although ms. 93 tends to denote it with an alef (‡). For example, the accusative marker ‰Ï‡ (15 0-1,3; 1b,4; 1b,5 and many more occurrences; 3 3,2; 3,21; 4,19; and more;1 3 74 1,11; 2,4; 2,13; and more) and ‰ÏÈ‡ (91 2b,2; 2b,3; 3a,3; and more) are spelled with a final he (‰) to mark the alif maqß¥ra bi-ß¥rat il-yå< in literal imitation of the Hebrew as part of the Hebraized orthography. On the other hand, mss. 93 (14,3; 15,12; 17,4; and more) and 1302 (1b,6; 1b,16; 2a,3; and many more 11

This spelling can also be considered a literal translation since it may be part of the Hebraized orthography, as influenced by the spelling of the Babylonian Talmud (Hary 1991a:124). 12 This spelling may, however, indicate final alef ( ‡), as ‡≤Î ( 1302 1b,18; 3a,6) appears in the manuscript, probably reflecting Arabic «c!. 13

‡Ï‡ in 3 7,9 represents /illå/ ‘except’ (with a final alif). In 74 6,1 even this word is spelled with a final he (‰) ‰Ï‡.

314

THE SEGMENT LEVEL: HEBREW-INFLUENCED ORTHOGRAPHY occurrences) exhibit ‡Ï‡ in a more interpretive mode as part of the Arabicized orthography. Ms. 91 shows inconsistency, where ‡ÏÈ‡ (91 3a,1; 3a,2) appears, as opposed to ‰ÏÈ‡, shown above. The same holds true for the preposition ‰ÏÚ (15 0-1,18; 2a,9; 2b,5; and many more occurrences; 3 7,9; 7,14; 9,4; and more; 74 5,19; 6,9; 7,1; and more; 91 3a,18; 4b,1; 5a,3; and more) ‘on,’ as opposed to ‡ÏÚ (1302 1b,3; 1b,9; 1b,10; and many more occurrences; 911 4 5b,10; 5b,11; 93 21,4; 21,13; 23,9; and more). The particle ‰˙Á (3 2,11; 3,4; 22,14; 74 1,12; 22,1; 91 5a,2; 10a,13-‰˙ÈÁ) ‘even’ is spelled with a final he (‰) to mark the alif maqß¥ra bi-ß¥rat il-yå< in literal imitation of the Hebrew as part of the Hebraized orthography. On the other hand, ‡˙Á appears in 1302 (4b,8), 74 (1,6), and 93 (13,2; 14,5; 23,8) in a more interpretive mode. The following verb follows the regular trend: ‰ˆÂ (15 6b,2; 9a,19; 15a,20; and many more occurrences; 3 4,13; 10,18; 74 2,10; 5,13) ‘command’ with a final he (‰) as opposed to Ø‡ˆÂ ‡ˆÂÂ‡Ø‡ˆÂ‡ (1302 2b,7; 3a,6; 6a,8; and more; 91 5b,3; 93 16,12; 25,13) with a final alef (‡). The verb ‰Ï‡ÚÂ˙Ø‰Ï‡Ú˙ ‘may (He) be exalted,’ on the other hand, appears only with a final he (‰) (15 22a,2; 3 4,6; 5,19; 9,3; and more; 91 2b,15; 5a,1; 93 16,4; 18,15; 23,7; and more). Ms 74 usually writes Â¢˙ (2,7; 8,5; 9,15) for /ta>ålå/, although ‰ÏÚ˙˙ with a final he (‰) also occurs (74 19,5). Unlike the above trend, the verb ‡ËÚ‡ ‘give’ consistently appears with a final alef ‡ in ms. 3 of the Haggadah (3 16,6; 16,7; 16,18; and more) in an interpretive mode. The same holds true for ms. 93 (36,1; 36,2; 36,13; and more), where ‡ËÚ‡ alternates with ‡ËÚ, both with a final alef ‡. Ms. 74 of the Haggadah, on the other hand, consistently denotes the verb 14

These examples are the exception in ms. 91, but they still show the ¡ar˙an‘s inconsistency in this manuscript.

315

CHAPTER NINE with a final he (‰) in a literal manner, following the trend mentioned above: ‰ËÚ (74 8,12; 8,14; 9,5; and many more occurrences). Ms. 91 is inconsistent again: ‡ËÚ‡ (91 7b,19; 8a,1; 8a,7) and ‰ËÚ‡ (91 7b,5). This verb follows the usual orthographic tendencies of the ¡ar˙ of Genesis with the final he (‰) in a literal mode: ‰ËÚ (15 3b,2; 3b,4; 4b,17; and many other cases) and that of the ¡ar˙ of Esther with the final alef (‡): ‡ËÚ (1302 2b,20; 3b,20; 5b,14; and more), as well as other variants: ‡ËÚ‡ (1302 3b,6; 3b,8; 4a,18; and more) ‘was given’ and ‡ËÚÈ (1302 2a,15; 2b,11; 5a,5; and more) ‘will be given.’ Atypical to ms. 3 is ‡¯‚ÂÓ (3 3,10) ‘event,’ which is spelled with a final alef (‡), representing the interpretive mode as part of the Arabicized orthography. ‰¯‚ÂÓ (74 1,15), on the other hand, follows the trend mentioned above for ms. 74. (13-6.4) Marking alif maqß¥ra bi-ß¥rat il-alif: ‰ÈÈÂ„Ø‰ÈÂ„ (15 37a,14; 3 4,3; 21,2; 74 2,5; 12,3) ‘world’ is spelled with a final he (‰) to mark the alif maqß¥ra bi-ß¥rat il-alif as part of the Hebraized orthography. This is done because the final /-a/ vowel of a word is usually spelled with a he (‰) in Hebrew, and in order to imitate this spelling behavior, the ¡ar˙an has done the same in Judeo-Arabic in a literal mode of translation. On the other hand, mss. 91 9b,10 and 93 15,15; 45,2 exhibit ‡ÈÈÂ„ in the Arabicized orthography in a more interpretive mode. These variant spellings of /dunyå/ continue the trend established above for the different manuscripts of the ¡ur¥˙. Other examples follow this trend. From Genesis: ‰È‡¯ÂØ‰È‡¯ÂÂ (15 2a,18; 3b,8; 14b,3) ‘behind me’ and ‰È‡È‡ (15 4a,15; 7a,6; 12a,7; and more) ‘me’ (direct object). From Esther: ‡„ (1302 2b,9; 2b,14) ‘neared’ and ‡ÿÏÚ (1302 3a,4; 4b,7) ‘elevated.’ (13-6.5) Marking alif mamd¥da (see 13-3.2 and 13-3.3): In Genesis, as seen above, the examples follow the Hebraized orthography, where the alif mamd¥da is marked with a he (‰) in a literal translation: ‰„˙·‡ (15 29b,4; 29b,7) ‘beginning’; 316

THE SEGMENT LEVEL: HEBREW-INFLUENCED ORTHOGRAPHY ‰¯ÊÂÂ (15 26b,8) ‘ministers’; ‰¯Â (15 4b,2; 16a,7; 19a,2; and more) ‘behind’; and ‰Î· (15 30a,3; 31b,6) ‘crying.’ In ‡ÂÂÒ (15 20a,9) ‘together’ the alif mamd¥da is denoted with an alef (‡), contrary to the trend in Genesis. In Esther the alif mamd¥da is marked with an alef (‡) in an interpretive mode. For example ‡¯ÊÂ (1302 1b,5; 6a,8) ‘ministers’; ‡Ò‡Â¯ (1302 1b,5; 1b,20; 1b,23; and many other occurrences) ‘officials’; ‡Ò (1302 2a,2; 2a,14; 2a,15; and many more examples) ‘women’; ‡µ (1302 1b,6) ‘wealth’; ‡ÏÎÂ (1302 2a,13) ‘deputies’; and ‡·˙Î (1302 3b,1) ‘scribes.’ Exceptions such as ‰ÒÂ¯ (1302 1b,16) ‘officials’ and ‰¯ÊÂ (1302 3b,2) ‘ministers’ also occur as part of the Hebraized orthography, an indication of a literal translation, as part of the literal/interpretive tension. The different manuscripts of the Haggadah follow, for the most part, earlier trends, but with more inconsistencies: ‰Ó‰Â¥ (3 3,5) ‘understanding’; ‰ÓÏÂÚ (3 3,5; 74 1,13; 93 14,5) ‘learned people’; ‰„˙·‡Ø‰„˙·È‡ (3 5,17; 91 2b,14) ‘beginning’; ‰Ó¯ÂÎ (3 20,9; 20,10; 93 43,9) ‘nobles’; and ‰¯ÂÂÏ (3 20,16; 20,18; 74 11,17) ‘backward.’ On the other hand, the following examples display the alif mamd¥da with a final alef (‡), as part of the Arabicized orthography in an interpretive mode: ‡Ó‰Â¥ (74 1,13; 93 14,6) ‘understanding’; ‡„˙·‡ (93 18,14) ‘beginning’; ‡ÂÂÒ (3 2,16; 74 1,9; 93 13,7; 62,2) ‘whether’; and ‡¯‡ÂÂÏØ‡¯ÂÂÏ (91 9b,3; 9b,5; 93 44,4; 44,6) ‘backward.’ (13-6.6) Marking the tå< marb¥†a: In general, the tå< marb¥†a is denoted with a he (‰) as part of the Arabicized orthography in an interpretive translation mode in Genesis, Esther, and mss. 3 and 74 of the Haggadah. Once again, mss. 91 and 93 show some inconsistencies, where the tå< marb¥†a can be marked alternatively with a he (‰) or with an alef (‡). For example, ‰ÏÈÏ (15 6a,20; 12a,1; 12a,4; and more; 1302 14b,14; 3 2,9; 2,11; 2,13; and more; 74 1,5; 1,6; 1,7; and more; 317

CHAPTER NINE 91 5b,13; 9b,12; 93 13.1; 13,3; 13,4; and more) ‘night’; ‰ƒ¥ (15 2a,2; 2a,3; 3b,3; and more; 1302 1b,9; 1b,10; 3a,17; and more; 3 23,12; 74 13,6; 91 10b,10; 93 64,9) ‘silver’; and ‰ÈÈ·ˆ (15 3a,12; 1302 2a,15; 2b,2; 2b,5; and more) ‘young woman.’ Infrequently, though, as part of a more literal translation, the tå< marb¥†a may be denoted with an alef (‡), probably influenced by the spelling conventions of the Babylonian Talmud and as part of the Hebraized orthography (Hary 1991a:124, 1996c:732). In Genesis this feature is rare: ‡ˆÈÏ¡˙ (15 31b,15) ‘deliverance’ and ‡‰È¯Î (15 30a,6) ‘calamity’ end with an alef (‡), the latter probably because of the wish to avoid two consecutive hes. The ¡ar˙an of Genesis has sometimes represented names with a final alef (‡), although in Hebrew they end in a he (‰): ‡‰Ï· (15 19b,8; 19b,11) ‘Bilha’; ‡ÊÈÓ (15 20a,17) ‘Miza’; and ‡Ú (15 20a,19; 20a,15; 20a,16; and more) ‘Ana’ (see also p. 324). In Esther: ‡¯Ó (1302 4a,3) ‘woman’ and ‡·ÈÈË (1302 5b,20) ‘good.’ ‡¯Î· (1302 4b,9) ‘tomorrow’ can be explained as standard Arabic, as in /bukra/ ‘early morning,’ but it can also reflect spoken /bukra/. In the Haggadah: ‡ÈÈ·‚‡ (91 2b,14) ‘foreign,’ as opposed to ‰ÈÈ·‚‡/‰ÈÈ·‡‚‡ (3 5,18; 93 18,14); ‡¯˙ÂÎ (91 4b,9; 93 22,8) ‘multitude,’ as opposed to ‰¯˙ÂÎ (3 8,4); ‡ÂÂ·¯ (91 4b,14) ‘myriad,’ as opposed to ‰ÂÂ·¯ (3 8,12; 93 22,15); ‡ÓÈ‰· (91 5b,14) ‘beast,’ as opposed to ‰ÓÈ‰· (3 11,15; 74 6,2; 93 27,2); ‡ÈÈ¯· (91 7b,12; 7b,13) ‘desert,’ as opposed to ‰ÈÈ¯· (3 16,14; 16,15; 17,12; 74 9,1; 9,3; 9,18; and more; 91 8a,9; 93 36,9; 36,10; 73,16; and more); ‡˜È„ (91 5b,6; 5b,7; 7a,4; and more; 93 26,3; 32,10; 32,11; and more) ‘anguish,’ as opposed to ‰˜Èƒ/ ‰˜È„ (3 11,3; 15,3; 15,4; 74 7,17; 18,1; 17,16); ‡¯„Â˜/‡¯ƒÂ˜ (91 6b,15; 6b,17; 31,13) ‘power,’ as opposed to ‰¯„Â˜ (3 14,10; 74 7,6); ‡ÈÈ„Â·Ú (91 9a,14) ‘slavery,’ as opposed to ‰ÈÈ„Â·Ú (3 5,6; 19,18; 74 18,11; 91 2b,7; 14a,1; 93 42,9; 58,12; 78,6; and more); ‡¯‡˘‡ (93 31,15) ‘signal,’ as opposed to ‰¯‡˘‡ (3 14,14); 318

THE SEGMENT LEVEL: HEBREW-INFLUENCED ORTHOGRAPHY ‡˜‡ÓÁ (93 32,10; 32,11; 33,8; and more) ‘anger,’ as opposed to ‰˜‡ÓÁ/‰˜ÓÁ (3 15,3; 15,4; 15,13; and more; 74 7,12; 7,13; 8,1; and more); and ‡ÚË˜ (93 46,5; 47,3; 47,11) ‘piece,’ as opposed to ‰ÚË˜ (93 46,6; 47,11a). Interesting is the spelling of ‡ÏÈË (91 10a,1) ‘wash (the hands)’ with a final alef (‡), as the lexeme is an imitation of the Hebrew ‰ÏÈË. (13-6.7) The spelling of independent pronouns: The orthography of the independent pronouns reflects the literal/interpretive linguistic tension (see also chapter 8, pp. 237–38, 6-1). The first person independent pronoun ‰‡ (15 2a,16; 3 3,18; 74 2,2; 91 5b,17; 93 97,12; and many other cases) follows verbatim the Hebraized orthography with its use of final he (‰) to indicate final alif, whereas the spelling ‡‡ (15 13a,11, plus a limited number of occurrences; 1302 4a,4 and many other cases; 3 6,19; and more; 91 10b,6; and more; 93 20,6; and more) ‘I’ indicates the interpretive direction, as it follows standard Arabic spelling. The spelling of the second singular masculine independent pronoun follows literal trends: ‰˙‡ (15 3b,16; 7a,10; 8a,4; and many more occurrences; 3 4,14; 5,8; 14,5; and more; 74 2,14; 12,3; 12,9; 91 2b,8; 6b,10) ‘you’ is spelled with a final he (‰) as an imitation of the Hebrew spelling and as part of the Hebraized orthography. The spelling of ‡˙‡ (1302 3a,8; 91 2b,3; 9b,10; 93 16,12; 18,2; 31,9; and more) with a final alef (‡), though, may have been influenced by the Babylonian Talmud. This spelling as well is part of the Hebraized orthography (Hary 1991a:124, 1996c:732), used in the literal translation. On the other hand, both spellings indicate a pronunciation of /anta/ or /inta/, which follows regular Arabic structure in interpretive mode (see also chapter 8, p. 237, 6-1.1). For the most part, the spelling of the third singular independent pronoun follows the Hebraized orthography in a literal 319

CHAPTER NINE translation, as it is an exact or almost exact copy of the Hebrew pronoun with the final alef (‡): ‡Â‰ (15 10b,7; 1302 1b,2; 3 3,2; 74 1,1; 93 12,5) and ‡ÂÂ‰ (15 4b,7; 1302 2a,7; 2b,17; 3a,12; and more; 74 2,16; 91 2b,6) ‘he’ (Hary 1991a:124; Blau 1980:57). On the other hand, Â‰ (3 2,3) is spelled in the Arabicized orthography in the direction of the interpretive mode. Similarly, ‡È‰ (15 2a,2 and more; 3 26,19) ‘she’ is also an exact imitation of the Hebrew spelling in a literal translation and, in addition, both ‡ÈÈ‰ (1302 1b,17; 2b,1; 2b,12; and more; 74 15,16; 93 31,12) and ‰ÈÈ‰ (91 12a,19) mark the final vowel /-a/ with an alef (‡) and he (‰) respectively, in a somewhat less literal mode. On the other hand, È‰ (only in 1302 2b,6) is spelled in the Arabicized orthography and, like Â‰, in the direction of the interpretive mode. The first plural independent pronoun ‰Á‡ (15 10a,16; 3 2,12; 3,3; 74 1,6; 10,9; 91 8b,1) ‘we’ is spelled with a final he (‰), indicating the Hebraized orthography in a literal translation; however, the pronunciation /i˙na/ is typical colloquial Egyptian Arabic in an interpretive mode (see chapter 8, p. 237, 6-1.3). (13-6.8) The spelling of personal and place names: It is clear from the different ¡ur¥˙ that, when translating proper names, the ¡ar˙anim were struggling along the literal/interpretive translation scale (chapter 3, pp. 83–85). In literal translation, at times the ¡ar˙anim copied the Hebrew names verbatim into the Judeo-Arabic translation. The following personal names are copied into the Judeo-Arabic ¡ar˙: In Genesis: „¯È (15 0-1,2; 1b,3; 1b,4) ‘Jared’; ÍÂÁ (15 0-1,3; 1b,5; 1b,6; and more) ‘Enoch’; ÁÏ˘Â˙Ó (15 0-1,6; 1b,9; 1b,10; and more) ‘Methushelah’; ÍÓÏ (15 0-1,10; 0-1,12; 0-1,15; and more) ‘Lamech’; ÌÁ (15 0-1,18) ‘Ham’; ˙ÙÈ (15 0-1,18) ‘Japheth’; Ì‰¯·‡ (15 2a,1; 2a,3; 2a,7; and more) ‘Abraham’; ‰˜·¯ (15 2b,15; 3a,13; 3a,15; and more) ‘Rebecca’; and ‰¯˘ (15 2a,9; 4b,9; 5a,4; and more) ‘Sarah.’ 320

THE SEGMENT LEVEL: HEBREW-INFLUENCED ORTHOGRAPHY In Esther: È˙˘Â (1302 1b,9; 1b,17; 1b,22 and more) ‘Vashti’; ÔÓÂ‰Ó, ‡˙Ê·, ‡Â·¯Á, ‡˙‚· (1302 1b,14) ‘Mehuman, Bizta, Harbona, Bigta’; ¯˙Ê (1302 1b,15) ‘Zetar’; ¯˙˘ (1302 1b,19) ‘Shetar’; and ‡˙Ó„‡, ˘È˘¯˙, Ò¯Ó, ‡Ò¯Ó (1302 1b,20) ‘Admata, Tarshish, Meres, Marsena’ all transcribe verbatim the Hebrew names. Notice that both ‡˙µ·‡ (1302 1b,14) ‘Avagta’ and Ô¡ÂÓÓ (1302 1b,20; 1b,22; 2a,9) ‘Memuhan’ indicate the fricative pronunciations /\/ and /x/ of the Hebrew gimel (‚) and kåf (Î) respectively, in accordance with Hebrew phonology. Moreover, the verbatim translation is so exact that even in the translation of the name ÒÎ¯¡ (1302 1b,15) ‘Carcas,’ the fricative pronunciation is indicated because the name in the Hebrew text comes after the conjunction våv ÒÃkYÃÎ"Â (Esth 1:10). It would have been accurate in the ¡ar˙ to translate the name ÒÎ¯Î ‘Carcas’ using the stop pronunciation according to Arabic phonology, but the need to transfer the names verbatim overrode Arabic phonology. In the Haggadah the following Hebrew and Aramaic personal names are inserted verbatim into the ¡ar˙: ÚÂ˘Â‰È/Ú˘Â‰È (3 3,10; 74 1,15; 91 2b,16; 93 14,14) ‘Joshua’; ÔÂ¥¯Ë (3 3,12; 74 1,16; 93 14,15) ‘Tarfon,’ with the supralinear dot above the pe to indicate the fricative nature of the fe; ‡ÓÂÊ (93 15,12) ‘Zoma,’ although ‰ÓÂÊ with a final he (‰) appears in 3 3,20 and in 74 2,3 as part of the Hebraized orthography; Á¯˙ (3 6,2; 91 2b,18; 93 19,1) ‘Terah’; Ì‰¯·‡ (3 6,2; 91 2b,18; 93 19,2; and more)1 5 ‘Abraham’; ¯ÂÁ (3 6,2; 91 2b,18; 93 19,2) ‘Nahor’; ˜ÁˆÈ (3 6,8; 91 3a,3; 93 19,9) ‘Isaac’; ·Â˜ÚÈ/·˜ÚÈ (3 6,8; 91 3a,3; 93 19,9) ‘Jacob’; ÂÈ˘Ú/ Â˘Ú1 6 (3 6,8; 91 3a,4; 93 19,9) ‘Esau’; and ‰Ú¯Ù/‰Ú¯¥ (3 9,9; 10,18; 14,9; 74 5,7; 93 25,13) ‘Pharaoh,’ although the more interpretive translation ÔÂÚ¯¥ is frequent (see below).

15 16

See above p. 320, for the occurrences of Ì‰¯·‡ in Genesis. Sometimes Â÷Ú has a s•n (÷) in it (93 19,9).

321

CHAPTER NINE As with the preceding personal names, the following place names are copied into the Judeo-Arabic ¡ar˙. In Genesis: ÔÂ¯·Á (15 2a,10; 19b,15; 22a,4) ‘Hebron’; ¯ÂÁ 1 7 (15 2b,8; 3a,8; 4a,1; and more) ‘Nahor’; ¯Â˘ (15 5a,15) ‘Shur’; and ¯Â˘‡ (15 5a,16) ‘Assyria.’ In Esther: Ô˘Â˘ (1302 1b,4; 1b,7; 2a,14; and more) ‘Shushan’ translates verbatim the Hebrew Ô˘Â˘ (Esth 1:2 and more); ÌÈÏ˘Â¯È (1302 2a,18) ‘Jerusalem’ translates the Hebrew ÌÈÏ˘Â¯È; and Ï·· (1302 2b,1) ‘Babylon’ renders the Hebrew name Ï··. In the Haggadah: Ï‡¯˘È (3 2,5; 74 1,3; 91 3a,6; 931 8 12,8; and more) ‘Israel’; ¯ÈÚ˘1 9 (3 6,9; 91 3a,4) ‘Se>•r’; ÔÚÎ2 0 (3 7,18; 91 4b,5; 93 21,16) ‘Canaan’; Ô˘Â‚ (3 7,19; 91 4b,6; 93 21,17) ‘Goshen’; and ÈÈÒ (3 16,20; 16,21; 74 9,7; 9,9; 91 7b,17; 7b,19; 93 36,15; 36,16) ‘Sinai.’ In a more literal mode, the ¡ar˙an sometimes translated Hebrew names verbatim into Judeo-Arabic in a calque translation. In Genesis: The place name ‡¯ÓÓ (Gen 23:17 and more) ‘Mamre’ is translated verbatim as ‚¯Ó (15 2a,6; 2a,9), which means ‘meadow.’ In another place in the text, the ¡ar˙an wrote ‚¯Ó ‰¯ÓÓ (15 19b,14) to translate ‡¯ÓÓ (Gen 35:27), where he both offered a calque translation (‚¯Ó) and a copy of the Hebrew name (‰¯ÓÓ), although with a final he (‰) in line with the Hebraized orthography. Finally, ‰¯ÓÓ ‘Mamre’ appears alone in 15 37b,1, indicating both verbatim and interpretive modes of translation (see chapter 6, p. 170). Furthermore, the Hebrew ‰ÏÙÎÓ‰ ˙¯ÚÓ (Gen 25:9) ‘the cave of Machpela,’ ‰ÏÙÎÓ‰ ‰„˘ ˙¯ÚÓ (Gen 23:19; 50:13) ‘the cave of the field of Machpela,’ and ‰ÏÙÎÓ‰ ‰„˘ (Gen 49:30) ‘the field of the Machpela’ are also rendered as calques 17

This place name is also used as a personal name. Sometimes Ï‡¯÷È has a s•n in it (93 19,1). 19 Sometimes ¯ÈÚ÷ has a s•n in it (93 19,10). 20 See below for Ô‡ÚÎ in an interpretive mode in Genesis. 18

322

THE SEGMENT LEVEL: HEBREW-INFLUENCED ORTHOGRAPHY in Judeo-Arabic: ‰È˙Ó Ï‡ ˙¯‡µÓ (15 5a,2) ‘the cave of the double,’ ‰È˙Ó Ï‡ ËÈµ ˙¯‡µÓ (15 2a,9; 5a,2) ‘the cave of the field of the double,’ and ‰È˙Ó Ï‡ ËÈµ (15 37b,1) ‘the field of the double’ respectively. In these examples the Hebrew ‰ÏÙÎÓ is translated verbatim by Judeo-Arabic ‰È˙Ó to indicate the meaning of the name. The Hebrew name ‰ÏÙÎÓ in Gen 23:17, where it is not accompanied by ‰¯ÚÓ ‘cave’ or ‰„˘ ‘field,’ is translated as ‰¯‡µÓ (15 2a,5) ‘cave’ because the Hebrew original lacks the word for cave or field. Finally in Genesis, È‡¯∂/¯∂‡ ÈÈÁ Ï‡ ¯È· (15 4b,3; 5a,6) ‘a well of the living, seeing’ is a clear calque translation of the place name È‡] ÈÁÏ ¯‡· (Gen 24:62; 25:11) ‘Beer Lachai Roi.’ The Judeo-Arabic phrase È‡¯∂/ÈÈÁ Ï‡ ¯È· ¯∂‡ is unclear, but it is a literal translation of the Hebrew place name. In Esther: ‡ÈÒ¯Ó (1302 2b,1) ‘Marsina’ (myrtle) translates verbatim the Hebrew name ‰Ò„‰ (in Esth 2:7) ‘Hadassah,’ as the name in Hebrew literally means ‘myrtle.’ In the Haggadah: The place name ˜¯· ¯ƒ‡·/˜¯· ¯„‡· /banå∂ir¢ braq/ (3 12-13; 74 1,16) ‘large towns, districts of Beraq’ translates verbatim the Hebrew town name ˜¯· È·. Moving along the literal/interpretive continuum, in a more interpretive translation of personal names, the following names in Genesis have alef (‡) in them to indicate long /å/ pronunciation: Ô‡¯ÓÊ /zimrån/ (15 4b,12) ‘Zimran’ (Hebrew Ô¯ÓÊ in Gen 25:2); Ô‡˘˜È /yoq¡ån/ (15 4b,12; 4b,13) ‘Jokshan’ (Hebrew Ô˘˜È in Gen 25:2); Ô‡„Ó /medån/ (15 4b,12) ‘Medan’ (Hebrew Ô„Ó in Gen 25:2); Ô‡È„Ó /midyån/ (15 4b,12; 4b,15) ‘Midian’ (Hebrew ÔÈ„Ó in Gen 25:2); and ˜‡·˘È /yi¡båq/ (15 4b,13) ‘Iishbak’ (Hebrew ˜·˘È in Gen 25:2). The personal name ÁÂ (15 0-1,13; 0-1,15; 0-1,17) is copied into the Judeo-Arabic ¡ar˙ in a scriptio plena translating Hebrew Á (Gen 5:29 and more). Similarly, the Hebrew name ÔÂ¯ÙÚ (Gen 23:14 and more) ‘Ephron’ seems to be copied

323

CHAPTER NINE verbatim into the ¡ar˙, but the translation adds the supralinear dot above the pe to indicate the fricative /f/: ÔÂ¯¥Ú />efrøn/ (15 2a,1; 2a,3; 2a,5; and more). The Hebrew name ˙Á ‘Heth’ (Gen 23:16; and more) is written in the ¡ar˙ with a yod to indicate the vowel /e/: ˙ÈÁ /˙et/ (15 2a,4; 2a,8; 2a,11; and more). Elsewhere in Genesis, the ¡ar˙an sometimes rendered names with a final alef (‡), although in Hebrew they end in a he (‰): ‡‰Ï· (15 19b,8; 19b,11) ‘Bilhah’; ‡ÊÈÓ (15 20a,17) ‘Mizzah’; and ‡Ú (15 20a,19; 20a,15; 20a,16; and more) ‘Anah’ (see above p. 318, 13-6.6). This could be viewed as an interpretive mode of translation since the ¡ar˙an has deviated from the Hebrew spelling, but it could be perceived as part of the Hebraized orthography, where the final alef (‡) may represent the orthographic influence of the Babylonian Talmud. In Esther: The Hebrew personal name ˘Â¯Â˘Á‡ ‘Ahasuerus, Xerex’ is usually translated in the ¡ar˙ as È¯È˘„Ê‡ Ï‡ /al-azda¡•r•/ (1302 1b,2; 1b,3; 1b,22; and more) in an interpretive mode. This name is usually used in Arabic to translate the names of a number of Persian kings.2 1 It may be a variant of Ardashir, which is the Middle Persian form of Old Persian Artakhshathra, the Greek form of which was Artaxerxes. Ardashir was the founder of the Sassanian dynasty and a just ruler according to Persian tradition. On the other hand, /azhdahaa/ is a dragon, from which the word /zahhaak/, denoting the evil ruler with snakes growing from his shoulders, is derived. In a few places in Esther, the name ˘Â¯Â˘Á‡ ‘Ahasuerus’ is copied verbatim into the Judeo-Arabic text as ˘Â¯Â˘Á‡ (1302 1b,15). In the Haggadah the personal name ‰Ú¯Ù ‘Pharaoh’ is translated interpretively into Judeo-Arabic ÔÂÚ¯¥ (3 2,19; 74 1,10; 91 13a,4; 93 14,1; and more) ‘Pharaoh,’ although see above for the literal translation of ‰Ú¯¥. Furthermore, in 3 14,5 the Hebrew name ÈÒÂÈ 21

See H. Massé, “Ardashir,” EI2, 1:626.

324

THE SEGMENT LEVEL: HEBREW-INFLUENCED ORTHOGRAPHY ÈÏÈÏ‚‰ is translated verbatim into ÈÏÈÏ‚ Ï‡ ÈÒÂÈ, whereas in 74 7,4 the translation reflects the Egyptian Judeo-Arabic pronunciation of the definite article ÈÏÈÏ‚ ‡, and thus is interpretive. Note that in Arabic the /l/ of the definite article /al/ or /il/ is fully assimilated to the first consonant of the noun if it is a coronal. These consonants are also known as /˙ur¥f ¡amsiyya/ ‘sun letters.’ The phoneme /g/ in Egyptian Arabic is an exception: although it is not a coronal, it is still considered a /˙arf ¡amsiyya/, thus /ig-gal•l•/. The same interpretive mode occurs for place names, where the spelling is helpful in understanding the Judeo-Arabic pronunciation. In Genesis the spelling of ÌÈÈ¯Â‰ Ì‡¯‡ /aråm nah¥rayim/ (15 2b,7) ‘Aram Naharaim’ indicates the long vowels in an interpretive mode; ¯ˆÓ /mißr/ (15 5a,16; 6a,6; 22b,2; and more) ‘Egypt’ translates Hebrew ÌÈ¯ˆÓ; the Hebrew place name ·‚‰ ı¯‡ (Gen 24:62) ‘the region of the Negev’ is translated interpretively as ÈÏ·˜ Ï‡ √¯‡ (15 4b,3) ‘the region of the south.’ Likewise, the Hebrew place name ¯¯‚ ÏÁ (Gen 26:17) is translated interpretively as ¯‡¯‚ È„‡ÂÂ (15 6b,10) ‘the valley of Grar,’ as the word ÏÁ is translated into È„‡ÂÂ ‘valley’ and the name ¯‡¯‚ has an alef (‡), to indicate the long /å/. Similarly, the place name Ú·˘ ¯‡· (Gen 26:23; 26:33) ‘Beer-Sheba’ is translated as Ú·‡˘ ¯È· /b•r såbi>/ (15 6b,20), literally ‘the well of Såbi> (seventh),’ and Ú·˘ ¯È· (15 7a,16): ¯‡· is translated as /b•r/ and the alef (‡) is inserted to indicate long /å/. Furthermore, the place name Ô‡ÚÎ (15 2a,10; 9a,10; 9a,20; and more) ‘Canaan’ translates slightly interpretively the Hebrew ÔÚÎ (Gen 23:19 and more), marking the long /å/; however, in the Haggadah the Hebrew ÔÚÎ is copied verbatim into the ¡ar˙: ÔÚÎ (3 6,7; 91 3a,2; 93 19,8) ‘Canaan’ (see above). Place names in Esther can be translated interpretively as well: Hebrew Â„Â‰ (Esth 1:1; 8:9) is translated „‰ Ï‡ (1302 1b,2; 6a,9), but „‰ Ï‡ (1302 1b,5) also translates Ò¯Ù (Esth 1:3). Ò¯Ù 325

CHAPTER NINE in Esth 1:14, on the other hand, is translated into Ì‚Ú Ï‡ (1302 1b,20), the Arabic name for Persia. A similar phenomenon occurs with the Hebrew place name ˘ÂÎ (Esth 1:1; 8:9), which is translated as ˘·Á Ï‡ (1302 1b,2; 6a,9), but ˘·Á Ï‡ (1302 1b,5) also translates È„Ó (Esth 1:3). È„Ó in Esth 1:14, on the other hand, is translated by È„‡Ó Ï‡ (1302 1b,20), although the inserted alef (‡) reflects an interpretive tendency. Place names in the Haggadah are also translated interpretively at times: ¯ˆÓ (3 2,3; 74 2,3; 91 3a,5; 93 15,13; and more) ‘Egypt’ translates the Hebrew ÌÈ¯ˆÓ; ÌÂÈ¥ (3 9,10; 74 5,2; 91 5a,7; 93 23,16) ‘Fayum’ translates Hebrew Ì&˙Ù ‘Pithom’; Hebrew ÒÒÓÚ¯ ‘Raamses’ is translated as ‰ÒÓ‰·/‰ÒÈ‰·/‰ÈÒ‰· (3 9,10; 74 5,2; 91 5a,7; 93 23,16) ‘Bahamsa (?)’; and Hebrew ¯›‡È is translated into Judeo-Arabic ÏÈ (3 10,19; 74 5,14; 91 5b,5; 93 25,14) ‘(the) Nile.’ Furthermore, in a clear deviation from the Hebrew-influenced orthography and in an interpretive Arabic mode, the name ¯ÊÚÏ‡ ‘Elazar’ is translated in one manuscript as ¯∂Ú Ï‡ /el>aΩar¢/ (93 15-b,10). The ¡ar˙an separated the Ï‡ from ¯∂Ú as if it were the Arabic definite article, where it is actually the Hebrew word ÏÕ‡ ‘God.’ Furthermore, the Judeo-Arabic spelling ¯∂Ú Ï‡, by using the emphatic /Ω/, indicates that the pronunciation of the name is changed in Egyptian Judeo-Arabic to be in accordance with Arabic phonology, yielding /el>aΩar¢/, where Hebrew /z/ became emphatic Judeo-Arabic /Ω/. Other manuscripts and even the same ¡ar˙an of 93 have copied the Hebrew name into the Judeo-Arabic translation in a clear verbatim mode: ¯ÊÚÏ‡ (3 3,11; 3,18; 74 1,1; 2,2; 93 14,14). The same analysis holds true for the name ‰ÈÈ¯∂Ú />aΩar¢ya/ (93 15,9) ‘Azariah,’ translating the Hebrew ‰È¯ÊÚ with the shift /z/ > /Ω/ in the environment of the emphatic [r¢]. As above, other manuscripts, including the ¡ar˙an of 93 in another place, have copied the Hebrew name into the

326

THE SEGMENT LEVEL: HEBREW-INFLUENCED ORTHOGRAPHY Judeo-Arabic translation: ‰È¯ÊÚ/‰ÈÈ¯ÊÚ (3 3,11; 3,18; 74 1,15; 2,2; 93 14,15).2 2 Finally, the following example demonstrates well the literal/ interpretive linguistic tension in which the ¡ar˙anim found themselves. On the surface the name ‰·È˜Ú ‘Akiva’ in the Judeo-Arabic text (3 3,12) looks as if it is translated in the literal mode, since it is spelled with a final he (‰), as part of the Hebraized orthography. A closer look, though, reveals that the Hebrew word is spelled with a final alef (‡), ‡·È˜Ú; therefore, Judeo-Arabic ‰·È˜Ú is actually in the interpretive mode. However, the ¡ar˙an heard the name Akiva, with its final /-a/ vowel, and therefore chose the final he (‰) in imitation of normative Hebrew orthography, thus translating it after all in a more literal mode. In another manuscript of the Haggadah ‡·È˜Ú (93 14,15) appears, which again can be considered either literal, as it is an exact copy of the Hebrew ‡·È˜Ú (or perhaps even influenced by the Aramacized spelling of the Babylonian Talmud with final alef [‡]) or interpretive, as it is not spelled with a final he (‰). Summary The last three chapters have provided selected data to exemplify the model of the analysis of the texts of the ¡ur¥˙ presented in chapter 3. The data have been scanned from head to toe through the phrase and the word levels (chapter 7), the morphosyntactic level (chapter 8), and the segmental level (chapter 9).

22

See also chapter 4, pp. 108, 1.11.4 and chapter 5, pp. 149–50, 1.4.2.

327

BIBLIOGRAPHY Arnold, Werner. 2006–7. “The Arabic Dialect of the Jews of Iskenderun.” RomanoArabica: Peripheral Arabic Dialects, n.s. 6–7:7–12. Avishur, Yitzhak. 1979. “The Folk Literature of the Jews of Iraq in Judeo-Arabic.” Pe‘amim 3:83–90. [in Hebrew] ———. 1988. “Modern Judeo-Arabic Translations of the Bible in the Orient: A Cultural-Linguistic Presentation.” In Studies in Jewish Languages: Bible Translations and Spoken Dialects, ed. M. Bar-Asher. Jerusalem: Misgav Yerushalayim. 39–54. [in Hebrew] ———. 1991. “Translations of the Old Testament into Judeo-Arabic in Iraq.” In Studies in the History and Culture of Iraqi Jewry, ed. Y. Avishur. Or Yehuda: The Babylonian Jewry Heritage Center, Institute for Research on Iraqi Jewry. 139–65. [in Hebrew] ———. 1992. “New Folk Tales Concerning R. Abraham Ibn Ezra (and His Son) from Egypt and Iraq.” In Studies in the Works of Abraham Ibn Ezra, ed. I. Levin and M. Itzhaki. Tel Aviv: Tel Aviv University. 163–92. [in Hebrew] ———. 2001. Studies in Judaeo-Arabic Translations of the Bible. Tel Aviv–Jaffa: Archaeological Center Publications. Avrahami, Y. 1994. “Ha-moqshiya: tirgum ha-miqra le->arvit yehudit shel tunis me-
BIBLIOGRAPHY Ayyar, L. V. Ramaswami. 1993. The Evolution of Malayalam Morphology. Thrissur: Kerala Sahitya Akademi. Azulai, Haim Yoseph David. N.d. “Haggadah shel pesa˙ ma>a ¡ar˙ bi-l- >arabi” [A Passover Haggadah with an Arabic ¡ar˙]. [in Hebrew and Judeo-Arabic] Badawi, El-Said. 1973. Mustawayåt l->arabiyyati l-mu>åßira f• mißr [Levels of contemporary Arabic in Egypt]. Cairo: Dår al-Ma>årif bi-Mißr. [in Arabic] Bar-Asher, Moshe. 1988. “The Shar˙ of the Maghreb: Judeo-Arabic Exegesis of the Bible and Other Jewish Literature: Its Nature and Formation.” In Studies in Jewish Languages: Bible Translations and Spoken Dialects, ed. M. Bar-Asher. Jerusalem: Misgav Yerushalayim. 3–34. [in Hebrew] ———. 1998. Traditions linguistiques des Juifs d’Afrique du Nord. Jerusalem: Bialik Institute. [in Hebrew] ———. 2001. Commentaire biblique leshon limmudim de Rabbi Raphael Berdugo. 3 vols. Jerusalem: Bialik Institute. [in Hebrew] ———. 2002. “Aspects of the Study of Jewish Languages and Literatures.” Pe‘amim 93:77–89. [in Hebrew] Behnstedt, Peter. 1978. “Zur Dialektgeographie des Nildeltas.” Zeitschrift der arabische Linguistik 1:64–92. Behnstedt, Peter, and Manfred Woidich. 1985. Die ägiptisch-arabischen Dialekte, Band 1, Einleitung und Anmerkungen zu den Karten. Wiesbaden: Dr. Ludwig Reichert. ———. 1985. Die ägiptisch-arabischen Dialekte, Band 2, Dialektatlas von Ägypten. Wiesbaden: Dr. Ludwig Reichert. ———. 1987. Die ägiptisch-arabischen Dialekte, Band 3, Texts, I. Delta-Dialekte, Arabisch-Deutsch. Wiesbaden: Dr. Ludwig Reichert. ———. 1988. Die ägiptisch-arabischen Dialekte, Band 3, Texts, II. Niltaldialekte; III. Oasendialekte. Wiesbaden: Dr. Ludwig Reichert. ———. 1994. Die ägiptisch-arabischen Dialekte, Band 4, Glossar, ArabischDeutsch. Wiesbaden: Dr. Ludwig Reichert. ———. 1999. Die ägiptisch-arabischen Dialekte, Band 5, Glossar, DeutschArabisch. Wiesbaden: Dr. Ludwig Reichert. Ben-Oren, Gershon. 2000. “Three Recent Bible Translations to Georgian.” Pe‘amim 84:78–87. [in Hebrew]

330

BIBLIOGRAPHY Ben Sasson, H. H. 1994. A History of the Jewish People. Cambridge, MA: Harvard University Press. Benabu, Isaac. 1988. “Contribution to a Symposium on ‘Bible Translation in Judeo-Arabic and Ladino in Recent Times: A Cultural Linguistic Presentation.’ ” In Studies in Jewish Languages: Bible Translations and Spoken Dialects, ed. M. Bar-Asher. Jerusalem: Misgav Yerushalayim. 47*–51*. Benor, Sarah. 2004. “Second Style Acquisition: The Linguistic Socialization of Newly-Orthodox Jews.” Ph.D. dissertation, Stanford University, Stanford, CA. ———. 2008. “Towards a New Understanding of Jewish Languages in the 21st Century.” Religion Compass. Vol. 3. ———. 2009. “Do American Jews Speak a ‘Jewish Language’? A Model of Jewish Linguistic Distinctiveness.” Jewish Quarterly Review. Forthcoming. Bickerton, Derek. 1975. Dynamics of a Creole System. Cambridge: Cambridge University Press. Birnbaum, Solomon Asher. 1971. “Jewish Languages.” Encyclopedia Judaica 10:66–69. ———. 1979. Yiddish: A Survey and a Grammar. Toronto: University of Toronto Press. Blanc, Haim. 1960. “Style Variations in Spoken Arabic: A Sample of Interdialectal Educated Conversation.” In Contributions to Arabic Linguistics, ed. C. Ferguson. Cambridge, MA: Harvard University Press. 79–161. ———. 1964. Communal Dialects in Baghdad. Cambridge, MA: Harvard University Press. ———. 1974. “The Nekteb-Nektebu Imperfect in a Variety of Cairene Arabic.” Israel Oriental Studies 4:206–26. ———. 1981. “Egyptian Arabic in the Seventeenth Century: Notes on the JudeoArabic Passages of Darxe No‘am (Venice, 1697).” In Studies in Judaism and Islam Presented to Shelomo Dov Goitein, ed. S. Morag, I. Ben-Ami, and N. Stillman. Jerusalem: The Magnes Press, The Hebrew University. 185–202. ———. 1985. “Egyptian Judeo-Arabic: More on the Subject of R. Mordekhai b. Yehuda Ha-Levi’s Sefer Darxe No‘am.” Sefunot n.s. 3/18:299–314. [in Hebrew] Blau, Joshua. 1970. On Pseudo-Corrections in Some Semitic Languages. Jerusalem: Israel Academy of Sciences and Humanities.

331

BIBLIOGRAPHY ———. 1991. “On the Multilayered Character of Medieval Judaeo-Arabic.” Sefunot n.s. 5/20:165–70. [in Hebrew] ———. 1995. A Grammar of Medieval Judeo-Arabic. 2nd ed. Jerusalem: The Magnes Press, The Hebrew University. [in Hebrew] ———. 1999. The Emergence and Linguistic Background of Judaeo-Arabic. 3rd ed. Jerusalem: Ben-Zvi Institute. Blau, Joshua, and Hopkins, Simon. 1984. “On Early Judaeo-Arabic Orthography.” Zeitschrift für arabische Linguistik 12:9–27. ———. 1987. “Judaeo-Arabic Papyri: Collected, Edited, Translated and Analysed.” Jerusalem Studies in Arabic and Islam 9:87–160. Brzezina, Maria. 1986. Polszczyzna Zydów. Warsaw: Panstwowe Wydawnictwo Naukowe. Bunis, David M. 1981. “A Comparative Linguistic Analysis of Judezmo and Yiddish.” International Journal of the Sociology of Language 30:49–70. Bussmann, Hadumod. 1996. Routledge Dictionary of Language and Linguistics. London: Routledge. Chambers, Jack K., and Peter Trudgill. 1998. Dialectology. 2nd ed. Cambridge: Cambridge University Press. Chetrit, Joseph. 2007. Diglossie, hybridation, et diversité intra-linguistique: Études socio-pragmatiques sur les langues juives, le judéo-arabe et le judéo-berbère. Leuven: Peeters. Comrie, Bernard. 1989. Language Universals and Linguistic Typology. 2nd ed. Oxford: Basil Blackwell. Corré, Alan. 1989. A ‘Diskionary’ and Chrestomathy of Modern Literary JudeoArabic. Milwaukee: The University of Wisconsin-Milwaukee, Language Resource Center Software. Crystal, David. 2003. A Dictionary of Linguistics and Phonetics. 5th ed. Oxford: Blackwell. Cuomo, Luisa-Ferretti. 2000. “Rashi in Italy: Echoes of Jewish Exegesis in the Translations of Song of Songs into Italian and Judeo-Italian during the Renaissance.” Pe‘amim 83:132–46. [in Hebrew] David, Noa. 2005. “The Book of Job (1-10): A Linguistic Analysis of a Shar˙,” M.A. thesis, Emory University, Atlanta, GA.

332

BIBLIOGRAPHY Davies, Humphrey. 1981. “17th-Century Egyptian Arabic: A Profile of the Colloquial Material in Y¥suf al-⁄irb•n•’s Hazz al-Qu˙¥f f • ¡ar˙ qaß•d Ab • ⁄åd¥f.” Ph.D. dissertation, University of California, Berkeley. ———. 2005. Y¥suf al-Shirb•n•’s Kitåb Hazz al-Qu ˙¥f bi-shar˙ qaß•d Ab• ⁄åd¥f [Brains Confounded by the Ode of Ab¥ ⁄åd¥f Expounded], vol. 1, Arabic Text . Leuven: Uitgeverij Peeters. DeCamp, David. 1971. “Toward a Generative Analysis of a Part-Creole Continuum.” In Pidginization and Creolization of Languages, ed. D. Hymes. Cambridge: Cambridge University Press. 349–70. Doron, David. 1979. “The Arabic Translation of the Torah by Issachar ben Susån Hamma‘ravi: A Linguistic Study.” Ph.D. dissertation, Bar-Ilan University, Ramat Gan, Israel. [in Hebrew] ———. 1995. “R. Mordechai Óai Diyyan’s Introduction to His Arabic Translation of Genesis.” In East and Maghreb: Researches in the History of the Jews in the Orient and North Africa VI, ed. S. Schwarzfuchs and E. Horowitz. Ramat-Gan: Bar-Ilan University Press. 131–60. Drory, Rina. 1992. “Words Beautifully Put: Hebrew versus Arabic in Tenth-Century Jewish Literature.” In Genizah Research after Ninety Years: The Case of JudaeoArabic, ed. J. Blau and S. C. Reif. Cambridge: Cambridge University Press. 53–66. ———. 2000. Models and Contacts: Arabic Literature and Its Impact on Medieval Jewish Culture. Leiden and Boston: Brill. Elqayam, Avraham, and Benjamin Hary. 1997. “A Judeo-Arabic Sabbatian Apocalyptic Hymn.” Kabbalah: Journal for the Study of Jewish Mystical Texts 2:105–41. Erez, Yehuda. 1959. “In the Eyes of an Emissary from Palestine.” In Encyclopedia of Diasporas: Subcarpathian Ruthenia, ed. Y. Erez. Jerusalem: “Encyclopedia of Diaspora” Publishers Inc. 231–56. [in Hebrew] Erwin, Wallace M. 2004. A Short Reference Grammar of Iraqi Arabic . Washington, DC: Georgetown University Press. Fischer, Wolfdietrich, and Otto Jastrow. 1980. Handbuch der arabischen Dialekte . Wiesbaden: Harrassowitz.

333

BIBLIOGRAPHY Fishman, Joshua A. 1981. “The Sociology of Jewish Languages from the Perspective of the General Sociology of Language: A Preliminary Formulation.” International Journal of the Sociology of Language 30:5–16. ———, ed. 1985a. Readings in the Sociology of Jewish Languages. Leiden: E. J. Brill. ———. 1985b. “The Sociology of Jewish Languages from a General Sociolinguistic Point of View.” In Readings in the Sociology of Jewish Languages, ed. J. Fishman. Leiden: E. J. Brill. 3–21. Ferguson, Charles. 1959. “Diglossia.” Word 15:325–40. ———. 1989. “Grammatical Agreement in Classical Arabic and the Modern Dialects: A Response to Versteegh’s Pidginization Hypothesis.” Al-‘Arabiyya 22:5–17. _____. 1996 [1991]. “Epilogue: Diglossia Revisited.” In Understanding Arabic: Essays in Contemporary Arabic Linguistics in Honor of El-Said Badawi, ed. A. Elgibali. Cairo: The American University in Cairo Press. 49–67. Futato, Mark D. 2003. Beginning Biblical Hebrew. Winona Lake, IN: Eisenbrauns. Gairdner, W. H. 1926. Egyptian Colloquial Arabic: A Conversation Grammar. 2nd ed. London: Oxford University Press. Galliner, Siegfried. 1903. Saadia al-Fajjûmi’s arabische Psalmenübersetzung und Commentar (Psalms 73–89). Berlin: H. Itzkowski. Gamliel, Ophira. 2008. “Oral and Written Literature in Jewish Malayalam (South India).” Paper presented at a conference on Jewish languages, The Hebrew University, Jerusalem, Israel. Geva-Kleinberger, Aharon. 2004. Die arabische Stadtdialekte von Haifa in der ersten Hälfte des 20. Jahrhunderts. Wiesbaden: Harrassowitz. ———. 2005. “The Last Informant: A Text in the Jewish Arabic Dialect of Peq•>•n.” Wiener Zeitschrift für die Kunde des Morgenlandes 95:45–61. ———. 2008. Autochthonous Texts in the Arabic Dialect of the Jews of Tiberias. Wiesbaden: Harrassowitz. Forthcoming. Gil, Moshe. 1984. “The Origin of the Jews of Yathrib.” Jerusalem Studies in Arabic and Islam 4:203–24. Goitein, Shlomo Dov. 1972. “Townsmen and Fellah: A Geniza Text from the 17th Century.” Asian and African Studies 8:257–61.

334

BIBLIOGRAPHY Gold, David. 1981. “Jewish Intralinguistics as a Field of Study.” International Journal of the Sociology of Language 30:31–46. ———. 1989. Jewish Linguistic Studies. Haifa: Association for the Study of Jewish Languages. Gottreich, Emily Benichou. 2008. “Historicizing the Concept of Arab Jews in the Maghrib.” Jewish Quarterly Review 98/4:433–51. Greenspahn, Frederick E. 2000. “Why Do Jews Translate the Bible?” Paper presented at the annual meeting of the Association for Jewish Studies, Boston, Massachusetts. ———. 2002. “How Jews Translate the Bible.” In Biblical Translation in Context, ed. F. W. Knobloch. Bethesda, MD: University Press of Maryland. 43–61. ———. 2006. “Why Jews Translate the Bible.” In Biblical Interpretation in Judaism and Christianity, ed. I. Kalimi and P. J. Haas. New York: T&T Clark. 179–95. Grotzfeld, Heinz. 1967. “Ein Zeugnis aus dem Jahre 688/1289 für die Aussprache des qåf also hamza in Kaireischen.” Zeitschrift des Deutschen Morgenländischen Gesellschaft 117:87–90. Hary, Benjamin. 1987. “Judeo-Arabic, Written and Spoken in Egypt in the Sixteenth and Seventeenth Centuries.” Ph.D. dissertation, University of California, Berkeley. ———. 1990. “The Importance of the Orthography in Judeo-Arabic Texts.” Proceedings of the Tenth World Congress of Jewish Studies, Division D, Volume 1. Jerusalem: World Union of Jewish Studies. 77–84. ———. 1991a. “The Tradition of Later Egyptian Judeo-Arabic Orthography.” Massorot 5–6:119–37. [in Hebrew] ———. 1991b. “On the Use of
335

BIBLIOGRAPHY ———. 1994b. “Linguistic Notes on an Egyptian Judeo-Arabic Passover Haggadah and the Study of the Egyptian ⁄ar˙.” Actes des premières journées internationales de dialectologie arabe de Paris, ed. D. Caubet and M. Vanhove. Paris: INALCO. 375–88. ———. 1995. “Judeo-Arabic in Its Sociolinguistic Setting.” Israel Oriental Studies 15:73–99. ———. 1996a. “The Importance of the Language Continuum in Arabic Multiglossia.” In Understanding Arabic: Essays in Contemporary Arabic Linguistics in Honor of El-Said Badawi, ed. A. Elgibali. Cairo: The American University in Cairo Press. 69–90. ———. 1996b. “The ±•m/G•m in Colloquial Urban Egyptian Arabic.” Israel Oriental Studies 16:153–68. ———. 1996c. “Adaptations of Hebrew Script.” In The World’s Writing Systems, ed. W. Bright and P. Daniels. Oxford: Oxford University Press. 727–34 and 741–42. ———. 1997a. “The Impact of the Cairo Genizah Documents on the Study of the History of Arabic.” Bulletin of the Israeli Academic Center in Cairo, Special Issue: The Cairo Genizah: One Hundred Years of Discovery 21:35–39. ———. 1997b. “On Later and Modern Egyptian Judeo-Arabic.” In Humanism, Culture, and Language in the Near East: Studies in Honor of George Krotkoff, ed. A. Afsaruddin and M. Zahniser. Winona Lake, IN: Eisenbrauns. 199–224. ———. 1999. “Hebrew Elements in Egyptian Judeo-Arabic Texts.” In Vena Hebraica in Judaeorum Linguis, ed. S. Morag, M. Bar-Asher, M. Mayer-Modena. Milano: Università delgi Studi di Milano. 67–91. ———. 2000a. “Egyptian Judeo-Arabic ¡ar˙: Bridging the Cultures of Hebrew and Arabic.” In Judaism and Islam: Boundaries, Communication and Interaction. Essays in Honor of William M. Brinner, ed. B. Hary, J. Hayes, and F. Astren. Leiden: Brill. 395–407. ———. 2000b. “Egyptian Judeo-Arabic ¡ar˙ of Genesis.” Proceedings of the Third International Conference of AÏDA (Association Internationale de Dialectologie Arabe), ed. Manwel Mifsud. Malta. 53-58. ———. 2003. “Judeo-Arabic: A Diachronic Reexamination.” International Journal for the Sociology of Language 163:61–75.

336

BIBLIOGRAPHY ———. 2004. “Jewish Languages, Are They Sacred?” In Lenguas en contacto de la Antiguedad a la Edad Media, ed. P. Bádenas de la Peña, S. Torallas Tovar, E. R. Luján, and M. A. Gallego. Madrid: CSIC. 225–44. ———. 2007. “Hypercorrection.” Encyclopedia of Arabic Language and Linguistics. Vol. 2. Leiden: Brill. 275–79. ———. 2009. Sacred Texts: The Tradition of the ¡ar˙ in Egyptian Judeo-Arabic, with Critical Editions and Translations of the Book of Genesis, the Book of Esther, and the Passover Haggadah. Leiden: Brill. Forthcoming. Hary, Benjamin, and María Ángeles Gallego. 1999. “La Versión Española de Maqre Dardeqe.” In Jewish Studies at the Turn of the Century, ed. J. T. Borrás and A. Sáenz-Badillos. Vol 1. Leiden: Brill. 57–64. Hary, Benjamin, and Martin J. Wein. 2008. “Religiolinguistics: Mapping the Impact of Religion on Linguistic Varieties.” Paper presented at the University of Haifa, Israel. Hatav, Galia. 2004. “Anchoring World and Time in Biblical Hebrew.” Journal of Linguistics 40:491–526. ———. 2007. “Teaching the Biblical Hebrew Verbal System.” Higher Hebrew Education 12:5–52. Hertz, Joseph H. 1936. Sermons, Addresses and Studies. 3 vols. London: Soncino. Hinds, Martin, and El-Said Badawi. 1986. A Dictionary of Egyptian Arabic: Arabic-English. Beirut: Librarie du Liban. Hopkins, Simon. 1984. Studies in the Grammar of Early Arabic. Oxford: Oxford University Press. Horsnell, Malcolm J. A. 1999. A Review and Reference Grammar for Biblical Hebrew. Hamilton, Ontario: McMaster University Press. Jacobs, Neil G. 2005. Yiddish: A Linguistic Introduction. Cambridge: Cambridge University Press. Jastrow, Otto. 1990. Der arabische Dialekt der Juden von >Aqra und Arb•l. Wiesbaden: Otto Harrassowitz. ———. 1992. “Der arabische Dialekt der Juden von Kirkuk.” Jerusalem Studies in Arabic and Islam 15:240–54.

337

BIBLIOGRAPHY ———. 1999. “Three Anecdotes in Jewish Maslawi.” In Linguistic and Cultural Studies on Arabic and Hebrew: Essays Presented to Moshe Piamenta for His Eightieth Birthday. Wiesbaden: Harrassowitz. 61–70. Jochnowitz, George. 1985. “Had Gadya in Judeo-Italian and Shuadit (JudeoProvençal).” In Readings in the Sociology of Jewish Languages, ed. J. A. Fishman. Leiden: E. J. Brill. 241–45. ———. 2001. “Sacred Texts in Judeo-Italian and Judeo-Provençal.” Paper presented at the annual meeting of the Association for Jewish Studies, Washington, DC. Israel, Jonathan I. 1989. European Jewry in the Age of Mercantilism 1550–1750. Oxford: Clarendon Press. Kaplan, Steven, and Meley Mulugetta. “Bible Translations into Ge>ez: Jewish and Christian Contexts.” Pe‘amim 83:132–46. [in Hebrew] Kasher, Rimon. 2000. “Aramaic Bible Translations.” Pe‘amim 83:70–107. [in Hebrew] Khan, Geoffrey. 1990. Karaite Bible Manuscripts from the Cairo Genizah. Cambridge: Cambridge University Press. ———. 1991. “A Study of the Judaeo-Arabic of Late Genizah Documents and Its Comparison with Classical Judaeo-Arabic.” Sefunot n.s. 5/20:223–34. [in Hebrew] ———. 1992. “Notes on the Grammar of a Late Egyptian Judaeo-Arabic Text.” Jerusalem Studies in Arabic and Islam 15:220–39. Lebedev, Viktor Vladimirovitch. 1965. “Skazka o zolotoj gazele: obrazçik arabskogo fol’klora XVII veka.” Semitskie Jazyki 2:521–32. [in Russian] ———. 1977. Pozdnij srednearabskij jazyk (XIII–XVIII VV). Moscow: Nauka. [in Russian] Lehiste, Ilse. 1988. Lectures on Language Contact. Cambridge, MA: The MIT Press. Levy, Lital. 2008. “Historicizing the Concept of Arab Jews in the Mashriq.” Jewish Quarterly Review 98/4:452–69. López-Morillas, Consulo. 1994. “Aljamiado and the Moriscos’ Islamicization of Spanish.” In Perspective on Arabic Linguistics VI , ed. M. Eid, V. Cantarino and K. Walters. Amsterdam: John Benjamins. 17–23. al-Ma\rib•, Y¥sef. 1968. Daf> al-ißr >an kalåm ahl mißr, ed. >Abd as-Salåm A˙mad >Awwåd. Moscow (ms. 1606).

338

BIBLIOGRAPHY Mahootian, Shahrzad. 2006. “Code Switching and Mixing.” Encyclopedia of Language and Linguistics, ed. K. Brown. 2nd ed. Vol. 2. Amsterdam: Elsevier. 511–27. Malter, Henry. 1921. Saadia Gaon, His Life and Works. Philadelphia: Jewish Publication Society. Maman, Aharon. 2000. “The Meghrebi Shar˙ of the Bible.” Pe‘amim 83:48–56. [in Hebrew] ———. 2004. “The Sefrou (Morocco) Version of al->a¡ar kalimåt.” Jerusalem Studies in Arabic and Islam 29:358–76. Mansour, Jacob. 1991. The Jewish Baghdadi Dialect: Studies and Texts in the Judaeo-Arabic Dialect of Baghdad. Or-Yehuda: The Babylonian Jewry Heritage Center. Matras, Yaron. 1991. “Zur Rekonstruktion des jüdischdeutschen Wortschatzes in den Mundarten ehemaliger ‘Judendörfer’ in Südwestdeutschland.” Zeitschrift für Dialektologie und Linguistik 58/3:267–93. McGuirk, Russel. 1986. Colloquial Arabic of Egypt. London: Routledge. Mejdell, Gunvor. 2006. Mixed Styles in Spoken Arabic in Egypt: Somewhere between Order and Chaos. Leiden: Brill. Mitchell, Terence Frederick. 1962. Colloquial Arabic: The Living Language of Egypt. London: The English Universities Press. ———. 1978. An Introduction to Egyptian Colloquial Arabic. Oxford: Clarendon Press. Myhill, John. 2004. Language in Jewish Society: Toward a New Understanding. Clevedon: Multilingual Matters. Newby, Gordon Darnell. 1971. “Observations about an Early Judaeo-Arabic.” Jewish Quarterly Review 61:212–21. ———. 1988. A History of the Jews of Arabia from Ancient Times to Their Eclipse Under Islam. Columbia: University of South Carolina Press. Norlin, Kjell. 1987. A Phonetic Study of Emphasis and Vowels in Egyptian Arabic. Lund: Lund University, Department of Linguistics. Ohala, John. 1993. “The Phonetics of Sound Change.” In Historical Linguistics: Problems and Perspectives, ed. C. Jones. London: Longman. 237–78.

339

BIBLIOGRAPHY Ornan, Uzzi. 1985: “Hebrew Is Not a Jewish Language.” Le¡onénu 48/49:199–206. [in Hebrew] Paper, Herbert, ed. 1978. Jewish Languages: Theme and Variations. Cambridge: Association for Jewish Studies. Piamenta, Moshe. 1988. “Remarks following up the Lectures of Y. Avishur and H. V. Sephiha.” In Studies in Jewish Languages: Bible Translations and Spoken Dialects, ed. M. Bar-Asher. Jerusalem: Misgav Yerushalayim. 75–78. [in Hebrew] ———. 1996. “Intra- and Intercommunal Appellations in Judeo-Yemeni.” In Proceedings of the Colloquium on Logos, Ethos, Mythos in the Middle East and North Africa, Budapest, September 1995, ed. K. Dévényi and T. Iványi. Part 1, Budapest: Eötvös Lórand University; Csoma de Korös Society. 19–30. Polliack, Meria. 1997. The Karaite Tradition of Arabic Bible Translation: A Linguistic & Exegetical Study of Karaite Translations of the Pentateuch from the Tenth & Eleventh Centuries CE. Leiden: E. J. Brill. Prager, Leonard. 1986. “A Preliminary Checklist of English Names of Jewish Lects.” Jewish Language Review 6. Rabin, Chaim. 1981. “What Constitutes a Jewish Language?” International Journal of the Sociology of Language 30:19–28. Rabin, Haim, et al. 1979. “Ha-leshonot ha-yehudiyot: ha-meshutaf, ha-meyu˙ad ve-ha-be>ayati” [Jewish Languages: The Common, the Unique and the Problematic]. Pe‘amim 1:40–66. [in Hebrew] Rabinowitz, Louis Isaac. 1971. “Proselytes.” Encyclopedia Judaica 13:1182–94. Reckendorf, Hermann. 1895–98. Die syntaktischen Verhältnisse des Arabischen. Leiden: E. J. Brill. Rickford, John R. 1987. Dimensions of a Creole Continuum: History, Texts, and Linguistic Analysis of Guyanese Creole. Stanford: Stanford University Press. Rosenbaum, Gabriel. 2002a. “The Particles ma and lam and Emphatic Negation in Egyptian Arabic.” In Sprich doch mit deinen Knechten aramäisch, wir verstehen es!, ed. W. Arnold and H. Bobzin. Wiesbaden: Harrassowitz. 583–98. ———. 2002b. “The Arabic Dialect of Jews in Modern Egypt.” Bulletin of the Israeli Academic Center in Cairo 25:35–46. ———. 2002c. “Spoken Jewish Arabic in Modern Egypt: Hebrew and Non-Standard Components.” Massorot 12:117–48. [in Hebrew]

340

BIBLIOGRAPHY ———. 2002d. “Hebrew Words and Karaite Goldsmiths’ Secret Language Used by Jews and Non-Jews in Modern Egypt.” Pe‘amim 90:115–53. [in Hebrew] ———. 2004. “Egyptian Arabic as a Written Language.” Jerusalem Studies in Arabic and Islam 29:281–340. ———. 2008. “Mixing Colloquial and Literary Arabic in Modern Egyptian Prose through the Use of Free Indirect Style and Interior Monologue.” In Moyen arabe et variétés moyennes de l’arabe à travers l’histoire (Actes du premier colloque international, Louvain-la-Neuve, 10–14 May 2004), ed. J. Lentin and J. Grand’Henry. Louvain-la-Neuve: Université catholique de Louvain, Institut orientaliste de Louvain. 391–404. Rosenbloom, Joseph R. 1978. Conversion to Judaism: From the Biblical Period to the Present. Cincinnati, OH: HUC Press. Rosenzweig, Franz. 1971. The Star of Redemption. Trans. William W. Hallo. New York: Holt, Rinehart and Winston. Sabar, Yona. 1983. The Book of Genesis in Neo-Aramaic. Publications of the Hebrew University Language Traditions Project, vol. 9. Jerusalem: The Magnes Press. The Hebrew University. ———. 1988. The Book of Exodus in Neo-Aramaic. Publications of the Hebrew University Language Traditions Project, vol. 12. Jerusalem: The Magnes Press. The Hebrew University. ———. 2000. “The Pentateuch in Neo-Aramaic: Overview.” Pe‘amim 83:107–17. [in Hebrew] Sasson, David Solomon. 1932. Ohel David: Descriptive Catalogue of the Hebrew and Samaritan Manuscripts in the Sasson Library, London. London: Oxford University Press. Schwarzvald (Rodrigue), Ora. 1989. The Ladino Translations of Pirke Aboth: Studies in the Translation of Mishnaic Hebrew into Judeo-Spanish. Publications of the Hebrew University Language Traditions Project, vol.13. Jerusalem: The Magnes Press. The Hebrew University. ———. 1992. “Linguistic Trends in the Development of Ladino Translations.” Paper presented at the conference, “Memories from Spain,” Paris. ———. 2001. “Judeo-Spanish Textbooks and Orthographic Issues.” Leshonenu La‘am LI–LII 2:81–85. [in Hebrew]

341

BIBLIOGRAPHY Sephiha, Haim Vidal. 1988. “Ladino: A Judeo-Spanish Calque.” In Studies in Jewish Languages: Bible Translations and Spoken Dialects, ed. M. Bar-Asher. Jerusalem: Misgav Yerushalayim. 35–37. [in Hebrew] Shenhav, Yehouda. 2003. The Arab-Jews: Nationalism, Religion and Ethnicity. Tel Aviv: Am Oved Publishers. [in Hebrew] Smith, Anthony D. 1993. “Ethnic Sources of Nationalism.” Survival V, 35/1. Snir, Reuven. 2005. Arabness, Jewishness, Zionism: A Clash of Identities in the Literature of Iraqi Jews. Jerusalem: Ben-Zvi Institute. Sole, Aryeh. 1959. “Between the Two World Wars.” In Encyclopedia of Diasporas: Subcarpathian Ruthenia, ed. Y. Erez. Jerusalem: “Encyclopedia of Diaspora” Publishers Inc. 139–226. [in Hebrew] Spiro, Socrates. 1923. Arabic-English Dictionary of the Modern Arabic of Egypt. 2nd ed. Cairo: Elias’ Modern Press. Spitta-Bey, Wilhelm. 1880. Grammatik des arabischen Vulgärdialektes von Ägypten. Leipzig: J. C. Hinrich’sche Buchhandlung. Spolsky, Bernard, and Sarah Bunin Benor. 2006. “Jewish Languages.” In Encyclopedia of Language and Linguistics, ed. K. Brown. 2nd ed. Vol. 6. Oxford: Elsevier. 120–24. Spolsky, Bernard, and Elana Shohamy. 1999. The Languages of Israel: Policy, Ideology and Practice. Clevedon: Multilingual Matters. Stern, Karen B. 2008. Inscribing Devotion and Death: Archaeological Evidence for Jewish Populations of North Africa. Leiden: Brill. Stewart, Devin. 1994. “A Contribution to the Lexicography of Egyptian Arabic.” Zeitschrift für arabische Linguistik 28:36–86. Stewart, William. 1965. “Urban Negro Speech: Sociolinguistic Factors Affecting English Teaching.” In Social Dialects and Language Learning, ed. R. Shuy. Champaign, IL: National Council of Teachers of English. 10–18. Stillman, Norman A. 1988. The Language and Culture of the Jews of Sefrou, Morocco: An Ethnolinguistic Study. [Manchster, England]: University of Manchester. ———. 1991. “Language Patterns in Islamic and Judaic Societies.” In Islam and Judaism: 1400 Years of Shared Values, ed. S. Wasserstrom. Portland, OR: The Institute for Judaic Studies in the Pacific Northwest. 41–55.

342

BIBLIOGRAPHY Tedghi, Joseph. 1994. “A Moroccan Ma˙zor in Judeo-Arabic.” Massorot 8:91–160. [in Hebrew] Thomsen, John. N.d. “Features of Faroese Diglossia.” Unpublished manuscript. Timm, Erika. 2007. “Der Einfluss aschkenasischen Bibelübersetzung-stätigkeit auf die jiddische Gemeinsprach.” In The Bible in/and Yiddish, ed. S. Berger. Amsterdam: Menasseh ben Israel Institute. 21–41. Tirosh-Becker, Ofra. 2006. “An Algerian Judeo-Arabic Translation of Piyyut ‘Mi Khamokha’ by Rabbi Yehuda Ha-Levi.” Massorot 13–14:315–69. [in Hebrew] Tobi, Yosef. 1996. “Another Popular Judeo-Arabic Translation of the Pentateuch.” In Studies in Hebrew and Jewish Languages, Presented to Shelomo Morag, ed. Moshe Bar-Asher. Jerusalem: The Bialik Institute. 481–501. ———. 2002. “Translations of Sa>adia Liturgical Works into Arabic.” In Studies in the History and Culture of the Jews in Babylonia, ed. Y. Avishur and Z. Yehuda. Or-Yehuda: The Babylonian Jewry Heritage Center. 203–30. Tomiche, Nada. 1964. Le parler arabe du Caire. The Hague: Mouton. ———. 1968. “La situation linguistique en Égypte.” In Le langage, ed. A. Martinet. Paris: Gallimard. 1173–87. Turniansky, Chava. 2007. “Reception and Rejection of Yiddish Renderings of the Bible.” In The Bible in/and Yiddish, ed. S. Berger. Amsterdam: Menasseh ben Israel Institute. 7–20. Verschik, Anna. 2007. “Jewish Russian and the Field of Ethnolect Study.” Language in Society 36/2:213–32. Versteegh, Kees. 1984. Pidginization and Creolization: The Case of Arabic. Amsterdam: John Benjamins Publishing Company. ———. 1997. The Arabic Language. Edinburgh: Edinburgh University Press. Wacholder, Ben Zion. 1956. “The Haskalah and the Proselytizing of Jews During the Gaonic Era.” Historia Judaica 18. Wagner, Esther-Miriam. 2007. “A Linguistic Analysis of Judaeo-Arabic Letters from the Cairo Genizah.” Ph.D. dissertation, University of Cambridge, Cambridge. Wehr, Hans. 1976. Arabic-English Dictionary, ed. J. Milton Cowan. 3rd ed. Ithaca, NY: Spoken Languages Services. ———. 1994. A Dictionary of Modern Written Arabic, ed. J. Milton Cowan. 4th ed. Ithaca: Spoken Languages Services.

343

BIBLIOGRAPHY Wein, Martin J. 2005. “Only Czecho-German Jews? A Response to Dimitri Shumsky’s Concept of Czecho-German Jews.” Zion 70/3:383–92. [in Hebrew] ———. 2007. “Nation-Cleansing and Wars of Authenticity: Czech Nationalism and Jewish History, 1914–1952.” Ph.D. dissertation, Ben-Gurion University, Beer-Sheva, Israel. ———. 2009. “Chosen Peoples—Holy Tongues: Religion, Language, Nationalism and Politics in Bohemia and Moravia in the Seventeenth to Twentieth Centuries.” Past and Present. Forthcoming. Weinreich, Max. 1945. “Der Yivo un di problemen fun undzer tsayt.” Yivo-bleter 25/1. ———. 1980. History of the Yiddish Language. Transl. S. Noble and J. Fishman. Chicago: University of Chicago Press. Weinreich, Uriel. 1979. Languages in Contact: Findings and Problems. The Hague: Mouton. Wexler, Paul. 1981. “Jewish Interlinguistics: Facts and Conceptual Framework.” Language 57/1:99–145. ———. 2006. Jewish and Non-Jewish Creators of ‘Jewish’ Languages with Special Attention to Judaized Arabic, Chinese, German, Greek, Persian, Portuguese, Slavic (Modern Hebrew/Yiddish), Spanish, and Karaite, and Semitic Hebrew/Ladino: A Collection of Reprinted Articles from across Four Decades with a Reassessment. Wiesbaden: Harrassowitz. Woidich, Manfred. “Negation and negative Sätze im Ägyptisch-arabischen.” Ph.D. dissertation, University of Munich, Munich. ———. 1990. Materialen zur Kenntnis des kairenisch-arabischen. Amsterdam: Instituut voor het Moderne Nabiije Oosten, Universiteit van Amsterdam. Wright, W. 1974. A Grammar of the Arabic Language, Translated from the German of Caspari and Edited with Numerous Additions and Corrections. 3rd ed. 2 vols. Beirut: Librairie du Liban. Yerushalmi, David. 2000. “Judeo-Persian Bible Translations.” Pe‘amim 84:21–39. [in Hebrew] Yu, Alan. 2006. “Sound Change: Phonetic.” Encyclopedia of Language and Linguistics, ed. K. Brown. 2nd ed. Vol. 11. Amsterdam: Elsevier. 525–28.

344

BIBLIOGRAPHY Zafrani, Haim. 1980. Littératures juives en Occident musulman. Paris: Geuthner. ———. 1988. “Jewish Languages in the Maghreb under Islam and Judeo-Arabic Translations of the Bible.” In Studies in Jewish Languages: Bible Translations and Spoken Dialects, ed. M. Bar-Asher. Jerusalem: Misgav Yerushalayim. 55–67. [in Hebrew]

Bible Editions, Lexicons, and Translations: Al-Kitåb al-Muqaddas ayy kutubu al->ahdi al-qad•mi wa-l->ahdi al-jad•di. 2004. Dår al-Kitåb al-Muqaddas f• al-⁄arq al-
345

INDEX >Abd al->Al, |åda, 97n8 >Abd al->Al•m, Óusayn, 97n8 Abraham ibn Ezra, 142 accusative case (Feature 8-1), 129, 176, 177, 257–64 accusative marker, 21, 184, 257–64, 314. See also direct object, definite adjective-noun agreement, 277–78, 280–81, 282, 283, 284–85 adjectives, 123 adverbs (Feature 2-2), 129, 131, 136, 192–98, 230 afikoman (defined), 116n53 agreement (Category 10), 127–28, 274–87 alif, final, 312–16 alif mamd¥da, 104, 306, 316–17 alif maqß¥ra bi-ß¥rat il-alif, 316 alif waßla, 306 Amidah, 66, 68 Arab Israelis, 31 Arab Jews, 30–32

Arab Muslim culture, 62 Arabic language colloquial, 87, 140. See also Egyptian Arabic, colloquial; Egyptian Judeo-Arabic, spoken Classical, 137–38 development of, 44 dialects of, 24 standard, 38–40, 44, 78, 87, 95, 97, 132, 141, 147, 184, 237, 279–80, 282, 294, 300, 311, 318–19 See also Egyptian JudeoArabic; Judeo-Arabic Arabic script, 9, 19, 55 Arabic-speaking Jews, 32 Arabicized orthography. See orthography, Arabicized >arabiyya, al-, 24, 40 Aramaic, 6, 25, 48, 51–54, 61–62, 64–65, 69, 85–86, 88, 93 See also Hebrew/Aramaic elements; Hebrew/ Aramaic loans

INDEX archaic forms, 23–24, 60, 134, 142 argot, borrowing of, 17–18 article. See definite article Ashkenazi Jewry, 45 assimilation (Feature 13-1), 75, 79, 85, 107, 110, 304 Association for Jewish Studies, 47 asyndetic clauses, 89, 130, 233 />ayn/, in Egyptian JudeoArabic, 105 Babylonian Jewry, 47 Bakr•, °mån, 97 Benjamin of Tudela, 47 Berberic, 26 Bird Milk, 97 Bukharic, 26 Bustanai b. Óaninai, 15n23 ‘Cairene Purim Scroll,’ 142 Cairo Collection, 63–65 Cairo Geniza, 95, 101, 142 calque translations, 163, 170, 178–81, 200, 201–2, 217, 224, 232, 322–23 Canaanic, 26 cases (Category 8), 128–29, 257–65 Catalanic, 26 Christian languages, 9, 17 complementizer, 242–43 348

conditional particles (Feature 5-3), 234–37 conjunctions, coordinating, 231–33 continuglossia, 37, 40–44 conversion, to Judaism, 13–15 coordinating particles and conjunctions (Feature 5-2), 166–67, 173, 231–33 Copts, 92n3 co-territorial language, 7n6 Cyrillic script, 20, 55 Czernovitz conference (1908), 10, 20 Darkei no>am, 142 deemphatization, 79, 108–9, 305–6. See also emphatization defective spelling, 98, 173n2 definite article, 67–69, 110–11, 125, 147, 151, 153, 181 adding where needed (Feature 9-1), 78, 175, 177–78, 266–70 adding where not needed (Feature 9-2), 270–71 deleting where needed (Feature 9-3), 271–72 deleting where not needed (Feature 9-4), 78, 169, 174, 272–73

INDEX definiteness (Category 9), 266–73 See also accusative marker; definite article derash, 67 diacritic marks (Feature 13-5), 78, 310–11 dialect (defined), 10–11 dialectalism, migrated or displaced, 22–23, 103, 114, 134 Diaspora, 13, 48, 59, 188 diglossia, 38n12. See also continuglossia diphthongs, contraction of, 103, 308 direct object, definite, 21, 58, 69, 165, 176, 238. See also accusative marker directional case (Feature 8-2), 264–65 dual, 123 dual agreement (Feature 10-1), 274 Ecclesiastes, book of, 64 Egyptian Arabic colloquial, 79, 89, 96, 97, 101, 106, 123, 150, 165, 170, 173, 174, 177. See also Arabic language, colloquial; Egyptian Judeo-Arabic, spoken

urban, 96 See also Egyptian JudeoArabic, spoken; JudeoArabic, Cairene Egyptian Christians, 92n3 Egyptian Judeo-Arabic, spoken 23, 34n8, 37, 64, 89, 91–136, 141–43, 172, 175. See also Arabic, colloquial; Egyptian Arabic, colloquial adjectives in, 123 adverbs in, 131 agreement in, 127–28 alif mamd¥da in, 104 assimilation in, 107 asyndetic clauses in, 130 />ayn/ in, 105 case in, 128–29 data for the study of, 92, 141–42 deemphatization in, 108–9. See also tarq•q definite article in, 110–11, 125 diphthongs in, 103 dual in, 123 emphatization in, 107–8. See also tafx•m features of, summarized, 134–36 genitive marker in, 116–17 /g•m/ in, 96, 105 glides in, 106 349

INDEX hamza in, 103–4 Hebrew loans in, 102, 106, 108 interdental shift in, 104–5 lexicon of, 131–33 loss of final consonants in, 110 morphology of, 112–24 negation in, 125–27 nouns in, 122, 127–28, 132–33 numerals in, 123–24, 129–30 /p/ in, 107 phonetics and phonology of, 100–11 prepositions in, 131 pronouns in, 112–16, 125, 128, 130 /qåf/ in, 106 reconstruction of, 93–99 sound plural in, 123 sound shifts in, 109–10 syntax of, 125–30 tafx•m in, 107–8. See also emphatization tarq•q in, 108–9. See also deemphatization verbs in, 117–22, 127, 132 vowel shifts in, 100–103 vowel shortening in, 102–3, 104 Egyptian Muslims, 92n3 350

elision (Feature 13-3), 306 emphatization, 107–8, 134–35, 140, 149–50, 208 and deemphatization (Feature 13-2), 305–6. See also deemphatization ethnicity, 12, 31 ethnolect (defined), 12 Ezekiel, book of, 64 Faroese, 41 Farsi, 26 Film Industry Regulations, 45 finite verbs (Feature 7-2), 164, 167, 171–72, 252, 255, 256 /g/, in urban Egyptian Arabic, 96 gemination, 307, 309 gender, in analysis of ¡ar˙ and tefilot latini, 61 gender agreement (Feature 10-3), 67, 69, 77, 116, 128, 164, 169, 170, 174, 177, 280–87 genitive marker, 116–17, 151–52, 286–87 /≠•m/, in Egyptian JudeoArabic, 96, 105 glides, 106, 208 orthographic marking of (Feature 13-4), 307–10

INDEX Gurjic, 26 Gyoto-Italian, 23 haf†arot, 52 Haggadah, 25, 52, 64, 65 sample linguistic analyis of, 75–80 Haketiya, 26 hamza, 103–4, 138n2, 212, 306 ˙art il-yah¥d, 33, 36 head-to-toe scanning, 74–80, 90 examples of, 166–78 Hebraism, 30n2 Hebraized orthography, see Hebrew-influenced orthography (Feature 13-6); orthography, Hebraized Hebrew, translation of into JudeoArabic, 57–59 translation into JudeoSpanish 59–60 Hebrew Bible, 19, 25, 52 as sacred text, 54–55 translation of, 53–57, 60 translations of, 48, 65, 72, 182 See also medieval biblical interpretation Hebrew script, 9, 13, 19, 19n30, 54, 55–56 Hebrew/Aramaic elements, 9, 13, 17–18, 20

in argot, 17–18 in Hebraized orthography, 43 in Jewish religiolects, 21, 137 in Judeo-Arabic, 29, 33, 35–38, 40, 44, 134, 144–55 in ¡ur¥˙, 54, 156–59 Hebrew/Aramaic loans, 6, 102, 106, 108, 148–51 Hebrew-influenced orthography (Feature 13-6), 167, 170, 172, 174, 176, 177bis, 311–27 Hertz, Joseph, 48, 53 Hessonian dialects, 17 Óusayn, ˇaha, 35n9 hyperaccommodation, 16 hypercorrections, 97, 138, 139–41 hypocorrections, 95, 138, 140, 141–43, 216–17 standardization of, 78, 94–95 I’d Like to Get Married, 97n8 i∂åfa construction, 122, 123, 129–30, 136, 172–73, 266n42, 269, 271, 272 \ayr ˙aq•qiyya, 94, 267 identity, religious, 5–6

351

INDEX infinitives (Feature 7-1), 88, 164, 171, 250–56 intelligibility, mutual, 7, 10–11, 22 interdentals, 104–5, 140 loss of, 263n39 interference of Hebrew/Aramaic on Judeo-Arabic, 145–46, 153–55 of Judeo-Arabic on Hebrew/Aramaic, 144–45, 146–53 morphological, 151–53, 154–55 phonological, 148–51, 154–55 resistance to, 147–48 syntactic, 153, 155 intermediate product (IP), 86–89. See also translation, process of Isaiah, book of, 64 Israeli Hebrew, 144 Israeli Postal Service, 45 Issachar ben Susan, 56–57, 61–62 Italkian, 26 jargon, borrowing of, 17 Jeremiah, book of, 64 Jewish assimilation, 54

352

Jewish English, 16, 17, 21, 25, 27 Orthodox, 11, 16n24 Jewish identity, 13, 59, 62, 68 Jewish languages, 5n1, 6, 7, 9, 16 definition of, 8–13 development of, 8 study of, 8 See also Jewish religiolects; Jewish varieties; and entries beginning with JudeoJewish linguistics, 7, 16 Jewish literature, 22 Jewish Malayalam, 21–22, 23, 24, 25, 27, 52 Jewish religiolects, 13, 16, 19 and archaic forms, 23–24, 142 and Hebrew/Aramaic elements, 21–22, 144 and interference, 146–47 defined, 12–13. See also religiolect, defined Jewish varieties, 5–6, 25–27 characteristics of, 19–25 development of, 13–16 Jewish-defined languages, 5n1, 18–19, 23 Jews, Arabic-speaking, 32 Jidyó, 26 Job, book of, 64

INDEX Judaism, conversion to, 13–15 Judeo-Alsatian, 26 Judeo-Arabic, 7, 8, 9, 15, 25–26 Baghdadi, 22 Cairene, 22–23, 69, 91n1. See also Egyptian Judeo-Arabic Classical, 34, 35, 36, 37, 39–40, 41–42, 43, 56, 61 Contemporary, 34, 37, 43, 45–49 defined, 29–30 Dialectal Spoken, 38 dialects of Haifa, Peq•>•n, and Tiberias,105n36 Early, 34, 35 Egyptian See Egyptian Judeo-Arabic extent of, 27 folk literature in, 36, 37 history of, 30–37 Iraqi, 37, 37n11 Late, 52 Later Egyptian literary, 169 Later, 34, 36, 42, 43, 56 Literary Written, 33 Maghrebi, 37 Moroccan, 24, 46, 49 number of speakers, 49 orthography of, 20, 33, 61 periodization of, 34 Pre-Islamic, 34, 35

spoken Baghdadi, 92 Standard Written, 38 structure of, 37–39 Syrian, 37 Tunisian, 98n13 Yemenite, 37 Judeo-Arabic continuglossia, 40–44 Judeo-Arabic continuum, 38–41 Judeo-Arabic folktales, 142 Judeo-Arabic letters, 95, 101 Judeo-Arabic literature, 29, 40 Judeo-Arabic speakers, 30–32, 49 Judeo-Aramaic, 14 Judeo-Berber, 8, 15, 26, 52 Judeo-Catalan, 26 Judeo-Crimchak, 26 Judeo-Czech, 26 Judeo-Dutch, 8, 27 Judeo-French, 26 currently spoken in France, 27 Judeo-Georgian, 25, 26, 52 Judeo-German, 8. See also Yiddish Judeo-Greek, 14, 26 Judeo-Italian, 6, 8, 17, 21, 23, 25, 26, 52, 61 Judeo-Kurdish, 26 Judeo-Neo-Aramaic, 25, 26, 52 Judeo-Persian, 8, 18, 26, 52 353

INDEX Judeo-Polish, 8 Judeo-Portugese, 26 Judeo-Provençal, 8, 26, 52 Judeo-Russian, 8, 27 Judeo-Slavic, 26 Judeo-Spanish, 7, 9, 18, 19n29, 24, 25, 26, 27, 44, 45, 47, 52, 60 currently spoken in Latin America, 27 Hebrew roots, 22 Judeo-Tajik, 26 Judeo-Tat, Judeo-Tatic, 9, 26 Judezmo, 26 Juhuri, 26 Knaanic, 26 Kulturbund, 9n12 Ladino, 25, 26, 45, 52, 59–60 /lam/ (negative particle), 94–95, 126–27, 138n2, 141–43, 215–19 See also verbal negation (Feature 4-2) language co-territorial, 7n5 defined, 9–10, 10–11, 17 majority, 7n6 language variety, 11. See also variety languages, classification of, 9 Latin alphabet, 19n30, 20, 55 354

lect (defined), 12n16 lexicon (Category 3), 200–12 linguistics, Jewish, 7 literal/interpretive continuum, 68–76 literal/interpretive tension, 57–60 in ¡ar˙, 166–78, 182 See also translation; word-for-word translation (Feature 1-1) liturgical/religious terms, in ¡ar˙ texts, 158–59 loan processes, 178–79 loan translations, 179 loans, 102, 106, 108, 148–51 locative. See directional case (Feature 8-2), 264–65 lu\a al-wuß†a, al-, 40 Maimonides, Moses, 39 majority dialect, 134 majority language, 7n6, 17, 23, 24, 25 Malayalam, 19n32. See also Jewish Malayalam Maltese, 44 il-maßri wir-r•fi, 142. See also ‘Townsmen and Fellah’ matres lectionis, 207–8 Mechanism A, 86–87, 89–90, 163, 182. See also translation, process of

INDEX Mechanism B, 87–90, 163, 166, 182. See also translation, process of medieval biblical interpretation, 186 Megillat p¥r•m il-mißriyy•n, 142 megillot, 52 mEllå˙, 33, 36 methodology, 66–80, 93–99, 165–66 Middle Arabic, Literary Written, 139 Midrashic literature, 25, 52, 55. See also medieval biblical interpretation migration, 13, 22–33. See also dialectalism, migrated or displaced migration language, 6 Minor Prophets, book of, 64 mistranslations, 190n1 mixed forms, 153 Mizrahi Democratic Rainbow, 31n4 Mizrahim, 30 Modern Hebrew, 18 mood (Feature 11-2), 293–95 Moroccan Judeo-Spanish, 26 morphosyntactic level (of analysis), 74, 166, 213–302 Muslim Arabic, Morrocan, 24 Muslim languages, 9, 17

Muslim Malayalam, 19, 19n32. See also Jewish Malayalam nationalism, 31–32 negation (Category 4), 125–27, 213–19 in analysis of shar˙, 69 of nouns. See nominal negation (Feature 4-1) of verbs. See verbal negation (Feature 4-2) nominal negation (Feature 4-1), 171, 213–15 non-Jewish (term), 6n2 nouns, 123, 127–28, 132–33 with numerals (Feature 12-1), 299–302 See also cases (Category 8); nominal negation noun-verb agreement, 274–77, 279, 280, 281, 282 number, in analysis of ¡ar˙, 69 number agreement, 274–80 numerals, 123–24, 129–30 (Category 12), 299–302 (Feature 2-3), 167, 199–200 with counted nouns (Feature 12-1), 299–302 object marker, 165 Onqelos, 61, 68, 87. See also Targum 355

INDEX Orthography Arabicized, 20, 35–36, 43, 61, 177, 309, 310, 312, 313, 315, 316, 317, 320 Hebraized, 20, 33, 36, 43, 44, 104, 167, 170, 172, 174, 177 Hebrew-influenced (Feature 13-6), 167, 170, 172, 174, 177bis, 311–27 hypercorrection and, 140–41 Jewish varieties and, 20–21 Phonetic, 20, 35 phonology and, 96–99 religious symbolism of, 55–56 Talmudic, 19 orthography/phonology (Category 13), 212, 305–27 overcorrectness. See pseudocorrections /p/, in Egyptian Judeo-Arabic, 107, 154 paraphrase, in analysis of ¡ar˙, 69 Parsic, 26 participles, 67–68, 70, 119, 127, 132, 168–69, 171, 173, 218, 273, 274, 277, 282, 285, 356

287–88, 290–93 negation of, 126, 218 passive, 296–97, 299 particles, 69 as accusative markers, 259–60 coordinating, 231–34 conditional (Feature 5-3), 234–37 passive voice (Feature 11-3), 295–99 Peq•>•n, 21, 105n36 personal names. See proper names peshat, 67 Phonetic orthography. See orthography, Phonetic phonetic spelling, 169 phonetics, of Egyptian JudeoArabic, 100–11 phonology of Egyptian Judeo-Arabic, 100–11 orthography and, 96–99 phrase (term), 73n25 phrase level (of analysis), 74, 166, 183–200 Pirkei Avot, 25, 52, 59 piyyu†im, 64 place names. See proper names plural, sound, 123 plural agreement (Feature 10-2), 168, 274–80

INDEX Portugesic, 26 prepositions (Feature 5-1), 131, 152, 164, 166, 168, 171, 173, 175, 185, 219–31 in analysis of ¡ar˙, 69 infinitives and, 250–56 interference and, 152 prepositions/particles (Category 5), 219–37 prestige language, 137–38 pronominal suffixes (Feature 6-2), 238–40 pronoun-noun agreement, 278, 279, 280–81, 282–83, 284–85, 286 pronouns (Category 6), 237–50 demonstrative (Feature 6-4), 164, 174–75, 177, 244–48 in Egyptian Judeo-Arabic, 111–14, 125, 128, 130 independent personal (Feature 6-1), 167, 172, 237–38 interference and, 152 interrogative (Feature 6-5), 98, 248–50 relative (Feature 6-3), 169, 240–44 spelling of, 319–20 proper names, 156–58, 268 literal/interpretive tension and, 170

spelling of, 320–27 translation of, 83–85, 200, 202, 205–8, 209–10 proselytism. See conversion, to Judaism Psalms, book of, 64 pseudoclassical forms, 141 pseudocorrections, 39, 40, 216–17, 250 defined, 138 in the ¡ar˙ tradition, 137–43 standardization of, 93, 95 See also hypercorrections; hypocorrections Puzzle, 97n8 /qåf/, in Egyptian Judeo-Arabic, 106 Qa>•d, Y¥suf al-, 97 religiolect, 29 defined, 12–13 Jewish, 52–53 religious identity, 5–6 Revivo, Haim, 46 root choice. See word (or root) choice Rosenzweig, Franz, 53 Ruth, book of, 64 Saadia Gaon, 48, 56–57, 61, 62, 70, 87, 93, 94, 182. See also 357

INDEX Saadia ibn Yosef al-Fayy¥m• Saadia ibn Yosef al-Fayy¥m•, 36. See also Saadia Gaon ¡adda, 307, 309 SamawÅdiyå<, as-, 35 ‘sanctity by association,’ 56 ¡ar>, 25, 52 ¡ar˙ (pl. ¡ur¥˙), 25, 30, 36, 37, 39, 52 Hebrew/Aramaic lexicon in, 156–59 linguistic analysis of, 139, 163–78. See also morphosyntactic level; phrase level; segment level; word level linguistic elements of, 93 origin of, 61–63 literal/interpretive tension in, 66–74 non-standard Arabic of, 57, 59, 66, 88, 185, 218, 223, 230–31 previous research on, 65–68 reconstruction of spoken Egyptian Judeo-Arabic and, 93–99 sources for, 63–65 unique grammar and structure of, 165, 259 ¡ar˙an (pl. ¡ar˙anim), translation principles of, 358

67–68, 69, 76, 85–90, 165–66 ¡ur¥˙. See ¡ar˙ (pl. ¡ur¥˙) script, Arabic. See Arabic script script, Cyrillic. See Cyrillic script script, Hebrew. See Hebrew script scriptio defectiva, 98, 173n2 scriptio plena, 323 segment level (of analysis), 74, 166, 304–27 Semitic languages, 9 Sephardi Jews, 52 ‘Sephardi/Mizrahi’ caucus, 47–48 Sephardim, 30 Septuagint, 61, 68, 205 Shuadit, 26 Shul˙an >Arukh, 15 Siddur, 25, 52 slaves, 14–15 sociolinguistics, 5–7, 9–10 ‘Story of Hannah, The,’ 64, 100 ‘Story of Joseph, The,’ 64 ‘Story of the Book of Esther, The,’ 64 ‘Story of the Destruction, The,’ 64, 100 ‘Story of the Ten Rabbis,’ 64, 100 ‘Story of Zechariah, The,’ 64 stress, Hebrew, 149n18

INDEX suffixes, pronominal (Feature 6-2), 238–40 syntactic adaptation (Feature 2-1), 188–92 syntax, of Egyptian JudeoArabic, 125–30 tå< marb¥†a, 317–23 Tafilalt tradition, 67 tafs•r, 56–57, 71. See also Saadia Gaon tafx•m, 107–8. See also emphatization Talmud, 25, 47, 52 Talmudic orthography, 307, 309 tamsir, 25, 52 tanw•n, 253, 261, 262 Targum, 61 tarq•q, 108–9. See also deemphatization tavsili, 25, 52 taytsh, 25, 52 tefila, 25, 52 tefilot latini, 61 tefilot vulgar, 61 tense-mood-aspect (T-M-A) (Category 11), 69, 287–299 tense/aspect (Feature 11-1), 167, 168bis, 170, 171, 173–74, 176, 287–93 This Rooster Is an Ox, 97

T-M-A. See tense-mood-aspect (T-M-A) (Category 11) ‘Townsmen and Fellah,’ 142. See also il-maßri wir-r•fi translation education and, 58–59, 61, 62, 63 etymological, 54 process of, 85–90 sacred texts and, 24–25, 53–57 word choice and, 71, 75 See also calque translations; literal/ interpretive tension; Mechanism A; Mechanism B; ¡ar˙an (pl. ¡ar˙anim), translation principles of translation continuum, 83–85 uvular stop (/q/), shift of, 96–97 variety (defined), 10, 11–12 See also Jewish varieties verb conjugation (Category 7), 250–56 verbal forms, 201 verbal negation (Feature 4-2), 174, 215–19 verbs, 117–22, 127, 131 finite (Feature 7-2), 164, 167, 172, 256 359

INDEX final-hamza, 121 See also noun-verb agreement; tense-moodaspect (T-M-A) (Category 11) vocative, 195, 197 voice, passive (Feature 11-3), 295–299 vowel lengthening, 98, 148–49, 207 vowel shifts, 100–103 vowel shortening, 104 ‘Ways of Pleasantness,’ 142 Weinreich, Max, 10 women as audience for ¡ur¥˙, 53–54, 61, 62 deemphatization and, 108–9n41 word (or root) choice, 71, 75 considerations of sound/appearance and (Feature 3-2), 76, 77, 168, 170, 171, 172, 176, 205–12, 219, 220, 228–29

semantic considerations and (Feature 3-1), 168, 173, 176, 200–205 word level (of analysis), 74, 166, 200–12 word order, 9, 69, 241–42, 244–48 word order (Category 2), 188–200 word-for-word translation (Feature 1-1), 166, 167, 168, 170, 171bis, 172, 176, 183–88 Yahuudiyya, al-, 35 Yédisch-Daïtsch, 26 Yeshivish, 11 Yevanic, 14, 26 Yiddish, 7, 8, 9, 10, 12n20, 15, 17, 18, 20, 24, 25, 26, 27, 45, 46, 52, 65 and linguistic interference, 147 Yinglish, 11 YIVO (Institute for Jewish Studies), 45 Zarphatic, 26

360

Translating Religion; Linguistic Analysis of Judeo-Arabic Sacred Texts From Egypt - Hary, Benjamin H

Recommend Documents