259 53 3MB
English Pages [295] Year 2021
The (Un)Translatability of Qur’anic Idiomatic Phrasal Verbs
Qur’anic idiomaticity, in its all aspects, poses a great deal of challenge to Qur’an readers, learners, commentators, and translators. One of the most challenging aspects of Qur’anic idiomaticity is Qur’anic idiomatic phrasal verbs, where significances of proper Arabic verbs are entirely fused with significances of prepositions following them to produce new significances that have nothing to do with the basic significances of those verbs and prepositions. By examining a corpus of ten of the most influential English translations of the Qur’an, this study scrutinizes how some translators of the Qur’an have dealt with the phenomenon of Qur’anic idiomatic phrasal verbs, the difficulties that they have encountered when translating them into English, and the strategies that they have employed in their attempts to overcome the inherent ambiguity of such expressions and provide their functional-pragmatic equivalents for English readership. The study proposes a working model for analysing and assessing the translation of the Qur’anic idiomatic phrasal verbs and provides a number of theory-based recommendations for translators in general and Qur’an translators in particular. Ali Yunis Aldahesh is a lecturer in Arabic language, literature, and culture at the University of Sydney, Department of Arabic Language and Cultures, School of Languages and Cultures, Faculty of Arts and Social Sciences. He holds a PhD in translation and linguistics, an MA in translation and linguistics, and an MA in Semitic studies. Dr Aldahesh’s area of research interest is translation and linguistics, with a special emphasis on the language, text, and translation of the Qur’an. His published works include English Translations of the Qur’an: A Descriptive Comparative Study in their Aspects of Disagreement (2020), The Concept of Taqwa in the Holy Qur’an as Understood by Muslim Commentators (2010), and Translating Idiomatic English Phrasal Verbs into Arabic: A Contrastive Linguistic Study (2009).
Routledge Advances in Translation and Interpreting Studies
58 African Perspectives on Literary Translation Edited by Judith Inggs and Ella Wehrmeyer 59 Interpreters and War Crimes Kayoko Takeda 60 Simultaneous Interpreting from a Signed Language into a Spoken Language Quality, Cognitive Overload, and Strategies Jihong Wang 61 Changing Paradigms and Approaches in Interpreter Training Perspectives from Central Europe Edited by Pavol Šveda 62 Empirical Studies of Translation and Interpreting The Post-Structuralist Approach Edited by Caiwen Wang and Binghan Zheng 63 English and Translation in the European Union Unity and Multiplicity in the Wake of Brexit Alice Leal 64 The (Un)Translatability of Qur’anic Idiomatic Phrasal Verbs A Contrastive Linguistic Study Ali Yunis Aldahesh 65 The Qur’an, Translation and the Media A Narrative Account Ahmed S. Elimam and Alysia S. Fletcher For more information about this series, please visit https://www.routledge. com/Routledge-Advances-in-Translation-and-Interpreting-Studies/bookseries/RTS.
The (Un)Translatability of Qur’anic Idiomatic Phrasal Verbs A Contrastive Linguistic Study Ali Yunis Aldahesh
First published 2022 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2022 Ali Yunis Aldahesh The right of Ali Yunis Aldahesh to be identified as author of this work has been asserted by him in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested ISBN: 978-0-367-13295-8 (hbk) ISBN: 978-1-032-03433-1 (pbk) ISBN: 978-0-429-02574-7 (ebk) Typeset in Galliard by Apex CoVantage, LLC
Contents
List of figures and tables Acknowledgements List of abbreviations Table of transliteration 1
Introduction 1.1 1.2 1.3 1.4
2
ix x xi xii 1
Statement of the problem 1 Aim of the study 3 Outline of the study 4 Limitations of the study 5
(Un)translatability of the Qur’an: A theoretical perspective
8
2.1 Introduction 8 2.2 The Qur’an’s (un)translatability 12 2.2.1 The Qur’an’s (un)translatability as viewed by Muslim intellectuals 13 2.2.2 The Qur’an’s (un)translatability as viewed by English translators 18 2.3 Conclusion 31 3
Arabic idiomatic phrasal verbs: Nailing down the phenomenon 3.1 Introduction 34 3.2 Establishing a parameter for AIPVs 35 3.3 AIPVs’ components 41 3.3.1 Arabic proper verbs 41 3.3.2 Arabic preposition 46 3.4 AIPVs’ syntactic and semantic properties 51 3.4.1 al-Ta’alluq (verb–preposition relation, attachment, or dependency) 55
34
vi
Contents 3.4.2 al-Ta’addī (verb transitivity) 56 3.4.3 al-Inābah (preposition substitution) and al-Taḍmīn (verb implication) 57 3.5 AIPVs’ register variations 60 3.6 Conclusion 61
4
Qur’anic idiomatic phrasal verbs
66
4.1 Introduction 66 4.2 Syntactic idiosyncrasies of QIPVs 68 4.2.1 QIPVs’ word order 69 4.2.1.1 Nonsplit QIPVs 70 4.2.1.2 Split QIPVs 72 4.3 Semantic idiosyncrasies of the QIPVs 76 4.3.1 QIPVs’ semantic fields 78 4.3.2 QIPVs’ semantic categories 78 4.3.2.1 Metaphorical QIPVs 78 4.3.2.2 Figurative QIPVs 82 4.4 Pragmatic idiosyncrasies of the QIPVs 83 4.4.1 Direct speech acts performed by QIPVs 90 4.4.2 Indirect speech acts performed by QIPVs (QIPVs’ implicatures) 91 4.5 Conclusion 94 5
Research methodology
97
5.1 5.2 5.3 5.4 5.5
Theoretical framework 97 Data collection 103 Research questions and data analysis 106 Research procedure 107 Towards a model for analysing and assessing the translation of QIPVs 107 5.5.1 On translating metaphorical and figurative expressions 108 5.5.2 Speech acts 110 5.5.3 Conversational implicatures 110 5.5.4 Adequacy/inadequacy 111 5.6 Model for analysing and assessing translations of QIPV 114 5.7 Conclusion 115 6
Qualitative data analysis, assessment, and discussion 6.1
Case study 1 118 6.1.1 Tertium comparationis 118
118
Contents 6.1.2 English equivalents provided 6.1.3 Analysis 122 6.2 Case study 2 124 6.2.1 Tertium comparationis 124 6.2.2 English equivalents provided 6.2.3 Analysis 127 6.3 Case study 3 129 6.3.1 Tertium comparationis 129 6.3.2 English equivalents provided 6.3.3 Analysis 134 6.4 Case study 4 135 6.4.1 Tertium comparationis 135 6.4.2 English equivalents provided 6.1.3 Analysis 138 6.5 Case study 5 140 6.5.1 Tertium comparationis 140 6.5.2 English equivalents provided 6.5.3 Analysis 142 6.6 Case study 6 146 6.6.1 Tertium comparationis 146 6.6.2 English equivalents provided 6.1.3 Analysis 148 6.7 Case study 7 150 6.7.1 Tertium comparationis 150 6.7.2 English equivalents provided 6.7.3 Analysis 155 6.8 Case study 8 156 6.8.1 Tertium comparationis 156 6.8.2 English equivalents provided 6.8.3 Analysis 159 6.9 Case study 9 161 6.9.1 Tertium comparationis 161 6.9.2 English equivalents provided 6.9.3 Analysis 165 6.10 Case study 10 166 6.10.1 Tertium comparationis 166 6.10.2 English equivalents provided 6.10.3 Analysis 169 6.11 Discussion 172
120
126
131
136
141
147
152
158
162
168
vii
viii 7
Contents Conclusion 7.1 7.2 7.3 7.4
195
Summary of the study 195 Findings of the study 197 Recommendations 201 Suggestions for further research 203
Appendix I: Concordance for the Qur’anic idiomatic phrasal verbs Appendix II: Frequencies Index
205 270 275
Figures and tables
Figures 3.1 5.1
The Interrelatedness of the AIPV four factors Model for analysing and assessing translations of QIPV
60 114
Tables 4.1 4.2 4.3 4.4 4.5 6.1
Types of Qur’anic verb-preposition combinations Nonsplit QIPVs Split QIPVs Major semantic fields of the QIPVs Direct speech acts performed by QIPVs Translators’ performance when dealing with the QIPV used in the Qur’an, 2:222 6.2 Translators’ performance when dealing with the QIPV used in the Qur’an, 3:21 6.3 Translators’ performance when dealing with the QIPV used in the Qur’an, 2:255 6.4 Translators’ performance when dealing with the QIPV used in the Qur’an, 28:11 6.5 Translators’ performance when dealing with the QIPV used in the Qur’an, 18:11 6.6 Translators’ performance when dealing with the QIPV used in the Qur’an, 9:12 6.7 Translators’ performance when dealing with the QIPV used in the Qur’an, 59:2 6.8 Translators’ performance when dealing with the QIPV used in the Qur’an, 2:130 6.9 Translators’ performance when dealing with the QIPV used in the Qur’an, 9:120 6.10 Translators’ performance when dealing with the QIPV used in the Qur’an, 94:8 6.11 Translators’ overall achievement of functional-pragmatic equivalence 6.12 Ranking procedures according to their frequency and usage adequacy
67 70 73 79 90 123 128 134 139 143 149 155 160 165 170 172 188
Acknowledgements
I thank the copyright holders, who kindly gave their permission to reproduce the articles I have published with them over the previous five years or so. I am also indebted to the anonymous reviewers of the proposal for this book. Their insightful feedback, constructive criticism, and valuable suggestions have been of great benefit in deepening the idea of this work and producing it in such a thoroughgoing form. To them, I will always remain appreciative. I also thank my colleagues and students at the Department of Arabic Language and Cultures at the University of Sydney. During the different stages of this project, I was privileged to have had many profound discussions that inspired a number of fruitful thoughts. Special thanks are due to my family members for their support and patience and for the creative environment that they provided. Without them, I wouldn’t have accomplished this achievement. Finally, responsibility for any weaknesses and shortcomings that the readers might find in this work lies with me alone.
Abbreviations
Anno Hegirae, Islamic year APV Arabic phrasal verb AIPV Arabic idiomatic phrasal verb CE Common Era cf Compare CP Cooperative principle d Death date EPV English phrasal verb EIPV English idiomatic phrasal verb IP Irony principle PP Politeness principle PV Phrasal verb QPV Qur’anic phrasal verb QIPV Qur’anic idiomatic phrasal verb SL Source language ST Source text TL Target language TT Target text n.d. No date AH
Table of transliteration
The Arabic transliteration system adopted in this book is based on the romanization table proposed by the American Library Association/Library of Congress (2012).
Arabic consonants Letters
Arabic Names
Transliteration
ء ﺏ ﺕ ﺙ ﺝ ﺡ ﺥ ﺩ ﺫ ﺭ ﺯ ﺱ ﺵ ﺹ ﺽ ﻁ ﻅ ﻉ ﻍ ﻑ ﻕ ﻙ ﻝ ﻡ ﻥ ﻩ ﻭ ﻱ
hamzah bāʾ tāʾ thāʾ jīm ḥāʾ khāʾ dāl dhāl rāʾ zāy sīn shīn ṣād ḍād ṭāʾ ẓāʾ ʿayn ghayn fāʾ qāf kāf lām mīm nūn hāʾ wāw yāʾ
ʾ b t th j ḥ kh d dh r z s sh ṣ ḍ ṭ ẓ ʿ gh f q k l m n h w y
Table of transliteration xiii
Arabic short vowels Letters
Arabic Names
Transliteration
َ
fatḥah ḍammah kasrah
a u i
Letters
Arabic Names
Transliteration
ﺍ ﻯ ﻭ ﻱ
alif mamdūdah alif maqṣūrah wāw yāʾ
ā á ū ī
Letters
Arabic Names
Transliteration
ﻭ ﻱ
fatḥah + wāw fatḥah + yāʾ
aw ay
ُ ِ
Arabic long vowels
Arabic diphthongs
Other major features • • • •
• • •
When the noun or adjective ending in ( ﺓtāʾ marbūṭah) is indefinite or is preceded by the definite article, ﺓis romanized h. When the word ending in ( ﺓtāʾ marbūṭah) is in the construct state (muḍāf wa-muḍāf ilayh), ﺓis romanized t. When the word ending in ( ﺓtāʾ marbūṭah) is used adverbially, ( ﺓvocalized ً )ﺓ is romanized tan. In initial position, whether at the beginning of a word, following a prefixed preposition or conjunction, or following the definite article, ( ءhamzah) is not represented in romanization. When medial or final, ءis romanized as’ (alif ). When (hamzah) ءis replaced by the waṣlah and then known as hamzat al-waṣl, it is not represented in romanization. ّ ُ ﻭ, representing the combination of a long vowel plus a consonant, is romanized ūw. ّ َ ﻭ, representing the combination of a diphthong plus a consonant, is romanized aww.
xiv • • • • • • • • •
•
• • •
Table of transliteration Medial ﻱ ّ ِ , representing the combination of a long vowel plus a consonant, is romanized īy. Final ﻱ ّ ِ , is romanized ī. Medial and final ﻱ ّ َ , representing the combination of a diphthong plus a consonant, is romanized ayy. Over other letters, ّ (shaddah) is represented in romanization by doubling the letter or digraph concerned. Tanwīn may take the written form of ٌ , ً , or ٍ , romanized un, an, and in respectively. Final vowels of separable prepositions and conjunctions are retained in romanization. Inseparable prepositions, conjunctions, and other prefixes use a hyphen to connect to what follows them. The romanized form al (the Arabic definite article) is connected with the following word by a hyphen. When al (the Arabic definite article) is initial in the word and when it follows an inseparable preposition or conjunction, it is always romanized al, regardless of whether the preceding word, as romanized, ends in a vowel or a consonant. The ﻝof the Arabic definite article is always romanized l, whether it is followed by a “sun letter” or not, i.e. regardless of whether it is assimilated in pronunciation to the initial consonant of the word to which it is attached. Rules for the capitalization of English are followed, except that the definite article al is given in lower case in all positions. ﺍﺑﻦand ﺑﻦare both romanized ibn in all positions. The hyphen is used between bin and the following element in personal names when they are written in Arabic as a single word.
A fuller version of the American Library Association/Library of Congress transliteration system is available at https://www.loc.gov/catdir/cpso/roman.html.
1
Introduction
1.1 Statement of the problem Idiomatic expressions, in any given language, constitute headaches for translators because of their syntactic, semantic, and pragmatic complexity. This complexity is compounded when the expressions occur in scriptures to be used idiosyncratically and encumbered with sociolinguistic connotations. This study focuses on one of the most significant aspects of idiomaticity in Arabic, namely Qur’anic idiomatic phrasal verbs (henceforth QIPVs). The current study examines the vital question of the (un)translatability of the Qur’an. It approaches this issue from a wholly new perspective: the (un)translatability of QIPVs. The notion of a phrasal verb (henceforth PV) dates back to the middle of the eighteenth century, when the phenomenon of collocating proper verbs with prepositions started to draw the attention of English linguists. They identified a bizarre fusing of verbs and prepositions to produce a significance that is completely different from the sum of their literal significances. Since then, the phenomenon has been of interest to English grammarians, linguists, pedagogues, and lexicographers. They named them English idiomatic phrasal verbs (henceforth EIPVs) (cf Aldahesh, 2009). IEPV are defined as a combination of two or three elements: a verb + a preposition, a verb + an adverb, or a verb + an adverb + a preposition. The key condition of this combination is to behave idiomatically— that is, to function as a single unit of meaning in which the combination meaning cannot be comprehended from the sum of the meanings of its individual parts. Examples of IEPV include to carry out, to carry on, to turn up, to turn on, to turn off, to sort out, to give up, to give in, to give away, to get away with, to black out, to tip off, to take off, to take in, to put up with, etc. (cf Aldahesh, 2009, 2016a). The Arabic language features a similar, albeit not identical, phenomenon. Arabic does not permit proper verbs to combine with adverbs. Yet it permits verbs to combine with prepositions. Thus, of the three aforesaid EIPV structures, Arabic has only one structure—that is, a verb + a preposition structure. Like the English structures, the Arabic structure satisfies the idiomaticity condition. That is, it functions as a single unit of meaning in a sense that the structure’s overall meaning cannot be grasped from the sum of the meanings of its individual components. Examples of Arabic idiomatic phrasal verbs (henceforth AIPVs) include
2
Introduction
ʾatā ʿalā (to destroy, to finish), ʾatā bi (to bring), naẓara fī (to investigate), etc. (Aldahesh, 2016b). Like their English counterparts, AIPVs are quite common and frequent in both written Arabic and spoken Arabic. Most importantly, they constitute one of the major structures in the Qur’anic discourse. Nevertheless, this phenomenon has not been given a specific academic label by classical Arabic linguists. This is because they do not include prepositions when studying Arabic proper verbs. They deal with prepositions separately, without paying attention to their function when collocating with verbs. This being said, modern Arabic linguists have paid due attention to this crucial structure, which they call al-ḍamā’im (enclosures) (Al-Shamsān, 1986, p. 252, p. 747; Aldahesh, 2016a, pp. 13–16). Yet the treatment of AIPVs by modern Arabic linguists varies from scholar to scholar. Several have come across AIPVs in passing without accounting for their peculiarities in detail (e.g. Ryding, 2005). Others have devoted book chapters and academic articles to elaborate on AIPVs (e.g. Lentzner, 1977; Heliel, 1994; Al-Shamsān, 1987; Aldahesh, 2016b). A few researchers, however, have attended to AIPVs profoundly and come up with several fruitful insights (Dāwood, 2002; Al-Shamsān, 1987; Aldahesh, 2016a). Most of these scholars make no distinction between the idiomatic and nonidiomatic types of the Arabic verb-preposition structure. They inaccurately have taken for granted that all verb-preposition combinations are ‘idioms’ (Aldahesh, 2016a). These idiomatic combinations constitute one of the key rhetorical features of the Qur’anic texture. Like other aspects of the Qur’an’s idiomaticity, QIPVs pose many challenges to readers, learners, and commentators of the Qur’an and most importantly to the Qur’an translators who admit to the Qur’an’s (un)translatability. In previous studies, scholars have dealt with the issue of the Qur’an’s (un)translatability from many angles. For example, they investigated the (un)translatability of Qur’anic cultural aspects (Abdul-Raof, 2005), the (un)translatability of the Qur’anic pun (Dastjerdi & Jamshidian, 2011), the (un)translatability of the structural pause in the Qur’an (Al-Azzam, 2009), and the (un)translatability of Qur’anic phonetic features (Al-Amri, 2007). Yet none to date has attended to the question of the translatability/untranslatability of QIPVs. The problems coming from dealing with the phenomenon of QIPVs by Qur’an readers, learners, commentators, and translators can be summarized in the following distinct yet correlated points: 1
2
The significance of the QIPVs—comprising two main constituent parts (i.e. verbs and prepositions)—may easily be taken literally to denote the sum of the significance of their constituent parts. However, the overall significance of the QIPVs is in fact profoundly different within their Qur’anic contextual behaviour. The first constituent part of the QIPVs (i.e. the verb) might have its own concrete significance but might also have significance as a different verb (al-Taḍmīn ‘verb implication’; see Chapter 3).
Introduction 3 3
4
5
6
The second constituent part of the QIPVs (i.e. the preposition) might have its own significance but might also have significance as a different preposition (al-Inābah ‘preposition substitution’; see Chapter 3). The proper verb in QIPVs might combine with only one preposition or combine with more than one preposition in the same structure (al-Taʿaddī ‘verb transitivity’; see Chapter 3). The QIPVs’ constituent parts might appear in their unmarked word order (i.e. verbs plus prepositions) or in a marked word order (i.e. prepositions plus verbs), to satisfy particular communicative purposes. Stakeholder confusion about the literal meanings of QIPVs with their intended contextual meanings is aggravated by the lack of in-depth scholarly study dedicated to describing the functional behaviour of this crucial phenomenon; differentiating its idiomatic type from its nonidiomatic types; attending to its syntactic, semantic, and pragmatic peculiarities; and addressing the thorny issues pertinent to its rendering into other languages.
This study, therefore, focuses on flling this gap. It is important and timely for four reasons. First, it is the frst of its kind to tackle the (un)translatability of the phenomenon of QIPVs, per se. Second, by focusing on Qur’anic idiomatic expressions, this study contributes to the current heated debate on the (un)translatability of the Qur’an. Third, this study proposes an eclectic working model for analysing and assessing the translation of QIPVs. Fourth, the fndings are unprecedented in this feld, and the theory-based recommendations will be, I hope, of great beneft to all stakeholders.
1.2 Aim of the study This study concerns itself with scrutinizing the (un)translatability of QIPVs. It sheds light on the enormity of the task of translating these problematic expressions, which is due to the idiosyncratic way in which they are employed in the Qur’an and the Qur’an-specific connotations with which they are encumbered. The major aim of this study is to investigate the difficulties experienced by Qur’an translators when rendering QIPVs into English. The other purpose is to propose several recommendations for Qur’an translators on the basis of the findings derived from the practical component of this study. More precisely, this study attempts to achieve the following aims: 1 2 3
4
To investigate how translators of the Qur’an deal with QIPVs; To explore the difficulties that they encounter when rendering these expressions into English; To highlight, scrutinize, and evaluate the procedures that they employ in their attempts to overcome the inherent complexity of such expressions when providing their English equivalents; To propose an eclectic working model for analysing and assessing the translation of QIPVs;
4 5
Introduction To provide a few theory-based recommendations for translators in general and translators of the Qur’an in particular.
To achieve these aims, a corpus of ten English translations of the Qur’an produced by leading scholars in this feld is closely analysed and assessed. The main thesis of this book is that Qur’an translators experience a range of diffculties when they render QIPVs into English. However, absolutely untranslatable aspects of QIPVs are considered rare in comparison to the vast majority of translatable and relatively translatable aspects. Therefore, the translatability of QIPVs rather than their untranslatability ought to be the focus of Qur’an translators and Qur’an translation stakeholders.
1.3 Outline of the study This study comprises seven chapters and two appendices as follows: Chapter 1 provides the context for the study by highlighting the statement of the problem, underscoring the aim and rational of the study, and outlining its layout. It also elaborates on the focus of the study by stating its limitations. Chapter 2 offers a general overview of untranslatability as a key concept in the field of translation studies and addresses the question of the Qur’an’s (un) translatability in more detail. It has been broken down into three parts. Part one investigates untranslatability as a construct dealt with by linguists and translation theorists; part two surveys the different views proposed by Muslim intellectuals on the Qur’an’s untranslatability; and part three scrutinizes the views, in relation to that question, expressed by a number of leading translators of the Qur’an into English. Chapter 3 investigates the phenomenon of idiomatic verb-preposition combinations. It explores the syntactic properties of this phenomenon and accounts for its underlying semantic subtleties. It also establishes a parameter for pinpointing the phenomenon and underscoring the key factors that control the syntactic and semantic relationships between its main two constituent parts. This chapter examines several illustrative authentic contexts of Arabic usage. It takes into consideration the view of both classical linguists and modern linguists, and it uses realistic examples taken from classical literature and modern literature: the Qur’an, poetry, novels, and press articles. Chapter 4 concerns itself with scrutinizing the main topic of the book: QIPVs. It tackles this vital phenomenon from three perspectives: syntactic, semantic, and pragmatic. Chapter 5 outlines the methodology used in the present study. It focuses on depicting a working model for analysing and assessing the translation of QIPVs. In addition, the chapter details the data collection, data analysis, and other research procedures implemented in this study. Finally, it outlines the theoretical framework used—that is, the contrastive analysis approach. Chapter 6 constitutes the practical component of this study, whereby ten case studies are taken as qualitative data. Ten Qur’anic verses containing QIPVs are
Introduction 5 presented as illustrative instances, glossed in view of most authoritative and influential Arab and Muslim exegetes and lexicographers and analysed in light of the prescribed theoretical framework (contrastive analysis). The renderings given to them by the selected translators are examined, compared, discussed, and assessed by employing the proposed model for analysing and assessing the translation of QIPVs. Chapter 7 concludes this study by summarizing its findings and proposing recommendations for translators in general and the Qur’an translators in particular, and it provides suggestions for further research. Appendix I is an exclusive concordance of QIPVs enclosed in the Qur’an. It lists the QIPVs according to the Arabic alphabetical order of the roots of their main verbs. It also provides grammatical descriptions of the QIPVs and an English translation of the Qur’anic verses in which a QIPV is included. Appendix II presents lists that detail the frequencies of the QIPVs used in this study. It offers information on their distribution in the Qur’an, Arabic roots, derivatives’ word classes, and degree of idiomaticity.
1.4 Limitations of the study This is by no means a prescriptive study. Rather, it is descriptive. As is evidenced in the forthcoming chapters, Qur’anic PVs are of two distinct types: idiomatic and nonidiomatic. The focus of this study is to investigate idiomatic Qur’anic PVs. The latter type, however, will not be included because its meaning is literal/ nonidiomatic, straightforward, and transparent and because it poses no problem for readers, commentators, and translators of the Qur’an. Further details on the differences between these types and their specific properties are attended to in the forthcoming chapters. For illumination, however, a number of illustrative examples of the nonidiomatic Qur’anic verb-preposition structures not included in this study are presented next, where the two constituent parts (i.e. verbs and prepositions) are written in bold: 1
Wa-tlū ʿalay-him nabaʾa ibnay ādama bi-al-Ḥaqqi idh qarrabā qurbānan fatuqubbila min aḥadihimā wa-lam yutaqabbal min al-ʾākhari qāla la ʾaqtullannaka qāla innmā yataqabbalu Allāhu mina al-muttaq īn (Qur’an, 5:27). “But recite unto them with truth the tale of the two sons of Adam, how they offered each a sacrifice, and it was accepted from the one of them and it was not accepted from the other. (The one) said: I will surly kill thee. (The other) answered: Allah accepteth only from those who ward off (evil)” (Pickthall, 1971, p. 140, my emphasis).
2
Wa-idhā jāʾūka qālū ʾāmannā wa-qad dakhlū bi-al-kufri wa-hum qad kharajū bi-hi wa-Allāhu aʿlamu bi-mā kānū yaktumūn (Qur’an, 5:61). “When they come to thee, they say: ‘We believe’: but in fact they enter with a mind against Faith, and they go out with the same but Allah knoweth fully all that they hide” (Yūsuf ʿAlī, 1991, p. 268, my emphasis).
6 3
Introduction Wa-idh qāla al-ḥawāryyūna yā ʿisá ibna Maryama hal yastaṭiʿu rabbuka an yunazzila ʿalay-nā māʾidatan min al-samāʾi qāla itttaqū Allāha in kuntum mūʾminīn (Qur’an, 5:112). “(And,) lo, the white-garbed ones said: ‘O Jesus, son of Mary! Could thy Sustainer send down unto us a repast from heaven?’ (Jesus) answered: ‘Be conscious of God if you are (truly) believers!’” (Asad, 2011, p. 202, my emphasis).
4
Laqad arsalnā Nūhan ilá qawmihi fa-qāla yāqawmī iʿbudū Allāha mā lakum min ilāhin ghayruhu innī akhāfu ʿalay-kum ʿadhāba yawmin ʿazīm (Qur’an, 7:59). “And We sent Noah to his people; and said, ‘O my people, serve God! You have no god other than He; truly, I fear for you the chastisement of a dreadful day’” (Arberry, 1955, p. 1/178, my emphasis).
5
Innamā yuridu al-shaytānu an yuqiʿa baynakum al-ʿadāwata wa-albaghḍāʾa fī al-khamri wa-al-maysiri wa-yaṣuddakum ʿan dhikri Allāhi wa-ʿan al-ṣalāti fa-hal antum muntahūn (Qur’an, 5:91). “With intoxicants and gambling, Satan seeks only to incite enmity and hatred among you, and to stop you remembering God and prayer. Will you not give them up?” (Abdel Haleem, 2010, p. 124, my emphasis).
Finally, this study aims to provide practical insights to Arabists, contrastivists, and pedagogues in the feld of teaching Arabic to nonnative speakers and to translators from or into Arabic. No claim to comprehensiveness is made here.
References Abdel Haleem, M. A. S. (2010). The Qur’an: English Translation and Parallel Arabic Text. Oxford University Press. Oxford. Abdul-Raof, H. (2005). Cultural Aspects in Qur’an Translation. In Lynne Long (Ed.), Translation and Religion: Holy Untranslatable. Multilingual Matters LTD. Clevedon, Buffalo, Toronto. Al-Amri, W. B. (2007). Qur’an Translatability at the Phonic Level. Perspectives: Studies in Translatology, 15 (3), 159–176. Al-Azzam, B. (2009). Translating Structural Pause in the Qur’an. Translation and Interpreting Studies, 4 (1), 67–84. Aldahesh, A. Y. (2009). Translating Idiomatic English Phrasal Verbs into Arabic. VDM Verlag Dr. Muller. Saarbrucken, Germany. Aldahesh, A. Y. (2016a). Pinning Down the Phenomenon of Phrasal Verbs in Arabic. International Journal of Language and Linguistics, 3 (1), 12–24. Aldahesh, A. Y. (2016b). Towards a Model for Analyzing and Assessing Translation of Qur’anic Idiomatic Phrasal Verbs. Arab World English Journal, Special Issue on Translation (5), 33–53. al-Shamsān, A. I. (1986). Al-Fiʿilu fī al-Qurʿani al-Karīmi Taʿadyatuhu wa-Luzūmuhu. The Kuwait University. Kuwait.
Introduction 7 al-Shamsān, A. I. (1987). Qaḍāyā al-Taʿadī wa-l-Luzūmi fī al-Darsi al-Naḥwī. Maṭbaʿatu al-Madanī. Jiddah, Saudi Arabia. Arberry, A. J. (1955). The Koran Interpreted. George Allen & Unwin LTD. London; The Macmillan Company. New York. Asad, M. (2011). The Message of the Qur’an. Islamic Book Trust. Kuala Lumpur. Dastjerdi, H. V., & Jamshidian, E. (2011). A Sacrament Wordplay: An Investigation of Pun Translatability in the Two English Translations of the Quran. Asian Social Science, 7 (1), 133–144. Retrieved from http://www.ccsenet.org/journal/index. php/ass/article/view/7132/6533 Dāwood, M. M. (2002). Al-Qurʿan al-Karīm wa-Tafāʿulu al-Maʿānī: Dirāsatun Dilāliyatun li-Taʿlluqi Ḥarfi al-Jarri bil-Fiʿili wa-ʾAtharihi fī al-Maʿná fī al-Qurʿani al-Karīm. Dār Gharīb. Cairo, Egypt. Heliel, M. H. (1994). Verb-Particle Combinations in English and Arabic: Problems for Arab Lexicographers and Translators. In R. De Beaugrand, A. Shunnaq, & M. H. Heliel (Eds.), Language, Discourse and Translation in the West and Middle East. John Benjamins Publishing Company. Amsterdam and Philadelphia. Lentzner, K. R. (1977). Semantic and Syntactic Aspects of Arabic Prepositions. Unpublished doctoral dissertation, Georgetown University, Washington, DC. Pickthall, Marmaduke (1971). The Meaning of the Glorious Qur’ân. Dar al-Kitab alLubnani. Beirut, Lebanon. Ryding, K. G. (2005). A Reference Grammar of Modern Standard Arabic. Cambridge University Press. Cambridge. Yūsuf ʿAlī, ʿA. (1991). The Meaning of the Holy Qur’ān. New Edition with Revised Translation and Commentary. Amana Corporation. Brentwood, MD.
2
(Un)translatability of the Qur’an A theoretical perspective
2.1 Introduction (Un)translatability as a construct can be clearly understood by juxtaposing it with the notion of translatability. Translatability is generally defined as “the capacity for some kind of meaning to be transferred from one language to another without undergoing radical change” (Pym & Turk, 2001, p. 273). It is, to use Hatim and Munday’s (2004) words, “a relative notion that has to do with the extent to which, despite obvious differences in linguistic structure (grammar, vocabulary, etc.), meaning can still be adequately expressed across languages” (p. 15). (Un)translatability is the opposite of this notion. Newmark (1988) defines untranslatable words as those that “have no ready one-to-one equivalent in the TL [target language]; they are likely to be qualities or actions—descriptive verbs, or mental words—words relating to the mind, that have no cognates in the TL” (p. 17). Current interest in (un)translatability, as a vital construct in translation studies, has generated a considerable body of literature. It constitutes one of the most debatable issues among scholars in such fields as philosophy, linguistics, and translation theory. In tackling the construct from a variety of perspectives, scholars differ on the possibility/impossibility of conveying a given text from one language into another. They also differ in how they categorize aspects of this construct. A close look at the literature reveals that scholars have adopted two approaches to the notion of (un)translatability: the monadist approach and the universalist approach (De Pedro, 1999). Scholars (e.g. Nida, Jakobson, Bausch, Hauge, and Ivir) who have adopted the first approach believe that translatability is ensured by the existence of linguistic universals (syntactic and semantic categories). This became the basis of Noam Chomsky’s transformational generative grammar (TGG). Scholars who have adopted the second approach claim that reality is interpreted in different ways by different linguistic communities; therefore, translatability is jeopardized (De Pedro, 1999). This is known as the Sapir–Whorf hypothesis, named after Edward Sapir and Benjamin Lee Whorf, who “exploited this hypothesis in [the US], where anthropological study of Native American cultures had opened new paths to linguistics” (De Pedro, 1999, p. 547). Sapir
(Un)translatability of the Qur’an 9 affirms that “the ‘real world’ is to a large extent unconsciously built up on the language habits of the group. . . . The worlds in which different societies live are distinct worlds, not merely the same world with different labels attached” (Steiner, 1992, p. 91; cited in De Pedro, 1999, p. 547). Throughout the 1930s and 1940s, this hypothesis developed into “a theory according to which the fact that each linguistic community has its own perception of the world, which differs from that of other linguistic communities, implies the existence of different worlds determined by language” (De Pedro, 1999, pp. 547–548). In addition, researchers cast light on a third approach to translatability, which emerged in France in the late 1960s: the deconstructionist approach. It questions the notion of translation as a transfer of meaning. Initiators of this approach (Andrew Benjamin, Michel Foucault, Paul de Man, and Jacques Derrida) claim that the way the translation of a text is perceived is affected by the translation, which involves a “rewriting” of the original text. Therefore, “[t]arget texts cease to be considered as subsidiaries of the original, which, in turn, becomes dependent on translation” (De Pedro, 1999, p. 554). Three important concepts have been reconceptualized by this approach: the translation process, originality, and the authorship of a text. First, the translation process is “a validation of the text that is being translated”. Second, originality “ceases to be a chronological concept (i.e. it is not about which text was produced first) and becomes a qualitative matter (i.e. it refers to the nature of the text which was conceived first)”. Third, the issue of authorship “is challenged and translation is seen as a process in which language is constantly modifying the source text” (De Pedro, 1999, p. 554). From a linguistic perspective, Catford (1965) proposes two types of (un)translatability: linguistic and cultural. He considers (un)translatability as a translational failure that occurs when the functionally relevant features of a given text are formal and there exists no formal correspondence of these features between the source language (SL) and the TL. Catford’s conception of linguistic (un)translatability is deemed straightforward in comparison with cultural (un)translatability, which is more problematic (Bassnett, 2002, p. 39). Cultural (un)translatability occurs when the functionally relevant features of a given text are cultural ones and there exists no cultural correspondence of these features between the SL and the TL. Catford writes, Translation fails—or untranslatability occurs—when it is impossible to build functionally relevant features of the situation into the contextual meaning of the TL text. Broadly speaking, the cases where this happens fall into two categories. Those where the difficulty is linguistic, and those where it is cultural. (p. 94, emphasis in original) Catford (1965) exemplifes linguistic (un)translatability in SL puns where ambiguity and play on words are the functionally relevant features (p. 94). However, he considers cultural untranslatability as “less ‘absolute’ than linguistic untranslatability” (p. 99) and exemplifes it in the various concepts given to the term bathroom in three distinct languages: English, Finnish, and Japanese.
10
(Un)translatability of the Qur’an
Considering the dynamic nature of language and culture, Bassnett (2002) goes beyond the aforesaid definition of cultural (un)translatability to proclaim that Catford rebuts this category by not considering its dynamic nature. She writes, Catford starts from different premises, and because he does not go far enough in considering the dynamic nature of language and culture, he invalidates his own category of cultural untranslatability. In so far as language is the primary modelling system within a culture, cultural untranslatability must be de facto implied in any process of translation. (p. 41) The basic division between linguistic (un)translatability and cultural (un)translatability has encouraged some scholars in the feld of translation to make a key assumption: “a dichotomy can be established between the translation diffculties that have their origin in the gap between source language and target language, and those which arise from the gap between source culture and target culture” (De Pedro, 1999, p. 551). Along these lines, (un)translatability has been categorized from a sociosemiotic perspective into three types: referential, pragmatic, and intralingual (Ke, 1999). These categorizations have been proposed in the light of the sociosemiotic view of meaning and translation. Scholars in this field perceive meaning as “[a] relationship between a sign and something outside itself. Such a relationship is fundamentally conventional, i.e. language specific” (Ke, 1999, p. 290). In addition, they identify the following three dimensions of sign relationships: (1) Semantic signifies “the relationship between signs and entities in the world which they refer to or describe” (p. 290); (2) Pragmatic denotes the relationship between “signs and their users (interpretants)” (p. 290); (3) Syntactic indicates the relationship between the signs (p. 290). According to these three kinds of semiotic relationships, such scholars propose three categories of sociosemiotic meaning: referential meaning, pragmatic meaning, and intralingual meaning (Ke, 1999). Hence, referential (un)translatability “occurs when a referential element in a source message is not known or readily comparable to a particular item in the target language” (Ke, 1999, p. 292). In contrast, pragmatic (un)translatability arises where some pragmatic meaning encoded in a source item is not encoded likewise in a functional comparable unit in the target language, or where the exact pragmatic meaning(s) carried by the source sign is/are unclear or indeterminable due to historical reasons or to the intentional equivocation on the part of the author (as may be found in some theological and mystic writings). (Ke, 1999, p. 293)
(Un)translatability of the Qur’an 11 Intralingual (un)translatability, however, occurs in “any situation in which the source expression is apparently not transferable due to some communicatively foregrounded linguistic peculiarity it contains. . . . Intralingual untranslatability accounts for a majority of cases of untranslatability” (Ke, 1999, pp. 293–294). What is more, “[i]f referential and pragmatic untranslatability are relative, intralingual untranslatability is usually ‘absolute’, since languages differ from each other more in their structure . . . than in the communicative functions they may be employed to perform” (Ke, 1999, p. 295). From a sociosemiotic point of view, two causes underlie the problem of (un) translatability (Ke, 1999, pp. 291–292): (1) The concurrence or combination of referential meaning . . . pragmatic meaning . . . and intralingual meaning . . . in a linguistic sign in different languages is a matter of convention. The three categories of socio-semiotic meaning carried by an expression in one language will often not coincide with those of a comparable expression in another language; (2) Annotation, which is capable of elucidating virtually any kind of linguistic or cultural peculiarities, cannot be unrestrictedly employed, at least not in most literary works, for the practical reason that it would make the translation longwinded and cumbersome. (Just consider the case of movie translation, where annotation or other forms of explanation are usually not possible owing to the time limit). Recognizing this, it is undeniable that genre plays an important role in the translatability of texts; some genres are more easily translatable than others (De Pedro, 1999). For instance, “a text with an aesthetic function will contain elements which will make its reproduction in a different language diffcult, whereas a text with a merely informative function will be easier to translate” (De Pedro, 1999, pp. 552–553). Texts have been categorized according to the degree of their translatability as follows (De Pedro, 1999, pp. 552–553): (1) Exclusively source-language oriented texts—relatively untranslatable. (2) Mainly source-language oriented texts (e.g. literary texts)—partially translatable. (3) Both source-language-texts and target-language-oriented texts (e.g. texts written in a language for specific purposes)—optimum translatability. (4) Mainly or solely target-language-oriented texts (e.g. propaganda)—optimum translatability. To this end, scholars have two positions on the translatability/(un)translatability of texts from a given SL into any TL. While some (e.g. Von Humboldt, Quine, and Virginia Woolf, among others) insist that translation is ultimately impossible, others (e.g. Newmark) believe that everything is translatable and can be translated either directly or indirectly into a TL (Ke, 1999). The latter position, I believe, is more sensible, realistic, and practical than the former one. There is a tendency among stakeholders to presuppose that most texts are translatable and
12
(Un)translatability of the Qur’an
that “absolute untranslatability, whether linguistic or cultural, does not exist” (De Pedro, 1999, pp. 556–557). They assert this since absolute “untranslatables” are considered the minority in comparison to the vast majority of “translatables and relative translatables” (Ke, 1999, p. 297). The expanded conceptualization of translation, and the many strategies that a translator can resort to when confronted with a linguistic or cultural gap between two languages, has meant that the debate on translatability versus (un)translatability has recently become unpopular and lost part of its validity (De Pedro, 1999). This does not mean that perfect translation is attainable; rather, it means that “[a] practical approach to translation must accept that since not everything that appears in the source text can be reproduced in the target text, an evaluation of potential losses has to be carried out” (De Pedro, 1999, pp. 556–557). My orientation is substantiated by the aforementioned monadist approach to the notion of (un)translatability. The initiators of this approach believe translatability is assured by the existence of linguistic universals. These universals constitute the crux of Noam Chomsky’s well-thought-of TGG theory.
2.2 The Qur’an’s (un)translatability A great deal of literature has been devoted to the question of the (un)translatability of the Qur’an. Scholars in such fields as Islamic studies, theology, and linguistics have accounted for this essential issue from a variety of perspectives. Muslim theologians, from the middle of the eighth century to the present, have debated this issue from the perspectives of the lawfulness and permissibility/ impermissibility of translating the Qur’an into other languages. In the early centuries of Islam, the following two main questions arose (Tibawi, 1962, p. 4): (1) Is it permissible to translate the Arabic Qur’an into another tongue? (2) Is it lawful to recite the translated Qur’an in prayer? These two controversial questions have been the focuses of great debate among Muslim scholars, clerics, and intellectuals throughout Islamic history. The debate has resulted in two distinct standpoints, each of which draws on the Qur’an and the Sunna ‘prophetic tradition’. On the one hand, one standpoint maintains the Qur’an should not be translated into other languages. This is because the divine nature of the scripture and its unique linguistic and cultural features make it an untranslatable text. On the other hand, another standpoint attests that the Qur’an should be translated into other languages despite its divine nature and linguistic and cultural uniqueness. This view is based on the belief that as “the fnal divine Word to [humankind], its message ought to be understood by all human beings regardless of their religious, linguistic, and cultural backgrounds, and there is no way to do so but to translate it into other languages” (Aldahesh, 2017, p. 178). However, because of space and time constraints, questions of the lawfulness and permissibility/impermissibility of translating the Qur’an, albeit
(Un)translatability of the Qur’an 13 pertinent the issue of (un)translatability, will not be addressed in this book. This signifcant area merits its own separate study. Scholars interested in such fields as Tafsīr al-Qur’an (exegesis of the Qur’an) and ʿUlūm al-Qur’an (sciences of the Qur’an) have also contributed significantly to the debate about the Qur’an’s (un)translatability. The perspective from which these scholars have approached the topic is the issue of Iʿjāz al-Qur’an (the miraculous inimitability of the Qur’an). What follows is a presentation of the major contributions to the debate of the Qur’an and its (un)translatability made by Muslim intellectuals. Then, the major contributions made by several translators of the Qur’an into English are discussed.
2.2.1 The Qur’an’s (un)translatability as viewed by Muslim intellectuals In ‘Translating the Qur’an’, Fazlur Rahman (1988) asserts that the inspired language of the Qur’an “can never be completely satisfactorily translated into another language” (p. 24). He posits two reasons for the difficulties in adequately translating the Qur’an into other languages: “the style and expression of the Qur’an” and the nature of the scripture. He writes, [T]he fact [is] that the Qur’an is not really a single “book”, because nobody ever “wrote” it: it is an assembly of all the passages revealed or communicated to Muhammad by the Agency of Revelation, which the Qur’an calls Gabriel and “The Trusted Spirit” or “The Holy Spirit”. This agency, according to the Qur’an itself, emanates from the “Preserved table”, the Book on High, and “descended upon your heart” (Qur’an, 2:97). Clearly, the divine messages broke through the consciousness of the Prophet from an agency whose source is God. (p. 24) Furthermore, Fazlur Rahman (1988) claims that modern Western scholars who have attempted to translate the Qur’an into their languages “unanimously agree on the untranslatability of the Book” (p. 24). The claim is substantiated by referring to two translations of the Qur’an: The Koran Interpreted, conducted by Arthur John Arberry, and The Meaning of the Glorious Qur’an, conducted by Mohammad Marmaduke Pickthall. One can infer from the wording of these titles that they are “intended to convey to the reader the idea that an adequate translation of the Qur’an is impossible” (p. 24). Along these lines, von Denffer (1983), in his book ‘Ulūm al-Qur’an: An Introduction to the Sciences of the Qur’an, initiates his input on the Qur’an’s (un)translatability by pointing out that translating the text means expressing its meanings in a language other than Arabic, to help those who are not familiar with the Arabic language to know about the Qur’an and to “understand Allah’s guidance and will” (p. 143). Von Denffer (1983) goes on to indicate that “[t]here is agreement among Muslim intellectuals that it is impossible to transfer the original Qur’an
14
(Un)translatability of the Qur’an
word by word in an identical fashion into another language” (p. 143). He proposes the following three reasons for this impossibility (p. 143): (1) A lack of equivalents, words of different languages may express specific concepts, but not all the shades of meanings of their counterparts; (2) Narrowing down the meaning of the Qur’an to specific concepts in a foreign language means leaving out other important dimensions; and (3) Presentations of the Qur’an in a different language would therefore result in confusion and misguidance. Therefore, a word-for-word Qur’an translation, according to von Denffer (1983), would not be enough. Further, on the basis of the comments of “good translators” who aim at frst determining the meaning of a Qur’anic passage and then translating it into the other language, he emphasizes that “translations of the Qur’an are actually expressions of meanings of the Qur’an in other languages” (p. 144). What is more, in his endeavour to elaborate on the limitations of Qur’an translations, von Denffer (1983) indicates that since the Qur’an, as ‘the word of Allah’, has been revealed in the Arabic language, all translations of it into other languages would not be ‘the word of Allah’. Another limitation von Denffer (1983) highlights is the loss of the “uniqueness and inimitability of the Qur’an (Iʿjāz al-Qur’an)”, which is linked by scholars to the Qur’anic expression in the Arabic language (p. 145). The last limitation is the inadequacy of the translation to express the meanings of the Qur’an in languages other than Arabic because of the distinctive shades of word meanings carried in different languages (p. 145). Additionally, von Denffer (1983) underlines the importance and benefits of translating the meanings of the Qur’an. Given that most Muslims are nonnative speakers of Arabic, translating the Qur’an into their mother tongue would help them to familiarize themselves with its meanings. Hence, translating the Qur’an is considered an effective daʿwah (missionary invitation) to both Muslims and non-Muslims. It “present[s] the message of Islam to non-Muslims and invite[s] them to ponder over the Qur’an” and “point[s] out to Muslims the revealed guidance and will of Allah to be observed by them” (pp. 145–146). Likewise, in the article ‘Translating the Untranslatable: A Survey of English Translations of the Quran’, Abdur Raheem Kidwai (1987) notes that the act of translating the Qur’an into other languages is a natural part of the exegetical effort conducted by Muslims. However, the emotional motives behind such an act have always been “looked upon with suspicion”. Moreover, in Bibliography of the Translations of the Meanings of the Glorious Qur’an into English 1649–2002: A Critical Study, Kidwai (2007) remarks on the growing body of literature devoted to the “twin issues of the Quran translation and translatability” (p. xx). He indicates that there have been two issues in this regard: “the debate on the desirability of a translation of the Quran” and “the specific linguistic and socio-cultural problems in translating the Quran into English in particular” (p. xx).
(Un)translatability of the Qur’an 15 Against this background, a major contribution to the debate surrounding the (un)translatability of the Qur’an has been made by Hussien Abdul-Raof (2001) in his valuable work Qur’an Translation: Discourse, Texture and Exegesis. The author outlines and substantiates the question of the Qur’an’s (un)translatability by providing Qur’anic examples at linguistic, rhetorical, micro, and macro levels. In his endeavour to determine what makes the Qur’an an untranslatable text, Abdul-Raof (2001) ably provides a comprehensive analysis of the limits of Qur’anic translatability. He does this by explaining the linguistic and rhetorical limitations that shackle the translator of the Qur’an. He tackles this issue from all its possible perspectives: [S]tyle, stylistic mechanism of stress, word order, cultural voids, problems of literal translation, syntactic and semantic ambiguity problems, emotive Qur’anic expressions, disagreement among Qur’an translators, different exegetical analyses, morphological patterns, semantico-syntactic interrelation, semantic functions of conjunctives, semantico-stylistic effects, prosodic and acoustic features, and most importantly the shackles imposed by the thorny problem of linguistic and rhetorical Qur’an-specific texture. (p. 1) Furthermore, Abdul-Raof (2001) indicates that the translation of the Qur’an is not, and should not be considered, the replacement of the original Arabic version, because “we cannot produce a Latin Qur’an no matter how accurate or professional the translator attempts to be” (p. 1), for two reasons. First, the Qur’an-bound expressions and structures “cannot be reproduced in an equivalent manner to the original in terms of structure, mystical effect on the reader, and intentionality of source text”. Thus, any Qur’an translation will inevitably emerge with inaccuracies and skew sensitive Qur’anic information as a by-product. Second, the divine nature of the Qur’an, as the word of God, “cannot be reproduced by the word of [humankind]” (p. 1). Abdul-Raof (2001) uses the Qur’an translation as a vehicle to examine the extent to which the translation theory is applicable and to examine the human capacity to interpret the meanings of the word of God to other nations of different tongues. He emphasizes that “the beauty of the Qur’an-specific language and style surpasses man’s faculty to reproduce the Qur’an in a translated form” (p. 2). He confirms, however, the possibility of producing a “crude approximation of the language, meanings and style of the Qur’an”—to enable nonspeakers of Arabic to appreciate the meaning of the text. Abdul-Raof (2001) substantiates such a thesis in referring to the reluctance of Qur’an translators to claim that their translations are typical or the equivalent of the Qur’an (p. 2). Abdul-Raof’s (2001) notion of crude approximation implies “a pragmatic translation of the surface meanings of the Qur’an and the provision of linguistic and rhetorical patterns suitable for the target language” (p. 2). He proposed this notion, which is, to use his words, “the most we can hope for” (p. 13), as a
16
(Un)translatability of the Qur’an
strategy for communicative purposes and as a practical solution to Qur’an translation problems. This is due to two reasons: [First], the gap between translation theory and practice remains unbridged and what applies as a solution to one language cannot apply to another. Second, there is, I believe, no possible theoretical or practical solution to Qur’an translation problems for Qur’anic expressions as well as linguistic/rhetorical features remain Qur’an-specific; to force them into a target language is to deform and sacrifice the linguistic architecture of the source text; the flow of sound is sacrificed to meaning while in the Qur’an sound and meaning are closely interrelated. (p. 2) What is important, according to Abdul-Raof (2001), is to frequently inform the TL readers that what they are reading is merely a ‘crude approximation’ of the Qur’an produced to assist them in reading and understanding the text but that it is not a substitute for the Qur’an itself (pp. 2–3). Hence, he fnds justifable the objection of Muslim intellectuals—both traditional and modern—who call the translated version of the Qur’an ‘Qur’an’. It is rather a “translation of the meanings of the Qur’an” (Abdul-Raof, 2001, p. 13). Further, in his attempt to account for the Qur’an-specific linguistic and stylistic aspects that defy translation and constitute the grounds for claims of the Qur’an’s (un)translatability, Abdul-Raof (2001) tackles the topic from six distinct but interrelated perspectives. These are form, word order, the use of transliteration, the special syntactic structures, Qur’anic particles, and Qur’anic style (pp. 42–54). Regarding the competence of translators who intend to translate the Qur’an, Abdul-Raof (2001) points out that sound linguistic competence in the Arabic and English languages is not enough. Rather, the translator has to have, in addition to that, an “advanced knowledge in Arabic syntax and rhetoric in order to appreciate the complex linguistic and rhetorical patterns of Qur’anic structures” (p. 2). Additionally, and most importantly, to derive and deliver the precise underlying meaning of the Qur’anic expressions and particles and prepositions, translators ought to compare and refer to the main Qur’an exegeses (p. 2). All things considered, Abdul-Raof (2001) concludes that the “Qur’anic discourse is inimitable and cannot be reproduced into a target language” (p. 3). Similarly, in Cultural Aspects in Qur’an Translation, Abdul-Raof (2005) refers to Muslim intellectuals who believe that “the Qur’an is untranslatable since it is a linguistic miracle with transcendental meanings that cannot be captured fully by human faculty. This is why we find titles like The Meanings of the Qur’an or The Message of the Qur’an, but The Qur’an are not used as a title for translated text” (p. 162). Drawing on Bassnett and Lefevere, Abdul-Raof (2005) adds that for these scholars, the Latin version of the Qur’an can never be considered a replacement of the original Qur’an for translation. According to them, it is “a traducement, a betrayal, an inferior copy of a prioritised original” (p. 162). By the same token, Abdul-Raof (2005) points out that since the Qur’an was revealed
(Un)translatability of the Qur’an 17 in the pre-Islamic Arabian context, which is completely distinct from other cultures outside the Arabian Peninsula, domesticating the “Qur’an-specific cultural expressions as well as Qur’an-specific linguistic patterns” by the linguistic norms of TLs is impossible (p. 162). Abdul-Raof (2001) asserts that (un)translatability as discussed by him is “much more complicated than the simplistic notion of untranslatability referred to by Tancock” (1958, p. 32). Tancock’s untranslatability may arise if a translator insists upon rendering the verb by a verb, and an adjective by an adjective, and so on (p. 40). Moreover, Dastjerdi and Jamshidian (2011) examined the strategies applied by two translators of the Qur’an (i.e. Arberry and Pickthall) when tackling the phenomenon of pun in the Qur’an. When investigating the question of the (un) translatability of such pun use, the authors concluded that even though the translators were well informed about pun use in the Qur’an, much of the aesthetic value of the Qur’anic puns is lost in their translations. Interestingly, they maintain that [o]ne of the most important factors in making the Quran untranslatable is its use of the untranslatable in its text. That is what makes the Quran unique. In some cases, two or three puns are mixed in such an elegant way that no translator can reproduce them in the target language. (p. 141) They conclude that “the results obtained from the present study established the notion of untranslatability of puns in the two English translations of the Quran” (p. 141). Qur’anic puns aside, in The Qur’an: An Introduction, Abdullah Saeed (2008) elaborates on Muslim discourse on translating the Qur’an. He cites several fatwas (Islamic opinions given by qualified scholars) in support of the traditional view that the Qur’an cannot be literally translated, in that the translation constitutes an equivalent to the Qur’an itself. This is neither possible nor permissible. Hence, Muslim intellectuals refer to a “translation of the meanings of the Qur’an” (p. 139). Further, Saeed (2008) highlights the theological and linguistic reasons provided by Muslim intellectuals for their viewpoint. The theological reason they propose is that the Qur’an is the Word of God and, hence, has a unique style that cannot be matched, even in Arabic. They argue that if a piece of writing like the Qur’an cannot be imitated in Arabic, it follows that it can never be replicated in an entirely different language. (p. 126) The linguistic argument, on the other hand, includes a number of reasons, including “the richness of the Arabic language”, “the existence of certain untranslatable terms”, and “the fact that a translation can never be completely exact or neutral” (p. 139).
18
(Un)translatability of the Qur’an
Muzaffar Iqbal (2000) affirms in his article ‘ʿAbdullah Yusuf Ali and Muhammad Asad: Two Approaches to the English Translation of the Qur’an’, that all those who embark on the task of translating the Qur’an have admitted the enormity of such a task and arrived at the conclusion that the text was untranslatable. Nevertheless, the Qur’an has been translated into almost all living languages. According to Iqbal, existing Qur’an translations echo the translators’ understandings of the Qur’an, their intellectual makeup and spiritual makeup, their linguistic and ideological limitations, and, to a great extent, their social, economic, and political backgrounds (Iqbal, 2000). To conclude thus far, Muslim intellectuals believe that the Qur’an is untranslatable into other languages. They draw attention to the issue of Iʿjāz al-Qur’an (the inimitability/miraculousness of the Qur’an) and attribute the Qur’an’s (un)translatability to a number of factors. They include, but are not limited to, the unique style of the Qur’an; the inevitable linguistic gaps among languages because of syntactic, semantic, and pragmatic specificities; the inevitable cultural voids; the linguistic and rhetorical Qur’an-specific texture; and the semantically, syntactically, and stylistically motivated Qur’an-bound morphological forms. Muslim intellectuals, on the other hand, assert that translating the meanings of the Qur’an into other languages is possible and preferable. It is considered an effective daʿwah (missionary invitation) to both Muslims and non-Muslims. Yet a word-for-word Qur’an translation is neither enough nor possible. Moreover, scholars of the Qur’an agree that only the original Arabic version is considered the word of God or muʿjiz (unique and miraculously inimitable). Since the Qur’an is a linguistic miracle, only a crude approximation of its language, meanings, and style is possible. That is, translations, no matter how accurate they are, must not be thought of as substitutions of the original Arabic version. Scholars also agree that reproducing the attractiveness and beauty of Qur’an-specific rhetoric is beyond humankind’s faculties. Thus, Qur’an-specific properties are lost in translation, and hence, inaccuracy and skewing are by-products of such translations. Finally, one can safely assume that for Muslim intellectuals, three types of (un)translatability exist when translating the Qur’an: linguistic, cultural, and theological. While the first and the second types, as explored by Catford (1965), are inevitable and pertinent to translating any genre of text, the last type is exclusive to the Arabic Qur’an. Moreover, it is driven by Muslims’ profound belief in the Qur’an as the word of God, which cannot be reproduced by the words of humankind. Having said that, let us review the attitudes of highly influential translators of the Qur’an into English. This will assist us in sketching a clear picture of the plausible reasons for the (un)translatability of the scripture.
2.2.2 The Qur’an’s (un)translatability as viewed by English translators This section scrutinizes introductory materials written by several prominent translators of the Qur’an into English. It explores their attitudes towards the
(Un)translatability of the Qur’an 19 issue of the (un)translatability of the Qur’an, the possible reasons for their perceptions of the (un)translatability of the Qur’an, and the mechanisms they use to deal with (un)translatability issues. In Muhammad Marmaduke Pickthall’s (1971) translation, The Meaning of the Glorious Qur’an: Text and Explanatory Translation (1930), he stresses the orthodox view that “the Qur’an cannot be translated” (p. i). This is made clear by the author in his foreword, and like many Muslims translators of the Qur’an, he makes no claim that what has been produced was an equivalent of the original Qur’an. He writes, The Qur’an cannot be translated. This is the belief of old-fashioned Sheykhs and the view of the present writer. The book is here rendered almost literally and every effort has been made to choose befitting language. But the result is not the Glorious Qur’an, that inimitable symphony, the very sound of which move men to tears and ecstasy. It is only the attempt to present the meaning of the Qur’an—and peradventure something of the charm—in English. It can never take the place of the Qur’an in Arabic, nor it is meant to do so. (p. i) Pickthall, however, stops short of providing details on why “the Qur’an cannot be translated”. Likewise, he does not elaborate on the types of diffculties that he encountered while undertaking the translation. Nonetheless, he refers to his determination to face the diffculties while conceding that “when diffculties were encountered the translator had recourse to perhaps the greatest living authority on the subject” (p. i). In addition, he lists the exegetical works and biographies of the Prophet Muhammad that he consulted, alongside modern scholars who helped him to clarify the “old meanings of Arabic words not to be found in dictionaries” (p. i). In his endeavour to translate as close as possible to the source text, Pickthall adheres to literal translation that accommodates the mainstream opinions of traditional commentators. When deviation from these opinions is inevitable, he resorts to using footnotes (p. i). In Abdullah Yusuf Ali’s (1991) translation, The Meaning of the Holy Qur’an (1934), he describes his work as an “English interpretation” of the Qur’an. He avoids a word-for-word rendering and opts to provide a translation that reflects both the broad and the specific meanings of the Arabic original as much as possible (p. xii). He believes that Qur’anic features, such as its rhythm, music, and exalted tone, should be reflected in the translation, albeit faintly. He concedes that “[i]t may be but a faint reflection, but such beauty and power as my pen can command shall be brought to its service” (p. xii). In so doing, Yusuf Ali gives a running commentary at the beginning of each sūrah (crudely translated as ‘Qur’anic chapter’), to create the atmosphere and to introduce the subject matter to the reader. For the short sūrahs, Yusuf Ali provides one or two paragraphs of rhythmic commentaries. The author acknowledges that in “choosing an English word for an Arabic word, a translator necessarily exercises his own judgement and
20
(Un)translatability of the Qur’an
may be unconsciously expressing a point of view, but that is inevitable” (p. xii). Nonetheless, he affirms that he airs no views of his own but instead follows the commentators. Moreover, when discrepancies occur among them, he relies on his judgement and chooses what appears to be “the most reasonable opinion from all points of view” (p. xii). In addition to his method of providing running commentaries, Yusuf Ali uses footnotes when the spirit of the original text is used and the literal translation is not provided: I have explained the literal meaning in the notes . . . [which] I have them as short as possible constantly with the object I have in view, vis, to give to the English reader, scholar as well as general reader, a fairly complete but concise view of what I understand to be the meaning of the text. (1991, pp. xii–xiii) In addition to the aforementioned reason, Yusuf Ali uses footnotes to elaborate on the questions of law for which the Qur’an provides mere general principles. Further, he uses them to state asbāb al-nuzūl (the occasions of revelation) for some āyāt (Qur’anic verses) when they facilitate understanding. He believes that “such notes are so important for a full understanding of the text. In many cases the Arabic words and phrases are so pregnant of meaning that a translator would be in despair unless he were allowed to explain all that he understands by them” (p. xx). Most importantly, Yusuf Ali lists fve factors as the main causes of diffculties in interpreting the Qur’an: (1) The phenomenon of semantic shift or transformation words occurs in Arabic as in all living languages. The meanings of the Qur’anic words are not the same meanings understood by the Prophet Muhammad and his contemporaries. To overcome this difficulty, Yusuf Ali suggests accepting the conclusions of the early commentators and philologists who “went into these matters with a very comprehensive grasp” (p. xvi). Should divergences arise among them, “we must use our judgment and adopt the interpretation of the authority which appeals to us most. We must not devise new verbal meanings” (p. xvi). (2) Developments in the Arabic language have led commentators to abandon the earlier interpretations, without providing satisfactory reasons. In exercising his own judgement when selecting interpretations, Yusuf Ali follows the rule of preferring the earlier to the later interpretation. However, “where a later writer has reviewed the earlier interpretations and given good reasons for his own view, he has an advantage which we must freely concede to him” (p. xvi). (3) The polysemy of some Arabic words makes the meaning of each root word “so comprehensive that it is difficult to interpret it in a modern analytical language word for word, or by the use of the same word in all places where the original word occurs in the text” (p. xvi). Yusuf Ali refers to the failure of European translators in this regard. Nonetheless, he argues that no human
(Un)translatability of the Qur’an 21 language can possibly be adequate for the expression of the highest spiritual thought. Such thought must be expressed symbolically in terse and comprehensive words, out of which people will perceive just as much light and colour as their spiritual eyes are capable of perceiving. (p. xvi) (4) Contrary to the previous point, special words in the Qur’an do differentiate between objects and ideas. Such words have no readily accurate equivalent in English, and as a result, they are rendered by using general words. For example, for the words raḥmān and raḥīm (crudely, ‘most merciful’) and the words ʿafu, ṣafaḥa, and ghafara (crudely, ‘to forgive’), the English words used as equivalents provide a limited idea of the attributions of Allah (p. xvi). (5) In comparison to Allah’s eternal purpose and the perfectness of his plan, human intelligence is limited and subject to decline according to the person’s power and experience. Therefore, different interpretations by people of different ages are enormous. Hence, Yusuf Ali believes in “progressive interpretation, in the need for understanding and explaining spiritual matters from different angles” (pp. xvi–xvii). Finally, Yusuf Ali (1991) summarizes the principles on which he based his work: In matters of philology and language I accept the best authority among those who were competent to deal with these questions: the older the better. In matters of narration, contemporary authorities are best, subject to such corrections as have to be applied for their points of view. As to the particular occasions on which particular verses were revealed, the information is interesting and valuable from a historical point of view, and our older writers have collected ample materials for it. But to lay too much stress on it today puts the picture out of all perspective. . . . In matters of remote history or folklore, we must take the results of the latest researchers. In interpreting Jewish or Christian legends or beliefs we go to Jewish or Christian sources, but by way of illustration only, not in the direction of incorporating such beliefs or systems. (p. xvii) In Mohammad Asad’s (1980) translation, The Message of the Qur’an, he makes clear that the Qur’an is “unique and untranslatable” (p. v). The main reason for its (un)translatability, according to him, is the exceptional, organic interconnection between its meaning and its linguistic presentation, as they “form one unbreakable whole” (p. v). This interconnection, Asad elaborates, is manifested in what follows: The position of individual words in a sentence; the rhythm and sound of its phrases and their syntactic construction, the manner in which a metaphor flows almost imperceptibly into a pragmatic statement, the use of acoustic
22
(Un)translatability of the Qur’an stress not merely in the service of rhetoric but as a means of alluding to unspoken but clearly implied ideas: all these makes the Qur’an, in last resort, unique and untranslatable—a fact that has been pointed out by many earlier translators and by all Arab scholars. (p. v)
Therefore, Asad declares, “I do not claim to have ‘translated’ the Qur’an in the sense in which, say, Plato or Shakespeare can be translated” (p. v). He further claims that none of the translators (Muslim or non-Muslim) “has so far brought the Qur’an nearer to the hearts or minds of people raised in a different religious and psychological climate and revealed something, however little, of its real depth and wisdom” (p. ii). Asad, however, believes that the impossibility of reproducing the Qur’an in another language does not mean that it is impossible to render its message to people who do not know Arabic (p. v). The possibility of translating the Qur’an, according to him, is subject to the following points, which translators must take into consideration: (1) [They] must be guided throughout by the linguistic usage prevalent at the time of the revelation of the Qur’an, and must always bear in mind that some of its expressions—especially [those] relate to abstract concepts—have in the course of time undergone a subtle change in the popular mind and should not, therefore, be translated in accordance with the sense given to them by post-classical usage (p. v); (2) [They] must take fully into account . . . the iʿjāz of the Qur’an: that inimitable ellipticism which often deliberately omits intermediate thought-clauses in order to express the final stage of an idea as pithily and concisely as is possible within the limitations of a human language (p. vi); (3) [They] must beware of rendering, in each and every case, the religious terms used in the Qur’an in the sense which they have acquired after Islam had become ‘institutionalized’ into a definite set of laws, tenets and practices (p. vi). Thus, Asad makes no claim that he has “reproduced anything of the indescribable rhythm and rhetoric of the Qur’an. No one who has truly experienced its majestic beauty could ever be presumptuous enough to make such a claim or even to embark upon such an attempt” (p. vii). He further admits that he is fully aware that his translation “does not and could not really ‘do justice’ to the Qur’an and the layers upon layers of its meaning” (p. vii). In Arthur John Arberry’s (1955) translation, The Koran Interpreted, he likewise believes that the Qur’an is untranslatable. By labelling his work as The Koran Interpreted, he accepts the aforementioned orthodox Muslim view and admits that the Qur’an cannot be translated. He clarifies this in the preface to his translation, where he indicates that “[i]n choosing to call the present work The Koran Interpreted I have conceded the relevancy of the orthodox Muslim view, of which Pickthall, for one, was so conscious, that the Koran is untranslatable” (p. 24, vol. 1). Arberry, however, attributes the (un)translatability of the Qur’an to the
(Un)translatability of the Qur’an 23 rhythmic and rhetorical features of Qur’anic Arabic in particular. He asserts that these features are “so characteristic, so powerful, so highly emotive, that any version whatsoever is bound in the nature of things to be but a poor copy of the glittering splendour of the original” (pp. 24–25, vol. 1). For Arberry, imitating the rhetorical and rhythmic patterns of the Qur’an was the main reason for conducting the translation. He writes, My chief reason for offering this new version of a book which has been “translated” many times already is that in no previous rendering has a serious attempt been made to imitate, however imperfectly, those rhetorical and rhythmical patterns which are the glory and the sublimity of the Koran. I am breaking new ground here. (p. 25 vol. 1) Therefore, Arberry disregards the accepted claim that the sūrahs of the Qur’an are “in many instances of a composite character, holding embedded in them fragments received by Muhammad at widely differing dates”. Consequently, he deals with each sūrah “as an artistic whole, its often incongruous arts constituting a rich and admirable pattern” (p. 25, vol. 1). Acknowledging the exalted eloquence of Qur’anic Arabic, Arberry (1953) writes in his earlier book, The Holy Koran: An Introduction with Selections, “if Arabic could and can never again be spoken as it was in the Qur’an, certainly the Arabic of the Koran defes adequate translation” (p. 28). Ahmed Ali (1984), in his translation The Qur’an: A Contemporary Translation, elaborates on the features of Qur’anic Arabic and stylistic elements, including “sublimity and excellencies of sound and eloquence, rhetoric and metaphor, assonance and alliteration, onomatopoeia and rhyme, ellipse and parallelism. Its cadences and sprung rhythm, pauses and stops, imply eloquent speech and duration” (p. 7). Ahmed Ali admits that some of these “stylistic beauties are untranslatable and can only be suggested” (p. 7). Therefore, he adopts the “form of metrical lines” in his translation to attend to the accent, “sprung rhythm and tonal structure [of] the sonority and rhythmic patterns of the Qur’anic language” (p. 7). In addition, Ahmed Ali’s attempt to demonstrate the celebratory effect of Qur’anic rhyme, which he claims “cannot be used in English without disastrous consequences” (p. 7), constantly employs “assonance, alliteration and internal rhyme” (p. 7). The author is fully aware of the complexity of the derivational mechanism of the Arabic language, where words are “derived from the same root branch off into different sets of meanings” (p. 7). Thus, he indicates that capturing the shade of meaning “can be fixed only with reference to the context and regard to instances of their similar use elsewhere in the book, as well as the logic and wider world view of the Qur’an” (p. 7). Furthermore, Ahmed Ali justifies his use of brackets in his translation by arguing that they “have been used mainly to give elucidations, differentiated or implied and extended meanings of words, or to fill elliptical gaps” (p. 8). He also indicates that he appends notes to his translation to elaborate on such issues as the truths presented by the Qur’an. As
24
(Un)translatability of the Qur’an
he states, they “have come to be recognized with advance of knowledge in our age as conforming to the laws of causation and effect which science itself is trying to understand” (p. 8). In the translation The Qur’an: A New Interpretation, Colin Turner (1997) perceives the issue of the Qur’an’s (un)translatability from the perspective of its miraculous nature—that is, its iʿjāz (inimitability). He indicates that translation is a form of imitation and asserts that the question of whether or not one should attempt a translation of the Quran should perhaps be seen in the context of the rather more complex issue of translatability in general, of whether or not translation—any translation—is possible at all. (p. x) Drawing on theoreticians’ and critics’ opinions that “all translation is ultimately impossible” (p. x), Turner points out that the notion of untranslatability operates on two distinct levels—the aestheticolinguistic and the religio-philosophical—but at the heart of both arguments lies the question of fidelity, of faithfulness to the text—and, by extension in the case of religious scripture, faithfulness to God himself. (p. x) Fidelity and translatability, according to Turner, are out of the question, because the source and target texts “are not, and can never be, equal”. He attributes this to four reasons, as follows (p. x): (1) (2) (3) (4)
[S]emantic differences between the source and target languages; [P]honetic differences between the source and target languages; [D]iscrepancies in literary traditions; [D]iscrepancies in cultural mindsets.
On the basis of the aforementioned reasons why everything is untranslatable, Turner asks, what do we (translators) do? In an attempt to answer his question, he declares, [w]e do what countless literary law-breakers before us have done: we accept this most unholy of principles and then we dive in and translate, accepting the truth that perfection, immutability and absolutes may pertain to mathematics, but only mythically to the arts, to most human experience, and to everything else in the cosmos—but not at all to literature and literary translation. (p. x) Turner affirms the orthodox view and indicates that the translation of the Qur’an “should not detract from the fact that the general consensus among
(Un)translatability of the Qur’an 25 Muslim intellectuals—including those who have attempted translations of the Quran into other languages—is that the Quran is ultimately untranslatable” (p. xiii). Accepting the claim that the Qur’an is untranslatable, he adds that this does not mean that it should never be translated. Rather, it means that when reading a translation of the Qur’an, the reader must bear in mind “what is lost in translation is the Quran itself” (p. xiii). Turner attributes the Qur’an’s (un)translatability to a complete divergence between the language of the Qur’an (Arabic) and any other language. This divergence is manifested in the syntax and structures of Qur’anic Arabic, its unique nuances and metaphorical uses of words, its excellences of sound and eloquence, of rhetoric and metaphor, of assonance and alliteration, of onomatopoeia and rhyme, of ellipsis and parallelism so sublime that all attempts to replicate its verses in tongues other than Arabic cannot but take on the form of well-intentioned parody. (p. xiii) Turner further declares that [w]hen one considers the complexities involved in translating a work such as the Quran, one often wonders whether it might not be easier for the whole English-speaking world to learn Arabic to read the Quran than for one translator to bring the Quran to the whole of the English-speaking world. (p. xiii) What is more, Turner titles his translation The Qur’an: A New Interpretation and admits that it is “not a straightforward translation”. Rather, it is an “exegetically led” reading that is based on Muhammad Baqir Behbudi’s work Ma‘ānī al-Qur’ān (The Meanings of the Qur’an). In other words, it is “a combination of translation and exegesis—tafsīr. As such, the verses of the Holy Book have been ‘opened out’ to reveal some of the layers of meaning expounded by the Prophet and transmitted through the ages by the Prophet’s family and companions” (p. xvi). This combination of translation and exegesis makes the translation different from other translations of the Qur’an into English. In the translation The Qur’an: English Translation and Parallel Arabic Text, Mohammad Abdel Haleem (2010) stresses the orthodox notion that only the Arabic text of the Qur’an is recognized as ‘the Qur’an’ and that no translation can be a substitute for it. He further concedes that any translation of the Qur’an “is no more than an interpretation or form of exegesis to attempt to explain, in the target language, what the Arabic says. . . . Like any human endeavour, all translations are open to improvement” (p. vx). Acknowledging the difficulties around translating the Qur’an, Abdel Haleem elaborates on the method that he employed in his translation “to enhance accuracy and clarity of meaning” (p. xxviii). Elements of his method are practical mechanisms that he put forward to overcome some of the aforementioned linguistic and cultural aspects of the
26
(Un)translatability of the Qur’an
(un)translatability of the Qur’an. Abdel Haleem addresses eight points in his method: intertextuality, context, identifying aspects of meaning, Arabic structure and idiom, pronouns, classical usage, paragraphing and punctuation, and footnotes and explanatory introductions (pp. xxviii–xxxiii). By ‘intertextuality’ (tafsīr al-Qur’an bi-l-Qur’an), Abdel Haleem refers to the use of one part of the Qur’an to understand other parts. He uses this technique in the footnotes of his translation in his attempt to explain the meaning of “ambiguous passages of the Qur’an” (p. xxviii). Context, according to Abdel Haleem, is “crucial in interpreting the meaning of any discourse, Qur’anic or otherwise” (p. xxviii). Not taking the context into consideration may result in an erroneous translation. To illustrate this issue, he presented his translation of the following Qur’anic passage: fa- anzala Allāhu sakinatahu ‘layhi wa- ayyadahu bi- junudin lam tarawhā wa- ja‘ala kalimata al-la-dhīna kafarū al-suflá wa- kalimatu Allāhi hiya al-‘ulyá (Qur’an, 9:40; God sent his calm down to him, aided him with forces invisible to you, and brought down the disbelievers’ plan). Abdel Haleem compared it with a translation of the same passage, this time translated by Dawood: “God caused his tranquillity to descend upon him and sent to his aid invisible warriors so that he routed the unbelievers and exalted the word of God”. Dawood neglected the context of the verse and mistakenly took the subject of the verb routed/brought down to be the Prophet Muhammad rather than God (p. xxix). Another mechanism used by Abdel Haleem is identifying aspects of meaning. By this, he means to tackle the Qur’anic terms that are frequently used but that have different meanings for different contexts. He argues against the idea of employing one word in translating a given key term for the sake of consistency. He maintains that “[i]t is important for the translator to recognize when it is appropriate to be consistent in the translation of a repeated term and when to reflect the context” (p. xxix). Concerning Arabic structure and idiom, Abdel Haleem makes the point that unnecessarily close adherence to the original Arabic structure and idioms should be avoided as literal translations. He asserts that this is because the Qur’an-specific style typically sounds odd and meaningless in English and “would produce a text incomprehensible to most readers” (p. xxx). Another mechanism employed by Abdel Haleem involves pronouns. Pronouns in the Qur’an sometimes shift in the same verse in a Qur’an-specific phenomenon referred to in Arabic as iltifāt. Not identifying the proper reference of a given pronoun may cause ambiguity and distortion of meaning. The Qur’an commonly features a shift from one personal pronoun or one verb tense to another. Because Arabic differentiates between you singular and you plural and because modern English allows you to signify both singular and plural, Abdel Haleem opts for inserting the word Prophet in his translation “where it is clear that it is he who is being addressed, to make the passages as clear in English as they are in Arabic” (p. xxxi). Abdel Haleem also refers to the classical usage of some Qur’anic terms. He confirms the importance of identifying the original meanings of these terms when
(Un)translatability of the Qur’an 27 used at the time of the Prophet, while also avoiding the use of their new meanings as applied in modern Arabic. This mechanism aligns with Asad’s perspective, presented earlier. To resolve this problem, Abdel Haleem draws on classical Arabic dictionaries such as the Lisān al-ʿArab, by Ibn Manẓūr; al-Qāmūs al-Muḥīṭ, by al-Fayrūzabādī; and al-Muʿjam al-Wasīṭ, by the Arabic Language Academy in Cairo (p. xxxi). Moreover, Abdel Haleem finds that the discrepancies between Arabic and English regarding how to apply paragraphing and punctuation are problematic. As a result, they need to be addressed to “clarify the meaning and structure of thoughts and to meet the expectation of modern readers” (p. xxxii). Therefore, he divides his target text into paragraphs and marks the beginning of each verse with its number. In so doing, Abdel Haleem diverges from Arabic conventions and traditional Qur’anic approaches, where verse numbers are added at the end and each sūrah is presented in one continuous paragraph irrespective of its length. He believes that his method is “extremely important for the referencing and cross-referencing which contributes so much to understanding the meaning of the text” (p. xxxii). Abdel Haleem also diverges from the Arabic conventions in terms of punctuation. While the Qur’an has a specific system of marking pauses, he employs commas, full stops, colons, semicolons, question marks, dashes, quotation marks, etc. that are not employed in the Qur’an. Abdel Haleem justifies this method by claiming that it helps with clarity and with resolving stylistic difficulties (p. xxxiii). Lastly, for the sake of not overburdening the target reader with wide-ranging commentaries, Abdel Haleem (2010) keeps the use of footnotes to a minimum. He (2010) gives the following reasons to justify his use of footnotes: to explain allusions, references, and cultural background only when it was felt these were absolutely necessary to clarify meaning and context. Sometimes the footnotes explain reasons for departing from accepted translations, give alternatives, or make cross-references. The footnotes also give explanations (where they are considered to be helpful or of interest) of ambiguous passages which are made clear in the Arabic commentaries on the Qur’an, classical and modern. (p. xxxiii) Although other translators of the Qur’an devote most of their respective introductions to their respective translations to explanations of the overall features of Islam, the Prophet Muhammad, and the Qur’an itself, they provide some profound insights into the question of the Qur’an’s (un)translatability. For instance, in the translation The Koran (1956), Nessim Joseph Dawood (2003) indicates that the text is “not only one of the most infuential books of prophetic literature but also a literary masterpiece in its own right” (p. 3). Such a masterpiece, according to him, contains many ambiguous statements that, “if not recognized as altogether obscure, lend themselves to more than one interpretation” (p. 4). Criticizing other translators who opt for “the rigidly literal rendering of Arabic
28
(Un)translatability of the Qur’an
idioms”, Dawood believes that the ambiguities caused by such idioms must be tackled by using explanatory footnotes. He writes, I have taken pains to reproduce these ambiguities wherever they occur, and have provided explanatory footnotes in order to avoid turning the text into interpretation rather than a translation. Throughout this rendering the standard commentaries of Al-Zamakhshari, Al-Baidawi, and Al-Jalalayn have been closely followed. (p. 4) Along these lines, T. B. Irving’s (1992) translation, The Noble Qur’an: The First American Translation and Commentary (1985), posits that the Qur’an is “literally untranslatable” because it is a living document that “each time one returns to it, [one] fnds new meanings and fresh ways of interpreting it; the messages are endless, for it is a living book” (p. xxxiii). Thus, he declares that “[t]his in fact is not a translation but a version, a modest tafsir [exegesis] for the English-speaking Muslim who has not been able to rely on Arabic” (p. xxxviiii). Similarly, in the translation The Bounteous Koran: A Translation of Meaning and Commentary, Mohammad M. Khatib (1984) indicates that any translation of the Qur’an is a mere translation of a particular meaning, which is far from revealing or reflecting its true spirit or beauty. For him, the original text includes two types of words: those that are apparent and those that are “figurative and are meant to be left to the imagination of [humankind] throughout the ages” (p. v). Khatib lists the following as the main difficulties that he encountered when he rendered the Qur’an into English (p. v): (1) The omissions, additions, and figurative words that are part of the beauty, eloquence, sequence, and rhythmic pattern of the book; (2) The commitment to an extreme precision in translating letter by letter and word by word while maintaining the exact sequence and construction of the Arabic verse; (3) Finding English words that precisely match the Arabic meaning. In his attempt to counteract these diffculties, Khatib uses footnotes to provide commentaries “intended to make it easier for the reader to understand the true meaning of the text, and as a reference to the difference between the Islamic laws and the pre-Islamic norms and patterns of social behaviour” (p. v). Such commentaries, he elaborates, were drawn from classic and modern books and studies on Islam such as Ibn Kathīr, al-Qurṭubī, al-Ṭabarī, al-Zamakhsharī, alFakhr al-Rāzī, al-Jalālayn, al-Nasafī, Ibn al-Jawzī, Ibn Saʿd, al-Wahīdī, and the al-Muṣḥaf al-Mufassar by Mohammad Farīd Wajdī (p. v). In the same way, Ahmad Zidan and Dina Zidan (2000) in their Translation of the Glorious Qur’an acknowledge that expressing “GOD’s Message” as he has done is a task beyond human faculty. Consequently, a translation of the Qur’an “can never achieve the perfection and degree of expression of the Arabic text in
(Un)translatability of the Qur’an 29 which it was revealed” (p. 6). Therefore, they notify the reader that their work is “only a translation of the meaning of the Qur’an, not a version, as there are no versions of the Qur’an” (p. 6). Likewise, Laleh Bakhtiar (2012) in her translation, The Sublime Quran, admits that neither her translation nor any other translation can be compared with the original Arabic version of the Qur’an in terms of beauty and style. She says, “[c]learly no translation of the Quran can compare in beauty and style with the original Arabic. . . . Recognizing this, a translation of a sacred text which will never equal the original still has certain criteria it should meet” (p. xiii). Furthermore, Bakhtiar confirms that only the Arabic version of the Qur’an is deemed to be the “eternal word of God”. In turn, any translation of that version is no more than an interpretation of the original and not the original itself (p. xxv). Conversely, although other translators of the Qur’an devote extended introductory materials to explaining their respective translations, they stop short of addressing the issue of the Qur’an’s (un)translatability. Muhammad Zafrulla Khan (1971) is a good example. He commences his translation with a lengthy introduction in which he covers a variety of issues pertinent to the Qur’an. He does not, however, elaborate on its (un)translatability or describe the difficulties that he encountered when undertaking the task of translation. The purpose of Zafrulla Khan’s introduction is, clearly, as the editors of the translation put it, to serve as a key to the study of the Quran and should itself be carefully studied. . . . it does not spell out anything in detail. It operates as a pointer for the reader’s mind and beckons it on to exercise, necessary and essential, for a profitable study of the Quran. By his study of the Introduction the reader will acquire some familiarity with the style, diction, the method of persuasion and the manner of reasoning of the Quran. (p. 8) In attending to the issues covered in his introduction, Zafrulla Khan (1971) offers his own understanding, which mirrors the dogma of the Ahmadi doctrine. Therefore, his opinions occasionally overlap with the beliefs of mainstream traditional and contemporary Muslim scholars, whereas at other times they diverge signifcantly. This orientation is manifested throughout his translation of the Qur’an, especially when it comes to rendering the Qur’anic verses containing miracles or Jinns matters. Similarly, Muhammad Habib Shakir (2011), in The Qur’an Translation, does not provide details on aspects of the Qur’an that he fnds (un)translatable, nor does he elaborate on the methods that he employs to translate the Qur’an. Translators of the Qur’an assert the orthodox view that the Qur’an is untranslatable because of its linguistic, cultural, and (for Muslims) theological idiosyncrasies. These views are also emphasized by Muslim intellectuals. However, contrary to the scholars who limit themselves to theorizing about the Qur’an-specific linguistic, cultural, and theological aspects of the Qur’an’s (un)translatability, Qur’an translators have themselves been involved in tackling the difficulties caused by such aspects. Indeed, they talk of the practical mechanisms that they
30
(Un)translatability of the Qur’an
used in their endeavour to provide a linguistically and culturally appropriate rendering of the Qur’an to the closest extent possible. Describing their works as ‘English interpretations’ of the Qur’an, the translators elaborate on the following thorny issues: (1) Translations of the Qur’an are not to be considered substitutions of the glorious Qur’an. They are no more than modest attempts to present the meanings and messages of the Qur’anic verses to English-speaking readers within the translator’s human faculty. (2) Some words with old meanings cannot be found in dictionaries. (3) The semantic shift or transformation that words undergo renders the contemporary meaning of Arabic words in the Qur’an dissimilar to the meaning understood by the Prophet Muhammad and his contemporaries. (4) Some Arabic words feature polysemy. (5) Some words have no readily accurate equivalent in English. (6) Stylistic beauties resist accurate translation. (7) The Qur’an’s meaning and its linguistic presentation are organically interconnected. (8) The Qur’an is a living document in which new meanings and fresh ways of interpreting it can be found. Translators of the Qur’an consider the following as practical mechanisms to deal with the aforementioned diffculties: (1) Resorting to traditional exegetical works and biographies of the Prophet Muhammad al-sīrah al-nabawīyah; (2) Consulting contemporary scholars and living authorities for clarification; (3) Adhering to literal translations that accommodate the opinions of most traditional commentators; (4) Using footnotes to provide literal translations when the spirit of the original text is targeted in the body of the translation; (5) Using footnotes or appended notes to elaborate on such issues as asbāb al-nuzūl (the occasions of revelation) of some verses and questions of law; (6) Accepting the conclusions of early commentators and philologists—though should divergences arise among them, using their judgement and historical sense to adopt the interpretation of that authority which appeals to them the most; (7) Not devising new verbal meanings; (8) Providing a general word when rendering Qur’anic terms that have no readily accurate equivalent in English; (9) Dealing with each Qur’anic sūrah as an artistic whole; (10) Using poetic language and metrical lines to imitate the Qur’anic style and to give the English reader a flavour of its stylistic beauties; (11) Considering the Qur’anic context and the use of a given word in different instances to capture implied meaning;
(Un)translatability of the Qur’an 31 (12) Using in-text parenthetical glosses to provide implied and extended meanings of words or to fill elliptical gaps; (13) Taking the phenomenon of īʿjāz (miraculous inimitability) of the Qur’an into consideration; (14) Avoiding word-by-word translation; (15) For some Qur’an translators (e.g. Yusuf Ali, Arberry, and Ahmed Ali), believing that Qur’anic features such as its rhythm, music, and exalted tone should also be reflected in the translation.
2.3 Conclusion (Un)translatability, at its core, implies the impossibility of conveying the meaning of words and structures from one language to another. The apparent (un) translatability of the Qur’an has been the focus of this chapter, which deals with aspects of this issue from two perspectives: the (un)translatability of the Qur’an as understood by Muslim intellectuals and the (un)translatability of the Qur’an as viewed by its translators into English. This chapter demonstrates that although both scholars and translators of the Qur’an agree that the Qur’an-bound linguistic and cultural aspects are untranslatable, they vary on what they prioritize. Moreover, translators differ in how they deal with untranslatables. Nonetheless, they both affirm the possibility of rendering the meanings of the Qur’an into other languages and confirm the necessity and nobility of such a task as part of da‘wah (missionary invitation) to Islam. Given this possibility, I argue that the Qur’an’s translatability, rather than (un) translatability, needs to be the main concern of stakeholders. This is because translatability seems to be more plausible and practical than (un)translatability. Indeed, absolute untranslatables are considered the minority in relation to the vast majority of “translatables and relative translatables” (Ke, 1999, p. 297). Having said that and given the nonexistence of ‘absolute equivalence’, the process of translating a text from a given SL into a TL inevitably involves a certain amount of loss. The quality or quantity of this loss varies according to several crucial variables. They include, but are not limited to, linguistic and cultural divergences between the two languages, the purpose of the translation, and the genre of the text at hand. By the same token, scholars have become more aware that “absolute untranslatability, whether linguistic or cultural, does not exist” (De Pedro, 1999, p. 557). When it comes to translating sacred scriptures in general and the Qur’an in particular, because of their spiritual, historical, theological, and linguistic statuses, such loss becomes more significant and questions both the legitimacy of translating these scriptures and their translatability. At the end of the day, translators are responsible for reaching, within their human faculty, a compromised ‘crude approximation’. Further, they should find appropriate ways to compensate for these losses, by resorting to practical procedures such as those mentioned in this chapter and those attended to in the forthcoming practical section. Hence, the prefixed morpheme un has been put between brackets when the key term (un)translatability is mentioned throughout this book.
32
(Un)translatability of the Qur’an
References Abdel Haleem, M. A. S. (2010). The Qur’an: English Translation and Parallel Arabic Text. Oxford University Press. Oxford. Abdul-Raof, H. (2001). Qur’an Translation: Discourse, Texture and Exegesis. Curzon. Richmond, Surrey, UK. Abdul-Raof, H. (2005). Cultural Aspects in Qur’an Translation. In Lynne Long (Ed.), Translation and Religion: Holy Untranslatable. Multilingual Matters LTD. Clevedon, Buffalo, Toronto. Aldahesh, A. Y. (2017). Translating the Qur’an: A Jurisprudential Approach. AlAmeed Journal, 6 (23), 176–218. Ali, A. (1984). The Qur’an: A Contemporary Translation. Princeton University Press. Princeton, NJ. Arberry, J. A. (1953). The Holy Koran: An Introduction with Selections. George Allen & Unwin LTD. London. Arberry, J. A. (1955). The Koran Interpreted. George Allan & Unwin LTD. London; The Macmillan Company. New York. Asad, M. (1980). The Message of the Qur’an. Dār al-Andalus. Gibraltar. Bakhtiar, L. (2012). The Sublime Quran: English Translation. Revised Fifteenth Edition. Kazi Publications. Chicago. Bassnett, S. (2002). Translation Studies. Third Edition. Routledge. London and New York. Catford, J. C. (1965). A Linguistic Theory of Translation. Oxford University Press. London. Dastjerdi, H. V., & Jamshidian, E. (2011). A Sacrament Wordplay: An Investigation of Pun Translatability in the Two English Translations of the Quran. Asian Social Science, 7 (1), 133–144. Retrieved from http://www.ccsenet.org/journal/index. php/ass/article/view/7132/6533 Dawood, N. J. (2003). The Koran. Penguin Books. Victoria, Australia. De Pedro, R. (1999). The Translatability of Texts: A Historical Overview. Meta, XLIV (4), 546–559. Fazlur Rahman, M. (1988). Translating the Qur’an. Religion and Literature, 20 (1), 23–30. Hatim, B., & Munday, J. (2004). Translation: An Advanced Resource Book. Routledge. London and New York. Iqbal, M. (2000). Abdullah Yusuf Ali and Muhammad Asad: Two Approaches to the English Translation of the Qur’an. Journal of Qur’anic Studies, 2 (1), 107–123. Irving, T. B. (1992). The Noble Qur’an: The First American Translation and Commentary. Amana Books. London. Ke, P. (1999). Translatability vs. Untranslatability: A Sociosemiotic Perspective. Bable, 45 (4), 289–300. Khatib, M. M. (1984). The Bounteous Koran: A Translation of Meaning and Commentary. Macmillan Press. London. Kidwai, A. R. (1987). Translating the Untranslatable: A Survey of English Translations of the Quran. Retrieved from http://www.soundvision.com/Info/quran/ english.asp Kidwai, A. R. (2007). Bibliography of the Translations of the Meanings of the Glorious Qur’an into English 1649–2002: A Critical Study. King Fahad Quran Printing Complex. Madinah, Saudi Arabia.
(Un)translatability of the Qur’an 33 Newmark, P. (1988). A Textbook of Translation. Prentice Hall. New York, London, Toronto, Sydney, and Tokyo. Pickthall, M. (1971). The Meaning of the Glorious Qur’an: Text and Explanatory Translation. Dār al-Kitāb Allubnānī. Beirut. Lebanon. Pym, A., & Turk, H. (2001). Translatability. In Mona Baker (Ed.), Routledge Encyclopaedia of Translation Studies. Routledge. London and New York. Saeed, A. (2008). The Qur’an: An Introduction. Routledge. London. Shakir, M. H. (2011). The Qur’an Translation. Tahrike Tarsile Qur’an, Inc. New York. Tibawi, A. L. (1962). Is the Qur’an Translatable? Early Muslim Opinion. The Muslim World, 52 (1), 4–16. Turner, C. (1997). The Qur’an: A New Interpretation. Curzon. Richmond. Von Denffer, A. (1983). ‘Ulum al-Qur’an: An Introduction to the Sciences of the Qur’an. The Islamic Foundation. Leicester, UK. Yusuf Ali, A. (1991). The Meaning of the Holy Qur’an. Amana Corporation. Brentwood, MD. Zafrulla Khan, M. (1971). The Quran: The Eternal Revelation Vouchsafed to Muhammad the Seal of the Prophets. Curzon Press. London and Dublin. Zidan, A., & Zidan, D. (2000). Translation of the Glorious Qur’an. A. S. Noordeen. Kuala Lumpure.
3
Arabic idiomatic phrasal verbs Nailing down the phenomenon
3.1 Introduction The key term in this chapter, phrasal verb (henceforth PV), is originally an English term. It was first coined by English grammarians and linguists who, since the eighteenth century CE, have scrutinized an unusual phenomenon in their language. That is, proper verbs and adverbial particles syntactically colligate and semantically collocate to fuse and produce a significance that is different from the total of their literal significances. After ample scrutiny, they called the phenomenon idiomatic English phrasal verbs (henceforth EPVs). EPVs have been given attention by scholars in such domains as grammar, linguistics, pedagogy, and lexicography (cf Bolinger, 1971; McArthur, 1975, 1979, 1989; Fraser, 1976). In their endeavour to provide a definition for this phenomenon, scholars have proposed a variety of broad definitions. The essence of their definitions suggests that they perceive an EPV as a combination of two or three elements (a verb + a preposition, a verb + an adverb, or a verb + an adverb + a preposition). This combination functions as a single unit of meaning in the sense that its significance cannot be deduced from the sum of the significances of its individual elements. Examples of EPVs include to carry out, to carry on, to turn up, to turn on, to turn off, to sort out, to give up, to give in, to give away, to get away with, to black out, to tip off, to take off, to take in, to put up with, etc. (cf Aldahesh, 2009a, 2016a). The Arabic language, by contrast, does not allow proper verbs to combine with adverbs. It does, however, allow them to combine with prepositions. Consequently, unlike English, which has three EPV combinations, Arabic has only one combination—that is, a verb + a preposition. Arabic verb-preposition combinations are exemplified in what follows: (1) (2) (3) (4) (5) (6) (7)
sahira ʿalá ‘to take care of’; taṭallaʿa ilá ‘to look forward to’; atá ʿalá ‘to destroy, to finish’; atá bi- ‘to bring’; māla ʿan ‘to avoid, to dislike’; māla ilá ‘to like, to sympathize’; waqaʿa ʿalá ‘to come across, to find’.
Arabic idiomatic phrasal verbs 35 These constructions are widely used in both written Arabic and spoken Arabic. Yet unlike English linguists, Arabic linguists—especially the classical ones—do not group such constructions under a specifc heading. Many modern Arabic linguists label these Arabic combinations as al-ḍamāʾim (enclosures) (cf al-Shamsān, 1986; Aldahesh, 2016a). Although some researchers (e.g. Ryding, 2005; al-Shamsān, 1986, 1987; Kharma & Hajjaj, 1989; Abboud & McCarus, 1968; Bataineh & Bataineh, 2002) have touched on the phenomenon of Arabic idiomatic phrasal verbs (henceforth AIPVs) in passing, without elaborating on their form and function criteria, a few others have tried to explore AIPVs in detail. The leading study is Lentzner’s doctoral dissertation Semantic and Syntactic Aspects of Arabic Prepositions, in which she assigns a chapter to exploring verb-preposition structures in Arabic (Lentzner, 1977, pp. 155–195). Another important study is Heliel’s VerbParticle Combinations in English and Arabic: Problems for Arab Lexicographers and Translators (Heliel, 1994). Furthermore, in the past 20 years or so, a number of Arab scholars have conducted studies in Arabic on such pertinent issues as the transitivity of Arabic verbs and the semantic behaviour of Arabic prepositions. Such scholars include Muhammad Dāwood (2002) in his book al-Qurʿan alKar īm wa-Tafāʿulu al-Maʿān ī: Dirāsatun Dilāliyatun li-Taʿlluqi Ḥarfi al-Jarri bil-Fiʿili wa-ʾAtharihi f ī al-Maʿná f ī al-Qurʿani al-Kar īm and Ahmad al-ʿaṭiyah (2008) in his article ‘Ḥurūfu al-jarri bayna al-nyābati wa-l-taḍmīn’. Except for the present researcher (cf Aldahesh, 2016a, 2016b, 2017, 2018), who draws a clear distinction between the idiomatic and the nonidiomatic Arabic verb-preposition structures, no such distinction has been made by scholars to previously investigate the phenomenon. They instead took for granted that every combination of an Arabic verb followed by a preposition was an ‘idiom’ and dealt with it as such. Lentzner (1977), for instance, labels every verb-preposition combination an Arabic verb-preposition idiom, which is not accurate, by any means. She confuses the idiomatic/metaphorical verb-preposition combination with nonidiomatic/literal ones. This confusion, in my view, is partly due to the approach that she adopted in her study. Lentzner approaches the point ably from the standpoint of prepositions rather than verbs, concentrating on the function of each preposition when it is combined with certain verbs. Despite tackling, albeit casually, the syntactic characteristics of the verbs, she ignores their semantic features and major input into the overall significance of the structure when they combine with prepositions. To the best of my knowledge and research, I know of no such specific analysis. Hence the importance of this study, which aims to bridge the gap by exploring the phenomenon of AIPVs per se, establishing a parameter for them and outlining the key factors that control the syntactic and semantic relationships between their main constituents.
3.2 Establishing a parameter for AIPVs Before establishing a parameter for AIPVs, this chapter explores the key notions of idioms and idiomaticity, which constitute the crux of this study. The aim
36
Arabic idiomatic phrasal verbs
here is to differentiate between the idiomatic and the nonidiomatic Arabic verbpreposition structures and, in turn, to take the characteristics of these two notions as platforms from which to establish a parameter for AIPVs. Linguists, grammarians, lexicographers, and pedagogues have proposed a range of definitions for idioms and idiomaticity. The main definitions are cited as follows: In his book The Verb-Particle Combination in English, Fraser (1976) defines idiom as “a single constituent or series of constituents, whose semantic interpretation is independent of the formatives which compose it” (p. v). By the same token, Richards et al. (1985), in their Longman Dictionary of Applied Linguistics, define idiom as “an expression which functions as a single unit and whose meaning cannot be worked out from its separate parts” (p. 134). Along these lines, in her book In Other Words, Baker (1992) defines idioms as “frozen patterns of language which allow little or no variation in form and . . . often carry meanings which cannot be deduced from their individual components” (p. 63). Likewise, in Dictionary of English Grammar, Trask (2000) defines idiom as a “fixed expression whose meaning is not guessable from the meaning of its parts” (p. 67). Yallor et al. (2005) in their Macquarie Dictionary define idiom as “a form of expression peculiar to a language, especially one having a significance other than its literal one” (p. 709). Moreover, Daud et al. (2003), define idiom in Linguistics Dictionary: English-Arabic Arabic-English as “[a]n expression which functions as a single unit [where its] meaning cannot be derived from the meaning of the individual elements” (p. 100). Ghazala (2003) asserts that idioms are “special, metaphorical, fixed phrases whose meanings and forms are not negotiable” (p. 204). In Baker’s (1992) attempt to elaborate on the idiosyncratic of this notion, she indicates that when it comes to idiom, language users cannot usually do the following things unless they are “consciously making a joke or attempting a play on words” (p. 63): (1) (2) (3) (4) (5)
Change the order of the words in it; Delete a word from it; Add a word to it; Replace a word with another; Change its grammatical structure.
Observing the form and meaning of idioms, Lattey (1986) maintains in her article ‘Pragmatic Classifcation of Idioms as an Aid for the Language Learning’ that “as far as the form of idioms is concerned, we have groups of words, and in terms of meanings, we can say that we are dealing with new, not readily apparent meanings when we confront idioms” (p. 219). Palmer defines idiomaticity as “a lexical feature, something to be dealt with in the lexicon or dictionary rather than the grammar” (Palmer, 1974, p. 213). Moreover, Ghazala (2003) defines idiomaticity in a more explicit manner. He
Arabic idiomatic phrasal verbs 37 points out that it is idioms’ “most special component [forms] which their metaphorical aspect . . . the heart of the matter of any idiomatic expression [and] the gist of any idiomatic phrase” (pp. 204–409). In the main, idioms are such important constituents that they exist in all languages to the extent that no language can afford to abandon them. Their importance in any given language “cannot be doubted [because their] ubiquity makes them anything but a marginal phenomenon” (Wallace, 1968, p. 112; cf Aldahesh, 2009a, 2013a). What is more, the notion of idiomaticity plays a vital role in the Arabic language. It by far constitutes the heart of Arabic eloquence and rhetorical styles. Classical Arabic linguists and modern Arabic linguists have paid a great deal of attention to this phenomenon. They approached idiomaticity from four main perspectives (cf Aldahesh, 2013a): (1) (2) (3) (4)
al-Majāz (metaphor); al-Tashb īh (simile); al-Istiʿārah (allegory); al-Kināyah (pun).
The main structural and semantic features of Arabic idioms are summarized by Abu Saʿad (1987) as follows (cited in Bataineh & Bataineh, 2002, pp. 47–48; cf Aldahesh, 2013a). (1) Idioms come in the following structural patterns: (a) A sentence that consists of two or more words; Example: to put the cart before the horse [yaḍʿ al-ʿarabata amāma al- ḥiṣān] (b) Genitive constructions whose individual meanings are familiar, while the result of their combination is unfamiliar; Example: Noah’s ark [saf īnat Nūḥ] ‘something that gathers many objects or species’ The patience of Job [ṣabr Ayūb] ‘real patience and tolerance’ (c) Individual idiomatic words; Example: He is an ear [huwa udhun] ‘He tells of what he hears without thinking’ (2) They are influenced by certain linguistic phenomena: (a) Synonymy [al-tarāduf], where different structures express the same meaning; (b) Homonymy [al-tabāyun], where one structure expresses different meanings; (c) Antonymy [al-taḍād], where one structure expresses opposite meanings.
38
Arabic idiomatic phrasal verbs
(3) Idioms are related to proverbs, and thus, the more common the proverb, the greater its chance of being an idiom. (4) Idioms derive their figures from the environment. Arabic, like English, is full of idiomatic expressions. Yet there are more of them in Arabic dialects than in Modern Standard Arabic. . . . Many of the Arabic idioms are easy to understand because their meanings are not that far from the sum total of their respective components. But others, just as in English, are difficult to understand, especially for nonnative speakers of Arabic, simply because their meanings are far from the sum of their components. The syntactic and semantic restrictions of Arabic idioms exhibit the same restrictions as their English counterparts. Three commonly used Arabic proverbs illustrate this point: (1) ḍaraba ʿuṣfūrayni bi-ḥajarin wāḥid. (Literally, ‘he hit two birds in one stone’). (Pragmatically, ‘he achieved two goals in one go’) [my translation]. In this Arabic proverb, we cannot substitute the verb iṣṭāda (he hunted) for the verb ḍaraba (he hit) or the noun raṣāṣah (a bullet) for the noun ḥajar (a stone). (2) saḥaba al-bisāṭa min taḥti aqdāmihi. (Literally, ‘he pulled the rug from under his feet’). (Pragmatically, ‘he stopped supporting him in critical circumstances’) [my translation]. In this Arabic proverb, we cannot substitute the verb jarra (he dragged) for the verb saḥaba (he pulled) or the noun baṭṭāniyah (blanket) for the noun al-bisāṭa (the rug). (3) ʿāda bi-khufay Ḥunayn. (Literally, ‘He returned with the slippers of Hunain’). (Pragmatically, ‘He returned empty-handed’) [my translation]. In this Arabic idiomatic expression, we cannot substitute waṣala (he arrived) or sāra (he walked) for the verb ʿāda (he returned) (cf Awwad, 1990, p. 58; Aldahesh, 2013a). In conclusion, Arabic idiomatic expressions are language and culture specifc. Moreover, the Arabic language, like English, relies heavily on them in its spoken and written forms (cf Aldahesh, 2009a, 2013a). Arabic verb-preposition structures fall into two distinct categories: (1) Idiomatic/metaphorical nontransparent structures; (2) Nonidiomatic/literal transparent structures. The frst category is a construction of Arabic verb-prepositions where the verb changes its fundamental signifcance to form with the preposition a single unit
Arabic idiomatic phrasal verbs 39 of meaning. This meaning is different from the total signifcances of the verb and the preposition. Let me exemplify this point by citing two excerpts from the literary works of the well-known Arabic writer Najīb Maḥfūẓ. To appreciate the word order of the Arabic structures, literal word-by-word glosses are provided frst, followed by pragmatic translations of the excerpts (cf Aldahesh, 2016a, 2016b): (1) wa-lisānuhā lā yumsiku ʿan al-tilāwati (Maḥfūẓ, 1991, p. 328). and-tongue-her not catch about the-reciting: ‘and her tongue does not stop reciting the Qur’an’ [my translation]. In this excerpt, the Arabic verb amsaka literally means ‘to catch something’. It abandons its basic significance and acquires a new metaphorical one when it combines with the Arabic preposition ʿan—that is, ‘to stop doing something’. (2) wa-kāna al-sayyd Ahmad qad faragha min al-ṣalati (Maḥfūẓ, 1991b, p. 591). and-was-he Mr. Ahmad became empty from the-prayer: ‘and Mr. Ahmad has had finished practising the prayer’ [my translation]. In this excerpt, the Arabic verb faragha literally means ‘to become empty’. It abandons its basic significance to obtain a new, metaphorical one when it combines with the Arabic preposition, min—that is, ‘to finish doing something’. The second category is a verb-preposition construction in which the verb and the preposition retain their basic signifcances. I exemplify this category by citing two excerpts, one from the works of Arabic novelist Ahlam Mosteghanemi and one by the eminent Arabic poet Adonis: (1) yufaḍḍilu ʿalá ḥuḍūrahā al-ʿābiri ghiyāban ṭawilan (Mosteghanemi, 2012, p. 11). Prefer-he on presence-her passing an-absence long: ‘He prefers her long absence on her casual presence’ [my translation]. In this excerpt, both the Arabic verb faḍḍala (to prefer) and the Arabic preposition ʿalá (on) keep their basic meanings. (2) wa-ahrubu min al-mawti (Adonis, 1996, p. 45). and-escape-I from the-death: ‘and I escape from the death’ [my translation]. In this excerpt, both the Arabic verb haraba (to escape from someone or something) and the Arabic preposition min (from) maintain their basic significances. The condition by which I distinguish between the frst category and the second category (i.e. idiomaticity) is of varying degrees. In other words, verb-preposition
40
Arabic idiomatic phrasal verbs
constructions can be somewhere between too literal and too idiomatic. To account for the EPV’s idiomaticity, Cowie and Mackin (1993) employ two testing procedures to determine whether a given EPV is idiomatic or nonidiomatic. Both testing procedures, however, relate to semantic meanings rather than to syntactic structures. The frst testing procedure is to ask whether the whole phrase can be substituted by one word. The second testing procedure is to ask whether the second word (preposition or adverb) can be deleted. To know, for instance, whether the EPV fall out—as in, ‘I was pleased with the way things had fallen out’—is an idiomatic construction, you need to apply one of these testing procedures. In fact, the answer of the frst test is, yes, we can substitute fall out for one word—that is, ‘happen’ or ‘occur’. The answer to the second test is, no, we cannot delete the second part of fall out, because the form of this phrase is fxed and cannot be broken up. Hence, it is an idiomatic EPV (cf Cowie & Mackin, 1993; Aldahesh, 2016a). These testing procedures are both applicable and relevant to this study. They can well be used to measure the degree of idiomaticity of Arabic verb-preposition combinations. Let me demonstrate this through examples (cf Aldahesh, 2016a). (1) Examine the idiomaticity of the Arabic verb-preposition combination haraba min. To apply the first testing procedure (i.e. substituting the whole phrase for one word), one may ask, can we substitute an Arabic phrase such as haraba min (to escape from) for one word? The answer is, no, we have no one word for this Arabic phrase. This is simply because it is made up of two distinct lexical constituents, each of which retains its own literal significance. Moreover, the whole significance of the phrase is so transparent that it can be easily deduced from the sum of the significances of its components. To apply the second testing procedure (i.e. to delete the second word of the phrase), again one may ask, can we delete the Arabic preposition without changing the whole meaning of the phrase? The answer is, yes, we can use haraba only to express the same meaning of the phrase ‘to escape’. Therefore, this Arabic combination is a nonidiomatic, literal, and transparent one. (2) Examine the idiomaticity of the Arabic verb-preposition combination insaḥabaʿalá. Let us apply the same testing procedures on the phrase insaḥabaʿalá (literally, ‘to withdraw . . . on’, ‘to retreat . . . on’) in the following example: wa-idhā mā insaḥaba al-ʾamru ʿalá ghayrihim aw-ʿalá madākh īl ukhrá fainna al-ḥāl sayatakarrar ayḍan (al-Wasat News, 2015). and-if withdrew the-matter on other-than-them and-on incomes others then-that the-situation will-repeat-itself too. ‘This situation will happen again if the matter is applied to other people or to other incomes as well’ [my translation].
Arabic idiomatic phrasal verbs 41 Can we substitute the whole phrase with a one-word synonym? Yes, we can. The phrase insaḥaba ʿalá can be substituted for the verb ṭubbiqa (to be applied). Can we delete the second component of the phrase (i.e. the preposition) ʿalá? No, we cannot. If we do so, the verb insaḥaba will retain its literal basic meaning ‘to withdraw’. Thus, this Arabic verb-preposition combination is an AIPV. (3) Examine the idiomaticity of the Arabic verb-preposition combination intahá ilá. Let me apply the producers of the test on this combination as it appears in the following excerpt: wa-intahá al-rajulu ilá mawqifihā (Maḥfūẓ, 1991a, p. 330). and-finish the-man to place-her. ‘and the man reached the place where she stands’ [my translation]. Can we substitute the entire combination with a one-word synonym? Yes, we can. The phrase intahá ilá can be substituted for the verb waṣala (to reach). Can we remove the second component of the phrase (i.e. the preposition) ilá? No, we cannot. If we remove the preposition, the verb intahá will retain its literal basic meaning ‘to finish’. Hence, this Arabic verb-preposition combination is certainly an AIPV. Here the context plays a vital role in determining the idiomaticity of the whole Arabic verb-preposition structure. That is, to understand whether a given Arabic verb-preposition structure is used literally or metaphorically, one needs to consider the context of that structure. Because the literal and idiomatic structure types share the same syntactic properties, the distinction between them is confned to their semantic peculiarities. Also, the nonidiomatic, literal type of structure is a mere verb + preposition combination. Thus, it is excluded as an AIPV and falls outside the scope of this study.
3.3 AIPVs’ components Before tackling the syntactic and semantic interactions of the main components of AIPVs, viz. the Arabic proper verbs and prepositions, let me briefly shed some light on the properties of each component.
3.3.1 Arabic proper verbs Grammatically speaking, al-fiʿil (verb) in Arabic signifies an action undertaken by al-fāʿil (agent or verb’s subject), either optionally—as in akala (to eat) qaʿada (to sit)—or by attributing the action to the agent—as in māta (to die) (al-Shamsān, 1986; Aldahesh, 2017).
42
Arabic idiomatic phrasal verbs
Morphologically speaking, Arabic verbs fall into two main categories: al-fiʿil al-thulāth ī (triliteral verb) and al-fiʿil al-rubāʿī (quadriliteral verb). The former refers to verbs with three-consonant roots (e.g. kataba [to write], ʿalima [to know], and nāma [to sleep]), whereas, the latter refers to verbs with four-consonant roots (e.g. tarjama [to translate], ṭamʾana [to reassure], and daḥraja [to roll]). On the basis of several morphological ʾawzān (patterns)—Western scholars who dealt with Arabic grammar labelled them ‘Arabic verb forms or patterns’— many forms can be derived from both triliteral verbs and quadriliteral verbs. The process of deriving new verbs by using different patterns constitutes a vital characteristic of Arabic derivational mechanism. In turn, this shows the richness of the word formation and productivity of the Arabic language (Aldahesh, 2008). Arabic linguists morphologically classify Arabic verbs into two distinct types (Ghalāyīnī, 1986, pp. 55–64): (1) al-fiʿil al-jāmid (invariable verb)—a frozen verb with a meaning that has nothing to do with time and characteristically restricted to one pattern. Examples include the verbs laysa (it is not), ʿasá (it may be), niʿma (how good), and biʾsa (how bad). (2) al-fiʿil al-mutaṣarrif (inflected verb)—a verb with a meaning that has something to do with time. Hence, it has the capacity to be formed in a variety of patterns according to the time in which its action happens. This type of verb falls into two subcategories: (i) Fully inflected verbs that can be formed into three verb patterns—that is, al-māḍi (past), al-muḍāriʿ (present), and al-ʾamr (imperative) (e.g. kataba [he wrote], yaktubu [he writes], uktub [write]); (ii) Partially inflected verbs that can be formed into only two verbs patterns— that is, either past and present (e.g. kāda [to be about to], yakādu [he is about to]) or present and imperative (e.g. yadaʿu [he lets], daʿ [let]). Like other Semitic languages, such as Hebrew and Aramaic, Arabic has a systematic derivational mechanism with a fxed paradigm of derivative forms by which newly derived vocabulary can be formed. The main ten derivative Arabic verb forms and their inherent meanings are listed as follows. There are several other forms not listed here, because they are characteristically rare: (1) faʿala, basic form (e.g. darasa [to study], Kasara [to break]). (2) faʿʿala, formed by doubling the second radical of the basic form (e.g. darrasa [to teach], kassara [to break in pieces]). This form denotes intensity, causative, and applicative connotations. (3) fāʿala, formed by adding a long vowel ʾalif after the first radical (e.g. jālasa [to sit down with someone], qātala [to fight]). This form implies an associative connotation. (4) afaʿala, formed by adding the sound hamza before the first radical (e.g. anṭaqa [to make someone talk], ajrā [to make, to conduct]). This form signifies causative and transitive connotations.
Arabic idiomatic phrasal verbs 43 (5) tafaʿʿala, formed by adding the consonant tāʾ in the front of the aforementioned second form faʿʿala (e.g. tadhakkara [to remember], takhawwafa [to be afraid]). This form denotes a reflexive connotation. (6) tafāʿala, formed by adding the consonant tāʾ in front of the aforementioned third form fāʿala (e.g. taʿāwana [to cooperate with someone], takhādaʿa [to pretend to be deceived]). This form signifies reflexive and reciprocity connotations. (7) infaʿala, formed by adding the sound hamza and the consonant nūn in front of the basic root radicals (e.g. inkasara [to be broken], inkashafa [to be uncovered]). This form implies passiveness and reflexive connotations. (8) iftaʿala, formed by adding the sound hamza in front of the first radical and the consonant tāʾ after it (e.g. iftaʿala [to fabricate something], iltamasa [to seek something]). This form denotes a reflexive connotation. (9) ifʿalla, formed by adding the sound hamza at the beginning and doubling the third radical (e.g. ikhḍarra [to turn green], iswadda [to turn black]). This form signifies a connotation of colour defects. (10) istafʿala, formed by adding the sound hamza and the consonant s īn at the beginning and the consonant tāʾ in front of the first radical and by placing a sukūn over the first root letter (e.g. istakhraja [to take out], istakhdama [to use]). This form implies causative, reflexive, and appointive connotations (cf Wright, 1981; Wightwick & Gaafar, 1998; Aldahesh, 2009a, 2008). The derivational mechanism is not the only way that the Arabic language sustains its enrichment and productivity. Arabic profoundly and creatively allows the employment of metaphors to generate countless newly formed structures and collocations with different semantic and pragmatic connotations. The following fve examples of Arabic metaphorical expressions suffce for the current purpose (cf Ḥammād, 1983; Aldahesh, 2009a): (1) (2) (3) (4) (5)
udhunu al-finjāni (the ear of the cup); rijlu al-kursi (the leg of the chair); ʿunuqu al-zujājati (the neck of the bottle); asnānu al-mishiṭi (the teeth of the comb); ʿaynu al-ibrati (the eye of the needle).
What is more, Arabic verbs are marked for person, gender, and number. In terms of person, there are three classes of Arabic verbs: (1) First person verbs (e.g. adrusu [I study], aʿlamu [I know], ʾaktubu [I write]); (2) Second person verbs (e.g. tadrusu [you study], taʿlamu [you know], taktubu [you write]); (3) Third person verbs (e.g. yadrusu [he studies], yaʿlamu [he knows], yaktubu [he writes]).
44
Arabic idiomatic phrasal verbs
In terms of gender, there exist two classes of Arabic verbs: (1) Masculine verbs (e.g. yadrusu [he studies], yaʿlamu [he knows], yaktubu [he writes]); (2) Feminine verbs (e.g. tadrusu [she studies], taʿlamu [she knows], taktubu [she writes]). In terms of number, there are three classes of Arabic verbs: (1) Singular verbs (e.g. yadrusu [he studies], yaʿlamu [he knows], yaktubu [he writes]); (2) Dual verbs (e.g. yadrusāni [they both study], yaʿlamāni [they both know], yaktubāni [they both write]); (3) Plural verbs (e.g. yadrusūna [they study], yaʿlamūna [they know], yaktubūna [they write]). Moreover, Arabic verbs can be conjugated into three classes in terms of tense: (1) Past tense verbs (e.g. darasa [he studied], ʿalima [he knew], kataba [he wrote]); (2) Present tense verbs (e.g. yadrusu [he studies], yaʿlamu [he knows], yaktubu [he writes]); (3) Future tense verbs, which is formed by prefixing the morpheme sa or the morpheme sawfa (e.g. sa-yadrusu [he will study], sawfa yaʿlamu [he will know], sawfa yaktubu [he will write]). Furthermore, Arabic verbs come in four major moods: (1) Indicative verbs (e.g. yadrusu [he studies], yaʿlamu [he knows], yaktubu [he writes]); (2) Subjunctive verbs (e.g. ʾan yadrusa [to study], ʾan yaʿlama [to know], ʾan yaktuba [to write]); (3) Jussive verbs (e.g. lam yadrus [he did not study], lam yaʿlam [he did not know], lam yaktub [he did not write]); (4) Imperative verbs (e.g. udrus [study], iʿlam [know], uktub [write]). In addition, Arabic verbs have two types of voice: (1) Active (e.g. yadrusu [he studies], yaʿlamu [he knows], yaktubu [he writes]); and (2) Passive (e.g. yudrasu [to be studied], yuʿlamu [to be known], yuktabu [to be written]). What is relevant to our main topic is that Arabic verbs, in all their classes, may well constitute the frst component of an AIPV. Two examples from the Qur’an illustrate this point:
Arabic idiomatic phrasal verbs 45 (1) ulā’ika alladh īna ṭabaʿa Allahu ʿalá qulubihim wa-samʿihim wa-abṣārihim (Qur’an, 16:108). ‘These are people whose hearts, hearing, and sight have been closed off by God’ (Abdel Haleem, 2010, p. 280). The verb ṭabaʿa in this verse is past tense, active voice form and constitutes the first element of the Qur’anic verb-preposition combination ṭabaʿa ʿalá. (2) dhālika bi-annahum āmanū thumma kafarū fa-ṭubiʿa ʿalá qulubihim fa-hum lā-yafqahūn (Qur’an, 63:3) ‘because they professed faith and then rejected it, so their hearts have been sealed and they do not understand’ (Abdel Haleem, 2010, p. 555). The verb ṭubiʿa in this verse is past tense, passive voice form, and it constitutes the first constituent of the Qur’anic verb-preposition combination ṭubiʿa ʿalá (cf Aldahesh, 2017). More examples of Qur’anic verb-preposition combinations are provided in Chapter 3, when their peculiarities are attended to in detail. Unlike EPVs, whose communicative function is mainly carried by particles (Side, 1990), the communicative function of AIPVs is typically carried by the verb. This is despite the preposition’s modifying, to a certain extent, the meaning of the verb that it combines with (see the two Qur’anic verses cited earlier for examples). In this connection, Arabic verbs are categorized into two groups according to their combinations with prepositions: (1) Restricted verbs, which have a tendency to combine with one preposition. One example from the Qur’an shows that the past tense verb abaqa (to flee) appears in the Qur’an with one preposition only—that is, the preposition ilá, in idh abaqa ilá al-fulki al-mashḥūn (Qur’an, 37:140; cf Dāwood, 2002, p. 1/7). “He fled to the overloaded ship” (Abdel Haleem, 2010, p. 452). (2) Nonrestricted verbs, which can combine with several prepositions. A good example is the past tense verb dakhala (to enter), which appeared in the Qur’an combined with four prepositions, as follows: (i) bi-’, as in wa-qad dakhalū bil-kufri wa-hum qad kharajū bi-hi (Qur’an, 5:61) “they come with the resolve to deny the truth and depart in the same state”. (Asad, 2011, p. 188); (ii) ʿalá, as in idkhulū ‘alayhim al-bāba (Qur’an, 5:23 and Qur’an, 13:23) “Enter upon them through the gate” (Asad, 2011, p. 175); (iii) min, as in lā-tadkhulū min bābin wāḥidin wa-dkhilū min abwābin mutafarriqah (Qur’an, 12:67) “Do not enter [the city all] by one gate, but enter by different gates” (Asad, 2011, p. 417); (iv) f ī, as in wa-ra’ayta al-nāsa yadkhulūna f ī d īni-illahi afwājā (Qur’an, 110:2) “and thou seest people enter God’s religion in hosts” (Asad, 2011, p. 1171). A verb may be ‘restricted’ in the Qur’anic discourse but ‘nonrestricted’ in the general usage of Arabic. The aforementioned verb abaqa, for instance, combines with the preposition min in general usage (Dāwood, 2002, p. 1/7), as in abaqa al-ʿabdu min sayydihi (Ibn Manẓūr, 1994) ‘the slave fed from his master’ (cf Aldahesh, 2016a).
46
Arabic idiomatic phrasal verbs
An important point here is that al-afʿāl al-nāqiṣah (defective verbs)—namely the verb kāna and its ‘sisters’ aṣbaḥa, ṣāra, amsá, etc.—and al-afʿāl al-jāmidah (invariable verbs)—namely laysa, mādāma, niʿma, biʾsa, etc.—are not included in this study. This is because they do not indicate actions (Dāwood, 2002, p. 1/5) and cannot combine with prepositions to form AIPVs.
3.3.2 Arabic preposition Arabic prepositions may well be defined as a few particles with vital syntactic and semantic functions. They characteristically link the significance of noun or pronouns to the significance of verbs and put those noun or pronouns into a genitive case. Arabic prepositions constitute a closed class of Arabic vocabulary. This is in the sense that they do not have referential meanings when they appear by themselves, do not allow for prefixes or suffixes to be added, and do not produce other derivational forms (Abdul-Raof, 2001; Aldahesh, 2013b). Arabic prepositions are itemized in what follows: ilá (to), ʿalá (on), f ī (in), ʿan (about), min (from), bi- (in, by, with), ka- (as), li- (for), wāw al-qasam (by), tāʾ al-qasam (by), mudh (for), mundhu (since), rubba (may be, probably, perhaps), wāw rubba (probably, perhaps), ḥattá (until), khalā (except), ḥāshā (except), ʿadā (except), kay (in order to), matá (when) (in the ancient Hudhayl language), laʿalla (perhaps), (in the ancient ʿUqayl language) (cf Aldahesh, 2009b). Unlike English linguists, who typically incorporate prepositions when studying English verbs, Arabic linguists (classical and modern) tend to devote a special part of their studies to Arabic prepositions or study them in separate monographs with other types of particles, which are commonly referred to as ḥurūfu al-maʿān ī (particles of meanings). The first medieval Arabic linguist who devoted a whole study to these vital particles was al-Kisāʾī (d. AH 189) followed by al-Akhfash (d. AH 215), al-Mubarrid (d. AH 285), al-Zajjājī (d. AH 337), al-Fārisī (d. AH 377), al-Rummānī (d. AH 412), al-Harawī (d. AH 415), Ibn Mālik (d. AH 672), al-Murādī (d. AH 749), and Ibn ʿaqīl (d. AH 769), among others. The main Arabic linguists who composed books in studying Arabic grammar and devoted special chapters for prepositions are al-Zamakhsharī (d. AH 538) in his book al-Mufaṣṣal and Ibn Hishām (d. AH 671) in his book Mughn ī al-Lab īb (for more details, see Dāwood, 2002; al-Shamsān, 1986; Aldahesh, 2016a). Arabic prepositions have been given two labels. Classical Arabic grammarians of the al-Kūfah school named them ḥurūfu al-iḍāfati, on the grounds that they are genitive particles that add the significance of verbs, which precede them to the significance of nouns or pronouns, which follow them (Haywood & Nahmad, 1965; Ghalāyīnī, 1986; al-Sāmarrāʾī, 1979). Classical grammarians of the al-Baṣrah school and modern linguists, on the other hand, named them ḥurūfu
Arabic idiomatic phrasal verbs 47 al-jarri or ḥurūfu al-khafḍi, on the grounds that they put the nouns or pronouns that they govern into the majrūr (genitive) case (Aldahesh, 2009b). Arabic prepositions are of two kinds, as far as their form is concerned (cf Wright, 1981; Aldahesh, 2009b): (1) Separable prepositions, which are written as separate and unattached words— e.g. ilá (to), ʿalá (on), f ī (in), ʿan (about), min (from); (2) Inseparable prepositions, which are written as attached morphemes prefixed to nouns or pronouns—e.g. bi- (in, by, with), ka- (as), li- (for), wāw alqasam (by), tāʾ al-qasam (by). Semantically speaking, Arabic prepositions fall into one of three types: (1) ḥurūfu al-jarri al-aṣliayti (authentic/genuine prepositions). These Arabic prepositions characteristically complete the signifcances of entities with which they are combined. In addition, they put nouns or pronouns that come after them in a genitive case, and they need a mutaʿalliq (relator). Arabic has 12 authentic/genuine prepositions, and they are itemized in what follows: min (from), bi- in, by, with), ka- (as), li- (for), ilá (to), ḥattá (until), ʿan (about), ʿalá (on), f ī (in), matá (when), ta- (by) and wa- (by). The function of this type of Arabic preposition is exemplifed in what follows. A word-by-word gloss of the example is provided frst, followed by a translation: sāfartu bil- ṭāʾirati. travelled-I by the-plane. ‘I travelled by plane’ [my translation]. The preposition bi- in this sentence is considered an authentic/genuine one in a sense that it completes the meanings of both the verb sāfara and the noun al-ṭāʾirati to the extent that the meaning of the whole sentence cannot add up without it. The preposition bi- in this example puts the following noun, al-ṭāʾirati, in the genitive and has a mutaʿalliq (relator)—that is, the verb sāfara. I must announce here that this type of Arabic prepositions constitutes the focus of this book. (2) ḥurūfu al-jarri al-zāʾidati (redundant prepositions). These Arabic propositions do not need a mutaʿalliq (relator). They are typically used to serve one communicative function—that is, to confrm the meaning of the sentence in which they are employed. Arabic has eight redundant prepositions, and they are itemized in what follows: min (from), bi- (in, by, with), ka- (as), li- (for), ilá (to), ʿan (about), ʿalá (on), f ī (in).
48
Arabic idiomatic phrasal verbs
This type of Arabic preposition is exemplifed in what follows. A word-by-word gloss of the example is provided frst, followed by a translation: mā qaraʾtu min kitābin. Not read-I from a book. ‘I didn’t read any book’ [my translation]. The preposition min in this sentence is a redundant one because it has no mutaʿalliq (relator) to be related to. Additionally, the meaning of the whole sentence can well add up without it. This preposition, however, has a rhetorical communicative function—that is, to emphasize the signifcance of the whole sentence. (3) ḥurūfu al-jarri al-shabihatu bil-zāʾidati (quasi-redundant prepositions). These Arabic prepositions usually add a new signifcance to the sentence in which they are employed, yet they do not need a mutaʿalliq (relator). The Arabic quasiredundant prepositions are as follows: khalā (except), ḥāshā (except), ʿadā (except), rubba (may be, perhaps), laʿalla (perhaps), and lawlā (had it not been for). This type of Arabic prepositions is exemplifed in what follows. A word-by-word gloss of the example is provided frst, followed by a translation: rubba sadīqin wafyyin. may be a friend faithful. ‘There may be a faithful friend’ [my translation]. The preposition rubba (maybe) in this sentence is a quasi-redundant one because it has no mutaʿalliq (relator) to relate to, yet it puts the noun sadīqin in the genitive. Moreover, it adds a new meaning to the sentence, such as likelihood, probability, or lessening (cf Zain al-ʿĀbidīn, 2008; Dāwood, 2002; Ghalāyīnī, 1986; Aldahesh, 2017). In his book Qaṭru al-nadá wa- ballu al-ṣadá, Ibn Hishām (n.d.) groups the Arabic prepositions in relation to their grammatical functions into six categories (p. 319). (1) Prepositions that put explicit nouns and pronouns in the genitive. This category includes seven Arabic prepositions: min (from), ilá (to) ʿan (about), ʿalá (on), fī (in), bi- (in, by, with), and li- (for). (2) Prepositions that put only explicit nouns in the genitive. This category includes three Arabic prepositions: ka- (as), ḥattá (until), and wāw al-qasam (by).
Arabic idiomatic phrasal verbs 49 (3) Prepositions that put two nouns in the genitive. This category includes one Arabic preposition only: tāʾ al-qasam (by). (4) Prepositions that put the interrogative particle mā (what) in the genitive. This category includes one Arabic preposition only: kay (in order to). (5) Prepositions that put ẓurūf al-zamān (adverbs of time) in the genitive. This category is of two Arabic prepositions: mudh (for), mundhu (since). (6) Prepositions that put explicit and implicit nouns in the genitive. This category has only one Arabic preposition: rubba (may be, probably, perhaps). Not all the aforementioned Arabic prepositions can combine with verbs to form AIPVs. The rule is that only the authentic, genuine, and ‘true’ prepositions— which have the ability to be employed in abstract senses—can be used for that communicative purpose (cf Lentzner, 1977, p. 33). The authentic, genuine, and ‘true’ prepositions can be classifed, according to their semantic properties, into two types. (1) Locative prepositions: Arabic prepositions of this type can specify where something is either in space or in time, and they are bi- (by, in, with), f ī (at, in), and ʿalá (on). Among them are the prepositions bi- (by, in, with), and ʿalá (on), which are deemed to be the most used in combination with Arabic verbs (Lentzner, 1977, p. 33). (2) Directional prepositions: Arabic prepositions of this type can specify a ‘change’ in location, or movement in space or time (Lentzner, 1977, p. 33), and they are prepositions that indicate direction with the connotation of ‘towards’—ilá (to, towards), li- (to), and ḥattá (until) (see Lentzner, 1977, p. 33)—and prepositions that signify direction with the connotation of ‘from’—min (from),ʿan (from, away from), and mundhū (since) (Lentzner, 1977, p. 33; cf Aldahesh, 2016a). Redundant Arabic prepositions and quasi-redundant Arabic prepositions fall outside the scope of this study because they have no ta‘alluq (attachment/relation) with the verb. Further details about al-ta‘alluq and its role in forming AIPVs are provided in the forthcoming section. In addition, the prepositions ḥattá (until), mundhū (since), wāw al-qasam (by), and tāʾ al-qasam (by) are also excluded, for two reasons: (1) They do not satisfy the criteria for forming AIPVs; (2) They do not combine with verbs in the Qur’anic discourse, which is the focus of this study. A vital point should be made here concerning the semantic characteristics of Arabic prepositions. Each Arabic preposition has its own functional signifcance, along with a number of other associated meanings that may overlap with the
50
Arabic idiomatic phrasal verbs
meanings of other prepositions (Dāwood, 2002, p. 1/29–31). Classical Arabic grammarians and modern Arabic grammarians have accounted for these meanings, with ample details. Al-Rummānī (2005), for instance, in Maʿānī al-Ḥurūf (signifcances of prepositions) attended to Arabic prepositions and their basic and associated meanings. Additionally, al-Shamsān (1986) provides a table for each Arabic preposition and the verbs with which it combines (pp. 224–253). He also elaborates on the semantic properties of seven prepositions (ilá, bī-, ʿalá, ʿan, f ī, li-, and min) (pp. 733–738). Arabic prepositions are context sensitive; that is, their functional significances cannot be grasped from their lexical, individual, and decontextualized meanings. They are instead understood by considering their micro-textual and macro-contextual factors. In addition, as is the case in many languages, Arabic prepositions are language specific. Arabic language users have their own distinct and special manner of employing them. This manner may or may not overlap with how prepositions are employed in other languages. Research revealed that the context sensitivity of Arabic prepositions and their Arabic language specificity constitute the main two difficulties related to mastering their usage by nonnative learners of Arabic language (Aldahesh, 2013b). The basic functional significances of the main authentic Arabic prepositions and their associated meanings are outlined as follows: (1) The Arabic preposition bi- primarily signifies al-ilṣāq (affixation). The associated meanings of b ī- include al-tabʿīḍ (partition), al-sababiyah (causation), al-muṣāḥabah (accompanying), al-taʿdiyah (transitivity), al-istiʿānah (seeking assistance), al-ẓarfiyah (adverbiality of time and place), al-istiʿlāʾ (superiority), al-mujāwazah (going beyond), al-badal (replacement), al-ʿawaḍ (compensation), al-taʿl īl (justification), al-qasam (oath), and intihāʾ al-ghāyah (end of destination). (cf Dāwood, 2002, p. 1:30–31; al-Shamsān, 1986, pp. 733–738; Aldahesh, 2016a). (2) The Arabic preposition f ī mainly signifies al-ẓarfiyah (adverbiality of time and place). The associated meanings of f ī include intihāʾ al-ghāyah (end of destination), al-istiʿlāʾ (superiority), al-ilṣāq (affixation), al-muṣāḥabah (accompanying), al-ʿawaḍ (compensation), al-taʿl īl (justification), al-muqāyasah (analogy), and al-sababiyah (causation). (cf Dāwood, 2002, p. 1:30–31; al-Shamsān, 1986, pp. 733–738; Aldahesh, 2016a). (3) The Arabic preposition ʿalá essentially denotes al-istiʿlāʾ (superiority). The associated meanings of ʿalá include al-mujāwazah (going beyond), al-ẓarfiyah (adverbiality of time and place), al-ilṣāq (affixation), al-taʿl īl (justification), maʾná baʿda (the meaning of after), and al-muṣāḥabah (accompanying). (cf Dāwood, 2002, p. 1:30–31; al-Shamsān, 1986, pp. 733–738; Aldahesh, 2016a). (4) The Arabic preposition li- mainly signifies al-ikhtiṣāṣ (habitual belonging). The associated meanings of l ī- include al-istiḥqāq (deserving), almulk (possession), al-taby īn (clarification), al-ṣayrūrah (act of becoming), al-taʿl īl (justification), al-nasab (attribution), al-tabl īgh (reporting),
Arabic idiomatic phrasal verbs 51 maʾná ʿinda (the meaning of having), intihāʾ al-ghāyah (end of destination), al-mujāwazah (going beyond), al-istiʿlāʾ (superiority), al-ẓarfiyah (adverbiality of time and place), and al-ta’diyah (transitivity). Classical Arabic grammarians and linguists have paid great attention to this particular preposition. For instance, al-Zajjajī (d. AH 337) dedicated a whole monograph to account for its properties. He distinguished 31 types of the letter lām, some of which overlap. Another example is Sibawayh (d. AH 180), who referred to the meaning of al-mulk and al-istiḥqāq in lām aljarr or lām al ʾiḍafah. It functions as “a particle making the noun or pronominal suffix annexed to it in ḥālat al-garr or the so-called genitive case” (cf Dāwood, 2002, p. 1:30–31; al-Shamsān, 1986, pp. 733–738; Shboul, 1983, pp. 24–47; Aldahesh, 2016a). (5) The Arabic preposition ilá primarily denotes intihāʾ al-ghāyah (end of destination). The associated meanings of ilá include ibtidāʾ al-ghāyah (start of destination), al-ẓarfiyah (adverbiality of time and place), al-ilṣāq (affixation), al-ikhtiṣāṣ (habitual belonging), al-taby īn (clarification), and al-muṣāḥabah (accompanying) (cf Dāwood, 2002, p. 1:30–31; al-Shamsān, 1986, pp. 733– 738; Aldahesh, 2016a). (6) The Arabic preposition min principally signifies ibtidāʾ al-ghāyah (start of destination). The associated meanings of min include intihāʾ al-ghāyah (end of destination), al-mujāwazah (going beyond), al-istiʿlāʾ (superiority), al-ẓarfiyah (adverbiality of time and place), al-ilṣāq (affixation), bayān al-jins (gender determination), al-badal (replacement), al-taby īn (clarification), al-taʿl īl (justification), al-faṣl (separation), and al-tabʿīḍ (partition) (cf Dāwood, 2002, p. 1:30–31; al-Shamsān, 1986, pp. 733–738; Aldahesh, 2016a). (7) The Arabic preposition ʿan primarily denotes al-mujāwazah (going beyond). The associated meanings of ʿan include al-istiʿlāʾ (superiority), al-ẓarfiyah (adverbiality of time and place), al-ilṣāq (affixation), al-istiʿānah (seeking assistance), al-sababiyah (causation), al-badal (replacement), al-taby īn (clarification), al-taʿl īl (justification), al-faṣl (separation), maʿná baʿda (the meaning of ‘after’), and al-tabʿīḍ (partition) (cf Dāwood, 2002, p. 1:30–31; al-Shamsān, 1986, pp. 733–738; Aldahesh, 2016a). The obvious overlap between the basic functional signifcances and the associated meanings of Arabic prepositions is the main reason for the phenomenon of al-inābah (preposition substitution) (Dāwood, 2002, p. 1/32). The phenomenon of al-inābah is attended to in detail in the forthcoming sections of this chapter.
3.4 AIPVs’ syntactic and semantic properties Having accounted for the characteristics of Arabic verbs and Arabic prepositions, I in the following attend to the issue of combining them in an idiomatic manner to produce AIVPs.
52
Arabic idiomatic phrasal verbs
Arabic verbs are classified into six categories according to their syntactic relationships with the prepositions with which they combine: (1) Arabic verbs that typically require Arabic prepositions (i.e. they cannot do without them); the verb intaqama (to take revenge on someone)—for example, requires the preposition min (from). Similarly, the verb takhallá (to abandon, to relinquish) typically requires the preposition ʿan (about). (2) Arabic transitive verbs that may optionally take prepositions—for example, the verb zāda (to increase), which may occur with the preposition min (from), and the verb baḥatha (to search), which may occur with the preposition ʿan (about). (3) Arabic intransitive verbs that use prepositions in construction with a noun phrase—for example, the verbs sahira ʿalá (to watch over someone/ something) and ḍaḥika min (to laugh at someone or something). (4) Arabic verbs that are passive (either in form or meaning), in which prepositions mark their underlying agents—for example, the verbs ukhidha bi- (to be influenced by someone or something) and uʿjiba bi- (to admire someone or something). (5) Arabic verbs that involve two noun phrases introduced by prepositions—for example, the verbs samiḥa bi- (to permit someone to do something) and amlā ʿalá (to dictate something to someone). (6) Arabic verbs that take one preposition for one meaning and another preposition for other meanings; for example, the verbs raghiba f ī (to desire someone/something) and raghiba ʿan (to detest, to turn away from someone/ something) (cf Lentzner, 1977; Aldahesh, 2016a, 2016b, 2017). Semantically speaking, AIPVs are the product of combining ḥurūf al-jarr (Arabic prepositions) with al-afʿāl (Arabic proper verbs). In AIPVs, the prepositions modify, to a certain degree, the basic signifcances of the verbs with which they combine. Therefore, changing a given preposition in an Arabic verb-preposition structure entails altering the meaning of the entire structure. Hence, in the essence of AIPVs exists a semantic interaction between the Arabic proper verb and the Arabic preposition. Such an interaction determines the idiomaticity degree of the whole Arabic verb-preposition structure. The semantic affect that Arabic prepositions have on Arabic verbs involves a ‘semantic transfer’. That is, they affect the proper actions of the verbs to the extent that they change their dictionary meanings (cf Dāwood, 2002; Lentzner, 1977; Aldahesh, 2017). Consider the following examples taken from the press: (1) Milyawn farans ī yuqliʿūna ʿan al-tadkh īni f ī ʿāmin wāḥid (BBC Arabic). ‘One million French quit smoking in one year’ [my translation]; (2) Ikhmād khams ḥarāʾiq f ī baldāti Akkār atat ʿalá masaḥātin min al-ashjāri wa-l-aʿshāb (VDL News). ‘Extinguishing five fires that ruined areas of trees and grass in Akkar towns’ [my translation].
Arabic idiomatic phrasal verbs 53 The preposition ʿan in the frst example changes the original signifcance of the verb yuqliʿūna (to take off) to a new signifcance—that is, ‘to quit’. Similarly, the preposition ʿalá in the second example changes the original significance of the verb atat (to come) to a new signifcance—that is, ‘to ruin, to destroy’. Some Western scholars of Arabic grammar (e.g. Wright, 1981; Cantarino, 1974; among others) have established a sort of systematicity, on the basis of semantic cohesion, between the classes of Arabic verbs and the prepositions with which they combine. They assumed, for instance, that because the Arabic preposition min signifies the casual point of departure, it is used with verbs meaning ‘to sell’ and ‘to give in marriage’. Furthermore, because the preposition ʿalá means ‘to be on, over’, it is employed with verbs meaning ‘to cover, to include’ and ‘domination power’ (Lentzner, 1977, p. 160). These analyses while often intuitively valid, are nonetheless restricted by two facts: first, such statements are only generalities and cannot account for all instances of verbpreposition occurrence. Second, they are not precise enough to be able to predict which preposition will be used where. (Lentzner, 1977, pp. 160–161) Nonetheless, researchers who have scrutinized the phenomenon of AIPVs in the Qur’an have identified a type of regularity between the prepositions and the classes of Arabic verbs with which they combine. This kind of regularity is attended to with more detail in the forthcoming chapter. The point to be made here is that the nontransparent, metaphorical significance (i.e. idiomaticity) of AIPVs is affected by four interrelated variables (cf al-Shamsān, 1986; Dāwood, 2002; Aldahesh, 2016a, 2016b): (1) The original significance of the verb; (2) The contextual significance of the verb; (3) The associated meaning of the preposition (i.e. the special meaning understood from the preposition as a result of its combination with a verb); (4) The object governed by the preposition, whether it is human or nonhuman, indicating time, indicating place, or indicating something else. Thus, the context determines the degree of idiomaticity of an AIPV. Consider the following illustrative examples: (1) atat Sārah min jāmiʿati Sydney. Came-she Sarah from the University of Sydney. ‘Sarah came from the University of Sydney’ [my translation]. (2) atat Sārah ʿalá al-ṭaʿāmi. Came-she Sarah on the-food. ‘Sarah ate the food up’ (i.e. entirely) [my translation].
54
Arabic idiomatic phrasal verbs
These two examples include two Arabic verb-preposition structures, namely atat min and atat ʿalá. Both structures involve the Arabic verb atat used with two different prepositions. Yet the frst is considered a mere literal Arabic verbpreposition structure because its constituents (i.e. verb and preposition) retain their direct, nonidiomatic, and transparent signifcances. The second structure is deemed an AIPV because its constituents (i.e. atat and ʿalá) sacrifce their direct meanings and fuse to produce one nontransparent, idiomatic unit of meaning— that is, ‘to eat up’, or ‘to consume entirely’. What is more, the object governed by the preposition plays a significant role in determining the idiomaticity degree of the Arabic verb-preposition structure. Let me illustrate this point with the following examples: (1) rabaṭat al-ḥizām ʿalá khiṣrihā. Fastened-she the-belt on waist-her. ‘She fastened the belt on her waist’ [my translation]. (2) Lawlā an rabaṭnā ʿalá qalbihā (Qur’an, 28:10). Had that fastened-we on heart-her. ‘Had we strengthened her heart’ (Shakir, 2011, p. 254). These two examples involve one Arabic verb-preposition structure, viz. rabaṭa ʿalá. Nevertheless, the frst is a plain Arabic verb-preposition structure, where its two components preserve their direct, nonidiomatic, and transparent signifcances. The second structure is an AIPV, where its two components (i.e. rabaṭa and ʿalá) abandon their direct meanings and combine to generate a new, nontransparent, idiomatic unit of meaning—that is, ‘to strengthen’. The main variable playing the vital role in this semantic shift is the difference in the objects governed by the preposition. The object of the preposition ʿalá in the frst example is the noun khiṣr (waist). In everyday Arabic, using this noun with the verb-preposition structure rabaṭa ʿalá is not uncommon to express the literal and direct meanings of both the verb rabaṭa and the preposition ʿalá. What is exceptional is the usage of the noun qalb (heart) as an object governed by the preposition ʿalá. This marked collocation gives the entire Arabic verbpreposition structure rabaṭa ʿalá its idiomatic nontransparent connotation. The AIPV atáʿalá is used in three contexts, which follow, to illustrate this marked collocation: (1) atáʿalayhi al-dahru. Came-it on-him the-time. ‘Time destroyed him’ (i.e. he died) [my translation]. (2) atáʿalá al-ʿamali. Came-he on the-work. ‘He completely finished the work’ [my translation].
Arabic idiomatic phrasal verbs 55 (3) atatʿalá al-jan īni tsʿatu shuhūrin. Came-they on the-embryo nine months. ‘The embryo is nine months old’ [my translation]. Thus, the relationship between the Arabic verb, the preposition, and the object of that preposition in an AIPV is multilayered. Between them is a subtle semanticosyntactic interrelation comprising two concurrent dimensions. To use Firth’s (1957–1986) terminologies, they are colligational and collocational dimensions. By colligation (al-intiẓām), we mean the grammatical conditions that regulate the combination of a given set of words. The English verb prefer, for example, colligates with the particle to, followed by an infinitive (e.g. I prefer to go). It does not, however, colligate with the -ing form (cf Daud et al., 2003). By collocation (al-muṣāḥabah or al-taḍām), we mean the common co-occurrence of words to have developed an idiomatic semantic relation (cf Daud et al., 2003). Or, as Dickins et al. (2002) put it, collocation is “an occurrence of one word in close proximity with another” (p. 71). The English verb perform, for instance, collocates with operation, dog collocates with bark, and rancid collocates with butter (cf Daud et al., 2003; Aldahesh, 2016a). This subtle and delicate relationship between Arabic verbs and Arabic prepositions has been approached by Arabic classical and modern grammarians by attending to the following four different, but utterly interconnected, factors: (1) (2) (3) (4)
al-Taʿalluq (verb–preposition relation, attachment, or dependency); al-Taʿadd ī (verb transitivity); al-Inābah (preposition substitution); al-Taḍm īn (verb implication).
To better investigate this intricate relationship, we need to account for these factors one by one in a more elaborate manner. The objective here is to identify the extent to which these factors affect the idiomaticity of the AIPVs.
3.4.1 al-Taʿalluq (verb–preposition relation, attachment, or dependency) In his book dalāʾil al-iʿjāz, Abdul Qāhir al-Jurjānī (d. AH 471) explains his prominent notion of al-naẓm (discourse arrangement). He maintains that al-naẓm is nothing but relating (taʿalluq) parts of speech to each other. If three parts of speech exist in the Arabic language (i.e. nouns, verbs, and particles), al-taʿalluq is of three categories: (1) Relating a noun to another noun (taʿalluqu al-isimi bil-ismi); (2) Relating a noun to a verb (taʿalluqu al-isimi bil-fiʿili); (3) Relating a particle to a noun or to a verb (taʿalluqu al-adāti bil- isimi aw bil-fiʿili).
56
Arabic idiomatic phrasal verbs
The latter category in turn has three kinds, one of which is relating a preposition to a verb (taʿalluqu ḥarf al-jarri bil-fʿili). This would give an intransitive verb a sense of transitivity (al-Jurjānī, 1992, pp. 4–7). The point to be made here is that al-taʿalluq indicates that the significance of the verb is completed by the preposition and the noun governed by that preposition (Heliel, 1994, p. 146; al-Daḥdaḥ, 2009, p. 21). In other words, al-taʿalluq constitutes a link that complements the significance of the verb and connects the preposition and the noun that it governs with that verb (al-Labādī, 1986, cited in Heliel, 1994, p. 146). Such a link makes the preposition and its object semantically related and bound to the verb “just as a part is bound to its whole, or a branch to its root” (Hassan, 1963, p. 2/405–406; Lentzner, 1977, p. 29). Grammarians call this verb mutaʿalliq (relator), and it may appear in one of two ways in a given sentence: (1) fiʿil ẓāhir (a stated, itemized, explicit verb), as in atat al-nāru ʿalá al-bayti. Came-it the-fire on the-house. ‘The fire completely devastated the house’ [my translation]. The mutaʿalliq ‘relator’ in this example is the stated verb atat. (2) fiʿil maḥdhūf (an unstated, hidden, implied verb), as in al-kitābu f ī al-maktabati. The-book in the-library. ‘The book is in the library’ [my translation]. The mutaʿalliq ‘relator’ in this example is the unstated yet implied verb (i.e. mawjūdun [exist]). Even though al-taʿalluq is a precondition for an AIPV, only the frst type of the mutaʿalliq (i.e. the stated, itemized, explicit verb) is relevant to this study. The second type (i.e. the unstated, hidden, implied verb) is irrelevant; thus, it has been excluded.
3.4.2 al-Taʿaddī (verb transitivity) Classical Arabic scholars to have studied al-taʿadd ī (transitivity) (e.g. Sibawayh, 1977, p. 1/34 and beyond; al-Mubarrid, 1965, p. 3/187 and beyond; Ibn al-Sarrāj, 1973, p. 1/203 and beyond; al-Jurjānī, 1982, p. 1/628 and beyond; Ibn Jinnī, 1954a, p. 2/214 and beyond; Ibn Jinnī, 1954b, p. 1/84 and beyond; Ibn Jinnī, 1972, p. 51; Ibn Hishām, 1953, p. 354 and beyond; al-Siūṭī, 1327 AH, p. 5/9 and beyond. Cf Qinano, 2005) and formulated two Arabic verb categories: (1) afʿāl mutaʿaddiyah bi-nafsihā (transitive verbs that pass on to their objects through themselves); (2) afʿāl lāzimah or afʿāl mutaʿaddiyah bi-ghayrihā (intransitive verbs that pass on to their objects through other means) (cf Heliel, 1994, p. 144).
Arabic idiomatic phrasal verbs 57 Some verbs, however, can be used both transitively and intransitively. In addition, doubly transitive verbs (Lentzner, 1977, p. 161) govern more than one direct object, such as raʾá (to see) and wajada (to fnd) (al-Shamsān, 1987, p. 39). Furthermore, al-afʿāl al-lāzimah (intransitive verbs) characteristically signify actions indicating movement and the transport of the agent, whereas al-afʿāl al-mutaʿadyah (transitive verbs) denote actions that surpass the agent to its surrounding world. They include actions indicating all activities undertaken by the agent within its environment (al-Shamsān, 1986, p. 17). However, the distinction between these two categories of the Arabic verb is not clear. Verbs may be used transitively and intransitively in Arabic, depending on the communicative purpose of their usage (al-Shamsān, 1986, p. 749). Arabic grammarians identify, as follows, three means by which a given Arabic intransitive verb can be made transitive: (1) Introducing hamzah to the verb—for example, dhahaba (to go) and adhhaba (to make something or someone to go); (2) Doubling the second radical letter of the verb—for example, fariḥa (to become happy) and farraḥa (to make someone happy); (3) Using prepositions—for example, dhahaba (to go) and dhaaba ilá (to go to) (for other transitivity particles see al-Shamsān, 1987, pp. 22–45). Consequently, one of the major functions of Arabic prepositions is to ‘trasitivize’ the intransitive verb. Therefore, combining a preposition with a verb might modify its functionality and change it from transitive to intransitive and vice versa. The verb dhahaba, for instance, is an intransitive verb. It turns out to be transitive when it combines with the preposition bi-, as in the following Qur’anic verse (Dāwood, 2002, p. 1/21): dhahaba Allahu bi-nūrihim (Qur’an, 2:17) ‘God takes away all their light’ (Abdel Haleem, 2010, p. 5). In contrast, the verb khālafa in the following example is a transitive and becomes intransitive when it combines with the preposition ʿan (Dāwood, 2002, p. 1/21): faliyaḥẓar-i-lladh īna yukhālifūna ʿan amrihi (Qur’an, 24:63) ‘And those who go against his order should beware’ (Abdel Haleem, 2010, p. 60).
3.4.3 al-Inābah (preposition substitution) and al-Taḍmīn (verb implication) The broad semantic range of Arabic verbs and prepositions on the one hand and the overlap that occurs among their significances on the other hand have made investigating al-inābah (preposition substitution) and al-taḍm īn (verb implication) of interest to classical and modern Arabic scholars alike. These two issues are interrelated to the extent that one cannot deal with them separately.
58
Arabic idiomatic phrasal verbs
al-inābah, also called al-taʿāqub (succession) (cf Dāwood, 2002, p. 1/32), means that a preposition might be a substitute for another preposition in some contexts. The word al-inābah is exemplified in the following Qur’anic structures: (1) Yaḥfazūnahu min amri-llāhi (Qur’an, 13:11). “that could preserve him from whatever God may have willed” (Asad, 2011, p. 431). In this verse, the preposition min replaces the preposition bi- (al-Mubarrid, 1965; cf al-ʿAṭiyah, 2008, p. 138). (2) Wa-la-ʾuṣallibannakum f ī judhūʿi al-nakhli (Qur’an, 20:71). “I shall most certainly crucify you in great numbers on trunks of palm trees” (Asad, 2011, p. 969). In this verse, the preposition f ī replaces the preposition alá (al-Mubarrid, 1965; cf al-ʿAṭiyah, 2008, p. 138). (3) Fa-raddū aydiyahum f ī afwāhihim (Qur’an, 14:9). “they covered their mouths with their hands” (Asad, 2011, p. 445). In this verse, the preposition f ī replaces the preposition ilá (al-Māliqī, 1985; cf al-ʿAṭiyah, 2008, p. 240). (4) Am la-hum sullamun yastamiʿūna f ī-hi (Qur’an, 52:38). “Or have they a ladder by which they could (ascend to ultimate truths and) listen (to what is beyond the reach of human perception)?” (Asad, 2011, p. 569). In this verse, the preposition f ī replaces the preposition ʿalá (al-Mubarrid, 1965; cf al-ʿAṭiyah, 2008, p. 138). Sibawayh (d. AH 180) gives two reasons for such replacements (1977, p. 4/226): (1) Diversity of Arabic dialects; (2) Similarity of the Arabic prepositions’ significances. Whereas al-taḍm īn means that the signifcance of a verb may include the signifcance of another verb, which entails employing a preposition not typically combined with that verb, it instead combines with the other verb whose signifcance is implied (Ibn Jinnī, 1954a, p. 2/308; al-Shamsān, 1987, pp. 68–72). The concept of al-taḍm īn is illustrated in the following Qur’anic structures: (1) Ihdinā al-ṣirāṭ al-mustaq īm (Qur’an, 1:6). “Guide us the straight way” (Asad, 2011, p. 2). In this verse, the verb Ihdinā (guide us) implies the significance of the verb aʿlimnā (let us know) (al-Farrāʾ, 1980; cf al-ʿAṭiyah, 2008, p. 241).
Arabic idiomatic phrasal verbs 59 (2) Wa-ihdinā ilá sawāʾi al-ṣirāṭ (Qur’an, 38:22). “and show (both of) us the way to rectitude” (Asad, 2011, p. 833). In this verse, the verb Ihdinā (guide us) implies the significance of the verb arshidnā (direct us) (al-Farrāʾ, 1980; cf al-ʿAṭiyah, 2008, p. 241). (3) Wa-lā taʾkulū amwālahum ilá amwālikum (Qur’an, 4:2). “and do not consume their possessions together with your own” (Asad, 2011, p. 122). In this verse, the verb taʾkulū (to consume) implies the significance of the verb taḍummū (include) (al-Farrāʾ, 1980; cf al-ʿAṭiyah, 2008, p. 241). Most classical Arabic linguists and most modern Arabic linguists are in favour of the notion of al-taḍm īn and accept as true that verbs may imply signifcances of other verbs in some contexts (for scholars in favour of the notion of al-taḍm īn, see al-Shamsān, 1987, p. 171; al-ʿAṭiyah, 2008, p. 259). Yet when it comes to al-inābah, we are faced with three different opinions: (1) Some Arabic linguists (mostly from the al-kūfah school) accept that prepositions may be substitutes for other prepositions, on some occasions; (2) Some Arabic linguists (mostly from the al-Baṣrah school) disagree and attribute the co-occurrences of some prepositions with verbs with which they do not usually combine to the aforementioned notion of al-taḍm īn; (3) A third group of Arabic linguists (such as Ibn al-Sarrāj d. AH 316) steer a middle ground, by accepting al-inābah on the condition that the prepositions in question have a similar significance. For instance, the prepositions f ī and bi- (Dāwood, 2002, p. 1/32–37) as well as li and ilá can be used interchangeably (for more details on these opinions, see Sibawayh, 1977, p. 4/217; al-Mubarrid, 1965, p. 1/45; al-Sāmarrāʾī, 2002, p. 3/7; Hassan, 1963, p. 2/496; Lentzner, 1977, p. 182). In addition, al-taḍm īn (verb implication) emphasizes that verbs imply new signifcances when they co-occur with prepositions with which they do not usually combine (al-ʿAṭiyah, 2008). Along these lines, Ibn Jinnī (d. AH 392) maintains that using one preposition instead of another is not without a communicative purpose. Violating Arabic grammar conventions by using a preposition other than the one typically combined with the verb at hand aims to communicate a subtle meaning and send a delicate message to readers or hearers (al-ʿAṭiyah, 2008). On the basis of the aforesaid facts, we can assume that the four factors (i.e. al-taʿalluq, al-taʿaddī, al-inābah, and al-taḍmīn) control, to a great extent, the significances of both the verbs and the prepositions in any given context. They also control the degree of idiomaticity of both constituents when combining to form AIPVs. Therefore, to better understand the meaning of any given AIPV, the four factors must be taken into consideration. First, one needs to know the mutaʿalliq (relator) of the given preposition and then the transitiveness of the verb at hand
60
Arabic idiomatic phrasal verbs
al-Ta alluq
al-In bah
‘Verb-preposition Relation’
‘Preposition Substitution’
AIPV
al-Ta add
al-Ta m n
‘Verb Transitivity’
‘Verb Implication’
Figure 3.1 The Interrelatedness of the AIPV four factors
(i.e. whether it is a transitive verb or an intransitive verb). This helps in determining the significance of the preposition with which it combines and in turn in determining whether the preposition in question typically combines with that verb or with a substituted one (al-inābah), because that verb carries the meaning of another verb (al-taḍm īn). The following diagram illustrates the interrelatedness of the four factors and their importance to AIPVs.
4.5 AIPVs’ register variations Unlike EPVs, which are less formal, less rhetorical, and commonly used by everyone in everyday contexts, AIPVs are far more formal and highly rhetorical. As demonstrated in this chapter, they are typically used in formal settings such as literary works, religious contexts, and press articles. Arab writers, poets, and artists typically use such metaphorical expressions to liven up their works or to avoid using taboo words or being rude. They employ words in a different way by widening or narrowing their semantic meanings according to the context and thoughts that they want to express. In other words, they impart new life into such words and inject the language with new items that can be used later in the everyday contexts (Ḥammād, 1983). AIPVs are common in both written Arabic and spoken Arabic and have productive natures. The metaphorical use of verbs and prepositions plays an important role in their productivity. Let us consider this issue through three examples, provided by Ḥammād (1983): (1) qāmat al-ḥarb ʿalá sāq. Literally: ‘the war has stood on a leg’. Metaphorically: ‘the war has erupted’ [my translation].
Arabic idiomatic phrasal verbs 61 (2) kashaft al-ḥarbu ʿan sāqihā. Literally: ‘the war has exhibited its leg’. Metaphorically: ‘the war has broken up’ [my translation]. (3) abdá al-sharruʿan nājidhayhi. Literally: ‘the evil has shown its molars’. Metaphorically: ‘the evil has been obviously revealed’ [my translation]. The AIPVs used productively in these examples are qāmat ʿalá, kashaft ʿan, and abdá ʿan respectively. Most of AIPVs are confned to the written mode and used only by educated people from certain backgrounds of society and education. However, hearing AIPVs used by laypeople is not uncommon in informal, colloquial Arabic, with some modifcation of verbs forms imposed by the dialect-specifc nuances. Three examples from the Egyptian colloquial suffce to illustrate this point: (1) Biyigr ī ʿalá kūmat ʿiyāl. Literally: ‘He runs on many kids’. Metaphorically: ‘He is the breadwinner for a troop of children’ [my translation]. (2) Bitimsh ī ʿalá ḥalli shaʿrahā. Literally: ‘She walks on release of her hair’. Metaphorically: ‘She is a scandalous woman’ [my translation]. (3) Yishrab min al-baḥr. Literally: ‘He drinks from the sea’. Metaphorically: ‘He can go stuff himself’ [my translation]. The AIPVs employed productively in these examples are yigr ī ʿalá, timsh ī ʿalá, and yishrab min, in that order. While the EPVs constitute a highly productive category, their Arabic counterparts can be productive only when used as parts of metaphorical expressions, as is the case in the illustrative examples cited earlier.
3.6 Conclusion This chapter investigated the phenomenon of AIPVs, established a parameter for their idiomaticity, and outlined the key factors governing the syntactic and semantic relationships between their constituents. Two types of Arabic verbpreposition structures were identified and discussed: nonidiomatic, literal, transparent structures and idiomatic, metaphorical, nontransparent structures. The feature by which one can differentiate between the two types is idiomaticity—the key feature of the idiomatic, metaphorical, and nontransparent structures. The idiomatic nature of the AIPVs makes significances of both the verb and the preposition nontransparent and cannot be taken literally. Both components
62
Arabic idiomatic phrasal verbs
rather sacrifice their genuine significances and fuse to produce a new metaphorical significance, which differs greatly from their basic significances. The nonidiomatic, literal, and transparent type, however, is a mere grammatical colligation between a verb and a preposition with no semantic collocability. Therefore, I decided that it should not be included as an AIPV. The idiomatic tendency of AIPVs and their context sensitivity and language specificity complicate the task of nonnative learners of Arabic in understanding AIPVs’ intended meanings in different contexts. This chapter has also shown that the idiomatic, nontransparent, metaphorical significance of AIPVs is affected by four variables: the original significance of the verb, the contextual significance of the verb, the associated meaning of the preposition, and the object governed by the preposition. Each Arabic preposition has its own original semantic significance and several other associated meanings, which may overlap with the meanings of other prepositions. Not all Arabic prepositions, however, can collocate with verbs to form AIPVs. Only authentic locative and directional prepositions can do so. This chapter has also demonstrated that four key factors control the syntactic and semantic relationships of Arabic verbs and prepositions: al-taʿalluq, al-taʿadd ī, al-inābah, and al-taḍm īn. These factors delineate not only the significances of both the verbs and prepositions but also the degree of their idiomaticity when merging to form an AIPV. Finally, while AIPVs are predominantly formal and rhetorical expressions used by highly educated people, they may also be used by laypeople in informal, colloquial Arabic.
References Abboud, P. F., & McCarus, E. N. (1968). Elementary Modern Standard Arabic. Cambridge University Press. Cambridge. Abdel Haleem, M. A. S. (2010). The Qur’an: English Translation and Parallel Arabic Text. Oxford University Press. Oxford. Abdul-Raof, H. (2001). Arabic stylistics: A Coursebook. Harrassowitz Verlog. Wiesbaden. Adonis, A. S. (1996). Al-Aʿmāl al-Shiʿriyah. Dār al-Madá. Damascus, Syria. al-ʿAṭiyah, A. M. (2008). Ḥurūf al-Jarr bayna al-Nyābati wa-l-Taḍmīn. Majallat al-Turāth al-ʿArab ī, year, 28 (112), 233–260. al-Daḥdaḥ, A. (2009). Miʿjam Qawāʿid al-Lughati al-ʿArabiyati f ī Jadāwil wa Lawḥāt. Maktabat Lubnān Nāshirūn. Beirut, Lebanon. Aldahesh, A. Y. (2008). Derivational Mechanisms and Productivity Aspects in English and Arabic: A Contrastive Linguistic Study. Translation Watch Quarterly, 4 (2), 49–63. Aldahesh, A. Y. (2009a). Translating Idiomatic English Phrasal Verbs into Arabic. VDM Verlag Dr. Muller. Saarbrucken, Germany. Aldahesh, A. Y. (2009b). Notes on Western Scholars’ Classification of Arabic Prepositions. Proceedings of the Conference “Arabic Language on Perspective of Social and Culture”. October 12–14, 2009. North Sumatra University, Indonesia, Medan. Aldahesh, A. Y. (2013a). On Idiomaticity in English and Arabic: A Cross—linguistic Study. Journal of Language and Culture, 4 (2), 23–29.
Arabic idiomatic phrasal verbs 63 Aldahesh, A. Y. (2013b). Context Sensitivity and Language Specificity of Arabic VerbPreposition Structure: The case study of English Learners of Arabic. International Journal of Linguistics, 5 (3), 177–197. Aldahesh, A. Y. (2016a). Pinning Down the Phenomenon of Phrasal Verbs in Arabic. International Journal of Language and Linguistics, 3 (1), 12–24. Aldahesh, A. Y. (2016b). Towards a Model for Analyzing and Assessing Translation of Qur’anic Idiomatic Phrasal Verbs. Arab World English Journal, Special Issue on Translation (5), 33–53. Aldahesh, A. Y. (2017). Qur’anic Idiomatic Phrasal Verbs: Their Syntactic and Semantic Properties. International Journal of Language and Linguistics, 4 (3), 12–27. Aldahesh, A. Y. (2018). On Pragmatic Idiosyncrasies of Qur’anic Idiomatic Phrasal Verbs. International Journal of Language and Linguistics, 5 (2), 110–118. al-Farrāʾ, Y. Z. (1980). Maʿān ī al-Qur’an. ʿālam al-Kutub. Beirut, Lebanon. al-Jurjānī, A. (1982). Al-Muqtaṣidu f ī Sharḥi al-ʾIḍāḥi. Wizārat al-Thaqāfati wa-lIʿlāmi al-ʿIrāqiyah. Baghdad, Iraq. al-Jurjānī, A. (1992). Dalāʾilu al-Iʿjāz. Maṭbaʿatu al-Madanī. Cairo, Egypt. al-Māliqī, A. A. (1985). Raṣfu al-Mabān ī f ī Ḥurūfi al-Maʿān ī. Dār al-Qalam. Damascus, Syria. al-Mubarrid, A. M. (1965). Al-Muqtaḍab. Al-Majlis al-Aʿlá lil-Shiʾūni al-Islāmiyah. Cairo, Egypt. al-Rummānī, A. (2005). Maʿān ī al-Ḥurūf. al-Maktabatu al-ʿAṣriyah. Beirut, Lebanon. al-Sāmarrāʾī, F. (2002). Maʿan ī al-Naḥū. Dār al-Fikr. Beirut, Lebanon. al-Sāmarrāʾī, I. (1979). Muqaddimatun f ī Tār īkhi al-ʿArabiyati. Dār al-Ḥurriyya. Baghdad, Iraq. al-Shamsān, A. I. (1986). Al-Fiʿilu f ī al-Qurʿani al-Kar īmi Taʿadyatuhu wa-Luzūmuhu. The Kuwait University. Kuwait. al-Shamsān, A. I. (1987). Qaḍāyā al-Taʿad ī wa-l-Luzūmi f ī al-Darsi al-Naḥw ī. Maṭbaʿatu al-Madanī. Jiddah, Saudi Arabia. al-Siūṭī, J. A. (AH 1327). Hamʿu al-Hawāmiʿi Sharḥu Jamʿi al-Jawāmiʿi. First Edition. Maṭbaʿatu al-saʿādah. Cairo, Egypt. al-Wasat (2015). Online Arabic Journal. Retrieved from http://www.alwasatnews. com/3213/news/read/569333/1.html Asad, M. (2011). The Message of the Qur’an. Islamic Book Trust. Kuala Lumpur. Awwad, M. (1990). Equivalence and Translatability of English and Arabic Idioms. Papers and Studies in Contrastive Linguistics, 26, 57–67. Baker, M. (1992). In Other Words: A Coursebook on Translation. Routledge. London and New York. Bataineh, R. F., & Bataineh, R. F. (2002). The Difficulties Jordanian Graduate Learners of English as a Second Language Face When Translating English Idioms into Arabic. RASK, 16, 33–83. BBC Arabic. Retrieved from www.bbc.com/arabic/science-and-tech-44284057/ Bolinger, D. (1971). The Phrasal Verb in English. Harvard University Press. Cambridge, MA. Cantarino, V. (1974). Syntax of Modern Arabic Prose. Indiana University Press. Bloomington and London. Cowie, A. P., & Mackin, R. (1993). Oxford Dictionary of Phrasal Verbs. Second Edition. Oxford University Press. Oxford. Daud, N., Dollah, N., & Zubir, B. (2003). Linguistics Dictionary: English-Arabic Arabic-English. A. S. Noordeen. Kuala Lampur.
64
Arabic idiomatic phrasal verbs
Dāwood, M. M. (2002). Al-Qurʿan al-Kar īm wa-Tafāʿulu al-Maʿān ī: Dirāsatun Dilāliyatun li-Taʿlluqi Ḥarfi al-Jarri bil-Fiʿili wa-ʾAtharihi f ī al-Maʿná f ī al-Qurʿani al-Kar īm. Dār Gharīb. Cairo, Egypt. Dickins, J., Hervey, S., & Higgins, I. (2002). Thinking Arabic Translation. Routledge. London and New York. Fraser, B. (1976). The Verb-Particle Combination in English. Academic Press. New York, San Francisco, and London. Ghalāyīnī, M. (1986). Jāmiʿu al-Durūsi al-ʿArabiyati. Al-Maktabatu al-ʿAṣriyah. Beirut, Lebanon. Ghazala, H. (2003). Idiomaticity Between Evasion and Invasion in Translation: Stylistic, Aesthetic and Connotative Considerations. Babel, 49 (3), 203–227. Ḥammād, A. (1983). ʿAwāmil al-Taṭawwuri al-lughaw ī. Dār al-ʾAndalus. Beirut, Lebanon. Hassan, A. (1963). Al-Naḥū al-Wāf ī. Third Edition. Dār al-Maʿārif. Cairo, Egypt. Haywood, J. A., & Nahmad, H. M. (1965). A New Arabic Grammar of the Written Language. Lund Humphries. London. Heliel, M. H. (1994). Verb-Particle Combinations in English and Arabic: Problems for Arab Lexicographers and Translators. In R. De Beaugrand, A. Shunnaq, & M. H. Heliel (Eds.), Language, Discourse and Translation in the West and Middle East. John Benjamins Publishing Company. Amsterdam and Philadelphia. Ibn al-Sarrāj, M. S. (1973). Al-Aṣulu f ī al-Naḥū. Maṭbaʿatu al-Aʿẓamī. Baghdad, Iraq. Ibn Hishām, A. (1953). Shudhūr al-dhahab. Sixth Edition. Al-Maktabatu al-Tijāriyatu al-Kubrá. Cairo, Egypt. Ibn Hishām, M. (n.d.). Qaṭru al-Nadá wa-Balu al-Ṣadá. Maktabatu al-Riyadh. Riyadh, Saudi Arabia. Ibn Jinnī, A. U. (1954a). Al-Khaṣāʾiṣ. Maṭbaʿatu al-Bābī al- Ḥalabī. Cairo, Egypt. Ibn Jinnī, A. U. (1954b). Al-Munṣif. Maṭbaʿatu al-Bābī al- Ḥalabī. Cairo, Egypt. Ibn Jinnī, A. U. (1972). Al-Lumaʿ. First Edition. Dār al-Kutubi al-Thaqāfiyah. Kuwait. Ibn Manẓūr, J. M. (1994). Lisānu al-ʿArab. Third Edition. Dār Ṣādir. Beirut, Lebanon. Kharma, N., & Hajjaj, A. (1989). Errors in English Among Arabic Speakers: Analysis and Remedy. Longman. London. Lattey, E. (1986). Pragmatic Classification of Idioms as an Aid for the Language Learning. IRAL, 24 (3), 217–233. Lentzner, K. R. (1977). Semantic and Syntactic Aspects of Arabic Prepositions. Unpublished doctoral dissertation, Georgetown University, Washington, DC. Maḥfūẓ, N. (1991a). Bayna al-Qaṣrayn. In al-Aʿmāl al-Kāmila. Matbaʿatu Lubnān. Beirut, Lebanon. Maḥfūẓ, N. (1991b). Qaṣru al-Shawq. In al-Aʿmāl al-Kāmila. Matbaʿatu Lubnān. Beirut, Lebanon. McArthur, T. (1975). Using Phrasal Verbs. Second Edition. Collins. London and Glasgow. McArthur, T. (1979). The Strange Cases of the English Phrasal Verb. Zielsprache Englisch, 9 (3), 24–26. McArthur, T. (1989). The Long-neglected Phrasal Verbs. English Today, 5 (2), 38–44. Mosteghanemi, A. (2012). al-Aswad Yal īqu Biki. Dār Nawfal. Beirut, Lebanon. Palmer, F. R. (1974). The English Verb. Second Edition. Longman. London. Qinano, A. (2005). Al-Mutqan: Miʿjamu Taʿadd ī al-Afʿāli f ī al-lughati al-ʿArabiyah. Dār al-Rātib al-Jāmiʿiyah. Beirut, Lebanon.
Arabic idiomatic phrasal verbs 65 Richards, J., Platt, J., & Weber, H. (1985). Longman Dictionary of Applied Linguistics. Longman Group UK Limited. Harlow, Essex. Ryding, K. G. (2005). A Reference Grammar of Modern Standard Arabic. Cambridge University Press. Cambridge. Shakir, M. H. (2011). The Qur’an. Seventh Edition. Tahrike Tarsile Qur’an, Inc. Elmhurst, New York, NY. Shboul, A. M. H. (1983). ‘Having’ in Arabic. Journal of Arabic Linguistics, Heft (11), 24–47. Sibawayh, A. Q. (1977). Al-Kitāb. Al-hayʾatu al-Maṣriyatu al-ʿāmatu lil kitāb. Cairo, Egypt. Side, R. (1990). Phrasal Verbs: Sorting them Out. ELT Journal, 44 (2), 144–152. Trask, R. L. (2000). Dictionary of English Grammar. Penguin Group. London. VDL News. Retrieved from https://vdlnews.com/lebanon-news/5-42 Wallace, L. C. (1968). Idiomaticity as an Anomaly in the Chomskyan Paradigm. Foundations of Language, 4, 109–127. Wightwick, J., & Gaafar, M. (1998). Arabic Verbs and Essentials of Grammar. Passport Books. Lincolwood, IL. Wright, W. (1981). A Grammar of the Arabic Language. Third Edition. Librairie Du Liban. Beirut, Lebanon. Yallor, C., Bernard, J., Blair, D., Butler, S., Delbridge, A., Peters, P., & Witton, N. (2005). Macquarie Dictionary. Fourth Edition. Macquarie University. NSW, Sydney, Australia. Zain al-ʿĀbidīn, B. N. (2008). Ḥurūfu al-Jarri al-Zāʾidatu wa-l-Shabīhatu bil-Zāʾidati: Dirāsatun Naḥwiyah. Majalatu al-ʿUlūmi wa-l-Thaqāfah, 9 (2), 1–13.
4
Qur’anic idiomatic phrasal verbs
4.1 Introduction Combining proper verbs with prepositions in an idiomatic manner constitutes a common feature of the Qur’anic discourse. In the two-volume book al-Qurʿan al-Karīm wa-Tafāʿulu al-Maʿānī: Dirāsatun Dilāliyatun li-Taʿlluqi Ḥarfi al-Jarri bi- al-Fiʿili wa-ʾAtharihi f ī al-Maʿná f ī al-Qurʿani al-Karīm, Dāwood (2002) makes the most significant contribution by attending to this phenomenon in the Qur’an. A significant contribution has also been made by al-Shamsān (1986) in his book al-Fiʿilu f ī al-Qurʿani al-Karīmi Taʿadyatuhu wa-Luzūmuh. Nonetheless, as mentioned in the previous chapter, the scholars who have tackled this phenomenon in the Arabic language do not differentiate between the idiomatic and nonidiomatic types of Arabic verb-preposition structures. Dāwood (2002) and al-Shamsān (1986) are not exceptions in confusing the two types of Arabic verb-preposition structures. In the former’s treatment of this phenomenon in the Qur’an, Dāwood (2002) thoroughly traces all verb-preposition occurrences; elaborates on their syntactic and semantic nuances; and provides quite fruitful insights, charts, and statistics. Still, like other researchers, he does not distinguish between idiomatic combinations and nonidiomatic combinations. This is, in my view, due to the perspective from which he tackles this phenomenon: Dirāsatun Dilāliyatun li-Taʿlluqi Ḥarfi al-Jarri bi- al-Fiʿili wa-ʾAtharihi f ī al-Maʿná f ī al-Qurʿan al-Karīm ‘a semantic study in the preposition-verb relation and its impact on meanings in the Qur’an’ [my translation]. To tackle the issue from such a wide-ranging view, Dāwood incorporates every combination of verb preposition in the Qur’an, and he does so without paying attention to the idiomaticity of some combinations and the nonidiomaticity of others. As discussed in previous chapter, al-taʿlluq (relation, attachment, or dependency) helps to shape the idiomaticity of the Arabic verb-preposition combinations; yet it is not the only factor. This factor may occur in many Arabic verb-preposition combinations but have no effect on their idiomaticity, as was demonstrated earlier. Likewise, al-Shamsān (1986), in his treatment of verbs in the Qur’an, provides interesting insights and several useful tables of verbs and the prepositions with which they typically combine. Nevertheless, he also stops short of differentiating between idiomatic combination and nonidiomatic combination types. This is due
Qur’anic idiomatic phrasal verbs 67 Table 4.1 Types of Qur’anic verb-preposition combinations Idiomatic Qur’anic verb-preposition combinations (QIPVs)
Nonidiomatic Qur’anic verb-preposition combinations
wa-lyaḍribna bi-khumurihinna ʿalá juyūbihinna (Qur’an, 24:31). “and let them wear their head-coverings over their bosoms” (Shakir, 2011, p. 230). wa-ākharūna yaḍribūna f ī al-arḍi (Qur’an, 73:20). “and others who travel in the land” (Shakir, 2011, p. 393). bi-mā fataḥa Allāhu ʿalay-kum (Qur’an, 2:76). “of what Allah has disclosed to you” (Shakir, 2011, p. 7).
wa-la yaḍribna bi-arjulihinna (Qur’an, 24:31). “and let them not strike their feet” (Shakir, 2011, p. 230).
fa-itūhunna min ḥaythu amarakum Allāhu (Qur’an, 2:222). “go in to them as Allah has commended you” (Shakir, 2011, p. 21). fa-in lam taʾtūnī bi-hi (Qur’an, 12:60). “But if you do not bring him to me” (Shakir, 2011, p. 152). fa-in atayna bi-fāḥishatin (Qur’an, 4:25). “then if they are guilty of indecency” (Shakir, 2011, p. 51). dhālika adná an yaʾtū bi- al-shahādati ʿalá wajhihā (Qur’an, 5:108). “This is more proper in order that they should give testimony truly” (Shakir, 2011, p. 77). fa-itū bi-hi ʿalá aʿyuni al-nāsi (Qur’an, 21:61). “Then bring him before the eyes of the people” (Shakir, 2011, p. 212). lā yaʾti bi-khayrin (Qur’an, 16:76). “he brings no good” (Shakir, 2011, p. 174). aw taʾtiya bi-llāhi (Qur’an, 17:92). “or bring Allah” (Shakir, 2011, p. 185).
iḍrib bi-ʿaṣaka al-ḥajara (Qur’an, 2:60). “Strike the rock with your staff” (Shakir, p. 5). wa-law fataḥnā ʿalay-him bāban min al-samāʾi (Qur’an, 15:14). “And even if We open to them a gateway of heaven” (Shakir, 2011, p. 165). nazala bi-hi al-rūḥu al-amīnu (Qur’an, 26:193). “The Faithful Spirit has descended with it” (Shakir, 2011, p. 246). aw-min al-layli fa-sjud la-hu (Qur’an, 76:26). “And during part of the night adore Him” (Shakir, 2011, p. 398). fa-aqbala baʿḍuhum ʿalá baʿḍin yatasāʾalūn (Qur’an, 37:50). “Then shall some of them advance to others, questioning each other” (Shakir, 2011, p. 296). idh qalā la-hum akhūhum Nūḥ (Qur’an, 26:106). “When their brother Nuh said to them” (Shakir, 2011, p. 243). lā- yusʾalu ʿan-mā yafʿalu (Qur’an, 21:23). “He cannot be questioned concerning what He does” (Shakir, 2011, p. 210). anzil ʿalay-nā māʾidatan min al-samāʾi (Qur’an, 5:114). “send down to us food from heaven” (Shakir, 2011, p. 77). tarmīhim bi-ḥijāratin min sijjīl (Qur’an, 105:4). “Casting against them stones of baked clay” (Shakir, 2011, p. 421).
68
Qur’anic idiomatic phrasal verbs
to the standpoint that he opts for to explore al-taʿaddī wal-luzūm (verb transitivity and intransitivity). Again, the previous chapter discussed the issue of verb transitivity or intransitivity as an important factor potentially contributing to the idiomaticity of Arabic verb-preposition combinations. Nonetheless, it is not the only factor, as was established earlier. An important point to confirm here is that the focus in this study is on the idiomatic Qur’anic verb-preposition combination, which has a figurative, metaphorical, and nontransparent significance. The other type does not fall within the scope of this study, because its constituents retain their basic, literal, and straightforward meanings. Hence, they pose no problem whatsoever to readers, commentators, and translators of the Qur’an. In distinguishing the idiomatic from the nonidiomatic Qur’anic verb-preposition combinations, I test the idiomaticity as discussed in the previous chapter on each verb-preposition combination in the Qur’an. Table 4.1 illustrates the two types, where the constituent parts of Qur’anic verbpreposition combinations (i.e. verbs and prepositions) are written in bold.
4.2 Syntactic idiosyncrasies of QIPVs Given that the Qur’an is an Arabic text characterized by its incredibly unique and superior qualities, the QIPVs share the syntactic and semantic idiosyncrasies of AIPVs, which were accounted for in the preceding chapter. In other words, QIPVs are Arabic verb-preposition combinations in which the components (i.e. verbs and prepositions) sacrifice their basic, literal, and transparent meanings when they fuse to produce new units of meanings divergent from the sum of the literal meanings of their components. The morphological, syntactic, and semantic features of the Arabic verbs and Arabic prepositions outlined in the previous chapter, the nuance and delicate semantico-syntactic interrelation between them, and the key factors that govern their functionality are all applicable to QIPVs. The only difference between AIPVs and QIPVs is that the latter are characterized by Qur’an-specific features. In this section, I attend to the syntactic Qur’anspecific features of QIPVs. The semantic and pragmatic Qur’an-specific features, however, are attended to in the forthcoming sections. In his book Qur’an Translation: Discourse, Texture and Exegesis, Abdul-Raof (2001) addresses the issue of semantico-syntactic interrelation, which he deems “a prototypical feature of Qur’anic discourse; this where the meaning of Qur’anic structure is signalled through syntactic elements like prepositions” (p. 48). To exemplify this organic interrelation, Abdul-Raof (2001) cites the following two Qur’anic verses: (1) wa-innā aw iyyākum la-ʿalá hudun aw f ī ḍalālin mubīn (Qur’an, 34:24). “And, behold, either we (who believe in Him) or you (who deny His oneness) are on the right path, or have clearly gone astray!” (Asad, 2011, p. 787). This Qur’anic structure has two prepositions—ʿalá and f ī. The change in preposition from the first to the second is semantically motivated to indicate
Qur’anic idiomatic phrasal verbs 69 a change in meaning—in other words, to alert the reader to the semantic change. Each preposition signals a distinct meaning. The first preposition, ʿalá (on), is employed before the word hudá ‘the right path’ to indicate elevated position that suits this emotive Qur’anic expression. This preposition also provides a further associative meaning. Specifically, people who are (on) this high place, which is referred to as ‘the right path’, can see the world better: They have a bird’s-eye view of everything; they are not narrow-minded; and the place they are at is not dark or limited. The second preposition, f ī ‘in’, however, is placed before the word ḍalāl ‘gone astray’ to indicate the low position that suits the world that accompanies it. This preposition also indicates that people ‘in’ these ‘low’ conditions are narrow-minded and cannot think properly, because of the ‘low’ dark place that they are ‘in’ (AbdulRaof, 2001, pp. 48–49). (2) innamā al-ṣadaqātu lil-fuqarāʾi wa- al-masākini wa- al-ʿāmilina ʿalay-hā wa- al-muʾallafati qulubuhum wa- f ī al-riqābi wa- al-ghārimīna wa- f ī sabīli-illhi wa- ibn al-sabīli (Qur’an, 9:60). “Alms are for the poor and the needy, and those employed to administer the (funds); for those whose hearts have been (recently) reconciled (to the Truth); for those in bondage and in debt; in the cause of Allah; and for the wayfarer” (Yūsuf ʿAlī, 1991, p. 456). In this structure, we have been alerted, by the change in prepositions used, to a change in meaning. In this structure are two categories of people who are entitled to alms (charity). The first preposition, li-, is used for the first category of people, whereas for the second category of people, the preposition f ī (in) is used. This change in preposition is meaning oriented; it indicates that the second category of people is more entitled to charity than the first category is. The associative meaning of the Arabic preposition ‘in’ refers to the ‘bowl’ in which charity money is dropped and that this ‘bowl’ is deeper, i.e. a reference to those people’s desperate financial needs (AbdulRaof, 2001, p. 49). These two examples illustrate the functional role of prepositions in signalling subtle aspects of meaning, which profoundly affect the overall signifcances of the Qur’anic structures.
4.2.1 QIPVs’ word order The word order patterns of QIPVs can be classified into two types—nonsplit QIPVs and split QIPVs—depending on the preposition’s location in each Qur’anic verb-preposition construction. The main patterns of the QIPVs are presented in this section. The purpose is not only to exhibit the Qur’an-specific structural aspects of the verb-preposition combinations but also to appreciate how meanings are contributed by Arabic prepositions when collocated with Arabic verbs to form QIPVs.
70
Qur’anic idiomatic phrasal verbs
4.2.1.1 Nonsplit QIPVs In nonsplit QIPVs, the preposition is located immediately after the verb with which it combines. This type of QIPV falls into several syntactic patterns, depending on the verb’s transitivity, tense, voice, and the number of prepositions with which it combines (cf Dāwood, 2002; Aldahesh, 2016b, 2017). The patterns are listed and illustrated in Table 4.2, where the two constituent parts of the QIPVs (i.e. verbs and prepositions) are written in bold. Table 4.2 Nonsplit QIPVs Syntactic pattern
Illustrative example
Transitive past passive verb + preposition + pronoun + preposition + pronoun + agent of the passive Transitive present active verb + preposition + noun + preposition + noun
Kutiba la-hum bi-hi ʿamalun ṣālihun (Qur’an, 9:120). “a good deed is recorded in their favour on account of it” (Abdel Haleem, 2010, p. 207). bal naqdhifu bi-l-ḥaqqi ʿalá al-bāṭili fa-yadmaghahu (Qur’an, 21:18). “No! We hurl the truth against falsehood, and truth obliterates it” (Abdel Haleem, 2010, p. 324). ḍuribat ʿalay-himu al-dhillatu (Qur’an, 3:112). “they are overshadowed by vulnerability” (Abdel Haleem, 2010, p. 65). Kutiba ʿalay-kum idhā ḥaḍara aḥadakum al-mawtu in taraka khayran al-waṣiyatu (Qur’an, 2:180). “When death approached one of you who leaves wealth, it is prescribed that he should make a proper bequest” (Abdel Haleem, 2010, p. 28).
Transitive past passive verb + preposition + pronoun + agent of the passive Transitive Past Passive Verb + Preposition + Pronoun + Conditional Particle + Verbal Sentence + interrogative particle + verbal sentence + agent of the passive Transitive present active verb + preposition + pronoun + adverb of time + governed clause + direct object Transitive past passive verb + preposition + pronoun + nominal sentence Transitive future verb + preposition + noun + governed relative clause + direct object Transitive past active verb + preposition + noun phrase + direct object Intransitive past active verb + preposition + pronoun + preposition + noun phrase + verb agent
fa-lā nuqīmu la-hum yawma al-qiyāmati waznan (Qur’an, 18:105). “On the Day of Resurrection We shall give them no weight” (Abdel Haleem, 2010, p. 305). Kutiba ʿalay-hi annahu man tawallāhu fa-annahu yuḍilluhu (Qur’an, 22:4). “fated to lead astray those who take his side” (Abdel Haleem, 2010, p. 333). sa-nulqī f ī qulūbi al-ladhīna kafarū al-ruʿba (Qur’an, 3:151). “We will strike panic into the disbelievers’ hearts” (Abdel Haleem, 2010, p. 70). Kataba ʿalá nafsihi al-raḥmata (6:12). “He has taken it upon Himself to be merciful” (Abdel Haleem, 2010, p. 130). Qāla qad waqaʿa ʿalay-kum min rabbikum rijsun wa-ghaḍabun (Qur’an, 7:71). “He said, ‘You are already set to receive your Lord’s loathing and anger’” (Abdel Haleem, 2010, p. 160).
Qur’anic idiomatic phrasal verbs 71 Intransitive present active verb + preposition + noun + preposition + noun
Intransitive present passive verb + preposition + pronoun Intransitive present active verb + preposition + noun Intransitive present passive verb + preposition + pronoun + agent of the passive Intransitive past active verb + preposition + noun
Intransitive past passive verb + preposition + noun phrase Intransitive imperative verb + preposition + noun phrase Transitive imperative verb + preposition + pronoun + preposition + noun phrase Transitive past active verb + preposition + noun
Transitive past passive verb + preposition + pronoun + agent of the passive Intransitive future verb + preposition + pronoun Transitive imperative verb + preposition + pronoun + direct object
Qāla Ibrahimu fa-inna Allaha yaʾti bi-al-shamsi min al-mashriqi fa-ʾti bi-ha min al-maghribi (Qur’an, 2:258). “Abraham said, ‘God brings the sun from the east; so bring it from the west’” (Abdel Haleem, 2010, p. 44). illā an yuḥaṭa bi-kum (Qur’an, 12: 66). “not unless you swear by God that you will bring him back to me if that is humanly possible” (Abdel Haleem, 2010, p. 244). yakādu sanā barqihi yadhhabu bi- al-abṣāri (Qur’an, 24:43). “the flash of its lightning almost snatches sight away” (Abdel Haleem, 2010, p. 356). yuʾfaku ʿan-hu man ʾufik (Qur’an, 51:9). “those who turn away from it are (truly) deceived” (Abdel Haleem, 2010, p. 522). huwa al-ladhī khalaqa la-kum mā f ī al-arḍi jamīʿan thumma istawá ilá al-samāʾi (Qur’an, 2:29). “It was He who created all that is on the earth for you, then turned to the sky” (Abdel Haleem, 2010, p. 6). wa-lammā suqiṭa f ī aydīhim (Qur’an, 7:149). “When, with much wringing of hands” (Abdel Haleem, 2010, p. 169). irkuḍ bi-rijlika hadhā mughtasalun bāridun wa-sharābun (Qur’an, 38:42). “Stam your foot! Here is cool water for you to wash in and drink” (Abdel Haleem, 2010, p. 456). wa-ajlib ʿalay-him bi-khaylika wa-rajilika (Qur’an, 17:64). “muster your cavalry and infantry against them” (Abdel Haleem, 2010, p. 289). wa-idh qulnā la-ka inna rabbaka aḥāṭa bi-al-nāsi (Qur’an, 17:60). “(Prophet), We have told you that your Lord knows all about human beings” (Abdel Haleem, 2010, p. 289). wa-ḍuribat ʿalay-himu al-dhillatu wa- al-maskantu (Qur’an, 2:61). “They were struck with humiliation and wretchedness” (Abdel Haleem, 2010, p. 10). sa-nafrughu la-kum ayyuhā al-thaqalāni (Qur’an, 55:31). “We shall attend to you two huge armies (of jinn and [humankind])” (Abdel Haleem, 2010, p. 533). an asri bi-ʿibadī fa-iḍrib la-hum ṭarīqan f ī al-baḥri yabasan (Qur’an, 20:77). “Go out at night with My servants and strike a dry path for them across the sea” (Abdel Haleem, 2010, p. 318). (Continued)
72
Qur’anic idiomatic phrasal verbs
Table 4.2 (Continued) Syntactic pattern
Illustrative example
Negative interrogation + transitive present active verb + preposition + pronoun Conditional particle + intransitive past passive verb + preposition + nominal sentence
alam nashraḥ la-ka ṣadraka (Qur’an, 94:1). “Did we not relieve your heart for you” (Abdel Haleem, 2010, p. 597). fa-in ʿuthira ʿalá annahumā istaḥaqqā ithman fa-ākharāni yaqūmāni maqāmahumā (Qur’an, 5:107). “If it is discovered that these two are guilty (of perjury), two of those whose rights have been usurped have a better right to bear witness in their place” (Abdel Haleem, 2010, p. 126).
4.2.1.2 Split QIPVs In split QIPVs, the preposition is not located immediately after the verb with which it combines. Rather, it is separated by one or more constituents. The split type falls into a number of syntactic patterns, depending on the verb type (i.e. transitivity, tense, and voice), the number of constituents, and the number of prepositions that follow the verb (cf Dāwood, 2002; Aldahesh, 2016b, 2017). These patterns are listed and illustrated in Table 4.3, where the verbs and prepositions are written in bold. That several other syntactic patterns of verb-preposition combinations appear in the Qur’an is telling. Nevertheless, I have excluded them because they are nonidiomatic combinations and thus irrelevant to the present study. In the Qur’anic discourse, seeing Arabic prepositions come before the verbs with which they combine is not uncommon. This point is illustrated by the following examples, where the prepositions and the verbs are written in bold: (1) al-Raḥmānu ʿalá al-ʿarshi istawá (Qur’an, 20:5). “the most Gracious, established on the throne of His almightiness” (Asad, 2011, p. 562). (2) in kuntum li- al-ruʾyā taʿburūn (Qur’an, 12:43). “if you are able to interpret dreams!” (Asad, 2011, p. 413). (3) wa-ilay-hi tuqalibūn (Qur’an, 29:21). “and unto Him you shall be made to return” (Asad, 2011, p. 727). (4) wa-ilá rabbika fa-irghab (Qur’an, 94:8). “and unto thy Sustainer turn with love” (Asad, 2011, p. 1147). (5) fa-bi-ma tubashshirūn (Qur’an, 15:54). “Of what (strange thing), then, are you giving me a tiding!” (Asad, 2011, p. 464).
Qur’anic idiomatic phrasal verbs 73 Table 4.3 Split QIPVs Syntactic pattern
Illustrative example
Transitive imperative verb + direct object + relative clause + preposition + noun Intransitive past active verb + verb subject + preposition + pronoun + preposition + particle + verbal sentence Intransitive past passive verb + adverb of time + pronoun + preposition + noun Intransitive present active verb + verb subject + adjective + exception particle + preposition + noun phrase Intransitive present active verb + agent + preposition + noun phrase
wa-bashshir al-ladhīna kafarū bi-ʿadhabin alimīn (Qur’an, 9:3). “And proclaim a grievous penalty to those who reject Faith” (Yūsuf ʿAlī, 1991, p. 437). wa-waqaʿa al-qawlu ʿalay-him bi-mā ẓalamū (Qur’an, 27:85). “And the Word will be fulfilled against them, because of their wrongdoing” (Yūsuf ʿAlī, 1991, p. 957). fa-ḍuriba baynahum bi-sūrin la-hu bābun (Qur’an, 57:13). “So a wall will be put up betwixt them, with a gate therein” (Yūsuf ʿAlī, 1991, p. 1423). wa-lā- yaḥīqu al-makru al-sayiʾu illā bi-ahlihi (Qur’an, 35:43). “But the plotting of Evil will hem in only the authors thereof” (Yūsuf ʿAlī, 1991, p. 1114).
Intransitive present active verb + agent + direct object + preposition + pronoun Negative particle + transitive present active verb + agent + preposition + noun phrase + preposition + noun phrase Transitive past active verb + agent + direct object + preposition + noun Transitive past active verb + agent + preposition + noun phrase Transitive present active verb + agent + direct object + preposition + noun
Qālū in hādhāni la-sāḥirāni yurīdāni an yukhrijākum min arḍikum bi-siḥrihimā wa-yadhhabā bi-ṭarīqatukum al-muthlá (Qur’an, 20:63). “Saying, ‘These two men are sorcerers. Their purpose is to drive you out of your land with their sorcery and put an end to your timehonoured way of life’” (Abdel Haleem, 2010, p. 316). fa-lā-tadhheb nafsuka ʿalay-him ḥasarāt (Qur’an, 35:8). “do not waste your soul away with regret for them” (Abdel Haleem, 2010, p. 436). wa-lā-yarghabū bi-anfusihim ʿan nafsihi (Qur’an, 9:120). “nor should they have cared about themselves more than him” (Abdel Haleem, 2010, p. 207). ḥattá idhā akhdhnā mutraf īhim bi-alʿadhābi idhā hum yajʾarūn (Qur’an, 23:64). “When We bring Our punishment on those corrupted with wealth, they will cry for help” (Abdel Haleem, 2010, p. 347). fa-ḍarabnā ʿalá ādhānihim f ī al-khafi sinīna ʿadadā (Qur’an, 18:11). “We sealed their ears (with sleep) in the cave for years” (Abdel Haleem, 2010, p. 295). wa-yaḍribu Allāhu al-amthāla li-al-nāsi laʿallahum yatadhakkarūn (Qur’an, 14:25). “God makes such comparisons for people so that they may reflect” (Abdel Haleem, 2010, p. 260). (Continued)
74
Qur’anic idiomatic phrasal verbs
Table 4.3 (Continued) Syntactic pattern
Illustrative example
Transitive past active verb + agent + direct object + preposition + noun phrase
jāʾat-hum rusuluhum bi-al-bayināti fa-raddū aydiyahum f ī afwāhihim (Qur’an, 14:9). “Their messengers came to them with clear proof, but they tried to silence them” (Abdel Haleem, 2010, p. 257). fa-istaqīmū ilay-hi wa-istaghfirūhu (Qur’an, 41:6). “Take the straight path to Him and seek His forgiveness” (Abdel Haleem, 2010, p. 478). wa-law tará idh wuqifu ʿalá al-nāri (Qur’an, 6:27). “If you could only see, when they are made to stand before the Fire” (Abdel Haleem, 2010, p. 131). wa-aqīmū al-wazna bi- alqisṭi wa-la-tukhsirū al-mizān (Qur’an, 55:9). “weight with justice and do not fall short in the balance” (Abdel Haleem, 2010, p. 532). an iqdhif īhi f ī al-tābūti fa-iqdhif īhi f ī al-yammi (Qur’an, 20:39). “Put your child into the chest, then place him in the river” (Abdel Haleem, 2010, p. 315). wa-itaqū yawman lā-tujzī nafsun ʿan nafsin shayʾan (Qur’an, 2:48). “Guard yourselves against a Day when no soul will replace another in any way” (Abdel Haleem, 2010, p. 8). wa-lā-taʾkilū amwālakum ilá amwālihim (Qur’an, 4:2). “and do not consume their property along with your own” (Abdel Haleem, 2010, p. 78). wa-lammā ḍuriba ibnu maryama mathalan idhā qawmuka min-hu yaṣiddūn (Qur’an, 43:57). “When the son of Mary is cited as an example, your [the Prophet’s] people laugh and jeer” (Abdel Haleem, 2010, p. 494). f ī-hā yufraqu kullu amrin ḥakīm (Qur’an, 44:4). “a night when every matter of wisdom was made distinct” (Abdel Haleem, 2010, p. 497). ʿalá al-nāri yuftanūn (Qur’an, 51:13). “On a Day when they will be punished by the Fire” (Abdel Haleem, 2010, p. 522).
Intransitive imperative verb + agent + preposition + pronoun Transitive past passive verb + agent + preposition + noun Transitive imperative verb + agent + direct object + preposition + noun Transitive imperative verb + direct object + preposition + noun Negative particle + transitive present active verb + agent + preposition + noun Negative particle + transitive present active verb + agent + direct object + preposition + noun phrase Preposition + pronoun + verbal sentence
Preposition + pronoun + present passive verb + agent of the passive Preposition + noun + verbal sentence
Furthermore, in Qur’anic discourse, seeing Arabic verbs combine with more than one preposition is also not uncommon. The following examples illustrate this, where the verbs and the prepositions are written in bold: (1) lawlā yaʾtūna ʿalay-him bi-sulṭānin bayyn (Qur’an, 18:15). “without being able to adduce any reasonable evidence in support of their beliefs” (Asad, 2011, p. 527).
Qur’anic idiomatic phrasal verbs 75 (2) lawlā jāʾū ʿalay-hi bi-arbaʿati shuhadāʾ (Qur’an, 24:13). “Why do they not (demand of the accusers that they) produce four witnesses to prove their allegation?” (Asad, 2011, p. 639). (3) fa-kayfa idhā jiʾnā min kulli ummatin bi-shāhidin wa-jiʾnā bi-ka ʿalá hāʾulāʾi shahidā (Qur’an, 4:41). “How, then, (will the sinners fare on Judgement Day) when We shall bring forward witnesses from within every community, and bring thee (O Prophet) witness against them?” (Asad, 2011, p. 134). Scholars, who investigated these patterns in the Qur’an, although they confused idiomatic combinations with nonidiomatic combinations, have come up with a number of valuable statistical insights that merit being taken into our consideration here: (1) Arabic prepositions combine with transitive verbs much more than they do with intransitive verbs throughout the Qur’an; (2) Patterns of most of the intransitive verb + preposition combinations are nonsplit, whereas most of the transitive verb + preposition combinations are split patterns; (3) Overall, the split patterns and the nonsplit patterns of verb-preposition combinations in the Qur’an are equal in number; (4) The patterns of verb + one preposition outnumber the patterns of verb + two prepositions; (5) The patterns of verb + three prepositions are quite rare (only three combinations in the whole scripture) (cf Dāwood, 2002, p. 1/67–68). The three Qur’anic verses in which the pattern of verb + three prepositions is used are Qur’an 24:36; Qur’an, 17:1; and Qur’an, 26:193–194. All combinations are nonidiomatic and therefore have been excluded from this study. Two types of Qur’an-specific features of the verb-preposition combinations are related to the diversity of the prepositions with which the verb may combine (cf Dāwood, 2002): (1) Vertical diversity whereby a given verb may combine with different prepositions in different Qur’anic contexts. For example, (i) wa-idhā mā unzilat sūratun naẓara baʿḍuhum ilá baʿḍin (Qur’an, 9:127) “Whenever there cometh down a Sūrah, they look at each other” (Yūsuf ʿAlī, 1991, p. 475); (ii) fa-naẓara naẓratan f ī al-nujūmi (Qur’an, 37:88) “Then did he cast A glance at the Stars” (Yūsuf ʿAlī, 1991, p. 1147);
76
Qur’anic idiomatic phrasal verbs (iii) yanẓurūna min ṭarfin khafiyin (Qur’an, 42:45) “looking with a stealthy glance” (Yūsuf ʿAlī, 1991, p. 1259).
In the foregoing three examples, the verb naẓara has combined with three different prepositions in three Qur’anic contexts. (2) Horizontal diversity whereby a given verb may combine with different prepositions in the same Qur’anic context. For example, subḥāna al-ladhī asrá bi-ʿabdihi laylan min al-masjid al-ḥarāmi ilá al-masjid al-aqṣá (Qur’an, 17:1) “Glory to (Allah) who did take His servant for a Journey by night from the Sacred Mosque to the Farthest Mosque” (Yūsuf ʿAlī, 1991, p. 673). As can be seen, the verb asrá has combined with three different prepositions in a single Qur’anic context (cf Dāwood, 2002, p. 1/8).
4.3 Semantic idiosyncrasies of the QIPVs Like the AIPVs, the QIPVs are an outcome of combining ḥurūf al-jarr ‘prepositions’ with al-afʿāl ‘proper verbs’. Prepositions notably transform the basic significances of the verbs in such a combination. Replacing a given preposition with another would trigger a semantic modification in the overall significance of the whole combination. This semantic interrelation between verbs and prepositions determines, to a certain extent, the degree of QIPV idiomaticity. The following examples suffice to illustrate this point: (1) wa-idhā ḍarabtum f ī al-arḍi (Qur’an, 4:101). “When you (believers) are travelling in the land” (Abdel Haleem, 2010, p. 95). The preposition f ī in this Qur’anic context has changed the original signifcance of the verb ḍaraba (to blow, to hit) into a new signifcance—that is, ‘to walk, to travel’. (2) wa-man yarghabu ʿan millati Ibrāhīm (Qur’an, 2:130). “Who but a fool would forsake the religion of Abraham?” (Abdel Haleem, 2010, p. 21). The preposition ʿan in this verse has changed the original signifcance of the verb raghiba (to like something and seek it) into a new signifcance—that is, ‘to dislike something and go away from it’. (3) bi-mā fataḥa Allāhu ʿalay-kum (Qur’an, 2:76). “of what Allah has disclosed to you” (Shakir, 2011, p. 7).
Qur’anic idiomatic phrasal verbs 77 The preposition ʿalá in this verse has transferred the original signifcance of the verb fataḥa (to open) to a new signifcance—that is, ‘to disclose something to someone’. (4) fa-in atayna bi-fāḥishatin (Qur’an, 4:25). “then if they are guilty of indecency” (Shakir, 2011, p. 51). The preposition bi- in this verse has shifted the original signifcance of the verb atá (to come) to a new signifcance—that is, ‘to be guilty of doing something’. (5) yakādu sanā barqihi yadhhabu bi-al-abṣāri (Qur’an, 24:43). “the flash of its lightning almost snatches sight away” (Abdel Haleem, 2010, p. 356). The preposition bi- in this Qur’anic context has changed the original signifcance of the verb yadhhab (to go) into a new signifcance—that is, ‘to snatch something away’. (6) wa-idh qulnā la-ka inna rabbaka aḥāṭa bi-al-nāsi (Qur’an, 17:60). “(Prophet), We have told you that your Lord knows all about human beings” (Abdel Haleem, 2010, p. 289). The preposition bi- in this structure has shifted the original signifcance of the verb aḥāṭa (to surround, to frame) to a new signifcance—that is, ‘to know all about something’. For more examples, see Dāwood, 2002 (p. 1/6). As stated in the previous chapter, Arabic classical linguists in their treatment of prepositions did not account for the semantic interaction between combined prepositions and verbs. Yet modern linguists to have studied the verb-preposition combinations in the Qur’an arrived at a sort of regularity between the prepositions and verbs types with which they tend to combine throughout the Qur’anic discourse. They declare that in the Qur’an, the preposition ʿalá typically collocates with verbs to signify al-inʿām wa- al-tafaḍḍul (giving and bestowing favour). For example, laqad manna Allāhu ʿalá al-Muʾminīn (Qur’an, 3:164). “God has been gracious to the believers” (Abdel Haleem, 2010, p. 72). Arabic classical linguists also assert that the preposition ʿan collocates with verbs to signify al-tajāwuz wa-al-ṣaf ḥ wa-al-musāmaḥah (pardon, amnesty, and forgiveness). For example, afanaḍribu ʿan-kumu al-dhikra ṣafḥan in kuntum qawman musrif īn (Qur’an, 43:5).
78
Qur’anic idiomatic phrasal verbs “Should We ignore you and turn this revelation away from you because you are an insolent people?” (Abdel Haleem, 2010, p. 490).
Moreover, the prepositions min and ilá collocate with verbs to signify motion in time and place. For example, wa-ilá rabbika farghab (Qur’an, 94:8). “and direct your requests to your Lord” (Abdel Haleem, 2010, p. 597; cf Dāwood, 2002; Aldahesh, 2016a, 2017).
4.3.1 QIPVs’ semantic fields Restricting the use of QIPVs in Qur’anic discourse to a limited number of semantic fields is difficult. This is mainly due to the nature of the Qur’an as a divine book of guidance that deals with a wide range of human activities, thoughts, feelings, beliefs, and connections with the surrounding world, let alone the unseen world. Thus, the QIPVs have been employed in diverse semantic fields throughout the Qur’an (cf Dāwood, 2002, p. 1/71–72; Aldahesh, 2016b, pp. 36–38). Table 4.4 lists and illustrates the major semantic fields of the QIPVs. The semantic field of MOVMENT in its concrete and abstract dimensions constitutes the largest semantic field of QIPVs. Arabic verbs denoting movement, derived from a variety of Arabic roots, combine with almost all Arabic authentic prepositions to deliver a wide range of movement denotations through the Qur’anic discourse (cf Dāwood, 2002, p. 1/75–78). The Arabic verbs derived from the root (ʾ-T-Y) are the most common in this regard. Other Arabic roots from which the verbs denoting movement are derived include (ʾ-KH-DH), (ḤW-Ṭ), (D-KH-L), (DH-H-B), (Ḍ-R-B), (Q-DH-F), (S-L-KH), and (S-Q-Ṭ), among others (see Appendix I for the occurrences of these Arabic roots in the Qur’an, and sees Appendix II for their frequencies and derivatives).
4.3.2 QIPVs’ semantic categories Before we proceed to the pragmatic properties of the QIPVs, an essential point needs to be made clear here. A thorough semantic analysis of the QIPVs reveals that they can be classified into two distinct categories according to their degree of idiomaticity: metaphorical QIPVs and figurative QIPVs (cf Aldahesh, 2016b). The difference between these two categories is covered in 4.3.2.1.
4.3.2.1 Metaphorical QIPVs One can safely declare that the majority of the QIPVs are metaphorical in their semantic nature. By metaphorical, I mean that the Qur’anic verb-prepositions combinations fall under Nida’s and Newmark’s definitions of ‘metaphor’. Nida (1975) defines metaphor as “a figurative expression used to make an explicit
Qur’anic idiomatic phrasal verbs 79 Table 4.4 Major semantic fields of the QIPVs Semantic field
Illustrative example
MOVEMENT (concrete)
wa-qālū li-ikhwānihim idhā ḍarabu fī al-arḍi (Qur’an, 3:156). “and said of their brothers who went out on a journey or raid” (Abdel Haleem, 2010, p. 71). lā-yaʾtīhi al-bāṭilu min bayni yadayhi wa-lā min khalfihi (Qur’an, 41:42). “which falsehood cannot touch from any angle” (Abdel Haleem, 2010, p. 482). fa-idhā dhahaba al-khawfu salaqūkum bi-alsinatin ḥidād (Qur’an, 33:19). “when fear has passed, they attack you with sharp tongues” (Abdel Haleem, 2010, p. 421). wa-mathalu al-ladhīna kafarū ka-mathali al-ladhī yanʿiqu bi-mā lā-yasmaʿu illā duʿāan wa-nidāan (Qur’an, 2:171). “Calling to disbelievers is like a herdsman calling to things that hear nothing but a shout and cry” (Abdel Haleem, 2010, p. 27). alam tara ilá al-ladhī ḥājja Ibrāhīma f ī rabbihi ʾan ātāhu Allāhu al-mulka (Qur’an, 2:258). “(Prophet), have you not thought about the man who disputed God had given him power to rule?” (Abdel Haleem, 2010, p. 44). wa-kadhālika aʿtharnā ʿalay-him li-yaʿlamu ʾanna waʿda Allāhi haqqun (Qur’an, 18:21). “In this way We brought them to people’s attention so that they might know that God’s promise (of resurrection) is true” (Abdel Haleem, 2010, p. 297). alam nashraḥ la-ka ṣadrak (Qur’an, 94:1). “Did we not relieve your heart for you” (Abdel Haleem, 2010, p. 597). wa-man yakfur bi-al-imāni fa-qad ḥabiṭa ʿamaluhu wa-huwa f ī al-ākhirati min al-khāsirīn (Qur’an, 5:5). “The deeds of anyone who rejects [the obligations of] faith will come to nothing, and in the Hereafter he will be one of the losers” (Abdel Haleem, 2010, p. 108). yāʾayuhā al-ladhīna āmanū kutiba ʿalay-kum al-ṣiyāmu ka-mā kutiba ʿalá al-ladhīna min qablikum laʿallakum tataqūn (Qur’an, 2:183). “You who believe, fasting is prescribed for you, as it was prescribed for those before you, so that you may be mindful of God” (Abdel Haleem, 2010, p. 29). wa-mā naqamū min-hum illā ʾan yuʾminū bi-llāhi al-ʿazīzi al-ḥamid (Qur’an, 85:8). “Their only grievance against them was their faith in God, the Mighty, the Praise worthy” (Abdel Haleem, 2010, p. 591). wa-man ʾaẓlamu min-man manaʿa masājida Allāhi ʾan yudhkara f ī-ha ismuhu wa-saʿá f ī khrābiha (Qur’an, 2:114). “Who could be more wicked than those who prohibit the mention of God’s name in His places of worship and strive to have them deserted?” (Abdel Haleem, 2010, p. 19).
MOVEMENT (abstract) SPEECH
NATURAL SOUNDS
COGNITION (mental)
COGNITION (physical)
FEELINGS DISOBEDIENCE
OBLIGATION
OBJECTION
ENDEAVOUR
(Continued)
80
Qur’anic idiomatic phrasal verbs
Table 4.4 (Continued) Semantic field
Illustrative example
DISCLOSURE
wa-ʾaṣbaḥa fuʾadu ummi mosá fārighan wa-in kādat la-tubdya bi-hi lawlá ʾan rabaṭnā ʿalá qalbihā li-takuna min al-muʾminīn (Qur’an, 28:10). “The next day, Moses’ mother felt a void in her heart—if We had not strengthened it to make her one of those who believe, she would have revealed everything about him” (Abdel Haleem, 2010, p. 387). khatama Allāhu ʿalá qulūbihim wa-ʿalá samʿihim wa-abṣārihim ghishāwatun wa-la-hum ʿadhābun ʿaẓīm (Qur’an, 2:7). “God has sealed their hearts and their ears, and their eyes are covered. They will have a great torment” (Abdel Haleem, 2010, p. 4). ahāʾulāʾi al-ladhīna aqsamtum lā-yanāluhum Allāhu birahmatihi (Qur’an, 7:49). “And are these the people you swore God would never bless?” (Abdel Haleem, 2010, p. 157). wa-kayfa taʾkhudhūnahu wa-qad afḍá baʿḍukum ilá baʿḍin wa-akhdhna ʿalay-kum mithāqan ghlīẓa (Qur’an, 4:21). “How could you take it when you have lain with each other and they have taken a solemn pledge from you?” (Abdel Haleem, 2010, p. 82). wa-man yaʿshu ʿan dhikri al-raḥmāni nuqayiḍ la-hu shayṭānan fa-huwa la-hu qarīn (Qur’an, 43:36). “We assign an evil one as a comrade for whoever turns away from the revelations of the Lord of Mercy” (Abdel Haleem, 2010, p. 493). inna al-ladhīna yaghḍūna aṣwātahum ʿinda rasūli illāhi ʾulāʾika al-ladhīna imtaḥana Allāhu qulūbahum li-altaqwá (Qur’an, 49:3). “It is those who lower their voices in the presence of God’s Messenger whose hearts God has proved to be aware” (Abdel Haleem, 2010, p. 516). wa-lā-taʾkulū amwālahum ilá amwālikum (Qur’an, 4:2). “and do not consume their property along with your own” (Abdel Haleem, 2010, p. 78). wa-in kabura ʿalayka iʿrāḍuhum fa-in istaṭaʿta ʾan tabtaghya nafaqan f ī al-arḍi aw sullaman f ī al-samāʾi fa-taʾtyahum bi-aytin (Qur’an, 6:35). “If you find rejection by the disbelievers so hard to bear, then seek a tunnel into the ground or a ladder into the sky, if you can, and bring them a sign” (Abdel Haleem, 2010, p. 179). wa-li-yarbiṭa ʿalá qulūbikum wa-yuthabbita aqdāmakum (Qur’an, 8:11). “to make your hearts strong and your feet firm” (Abdel Haleem, 2010, p. 179). Qulnā iḥmil f ī-hā min kulli zawjayni ithnayni wa-ahlaka illā man sabaqa ʿalay-hi al-qawl (Qur’an, 11:40). “We said, Place on board this Ark a pair of each species, and your own family—except those against whom the sentence has already been passed” (Abdel Haleem, 2010, p. 227).
CONCEALING
PREVENTION
HUMAN RELATIONS
ASTRAY
PREPARATION
CONSUMING DIFFICULTY
SUPPORT
DENSITY
Qur’anic idiomatic phrasal verbs 81 comparison between items” (p. 231). Along these lines, Newmark (1988) delineates metaphor as any figurative expression: the transferred sense of a physical word. . . . the personification of an abstraction. . . . the application of a word or collocation to what it does not literally denote, i.e. to describe one thing in terms of another. All polysemous words . . . and most English phrasal verbs . . . are potentially metaphorical. (p. 104) In saying that “most English phrasal verbs . . . are potentially metaphorical”, Newmark excludes the EPVs that denote literally and those that do not satisfy the criteria of metaphor. This is quite relevant to QIPVs, as explained later on. The essence of the comparison between two items expressed metaphorically is the similarity between the items. That give the sense that one item is transferred to express the sense of the other because of a similarity between them (El-Zeiny, 2011, p. 248). Moreover, the metaphorical QIPVs have the three key ‘metaphor’ components, proposed by Newmark (1988): object, image, and sense. The object refers to “what is described or qualified by the metaphor” (p. 105); the image refers to “the picture conjured up by the metaphor, which may be universal . . . cultural . . . or individual” (p. 105); and the sense refers to “the literal meaning of the metaphor; the resemblance or the semantic area overlapping object and image; usually this consists of more than one sense component—otherwise literal language would do” (p. 105). The last component (i.e. sense) constitutes the similarity between the object and the image (El-Zeiny, 2011, p. 248). Furthermore, the metaphorical QIPVs satisfy the two purposes of metaphor identified by Newmark (1988): referential and pragmatic. The referential purpose of metaphor, according to Newmark, is a cognitive one, and it is “to describe a mental process or state, a concept, a person, an object, a quality or an action more comprehensively and concisely than is possible in literal or physical language” (p. 104). The pragmatic purpose of metaphor, however, is an aesthetic one, and it is “to appeal to the senses, to interest, to clarify ‘graphically’, to please, to delight, to surprise” (p. 104). This being said, the metaphorical QIPVs can be considered, to use Newmark’s term, a stoke or standard type of metaphor referring to “an efficient and concise method of covering a physical and/or mental situation both referentially and pragmatically” (Newmark, 1988, p. 108). The metaphorical QIPVs are exemplified in the following Qur’anic structures: (1) wa-āyatun la-humu al-laylu naslakhu min-hu al-nahāra fa-idhā hum muẓlimūn (Qur’an, 36:37). “The night is also a sign for them: We strip the daylight from it, and—lo and behold!-they are in darkness” (Abdel Haleem, 2010, p. 443).
82
Qur’anic idiomatic phrasal verbs
The metaphorical QIPV used in this verse is the verb-preposition combination salakha min. The object of the metaphor here is presenting the alternating of the day and night as one of God’s signs. The image of the metaphor is stripping off the skin. The sense of the metaphor is extracting the day from the night. (2) khatama Allāhu ʿalá qulubihim wa-ʿalá samʿihim wa-abṣārihim ghishāwatun wa-la-hum ʿathābun ʿaẓīm (Qur’an, 2:7). “God has sealed their hearts and their ears, and their eyes are covered. They will have great torment” (Abdel Haleem, 2010, p. 4). The metaphorical QIPV used in this verse is the verb-preposition combination khatama ʿalá. The object of the metaphor is the denying hearts of disbelievers. The image of the metaphor is closing hearts by a seal. The sense of the metaphor is refusing to observe the truth. (3) kadhālika wa-qad aḥaṭnā bi-mā ladayhi khubrā (Qur’an, 18:91). “And so it was: We knew all about him” (Abdel Haleem, 2010, p. 304). The metaphorical QIPV used in this verse is the verb-preposition combination aḥāṭa bi-. The object of the metaphor is confrming the thorough knowledge of something. The image of the metaphor is the surroundings or fencing. The sense of the metaphor is profoundly knowing or fully understanding the subject matter. (4) bal naqdhifu bi-l-ḥaqqi ʿalá al-bāṭili fa-yadmaghahu (Qur’an, 21:18). “No! We hurl the truth against falsehood, and truth obliterates it” (Abdel Haleem, 2010, p. 324). The metaphorical QIPV used in this verse is the verb-preposition combination qadhafa bi-. The object of the metaphor is God’s will to make the truth dominant. The image of the metaphor is throwing the truth from its high and exalted position onto the falsehood in its lower and inferior position. The sense of the metaphor is enabling the truth to prevail.
4.3.2.2 Figurative QIPVs Figurative QIPVs are those that do not fit into the category of metaphor, because they do not have the aforementioned components of a metaphor (i.e. object, image, and sense). Nonetheless, they remain semantically nontransparent in that their intended meanings cannot be deduced from the total meanings of their constituent parts (i.e. verb and preposition) (cf Aldahesh, 2016b). The figurative QIPVs are exemplified in the following Qur’anic structures: (1) sa-nafrughu la-kum ayyuha al-thaqalān (Qur’an, 55:31). “We shall attend to you two huge armies [of jinn and humankind]” (Abdel Haleem, 2010, p. 533).
Qur’anic idiomatic phrasal verbs 83 (2) lā yaʾti bi-khayrin (Qur’an, 16:76). “he brings no good” (Shakir, 2011, p. 174). (3) yakādu sanā barqihi yadhhabu bi- al-abṣāri (Qur’an, 24:43). “the flash of its lightning almost snatches sight away” (Abdel Haleem, 2010, p. 356). (4) irkuḍ bi-rijlika hadhā mughtasalun bāridun wa-sharābun (Qur’an, 38:42). “Stamp your foot! Here is cool water for you to wash in and drink” (Abdel Haleem, 2010, p. 456). The fgurative QIPVs used in these verses are the verb-preposition combinations faragha li, atá bi-, dhahaba bi-, and rakaḍ bi- respectively. These QIPVs lack the three metaphor components, yet semantically speaking, they are nontransparent combinations. This is because their intended meanings cannot be inferred from the total meanings of their constituent parts.
4.4 Pragmatic idiosyncrasies of the QIPVs Looking at the phenomenon of QIPVs from a pragmatic perspective requires us to draw on several relevant pragmatics theories. Contrary to semantics, which focuses on the literal, propositional, and linguistically encoded aspects of meaning, pragmatics is the study of language in use, with particular focus on the functional communicative aspects of meaning. It is the study of “the way utterances are used in communicative situations and the way we interpret them in context. [It is] the study of meaning not as generated by linguistic system but as conveyed and manipulated by participants in a communicative situation” (Baker, 1992, p. 217). In determining the quality of meaning, pragmatics considers both linguistic/textual and extralinguistic/contextual peculiarities. Thus, the main interest in pragmatics is the “relations between language and context that are grammaticalized or encoded in the structure of a language [i.e.] the interrelation of language structure and principles of language usage” (Levinson, 1983, p. 9). Pragmatic studies scrutinize how meaning is produced by the addresser(s) and how it is subsequently understood by the addressee(s) in a specific situational context. Also, the field has grown and flourished by proposing speech act theory. In 1962, J. L. Austin introduced speech act theory in his book How to Do Things with Words. The theory was later advanced by Searle (1969) in his book Speech Acts: An Essay in the Philosophy of Language. The two language philosophers attempted “to go beyond the literal meaning of words and structures by classifying utterances according to their implicit rather than explicit functions” (Baker, 1992, p. 259). Austin (1975) indicates that when people say words, they perform speech acts, such as COMMAND, PROMISE, REQUEST, and APOLOGY. We perform these acts “when, for example, we make a complaint or a request, apologize or pay someone a compliment” (Hatim, 2001, p. 179). Therefore, speech acts for
84
Qur’anic idiomatic phrasal verbs
Austin (1975) are “the total situation in which the utterance is issued” (p. 52). Further, he classifies utterances into two types: constative and performative. Constative utterances refer to statements that may be judged as true or false (p. 3), whereas performative utterances refer to statements that cannot be judged true or false. Austin indicates, however, that most utterances are performative in type (pp. 4–7). Most importantly, Austin (1975) identifies three correlated speech acts dimensions: (1) Locutionary; (2) Illocutionary; (3) Perlocutionary. Austin delineates locutionary acts as performing something in the “full normal sense” (1975, p. 94). It is, he elaborates, “roughly equivalent to uttering a certain sentence with a certain sense and reference, which again is roughly equivalent to ‘meaning’ in the traditional sense” (1975, p. 109). Conversely, Austin defnes illocutionary acts as “performance of an act in saying something as opposed to performance of an act of saying something” (1975, pp. 99–100, emphasis in original). He illustrates this dimension in “informing, ordering, warning, undertaking . . . i.e. utterances which have a certain (conventional) force” (1975, p. 109). In other words, the meaning “has to do with the speaker’s intentions rather than his/her actual words” (Baker, 1992, p. 259). Furthermore, Austin delineates perlocutionary acts as “what we bring about or achieve by saying something, such as convincing, persuading, deterring, and even, say, surprising or misleading” (1975, p. 109, emphasis in original). The three act dimensions—locutionary, illocutionary, and perlocutionary—are concurrent features in any given speech act, where the locutionary signifies the utterance itself, the illocutionary signifies the intended communicative meaning behind the utterance, and the perlocutionary signifies the impact of the utterance on the addressee (Hale, 2004). Scholars refer to these three utterance dimensions as sense, force, and effect, respectively (Marogy, 2010; Hatim, 2001). The concurrent aspect of these dimensions is summarized as follows: Speakers utter things (utterance act), and in uttering things they often say things (locutionary act), and in saying things they often do things (illocutionary act). And typically, speakers bother with all this because they want to communicate something to a hearer, and even have some effect on the thought and/or action of that hearer (perlocutionary act). (Harnish, 2010, p. 6) Further, Grice (1975) draws on speech act theory to propose the notion of the cooperative principle (henceforth CP) and its maxims. Grice (1975) asserts that participants (i.e. addresser and addressee) in any given communicative situation must adhere to a principle and a set of maxims to accomplish certain communicative purposes (cf James, 1980; Bell, 1991; Baker, 1992; Aldahesh, 2018). The
Qur’anic idiomatic phrasal verbs 85 CP participants adhere to is formulated by Grice (1975) as follows: “Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged” (p. 45). Grice (1975) classifes this principle into four categories, each with several maxims. The principle’s categories and their maxims are listed as follows: (1) Quantity (i) Make your contribution as informative as is required (for the current purposes of the exchange); (ii) Do not make your contribution more informative than is required. (2) Quality (i) Do not say that for which you lack adequate evidence; (ii) Do not say what you believe to be false. (3) Relation (i) Be relevant. (4) Manner (i) (ii) (iii) (iv)
Avoid obscurity of expression; Avoid ambiguity; Be brief (avoid unnecessary prolixity); Be orderly.
Interlocutors in everyday conversations quite often fout and infringe these maxims. In so doing, they presume that addressees will observe the infringements and draw conclusions. When the addressees observe the infringements, they continue to assume that the interlocutors are “making infringements for a good reason. These conclusions are referred to by Grice as conversational implicatures” (James, 1980, p. 128, emphasis in original). The term conversational implicatures has been employed in pragmatics to denote “what the speaker means or implies rather than what s/he literally says” (Baker, 1992, p. 285). Thus, the main concern here is “the question of how it is that we come to understand more than is actually said” (Baker, 1992, p. 223). The pragmatic analysis of speech acts entails looking at the functions of all utterances from two perspectives: “‘stating’ and ‘doing things’, of having a meaning and a force” (Hatim, 2001, p. 179). In this respect, Grice (1975) proposes two levels of analysis: what is said and what is implicated. If the addresser flouts a maxim at the level of what is said, the addressee is “entitled to assume that that maxim, or at least the overall Cooperative Principle, is observed at the level of what is implicated” (Grice, 1975, p. 52). In observing or violating one or more of the conversational maxims, interlocutors are doing so to achieve a communicative purpose. Different purposes can be achieved in different communicative situations and by different participants. Some communicative purposes include
86
Qur’anic idiomatic phrasal verbs
“conveying information, influencing the opinion or emotions of hearers, directing their actions, and so on” (Baker, 1992, p. 226). Interlocutors may try to flout one or more of the conversational maxims for personal reasons (e.g. when the addresser wishes to mislead the addressee); for rhetorical purposes such as exaggeration and irony (Marogy, 2010); or for evading a topic or a question, as is regularly seen in interviews with politicians (Baker, 1992). It is crucial and relevant to the topic at hand (i.e. QIPVs) to differentiate between the notion of implicature and the notion of idiomaticity. As discussed in the previous chapter, idiomaticity concerns meaning transparency of a given lexical expression whereby the overall meaning is not inferable from the meanings of its separate constituents. In this sense, idiomatic meaning is a nonliteral conventional type of meaning, and its interpretation first and foremost “depends on a good mastery of the linguistic system in question” (Baker, 1992, p. 223). Hence, idiomatic meaning should not be confused with implied meaning as conveyed via conversational implicatures, because conversational implicatures depend in their appreciation on “a successful interpretation of a particular speaker’s intended or implied meaning in a given context” (Baker, 1992, p. 223). This differentiation is pertinent to the distinction made earlier between metaphorical QIPVs and figurative QIPVs (see 4.3.2). Metaphorical QIPVs—apart from satisfying the definition of metaphor and fulfilling its two purposes and its three components— have implied meanings expressed through conversational implicatures. Figurative QIPVs—apart from stopping short of fulfilling the definition of metaphor and meeting its components and purpose criteria—express nonliteral and nontransparent conventional meanings not deducible from the meanings of their separate constituents. In other words, appreciating the overall meanings of figurative QIPVs depends on the speaker’s mastery of the Arabic linguistic system rather than on a successful interpretation of the speaker’s implied meanings delivered via conversational implicatures. Nonetheless, there is a strong correlation between idiomaticity and implicatures as a type of indirect speech act. This correlation prompted Searle (1975) to set idiomaticity as a proviso for a given sentence to be “a plausible candidate for an utterance as an indirect speech act” (p. 76). He writes, In order to be a plausible candidate for an utterance as an indirect speech act, a sentence has to be idiomatic to start with . . . I think the explanation for this fact may derive from another maxim of conversation having to do with speaking idiomatically. In general, if one speaks unidiomatically, hearers assume that there must be a special reason for it, and in consequence, various assumptions of normal speech are suspended. . . . Beside the maxims proposed by Grice, there seems to be an additional maxim of conversation that could be expressed as follows: Speak idiomatically unless is some special reason not to. For this reason, the normal conversational assumptions on which the possibility of indirect speech acts rests are in large part suspended in the nonidiomatic cases. (Searle, 1975, pp. 76–77)
Qur’anic idiomatic phrasal verbs 87 The QIPVs are taken up in this study at all levels (i.e. syntactically, semantically, and pragmatically) and in an integrated manner. Leech (1983) rightly argues that “grammar (the abstract formal system of language) and pragmatics (the principles of language use) are complementary domains within linguistics” (p. 4). He accordingly formulates the notion of complementarity, which is relevant to this study, as follows: In essence, the claim will be that grammar . . . and pragmatics . . . are complementary domains within linguistics. We cannot understand the nature of language without studying both these domains, and the interaction between them. The consequences of this view include an affirmation of the centrality of formal linguistics in the sense of Chomsky’s “competence”, but a recognition that this must be fitted into, and made answerable to, a more comprehensive framework which combines functional with formal explanation. (Leech, 1983, p. 4) Leech’s complementarity approach to language is often used for a reliable analysis of language. It takes into its consideration both formal and functional aspects of language (Marogy, 2010, p. 56). The crux of this communicative approach is portrayed as follows: [It] involves a fruitful interaction among the syntactic, semantics and pragmatic levels underlying the language. In fact, the whole thrust of this account is that it studies language as a system of human communication and gives prominence to the complementarity of syntax and pragmatics in such a way that it escapes a rigid either-or approach to language. Within the framework of a complementarity approach, syntax and pragmatics are equally important factors in the analysis of language. (Marogy, 2010, p. 48) Along these lines, Halliday pinpoints the organic correlation between grammar and discourse analysis. He maintains that “a discourse analysis that is not based on grammar is not an analysis at all, but simply a running commentary on a text” (p. xvii, cited in Hoey & Houghton, 2001, p. 47). In addition, Halliday asserts that in spite of the fact that “a text is a semantic unit, not a grammatical one, meanings are realized through wordings; and without a theory of wording—that is, a grammar—there is no way of making explicit one’s interpretation of meaning of a text” (p. xvii, cited in Hoey & Houghton, 2001, p. 47). Other pragmatic principles relevant to the topic at hand are the politeness principle (henceforth PP) and the irony principle (henceforth IP), proposed by Leech (1983). The CPs and their maxims are, as acknowledged by Grice, not comprehensive, and other maxims can be added. One of the maxims to add, as suggested by Grice, is ‘be polite’, which in some cultures “indeed seems to override all other maxims” (Baker, 1992, p. 233). Loveday (1982, p. 364) points out
88
Qur’anic idiomatic phrasal verbs
that the word no “almost constitutes a term of abuse in Japanese, and equivocation, exiting, or even lying is preferred to its use” (cited in Baker, 1992, p. 233). This point has led some researchers to assert that “maxims of Quality and Manner are easily overridden by considerations of politeness in some cultures” (Baker, 1992, p. 233). Politeness is broadly defined as one manifestation of the wider concept of etiquette, or appropriate behaviour. [It is] a way of encoding distance between speakers and their addresses. . . . In being “polite”, a speaker is attempting to create an implicated context . . . to imply the most appropriate speaker-addressee relationship. (Grundy, 1995, pp. 127–128) As far as the PP is concerned, a close interaction exists between Leech’s PP on one hand and Grice’s CP and its conversation maxims on the other. The latter is supplemented by the former, which go hand in hand (Farghal, 1995). The PP, as explained by Leech (1983), “concerns a relationship between two participants . . . self and other” (p. 131, emphasis in original). Self, in conversation, is typically identified as the speaker, and other is identified as the hearer. In addition, politeness is shown by the speaker to third parties “who may or may not be present in the speech situation” (p. 131). Leech (1983) proposed the following six maxims for his PP theory (p. 132, emphasis in original): (1) TACT MAXIM (a) Minimize cost to other; (b) Maximize benefit to other. (2) GENEROCITY MAXIM (a) Minimize benefit to self; (b) Maximize cost to self. (3) APPROBATION MAXIM (a) Minimize dispraise of other; (b) Maximize praise of other. (4) MODESTY MAXIM (a) Minimize praise of self; (b) Maximize dispraise of self. (5) AGREEMENT MAXIM (a) Minimize disagreement between self and other; (b) Maximize agreement between self and other.
Qur’anic idiomatic phrasal verbs 89 (6) SYMPATHY MAXIM (a) Minimize antipathy between self and other; (b) Maximize sympathy between self and other. Politeness, for Searle (1975), seems to be “the most prominent motivation for indirectness in requests” (pp. 76–77). Along these lines, he suggests that directives are the most useful to study in the feld of indirect illocutionary acts, “because ordinary conversational requirements of politeness normally make it awkward to issue fat imperative sentences (e.g. Leave the room) or explicit performatives (e.g. I order you to leave the room), and we therefore seek to fnd indirect means to our illocutionary ends (e.g. I wonder if you would mind leaving the room)” (Searle, 1975, p. 64). Moreover, Searle (1975) exemplifes ‘directive’ illocutionary acts as “ordering, commanding, requesting, pleading, begging, praying, entreating, instructing, forbidding, and others” (p. 61). Moving on to IP, Leech (1983) maintains that “[i]f you must cause offence, at least do so in a way which doesn’t overtly conflict with the PP but [which] allows the listener to arrive at the offensive point of your remark indirectly, by way of implicature” (p. 82). The IP for Leach is a “substitute for impoliteness” and is a “second-order principle” (p. 142). It allows “a speaker to be impolite while seeming to be polite; it does so by superficially breaking the CP, but ultimately upholding it” (p. 142). To this end, IP is complementary to CP and PP. It denotes that a given listener should indirectly arrive to an offensive remark, if such a remark cannot be avoided. [Therefore, it does] not only allows the speaker to circumvent the rules of politeness and social correctness, but also provides him with a tool to express his unconcealed false politeness that mounts to mockery without being openly rude. (Marogy, 2010, p. 63) The point to confirm here is that all the aforesaid indirect speech forms are language- and culture-specific forms. In other words, differences in this regard occur “from one language to another. The mechanisms are not peculiar to this language or that, but at the same time the standard forms from one language will not always maintain their indirect speech act potential when translated from one language to another” (Searle, 1975, p. 76). The Qur’an is a divine source of guidance that addresses humankind throughout all generations. Its divine message is not limited to the contemporaries of its time of revelation living in the Arabian Peninsula. Therefore, its discourse is characterized by employing different means to perform a variety of speech acts to fulfil different purposes. A close look at the Qur’anic discourse reveals that the QIPVs are employed in a range of Qur’anic structures to perform both direct speech acts and indirect speech acts. In what follows, I use the aforementioned
90
Qur’anic idiomatic phrasal verbs
method to investigate the QIPVs employed in the Qur’an to perform direct speech acts followed by those employed to perform indirect speech acts (i.e. conversational implicatures).
4.4.1 Direct speech acts performed by QIPVs Before embarking on the direct speech acts achieved by the QIPVs, let me start this subsection by citing Searle’s definition of direct speech acts: The simplest cases of meaning are those in which the speaker utters a sentence and means exactly and literally what he says. In such cases the speaker intends to produce a certain illocutionary effect in the hearer, and he intends to produce this effect by getting the hearer to recognize his intention to produce it, and he intends to get the hearer to recognize his intention in virtue of the hearer’s knowledge of the rules that govern the utterance of the sentence. (Searle, 1975, p. 59) The QIPVs are used in the Qur’an to perform several direct speech acts for a wide range of purposes. The direct speech acts performed, along with their illustrative examples, are shown in Table 4.5.
Table 4.5 Direct speech acts performed by QIPVs Direct speech act
Illustrative example
COMMANDING
wa-itū al-biyūta min abwābihā (Qur’an, 189). “So enter your houses by their [main] doors” (Abdel Haleem, 2010, p. 30). yā-ayyuhā al-ladhīna āmanū idhā nūdiya li-al-ṣalāti min yawmi al-jumuʿati fa-isʿaw ilá dhikri-illāhi (Qur’an, 62:9). “Believers! When the call to prayer is made on the day of congregation, hurry towards the reminder of God” (Abdel Haleem, 2010, p. 555). Qāla rabbi ishraḥ lī ṣadrī (Qur’an, 20:25). “Moses said, ‘Lord, lift up my heart’” (Abdel Haleem, 2010, p. 314). wa-aktub la-nā f ī hadhihi al-dunyā ḥasanatan wa-f ī al-ālkhirati innā hudnā ilay-ka (Qur’an, 7:156). “Grant us good things in this world and in the life to come. We return to you” (Abdel Haleem, 2010, p. 171). wa-ittakhidhū min maqāmi Ibrāhima muṣallá (Qur’an, 2:125). “Take the spot where Abraham stood as your lace of prayer” (Abdel Haleem, 2010, p. 20). wa-li-yarbiṭa ʿalá qulūbikum wa-yuthabbita bi-hi al-aqdāma (Qur’an, 8:11). “to make your hearts strong and your feet firm” (Abdel Haleem, 2010, p. 179).
REQUESTING
PLEADING PRAYING
INSTRUCTING SUPPORTING
Qur’anic idiomatic phrasal verbs 91 FORBIDDING PRAISING
WONDERING
THREATINING
REGRETTING
SYMPATHIZING PROMISE
CHALLENGING
wa-lā-taʾkulū amwālahu ilá amwālikum (Qur’an, 4:2). “and do not consume their property along with your own” (Abdel Haleem, 2010, p. 78). Inna al-ladhīna yaghuḍḍūna aṣwātahum ʿinda rasūli-illāhi ulāʾika al-aldhīna imtaḥana Allāhu qulūbahu li-l-taqwá (Qur’an, 49:3). “It is those who lower their voices in the presence of God’s Messenger whose hearts God has proved to be aware” (Abdel Haleem, 2010, p. 516). alam tara ilá al-ladhīna utū naṣīban min al-kitābi yuʾminūna bi-al-jubti wa-al-ṭāghūti (Qur’an, 4:51). “Do you not see how those given a share of the Scripture, (evidently) now believe in idols and evil powers?” (Abdel Haleem, 2010, p. 87). wa-law shāʾa Allāhu la-dhahaba bi-samʿihim wa-abṣārihim (Qur’an, 2:20). “If God so willed, He could take away their hearing and sight” (Abdel Haleem, 2010, p. 5). wa-lammā suqiṭa f ī aydihim wa-raʾaw annahum qad ḍallū qālū la-in lam yarḥamna rabbunā wa-yaghfir la-nā la-nakūnanna min al-khāsirīn (Qur’an, 7:149). “When with much wringing of hands, they perceived that they were doing wrong, they said, ‘If our Lord does not have mercy on us and forgive us, we shall be the losers’” (Abdel Haleem, 2010, p. 169). wa-mā uridu an ashuqqa ʿalay-ka (Qur’an, 28:27). “I do not intend to make things difficult for you” (Abdel Haleem, 2010, p. 389). thumma la-ātiyannahum min bayni aydihim wa-min khalfihim (Qur’an, 7:17). “I will come at them—from their front and their back” (Abdel Haleem, 2010, p. 153). fa-inna Allāha yaʾtī bi-alshamsi min al-mashriqi fa-ʾti bi-hā min al-maghribi (Qur’an, 2:258). “God brings the sun from the east; so bring it from the west” (Abdel Haleem, 2010, p. 44).
4.4.2 Indirect speech acts performed by QIPVs (QIPVs’ implicatures) Indirect speech acts, as defined by Searle (1975), are “cases in which one illocutionary act is performed indirectly by way of performing another” (p. 60). The QIPVs are typically heavy-loaded structures with multilayered significances. They are characterized as carrying both linguistically encoded significances and several implied significances, which are usually the significances intended by the addresser. To perform a wide range of conversational implicatures by flouting several of Grice’s maxims to achieve different communicative purposes, the QIPVs are employed extensively throughout the Qur’anic discourse. I must assert here that accounting for the pragmatic properties of all the QIPVs and tracing the conversational implicatures performed by them go beyond the
92
Qur’anic idiomatic phrasal verbs
scope of this study. Thus, three Qur’anic verses are selected here as illustrative data. More Qur’anic verses are fully attended to in the practical section of this book (see Chapter 6), where the QIPVs are scrutinized from syntactic, semantic, and pragmatic standpoints. For the purposes of this chapter, the three illustrative examples are investigated from the following perspectives: (1) The context of each verse; (2) The QIPV as used in that verse, focusing on its syntactic and semantic features; (3) Pragmatic principles used, maxims flouted, and implicatures intended. To do justice to the third perspective, the two levels of analysis proposed by Grice (1975)—that is, what is said and what is implicated—are considered. Scrutinizing the data at the frst level enables me to account for its propositional, linguistically encoded, semantic meaning (i.e. the locutionary act). Doing so at the second level enables me to underscore its pragmatic properties and in turn to determine the intended communicative meaning (i.e. the illocutionary force). The conversational implicatures performed by the QIPVs are exemplifed in the following Qur’anic contexts: (1) sa-nafrughu la-kum ayyuhā al-thaqalāni (Qur’an, 55:31) “We shall attend to you two huge armies [of jinn and humankind]” (Abdel Haleem, 2010, p. 533). The addresser (i.e. God) in this Qur’anic verse is addressing both jinn and [humankind] telling them that he will take them to task. The QIPV used in this verse is faragha li-. The verb faragha literally means ‘to become vacated’, ‘to take leisure’, ‘to complete an assignment’, ‘to become free of tasks’, or ‘to free oneself of all duties’ (Badawi & Abdel Haleem, 2008, p. 703). The preposition li- in this context signifes al-ikhtiṣāṣ (habitual belonging) (Dāwood, 2002, p. 1/293). Its principal function is to specify the indirect object to which the action performed by the verb is intended. At the what-is-said level, the locutionary act/propositional meaning of this verse is that God will free himself exclusively to jinn and humankind, attend to them, and take them to task (Badawi & Abdel Haleem, 2008, p. 703; Abdel Haleem, 2010, p. 533). At the what-is-implicated level, however, the addresser intentionally violates the maxim of MANNER by making his contribution ambiguous. The source of ambiguity in this verse is the use of the verb faragha. God commonly cannot be imagined to be too busy doing something that might prevent him from doing something else. This proposition contradicts how he typically describes himself on many occasions throughout the Qur’an, such as he “has power over everything” (cf Qur’an, 2:20, 106, 109, 148, 259, 284, among others). By purposely flouting the maxim of MANNER, the addresser aims to get across the intended implicature—that is, threatening the addressees
Qur’anic idiomatic phrasal verbs 93 (i.e. the jinn and humankind) (cf al-Rāzī, 1981, p. 29/111; al-Zamakhsharī, 1998, p. 6/13). This is shown by the usage of the future prefix sa- and by the metaphorical employment of the QIPV faragha li- in that context. (2) innā aʿtadnā li-al-ẓālimīna nāran aḥāṭa bi-him surādiquhā wa-in yastaghīthū yughāthū bi- māʾin ka-al-muhli yashwī al-wujūha (Qur’an, 18:29). “We have prepared a Fire for the wrongdoers that will envelop them from all sides. If they call for relief, they will be relieved with water like molten metal, scalding their faces” (Abdel Haleem, 2010, p. 298). This Qur’anic verse describes the fre that God prepares for wrongdoers. It accounts for two aspects of that fre. The frst is that it will surround them from everywhere. The second is that if the wrongdoers were to call for relief, they would be relieved with water like molten metal (cf al-Rāzī, 1981, p. 21/121; al-Zamakhsharī, 1998, p. 3/583). The QIPV used in this structure is yughāthū bi-. The verb yughāthū is passive present and means ‘to be succoured, to be bailed out, or to be relieved’ (Badawi & Abdel Haleem, 2008, p. 678). The preposition bi- in this context denotes the meaning ‘with’ and serves as an instrumental particle. Its main function is to show how the verb is performed (Da wood, 2002, p. 1/170). At the what-is-said level, the locutionary act/propositional meaning of the QIPV employed in this verse (i.e. yughāthū bi-) is ‘to be succoured, bailed out, relieved with’. Yet this propositional meaning evidently contradicts the meaning of the object of the preposition bi- and its adjective (mā in ka-al-muhli) ‘water like molten metal’. Succour and relief must be with something, which would help satisfy the needs of those who call for relief rather than something that would increase their torment. At the what-is-implicated level, as with the verse discussed in (1) from earlier, the metaphorical employment of the QIPV in this Qur’anic context is not without a communicative purpose. The addresser intentionally violates the maxim of RELATION by being irrelevant, in order to get across the intended implicature, viz. IRONY. The addresser wants to mock those who are addressed in this Qur’anic discourse (i.e. the wrongdoers), by employing the QIPV yughāthū bi- (to be relieved with) in a context where the ‘water like molten metal’ is the substance with which they are ‘relieved’. (3) wa-idhā massa al-insāna ḍurrun daʿā rabbahu muniban ilay-hi thumma idhā khawwalahu niʿmatan min-hu nasiya mā kāna yadʿu ilay-hi min qablu wa-jaʿala li-lāhi andānan li-yuḍilla ʿan sabīlihi qul tamattaʿ bi-kufrika qalīlan innaka min aṣhābi al-nāri (Qur’an, 39:8) “When man suffers some affliction, he prays to his Lord and turns to Him, but once he has been granted a favour from God, he forgets the One he had been praying to and sets up rivals to God, to make others stray from His path. Say, ‘Enjoy your ingratitude for a little while: you will be one of the inhabitants of the Fire’” (Abdel Haleem, 2010, p. 460).
94
Qur’anic idiomatic phrasal verbs
This Qur’anic verse talks about the contradictory and inconsistent ways adopted by idol worshippers. They turn to God and pray to him only when they suffer some affiction in their life. Yet when God saves them from that affiction, they forget his favour and go back to their old ways (i.e. idol worship) (cf al-Rāzī, 1981, p. 2/295; al-Zamakhsharī, 1998, p. 5/292). The QIPV used in this structure is tamattaʿ bi-. The basic meaning of the verb tamattaʿa is ‘to enjoy’, ‘to beneft’ (Badawi & Abdel Haleem, 2008, p. 866). The preposition bi- in this context denotes al-ilṣāq al-maʿnawī (abstract affxation) (Dāwood, 2002, p. 2/180). Its core function is to express the ascribing of the action performed by the verb to its indirect object. At the what-is-said level, the locutionary act/propositional meaning of the expression in which the QIPV is employed is that God is asking an idol worshipper to enjoy their ingratitude for a little while: ‘tamattaʿ bi-kufrika’. At the what-is-implicated level, however, the addresser intentionally flouts the maxim of QUALITY by saying what they believe to be false. The source of this falsity is the metaphorical use of the QIPV tamattaʿ bi- in that context. The propositional meaning of tamattaʿ bi- obviously contradicts the meaning of the object of the preposition bi- (i.e. kufr, or ‘ingratitude/disbelief’) because one normally enjoys good/positive deeds and does not enjoy bad/negative deeds such as ingratitude and disbelief. The maxim of QUALITY is flouted here by the addresser to get across the intended implicature—that is, threatening the addressee and letting them down (cf al-Zamakhsharī, 1998, p. 5/292; al-Aṣfahānī, n.d., p. 2/295).
4.5 Conclusion Thus far, except for their Qur’an-specific features, QIPVs share the morphological, syntactic, and semantic idiosyncrasies of the AIPVs explored in the previous chapter. The QIPVs and AIPVs also share the subtle semantico-syntactic interrelation between their constituent parts (i.e. verbs and prepositions), along with the fundamental factors that govern their functionality. Syntactically speaking, the QIPVs are of one of two types, depending on the preposition’s location (i.e. split QIPVs or nonsplit QIPVs). Each type falls into several syntactic patterns according to the verbs’ transitivity, tense, and voice and according to the number of prepositions with which they combine. Semantically speaking, the QIPVs are used in different semantic fields throughout the Qur’anic discourse and fall into one of two categories, according to their semantic properties. First, metaphorical QIPVs are characterized as nonliteral and nontransparent and fall under the definition of metaphor devised by Nida (1975) and Newmark (1988). They fulfil the two purposes of ‘metaphor’ (referential and pragmatic) and have three components (object, image, and sense). Second, figurative QIPVs are characterized as stopping short of satisfying the definition of metaphor and meeting its components and purpose criteria. Still, they are semantically nonliteral and nontransparent combinations. Pragmatically speaking, the QIPVs have been employed in the Qur’anic context to perform both direct speech acts and indirect speech acts. They are also employed to flout several conversation maxims, to bring forth several conversational implicatures.
Qur’anic idiomatic phrasal verbs 95
References Abdel Haleem, M. A. S. (2010). The Qur’an: English Translation and Parallel Arabic Text. Oxford University Press. Oxford. Abdul-Raof, Hussien (2001). Qur’an Translation: Discourse, Texture and Exegesis. Curzon. Richmond, Surrey, UK. al-Aṣfahānī, A. (n.d.). Al-Mufradāt fī ghrīb al-Qur’ān. Maktabat Nizār Musṭafā al-Bāz. Mecca, Saudi Arabia. Aldahesh, A. Y. (2009a). Translating Idiomatic English Phrasal Verbs into Arabic. VDM Verlag Dr. Muller. Saarbrucken, Germany. Aldahesh, A. Y. (2016a). Pinning Down the Phenomenon of Phrasal Verbs in Arabic. International Journal of language and Linguistics, 3 (1), 12–24. Aldahesh, A. Y. (2016b). Towards a Model for Analyzing and Assessing Translation of Qur’anic Idiomatic Phrasal Verbs. Arab World Journal, Special Issue on Translation (5), 33–53. Aldahesh, A. Y. (2017). Qur’anic Idiomatic Phrasal Verbs: Their Syntactic and Semantic Properties. International Journal of Language and Linguistics, 4 (3), 12–27. Aldahesh, A. Y. (2018). On Pragmatic Idiosyncrasies of Qur’anic Idiomatic Phrasal Verbs. International Journal of Language and Linguistics, 5 (2), 110–118. al-Rāzī, M. F. (1981). Mafātīḥ al-Ghayb. First Edition. Dār al-Fikr lil Tibāʿati wa alNashri wa al-Tawīzʿ. Beirut, Lebanon. al-Shamsān, A. I. (1986). Al-Fiʿilu fī al-Qurʿani al-Karīmi Taʿadyatuhu wa-Luzūmuhu. The Kuwait University. Kuwait. al-Zamakhsharī, J. A. M. (1998). Al-Kashshāf ʿan Ḥaqāʾiq ghawamid al-Tanzīl wa ʿuyūn al-ʾaqāwīl fī wujūi al-Tʾwīl. First Edition. Maktabat Al-ʿubaykān. Riyadh, Saudi Arabia. Asad, M. (2011). The Message of the Qur’an. Islamic Book Trust. Kuala Lumpur. Austin, J. L. (1975). How to Do things with Words. Clearndon Press. Oxford. Badawi, E. M., & Abdel Haleem, M. (2008). Arabic-English Dictionary of the Qur’anic Usage. Brill. Leiden and Boston. Baker, M. (1992). In Other Words: A Coursebook on Translation. Routledge. London and New York. Bell, R. (1991). Translation and Translating: Theory and Practice. Longman. London and New York. Dāwood, M. M. (2002). Al-Qurʿan al-Karīm wa-Tafāʿulu al-Maʿānī: Dirāsatun Dilāliyatun li-Taʿlluqi Ḥarfi al-Jarri bi- al-Fiʿili wa-ʾAtharihi f ī al-Maʿná f ī al-Qurʿan al-Karīm. Dār Gharīb. Cairo, Egypt. El-Zeiny, I. (2011). Criteria for the Translation and Assessment of Qur’anic Metaphor: A Contrastive Analytical Approach. Bable, 57 (3), 247–268. https://doi. org/10.1075/bable.57.3.01zei Farghal, M. (1995). Euphemism in Arabic: A Gricean Interpretation. Anthropological Linguistics, 37 (3), 366–378. Grice, H. P. (1975). Logic and Conversation. In P. Cole & J. L. Margon (Eds.), Syntax and Semantics, Volume 3: Speech Acts, pp. 41–58. Academic Press. London. Grundy, P. (1995). Doing Pragmatics. Edward Arnold. London, New York, Sydney, and Auckland. Hale, S. B. (2004). The Discourse of Court Interpreting: Discourse Practices of the Law, the Witness and the Interpreter. John Benjamins Publishing Company. Amsterdam and Philadelphia. Harnish, Robert M. (2010). Fragments and Speech Acts. In Iwona Witczak-Plisiecka (Ed.), Pragmatic Perspectives on Language and Linguistics Volume I: Speech Actions
96
Qur’anic idiomatic phrasal verbs
in Theory and Practice Studies. Cambridge Scholars Publishing. Newcastle upon Tyne, UK. Hatim, B. (2001). Pragmatics and Translation. In M. Baker (Ed.), Routledge Encyclopedia of Translation Studies. Routledge. London and New York. Hoey, M., & Houghton, D. (2001). Contrastive Analysis and Translation. In M. Baker (Ed.), Routledge Encyclopedia of Translation Studies. Routledge. London and New York. James, C. (1980). Contrastive Analysis. Longman. London. Leech, G. (1983). Principles of Pragmatics. Longman. London. Levinson, S. C. (1983). Pragmatics. Cambridge University Press. Cambridge. Marogy, A. E. (2010). Kitab Sibawayhi: Syntax and Pragmatics. Brill. Leiden and Boston. Newmark, P. (1988). A Textbook of Translation. Prentice Hall. New York, London, Toronto, Sydney, and Tokyo. Nida, Eugene. (1975). Componential Analysis of Meaning. Approaches to Semiotics 5 [AS]. Mouton, Hague. Searle, J. R. (1969). Speech Acts: An Essay in The Philosophy of Language. Alden & Mowbray Ltd. Oxford. Searle, J. R. (1975). Indirect Speech Acts. Syntax and Semantics, 3, 59–82. Shakir, Muhammad H. (2011). The Qur’an: Translation. Tahrike Tarsile Qur’an, Inc. Elmhurst and New York. Yūsuf ʿAlī, ʿA. (1991). The Meaning of the Holy Qur’ān. New Edition with Revised Translation and Commentary. Amana Corporation. Brentwood, MD.
5
Research methodology1
5.1 Theoretical framework Linguistic contrastive analysis theory (as introduced and implemented by Carl James, 1980) is used as a theoretical framework in this study. The QIPVs and their provided translations are described, compared, and assessed at both the micro-linguistic and the macro-linguistic levels. Contrastive analysis, as delineated by James (1980), is “a linguistic enterprise aimed at producing inverted (i.e. contrastive not comparative) two valued typologies . . . founded on the assumption that languages can be compared” (p. 3). It emerged in the US after World War Two and was developed as a practical tool for teaching second and foreign languages. Contrastive analysis has come to be understood as “a linguistic study of two languages aiming to identify differences between them in general selected areas” (Hoey & Houghton, 2001, p. 46; c.f. Aldahesh, 2009). In the wake of contrastive analysis’s popularity in Europe in the 1970s, it was employed to conduct several successful projects on proposing and applying models for translation studies. The relevance of contrastive analysis to translation studies is considered from both a practical perspective and a global perspective. From the practical perspective, contrastive analysis is “most useful in pointing out areas where direct translation of a term or phrase will not convey accurately in the second language the intended meaning of the first” (Hoey & Houghton, 2001, p. 47). From the global perspective, however, contrastive analysis “leads the translator to look at broader issues such as whether the structure of the discourse for a given text-type is the same in both languages” (Hoey & Houghton, 2001, p. 47). In addition, the relationship between translation studies and contrastive analysis has been referred to as “bidirectional” in that “the translation may provide the data for [contrastive analysis], as in Gleason (1965), Krzeszowski (1990), and James (1980). On the other hand, [contrastive analysis] may provide explanations of difficulties encountered in translation [as in] Nida (1964), Beekman and Callow (1974), Yebra (1982), Enkvist (1978) and Baker (1992)” (Hoey & Houghton, 2001, p. 49; c.f. Aldahesh, 2009). Along these lines, House (2009) writes the following in her attempt to account for the differences between translation and contrastive linguistics:
98
Research methodology While contrastive linguists are interested in equivalences of linguistic categories within and cross languages, translation scholars focus on equivalence in text, in the actual use of the languages and their component parts in communicative situation. . . . Translation does, however, draw on the findings of contrastive linguistics: many translation problems can be described on the basis of differences across linguistic systems. For instance, descriptions of how concepts such as time are encoded in different tense system can inform translation strategies used for dealing with descriptions of events in another language. (p. 15)
With this bidirectional correlation between contrastive analysis and translation studies in mind, theorists of the two disciplines “have expanded their focus of attention towards each other, and some scholars have openly sought to establish conceptual bridges between the two disciplines” (Chesterman, 1998, pp. 27–28). What is more, James (1980) proposes two approaches to his contrastive analysis theory: (1) Micro-linguistics; (2) Macro-linguistics Micro-linguistics, also called code linguistics and code oriented by James, refers to “the description of the linguistic code, without making reference to the uses to which the code is put, or how messages carried by this code are modifed by the context in which they occur” (James, 1980, p. 27). In linguists’ attempts to “specify the universal and particular properties of human languages”, they have employed this level of contrastive analysis to account for the abstract formal systems of language—that is, phonology, lexis, and grammar (James, 1980, pp. 98–100). Macro-linguistics, on the other hand, refers to the “contextual determination of messages and their interpretation” (James, 1980, p. 27). Attention in contrastive analysis is “shifted from the code to a process: the process of communication” (James, 1980, p. 100). The central objective of macro-linguistics is to attend to the six variables put forward by James in the following formula: “who says what to whom, where and when, how and why” (James, 1980, pp. 100–101, emphasis in original; cf Aldahesh, 2009). I must emphasize the relevance of complementarity (Leech, 1983) to the study at hand. In the previous chapter, I demonstrated such relevance and stated that the QIPVs are taken up in this study at the syntactic, semantic, and pragmatic levels in an integrated manner. Because “both formal and functional aspects are necessary for a reliable account of language, as pragmatics can offer answers to questions that formal grammar fails to satisfy” (Marogy, 2010, p. 56). Thus, the contrastive analysis approach here is not confined to dealing with the lexicalsemantic level. Rather, it is also employed to deal with the pragmatic aspects involved in translating the QIPVs into English. House (2009), in this respect, indicates that
Research methodology 99 Contrastive linguistics becomes even more relevant when it extends its scope to deal not only with linguistic forms but with the way they are typically used in different languages to organize information, perform communicative acts, or express attitude. How far languages do and do not correspond in general across these different levels obviously gives crucial guidance to the translator when it comes to establishing correspondences across particular texts in different languages. (p. 16) To this end, James (1980) proposes two principal procedures for conducting a contrastive analysis study: description and comparison. He also introduces the notion of tertium comparationis as a yardstick and as a point of comparison. According to James (1980), tertium comparationis is the sameness/constant that constitutes the basis of comparison between the two texts on which the contrastive analysis is to be conducted. House (2009), in this regard, defnes tertium comparationis as “the third element or factor that is the common ground between two elements being compared” (p. 31). Along these lines, Hoey and Houghton (2001) indicate that “[a]ll comparisons require that there be a common ground against which variation may be noted, a constant that underlies and makes possible the variables that are identifed; this is known as the tertium comparationis” (p. 47, italics in original). Likewise, Munday (2001) asserts that “the biggest bone of contention in the comparison of [a ST] and a [TT] is the so-called tertium comparationis, an invariant against which two text segments can be measured to gauge variation” (p. 49). The factors that constitute the common ground (i.e. the tertium comparationis) in this study are the formal and functional significances of the QIPVs. These are determined and interpreted by the main authoritative classical and contemporary exegetes of the Qur’an and by the explanations that leading influential Arabic lexicographers have provided. One of the research questions put forward in this study (see section 5.3) is, to what extent are the selected translators of the Qur’an successful in providing the English functional-pragmatic equivalence of the QIPVs? Thus, here I shed some light on this equivalence type and its relevance to this study. The notion of functional-pragmatic equivalence has long been established and has gained increasing acceptance in the field of contrastive linguistics (House, 2001a, 2001b). The functional-pragmatic equivalence, as defined by House (2001b), is “the preservation of ‘meaning’ across two different languages and cultures” (House, 2001b, p. 247). It is, she elaborates, “the type of equivalence which is most appropriate for describing relations between original and translation” (House, 2001b, p. 247). What is more, this equivalence type was used by House in her A Model for Translation Quality Assessment, proposed in 1977. It was subsequently revised in 1981, 1997, and 2001 and presented under the title Functional-Pragmatic Model of Translation Evaluation (House, 2001a, 2001b). Given the relevance of House’s model and the close appropriateness of her functional-pragmatic equivalence to this study, the gist of her approach is summed up next.
100
Research methodology
House (2001b) maintains that three approaches can be taken for translation quality assessment: mentalist views, response-based approaches (behaviouristic views and the functionalistic “skopos”-related approach), and text- and discoursebased approaches (literature-oriented approaches, descriptive translation studies, postmodernist and deconstructionist thinking, and linguistically oriented approaches). House (2001b) expresses her disagreement with the first two approaches on the grounds that the first approach (i.e. mentalist views) is characterized as being rather subjective and intuitive. She principally argues against its stance that “the relativisation of ‘content’ and ‘meaning’ . . . is particularly inappropriate for the evaluation business of making argued statements about when, how and why a translation is good” (p. 244). House (2001b) also disagrees with taking the second approach (i.e. response-based approach) on the grounds that responses cannot be taken as criteria for translation assessment, because views and opinions cannot be measured and evaluated (p. 244). House supports the third approach (specifically the linguistically oriented approach) established and developed by the groundbreaking works of Catford, Reiss and Wilss, and extended by the scholarly works of Baker, Doherty, Hatim and Mason and Hickey among others (cf House, 2001a, 2001b; Aldahesh, 2009). House considers this approach as promising, and she therefore formulates her model of translation quality assessment on the basis of this approach. House describes Functional-Pragmatic Model of Translation Evaluation in the following manner: [The model is] based on Hallidyan systematic-functional theory, but also draws eclectically on Prague school ideas, speech act theory, pragmatics, discourse analysis and corpus-based distinction between spoken and written language. It provides for the analysis and comparison of an original and its translation on three different levels: the levels of Language/Text, register (Field, Mode and Tenor) and Genre. (House, 2001b, p. 247) In this model, House writes that ‘Text’ and ‘Register’ (‘context of situation’ or ‘situational dimension’, as she sometimes refers to it) “should not be viewed as separate entities. Rather, the context in which the text unfolds is encapsulated in the text through a systematic relationship between the social environment on the one hand and the functional organization of language on the other” (House, 2001b, p. 248). She adopts the established taxonomy of register, which breaks it into the three “manageable parts”: feld, mode, and tenor. She defnes ‘Field’ as the parameter that “captures social activity, subject matter or topic, including differentiations of degrees of generality, specifcity or ‘granularity’ in lexical items according to rubrics of specialized, general and popular” (House, 2001b, p. 248). House refers to ‘Tenor’ as the parameter that “captures ‘social attitude’, i.e. different styles (formal, consultative and informal) [and] refers to the nature of the participants, the addresser and the addressees, and the relationship between them in terms of social power and social distance, as well as degree of emotional charge” (House, 2001b, pp. 247–248). Finally, she writes that ‘Mode’ “refers
Research methodology 101 to both the channel—spoken or written . . . and the degree to which potential or real participation is allowed for between writer and reader” (House, 2001b, p. 248). In A Model for Translation Quality Assessment, House (1977) writes that the core of translation “lies in the preservation of ‘meaning’ across two different languages” (p. 25). Accordingly, she highlights three aspects of ‘meaning’: semantic, pragmatic, and textual. The semantic aspect of ‘meaning’ is referential and “consists of relationship of reference or denotative, i.e. the relationship of linguistics units or symbols to their referents in some possible world” (p. 25). House asserts that the semantic aspect of ‘meaning’ is “(a) most readily accessible and [something] for which (b) equivalence in translation can most easily be seen to be present or absent” (House, 1977, p. 26). Pragmatic ‘meaning’ is described by House as the type of meaning that is shown in the difference between the two fields of pragmatics and semantics, where pragmatics “relates to the correlation between linguistic units and the user(s) of these units in a given communicative situation” (House, 1977, p. 27). Such a distinction, House elaborates, constitutes the essence of speech act theory, which was first proposed by Austin (in 1962) and subsequently developed by Searle (1969). Pragmatic meaning here is “the illocutionary force that an utterance is said to have, i.e. the particular use of an expression on a specific occasion” (House, 1977, p. 27). By the textual aspect of ‘meaning’, House means the ways that a text is composed. Such ways include occurrences of pro-forms, substitutions, coreferences, ellipses, and anaphora, each of which contributes to the overall meaning of the text to varying degrees. Hence, they need to be observed and kept equivalent in translation because translation is “a textual phenomenon” (House, 1977, pp. 28–29). An adequately translated text, for House, is thus “a semantically and pragmatically equivalent one [and] as a first requirement for semantic-pragmatic equivalence [she posits] that the translation text have a function equivalent to that of its source text” (House, 1977, p. 30, emphasis in original). In her model, House emphasizes the notion of ‘function’. For her, the function of a text is “the application . . . or use which the text has in the particular context of a situation. [Thus,] in order to characterize the function of a text precisely, we must analyze the text in detail” (p. 37). She adopts Halliday’s dichotomy of the function’s types (i.e. ideational and interpersonal). The ideational type signifies the content-oriented function in which “language expresses content: the speaker’s vision of the external world as well as the experience of the internal world of his own consciousness” (House, 1977, p. 34). The interpersonal function, however, denotes the nonreferential functional component in which “language serves as a means for conveying the speaker’s relationship with interlocutor(s) and for expressing social roles including communication roles such as questioner and respondent” (House, 1977, p. 34). House employs this dichotomy of functional types in categorizing that the texts she collects as samples. Accordingly, she lists scientific texts, commercial texts, journalistic articles, and tourist information booklets under the ideational function. In contrast, she lists religious sermons, political speeches, moral anecdotes, and comedy dialogue under the interpersonal function.
102
Research methodology
With all these in mind, House devises her own eclectic model for translation quality assessment as “explicit practical guidelines for a coherent analysis and evaluation of a translation” (House, 1977, p. 2). She endeavours to characterize “the linguistic-situational peculiarities of the source text, comparing source and translation texts, and making objective statement about the relative match of the two texts” (House, 1977, p. 2). The cornerstone of House’s model is the functional-pragmatic equivalence. Hence, a given translation text “should have a function—consisting of two functional components, the ideational and the interpersonal—which is equivalent to ST’s function, and that TT should employ equivalent pragmatic means for achieving that function” (House, 1977, p. 244). In House’s attempt to establish an explicit way to determine the function of a given text, she proposes the term situational dimensions, in which she breaks down the notion of situation into eight manageable elements. In so doing, House offers two situational dimension categories and establishes two linguistic correlates for them (House, 1977, pp. 37–50): (1) Language use comprising the three parameters of geographical origin, social class, and time; (2) Language use comprising the five parameters of medium, participation, social role relationship, social attitude, and province. The first step in applying House’s model for translation quality assessment is to sketch a textual profile depicting the function of a source text. This is done by analysing the text in relation to the aforesaid eight situational dimensions. The second step is to use this textual profile as a yardstick against which to measure the translation text to portray its textual profile. The third step is to compare the two resultant textual profiles (i.e. source text and its translation text profiles, which describe their functions). Thus, the matches and mismatches between the two texts are explored. The last step in applying House’s model for translation quality assessment is to provide an objective statement of the relative match between the ideational function and the interpersonal function. Against this background, House (2001b) regards translation as “the recontextualization of a text in SL by a semantically and pragmatically equivalent text in TL” (p. 247). Further, she considers the notion of equivalence “as a concept constitutive of translation[; it] is the fundamental criterion of translation quality” (p. 247)—hence the centrality of the functional-pragmatic equivalence in her functional-pragmatic model for translation quality assessment (cf Aldahesh, 2009). Another relevant aspect of House’s approach to this study is the way she differentiates between two types of mismatches or errors. She proposes in this respect the terms “covertly erroneous errors” and “overtly erroneous errors” (House, 1977, pp. 56–57). Covertly erroneous errors refer to dimensional mismatches, including the mismatch between any situational dimensions. Overtly erroneous errors refer to nondimensional mismatches, including “both mismatches of the denotative meanings of ST and TT elements and breaches of the target language system” (House, 1977, p. 245).
Research methodology 103
5.2 Data collection A corpus of the ten most common and circulated English-language translations of the Qur’an is scrutinized, compared, and assessed in this study. The aim is to determine the extent to which each translation achieves the functional-pragmatic equivalent of the QIPVs. Muslim and non-Muslim, Arab and non-Arab, and male and female translators have translated the Qur’an into English. Notwithstanding that these are popular translations, the reason for selecting them rather than others is to represent the translators’ different religious, linguistic, and gender backgrounds. This study investigates ten English translations of the Qur’an, which follow. 1
The Meanings of the Glorious Qur’an, by Muhammad Marmaduke Pickthall (first published in 1930).
Pickthall is a native English speaker, British novelist, and the son of a vicar who became a Muslim (Abdel Haleem, 2010). Pickthall’s translation of the Qur’an is said to be the frst English translation conducted by a Muslim translator. His translation adheres to the “conventional mainstream Muslim stance” (Kidwai, 2007, p. 274). It has been described as the “earliest Muslim translation to gain wide acceptance. . . . it generally gives a fairly literal rendering of the Arabic” (Robinson, 1977, p. 261), and it remains well liked in the Arab and Muslim worlds (Abdel Haleem, 2010). 2
The Meaning of the Holy Qur’an, by ʿAbdullah Yūsuf ʿAlī (first published in 1934).
Yūsuf ʿAlī is an Indian Muslim with an Ismāʿilī background. He is a native speaker of neither Arabic nor English. With his British wife, he spent most his life between India and Britain and “went through personal tragedies and fnally sought solace in the Qur’an by embarking upon its translation” (Iqbal, 2000, p. 107). His translation is a “paraphrase of the meaning of the Qur’an rather than a strict, literal translation” (Kidwai, 2007, p. 5). In addition, Yūsuf ʿAlī’s translation occupies a high status in the feld of Qur’anic literature because of the thousands of footnotes and running commentaries that it includes. They provide the reader with detailed insights into different Qur’anic matters. This is in addition to the rhythmic introductions that he provides for each Qur’anic sūrah in his attempt to communicate the music and richness of Qur’anic Arabic to the English readership. 3
The Koran Interpreted, by Arthur John Arberry (first published in 1955).
Arberry is a non-Muslim British scholar and native English speaker. After graduating from the University of Oxford, he became head of the Department of Classics at Cairo University. This led him to appreciate the richness of Islamic culture
104
Research methodology
and to gain mastery over East, Far East, and Southeast Asian languages (Kidwai, 2007, p. 117). What makes Arberry’s translation “one of the most respected translations of the Qur’an in English [and] a popular version of the text, particularly in academic circles” (Abdel Haleem, 2010, p. xxvii) is the great respect that he shows “towards the language of the Qur’an, particularly its musical effects. His careful observation of Arabic sentence structure and phraseology makes his translation remarkably close to the Arabic original in grammatical terms” (Abdel Haleem, 2010, p. xxvii). 4
The Qur’an: Translation, by Muhammad Habib Shakir (first published in 1968).
Shakir is a Pakistani Muslim with a background as a Shiʿite scholar. He is a native speaker of neither Arabic nor English. Little is known about Shakir despite the many editions of his translation. His translation offers the Shiʿite understanding of Qur’anic matters and is thus important for those who would like to study the Qur’an from that perspective (Kidwai, 2007, pp. 428–429). 5
The Quran: The Eternal Revelation Vouchsafed to Muhammad the Seal of the Prophets, by Muhammad Zafrulla Khan (first published in 1971).
Zafrulla Khan is a Pakistani Ahmadi Qadiani scholar. He is a native speaker of neither Arabic nor English. Zafrulla Khan studied law at a US mission school in King’s College and then at Lincoln’s Inn, London. He was appointed as Pakistan’s frst foreign minister, in 1947. In 1962, Zafrulla Khan was elected as the president of the United Nations General Assembly. Despite the carefully selected subtitle The Quran: The Eternal Revelation vouchsafed to Muhammad the Seal of the Prophets, Zafrulla Khan’s translation of the Qur’an is “a typical Qadyani translation” (Kidwai, 2007, pp. 229–230). 6
The Noble Qur’an: English Translation of the Meanings and Commentary, by Muhammad Taqī-ud-Dīn al-Hilālī & Muhammad Muhsin Khān (first published in 1977).
Al-Hilālī is a Moroccan scholar, a native speaker of Arabic, and a professor of Islamic faith and teaching at the Islamic University in Madinah, Saudi Arabia (Kidwai, 2007, p. 200). Khān is a Pakistani physician and a native speaker of neither Arabic nor English. For years, he worked “for the Ministry of Health, Saudi Arabia, and developed a keen interest in producing literature on Islam” (Kidwai, 2007, p. 200). Al-Hilālī and Khan’s translation is an English summary of three major classical tafs īr (exegeses of the Qur’an): tafsīr ibn Kathir, tafsīr al-Qurṭubī, and tafsīr al-Ṭabarī (Kidwai, 2007, p. 200). The translation is sponsored and disseminated by the Saudi authorities and adheres to the mainstream traditional view of the Islamic faith. 7
The Message of the Qur’an, by Muhammad Asad (first published in 1980).
Research methodology 105 Asad is an Austrian converted Muslim who is a native speaker of neither Arabic nor English. He was born in 1900 in a Jewish family as Leopold Wiess and studied the history of art and philosophy at Vienna University. Asad converted to Islam in 1926 after studying at al-Azhar in Cairo, Egypt (Kidwai, 2007, p. 131). Having spent six years in Arabia learning Arabic, Asad is deemed to be the only European translator of the Qur’an to have learned Arabic directly from speakers of the language of the Qur’an (Iqbal, 2000, p. 108). In addition, he is “one of the most original translators, who did the background research for himself in the original lengthy Arabic exegeses” (Abdel Haleem, 2010, p. xxvii). Using idiomatic English in his translation, Asad was aware that “he was rendering into English a very diffcult text and he was translating for a specifc readership: the modern Western reader” (Iqbal, 2000, p. 112). Such a rich experience and “his study of the early history of Islam, familiarity with biblical sources and understanding of Western civilization are some of the important factors which have contributed to the uniqueness of his translation” (Iqbal, 2000, p. 108). 8 The Qur’an: English Translation and Parallel Arabic Text, by Muhammad A. S. Abdul Haleem (first published in 2004). Abdul Haleem is an Egyptian Muslim scholar. He is a native Arabic speaker who learned the Qur’an by heart and graduated from al-Azhar in Cairo, Egypt. He is also a professor of Islamic studies at SOAS, University of London, England. In addition, Abdul Haleem is the editor of the well-known Journal of Qur’anic Studies. His translation renders the Qur’an in modern English and thus avoids the archaic language used by previous translators. 9 The Qur’an with a Phrase-by-Phrase English Translation, by ʿAlī Qulī Qarāʾī (first published in 2004). Qarāʾī is an Iranian Muslim of Shiʿite background who is a native speaker of neither Arabic nor English. Qarāʾī is renowned for producing outstanding text translations in the felds of Islamic philosophy and mysticism (Rizvi, 2006). In his Qur’an translation, Qarāʾī adopts a phrase-by-phrase approach as a ‘mirrorparaphrasing’, to allow for “idiomatic and grammatical shifts in favour of meaningful expressions” (Rizvi, 2006, p. 129). Qarāʾī’s translation communicates a Shiʿite understanding of the Qur’an. Nonetheless, he “deserves credit for his pious commitment to Qur’anic scholarship” (Kidwai, 2005, p. 270). 10 The Sublime Quran, by Laleh Bakhtiar (first published in 2007). Bakhtiar is an Iranian US-American Muslim woman who is a native English speaker. She studied classical Arabic at “Tehran University in a Ph.D. program and later with a private tutor for three years” (Bakhtiar, 2012, p. xxii). Before embarking on the task of translating the Qur’an, Bakhtiar compiled the Concordance of the Sublime Qur’an (2011), which enabled her to appreciate the
106
Research methodology
semantic nuances of the Qur’anic vocabulary and to realize the “lack of internal consistency and reliability” of the previous Qur’an translations (Bakhtiar, 2012, p. xiii). Bakhtiar provides a feminist translation of Qur’anic verses concerning women whereby her feminist mindset aims to “challenge the over 1400 years of male interpretation of the Quran” (Bakhtiar, 2012, p. xiii).
5.3 Research questions and data analysis This study aims to answer two sets of research questions. The first set accounts for the relevant issues covered in the theoretical part of this study (i.e. Chapters 2, 3, 4, and 5). The second set aims to address several crucial issues pertinent to the practical component of this study (i.e. Chapter 6). The first set of research questions are as follows: 1 2 3 4 5
6 7 8
What does the notion of untranslatability mean in the fields of linguistics and translation studies? What does Qur’an (un)translatability mean to Muslim intellectuals in general and translators of the Qur’an into English in particular? What is the phenomenon of AIPVs? What are the syntactic and semantic peculiarities of AIPVs? How can we devise a parameter for pinpointing the phenomenon of AIPVs and for underscoring the key factors that control the syntactic and semantic relationships between its main two constituent parts? What is the phenomenon of QIPVs? What are the syntactic, semantic, and pragmatic idiosyncrasies of QIPVs? What are the relevant parameters for depicting a working model for analysing and assessing translations of QIPVs?
By employing the two principal paradigms proposed by James (1980) (i.e. description and comparison), data collected from the aforementioned translations of the Qur’an are described and compared in view of the linguistic contrastive analysis approach. Using the proposed model for analysing and assessing the translations of the QIPVs (see section 5.6), the data analysis conducted addresses the following second set of research questions: 1 2 3 4 5
To what extent are the selected translators of the Qur’an successful in providing the English functional-pragmatic equivalence of the QIPVs? What are the types of translational pitfalls made by the selected translators of the Qur’an when tackling the issue of QIPVs? What are the procedures employed by the selected translators when dealing with QIPVs? What are the most adequate and recommendable procedures employed by the selected translators in rendering the QIPVs? To what extent can the QIPVs be described as translatable/untranslatable?
Research methodology 107 As previously mentioned, the QIPV’s degree of idiomaticity ranges from transparent to opaque. The transparent, literal structures of the QIPVs are excluded from this study because their signifcances are straightforward and cause no diffculties in translation.
5.4 Research procedure The procedure used to conduct the present study is outlined as follows: 1 A thorough analytical reading of the Qur’an is conducted to pinpoint the Āyāt ‘Qur’anic verses’ in which the QIPVs are used; 2 The selected QIPVs are analysed according to the aforesaid theoretical framework, to highlight their syntactic, semantic, and pragmatic idiosyncrasies; 3 English equivalents provided for these QIPVs in the corpus of ten English translations of the Qur’an are scrutinized and presented; 4 The English equivalents of the QIPVs are described and compared at both the micro-linguistic and the macro-linguistic levels; 5 The formal and functional significances of the QIPVs—as ascribed by the authoritative Qur’anic exegetical and lexicographical works—are identified and taken as the basis for comparison (i.e. tertium comparationis); 6 The QIPV English translations are analysed, compared, and assessed in the light of the model proposed in this study; 7 The QIPV English translations are examined to underscore the procedures opted for by the selected translators in rendering them; 8 The procedures employed are analysed, compared, and assessed in view of the prescribed model, to determine the most adequate procedure for translating the QIPVs in question; 9 A comprehensive concordance of the QIPVs enclosed in the Qur’an is provided. The concordance itemizes every QIPV in the Qur’an, and the QIPVs are listed according to the Arabic alphabetical order of the roots of their main verbs (see Appendix I); 10 Details on the frequencies of the QIPVs used in this study are presented, including their distribution in the Qur’an, Arabic roots, derivative word classes, and degree of idiomaticity (see Appendix II).
5.5 Towards a model for analysing and assessing the translation of QIPVs Drawing on the pertinent pragmatics theories (i.e. speech acts, CP, PP, and IP) discussed in Chapter 4 and implementing the contrastive analysis approach as a theoretical framework, this subsection focuses on identifying the relevant parameters to assist in proposing a working model for analysing and assessing QIPV translations.
108
Research methodology
5.5.1 On translating metaphorical and figurative expressions As demonstrated in Chapter 4, QIPVs fall into two semantic categories: metaphorical and figurative. The metaphorical QIPVs were defined as nonliteral and nontransparent and as falling under the definition of metaphor devised by Nida (1975) and Newmark (1988), fulfilling its two purposes (referential and pragmatic) and satisfying its three components (object, image, and sense). The figurative QIPVs were described as those that stop short of fulfilling the definition of metaphor and meeting its components and purpose criteria but that remain semantically nonliteral and nontransparent combinations. Chapter 4 also demonstrated that appreciating the overall significances of the figurative QIPVs depends primarily on the speaker’s mastery of the Arabic linguistic system. This is because their idiomatic significance is of a nonliteral conventional kind. Appreciating the overall significances of the metaphorical QIPVs, however, predominantly depends on the successful interpretation of the speakers’ implied meanings delivered via conversational implicatures. When it comes to translation, metaphors constitute “the most important particular problem” (Newmark, 1988, p. 104), for two reasons. First, between the three levels of meaning (i.e. aesthetic, expressive, and informative) in the target text is an inherent tension and conflict. Such conflict inevitably leads to sacrificing one level of meaning for the sake of the other levels (El-Zeiny, 2011). Second, a decision must be made regarding “how much space to allot to the criss-crossed area of sense, and further to determine whether this area is: (a) positive or negative; (b) connotative or denotative” (Newmark, 1988, p. 105). Wonderly (1971), Larson (1984), and Newmark (1981, 1988) proposed several procedures for translating metaphor. Given their relevance to our study, these procedures are summarized as follows (cf El-Zeiny, 2011, pp. 149–250): 1 2 3 4 5 6 7
To reproduce the metaphor provided the TL permit; To reproduce the SL by an equivalent TL metaphor; To reduce metaphor to sense; To translate the same metaphor plus sense; To substitute the metaphor by using a simile; Deleting the metaphor if the text is not authoritative or expressive and if the metaphor’s function is fulfilled elsewhere; Replacing the metaphor by using a simile plus sense.
Scholars hold diverse views on how to classify metaphors, because of the different scales that they adopt (Zahid, 2009). Fowler (1926), for instance, adopts a ‘mental’ scale and classifes metaphors as live or dead. Newmark (1988), on the other hand, adopts a ‘multidimensional’ scale and classifes metaphor into six types: dead, clichéd, stoke or standard, recent, original, and adopted. Dickins et al. (2002) adopt a ‘lexical’ scale and classify metaphors as lexicalized and nonlexicalized. This study does not intend to examine these defnitions. What is relevant to our topic is that the metaphorical QIPVs can be considered as, to use Newmark’s
Research methodology 109 term, stoke or standard metaphors. These are established metaphors that refect “an effcient and concise method of covering a physical and/or mental situation both referentially and pragmatically” (Newmark, 1988, p. 108). Translating figurative expressions requires the translator to “tease out the meaning of each word in a figurative meaning by matching its primary meaning against its linguistic, situational and cultural contexts” (Newmark, 1988, p. 106). Elaborating on this point, Zahid (2009) asserts that “[a] translator is called upon to not bind himself within the grammatical structure and the denotative meaning but dig beyond the first meaning into the ‘meaning of meaning’ instead” (p. 8). Recognizing this, Qur’anic idioms—whether metaphorical or figurative—have been classified into the following three categories in terms of their translatability/ untranslatability into English (Qarāʾī, 2004, pp. xviii–xx): 1
“Arabic idioms which though unfamiliar to the English-speaking audience are not difficult to understand when translated literally. These have been rendered literally” (Qarāʾī, 2004, p. xviii). Illustrative examples of this type of the Qur’anic idioms include (i) ushdud bi-hi azrī (Strengthen my back through him) (Qur’an, 20:3); (ii) tatajāfá junūbuhum ʿan al-maḍājiʿi (their sides vacate their beds) (Qur’an, 32:16).
2
Arabic idioms which are “unintelligible when translated literally. These have to be paraphrased appropriately in order to be understood” (Qarāʾī, 2004, p. xix). Illustrative examples of this type of the Qur’anic idioms include (i) fa-reddū aydīyahum fī afwāhihim (but they did not respond to them) (Qur’an, 14:9); (ii) wa-man yuslim wajhahu ilá Allāhi (whoever surrenders his heart to Allah) (Qur’an, 31:22); (iii) yawma yukshafu ʿan sāq (the day when the catastrophe occurs) (Qur’an, 68:42).
3
“In certain cases, it may be possible to substitute an English idiom” (Qarāʾī, 2004, p. xx). Illustrative examples of this type of the Qur’anic idioms include (i) fa-inna dhālika min ʿazmi al-umūri (that is indeed the steadiest of courses) (Qur’an, 3:186); (ii) fa-lā tadhhab nafsuka ʿalayhim ḥasarāt (so do not fret yourself to death regretting them) (Qur’an, 35:8); (iii) kulū wa-ishrabū han īʾan (enjoy your food and drink) (Qur’an, 52:19); (iv) Inna qawm ī ittakhdhū hādhā al-Qur’an mahjūrā (Indeed, my people consigned this Qur’an to oblivion) (Qur’an, 25:30); (v) fa-abá aktharu al-nāsi illā kufūrā (But most people are only intent on ingratitude) (Qur’an, 25:50); (vi) qabla an yartadda ilayka ṭarfuka (in the twinkling of an eye) (Qur’an, 27:40).
110
Research methodology
5.5.2 Speech acts Speech acts and their peculiarities as outlined in Chapter 4 are relevant to the issue of translating QIPVs in three ways: 1
2
3
QIPVs are employed in the Qur’anic discourse to perform different types of speech acts. The fact remains, however, that QIPVs have to be examined in their communicative contexts to determine their intended communicative values. Focusing on the locutionary/semantic aspects of meaning rather than the illocutionary/pragmatic aspects of meanings when QIPVs are translated distorts the meaning of the QIPVs by miscalculating the intended implicatures. In other words, the QIPVs should not be taken literally by Qur’an readers, commentators, and translators. Taking them literally entails matching their locutionary act rather than their illocutionary and perlocutionary acts, which constitute the final communicative goals. Many QIPVs are polysemic in that they denote different meanings in different contexts. Once more, targeting the pragmatic meaning of the QIPV entails distinguishing between their propositional and semantic significance on one hand and their illocutionary force on the other.
In this regard, House (1977) writes, In effect, translation operates not with sentences but with utterances, i.e. units of discourse characterized by their use-value in communication. In translation, it is always necessary to aim at equivalence of pragmatic meaning, if necessary at the expense of semantic equivalence. Pragmatic meaning thus overrides semantic meaning. We may therefore consider a translation to be primarily a pragmatic reconstruction of its source text. (p. 28) Thus, Qur’anic contexts in which the QIPVs are employed play a vital role in determining which illocutionary force emerges in each context. The role of context in appreciating the intended meaning of a given language structure is so sensitive that the same language structure “can have opposite interpretations, depending on the context in which it is processed” (Gutt, 1998, p. 49).
5.5.3 Conversational implicatures The theories of conversational maxims and conversational implicatures are relevant to the topic at hand because they assist in the perception of QIPV idiomaticity. In addition, the knowledge of conversational maxims and conversational implicatures, according to Fawcett (2001), should constitute part of the translator’s competence because “different languages will apply the principles in different ways in different situations” (p. 124).
Research methodology 111 As discussed in Chapter 4, Grice proposes two levels for consideration when inferring a given conversational implicature: the level of what is said and the level of what is implicated. When the addresser violates a maxim at the level of what is said, the addressee is “entitled to assume that that maxim, or at least the overall cooperative principle, is observed at the level of what is implicated” (Grice, 1975, p. 52). To do justice to this perspective, Grice’s two levels of analysis are considered in this book’s study. Scrutinizing the data at the first level enables us to account for its propositional, linguistically encoded, semantic meaning (i.e. the locutionary act), whereas doing so at the second level allows us to underscore its pragmatic properties and, in turn, to determine the intended communicative meaning (i.e. the illocutionary force).
5.5.4 Adequacy/inadequacy Translation first and foremost is “concerned with meaning” (Dickins et al., 2002, p. 52). The crux of it “lies in the preservation of ‘meaning’ across two different languages” (House, 1977, p. 25), and the “relevance question for translation is how far the meaning of a text produced in language A can be expressed in a text in language B” (House, 2009, p. 15). Given this and the multilayered nature of textual, semantic, and pragmatic meanings of QIPVs, I believe that analysing and assessing the adequacy/inadequacy of the equivalents provided for them ought to be measured in a multidimensional manner. At the semantic level of the analysis and assessment, the metaphorical QIPVs must be analysed and assessed according to the framework proposed by Newmark (1988) (i.e. object, image, and sense). This framework has been adopted by some researchers translating Qur’anic metaphors into English (cf El-Zeiny, 2011; Zahid, 2009). The figurative QIPVs, however, should be analysed and assessed by attending to their literal/transparent meanings and figurative/nontransparent meanings. Hence, at this level, the chosen metaphorical and figurative QIPVs are compared to their TL ‘equivalents’, provided by the selected translators. The procedures employed by these translators in rendering the QIPVs into English are identified, the adequacy or inadequacy of the ‘equivalents’ provided is measured, and the most adequate procedure is recommended. However, at the pragmatic level of the QIPVs’ analysis and assessment, when the focus is on contextual and communicative properties, the QIPVs ought to be analysed and assessed in light of the speech acts and conversational implicature theories. The chosen QIPVs are then compared to their TL ‘equivalents’, provided by the selected translators, to reveal the procedures employed by the translators in rendering the QIPVs into English. The most adequate procedure will be recommended. At both the semantic level and the pragmatic level of analysis, the adequacy or inadequacy is assessed according to closeness to the intended meaning as ascribed by the authoritative exegetical and lexicographical works (i.e. tertium comparationis). At the semantic level, closeness to the denotative meaning is considered. By denotative meaning, we mean the cognitive, propositional, literal, or referential
112
Research methodology
meaning (cf Dickins et al, 2002; Crystal, 1997). The notion of denotation from which this type of meaning is derived signifies an objective “relationship between a word and the reality to which it refers” (Crystal, 1997, p. 418). Denotative meaning, as delineated by Dickins et al. (2002), is “that kind of meaning which is fully supported by ordinary semantic conventions, such as the convention that ‘window’ refers to a particular kind of aperture in a wall or a roof” (p. 52). Moreover, Dickins et al. (2002) suggest that denotative meaning is the semantic equivalence or synonymy of words or expressions when they are synonyms of each other. This holds true not only between words and expressions within a single language but also between words and expressions from two or more languages. They exemplify this point by juxtaposing the English expression maternal uncle with the Arabic word khāl (in its one sense), which “covers exactly the same range of meanings and [is] therefore fully synonymous” (p. 53). More importantly, denotative meaning reflects the informative level of meaning. When engaging with sacred scriptures in general and the Qur’an in particular, this level of meaning must be given priority over the other levels (i.e. expressive and aesthetic) when a conflict among them occurs (El-Zeiny, 2011). Scholars (cf Newmark, 1988; El-Zeiny, 2011) have proposed three parameters to measure faithfulness to the denotative meaning: semantic deviation, over-translation, and under-translation. Semantic deviation means “giving the wrong denotative meaning” (El-Zeiny, 2011, p. 257). Over-translation, however, means a “translation that gives more detail than its corresponding SL unit. Often a more specific word” (Newmark, 1988, p. 284). Lastly, under-translation means a translation that “gives less detail and is more general than the original” (Newmark, 1988, p. 285). Further, Newmark (1988) asserts that “[m]ost translations are undertranslations, but their degree of under-translation is too high” (p. 285). At the pragmatic level, the adequacy or inadequacy of equivalents given to the QIPVs should be assessed according to the faithfulness of equivalents to the connotative meaning of the QIPVs. The notion of connotation from which connotative meaning is derived denotes the “personal associations aroused by words” (Crystal, 1997, p. 417). Thus, by connotative meaning, we mean the “associations, which over and above the denotative meaning of an expression, form part of its overall meaning” (Dickins et al., 2002, p. 66). Dickins et al. (2002) point out that synonyms typically have different overall semantic effects even in a single language. A given expression has its own overtones that differentiate it from its synonym, which has its own overtones. These overtones constitute the connotative meaning. On this, Dickins et al. (2002), write, The meaning of a text comprises a number of different layers: referential content, emotional colouring, cultural associations, social and personal connotation, and so on. The many-layered nature of meaning is something translators must never forget . . . in fact, of course, connotative meanings are many and varied, and it is common for a single piece of text, or even a single expression, to combine more than one kind into a single overall effect. (p. 66)
Research methodology 113 Here I cite Newmark’s theoretical differentiation between a literary text and a nonliterary text, which he bases on the text’s denotations and connotations and the priority given to either of them by a translator. Newmark (1988) writes, [I]n a non-literary text the denotations of a word normally come before its connotations. But in a literary text, you have to give precedence to its connotations, since, if it is any good, it is an allegory, a comment on society, at the time and now, as well as on its strict setting. From a translator’s point of view this is the only theoretical distinction between a non-literary and a literary text. In fact, the greater the quantity of a language’s resources (e.g. polysemy, word-play, sound-effect, metre, rhyme) expended on a text, the more difficult it is likely to be to translate, and the more worthwhile. (pp. 16–17) Given all these, I must emphasize two essential things: frst, the centrality of context in determining which kind of meaning is in play. Dickins et al. (2002) assert that “as with denotative meaning, being receptive of connotative meaning is a matter of considering words and phrases within the particular context in which they occur” (p. 73), and second, the relevance of appreciating and identifying which kind of meaning is in use to our proposed model for analysing and assessing the translation of QIPVs. Identifying the specifc type of meaning will help us in the process of analysing and assessing the adequacy or inadequacy of the equivalents provided for the QIPVs, which need to be carefully evaluated in a multidimensional way, as mentioned earlier. At this point, I should confirm that in my endeavour to account for the (un)translatability of the QIPVs, the extent to which translators of the Qur’an are successful in providing the English functional-pragmatic equivalent of the QIPVs is my main concern (see section 5.1). Different scholars have proposed different views on the controversial notion of equivalence. Scholars such as Catford (1965), Nida and Taber (1969), Toury (1980), Koller (1995), and House (2001a) have argued for the notion and highlighted its importance and relevance to translation. Other scholars, such as Snell-Hornby (1988), have argued against it, declaring its inappropriateness and irrelevance to translation. That being said, several types of equivalence have been proposed by translation theorists. To name just a few, Nida (1974) proposed the formal equivalence and the dynamic equivalence; Catford (1965) proposed formal equivalence, semantic equivalence, and situational equivalence; and House (1977, 2001a, 2001b) proposed the functional-pragmatic equivalent (cf Aldahesh, 2009). Among these types of equivalence, House’s functional-pragmatic equivalence is the most appropriate type for this study. This is because it comprises the crux of many of the aforementioned types of equivalence and because it allows for the “preservation of ‘meaning’ across two different languages and cultures” (House, 2001b, p. 247).
114
Research methodology
5.6 Model for analysing and assessing translations of QIPV Drawing on the aforementioned theories, I propose the following model for analysing and assessing translations of QIPVs.
Qur'anic Idiomatic Phrasal Verbs Literal (Not included)
Idiomatic
Figurative Metaphorical
Analysis/Assessment
At Semantic Level (Literal vs. Figurative Meanings)
Assessment (Faithfulness to the Denotative Meaning)
At Pragmatic Level (Speech Acts Preformed)
Assessment (Faithfulness to the Connotative Meaning)
Analysis/Assessment
At Pragmatic Level (Speech Acts Preformed)
At Semantic Level (Object, Image, and Sense)
Assessment (Faithfulness to the Connotative Meaning)
Figure 5.1 Model for analysing and assessing translations of QIPV
Assessment (Faithfulness to the Denotative Meaning)
Research methodology 115
5.7 Conclusion This chapter explained the methodology employed in this study and devised a working model for analysing and assessing translations of QIPVs. The model incorporates the two categories of QIPVs: metaphorical QIPVs and figurative QIPVs. Metaphorical QIPVs fall under the definition of metaphor and fulfil its two purposes (referential and pragmatic) and three components (object, image, and sense). Figurative QIPVs do not meet these criteria and yet remain semantically nontransparent. The contrastive analysis approach (as introduced and implemented by James, 1980) is adopted as a theoretical framework. Using its two dimensions (i.e. description and comparison) and employing its two levels of analysis (i.e. microlinguistics and macro-linguistics) means that the framework’s application in this study is not confined to the lexical-semantic level. Rather, it covers the pragmatic aspects involved in translating QIPVs into English. Drawing on speech act theory and adopting the contrastive analysis approach, the proposed model attends to the QIPVs’ categories from the semantic and pragmatic perspectives. This chapter argues that the adequacy of the equivalents given to the QIPVs should be determined according to closeness to the intended meaning ascribed by the authoritative exegetical and lexicographical works (i.e. tertium comparationis). Semantically speaking, closeness to the denotative meaning should be considered because it reflects the informative level of meaning and must be given priority over the other levels of meaning (i.e. expressive and aesthetic). Pragmatically speaking, adequacy should be measured according to its faithfulness to the connotative meaning. The goal of the devised model is to gauge the extent to which the selected translators of the Qur’an are successful in providing the English functional-pragmatic equivalence of the QIPVs.
Note 1 This chapter was originally published as a peer-reviewed article appearing on No. 5 of the AWELJ (2016). The AWELJ is the only source for citation purposes. Major amendments, however, have been made to the previous version.
References Abdel Haleem, M. A. S. (2010). The Qur’an: English Translation and Parallel Arabic Text. Oxford University Press. Oxford. Aldahesh, A. Y. (2009). Translating Idiomatic English Phrasal Verbs into Arabic. VDM Verlag Dr. Muller. Saarbrucken, Germany. Baker, M. (1992). In Other Words: A Coursebook on Translation. Routledge. London and New York. Bakhtiar, L. (2011). Concordance of the Sublime Qur’an. Library of Islam. Kazi Publications. Chicago. Bakhtiar, L. (2012). The Sublime Quran. Fifteenth Revised Edition. Kazi Publications. Chicago.
116
Research methodology
Catford, J. (1965). A Linguistic Theory of Translation. Oxford University Press. Oxford. Chesterman, A. (1998). Contrastive Functional Analysis. John Benjamins Publishing Company. Amsterdam and Philadelphia. Crystal, D. (1997). The Cambridge Encyclopedia of the English Language. Cambridge University Press. Cambridge. Dickins, J., Hervey, S., & Higgins, I. (2002). Thinking Arabic Translation. Routledge. London and New York. El-Zeiny, I. (2011). Criteria for the Translation and Assessment of Qur’anic Metaphor: A Contrastive Analytical Approach. Bable, 57 (3), 247–268. Fawcett, P. (2001). Linguistic Approaches. In M. Baker (Ed.), Routledge Encyclopedia of Translation Studies. Routledge. London and New York. Fowler, H. W. (1926). A Dictionary of Modern English Usage. Oxford University Press. Oxford. Grice, H. P. (1975). Logic and Conversation. In P. Cole & J. L. Margon (Eds.), Syntax and Semantics, Volume 3: Speech Acts, pp. 41–58. Academic Press. London. Gutt, E. A. (1998). Pragmatic Aspects of Translation: Some Relevance Theory Observation. In Leo Hickey (Ed.), The Pragmatics of Translation. Cromwell Press. Trowbridge, UK. Hoey, M., & Houghton, D. (2001). Contrastive Analysis and Translation. In M. Baker (Ed.), Routledge Encyclopedia of Translation Studies. Routledge. London and New York. House, J. (1977). A Model for Translation Quality Assessment. TBL Verlag Gunter Narr. Tubingen, Germany. House, J. (2001a). Quality of Translation. In M. Baker (Ed.), Routledge Encyclopedia of Translation Studies. Routledge. London and New York. House, J. (2001b). Translation Quality Assessment: Linguistic Description versus Social Evaluation. Meta, XLVI (2), 243–257. House, J. (2009). Translation. Oxford University Press. Oxford. Iqbal, M. (2000). ʿAbdullah Yūsuf ʿAlī & Muhammad Asad: Two Approaches to the English Translation of the Qur’an. Journal of Quranic Studies, 2 (1), 107–123. James, C. (1980). Contrastive Analysis. Longman. London. Kidwai, A. R. (2005). Review of ‘The Qur’an with a Phrase-by-Phrase English Translation’ by Ali Quli Qara’i. Islamic Studies, 44 (2), 270–272. Kidwai, A. R. (2007). Bibliography of the Translations of the Meanings of the Glorious Qur’an into English1649–2002: A Critical Study. King Fahd Qur’an Printing Complex. Madinah, Saudi Arabia. Koller, W. (1995). The Concept of Equivalence and the Object of Translation Studies. Target, 7 (2), 191–222. Larson, M. (1984). Meaning-Based Translation: A Guide to Cross Language Equivalence. University Press of America. New York. Leech, G. (1983). Principles of Pragmatics. Longman. London. Marogy, A. E. (2010). Kitab Sibawayhi: Syntax and Pragmatics. Brill. Leiden and Boston. Munday, J. (2001). Introducing Translation Studies: Theories and Applications. Routledge. London and New York. Newmark, P. (1981). Approaches to Translation. Pergamon Press. Oxford. Newmark, P. (1988). A Textbook for Translation. Prentice Hall. New York, London, Sydney, Toronto, and Tokyo. Nida, E. (1964). Towards a Science of Translating: With Special Reference to Principles and Procedures Involved in Bible Translating. Brill. Leiden, Netherlands.
Research methodology 117 Nida, E. (1974). Words and Thoughts. The Bible Translators, 25 (3), 339–343. Nida, E. (1975). Componential Analysis of Meaning. Approaches to Semiotics 5 [AS]. Mouton. The Hague. Nida, E., & Taber, C. (1969). The Theory and Practice of Translation. Brill. Leiden. Qarāʾī, A. Q. (2004). The Qur’an With a Phrase-by-Phrase English Translation. ICAS Press. London. Rizvi, S. H. (2006). Review of ‘The Qur’an with a Phrase-by-Phrase English Translation’ by Ali Quli Qara’i. Journal of Qur’anic Studies, 8 (2), 128–131. Robinson, N. (1977). Sectarian and Ideological Bias in Muslim Translations of the Qur’an. Islam and Christian-Muslim relations, 8 (3), 261–278. Searle, J. R. (1969). Speech Acts: An Essay in the Philosophy of Language. Alden & Mowbray Ltd. Oxford. Snell-Hornby, M. (1988). Translation Studies: An Integrated Approach. John Benjamins Publishing Company. Amsterdam and Philadelphia. Toury, G. (1980). In Search of a Theory of Translation. The Porter Institute for Poetics and Semiotics. Tel Aviv University. Tel Aviv. Wonderly, W. L. (1971). Bible Translations for Popular Use. Helps for Translators VII. United Bible Society. New York. Zafrulla Khan, M. (1971). The Quran: The Eternal Revelation Vouchsafed to Muhammad the Seal of the Prophets. Curzon Press. London. Zahid, A. (2009). A Model for Metaphor Translation: Evidence from the Holy Qur’an. Retrieved from http://zahid66.arabblog.com/archive/2009/808744.html
6
Qualitative data analysis, assessment, and discussion
6.1 Case study 1 wa-yasʾalūnaka ʿan al-maḥīḍi qul huwa adhan fa-iʿtazilū al-nisāʾa f ī al-maḥīḍi wa-lā-taqrabūhunna ḥattá yaṭhurna fa-idhā taṭahharna fa-ʾtūhunna min ḥaythu amarakum Allāhu inna Allāha yuḥibbu al-tawwabīna wa-yuḥibbu al-mutaṭahhirīna. (Qur’an, 2:222)
6.1.1 Tertium comparationis This verse refers to sexual relations by a husband and a wife during the wife’s menstruation. It was revealed when one of the Prophet’s companions, Abu al-Daḥdāḥ (d. AH 3/624 CE), asked him, what should we do with women during their menses? (al-Nisābūrī, 1994, p. 67). According to the narration, Arabs in the pre-Islamic era would emulate Jews and Zoroastrians as to their treatment of women during their menses. They would stop eating, drinking, and living with women under one roof. Hence, this verse describes menstruation as a painful condition and prohibits husbands from having sexual intercourse with their wives during that time. The verse also permits husbands to have sexual intercourse with their wives when the menses are over (cf al-Zamakhsharī, 1998, p. 1/432; al-Rāzī, 1981, p. 6/67; al-Ṭabrasī, 1960, p. 2/319; Ayoub, 1984, p. 1/225). The verse at hand is interpreted by al-Maḥallī and al-Suyūṭī (2007) as follows: They will ask you about the monthly period, that is, menstruation and the spot in which it occurs, and how should one treat women during it. Say: ‘It is an ailment, filth, or the place whence it issues is so; so part with women, refrain from sexual intercourse with them, in the monthly period, in this time, or in the part affected; and do not approach them, for sexual intercourse, until they are pure . . . that is, until they have cleansed themselves after its cessation; when they have cleansed themselves, then come to them, in sexual intercourse, as God has commanded you’, by avoiding it, the female organ, during menstruation and not resorting to any other part. Truly, God
Data analysis, assessment, and discussion 119 loves, that is, He rewards and honours, those who repent, of sins, and He loves those who cleanse themselves, from impurities. (p. 40) The QIPV used in this verse is iʾtūhunna min, which is a metaphorical combination of the verb ʾatá and the preposition min. The verb iʾtūhunna is an imperative mood of the verb ʾatá, which is derived from the root (A-T-Y). Derivatives of the root (A-T-Y) occur 549 times in the Qur’an. They occur 534 times as verbs, ten times as active participles, two times as passive participles, and three times as verbal nouns. The 534 verbs derived from this root appear with and without prepositions in the Qur’anic texture. They occur 387 times without prepositions and 147 times with prepositions. Of the 147 verb-preposition structures, 111 are QIPVs and 36 are nonidiomatic. The verbs derived from the root (A-T-Y) combine with four different prepositions (i.e. baʾ, ʿalá, min, and f ī) to form QIPVs. They combine with the preposition baʾ 97 times, with the preposition ʿalá twice, with the preposition min twice, with the preposition ʿan once, and with the preposition f ī once. For further details on the frequencies of the root (A-T-Y) in the Qur’an, see Appendix II. The high frequency of the root (A-T-Y) and its derivatives indicates the important status of this root and the concepts derived from it in the Qur’an. Badawi and Abdel Haleem (2008) point out that The comparison of the frequency of occurrence of the various derivatives of a certain root and also of the total number of occurrences of all derivatives of this root with other roots should be an indication of the relative importance the Qur’an ascribes to concepts covered by the various roots. (p. xx) The verb ʾatá essentially denotes ‘to come’. This verb occurs in the Qur’an in different contexts to signify different meanings. These meanings include ‘to come’ (Qur’an, 61:6), ‘to be (in a place)’ (Qur’an, 20:69), ‘to revert’ (Qur’an, 12:93), ‘to come upon’ (Qur’an, 51:42), ‘to bring out’ (Qur’an, 5:108), and ‘to come up with’ (Qur’an, 21:47), among other meanings (cf Badawi & Abdel Haleem, 2008, p. 8). For a fuller description of the Qur’anic meanings of this verb and its derivatives, see Appendix I. In this context, however, the preposition min denotes ibtidāʾ al-ghāyah (start of destination). That is, it indicates the place from which a husband should start the act of sexual intercourse (Dāwood, 2002, p. 1/91). The semantic properties of the metaphorical QIPV at hand unfold as follows: The object of the metaphor instructs Muslim men about the right way to have sexual intercourse with their respective wives when the wives’ respective menses are over. The image of the metaphor depicts husbands’ approaching their wives.
120
Data analysis, assessment, and discussion The sense of the metaphor includes husbands having sexual intercourse with their respective wives when the wives’ respective menses are over, as God has directed them.
Moreover, the pragmatic properties of this QIPV unfold as follows: At the what-is-said level (locutionary act), O husbands, come to your wives from where God has directed you. At the what-is-implicated level (illocutionary force), employing the QIPV iʾtūhunna min (literally, ‘come to them [your wives] from’), in this husband– wife sexual relationship context fouts the maxim of QUANTITY (make your contribution as informative as required for the current purposes of the exchange). By using the verb iʾtūhunna (come to them [your wives]), the addresser purposefully makes his contribution less informative than is required for the purpose of the context at hand. The communicative purpose of violating this maxim is to avoid explicit reference to such a sensitive issue (i.e. sexual intercourse between a husband and a wife). The intended implicature here (i.e. EUPHEMISIM) is played to implicate that “the explicit mention of sexual intercourse is taboo and incongruent with the sanctity of the Qur’an” (Farghal, 1995, p. 372). Hence, the implicature in this verse is to achieve politeness, which constitutes one of the key Qur’anic discourse characteristics (e.g. see Qur’an, 4:23 and 43). The aim is to protect the modesty of the addressee and to prevent loss of face. Therefore, the illocutionary force of this Qur’anic context may be presented as follows: O husbands, have sexual intercourse with your wives as God has directed you when their respective menses are over.
6.1.2 English equivalents provided Pickthall (1971, p. 43): They question thee (O Muhammad) concerning menstruation. Say: It is an illness, so let women alone at such times and go not in unto them till they are cleansed. And when they have purified themselves, then go in unto them as Allah hath enjoined upon you. Truly Allah loveth those who turn unto Him, and loveth those who have a care for cleanness. Yūsuf ʿAlī (1991, pp. 89–90): They ask thee concerning women’s courses. Say: They are a hurt and a pollution: So keep away from women in their courses, and do not approach them until they are clean. But when they have purified themselves, ye may approach them in any manner, time, or place ordained for you by Allah. For Allah loves those who turn to Him constantly and He loves those who keep themselves pure and clean.
Data analysis, assessment, and discussion 121 Arberry (1955, 1/59): They will question thee concerning the monthly course. Say: ‘It is hurt; so, go apart from women during the monthly course, and do not approach them till they are clean. When they have cleansed themselves, then come unto them as God has commanded you.’ Truly, God loves those who repent, and He loves those who cleanse themselves. Shakir (2011, p. 21): And they ask you about menstruation. Say: It is a discomfort; therefore, keep aloof from the women during the menstrual discharge and do not go near them until they have become clean; then when they have cleansed themselves, go in to them as Allah has commanded you; surely Allah loves those who turn much (to Him), and He loves those who purify themselves. Zafrulla Khan (1971, p. 35): They enquire from thee as to consorting with their wives during their monthly courses. Tell them: It is harmful, so keep away from women during their monthly courses and do not consort with them until they are clean. But when they have washed themselves clean, consort with them as Allah has commanded you. Indeed, Allah loves those who turn to Him often and He loves those who are clean and pure. al-Hilālī and Khān (1997, p. 48): They ask you concerning menstruation. Say: that is an Adhá (a harmful thing for a husband to have a sexual intercourse with his wife while she is having her menses), therefore keep away from women during menses and go not unto them till they have purified (from menses and have taken a bath). And when they have purified themselves, then go in unto them as Allah has ordained for you (go in unto them in any manner as long as it is in their vagina). Truly, Allah loves those who turn unto Him in repentance and loves those who purify themselves (by taking a bath and cleaning and washing thoroughly their private parts, bodies, for their prayers, etc.). Asad (2011, p. 59): And they will ask thee about (women’s) monthly courses. Say: “It is a vulnerable condition. Keep, therefore, aloof from women during their monthly courses, and do not draw near unto them until they are cleansed; and when they are cleansed, go in unto them as God has bidden you to do”. Verily, God loves those who turn unto Him in repentance, and He loves those who keep themselves pure.
122
Data analysis, assessment, and discussion
Abdel Haleem (2010, p. 36): They ask you (Prophet) about menstruation. Say ‘Menstruation is a painful condition, so keep away from women during it. Do not approach them until they are cleansed; when they are cleansed, you may approach them as God has directed you. God loves those who turn to Him, and He loves those who keep themselves clean’. Qarāʾī (2004, p. 49): They ask you concerning (intercourse during) menses. Say, ‘It is hurtful’ [Footnote: Or ‘offensive’]. So keep away from wives during the menses [Footnote: That is, ‘refrain from sexual intercourse’], and do not approach them till they are clean. And when they become clean, go into them as Allah has commanded you. Indeed, Allah loves the penitent and He loves those who keep clean. Bakhtiar (2012, p. 32): They ask you about menstruation. Say: It is an impurity, so withdraw from your wives during menstruation. Come not near them (f) until they cleanse themselves. And then when they (f) cleanse themselves, approach them (f) as God commanded you. Truly, God loves the contrite and He loves the ones who cleanse themselves.
6.1.3 Analysis Table 6.1 outlines the performance of the selected translators in dealing with the QIPV employed in the verse at hand. Except for Arberry, who reproduces the metaphor in the TL without adding a sense to it, all the selected translators opted to reduce the metaphor to sense. By reproducing the metaphor in the TL without adding a sense to it, Arberry mistakes the literal meaning of the QIPV for its metaphorical meaning. This gives rise to deviation from the functional-pragmatic equivalence. More importantly, all the translators (except for al-Hilālī and Khān and for Qarāʾī) avoided explicit reference to the implied sense of a husband and a wife having sexual intercourse. In so doing, they well achieved the intended communicative purpose (i.e. politeness) of employing the QIPV at hand in such a delicate context. The exception here is the translation of al-Hilālī and Khān and that of Qarāʾī. Al-Hilālī and Khān added an in-text gloss to their provided equivalent (i.e. ‘go in unto them in any manner as long as it is in their vagina’). Similarly, Qarāʾī overtly employed the terms intercourse and sexual intercourse in his translation of this verse, in-text gloss, and footnote, respectively. The added glosses, although further illuminating the meaning, have resulted in translations’ failing to protect the modesty of
Data analysis, assessment, and discussion 123 Table 6.1 Translators’ performance when dealing with the QIPV used in the Qur’an, 2:222 Translator
Adequacy (functionalpragmatic equivalence)
Type of translational pitfall
Procedure
Pickthall
Inadequate
Reducing the metaphor to sense
Yūsuf ʿAlī
Adequate
Covertly erroneous error (register shift—providing EIPV) None
Arberry
Inadequate
Shakir
Inadequate
Zafrulla Khan al-Hilālī and Khān Asad
Inadequate Inadequate Inadequate Adequate
Overtly erroneous error (Mistaking the literal for the metaphorical meaning) Covertly erroneous error (register shift—providing EIPV) Overtly erroneous error (semantic deviation) Covertly erroneous error (register shift—providing EIPV) Covertly erroneous error (register shift—providing EIPV) None
Abdel Haleem Qarāʾī
Inadequate
Covertly erroneous error (register shift—providing EIPV)
Bakhtiar
Adequate
None
Reducing the metaphor to sense Reproducing the metaphor in the TL without adding a sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense + providing parenthetical glosses Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense + providing parenthetical glosses Providing footnotes Reducing the metaphor to sense
the addressees. Hence, they have failed to attend to the aforementioned communicative purpose. The selected translators, however, have exhibited varying degrees of faithfulness to the denotative meaning of the QIPV in question. Zafrulla Khan provided ‘consort with them’, where the obsolete verb ‘consort’ is used knowing that it denotes the sense of ‘to associate’, ‘to agree or harmonies’, and ‘to accompany’ (cf Macquarie Dictionary, 2005, p. 314). All these senses are more general than the original and stop short of capturing its denotative meaning—that is, ‘to sexually intercourse’. Hence, a semantic deviation pitfall has been fallen into here. Alternatively, Shakir, al-Hilālī and Khān, Asad, and Qarāʾī have drawn on Pickthall and provided ‘then go in (un)to them’ as an equivalent for fa-ʾtūhunna min. Using the informal EIPV ‘go in (un)to’ as an equivalent of the formal- and
124
Data analysis, assessment, and discussion
high-register QIPV has meant that the translators experienced a register shift despite partially attending to the denotative meaning. This caused to a mismatch of the functional-pragmatic equivalence. Yūsuf ʿAlī, Abdel Haleem, and Bakhtiar provided ‘approach them’. This translation sounds adequate and matches the functional-pragmatic equivalence because the translators successfully captured both the denotative meanings and the connotative meaning ascribed earlier. However, the footnote by Abdel Haleem that alerted the reader that “The Arabic expressions used here are clear euphemisms” has rendered his translation the most adequate at the informative level of meaning.
6.2 Case study 2 inna al-ladh īna yakfurūna bi-āyāti-illāhi wa-yaqtulūna al-nabiyyīna bighayri ḥaqqin wa-yaqtulūna al-ladhīna yaʾmurūna bi-al-qisiṭi min al-nāsi fabashshirhum bi-ʿdhābin alīm. (Qur’an, 3:21)
6.2.1 Tertium comparationis This verse refers to the disbelievers who commit crimes of unjustly killing prophets, killing people, and killing those who command that justice be done. It provides the disbelievers with ‘good news’ of a painful torment as a result of their wrongdoings. The verse is interpreted by Qur’an exegetes as follows: Those who disbelieve in the signs of God and slay (yaqtulūna, is also read as yuqātilūna, ‘they fight against’) the prophets without right, and slay those who enjoin to equity, to justice, and these are the Jews, who are reported to have killed forty-three prophets and to have been forbidden this by a hundred and seventy devout worshippers among them, each of whom was killed immediately. So, give them good tidings, let them know, of a painful chastisement. The use of ‘good tidings’ here is meant as a sarcastic ridicule of them. (al-Maḥallī & al-Suyūṭī, 2007, p. 57) The QIPV used in this verse is bashshirhum bi-. It is employed two other times in the Qur’an: 9:3 and 84:24. The QIPV is a metaphorical combination of the verb bashshara and the preposition baʾ. The verb bashshara is derived from the root (B-SH-R), whose derivatives occur 123 times in the Qur’an. They occur 48 times as verbs, 19 times as active participles, 19 times as verbal nouns, and 37 times as proper nouns. The 48 verbs derived from this root appear with and without prepositions in the Qur’anic texture. They appear 19 times without prepositions and 29 times with prepositions. Of the 29 verb-preposition structures, 20 are nonidiomatic and nine are QIPVs. The verbs derived from the root (B-SH-R) combine with two different prepositions (i.e. baʾ and ʿalá,) to form QIPVs. They combine with the preposition baʾ eight times and with the preposition ʿalá once.
Data analysis, assessment, and discussion 125 For further details on the frequencies of the root (B-SH-R) in the Qur’an, see Appendix II. The primary significance of the verb bashshara is ‘to convey good news’ (cf Badawi & Abdel Haleem, 2008, p. 93). For fuller descriptions of the Qur’anic meanings of this verb and its derivatives, see Appendix I. The preposition baʾ in this context, however, signifies al-ilṣāq al-maʿnawī (abstract affixation). It abstractly affixes the noun phrase governed by it (i.e. painful torment) to the object of the verb (i.e. the pronoun hum [the disbelievers]) (Dāwood, 2002, p. 1/513). The semantic properties of this metaphorical QIPV unfold as follows: The object of the metaphor tells the disbelievers about the serious consequences of their wrongdoings. The image of the metaphor conveys good news to them. The sense of the metaphor gives them the bad news of painful torment. Furthermore, the pragmatic properties of this QIPV unfold as follows: At the what-is-said level (locutionary act), O Muhammad, give the disbelievers the good news of painful torment. At the what-is-implicated level (illocutionary force): taking the aforementioned propositional linguistically encoded meaning would semantically contradict the meaning of the object of the preposition baʾ and its adjective (i.e. ʿadhābin alīm [a painful torment]). This is because the verb bashshara is typically used to announce good tidings (not bad tidings such as painful torment). This verb normally signifes ‘plus positive news’. It is, however, markedly employed in this Qur’anic context to signify ‘minus positive news’ (Dweik & Abu Shakra, 2011, p. 12). Pragmatically speaking, such a superficial contradiction is made by the addresser to flout the maxim of RELATION (be relevant). This infringement of the RELATION maxim is done for the communicative purpose of getting across the implied meaning, viz. IRONY. Arabic linguists refer to this usage as al-istiʿāratu al-tahakkumiyah (ironic allegory) (cf al-Rāzī, 1981, p. 7/233; Dāwood, 2002, p. 1/512). The implicature played here is that the addresser (God) intends through the addressee (the Prophet) to scorn and ridicule those who ignore his revelations, unjustifiably kill prophets, and kill people who command that justice be done. In so doing, the addresser employs the QIPV bashshara bi- (to convey good tidings) in a context where a painful torment is the ‘good tidings’ being referred to. In addition, the QIPV is used here in its imperative mood. Use of the imperative mood to perform irony is quite common in the Qur’anic discourse (Marogy, 2010, p. 67). Irony is a rhetorical tool frequently employed in the Qur’an to threaten “unbelievers with the punishment of hell and all sorts of pains awaiting them” (Marogy, 2010, p. 67). Accordingly, the illocutionary force of this Qur’anic context can be presented as follows: O Muhammad, threaten the disbelievers with a painful torment.
126
Data analysis, assessment, and discussion
6.2.2 English equivalents provided Pickthall (1971, p. 65): Lo! those who disbelieve the revelations of Allah, and slay the prophets wrongfully, and slay those of mankind who enjoin equity: promise them a painful doom. Yūsuf ʿAlī (1991, pp. 131–132): As to those who deny the Signs of Allah and in defiance of right, slay the prophets, and slay those who teach just dealing with mankind, announce to them a grievous penalty. Arberry (1955, p. 1/75): Those who disbelieve in the signs of God, and slay the Prophets without right, and slay such men as bid to justice—do thou give them the good tidings of a painful chastisement. Shakir (2011, p. 32): Surely (as for) those who disbelieve in the communications of Allah and slay the prophets unjustly and slay those among men who enjoin justice, announce to them a painful chastisement. Zafrulla Khan (1971, p. 50): To those who reject the Signs of Allah and seek to kill the Prophets and those from among the people who enjoin equity, without just cause, announce thou a painful chastisement. al-Hilālī and Khān (1997, p. 70): Verily! Those who disbelieve in the Ayāt (proofs, evidences, verses, lessons, signs, revelations, etc.) of Allāh and kill the Prophets without right, and kill those men who order just dealings, . . . then announce to them a painful torment. Asad (2011, p. 84): Verily, as for those who deny the truth of God’s messages, and slay the prophets against all right, and slay people who enjoin equity—announce unto them a grievous chastisement.
Data analysis, assessment, and discussion 127 Abdel Haleem (2010, p. 53): Give news of an agonizing torment to those who ignore God’s revelations, who unjustifiably kill prophets, who kill those who command that justice is done. Qarāʾī (2004, p. 73, emphasis in original): Those who defy Allah’s signs and kill the prophets unjustly, and kill those who call for justice from among the people, inform them of a painful punishment. Bakhtiar (2012, p. 47): Truly, those who are ungrateful for the signs of God and kill the Prophets without right and kill those who command to equity from among humanity, then, give you to them the good tidings of a painful punishment.
6.2.3 Analysis Table 6.2 outlines the performance of the selected translators in dealing with the QIPV employed in the verse at hand. The selected translators adopt two approaches to rendering the QIPV at hand. The approach by Arberry and Bakhtiar was to reproduce the metaphor in the TL without adding a sense to it. By adopting this source-text-oriented procedure, they have mistaken the literal meaning of the QIPV for its metaphorical meaning and have lapsed into a semantic deviation. This is because the TL does not allow for the transference of the metaphor in question. The nonsense of this translation results from imitating the superficial contradiction involved in the idiomatic expression of giving ‘good tidings of a painful punishment’. This causes a mismatching of illocutionary force and, in turn, the functional-pragmatic equivalence. The translators needed to provide more details to alert the TT reader to the metaphorical use of the verb and preposition in the QIPV. This may be achieved by adding sense to the metaphor, either by inserting a parenthetical gloss or by providing an explanatory footnote. The approach by the remaining selected translators (i.e. Pickthall, Yūsuf ʿAlī, Shakir, Zafrulla Khan, al-Hilālī and Khān, Asad, Abdel Haleem, and Qarāʾī) was to reduce the metaphor to sense. They provided various lexical items as equivalents to the QIPV at hand. Most of them (i.e. Yūsuf ʿAlī, Shakir, Zafrulla Khan, al-Hilālī and Khān, and Asad) provided ‘announce to them’. The verb ‘to announce’ signifies “to make [something] known publicly, in an official way; to say [something] in a loud voice or in an aggressive way” (Oxford Wordpower, 2000, p. 27). As such, it stops short of echoing the ironic tone expressed by the original—hence the under-translation pitfall and the mismatch of the functional-pragmatic equivalence. Similarly, Qarāʾī rendered the QIPV in
128
Data analysis, assessment, and discussion
Table 6.2 Translators’ performance when dealing with the QIPV used in the Qur’an, 3:21 Translator
Adequacy (functionalpragmatic equivalence)
Type of translational pitfall
Procedure
Pickthall
Adequate
None
Yūsuf ʿAlī
Inadequate
Arberry
Inadequate
Shakir
Inadequate
Zafrulla Khan
Inadequate
al-Hilālī and Khān
Inadequate
Asad
Inadequate
Abdel Haleem
Inadequate
Qarāʾī
Inadequate
Bakhtiar
Inadequate
Covertly erroneous error (failure to capture the illocutionary force of the original) Overtly erroneous error (mistaking the literal for the metaphorical meaning) and covertly erroneous error (failure to capture the illocutionary force of the original) Covertly erroneous error (failure to capture the illocutionary force of the original) Covertly erroneous error (failure to capture the illocutionary force of the original) Covertly erroneous error (failure to capture the illocutionary force of the original) Covertly erroneous error (failure to capture the illocutionary force of the original) Covertly erroneous error (failure to capture the illocutionary force of the original) Covertly erroneous error (failure to capture the illocutionary force of the original) Overtly erroneous error (mistaking the literal for the metaphorical meaning of the original)
Reducing the metaphor to sense Reducing the metaphor to sense Reproducing the metaphor in the TL without adding a sense
Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense Reproducing the metaphor in the TL without adding a sense
Data analysis, assessment, and discussion 129 question as ‘inform them’, italicizing the word inform as such, which signifies “to give [somebody] information about [something]” (Oxford Wordpower, 2000, p. 393). Presenting the word inform in italics assigns emphasis. Nevertheless, the intended connotative meaning in the original has not been adequately delivered. Similarly, Abdel Haleem provided ‘give news’, which suggests the neutral sense of telling someone something. The aforementioned drawbacks of ‘announce to them’ and ‘inform them’ are applicable to Abdel Haleem’s ‘give news’. Pickthall, however, provided ‘promise them a painful doom’. The verb ‘promise’ here denotes “to say definitely that you will or will not do [something]; to show signs of [something], so that you expect it to happen” (Oxford Wordpower, 2000, p. 590). In turn, its use by Pickthall in his translation appears to be more suitable than the previous uses are. This is because it captures the intended denotative meaning stated earlier. That being said, even though the selected translators have attended, in varying degrees, to the denotative meaning of the QIPV at hand, none has successfully conveyed its connotative meaning. Failure to capture the aforesaid illocutionary force of the original results in mismatching the functionalpragmatic equivalence of the QIPV. In turn, this means sacrificing a great deal of the meaning not only at the informative level but also at the expressive and aesthetic levels.
6.3 Case study 3 Allāhu lā-ilāha illā huwa al-ḥayyu al-qayyūmu lā-taʾkhudhuhu sinatun wa-lānawmun la-hu mā f ī al-samāwāti wa-mā f ī al-arḍi man dhā al-ladhī yashfaʿu ʿindahu illā bi-idhnihi yaʿlamu mā bayna aydīhim wa-mā khalfahum wa-lāyuḥīṭūna bi-shayʾin min ʿilmihi illā bi-mā shāʾa wasiʿa kursīyyuhu al-samāwāti wa-al-arḍa wa-lā-yaʾūduhu ḥifzuhumā wa-huwa al-ʿaliyyu al-ʿazīmu (Qur’an, 2:255).
6.3.1 Tertium comparationis This verse is called Āyatu al-Kursī (literally, ‘the verse of the Throne’). It has special status in the Qur’an, other Islamic literature, and among Muslims. It is a well-known, typically memorized, and often-recited Qur’anic verse. This is because Āyatu al-Kursī is associated with incredible virtues. It was described by the Prophet Muhammad, in an authentic Hadith ‘tradition’, as “the greatest Āyah in the book of Allah” (Ibn Kathīr, 2003, p. 2/21). The topic of Āyatu al-Kursī is God and his oneness, attributes, ownership, knowledge, and power. It begins by confirming the oneness of God with the statement Allāhu lā-ilāha illā huwa, or “God, there is no god, that is, there is none worthy of being worshipped in (all) existence, except Him” (al-Maḥallī & al-Suyūṭī, 2007, p. 47). It then proceeds to highlight two of his outstanding attributions: al-ḥayyu (the living, the everlasting) and al-qayyūmu (the eternal sustainer) (al-Maḥallī & al-Suyūṭī, 2007, p. 47). Further, the verse affirms God’s ownership by emphasizing that la-hu mā f ī al-samāwāti wa-mā f ī al-arḍi, or “to Him belongs all that is in the heavens and
130
Data analysis, assessment, and discussion
the earth, as possessions, creatures and servants” (al-Maḥallī & al-Suyūṭī, 2007, p. 47). The verse then indicates God’s complete and exclusive knowledge in the declaration yaʿlamu mā bayna aydīhim wa-mā khalfahum wa-lā-yuḥīṭūna bi-shayʾin min ʿilmihi illā bi-mā shāʾa (He knows what lies before them, that is, creation, and what is after them, of the affairs of this world and the Hereafter; and they encompass nothing of His knowledge, that is, they know nothing of what He knows, save such as He wills, to inform of it by way of His messengers). (al-Maḥallī & al-Suyūṭī, 2007, p. 47) The verse also refers to God’s kursī (throne), describing it as wasiʿa al-samāwāti wa-al-arḍa (incorporates the heavens and the earth). Āyatu al-Kursī ends by stressing God’s power and emphasizing that wa-lā-yaʾūduhu ḥifzuhumā wahuwa al-ʿaliyyu al-ʿazīmu, or the preserving of heaven and earth “wearies Him not, does not burden Him; He is the Sublime, above His creation by virtue of His subjugation (of them), the Tremendous, the Great” (al-Maḥallī & al-Suyūṭī, 2007, p. 47). What concerns us about this verse is the employment of the QIPV yuḥīṭūna bi-, a metaphorical combination of the verb yuḥīṭu and the preposition bāʾ. The verb yuḥīṭu is derived from the root (Ḥ-W-Ṭ), whose derivatives occur 28 times in the Qur’an. They occur 17 times as verbs and 11 times as active participles. All the 17 verbs derived from this root appear in combination with the preposition baʾ in the Qur’anic texture to form QIPVs. For further details on the frequencies of the root (Ḥ-W-Ṭ) in the Qur’an, see Appendix II. The verb yuḥīṭu is an active present tense form of the verb aḥāṭa (literally, ‘to besiege or encircle from all directions’). It occurs in the Qur’anic discourse to denote different meanings in different contexts. Its Qur’anic significances include ‘to surround’, ‘to enclose’, ‘to envelop from all sides’ (Qur’an, 18:29), ‘to learn’, ‘to comprehend’, ‘to gain full knowledge of’ (Qur’an, 27:22), ‘to overwhelm’, ‘to engulf’, ‘to take control’ (Qur’an, 2:81), ‘to be encircled from all directions’, ‘to become trapped’, ‘to be in mortal danger’ (Qur’an, 10:22), ‘to be prevented’, ‘to be incapacitated’, and ‘to be besieged’ (Qur’an, 12:66), among other meanings (cf Badawi & Abdel Haleem, 2008, p. 243). For fuller descriptions of the Qur’anic meanings of this verb and its derivatives, see Appendix I. The functional significance of the preposition bāʾ in this context is to indicate al-ilṣāq al-maʿnawy (abstract affixation). That is to say, the main function of the preposition bāʾ is to affix the meaning of the verb yuḥīṭu (which is preceded by the negation particle lā-) to the meaning of its object—that is, shayʾin min ʿilmihi (something of his knowledge) to indicate in an inclusive manner the impossibility of people attaining even a part of God’s knowledge (Dāwood, 2002, p. 1/162). Likewise, al-Zamakhsharī (1998b) indicates that the structure aḥāṭa + bāʾ + ʿilm (knowledge) constitutes a metaphorical combination to signify obtaining full knowledge of something—that is, to entirely appreciate it from all its sides (p. 1/223).
Data analysis, assessment, and discussion 131 In an attempt to elaborate on the meaning of wa-lā-yuḥīṭūna bi-shayʾin min ʿilmihi, al-Rāzī (1981) points out that God, having clarified his ownership of the heavens and earth, also clarified that his ownership of whatever is beyond the heaven and earth is by far greater. This indicates that he is the only one who encompasses such knowledge, which is beyond imagination (p. 7/6). Along these lines, ibn Kathīr (2003) claims that this part of Āyatu al-Kursī “indicates that no one ever acquires knowledge of Allah and His Attributes, except what He conveys to them” (p. 2/28). On the basis of these explanations, we can assume that the semantic properties of the metaphorical QIPV in this context unfold as follows: The object of the metaphor is to realize the knowledge of God. The image of the metaphor depicts besieging or encircling from all directions. The sense of the metaphor is to appreciate the knowledge of God is utterly beyond people’s faculty. In addition, the pragmatic properties of the QIPV unfold as follows: At the what-is-said level (locutionary act), people cannot besiege or encircle God’s knowledge from all its directions. At the what-is-implicated level (illocutionary force), the phrase shayʾin min ʿilmihi (something of his knowledge) is governed by the preposition bāʾ, which links to the verb yuḥīṭūna (literally, ‘to besiege or encircle’). This marks the fouting of the maxims of RELATION (be relative) and MANNER (avoid obscurity of expression). This is apparently because knowledge cannot be materially besieged or encircled. Rather, it is appreciated, perceived, and comprehended to varying degrees. Flouting these maxims occurs not without a communicative purpose, which here is to implicate the sense of INCAPACITATION on the part of people to ‘besiege’ his knowledge from all its directions. Thus, the illocutionary force of this Qur’anic context may be presented as follows: Humans are incapable of encompassing anything of God’s knowledge.
6.3.2 English equivalents provided Pickthall (1971, p. 52): Allah! There is no deity save Him, the Alive, the Eternal. Neither slumber nor sleep overtaketh Him. Unto Him belongeth whatsoever is in the heavens and whatsoever is in the earth. Who is he that intercedeth with Him save by His leave? He knoweth that which is in front of them and that which is behind them, while they encompass nothing of His knowledge save what He will. His throne includeth the heavens and the earth, and He is never weary of preserving them. He is the Sublime, the Tremendous.
132
Data analysis, assessment, and discussion
Yūsuf ʿAlī (1991, pp. 105–106): Allah! There is no god but He—the Living, the Self-subsisting, Eternal. No slumber can seize Him nor sleep. His are all things in the heavens and on earth. Who is there can intercede in His presence except as He permitteth? He knoweth what (appeareth to His creatures as) before or after or behind them. Nor shall they compass aught of His knowledge except as He willeth. His Throne doth extend over the heavens and the earth, and He feeleth no fatigue in guarding and preserving them for He is the Most High, the Supreme (in glory). Arberry (1955, p. 1/65): God there is no god but He, the Living, the Everlasting. Slumber seizes Him not, neither sleep; to Him belongs all that is in the heavens and the earth. Who is there that shall intercede with Him save by His leave? He knows what lies before them and what is after them, and they comprehend not anything of His knowledge save such as He wills. His Throne comprises the heavens and earth; the preserving of them oppresses Him not; He is the All-high, the All-glorious. Shakir (2011, p. 26): Allah is He besides Whom there is no god, the Ever-living, the Self-subsisting by Whom all subsist; slumber does not overtake Him nor sleep; whatever is in the heavens and whatever is in the earth is His; who is he that can intercede with Him but by His permission? He knows what is before them and what is behind them, and they cannot comprehend anything out of His knowledge except what He pleases, His knowledge extends over the heavens and the earth, and the preservation of them both tires Him not, and He is the Most High, the Great. Zafrulla Khan (1971, p. 41): Allah is He save Whom none is worthy of worship, the Ever-Living, the SelfSubsisting and All-Sustaining. Slumber seizes Him not, nor sleep. To Him belongs whatsoever is in the heavens and whatsoever is in the earth. Who is he that dare intercede with Him, except by His permission? He knows all that is before them and all that is behind them, and they cannot encompass aught of His knowledge, except that which He pleases. His knowledge extends over the heavens and the earth, and the care of them wearies Him not. He is the Most High, the Most Great. al-Hilālī and Khān (1997, pp. 57–58): Allāh! Lā ilāha illā Huwa (none has the right to be worshipped but He), Al-Hayyul-Qayyuum (the Ever Living, the One Who sustains and protects all that exists). Neither slumber, nor sleep overtake Him. To Him belongs
Data analysis, assessment, and discussion 133 whatever is in the heavens and whatever is on the earth. Who is he that can intercede with Him except with His Permission? He knows what happens to them (His creatures) in this world, and what will happen to them in the Hereafter. And they will never compass anything of His Knowledge except that which He wills. His Kursī extends over the heavens and the earth, and He feels no fatigue in guarding and preserving them. And He is the Most High, the Most Great. Asad (2011, p. 69): GOD—there is no deity save Him, the Ever-Living, the Self-Subsistent Fount of All Being. Neither slumber overtakes Him, nor sleep. His is all that is in the heavens and all that is on earth. Who is there that could intercede with Him, unless it be by His leave? He knows all that lies open before men and all that is hidden from them, whereas they cannot attain to aught of His knowledge save that which He wills [them to attain]. His eternal power overspreads the heavens and the earth, and their upholding wearies Him not. And he alone is truly exalted, tremendous. Abdel Haleem (2010, p. 43): God: there is no god but Him, the Ever Living, the Ever Watchful. Neither slumber nor sleep overtakes Him. All that is in the heavens and in the earth belongs to Him. Who is there that can intercede with Him except by His leave? He knows what is before them and what is behind them, but they do not comprehend any of His knowledge except what He wills. His throne extends over the heavens and the earth; it does not weary Him to preserve them both. He is the Most High, the Tremendous. Qarāʾī (2004, p. 59): Allah—there is no god except Him—is the Living One, the All-sustainer. Neither drowsiness befalls Him nor sleep. To Him belongs whatever is in the heavens and whatever is on the earth. Who is it that may intercede with Him except with His permission? He knows that which is before them and that which is behind them, and they do not comprehend anything of His knowledge except what He wishes. His seat embraces the heavens and the earth, and He is not wearied by their preservation, and He is the All-exalted, the All-supreme. Bakhtiar (2012, pp. 38–39): God! There is no god but He, The Living, The Eternal. Neither slumber takes Him nor sleep. To Him belongs whatever is in the heavens, and whatever is in and on the earth. Who will intercede with Him but with His permission? He knows what is in front of them, and what is behind them. And
134
Data analysis, assessment, and discussion they will not comprehend anything of His knowledge, but what He willed. His Seat encompassed the heavens and the earth, and He is not hampered by their safe-keeping. And He is The Lofty, The Sublime.
6.3.3 Analysis Table 6.3 outlines the performance of the selected translators in dealing with the QIPV employed in the verse at hand. As evidenced in the aforementioned translations of this verse, the translators represented four approaches to rendering the QIPV in question: 1 2
‘they encompass nothing of His knowledge’ (Pickthall, Zafrulla Khan); ‘they (will never) compass aught of His knowledge’ (Yūsuf ʿAlī, and al-Hilālī and Khān);
Table 6.3 Translators’ performance when dealing with the QIPV used in the Qur’an, 2:255 Translator
Adequacy (functionalpragmatic equivalence)
Type of translational pitfall
Procedure
Pickthall
Adequate
None
Yūsuf ʿAlī
Adequate
None
Arberry
Inadequate
Shakir
Inadequate
Zafrulla Khan
Adequate
Overtly erroneous error (under-translation) Overtly erroneous error (under-translation) None
al-Hilālī and Khān
Adequate
None
Asad
Inadequate
Abdel Haleem
Inadequate
Qarāʾī
Inadequate
Bakhtiar
Inadequate
Overtly erroneous error (under-translation) Overtly erroneous error (under-translation) Overtly erroneous error (under-translation) Overtly erroneous error (under-translation)
Reproducing the metaphor in the TL without adding a sense Reproducing the metaphor in the TL without adding a sense Reducing the metaphor to sense Reducing the metaphor to sense Reproducing the metaphor in the TL without adding a sense Reproducing the metaphor in the TL without adding a sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense
Data analysis, assessment, and discussion 135 3 4
‘they (do not, will not) comprehend not anything of His knowledge’ (Arberry, Shakir, Abdel Haleem, Qarāʾī, Bakhtiar); ‘they cannot attain to aught of His knowledge’ (Asad).
Pickthall, Yūsuf ʿAlī, al-Hilālī and Khān, and Zafrulla Khan reproduced the Qur’anic metaphor in the TL without adding a sense to it. Conversely, Arberry, Shakir, Abdel Haleem, Qarāʾī, Bakhtiar, and Asad reduced the metaphor to a sense. Those who reproduced the metaphor in the TL were relatively successful in capturing the functional-pragmatic equivalence of the original. This is because the TL does permit transference of the metaphor at hand. The translators who opted to reduce the metaphor to a sense, however, lapsed into undertranslation. In other words, by providing ‘they comprehend’ and ‘they attain to’ as equivalents, they provided less and more general details than the intended meaning of the original QIPV. Thus, their translations—albeit partially capturing the sense of incapacitation—lacked the sense of inclusiveness intended in the original. This caused the mismatching of the functional-pragmatic equivalence of the original.
6.4 Case study 4 wa-aṣbaḥa fuʾādu ummi mousá fārighan in kādat la-tubdī bi-hi lawlā an rabaṭnā ʿalá qalbihā li-takūna min al-mū’minīn. (Qur’an, 28:11)
6.4.1 Tertium comparationis This verse refers to the mother of Moses. It portrays the emotional story of the infant Moses and his shocked mother. In her attempt to save her baby from imminent death—the Pharaoh has commanded Egyptian midwives to kill all new born Israelite babies—Moses’s mother placed him into a box and cast him in the river Nile. The box then was retrieved by a member of the Pharaoh’s household. The verse depicts the critical moment when she discovered that her baby had been picked up, at which point the verse says, wa-aṣbaḥa fuʾādu ummi mousá fārighan, or “the heart of Moses’s mother . . . became empty, of everything other than him. [in kādat la-tubdī bi-hi] Indeed . . . she was about to expose him, that is, as being her son, [lawlā an rabaṭnā ʿalá qalbihā] had We not fortified her heart, with patience, that is, We made it at peace, [li-takūna min al-mū’minīn] that she might be of the believers, of those who have faith in God’s promise” (al-Maḥallī & al-Suyūṭī, 2007, p. 442; cf al-Zamakhsharī, 1998a, p. 4/486; al-Rāzī, 1981, p. 24/229; Ibn Qutaybah, 1978, pp. 328–329; al-Aṣfahānī, n.d., p. 1/247). The QIPV used in this Qur’anic context is rabaṭnā ʿalá, a metaphorical combination of the active past tense verb rabaṭa and the preposition ʿalá. The verb rabaṭa is derived from the root (R-B-Ṭ), whose derivatives occur five times in the Qur’an. They occur four times as verbs and once as an active participle. The four verbs derived from this root appear with and without prepositions in the Qur’anic texture. They occur three times with prepositions and once without preposition.
136
Data analysis, assessment, and discussion
All three verb-preposition structures are QIPVs. The verbs derived from the root (R-B-Ṭ) combine with only one preposition—that is, ʿalá—to form QIPVs. For further details on the frequencies of the root (R-B-Ṭ) in the Qur’an, see Appendix II. The primary significance of the verb rabaṭa is “to tie, to tie up, to connect, to unite; to station, garrison; to line up, (of an army) to take up a position; to conclude an agreement; a band, fetters, stables” (Badawi & Abdel Haleem, 2008, p. 344). This verb is used in the Qur’anic discourse to denote ‘to strengthen’ or ‘to fortify’ (cf Badawi & Abdel Haleem, 2008, p. 344). For further descriptions of the Qur’anic meanings of this verb and its derivatives, see Appendix I. The preposition ʿalá in this context signifies al-istiʿlāʾ al-maʿnawy (abstract superiority) (Dāwood, 2002, p. 2/366). It is employed here to imply that ‘strength’ or ‘fortification’ was descended on the heart of Moses’s mother from the exalted God. The QIPV used in this context is metaphorical (cf al-Zamakhsharī, 1998b, p. 1/331), and its semantic properties unfold as follows: The object of the metaphor describes the emotional status of Moses’s mother at that critical moment. The image of the metaphor depicts tying up the heart of Moses’s mother. The sense of the metaphor is about strengthening the heart of Moses’s mother with patience to make it at peace. The pragmatic properties of the QIPV, however, unfold as follows: At the what-is-said level (locutionary act), Moses’s mother was about to disclose the secret had we tied up on her heart. At the what-is-implicated level (illocutionary force), use of the phrase qalbihā (her heart) as an indirect object of the verb rabaṭnā (to tie up) marks the fouting of the maxim of RELATION (be relative). This is because hearts cannot be physically tied up. One can infer that the fouting of this maxim is done for a communicative purpose—that is, to imply the sense of endowing her heart with full REASSURANCE, to fortify it from fear as if it were materially tied up to settle her. Therefore, the illocutionary force of this Qur’anic context may be presented as follows: Moses’s mother was about to disclose the secret had we not endowed reassurance and patience on her heart to strengthen it and make it at peace.
6.4.2 English equivalents provided Pickthall (1971, p. 506): And the heart of the mother of Moses became void, and she would have betrayed him if We had not fortified her heart, that she might be of the believers.
Data analysis, assessment, and discussion 137 Yūsuf ʿAlī (1991, p. 963): But there came to be a void in the heart of the mother of Moses: She was going almost to disclose his (case), had We not strengthened her heart (with faith), so that she might remain a (firm) believer. Arberry (1955, p. 2/87): On the morrow, the heart of Moses’ mother became empty, and she wellnigh disclosed him had We not strengthened her heart, that she might be among the believers. Shakir (2011, p. 254): And the heart of Musa’s mother was free (from anxiety); she would have almost disclosed it had We not strengthened her heart so that she might be of the believers. Zafrulla Khan (1971, p. 378): The heart of the mother of Moses became free from anxiety. Had We not strengthened her heart so that she might be a firm believer in Our grace and mercy, she would have disclosed his identity. al-Hilālī and Khān (1997, p. 517): And the heart of the mother of Mūsā (Moses) became empty (from every thought, except the thought of Mūsā (Mouses)). She was very near to disclose his (case, i.e. the child is her son), had We not strengthened her heart (with faith), so that she might remain as one of the believers. Asad (2011, p. 705): On the morning, however, an aching void grew up in the heart of the mother of Moses, and she would indeed have disclosed all about him had We not endowed her heart with enough strength to keep alive her faith (in Our promise). Abdel Haleem (2010, p. 387): The next day, Moses’ mother felt a void in her heart—if We had not strengthened it to make her one of those who believe, she would have revealed everything about him-
138
Data analysis, assessment, and discussion
Qarāʾī (2004, p. 539): The heart of Moses’ mother became desolate, and indeed she was about to divulge it had We not fortified her heart so that she might have faith (in Allah’s promise). Bakhtiar (2012, p. 367): And it came to be in the morning that the mind of the mother of Moses was that which is empty. Truly, she was about to show him, if We had not invigorated her heart so that she became among the ones who believe.
6.4.3 Analysis Table 6.4 outlines the performance of the selected translators in dealing with the QIPV employed in the verse at hand. The English equivalents assigned to the QIPV in question show that the translators adopted five approaches to rendering the structure lawlā an rabaṭnā ʿalá qalbihā: 1 2 3 4 5
‘if We had not fortified her heart’ (Pickthall and Qarāʾī); ‘had We not strengthened her heart’ (Arberry, Shakir, Zafrulla Khan, and Abdel Haleem); ‘had We not strengthened her heart (with faith)’ (Yūsuf ʿAlī, and al-Hilālī and Khān); ‘had We not endowed her heart with enough strength’ (Asad); ‘if We had not invigorated her heart’ (Bakhtiar).
All the selected translators reduced the metaphor to sense in the TL. This procedure is justifable given that the metaphoric ‘image’ depicted in the Qur’anic expression rabaṭnā ʿalá qalbihā cannot be reproduced in the TL, which lacks such a metaphorical expression. As a result, the original QIPV’s expressive and aesthetic levels of meaning have been acceptably sacrifced for the sake of the informative level. The translators satisfy the informative level of meaning to varying degrees. The phrases ‘to fortify her heart’ and ‘to strengthen her heart’ provided as equivalents by Pickthall, Qarāʾī, Arberry, Shakir, Zafrulla Khan, and Abdel Haleem are not entirely faithful to the original. They lapse into under-translation by being more general and less detailed. Conversely, Yūsuf ʿAlī on one hand and al-Hilālī and Khān on the other provided ‘to strengthen her heart (with faith)’ as an equivalent. As such, they lapsed into over-translation by being more specific and by providing additional details. This is evident from the parenthetical gloss they provide (i.e. ‘with faith’). Yūsuf ʿAlī, however, is more successful than al-Hilālī and Khān because he illuminates the connotative meaning in a footnote: “The mother’s heart felt the gaping void at parting from her son; but her faith in Allah’s Providence kept her from betraying herself” (1991, p. 963). Therefore,
Data analysis, assessment, and discussion 139 Table 6.4 Translators’ performance when dealing with the QIPV used in the Qur’an, 28:11 Translator
Adequacy (functionalpragmatic equivalence)
Type of translational pitfall
Procedure
Pickthall
Inadequate
Yūsuf ʿAlī
Inadequate
Overtly erroneous error (under-translation) Overtly erroneous error (over-translation)
Arberry
Inadequate
Shakir
Inadequate
Zafrulla Khan
Inadequate
al-Hilālī and Khān
Inadequate
Reducing the metaphor to sense Reducing the metaphor to sense + providing parenthetical glosses + providing a footnote Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense + providing parenthetical glosses
Asad
Adequate
Abdel Haleem
Inadequate
Qarāʾī
Inadequate
Bakhtiar
Inadequate
Overtly erroneous error (under-translation) Overtly erroneous error (under-translation) Overtly erroneous error (under-translation) Covertly erroneous error (register shift) and overtly erroneous error (over-translation) None Overtly erroneous error (under-translation) Overtly erroneous error (under-translation) Overtly erroneous error (semantic deviation)
Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense
Yūsuf ʿAlī’s translation is closer to the functional-pragmatic equivalence of the original. That being said, Asad’s translation ‘had We not endowed her heart with enough strength’ appears to be the most faithful to the original. Despite the justifiable mismatching of the original QIPV at the expressive and aesthetic levels of meaning, it nonetheless achieved functional-pragmatic equivalence. Most aspects of the intended meaning are covered in this translation. The use of the word endowed provides an air to the sense of the ‘abstract superiority’ implied in the preposition ʿalá. In addition, modifying the word strength with the preceding quantifier enough imparts a tone of ‘firmness’ implied in the verb rabaṭnā. Finally, the phrase invigorated her heart provided by Bakhtiar as an equivalent to the QIPV rabaṭnā ʿalá qalbihā lapses into semantic deviation. This is because the term to invigorate, though it partially fulfils the denotative meaning, stops short of delivering the connotative meaning. It lacks the tone of ‘firmness’ inferred partially from “to strengthen her heart” and “to fortify her heart” and entirely from Asad’s “to endow her heart with enough strength”.
140
Data analysis, assessment, and discussion
6.5 Case study 5 fa-ḍarabnā ʿalá ādhānihim f ī al-kahfi sīnīna ʿadadā. (Qur’an, 18:11)
6.5.1 Tertium comparationis This verse refers to ahlu al-kahf (the men of the cave), whose tale is recounted in the eighteenth sūrah of the Qur’an, named after the tale as Sūrat al-kahf. The tale of ahlu al-kahf is recapitulated as follows: In the reign of a Roman Emperor [Decius] who persecuted the Christians, seven Christian youths of Ephesus left the town and hid themselves in a cave in a mountain nearby. They fell asleep and remained asleep for some generations or centuries. When the wall which sealed up the cave was being demolished, the youths awoke. They still thought of the world in which they had previously lived. They had no idea of the duration of time. But when one of them went to the town to purchase provisions, he found that the whole world had changed. The Christian religion, instead of being persecuted was fashionable; in fact, it was now the State religion. His dress and speech, and the money which he brought, seemed to belong to another world. This attracted attention. The great ones of the land visited the Cave and verified the tale by questioning the man’s Companions. (Yūsuf ʿAlī, 1991, p. 709) In commentators’ endeavours to interpret the Qur’anic verse in question, they agree that it means ‘We sealed their ears with a veil [ḥijāban]”, or For a fxed number of years, we put them into a deep sleep in the cave, where sounds could not awaken them (cf al-Zamakhsharı̄ , 1998a, p. 3/567; al-Rāzı̄ , 1981, p. 21/84; Ibn Qutaybah, 1978, p. 264; al-Maḥallı̄ & al-Suyūṭı̄ , 2007, p. 312). The QIPV ḍarabnā ʿalá is used in this Qur’anic context. It is a metaphorical combination of the verb ḍaraba and the preposition ʿalá. The verb ḍaraba is derived from the root (Ḍ-R-B), whose derivatives occur 58 times in the Qur’an. They occur 55 times as verbs and three times as verbal nouns. The 55 verbs derived from this root appear with and without prepositions in the Qur’anic texture. They appear 13 times without prepositions and 42 times with prepositions. Of the 42 verb-preposition structures, 34 are QIPVs and eight are nonidiomatic. The verbs derived from the root (Ḍ-R-B) combine with five prepositions (i.e. lām, f ī, baʾ,ʿalá, and ʿan) to form QIPVs. They combine with the preposition lām 25 times, f ī five times, baʾ twice, ʿan once, and ʿalá once. For further details on the frequencies of the root (Ḍ-R-B) in the Qur’an, see Appendix II. Ḍaraba is an active past tense verb that literally means ‘to beat’, ‘to strike’, ‘to battle’, ‘to sting’ (Badawi & Abdel Haleem, 2008, p. 547). It appears in the Qur’anic discourse to signify various meanings in different contexts. Its Qur’anic significances include ‘to beat’ (Qur’an, 47:27 and 43:5); ‘to hit’, ‘to strike’ (Qur’an, 4:34); ‘to make a clanging sound’, ‘to stamp’ (Qur’an, 24:31);
Data analysis, assessment, and discussion 141 ‘to travel about’, ‘to hit the road’ (Qur’an, 3:156); ‘to cut through’ (Qur’an, 20:77); ‘to set forth (a parable)’ (Qur’an, 14:24–5); ‘to compare’, ‘to contrast’ (Qur’an, 13:17); ‘to secure’, ‘to tighten’ (Qur’an, 24:31); ‘to seal’, ‘to raise’, ‘to erect’ (Qur’an, 57:13); and ‘to brand’, ‘to stamp’ (Qur’an, 2:61), among other meanings (cf Badawi & Abdel Haleem, 2008, pp. 547–548). For a fuller description of the Qur’anic meanings of this verb and its derivatives, see Appendix I. Arabic linguists assert that this Qur’anic structure contains an ellipsis. The direct object of the verb ḍaraba (i.e. ḥijāban [a veil]) is omitted here. This ellipsis, they justify, is quite common in idiomatic Arabic to serve rhetorical purposes. They illustrate this phenomenon in the following example: baná ʿalá imraʾatihi (literally, ‘He built on his woman’, pragmatically: ‘He married her’) where the direct object of the verb baná (i.e. Qubbatan [dome, cupola]) is omitted. (cf al-Zamakhsharī, 1998a, p. 3/567; al-Rāzī, 1981, p. 21/84; Ibn Qutaybah, 1978, p. 264). The preposition ʿalá in this Qur’anic context denotes al-istiʿlāʾ al-maʿnawy (abstract superiority). It is used here to indicate the ‘firmness’ of that ḥijāb (veil) on Men’s ears, which put them in deep sleep (cf Dāwood, 2002, p. 1/261–262). The QIPV in this context is metaphorical and its semantic properties unfold as follows: The object of the metaphor describes the status of ahlu al-kahf ‘the men of the cave’ as they hid in the cave. The image of the metaphor depicts beating on ears. The sense of the metaphor puts ahlu al-kahf ‘the men of the cave’ in a deep sleep for many years by sealing their ears. The pragmatic properties of the QIPV, however, unfold as follows: At the what-is-said level (locutionary act) the men of the cave were beaten on their ears for many years while they were hiding in the cave. At the what-is-implicated level (illocutionary force), the inherent ellipsis of the verb’s direct object (i.e. ḥijāban ‘a veil’) infers a violation of the maxim of MANNER (avoid ambiguity, avoid obscurity of expression). In addition, the indirect object of the preposition ʿalá (i.e. ādhānihim ‘their ears’) fouts the maxim of QUALITY (do not say that for which you lack adequate evidence). The communicative purpose of violating these maxims is to implicate the sense of MIRACLE on the part of God. Thus, the illocutionary force of this Qur’anic context may be presented as follows: We miraculously put ahlu al-kahf (the men of the cave) in a deep sleep by sealing their ears with a ḥijāban (veil) in the cave for a fixed number of years where sounds had no effect on waking them.
6.5.2 English equivalents provided Pickthall (1971, p. 380). Then We sealed up their hearing in the Cave for a number of years.
142
Data analysis, assessment, and discussion
Yūsuf ʿAlī (1991, pp. 709–710). Then We drew (a veil) over their ears, for a number of years, in the Cave, (so that they heard not). Arberry (1955, p. 1/317): Then We smote their ears many years in the Cave. Shakir (2011, p. 187): So We prevented them from hearing in the cave for a number of years. Zafrulla Khan (1971, pp. 276–277): So for a number of years We deprived them of news of the outside world. al-Hilālī and Khān (1997, p. 386): Therefore We covered up their (sense of) hearing (causing them, to go in deep sleep) in the Cave for a number of years. Asad (2011, p. 526): And thereupon We veiled their ears in the cave for many a year. Abdel Haleem (2010, p. 295): We sealed their ears (with sleep) in the cave for years. Qarāʾī (2004, p. 406): So We put them to sleep in the cave for several years. [Footnote: “Literally: ‘struck on their ears’ or ‘drew a curtain (or veil) on their ears’”]. Bakhtiar (2012, p. 274): So We sealed their ears in the Cave for a number of years.
6.5.3 Analysis Table 6.5 outlines the performance of the selected translators in dealing with the QIPV employed in the verse at hand.
Data analysis, assessment, and discussion 143 Table 6.5 Translators’ performance when dealing with the QIPV used in the Qur’an, 18:11 Translator
Adequacy (functionalpragmatic equivalence)
Type of translational pitfall
Procedure
Pickthall
Inadequate
Reducing the metaphor to sense
Yūsuf ʿAlī
Adequate
Covertly erroneous error (register shift and Failure to capture the illocutionary force of the original) None
Arberry
Inadequate
Overtly erroneous error (semantic deviation)
Shakir
Inadequate
Zafrulla Khan
Inadequate
al-Hilālī and Khān
Inadequate
Asad
Adequate
Covertly erroneous error (failure to capture the illocutionary force of the original) Covertly erroneous error (failure to capture the illocutionary force of the original) Overtly erroneous error (semantic deviation) Covertly erroneous error (register shift and Failure to capture the illocutionary force of the original) None
Abdel Haleem
Inadequate
Overtly erroneous error (under-translation)
Qarāʾī
Inadequate
Overtly erroneous error (under-translation)
Bakhtiar
Inadequate
Covertly erroneous error (failure to capture the illocutionary force of the original)
Reducing the metaphor to sense + providing parenthetical glosses Reproducing the metaphor in the TL without adding a sense Reducing the metaphor to sense
Reducing the metaphor to sense
Reducing the metaphor to sense + providing parenthetical glosses
Reducing the metaphor to sense Reducing the metaphor to sense + providing a parenthetical gloss Reducing the metaphor to sense + providing a footnote Reducing the metaphor to sense
144
Data analysis, assessment, and discussion
The foregoing translations of the verse in question indicate that the translators adopted seven approaches to rendering the idiomatic expression ḍarabnā ʿalá ādhānihim into English: 1 2 3 4 5 6 7
‘We sealed up their hearing/We sealed their ears/We sealed their ears (with sleep)’ (Pickthall, Bakhtiar, and Abdel Haleem); ‘We drew (a veil) over their ears/We veiled their ears’ (Yūsuf ʿAlī and Asad); ‘We smote their ears’ (Arberry); ‘We prevented them from hearing’ (Shakir); ‘We deprived them of news of the outside world’ (Zafrulla Khan); ‘We covered up their (sense of) hearing (causing them, to go in deep sleep)’ (al-Hilālī and Khān); ‘We put them to sleep’ with a footnote (Qarāʾī).
Arberry attempted to reproduce the metaphor in the TL without adding a sense to it. In so doing, he lapsed into semantic deviation by delivering the literal (locutionary act/prepositional) meaning of the QIPV at hand. This is because the TL does not permit such a reproduction. Hence, he failed to attend to the aforementioned intended meaning of the original and, in turn, to fulfil the functional-pragmatic equivalence. Pickthall, Bakhtiar, and Shakir, however, reduced the metaphor to sense, though the paraphrases that they provided stopped short of fulfilling the aforesaid illocutionary force. Consequently, they lapsed into under-translation, which mismatched the functional-pragmatic equivalence of the original QIPV. Further, Yūsuf ʿAlī, Asad, Abdel Haleem, Zafrulla Khan, Qarāʾī, and al-Hilālī and Khān reduced the metaphor and added a sense to it. They varied to some extent in the closeness of their paraphrases to the intended meaning of the Qur’anic structure. Abdel Haleem’s parenthetical gloss (i.e. ‘with sleep’) brought his paraphrase (i.e. ‘We sealed their ears’) closer to the intended meaning; yet he demonstrated under-translation. This is because what has been lost in Abdel Haleem’s translation, other than the expressive and aesthetic levels of meaning in the original, was the sense of ‘depth’ to which the men slept in the cave. Hence, there is a mismatch in the functional-pragmatic equivalence. In Yūsuf ʿAlī’s, Asad’s, and Qarāʾī’s attempts to compensate for the loss of denotative meaning and connotative meaning, they provided rather lengthy footnotes to elaborate on their translations. Such footnotes bring their translations closer to the functional-pragmatic equivalence of the original. Yūsuf ʿAlī attempted to bring the ‘sense’ of the metaphor closer to the mind of the target reader through examples: Drew (a veil) over their ears i.e. sealed their ears so that they heard nothing. As they were in the Cave they saw nothing. So they were completely cut off from the outer world. It was as if they had died, with their knowledge and ideas remaining at the point of time when they entered the Cave. It is as if a
Data analysis, assessment, and discussion 145 watch stops at the exact moment of some accident, and anyone taking it up afterwards can precisely fix the time of the accident. (1991, pp. 709–710, emphasis in original) Asad, however, has not confned his footnote to details on the literal meaning provided as an equivalent to the Qur’anic structure. He also provides the crux of the classical exegetes’ opinions on this issue: “God caused them to remain cut off—physically or metaphorically—from sounds and the bustle of the outside world. The classical commentators take the above phrase to mean that God ‘veiled their ears with sleep’” (2011, p. 526). Qarāʾī, on contrast to Yūsuf ʿAlī and Asad, paraphrased the verse and elaborated on its literal meaning in a footnote: “literally: ‘struck on their ears’ or ‘drew a curtain (or veil) on their ears in the Cave for several years’” (2004, p. 406). Yet neither Qarāʾī’s paraphrase nor his footnote achieves the functional-pragmatic equivalence. He stopped short of fulflling the informative level of meaning. Hence, the argument proposed earlier against Abdel Haleem’s under-translation applies to Qarāʾī’s translation also. By providing ‘We deprived them of news of the outside world’ as an equivalent to this Qur’anic structure, Zafrulla Khan exercised absolute semantic deviation. His rendering is inadequate at all levels of meaning, including the most-needed informative level. Depriving the men of the cave of news of the outside world does not necessarily mean putting them to sleep, which is the intended meaning (illocutionary force) of the original. This intended meaning is unmistakably substantiated by the macro distance Qur’anic context (i.e. Qur’an, 18:18). Therefore, it must be adequately communicated in the TL. Verse 18 of the same sūrah evidently describes the men’s status while sleeping in the cave: Wa-taḥsabuhum ayqādan wa-hum ruqūdun wa-nuqallibuhum dhāta al-yamīni wa-dhāta al-shimāli wa-kalbuhum bāsitun dhirāʿayhi bi-al-waṣīdi law iṬṬalaʿta ʿalay-him la-wallayta min-hum firāran wa-la-muliʾta min-hum ruʿban. (Qur’an, 18:18) And you would have thought them awake, whereas they were asleep. And We turned them on their right and on their left sides, and their dog stretching forth his two forelegs at the entrance (of the Cave or in the space near to the entrance of the Cave as a guard at the gate). Had you looked at them, you would certainly have turned back from them in flight, and would certainly have been filled with awe of them. (al-Hilālī and Khān, 1997, p. 387) Recognizing this, al-Hilālī and Khān’s rendering of the Qur’anic structure in question appears to be the most adequate. Inserting in-text phrases (i.e. ‘causing them, to go in deep sleep’) has enabled them to satisfactorily attend to the informative level of meaning. Yet like Pickthall, who provided (‘sealed up’), al-Hilālī and Khān experienced a register shift by providing the informal EIPV (‘covered up’) as an equivalent to the highly formal Qur’anic expression ḍarabnā ʿalá
146
Data analysis, assessment, and discussion
ādhānihim. Hence, there is a mismatch of the functional-pragmatic equivalence with the original.
6.6 Case study 6 wa-in nakathū aymānahum min baʿdi ʿahdihim wa-ṭaʿanū f ī dīnikum fa-qātilū aʾimmata al-kufri inna-hum lā-aymāna la-hum laʿallahum yantahūn. (Qur’an, 9:12)
6.6.1 Tertium comparationis This verse refers to the disbelievers of the pagan Quraysh, who according to the verse broke their oaths after concluding a truce with the Muslims in Hudaybiyyah in AH 6 (628 CE). This led to the conquest of Makkah by the Muslims two years later (Asad, 2011). The verse declares that if the disbelievers break “their oaths, their covenants, after (making) their pact and assail your religion, slander it, then fight the leaders of unbelief . . . —verily they have no (binding) oaths, (no) pacts . . . —so that they might desist, from unbelief” (al-Maḥallī & al-Suyūṭī, 2007, p. 195). The QIPV used in this context is ṭaʿanū f ī. It is a metaphorical combination of the verb ṭaʿanū and the preposition f ī. The verb ṭaʿanū is derived from the root (Ṭ-ʿ-N), whose derivatives occur two times in the Qur’an. They occur once as a verb and once as a verbal noun. The verb derived from this root appears in combination with the preposition f ī in the Qur’an to form a QIPV. For further details on the frequencies of the root (Ṭ-ʿ-N) in the Qur’an, see Appendix II. The verb ṭaʿanū is the past tense plural form of the verb ṭaʿana, which primarily signifies ‘to stab’, ‘to penetrate with a blade’, ‘to thrust’ (cf Badawi & Abdel Haleem, 2008, p. 565). For a more comprehensive explanation of the Qur’anic meanings of this verb and its derivatives, see Appendix I. The semantic properties of the metaphorical QIPV in this context unfold as follows: The object of the metaphor describes the conduct of the disbelievers in relation to their violation of the truce. The image of the metaphor depicts stabbing the religion of Islam. The sense of the metaphor is about slandering the religion of Islam. The pragmatic properties of the QIPV, however, unfold as follows: At the what-is-said level (locutionary act), if the disbelievers violate their oaths and stab your faith, then fight them. At the what-is-implicated level (illocutionary force), use of the phrase dīnikum (your faith) as an indirect object of the verb ṭaʿanū (to stab) marks fouting the maxim of MANNER (avoid obscurity of expression/avoid ambiguity). This is because the faith cannot be physically stabbed. It instead can be
Data analysis, assessment, and discussion 147 attacked, assaulted, or slandered. Flouting this maxim is intended to implicate the sense of al-waqīʿah ‘BACKBITING/SLENDER/CALUMNIATION’ (cf al-Zamakhsharı̄ , 1998a, p. 3/17; al-Rāzı̄ , 1981, p. 15/242; al-Aṣfahānı̄ , n.d., p. 2/397). Thus, the illocutionary force of this Qur’anic context may be presented as follows: if disbelievers violate their oaths and backbite/slender/calumniate your faith, then fght them.
6.6.2 English equivalents provided Pickthall (1971, p. 240): And if they break their pledges after their treaty (hath been made with you) and assail your religion, then fight the heads of disbelief—Lo! they have no binding oaths—in order that they may desist. Yūsuf ʿAlī (1991, p. 440): But if they violate their oaths after their covenant and taunt you for your Faith—fight ye the chiefs of Unfaith: for their oaths are nothing to them: that thus they may be restrained. Arberry (1955, p. 1/208): But if they break their oaths after their covenant and thrust at your religion, then fight the leaders of unbelief; they have no sacred oaths; haply they will give over. Shakir (2011, p. 117): And if they break their oaths after their agreement and (openly) revile your religion, then fight the leaders of unbelief—surely their oaths are nothing— so that they may desist. Zafrulla Khan (1971, p. 174): But if they break faith after pledging it and ridicule your religion, then fight these leaders of disbelief that they may desist. al-Hilālī and Khān (1997, p. 245): But if they violate their oaths after their covenant, and attack your religion with disapproval and criticism then fight (you) the leaders of disbelief (chiefs of Quraish pagans of Makkah)—for surely their oaths are nothing to them— so that they may stop (evil actions).
148
Data analysis, assessment, and discussion
Asad (2011, p. 309): But if they break their solemn pledges after having concluded a covenant, and revile your religion, then fight against these archetypes of faithlessness who, behold, have no (regard for their own) pledges, so that they might desist (from aggression). Abdel Haleem (2010, p. 189): But if they break their oath after having made an agreement with you and revile your religion, then fight these leaders of disbelief—oaths mean nothing to them—so that they may stop. Qarāʾī (2004, p. 255): But if they break their pledges after their having made a treaty and revile your religion, then fight the leaders of unfaith—indeed they have no (commitment to) pledges—maybe they will relinquish. Bakhtiar (2012, p. 173): But if they broke their sworn oaths after their compact and discredited your way of life, then, fight the leaders of ingratitude. Truly, they, their sworn oaths are nothing to them, so that perhaps they will refrain themselves.
6.6.3 Analysis Table 6.6 outlines the performance of the selected translators in dealing with the QIPV employed in the verse at hand. Except for Arberry, all the selected translators reduced the metaphor to sense. Arberry, however, attempted to reproduce the metaphor in the TL but did not include a sense. This has been done by providing ‘thrust at’ as an equivalent to the QIPV at hand. Yet ‘thrust at’, which denotes “to make a sudden forward movement with a knife, etc.” (Oxford Wordpower, 2000, p. 781) echoes the literal meaning of the QIPV, which is certainly not the intended one in this Qur’anic context. The TL does not permit the reproduction of this metaphor. Hence, there is semantic deviation from both the denotative and the connotative meanings of the QIPV and a mismatch of functional-pragmatic equivalence. In the other nine translators respective attempts to paraphrase the QIPV in question, they have given six different equivalents to ṭaʿanū f ī, being: 1
Asad, Abdel Haleem, Shakir, and Qarāʾī provided ‘to revile’, which means “to assail with contemptuous or opprobrious language; address, or speak of, abusively” (Macquarie Dictionary, 2005, p. 1212);
Data analysis, assessment, and discussion 149 Table 6.6 Translators’ performance when dealing with the QIPV used in the Qur’an, 9:12 Translator
Adequacy (functionalpragmatic equivalence)
Type of translational pitfall
Procedure
Pickthall
Adequate
None
Yūsuf ʿAlī
Adequate
None
Arberry
Inadequate
Shakir
Adequate
Overtly erroneous error (mistaking the literal for the metaphorical meaning and Semantic deviation) None
Reducing the metaphor to sense Reducing the metaphor to sense Reproducing the metaphor in the TL without adding a sense
Zafrulla Khan al-Hilālī and Khān Asad
Adequate
None
Adequate
None
Adequate
None
Abdel Haleem Qarāʾī
Adequate
None
Adequate
None
Bakhtiar
Adequate
None
2
3
4 5
6
Reducing the metaphor to sense Providing a parenthetical gloss Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense Reducing the metaphor to sense
Pickthall provided ‘to assail’, which means “to set upon with violence; assault; to set upon vigorously with argument, entreaties, abuse, etc.” (Macquarie Dictionary, 2005, p. 78); Yūsuf ʿAlī provided ‘to taunt’, which means “to try to make [somebody] angry or unhappy by saying unpleasant or cruel things” (Oxford Wordpower, 2000, p. 767); Zafrulla Khan provided ‘to ridicule’, which means “to laugh at [somebody or something] in an unkind way” (Oxford Wordpower, 2000, pp. 640–641); al-Hilālī and Khān provided ‘to attack’, which means “to say strongly that you do not like or agree with [somebody or something]; to try to hurt or defeat [somebody or something] by using force” (Oxford Wordpower, 2000, p. 42); Bakhtiar provided ‘to discredit’, which means “to cause [somebody or something] to lose people’s trust; to damage the reputation of [somebody or something]” (Oxford Wordpower, 2000, p. 216).
150
Data analysis, assessment, and discussion
All the aforesaid six provided equivalents are faithful to the illocutionary force of this Qur’anic context—that is, to communicate the sense of ‘to backbite/ slender/calumniate’. Yet some translators were more faithful than others. Yūsuf ʿAlī, for instance, elaborated on the meaning by providing a clarifying footnote: Not only did the enemies break their oaths shamelessly, but they even taunted the Muslims on their Faith and the “simple-minded” way in which they continued to respect their part of the treaty, as if they were afraid to fight. (1991, p. 440) Shakir inserted the adverb openly, bringing his rendering even closer to the intended meaning. However, al-Hilālī and Khān provided what seems to be the most adequate functional-pragmatic equivalence, in that they added to their translation (i.e. attack your religion) the parenthetical gloss “with disapproval and criticism”, which perfectly captures the aforesaid illocutionary force of the original.
6.7 Case study 7 huwa al-ladhī akhraja al-ladhīna kafarū min ahli al-kitābi min dyārihim liawwali al-ḥashri mā ẓanantum an yakhrijū wa-ẓannū anna-hum māniʿatuhum ḥuṣūnuhum min Allāhi fa-atāhum Allāhu min ḥaythu lam yaḥtasibū waqadhafa f ī qulūbihimu al-ruʿba yukhribūna buyūtahum bi-aydīhim wa-aydī al-mu’minīna fa-iʿtabirū yā-ʾūlī al-abṣāri. (Qur’an, 59:2)
6.7.1 Tertium comparationis This verse refers to a historical event that took place shortly after the hijrah (emigration) of the Prophet from Makkah to Madīnah. It discusses the disbelievers of ahli al-kitāb (People of the Book), particularly the Jewish tribe of Banū al-Naḍīr. This tribe inhabited in the outskirts of Madīnah and is in conflict with the Muslim community. This conflict and its aftermath are amply summarized as follows: [T]he Prophet concluded a treaty with Banū al-Naḍīr, according to which they pledged themselves to neutrality in the hostilities between the Muslims and the pagan Quraysh. After the Muslim victory in the battle of Badr, in the year [AH] 2, the leaders of that Jewish tribe spontaneously declared that Muhammad was indeed the Prophet whose coming had been predicted in the Torah; but one year later, after the near-defeat of Muslims at Uhud . . . the Banū al-Naḍīr treacherously broke their compact with the Prophet Muhammad and entered into an alliance with the Makkan Quraysh with a view of destroying the Muslim community once and for all. Thereupon the Prophet placed before them an alternative: either war or departure from Madinah with all their possessions. If they accepted this latter proposition, they would be
Data analysis, assessment, and discussion 151 allowed to return every year to gather the produce of their date groves, which would thus remain their property. Ostensibly agreeing to the second alternative, the Banū al-Naḍīr asked for—and were granted—ten days of respite. In the meantime, they secretly conspired with the hypocrites among the Arabs of Madinah, led by Abdullah ibn Ubbay, who promised them armed support by two thousands warriors in case they decided to remain in their fortified settlements on the outskirts of the town: ‘Hence, do not leave your homes; if the Muslims fight against you, we shall fight side by side with you’. The Banū al-Naḍīr followed this advice, defied the Prophet and took up arms. In the ensuing conflict, their forts were besieged by the Muslims—though without actual fighting—for twenty-one days; but when the promised help of Abdullah ibn Ubayy’s followers did not materialized, the [Banū al-] Naḍīr surrendered in the month of Rabīʿ al-Awwal, [AH] 4, and sued for peace. This they were granted on condition that they would leave Madinah. Taking with them all their movable properties, but not their arms. Most of them emigrated to Syria in caravan of about six hundred camels; only two families chose to settle in the oasis of Khaybar, while a few individuals went as far as al-Ḥīrah on lower Mesopotamia. . . . their fields and plantations were forfeited; most of them were divided among needy Muslims, and the reminder was reserved for the requirements of the Islamic community as a whole. (Asad, 2011, p. 1015) The verse at hand sums up this historical event. Al-Maḥallı̄ and al-Suyūṭı̄ (2007) interpret the verse as follows: It is He Who expelled those who disbelieved of the People of the Scripture, namely, the Jews of the Banū al-Naḍīr, from their homelands, (from) their dwellings at Medina, at the first exile, that is, their exile to Syria, the last (exile) being their banishment to Khaybar by ʿUmar during his caliphate. You did not think, O believers, that they would go forth, and they thought that they would be protected . . . by their fortresses . . . from God, from His chastisement. But God, His command and His chastisement, came at them from whence they had not reckoned, (from whence) had never occurred to them, from the part of the believers, and He cast terror . . . into their hearts, by having their chief Kaʿb bin al-Ashraf slain, destroying (as they did) . . . their houses, in order to take away with them what they valued of wood and so on, with their own hands and the hands of the believers. So, take heed, O you who have eyes! (p. 658) The QIPV qadhafa f ī is employed in this verse. It is a metaphorical combination of the verb qadhafa and the preposition f ī and derived from the root (Q-DH-F), whose derivatives occur nine times in the Qur’an. All of them appear as verbs. In the Qur’anic texture, the nine verbs derived from this root appear with and without prepositions. They occur eight times with a preposition and once without a preposition. Overall, five of the eight verb-preposition structures are QIPVs, and
152
Data analysis, assessment, and discussion
three are nonidiomatic. The verbs derived from the root (Q-DH-F) combine with two different prepositions (i.e. f ī and baʾ) to form QIPVs. They combine with the preposition baʾ three times and with the preposition f ī twice. For further details on the frequencies of the root (Q-DH-F) in the Qur’an, see Appendix II. The verb qadhafa basically denotes ‘to cast away’, ‘to throw’, or ‘to shoot’ (Badawi & Abdel Haleem, 2008, p. 745). It is used in the Qur’anic discourse to mean ‘to cast’, ‘to throw’, ‘to fling’ (Qur’an, 20:39) and ‘to pelt’, ‘to hurl’ (Qur’an, 21:18 and 34:53) (cf Badawi & Abdel Haleem, 2008, p. 745). For a more comprehensive explanation of the Qur’anic meanings of this verb and its derivatives, see Appendix I. The preposition fī denotes in this context al-ẓarfiyah al-makāniyah (adverbiality of place), indicating the place where the terror is located. Consequently, the combination of the verb qadhafa and the preposition f ī signifies both the action (i.e. casting the terror) and its location (i.e. their hearts) (cf Dāwood, 2002, p. 1/313). In tackling this verse, Qur’an exegetes assert that employing the verb qadhafa in such a metaphorical way hints at the significance of ‘hurling’ al-ruʿba (terror) into the hearts of the People of the Book and of making it ‘steady and fixed’ in their hearts (cf al-Zamakhsharī, 1998a, p. 6/74; al-Rāzī, 1981, p. 29/281). Therefore, the semantic particulars of the metaphorical QIPV qadhafa f ī in this specific context unfold as follows: The object of the metaphor describes the status of the ahli al-kitāb (People of the Book) when being exiled from their dwellings. The image of the metaphor depicts hurling terror into their hearts. The sense of the metaphor is about causing a steady and fixed terror to grow in their hearts. The pragmatic particulars of the QIPV, however, unfold as follows: At the what-is-said level (locutionary act), God hurled terror into the hearts of People of the Book. At the what-is-implicated level (illocutionary force), using the word al-ruʿba (terror) as an indirect object of the verb qadhafa (to hurl) violates the maxim of RELATION (be relevant). The communicative purpose of violating this maxim is to implicate the sense of STEADINESS and FIXEDNESS of the terror being hurled into their hearts. Accordingly, one may present the illocutionary force of this Qur’anic context in this manner: God hurled steady and fxed terror into the hearts of the People of the Book.
6.7.2 English equivalents provided Pickthall (1971, pp. 728–729): He it is Who hath caused those of the People of the Scripture who disbelieved to go forth from their homes unto the first exile. Ye deemed not that they
Data analysis, assessment, and discussion 153 would go forth, while they deemed that their strongholds would protect them from Allah. But Allah reached them from a place whereof they recked not, and cast terror in their hearts so that they ruined their houses with their own hands and the hands of the believers. So learn a lesson, O ye who have eyes! Yūsuf ʿAlī (1991, pp. 1442–1443): It is He Who got out the Unbelievers among the People of the Book from their homes at the first gathering (of the forces). Little did ye think that they would get out: And they thought that their fortresses would defend them from Allah! But the (Wrath of) Allah came to them from quarters from which they little expected (it), and cast terror into their hearts, so that they destroyed their dwellings by their own hands and the hands of the Believers, take warning, then, O ye with eyes (to see)! Arberry (1955, p. 2/267): It is He who expelled from their habitations the unbelievers among the People of the Book at the first mustering. You did not think that they would go forth, and they thought that their fortresses would defend them against God; then God came upon them from whence they had not reckoned, and He cast terror into their hearts as they destroyed their houses with their own hands, and the hands of the believers; therefore, take heed, you who have eyes! Shakir (2011, p. 370): He it is Who caused those who disbelieved of the followers of the Book to go forth from their homes at the first banishment; you did not think that they would go forth, while they were certain that their fortresses would defend them against Allah; but Allah came to them whence they did not expect, and cast terror into their hearts; they demolished their houses with their own hands and the hands of the believers; therefore, take a lesson, O you who have eyes! Zafrulla Khan (1971, p. 554): He it is Who turned out the disbelievers from among the People of the Book from their homes at the time of the first banishment. You did not think that they would go forth and they thought that their fortresses would protect them against Allah. But Allah came upon them whence they did not expect and cast terror into their hearts, so that they destroyed their homes with their own hands and the hands of the believers. So, take warning, O ye who possess understanding. al-Hilālī and Khān (1997, p. 749): He it is Who drove out the disbelievers among the people of the Scripture (i.e. the Jews of the tribe of Banū An-Naḍīr) from their homes at the first
154
Data analysis, assessment, and discussion gathering. You did not think that they would get out. And they thought that their fortresses would defend them from Allāh! But Allāh’s (Torment) reached them from a place whereof they expected it not, and He cast terror into their hearts, so that they destroyed their own dwellings with their own hands and the hands of the believers. Then take admonition, O you with eyes (to see).
Asad (2011, p. 1016): He it is who turned out of their homes, at the time of (their) first gathering (for war), such of the followers of earlier revelation as were bent on denying the truth. You did not think (O believers) that they would depart (without resistance)—just as they thought that their strongholds would protect them against God: but God came upon them in a manner which they had not expected, and cast terror into their hearts; (and thus) they destroyed their homes by their own hands as well as the hands of the believers. Learn a lesson, then, O you who are endowed with insight! Abdel Haleem (2010, p. 546): It was He who drove those of the People of the Book who broke faith out from their homes at the first gathering of forces—you (believers) never thought they would go, and they themselves thought their fortifications would protect them against God. God came up on them from where they least expected and put panic into their hearts: their homes were destroyed by their own hands, and the hands of all the believers. Learn from this, all of you with insight! Qarāʾī (2004, p. 775): It is He who expelled the faithless belonging to the People of the Book from their homes at the outset of (their) en masse banishment. You did not think that they would go out, and they thought their fortresses would protect them from Allah. But All came at them from whence they did not reckon and He cast terror into their hearts. They demolished their houses with their own hands and the hands of the faithful. Bakhtiar (2012, p. 532): It is He Who drove out those who were ungrateful—among the People of the Book—from their abodes at the first assembling. You thought that they would not go forth. And they thought that they are ones who are secure in their fortresses from God. But God approached them from where they anticipate not. And He hurled alarm into their hearts. They devastate their own houses with their own hands and the hands of the ones who believe. Then, take warning, O those imbued with insight!
Data analysis, assessment, and discussion 155
6.7.3 Analysis Table 6.7 outlines the performance of the selected translators in dealing with the QIPV employed in the verse at hand. Except for Abdel Haleem, who reduced the metaphor to sense, all the selected translators opted for a source-text-oriented approach and attempted to reproduce the metaphor in the TL without adding a sense. Regarding choice of words, most of the selected translators (Pickthall, Yūsuf ʿAlī, Arberry, Shakir, Zafrulla Khan, al-Hilālī and Khān, Asad, and Qarāʾī) provided as an equivalent to the original: ‘He cast terror into their hearts’. Abdel Haleem and Bakhtiar, however, provided ‘[He] put panic into their hearts’ and ‘He hurled alarm into their hearts’, respectively. Providing ‘to cast . . . into’, ‘to hurl . . . into’, and ‘to put . . . into’ as equivalents to the original QIPV demonstrates a less-than-perfect translation. This is because
Table 6.7 Translators’ performance when dealing with the QIPV used in the Qur’an, 59:2 Translator
Adequacy (functionalpragmatic equivalence)
Type of translational pitfall
Procedure
Pickthall
Inadequate
Overtly erroneous error (under-translation)
Yūsuf ʿAlī
Inadequate
Overtly erroneous error (under-translation)
Arberry
Inadequate
Overtly erroneous error (under-translation)
Shakir
Inadequate
Overtly erroneous error (under-translation)
Zafrulla Khan
Inadequate
Overtly erroneous error (under-translation)
al-Hilālī and Khān
Inadequate
Overtly erroneous error (under-translation)
Asad
Inadequate
Overtly erroneous error (under-translation)
Abdel Haleem
Inadequate
Qarāʾī
Inadequate
Overtly erroneous error (under-translation) Overtly erroneous error (under-translation)
Bakhtiar
Inadequate
Reproducing the metaphor in the TL without adding a sense Reproducing the metaphor in the TL without adding a sense Reproducing the metaphor in the TL without adding a sense Reproducing the metaphor in the TL without adding a sense Reproducing the metaphor in the TL without adding a sense Reproducing the metaphor in the TL without adding a sense Reproducing the metaphor in the TL without adding a sense Reducing the metaphor to sense Reproducing the metaphor in the TL without adding a sense Reproducing the metaphor in the TL without adding a sense
Overtly erroneous error (under-translation)
156
Data analysis, assessment, and discussion
the neutral term ‘to put’ and the old-fashioned term ‘to cast’, which denotes “to throw” something (Oxford Wordpower, 2000, p. 109), both stop short of capturing the tone of steadiness and fixedness associated with the original qadhafa. Bakhtiar’s choice, ‘to hurl’, denotes “to throw [something] with force” (Oxford Wordpower, 2000, p. 374) is closer to the intended meaning because of its association with ‘force’. Yet Bakhtiar’s ‘to hurl’ is still not adequate to encompass the complete meaning of the original text, let alone the semantic deviation created by providing ‘alarm’ as an equivalent to the original wording, al-ruʿba (terror). At any rate, none of the selected translators was successful in capturing the illocutionary force of the metaphorical expression qadhafa f ī qulūbihimu al-ruʿba (i.e. God caused steady and fastened terror into the hearts of People of the Book). This is confirmed by the authoritative Qur’an exegetes (cf al-Zamakhsharī, 1998a, p. 6/74; al-Rāzī, 1981, p. 29/281). Therefore, all the selected translators exercised under-translation by providing equivalents that communicate fewer details than the original communicates. Hence, there is a mismatch of the functionalidiomatic equivalence.
6.8 Case study 8 wa-man yarghabu ʿan millati Ibrāhīma illā man safiha nafsahu wa-laqad iṣṭafaynāhu f ī al-dunyā wa-innahu f ī al-ākhirati la-min al-ṣṣaliḥīn. (Qur’an, 2:130)
6.8.1 Tertium comparationis The Qur’anic macro structure of this verse shows that it occurs after presenting the arguments for the children of Israel and for the Arabs “as succeeding to the spiritual inheritance of Abraham” (Yūsuf ʿAlī, 1991, p. 52). The QIPV yarghabu ʿan is used in this Qur’anic context. It is a figurative combination of the verb yarghabu and the preposition ʿan. The verb yarghabu is an active present tense form of the verb raghiba, derived from the root (R-GH-B), whose derivatives occur eight times in the Qur’an. They occur four times as verbs, three times as active participles, and once as a verbal noun. In the Qur’anic texture, the four verbs derived from this root appear with and without prepositions: three times with prepositions and once without prepositions. All three verbpreposition structures are QIPVs. The verbs derived from the root (R-GH-B) combine with three different prepositions (i.e. baʾ, ilá, and ʿan) to form QIPVs. They combine with the preposition baʾ once, with the preposition ʿalá once, and with the preposition ʿan once. For further details on the frequencies of the root (R-GH-B) in the Qur’an, see Appendix II. The verbs derived from this root across all combinations with different prepositions constitute good examples of the figurative QIPV type. The difference between the figurative QIPVs and the metaphorical QIPVs has been accounted for in previous chapters (see 4.3.2 and 5.5.1 in specific). We have demonstrated that unlike
Data analysis, assessment, and discussion 157 metaphorical QIPVs that satisfy the definition of metaphor, the figurative QIPVs stop short of fulfilling the definition and meeting its components and purpose. Yet they express nonliteral and nontransparent conventional meanings that have nothing to do with the meanings of their individual constituents. What is more, we established that metaphorical QIPVs are different from figurative QIPVs in how their meaning is determined. That is, the meanings of metaphorical QIPVs are appreciated only when a successful interpretation of the speakers’ implied meaning delivered via conversational implicatures in given situational context is achieved. Conversely, grasping the overall meanings of figurative QIPVs depends on the speakers’ mastery of the Arabic linguistic system because conventional linguistic idiomaticity rather than conversational implicatures is the core issue here. Having devoted the previous seven case studies to illustrating metaphorical QIPVs, this case study and the forthcoming two case studies are devoted to illustrating figurative QIPVs. We take up as illustrative examples three Qur’anic contexts containing figurative QIPVs in which the verbs derived from the root (R-GH-B) combine with three different prepositions (i.e. baʾ, ilá, and ʿan). The verb raghiba basically denotes “‘to desire’, ‘to wish’” (Badawi & Abdel Haleem, 2008, p. 372). It appears in the Qur’an to signify several meanings in several contexts. Its Qur’anic meanings include ‘to ask humbly’, ‘to implore’, ‘to petition’ (Qur’an, 94:8), ‘to reject’, ‘to disdain’, ‘to shun’ (Qur’an, 2:130), ‘to consider oneself too good for’ (Qur’an, 9:120), and ‘to wish’, ‘to desire’ (Qur’an, 4:127) (cf Badawi & Abdel Haleem, 2008, p. 372). For a more comprehensive explanation of the Qur’anic meanings of this verb and its derivatives, see Appendix I. In this context, however, the preposition ʿan signifies al-mujāwazah (going beyond something). In its combination with the verb raghiba, it indicates al-inṣiraf wa-ʿadam al-raghbah (abandonment and disinclination). This combination of raghiba + ʿan + something denotes ‘a person’s desire has abandoned that thing and gone beyond it’ (cf al-Aṣfahānī, n.d., p. 1/262–263; Dāwood, 2002, p. 2/63–64). Therefore, wa-man yarghabu ʿan millati Ibrāhīma means who “abandons [Abraham’s] path, way and method” (Ibn Kathir, 2003, p. 1/403). The Qur’an exegetes interpret the verse at hand as follows: Who therefore, meaning ‘none’, shrinks from the religion of Abraham, abandoning it, except he who fools himself? that is to say, either the one who ignores that his soul has been created for God, and that it is obliged to worship Him, or the one who treats it with frivolity and degrades it. Indeed, We chose him, We elected him, in this world, for prophethood and friendship, and in the Hereafter he shall be among the righteous, those of the high stations. (al-Maḥallī & al-Suyūṭī, 2007, p. 23) Pragmatically speaking, initiating this verse with the interrogative particle man (who) followed by the exceptional particle illā (except) denotes the status of the istifhām bi-maʿná al-inkār (interrogating implying denial) on the part of God. By using this structure, the speech act of tawbīkh ‘REBUKE’ is performed in this context (cf al-Rāzı̄ , 1981, p. 4/75). This speech act is preformed to clearly
158
Data analysis, assessment, and discussion
communicate the idea that “Only the Fools deviate from Ibrahim’s Religion” (Ibn Kathir, 2003, p. 1/402).
6.8.2 English equivalents provided Pickthall (1971, p. 24): And who forsaketh the religion of Abraham save him who befooleth himself? Verily We chose him in the world, and lo! in the Hereafter he is among the righteous. Yūsuf ʿAlī (1991, p. 54): And who turns away from the religion of Abraham but such as debase their souls with folly? Him We chose and rendered pure in this world: And he will be in the Hereafter in the ranks of the Righteous. Arberry (1955, p. 1/44): Who therefore shrinks from the religion of Abraham, except the foolishminded? Indeed, We chose him in the present world, and in the world to come he shall be among the righteous. Shakir (2011, p. 12): And who forsakes the religion of Ibrahim but he who makes himself a fool, and most certainly We chose him in this world, and in the hereafter he is most surely among the righteous. Zafrulla Khan (1971, p. 22): Thus who will turn away from the religion of Abraham but he who has ruined his soul? Of a surely, We exalted him in this world, and in the next also he will be among the righteous. al-Hilālī and Khān (1997, p. 25): And who turns away from the religion of Ibrāhīm (Abraham) (i.e. Islāmic Monotheism) except him who befools himself? Truly, We chose him in this world and verily, in the Hereafter he will be among the righteous. Asad (2011, p. 33): And who, unless he be weak of mind, would want to abandon Abraham’s creed, seeing that We have indeed raised him high in this world, and that, verily, in the life to come he shall be among the righteous?
Data analysis, assessment, and discussion 159 Abdel Haleem (2010, p. 21): Who but a fool would forsake the religion of Abraham? We have chosen him in this world and he will rank among the righteous in the Hereafter. Qarāʾī (2004, p. 28): And who will (ever) renounce Abraham’s creed except one who fools himself? We certainly chose him in the (present) world, and in the Hereafter he will indeed be among the Righteous Bakhtiar (2012, p. 18): And who shrinks from the creed of Abraham, but he who fooled himself? And, certainly, We favored him in the present. And, truly, in the world to come he will be among the ones in accord with morality.
6.8.3 Analysis Table 6.8 outlines the performance of the selected translators in dealing with the QIPV employed in the verse at hand. The English equivalents given to the QIPV in question show that the selected translators adopt fve approaches to rendering the context wa-man yarghabu ʿan millati Ibrāhīma: 1 2 3 4 5
‘who forsake(th) the religion of Abraham’ (Pickthall, Shakir, and Abdel Haleem); ‘who (will) turn(s) away from the religion of Abraham’ (Yūsuf ʿAlī, al-Hilālī and Khān, and Zafrulla Khan); ‘who (therefore) shrinks from the religion of Abraham’ (Arberry and Bakhtiar); ‘would want to abandon Abraham’s creed’ (Asad); ‘who will (ever) renounce Abraham’s creed’ (Qarāʾī).
Apparently, all the selected translators tried to imitate the original syntactic structure by initiating their renderings with the English interrogative particle ‘who’, followed by an exceptional particle: ‘put, except, save, unless’. In so doing, the speech act of REBUKE (tawbīkh) performed in the original has been faithfully reflected in the TL. Thus, the connotative meaning has been attended to. Regarding the choice of words, the selected translators provided five expressions as English equivalents to the QIPV yarghabu ʿan, all of which convey its denotative meaning to varying degrees. Pickthall, Shakir, and Abdel Haleem provided the word ‘forsake(th)’, defined as “to leave a person or a place for ever (especially when you should stay)” (Oxford Wordpower, 2000, p. 301). This constitutes a plausibly adequate English functional-pragmatic equivalence for yarghabu ʿan in that it perfectly reflects its informative level of meaning.
160
Data analysis, assessment, and discussion
Table 6.8 Translators’ performance when dealing with the QIPV used in the Qur’an, 2:130 Translator
Adequacy (functionalpragmatic equivalence)
Type of translational pitfall
Procedure
Pickthall
Adequate
None
Yūsuf ʿAlī
Inadequate
Arberry
Inadequate
Shakir
Adequate
Covertly erroneous error (register shift) Overtly erroneous error (under-translation) Covertly erroneous error (register shift— providing EIPV) None
Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure
Zafrulla Khan
Inadequate
al-Hilālī and Khān
Inadequate
Asad
Inadequate
Overtly erroneous error (over-translation)
Abdel Haleem
Adequate
None
Qarāʾī
Inadequate
Overtly erroneous error (over-translation)
Bakhtiar
Inadequate
Covertly erroneous error (register shift— providing EIPV)
Covertly erroneous error (register shift) Overtly erroneous error (under-translation) Covertly erroneous error (register shift) Overtly erroneous error (under-translation)
Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure + providing parenthetical glosses Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure + providing a parenthetical gloss Imitating the original’s marked syntactic structure
Likewise, Qarāʾī’s rendering attended to the informative level of meaning by providing the word ‘renounce’, a formal word meaning “to say formally that you no longer want to be connected with [something]” (Oxford Wordpower, 2000, p. 628). What makes Qarāʾī’s rendering less adequate is his addition of the modal auxiliary verb ‘will’ and the adverb ‘ever’. This addition is unjustified and constitutes over-translation by being too specific and providing details not
Data analysis, assessment, and discussion 161 expressed in the original. Asad, however, provided the word ‘abandon’, which is defined as “to leave [somebody or something] that you are responsible for, usually permanently [and] to stop doing [something] without finishing it or without achieving what you wanted to do” (Oxford Wordpower, 2000, p. 1). The word implies the original sense of leaving somebody or something permanently but appears less accurate. This is because it implies the sense of ‘you are responsible for’ and the sense of ‘without finishing it or without achieving what you wanted to do’, which are not intended in the original. Thus, an over-translation is demonstrated here. Strikingly, the EIPV ‘turn away’, meaning “to stop looking at [somebody or something]” (Oxford Wordpower, 2000, p. 801) was provided by Yūsuf ʿAlī, al-Hilālī and Khān, and Zafrulla Khan as an equivalent to the QIPV at hand. This translation, apart from its register shift in using the informal EIPV as an equivalent to the formal QIPV, fails to satisfy the informative level of meaning by being too general. Thus, under-translation is demonstrated here. Similarly, Arberry, and Bakhtiar experienced a register shift by providing the EIPV ‘shrink from’ as an equivalent to the QIPV in question. Otherwise, this EIPV, defined as “to be unwilling to do [something] because you find it unpleasant” (Oxford Wordpower, 2000, p. 691), captures the denotative meaning of the original.
6.9 Case study 9 mā kāna li-ahli al-madīnati wa-man ḥawla-hum min al-aʿrābi an yatakhallafū ʿan rasūli Allāhi wa-lā-yarghbū bi-anfusihim ʿan nafsihi dhālika bi-anna-hum lā-yuṣībuhum ẓamaʾun wa-lā-naṣabun wa-lā-makhmaṣatun f ī sabīli-llāhi wa-lā-yaṭaʾūna mawṭiʾan yaghīẓu al-kuffāra wa-lā-yanālūna min ʿaduwin naylan illā kutiba la-hum bi-hi ʿamalun ṣāliḥun inna Allāha lā-yuḍīʿu ajra al-muḥsinīn. (Qur’an, 9:120)
6.9.1 Tertium comparationis This verse refers to ahli al-madīnati (the people of the Prophet’s City) and al-aʿrāb (the Bedouin) who dwell around them. The commentators believe that this Qur’anic discourse is relevant and applicable to all Muslims at all times and that the “specific reference to ‘the Prophet’s City’ is due to the fact that it was the place where the revelation of the Qur’an was completed and Islam came to its full fruition under the Prophet’s guidance” (Asad, 2011, pp. 341–342). The QIPV yarghabu bi- ʿan is employed in this context. It is a figurative combination of the verb yarghabu and the prepositions baʾ and ʿan. The literal and contextual significances of yarghabu were accounted for in case study 8. Regarding the prepositions, the verb yarghabu is combined with two prepositions, each with its own functional significance, to contribute to the meaning of the QIPV. The first preposition, baʾ, denotes in this context the significance of al-mulābasah (attendant/association), according to Arabic linguists (Dāwood, 2002, p. 2/64).
162
Data analysis, assessment, and discussion
The second preposition, ʿan, combines with the verb yarghabu to denote the significance of al-inṣiraf wa-ʿadam al-raghbah (abandonment and disinclination). Accordingly, the combination of yarghabu + baʾ + ʿan signifies al-inṣiraf wa-mā yulābisuhu (abandonment and whatever is associated with it) (cf Dāwood, 2002, p. 2/63–64). Hence, the meaning of this QIPV reflects the following notion: They should not abandon the Prophet for the sake of caring more for themselves than for him. In this verse, al-Zamakhsharī (1998a) asserts that Muslims were asked to be with the Prophet in adversity and hardship: to go with him happily and enthusiastically through all kinds of horrors and to let their souls experience difficulties as does his own soul. This is bearing in mind that the Prophet’s soul is the most honourable soul to Allah. Therefore, once it has been exposed to adversity and hardship, all other souls must be generously and bigheartedly scarified for him (p. 3/105). In an attempt to illuminate the meaning of this section of verse, al-Rāzī (1981) indicates that they should not ask for protection and comfort for themselves when the Prophet is enduring hardship and discomfort (p. 16/229). Thus, the interpretation of the entire verse from al-Maḥallī and al-Suyūṭī (2007) is as follows: It is not for the people of Medina and for the Bedouins (who dwell) around them to stay behind God’s Messenger, when he sets out on a campaign, and to prefer their lives to his life, by guarding them against hardships which he (the Prophet) is content (to suffer) himself (this statement is a prohibition expressed as a predicate); that, prohibition against staying behind, is because neither thirst nor toil nor hunger afflicts them in the way of God, nor tread they any tread . . . that enrages the disbelievers, nor gain any gain from the enemy, of God, be it through slaughter, capture or plunder, but a righteous deed is therefore recorded for them, that they may be rewarded for it. Truly God does not leave the wage of the virtuous to go to waste, that is, the wage of those (mentioned), rather He rewards them. (p. 211) Thus, pragmatically speaking, the speech act of manʿ ‘PROHIBITION’ is performed in this context. In addition, al-Zamakhsharı̄ (1998a) maintains that this is an eloquent prohibition associated with taqbīḥ ‘DEMONIZING’ their attitude, tawbīkh ‘REBUKING’ them for having such an attitude, and tahyīj ‘ENCOURAGING’ them to follow the Prophet with self-respect and enthusiasm (al-Zamakhsharı̄ , 1998a, p. 3/105).
6.9.2 English equivalents provided Pickthall (1971, p. 262): It is not for the townsfolk of Al-Madinah and for those around them of the wandering Arabs so stay behind the messenger of Allah and prefer their lives
Data analysis, assessment, and discussion 163 to his life. That is because neither thirst nor toil nor hunger afflicteth them in the way of Allah, nor step they any step that angereth the disbelievers, nor gain they from the enemy a gain, but a good deed is recorded for them therefor. Lo! Allah loseth not the wages of the good. Yūsuf ʿAlī (1991, pp. 473–474): It was not fitting for the people of Medīnah and the Bedouin Arabs of the neighbourhood, to refuse to follow Allah’s Messenger, nor to prefer their own lives to his: because nothing could they suffer or do, but was reckoned to their credit as a deed of righteousness—whether they suffered thirst, or fatigue, or hunger, in the cause of Allah, or trod paths to raise the ire of the Unbelievers, or received any injury whatever from an enemy: for Allah suffereth not the reward to be lost of those who do good. Arberry (1955, 1/221–222): It is not for the people of the City and for the Bedouins who dwell around them to stay behind God’s Messenger, and to prefer their lives to his; that is because they are smitten neither by thirst, nor fatigue, nor emptiness in the way of God, neither tread they any tread enraging the unbelievers, nor gain any gain from any enemy, but a righteous deed is thereby written to their account: God leaves not to waste the wage of the good-doers. Shakir (2011, p. 128): It did not beseem the people of Medina and those round about them of the dwellers of the desert to remain behind the Apostle of Allah, nor should they desire (anything) for themselves in preference to him; this is because there afflicts them not thirst or fatigue or hunger in Allah’s way, nor do they tread a path which enrages the unbelievers, nor do they attain from the enemy what they attain, but a good work is written down to them on account of it; surely Allah does not waste the reward of the doers of good. Zafrulla Khan (1971, p. 189): It was not proper for the people of Medina and those around them from among the Arabs of the desert to remain behind, leaving the Messenger of Allah to go forth, or to prefer their own security to his. It is so ordained because they endure no thirst, or fatigue or hunger in the cause of Allah, nor do they tread a track which chagrins the disbelievers, nor do they wrest an advantage from the enemy, but there is written down for them a good deed on account of it. Surely, Allah suffers not the reward of those who carry out their duty in the best manner to be lost.
164
Data analysis, assessment, and discussion
al-Hilālī and Khān (1997, p. 267): It was not becoming of the people of Al-Madinah and the bedouins of the neighbourhood to remain behind Allāh’s Messenger (Muhammad ṣallá Allāhu ʿalay-hi wa-sallam when fighting in Allāh’s Cause) and (it was not becoming of them) to prefer their own lives to his life. That is because they suffer neither thirst nor fatigue nor hunger in the Cause of Allāh, nor they take any step to raise the anger of disbelievers nor inflict any injury upon an enemy but is written to their credit as a deed of righteousness. Surely, Allāh wastes not the reward of the Muhsinūn. Asad (2011, pp. 341–342): It does not behove the people of the (Prophet’s) City and the bedouin (who live) around them to hold back from following God’s Apostle, or to care for their own selves more than for him—for, whenever they suffer from thirst or weariness or hunger in God’s cause, and whenever they take any step which confounds those who deny the truth, and whenever there comes to them from the enemy whatever may be destined for them—(whenever anything thereof comes to pass), a good deed is recorded in their favour. Verily, God does not fail to requite the doers of good! Abdel Haleem (2010, p. 207): The people of Medina and their neighbouring desert Arabs should not have held back from following God’s Messenger, nor should they have cared about themselves more than him: if ever they suffer any thirst, weariness, or hunger in God’s cause, take any step that angers the disbelievers, or gain any advantage over an enemy, a good deed is recorded in their favour on account of it—God never wastes the reward of those who do goodQarāʾī (2004, p. 278): It is not fitting for the people of Madinah and the Bedouins around them to hang back behind the Apostle of Allah and prefer their own lives to his life. That is because there does experience any thirst, nor fatigue, nor hunger, in the way of Allah, nor do they tread any ground enraging the faithless, nor do they gain any ground against an enemy but a righteous deed is written for them on its account. Indeed Allah does not waste the reward of the virtuous. Bakhtiar (2012, pp. 188–189): It had not been for the people of the city and among the nomads around them to stay behind from the Messenger of God nor prefer themselves more than himself. That is because they were neither lit on by thirst nor fatigue nor
Data analysis, assessment, and discussion 165 emptiness in the way of God nor tread they any treading on any ground— enraging the ones who are ungrateful—nor glean any gleaning of ground against the enemy but as an action in accord with morality written down for them. Truly, God wastes not the compensation of the ones who are doers of good.
6.9.3 Analysis Table 6.9 outlines the performance of the selected translators in dealing with the QIPV employed in the verse at hand. Table 6.9 Translators’ performance when dealing with the QIPV used in the Qur’an, 9:120 Translator
Adequacy (functionalpragmatic equivalence)
Type of translational pitfall
Procedure
Pickthall
Adequate
None
Yūsuf ʿAlī
Adequate
None
Arberry
Adequate
None
Shakir
Inadequate
Zafrulla Khan
Adequate
Overtly erroneous error (semantic deviation) None
al-Hilālī and Khān
Adequate
None
Asad
Inadequate
Abdel Haleem
Inadequate
Qarāʾī
Adequate
Covertly erroneous error (register shift— providing EIPV) Covertly erroneous error (register shift— providing EIPV) None
Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure
Bakhtiar
Adequate
None
Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure Imitating the original’s marked syntactic structure
166
Data analysis, assessment, and discussion
A close examination of the English equivalents provided for the verse reveals that by adhering to the syntactic structure of the source text, all the selected translators conveyed the communicative function intended in this Qur’anic discourse: to deliver the speech acts of ‘PROHIBITION’, ‘DEMONIZING’, ‘REBUKE’, and ‘ENCOURAGING’ as preformed in the original. Consequently, the connotative meaning has been fulfilled. Regarding the equivalents given to wa-lā-yarghbū bi-anfusihim ʿan nafsihi where the QIPV is employed, the selected translators adopted three approaches in their respective renderings: 1
2 3
‘prefer their (own) lives (security) to his life, prefer themselves more than himself’ (Pickthal, Yūsuf ʿAlī, Arberry, Zafrulla Khan, al-Hilālī and Khān, Qarāʾī, and Bakhtiar); ‘to care for their own selves (themselves) more than for him’ (Asad, and Abdel Haleem); ‘nor should they desire (anything) for themselves in preference to him’ (Shakir).
The words selected by the translators show that they provided ‘to prefer’, ‘to care for/about’, and ‘to desire’ as equivalents to the verb yarghbū. As indicated earlier, this verb is semantically modified by two prepositions that play vital roles in creating its figurative meaning. This figurative meaning, rather than the verb’s literal meaning, ought to be considered by the translators. Therefore, by providing the verb ‘to desire’ as an equivalent, Shaker mismatches the figurative meaning and thus semantically deviates from the original. Regarding the other equivalents provided by the selected translators, they all satisfied the denotative meaning of the original. Yet the equivalents given by Asad and Abdel Haleem sound less faithful to the original. This is due to their use of the informal EIPVs ‘to care for’ and ‘to care about’ as equivalents to the formal QIPV at hand. Thus, a register shift has taken place here.
6.10 Case study 10 alam nashraḥ la-ka ṣadraka. wa-waḍʿnā ʿan-ka wizraka. al-ladhī anqaḍa zahraka. wa-rafaʿnā la-ka dhikraka. fa-inna maʿa al-ʿusri yusran. inna maʿa al-ʿusri yusran. fa-idhā faraghta fa-inṣab. wa-ilá rabbika fa-irghab. (Qur’an, 94:1–8)
6.10.1 Tertium comparationis This short sūrah—called Surat al-Sharḥ or al-Inshirāḥ—is the 94th sūrah of the Qur’an. It addresses the Prophet in the first instance and, “through him, every true follower of the Qur’an” (Asad, 2011, p. 1147). It sends the Prophet and his followers “a message of hope and encouragement in a time of darkness and difficulty” (Yūsuf ʿAlī, 1991, p. 1665). The sūrah begins with the interrogative particle hamzah, followed by the negation particle lam. In the Arabic language,
Data analysis, assessment, and discussion 167 this interrogative form denotes ‘affirmation’. Therefore, the first verse of the sūrah, alam nashraḥ la-ka ṣadraka, or “Did We not expand your breast for you, O Muhammad”, actually means “We did (indeed) expand your breast for you, O Muhammad”. This expansion is manifested through prophethood and in other ways (al-Maḥallī & al-Suyūṭī, 2007, p. 753). The sūrah continues listing the bounties of God bestowed upon the Prophet by using the same interrogative/ affirmative structure: wa-waḍʿnā ʿan-ka wizraka, or “and relieve you of your burden”; al-ladhī anqaḍa zahraka, or “that which weighed down your back”; wa-rafaʿnā la-ka dhikraka, or “Did We not exalt your mention [fame]? For you are mentioned where I (God) am mentioned in the call announcing (the time for) prayer adhān, in the (second) call to perform the prayer iqāma, in the witnessing (‘there is no god but God, Muhammad is His Messenger’) tashahhud, in the Friday sermon and in other instances” (al-Maḥallī & al-Suyūṭī, 2007, p. 753). The sūrah then turns to offer hope and encouragement to the Prophet by confirming that fa-inna maʿa al-ʿusri yusran. inna maʿa al-ʿusri yusran, or “For truly with hardship comes ease. Truly with hardship comes ease” (al-Maḥallī & al-Suyūṭī, 2007, p. 754). The sūrah concludes with a lesson for the Prophet about what is best to be done: fa-idhā faraghta fa-inṣab, or “So when you are finished, from (performing) prayer, toil, exert yourself in supplication (to God)”, and wa-ilá rabbika fa-irghab, or “and seek, devote yourself humbly to, your Lord” (al-Maḥallī & al-Suyūṭī, 2007, p. 754). What concerns us in this sūrah is the use of the QIPV irghab ilá in the final verse. The verb irghab is an imperative mood of the verb raghiba, whose primary and contextual significances were taken up in the two previous case studies (case studies 8 and 9). When the verb raghiba combines with the preposition ilá it means ‘to ask humbly’, ‘to request’, ‘to hope’, or ‘to supplicate’ (Badawi & Abdel Haleem, 2008, p. 372). The preposition ilá, however, denotes intihāʾ al-ghāyah (end of destination). It implies ‘to head for’ or ‘to turn towards’ when combined with the verb raghiba. Hence, the QIPV irghab ilá denotes something like ‘to turn your request, hope, or supplication’ towards someone or something. This depends, however, on the noun governed by the preposition ilá, which is rabbika ‘your Lord’ in the Qur’anic context at hand (cf Badawi & Abdel Haleem, 2008, p. 372). Interestingly, this Qur’anic context exhibits a marked word order whereby the preposition ilá is uncommonly fronted and the verb irghab is backgrounded. This semantically oriented structure is not without its communicative purpose. It implies the meaning of al-ikhtiṣāṣ (exclusivity). In other words, fronting the preposition ilá with the noun that it governs (rabbika) entails inferring the connotative meaning of ‘devoting your request to your Lord exclusively and not to someone else’. This connotative meaning constitutes the essence of the Qur’an exegete’s interpretation of the verse at hand (cf al-Zamakhsharī, 1998a, p. 10/399; al-Rāzī, 1981, p. 32/7; Ibn Qutaybah, 1978, p. 532; al-Maḥallī & al-Suyūṭī, 2007, p. 754; among others). In an endeavour to elaborate on the reason for this exclusivity, Yūsuf ʿAlī (1991) provides the following explanatory footnote to his translation of this verse:
168
Data analysis, assessment, and discussion The kingdom of Allah is everything. Other things are incidental, and really do not matter. Worldly greatness or success may be a means to an end, but it may also be a hinderance to true spiritual greatness. Allah is the goal of the righteous man’s whole attention and desire. (p. 1667)
Pragmatically speaking, the speech act of IMPERATIVE is performed in this Quranic context. This is manifested in the use of the imperative mood of the verb (i.e. irghab). In addition, and most importantly, this speech act is associated with the sense of EXLUSIVITY, which is manifested by the verse’s marked word order.
6.10.2 English equivalents provided Pickthall (1971, p. 812): And strive to please thy Lord. Yūsuf ʿAlī (1991, p. 1667): And to thy Lord turn (all) thy attention. Arberry (1955, p. 2/343): And let thy Lord be thy Quest. Shakir (2011, p. 416): And make your Lord your exclusive object. Zafrulla Khan (1971, p. 623): and supplicate thy Lord earnestly. al-Hilālī and Khān (1997, p. 841): And to your Lord (Alone) turn (all your) intentions and hopes. Asad (2011, p. 1147): and unto thy Sustainer turn with love. Abdel Haleem (2010, p. 597): and direct your requests to your Lord.
Data analysis, assessment, and discussion 169 Qarāʾī (2004, p. 854, italics in original): And turn eagerly to your Lord. Bakhtiar (2012, p. 592, boldface in original): and quest your Lord.
6.10.3 Analysis Table 6.10 outlines the performance of the selected translators in dealing with the QIPV employed in the verse at hand. The English equivalents given to wa-ilá rabbika fa-irghab exhibit a complete divergence from the selected translators to the extent that no two provided a translation similar enough to enable me to group them together, whereas I was able to do so in the previous case studies. Consequently, the translations here are investigated individually. Nevertheless, the selected translators fall into two groups in their adhering to the source-text marked word order. Four of them (Yūsuf ʿAlī, Shakir, al-Hilālī and Khān, and Asad) opted for a source-text-oriented approach and reproduced the original marked word order. The other six (Pickthall, Arberry, Zafrulla Khan, Abdel Haleem, Qarāʾī, and Bakhtiar) opted for a target-text-oriented approach and completely abandoned the original marked word order. They instead used the ‘rephrasing’ strategy where “a problematic source language text segment is tackled by changing its form (surface structure) but keeping its meaning invariant to the greatest possible extent” (Abdul-Raof, 2001, p. 44). Lastly, all the selected translators conveyed satisfactorily the intended connotative meaning of IMPERATIVE by providing different verbs—as equivalents to the original verb irghab—in their imperative mood. However, those who abandoned the original marked word order failed to convey the intended connotative meaning of EXCLUSIVITY signalled by that marked grammatical structure. As far as capturing the denotative meaning of the verse in question, the choice of words reveals that the translators have completely diverged from each other. Apart from mismatching the connotative meaning, Pickthall’s translation demonstrates under-translation by being more general than the original is. Providing ‘strive to please’ as an equivalent to the QIPV irghab ilá, even though it fulfils part of the informative level of meaning, stops short of completely mapping the aforementioned intended meaning specified by the Qur’an exegetes. Yūsuf ʿAlī’s translation, however, appears more accurate than Pickthall’s in terms of capturing more aspects of the denotative meaning. Inserting the in-text modifier (all) has imparted the equivalent that he provided to the original irghab (i.e. turn thy attention) a sense of inclusivity. Thus, it has brought it closer to the denotative meaning.
170
Data analysis, assessment, and discussion
Table 6.10 Translators’ performance when dealing with the QIPV used in the Qur’an, 94:8 Translator
Adequacy (functionalpragmatic equivalence)
Type of translational pitfall
Procedure
Pickthall
Inadequate
Yūsuf ʿAlī
Adequate
Overtly erroneous error (under-translation) None
Arberry
Inadequate
Shakir
Adequate
Zafrulla Khan
Inadequate
Abandoning the original’s marked syntactic structure Imitating the original’s marked syntactic structure + providing a parenthetical gloss Abandoning the original’s marked syntactic structure Imitating the original’s marked syntactic structure Abandoning the original’s marked syntactic structure
al-Hilālī and Khān
Adequate
Asad
Inadequate
Abdel Haleem
Inadequate
Qarāʾī
Inadequate
Bakhtiar
Inadequate
Overtly erroneous error (semantic deviation) None Covertly erroneous error (failure to capture the illocutionary force of the original) None
Overtly erroneous error (over-translation) Covertly erroneous error (failure to capture the illocutionary force of the original) Overtly erroneous error (under-translation) Covertly erroneous error (failure to capture the illocutionary force of the original) and Overtly erroneous error (semantic deviation)
Imitating the original’s marked syntactic structure + providing parenthetical glosses Imitating the original’s marked syntactic structure Abandoning the original’s marked syntactic structure Abandoning the original’s marked syntactic structure Abandoning the original’s marked syntactic structure
Data analysis, assessment, and discussion 171 Arberry’s translation seems to be less accurate. By paraphrasing the whole verse and discarding the marked word order, Arberry has failed to convey the denotative and connotative meanings of the original. The word quest that he provides suggests the significances of “a long or arduous search” (Concise Oxford English Dictionary, 2011, p. 1177), “a long search for [something] that is difficult to find” (Cambridge Advanced Learner’s Dictionary, 2005, p. 1037; Oxford Wordpower, 2000, p. 604); “an attempt to achieve something difficult” (Cambridge Advanced Learner’s Dictionary, 2005, p. 1037); and “a search or pursuit made in order to find or obtain something” (Macquarie International English Dictionary, 2005, p. 1163). This deviates semantically from the intended meaning of the original, where no long search, arduousness, and difficulties are implied in turning requests, hope, attention, and supplication to God. Capitalizing the initial letter of the word quest does not compensate for the loss of meaning. Shakir’s translation sounds more faithful than the previous ones did in that his insertion of the modifier ‘exclusive’ successfully captures the connotative meaning. Zafrulla Khan’s rendering of the original irghab as ‘supplicate earnestly’ satisfies to some extent the denotative meaning. Yet because it abandons the original marked word order, it stops short of capturing the connotative meaning. Employing the modifier ‘earnestly’ seems inadequate to convey the intended meaning of EXLUSIVITY. Asad’s rendering exhibits over-translation. Providing ‘turn with love’ as an equivalent to the original irghab does not match the denotative meaning. It involves an over-translation by being more specific and more emotionally loaded than the original. Thus, it signals an erroneous expressive meaning. Abdel Haleem’s translation captures, to some extent, the denotative meaning of the original. Yet it is less than perfect, because by abandoning the original marked word order, it fails to deliver the intended connotative meaning. Qarāʾī’s rendering shows an obvious under-translation. Providing ‘turn eagerly to’ as an equivalent to the QIPV at hand suggests significant loss in the denotative meaning—let alone the connotative meaning, which was mismatched by discarding the original marked word order. Emphasizing the words ‘turn’ and ‘your’ by presenting them in italics and inserting the modifier ‘eagerly’ do not compensate for this loss. Bakhtiar’s translation imitates Arberry’s translation to some degree in terms of providing the term ‘quest’ as an equivalent to the original irghab, thus mismatching the denotative meaning. What is more, like Arberry’s translation, Bakhtiar’s translation mismatches the connotative meaning by failing to convey the intended sense of EXLUSIVITY syntactically (i.e. word order) and semantically (i.e. inserting modifiers). Emphasizing the pronoun ‘your’ by presenting it in bold does not compensate for this loss. All that being said, al-Hilālī and Khān’s rendering seems to be the most adequate. By inserting the in-text glosses and adhering to the original marked word order, they successfully attend to both the connotative and the denotative meanings. Inserting the word (Alone) and capitalizing its initial letter significantly
172
Data analysis, assessment, and discussion
enhances the intended meaning of EXLUSIVITY. In addition, inserting the phrase (all your) as a modifier to (intentions and hopes)—given as an equivalent to the original irghab—satisfies the denotative meaning as understood by the Qur’an exegetes. Hence, they achieve functional-pragmatic equivalence.
6.11 Discussion Having generally overviewed the selected translators’ performance in tackling the QIPVs, it is time now to critique the critical characteristics of their performance. In this section, I investigate the performance of the selected translators in relation to the five research questions that address the practical component of this study. The aim here is to approach the data from different viewpoints, to answer the research questions and to determine several theory-based recommendations. I start by tackling the first research question: To what extent are the selected translators of the Qur’an successful in providing the English functional-pragmatic equivalents of the QIPVs? Table 6.11 summarizes the selected translators’ overall achievement of the functional-pragmatic equivalents of the QIPVs. A close examination of the selected translators’ respective performances reveals that they adopted different procedures when rendering the QIPVs. They all experienced difficulties, however, in applying their approaches. This is manifested in their demanding overall achievement of functional-pragmatic equivalents of the QIPVs. Table 6.11 shows that the average score for the adequacy of the functional-pragmatic equivalents provided for the QIPVs was quite low (32%). This clearly supports the thesis of this book and demonstrates that the QIPVs posed translation difficulties for the selected translators. In what follows, to attend to the research question at hand, I explore in more detail the translators’ efforts in dealing with the QIPVs. Table 6.11 Translators’ overall achievement of functional-pragmatic equivalence Translator
Adequate functional- Inadequate Average of Average of pragmatic functional-pragmatic adequacy inadequacy equivalence equivalence
Pickthall 5 Yūsuf ʿAlī 6 Arberry 1 Shakir 3 Zafrulla Khan 3 al-Hilālī and 4 Khān Asad 3 Abdel Haleem 3 Qarāʾī 2 Bakhtiar 2 Total 32
5 4 9 7 7 6
50% 60% 10% 30% 30% 40%
50% 40% 90% 70% 70% 60%
7 7 8 8 68
30% 30% 20% 20% 32%
70% 70% 80% 80% 77%
Data analysis, assessment, and discussion 173 Pickthall scored an average of 50% adequacy on the functional-pragmatic equivalents provided for the QIPVs (see Table 6.11). In rendering the metaphorical QIPVs, he reduced the metaphor to sense in most instances (see case studies 1, 2, 4, 5, and 6). In so doing, he has successfully represented the informative meaning. He did fail, however, to capture the intended illocutionary force of the original (see case studies 2 and 5). On two occasions, Pickthall endeavoured to reproduce the metaphor in the TL without adding a sense (see case studies 3 and 7). Although he managed to adequately deliver the intended meaning of one QIPV (see case study 3), he under-translated the other (see case study 7). In rendering the figurative QIPVs, however, Pickthall exhibited different levels of adequacy (see case studies 8, 9, and 10). By using a source-text-oriented procedure and by adhering to the original syntactic structure, he succeeded in delivering both the denotative meaning and the connotative meaning of two of the figurative QIPVs (see case studies 8 and 9). Yet by applying a target-text-oriented procedure and by abandoning the original syntactic structure, Pickthall under-translated the third figurative QIPV and mismatched its connotative meaning (see case study 10). This being said, some of the procedures that Pickthall adopted to render the QIPVs did not support his intended purpose—to embark on a new translation of the Qur’an—or the approach he indicated he would follow. Pickthall explicated the purpose of his translation of the Qur’an as follows: The aim of this work is to present to English readers what Muslims the world over hold to be the meaning of the words of the Qur’an, and the nature of the Book, in not unworthy language and concisely, with a view to the requirements of English Muslims. (p. i) He expounded his approach as follows: The Book is here rendered almost literally and very effort has been made to choose befitting language. . . . Every care has been taken to avoid unwarrantable renderings. On the one or two occasions where there is departure from the traditional interpretation, the traditional rendering will be found in a footnote. (p. i) Pickthall’s purpose and approach can well be understood as seeking to provide maximum equivalence at the informative level given that he pledged to keep as closely as possible to the original. The aforesaid analysis shows that the informative level of meaning has been relatively attended to by him when the procedures to reduce the metaphor to sense and to adhere to the source text grammatical structures are adopted. Nonetheless, this level of meaning was sacrifced when he reproduced the metaphor in the TL without adding a sense and by abandoning the source text grammatical structures. This is in addition to the mismatching of connotative meanings and his sacrifcing the aesthetic and expressive levels of meaning on several occasions. This critical point leads us to discard the generality
174
Data analysis, assessment, and discussion
of the statements proposed by Kidwai (2007) and Robinson (1997) on the faithfulness and accuracy of Pickthall’s translation. Kidwai declares, A distinctive feature of Pickthall’s is its faithfulness to the original. . . . Unlike his contemporary, Abdulla Yusuf Ali, he does not paraphrase the meaning. Rather, he adheres closely to the Qur’anic text and in so doing he manages to avoid the shortcomings of literal translation. (pp. 273–274) In turn, this ‘faithfulness’ convinced Robinson to take Pickthall’s translation as the basis of comparison when evaluating other translations of the Qur’an. Robinson makes the point that Pickthall’s translation “generally gives a fairly literal rendering of the Arabic. I shall therefore use it as the basis for comparison” (p. 261). Yūsuf ʿAlī scored an average of 60% adequacy on the functional-pragmatic equivalents provided for the QIPVs (see Table 6.11). In translating the metaphorical QIPVs, he has attempted to reproduce the metaphor in the TL in two instances without adding a sense (see case studies 3 and 7). This procedure worked well in case study 3 in that the TL allows the image in this expression to be transferred. Conversely, the procedure did not work in the other case study (i.e., case study 7); there he lapsed into under-translation. On other occasions, Yūsuf ʿAlī reduced the metaphor to sense (see case studies 1, 2, 4, 5, and 6). Employing this procedure allowed him to successfully convey the informative level of meaning in the original. Yet this was at the expense of the aesthetic and expressive levels of meaning. In translating the figurative QIPVs and by opting for a source-text-oriented procedure and replicating the original marked word order of the QIPVs, Yūsuf ʿAlī came close to the denotative and connotative meanings of the original (see case studies 8, 9, and 10). Adopting the aforesaid different procedures—whether successfully or not and regardless of their effectiveness in conveying the intended denotative and connotative aspects of the QIPVs’ meanings—Yūsuf ʿAlī reflects his stated purpose for translating the Qur’an: to attend to all levels of meaning in the original. In the preface to his translation of the Qur’an, Yūsuf ʿAlī outlines this purpose: “to present it in a fitting garb in English” (p. xi) and to help nonnative Arabic readers understand its meaning and appreciate its beauty and grandeur. He writes, If [the readers] find that I have helped them even the least bit further in understanding [the Qur’an’s] meaning, or appreciating its beauty, or catching something of the grandeur of the original, I would claim that my humble attempt is justified. (p. xi) Yūsuf ʿAlī expounded his approach as follows: Gentle and discerning reader! What I wish to present to you is an English interpretation, side by side with the Arabic Text. English shall be not a mere
Data analysis, assessment, and discussion 175 substitution of one word for another, but the best expression I can give to the fullest meaning which I can understand from the Arabic text. The rhythm, music and exalted tone of the original should be reflected in the English interpretation. It may be but a faint reflection, but such beauty and power as my pen can command shall be brought to its service. (p. xii) Yūsuf ʿAlī attempted to communicate the poetic features of the Qur’an by beginning his translation with a running poetic commentary to “prepare the atmosphere” (p. xii). He also initiates each Qur’anic sūrah with a rhythmic commentary, to “introduce the subject matter” (p. xii). Given all this, Yūsuf ʿAlī’s approach may well aim to provide maximum equivalence at all levels, because he sought not only to echo the poetic and musical qualities of the original (i.e. the aesthetic and expressive levels of meaning) but also to convey the ‘literal’ meaning of the original (i.e. the informative level of meaning). On this, he wrote, “Where I have departed from the literal translation in order to express the spirit of the original better in English, I have explained the literal meaning in the Notes” (p. xii). Therefore, all levels of meaning are signifcant to Yūsuf ʿAlī and ought to be reproduced in the TL. They are so important to him that “when loss is unavoidable at any level of meaning, he compensates for it at any other level” (El-Zeiny, 2011, p. 256). Hence, Yūsuf ʿAlī’s approach in translating the QIPVs may well be described as paraphrasing. This aligns with Kidwai’s description of Yūsuf ʿAlī’s translation of the Qur’an. He indicates that “it offers a paraphrase of the meaning of the Qur’an in English rather than a strict, literal translation” (Kidwai, 2007, p. 5). Arberry scored an average of 10% adequacy on the functional-pragmatic equivalents provided for the QIPVs (see Table 6.11). In dealing with the metaphorical QIPVs, he opted in most instances to reproduce the metaphor in the TL without adding a sense (see case studies 1, 2, 5, 6, and 7). Although this procedure is the most appropriate in rendering metaphors when the TL permits transferring the metaphor, it sometimes results in complete nonsense and ambiguity (ElZeiny, 2011). In all the illustrative examples, Arberry’s procedure to reproduce the metaphor in the TL without adding a sense to it was unsuccessful. It caused him to lapse into mistaking the literal denotative meaning for the metaphorical one (see case studies 1, 2, 6, and 7), miscalculating the intended implicature (see case study 2) and under-translating the metaphor by conveying it in a general sense (see case study 7). Hence, he deviated from the informative, aesthetic, and expressive levels of meaning. Arberry’s inclination towards reproducing the metaphor in the TL may well be attributed to his purpose in attempting a new translation of the Qur’an and to his predetermined approach. In the preface of his translation, he expounded his purpose as follows: [The] chief reason for offering this new version of a book which has been ‘translated’ many times already is that in no previous rendering has a serious attempt been made to imitate, however imperfectly, those rhetorical and
176
Data analysis, assessment, and discussion rhythmical patterns which are the glory and the sublimity of the Koran. I am breaking new ground here; it may therefore be thought appropriate to explain in short my intentions and my method. (p. 1:25)
Hence, Arberry embarked on the task of translating the Qur’an, aiming to echo the scripture’s rhetorical and aesthetic qualities by rendering its verses in a poetic style. In adopting this approach, Arberry aspired “to give, relatively, maximum equivalence at the expressive and aesthetic levels because he [wanted] to echo the music and rhetorical grandeur of the original” (El-Zeiny, 2011, p. 256). That being said, when translating the QIPVs quoted earlier, Arberry adhered to this approach. His focus was to imitate the rhetorical aspects of the original. In so doing, he predominantly attended to the aesthetic and expressive levels of meaning typically at the expense of the informative level. In two instances, however, Arberry opted to reduce the metaphor to sense (see case studies 3 and 4). This procedure appears to be justifable in that the TL does not allow for a reproduction of the metaphors. Yet this also resulted in his translation’s being less faithful to the original and resulted in under-translation because it provided fewer detailed equivalents. On the other hand, when tackling the fgurative QIPVs, Arberry demonstrated varying degrees of faithfulness to the original (see case studies 8, 9, and 10). He, for instance, adequately conveyed both the denotative and the connotative meanings of the fgurative QIPV, by employing a source-text-oriented procedure and imitating the original syntactic structure (see case study 9). Nevertheless, by providing the EIPV ‘shrink from’ as an equivalent to the QIPV, he committed a register shift and mismatched the connotative meaning (see case study 8). In another instance, he partially departed from the denotative and connotative meanings of the fgurative QIPV by adopting a target-text-oriented procedure, abandoning the original marked structure and using a less accurate equivalent (see case study 10). Arberry’s translational drawbacks might well be ascribed to his approach to abandon useful procedures such as using introduction notes, footnotes, and parenthetical glosses—as done so by other translators of the Qur’an (cf Pickthall, 1971; Yūsuf ʿAlī, 1991; Asad, 2011; al-Hilālī and Khān, 1997). Had he employed one or more of these procedures, he could have not only compensated for the loss at the informative level of meaning but also delineated the connotative meanings and conveyed the illocutionary force of the given QIPVs. To this end, Arberry’s approach in rendering the QIPVs is consistent with his predetermined approach to translate the Qur’an in general and to his approach to translating the Qur’anic metaphors. In the article Criteria for the Translation and Assessment of Qur’anic Metaphor: A Contrastive Analytical Approach, El-Zeiny (2011) arrives at a similar conclusion. She writes, Arberry’s approach to metaphor is consistent to his will of echoing the rhetoric of the original. He reproduces metaphor except when this approach leads to nonsense. . . . However, Arberry sometimes lapses into semantic
Data analysis, assessment, and discussion 177 deviation by reproducing metaphors that should be reduced to sense. . . . The rhetoric of the original is not always properly echoed because it clashes with the rhetoric of the TL. Thus, Arberry’s over-concern with reproducing the aesthetic and expressive meanings sometimes affects the informative meaning. (p. 264) Shaker scored an average of 30% adequacy on the functional-pragmatic equivalents provided for the QIPVs (see Table 6.11). In tackling the metaphorical QIPVs, he resorted to reducing the metaphor to sense in almost all instances (see case studies 1, 2, 3, 4, 5, and 6). In one instance, he reproduced the metaphor in the TL without adding a sense (see case study 7). In using this procedure, he managed to capture the intended meanings of one QIPV (see case studies 6). Yet he committed a register shift, under-translated, and mismatched the illocutionary force of other QIPVs (see case studies 1, 2, 3, 4, and 5). In using the latter procedure, Shakir lapsed into under-translation (see case study 7). When dealing with the figurative QIPVs, however, Shakir adhered to the original’s marked syntactic structure and accurate word choice in two instances (see case studies 8 and 10). He exhibited semantic deviation from the original in one instance (see case study 19). In his translation of the Qur’an, Shakir stated neither his purpose in attempting a new translation nor the approach that he intended to adopt. However, his performance in dealing with the QIPVs shows that he tended to give priority to the informative level of meaning over the other levels of meaning in the QIPVs. Zafrulla Khan scored an average of 30% adequacy on the functional-pragmatic equivalents provided for the QIPVs (see Table 6.11). In rendering the metaphorical QIPVs, he predominantly reduced the metaphor to sense (see case studies 1, 2, 4, 5, and 6). In so doing, he succeeded in capturing the intended meaning in one case only (see case study 6). In other cases, however, he lapsed into semantic deviation, mismatching the illocutionary force (see case studies 1 and 5) and under-translation (see case studies 4 and 6). In other instances, Zafrulla Khan opted to reproduce the metaphor in the TL without adding a sense (see case studies 3 and 7). Thus, he successfully conveyed the intended meaning in one case, where the TL permits transferring the metaphor (see case study 3). Yet he failed to successfully convey the intended meaning in the other cases, wherein he lapsed into under-translation (see case study 7). In dealing with the figurative QIPVs, however, Zafrulla Khan exhibited adherence to the marked syntactic structures of the original in two instances (see case studies 8 and 9). This procedure helped him to convey the denotative and connotative meanings of the original in one instance (see case study 9). His inaccurate choice of words, however, caused him to lapse into a register shift and under-translation in the other instance (see case study 8). Zafrulla Khan succeeded in capturing the denotative meaning of the original in the third instance, but he failed to attend to the connotative meaning because of his abandoning the original’s marked word order (see case study10). Zafrulla Khan did not state his purpose for producing a new translation of the Qur’an, nor did he mention the approach that he adopted for
178
Data analysis, assessment, and discussion
his translation. In the preface to the translation, the editors did, however, allude to Zafrulla Khan’s aim and approach: This handy volume is a humble and modest effort at making a beginner’s study of the Quran a cheering, even an absorbing and profitable experience. The translation is strictly faithful in meaning to the text though it spells out, where it is inescapable, the inherent implications which would be obvious to one familiar with the language of the original and the context of the events referred to, but would escape those who do not possess that advantage. Even the Arabic idioms has been retained except where adherence to it would make the meaning elusive and difficult to discover. (Preface, p. 8) The editors also described Zafrulla Khan’ work as “an excellent translation, faithful and literal, which by the adoption of certain simple devices has made it comparatively easy to discover the meaning, so far as a translation is capable of conveying it” (Preface, p. 7). Given this and the procedures that Zafrulla Khan used when rendering the QIPVs, we can infer that he aimed to provide maximum equivalence at all levels of meaning by achieving a balance between them. While he succeeded in doing so in several instances, he failed in others. These failures may well be attributed to two distinct factors: one of a linguistic nature manifested in the translational pitfalls presented and discussed earlier and one of a dogmatic nature. This applies to Zafrulla Khan’s semantic deviation from the QIPV in case study 5. His odd rendering of this QIPV is attributed to his being a staunch follower of the Ahmadi Qadyani doctrine, which dogmatically rejects the notion of miracles. According to this dogmatic principle, all miracles cited in the Qur’an are denied, twisted, and deciphered in an allegoric or symbolic way, not only by Zafrulla Khan but also by his fellow Ahmadi Qadyani Qur’an translators (cf Kidwai, 2007, pp. 88, 107, 230, 412; Robinson, 1997, pp. 265–266). Along these lines, Kidwai (2007) makes the point that “Khan is an unmistakably Qadyani translator who unscrupulously distorts the import of several ayahs with a view to lending credence to the Qadyani stance” (p. 230). Al-Hilālī and Khān scored an average of 40% adequacy on the functionalpragmatic equivalents provided for the QIPVs (see Table 6.11). Their main procedure in dealing with the metaphorical QIPVs was to reduce the metaphor to sense (see case studies 1, 2, 4, 5, and 6). By using this procedure—though in some instances it resulted in adequately capturing the intended meanings of the QIPVs (see case studies 3, 5, and 6)—al-Hilālī and Khān lapsed into mismatching the illocutionary force and into over-translation for other QIPVs (see case studies 2 and 4). Al-Hilālī and Khān’s second procedure to deal with the metaphorical QIPVs was to reproduce the metaphor in the TL without adding a sense (see case studies 3 and 7). In one instance, they succeeded in attending to the intended meaning of the original because the TL permits the transference of that metaphor (see case study 3). They failed to do so in the
Data analysis, assessment, and discussion 179 other instance, however, where they lapsed into under-translation (see case study 7). In tackling the figurative QIPVs, al-Hilālī and Khān adhered to the original’s syntactic structures by adopting a source-text-oriented procedure. The equivalents that they provided were more adequate than those provided by other translators in two instances (see case studies 9 and 10). Yet they exhibited less adequacy in the third instance, as evidenced by under-translation and a register shift (see case study 8). The register shift resulted from their tendency to provide informal EIPVs as equivalents to several formal QIPVs (see case studies 1, 5, and 8). This being said, al-Hilālī and Khān declared that their principal purposes for attempting a new translation of the Qur’an were to correct the “serious errors which the previous translators had committed” and to clarify “the obscure sentences which the previous translators have not done, so as to leave the English reader in perplexity” (Kidwai, 2007, p. 201). Al-Hilālī and Khān did not elaborate on the approach that they took in their translation. Nonetheless, a close examination of their work reveals that they employed an interpretive approach, depicting “an explanatory translation” (El-Zeiny, 2011, p. 256). This approach is characterized by the heavy employment of in-text parenthetical glosses, transliterations, and footnotes. Hence, they seemingly aimed to provide “maximum equivalence at the informative level saving no effort to echo all the other levels of meaning” (El-Zeiny, 2011, p. 256). Given all this, their excessive employment of the parenthetical explanatory glosses helped, to a certain degree, to illuminate the QIPV meanings and in turn deliver adequate equivalents (see case studies 6, 9, and 10). The excessive employment resulted, however, in a distortion of the communicative purpose of the QIPV (see case study 1) and lapsed into over-translation (see case study 4). Therefore, the QIPVs’ informative level of meaning was not always attended to by al-Hilālī and Khān. Asad scored an average of 30% adequacy on the functional-pragmatic equivalents provided for the QIPVs (see Table 6.11). His approach in dealing with the metaphorical QIPVs was predominantly to reduce the metaphor to sense (see case studies 1, 2, 3, 4, 5, and 6). At times, he sacrificed the aesthetic and expressive levels of meaning and lapsed into register shift, failed to capture the illocutionary force, and demonstrated under-translation (see case studies 1, 2, 3, and 7). However, he avoided mistaking the literal meaning for the metaphorical one, satisfied the informative level of meaning, and captured the illocutionary force of the QIPVs in other instances (see case studies 4, 5, and 6). In one instance, Asad reproduced the metaphor in the TL without adding a sense. In so doing, he under-translated the QIPV and sacrificed the informative, expressive and aesthetic levels of meaning (see case study 7). In dealing with the figurative QIPVs, Asad adhered to the original’s marked word order and adopted a source-text-oriented procedure. This allowed him to remain close to the denotative and connotative meanings of the QIPVs (see case studies 8, 9, and 10). Yet such a closeness was undermined by over-translation on occasions (see case studies 8 and 10) and a register shift once (see case study 9). Asad’s approach to dealing with the QIPVs goes hand in hand with his ultimate purpose in translating the Qur’an: to bridge
180
Data analysis, assessment, and discussion
the gap left by previous translators and bring the scripture closer to the hearts and minds of a non-Arab and non-Muslim readership: [N]one of these translations—whether done by Muslims or by nonMuslims—has so far brought the Qur’an nearer to the hearts or minds of people raised in a different religious and psychological climate and revealed something, however little, of its real depth and wisdom . . . the Qur’an itself has never yet been presented in any European language in a manner which would make it truly comprehensible. (p. xvi, italics in original) Asad elaborated on his approach to bridging the gap: The work which I am now placing before the public is based on a lifetime of study and of many years spent in Arabia. It is an attempt—perhaps the first attempt—at a really idiomatic, explanatory rendition of the Qur’anic message into European language. (p. xix) Unlike Yūsuf ʿAlī and Arberry, Asad made no claim to imitate the poetic and rhetorical characteristics of the Qur’an: I make no claim to having reproduced anything of the indescribable rhythm and rhetoric of the Qur’an. No one who has truly experienced its majestic beauty could ever be presumptuous enough to make such a claim or even to embark upon such an attempt. I am fully aware that my rendering does not and could not really “do justice” to the Qur’an and the layers upon layers of its meaning. (p. xxiii) More importantly, Asad stated the principle that he applied to perceiving and translating the message of the Qur’an: [I]f it is to be truly comprehensible in another language, the message of the Qur’an must be rendered in such a way as to reproduce, as closely as possible, the sense which it had for the people who were as yet unburdened by the conceptual images of later Islamic developments: and this has been the overriding principle which has guided me throughout my work. (p. xxi) Therefore, Asad’s aim in translating the Qur’an may be interpreted linguistically as to provide maximum equivalence at the informative level of meaning. This approach is evidently embodied in his treatment of the QIPVs, where he endeavoured to unfold that level of meaning. Asad achieved this goal, to a certain
Data analysis, assessment, and discussion 181 degree, by using the aforesaid procedures and by using footnotes to further elucidate the QIPVs’ layers of meaning in light of the classical exegetes’ views (see case study 5). Abdel Haleem scored an average of 30% adequacy on the functional-pragmatic equivalents provided for the QIPVs (see Table 6.11). He adopted the procedure to reduce the metaphor to sense in all metaphorical QIPVs (see case studies 1, 2, 3, 4, 5, 6, and 7). Although he provided an adequate translation in several instances, he lapsed into under-translation and mismatched the illocutionary force in most of them. In dealing with the figurative QIPVs, however, Abdel Haleem remained more faithful to the original by adhering to the source-text marked syntactic structures in two instances (see case studies 8 and 9). He in one instance succeeded in capturing the denotative and the connotative meanings (see case study 8) but in another instance failed to do so and lapsed into register shift (see case study 9). He also failed to convey the connotative meaning of the QIPV in case study 10 because of his abandoning the original’s marked word order. Abdel Haleem’s tendency to paraphrase and to reduce the metaphor to sense aligns with his purpose for attempting a new translation of the Qur’an and with the approach that he adopted to deal with Arabic idioms. His purpose was primarily to “make things clear” for modern readers of English by presenting the Qur’an into “clear and lucid modern English, free of archaisms and literal Arabisms that have been a source of obscurity for modern readers” (Abdel Haleem, 2010, p. v). He described his intention as follows: This translation is intended to go further than previous works in accuracy, clarity, flow, and currency of language. It is written in a modern, easy style, avoiding where possible the use of cryptic language or archaisms that tend to obscure meaning. The intention is to make the Qur’an accessible to everyone who speaks English, Muslims or otherwise, including the millions of people all over the world for whom the English language has become a lingua franca. (p. xxviii) Regarding his management of Arabic idioms, Abdel Haleem indicated that “Throughout this translation, care has been taken to avoid unnecessarily close adherence to the original Arabic structures and idioms, which almost always sound unnatural in English. Literal translations of Arabic idioms often result in meaningless English” (p. xxx). What is more, he provided readers with the following advice when they encountered an ‘unnatural’ rendering: If readers think an interpretation is unusual, they are advised first to consider the meaning in context, and the requirements of idiomatic English, and to look at important classical Arabic dictionaries and works of exegesis to see that nothing is included here without solid linguistic and exegetical bases. Like any human endeavour, all translations are open to improvement. (p. vi)
182
Data analysis, assessment, and discussion
Given all this, Abdel Haleem clearly sought to provide maximum equivalence at the informative level of meaning. This aim was fulflled by reducing the metaphor to sense and by coupling it with a footnote and in-text glosses. Qarāʾī scored an average of 20% adequacy on the functional-pragmatic equivalents provided for the QIPVs (see Table 6.11). In tackling the metaphorical QIPVs, he was inclined in most instances to reduce the metaphor to sense (see case studies 1, 2, 3, 4, 5, and 6). In that endeavour, he achieved varying degrees of adequacy. In some instances, he succeeded in conveying the intended meaning of the original (see case studies 6), lapsed into register shift (see case study 1), mismatched the illocutionary force (see case study 2), and engaged in under-translation (see case studies 3, 4, and 5). In other instances, he opted to reproduce the metaphor in the TL without adding a sense. This provided a less-than-perfect translation as a result of its being more general than the original (see case study 7). In dealing with the figurative QIPVs, Qarāʾī remained faithful to the original by employing a source-text-oriented procedure and by adhering to the original’s syntactic structure in two instances (see case studies 8 and 9). He adequately conveyed the denotative and connotative meanings of the original in one instance (see case study 9) and failed to do so in the other instance (see case study 8). In case study 10, his translation was less faithful to the original in that he adopted a target-text-oriented procedure and demonstrated under-translation for the QIPV. Hence, he failed to adequately provide the functional-pragmatic equivalence of the original. Nonetheless, all the aforesaid procedures used by Qarāʾī in dealing with the QIPVs were in line with his aim of translating the Qur’an and the approach that he proclaimed to take. Qarāʾī presented his aim and approach as follows: As my principal aim was to provide a translation affording direct access to the Arabic Qur’an, I have tried, so far as possible, to maintain a formal equivalence between the phrases and clauses of the source and the target text, but I have not hesitated to make adjustments when required by the need for intelligibility, clarity and naturalness of expression, so far as permitted by the constraint imposed by the method of “mirror-paraphrasing”. (pp. xvii–xviii) By mirror-paraphrasing, Qarāʾī meant the phrase-by-phrase approach that he adopted in his translation of the Qur’an. Here, he “intended to bring some of the advantage of the interlinear translations to English-speaking readers of the Holy Qur’an” (Qarāʾī, 2004, p. xvii). By taking this approach, Qarāʾī elaborated, “the translation of the source text develops phrase by phrase, with the translation appearing opposite the corresponding phrase in Arabic. Each phrase in the target or receptor language mirrors the semantic import of the phrase in the [ST]” (Qarāʾī, 2004, p. xvii, emphasis in original). When implementing this approach, Qarāʾī restricted himself to two constant constrains: [First], of having to cover the complete meaning of each phrase of the [ST] in corresponding phrase of the [TT], and second, of connecting the
Data analysis, assessment, and discussion 183 successive phrase in such a manner as to generate, so far as possible, a fluent, clear, intelligible, natural and stylistically acceptable prose. (p. xvii) Interestingly, Qarāʾī paid special attention to the Qur’anic idioms and classifed them into three categories according to their translatability into English. He also specifed the procedure that he adopted when dealing with each category. The categories of Qur’anic idioms and Qarāʾī’s proposed procedures were as follows: (1) “Arabic idioms which though unfamiliar to the English-speaking audience are not difficult to understand when translated literally. These have been rendered literally” (p. xviii). (2) “Some [Arabic idioms which] are unintelligible when translated literally. These have to be paraphrased appropriately in order to be understood” (p. xix). (3) “In certain cases it may be possible to substitute an English idiom” (p. xx). Qarāʾī clearly sought to provide maximum equivalence at all levels of meaning, yet when a tension occurred among them, he prioritized the informative level of meaning, in an attempt to ‘mirror-paraphrase’ the original. At this point, I should say that this is the recommended best approach for tackling Qur’anic idioms in general and QIPVs in particular. This is especially true when it is coupled with the working procedure to provide footnotes, as done by Qarāʾī (see case study 5). In applying this approach to translating the QIPVs, Qarāʾī successfully conveyed their denotative and connotative imports to a certain extent. Yet this was not without lapsing into the translational pitfalls mentioned earlier. These pitfalls may well be attributed to his adhering so closely to the what-is-said level of meaning. Qarāʾī’ and other Qur’an translators need to tend towards the what-is-implicated level of meaning if they are to achieve the functional-pragmatic equivalence and strike a balance between the original’s informative, aesthetic, and expressive levels of meaning. Bakhtiar scored an average of 20% adequacy on the functional-pragmatic equivalents provided for the QIPVs (see Table 6.11). Her approach in translating the metaphorical QIPVs was primarily to reduce the metaphor to sense (see case studies 1, 3, 4, 5, and 6). In so doing, she achieved the functionalpragmatic equivalence in only two instances (see case studies 1 and 6). In other instances, Bakhtiar lapsed into under-translation, semantic deviation, mistaking the literal for the metaphorical meaning, and failing to capture the illocutionary force of the original (see case studies 3, 4, and 5). In two other instances, Bakhtiar reproduced the metaphor without adding a sense (see case studies 2 and 7). In so doing, she failed to capture the functional-pragmatic equivalents of the QIPVs, because of under-translation and by mistaking the literal meaning for the metaphorical meaning of the original. In tackling the figurative QIPVs, Bakhtiar adhered to the original marked word order and adopted a source-textoriented procedure in two instances (see case studies 8, and 9). In so doing, she
184
Data analysis, assessment, and discussion
successfully delivered the denotative and connotative meanings of the original in one instance (see case study 9) and failed to do so in the other instance (See case study 8). Bakhtiar abandoned the original marked word order in case study 10 when she adopted a target-text-oriented procedure, which resulted in mismatching the denotative and connotative meanings of the QIPV. Her inconsistencies in dealing with the QIPVs contradicted her stated purpose for translating the Qur’an and the approach that she adopted in her translation. Having found that previous English translations “lack internal consistency and reliability” (Bakhtiar, 2012, p. xiii), Bakhtiar aimed to bridge the gap by attempting a new translation of the Qur’an: This translation differs from previous English translations in that there has been a conscious attempt to present a translation of the sacred text that has internal consistency and reliability. It is the first critical English translation by a woman who includes the view of women in the Signs (verses) wherever relevant. The translation is consciously a universal, inclusive, one widening the relevance of the sacred text to larger community. (p. xxi) By recognizing that “no translation of the Quran can compare in beauty and style with the original Arabic” (p. xiii), Bakhtiar wasted no effort on echoing the stylistic and poetic qualities of the Qur’an. She instead opted for “formal equivalence in order to be as close to the original as possible” (p. xiv). Therefore, she focused on “[i]ntroducing the non-Arabic speaker to the words of the revelation without any commentary” (p. xv) and with “no parenthetical phrases . . . further interpreting and elaborating a verse, thus allowing the translation, as the Quran itself is, to be free of any transient political, denominational or doctrinal bias” (p. xvi). To this end, Bakhtiar’s aim in translating the Qur’an might well have been to provide maximum equivalence at the informative level, but at the expense of the other levels of meaning. Strikingly, this has not been the case in her translation of the QIPVs. On a few occasions, she succeeded in satisfying the informative level of meaning by attending to the illocutionary force and adhering to the original marked word order (see case studies 1, 6, and 9). In the main, however, Bakhtiar was unsuccessful in her attempt to provide a translation “as close to the original as possible”. She lapsed into taking an unnatural approach (see case study 2), committing semantic deviation (see case study 4), engaging in under-translation (see case studies 3 and 7), and mismatching the denotative and connotative meanings (see case study 10). Such translational pitfalls could well be attributed to Bakhtiar’s predetermined approach to not provide parenthetical glosses, commentaries, or footnotes. Undeniably, adopting these procedures would have helped to illuminate the intended meanings of the complex QIPV expressions. These procedures are useful in compensating for the loss of meaning not only at the informative level but also at the aesthetic and expressive levels.
Data analysis, assessment, and discussion 185 Almost all the selected translators (except for those who aimed to imitate the stylistic and poetic qualities of the Qur’an) pledged to produce translations as close as possible to the original’s intended meaning: to provide maximum equivalence at the informative level of meaning. In their respective attempts to echo the aesthetic and expressive levels of meaning, however, all the selected translators sacrificed the informative level in several instances and stumbled into translational pitfalls. As confirmed earlier, when translating a sensitive text such as the Qur’an, the informative level of meaning must be given priority, because, “though a linguistic miracle, the primary aim of the Qur’an is to ‘teach and preach’ then to ‘please’ those who read it. Therefore, the translator should act as a ‘message transmitter’ not as an ‘imitator’” (El-Zeiny, 2011, p. 256). Thus, the translators should put no effort into echoing the stylistic and poetic qualities of the Qur’an, because they are Qur’an-specific properties and surpass humankind’s “faculty to reproduce [them] in a translated from” (Abdul-Raof, 2001, p. 2). What is more, in their endeavours to echo the aesthetic and expressive levels of meaning by reproducing the metaphor in the TL without adding a sense, most of the translators mistook the literal meaning of the QIPVs for the QIPVs’ metaphorical meanings. Literal translations entail unjustifiable semantic deviations and in turn constitute a deviation from the informative level of meaning as well as from the aesthetic and expressive levels. Therefore, they should be avoided if the translations are to do justice to the original. This point is pertinent not only to translating the QIPVs but also to dealing with other aspects of the Qur’anic discourse. Let us now move on to the second research question that addresses the practical component of this study: What are the types of translational pitfalls that the selected translators of the Qur’an stumbled into when tackling the issue of QIPVs? As outlined in Tables 6.1–6.10, the selected translators fell into several translational pitfalls. Using House’s (1977) taxonomy, those pitfalls may well be categorized into two types: overtly erroneous errors and covertly erroneous errors. Overtly erroneous errors were made by the selected translators 45 times (60.81%) (see Tables 6.1–6.10). They were of the following four types: (1) Literal translation was demonstrated by the selected translators four times (see Tables 6.1–6.10). This translational pitfall is manifested in reproducing the metaphor in the TL without adding a sense when the TL does not permit such a reproduction. It occurred mainly when dealing with the metaphorical QIPVs whereby the selected translators mistook the literal meaning for the metaphorical meaning, which happened for two reasons: the selected translators’ pretext of being too close to the original and/or the exaggerated efforts of the translators to echo its aesthetic qualities. Literal translation “does not produce the appropriate sense” (Newmark, 1988, p. 285) and must therefore be avoided to achieve functional-pragmatic equivalence. This is because literal translations sometimes succeed in conveying
186
Data analysis, assessment, and discussion
the locutionary act/prepositional meaning of the original. Yet it is certainly the main reason for mismatching its illocutionary and perlocutionary acts (cf Hale, 1996; Aldahesh, 2009). (2) Under-translation was manifested in providing the general denotative and/ or connotative meanings of both the metaphorical and the figurative QIPVs. The selected translators fell into this pitfall 29 times (see Tables 6.1–6.10). This high number confirms Newmark’s (1988) assertion that “Most translations are under-translations, but their degree of under-translation is too high” (p. 285). (3) Over-translation was demonstrated by the selected translators four times (see Tables 6.1–6.10). This translational pitfall manifests in being more specific than the original by unjustifiably adding a sense to the denotative and/or connotative meanings of both the metaphorical and the figurative QIPVs. (4) Semantic deviation was demonstrated by the selected translators seven times (see Tables 6.1–6.10). This translational pitfall occurs when providing an erroneous denotative meaning and/or connotative meaning of both the metaphorical and the figurative QIPVs. Most cases of semantic deviation are ascribed to the translators’ wrong word choice and the use of inappropriate procedures. However, it is sometimes ascribed to their misguided dogmatic principles, as manifested in Zafrulla Khan’s odd translation of Qur’an, 18:11, in case study 5. Covertly erroneous errors were made by the selected translators 29 times (39.19%) (see Tables 6.1–6.10). They were of the following two types: (1) Register shift was demonstrated by the selected translators 13 times (see Tables 6.1–6.10). This translational pitfall is manifested in providing EIPVs as equivalents to the QIPVs. Unlike the highly formal QIPVs, EIPVs are commonly used in everyday informal spoken English. Therefore, translators should consider that unacceptable register variations might convey an incorrect message and distort the intended meaning of the original (cf Hale, 1997; Hale, 2002; Aldahesh, 2009). (2) The incorrect delivery of speech acts was demonstrated by the selected translators 16 times (see Tables 6.1–6.10). This translational pitfall is manifested in the failure to capture the intended illocutionary force. The high number of incorrectly delivered speech acts is attributed to the fact that all the aforesaid translational pitfalls and other kinds of mistranslations of the ST words and structures might well deliver erroneous speech acts and “affect the calculability of implicatures in the [TT]” (Baker, 1992, p. 229). According to all the claims just presented, the selected translators made overtly erroneous errors more often than they made covertly erroneous errors. Of the overtly erroneous errors, under-translation errors occurred most frequently, followed by semantic deviations, then literal translation errors, and fnally
Data analysis, assessment, and discussion 187 over-translation errors. Of the covertly erroneous errors, failure to match the illocutionary force errors occurred most frequently, followed by register shift errors. Such a diverse range of translational pitfalls stumbled into by the selected translators leads us to attend to the third research question that addresses the practical component of this study: What were the procedures adopted by the selected translators when they dealt with QIPVs? Tables 6.1–6.10 indicate that when dealing with the metaphorical QIPVs, the selected translators did not use all the aforesaid procedures proposed by Wonderly (1971), Larson (1984), and Newmark (1988) for translating metaphors (see 5.5.1). Rather, they confined themselves to two procedures: reducing the metaphor to sense and reproducing the metaphor in the TL without adding a sense. In dealing with the figurative QIPVs, the selected translators used two procedures: opting for a source-text-oriented approach by imitating the original’s marked syntactic structure and opting for a target-text-oriented approach by abandoning the original’s marked syntactic structure. Nonetheless, several of the selected translators coupled these procedures with useful procedures, such as inserting parenthetical glosses and/or providing explanatory footnotes. Moreover, having put forward different purposes and having adopted different approaches to translating the Qur’an, the selected translators differed significantly in their providing “maximum equivalence” at one level of meaning or another when rendering the QIPVs. This point aligns with previous theory-based results reported in a study of Qur’anic metaphor translations. The study’s results showed that “translators have given ‘maximum equivalence’ at different levels of meaning according to their scale of preferences. The maximization applies to their overall approaches to translating the Qur’an including metaphor” (ElZeiny, 2011, 256). Furthermore, in dealing with the phenomenon of QIPVs, each selected translator did not confine themselves to using only one procedure. Rather, they made use of several procedures. Yūsuf ʿAlī, for instance, employed two procedures when translating the metaphorical QIPVs (i.e. reducing the metaphor to sense and reproducing the metaphor in the TL without adding a sense). In addition, he coupled them with parenthetical glosses and footnotes. This affirms El-Zeiny’s (2011) point that “even in translating a single linguistic phenomenon, [a translator’s] procedures differ according to the context. In other words, there are approaches within approaches” (p. 264). This leads us to the fourth research question that addresses the practical component of this study: What are the most adequate and recommendable procedures employed by the selected translators in rendering the QIPVs? Table 6.12 presents the ranking of the procedures used by the selected translators in achieving the functional-pragmatic equivalents of the QIPVs according to their frequency and usage adequacy. As can be seen in Table 6.12, the procedure of reducing the metaphor to sense was the most frequently used by the selected translators to deal with the metaphorical QIPVs. Yet the procedure achieved an adequate functional-pragmatic
188
Data analysis, assessment, and discussion
Table 6.12 Ranking procedures according to their frequency and usage adequacy Procedures
Frequency of usage
Adequate usage
Inadequate usage
Reducing the metaphor to sense Imitating the original’s marked syntactic structure Reproducing the metaphor in the TL without adding a sense Abandoning the original’s marked syntactic structure
52
16
36
24
13
11
18
4
14
6
0
6
equivalence of the QIPVs only 16 times. Despite this procedure’s high level of frequency, it also constituted the second-most imperfectly used procedure (36 times). The high inadequacy ratio for this procedure is attributed to the aforesaid covertly erroneous errors and overtly erroneous errors made by the selected translators. One of the main reasons for the errors was the translators’ not considering the Qur’anic contexts in which the QIPVs were employed. Of course, translators ought to pay due attention to both the short micro and distant macro Qur’anic contexts if they are to properly penetrate the meanings of the QIPVs and achieve functional-pragmatic equivalence. Failure to do so results in a distorted translation, no matter how successful the procedure they use. Gutt (1998) rightly claims that “The same utterance can have opposite interpretations, depending on the context in which it processed” (p. 49). Moreover, Sperber and Wilson (1986) assert that “the context does much more than filter out inappropriate interpretations; it provides premises without which the implicature cannot be inferred at all” (p. 37, as cited in Baker, 1992, p. 238). The procedure to reproduce the metaphor in the TL without adding a sense emerged as the second-most frequently used procedure by the selected translators to deal with the metaphorical QIPVs. It was inadequately used 14 times and adequately used only four times. This low level of adequacy is attributed mainly to the significant sociolinguistic differences between the SL and the TL. Such a difference makes reproducing the metaphor in the TL difficult. Regarding the figurative QIPVs, the procedure to imitate the original’s marked syntactic structure was most frequently used by the selected translators. The procedure was perfectly used 13 times and improperly used 11 times. The procedure to abandon the original’s marked syntactic structure was the most incorrectly used to deal with the figurative QIPVs, with zero adequate usage and (of course) 100% inadequate usage. With all the aforesaid points in mind, I assert that the procedure to reproduce the metaphor in the TL was relatively most successful and is thus recommended
Data analysis, assessment, and discussion 189 for dealing with the metaphorical QIPVs. This is provided that the TL permits the transference of the metaphor. Reducing the metaphor to sense was the second-most adequately applied procedure and is also recommended. The successful use of this commonly employed procedure depends on the competencies of the translators and their ability to avoid the aforementioned covert and overt errors. When translating the figurative QIPVs, the translators are best advised to employ a source-text-oriented procedure by adhering to the original’s marked syntactically and semantically oriented structures. Maintaining such marked structures is a must in order to perceive, interpret, and translate sacred texts in general and the Qur’an in particular where the form and content are organically interrelated. Newmark (1988) rightly defines sacred text as “An authoritative or expressive text where the manner is as important as the matter” (p. 184). Therefore, these marked structures are so important that abandoning them does not do justice to the semantico-syntactic organic interrelations between the QIPV’s constituents that embody their idiomaticity and, most importantly, are not without their communicative purposes. Unlike the metaphorical QIPVs, which to grasp their meaning depends on the successful interpretation of their implied meanings delivered via conversational implicatures (see 4.4), appreciating the overall meanings of the figurative QIPVs depends on elucidating their conventional linguistic idiomaticity as a key aspect of the Arabic linguistic system. Before proceeding to the next research question, two vital points need to be made clear. First, the usage of inappropriate translation procedures is undeniably the main reason for many of the translational languishes discussed so far. Therefore, the recommendation is for translators to provide maximum equivalence at all levels of meaning. When a tension occurs among them, however, they ought to give priority to the informative level of meaning. This idea holds true not only when tackling QIPVs but also when dealing with other types of Qur’anic idiomatic expressions. Second, translators need to use the ‘couplets’ technique when dealing with complex and multifaceted phrases such as the QIPVs. The ‘couplets’ technique— as proposed and defined by Newmark (1988)—requires the use of two or more procedures when “dealing with a single problem” (p. 91). Newmark describes this technique as “two or more bites at one cherry” (p. 91). The need to apply the ‘couplets’ technique is manifested in the successful translations provided by a number of the selected translators. As discussed, they combined the aforementioned recommended procedures with one or more of other practical procedures, such as using parenthetical glosses and footnotes. Making use of the ‘couplets’ technique when dealing with the QIPVs has been proven to be the best practical way to compensate for the loss at the informative level of meaning and at the aesthetic and expressive levels. As previously indicated, A sound Qur’anic translation strategy is to interpret the macro co-text and if there is no solution because of target language syntactic or stylistic
190
Data analysis, assessment, and discussion limitations, then an illumination marginal note can be very useful and informative to the target language reader. (Abdul-Raof, 2001, p. 51)
Thus, confning the translation to the use of one particular procedure inevitably prevents the translator from satisfactorily attending to the functionalpragmatic equivalence of the QIPVs because of their inherent multifaceted complexities. Finally, let us now address the last research question: To what extent are the QIPVs translatable/untranslatable? The performance of the selected translators when dealing with the QIPVs confirmed the thesis of this book and demonstrated that the translators experienced various difficulties when rendering the Qur’anic problematic expressions into English. This is evident from the high percentage (77%) of inadequate translations provided as equivalents to the QIPVs (see Table 6.11). Conversely, the selected translators’ performances also demonstrated that the QIPVs are in fact relatively translatable. This is obvious from the percentage (32%) of adequate translations provided by the selected translators. Despite the lower number of adequate translations, the results nonetheless indicate, to a certain degree, the possibility of satisfactorily translating the QIPVs into English should the appropriate procedures be mindfully followed. Owing to the complex idiomatic, metaphorical, and figurative nature of the QIPVs, the adequate functional-pragmatic equivalents provided by the selected translators satisfied the informative level of meanings. Yet this was achieved at the expense of the aesthetic and expressive levels of meaning. This outcome, I must say, is quite justifiable given that the QIPVs are typically interweaved into the Qur’anic texture in many different and highly rhetorical ways to form an integral part of the Qur’an-specific discourse. This discourse “cannot be reproduced in an equivalent manner to the original in terms of structure, mystical effect on the reader, and intentionality of [ST]” (Abdul-Raof, 2001, p. 1). Against this background, imprecisions and “skewing of sensitive Qur’anic information will always be the by-product of any Qur’anic translation” (Abdul-Raof, 2001, p. 1). Like other Qur’anspecific features, the QIPVs have several unique rhetorical characteristics that do not easily lend themselves to translation. Therefore, a certain degree of loss at all levels of meaning (especially the aesthetic and expressive levels) must be accepted. Along these lines, De Pedro (1999) points out that “[a] practical approach to translation must accept that, since not everything that appears in the [ST] can be reproduced in the [TT], an evaluation of potential losses has to be carried out” (pp. 556–557). Notably, the relative translatability of the QIPVs demonstrated in this study aligns with the notion of a “crude approximation of the language, meanings and style of the Qur’an” suggested by Abdul-Raof (2001, p. 2). This is to say that a perfect translation of the QIPVs—one that accounts for the all levels of meanings—is not attainable. Rather, providing a crude approximation of their
Data analysis, assessment, and discussion 191 significations is possible and quite acceptable on the proviso that it attends to the functional-pragmatic properties of such complex Qur’anic combinations. Given this and given that “the quality of a translated text is a reflection of the translator’s competence” (Campbell, 1998, p. 8), Qur’an translator remains the one responsible for advancing their communicative competence to the utmost level, to achieve the acceptable crude approximation and to use suitable procedures. In turn, this helps to bring into play appropriate mechanisms to compensate for the by-product losses. Communicative competence in this regard is taken to mean “the knowledge and ability possessed by the translator which permits him/her to create communicative acts—discourse—which are not only (and not necessarily) grammatical but [also] socially appropriate” (Hymes, 1971, p. 23, as cited in Bell, 1991, p. 42). The translators require both theoretical and practical knowledge which comprise the following five aspects proposed by Johnson and Whitlock (1987): “[TL] knowledge; text-type knowledge; [SL] knowledge; subject area (‘realworld’) knowledge; and contrastive knowledge” (p. 137 as cited in Bell, 1991, p. 36). In other words, translators must obtain semantic, syntactic, and pragmatic knowledge of the text on which they are working. Indeed, Bell (1991) asserts that translators need to know three things: (a) how propositions are structured (semantic knowledge), (b) how clauses can be synthesized to carry propositional content and analysed to retrieve the content embedded in them (syntactic knowledge), and (c) how the clause can be realized as information-bearing text and the text decomposed into the clause (pragmatic knowledge). Lack of knowledge or control in any of the three cases would mean that the translator could not translate. (pp. 36–37; cf Aldahesh, 2009) Hence, the communicative competence of the Qur’an translator must not be confned to mastery of the linguistic systems of the Arabic and English languages. Rather, it must refect “advanced knowledge in Arabic syntax and rhetoric in order to appreciate the complex linguistic and rhetorical patterns of Qur’anic structures” (Abdul-Raof, 2001, p. 2). What is more, Qur’an translators—whether Muslim or non-Muslim, or native or nonnative speakers of Arabic—are required to consult the main authoritative classical and contemporary exegetical works on the Qur’an (as most of the selected translators in this study did). They must also consult the infuential works of Arabic lexicographers, in order to thoroughly encompass the multilayered meanings of the QIPVs, counteract their diffculties, and achieve functional-pragmatic equivalence. Before bringing this chapter to an end, I must indicate here that my aforesaid comments about the translators’ shortcomings when dealing with the QIPVs were not meant to belittle the great efforts that they put into the huge task of translating the Qur’an. Nor were they meant to undervalue the
192
Data analysis, assessment, and discussion
qualities of their great works or to underestimate their scholarly contributions to the field of Qur’an translations. The comments were simply meant to bring the translators’ and the readers’ attention to the delicate idiosyncrasies of the QIPVs and the careful treatment that they require when being translated into English.
References Abdel Haleem, M. A. S. (2010). The Qur’an: English Translation and Parallel Arabic Text. Oxford University Press. Oxford. Abdul-Raof, H. (2001). Qur’an Translation: Discourse, Texture and Exegesis. Curzon. Richmond, Surrey, UK. al-Aṣfahānī, A. (n.d.). Al-Mufradāt fī ghrīb al-Qur’ān. Maktabat Nizār Musṭafā al-Bāz. Mecca, Saudi Arabia. Aldahesh, A. Y. (2009). Translating Idiomatic English Phrasal Verbs into Arabic. VDM Verlag Dr. Muller. Saarbrucken, Germany. al-Hilālī, M. T., & Muhsin Khān, M. (1997). The Noble Qur’an: English Translation of the Meanings and Commentary. King Fahd Complex for the Printing of the Holy Qur’an. Madinah, Saudi Arabia. al-Maḥallī, J. D., & al-Suyūṭī, J. D. (2007). Tafsīr al-Jalālayn. Translated by Feras Hamza. Royal Aal al-Bayt Institute for Islamic Thought. Amman, Jordan. al-Nisābūrī, A. A. (1994). Asbābu al-Nuzūl. Dār al-Kitāb al-ʿArabī. Beirut, Lebanon. al-Rāzī, M. F. (1981). Mafātīḥ al-Ghayb. First Edition. Dār al-Fikr lil Tibāʿati wa alNashri wa al-Tawīzʿ. Beirut, Lebanon. al-Ṭabrasī, A. A. (1960). Majmaʿu al-Bayān f ī Tafsīri al-Qur’an. Dār Iḥyāʾ al-Turāth. Beirut, Lebanon. al-Zamakhsharī, J. A. M. (1998a). Al-Kashshāf ʿan Ḥaqāʾiq ghawamid al-Tanzīl wa ʿuyūn al-ʾaqāwīl fī wujūi al-Tʾwīl. First Edition. Maktabat Al-ʿubaykān. Riyadh, Saudi Arabia. al-Zamakhsharī, J. A. M. (1998b). Asāsu al-Balāghati. First Edition. Dār al-Kutub al-ʿilmiyyah. Beirut, Lebanon. Arberry, A. J. (1955). The Koran Interpreted. George Allen & Unwin LTD. London; The Macmillan Company. New York, NY. Asad, M. (2011). The Message of the Qur’an. Islamic Book Trust. Kuala Lumpur. Ayoub, M. (1984). The Qur’an and Its Interpreters. State University of New York Press. Albany. Badawi, E. M., & Abdel Haleem, M. (2008). Arabic-English Dictionary of the Qur’anic Usage. Brill. Leiden and Boston. Baker, M. (1992). In Other Words: A Coursebook on Translation. Routledge. London and New York. Bakhtiar, L. (2012). The Sublime Quran. Fifteenth Revised Edition. Kazi Publications. Chicago. Bell, R. (1991). Translation and Translating: Theory and Practice. Longman. London and New York. Cambridge Advanced Learner’s Dictionary. (2005). Second Edition. Cambridge University Press. Cambridge, UK. Campbell, S. (1998). Translation into the Second Language. Longman. London and New York.
Data analysis, assessment, and discussion 193 Concise Oxford English Dictionary. (2011). Twelfth Edition. Oxford University Press Inc. New York, NY. Dāwood, M. M. (2002). Al-Qurʿan al-Karīm wa-Tafāʿulu al-Maʿānī: Dirāsatun Dilāliyatun li-Taʿlluqi Ḥarfi al-Jarri bi- al-Fiʿili wa-ʾAtharihi f ī al-Maʿná f ī al-Qurʿan al-Karīm. Dār Gharīb. Cairo, Egypt. De Pedro, R. (1999). The Translatability of Texts: A Historical Overview. Meta, XLIV (4), 546–559. Dweik, B. S., & Abu Shakra, M. M. (2011). Problems in Translating Collocations in Religious Texts from Arabic into English. The Linguistics Journal, 5 (1), 5–41. El-Zeiny, I. (2011). Criteria for the Translation and Assessment of Qur’anic Metaphor: A Contrastive Analytical Approach. Bable, 57 (3), 247–268. Farghal, M. (1995). Euphemism in Arabic: A Gricean Interpretation. Anthropological Linguistics, 37 (3), 366–378. Gutt, Ernst-August. (1998). Pragmatic Aspects of Translation: Some Relevance Theory Observation. In Leo Hickey (Ed.), The Pragmatics of Translation, pp. 41–53. Cromwell Press. Trowbridge, UK. Hale, S. (1996). Pragmatic Considerations in Court Interpreting. ARAL, 19 (1), 61–72. Hale, S. (1997). The Treatment of Register Variation in Court Interpreting. The Translator, 3 (1), 39–54. Hale, S. (2002). How Faithfully Do Court Interpreters Render the Style of NonEnglish Speaking Witnesses’ Testimonies? A Data-based Study of Spanish-English Bilingual Proceedings. Discourse Studies, 4 (1), 25–47. House, J. (1977). A Model for Translation Quality Assessment. TBL Verlag Gunter Narr. Tubingen, Germany. Ibn Kathīr, I. U. (2003). Tafsīr Ibn Kathīr (Abridged). Second Edition. Darussalam Publishers & Distributers. Riyadh, Houston, New York, and Lahore. Ibn Qutaybah, A. M. (1978). Tafsīr Gharīb al-Qur’an. Da r al-Kutub al-ʿilmiyyah. Beirut, Lebanon. Kidwai, A. R. (2007). Bibliography of the Translations of the Meanings of the Glorious Qur’an into English 1649–2002: A Critical Study. King Fahad Quran Printing Complex. Madinah, Saudi Arabia. Larson, M. L. (1984). Meaning-Based Translations of the Qur’an: A Guide to Cross Language Equivalence. University Press of America. New York, NY. The Macquarie Dictionary. (2005). Fourth Edition. Macquarie Library. North Ryde, Sydney, NSW, Australia. Marogy, A. E. (2010). Kitab Sibawayhi: Syntax and Pragmatics. Brill. Leiden and Boston. Newmark, Peter. (1988). A Textbook for Translation. Prentice Hall. New York, London, Sydney, Toronto, and Tokyo. Oxford Wordpower: English-English-Arabic Dictionary. (2000). Third Edition. Oxford University Press. Oxford. Pickthall, M. (1971). The Meaning of the Glorious Qur’ân. Dar al-Kitab al-Lubnani. Beirut, Lebanon. Qarāʾī, A. Q. (2004). The Qur’an With a Phrase-by-Phrase English Translation. ICAS Press. London. Robinson, N. (1997). Sectarian and Ideological Bias in Muslim Translation of the Qur’an. Islam and Christian-Muslim Relations, 8 (3), 216–278.
194
Data analysis, assessment, and discussion
Shakir, M. H. (2011). The Qur’an: Translation. Tahrike Tarsile Qur’an, Inc. Elmhurst, NY. Wonderly, W. L. (1971). Bible Translations for Popular Use, (Helps for Translators). United Bible Societies. New York, NY. Yūsuf ʿAlī, ʿA. (1991). The Meaning of the Holy Qur’ān. New Edition with Revised Translation and Commentary. Amana Corporation. Brentwood, MD. Zafrulla Khan, M. (1971). The Quran: The Eternal Revelation Vouchsafed to Muhammad the Seal of the Prophets. Curzon Press. London.
7
Conclusion
7.1 Summary of the study The main purpose of this book is to explore the (un)translatability of the QIPVs in the Qur’an. The other purposes include investigating the difficulties experienced by Qur’an translators when rendering the QIPVs into English and proposing recommendations for Qur’an translators on the basis of the findings reported from the practical component of the study. The principal thesis subsequently developed in this book is that translators encounter a range of difficulties when trying to render QIPVs into English, as evidenced in the difficulties experienced by the selected translators. To achieve the stated purposes of this study, two sets of research questions were proposed. The first set of questions addresses the fundamental issues pertinent to the theoretical component of this study. The second set of questions accounts for the practical component of this study: to analyse and discuss the illustrative data collected from the corpus of selected Qur’an translations. Chapter 1 provided the context for this study by highlighting the statement of the problem, delineating the aim and rational of the study, and outlining its layout. It also elaborated on the focus of the study by asserting its limitations. Chapter 2 imparted a general overview of the concept of (un)translatability as dealt with in the field of translation studies. It also addressed, in a more detailed manner, the issue of the Qur’an’s (un)translatability. The chapter comprised three parts. The first part explored the notion of (un)translatability as dealt with by linguists and translation theorists; the second part reviewed the perspectives of Muslim intellectuals concerning the issue of the Qur’an’s (un)translatability; and the third part scrutinized the opinions of several leading translators of the Qur’an about this particular issue. The main aim of the chapter was—apart from advancing our understanding of the notion of (un)translatability in general and the (un) translatability of the Qur’an in particular—to scrutinize aspects of the (un)translatability of the Qur’an and to identify the conceivable reasons for these aspects as stated by both Muslim intellectuals and Qur’an translators. The chapter demonstrated that, according to both Muslim intellectuals and Qur’an translators, three types of (un)translatability come up when translating the Qur’an: linguistic (un)translatability, cultural (un)translatability, and theological (un)translatability.
196
Conclusion
The chapter also demonstrated that although Muslim intellectuals and translators of the Qur’an agree that the Qur’an-specific linguistic and cultural aspects are untranslatable, there are variations in the way Muslim intellectuals prioritize them and in the way translators deal with them. However, both groups confirm the possibility of rendering the meanings of the Qur’an into other languages and affirm the inevitability of this process. Moreover, Chapter 2 argued that the notion of the Qur’an’s translatability, rather than untranslatability, needs to be the main concern of stakeholders. Chapter 3 presented a detailed and systematic scrutiny of the phenomenon of AIPVs. It attended to their syntactic and semantic peculiarities and accounted for their underlying subtleties. The chapter also established a parameter for nailing down this phenomenon in the Arabic language and highlighted the key factors that control the syntactic and semantic relationships between its two constituents. By considering the views of both classical linguists and modern linguists, the chapter presented genuine instances taken from classical and modern literature (e.g. the Qur’an, Arabic literary works, and press articles) and several illustrative authentic contexts of Arabic usage. Chapter 4 scrutinized the phenomenon of idiomatic PVs in the Qur’an. The overarching aim of the chapter was to initiate a discussion on the idiosyncrasies of Qur’anic PVs. This was undertaken to enhance our understanding of the Qur’anic discourse by providing insights into how these complex idiomatic expressions function in the texture of the Qur’an. Moreover, the aim was to sketch a detailed syntactic, semantic and pragmatic depiction of QIPVs. The chapter tackled this vital phenomenon from three perspectives: syntactic, semantic, and pragmatic. By considering relevant pragmatics theories (speech acts, CP, PP, and IP), Chapter 4 demonstrated that both direct speech acts and indirect speech acts are performed by employing the QIPVs. The chapter also demonstrated that the QIPVs are typically used to flout several of the maxims of conversation, to communicate various conversational implicatures. Chapter 5 delineated the methodology used in this study. It presented the theoretical framework, data collection and data analysis procedures, and the other research procedure applied in this study. The chapter focused on depicting a working model for analysing and assessing translations of the QIPVs. The proposed model was employed as a yardstick against which the (un)translatability of QIPVs was measured. Drawing on speech act theory and employing the contrastive analysis approach, the model attended to the two main categories of QIPVs (i.e. metaphorical and figurative) from both the semantic perspective and the pragmatic perspective. The chapter argued that the adequacy of the equivalents given to the QIPVs should be determined according to the closeness to the intended meaning ascribed by the authoritative exegetical and lexicographical works (tertium comparationis). Semantically speaking, closeness to the denotative meaning should be considered because it reflects the informative level of meaning. It also must be given priority over the other levels of meaning (i.e. expressive and aesthetic). Pragmatically speaking, equivalence adequacy should be measured according to faithfulness to the connotative meaning. The contrastive analysis approach was
Conclusion
197
employed as a theoretical framework. The QIPVs and their provided translations were described, compared, and assessed at both the micro-linguistic level and the macro-linguistic level. The contrastive analysis approach used in this study was not confined, however, to the lexical-semantic level. Rather, it covered the pragmatic aspects involved in translating the QIPVs into English. A corpus of the ten most common and most circulated English-language translations of the Qur’an was scrutinized, compared, and assessed to determine the extent to which each translation achieved functional-pragmatic equivalence for the QIPVs. The reason for selecting these translations rather than others was, apart from their being the most popular ones, to represent the religious, gender, and linguistic background differences among the translators. This was deemed important given that the Qur’an has been translated by Muslim and non-Muslim, Arab and non-Arab, and male and female translators. Chapter 6 constituted the practical component of this study. To adopt a case study analysis approach, ten Qur’anic verses containing QIPVs were taken as illustrative instances. The verses were interpreted in view of the most authoritative Muslim exegetes and Arabic lexicographers and analysed against the adopted theoretical framework (contrastive analysis approach). The equivalents provided to the QIPVs incorporated in these verses by the selected translators were examined, compared, discussed, and assessed by using the proposed model. In addition, the chapter attended to this study’s second set of research questions to account for the practical component of the study. The chapter demonstrated that the selected translators used different procedures and experienced many difficulties when dealing with the QIPVs. Chapter 6 also demonstrated that the selected translators fell into two types of translational pitfalls: overtly erroneous errors (i.e. literal translation, under-translation, over-translation, and semantic deviation) and covertly erroneous errors (i.e. register shift and failure to capture the illocutionary force of the QIPVs). Furthermore, the chapter argued that the adequate functional-pragmatic equivalents provided to the QIPVs, despite being relatively few, revealed that the Qur’anic expressions are in fact relatively translatable if suitable procedures are followed.
7.2 Findings of the study The findings to emerge from the theoretical and practical components of this study are as follows: (1) Arabic verb-preposition structures fall into two distinct types: nonidiomatic and idiomatic. The former is a literal and transparent structure reflecting a plain grammatical colligation between a verb and a preposition with no semantic collocability. The latter is a metaphorical and nontransparent structure. (2) The principal parameter distinguishing between the idiomatic type and the nonidiomatic type of Arabic verb-preposition structures is the degree of idiomaticity. This parameter makes the significances of the AIPVs’ components
198
(3)
(4) (5)
(6)
(7)
(8)
(9)
(10)
(11)
Conclusion submit their basic imports and combine to produce new metaphorical significance. This new form is typically nontransparent and cannot be taken literally. Grasping the overall significance of the idiomatic, nontransparent, and metaphorical AIPVs is subject to four variables: the original significance of the verb, the contextual significance of the verb, the associated meaning of the preposition, and the object governed by the preposition. Only the authentic locative and directional Arabic prepositions can collocate with verbs to function as a second constituent part of the AIPVs. Every Arabic preposition has two types of signification: its own basic signification and several other associated meanings that might overlap with the basic and associated meanings of other Arabic prepositions. Four key factors govern the syntactic and semantic relationships of Arabic verbs and prepositions, delineate their significances, and determine their degree of idiomaticity when combining to form an AIPV: al-taʿalluq, al-taʿaddī, al-inābah, and al-taḍmīn. The AIPVs, as far as their register variations are concerned, are principally formal and rhetorical expressions. They are typically used in high-register literary text types by educated people. Yet they may also be used in lowerregister text types and by laypeople in informal spoken Arabic. All the aforesaid morphological, syntactic, and semantic idiosyncrasies of the AIPVs, the delicate semantico-syntactic interrelation between their constituent parts (i.e. verbs and prepositions), and the vital factors that control their functionality are perfectly pertinent and applicable to the QIPVs. The only exception in this regard is the Qur’an-specific qualities of the latter that are acquired from their unique usage in the Qur’anic discourse. When interweaved into the Qur’anic texture, the QIPVs are employed in different marked word orders and in a variety of semantic fields to serve diverse communicative purposes by preforming various direct and indirect speech acts. The QIPVs, syntactically speaking, fall into two types in terms of the location of their second constituent part (i.e. preposition): split QIPVs and nonsplit QIPVs. Depending on the verbs’ transitivity, tense, and voice and on the number of prepositions with which they combine, each grammatical type of the QIPVs appears in several syntactic patterns throughout the Qur’anic texture. The QIPVs, semantically speaking, are of two categories: metaphorical and figurative. The former category is characterized as nonliteral and nontransparent. In addition, it falls under the definition of metaphor and fulfils its two purposes (referential and pragmatic) and three components (object, image, and sense). The latter QIPV category does not satisfy these criteria, and yet it is characterized as nonliteral and nontransparent. The QIPVs, pragmatically speaking, are used in the Qur’anic context to perform both direct speech acts and indirect speech acts; to violate several maxims of conversation, to bring forth various conversational implicatures; and to achieve different communicative purposes.
Conclusion 199 (12) The selected translators experienced many difficulties when dealing with the QIPVs. Their overall achievements regarding the functional-pragmatic equivalents of the QIPVs supported the thesis of this book. (13) The selected translators fell into two types of translational pitfalls: overtly erroneous errors and covertly erroneous errors. The former includes literal translation, under-translation, over-translation, and semantic deviation. The latter includes register shift and failure to capture the illocutionary force of the QIPVs. (14) The overtly erroneous errors made by the selected translators outnumbered the covertly erroneous errors. Of the overtly erroneous errors made, under-translation errors were the most common, followed by semantic deviations, then literal translation errors, and finally over-translation errors. (15) Of the covertly erroneous errors made by the selected translators, errors of failure to match the illocutionary force were the most common, followed by register shift errors. (16) Literal translation occurred mainly when dealing with the metaphorical QIPVs. It was a result of the selected translators’ respective attempts to reproduce the metaphor in the TL without adding a sense when the TL does not permit such a reproduction. Literal translation occurred for two reasons: the selected translators’ respective endeavours to keep as close as possible to the original and/or their insistence on echoing its aesthetic values. (17) Under-translation was attributed to the selected translators’ tendency to provide general denotative and/or connotative meanings for both the metaphorical QIPVs and the figurative QIPVs. (18) Over-translation was attributed to the selected translators’ tendency to provide more-specific meaning than the original provided. This was performed by unjustifiably adding a sense to the denotative and/or connotative meanings of both the metaphorical QIPVs and the figurative QIPVs. (19) Semantic deviation manifested in the provision of erroneous denotative and/or connotative meanings for both the metaphorical and figurative QIPVs. It was ascribed mainly to the translators’ wrong choice of words and the inappropriate procedures that they used. It was also ascribed to their odd dogmatic principles. (20) Register shift typically manifested in the provision of the informal EIPVs as an equivalent to the highly formal QIPVs. (21) The incorrect delivery of speech acts manifested in the failure to capture the intended illocutionary force. It was attributed to the wrong calculability of the intended implicatures conducted by the selected translators when they rendered the QIPVs. (22) Most selected translators vowed to provide the maximum equivalence at the informative level of meaning by remaining as close as possible to the original’s intended meaning. Nonetheless, they sacrificed the informative level of meaning in a number of instances and stumbled into translational pitfalls in their attempts to echo the QIPVs’ respective aesthetic and expressive levels of meaning.
200
Conclusion
(23) Most selected translators mistook the literal meaning of the QIPVs for its metaphorical meaning when they reproduced the metaphor in the TL without adding a sense. This was attributed to their attempts to echo the QIPVs’ aesthetic and expressive levels of meanings. (24) When dealing with the metaphorical QIPVs, the selected translators used two procedures: (i) Reducing the metaphor to sense; (ii) Reproducing the metaphor in the TL without adding a sense to it. (25) When dealing with the figurative QIPVs, the selected translators used two procedures: (i) Taking a source-text-oriented approach by imitating the original’s marked syntactic structure; (ii) Taking a target-text-oriented approach by abandoning the original’s marked syntactic structure. (26) Several selected translators coupled the aforementioned procedures with one or both of the following procedures: (i) Inserting parenthetical glosses; (ii) Offering explanatory footnotes. (27) Each of the selected translators often used more than one procedure when they dealt with the QIPVs. A translator’s procedures varied according to the context in which the QIPVs are employed. (28) The procedure most frequently used by the selected translators when dealing with the metaphorical QIPVs was to reduce the metaphor to sense. Despite its high frequency, this still constituted the most inadequately used procedure. The inadequacy was mainly attributed to the translators’ ignoring the short micro and the distant macro Qur’anic contexts in which the QIPVs were employed. This was the main reason behind making the aforesaid covert and overt errors. (29) The selected translators’ second-most frequently used procedure when dealing with the metaphorical QIPVs was to reproduce the metaphor in the TL without adding a sense. Notwithstanding, this formed also the second-most inadequately used procedure. Its low adequacy percentage was attributed to the significant sociolinguistic discrepancies between the Arabic language and the English language, which makes reproducing Arabic metaphors in English difficult to achieve. (30) The procedure most frequently used by the selected translators when dealing with the figurative QIPVs was to imitate the original’s marked syntactic structure. The procedure to abandon the original’s marked syntactic structure was the least frequently applied and the most inadequate procedure. (31) Qur’an translators and Muslim intellectuals have confirmed that the meanings of the Qur’an can be rendered into other languages. Moreover, they have affirmed the inevitability of such an assignment for the purpose of
Conclusion 201
(32) (33) (34)
(35)
(36)
da‘wah (missionary invitation) to the Islamic faith. Yet they believe that the Qur’an-specific linguistic and cultural qualities are untranslatable. Muslim intellectuals and Qur’an translators differ in how they prioritize the untranslatable Qur’an-specific aspects. Qur’an translators varied in how they dealt with the untranslatable aspects of the Qur’an. Given that absolute ‘untranslatables’ in any text type are far fewer than the ‘translatables’ and ‘relative translatables’, the translatability of the Qur’an as a whole and the translatability of the QIPVs in particular seem to be more plausible and practical than their untranslatability. Therefore, the notion of the translatability of the Qur’an rather than that of its untranslatability should be the main concern of stakeholders. The adequate functional-pragmatic equivalents provided to the QIPVs, despite their low percentage, revealed that these Qur’anic expressions are relatively translatable if suitable procedures are appropriately employed. The relative translatability of the QIPVs was manifested in the selected translators’ adequate achievement of the informative level of meaning at the expense of the aesthetic and expressive levels of meaning. This translation loss was deemed as acceptable because of the complex idiomatic, metaphorical, and figurative nature of the QIPVs and their Qur’an-specific rhetorical values. These cannot be reproduced in the TL without a certain amount of skewing and distortion, particularly at the aesthetic and expressive levels of meaning. Consequently, the QIPVs are neither untranslatable nor perfectly translatable. They are, in fact, relatively translatable in the sense that a crude approximation of their imports is rather achievable and entirely endurable as far as it attends to their functional-pragmatic idiosyncrasies.
7.3 Recommendations Upon consideration of the aforesaid findings and the various reasons to which they were attributed, several indispensable and applicable recommendations for Qur’an translators are proposed as follows: (1) Given that the pursuit of a literal translation was the main reason for the unjustifiable semantic deviations; the mismatching of the illocutionary and perlocutionary acts; and the departure from informative, aesthetic, and expressive levels of meaning by the translators, such a pursuit ought to be avoided when dealing with the Qur’anic discourse in general and with the QIPVs in particular. Therefore, if Qur’an translators are to do justice to the original, they should refrain from hurriedly and thoughtlessly rendering them literally and consider their constituent parts separately. They typically need to keep in mind that every item of the QIPVs form is one distinct unit of meaning with an idiomatic, nontransparent inclination. (2) If translators are to achieve functional-pragmatic equivalence for the QIPVs, they ought to distance themselves from the direct prepositional import of
202
(3)
(4)
(5)
(6)
Conclusion the QIPVs and target instead their illocutionary force. This can be accomplished only by considering the close micro and the distant macro Qur’anic contexts in which the QIPVs are used. If translators do so, they can then attend to the speech acts performed in given Qur’anic contexts where the QIPVs are employed. Moreover, they can avoid catastrophic consequences caused by miscalculating the conversational implicatures. The informative level of meaning must be given priority when translating a sensitive text such as the Qur’an. Given that the stylistic and poetic qualities of the Qur’an are Qur’an-specific properties and that reproducing them surpasses human faculties, no effort should be made by translators to echo these qualities. To achieve the English functional-pragmatic equivalence of the QIPVs, the Qur’an translators should appreciate the formal register of the demanding Qur’anic expressions. They should avoid any shifts of register as a result of providing EIPVs as equivalents to the QIPVs. Given that the EIPVs are informal expressions commonly used in everyday spoken English, providing them as equivalents to the highly formal QIPVs constitutes unacceptable register variation. Such a variation forms one of the main reasons behind an incorrect message being conveyed and the distortion of the original’s intended meaning. Hence, the register variations ought to be typically kept in check by Qur’an translators. To accurately appreciate the meanings of the QIPVs, achieve functional-pragmatic equivalence, and avoid making overtly erroneous and covertly erroneous errors, Qur’an translators must pay due attention to the short micro and distant macro Qur’anic contexts in which the QIPVs are employed. When rendering the metaphorical QIPVs into English, Qur’an translators should use the following procedures to achieve accuracy, adequacy, and faithfulness to the original: (i) Reproduce the metaphor in the TL without adding a sense. This procedure is the most adequate and recommended when the TL permits the transferring of the metaphor. (ii) If the TL does not permit the transferring of the metaphor, reducing the metaphor to sense should be in play. This was the procedure that the selected translators most commonly employed. To achieve functional-pragmatic equivalence of the QIPVs when using this procedure, the translator must be competent at appreciating the aforesaid idiosyncrasies of the QIPVs and at avoiding the aforementioned overtly erroneous and covertly erroneous errors.
(7) When rendering the figurative QIPVs into English, translators should use a source-text-oriented procedure, by adhering to the original’s marked syntactic structures. Grasping the meanings of the figurative QIPVs depends on elucidating their conventional linguistic idiomaticity (unlike their metaphorical counterparts). In turn, that the form and content of the QIPVs are organically interrelated and that the semantico-syntactic marked structures
Conclusion 203 embody the QIPVs’ idiomaticity (which are not without their communicative purposes) mean that in translating the figurative QIPVs, the original’s marked syntactic structures must be adhered to. (8) Using inappropriate translation procedures was the main reason why the Qur’an translators fell into many of the translational pitfalls. Consequently, they should give maximum equivalence at all levels of meaning. When a tension emerges among them, the priority should be the informative level of meaning. This idea holds true not only in tackling the QIPVs but also in dealing with other types of Qur’anic idiomatic expressions. (9) The best way to fully attend to the aforesaid enquiry is to employ the ‘couplets’ technique. Given the complex nature of the QIPVs, Qur’an translators should combine the aforementioned procedures with one or more other practical procedures, such as inserting parenthetical glosses and offering explanatory footnotes. The findings of this study have demonstrated that employing the ‘couplets’ technique was the best practical way to compensate for the loss of meaning at the informative level and at the aesthetic and expressive levels also. (10) Qur’an translators should enhance their communicative competence and boost their syntactic, semantic, and pragmatic knowledge, to better manage the complexities of the QIPVs. Their duties are to acquire advanced knowledge about the QIPVs’ idiosyncrasies and to consult authoritative exegetical and lexicographical works, to produce crude approximations of the QIPVs, to make use of the appropriate procedures, and to identify suitable ways to compensate for translation losses as by-products.
7.4 Suggestions for further research This study focused predominantly on translating Qur’anic QIPVs into English. Despite its inclusive treatment of the idiosyncrasies of QIPVs and the difficulties that they pose to translators of the Qur’an, it was restricted to scrutinizing, comparing, and assessing a number of Qur’anic verses containing QIPVs as illustrative instances taken from a corpus of ten English-language translations of the Qur’an. The selected translations are only the most common and most circulated English-language translations of the Qur’an and represent different religious, gender, and linguistic backgrounds. Replicating this study design to scrutinize other translations of the Qur’an with more illustrative instances would therefore help to shed more light on the question of the (un)translatability of the QIPVs. Furthermore, such a study would provide additional research evidence regarding the difficulties experienced by Qur’an translators and the procedures that they adopted when rendering the QIPVs into English. This study was limited to one language pair—that is, translating the QIPVs from Arabic into English. Replicating its design to investigate the translation of QIPVs into languages other than English would be even more thought-provoking and fruitful. The eclectic model for analysing and assessing translations of the
204
Conclusion
QIPVs proposed in this study can be used to conduct parallel studies to analyse and assess the translation of these problematic expressions into other languages. Finally, we urgently need an inclusive and theory-based bilingual Arabic–English contextual dictionary of QIPVs. Compiling such a dictionary is feasible in light of the findings reported and the recommendations proposed in this book. Indeed, such is the planned forthcoming project of this author.
Appendix I
Concordance for the Qur’anic idiomatic phrasal verbs This concordance is an exclusive representation of the Qur’anic idiomatic phrasal verbs (QIPVs) in the Qur’an. It lists every QIPV found in the Qur’an according to the Arabic alphabetical order of the roots (three or four radical letters) of their main verbs. The entries of this concordance are presented in the following manner: • • • •
•
• •
• • •
Roots appear in bold, centred at the top, and arranged by the Arabic alphabetical order. The transliterated QIPVs (verbs + prepositions) derived from that root are listed to the left in bold. The verbs of QIPVs are listed as they appear in the Qur’an (i.e. with their prefixes, infixes, and suffixes). A grammatical description of each verb (i.e. its morphological form and whether it is perfect/past tense, imperfect/present tense, future tense, imperative, perfect passive tense, imperfect passive voice, or active voice) is specified. Given that the morphological form of an Arabic verb implies the verb agent and sometimes the direct object, the agent’s and the direct object’s gender (masculine or feminine) and number (singular, dual, or plural) of the listed Arabic verbs are indicated in the provided grammatical description. An English translation of each QIPV is provided. Numbers of the Qur’anic chapters (Sūwar) and verses (Āyāt) are presented to the left respectively. The numbering used is consistent with the reading of Ḥafṣ ʿan ʿĀṣim. This reading of the Qur’an is the most widespread reading in the Islamic world. This is followed by a translation of the Qur’anic verses in which a QIPV is included. The English translation used in this concordance is that of Muhammad Asad’s The Message of the Quran (2011). When a verse contains more than one QIPV in different grammatical forms, the verse occurs twice, each time under a particular grammatical form.
206 •
Appendix I Nonidiomatic Qur’anic phrasal verbs are not included in this concordance as they fall out of the focus of this study. For the differences between the nonidiomatic Qur’anic phrasal verbs and the QIPVs, see Chapters 3 and 4 of this book.
A-T-Y ATÁ + MIN—(I)—(perfect)—(agent: masculine, singular)—to visit with destruction, to come upon. 16:26 God visited with destruction all that they had ever built. 59:2
God came upon them.
ATAT + ʿALÁ—(I)—(perfect)—(agent: feminine, singular)—to come upon. 51:42 which spared nothing of what it came upon.
ATAW + ʿALÁ—(I)—(perfect)—(agent: masculine, plural)—to come upon, to come across. 7:138 they came upon people. 25:40 must surely have come across that town. 27:18 when they came upon a valley [full] of ants.
ATAYTA + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—to place. 2:145 even if thou wert to place all evidence.
ATAYNA + BĀʾ—(I)—(perfect)—(agent: feminine, plural)—(direct object: feminine, plural)—to become guilty. 4:25
and thereafter become guilty of immoral conduct.
ATAYNĀ + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—to bring something or someone forth. 21:47 We shall bring it forth.
ATAYNĀKA + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to bring something or someone. 15:64 and we are bringing thee the certainty.
Appendix I 207 ATAYNĀHUM + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to conveyed unto. 23:71 We have conveyed unto them. 23:90 We have conveyed unto them.
ĀTĪKA + BĀʾ—(IV)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, singular)—to bring something or someone. 27:39 I shall bring it to thee. 27:40 I shall bring it to thee.
ĀTĪKUM + BĀʾ—(IV)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to bring something or someone. 20:10 I can bring you a brand therefrom. 27:7
I may bring you from there some tiding.
27:7
bring you [at least] a burning brand.
28:29 I may bring you from there some tiding.
LA-ĀTIYANNAHUM + MIN—(IV)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to fall upon. 7:17
and shall most certainly fall upon them.
TAʾTINĀ + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to produce. 7:132 Whatever sign thou mayest produce before us.
TAʾTIHIM + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to produce. 7:203 dost not produce any miracle for them.
TAʾTŪ + MIN—(I)—(imperfect)—(agent: masculine, plural)—to enter. 2:189 piety does not consist in your entering houses from the rear.
TAʾTŪNA + FĪ—(I)—(imperfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to commit.
208
Appendix I
29:29 And must you commit these shameful deeds in your [open] assemblies.
TAʾTŪNANĀ + ʿAN—(I)—(imperfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to approach. 37:28 you were wont to approach us [deceptively] from the right.
TAʾTŪNĪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to bring. 12:60 But if you do not bring him unto me.
TAʾTĪA + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—to bring, to bring forth. 17:92 or [till] thou bring God. 44:10 the skies shall bring forth a pall of smoke.
TAʾTĪKUM + BĀʾ—(I)—(imperfect)—(agent: feminine, singular)—(direct object: masculine, plural)—to come with. 40:50 your apostles came unto you with all evidence of the truth.
TAʾTĪNĀ + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to bring. 15:7
Why dost thou not bring before us angels?
FA-TAʾTĪYAHUM + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to bring. 6:35
to bring them a [yet more convincing] message.
TAʾTĪHIM + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to come with. 40:22 their apostles had come to them with all evidence of the truth. 64:6
came unto them their apostles with all evidence of the truth.
NAʾTI + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to replace. 2:106 We replace with a better or a similar one.
Appendix I 209 NAʾTĪYAKUM + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to bring. 14:11 to bring you a proof.
FA-LA-NAʾTIYANNAKA + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to produce. 20:58 we shall most certainly produce before thee the like thereof!
FA-LA-NAʾTIYANNAHUM + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to come with. 27:37 We shall most certainly come upon them with forces.
YAʾTI + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—to gather, to be faced with, to bring, to bring forth, to accomplish, to become, to produce. 2:148 God will gather you all unto Himself. 3:161 shall be faced with his deceit on the Day of Resurrection. 4:133 and bring forth other beings. 14:19 and bring forth a new mankind. 16:76 he accomplishes no good. 31:16 God will bring it to light. 33:30 to become guilty of manifestly immoral conduct. 35:16 and bring forth a new mankind. 52:38 produce a manifest proof.
YAʾTI + MIN—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, singular)—to attain. 41:42 no falsehood can ever attain to it openly.
FA-L-YAʾTIKUM + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to bring. 18:19 bring you thereof [some] provisions.
FA-L-YAʾTINĀ + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to come with.
210 21:5
Appendix I come unto us with a miracle.
YAʾTŪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to offer, to produce. 5:108 people will offer testimony in accordance with the truth. 17:88 producing the like of this Qur’an. 24:4
unable to produce four witnesses.
24:13 if they do not produce such witnesses. 52:34 let them produce another discourse like it. 68:41 let them produce those supporters of theirs.
YAʾTŪKA + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to bring, to assemble. 7:112 who shall bring before thee. 26:37 who shall assemble before thee.
YAʾTŪNA + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to produce, to adduce. 17:88 they could not produce its like. 18:15 to adduce any reasonable evidence.
YAʾTŪNAKA + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to taunt. 25:33 never taunt thee with any deceptive half-truth.
YAʾTIYA + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—to make manifest, to cause, to bring about, to bring forth, to produce. 2:109 God shall make manifest His will. 2:258 God causes the sun to rise in the east. 5:52
God may well bring about good fortune.
5:54
God will in time bring forth [in your stead] people.
9:24
God makes manifest His will.
Appendix I 211 13:38 to produce a miracle save at God’s behest. 40:78 to bring forth a miracle.
YAʾTIKUM + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to bring, to bring back, to provide. 6:46
could bring it all back to you.
11:33 Only God can bring it upon you. 28:71 could bring you light. 28:72 could bring you [the darkness of] night. 67:30 could provide you with water.
YAʾTĪNA + BĀʾ—(I)—(imperfect)—(agent: feminine, plural)—to become, to indulge. 4:19
they have become guilty.
60:12 and would not indulge in slander. 65:1
unless they become openly guilty.
YAʾTIYANĀ + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to come with, to produce. 3:183
unless he comes unto us with burnt offerings.
20:133 produce for us a miracle from his Sustainer.
LA-YAʾTIYANNĪ + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, singular)—to bring. 27:21 unless he brings me a convincing excuse.
YAʾTIYANĪ + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, singular)—to bring. 12:83 God may well bring them all [back] unto me.
YAʾTĪNĪ + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, singular)—to bring. 27:38 Which of you can bring me her throne?
FA-ʾTI + BĀʾ—(I)—(imperative)—(agent: masculine, singular)—to cause, to produce, to come with.
212
Appendix I
2:258
cause it, then, to rise in the west!
7:106
produce it—if thou art a man of truth!
26:31
Produce it, then, if thou art a man of truth!
26:154
Come, then, forward with a token!
IʾTI + BĀʾ—(I)—(imperative)—(agent: masculine, singular)—to bring. 10:15 Bring us a discourse other than this!
IʾTINĀ + BĀʾ—(I)—(imperative)—(agent: masculine, singular)—(direct object: masculine, plural)—to bring, to bring about, to inflict, to bring upon. 7:70
Bring about, then, that [punishment] with which thou hast threatened us!
7:77
Bring about, then, that [punishment] with which thou hast threatened us!
8:32
inflict [some other] grievous suffering on us!
11:32 bring upon us, therefore, that with which thou dost threaten us! 29:29 Bring down upon us God’s chastisement! 46:22 Bring, then, upon us that [doom] with which thou threatenest us!
IʾTŪ + BĀʾ—(I)—(imperative)—(agent: masculine, plural)—to produce, to enter, to come forward with, to bring forth. 2:23
produce a surah of similar merit!
2:189
enter houses through their doors.
3:93
Come forward, then, with the Torah!
10:38
Produce, then, a surah of similar merit!
11:13
Produce, then, ten surahs of similar merit!
21:61
bring him before the people’s eyes!
28:49
Produce, then, [another] revelation from God!
37:157
Produce, then, that divine writ of yours!
44:36
bring forth our forefathers!
45:25
Bring forth our forefathers!
FA-ʾTŪNĀ + BĀʾ—(I)—(imperative)—(agent: masculine, plural)—(direct object: masculine, plural)—to bring. 14:10 bring us a clear proof!
IʾTŪNĪ + BĀʾ—(I)—(imperative)—(agent: masculine, plural)—(direct object: masculine, singular)—to bring, to come with.
Appendix I 213 10:79 Bring before me every sorcerer of great knowledge! 12:50 Bring him before me! 12:54 Bring him unto me! 12:59 bring unto me that brother of yours from your father’s side. 12:93 come [back] to me with all your family. 46:4
bring me any divine writ preceding this one!
FA-ʾTŪHUNNA + MIN—(I)—(imperative)—(agent: masculine, plural)— (direct object: feminine, plural)—to go in. 2:222 go in unto them as God has bidden you to do.
ʾUTŪ + BĀʾ—(I)—(perfect passive)—(agent of the passive: masculine, plural)— to be given. 2:25
they shall be given something.
A-KH-DH AKHADHNĀ + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—to try, to overwhelm, to take to task. 7:94
without trying its people with misfortune and hardship.
7:130 We overwhelm Pharaoh’s people with drought and scarcity of fruits. 7:165 overwhelmed those who had been bent on evildoing with dreadful suffering. 23:64 We shall have taken to task. 29:40 We take to task for his sin.
AKHADHNĀ + MIN—(I)—(perfect)—(agent: masculine, plural)—to accept, to seize. 33:7
We did accept a solemn pledge from all the prophets.
33:7
We accepted a most weighty, solemn pledge from [all of] them.
69:45 We would indeed have seized him by his right hand.
AKHADHNĀHUM + BĀʾ—(I)—(perfect)—(agent: masculine, plural)— (direct object: masculine, plural)—to visit, to take to task, to test. 6:42
visited them with misfortune and hardship.
7:96
We took them to task through what they [themselves] had been doing.
214
Appendix I
23:76 We tested them through suffering. 43:48 We took them to task through suffering.
AKHADHAHUM + BĀʾ—(I)—(perfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to take to task. 3:11
God took them to task for their sins.
8:52
God took them to task for their sins.
40:21 God took them to task for their sins.
TAʾKHUDHKUM + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to keep carrying out something. 24:2
and let not compassion with them keep you from [carrying out] this law of God.
YAʾKHUDHAHUM + FĪ—(I)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to take to task. 16:46 He will not take them to task [suddenly].
YAʾKHUDHAHUM + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to take to task. 16:47 or take them to task through slow decay.
ʾUKHIDHŪ + MIN—(I)—(perfect passive)—(agent of the passive: masculine, plural)—to be seized. 34:51 they will have been seized from so close nearby.
YUʾKHADH + ʿALÁ—(I)—(imperfect passive)—(agent of the passive: masculine, singular)—to be pledged. 7:169 Have they not been solemnly pledged through the divine writ?
A-DH-N FA-ʾDHANŪ + BĀʾ—(I)—(imperative)—(agent: masculine, plural)—to know. 2:279 then know that you are at war with God and His Apostle.
Appendix I 215
A-K-L TAʾKULŪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to devour. 2:188 Devour not one another’s possessions wrongfully. 2:188 with a view to devouring sinfully. 4:29
Do not devour one another’s possessions wrongfully!
TAʾKULŪ + ILÁ—(I)—(imperfect)—(agent: masculine, plural)—to consume. 4:2
and do not consume their possessions together with your own.
A-N-S ĀNASTUM + MIN—(IV)—(perfect)—(agent: masculine, plural)—to find. 4:6
if you find them to be mature of mind.
A-W-Y ĀWAYNĀHUMĀ + ILÁ—(IV)—(perfect)—(agent: masculine, plural)—(direct object: masculine, dual)—to provide. 23:50 and provided for both an abode.
B-D-Y LA-TUBDĪ + BĀʾ—(I)—(imperfect)—(agent: feminine, singular)—to disclose. 28:10 she would indeed have disclosed all about him.
YUBDIHĀ + LĀM—(I)—(imperfect)—(agent: masculine, singular)—(direct object: feminine, singular)—to reveal. 12:77 without revealing his thought to them.
YUBDŪNA + LĀM—(I)—(imperfect)—(agent: masculine, plural)—to reveal. 3:154 which they would not reveal unto thee.
LI-YUBDIYA + LĀM—(I)—(imperfect)—(agent: masculine, singular)—to make someone conscious of something.
216
Appendix I
7:20
with a view to making them conscious of their nakedness.
B-S-L UBSILŪ+ BĀʾ—(I)—(perfect passive)—(agent of the passive: masculine, plural)—to be held in pledge. 6:70
every human being shall be held in pledge for whatever wrong he has done.
TUBSALA+ BĀʾ—(I)—(imperfect passive)—(agent of the passive: feminine, singular)—to be held in pledge. 6:70
shall be held in pledge for the wrong they have done.
B-SH-R BASHSHIR + BĀʾ—(II)—(imperative)—(agent: masculine, singular)—to announce, to give a tiding. 4:138 Announce thou to such hypocrites that grievous suffering awaits them. 9:3
give thou [O Prophet] the tiding of grievous chastisement.
BASHSHIRHU + BĀʾ—(II)—(imperative)—(agent: masculine, singular)— (direct object: masculine, singular)—to announce, to give a tiding. 31:7
give him, then, the tiding of grievous suffering.
45:8
announce unto him grievous suffering.
BASHSHIRHUM + BĀʾ—(II)—(imperative)—(agent: masculine, singular)— (direct object: masculine, plural)—to announce, to give a tiding. 3:21
announce unto them a grievous chastisement.
9:34
give them the tiding of grievous suffering.
84:24 give them the tiding of grievous suffering.
ABASHSHARTUMŪNĪ + ʿALÁ—(II)—(perfect)—(agent: masculine, plural)— (direct object: masculine, singular)—to give a glad tiding. 15:54 Do you give me this glad tiding?
TUBASHSHIRŪNA + BĀʾ—(II)—(imperfect)—(agent: masculine, plural)— to give a tiding.
Appendix I 217 15:54 are you giving me a tiding?
B-Ṣ-R BAṢARTU + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—to gain insight. 20:96 I have gained insight into something.
YUBṢIRŪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to see. 20:96 which they were unable to see.
B-W-ʾ BĀʾA + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—to earn. 3:162 who has earned the burden of God’s condemnation. 8:16
shall indeed have earned the burden of God’s condemnation.
BĀʾŪ + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—to earn. 2:61
they earned the burden of God’s condemnation.
2:90
have they earned the burden of God’s condemnation.
3:112 they have earned the burden of God’s condemnation.
TABŪʾA + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—to bear. 5:29
for thee to bear [the burden of] all the sins.
J-R-Ḥ JARAḤTUM + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—to work. 6:60
and knows what you work in daytime
J-Y-ʾ JĀʾA + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—to bring, to bring forth, to come, to come with, to produce, to be rightly guided. 6:91
which Moses brought unto men?
218
Appendix I
6:160
Whoever shall come [before God] with a good deed.
6:160
whoever shall come with an evil deed.
11:69
there came unto Abraham Our [heavenly] messengers.
12:72
he who produces it shall receive a camel-load [of grain as reward]!
12:100
He brought you [all unto me] from the desert.
27:89
Whoever shall come [before Him] with a good deed.
27:90
they who shall come with evil deeds.
28:37
who comes with guidance from Him.
28:84
Whosoever shall come [before God] with a good deed.
28:84
who shall come with an evil deed.
28:85
who is right-guided.
37:37
he [whom you call a mad poet] has brought the truth.
39:33
he who brings the truth.
43:63
Jesus came [to his people] with all evidence of the truth.
51:26
and brought forth a fat [roasted] calf.
JĀʾATHUM + BĀʾ—(I)—(perfect)—(agent: feminine, plural)—(direct object: masculine, plural)—to bring, to come with. 10:13 brought them all evidence of the truth. 14:9
There came unto them their apostles with all evidence of the truth.
30:9
to them [too] came their apostles with all evidence of the truth.
40:83 their apostles came to them with all evidence of the truth.
JĀʾAKUM + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to come with. 2:92
there came unto you Moses with all evidence of the truth.
3:183 there came unto you apostles with all evidence of the truth. 4:170 The Apostle has now come unto you with the truth from your Sustainer. 40:28 he has brought you all evidence of this truth from your Sustainer. 40:34 Joseph came aforetime with all evidence of the truth. 49:6
If any iniquitous person comes to you with a [slanderous] tale.
JĀʾAHUM + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to come with,
Appendix I 219 28:36 Moses came unto them with Our clear messages. 29:39 to them had come Moses with all evidence of the truth. 40:25 when he came to them. 43:47 he came before them with Our [miraculous] signs. 61:6
he came unto them with all evidence of the truth.
JĀʾŪ + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—to come with, to cause, to produce. 3:184 they came with all evidence of the truth. 24:11 those who would falsely accuse others of unchastity. 24:13 produce four witnesses to prove their allegation.
JĀʾŪHUM + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to bring. 10:74 they brought them all evidence of the truth. 30:47 they brought them all evidence of the truth.
JIʾTA + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—to bring, to come with. 2:71
At last thou hast brought out the truth!
7:106 If thou hast come with a sign, produce it!
JIʾTUKA + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—(direct object: masculine, singular)—to bring about, to come with. 26:30 Even if I should bring about before thee something that clearly shows the truth? 27:22 I have come to thee from Sheba with a tiding sure!
JIʾTUKUM + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to bring, to come with. 3:49
I have come unto you with a message from your Sustainer.
3:50
I have come unto you with a message from your Sustainer.
7:105 I have now come unto you with a clear evidence from your Sustainer. 43:24 I bring you a guidance better than that.
220
Appendix I
43:63 I have now come unto you with wisdom.
JIʾTUM + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—to contrive. 10:81 What you have contrived is [but] sorcery.
JIʾTANĀ + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to bring, to come with. 11:53 Thou hast brought us no clear evidence. 21:55 Hast thou come unto us [with this claim] in all earnest?
JIʾTAHUM + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to come with, to approach with. 5:110 when thou camest unto them with all evidence of the truth. 30:58 if thou approach them with any [such] message.
JIʾNĀ + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—to bring, to bring forth, to add. 4:41
and bring thee [O Prophet] as witness against them.
12:88
so we have brought but scanty merchandise.
16:89
We brought forth to bear witness regarding those.
17:104 We will bring you forth as [parts of] a motley crowd. 18:109 if we were to add to it sea upon sea.
JIʾNĀKA + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to come with, to convey. 15:63 but we have come unto thee with [the announcement of] something. 20:47 We have now come unto thee with a message from thy Sustainer. 25:33 without Our conveying to thee the [full] truth.
JIʾNĀHUM + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to convey. 7:52
We did convey unto them a divine writ.
JIʾNĀKUM + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to convey.
Appendix I 221 43:78 We have conveyed the truth unto you.
JĪʾA + BĀʾ—(I)—(perfect passive)—(agent of the passive: masculine, plural)—to be brought forward. 39:69 all the prophets will be brought forward. 89:23 hell will be brought [within sight].
Ḥ-B-Ṭ ḤABIṬA + ʿAN—(I)—(perfect)—(agent: masculine, singular)—to become in vain. 6:88
in vain, indeed, would have been all [the good] that they ever did.
Ḥ-D-TH ʾUḤDITHA + LĀM—(I)—(imperfect)—(agent: masculine, singular)—to give. 18:70 until I myself give thee an account thereof.
YUḤDITHU + LĀM—(I)—(imperfect)—(agent: masculine, singular)—to give rise to something. 20:113 or that it gives rise to a new awareness in them.
Ḥ-L-L YAḤILLU + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)—to alight upon. 11:39 upon whom long-lasting suffering shall alight. 39:40 upon whom long-lasting suffering shall alight.
YAḤILLA + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)—to fall upon. 20:81 My condemnation falls upon you. 20:86 your Sustainer’s condemnation falls upon you.
YAḤLIL + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)—to fall upon. 20:81 he upon whom My condemnation falls has indeed thrown himself into utter ruin!
222
Appendix I
Ḥ-M-L TAḤMIL + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)—to approach threateningly. 7:176 if thou approach him threateningly, he will pant with his tongue lolling.
Ḥ-W-Ṭ AḤĀṬA + BĀʾ—(IV)—(perfect)—(agent: masculine, singular)—to encompass. 17:60 thy Sustainer encompasses all mankind. 18:29 a fire whose billowing folds will encompass them from all sides. 48:21 which God has already encompassed [for you]. 65:12 God encompasses all things with His knowledge. 72:28 He who encompasses [with His knowledge] all that they have [to say].
AḤĀṬAT + BĀʾ—(IV)—(perfect)—(agent: feminine, singular)—to engulf. 2:81
Those who earn evil and by their sinfulness are engulfed.
AḤAṬTU + BĀʾ—(IV)—(perfect)—(agent: masculine, singular)—to encompass. 27:22 I have encompassed [with my knowledge] something that thou hast never yet encompassed [with thine].
AḤAṬNĀ + BĀʾ—(IV)—(perfect)—(agent: masculine, plural)—to encompass. 18:91 We did encompass with Our knowledge all that he had in mind.
TUḤIṬ + BĀʾ—(IV)—(imperfect)—(agent: masculine, singular)—to encompass, to comprehend. 18:68 how couldst thou be patient about something that thou canst not comprehend within the compass of [thy] experience? 27:22 I have encompassed [with my knowledge] something that thou hast never yet encompassed.
TUḤĪṬŪ + BĀʾ—(IV)—(imperfect)—(agent: masculine, plural)—to encompass. 27:84 Did you give the lie to My messages even though you failed to encompass them with [your] knowledge?
Appendix I 223 YUḤĪṬŪ + BĀʾ—(IV)—(imperfect)—(agent: masculine, plural)—to comprehend. 10:39 Nay, but they are bent on giving the lie to everything the wisdom whereof they do not comprehend.
YUḤĪṬŪNA + BĀʾ—(IV)—(imperfect)—(agent: masculine, plural)—to encompass, to attain. 2:255
whereas they cannot attain to aught of His knowledge.
20:110
whereas they cannot encompass Him with their knowledge.
ʾUḤĪṬA + BĀʾ—(IV)—(perfect passive)—(agent of the passive: masculine, plural)—to be encompassed. 10:22 so that they believe themselves to be encompassed [by death]. 18:42 his fruitful gardens were encompassed [by ruin].
YUḤĀṬA + BĀʾ—(IV)—(imperfect)—(agent of the passive: masculine, plural)—to encompass. 12:66 unless you yourselves be encompassed [by death]!
Ḥ-Y-D TAḤĪD + MIN—(I)—(imperfect)—(agent: masculine, singular)—to look away. 50:19 from which thou wouldst always look away.
Ḥ-Y-Q ḤĀQA + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—to be overwhelmed, to be encompassed. 6:10
overwhelmed by the very thing which they were wont to deride.
11:8
they shall be overwhelmed by the very thing which they were wont to deride.
21:41 overwhelmed by the very thing which they had been wont to deride. 39:48 shall they be overwhelmed by the very truth which they were wont to deride. 40:45 whereas suffering vile was to encompass Pharaoh’s folk. 40:83 they were overwhelmed by the very thing which they were wont to deride.
224
Appendix I
45:33 they will be overwhelmed by the very thing which they were wont to deride. 46:26 they were overwhelmed by the very thing which they had been wont to deride.
YAḤIQU + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—to engulf. 35:43 Yet [in the end,] such evil scheming will engulf none but its authors.
KH-T-M KHATAMA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to seal. 2:7
God has sealed their hearts.
6:46
and seal your hearts.
45:23 and whose hearing and heart He has sealed.
NAKHTIMU + ʿALÁ—(I)—(imperfect)—(agent: masculine, plural)—to set a seal. 36:65 We shall set a seal on their mouths.
YAKHTIMU + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)—to seal. 42:24 He could have sealed thy heart [forever].
KH-R-Q KHARAQŪ + LĀM—(I)—(perfect)—(agent: masculine, plural)—to invent. 6:100 they have invented for Him sons and daughters!
KH-L-Ṣ AKHLAṢNĀHUMU + BĀʾ—(IV)—(perfect)—(agent: masculine, plural)— (direct object: masculine, plural)—to purify. 38:46 We purified them by means of a thought most pure.
KH-L-W KHALĀ + ILÁ—(I)—(perfect)—(agent: masculine, singular)—to be alone with someone.
Appendix I 225 2:76
but when they find themselves alone with one another.
KHALĀ + FĪ—(I)—(perfect)—(agent: masculine, singular)—to bass away. 35:24 a warner has [lived and] passed away in its midst.
KHALAW + ILÁ—(I)—(perfect)—(agent: masculine, plural)—to be alone with someone. 2:14
but when they find themselves alone with their evil impulses.
YAKHLU + LĀM—(I)—(imperfect)—(agent: masculine, singular)—to be for someone alone. 12:9
so that your father’s regard may be for you alone.
KH-W-Ḍ YAKHŪḌŪ + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to begin to talk. 4:140 until they begin to talk of other things. 6:68
until they begin to talk of other things.
YAKHŪḌŪNA + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to talk. 6:68
whenever thou meet such as indulge in [blasphemous] talk about Our messages.
D-KH-L DAKHALTUM + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to consummate marriage. 4:23
with whom you have consummated your marriage.
4:23
but if you have not consummated your marriage.
D-L-W TUDLŪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to employ legal artifices. 2:188 and neither employ legal artifices.
226
Appendix I
DH-H-B DHAHABA + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—to take away, to stand apart. 2:17
God takes away their light.
2:20
He could indeed take away their hearing and their sight.
23:91 Each deity would surely have stood apart [from the others] in whatever it had created.
DHAHABA + ʿAN—(I)—(perfect)—(agent: feminine, plural)—to go, to leave. 11:10 Gone is all affliction from me. 11:74 And when the fear had left Abraham.
TADHHAB + ʿALÁ—(I)—(imperfect)—(agent: feminine, singular)—to waste. 35:8
do not waste thyself in sorrowing over them!
TADHHABŪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to take away. 4:19
with a view to taking away anything of what you may have given them.
NADHHABANNA + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to take away. 17:86 We could indeed take away whatever We have revealed unto thee. 43:41 But whether We do [or do not] take thee away.
YADHHABU + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—to deprive. 24:43 the flash of His lightning well-nigh deprives [men of their] sight!
YADHHABĀ + BĀʾ—(I)—(imperfect)—(agent: masculine, dual)—to do away with. 20:63 and on doing away with your time-honoured way of life.
ADHHABA + ʿAN—(I)—(imperfect)—(agent: masculine, singular)—to cause something to leave.
Appendix I 227 35:34 who has caused all sorrow to leave us.
YUDHHIBA + ʿAN—(I)—(imperfect)—(agent: masculine, singular)—to free, to remove. 8:11
and free you from Satan’s unclean whisperings.
33:33 for God only wants to remove from you all that might be loathsome.
R-B-Ṭ RABAṬNĀ + ʿALÁ—(I)—(perfect)—(agent: masculine, plural)—to endow strength. 18:14 and endowed their hearts with strength. 28:10 had We not endowed her heart with enough strength.
YARBIṬA + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)—to strengthen. 8:11
and strengthen your hearts.
R-GH-B YARGHABU + ʿAN—(I)—(imperfect)—(agent: masculine, singular)—to abandon. 2:130 who, unless he be weak of mind, would want to abandon Abraham’s creed?
YARGHABŪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to care. 9:120 or to care for their own selves more than for him.
IRGHAB + ILÁ—(I)—(imperative)—(agent: masculine, singular)—to turn with love. 94:8
and unto thy Sustainer turn with love.
R-K-Ḍ IRKUḌ + BĀʾ—(I)—(imperative)—(agent: masculine, singular)—to strike. 38:42 Strike [the ground] with thy foot!
228
Appendix I
R-W-D RĀWADTTUNNA + ʿAN—(III)—(perfect)—(agent: feminine, plural)—to seek. 12:51 when you sought to make Joseph yield himself unto you.
RĀWADATNĪ + ʿAN—(III)—(perfect)—(agent: feminine, singular)—(direct object: masculine, singular)—to seek. 12:26 It was she who sought to make me yield myself unto her!
RĀWADATHU + ʿAN—(III)—(perfect)—(agent: feminine, singular)—(direct object: masculine, singular)—to seek. 12:23 sought to make him yield himself unto her.
RĀWADTTUHU + ʿAN—(III)—(perfect)—(agent: feminine, singular)— (direct object: masculine, singular)—to try, to seek. 12:32 I did try to make him yield himself unto me. 12:51 It was I who sought to make him yield himself unto me.
RĀWADŪHU + ʿAN—(III)—(perfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to demand. 54:37 and even demanded that he give up his guests [to them].
TURĀWIDU + ʿAN—(III)—(imperfect)—(agent: feminine, singular)—to try to induce. 12:30 trying to induce her slave-boy to yield himself unto her!
SA-NURĀWIDU + ʿAN—(III)—(future)—(agent: masculine, plural)—to try to persuade. 12:61 We shall try to persuade his father to part with him.
R-W-GH RĀGHA + ILÁ—(I)—(perfect)—(agent: masculine, singular)—to approach stealthily, to turn quietly.
Appendix I 229 37:91 he approached their gods stealthily. 51:26 he turned quietly to his household.
RĀGHA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to fall upon. 37:93 he fell upon them, smiting them with his right hand.
R-Y-N RĀNA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to become corroded. 83:14 but their hearts are corroded by all [the evil] that they were wont to do!
S-B-GH ASBAGHA + ʿALÁ—(IV)—(imperfect)—(agent: masculine, singular)—to lavish. 31:20 and has lavished upon you His blessings.
S-Ṭ-W YASṬŪNA + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to assault. 22:72 they would almost assault those who convey Our messages unto them!
S-ʿ-Y SAʿÁ + FĪ—(I)—(perfect)—(agent: masculine, singular)—to strive for, to go about. 2:114 and strive for their ruin. 2:205 he goes about the earth spreading corruption. 17:19 and strive for it as it ought to be striven for.
SAʿAW + FĪ—(I)—(perfect)—(agent: masculine, plural)—to strive against. 22:51 those who strive against Our messages. 34:5
for those who strive against Our messages.
230
Appendix I
YASʿAWNA + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to endeavour, to labour, to strive against. 5:33
endeavour to spread corruption on earth.
5:64
and they labour hard to spread corruption on earth.
34:38 whereas all who strive against Our messages.
FA-ISʿAW + ILÁ—(I)—(imperfect)—(agent: masculine, plural)—to hasten. 62:9
hasten to the remembrance of God.
S-Q-Ṭ SUQIṬA+ FĪ—(I)—(perfect passive)—(agent of the passive: masculine, plural)— to smite hands in remorse. 7:149 when they would smite their hands in remorse.
S-K-T SAKATA + ʿAN—(I)—(perfect)—(agent: masculine, singular)—to become still. 7:154 And when Moses’s wrath as stilled.
S-K-N LI-TASKUNŪ + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to rest. 10:67 so that you might have rest therein. 28:73 so that you might rest therein. 40:61 so that you might rest therein.
LI-TASKUNŪ + ILÁ—(I)—(imperfect)—(agent: masculine, plural)—to incline towards. 30:21 so that you might incline towards them.
TASKUNŪNA + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to rest. 28:72 wherein you might rest.
LI-YASKUNA + ILÁ—(I)—(imperfect)—(agent: masculine, singular)—to incline towards.
Appendix I 231 7:189 so that man might incline [with love] towards woman.
LI-YASKUNŪ + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to rest. 27:86 so that they might rest therein.
S-L-KH NASLAKHU + MIN—(I)—(imperfect)—(agent: masculine, plural)—to withdraw. 36:37 We withdraw from it the [light of] day.
INSALAKHA + MIN—(VII)—(perfect)—(agent: masculine, singular)—to discard. 7:175 and who then discards them.
S-L-Q SALAQŪKUM + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to assail. 33:19 they will assail you [believers] with sharp tongues.
S-L-K SALAKA + FĪ—(I)—(perfect)—(agent: masculine, singular)—to trace out. 20:53 and has traced out for your ways [of livelihood] thereon.
SALAKAKUM + FĪ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to bring. 74:42 What has brought you into hellfire?
SALAKNĀHU + FĪ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to cause something to pass through something else. 26:200
We caused this [message] to pass [unheeded] through the hearts.
LI-TASLIKŪ + MIN—(I)—(imperfect)—(agent: masculine, plural)—to walk. 71:20 so that you might walk thereon on spacious paths.
232
Appendix I
NASLUKUHU + FĪ—(I)—(imperfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to cause to pervade. 15:12 We [now] cause this [scorn of Our message] to pervade the hearts.
YASLUKU + MIN—(I)—(imperfect)—(agent: masculine, singular)—to send forth. 72:27 He sends forth [the forces of heaven].
ISLUK + FĪ—(I)—(imperative)—(agent: masculine, singular)—to place, to put. 23:27 place on board of this [ark]! 28:32 put thy hand into thy bosom!
S-W-Y ISTAWÁ + ILÁ—(VIII)—(perfect)—(agent: masculine, singular)—to apply design. 2:29
and has applied His design to the heavens.
41:11 He [it is who] applied His design to the skies.
ISTAWÁ + ʿALÁ—(VIII)—(perfect)—(agent: masculine, singular)—to establish, to stand firm upon. 7:54
and is established on the throne of His almightiness.
10:3
and is established on the throne of His almightiness.
13:2
and is established on the throne of His almightiness.
20:5
established on the throne of His almightiness.
25:59 is established on the throne of His almightiness. 32:4
and is established on the throne of His almightiness.
48:29
stands firm upon its stem.
57:4
and is established on the throne of His almightiness.
ISTAWAT + ʿALÁ—(VIII)—(perfect)—(agent: feminine, singular)—to rest. 11:44 and the ark came to rest on Mount Jūdī.
ISTAWAYTA + ʿALÁ—(VIII)—(perfect)—(agent: masculine, singular)—to settle.
Appendix I 233 23:28 thou and those who are with thee are settled in the ark.
ISTAWAYTUM + ʿALÁ—(VIII)—(perfect)—(agent: masculine, plural)—to master. 43:13 whenever you have mastered them.
LITASTAWŪ + ʿALÁ—(VIII)—(imperfect)—(agent: masculine, plural)—to gain mastery. 43:13 in order that you might gain mastery over them.
TUSAWWÁ + BĀʾ—(II)—(imperfect passive)—(agent of the passive: masculine, plural)—to swallow. 4:42
the earth would swallow them.
Sh-D-D ISHDUD + ʿALÁ—(I)—(imperative)—(agent: masculine, singular)—to harden, to strength. 10:88 and harden their hearts! 20:31 Add Thou through him to my strength.
SH-R-B ʾUSHRIBŪ + FĪ—(I)—(perfect passive)—(agent of the passive: masculine, plural)—to be filled with love. 2:93
for their hearts are filled to overflowing with love of the [golden] calf.
Ṣ-B-B FA-ṢABBA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to let loose. 89:13 thy Sustainer let loose upon them a scourge of suffering.
Ṣ-R-F ṢARAFA + ʿAN—(I)—(perfect)—(agent: masculine, singular)—to free. 12:34 and freed him from the threat of their guile.
234
Appendix I
ṢARAFAKUM + ʿAN—(I)—(perfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to prevent. 3:152 He prevented you from defeating your foes.
ṢARAFNĀ + ILÁ—(I)—(perfect)—(agent: masculine, plural)—to cause to incline. 46:29 We caused a group of unseen beings to incline towards thee.
SA-ʾAṢRIFU + ʿAN—(I)—(future)—(agent: masculine, singular)—to cause to turn away. 7:146 From My messages shall I cause to turn away.
TAṢRIF + ʿAN—(I)—(imperfect)—(agent: masculine, singular)—to turn away. 12:33 unless Thou turn away their guile from me.
LI-NAṢRIFA + ʿAN—(I)—(perfect)—(agent: masculine, plural)—to avert. 12:24 We might avert from him all evil.
YAṢRIFUHU + ʿAN—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, singular)—to avert. 24:43 and averting it from whomever He wills.
IṢRIF + ʿAN—(I)—(imperative)—(agent: masculine, singular)—to avert. 25:65 avert from us the suffering of hell!
YUṢRAF + ʿAN—(I)—(imperfect passive)—(agent of the passive: masculine, singular)—to be spared. 6:16
Upon him who shall be spared on that Day.
ṢARRAFNĀ + FĪ—(I)—(perfect)—(agent: masculine, plural)—to give. 17:41
many facets have We given [to Our message] in this Qur’an.
18:54
have We given in this Qur’an many facets to every kind of lesson.
Appendix I 235 20:113 have given therein many facets to all manner of warnings.
ṢARRAFNĀ + LĀM—(I)—(perfect)—(agent: masculine, plural)—to give. 17:89 many facets have We given in this Qur’an to every kind of lesson.
Ṣ-N-ʿ LI-TUṢNAʿA + ʿALÁ—(I)—(imperfect passive)—(agent of the passive: masculine, singular)—to be formed. 20:39 in order that thou might be formed under Mine eye.
IṢṬANAʿTUKA + LĀM—(VIII)—(perfect)—(agent: masculine, singular)— (direct object: masculine, singular)—to choose. 20:41 I have chosen thee for Mine Own service.
Ṣ-W-B AṢĀBA + BĀʾ—(IV)—(perfect)—(agent: masculine, singular)—to cause to fall upon. 30:48 and as soon as He causes it to fall upon whomever He wills of His servants.
AṢABNĀHUM + BĀʾ—(IV)—(perfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to smite. 7:100 We could smite them [too] by means of their sins.
ʾUṢĪBU + BĀʾ—(IV)—(imperfect)—(agent: masculine, singular)—to afflict. 7:156 With My chastisement do I afflict whom I will.
TUṢĪBŪ + BĀʾ—(IV)—(imperfect)—(agent: masculine, plural)—to hurt. 49:6
lest you hurt people unwittingly.
NUṢĪBU + BĀʾ—(IV)—(imperfect)—(agent: masculine, plural)—to cause to alight.
236
Appendix I
12:56 We cause Our grace to alight upon whomever We will.
YUṢĪBU + BĀʾ—(IV)—(imperfect)—(agent: masculine, singular)—to cause to alight, to strike. 10:107
He causes it to alight upon whomsoever He wills of His servants
13:13
and strikes with them whom He wills.
24:43
striking therewith whomever He wills.
YUṢĪBAKUM + BĀʾ—(IV)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to inflict. 9:52
We are hopefully waiting for God to inflict chastisement upon you.
YUṢĪBAHUM + BĀʾ—(IV)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to afflict. 5:49
God’s will [thus] to afflict them for some of their sins.
Ḍ-R-B ḌARABA + LĀM—(I)—(perfect)—(agent: masculine, singular)—to propound, to argue, to attribute. 30:28 He propounds unto you a parable. 36:78 And [now] he [argues about Us]. 43:17 what he so readily attributes to the Most Gracious. 66:10 God has propounded a parable. 66:11 God has propounded a parable.
ḌARABTUM + FĪ—(I)—(perfect)—(agent: masculine, plural)—to go forth, to travel. 4:94
when you go forth [to war] in God’s cause.
4:101 when you go forth [to war] on earth. 5:106 while you are travelling far from home.
ḌARABNĀ + LĀM—(I)—(perfect)—(agent: masculine, plural)—to set forth, to proffer, to propound. 14:45 We have set forth unto you many a parable.
Appendix I 237 25:39 and unto each of them did We proffer lessons. 30:58 We have propounded unto men all kinds of parables. 39:27 We propounded unto men all kinds of parables.
ḌARABNĀ + ʿALÁ—(I)—(perfect)—(agent: masculine, plural)—to veil. 18:11 We veiled their ears in the cave for many a year.
ḌARABŪ + FĪ—(I)—(perfect)—(agent: masculine, plural)—to set out on a journey. 3:156 after having set out on a journey to faraway places.
ḌARABŪ + LĀM—(I)—(perfect)—(agent: masculine, plural)—to liken. 17:48 See to what they liken thee! 25:9
See what they liken to thee!
ḌARABŪHU + LĀM—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to put a comparison. 43:58 they put this comparison before thee.
TAḌRIBŪ + LĀM—(I)—(imperfect)—(agent: masculine, plural)—to coin. 16:74 do not coin any similitudes for God!
NAḌRIBU + ʿAN—(I)—(imperfect)—(agent: masculine, plural)—to withdraw. 43:5
Should We, perchance, withdraw this reminder from you altogether?
NAḌRIBUHĀ + LĀM—(I)—(imperfect)—(agent: masculine, plural)—(direct object: feminine, plural)—to propound. 29:43 We propound these parables unto man. 59:21 And [all] such parables We propound unto men.
YAḌRIBU + LĀM—(I)—(imperfect)—(agent: masculine, singular)—to propound, to set forth. 14:25 God propounds parables unto men.
238
Appendix I
24:35 God propounds parables unto men. 47:3
foes God set forth unto man the parables.
YAḌRIBNA + BĀʾ—(I)—(imperfect)—(agent: feminine, plural)—to draw. 24:31 let them draw their head coverings over their bosoms.
YAḌRIBŪNA + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to go about. 73:20 who will go about the land in search of God’s bounty.
IḌRIB + LĀM—(I)—(imperative)—(agent: masculine, singular)—– to propound, to strike out, to set forth. 18:32 And propound unto them the parable of two men. 18:45 And propound unto them the parable of the life of this world. 20:77 and strike out for them a dry path through the sea. 36:13 And set forth unto them a parable.
ḌURIBA + BĀʾ—(I)—(perfect passive)—(agent of the passive: masculine, singular)—to be raised. 57:13 a wall will be raised between them [and the believers].
ḌURIBAT + ʿALÁ—(I)—(perfect passive)—(agent of the passive: feminine, singular)—to overshadow. 2:61
ignominy and humiliation over shadowed them.
3:112 Overshadowed by ignominy are they wherever they may be saved. 3:112 and are overshadowed by humiliation.
Ḍ-L-L ḌALLA + ʿAN—(I)—(perfect)—(agent: masculine, singular)—to forsaken. 6:24
and [how] their false imagery has forsaken them!
6:94
and all your former fancies have forsaken you!
7:53
and all their false imagery will have forsaken them.
10:30 and all their false imagery will have forsaken them. 11:21 all their false imagery will have forsaken them.
Appendix I 239 16:87 and all their false imagery will have forsaken them. 28:75 and all their false imagery will have forsaken them. 41:48 all that they were wont to invoke aforetime will have forsaken them.
ḌALLŪ + ʿAN—(I)—(perfect)—(agent: masculine, plural)—to forsake. 46:28 Nay, they forsook them.
Ṭ-B-ʿ ṬABAʿA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to seal. 4:155
but God has sealed their hearts.
9:93
God has sealed their hearts.
16:108
They whose hears and whose hearing and whose sight God has sealed.
47:16
those whose hearts God has sealed.
NAṬBAʿU + ʿALÁ—(I)—(imperfect)—(agent: masculine, plural)—to seal. 7:100 sealing their hearts. 10:74 We seal the hearts.
YAṬBAʿU + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)—to seal, to set a seal. 7:101 God seals the hearts. 30:59 God seal the hearts. 40:35 God sets a seal on every arrogant, self-exalting heart.
ṬUBIʿA + ʿALÁ—(I)—(perfect passive)—(agent of the passive: masculine, singular)—to be sealed. 9:87
their hearts have been sealed.
63:3
a seal has been set on their hearts.
Ṭ-ʿ-N ṬAʿANŪ + FĪ—(I)—(perfect)—(agent: masculine, plural)—to revile. 9:12
and revile your religion.
240
Appendix I
Ṭ-GH-Y TAṬGHAW + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to transgress. 20:81 do not transgress therein the bounds of equity. 55:8
never transgress the measure [of what is right].
Ṭ-L-ʿ TAṬṬALIʿU + ʿALÁ—(VIII)—(imperfect)—(agent: feminine, plural)—to rise over. 104:7 which will rise over the [guilty] hearts.
Ṭ-M-S ṬAMASNĀ + ʿALÁ—(I)—(perfect)—(agent: masculine, plural)—to deprive. 36:66 We could surely have deprived them of their sight.
IṬMIS + ʿALÁ—(I)—(imperative)—(agent: masculine, singular)—to wipe out. 10:88 O our Sustainer! Wipe out their riches.
Ṭ-Y-B ṬIBNA + LĀM—(I)—(perfect)—(agent: feminine, plural)—to give up. 4:4
give up unto you aught thereof.
Ẓ-L-M ẒALAMŪ + BĀʾ—(I)—(perfect)—(agent: masculine, plural)—to sin, to reject. 7:103 they willfully rejected them. 17:59 they sinned against it.
TAẒLIM + MIN—(I)—(imperfect)—(agent: feminine, singular)—to fail. 18:33 and never failed therein in any way.
Ẓ-H-R YAẒHARŪ + ʿALÁ—(I)—(imperfect)—(agent: masculine, plural)—to overcome, to come to know, to be unaware.
Appendix I 241 9:8
were to overcome you.
18:20 if they should come to know of you. 24:31 children that are as yet unaware of women’s nakedness.
YAẒHARŪNA + ʿALÁ—(I)—(imperfect)—(agent: masculine, plural)—to ascend. 43:33 stairways whereon to ascend.
ẒAHARŪ + ʿALÁ—(III)—(perfect)—(agent: masculine, plural)—to aid. 60:9
or aid [others] in driving you forth.
YUẒAHIRŪ + ʿALÁ—(III)—(imperfect)—(agent: masculine, plural)—to aid. 9:4
and neither have aided anyone against you.
AẒHARAHU + ʿALÁ—(IV)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, singular)—to make something or someone known. 66:3
and God made this known to him.
YUẒHIRU + ʿALÁ—(IV)—(imperfect)—(agent: masculine, singular)—to disclose. 72:26 to none does He disclose aught of the mysteries of His Own unfathomable knowledge.
LI-YUẒHIRAHU + ʿALÁ—(VI)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, singular)—to cause to prevail. 9:33
He may cause it to prevail over all [false] religion.
48:28 He makes it prevail over every [false] religion. 61:9
He makes it prevail over all [false] religion.
TAẒĀHRĀ + ʿALÁ—(VI)—(perfect)—(agent: masculine, dual)—to uphold each other. 66:4
And if you uphold each other against him.
TAẒAHARŪNA + ʿALÁ—(VI)—(imperfect)—(agent: masculine, plural)—to aid one another.
242
Appendix I
2:85
aiding one another against them in sin and hatred.
ʿ-B-R TAʿBURŪNA + LĀM—(I)—(imperfect)—(agent: masculine, plural)—to interpret. 12:43 if you are able to interpret dreams.
ʿ-SH-W YAʿSHU + ʿAN—(I)—(imperfect)—(agent: masculine, singular)—to remain blind. 43:36 who chooses to remain blind to the remembrance of the Most Gracious.
ʿ-L-W TAʿLŪ + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)—to exalt. 27:31 Exalt not yourselves against Me. 44:19 exalt not yourselves against God.
ʿ-M-Y ʿAMIYAT + ʿALÁ—(II)—(perfect)—(agent: feminine, singular)—to erase. 28:66 have been erased from their minds.
ʿUMMIYAT + ʿALÁ—(I)—(perfect passive)—(agent of the passive: feminine, singular)—to remain blind. 11:28 to which you have remained blind.
GH-D-R NUGHĀDIR + MIN—(III)—(imperfect)—(agent: masculine, plural)—to leave out. 18:47 leaving out none of them.
GH-R-R GHARRAKA + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—to lure away.
Appendix I 243 82:6
What is it that lures thee away from thy bountiful Sustainer?
GHARRAHUM + FĪ—(I)—(perfect)—(agent: masculine, plural)—to cause to betray. 3:24
caused them to betray their faith.
GH-R-Y LA-NUGHRIYANNAKA + BĀʾ—(IV)—(imperfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to give mastery over something or someone. 33:60 We shall indeed give thee mastery over them.
GH-L-Ẓ ʾIGHLUẒ + ʿALÁ—(I)—(imperative)—(agent: masculine, singular)—to be adamant. 9:73
and be adamant with them.
66:9
and be adamant with them.
GH-M-Ḍ TUGHMIḌŪ + FĪ—(IV)—(imperfect)—(agent: masculine, plural)—to avert eyes in disdain. 2:267 without averting your eyes in disdain.
GH-N-Y YAGHNAW + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to live. 11:68 as though they had never lived there. 11:95 as though they had never lived there.
AGHNÁ + ʿAN—(IV)—(perfect)—(agent: masculine, singular)—to avail. 7:48
What has your amassing [of wealth] availed you?
15:84
and of no avail to them.
26:207
of what avail to them will be all their past enjoyments.
244
Appendix I
39:50
but of no avail to them was all that they had ever achieved.
40:82
but all that they ever achieved was of no avail to them
46:26
were of the least avail to them.
69:28
Of no avail to me is all that I have [ever] possessed.
111:2
What will his wealth avail him?
AGHNAT + ʿAN—(IV)—(perfect)—(agent: feminine, singular)—to avail. 11:101
proved of no avail whatever to them.
ʾUGHNĪ + ʿAN—(IV)—(imperfect)—(agent: masculine, singular)—to avail. 12:67 I can be of no avail whatever to you.
TUGHNI + ʿAN—(IV)—(imperfect)—(agent: feminine, singular)—to avail. 9:25
they proved of no avail whatever to you.
36:23 their intercession could not in the least avail me.
TUGHNIYA + ʿAN—(IV)—(imperfect)—(agent: feminine, singular)—to avail. 3:10
will in the least avail them against God.
3:116
will in the least avail them against God.
8:19
never will your community be of any avail to you.
10:101
But of what avail could all the messages and all the warnings be to people who will not believe?
58:17
will be of the least avail to them against God.
YUGHNŪ + ʿAN—(IV)—(imperfect)—(agent: masculine, plural)—to avail. 45:19 they could never be of any avail to thee.
YUGHNĪ + MIN—(IV)—(imperfect)—(agent: masculine, singular)—to avail, to be a substitute, to take place, to still. 10:36 conjecture can never be a substitute for truth. 53:28 never can surmise take the place of truth. 77:31 will be of no avail against the flame 88:7
which gives no strength and neither stills hunger.
Appendix I 245 YUGHNĪ + ʿAN—(IV)—(imperfect)—(agent: masculine, singular)—to avail. 12:68 this proved of no avail whatever to them against [the plan of] God. 19:42 and can be of no avail whatever to thee. 44:41 the Day when no friend shall be of the least avail to his friend. 45:10 shall be of no avail whatever to them. 52:46 the Day when none of their scheming will be of the least avail to them. 92:11 and what will his wealth avail him when he goes down [to his grave]?
YUGHNIYĀ + ʿAN—(IV)—(imperfect)—(agent: masculine, dual)—to avail. 66:10 and neither of the two [husbands] will be of any avail to these two women.
GH-W-TH YUGHĀTHŪ + BĀʾ—(IV)—(imperfect passive)—(agent of the passive: masculine, plural)—to be given. 18:29 they will be given water [hot] like molten lead.
F-T-Ḥ FATAḤA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to disclose. 2:76
Do you inform them of what God has disclosed to you?
FATAḤNĀ + LĀM—(I)—(perfect)—(agent: masculine, plural)—to lay open. 48:1
We have laid open before thee a manifest victory.
YAFTAḤU + LĀM—(I)—(imperfect)—(agent: masculine, singular)—to open up. 35:2
Whatever grace God opens up to man.
YASTAFTIḤŪNA + ʿALÁ—(X)—(imperfect)—(agent: masculine, plural)—to pray for victory. 2:89
they used to pray for victory over those who were bent on denying the truth.
246
Appendix I
F-R-Ṭ YAFRUṬA + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)—to act hastily. 20:45 we fear lest he act hastily with regard to us.
FARRAṬTU + FĪ—(II)—(perfect)—(agent: masculine, singular)—to be remiss. 39:56 Alas for me for having been remiss in what is due to God.
FARRAṬTUM + FĪ—(II)—(perfect)—(agent: masculine, plural)—to fail. 12:80 you had failed with regard to Joseph.
FARRAṬNĀ + FĪ—(II)—(perfect)—(agent: masculine, plural)—to disregard, to neglect. 6:31
Alas for us, that we disregarded it!
6:38
no single thing have We neglected in Our decree.
F-R-GH SA-NAFRAGHU + LĀM—(I)—(future)—(agent: masculine, plural)—to take task. 55:31 we shall take you to task.
AFRIGH + ʿALÁ—(IV)—(imperative)—(agent: masculine, singular)—to shower. 2:250 Shower us with patience in adversity. 7:126 Shower us with patience in adversity.
F-S-Q FASAQA + ʿAN—(I)—(perfect)—(agent: masculine, singular)—to turn away. 18:50 he turned away from his Sustainer’s command.
F-Ṣ-L FAṢALA + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—to set out.
Appendix I 247 2:249 And when Saul set out with his forces.
F-Ḍ-Y AFḌÁ + ILÁ—(IV)—(perfect)—(agent: masculine, singular)—to give oneself. 4:21
after you have given yourselves to one another.
F-Y-Ḍ AFAḌTTUM + FĪ—(IV)—(perfect)—(agent: masculine, plural)—to indulge. 24:14 in result of all [the calumny] in which you indulge.
TUFIḌŪNA + FĪ—(IV)—(imperfect)—(agent: masculine, plural)—to enter upon, to plunge into. 10:61 when you enter upon it. 46:8
into which you so recklessly plunge.
AFIḌŪ + ʿALÁ—(IV)—(imperative)—(agent: masculine, plural)—to pour. 7:50
Pour some water upon us.
Q-B-Ḍ QABAḌNĀHU + ILÁ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to draw. 25:46 We draw it in towards Ourselves.
Q-D-R QADARA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to straiten. 89:16 He tries him by straitening his means of livelihood.
QUDIRA + ʿALÁ—(I)—(perfect passive)—(agent of the passive: masculine, singular)—to be scanty. 65:7
whose means of subsistence are scanty.
248
Appendix I
Q-D-M QADIMNĀ + ILÁ—(I)—(perfect)—(agent: masculine, plural)—to turn. 25:23 We shall have turned towards all the [supposedly good] deeds.
Q-DH-F QADHAFA + FĪ—(I)—(perfect)—(agent: masculine, singular)—to cast. 33:26 and cast terror into their hearts. 59:2
and cast terror into their hearts.
NAQDHIFU + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to hurl. 21:18 We hurl the truth against falsehood.
YAQDHIFU + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—to hurl. 34:48 Verily, my Sustainer hurls the truth.
YAQDHIFŪNA + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to cast. 34:53 and had been wont to cast scorn.
Q-Ḍ-Y QAḌAYNĀ + ILÁ—(I)—(perfect)—(agent: masculine, plural)—to reveal, to make something known, to impose, to decree. 15:66 We revealed unto him this decree. 17:4
And we made [this] known to the children of Israel.
28:44 We imposed the Law upon Moses. 34:14 We decreed that he should die.
IQḌŪ + ILÁ—(I)—(imperative)—(agent: masculine, plural)—to carry out. 10:71 and then carry out against me!
QUḌYA + ILÁ—(I)—(perfect passive)—(agent of the passive: masculine, singular)—to come forthwith. 10:11 their end would indeed come forthwith!
Appendix I 249 YUQḌÁ + ILÁ—(I)—(imperfect passive)—(agent of the passive: masculine, singular)—to be revealed. 20:114
ere it has been revealed unto thee in full.
Q-L-B TUQLABŪNA + ILÁ—(I)—(imperfect passive)—(agent of the passive: masculine, plural)—to be made to return. 29:21 and unto Him you shall be made to return.
QALLABŪ + LĀM—(II)—(perfect)—(agent: masculine, plural)—to devise. 9:48
and devised all manner of plots against thee.
TATAQALLABU + FĪ—(V)—(imperfect)—(agent: feminine, plural)—to convulse. 24:37 all hearts and eyes will be convulsed.
INQALABA + ILÁ—(VII)—(perfect)—(agent: masculine, singular)—to turn away. 22:11 he turns away utterly.
INQALABTUM + ILÁ—(VII)—(perfect)—(agent: masculine, plural)—to turn about, to return. 3:144 will you turn about on your heels. 9:95
When you will have returned to them.
INQALABŪ + BĀʾ—(VII)—(perfect)—(agent: masculine, plural)—to return. 3:174 and returned [from the battle] with God’s blessings and bounty.
INQALABŪ + ILÁ—(VII)—(perfect)—(agent: masculine, plural)—to come, to return. 12:62 when they come home. 83:31 they return to people of their own kind.
YANQALIB + ILÁ—(VII)—(imperfect)—(agent: masculine, singular)—to return, to fall back upon, to turn.
250
Appendix I
48:12 would never return to their kith and kin. 67:4
thy vision will fall back upon thee.
84:9
and will [be able to] turn joyfully to those of his own kind.
YANQALIB + ʿALÁ—(VII)—(imperfect)—(agent: masculine, singular)—to turn about. 2:143 those who turn about on their heels. 3:144 But he that turns about on his heels.
Q-W-M TAQŪMŪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to treat. 4:127 and about your duty to treat orphans with equity.
YAQŪMA + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to behave. 57:25 so that men might behave with equity.
NUQIMU + LĀM—(IV)—(imperfect)—(agent: masculine, plural)—to assign. 18:105
and no weight shall We assign to them.
AQIM + LĀM—(IV)—(imperative)—(agent: masculine, singular)—to set. 10:105 set thy face steadfastly towards the [true] faith. 30:30
set thy face steadfastly towards the [one ever-true] faith.
30:43
Set, then, thy face steadfastly towards the one ever-true faith.
ISTAQĀMŪ + LĀM—(X)—(perfect)—(agent: masculine, plural)—to remain true. 9:7
so long as they remain true to you.
ISTAQĪMŪ + LĀM—(X)—(imperative)—(agent: masculine, plural)—to be true. 9:7
be true to them.
ISTAQĪMŪ + ILÁ—(X)—(imperative)—(agent: masculine, plural)—to go straight towards.
Appendix I 251 41:6
go, then, straight towards Him.
Q-Y-Ḍ QAYYAḌNĀ + LĀM—(II)—(perfect)—(agent: masculine, plural)—to assign. 41:25 We assigned to them.
NUQAYYIḌ + LĀM—(II)—(imperfect)—(agent: masculine, plural)—to assign 43:36 to him We assign an [enduring] evil impulse.
K-B-R KABURA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to distress, to repugnant, to appear oppressive. 6:35
And if it distresses thee.
10:71 If my presence [among you] and my announcement of God’s messages are repugnant to you. 42:13 appears oppressive to those who are wont to ascribe to other beings.
YAKBURU + FĪ—(I)—(imperfect)—(agent: masculine, singular)—to appear. 17:51 appears yet farther removed [from life].
K-T-B KATABA + FĪ—(I)—(perfect)—(agent: masculine, singular)—to inscribe. 58:22 He has inscribed faith.
KATABA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to will upon, to ordain. 6:12
Unto God, who has willed upon Himself the law of grace and mercy
6:54
Your Sustainer has willed upon Himself the law of grace and mercy.
59:3
And had it not been for God’s having ordained banishment for them.
KATABA + LĀM—(I)—(perfect)—(agent: masculine, singular)—to ordain, to promise, to decree.
252
Appendix I
2:187 and avail yourselves of that which God has ordained for you. 5:21
Enter the holy land which God has promised you.
9:51
save what God has decreed!
KATABTA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to ordain. 4:77
Why hast Thou ordained fighting for us?
KATABNĀ + ʿALÁ—(I)—(perfect)—(agent: masculine, plural)—to ordain. 4:66
Yet if We were to ordain for them.
5:32
We ordain unto the children of Israel.
5:45
We ordained for them in that [Torah].
KATABNĀ + FĪ—(I)—(perfect)—(agent: masculine, plural)—to lay down. 21:105
We laid it down.
KATABNĀ + LĀM—(I)—(perfect)—(agent: masculine, plural)—to ordain. 7:145 We ordained for him.
KATABNĀHĀ + ʿALÁ—(I)—(perfect)—(agent: masculine, plural)—(direct object: feminine, singular)—to enjoin upon. 57:27 We did not enjoin it upon them.
SA-ʾAKTUBUHĀ + LĀM—(I)—(future)—(agent: masculine, singular)— (direct object: feminine, singular)—to confer. 7:156 I shall confer it on those who are conscious.
WA-ʾKTUB + LĀM—(I)—(imperative)—(agent: masculine, singular)—to ordain. 7:156 And ordain Thou for us.
KUTIBA + ʿALÁ—(I)—(perfect passive)—(agent of the passive: masculine, singular)—to ordain, to decree. 2:178 Just retribution is ordained for you in cases of killing.
Appendix I 253 2:180 It is ordained for you. 2:183 Fasting is ordained for you as it was ordained for those before you. 2:216 Fighting is ordained for you. 2:246 fighting is ordained for you. 2:246 when fighting was ordained for them. 3:154 those [of you] whose death had been ordained. 4:77
fighting [in God’s cause] is ordained for them.
22:4
about which it has been decreed.
KUTIBA + LĀM—(I)—(perfect passive)—(agent of the passive: masculine, singular)—to be conveyed, to be recorded 4:127 in what is being conveyed unto you. 9:120 a good deed is recorded in their favour. 9:121 it is recorded in their favour.
K-SH-F KASHAFA + ʿAN—(I)—(perfect)—(agent: masculine, singular)—to remove. 16:54 He has removed the harm from you.
KASHAFTA + ʿAN—(I)—(perfect)—(agent: masculine, singular)—to remove. 7:134 If thou remove this plague from us.
KASHAFAT + ʿAN—(I)—(perfect)—(agent: feminine, singular)—to bare. 27:44 she bared her legs.
KASHAFNĀ + ʿAN—(I)—(perfect)—(agent: masculine, plural)—to remove, to free, to lift. 7:135 We removed the plague from them. 10:12 We have freed him of his affliction. 10:98 We removed from them the suffering of disgrace. 43:50 We removed the suffering from them. 50:22 We have lifted from thee thy veil.
ʾIKSHIF + ʿAN—(I)—(imperative)—(agent: masculine, singular)—to relieve.
254
Appendix I
44:12 relieve us of suffering!
YUKSHAFU + ʿAN—(I)—(imperfect passive)—(agent of the passive: masculine, singular)—to be bared. 68:42 shall be bared to the bone.
K-F-F KAFFA + ʿAN—(I)—(perfect)—(agent: masculine, singular)—to stay. 5:11
He stayed their hands from you.
48:20 and has stayed from you the hands of [hostile] people. 48:24 stayed their hands from you.
KAFAFTU + ʿAN—(I)—(perfect)—(agent: masculine, singular)—to prevent 5:110 I prevented the children of Israel from harming thee.
YAKUFFŪNA + ʿAN—(I)—(imperfect)—(agent: masculine, plural)—to ward off. 21:39 they will not be able to ward off the fire from their faces!
K-W-R YUKAWWIRU + ʿALÁ—(II)—(imperfect)—(agent: masculine, singular)—to cause to flow. 39:5
He causes the night to flow into the day.
39:5
and causes the day to flow into the night.
L-B-TH TALABBATHŪ + BĀʾ—(V)—(imperfect)—(agent: masculine, plural)—to delay. 33:14 would have done so without much delay.
L-B-S LABASNĀ + ʿALÁ—(I)—(perfect)—(agent: masculine, plural)—to confuse.
Appendix I 255 6:9
We would only have confused them.
TALBISŪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to overlay. 2:42
And do not overlay the truth with falsehood.
TALBUSŪNA + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to cloak. 3:71
Why do you cloak the truth with falsehood?
YALBISŪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to obscure. 6:82
who have not obscured their faith by wrongdoing.
YALBISŪ + ʿALÁ—(I)—(imperfect)—(agent: masculine, plural)—to confuse. 6:137 confuse them in their faith.
L-J-J LAJJŪ + FĪ—(I)—(perfect)—(agent: masculine, plural)—to persist. 23:75 they would still persist in their overweening arrogance. 67:21 persist in their disdain.
L-Ḥ-D YULḤIDŪNA + ILÁ—(IV)—(imperfect)—(agent: masculine, plural)—to point. 16:103 the tongue of him to whom they so maliciously point is wholly outlandish.
YULḤIDŪNA + FĪ—(IV)—(imperfect)—(agent: masculine, plural)—to distort. 7:180 who distort the meaning of His attributes. 41:40 they who distort the meaning of Our messages.
L-F-T TALFITANĀ + ʿAN—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to turn away.
256
Appendix I
10:78 Hast thou come to turn us away from what we found our forefathers believing in.
L-Q-Y ALQÁ + ILÁ—(IV)—(perfect)—(agent: masculine, singular)—to offer. 4:94
who offers you the greeting of peace.
ALQÁ + FĪ—(IV)—(perfect)—(agent: masculine, singular)—to cast. 22:52 Satan would cast an aspersion on his innermost aims.
ALQĀHĀ + ILÁ—(IV)—(perfect)—(agent: masculine, singular)—(direct object: feminine, singular)—to convey. 4:171 He had conveyed unto Mary.
ALQAW + ILÁ—(IV)—(perfect)—(agent: masculine, plural)—to offer, to fling, to proffer. 4:90
and offer you peace.
16:86 whereupon [those beings] will fling at them the retort. 16:87 proffer their surrender to God.
ALQAYTU + ʿALÁ—(IV)—(perfect)—(agent: masculine, singular)—to spread. 20:39 I spread Mine Own love over thee.
SA-ʾULQĪ + FĪ—(IV)—(future)—(agent: masculine, singular)—to cast. 8:12
I shall cast terror into the hearts of those who are bent on denying the truth.
TULQŪ + BĀʾ—(IV)—(imperfect)—(agent: masculine, plural)—to throw. 2:195 and let not your own hands throw you into destruction.
TULQŪNA + ILÁ—(IV)—(imperfect)—(agent: masculine, plural)—to show. 60:1
showing them affection.
Appendix I 257 SA-NULQĪ + FĪ—(IV)—(future)—(agent: masculine, plural)—to cast. 3:151 We shall cast dread.
SA-NULQĪ + ʿALÁ—(IV)—(future)—(agent: masculine, plural)—to bestow. 73:5
We shall bestow upon thee a weighty message.
YULQŪ + ILÁ—(IV)—(imperfect)—(agent: masculine, plural)—to offer. 4:91
and do not offer you peace.
YULQĪ + ʿALÁ—(IV)—(imperfect)—(agent: masculine, singular)—to bestow. 40:15 He bestows inspiration upon whomever He wills of His servants.
ULQIYA + ʿALÁ—(IV)—(perfect passive)—(agent of the passive: masculine, singular)—to be bestowed. 54:25 a [divine] reminder have been bestowed.
YULQÁ + ILÁ—(IV)—(imperfect passive)—(agent of the passive: masculine, singular)—to be offered. 28:86 this divine writ would [one day] be offered to thee.
TALAQQÁ + MIN—(V)—(perfect)—(agent: masculine, singular)—to receive. 2:37
Adam received words [of guidance] from his Sustainer.
TALAQQAWNAHU + BĀʾ—(V)—(imperfect)—(agent: masculine, plural)— (direct object: masculine, singular)—to take up. 24:15 when you take it up with your tongues.
L-W-Y TALWŪNA + ʿALÁ—(I)—(imperfect)—(agent: masculine, plural)—to pay heed. 3:153 paying no heed to anyone.
258
Appendix I
M-T-ʿ TAMATTAʿ + BĀʾ—(V)—(imperative)—(agent: masculine, singular)—to enjoy. 39:8
Enjoy thyself for a while in this thy denial of the truth.
M-D-D TAMUDDANNA + ILÁ—(I)—(imperfect)—(agent: masculine, singular)—to turn. 15:88
turn, not thine eyes [longingly] towards the worldly benefits.
20:131
never turn thine, eyes [with longing] towards whatever splendour of this world’s life.
NAMUDDU + LĀM—(I)—(imperfect)—(agent: masculine, plural)—to lengthen. 19:79 We shall lengthen the length of his suffering.
YAMDUD + LĀM—(I)—(imperfect)—(agent: masculine, singular)—to lengthen. 19:75 may the Most Gracious lengthen the span of his life.
YAMUDDUHUM + FĪ—(I)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to leave. 2:15
and will leave them for a while in their overweening arrogance.
YAMUDDUNAHUM + FĪ—(I)—(imperfect)—(agent: masculine, plural)— (direct object: masculine, plural)—to draw. 7:202 even though their [godless] brethren would [like to] draw them into error.
AMADDAKUM + BĀʾ—(IV)—(perfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to provide. 26:132
who has [so] amply provided you with all [the good] that you might think of.
26:133
amply provided you with flocks, and children.
AMDADNĀKUM + BĀʾ—(IV)—(perfect)—(agent: masculine, plural)— (direct object: masculine, plural)—to aid.
Appendix I 259 17:6
and aided you with wealth and offspring.
AMDADNĀHUM + BĀʾ—(IV)—(perfect)—(agent: masculine, plural)— (direct object: masculine, plural)—to bestow. 52:22 We shall bestow on them fruit and meat in abundance.
ATUMIDDUNANI + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)— (direct object: masculine, singular)—to add. 27:36 you people mean to add to my wealth?
NUMIDDUHUM + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)— (direct object: masculine, plural)—to provide. 23:55 We provide them.
YUMDIDKUM + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to aid. 3:125 will aid you with five thousand angels swooping down. 71:12 will aid you with worldly goods and children.
YUMIDDAKUM + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)— (direct object: masculine, plural)—to aid. 3:124 will aid you with three thousand angels.
M-R-R MARRAT + BĀʾ—(I)—(perfect)—(agent: feminine, singular)—to continue to bear. 7:189 and continues to bear it.
M-S-S MASSANĪ + BĀʾ—(I)—(perfect)—(agent: masculine, singular)—(direct object: masculine, singular)—to afflict. 38:41 Satan has afflicted me with [utter] weariness and suffering.
260
Appendix I
TAMASSŪHĀ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—(direct object: feminine, singular)—to do. 7:73
do her no harm.
11:64
do her no harm.
26:156 do her no harm.
YAMSASKA + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—(direct object: masculine, singular)—to touch. 6:17
if God should touch thee with misfortune.
6:17
if He should touch thee with good fortune.
10:107
if God should touch thee with misfortune.
M-S-K YUMASSIKŪNA + BĀʾ—(II)—(imperfect)—(agent: masculine, plural)—to hold fast. 7:170 those who hold fast to the divine writ.
TUMSIKŪ + BĀʾ—(I)—(imperfect)—(agent: masculine, plural)—to hold. 60:10 hold not to the marriage-tie with women who [continue to] deny the truth.
AMSIK + ʿALÁ—(IV)—(imperative)—(agent: masculine, singular)—to hold on. 33:37 Hold on to thy wife.
AMSIKŪHUNNA + BĀʾ—(IV)—(imperative)—(agent: masculine, plural)— (direct object: feminine, plural)—to retain. 2:231 then either retain them in a fair manner. 65:2
either retain them in a fair manner.
FA-ʾAMSIKŪHUNNA + FĪ—(IV)—(imperative)—(agent: masculine, plural)— (direct object: feminine, plural)—to confine. 4:15
confine the guilty women to their houses.
Appendix I 261
M-L-Y AMLÁ + LĀM—(IV)—(perfect)—(agent: masculine, singular)—to fill with false hopes. 47:25 and filled them with false hopes.
AMLAYTU + LĀM—(IV)—(perfect)—(agent: masculine, singular)—to give rein for a while. 13:32 and for a while I gave rein to those who were [thus] bent on denying the truth. 22:44 gave rein, for a while, to the deniers of the truth. 22:48 have I given rein for a while.
ʾUMLĪ + LĀM—(IV)—(imperfect)—(agent: masculine, singular)—to give rein for a while. 7:183 I may give them rein for a while. 68:45 I may give them rein for a while.
NUMLĪ + LĀM—(IV)—(imperfect)—(agent: masculine, plural)—to give rein. 3:178 Our giving them rein is good for them. 3:178 We give them rein.
M-W-J YAMŪJU + FĪ—(I)—(imperfect)—(agent: masculine, singular)—to surge like waves. 18:99 to surge like waves [that dash] against one another.
M-Y-Z TAMAYYAZU + MIN—(V)—(imperfect)—(agent: masculine, singular)—to burst. 67:8
well-nigh bursting with fury.
M-Y-L YAMĪLŪNA + ʿALÁ—(I)—(imperfect)—(agent: masculine, plural)—to fall upon.
262
Appendix I
4:102 they might fall upon you in a surprise attack.
N-B-DH NABDHNĀHU + BI—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, singular)—to be cast forth. 37:145 but We caused him to be cast forth on a desert shore.
NABDHNĀHUM + FĪ—(I)—(perfect)—(agent: masculine, plural)—(direct object: masculine, plural)—to cast. 28:40 and cast them into the sea. 51:40 and cast them all into the sea.
ʾINBIDH + ʿALÁ—(I)—(imperative)—(agent: masculine, singular)—to cast. 8:58
cast it back at them in an equitable manner!
NUBIDHA + BĀʾ—(I)—(perfect passive)—(agent of the passive: masculine, singular)—to be cast forth. 68:49 he would indeed have been cast forth upon that barren shore.
LA-YUNABDHANNA + FĪ—(I)—(imperfect passive)—(agent of the passive: masculine, singular)—to be abandoned. 104:4 he shall indeed be abandoned to crushing torment!
INTABADHAT + BĀʾ—(VII)—(perfect)—(agent: feminine, singular)—to withdraw. 19:22 she withdrew with him to a far-off place.
INTABADHAT + MIN—(VII)—(perfect)—(agent: feminine, singular)—to withdraw. 19:16 She withdrew from her family to an eastern place.
N-Z-ʿ NAZAʿNĀ + MIN—(I)—(perfect)—(agent: masculine, plural)—to call forth.
Appendix I 263 28:75 We will have called forth witnesses from within every community.
NAZAʿNĀHĀ + MIN—(I)—(perfect)—(agent: masculine, plural)—(direct object: feminine, singular)—to take away. 11:9
and then take it away from him.
TANZIʿU + MIN—(I)—(imperfect)—(agent: masculine, singular)—to take away. 3:26
and takest away dominion from whom Thou willest.
LA-NANZIʿANNA + MIN—(I)—(imperfect)—(agent: masculine, plural)—to draw forth. 19:69 We shall, indeed, draw forth from every group.
YATANĀZAʿŪNA + FĪ—(VI)—(imperfect)—(agent: masculine, plural)—to pass on. 52:23 they shall pass on to one another a cup.
N-Z-F YUNZAFŪN + ʿAN—(I)—(imperfect passive)—(agent of the passive: masculine, plural)—to get drunk. 37:47 and they will not get drunk thereon.
N-Z-L NAZALA + BI—(I)—(perfect)—(agent: masculine, singular)—to alight. 37:177 once it alights upon them.
NAZZALA + ʿALÁ—(II)—(perfect)—(agent: masculine, singular)—to enjoin. 4:140 He has enjoined upon you in this divine writ.
ANZALA + LĀM—(IV)—(perfect)—(agent: masculine, singular)—to bestow. 10:59 which God has bestowed upon you from on high?
264
Appendix I
39:6
and he has bestowed upon you.
ANZALTA + LĀM—(IV)—(perfect)—(agent: masculine, singular)—to bestow. 28:24 which Thou mayest bestow upon me.
ANZALNĀ + ʿALÁ—(IV)—(perfect)—(agent: masculine, plural)—to bestow. 7:26
We have bestowed upon you.
N-SH-R YANSHUR + LĀM—(I)—(imperfect)—(agent: masculine, singular)—to spread. 18:16 God will spread His grace over you.
ANSHARNĀ + BĀʾ—(IV)—(perfect)—(agent: masculine, plural)—to raise. 43:11 We raise therewith dead land to life.
N-Ṭ-Q YANṬIQU + BĀʾ—(I)—(imperfect)—(agent: masculine, singular)—to speak. 23:62 for with Us is a record that speaks the truth. 45:29 This Our record speaks you in all truth.
N-Ẓ-R NAẒARA + FĪ—(I)—(perfect)—(agent: masculine, singular)—to cast a glance. 37:88 Then he cast a glance at the stars.
YANẒURU + ILÁ—(I)—(imperfect)—(agent: masculine, singular)—to look. 3:77
nor look upon them.
10:43 And there are among them such as [pretend to] look towards thee.
YANẒUR + ILÁ—(I)—(imperfect)—(agent: masculine, singular)—to consider. 80:24 Let man, then, consider [the sources of] his food.
Appendix I 265 YANẒURŪ + ILÁ—(I)—(imperfect)—(agent: masculine, plural)—to look. 50:6
Do they not look at the sky above them?
YANẒURŪ + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to consider. 7:185 Have they, then, never considered [God’s] mighty dominion?
YANẒURŪNA + ILÁ—(I)—(imperfect)—(agent: masculine, plural)—to gaze. 88:17 gaze at the clouds pregnant with water.
ʾUNẒURŪ + ILÁ—(I)—(imperative)—(agent: masculine, plural)—to behold. 6:99
Behold their fruit!
N-F-SH NAFASHAT + FĪ—(I)—(perfect)—(agent: feminine, plural)—to stray by night. 21:78 into which some people’s sheep had strayed by night and pastured therein.
N-Q-B NAQQABŪ + FĪ—(II)—(perfect)—(agent: masculine, plural)—to become wanderer. 50:36 they became wanderers on the face of the earth.
N-Q-R NUQIRA + FĪ—(I)—(perfect passive)—(agent of the passive: masculine, singular)—to be sounded. 74:8
when the trumpet-call [of resurrection] is sounded.
N-K-S NUKISŪ + ʿALÁ—(I)—(perfect passive)—(agent of the passive: masculine, plural)—to relapse. 21:65 they relapsed into their former way of thinking.
266
Appendix I
N-K-Ṣ NAKAṢA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to turn. 8:48
he turned on his heels.
TANKUṢŪNA + ʿALÁ—(I)—(imperfect)—(agent: masculine, plural)—to turn about. 23:66 you would turn about on your heels
N-Y-B ANĀBA + ILÁ—(IV)—(perfect)—(agent: masculine, singular)—to turn towards. 31:15 an those who turn towards Me.
ANĀBŪ + ILÁ—(IV)—(perfect)—(agent: masculine, plural)—to turn unto. 39:17 and turn unto God instead.
ANABNĀ + ILÁ—(IV)—(perfect)—(agent: masculine, plural)—to turn unto. 60:4
and unto Thee do we turn.
ʾUNĪBŪ + ILÁ—(IV)—(imperfect)—(agent: masculine, singular)—to turn unto. 11:88 and unto Him do I always turn! 42:10 and unto Him do I always turn!
ANĪBŪ + ILÁ—(IV)—(imperative)—(agent: masculine, plural)—turn towards. 39:54 turn towards your Sustainer.
H-R-ʿ YUHRAʿŪNA + ILÁ—(I)—(imperfect passive)—(agent of the passive: masculine, plural)—to impel. 11:78 impelled towards his house.
YUHRAʿŪNA + ʿALÁ—(I)—(imperfect passive)—(agent of the passive: masculine, plural)—to make haste.
Appendix I 267 37:70 they make haste to follow in their footsteps.
H-SH-SH AHISHSHU + ʿALÁ—(I)—(imperfect)—(agent: masculine, singular)—to beat down leaves. 20:18 and with it I beat down leaves for my sheep.
H-L-K HALAKA + ʿAN—(I)—(perfect)—(agent: masculine, singular)—to die away. 69:29 all my power of argument has died away from me.
H-L-L ʾUHILLA + LĀM—(IV)—(perfect passive)—(agent of the passive: masculine, singular)—to be invoked 2:173
and that over which any name other than God’s has been invoked.
5:3
and that over which any name other than God’s has been invoked.
6:145
or a sinful offering over which any name other than God’s has been invoked.
16:115 and that over which any name other than God’s has been invoked.
H-W-Y TAHWĪ + ILÁ—(I)—(imperfect)—(agent: feminine, plural)—to incline. 14:37 cause Thou, therefore, people’s hearts to incline towards them.
TAHWĪ + BĀʾ—(I)—(imperfect)—(agent: feminine, plural)—to blow away. 22:31 or the wind blows him away onto a far-off place.
H-Y-M YAHĪMŪNA + FĪ—(I)—(imperfect)—(agent: masculine, plural)—to roam confusedly. 26:225
they roam confusedly through all the valleys.
268
Appendix I
W-J-F AWJAFTUM + ʿALÁ—(IV)—(perfect)—(agent: masculine, plural)—to spur. 59:6
you did not have to spur horse or riding-camel for its sake.
W-S-M SA-NASIMUHU + ʿALÁ—(I)—(future)—(agent: masculine, plural)—(direct object: masculine, singular)—to brand with indelible disgrace. 68:16 We shall brand him with indelible disgrace!
W-Ḍ-ʿ WAḌʿNĀ + ʿAN—(I)—(perfect)—(agent: masculine, plural)—to lift. 94:2
and lifted from thee the burden.
YAḌAʿU + ʿAN—(I)—(imperfect)—(agent: masculine, singular)—to lift. 7:157 and lift from them their burdens.
W-F-Y NUWAFFI + ILÁ—(II)—(imperfect)—(agent: masculine, plural)—to repay. 11:15 We shall repay them in full.
YUWAFFA + ILÁ—(II)—(imperfect passive)—(agent of the passive: masculine, singular)—to be repaid. 2:272 will be repaid unto you in full. 8:60
shall be repaid to you in full.
W-Q-ʿ WAQAʿA + ʿALÁ—(I)—(perfect)—(agent: masculine, singular)—to become ready, to become beset, to struck, to stand revealed. 4:100 his reward is ready with God. 7:71
You are already beset by loathsome evils.
Appendix I 269 7:134 And whenever a plague struck them. 27:82 when the word [of truth] stands revealed against them. 27:85 And the word [of truth] will stand revealed against them.
W-Q-F WUQIFŪ + ʿALÁ—(I)—(perfect passive)—(agent of the passive: masculine, plural)—to be made to stand. 6:27
when they will be made to stand before the fire.
6:30
when they shall be made to stand before their Sustainer.
W-L-Y WALLĀHUM + ʿAN—(II)—(perfect)—(agent: masculine, singular)—(direct object: masculine, plural)—to turn away. 2:142 What has turned them away from the direction of prayer?
Y-S-R NUYASSIRUKA + LĀM—(II)—(imperfect)—(agent: masculine, plural)— (direct object: masculine, singular)—to make something easy. 87:8
shall We make easy for thee the path towards [ultimate] ease.
SA-NUYASSIRUHU + LI—(II)—(future)—(agent: masculine, plural)— (direct object: masculine, singular)—to make something easy. 92:7
for him shall We make easy the path towards [ultimate] ease.
92:10 for him shall We make easy the path towards hardship.
Appendix II Frequencies
Table 1 Frequencies of the root (ʾ-T-Y) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total idiomatic verbs Total nonidiomatic verbs Total nonidiomatic active participles Total nonidiomatic passive participles Total nonidiomatic verbal nouns
549 111 36 95 292 10 2 3
Table 2 Frequencies of the root (ʾ-KH-DH) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total idiomatic verbs Total nonidiomatic verbs Total idiomatic verbal nouns Total nonidiomatic verbal nouns Total idiomatic active participles Total nonidiomatic active participles
273 20 58 63 119 5 2 1 5
Table 3 Frequencies of the root (B-SH-R) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic verbs Total nonidiomatic active participles Total nonidiomatic verbal nouns Total nonidiomatic proper nouns
123 9 20 19 19 19 37
Appendix II 271 Table 4 Frequencies the root (Ḥ-W-Ṭ) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total idiomatic active participles
28 17 0 11
Table 5 Frequencies of the root (KH-T-M) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic verbal nouns Total nonidiomatic active participles Total nonidiomatic passive participles
8 5 0 1 1 1
Table 6 Frequencies of the root (D-KH-L) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic verbs Total nonidiomatic active participles Total nonidiomatic passive participles Total idiomatic verbal nouns
126 2 37 79 2 4 2
Table 7 Frequencies the root (DH-H-B) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total idiomatic verbs Total nonidiomatic verbs Total idiomatic verbal nouns Total nonidiomatic active participles Total nonidiomatic proper nouns
57 14 11 10 12 1 1 8
Table 8 Frequencies of the root (R-B-Ṭ) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic verbs Total idiomatic verbal nouns
5 3 0 1 1
272
Appendix II
Table 9 Frequencies of the root (R-GH-B) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic verbs Total nonidiomatic verbal nouns Total idiomatic active participles
8 3 0 1 1 3
Table 10 Frequencies of the root (S-Q-Ṭ) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic active participles
8 1 6 1
Table 11 Frequencies of the root (S-L-KH) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total idiomatic verbs
3 2 0 1
Table 12 Frequencies of the root (S-W-Y) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total idiomatic verbs Total nonidiomatic verbs Total nonidiomatic verbal nouns Total nonidiomatic active participles
83 15 2 2 31 28 5
Table 13 Frequencies the root (Ḍ-R-B) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total idiomatic verbs Total nonidiomatic verbs Total nonidiomatic verbal nouns
58 34 8 10 3 3
Table 14 Frequencies of the root (Ṭ-B-ʿ) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs
11 11 0
Appendix II 273 Table 15 Frequencies of the root (Ṭ-ʿ- N) and its derivatives in the Qur’an Total occurrences Total QIPVs Total idiomatic verbal nouns
2 1 1
Table 16 Frequencies of the root (GH-W-TH) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic verbs
5 1 0 4
Table 17 Frequencies of the root (F-R-GH) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic verbs Total nonidiomatic active participles
6 3 1 1 1
Table 18 Frequencies of the root (F-Ḍ-Y) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs
1 1 0
Table 19 Frequencies of the root (Q-DH-F) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic verbs
9 5 3 1
Table 20 Frequencies of the root (Q-L-B) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total idiomatic verbs Total idiomatic verbal nouns Total idiomatic active participles Total idiomatic passive participles Total nonidiomatic verbs Total nonidiomatic proper nouns
168 14 1 9 5 3 3 1 132
274
Appendix II
Table 21 Frequencies of the root (K-W-R) and its derivatives in the Qur’an Total occurrences Total QIPVs Total idiomatic verbs
3 2 1
Table 22 Frequencies of the root (M-T-ʿ) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic verbs Total nonidiomatic proper nouns
70 1 12 22 35
Table 23 Frequencies of the root (W-S-M) and its derivatives in the Qur’an Total occurrences Total QIPVs Total nonidiomatic QIPVs Total nonidiomatic active participles
2 1 0 1
Index
Note: Page numbers in italics indicate a figure and page numbers in bold indicate a table on the corresponding page. Abdul-Raof, H. 15–17, 190 Abu Saʿad 37 adequacy/inadequacy 111–113 Adonis, A. S. 39 adverbs of time 49 Agency of Revelation 13 Ahmadi Qadyani doctrine 178 al-Akhfash 46 Alī, Y. 120, 124, 126, 135, 137, 138–139, 142, 144, 147, 149, 150, 153, 158, 161, 163, 167–168, 169, 180, 187 Ali, A. Y. 19–21, 23, 132, 174–175 allegory 37, 125 annotation 11 antonymy 37 Arabic idiomatic phrasal verbs (AIPVs): components 41–51; establishing a parameter for 35–41; register variations 60–61; syntactic and semantic properties 51–60 Arabic structure and idioms 25 Arabs of Madinah 151 Arberry, A. J. 13, 121, 122, 126, 132, 135, 137, 138, 142, 144, 147, 148, 153, 158, 161, 163, 168, 171, 175, 176, 177, 180 Asad, M. 21–22, 121, 123, 126, 133, 135, 137, 139, 142, 144–145, 148, 154, 158, 164, 168, 171, 179, 180–181 al-ʿaṭiyah, A. 35 Austin, J. L. 83–84, 101 al-Awwal, R. 151 Badawi, E. M. 119 Baker, M. 36, 97, 100
Bakhtiar, L. 29, 122, 124, 126, 133–134, 135, 138, 142, 144, 148, 149, 154, 155–156, 159, 161, 164–165, 169, 171, 183, 184 Banū al-Naḍīr 150–151 al-Baṣrah school 46, 59 Bassnett, S. 10, 16 Beekman, J. 97 Behbudi, M. B. 25 Bell, R. 191 Bibliography of the Translations of the Meanings of the Glorious Qur’an into English 1649–2002: A Critical Study (Kidwai) 14 Bounteous Koran: A Translation of Meaning and Commentary, The 28 Callow, J. 97 Catford, J. 9–10, 18, 100, 113 Chomsky, N. 8 Christian religion 140 code linguistics 98 communicative competence 166, 191 communicative meaning 84, 92, 111 connotation 42–43 connotative meaning 112–113, 129, 139, 144, 159, 166, 167, 169, 171, 173, 174 consistency in translation 26, 106, 184 constative utterances 84 context in translation 26 contextual significances 53, 62, 161, 167 contrastive linguistics 97–99 conversational implicatures 85–86, 90–92, 94, 108, 110–111 conversational maxims 85–86, 110
276
Index
Cooperative Principle 85–86 couplets technique 189 Cowie, A. P. 40 Cultural Aspects in Qur’an Translation (Abdul-Raof) 16 cultural (un)translatability 9–10 dalāʾil al-iʿjāz (al-Jurjānī) 55 Dastjerdi, H. V. 17 Daud, N. 36 Dawood, N. J. 26, 27–28 deconstructionist approach 9 denotative meaning 111–113, 123–124, 129, 139, 144, 169, 171, 173, 174, 177, 184 De Pedro, R. 190 Dickins, J. 55, 112–113 Dictionary of English Grammar (Trask) 36 directional prepositions 49, 62 direct speech acts 90, 91 disbelief 94, 147–148 Doherty, S. 100 dual verbs 44 Dāwood, M. 35, 66
177, 178, 179, 181, 182, 183, 188, 190 Functional-Pragmatic Model of Translation Evaluation (House) 99 gender and verbs 44 Ghazala, H. 36, 37 Gleason, H. A. 97 Grice, H. P. 84–85, 87–88, 91–92, 111 Gutt, E. A. 188
Egyptian colloquial 61 EIPV (English idiomatic phrasal verb) 1, 123, 145, 160, 161, 165, 176 El-Zeiny, I. 176–177, 187 English interrogative particle 159 English phrasal verbs 81 Enkvist, N. E. 97 exegesis 13, 25, 28 expressive meaning 171, 177
Hadith tradition 129 Haleem, A. 25–27, 119, 122, 124, 126, 129, 133, 135, 137, 138, 142, 144, 148, 154, 155, 164, 168, 171, 181–182 Halliday, M. 87, 101 Ḥammād 60–61 al-Harawī 46 Hatim, B. 8, 100 Heliel, M. H. 35 Hickey, L. 100 al-Hilālī 121, 122, 123, 126, 132–133, 135, 137, 138, 142, 144, 145, 147, 149, 150, 153–154, 158, 161, 164, 168, 178, 179 Hoey, M. 99 Holy Koran: An Introduction with Selections, The (Arberry) 23 homonymy 37 Houghton, D. 99 House, J. 99–101, 102, 110, 113, 185 How to Do Things with Words (Austin) 83 Hudhayl language 46 ḥurūfu al-iḍāfati 46
al-Fakhr al-Rāzī 28 al-Fārisī 46 fatwas (Islamic opinions given by qualified scholars) 17 Fawcett, P. 110 figurative meaning 108–109, 114, 161, 166 figurative QIPVs 82–83, 176 figurative vs metaphorical QIPVs 156–157 al-Fiʿilu f ī al-Qurʿani al-Karīmi Taʿadyatuhu wa-Luzūmuh (al-Shamsān) 66 Firth, J. R. 55 France, translatability in 9 functional-pragmatic equivalence 99, 144, 146, 148, 159, 172–173, 175,
Ibn al-Jawzī 28 Ibnʿaqīl 46 Ibn Hishām 46, 48 Ibn Jinn 59 Ibn Kathīr 28, 131 Ibn Mālik 46 Ibn Saʿd 28 ibn Ubbay, A. 151 idiomaticity 36–37, 39–41, 52–55, 59, 86 idioms: and Arabic structures 25; defined 36; influence of linguistic phenomena on 37–38; structural patterns 37 illocutionary acts 84, 89, 91 illocutionary force 101, 110, 111, 120, 125, 127, 128, 131, 136, 141, 143, 144, 145, 146, 147, 150, 152, 156,
Index 170, 173, 176–179, 177, 181–184, 186–187 illustrative instances 118–192 imperative verbs 44 indicative verbs 44 indirect speech acts 91–94 informative level of meaning 145, 173, 177, 179, 183, 184–185 interlinear translations 182 interpersonal function 101–102 interrogative/ affirmative structure 167 interrogative particle 49, 70, 157, 159, 166–167 intertextuality 26 intralingual (un)translatability 11 intransitive verb 56–57, 75 Iqbal, M. 18 Irving, T. B. 28 al-Jalālayn 28 James, C. 97–98, 99, 106 Jamshidian, E. 17 al-Jurjānī, A. Q. 55 jussive verbs 44 Khan, M. Z. 29, 121, 123, 126, 132, 135, 137, 138, 142, 144, 145, 147, 149, 153, 158, 161, 163, 168, 171, 177–178, 186 Khatib, M. M. 28 Khān 121, 122, 123, 126, 132–133, 135, 137, 138, 142, 144, 145, 147, 149, 150, 153–154, 158, 161, 164, 168, 178, 179 Kidwai, A. R. 14, 174, 175 al-Kisāʾ ī 46 Koller, W. 113 Koran, The 27 Koran Interpreted, The (Arberry) 13, 22–23, 103–104 al-Kūfah school 46, 59 Larson, M. L. 187 Lattey, E. 36 Leech, G. 87, 88 Lefevere, A. A. 16 Lentzner, K. R. 35 limitation in translation, von Denffer on 14 linguistically oriented approach 100 Linguistics Dictionary: English-Arabic Arabic-English (Daud) 36 linguistic (un)translatability 9–10
277
Lisān al-ʿArab (Ibn Manẓūr) 27 literal translation 15, 19, 20, 26, 30, 103, 174, 175, 181, 185–186 locative prepositions 49 locutionary acts 84, 111 locutionary/semantic aspects of meaning 110 Longman Dictionary of Applied Linguistics (Richards) 36 Loveday, L. 87–88 Mackin, R. 40 Macquarie Dictionary (Yallor) 36 macro-linguistics 98 al-Maḥallī 151, 162 Maḥfūẓ, N. 39 Ma‘ānīal-Qur’ān (Behbudi) 25 Mason, I. 100 Meaning of the Glorious Qur’an: Text and Explanatory Translation, The (Pickthall) 13, 19, 103 Meaning of the Holy Qur’an, The (Ali) 19, 103 mentalist views for translation quality assessment 100 Message of the Qur’an, The (Asad) 21, 104–105 metaphor 37, 43, 78, 81–82, 111, 119–120, 122, 125, 126, 130, 135, 136, 138, 140, 141, 144, 146, 152, 156–157, 173, 175, 176, 177, 183, 188 micro-linguistics 98 mirror-paraphrasing 182 modal auxiliary verb 160–161 Model for Translation Quality Assessment, A (House) 99, 101 modifier 169, 171–172 Mosteghanemi, A. 39 al-Mubarrid 46 al-Mufaṣṣal (al-Zamakhsharī) 46 Mughnī al-Labīb (Ibn Hishām) 46 al-Murādī 46 al-Muṣḥaf al-Mufassar 28 Munday, J. 8, 99 al-Nasaf ī 28 Newmark, P. 8, 78, 81, 111, 112–113, 186, 187, 189 Nida, E. 78, 113 Noble Qur’an: English Translation of the Meanings and Commentary, The (Taqī-ud-Dīn & Khān) 104
278
Index
Noble Qur’an: The First American Translation and Commentary, The 28 nonrestricted verbs 45 nonsplit QIPVs 70–72 nouns 47, 140, 146, 156 over-translation 112, 161, 171, 186 pagan Quraysh 146 Palmer, F. R. 36 paraphrasing 171, 175, 182 past tense verbs 135, 140, 146 performative utterances 84 perlocutionary acts 84, 110, 186, 201 Pickthall, M. M. 13, 19, 23–24, 120, 123, 126, 129, 131, 135, 136, 138, 141, 144, 145, 147, 149, 152–153, 158, 159, 162–163, 168, 169, 173–174 plural verbs 44 politeness 88–89 PP theory 88–89 pragmatic idiosyncrasies 83–94 pragmatic (un)translatability 10 prepositions 34, 46–51, 69, 75, 119, 124–125, 130, 135–136, 139, 140, 141, 146, 151–152; see also verbpreposition combinations preposition substitution 51, 57–60 pronouns 25 proper verbs, Arabic 41–46 proverbs 38 pun 37 punctuation 27 al-Qāmūs al-Muḥīṭ (al-Fayrūzabādī) 27 Qadyani translation 104 Qarāʾī 122, 123, 126–127, 133, 135, 138, 142, 144, 145, 148, 154, 159, 160, 164, 169, 171, 182–183 Qaṭru al-nadá wa- ballu al-ṣadá (Ibn Hishām) 48 QIPVs see Qur’anic idiomatic phrasal verbs (QIPVs) Qur’an: A Contemporary Translation, The (Ali) 23 al-Qurʿan al-Kar īm wa-Tafāʿulu al-Maʿān ī: Dirāsatun Dilāliyatun li-Taʿlluqi Ḥarfi al-Jarri bil-Fiʿili wa-ʾAtharihi f ī al-Maʿná f ī al-Qurʿani al-Kar īm (Dāwood) 35, 66 Qur’an: A New Interpretation, The (Turner) 24–25
Qur’an: An Introduction, The (Saeed) 17 Qur’an: English Translation and Parallel Arabic Text, The (Haleem) 25–26, 105 Quran: The Eternal Revelation Vouchsafed to Muhammad the Seal of the Prophets, The (Khan) 104 Qur’an: Translation, The (Shakir) 104 Qur’anic discourse 130, 136, 140, 152, 161, 166 Qur’anic idiomatic phrasal verbs (QIPVs) 66–94; pragmatic idiosyncrasies of 83–94; semantic idiosyncrasies of 76–78, 81–83; syntactic idiosyncrasies of 68–70, 72, 74–76 Qur’anic verses 20, 92–94 Qur’an’s (un)translatability 12–31; English translators on 18–31; Muslim intellectuals on 13–18; overview 8–12 Qur’an Translation: Discourse, Texture and Exegesis (Abdul-Raof) 15–16, 68 Qur’an Translation, The 29 Qur’an with a Phrase-by-Phrase English Translation, The (Qarāʾī) 105 al-Qurṭub 28 Rahman, F. 13 al-Rāzī 162 referential (un)translatability 10 Reiss, K. 100 rephrasing strategy 169 research methodology 97–115; data collection 103–106; model for analysing and assessing QIPV translations 114; research procedure 107; research questions and data analysis 106–107; theoretical framework 97–102; working model for analysing and assessing QIPV translations 107–113 response-based approach for translation quality assessment 100 restricted verbs 45 Richards, J. 36 Robinson, N. 174 al-Rummānī 46, 50 sacred text, defined 189 Saeed, A. 17 Sapir, E. 8–9 Sapir–Whorf hypothesis 8 Searle, J. R. 83, 86, 89, 91, 101
Index semantic affect of Arabic prepositions on Arabic verbs 52–53 semantically oriented structures 167, 189 Semantic and Syntactic Aspects of Arabic Prepositions (Lentzner) 35 semantic aspect of ‘meaning’ 101 semantic characteristics of Arabic prepositions 49–50 semantic deviation 112, 186 semantic idiosyncrasies 76–78, 81–83 semantico-syntactic organic interrelations 189 semantic relationships 10 semiotic relationships 10 Shaker 166, 177 Shakir, M. H. 29, 121, 123, 126, 132, 135, 137, 142, 144, 147, 148, 150, 153, 158, 159, 163, 168, 171, 177 Al-Shamsān 50, 66 Sibawayh, A. Q. 51, 58 sign relationships 10 simile 37 singular verbs 44 situational dimensions 102 Snell-Hornby, M. 113 social correctness 89 social environment 100 sociosemiotic perspective 10–11 source-language oriented texts 11 source-text marked syntactic structures 181 source-text-oriented approach 126, 155, 169, 174, 176, 179, 183–184, 187, 189 speech acts 110, 157–158, 159, 162, 168 Speech Acts: An Essay in the Philosophy of Language (Searle) 83 speech act theory 83–84, 101 Sperber, D. 188 split QIPVs 72, 74–76 sūrah 19, 23, 30, 103, 166–167, 175 al-sīrah al-nabawīyah 30 subjunctive verbs 44 Sublime Quran, The (Bakhtiar) 29, 105–106 Sunna ‘prophetic tradition’ 12 al-Suyūṭī 151, 162 synonymy 37 syntactically oriented structures 189 syntactic idiosyncrasies 68–70, 72, 74–76 syntactic relationships 10, 52, 176
279
al-Ṭabarī, 28 Taber, C. 113 Tancock, L. W. 17 target-language-oriented texts 11 target-text-oriented approach 187 tenses and verbs 44 tertium comparationis 99, 107, 111, 118 – 120, 124–126, 129–131, 135–136, 140–141, 146–147, 150–152, 156–158, 161–162, 166–168 textual aspect of meaning 101 Toury, G. 113 transformational generative grammar (TGG) 8 transitive verbs 52, 56–57, 75 translatability 8, 9, 11, 12, 15, 17, 24, 31 ‘Translating the Untranslatable: A Survey of English Translations of the Quran’ (Kidwai) 14 Translation of the Glorious Qur’an 28 translation quality assessment 100, 102 translators’ performance: and functionalpragmatic equivalence 172; and QIPV 123, 128, 134, 139, 143, 149, 155, 160, 165, 170 Trask, R. L. 36 Turner, C. 24–25 ‘Ulūm al-Qur’an: An Introduction to the Sciences of the Qur’an (von Denffer) 13 under-translation 112, 186 untranslatability see Qur’an’s (un)translatability Uqayl language 46 verb(s): classes of 43–44; dual 44; English phrasal 81; feminine/ masculine 44; imperative 44; implicative 57–60, 59; indicative 44; inflected 42; intransitive 56–57; invariable 42; jussive 44; modal auxiliary verb 160–161; mood 44, 119, 125; nonrestricted 45; nouns 140, 146, 156; past tense 135; plural 44; syntactic relationships with prepositions 52; transitivity 56–57 Verb-Particle Combination in English, The (Fraser) 36 Verb-Particle Combinations in English and Arabic: Problems for Arab Lexicographers and Translators (Heliel) 35
280
Index
verb-preposition combinations 34–35, 38–41, 45–46, 52, 67, 75, 136, 140, 151–152 verb-preposition relation, attachment, or dependency 55–56 von Denffer, A. 13–14
Wonderly, W. L. 187 word order 69–70, 72, 74–76, 169, 171, 174, 179
al-Wahīdī 28 Wajd, M. F. 28 Whorf, B. Lee 8 Wilson, D. 188 Wilss, W. 100
Zidan, A. 28 Zidan, D. 28 al-Zajjājī 46 al-Zamakhsharī 28, 46, 93, 130, 162
Yallor, C. 36 Yebra, G. 97