arabic - a linguistic introduction

C:/ITOOLS/WMS/CUP-NEW/4460944/WORKINGFOLDER/RARA/9781107023314HTL.3D i [1–2] 24.10.2013 9:30AM

Arabic

This lively introduction to the linguistics of Arabic provides students with aconcise overview of the language’s structure and its various components: itsphonology, morphology, and syntax. Through exercises, discussion points andassignments built into every chapter, the book presents the Arabic language invivid and engaging terms, encouraging students to grasp the complexity of itslinguistic situation. It presents key linguistic concepts and theories related toArabic in a coherent way, helping to build students’ analytical and critical skills.Key features:

� study questions, exercises, and discussion points in every chapter encouragestudents to engage with the material and undertake specific assignments;

� suggestions for further reading in every chapter allow readers to engage inmore extensive research on relevant topics; and

� technical terminology is explained in a helpful glossary.

karin c . ryd ing is Sultan Qaboos bin Said Professor Emerita of Arabiclinguistics at Georgetown University, where she taught Arabic linguistics forover twenty years.

C:/ITOOLS/WMS/CUP-NEW/4460944/WORKINGFOLDER/RARA/9781107023314HTL.3D ii [1–2] 24.10.2013 9:30AM

C:/ITOOLS/WMS/CUP-NEW/4457415/WORKINGFOLDER/RARA/9781107023314TTL.3D iii [3–3] 23.10.2013 2:31PM

ArabicA Linguistic Introduction

KARIN C. RYDING

C:/ITOOLS/WMS/CUP-NEW/4465676/WORKINGFOLDER/RARA/9781107023314IMP.3D iv [4–4] 25.10.2013 5:26PM

University Printing House, Cambridge CB2 8BS, United Kingdom

Published in the United States of America by Cambridge University Press, New York

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit ofeducation, learning and research at the highest international levels of excellence.

www.cambridge.orgInformation on this title: www.cambridge.org/9781107606944

© Karin C. Ryding 2014

This publication is in copyright. Subject to statutory exceptionand to the provisions of relevant collective licensing agreements,no reproduction of any part may take place without the writtenpermission of Cambridge University Press.

First published 2014

Printed in the United Kingdom by MPG Printgroup Ltd, Cambridge

A catalogue record for this publication is available from the British Library

Library of Congress Cataloguing in Publication data

ISBN 978-1-107-02331-4 HardbackISBN 978-1-107-60694-4 Paperback

Cambridge University Press has no responsibility for the persistence or accuracy ofURLs for external or third-party internet websites referred to in this publication,and does not guarantee that any content on such websites is, or will remain,accurate or appropriate.

http://www.cambridge.org

http://www.cambridge.org/9781107606944

C:/ITOOLS/WMS/CUP-NEW/4460944/WORKINGFOLDER/RARA/9781107023314TOC.3D v [5–6] 24.10.2013 9:26AM

Contents

Preface page viiAcknowledgments ixAbbreviations and symbols used in this book xi

1 Arabic linguistics: overview and history 1

2 Arabic phonology 13

3 Arabic phonotactics and morphophonology 23

4 Arabic syllable structure and stress 33

5 Introduction to Arabic morphology 41

6 Derivational morphology: the root/pattern system 55

7 Non-root/pattern morphology and the Arabic lexicon 79

8 Arabic inflectional morphology 89

9 Syntactic analysis and Arabic 107

10 Arabic syntax I: phrase structure 119

11 Arabic syntax II: clause structure 127

AppendicesA Fields of linguistics and Arabic 141B Arabic transcription/transliteration/romanization 145C Arabic nominal declensions 149

Glossary of technical terms 157References 167Index 181

v

C:/ITOOLS/WMS/CUP-NEW/4460944/WORKINGFOLDER/RARA/9781107023314TOC.3D vi [5–6] 24.10.2013 9:26AM

C:/ITOOLS/WMS/CUP-NEW/4460944/WORKINGFOLDER/RARA/9781107023314PRF.3D vii [7–8] 24.10.2013 9:33AM

Preface

Despite widening international interest in Arabic language and culture,few resources exist for a systematic introduction to the linguistics of Arabic and forteaching the basics of phonology, morphology, syntax, semantics. This is truedespite the fact that distinguished works exist in Arabic, English, and otherlanguages examining and documenting Arabic language history, structure, andprocesses. Works by Aoun, Badawi, Bateson, Beeston, Bohas, Carter, Eid, Holes,Owens, Parkinson, Stetkevych, Talmon, Versteegh and others have contributedvastly to understanding the linguistics of Arabic. However, there is a place for anorganized overview, both as a reference tool and as a foundational textbook forlearning about the field.For teaching courses on Arabic linguistics, I have used books and articles by all

the above-mentioned authors. In particular, I have found that Bateson’s ArabicLanguage Handbook, Beeston’s The Arabic Language Today, and Stetkevych’sTheModern Arabic Literary Language useful for concise summaries of key topics.These books originally date from 1967 (Bateson) and 1970 (Beeston andStetkevych). Holes’ Modern Arabic (2004) is a more modern and comprehensiveapproach, but I have found that it is less useful as a textbook than as a referencework, and I usually assign only certain parts of it. Versteegh’s book The ArabicLanguage (1997), provides historical background for key developments in theArabic language but does not analyze the actual linguistic structures and processesof contemporary modern standard Arabic (MSA). Owens’ many excellent workson the history of Arabic and of Arabic grammatical theory are focused primarily onpremodern developments. Thus none of these books – despite their many merits –forms by itself a framework for a course in contemporary Arabic linguistics, andthere is a distinct need for a more pedagogically focused work that includesdiscussion topics, questions, and suggestions for further readings on specificsubjects. This book aims to meet the challenges of teaching elements of Arabiclinguistics to students and teachers-in-training who may know little about linguis-tic theory, and for classes where there are mixed levels of ability in the languageand in academic background.

vii

C:/ITOOLS/WMS/CUP-NEW/4460944/WORKINGFOLDER/RARA/9781107023314PRF.3D viii [7–8] 24.10.2013 9:33AM

In 2005, I published AReference Grammar ofModern Standard Arabic based ona corpus of data from contemporary Arabic newspapers and other types of expos-itory prose. It was intended for audiences of Arabic learners and teachers, as well asthose with a general interest in the grammatical features of the written language.The present book is a more technical introduction to the structures and processesthat characterize Arabic linguistics, aiming to gather in one place current scholarlyresources and theories for study and further research. It has emerged and beendistilled from the content of graduate courses that I have taught at GeorgetownUniversity during the past thirty years. Rather than adopt one particular theoreticalstance, I have chosen to be as objectively descriptive as possible, introducingtheories of varying levels of formality and indicating where readers may want topursue further reading on particular topics. Due to length limitations, I have had toomit a considerable amount of interesting and relevant research; likewise I haveomitted extended descriptions of grammatical structures because this is not agrammar of Arabic, but an introduction to linguistics as applied to Arabic.

A key factor motivating the writing of this book is the need for more extensiveprofessional resources for teachers of Arabic as a foreign language, especially withthe steady demand for knowledge of Arabic language both as a professional skilland as a discipline within the fields of humanities and social sciences. Teachingpractical knowledge of Arabic relies on the sophistication and depth of under-standing that teachers bring to their classes – understanding not only the rules oflanguage structure, but the theoretical underpinnings of the language, its intellec-tual and scholarly heritage, and the ways in which its grammatical system can beelegantly and efficiently portrayed.

This book may serve as a text in courses on Arabic language and linguistics, orcourses on Arabic pedagogy, or it may serve to give non-specialists a generalpicture of linguistic issues in MSA. In designing this book, I have assumed anaudience with some knowledge of the Arabic language, but little knowledge oftechnical linguistic terms, theories, or approaches. There are bound to be those whowill find shortcomings and gaps in this overview, and I assume total responsibilityfor any errors or deficiencies. I hope that this book will constitute a useful first stepin conveying the enormous wealth of meaningful data, methods of linguisticresearch, and critical insights into language systems that have made progressthrough close analysis of Arabic language structures and processes.

viii Preface

C:/ITOOLS/WMS/CUP-NEW/4464129/WORKINGFOLDER/RARA/9781107023314ACK.3D ix [9–10] 25.10.2013 9:38AM

Acknowledgments

I would like to thank and acknowledge the following people, past andpresent, who inspired me, helped me form my ideas for this book, and assisted metoward its completion:

my Georgetown professors Wallace M. Erwin and Michael Zarechnak,who led me to and through Arabic linguistics;

my students, who challenged and stimulated my thinking on points ofArabic linguistics;

my colleagues at Georgetown, outstanding Arabists and linguists;at Cambridge University Press, Andrew Winnard, for his unfailing

support and encouragement every step of the way, and to HelenaDowson for her attentive and patient help in finalizing the manuscript;

my husband, Victor Litwinski, for being a vital interlocutor on all thingslinguistic, and for his unstinting professional and emotional support;and

His Majesty Sultan Qaboos bin Said, Sultan of Oman, whose patronageand encouragement of Arabic language study has been a great boon tothe development of Arabic linguistics, transcultural communication,and intercultural understanding.

ix

C:/ITOOLS/WMS/CUP-NEW/4464129/WORKINGFOLDER/RARA/9781107023314ACK.3D x [9–10] 25.10.2013 9:38AM

C:/ITOOLS/WMS/CUP-NEW/4460944/WORKINGFOLDER/RARA/9781107023314LOA.3D xi [11–12] 24.10.2013 9:37AM

Abbreviations and symbols used in this book

Additional abbreviations used specifically in syntactic theory are listed atgreater length in Chapter 9.

acc. accusativeadj. adjectiveadv. adverbAP active participleC any consonant (phonology); complement, complementizer

(syntax)dat. dativedef. definitedu. dualEALL Encyclopedia of Arabic Language and LinguisticsESA educated spoken Arabicf./fem. femininefut. futuregen. genitiveIC immediate constituentimp. imperfectindef. indefiniteindic. indicativem./masc. masculineMSA Modern Standard ArabicN nounno. numbernom. nominativeNP noun phraseO objectpl. pluralPP passive participlepron. pronoun

xi

C:/ITOOLS/WMS/CUP-NEW/4460944/WORKINGFOLDER/RARA/9781107023314LOA.3D xii [11–12] 24.10.2013 9:37AM

S subject; sentencesing. singularsubj. subject; subjunctiveUG universal grammarV any short vowel (phonology); verb (syntax)vd voicedvls voicelessVN verbal nounVP verb phraseVV any long vowelWFR word formation rule# word boundary- morpheme boundary{ } encloses morpheme/ / encloses phonemic transcription[ ] encloses phonetic transcription“ ” encloses glosses* indicates a hypothetical or nonstandard form~ ‘alternates with; or’small caps indicate morphemic structure

xii Abbreviations and symbols

C:/ITOOLS/WMS/CUP-NEW/4466921/WORKINGFOLDER/RARA/9781107023314C01.3D 1 [1–12] 26.10.2013 10:12AM

1

Arabic linguistics: overview and history

1. Introduction

In approaching the study of human language in general, if the aim is tocategorize, classify, and identify how languages work, then these functions must bebased on clearly documented empirical observations. This kind of activity sepa-rates linguistics from anecdotal, philosophical, impressionistic, or speculativeobservations about language that may come from anyone anywhere. Linguisticscan be defined as follows.

(1) Linguistics is “the study of language as a system of human communica-tion” (Richards and Schmidt 2010: 343).

(2) Linguistics is “a natural science, on a par with geology, biology, physics,and chemistry.” And “the task of linguistics is to explain the nature ofhuman language, through active involvement in the description of lan-guage – each viewed as an integrated system – together with explanationof why each language is the way it is, allied to the further scientificpursuits of prediction and evaluation” (Dixon 2010: 1).

(3) “For the beginning linguist, saying that linguistics is a science can beinterpreted as implying careful observation of the relevant real-worldphenomena, classification of those phenomena, and the search for usefulpatterns in the phenomena observed and classified. For the moreadvanced linguist, saying that linguistics is a science is a matter of seekingexplanations for the phenomena of language and building theories whichwill help explain why observed phenomena occur while phenomenawhich are not observed should not occur” (Bauer 2007: 17).

(4) “Linguists believe that their field is a science because they sharethe goals of scientific inquiry, which is objective (or more properlyintersubjectively accessible) understanding” (Aronoff and Rees-Miller2001: xiv).

(5) “The task of linguistics is to explain the nature of human language,through active involvement in the description of languages – each viewedas an integrated system – together with explanation of why each language

1


is the way it is, allied to the further scientific pursuits of prediction andevaluation” (Dixon 2010a: 1).

The field of linguistics is therefore seen as a scientific approach to language in allits diversity: spoken and written, formal and informal, internal and external. Itconcerns the analysis of language in use (such as conversation analysis), languageas a universal form of human cognition (e.g., universal grammar), theories oflanguage structure, and language acquisition in its various forms. Linguistics isdescriptive rather than prescriptive; it aims to document and explain language as itis, rather than to prescribe rules of performance.1

1.1. Linguistics and grammar

It is important to distinguish the realm of linguistics from the moresubordinate concept of ‘grammar.’ In fact, it is important to delineate exactlywhat ‘grammar’ denotes. Usually, the term ‘grammar’ refers to the study of bothmorphology and syntax: word structure and clause structure. Because morphol-ogy and syntax often interact, a core component of grammar is morphosyntax.One definition of grammar states that “a grammar consists of a number of closedsystems – categories such as tense, gender, and evidentiality – and a number ofconstruction types, or ways of relating together words into phrases, clauses,sentences, and utterances” (Dixon 2010a: 23). A linguist’s way of looking atgrammar is as a “descriptive”mechanism that accounts for all the morphologicaland syntactic phenomena in a language.2 A more didactic view of grammar is“prescriptive,” i.e., a grammar indicates what is correct and incorrect usage. Theformer takes language as it is and describes it; the latter takes an idealizedstandard of language and provides rules for adherence to that standard. Bothare useful in terms of language pedagogy, but it is important to know thatlinguists rarely see language in black and white – correct or incorrect; rather,they view language as a feature of human cognition and behavior, and try tocharacterize that behavior (or cognition, as it may be) as accurately and empiri-cally as possible.

2. Linguistics and Arabic

Arabic linguistics is a vast field combining study of the Arabic languagewith the analytical disciplines that constitute the field of linguistics. Linguistictheories, methods, and concepts are used to analyze the structure and processes ofArabic; but at the same time, Arabic with its millennium-long intellectual

2 Arabic linguistics: overview and history


traditions, its complex morphology, and its current broad diversity of registers,informs linguistic theory. Many linguistic approaches to Arabic language anal-ysis have been applied over the past fifty years both within the Arab world andfrom the point of view of western scholars. These approaches and their discipli-nary procedures are both varied and convergent, covering a wealth of data butalso coming to terms with central issues of concern to Arabic linguistics that hadbeen neglected in the past, such as validating the prominent role of vernacularArabic and variation theory in Arabic society and culture. Arabic linguistics isnow an active subfield in sociolinguistics, corpus linguistics, and computationallinguistics as well as theoretical and applied linguistics. Both traditional and newgenres of Arabic writing are now being examined within postmodern frameworksof literary theory and linguistic analysis. Media Arabic studies is a new andrapidly growing field; medieval texts are being re-examined in the light of newphilology and discourse analysis; previously ignored forms of popular culturesuch as songs, advertisements, oral poetry, vernacular writing, letters, email, andblogs are now legitimate grist for the linguistics mill.The discipline of linguistics has a growing number of subfields. The traditional

four core divisions usually include theoretical linguistics, applied linguistics,sociolinguistics, and computational linguistics. Each of these has developed newapplications, perspectives, hypotheses, and discoveries that extend their analyticalpower in novel ways, such as cognitive linguistics in theoretical linguistics, secondlanguage acquisition in applied linguistics, corpus linguistics in the computationalfield, and discourse analysis in sociolinguistics. When these perspectives andtheories are applied to Arabic, the findings can be revealing, satisfying, or puz-zling, but generally lead toward greater understanding of how languages work,how they resemble each other, and how they differ. The field of computationallinguistics has provided ways to develop extensive corpora of spoken and writtenArabic that can be used for pioneering research and analysis of language in use. Anactive subfield of linguistics – history of linguistics – examines linguistic histor-iography, the development of language analysis over time, and the evolution ofgrammatical theory in different cultures.The phonological, morphological, and syntactic structures of Arabic reflect its

Semitic origins and its essential differences from Indo-European languages. Thesedifferences and their cultural embeddedness are what make Arabic of interest toresearch in many fields of linguistics. For example, the particularly well-definedand elaborated verb system and its derivations reflect an aspect of Classical Arabicthat is both fascinating and rigorous in its structure and linguistic logic.3 As anotherexample, the contrasts between vernacular and written language, their differentroles within Arab society, and the tensions between local and regional linguisticidentities, form areas of sociolinguistics that pose particular challenges to data

Linguistics and Arabic 3


collection, empirical study, and objective analysis. Many research challenges andopportunities still lie ahead in this regard.4

2.1. Theoretical linguistics

In a very real sense theoretical linguistics is the mother of all branches oflinguistic science and it is often referred to as “general linguistics” because of itswide range of coverage.

Prior to the emergence of the field of modern linguistics, philology was the termused for the study of language structure and literary tradition, with special focus onhistorical developments and relationships among cognate languages (comparativephilology). The examination and analysis of language families, their relationshipsand development is referred to as diachronic analysis (analysis of language struc-ture and growth over time).5

2.1.1. Background

The nineteenth century witnessed a shift in perspective away from dia-chronic analysis to synchronic analysis; that is, the examination of language as it isat any point in time, especially contemporary language. In pinning down languageas an object of study, one of the first steps of early linguists was to establish itssystematic nature and the difference between abstract language-as-a-system andconcrete language-in-use (see Swiss linguist Ferdinand de Saussure’s distinctionbetween langue and parole).6 It was language-as-a-system that early twentieth-century structural linguists such as Jespersen, Sapir, and Bloomfield believedwould yield the most fruitful research results because it was an objective realitymeasurable in reliable, empirical ways. Central to the structural linguistic approachis the difference between descriptive grammar and prescriptive grammar, the ideathat theories focus on discovering and describing the structures and processes oflanguage as it is, rather than on placing particular values on the type or register oflanguage involved, or on prescribing rules for “correct” language use.

A turning point in theoretical linguistics was reached in the mid 1960s, whenNoam Chomsky, in his seminal text on generative grammar, Aspects of the Theoryof Syntax, offered a distinction between human beings’ knowledge of language(“competence”) and their actual use of language (“performance”) (1965: 4). Thefocus was still on language as a system, only Chomsky’s theory crucially includedcognition as a key component of language systems and processes. He stressed that“linguistic theory . . . is concerned with discovering a mental reality underlyingactual behavior” (1965: 4).7 In Chomsky’s view, syntax – the structures andprocesses of sentence-building – is the key to revealing that mental reality. Inaddition to placing syntactic structure at the center of linguistic theory, Chomsky



posited the existence of linguistic universals, structures, and concepts that arecommon to all human communication, and which indicate that human beings areuniquely endowed with a shared cognitive capacity to learn and use language.

2.1.2. Generative grammar and beyond

The notion of generative grammar within the study of linguistics is welldefined by Haegeman, who states: “The total of all the rules and principles thathave been formulated with respect to a language constitutes the grammar of thatlanguage. A grammar of a language is a coherent system of rules and principlesthat are at the basis of the grammatical sentences of a language. We say that agrammar generates the sentences of a language” (1994: 5) (emphasis in original).The concept of generative grammar is thus based on sentence grammar, howhumans construct their syntactic rule-systems, and what those rule-systems are.Throughout the twentieth century and into the twenty-first, debate and conceptualdevelopments in theoretical linguistics have flourished, different theoreticalapproaches yielding different types of analysis, from the detailed descriptionsdone in terms of structuralism to the powerful formalisms of generative syntactictheory. In recent years various theoretical approaches to the study of language havedeveloped in addition to generative theory, such as relational grammar, lexical–functional grammar, cognitive linguistics, construction grammar, functionallinguistics, lexical semantics, and others.

2.1.3. Basic linguistic theory and relational grammar

One approach that has been fruitful for the discussion and description ofmany languages is Dixon’s “Basic Linguistic Theory” (BLT) (Dixon 2010). BLT“consists in study and comparison of the grammatical patterns of individuallanguages” (Dixon 2010: 5), and centers on the fact that “every grammar is anintegrated system. Each part relates to the whole; its role can only be understoodand appreciated in terms of the overall system to which it belongs” (Dixon 2010:24). Along these lines, Dixon also characterizes grammar as “an abstract system ofinterlocking elements” (2010: 34). This concept of language structure helps tofocus analysis not just on individual components of language (e.g., morphology,phonology, syntax), but how those parts interrelate with the whole; that is, howvarious language systems and sub-systems synchronize and synthesize to create acomplex and effective network of communication.8

Relational grammar (RG) emerged as an alternative to transformational/gener-ative grammar in the 1970s. “RG sought to do justice to the interaction betweengrammatical relations, case relations, and thematic roles across language” (Butt2006: 33). I have found that RG is useful in analyzing Arabic syntax and semantics,

Linguistics and Arabic 5


in part because “RG implicitly assumes a relationship between overt case markingand grammatical relations,” (Butt 2006: 36) but also because of its compatibilitywith traditional categories of Arabic dependency relations. In addition, conceptsfrom lexical semantics (formerly “generative” semantics) and lexical decomposi-tion (especially predicate decomposition) are well suited to the analysis of Arabicsyntax and especially morphosyntax.9

2.1.4. Arabic linguistics

As applied to Arabic, linguistic theory has yieldedmany insightful studiesand also ways of approaching the language with precise, well-delineated analyt-ical, and discovery procedures. Particularly in the area of derivational morphology,Arabic offers a highly systematic and even exemplary perspective on languagestructure. In an overview of Arabic linguistics, Eid notes that:

Two approaches are identified as being dominant in research in theoreticallinguistics. One is more focused on developing a theory, or a part thereof,with data from individual language(s) serving as a testing ground for aspecific model being developed or an argument being made. The other ismore focused on analyzing linguistic data and discovering principles under-lying a linguistic system, with the theory being a means of approaching thedata . . . Both approaches are well represented in the literature on Arabictheoretical linguistics. (1990: 12–13)

Modern theoretical linguistics focuses to a large extent on syntax: phrase structureand clause structure. Much of the linguistic work on Arabic in recent years hascentered around word order, subordination, coordination, conjunction, agreement,relative clauses, prepositional phrases, transitivity, argument structure, and othercomponents of syntax and morphosyntax. The John Benjamins series,Perspectives on Arabic Linguistics (now amounting to more than twenty volumes),reproduces selected papers from the annual meeting of the Arabic LinguisticsSociety, and is a key resource for anyone interested in current theoretical thinkingabout Arabic.

As well as analyzing Classical Arabic and MSA, theoretical linguistics hassignificantly improved the understanding of vernacular Arabic grammatical struc-tures through the results of persistent and painstaking fieldwork. The subdisciplineof Arabic dialectology has produced extensive and valuable descriptive studies ofcolloquial Arabic in numerous regions in the Arab world and sponsored confer-ences on that topic (see, for example, the web site of AIDA: AssociationInternationale de Dialectologie Arabe at www.aida.org.at). Publications such asthe journal Zeitschrift für Arabische Linguistik regularly provide a range of articleson Arabic linguistics, examining both standard and spoken Arabic variants. Brill’s


http://www.aida.org.at


recent publication of the five-volume Encyclopedia of Arabic Language andLinguistics (2006–2009)has provided a much-needed reference work for research-ers in Arabic linguistics.

3. The Arabic grammatical tradition

“Every scientific discipline has a chronologically earliest paradigm, inother words, there is a definite point in time when a field achieves scientificmaturity” (Percival 1976: 287). For Arabic, the earliest paradigm for languageanalysis dates to the days of the young Islamic empire. The examination andanalysis of Arabic language structure do not start in the contemporary era.Centuries of indigenous erudition have preceded the application of current linguis-tic approaches to the description of Arabic, providing a powerful intellectuallineage for those who study Arabic today. The Arabic linguist who is not familiarwith the key conceptual insights of the great Arabic grammarians is bound to seeonly part of the picture of Arabic language analysis. Because of the centralimportance of the Qur’an and its message, and because of the more practical butessential role of Arabic literacy in building and administrating an internationalpolitical and religious power, Arabic language sciences were among the earliestdisciplines to emerge in the context of the Islamic empire, starting as early as theseventh century ad.10 Sociolinguist and historian of linguistics Dell Hymes pro-posed that the rise of linguistic analysis in any society is based on two factors: first,the existence of a corpus of written material; and, second, the recognition oflanguage change – the awareness of discrepancy – either synchronic or diachronic,i.e., language differences emerging within a speech community, language changebecause of contact with other language groups, or a recognition of differencebetween the current stage of a language and a previous one. It is particularly this“consciousness of imminent loss” of a valued form of language that appears to be adriving force in the growth of conscious awareness of language structure (Hymes1974: 5). In the case of Arabic, the language of the Qur’an was not only revered,but sacred, an “inimitable” rhetorical gift. Its preservation, therefore, and theanalysis of its linguistic processes and structures became a foundational discipli-nary activity in early Islam. The earliest Arabic grammarians used not only theQur’an but also the highly regarded genre of pre-Islamic and early Islamic poetryas the cornerstones of eloquence and correct usage for Arabic. As the context ofArab society shifted to greater horizons, and with the passing of time, the languageof the Qur’an and of the old poetic tradition became distanced from everydayspoken vernaculars, and the need for literacy in the written language became more

The Arabic grammatical tradition 7


acute in order to maintain cultural awareness of the Arabic word – both sacred andaesthetic.

To Hymes’ two factors for the development of language analysis I would addtwo more that apply, especially to the situation of classical Arabic: first, the needfor transference of language skills to other groups, i.e., the need to teach Arabic as aforeign language; and, second, the demand for translations and translators. Thesefour factors were all at play during the early days of the Islamic empire as it spreadits culture, religion, and language over a vast expanse of territory, encouragedpopular conversion to Islam, and developed a sophisticated cultural/political basein and around the Abbasid capital, Baghdad. With the establishment of a definitivewritten version of the Qur’an during the reign of Uthman, the third Caliph (644–656) had come the need to define principles of Arabic orthography, and with thestabilization of orthography came increased attention to grammar and lexicon.11

Scholarly momentum and literacy burgeoned during the first hundred years ofIslam, and a great thirst for systematized knowledge pervaded the Muslim world inthe eighth and ninth centuries. Language disciplines were leading components ofthe surge in translation, commentary, exegesis, documentation, education, andlegislation that were needed to form the foundations of Muslim culture, civilsociety, science, and governance. At the same time, other disciplines – medicine,alchemy, music, astronomy, and mathematics to name a few – began to flourish andform principles of practice, each with their own needs for taxonomies and technicalterms, translations, forms of education, and transmission of knowledge.12

The foundations for Classical Arabic grammar and lexicography were set by theend of the eighth century ad, with the extraordinary lexicographical legacy of Al-Khalil ibn Ahmad (Kitaab al-ʕAyn), and the evergreen grammatical masterwork ofSibawayhi (Al-Kitaab). The Arabic grammatical tradition thus consolidated itsfundamentals in written form over a thousand years ago, and constituted a back-ground against which disciplinary progress could be initiated, taxonomies could becompiled, terminology could be refined, and theoretical speculation could beengaged in – a matrix of information, analysis, and procedure that fostered thedevelopment of a lasting research tradition in Arabic language study.

The story of the development and elaboration of Arabic grammatical theory is along, intellectually fascinating, and distinguished one. Further readings in this areaare listed at the end of this chapter. I encourage those who have not yet had theexperience of dealing with Arabic primary sources from the late classical period/early Islamic period to try their hands at reading Sibawayi, at reading Ibn Jinni,Al-Khalil, or many of the prominent grammarians of early Islamic times, probingtheir architectures of linguistic complexity. Careful, close reading of originalsources helps us contemporary readers to integrate the intellectual discourses ofthe past into our epistemological frameworks and to prepare us for grounded,



thoughtful, and coherent analyses and queries not only about Arabic language butof linguistics as applied to the Arabic language. Foundational questions may nowyield transformational answers.

4. Aims of this book

This work aims to provide readers with a systematic introduction to thedescriptive methods and terminologies of contemporary Arabic linguistics, espe-cially as regards modern standard Arabic (al-fuşħaa). This book deals with levelsof linguistic analysis beginning with the sound system (phonology), progressingthrough morphology (derivational and inflectional), and to syntax. At each level,descriptive analysis is provided as well as an introduction to various theoreticalapproaches and intellectual trends. In addition, each chapter includes review anddiscussion questions as well as suggestions for further reading.

Questions and discussion points

(1) Some definitions of linguistics are given at the beginning of this chapter.Do they all agree? Do you know of others? Look up two more definitionsof linguistics and see how they compare.

(2) How does linguistics differ from the traditional field of philology? Howdoes it differ from the study of “grammar”?

(3) The terms “diachronic” and “synchronic” linguistics make a key distinc-tion in how the study of language structure is approached. Discuss thisdifference and its implications for Arabic.

(4) A further key distinction is between Chomsky’s use of the terms “com-petence” and “performance.”How do these terms relate to each other, andwhat do they imply for the study of Arabic linguistics?

(5) Discuss and evaluate the four factors mentioned in this chapter that areprerequisites for the initiation of language analysis within a particularculture. Do you agree that these are foundational? Do you think that thereare other factors that were especially pertinent for the emergence ofArabic language science?

Further reading

Bauer, Laurie. 2007. The Linguistics Student’s Handbook. Oxford: Oxford University Press.Bohas, Georges, Jean-Patrick Guillaume, and Djamal Eddine Kouloughli. 1990. The Arabic

Linguistic Tradition. London and New York: Routledge.

Aims of this book 9


Carter, M.C. 1972. Les origines de la grammaire arabe. Revue des Études Islamiques.40: 69–97.

1973. An Arab grammarian of the eighth century, A.D. A contribution to the history oflinguistics. Journal of the American Oriental Society 93: 146–157.

1981. Arab Linguistics. Amsterdam: John Benjamins.1985. When did the Arabic word naħw first come to denote grammar? Language &

Communication 5(4): 265–272.Eid, Mushira. 1990. Arabic linguistics: The current scene. In Perspectives on Arabic

Linguistics, vol. I, ed. Mushira Eid, 3–37. Amsterdam/Philadelphia: John Benjamins.Ferguson, Charles A. 1959. Myths about Arabic.Georgetown University Monograph Series

on Languages and Linguistics 12, 75–82. Washington, DC: Georgetown UniversityPress.

1962. Glossary of Terms Related to Languages of the Middle East. Washington, DC:Center for Applied Linguistics.

1990. “Come forth with a surah like it:” Arabic as a measure of Arab society. InPerspectives on Arabic Linguistics, vol. I, ed. Mushira Eid, 39–51. Amsterdam/Philadelphia: John Benjamins.

Owens, Jonathan. 1997. The Arabic grammatical tradition. In The Semitic Languages, ed.Robert Hetzron, 46–58. London: Routledge.

2006. A Linguistic History of Arabic. Oxford: Oxford University Press.2013b. History. In The Oxford Handbook of Arabic Linguistics, ed. J. Owens, 451–471.

Oxford: Oxford University Press.Retsö, Jan. 2013. What is Arabic? In The Oxford Handbook of Arabic Linguistics, ed.

J. Owens, 433–450. Oxford: Oxford University Press.Talmon, Rafael. 1985. Who was the first Arab grammarian? A new approach to an old

problem. ZAL 15: 128–145.Versteegh, C.H.M. (Kees). 1978. The Arabic terminology of syntactic position. Arabica

25(3): 261–281.1980. The origin of the term ‘qiyās’ in Arabic grammar. Zeitschrift für Arabische

Linguistik 4: 7–30.

Notes

1. “Linguistics is an empirical science, like biology or physics or astronomy. As such, itsgoal is the structure of explanatory hypotheses: empirically vulnerable accounts (theories)of observed phenomena” (Green and Morgan 1996: 37).

2. Dixon states further that “the grammar of a language has two components, syntax andmorphology. Some linguists treat phonology as a third part of a grammar; others regardphonology as distinct from grammar, but linked to it. A feature can be called ‘morpho-syntactic’ if it both occurs in a morphological paradigm and marks syntactic function; forexample a system of case affixes” (2010a: 93).

3. See Danks 2011, for an extensive analysis of the Arabic verb system.4. See Appendix A for an outline of sociolinguistics, applied linguistics, and computational

linguistics as applied to Arabic.



5. Study of language history and development from an empirical point of view is nowprimarily referred to as historical or comparative linguistics.

6. de Saussure (1857–1913) is considered the founder of modern structural linguistics,mainly by virtue of hisCourse in General Linguistics, given at the University of Geneva(1907–1911), and later published as a text through the efforts of certain of his students.See de Saussure 1972, 1983 [2009]: 9 “The linguist must take the study of linguisticstructure as his primary concern, and relate all other manifestations of language to it,”and 19: “The study of language thus comprises two parts. The essential part takes for itsobject the language itself” (i.e., langue). “The subsidiary part takes as its object theindividual part of language, which means speech” (i.e., parole). See also Culler 1986,for de Saussure’s intellectual biography and his theory of language.

7. van Valin in his article, “Functional linguistics” (2001: 325–327) has a useful summaryof these issues.

8. But note that Dixon also observes that “divisions within a grammar are seldom neat andtidy, and . . . one morphological form may have several roles in the syntax of alanguage” (2010a: 97).

9. For lexical semantics, see Cruse 1986 and 2001; for lexical semantics and predicatedecomposition, see Levin and Rappaport Hovav 1998; for a foundational study oflexical decomposition, see Gruber 1965; see also Jackendoff 1972; and for generativesemantics, see Parisi and Antinucci 1976.

10. “Face aux peuples civilisés, détenteurs de cultures supérieures à la leur, les Arabesprirent conscience de la nécessité de faire la conquête de leur propre langage par laréflexion sur ses structures, afin de découvrir toutes les richesses et mieux saisir toutesles significations du texte sacré” (Belguedj 1973: 169).

11. “The Qur’an conformed to Arab speech: it provided the reason for codifying Arabicgrammar and stylistics and was used as a criterion for these disciplines” (Abdel Haleem2009: 21).

12. “A thorough knowledge of grammar . . . was considered as a fundamental prerequisitefor any other intellectual pursuit, religious or secular” (Bohas, Guillaume, andKouloughli 1990: 49). See Baalbaki (2013) and Larcher (2013) for authoritativesurveys of the Arabic grammatical tradition.

Aims of this book 11

C:/ITOOLS/WMS/CUP-NEW/4464129/WORKINGFOLDER/RARA/9781107023314C02.3D 13 [13–22] 25.10.2013 1:59PM

2

Arabic phonology

Phonology is the analysis of the sound system of a language, including thestudy of the individual sounds themselves and how they are articulated (articulatoryphonetics) and how they are perceived (acoustic phonetics). Phonology also includesthe analysis of meaningful segments of sound (phonemes): their composition,distribution, and function (phonemics). This chapter will concern itself primarilywith the phonemics of Arabic, that is, analysis of distinctive MSA sounds and theirvariants (allophones). This analysis involves detailed description of the phonemesthemselves as well as description of processes that can be “phonemic,” (meaningful)such as vowel lengthening and consonant doubling (gemination).

1. Phonemics

The study of phonemics is concerned with the sounds of a language thatmake a difference in meaning; phonemes can be described as the semanticallysignificant sounds of a language. In order to establish a sound’s status as aphoneme, linguists look for environments or contexts in which everything isidentical except for one sound or a particular feature of a sound (such as voicing).If that sound or feature of a sound carries a difference in meaning and it contrastswith another sound in the same position, it is established as a phoneme. Phonemesare said for this reason to be in “contrastive” distribution.1 By contrasting sounds inpaired contexts, the identity of a phoneme can be established. This kind ofcontrastive comparison is called minimal pair analysis.2 For example, in Englishthe pair of words pet and bet, are exactly the same except for the initial sound, butthey are completely different in meaning. This contrast in meaning establishes thatthe sounds /p/ and /b/ are separate phonemes.3 Note that phonemes are conven-tionally written between two forward slashes, e.g., /k/.In Arabic, one can also come up with significant minimal pairs that establish

meaning differences between sounds (phonemes). Some of the most commonlyused examples are:

13


(1) kalb ‘dog’ / qalb ‘heart’ (both /k/ and /q/ are voiceless stops; the differ-ence is in point of articulation: /k/ is velar and the /q/ uvular.)

(2) sayf ‘sword’/ şayf ‘summer’ (both /s/ and /ş/ are voiceless alveolarfricatives, but the /ş/ has an added feature, velarization, that makes it adistinctive, separate phoneme.)

(3) tiin ‘fig’ / ŧiin ‘clay’ (both /t/ and /ŧ/ are voiceless alveolar stops, but /ŧ/ hasthe added feature of velarization.)

In the above example, the sounds /k/ and /q/ are established as phonemes, as well as/s/ and /ş/, and /t/ and /ŧ/. These sounds, in addition to being recognizably differentfrom each other are also thus established as theoretical constructs – components ofthe phonemic system of Arabic.

Each language has a particular sound inventory, or phoneme inventory thatrepresents all the meaningful sounds of the language. “In Modern LiteraryArabic, we find a very small vowel inventory . . . but a very rich consonantinventory” (Cohn 2001: 182). In an ideal situation, the phoneme inventory willcorrespond precisely with the symbols of an alphabetic writing system. This is notthe case for English (which has a complex and historically divergent system forspelling), but it is largely true of Arabic, where the letters of the alphabet essentiallyrepresent all the consonant phonemes of the language.4

1.2. Phoneme feature matrix

Phonemes each consist of a matrix of features that characterize them, andthey are described in these terms. The matrix consists of two basic sets of features:(1) place (or point) of articulation and (2) manner of articulation, i.e., where andhow the sound is produced in the vocal tract. A third feature, voicing (vibration ofvocal chords), is also a distinctive feature for many phonemes. For example, thephoneme /b/ is described as a voiced bilabial stop; the phoneme /s/ is described as avoiceless alveolar fricative, and the Arabic phoneme /ʕ/ is described as a voicedpharyngeal fricative. Note that sometimes a sound exists as a phoneme in onelanguage but not in another. For example, the glottal stop (usually transliterated as/’/ or /ʔ/ ) is a phoneme in Arabic, but not in English (even though it exists in manydialects).5

1.3. Consonants of modern standard Arabic (MSA)

The phoneme inventory of a language is usually represented in the formof a chart that indicates point of articulation, manner of articulation, and voicing.Arabic has twenty-eight consonants: eight stops: /b/, /t/, /ŧ/, /d/, /đ/, /k/, /q/, /ʔ/,thirteen fricatives: /f/, /th/, /dh/, /Z/, /s/, /ş/, /z/, /sh/, /x/, /gh/, /ħ/, /ʕ/, / h/, one

14 Arabic phonology


affricate: /j /, two nasals; /m/ and /n/, one lateral: /l/, one flap: /r/, and twosemivowels: /w/ and /y/. They can be charted as follows. Note that the point orplace of articulation is represented in the line across the top of the chart, proceedingfrom the front-most point of articulation at the lips to the farthest-back point, nearthe trachea (windpipe). The manner of articulation is noted in the far-left columnand indicate the degree of stricture or closing of the articulators. Generally, thedegrees of stricture include stops (where the airflow is blocked completely),affricates (where the airflow is blocked and then released into a fricative), fricatives(where the airflow is restricted but allowed through), and resonants (consonantsounds where the air flows smoothly – nasals, laterals, and semivowels).6

1.3.1 Arabic consonant phonemes

Descriptions of Arabic consonants are usually expressed in the followingtechnical terms:

hamza (ʔ) voiceless glottal stopbaaʔ (b) voiced bilabial stop

Labial

Stops

Voiceless

Voiced

Affricates

Voiceless

Voiced

Fricatives

Nasals

Laterals

Flaps

Semivowels

(approximants)

w

m

bt T

d

s sh

gh

x H h

c

c

z

n

l

r

y

S

j

k q

D

f th

dh Z

Voiceless

Voiced

Labio-dental Interdental Alveolar Palatal Velar Uvular Pharyngeal Glottal

Figure 1 Phonemic chart of MSA consonantsSource: Ryding 2005

Phonemics 15


taaʔ (t) voiceless alveolar stopthaaʔ (th) voiceless interdental fricativejiim (j) Three standard variants:

(1) voiced alveopalatal affricate; /j/ as in “judge”(2) voiced alveopalatal fricative /zh/: as in “rouge”(3) voiced velar stop; /g/ as in “guy”

ħaaʔ (ħ) voiceless pharyngeal fricativexaaʔ (x/ kh) voiceless velar fricative; also described as a uvular fricativedaal (d) voiced alveolar stopdhaal (dh) voiced interdental fricative: /ð/ or /dh/ pronounced like the /th/ in

“there”raaʔ (r) voiced alveolar flap or trill7

zayn or zaay(z)

voiced alveolar fricative

siin (s) voiceless alveolar fricativeshiin (sh) voiceless palatal fricativeşaad (ş) voiceless velarized alveolar fricative: /s/ pronounced with a

retracted tongue rootđaad (đ) voiced velarized alveolar stop: /d/ pronounced with a retracted

tongue rootŧaaʔ (ŧ) voiceless velarized alveolar stop: /t/ pronounced with a retracted

tongue rootZaaʔ (Z) Two standard variants:

(1) voiced velarized interdental fricative: /dh/ as in “there”pronounced with a retracted tongue root

(2) voiced velarized alveolar fricative: /z/ pronounced with aretracted tongue root

ʕayn (ʕ) voiced pharyngeal fricativeghayn (gh) voiced velar fricative; also described as a uvular fricativefaaʔ (f) voiceless labiodental fricativeqaaf (q) voiceless uvular stopkaaf (k) voiceless velar stoplaam (l) voiced lateral: this has two realizations:

(1) the “dark” /l/ as in “wall” or “bull” (back or “dark”/l/)(2) the “clear” /l /as in “lift” or “leaf” (fronted or “light” /l/)

miim (m) voiced biblabial continuantnuun (n) voiced nasal continuanthaaʔ (h) voiceless glottal fricativewaaw (w) or

(uu):bilabial semivowel: /w/ as in “wind” or long vowel /uu/pronounced like the “oo” in “boot”

yaaʔ (y) or(ii):

palatal semivowel: /y/ as in “yellow” or long vowel /ii/pronounced like the /i/ in “machine.”8

The consonant inventory of Arabic (listed above in standard alphabetical order)is characterized by two things in particular that distinguish it from the consonantinventory of English: consonants with secondary articulation (the velarized

16 Arabic phonology


consonants), and what are often termed “guttural” consonants: the velars, uvulars,and pharyngeals. Other components of the sound system vary as well, of course,but these two sets of phonemes are distinctive.In this book, I have chosen to represent word-initial non-elidable glottal stop

with the symbol /ʔ/ because it forms either part of the lexical root or part of thepattern. Often, word-initial glottal stop is omitted entirely from romanization intextbooks, but in that case there is no distinction between strong hamza (hamzat al-qaŧʕ) and weak or epenthetic hamza (hamzat al-waşl).

1.4. Vowels of MSA

The Arabic vowel system is straightforward: three different vowelqualities, each with a short and long variant. The difference in vowel length inArabic is not a difference in vowel quality, but in duration. This is similar to thedifference in duration of musical notes, where a half-note, for example, is heldtwice as long as a quarter note, and so on. There are two ways of analyzing thenumber of vowel phonemes in Arabic: they can be seen as six: three short andthree long, or they can be seen as four: three short vowels plus one “length”phoneme that can be added to each one. Either way is acceptable. Transliteratedrepresentation of long vowels varies from system to system. For my own use andin agreement with a number of other scholars, I prefer to represent long vowels asdoubled short vowels (i.e., aa, ii, uu) rather than use a macron (ā, ī, ū) or thedotted length notation ( a:, i:, u:).9

2. Phonemic processes

In addition to the consonant and vowel phoneme inventory of Arabic,there are also two phonemic processes, (1) vowel lengthening (as in kataba ‘ hewrote’ → kaataba ‘he corresponded’) and (2) gemination, or doubling (tashdiid),

Front Central Back

high i/ii u/uu

mid

low a/aa

Figure 2 Phonemic chart of MSA vowels

Phonemic processes 17


as in darasa ‘he studied’ → darrasa ‘he taught). Each of these processes contrib-utes to the derivation of words from a lexical root and forms a key component ofthe derivational system of Arabic.

It is important to note, however, that various forms of non-phonemic doublingalso occur. For example, in certain forms of preposition plus first person pronounsuffix: instead of *ʕan-ii ‘away fromme; about me’ the form is ʕannii; likewise, for*min-ii ‘from me; than me’ the form is minnii. In both of these cases the nuun isdoubled, but this does not change the meaning of the expression. Various con-textual situations also cause assimilation to occur, which often results in doublingthe pronunciation of a consonant, as in the assimilation of the laam of the definitearticle to the first sound of a word beginning with a “sun letter” (ħarf shamsi), e.g.,al-nuur is pronounced an-nuur; al-salaam is pronounced as-salaam. The pro-cesses of assimilation are discussed more extensively in the following chapter.

3. Allophones

An allophone is a contextually caused pronunciation variant of a pho-neme. That is, the pronunciation of the sound varies because of the environmentthat it is in. An allophonic variation of a phoneme is not a separate or independentphoneme because it does not carry or cause a difference in meaning. For example,American English has two kinds of pronunciation for /l/ – the fronted or “light” /l/of ‘leaf’ and the backed or “dark” /l/ of ‘well.’ The fronted /l/ occurs pre-vocalically and the backed /l/ occurs after a vowel. These two slightly differentrealizations of /l/ are still the same phoneme, just variant in pronunciation.

3.1. Allophones of laam

The /l/ phoneme in Arabic is usually realized as a fronted, “light” /l/sound. This Arabic /l/ is fronted and palatalized, approximating French /l/ as in“belle.”10 In certain restricted contexts, it is pronounced farther back in thearticulatory tract, with a raised tongue, as a “darker” /l/ sound. Thus the /l/ of fiil‘elephant’ and the /l/ of wallaah ‘by God’ are different, but not separate phonemes;they are allophones of the phoneme /l/.11

3.2. Allophones of jiim

The most variable consonant phoneme in MSA is /j/, represented by theletter jiim. Acceptable MSA pronunciations of jiim include the palatal voicedfricative /zh/ (as in the English word ‘measure’ or the French word ‘je’ ‘I’), or itmay be the voiced velar stop, /g/ (as in ‘good’) if the speaker’s regional spoken

18 Arabic phonology


variant is Egyptian or Sudanese, or it may be a voiced alveopalatal affricate, /j/ as inEnglish ‘judge.’12 Some Arabic speakers use both the /j/ and /zh/ variants. Thesepronunciations are all allophones of the MSA phoneme /j/.13

3.3. The issue of taaʔ marbuuta

The special form of taaʔ which can be written as word-final (taaʔmarbuuŧa) shows contextual variation as well, but it is unclear if this is trulyallophonic. In pause form, if a word spelled with final taaʔmarbuuŧa is the first partof an ʔiđaafa, or construct phrase, then the taaʔ is usually pronounced as avoiceless alveolar stop (e.g., madiinat bayruut, ‘the city of Beirut’). Elsewhere inpause form, it is pronounced as the voiceless glottal fricative /h/ (fii l-madiinah ‘inthe city’;madiinah kabiirah ‘a big city’). Because of its word-final position, the /h/is sometimes difficult to hear at all, and it sounds as though the word ends with thevowel /a/. There are also optional pronunciations of taaʔmarbuuŧawhen it followsthe long vowel /aa/, as in the word Hayaat/Hayaah ‘life.’ This particular variationprocess is complex because it is bound up with historical linguistics, grammaticalstructure, orthographic conventions, morphology, and regional differences inpronunciation.

3.4. Arabic vowel allophones

The pronunciation of Arabic vowels, especially /aa/ and /a/, but also /ii/and /i/ varies over a rather wide range, depending on word structure and theinfluence of adjacent consonants, but also on regional variations in pronuncia-tion.14 The primary cause of “backed” vowel allophones is the presence of anemphatic (velarized) consonant in a word.Ranges of vowel variation (front to back):

for /i/: min, xiffa, qiŧŧ, ŧ ifl, nişffor /a/: hal, sakan, qatal, şaff, tafađđalfor /aa/: islaam, waadii, ŧaalib, bayđaatfor /ii/ : diin, ŧiin, tamshiiŧ

The transference of the feature of velarization or emphasis from consonants toadjacent or even non-adjacent vowels and other consonants is referred to astafxiim, or “emphasis spread.”15 The retraction and raising of the tongue roottoward the velum (soft palate) or even farther back toward the uvula or phar-yngeal area for the pronunciation of the consonant, causes “backing” of thevowel.16 Postvocalic raaʔ also has a “backing” effect on /aa/ e.g., naar ‘fire’,

Allophones 19


daar ‘house’; so does the prevocalic uvular stop qaaf, e.g., qaadir ‘mighty,’maqaal ‘article.’

To sum up, the consonant system of modern standard Arabic shows littleallophonic variation except for jiim and laam, whereas the pronunciation ofvowels – especially /a/ and /aa/ – shows considerable allophonic variance dueparticularly to the spread of emphasis.


(1) Where are the major contrasts between the Arabic phonemic inventoryand the English phonemic inventory? What kind of impact might thesedifferences have on English speakers who are learning Arabic as a foreignlanguage? On Arabic speakers learning English?

(2) Come up with ten minimal pairs in Arabic, establishing certain sounds asphonemes. Remember, the sounds must occur in identical environmentsand be similar in terms of articulatory features, such as comparing thewords ħaal ‘condition’ and xaal ‘maternal uncle’, or ʔaxŧar ‘more danger-ous’ and ʔaxđar ‘green.’

(3) Read the article on “emphatic /l/” by Ferguson and discuss his ideas aboutthe phonemic status of “dark” /l/. Does he make a convincing argument?Why or why not?

Further reading

For book-length studies of Arabic phonology, see Al-Ani 1970, Gairdner 1925, and Semaan1968.Al-Ani, Salman H. 2008. Phonetics. In Encyclopedia of Arabic language and Linguistics,

vol. III, ed. Kees Versteegh, 593–603. Leiden: Brill.Bauer, Laurie. 2007. The Linguistics Student’s Handbook. Oxford: Oxford University Press.

(Especially 127–136 on the International Phonetic Association (IPA) and phoneticsymbols.)

Cohn, Abigail. 2001. Phonology. In The Handbook of Linguistics, eds. Mark Aronoff andJanie Rees-Miller, 180–212. Oxford: Blackwell.

Ferguson, Charles. 1956. The emphatic l in Arabic. Language 32: 486–552.Gordon, Cyrus. 1970. The accidental invention of the phonemic alphabet. Journal of Near

Eastern Studies 29:3: 193–197.Ladefoged, Peter. 1997. Linguistic phonetic descriptions. In The Handbook of Phonetic

Sciences, eds. William J. Hardcastle and John Laver, 589–618. Oxford: Blackwell.Laver, John. 2001, 2003. Linguistic phonetics. In The Handbook of Linguistics, eds.

Mark Aronoff and Janie Rees-Miller, 150–179. Oxford: Blackwell.Mitchell, T. F. 1990. Pronouncing Arabic. vol. I. Oxford: Clarendon Press.

20 Arabic phonology


Notes

1. Also called “parallel” distribution.2. “Minimal pairs (or sets) are words with distinct meanings differing only in one sound”

(Cohn 2001: 186).3. In fact, /p/ and /b/ are similar in both point and manner of articulation (bilabial stops) but

differ in one feature, voicing (/p/ is unvoiced, /b/ is voiced/). It is actually this singlefeature, voicing, that distinguishes the two words and the two phonemes.

4. For an interesting article on the correspondence between the Arabic phoneme inventoryand the Arabic alphabet, see Gordon 1970.

5. The glottal stop exists in English in various positions and in various dialects, but it is notconsidered an English phoneme. A Scotsman might pronounce the word ‘bottle’ as[boʔl]’; I myself pronounce the word ‘kitten’ as [kiʔn] and the word ‘satin’ as [saeʔn] inmyMichigan dialect. In the clear enunciation of vowel-initial words in English, a glottalstop is sometimes inserted to clarify word boundaries: “I said ‘an ʔice house,’ not ‘a nicehouse.’” Parrish, in his classic book on English pronunciation, Reading Aloud, color-fully refers to this as a “glottal attack” (1932: 160–161). The English expression ‘oh-oh’normally has two glottal stops, one before each ‘oh.’

6. See Laver 2001: 168 for a concise description of degree of stricture.7. “The conventional use of r in the transcription of Arabic and English completely

obscures the fact that the sounds so symbolized in the two languages are entirelydifferent; in Arabic r represents an apical trill, in English a slightly retroflex resonantcontinuant (a vocoid)” (Lehn and Slager 1983: 35).

8. “Certain consonants have some of the phonetic properties of vowels. . . they are usuallyreferred to as approximants (or frictionless continuants), though [/w/ and /y/] arecommonly called semi-vowels, as they have exactly the same articulation as vowelglides. although phonetically vowel-like, these sounds are usually classified along withconsonants on functional grounds” (Crystal 1997b: 159).

9. “Long vowels are best analyzed . . . as two identical short vowels. A universal rule,which need not be included in the grammar . . . accounts for this” (Abdo 1969: 9).

10. See Gairdner 1925: 17–19 for discussion of ‘dark’ and ‘light’ /l/.11. See Ferguson 1956 for an analysis arguing that “dark” Arabic /l/ is indeed a separate

phoneme.12. For this reason, the phoneme is often transliterated as /dj/, to indicate affrication (i.e.,

stop + fricative).13. Note that these variants are regionally determined, rather than caused by linguistic

context. These widely accepted MSA variants do not include the /y/ variant of vernac-ular Gulf Arabic because it is considered nonstandard.

14. Arabic allophonic variations were recognized by the Arab grammarians long ago, andgiven technical names referring to the manner of articulation: (1) imaala (‘leaning’ or‘inclination’ i.e., toward the high front part of the mouth), e.g., /aa/ or /a /→ fronted to[ae] or [e]; (2) tafxiim ‘thickening’ (or ‘heavying’ – pronouncing the vowel farthertoward the back of the vocal tract). Consonants that cause the latter pronunciation arecalledmufaxxama in Arabic, ‘emphatic’ or ‘velarized’ consonants. See Bakalla 2009 formore on tafxim (tafkhiim), and Barkat 2009 for acoustic analysis of vowel backing inArabic dialects.

Allophones 21


15. See Davis 2009: 636–637 for a discussion of “velarization” and “emphasis.” Note thatthe term “pharyngealization” is also used to describe the coarticulation feature of theseconsonants. Abdo and Hilu state that “a constriction in the throat, a sort of tightening ofthe muscles, accompanies the articulation of emphatics” (1968: 61). As Davis notesabout velarization and pharyngealization, “there is very little acoustic distinctionbetween the two” (2009: 636).

16. “Articulatorily, backed vowels are characterized by a constriction in the pharyngealcavity caused by Retracted Tongue Root” (Barkat 2009: 670). The term “retractedtongue root” is abbreviated as RTR. See Davis 1993: 150.

22 Arabic phonology


3

Arabic phonotactics and morphophonology

The term “phonotactics” refers to the study of sound distribution patternsand distribution restrictions within words (and sometimes across word bounda-ries).1 Phonotactic rules influence Arabic word structure in derivational and inflec-tional morphology, and also in lexical root structure.2 Most of these rules andrestrictions were discovered and described by the Arabic grammarians over athousand years ago (in particular by Al-Khalil ibn Ahmad, but also Sibawayhiand others). They are organized and described in this book in technical linguisticterms as they apply to MSA, using formalizations whenever possible. There are atleast two aspects to Arabic phonotactics: the phonotactics of root morphology andthe phonotactics of derivational and inflectional morphology. The study of mor-phological processes interfacing with phonological structures and rules is termedmorphophonemics. In Arabic the study of phonotactics and morphophonology areclosely interrelated. Four phonological processes are essential to the Arabic soundsystem: assimilation (one sound absorbing or affecting another), epenthesis (vowelinsertion), deletion (of vowel or semivowel), and vowel shift.

1. Assimilation (iddighaam/idghaam)

Assimilation can be defined as a change or spread of phonetic featurevalues (such as voicing or velarization) that makes segments more similar, or evenidentical. In Arabic it often occurs as a result of phonological rules applying at theintersection of morphological processes, for example, as a result of a pattern (wazn)applying to a particular lexical root (jidhr). Assimilation is normally non-phonemic; i.e., it does not affect the meaning of a segment or word. It may beprogressive or regressive.

1.1. Progressive assimilation

Progressive assimilation refers to the influence of a previously articu-lated sound on following sound, i.e., A→ B, A influences B. The most prevalent

23


examples of this result from the infixed /t/ of Form VIII verbs: the iftaʕala form,changing the nature of the infixed taaʔ to a pronunciation closer to that of the firstphoneme of the lexical root. It usually involves an orthographical change as wellas a change in sound.

1.1.1. Partial assimilation of infixed taaʔ (spelling variant)

işŧadama t → ŧ to collide, clashizdaada t → d to increaseizdawaja t → d to be double, to pair

1.1.2. Full assimilation of infixed taaʔ to preceding consonant (shadda inspelling)

iŧŧalaʕa t → ŧ to view, be informediZZalama t → Z to suffer injusticeiththaʔara t → th to be avengediddaʕaa t → d to allege, claimiddaghama t → d to be assimilated

1.1.3. Reciprocal (or mutual) assimilation (infixed taaʔ and precedingconsonant both change to a different consonant)

iddakara t + dh → dd to rememberiddaxara t + dh → dd to accumulate, preserve

1.1.4. Assimilation of initial waaw in Form IV verbal noun

Because of an Arabic phonological rule that prohibits the occurrence of thesequence [i + w] within a syllable, in the Form IV waaw-initial verbal noun, thekasra of the ʔifʕaal pattern assimilates thewaaw and lengthens into a long vowel /ii/.This assimilation and lengthening is reflected in spelling as well as pronunciation.

*ʔiwdaah → ʔiidaah clarification*ʔiwqaaʕ → ʔiiqaaʕ rhythm

1.1.5. Assimilation of initial hamza in Form IV verbal noun

Because of a phonological rule that prohibits the sequence [ʔ + V + ʔ](hamza – vowel – hamza), the hamza of Form IV hamza-initial verbs is assimilatedto the kasra of the ʔifʕaal pattern.3

*ʔiʔmaan → ʔiimaan faith*ʔiʔdhaan → ʔiidhaan proclamation

24 Arabic phonotactics and morphophonology


1.1.6. Vowel assimilation/assimilation at a distance/vowel harmony

The vowel /i/, whether long or short, as well as the semi-vowel /y/,assimilate the short vowel /u/ when it occurs after the third person pronominalsuffix /-h/. That is, the suffixes -hu, -hum, -hunna, -humaa all convert to -hi, -him,-hinna, -himaa when preceded by a high front vowel or semi-vowel. This kind ofassimilation at a distance is usually referred to as ‘vowel harmony.’

*fii maktab-i-hu → fii maktab-i-hi in his office*fii maktab-i-hum → fii maktab-i-him in their (m.) office*al-salaam-u ʕalay-hu → al-salaam-u ʕalay-hi peace be upon him

Note that this kind of assimilation occurs only with third person pronounsuffixes and is therefore conditioned by grammatical rules. In other contexts,such as when the /haaʔ/ forms a part of the lexical root, it does not occur, e.g.

tawjiih-u-naa our guidancemutanizzih-uuna walkers, promenaders

1.2. Regressive assimilation

Regressive assimilation results from the influence of a following soundon previous sound, i.e., A ← B. The most prevalent occurrence of this form ofassimilation is with the assimilation of the laam of the definite article (al-) to thesound of the first phoneme in a word; it also occurs with certain types of Form IVand Form VIII verbs.

1.2.1. Assimilation of laam to first sound of word

This assimilation occurs in the context of what are termed “sun letters” ofthe Arabic alphabet.4 It is important to note that the laam of the definite articleremains in the spelling of such words, but that the first sound in the word is doubledin pronunciation and spelled with shadda.

*al-daftar → ad-daftar the notebook*al-shams → ash-shams the sun*al-raabiʕ → ar-raabiʕ the fourth

1.2.2. Assimilation of initial semi-vowel waaw in Form VIII verbs:

Due to the phonological rule that prohibits the occurrence of the sequence[i + w], the waaw of waaw-initial verbs is assimilated to the infixed taaʔ of Form

Assimilation (iddighaam/idghaam) 25


VIII. This is a case of total assimilation and spelling change. The waaw disappearsin orthography as well as in pronunciation.

*iwtaşala → ittaşala to contact*iwtaħada → ittaħada to be united

1.2.3. Assimilation of affixed nuun of Form VII

FormVII verbs whose initial consonant ismiimmay optionally assimilatethe nuun of the form VII pattern (infaʕala). They may be spelled either way.

inmaħaa/ immaħaa to be erasedinmaħaqa/ immaħaqa to be destroyed

1.2.4. Assimilation across word boundaries

This process of regressive assimilation occurs with prepositions ending in/n/ followed by a word starting with /m/. The bilabial nasal /m/ assimilates thecoronal nasal /n/ and doubles, usually causing the words to fuse orthographically aswell as phonetically:

min + maa → mimmaa from whichʕan + maa → ʕammaa about which

1.3. Partial assimilation in pronunciation but not spelling (feature spread)

Avelarized consonant occurring anywhere in a word may lead to assimi-lated velarization in the entire word or in surrounding syllables.5 “Awell-knownfact about emphasis is that it spreads from an emphatic consonant to neighboringsegments” (Younes 1993: 119). This process of “feature spread,” is also known inparticular for Arabic velarized consonants as “emphasis spread.”6 For example, thefollowing words may be pronounced identically although the velarized consonantis in different positions:

şawt voicesawŧ whip

The degree of emphasis spread varies among speakers of Arabic, and variesconsiderably between vernacular speech and careful MSA pronunciation.7 Thevelarized consonant may affect the pronunciation of a whole word, even if thatconsonant is word-final. Or it may affect one syllable only. To some extent, thespread of emphasis beyond the velarized syllable is in free variation, that is, it is an



optional phonetic feature that does not affect word meaning.8 For example, differ-ent native speakers may pronounce the following words with a greater or lesserdegree of emphasis spread:

ixlaaş sincerityniZaam systemşabaaħ morningŧalabaat demands

2. Vowel insertion: epenthesis

The term “epenthesis” refers to the insertion of a sound into a sequence inorder to ease pronunciation and facilitate transition from one sound to the next.9

For this reason in Arabic these epenthetic vowels are often called “helpingvowels.” Because of Arabic phonotactic rules prohibiting the occurrence of threeconsonants in a row, vowel insertion is used to prevent consonant clusters.10 Theseconsonant clusters result from the interaction of lexicon, morphology, and syntax.All three short vowels (/u/, /i/, and /a/) are used in Arabic epenthesis, each vowelwith its own rules of occurrence.

2.1. Medial clusters

In medial clusters of three consonants, helping vowels are added accord-ing to the rules as set out below.

2.1.1. The vowel /u/ (đamma) or /uu/ (waaw)

This may be inserted to ease pronunciation after the occurrence ofsuffixed past tense second person masculine plural marker -tum. If -tum is followedby a noun with the definite article, the short vowel /u/ is inserted

hal katab-tum-u l-maqaaal-a? Have you (m. pl.) written the article?

When the -tum suffix is followed by a suffixed pronoun object, then the longvowel /uu/ is inserted between the inflectional suffix -tum and the object pronoun:

hal katab-tum-uu-hu? Have you (m. pl.) written it?mataa raʔay-tum-uu-hum? When did you (m. pl.) see them (m.)?

2.1.2. The vowel /a/ (fatħa)

This is inserted as a helping vowel in one case: after the word min ‘from’

Vowel insertion: epenthesis 27


hiya min-a l-yaman-i. She is from Yemen.min-a sh-sharq-i ʔilaa l-gharb-i from the east to the west

2.1.3. The vowel /i/

This is inserted in all other instances

man-i l-munassiq-u? Who is the coordinator?saʔal-at-i s-suʔaal-a. She asked the question.

2.2. Initial clusters: vowel prosthesis

The procedure of “prosthesis” refers to adding a vowel to a word,usually through prefixing. Because there is an Arabic phonological rule prohib-iting utterance-initial or syllable-initial consonant clusters (i.e., two or moreconsonants), Arabic allows the addition of a vowel prefix in order to make anutterance pronounceable.11 However, Arabic also has an even stronger rule thatno utterance or syllable may start with a vowel. It is therefore necessary to addhamza plus vowel in initial position in order to provide a glottal onset for startingan utterance. The hamza in this position is elidable (hamzat al-waşl) because it isneither a root consonant nor part of a morphological pattern; it is used purely forphonological reasons to ease pronunciation. Nonetheless, the ʔalif “seat” for theelidable hamza remains present in orthography in most cases. Elidable hamzaplus vowel occurs in the following situations: (1) with the laam of the definitearticle (vowel /a/) (e.g., al-burhaan ‘the proof’); (2) with the small group ofcommon biliteral nouns (vowel /i/) (e.g., -bn ‘son,’ -sm ‘name’); (3) with FormsVII–X verbal nouns, imperatives, and past tense verbs (vowel /i/) (e.g., inʕikaas‘reflection,’ istanbaŧa ‘to discover; to invent,’ istamiʕ! ‘listen!’); (4) with Form Iimperatives (vowel /u/ if verb stem vowel is /u/; otherwise, /i/) (e.g., uktub!‘write!’, iftaħ! ‘open!’); (5) with borrowed words that start with consonantclusters (vowel/i/) such as istuudiyuu ‘studio.’

3. Deletion (hadhf)

Deletion of vowels and semivowels occurs in Arabic, but rarely if ever,deletion of consonants. For example, the semivowel waaw of waaw-initial verbs(termed ‘assimilated’ verbs in English and ʔafʕaal al-mithaal in Arabic) may bedeleted in the present tense (e.g., waqaf-a / ya-qif-u ‘to stand’; wađaʕ-a/ya-đaʕ-u‘to put’).12



Short vowel deletion occurs under the following circumstances: where a vowelwould occur (based on a pattern template) between identical consonants precededby a short vowel:

marar-a → marr-a he passed by (compare: marar-tu I passed by)*radad-uu → radd-uu they replied (compare: radad-tum you (m. pl.) replied)

This rule may be expressed in formal notation, using the category symbols C forany consonant and V for any short vowel, as follows:

C1VC2VC3V ! C1VC2C3V=C2¼ C3

That is, the sequence CVCVCV will become CVCCV if and when the secondand third consonants are identical and followed by a vowel.13 In general,mathematical-style formalizations are considered strong evidence for the statusof rule-application because they summarize regularities and can be tested andevaluated for their predictive power.14 They strengthen the theoretical rigor ofobservations about language.

4. Vowel shift (qalb)

In addition to complete vowel deletion, Arabic phonological rules mayrequire vowel shift, i.e., displacement of a vowel from one position in a word toanother, as a result of either derivational or inflectional morphology. This is a sub-rule that applies after the application of the vowel-deletion rule in cases wherevowel deletion would result in a three-consonant cluster. This rule applies withgeminate roots when a short vowel occurs between identical consonants precededby another consonant:

Form IV: *ʔaħbab-a → ʔaħabb-a he loved

(The /a/ between the /b/’s is deleted according to the vowel-deletion rule, andshifts to the position before the /b/s in order to avoid a three-consonant cluster*ʔaħbb-a.)

Form X: *na-staʕdid-u → nastaʕidd-u we are getting ready

(In the same way, the /i/ between the /d/’s shifts to the position before the /d/’s inorder to avoid the unpronounceable *na-staʕdd-u.)In the elative form of adjectives derived from geminate lexical roots, this rule

also applies:

Vowel shift (qalb) 29


*ʔajdad-u → ʔajadd-u newer*ʔaqlal-u → ʔaqall-u fewer

And in certain plural patterns of nouns and adjectives from geminate lexical roots:

*ʔaŧbibaaʔ → ʔaTibbaaʔ doctors*ʔaʕzizaaʔ → ʔaʕizzaaʔ dear (pl.)

A formalization of this rule might read:

V1C1C2V2C3V3 ! V1C1V2C2C3V3=C2 ¼ C3

That is, a vowel between two identical consonants shifts to the position precedingthose consonants if the deletion of the vowel results in a three-consonant cluster.

Thus it is evident that there are strong phonotactic rules in Arabic that affectword structure, and these rules apply with regard to both derivational and inflec-tional processes. The study of this intersection of the rules of phonology withmorphological processes is referred to as morphophonology, and will be dealt withmore extensively in later sections of this book as we delve into morphologicalprocesses.


(1) Find ten examples each of progressive and regressive assimilation, otherthan the ones listed in this chapter. Note when orthography changes as aresult of these rules.

(2) Avoidance of consonant clusters is a key rule of Arabic phonotactics. Arethere any times or occasions when certain clusters (two consonants orthree consonants) are allowed to occur? Think of word- or utterance-initial, medial and final positions. Consider pause form as well as full-form pronunciation.

(3) Vowels and semivowels are sometimes incompatible in Arabic. We haveseen that the combination [i + w ] is prohibited by Arabic phonotacticrules, but not [w + i ] (e.g., wikaala ‘agency’). What other combinationsof vowels and semivowels are permitted or prohibited? Provide fiveexamples of these combinations or combination restrictions. Can youproduce a formal rule or rules for the structure of these combinations?

(4) The use of formalizations adds to the elegance and explanatory power oflinguistic observations.What areas of Arabic linguistics do you think lendthemselves to formalized rules? Could you improve on the formalized



rules written in this chapter? Write a rule for the assimilation of /u/ to /i/when it occurs in a third person pronoun suffix (1.1.6.).

Further reading

Bakalla, Muhammad Hasan. 2009. Tafxīm. Encyclopedia of Arabic Language andLinguistics, vol. IV, ed. Kees Versteegh, 421–424. Leiden: Brill.

Bauer, Laurie. 2003. Introducing Linguistic Morphology. Washington, DC: GeorgetownUniversity Press (especially sections on affixes and bases, pp. 146–156).

2007. The Linguistics Student’s Handbook. Oxford: Oxford University Press. (Forexplanation of general linguistic material and especially 95–103 on “notationalconventions.”)

Broselow, Ellen. 2008. Phonology. In Encyclopedia of Arabic Language and Linguistics,vol. III, ed. Kees Versteegh, 607–615. Leiden: Brill.

Frisch, Stefan A. 2008. Phonotactics. In Encyclopedia of Arabic Language and Linguistics,vol. III, ed. Kees Versteegh, 624–628. Leiden: Brill.

Ryding, Karin C. 2005. A Reference Grammar of Modern Standard Arabic. Cambridge:Cambridge University Press (especially pages 19–34 on phonology).

Zemánek, Petr. 2006. Assimilation. In Encyclopedia of Arabic Language and Linguistics,vol. I, ed. Kees Versteegh, 204–206. Leiden: Brill.

Notes

1. “An essential component of the phonological description of a language is statement ofwhich systems and subsystems of consonants and vowels correspond to the various slotsin the structure of syllable and of phonological word. This is ‘phonotactics’” (Dixon2010a: 275).

2. “Arabic has phonotactic restrictions between consonants within the verbal roots that haveplayed an important role in the development of phonological theory” (Frisch 2008: 624).

3. This rule applies widely, but usually when the hamza is fixed (hamzat al-qaŧʕ), not whenthe first hamza is weak (hamzat al-waşl). Thus, sequences such as the Form VIII verbalnoun ʔiʔtilaaf ‘agreement; coalition’ are permitted.

4. The sounds represented by the ‘sun letters’ (ħuruuf shamsiyya) include “dentals, sibi-lants, and liquids” (Zemánek 2006: 204). These include t, th, d, dh, r, z, s, sh, ş, đ, ŧ, Z, l, n.

5. “The emphatic consonant influences its neighborhood (the minimal domain of emphasisis the syllable, but in many cases, especially in the Arabic dialects, its domain can be awhole word” (Zemánek 2006: 205).

6. Shahin refers to emphasis spread in Palestinian Arabic in different terms, as “postvelarharmonies: pharyngealization harmony and uvularization harmony” (1996: 131).

7. See Bakalla (2009) and Younes (1993) for discussion of this point.8. This does not mean that emphasis spread is random, especially in vernacular Arabic,

where it may play key roles in discourse and identity. See Younes (1993) and Davis(1993).

Vowel shift (qalb) 31


9. The inserted sound itself is described as “anaptyctic” or “excrescent.”10. As my very first Arabic professor constantly reminded us, “Two sukuuns never meet.”11. The existence of consonant clusters at the start of an Arabic word is usually the result of

morphological processes; it may also occur when a lexical item is borrowed fromanother language.

12. This is true of verbs whose stem vowel is fatħa in both past and present tenses, and verbswhose past tense stem vowel is fatħa and present stem vowel is kasra. Other forms ofassimilated verb roots do not allow semivowel deletion. See Ryding (2005: 460–461)for more explicit rules and examples.

13. For a detailed analysis of this phenomenon based in prosodic morphology (McCarthyand Prince 1990), see Moore (1990).

14. The arrows of such rules are usually interpreted as “becomes,” “changes into.” Theforward slash is usually read as “in the environment of” or “when.” (See Bauer 2007 34–35 for more description of linguistic rule formats.)



4

Arabic syllable structure and stress

Syllable structure constitutes the component of phonological worddivision focused on pronounceable segments of words and how they are com-posed, divided, and distributed. Syllable structure is also a subdivision of thestudy of phonotactics, or the rules of sound distribution, the specific sequencesof sound that occur in a language. And, third, the study of syllables in Arabicinvolves the analysis of lexical stress. Although syllables themselves are linearand segmental in nature, word stress (the loudness or emphasis placed on asyllable) is suprasegmental; that is, it occurs at the same time as the pronunci-ation of the segment, adding a dimension of complexity to the syllable itself.MSA has explicit structural restrictions on syllables, as well as predictable rule-based stress based on syllable strength.1 Although not a spontaneous spokenregister of Arabic, MSA is nonetheless spoken on formal occasions (usuallyscripted) and in broadcast news and information formats, and adheres to estab-lished norms of stress placement. Recent published work on the stress system ofMSA has largely been done within the theoretical framework of prosodicmorphology.2 The discussion set forth here uses a basic descriptive approachsimilar to the one used in Ryding 2005 (36–39), Mitchell 1990 (19–21), andMcCarus and Rammuny (1974: 7–8, 23).

1. Syllable structure

In general, the core of a syllable is a vowel; in addition to a vowel, asyllable has “margins” that consist of consonants – either prevocalic or post-vocalic or both. The vowel core of a syllable is referred to as the syllablenucleus.3 In addition to the nucleus, a syllable has an onset or initial consonant,and may have a final consonant or consonants, termed the coda. Therefore in anArabic word such as min ‘from’, the onset would be /m/, the nucleus /i/, and thecoda /n/.

33


2. Arabic syllable constraints

There are clear phonotactic constraints on syllable structure in Arabic.4

Two sets of rules apply toMSA: one for full-form pronunciation and one for pause-form pronunciation.

2.1. For full-form pronunciation

The first constraint in the segmentation of MSA syllables is that nosyllable may start with a vowel. Second, no syllable may start with a consonantcluster (two or more consonants). Taken together these two rules yield the resultthat all Arabic syllables start with CV (consonant–short vowel) or CVV (conso-nant–long vowel). A third rule is that syllables must nowhere contain a cluster ofthree or more consonants. Therefore the permissible syllable types in full-formpronunciation MSA are usually three, each assigned a metrical value: either“weak” or “strong.”

(1) CV (consonant plus short vowel)“weak” or “light” syllable/wa/ /li/ /fa/ /mu/

(2) CVV (consonant plus long vowel)“strong” or “heavy” syllable/maa/ /tii/ /Duu/

(3) CVC (consonant-short vowel-consonant)“strong” or “heavy” syllable/mak/ /ras/ /tin/ /tub/

For example: here are some full-form Arabic words broken down by syllablesseparated by hyphens. In these examples, the hyphens do not indicate morphemeboundaries, but syllable boundaries.5 Stressed syllables are in bold.

shuk-ran thanksmu-qab-bi-laa-tun appetizersdha-ha-bun goldħaf-la-tun partyfaa-zuu they (m. pl.) won

yu-dar-ri-saa-ni they (two, m.) are teachingnak-tu-bu we are writingna-jaħ-naa we succeededta-waq-qa-ʕat she expected

34 Arabic syllable structure and stress


(4) CVVC: in restricted circumstances a CVVC syllable may occur within anMSA word. This is most often the result of the morphophonology ofgeminate roots, where the active participles of Form I verbs results inwords such as:

jaaf-fun dryxaaş-şun special, privatemaad-da-tun substance, material

2.2. Pause-form pronunciation

In addition to the syllabic sequences listed above, pause-form (omit-ting final short vowels) phonotactic rules allow for a word-final syllable to beeither CVVC or CVCC. These syllables are considered superstrong or“superheavy.”

(5) CVVC (consonant–long vowel–consonant): superstrong/-liin #/ /-riim # / /-suun #/ /-maan # /6

(6) CVCC (consonant–short vowel–consonant–consonant): superstrong/-rast # / /-rudd # / /-milt # /

Examples include:

yaʕ-ta-qi-duun they believenu-ħibb we like, we loveħa-milt I carriedmu-raa-si-liin reportersma-li-ka-taan two queensmab-niyy built

3. Formalization of syllable structure

Mitchell (1990) provides a concise formalization of Arabic syllablestructure, stating that “any [Arabic] syllable is derivable from the expression

CV Vð ÞC0=1=2

where the [parentheses] enclose a potential increment of vowel length andzero indicates the non-occurrence of a final consonant in the structure of asyllable. If C0 occurs, then the syllable is open; otherwise it is closed” (20).

Formalization of syllable structure 35


This rule tidily covers a range of lexical stress constraints. Other morespecific rules follow.

4. Lexical stress

Stress is a prosodic or suprasegmental (non-linear) feature of pronuncia-tion dependent in Arabic for its placement upon the nature of the syllable structurewithin a word. The placement of lexical stress in Arabic is predictable and non-phonemic; in fact, one author states that Arabic “stress in most instances is triviallypredictable” (Brame 1971: 556). Syllable structure and stress interrelate becauseone determines the other in Arabic, and it is therefore useful to be able to describethe system, especially when it comes to conveying to learners of Arabic a rule ofthumb for stress placement.7 Although McCarthy and Prince (1990b) state that“there is inconsistency in the stressing of standard Arabic words between differentareas of the Arab world” (252), they also admit that there is a “nearly universalnorm” which they summarize as follows: “The stress system is obviously weight-sensitive: final syllables are stressed if superheavy CvvC or CvCC; penults arestressed if heavy Cvv or Cvc; otherwise the antepenult is stressed” (252).8 Thissentence nicely summarizes the prevalent system of Arabic stress and I will unpackit as follows.

4.1. Basic rules of lexical stress9

Stress is always measured from the end of an Arabic word. A secondfeature of Arabic stress is that it never falls farther back than the third syllable fromthe end of a word (the antepenult). Stress rules differ slightly in full-form andpause-form pronunciation.

4.1.1. Full-form stress

When Arabic is pronounced in “full form,’ i.e., including all desinentialinflection markers, there are three basic stress constraints.

(1) Stress does not fall on a final syllable. In a word of two syllables, ittherefore falls on the first, no matter whether that first syllable is strong orweak:

ħat-taa untilqab-la beforehu-naa here



hi-ya shenaħ-nu we

(2) Stress is assigned to the second syllable from the end (the penult) if it is astrong/heavy syllable:

hu-naa-ka thereyad-ru-suu-na they are studyingmu-ta-ŧaw-wi-ʕii-na volunteerssa-fii-na-taa-ni two shipsqa-raʔ-tum you (m. pl.) read

(3) Stress is on the third syllable from the end of the word (the antepenult) ifthe second syllable from the end is weak/light:

mux-ta-li-fun differentda-ra-sat she studiedma-dii-na-tun citymad-ra-sa-tun schoolta-kal-la-muu they (m.) spokeʕuy-yi-na he was appointed

4.1.2. Pause form

An additional stress rule applies in pause-form pronunciation (wheredesinential inflection is omitted): that stress falls on the final syllable if it isCVVC or CVCC.

ka-riim Karim (man’s name)fa-naa-jiin cupstar-jamt I translatedruk-kaab ridersnaʕ-saan sleepymus-ta-ħiqq worthy, entitledyu-ħaa-wi-luun they (m.) try

4.1.3. Stress shift

As noted, words in Arabic may be pronounced in full form or in pauseform, depending on circumstances and context. Arabic words may also includepronoun suffixes that extend the length of the word and as a consequence (sincestress is calculated from the end of a word), shift the stress. The rules stated above

Lexical stress 37


still apply; it is the length or pronunciation style of a word that conditions theapplication of those rules. For example:

word form gloss(1) ti-jaa-ra pause form commerce, trade

ti-jaa-ra-tun full form commerce, tradeti-jaa-ra-tu-naa suffixed pronoun our commerce, our trade

(2) mu-ħaa-đa-ra pause form lecturemu-ħaa-đa-ra-tun full form lecturemu-ħaa-đa-ra-tu-haa suffixed pronoun her lecture

(3) jaa-mi-ʕa pause form universityjaa-mi-ʕa-tun full form universityjaa-mi-ʕa-tu-hum suffixed pronoun their (m.) universityjaa-mi-ʕa-tu-hun-na suffixed pronoun their (f.) university

(4) mad-ra-sa pause form schoolmad-ra-sa-tun full form schoolmad-ra-sa-tu-haa suffixed pronoun her school

(5) ħa-ma-la full form he carriedħa-ma-la-hum suffixed pronoun he carried them (m.)

(6) raa-qab-naa full form we observedraa-qab-naa-hum suffixed pronoun we observed them (m.)


(1) Select twenty Arabic words at random from a newspaper and put theminto phonological transliteration. Then assign the syllable boundarieswithin each word.

(2) After having done the first exercise, note where stress falls in each word,according to full-form pronunciation and then according to pause-formpronunciation.

(3) Write a brief introduction to syllables and stress for a group of studentsstudying first-year Arabic, and provide them with examples from thetextbook they would be using.

(4) Try writing formal rules for Arabic stress placement. For example, “stressis on the penult (second syllable from the end) if it is strong (CVV orCVC)” might look something like this:

CVVf g CVf gCVC=f g �� CVVf g #

CVCf g



The bold indicates stress; the forward slash indicates “in the environmentof”; the lank indicates where the bolded items would occur; hashmarkindicates word boundary.

(5) How would you compare the role of stress placement in English to therole of stress placement in Arabic? Are they equally important? If some-one pronounces Arabic using mistaken stress placement, is it still under-standable to most speakers?

Further reading

Abdo, Daud A. 1969. On Stress and Arabic Phonology: A generative approach. Beirut:Khayats (especially ch. IV, which deals with stress rules in Classical Arabic).

Jesry, Maher. 2009. Syllable structure. In Encyclopedia of Arabic Language and Linguistics,vol. IV, ed. Kees Versteegh, 387–389. Leiden: Brill.

Mitchell, T. F. 1990–1993. Pronouncing Arabic. 2 vols. Oxford: Clarendon Press (especially1990, vol. I, 19–21).

Ryding, Karin. 2005. A Reference Grammar of Modern Standard Arabic. Cambridge:Cambridge University Press (especially 36–40).

Notes

1. “In English, the incidence of the accent is unpredictable, having to be learned forindividual words, whereas in Classical Arabic it is dependent upon the syllable structureof the total word-form” (Mitchell 1990: 19).

2. See McCarthy and Prince 1990a and 1990b.3. Sometimes in other languages, a sonoric consonant (‘sonorant’) such as /l/, /r/, or /n/ will

serve as a syllable nucleus, like the /l/ in the English word “edible.”4. “Classical Arabic syllables are delimitable by the fact of their beginning with a consonant

and containing a vocalic nucleus, as well as by the inadmissibility of syllable-initialclusters and of sequences of more than two consonants” (Mitchell 1990: 19–20).

5. Because of the incomplete nature of Arabic script, it is customary to use transliteration forsyllable analysis and discussion.

6. The # symbol indicates the end of a word.7. English-speaking learners tend to be uncertain about Arabic stress placement because

lexical stress is not predictable in English. Sometimes they will borrow stress rules fromanother L2 that they have studied (e.g., French or Spanish) and apply to them to Arabic;but this experimentation usually leads to error and even more uncertainty. It is useful toprovide them with the rules, even if those rules are not immediately mastered. They canserve as a resource and a rule of thumb for stabilizing pronunciation progress.

8. Whereas this holds true in most Arab countries, Egyptian stress patterns tend to beaffected by Egyptian dialect stress, which is subject to a different set of rules.

9. These stress rules apply for Eastern forms of pronunciation ofMSA; they do not apply forEgyptian pronunciation.

Lexical stress 39


5

Introduction to Arabic morphology

1. Word structure

Ask most anyone, and they will say that words are at the heart oflanguage. Words are definitely at the center of discourse, and single words arethe first language elements that infants normally acquire. But seemingly simplequestions such as “what is a word?” have been surprisingly difficult to answer.Distinctions can be made according to various criteria. Three general aspects of“word” can be listed: the phonological word or word as a phonological unit; thelexeme, or content word with a dictionary meaning; and the “grammaticalword,” the word stem that serves as a base for grammatical/inflectional markers.One definition of ‘word’ is “a unit of expression which has universal intuitiverecognition by native-speakers, in both spoken and written language” (Crystal1997b: 419). However, the concept of “intuitive recognition” is neither empiri-cal nor rigorous. Another definition is “the smallest of the linguistic units whichcan occur on its own in speech or writing” (Richards and Schmidt 2010: 636).Here again, the concept of “on its own” is open to discussion. As Richards andSchmidt note, “it is difficult to apply this criterion consistently” (2010: 636).Morphology, the study of word structure, examines systematically the nature ofwords, their forms, their components, their interactions, and – to some extent –their meanings.

2. What is linguistic morphology?

Morphology in linguistics deals with the structure of words: howthey are formed and the identity and character of their component features.Sometimes words consist of solid stems (such as the Arabic noun yad ‘hand’or the English word book), but more often (especially in Arabic) words arecomposed of more than one morpheme (such as the English words books,bookshelf, booked; or the Arabic word maktab ‘office’ consisting of the

41


lexical root morpheme { k-t-b} ‘write’ and the grammatical pattern morphemespecifying “place,” { ma __ __ a __ }).1 A morpheme, then, can be definedas a minimum unit of form endowed with an independent meaning.2 Anotherdefinition is that a morpheme is “a minimal distinctive unit of grammar, andthe central concern of morphology” (Crystal 1997b: 248). Morphemes may befree, meaning that they can stand alone as words, or they may be bound,meaning that they exist only as components of words. In Arabic, most wordsare morphologically complex, that is, they consist of more than onemorpheme.

3. Where is morphology?

The location of morphological information and processes within the-ories of language has been for some time a matter of discussion and debate.3

Central to this debate is the question of the nature of morphological structure:are morphemes things (‘items”) or processes, or both?4 For structural descrip-tive linguistics, morphology is often represented as a tier or layer of languageanalysis situated between the phonological level and the syntactic level, andinteracting with both. Morphology is an area of extensive calibration betweensound and meaning, interfacing with both sound systems and the rules ofsyntax through the medium of word formation. For transformational–generativelinguistics, morphology is not a separate level, but functions within the gen-erative component of the grammar as well as within the lexical component. In“generative morphology” (Aronoff 1992a: 2), morphology is seen as autono-mous, and “should be dealt with on its own terms . . . it is different fromphonology and syntax. . . although it inevitably must interact with the rest oflanguage.” In fact Aronoff states that morphology “lies at the center of lan-guage” (1992a: 3).

Because of its interfaces with phonology and syntax, the role of morphologywithin the architecture of grammar is still open to many questions, such as: howto determine and characterize the ways in which morphology, phonology, andsyntax interact; how to distinguish complex or compound words from phrases;where are the boundaries between morphology and syntax (or between mor-phology and phonology) and how can they be identified? For Arabic in partic-ular these are important theoretical questions, and ones whose answers may berevealing for the field of morphology and morphological analysis inasmuch asArabic represents a mixture of non-concatenative and concatenative morpho-logical processes.

42 Introduction to Arabic morphology


4. Derivation and inflection

Morphology is traditionally seen as composed of two subfields: deriva-tion and inflection.5 It entails the examination of both the way that word stems orlexemes are formed (lexical or derivational morphology) and the ways that wordsbehave in context (inflectional morphology).6 Certain scholars, such as Aronoff,consider morphology to be an autonomous component of the grammar.7 For others,derivation and inflection are seen as segments of grammatical structure that aredistributed on an interactive spectrum or gradient, one end of which interfaces withthe lexicon and the other with syntax.8 Derivation is closer to the lexical end of thespectrum because its processes result in lexeme formation (e.g., derivation ofparticiples and verbal nouns from lexical roots). Inflection ranks closer to thesyntax end of that spectrum because of its greater degree of interface with syntacticprocesses (e.g., agreement and government).9

Derivation and inflection differ from each other in that derivation creates wordstems that are protected or “buffered” from the effects of grammatical operations,whereas inflection affects words by marking them for particular grammaticalcategories (e.g., case and mood; in Arabic, ʔiʕraab). The word stem (such asmaktab-) is a lexical unit, can be looked up in a dictionary, and carries semanticas well as grammatical information. When it is used in a sentence, it takes uponitself a role (e.g., subject of a verb, object of a verb, object of a preposition) and thatrole is marked by an inflectional suffix (for example, fii l-maktab-i). According toChomsky (1970) and Anderson (1988), the integrity of derived word stems givesrise to the “Lexicalist Hypothesis:” “The syntax neither manipulates nor has accessto the internal form of words” (Anderson 1988: 23).

Most current theories of grammar limit the interaction between syntax and lexicalmorphology. These theories differ, though, on whether they place the mechanismfor composing inflected forms in the lexicon along with derivational word for-mation rules (Strong Lexicalist Hypothesis, SLH) or in some syntactic or phono-logical component (Weak Lexicalist Hypothesis, WLH).

(Baedecker and Caramazza 1989: 109)10

For Arabic these morphological distinctions form an interesting field of study,especially as they relate to distinctions found in other languages. In a previousarticle, I noted the following:

That [Arabic] case and mood, as separate from number and gender, are moreclosely related to the syntax end of the grammatical spectrum, and that there maybe two distinct subcategories definable within inflection: the syntactically relevantcategories of number and gender [e.g., agreement] relating more closely to

Derivation and inflection 43


semantic and lexical specifications and the syntactically determined categories ofcase and mood more closely bound to syntactic processes.

If this distinction is relevant for Arabic, then it lends credence to Perlmutter’ssplit morphology hypothesis because it posits an even farther distance betweenderivation and desinential inflection; furthermore, it lends support to the contin-uum concept proposed by Bybee because of the differentiations and distinctionsof categories within the inflectional component of morphology, some categories(e.g., number) being more closely bound to semantics and some (e.g., case) tosyntax. (Ryding 1993: 177–178)

I would go as far as to say that morphology occupies the arena of prime importancein Arabic linguistic analysis. As opposed to syntax (considered central in standardtheories of generative grammar) morphology and its interfaces (morphophonologyand morphosyntax) constitute the most significant components of the powerfulinterlocking grammatical system that is Arabic. The key role of morphology inSemitic grammar has been noted by Aronoff, who notes the influence of this fieldonWestern grammatical analysis and even proposes that “it may. . .well be that allWestern linguistic morphology is directly rooted in the Semitic grammaticaltradition” (Aronoff 1994: 3).

5. Morphological models

Several morphological models have been central to the field of theoreticalmorphology. Five of the most important ones for Arabic are listed here:

(1) Item-and-arrangement model: a “model of morphological structure inwhich a complex word consists of a sequence of concatenatedmorphemes” (Booij 2005, 2007: 315).

(2) Item-and-process model: a “model of morphological structure in whicheach complex word is the output of one or more morphological processessuch as affixation and internal modification” (Booij 2005, 2007: 315). It“sets up one underlying form for alternating allomorphs and derives thesurface forms by applying feature-changing rules to the underlying form”(Bybee 1988: 119).

(3) Word-and-paradigm model: a “theoretical model of inflection that takesthe lexeme and its paradigm of cells as the starting point for the analysis ofinflectional systems” (Booij 2005, 2007: 324). “Word-and-paradigm is anapproach to morphology which gives theoretical centrality to the notionof the paradigm and which derives the word-forms representing lexemesby a complex series of ordered rules which do not assume that the word-form will be easily analyzable into morphs or that each morph will realize



a single morpheme. It is also known as a-morphous morphology”(Bauer 2003: 344).

(4) Lexical Phonology and Morphology (LPM): a theory of the interaction ofmorphology and phonology, especially as pertaining to issues of cyclicityand sequencing of morphophonological processes.

(5) Lexeme–Morpheme Base Morphology (LMBM): “is a theory of mor-phology which claims that lexical morphemes, called Lexemes, andgrammatical morphemes, Morphemes, are radically different linguisticphenomena” (Beard and Volpe 2005: 189) (emphasis in original).Essentially, lexemes are word stems of the major syntactic categories(nouns, verbs, adjectives), whereas grammatical morphemes such astense, person, plural, and so forth, are non-lexical. This theory reflects arather long-standing distinction between lexical and grammatical mor-phemes, although some theories of morphology consider all morphemes,whether lexical or bound, to be of essentially the same nature: meaning-bearing minimal forms.

6. Morphophonology and morphosyntax

In Arabic (as in many other languages) it is often the case that morphol-ogy overlaps or interfaces with phonology – e.g., through application of phono-logical rules – (morphophonology) and also with syntax, through application ofsyntactic requirements such as government or agreement rules (morphosyntax). Asfar as possible, linguists seek to identify morphemes, morpheme boundaries, andmorphological rules in order to be able to characterize fully the structures andfunctions that operate within Arabic morphological theory, and to systematicallyrelate sound to meaning. Sometimes morpheme boundaries are “fuzzy,” that is,they fuse into each other. One step in analyzing word structure, then, is to be able todistinguish all the morphemes that compose a particular word as well as variants ofthe base form or word stem which may occur as a result of derivational or inflec-tional processes. For example, hollow verbs in Arabic, such as qaal-a ‘to say’ havetwo stems in the past tense: qul- and qaal-. This is referred to as “stem allomor-phy,” or stem variance, resulting from phonological rules that apply as a result ofmorphological processes such as inflection or derivation. Arabic is a “synthetic” or“fusional” type of language (like Latin) wherein several morphemes may “fuse”together in one word, indicating various kinds of grammatical and lexical infor-mation.11 Sometimes these morphemes are relatively easy to identify and sort out,and sometimes not. A knowledge of Arabic root types and their variants as well as aknowledge of morphological processes need to be combined in order to deal withthe complexities of Arabic word formation.

Morphophonology and morphosyntax 45


7. Analytical procedures

Morphological analysis usually entails three basic steps:

(1) reducing word components to the smallest units of meaning(decomposition);12

(2) studying and categorizing recurrent elements of word structure; and(3) seeking generalizations that lead to “efficient, economical, and elegant”

explanations of word formation.13

7.1. Noun decomposition

In an Arabic word such as maktab-un, ‘an office,’ one can find at leastfour overt morphemes, realized by particular morphs:

(1) the lexical root {k-t-b} ‘write’(2) the noun-of-place lexical pattern {ma __ __a __}(3) the nominative case-marker suffix { -u }(4) the indefinite marker suffix {-n }

For the word maktab, one can also specify the morphological features number(singular) and gender (masculine). These features are grammatical components ofpoint number (2), the derivational pattern {ma __ __a __ }. In Arabic, themorphological properties “masculine” and “singular” are often unmarked, that is,without external features, and therefore more subtle in their presence.14

Nonetheless, they are part of the word’s essential feature matrix and play keyroles in agreement structures. For example, in order to say “a new office,” maktab-un jadiid-un, one needs to know that the modifier must be masculine, singular,indefinite, and nominative in order to be in strict concord with the noun.Altogether, the single word maktab-un has six morphemes: two derivational (rootand pattern) and four inflectional: number, gender, case, and definiteness.

7.2. Verb decomposition

In an Arabic word such as ya-ktub-u ‘he writes, he is writing’ one canidentify eight morphemes realized through the following morphs:

(1) the lexical root {k-t-b} ‘write’(2) the verb stem pattern { __ __ u __}, which also marks(3) the present tense and active voice(4) the person and gender inflectional prefix { ya- } (third person masculine)



(5) the mood-marking and number-marking inflectional suffix for indicativesingular: {-u}.

In this (and any) Arabic verb, these eight morphemes occur: root, stem pattern,tense (or aspect), person, gender, mood, number, and voice.15 Two of thesemorphemes are derivational (root and pattern); the other six are inflectional.

8. Words, morphemes, boundaries

Arabic word structure falls into several categories and subcategories. Todefine a word in Arabic often means thinking across morphological boundaries, asmost words are composed of at least two derivational morphemes: a lexical rootand a pattern template, as well as inflectional affixes. Because of the nature ofArabic pronouns, some words (nouns and verbs) can be extended to include suffixpronouns.16 It is therefore possible, in Arabic, that one word will be the equivalentof a clause, or full predication. If one takes a transitive verb, for example, inflects itfor tense, voice, subject, person, number, and mood and adds a pronoun objectsuffix, it may yield a complete one-word predication:

ya-staʕmil-uuna-haa. They are using it.sa-na-ntaxib-u-hum. We will elect them.

Word boundaries in Arabic are flexible, but obey specific grammatical and ortho-graphical rules. We know that the above expressions are considered words becauseof the application of word stress rules to the entire expression (phonologicalcriterion) and because of Arabic spelling conventions (orthographical criterion).They are nonetheless full predications.

9. sarf and nahw

şarf is the indigenous term used to refer to most of the morphology ofArabic, derivational (above the root level), inflectional categories such as gender,number, and person, and suffixal inflection on the past tense verb, and morpho-phonological change. “In relation to language, the science of taşriif is usuallycalled ʕilm al-şarf. Both indicate a change in the form of words, and both are usedindiscriminately to designate the science of morphology” (Åkesson 2009: 119).Neither şarf nor taşriif cover the inflectional categories of case and mood (desi-nence), however. Desinential inflection is normally considered a morphosyntacticcategory classified under the rubric of naħw, or syntax.

sarf and nahw 47


10. Other forms of lexical expansion

Although Arabic vocabulary resources are solidly anchored in the root/pattern system of word formation, other processes of lexical expansion also exist inArabic, such as borrowing words from a foreign language, coining new words,lexical blending, and compounding. Such processes have increased in frequencyfor Arabic in recent years due to the ever-growing need for new technical terms,and also due to the increased rapidity of communication and exchange of ideas at aglobal and international level. These non-root-and-pattern word-formation pro-cesses often originate in vernacular Arabic but are also developed in media Arabic,both spoken and written. These are dealt with in greater detail in Chapter 7.

11. Morphological terminology

“Avital requirement for all scientific work is a set of terminology which isclear and unequivocal” (Dixon 2010: 75). Here are some terms frequently used inthe discussion of morphology as it applies to Arabic.17 A more comprehensiveglossary is found at the end of this book.

affix: “A letter or sound, or group of letters or sounds (= a morpheme),which is added to a word, and which changes the meaning or functionof the word” (Richards and Schmidt 2010: 17). This includes prefixes,circumfixes, infixes, and suffixes. Another definition states that “anaffix is a grammatical element, belonging to a closed set, which canonly function as a component of a word” (Cruse 1986: 77).

allomorph: a variant of a morpheme that does not alter its basic identityor function, e.g., different forms for the English plural morpheme suchas: books, dogs, houses, oxen, children, sheep; or the contextual var-iation of the indefinite article a/an that depends on the initial sound ofthe following word. In Arabic also, the plural morpheme takes onvarious shapes: sound feminine plural (-aat), sound masculine plural(-uuna/ -iina), and the many variations of broken or internal plurals.The laam of the definite article (al-) has a wide range of allomorphicshapes because of the fact that “sun letters” assimilate it, and change itsrealization (e.g., al-dhahab ‘gold’ is pronounced adh-dhahab).

bound morpheme: a grammatical formative that cannot occur on itsown as a word (e.g., in English word parts such as -ish, -ment, un-, -ly, -s, -ed). In Arabic, a lexical root (jidhr) such as {k-t-b} or {d-r-s}, aword pattern or template such as {ma__ __ a __ } for noun of place, or



case-markers such as -u, -i and -a). Arabic abounds in boundmorphemes.

circumfix: a letter or sound, or group of letters or sounds which is addedat both ends of a word (i.e., prefix and suffix together), and whichchanges the meaning or function of the word, as in the Arabic presenttense structure ya-ktub-na ‘they f. write.’

citation form: the basic word stem listed in a dictionary.concatenative morphology: the formation of words through combina-

tion of elements into a linear sequence.derivational morphology: the creation of lexical items, word stems.discontinuous morphology: splitting one morpheme by insertion of

another unit, such as the interlocking of grammatical patterns andlexical roots in Arabic.

formative: any element entering into word formation, either derivationalor inflectional.

free morpheme: can function independently as a word (e.g., Arabic min‘from’).

infix: a letter or sound, or group of letters or sounds, which is addedinside a word or morpheme, and which changes its meaning orfunction.

inflectional morphology: the study of word variation in context (e.g.,number, gender, case, definiteness, tense, voice, and person are somecategories of inflectional morphology). Inflections are applied to wordstems.

lexeme: “A lexeme is a (potential or actual) decontextualized vocabularyword, a member of a major lexical category: noun (N), verb (V) oradjective/adverb (A)” (Aronoff 1992b: 13). Another definition is “theabstract unit that stands for the common properties of all the forms of aword” (Booij 2005, 2007: 316). Cruse refers to lexemes as “the itemslisted in the lexicon, or ‘ideal dictionary,’ of a language” (1986: 49).18

Aronoff 1994 states, “a lexeme, at least in its extrasyntactic state, isuninflected, both abstractly and concretely” (1994: 11).19

lexicon: essentially, the set of all words and idioms known to a nativespeaker of a language, including information on a word’s syntacticcategory (sometimes called “lexical category”), its grammatical func-tions and patterning, and its meaning/s.

morph: “A morph is a constituent element of a word-form. It is therealization of a morpheme (or sometimes more than one, see portman-teau morph)” (Bauer 2003: 334).

Morphological terminology 49


morpheme: a minimum unit of form having an independent lexical orgrammatical meaning. “Themorpheme is an abstract unit realized. . .bymorphs, or . . . allomorphs” (Bauer 2003: 334–335). Some theories ofmorphology crucially distinguish between lexical and grammaticalmorphemes (e.g., lexeme–morpheme base morphology).

morphology: the study of word structure and word formation, especiallyin terms of morphemes and morphological processes.

morphophonology: the study of the interaction/interface between mor-phology and phonology.

morphosyntax: the study of the interface between morphology andsyntax.

non-concatenative morphology: “morphology that makes use of otherprocesses than affixation or compounding to create new words or wordforms” (Booij 2007: 318). The root/pattern morphology of Arabic is anexample of non-concatenative morphology because the root mor-phemes and the pattern morphemes are discontinuous and are com-bined through interlinking rather than linear affixation.

paradigm: “A set of forms, corresponding to some subset (defined interms of a particular morphological category) of the grammatical wordsfrom a single lexeme. Paradigms are frequently presented in tabularform” (Bauer 2003: 337). For example, in Arabic, the possible forms ofa word can be listed in a table consisting of “cells” that constitute therange of word-form options possible in the language. For example, atriptote or fully inflectable (muʕrab) noun paradigm would look likethis, showing both case and definiteness within six cells:

Word-stem; najm- ‘tower’

Case Definite Indefinitenominative najm-u najm-u-ngenitive najm-i najm-i-naccusative najm-a najm-a-n

pattern (Arabic): a pattern is a bound and in many cases, discontinuousmorpheme consisting of one or more vowels and slots for root pho-nemes (radicals), which either alone or in combination with one tothree derivational affixes, interlocks with a root to form a stem, andwhich generally has grammatical meaning.20

portmanteau morph: incorporates two (or more) meanings in one mor-pheme, such as the dual suffix in Arabic:

-aani = number (dual) and case (nominative)-ayni = number (dual) and case (accusative/genitive)



prefix: a letter or sound, or group of letters or sounds which is added tothe beginning of a word, and which changes the meaning or functionof the word.

root (Arabic): a root is a relatively invariable discontinuous boundmorpheme, represented by two to five phonemes, typically three con-sonants in a certain order, which interlocks with a pattern to form astem and which has lexical meaning. In this book, usually referred to asa “lexical root.” Aronoff defines a root in more vivid terms: “A root iswhat is left when all the morphological structure has been wrung out ofa form. This is the sense of the term in Semitic grammar” (1992b: 15).

stem or word stem: the base or bare form of a word without inflectionalaffixes. A lexeme may have more than one stem.

stem allomorphy: the variation of a stem resulting from interaction withinflectional or derivational morphology.

suffix: a letter or sound, or group of letters or sounds which is added to theend of a word, and which changes the meaning or function of the word.

WFR: word formation rule. A rule that applies in derivationalmorphology.


(1) Morphological analysis. Analyze the following Arabic words into theirminimal morphological components by identifying the morphs (realiza-tions) of those categories. Separate the inflectional morphemes fromderivational morphemes. How many do you come up with? Are thereany that are difficult to identify?

(a) jaamiʕatun university(b) sayuTaalibuuna they (m.) will demand(c) ilayhimaa to the two of them(d) muħaamiihinna their (f.) lawyers(e) taʕaaluu come!(f) naʕam yes(g) mufiidun beneficial(h) zilzaalun earthquake(i) al-muxtabaru the laboratory(j) nataħaddathu we are speaking(k) tamtaddu it extends, spreads out(l) tastaʕdidna you (f. pl.) are preparing

(2) Arabicmorphology is said to be essentially “non-concatenative.”However,it also has concatenative (linear) processes of affixation, especially in



inflectional systems (for example, the suffixation of sound plural mor-phemes, e.g., muhandis-uuna). What would be some other types of linearor concatenative morphology in Arabic?

(3) Discuss the concepts of derivational and inflectional morphology as theyapply to Arabic. Are the boundaries between the two clear and distinctive,or do they ever blur? Does the Arabic term şarf cover both derivationaland inflectional morphology? If you are unsure, find an Arabic definitionof şarf and compare it to the concepts of derivational and inflectionalmorphology.

(4) List five free and five bound morphemes in Arabic.(5) A “portmanteau”morph realizesmore than onemorpheme, as noted above.

Make a list of four other examples of “portmanteau” morphs in Arabic.(6) In note 13, I quoted as follows: “In science, elegance aligns with preci-

sion, concision, and ‘ingenious simplicity’: an elegant solution is the onethat maps the most efficient route through complex terrain” (Sword 2012:165). Do you think that this definition of elegance applies in general, orjust to sciences such as linguistics?

Further reading

Bauer, Laurie. 2003. Introducing Linguistic Morphology. Washington, DC: GeorgetownUniversity Press. (This book has an excellent glossary of technical terms inAppendix C.)

Booij, Geert. 2005, 2007. The Grammar of Words: An introduction to morphology. Oxford:Oxford University Press. (This book also includes a useful glossary of technical termsin morphology.)

McCarthy, John. 2008. Morphology. In Encyclopedia of Arabic Language and Linguistics,vol. III, ed. Kees Versteegh, 297–307. Leiden: Brill.

Ratcliffe, Robert. 2013. Morphology. In The Oxford Handbook of Arabic Linguistics, ed.J. Owens, 71–91. Oxford: Oxford University Press.

Ryding, Karin C. 1993. Case/mood syncretism in Arabic grammatical theory: Evidence for thesplit morphology hypothesis and the continuum hypothesis. In Investigating Arabic:Linguistic, pedagogical and literary studies in honor of Ernest N. McCarus, RajiM. Rammuny, and Dilworth B. Parkinson, eds., 173–179. Columbus, OH: Greydon Press

Notes

1. A standard notational convention is to enclose morphemes in curly brackets { } (alsocalled ‘braces’). The slots in the pattern morpheme stand for the phonemes that constitutethe lexical root.



2. “A morpheme cannot be divided without altering or destroying its meaning” (Richardsand Schmidt 2010: 375).

3. “There are essentially three ways of thinking about morphology. One is to treat it as anautonomous ‘module,’ some of whose primes and principles are entirely independent ofother aspects of language (specifically, syntax, phonology, semantics or conceptualstructure): morphology-by-itself. Another is to think of morphology as a ragbag ofidiosyncratic phenomena . . .whose main interest for linguistics lies in the way it relatesto ‘genuinely’ linguistic levels of representation: morphology as merely a set of inter-face phenomena. The third tack is to admit that there are linguistically interestingphenomena in morphology (such as affix order or stem allomorphy), but to claim thatthese are reducible to principles of other models, e.g., syntax and phonology, respec-tively: reductionism (in the form of classical generative, SPE-type approach to allo-morphy, or to the more recent ‘syntax-all-the-way-down’ approaches to morphosyntax).To a large extent, current ideology favors the second and/or third position” (Spencer1994: 811–812). NB: The abbreviation “SPE” refers to The Sound Pattern of English, alandmark text in generative phonology by Noam Chomsky and Morris Halle, originallypublished in 1968.

4. “To simplify the polar positions [thing or process], in the former case we viewmorphemes as Saussurian signs – each a meaning paired with an identifiable form,presumably worthy of a lexical entry; strategies of word-formation like reduplication,ablaut, truncation, and metathesis then require some explanation. In the latter caseconcatenative and nonconcatenative morphology alike are deemed processual; ordinaryaffixation, like reduplication, ablaut, etc., is treated as a process” (Lieber 1996: 130).

5. “The distinction is delicate, and sometimes elusive, but nonetheless important”(Aronoff 1976: 2). Aronoff later notes: “Derivation and inflection are not kinds ofmorphology but rather uses of morphology: inflection is the morphological realizationof syntax, while derivation is the morphological realization of lexeme formation”(Aronoff 1994: 126).

6. See Ryding 1993 for an analysis of Arabic derivational and inflectional morphology.7. “In a number of recent publications on morphology, attention has been drawn to the

autonomy of morphology in the sense that the formal expression of inflection and wordformation is not always related to its content in a simple one-to-one fashion” (Booij1996: 812).

8. Perlmutter’s “split morphology” hypothesis proposes that derivational morphology ismuch more tightly bound to the lexicon, or the lexical end of the spectrum, than isinflectional morphology, and that “only syntactically relevant morphology can beextralexical” (1988: 94).

9. Carstairs refers to “a kind of spectrum of morphological behavior with ‘derivational’and ‘inflexional’ extremes” (1987: 4). Baedecker and Caramazza argue that “inflec-tional and derivational processes or representations are distinguished in the ‘cognitivelexicon’” (1989: 114).

10. Putting this another way, one scholar states that “there is some morphology (namelyinflection) which is integrated with the syntax in a crucial way; while other aspects ofmorphology (derivation) are primarily tied up with meaning” (Anderson 1988: 23).

11. See Dixon 2010: 226–227 for discussion and analysis of the terminology used inclassifying languages by morphological type.

12. Booij refers to these components as “atoms of words” (2005, 2007: 27).



13. Chomsky (1974: 3). In his 1974 work on modern Hebrew morphophonolgy, Chomskyemphasizes the concept of a “maximally simple grammar,” and states that “elegance” isa key factor in grammatical explanation, as well as simplicity (1974: 4). The followingdefinition of elegance is a very apt one: “In science, elegance aligns with precision,concision, and ‘ingenious simplicity’: an elegant solution is the one that maps the mostefficient route through complex terrain” (Sword 2012: 165).

14. Marked vs. unmarked features are components of “markedness theory,” which sees“certain linguistic elements . . . as unmarked, i.e., simple, core, or prototypical, whileothers are seen as marked, i.e., complex, peripheral, or exceptional” (Richards andSchmidt 2010: 352). Typically, in Arabic, masculine gender in substantives is seen as“unmarked” or most basic, whereas feminine gender often carries an overt feminine-marking morpheme, and is considered “marked.” Essentially the same argument can bemade for singular and plural number, singular being considered as the most basic, or“unmarked” category.

15. In this book I use the term “tense” rather than “aspect,” as I believe it is maximallyinformative for those readers new to linguistic analysis. In fact, tense and aspect seem tobe interwoven in Arabic verbs. See Ryding (2005: 51–52).

16. Bound pronouns are considered clitics, words/morphemes that are bound to other wordsand do not stand on their own. “A simple clitic differs from other lexical items in lackingthe prosodic status of ‘word:’ it has segmental, and possibly syllabic and even footstructure, but it is not a word” (Anderson 1988: 24).

17. See Carstairs-McCarthy (2005) for a concise introduction to and explanation of mor-phological terminology.

18. Cruse makes a distinction between lexemes and “lexical units,” especially as theseterms are used in lexical semantics (1986: 49–50).

19. Aronoff adds that “a lexeme is a (potential or actual) member of a major lexicalcategory, having both form and meaning but being neither, and existing outside ofany particular syntactic context” (1994: 11).

20. I am indebted to my mentor, Professor Wallace Erwin, for this definition.



6

Derivational morphology: the root/patternsystem

1. Introduction

Derivational morphology creates word stems, or lexemes.1 It builds andenlarges the lexicon so that concepts may find expression within a language.Sometimes the process of derivation changes a word’s form class (e.g., creatingan adjective from a noun, such as tuunis-iyy ‘Tunisian’ from tuunis “Tunisia’);sometimes it changes the subclass of a word (creating a transitive verb from anintransitive base, e.g., ʔadxala ‘to insert’ from daxala ‘to enter’). It affects almostall form classes or syntactic categories except those that are closed, such asfunction words (prepositions, conjunctions, particles).2 In Arabic, systematic der-ivation of words from lexical roots is at the heart of the word-creation system, andremains the distinctive feature of Arabic morphology. The fact that Arabic wordstems consist primarily of discontinuous morphemes (interlocked roots and pat-terns) has been of substantial interest to morphological theory in general.3

Derivational morphology can be expressed in terms of Word Formation Rules(WFRs). “A basic assumption . . . is that WFRs are rules of the lexicon, and as suchoperate totally within the lexicon. They are totally separate from the other rules ofthe grammar, though not from the other components of the grammar. AWFR maymake reference to syntactic, semantic, and phonological properties of words, butnot to syntactic, semantic, or phonological rules” (Aronoff 1976: 46). It is also thecase that “derivational markers will be encompassed within inflectional markers”(Aronoff 1976: 2). That is, derivation applies to word-stem formation, creating alexical unit. Inflectional markers are subsequently added to word stems whenwords are used in context. Derivation – in other words – is prior to inflection.Or, as Aronoff remarks, “Lexeme formation intrinsically feeds inflection”(1994: 127).In Arabic, the predominant word-formation process is based on root/pattern

morphology. However, other forms of lexical creation and innovation also exist,such as borrowing and compounding, and these have characteristics of their own.

55


In this chapter traditional Arabic derivational morphology is dealt with, and in thefollowing chapter, other forms of lexemes, lexical innovation, and expansion.

2. Paronymy in Arabic

The lexical root in Arabic is the source of semantic information uponwhich the Arabic meaning system is built; it is the key lexical element in wordformation. Usually consisting of three or four consonants (rarely, two), the root isan abstract concatenation of phonemes in a particular order which has lexical ordictionary meaning (such as {k-t-b} ‘write’ or {b-l-w-r} ‘crystal’). By virtue of thisfact, Arabic dictionaries are constructed according to roots, and within the root,according to the order of root phonemes – not according to word orthography.4

This is a key reason for the importance of understanding derivational morphology,for all words related to a particular root are clustered under that root, not by theirspelling.5 The technical term for words derived from a lexical base is “paronym.”

The relationship between one word and another belonging to a different syntacticcategory and produced from the first by some process of derivation . . . [is]paronymy; the derivationally primitive item [is] called the base, and the derivedform the paronym. (Cruse 1986: 130) (emphasis in original)

Paronymy largely characterizes the lexical resources of Arabic, with each base orlexical root extending its rich semantic power through morphological modificationsto create paronyms that cover the entire range of Arabic syntactic categories. Thisprocess of analogical modeling and building lexical resources through the applica-tion of analogous patterns and processes is a foundational feature of Semitic linguis-tics.6 In this regard, one of the key accomplishments of traditional Arabic grammarwas the ability to express any potential word-stem by fitting a grammatical pattern ortemplate into a specific model root: { f-ʕ-l }, thus giving form to both the abstractconcept of lexical root and the equally abstract concept of grammatical pattern.

By doing this, they provided a method for referring to patterns and a basis fordiscussion ofmorphological structures (e.g., faʕiil or tafaʕʕul). “The use of the root f-ʕ-las the prime exemplar for Arabic words is a powerful symbolic formalization thatprovides amodel of anymorphological template orwordpattern” (Ryding2005: 436).7

3. The Arabic verb system and its derivations

Verbs are the quintessential example of the systematicity of Arabic deriva-tional morphology and its basis within analogy, qiyaas. At the heart of the verb--creation process is the lexical root which interlocks with a set of “patterns” –

56 Derivational morphology: the root/pattern system


intercalcated vowels and consonants – in a fixed order, which have grammaticalmeaning.8 Moreover, within the verb derivation system, the various prepatterned“Forms” of the verb (ʔawzaan), or verb stem templates, convey a certain amount ofsemantic information such as intensity, association, or reflexivity. Each triliteral lexicalroot inArabic can potentially produce ten tofifteen verb Forms, and eachverb Form inturn produces participles (active and passive) and a verbal noun (maşdar).A practice of referring to these ʔawzaan, or Forms of the verb, using the Hebrew

term, binyan (pl. binyanim) has emerged in theoretical morphology research,especially autosegmental morphology and templatic morphology (see for exampleAronoff 1994: 123–164; Bauer 2003: 216–217; Booij 2005, 2007: 61; Durand1990: 258–259; Spencer 1991: 17 and 1994: 816). In this book I will refer to theseelements using the Arabic terms (wazn/ʔawzaan) or the established technicalEnglish term “Form” (capitalized) along with the reference numbers used inwestern scholarship and noted by roman numerals (I–XV). The British usage ofthe term “measure” to capture the meaning ofwazn is close to its original sense, butis not used here. In traditional Arabic grammatical analysis, the term ishtiqaaq isused to refer to derivational systems of paronymy, but this term means more thanderivation in the narrow sense: it is also used as an equivalent of the English term“etymology.” The kind of etymology involved here, however, does not trace wordhistories and their derivations from other languages or stages of language (as theEnglish term usually denotes); it refers to the system of root/pattern derivation,especially verb-stem templatic morphology, that characterizes Semitic languagesand Arabic in particular.9 This has led to the use of the term “derivationaletymology” to express the concept of ishtiqaaq in English.10

3.1. Verb morphology

Arabic verb stemsmust conform to certain stem templates or patterns, andif a notion needs to be borrowed into Arabic as a verb from another language, theforeign word must be adjusted to conform to one of the permitted verb-stem forms(e.g., talfana – ‘to phone’ based on a Form I quadriliteral pattern – faʕlala). Arabicverbs may be based only on triliteral or quadriliteral roots. If a biliteral rootmeaning is to be expressed as a verb (such as {b-n} ‘son’), it must be convertedinto a triliteral stem based on analogy with other triliteral stems (in the case of{b-n}, converted into a Form V defective verb stem, tabannaa ‘to adopt as a son’).There are ten essential verb-stem templates for triliteral verbs, each customarily

denoted by a roman numeral (I–X).11 In addition, there are five extended butrelatively rare verb-stem templates (XI–XV), taking the number of possible Formsup to fifteen. There are also four verb-stem templates for quadriliteral verbs (QI–QIV). Thus the complete verbal system of Arabic consists of nineteen stem

The Arabic verb system and its derivations 57


templates. A particular Form or wazn (or binyan) has been defined as “a functionthat maps a root onto a corresponding set of templates. The result of this mapping isa stem that undergoes inflectional morphology” (Aronoff 1994: 138). I would addthat the particular Arabic verb stem created by this process also undergoes deriva-tional morphology in that it produces a specific verbal noun (maşdar), an activeparticiple (ism faaʕil), and (for transitive verbs) a passive participle (ism mafʕuul),all of which exhibit characteristic features that distinguish them as members of anindividual verb-stem templatic class, wazn. Thus the process of verb-stem creationin Arabic is particularly rich, systematic, and productive.

3.2. Verb semantics

Each verb stem template or carries with it semantic implications orinformation conveyed by its pattern. This semantic information – or, as I havepreviously referred to it, its “semantic slant” (Ryding 2005: 434) – is interpretedwithin the sphere of the lexical information provided by the root morpheme,yielding a distinct lexical entity composed of both lexical and grammatical/seman-tic information. These semantic components include elements such as associativeaction (i.e., action involving another) (Form III), repeated action (Forms II and III),reciprocal action (Form VI), intensive action (Form II), causation (Forms II andIV), reflexive action (Forms Vand VIII), resultative (Forms VII and VIII), acquis-ition of a trait (Form IX), requestative or estimative (Form X).12 These categoriesdo not exhaust the possibilities of semantic modification of the root, but are someof the most frequent.13 One of the most pithy summaries of Arabic verb-stemlexeme formation is given by French linguist, Gérard Lecomte:

Si l’on met à part la forme dérivée IX, qui est nettement en marge du système, et laforme VII, commune à tout le domaine sémitique et de constitution claire, on peutexpliquer comme suit la formation des autres formes dérivées: les formes I, II, IIIet IV sont les quatre formes de base, auxquelles correspondent respectivement lesformes VIII, V, VI et X, obtenues en principe par préfixation d’un t-, qui leurconfère une valeur réfléchie-passive. Le principe est appliqué sans altération dansles formes dérivées Vet VI. Dans la forme dérivée VIII, on observe une métathèseimmédiatement perceptible. La forme dérivée X est issue non de la forme dérivéeIV à préfixe hamza, mais d’une forme dérivée IV à préfixe s- qui a existé dansd’autres langues sémitiques (ex. assyrien tardif). (1968: 34)

If one sets aside derived Form IX, which is clearly marginal to the system, andFormVII which is common to all Semitic languages and of clear constituency, onecan explain as follows the formation of other derived forms: Form I, II, III, and IVare the four base forms, to which correspond respectively Forms VIII, V, VI, andX, obtained in principle by prefixing of a /t-/, which confers on them a reflexive–passive value. The principle is applied without alternation in the derived Forms Vand VI. In derived Form VIII, one sees an immediately perceptible metathesis.



The derived Form X issues not from derived Form IV with the prefix hamza, butfrom a derived Form IV prefixed by /s-/ which has existed in other Semiticlanguages (e.g., late Assyrian).14

An Arabic root morpheme keyed into verb stem templates will therefore berealized as lexemes with different forms of transitivity: intransitive, transitive,and ditransitive. Figure 3 shows the typical templetic patterns, or Foms of the

Arabic verb forms

ActiveParticiple

VerbalNoun

PresentTense

PastTense Form

I

II

Unpredictable

III

IV

V

VI

VII

VIII

IX

X

PassiveParticiple

Figure 3 Arabic verb formsSource: Copyright 2005 of Georgetown University Press. ‘Arabic verbforms’. From Formal Spoken Arabic: Basic course with MP3 files, 2ndedn, Karin C. Ryding and David J. Mehall, 263. Reprinted withpermission (www.press.georgetown.edu)

The Arabic verb system and its derivations 59

http://www.press.georgetown.edu


triliteral lexical root. This transitivity shift may also be viewed as a shift in valency,with the verbal predicate taking one, two, or three arguments, or semantic roleswithin a predication.15

4. Arabic lexical roots and root types

As noted above, the traditional Arabic root typically consists of threephonemes in a certain order which convey lexical or dictionary-type information.This three-consonant minimal core is the lexical anchor that provides the keysemantic notion around which its derivatives are structured and elaborated. It isimportant to note that the sequence of consonants within a root is as important asthe root phonemes themselves, and that their sequence is considered a componentof a root’s lexical meaning. In mathematical terms, an Arabic lexical root may beseen as an “ordered set” of phonemes, e.g., < k, m, l >.16 The same members of theset in a different sequence convey substantially different semantic information(e.g., <k-m-l> ‘complete’ vs. <m-l-k> ‘possess,’ <k-l-m> ‘speak’).17 Each set isuniquely identified by its members and by the particular sequence of those mem-bers. Each Arabic root is, in other words, very much like a formula for meaning,expressed in terms of phonemes in sequence.

In terms of root-set membership, there are explicit cooccurrence restrictions onroot phoneme combinations, constraining the presence of homorganic consonants(phonemes that share point and/or manner of articulation, such as ħaaʔ and ʕayn –both pharyngeal fricatives) in a triconsonantal Arabic root, especially in adjacentpositions. This key determination about co-occurrence restrictions within rootmorphemes was discovered and described at length as early as the eighth centuryby Arabic grammarian Al-Khalil ibn Aħmad (d.791), in the introduction to hismulti-volume dictionary, Kitaab al-ʕayn, and later explored and described byGreenberg in his classic 1950 article, “The patterning of root morphemes inSemitic.”18 More recent studies, such as that of Pierrehumbert (1992) andBachra (2001), have explored the nature of these restrictions through applicationof the Obligatory Contour Principle (OCP). Bachra makes a case for “co-occurrence preferences” as well as co-occurrence restrictions in both Arabic andHebrew lexical roots (2001: 80–111).

It is the nature of the stem interacting with the inflectional systems of Arabic thatlead to a wide variance in “stem allomorphy” – that is, the modifications to a wordstem that occur as a result of inflectional or derivational morphological processes,such as the short and long stems for hollow verbs (e.g., qaal-; qul-; -quul-), theshifts in stem morphology that take place in assimilated verbs (e.g., wajad-; -jid-),or in verbs and other form classes derived from doubled verbs (e.g., radd-; radad-;



-rudd-; -rdud-). For this reason, knowing possible root-types is essential to pre-dicting possible form variants.

4.1. Biliteral roots

The role of biliteral roots in Arabic is deeply important in terms of boththe evolution of the language and in terms of its current structure and semantics.Although standard Arabic is clearly built around triliteral roots and their var-iants, there is a strong research tradition proposing that the original archaicSemitic (and Afro-Asiatic) root system may well have been biliteral.19 Thisessential biliterality is evidenced in the number of “weak” roots (containingwaaw or yaaʔ as a root consonant) and geminate roots that exist within thetriconsonantal system, as well as in the two-consonant commonalities sharedby many different roots (for example, {w-ş-l}, {b-ş-l}, {f-ş-l}, {ş-l-w}, {ş-l-ħ},{ş-l-b}, and many others).20

Some biliteral roots persist to this day, in very common words, almost alwaysnouns:

ʔumm motherʔab fatheryad handfam/fuu mouthʔax brother(i)sm name(i)bn son

4.2. Triliteral root (al-fiʕl al-thulaathiyy)

It is with the use of triliteral lexical roots that the Arabic verbalsystem has flourished and found profound signifying delicacy and expressivepower. Within this system, the lexical roots interlock with grammatical orform class patterns that elaborate a root-based reservoir of actual and potentialmeaning. The nature of root phonemes and their interactions with patterns ofderivational and inflectional morphology are key to the structure of the verbalsystem in Arabic, as well as knowledge of word stems and word-stemvariants (stem allomorphy). As Aronoff notes about Hebrew, “most irregu-larities in the forms belonging to the paradigm of a given verb can becharacterized in terms of the phonological peculiarities of certain root con-sonants. . . Once the nature and position of the weak consonant is given, theproperties of the verb paradigm follow automatically” (1994: 190–191). It istherefore useful to quantify the triliteral root-types in order to classify their

Arabic lexical roots and root types 61


paradigm variations. The various forms of Arabic lexical root weaknessinvolve gemination, hamza, waaw, and yaaʔ as phonemes in various positionswithin the root.21

4.2.1. Strong/regular root (saalim)

Three different consonants, none of which are waaw, yaaʔ, or hamza:

write k-t-bfasten; knot ʕ-q-dlift up r-f-ʕ

4.2.2. Geminate root (muDaʕʕaf)

Two consonants, the second of which is geminate, or doubled:

reply r-d-dhappy; secret s-r-rsolve ħ-l-l

4.2.3. Hamzated root (mahmuuz)

The glottal stop, hamza, interacts with morphology in various ways inboth orthography and in phonetic shape. It occurs in root-initial, medial or finalposition.

(1) hamza-initialeat ʔ-k-ltake ʔ-x-dhsorry ʔ-s-f

(2) hamza-medialask s-ʔ-lmend l-ʔ-mbe pessimistic sh-ʔ-m

(3) hamza-finalread q-r-ʔbegin b-d-ʔhide x-b-ʔ



4.2.4. Assimilated root (al-fiʕl al-mithaal)

Assimilated roots are ones which start with waaw or yaaʔ:

find w-j-darise y-q-Zput w-đ-ʕ

4.2.5. Hollow root (al-fiʕl al-ʔajwaf)

Hollow roots have “weakness” in the middle, in the medial radical, in theform of either waaw or yaaʔ:

sell b-y-ʕsay q-w-lsleep n-w-m

4.2.6. Defective root (al-fiʕl al-naaqis)

Defective roots have “weakness” in the third radical, the final radical:

appear b-d-winvite d-ʕ-wthrow r-m-y

4.3. Arabic quadriliteral root types: (al-fiʕl al-rubaaʕiyy)

Arabic quadriliteral roots are of five types: sound, reduplicated, com-pound (blended), acronymic, or borrowed.

4.3.1. Sound quadriliterals

Sound roots consist of four different consonants, one of which is usually aliquid or continuant: raaʔ, laam, nuun, or waaw.22

translate t-r-j-madorn z-r-q-shoverturn d-h-w-r

4.3.2. Reduplicated quadriliterals

Reduplicated quadriliterals consist of a repeated CVC pattern. Theseverbs are usually onomatopoeic, reflecting a sound or repeated movement.23

Arabic lexical roots and root types 63


neigh h-m-h-mflutter r-f-r-fshake z-l-z-l

4.3.3. Compound (blended) quadriliterals

Compound quadriliterals are the result of blending elements of two tri-literal roots into one, such as

jalmada ‘to be petrified’ (from the roots j-l-d ‘freeze’ and j-m-d ‘harden’)

4.3.4. Acronymic quadriliterals

Acronymic quadriliterals result from using the initials of words in a setphrase, such as basmala (from bi-sm-i llaah ‘in the name of God’), and refer to thesaying of that phrase. Other examples include:

ħamdala, ‘to say al-ħamd-u li-llaah’ ‘Praise be to God.’fadhlaka, ‘to say fa-dhaalika kadhaa wa-kadhaa, ‘and that is thusand so. . .’

4.3.5. Borrowed quadriliterals

Many contemporary quadriliterals are the result of borrowings from otherlanguages, such as:

talfana to telephonefalsafa philosophize

4.4. Quinquiliteral roots

Five-radical “roots” are items borrowed from other languages. They areused only as nominals, not as the basis for deriving verbs.

banafsaj violetshaŧranj chess

5. Derivational affixes in Arabic (zaaʔida/zawaaʔid: ‘augments')

Like other languages, Arabic makes use of certain sounds and sequences ofsounds to convey grammatical and lexical information. Arabic uses a subset of



phonemes in order to create patterns or templates into which root lexemes fit, and bymeans of which word stems are created and/or inflected. This subset of sounds arecalled zawaaʔid ‘augments’ in Arabic (singular zaaʔida), and are usually referred toas formatives in technical English terms, that is, they participate systematically inword formation. Onemorphologist defines formative as “serving to formwords: saidchiefly of flexional and derivative suffixes or prefixes” (Aronoff 1994: 2). The set ofitems used in Arabic derivational morphology overlaps with the set used in inflec-tional morphology.24 In this chapter derivational affixes and their uses are described;in the following chapter, inflectional affixes will be described.25

5.1. Arabic derivational formatives

In Arabic, formatives may be used in prefixes, suffixes, infixes or evencircumfixes, and in derivational morphology they work as components or sub-components of pattern formation. For example, the prefix mu- is a formative usedwith participles of derived forms of the verb, both active and passive. It constitutespart of many patterns, such as mufʕil (e.g., mumkin ‘possible,’ or mushrif ‘super-vising’) or muftaʕal (e.g., muntadaħ ‘alternative,’ or muħtaram ‘respected’). Thesystematicity of word-building strategies and features in Arabic and other Semiticlanguages exemplify the concept of derivational morphology in a particularly clearand compelling way.

The formative nature of linguistic morphology is especially clear when we look atSemitic languages, where roots are mere collections of consonants from which allindividual word-forms are quite dramatically given form by the laying on oftemplates and affixes. (Aronoff 1994: 3)

Semitic vowels are well known as components of derived word-stems including thefull vowel repertoire, three long and three short: /aa/, /ii/, /uu/, /a/, /i/, /u/. Theconsonantal affixes, on the other hand, have received less attention, and certainlyless systematic analysis. The consonantal derivational affixes used in Arabic areseven: hamza, taaʔ,miim, nuun, siin, yaaʔ, andwaaw. These thirteen formatives (sixvowels and seven consonants) constitute the phonemic inventory for pattern-formation in Arabic. (Note that this set of derivational affixes is not identical withinflectional affixes, although it contains some of the same items.) In addition to thisset of phonemes, Arabic also employs a derivational process: gemination.

6. Derivational consonant formatives

The following formatives are used to create word stems in Arabic ascomponents of particular patterns or templates that intersect with lexical roots.

Derivational consonant formatives 65


I cite the phoneme and give examples of its use in three positions: word-initial,word-medial, and word-final. These lists are not exhaustive, but contain commonexamples.

6.1. hamza

hamza occurs frequently as part of derivational patterns. This function ofhamza is not to be confused with its role in inflectional morphology (for example,the occurrence of hamza as part of the plural inflection in patterns such as qabaaʔilor jaraaʔid, or its use marking the first person singular inflection of an imperfectiveverb, e.g., ʔa-drus-u, ‘I study’). This use of hamza is strictly limited to its role inlexeme formation.

6.1.1. Word-initial hamza

Derivational word-inital hamza occurs as follows:

(a) In the stem class of Form IV verbs and verbal nouns (wazn ʔafʕal), e.g.

ʔaʕlana/ʔiʕlaan to announce/announcementʔaqaama/ʔiqaama to establish/establishmentʔakmala/ʔikmaal to complete/completion

(b) In the creation of nouns using the patterns ʔufʕuul and ʔufʕuula:

ʔusbuuʕ weekʔusluub styleʔuŧruuħa dissertation

(c) In the pattern-formation of the names of colors and physical characteristics:

ʔaħmar redʔazraq blueʔashqar blondʔaʕmaa blind

6.1.2. Word-medial hamza

Medial hamza is used in the derivation of the active participle of Form Ihollow verbs:

kaaʔin beingraaʔid pioneerzaaʔir visiting, visitor



6.1.3. Word-final hamza

(1) hamza occurs as a component of verbal noun patterns for defective verbs:

binaaʔ buildingilghaaʔ eliminationinqiđaaʔ expirationintihaaʔ endistiftaaʔ referendum

(2) hamza also occurs in final position in some singular noun patterns, e.g.

şaħraaʔ desertthulathaaʔ Tuesday

6.2. taaʔ

The formative taaʔ is used widely in the creation of Arabic word-stems/lexemes, and “is the only marker of the reflexive” (Larcher 2009: 642).

6.2.1. Word-intitial taaʔ

Word-initial taaʔ is used to form the following stem classes:

(1) Verb Forms V and VI, and their verbal nouns:

tamarrada/tamarrud to rebel/rebelliontanaffasa/tanaffus to breathe/breathingtaʕaawana/taʕaawun to cooperate/cooperationtakaafaʔa/takaafuʔ to be equal/equivalence

(2) the verbal nouns of Form II:

takraar repetitiontartiib arrangementtakwiin creation

(3) the Form II quadriliteral verb and verbal noun

tadahwara/tadahwur to tumble/tumblingtabalwara/tabalwur to crystallize/crystallization



6.2.2. Word-medial taaʔ

(1) Taaʔ is inserted after the first root consonant in the verb pattern for FormVIII, its verbal nouns, and its participles:

istamaʕa/istimaaʕ to listen/ listeningmustami/mustamaʕ listener/ listened-to, heardiktasaba/iktisaab to earn/earningmuktasib/mutktasab earning/earned, attained

(2) Taaʔ is also added to stem classes of Form X verbs, along with theformative /s/. These two are dealt with together in section 6.5.

6.2.3. Word-final taaʔ

As a word-final derivational element, taaʔ is not used with verbs. Usedwith nouns, it most often takes the form of taaʔ marbuuŧa. Many noun patternscontain the taaʔ marbuuŧa formative as part of their lexical formation. This is, ofcourse, distinct from its use as an inflectional gender marker, i.e., marking amodifier as feminine (e.g., maħZuuZa (t) – ‘fortunate, lucky.’). In the translitera-tion used here, the taaʔ marbuuŧa is noted in parentheses, as it is not normallypronounced in pause form. Some examples include:

(1) Form IV and form X hollow verbal nouns:

ʔidhaaʕa (t) broadcastingistifaada (t) benefit

(2) Form III verbal nouns:

musaaʕada (t) help, assistancemuħaawala (t) attempt

(3) Forms I and II verbal nouns of defective verbs:

daʕwa (t) invitationtarbiya (t) education, upbringing

(4) Some verbal nouns of assimilated verbs:

thiqa (t) trust, confidenceşifa (t) characteristic; adjective

(5) A number of basic noun formations, for example:

janna (t) gardenfikra (t) thought, ideaʕalaaqa (t) relationship



(6) As a marker of singularity on unit nouns and instance nouns:

mawza (t) a bananaraqşa (t) a dance

(7) As a derivational suffix on nouns of instrument:26

fattaaħa (t) openerthallaaja (t) freezer

(8) As a derivational suffix on nouns of place and participles:

maktab ‘office’/maktaba(t) libraryjaamiʕ ‘mosque’/jaamiʕa(t) university

(9) And as part of the derivational suffix -iyya (t) used to refer to concepts:

qawmiyya (t) nationalismfuruusiyya (t) horsemanshipnujuumiyya (t) stardomhuwiyya (t) identity

This listing of the derivational uses of taaʔ marbuuŧa is by no means complete,but it provides an idea of the wide-ranging functions of this particular formative inArabic derivational morphology.

6.3. miim

The derivational formativemiim is used only in word-initial position. It isnot used with verbs, only nouns (verbal nouns and participles). It may be used withshort vowels đamma, fatħa, or kasrah as a prefix:

(1) In participles of derived forms of the verb /mu-/:

muntadaħ alternative, choice (PP VIII)muslim Muslim (AP IV)mustaʕmar colony (PP X)

(2) In the verbal noun of Form III verbs /mu-/:

muħaađara lecturemusaaʕada help, assistancemubaadara initiative

(3) With the passive participle (PP) of Form I verbs /ma-/:

mawđuuʕ subject, topicmanduuħa alternative, choicemawjuud found; present



(4) With nouns of place /ma-/:

malʕab playgroundmarkaz centermaxraj exit

(5) With the miimi maşdar, a verbal noun of Form I that starts with miim/ma-/:

maʕrifa knowledgemaşiir destinymaʕnan meaning

(6) With nouns of instrument /mi-/:

miftaaħ keymiSʕad elevatormimsaħa cleaning cloth; (board) eraser

6.4. nuun

Nuun is used as a derivational formative in several ways, as set out below.

(1) As a prefix marking Form VII verbs and their derivatives. In the pasttense, the word stem result of this prefixation process is -nfaʕal-, unpro-nounceable because of the resultant initial consonant cluster. An epen-thetic short vowel /i/ is therefore prefixed, preceded by hamzat al-waşlwhen necessary.

(i)nfajara to explode(i)nbasaŧa to be content, glad(i)ndamaja be absorbed; incorporated

(2) (Much more rarely) as an infix in the extended Forms XIV (ifʕanlala) andXV (ifʕanlaa) of the verb:

isħankaka to be dark (Form XIV)israndaa to conquer, vanquish (Form XV)

(3) And as an infix in the Form III quadriliteral verb (ifʕanlala):27

ibranshaqa to bloom, flourish

(4) In word-final position, nuun is a component of the nominal derivationalsuffix -aan on deverbal nouns such as:

fiqdaan lossghufraan forgivenessfayađaan flood



(5) And as a component of derived adjectives of the form faʕlaan:28

farħaan glad, happykaslaan lazynaʕsaan sleepy

(6) A possible sixth instance of nuun as part of a derivational suffix could be itsuse in deriving adverbials from nouns and adjectives through use of theinflectional accusative indefinite nunation suffix, -an. Although nunation isnot normally considered a derivational suffix, I propose that this increas-ingly standardized process is one way to see the manner in which inflec-tional morphemes may become derivational over time, changing a wordfrom one form class to another. Unlike other forms of nunation, theadverbial accusative /-an/ is pronounced, even in colloquial Arabic.

ʔabadan neverʔaħyaanan sometimesmuŧlaqan absolutelyŧabʕan of course, naturally

6.5. /-st-/

The /-st-/ formative is used as a prefix for Form X verbs and theirderivatives. Because this formative is a consonant cluster, an epenthetic /i/ isprefixed to the past-tense stem and to the verbal noun to ease pronunciation.

(i)stawrada to import(i)stankara to disdain, detest(i)stithmaar investment

6.6. yaaʔ

The formative yaaʔ is used extensively in Arabic derivational morphol-ogy. Aside from its use as a long vowel in many patterns of derived verbs, nouns,and adjectives, it is also used as a consonant, as follows.

(1) Word initially, it is used in the derivational noun pattern yafʕuul:

yarbuuʕ jerboayaxđuur chlorophyllyanbuuʕ spring, source, well

(2) Word medially, yaaʔ surfaces as a consonant in a number of nounand adjective patterns, such as in verbal nouns of derived forms of



hollow roots and in the derivation of diminutive nouns, adjectives,and adverbs:

ixtiyaar choice (VN VIII)inqiyaad compliance, submission (VN VII)buħayra lakequbayla shortly before

(3) But by far the greatest use of consonantal yaaʔ in standard Arabic isas a word-final derivational formative for the process of conversionfrom one syntactic category to another {-iyy}.29 It is used extensively(with shadda) as a suffix to convert nouns to relative adjectives (al-nisba).30

tuunisiyy Tunisianyuunaaniyy Greeksharqiyy easterntaariixiyy historicalqamariyy lunarburtuqaaliyy orange

(4) In addition to its important role in the nisba suffix creating adjectives, theformative yaaʔ is also a component of the derivational suffix {-iyya}(along with taaʔmarbuuŧa), creating abstract nouns from a range of otherform classes:31

ʔahammiyya importancekammiyya quantitymasʔuuliyya responsibilityʔafđaliyya priority

6.7. waaw

The formativewaaw, like yaaʔ, is also used extensively as a long vowel inArabic derivational morphology, but in addition, it is used as a consonant thatexpands and regularizes biliteral or defective stems thereby allowing them to takederivational and inflectional suffixes:

sanawiyy annualʔaxawiyy fraternalyadawiyy manualʔaxawaani two brotherssanawaat years

Waaw is also used, much more rarely, in the derivation of two extended verb-stem templates:



(1) Form XII verbs on the pattern of ifʕawʕala:

iħdawdaba to become convex; humpbacked

(2) And doubled, in the derivation of Form XIII verbs, on the pattern ofifʕawwala:

ixrawwaŧa to last long

6.8. Gemination

In addition to these derivational phoneme formatives, a derivationalprocess is also used in Arabic: gemination/doubling. It does not occur in word-initial position, but it occurs frequently in medial position, and to some extent infinal position.

(1) In medial position:(a) In the creation of Form II and Form V verbal templates and their

derivatives through doubling of the medial radical:

Form II:rattaba to arrangeqaddara to appreciateʔajjala to delay

Form V:taŧawwaʕa to volunteertamannaa to wishtanabbaʔa to predict

(b) In the derivation of nouns and adjectives of intensity or professionthrough doubling of the medial radical:

nashshaafa dryerjarraaħ surgeonbaqqaal grocerqadduus most holy

(2) In stem-final position:(a) In the Form IX verb, through doubling of the final radical:

ixđarra to turn greeniħmarra to become red

(b) In the Form XI verb based on the pattern ifʕaalla:

ismaarra to be dark brownişfaarra to turn temporarily yellow



(c) In the Form XIV verb, based on the pattern ifʕanlala. In this pattern, avowel is inserted between the doubled final consonants.

isħankaka to be dark

(d) In the Form IV quadriliteral verb, on the pattern of ifʕalalla:32

iŧmaʔanna to be calm, reassurediqshaʕarra to shudder

7. Summary

The above phonemes and formative processes constitute the inventory ofconsonantal derivational formatives in Arabic. They have been described in somedetail in order to emphasize the systematicity of Arabic pattern-formation, the poolof linguistic resources put to use in word-formation, and the many varied uses towhich consonants as well as vowels are put in Arabic derivational morphology.Some morphologists downplay the key role of particular consonant affixes inArabic/Semitic, focusing more on consonant–vowel (C–V) prosody (in terms oftiers, melodies, prosodic templates, stem prosody).33 In producing a comprehen-sive inventory of the formatives, we have also introduced many types of derivationthat take place in Arabic, and have begun to estimate the power of Arabic pattern-formation resources. For a more complete description of Arabic form classes, theirtypes, and their morphology, see Ryding (2005).


(1) Arabic verb derivational systems are justly famous for their systematicitybased in analogy (qiyaas). What is your opinion about nominal derivationin Arabic? Do you think it is just as systematic? Why or why not?

(2) Word-formation rules (WFRs) are key components of Arabic derivationalmorphology. List five WFRs used in Arabic. Make your statements ofthese rules as explicit and as concise as possible. Compare your list withothers in your class.

(3) Read the articles “Stem” (Gafos) and “Root” (Zemánek) (see Furtherreading) in the Encyclopedia of Arabic language and Linguistics, vol. IV,and write a two-page summary of the most important points.

(4) Read the 1950 article by Greenberg and compare it to the work ofAl-Khalil ibn Ahmad, as published by Sara (1991). Write a five-pagepaper discussing and comparing their insights.



Further reading

Aronoff, Mark. 1992. Stems in Latin verbal morphology. In Morphology Now,ed. Mark Aronoff, 5–32. Albany: State University of New York Press. (This articledeals with Latin, but much of the analysis can be related by analogy to Arabic stemmorphology.)

Bachra, Bernard M. 2001. The Phonological Structure of the Verbal Roots in Arabic andHebrew. Leiden: Brill.

Gafos, Adamantios I. 2009. Stem. In Encyclopedia of Arabic language and Linguistics,vol. iv, ed. Kees Versteegh, 338–344. Leiden: Brill.

Greenberg, Joseph. 1950. The patterning of root morphemes in Semitic. Word 6: 162–181.Larcher, Pierre. 2009. Verb. In Encyclopedia of Arabic Language and Linguistics, vol. iv,

ed. Kees Versteegh, 638–645. Leiden: Brill.Sara, Solomon. 1991. Al-Khalil, the first Arab phonologist. International Journal of Islamic

and Arabic Studies 8(1): 1–57.Stetkevych, Jaroslav. 1970, 2006. The Modern Arabic Literary Language. Washington, DC:

Georgetown University Press. Especially ch. 1, on qiyaas.Zemánek, Petr. 2009. Root. In Encyclopedia of Arabic Language and Linguistics, vol. iv,

ed. Kees Versteegh, 93–100. Leiden: Brill.

Notes

1. “Derivational affixes produce new lexical units” (Cruse 1986: 77).2. But note, for example, that the derivational suffix {-iyya} may be used even with

pronouns and particles, e.g., huwiyya, ‘identity’ from huwa ‘he’; kammiyya ‘quantity’from kam ‘how much?’ or kayfiyya ‘quality’ from kayfa ‘how.’

3. “The formative nature of linguistic morphology is especially clear when we look atSemitic languages, where roots are mere collections of consonants from which allindividual word-forms are quite dramatically given form by the laying on of templatesand affixes” (Aronoff 1994: 3). A caveat: western morphologists who are not wellacquainted with Arabic have sometimes reported inaccurate data on Arabic verbForms. For example, an incorrect representation of the Form IV verb (wazn ʔafʕal)imperfective stem as “ʔu-ʔaktib” instead of “ʔu-ktib” is adduced in several linguisticstexts (McCarthy 1982: 134; Durand 1990: 258). See also Bauer (2003: 217); Spencer(1991: 17).

4. That is, Arabic dictionaries are organized alphabetically by root, not by word spelling.The root, of course, is an abstraction (e.g., {k-t-b}). The standard “citation form” forArabic lexemes is the third person masculine singular past tense, e.g., katab-a, used whendiscussing or listing lexemes.

5. For learners of Arabic, in particular, dictionary usage is important, and the process needsto be formally instructed, for it depends on morphological knowledge rather thanorthography and on declarative knowledge of the derivational systems of Arabic.

6. Stetkevych notes that “Qiyās as a linguistic concept and as method germinated anddefined itself in the relatively short span of time between ʽAbd al-Lāh Ibn Abī Isħāq (diedA.H. 117) and al-Khalil ibn Aħmad (died A.H. 175)” (1970: 2). He also observes that

Summary 75


among the early Arabic grammarians (particularly the Basran school), “analogy itselfwas turned into a binding rule, powerful enough not only to explain, but also to correctand to form” (1970: 3).

7. McCarthy states that the “basic insight” of early Arabic and Hebrew grammarians “wasto abstract away from the particular root, but not to any richer understanding of themorphological system than this” (1982: 117). He also refers to their understanding ofthe elaboration of ishtiqaaq as “rudimentary.” This conclusion seems questionable,given the extensive and centuries-long body of work – both speculative and pragmatic –by Arab grammarians, especially Ibn Jinni (d. ad 1002).

8. The use of the term ‘molds’ (a translation of the Arabic term qawaalib, sg. qaalib) todenote patterns within the pattern-based analogical system of Arabic derivation, ispreferred by Stetkevych, who states, “in the practical application of the analogicalmethod of derivation, we find the organizing criterion to be that of the linguisticmolds or qawaalib. All neologisms have to obey this criterion” (1970: 14).

9. See Stetkevych (1970: 7–47) for a classic analysis of ishtiqaaq and qiyaas.10. See Wright (2000).11. The original use of roman numerals to denote Arabic verb Forms dates to the seven-

teenth century and is attributed to Thomas Erpenius, author of a famous Latin grammarof Arabic and professor at Leiden University in the Netherlands.

12. Larcher states that although both II and IV may be causative, “when Forms IV and IIboth occur, there is always a difference of meaning between the two” (Larcher 2009:642). Note that the “resultative” FormVII may be interpreted as “unaccusative” in somerespects. See Levin and Rappaport Hovav 1995: 33–78 for discussion of resultativesand unaccusativity.

13. For a more comprehensive analysis of this topic as well as detailed morphology of theverb system, see Ryding (2005: 429–605).

14. See also Schramm (1962) for a detailed outline of the Arabic verb system.my translationof French original.

15. For valency see Chapter 9.16. The angle brackets < > indicate an ordered set.17. The Arabic grammarians – in particular Ibn Jinni – were aware of the combinatory

aspects of phonemes within particular lexical roots (or sets) and explored the idea ofvaried phoneme sequences in the study of what they termed “al-ishtiqaaq al-ʔakbar,” or“greater derivational etymology.” See Wright (2000). See also Bohas (1997) for morerecent analysis of root morphemes and their semantic implications.

18. For an English translation alongside the Arabic text of Al-Khalil’s analysis of Arabicroot phonology and phonotactics, see Sara (1991). Interestingly and unfortunately,Greenberg seems to have been unaware of Al-Khalil’s work. For more recent analysesof co-occurrence restrictions within Arabic lexical roots, see Mrayati (1987),Pierrehumbert (1992), and Bachra (2001).

19. See Ehret (1989) and also Bohas (1997) for more on the biconsonantal underlayment ofSemitic root structure.

20. For foundational work on this topic in both French and English see Bohas (1997), andBohas and Saguer (2006), (2007).

21. Nydell 1967 notes that there are 3,337 verb roots in the Hans Wehr Dictionary ofModern Written Arabic, of which about half are sound, about 25 percent are “weak”



(hollow, defective, or assimilated), 8 percent are geminate, 4 percent are hamzated, and10 percent quadriliteral.

22. “Certain quadriliteral verbs seem to be expanded triliterals, with liquid or continuantphonemes . . . added to the root” (Ryding 2005: 600). For more on quadriliterals seeRyding (2005: 599–605).

23. See Procházka (1993) for a discussion of reduplicated quadriliteral roots.24. “Morphology can be put to either derivational or inflectional ends, and the same

morphology can sometimes serve both” (Aronoff 1994: 127).25. Arabic uses a traditional mnemonic device that contains all the morphological compo-

nents of word structure in the form of an invented word: sa’altumuuniihaa, ‘you askedme it.’ See Ryding (2005: 48).

26. See Ryding (2005: 87–89) for more examples of this usage.27. Note that the Form XIV triliteral pattern is identical with the Form III quadriliteral

pattern, but when this pattern is plugged into the different roots, it yield different results,the Form XIV showing a doubled final radical and the Form III quadriliteral showingfour different root consonants.

28. Note that this -aan suffix is not the same as the dual suffix, -aani, which is inflectional.29. Note that although this suffix is often pronounced as a long /ii/ in pause form (thus

losing its consonantal nature), it formally consists of a short vowel /i/ plus the geminatedyaaʔ: {-iyy}.

30. But see Kouloughli (2007) for a different interpretation of the yaaʔ of nisba, which heconsiders inflectional, rather than derivational.

31. For more extensive examples see Ryding (2005: 90–92).32. See Ryding on extended forms of the triliteral verb and on extended forms of the

quadriliteral verb (2005: 596–605).33. See McCarthy (1982) McCarthy and Prince (1990a, 1990b).

Summary 77


7

Non-root/pattern morphology and theArabic lexicon

1. The Arabic lexicon

The Arabic lexicon, the word-stock of the language, consists primarilyof words derived through the dominant paronymic root/pattern system ofderivational morphology. However, a substantial segment of the lexicon con-sists of non-root/pattern-based lexemes. These items include solid stems thatdate back to the earliest forms of Arabic (such as laa ‘no,’ or hum ‘they m.’),borrowed foreign words and expressions, and the results of non-root/patternprocesses such as suffixation and compounding for word-creation and lexicalexpansion. This chapter examines both solid stems and the processes forexpansion of the lexicon which supplement the richness of root/pattern Arabicmorphology.

2. Solid stems

Solid stems are words which cannot be reduced morphologically oranalyzed in the typical root-and-pattern system. They consist of primarily foursets in Arabic: function words, pronouns, adverbs, and loanwords. Unlike wordsbased on lexical roots, solid-stem words are normally listed according to theirorthography in Arabic dictionaries.

2.1. Function words

A common subset of solid stems consists of Arabic function words –such as prepositions and conjunctions. These are high-frequency items, and interms of their structure, they are usually short or even monosyllabic. Theyinclude, for example, items such as fii,‘in, at’; ʔilaa, ‘to, towards;’ wa- ‘and;’fa- ‘so, and then;’, min ‘from;’ negation markers lam and lan; kay ‘in order that;’ʔanna ‘that;’ lakinna ‘but.’

79


2.2. Pronouns

A second solid-stem subset consists of Arabic pronouns, including per-sonal pronouns, demonstrative pronouns, and relative pronouns. These categoriesdo not fit into the standard root-and-pattern system, although they show definitephonological relationships to each other within their categories, such as the relationbetween haadhaa ‘this (m.)’ and haadhihi ‘this (f.).’

2.3. Basic adverbs

The number of word-stems in this class is small, but of relatively highfrequency, e.g., hunaa ‘here;’ hunaaka ‘there;’ hunaalika ‘(over) there’; faqaŧ‘only,’ haakadhaa ‘thus,’ ħaythu ‘where.’

2.4. Loanwords

There are also many loanwords (primarily nouns) in MSA that areborrowed from other languages, and these are considered, for the most part, tohave solid stems, e.g., they cannot be broken down into root-and-pattern mor-phemes (some of them may take broken plurals, however, if the singular stemreflects a typical Arabic noun pattern, such as bank/bunuuk and film/ʔaflaam).1

This category of words is a large and growing one, including words such asraadiyuu ‘radio’ kumbiyuutir, ‘computer,’ and siinamaa ‘cinema,’ ‘movies.’2

3. Lexical expansion through morphological processes: suffixation,compounding, blending, acronyms, and semantic shift

3.1. Suffixation: {-iyy } and {-iyya}

The yaaʔ of nisba {-iyy}, which creates relative adjectives, and thenominalizing suffix {-iyya} which creates abstract nouns, are both highly produc-tive derivational suffixes in Arabic. The yaaʔ of nisbamay be attached to nouns ofall types (even compound nouns and noun phrases) in order to convert them intomodifiers.3

maghribiyy Moroccanjanuubiyy southernħaaliyy currentduwaliyy internationallaa-nihaaʔiyy never-endingsharq ʔawsaŧiyy Middle Eastern

80 Non-root/pattern morphology and the Arabic lexicon


The feminine nisba suffix {-iyya} derives abstract nouns from a range of stems,including singular and plural nouns, adjectives, particles, and pronouns:

ʕamaliyya operationnujuumyya stardomdiibluumaasiyya diplomacyħurriyya freedomʔakthariyya majoritymawđuuʕiyya objectivitykayfiyya qualityhuwiyya identity

Note that the fact that this particular derivational affix may be applied subse-quent to inflections for pluralization or comparativeness contravenes the generalmorphological principle that “words inflected for number usually do not feed wordformation”, and that “inflection is peripheral to derivation” (Booij 1996: 814). Thisfact makes the {-iyya} suffixation process in Arabic of particular interest tomorphological theory.

3.2. Compounding

This refers to the derivation of new lexical items (single word-stems) byputting two (or more) words together, such as English laptop, has-been, sunburn,outlaw, schoolboy, snowflake, football, handyman, update, hand-me-downs, andso forth. In English compounding may also apply to modifiers, such as: would-be,middle-of-the-road, glow-in-the-dark. In Arabic, compounding is of several types:ʔiđaafa-based, negation-based, and phrase-based.4 In Arabic the process is usuallyknown as tarkiib:

3.2.1. One-word compounds

(1) From iđaafa structures:raʔsmaal capitalʕarđħaal petitionqaaʔimaqaam district official

(2) From negation structures:laa-markaziyya decentralizationlaa-faqaariyy invertebratelaa-ʔadriyya skepticism, agnosticismlaa-qaanuuniyy illegal

(3) From common phrases:maa-jaraa course of eventsyaa-nasiib lottery

Lexical expansion through morphological processes 81


(4) From coordinated noun phrases:barmaaʔiyy amphibian

(5) From adverbials plus the indefinite pronoun, maa:qabl-a-maa before; prior to + verbħayth-u-maa wherever

3.2.2. Two-word compounds

Noun–noun phrases: ʔiđaafaFrequently used two-word compounds, although orthographically sepa-

rate words, may come to be lexicalized as a single concept, and cohere as a singlesyntactic or lexical unit. For example, the expression jawaaz safar ‘passport’ ispluralized as jawaazaat safar, indicating the separate identity of the first noun (al-muđaaf ). However, when this concept is used with a possessive pronoun, thatpronoun is suffixed to the end of the expression: jawaaz safar-ii ‘my passport,’treating the ʔiđaafa as a morphological unit and stem for possessive purposes.

Other examples include:

rawđat ʔaŧfaal kindergartenradd fiʕl reactionrajul ʔaʕmaal businessman/mensuuʔ tafaahum misunderstanding

Certain verbal nouns used as the first term of an ʔiđaafa, have acquiredlexicalizing functions: ʕadam used as a negativizing prefix, and ʔiʕaada as a prefixindicating repetition or renewal. To a great extent in MSA, these two words arebecoming ‘grammaticalized,’ that is, they are shifting from being solely contentwords to being items that carry a specific grammatical function. This is especiallythe case with loan-translations, or calques, where each morpheme or part of asource word is converted into an Arabic equivalent.

ʕadam wujuud non-existenceʕadam istiqraar instabilityʕadam al-inħiyaaz neutrality, non-alignmentʔiʕaadat farđ-i l-ʕuquubaat-i re-imposition of sanctionsʔiʕaadat taʕyiin-i l-waziir-i reappointment of the minister

Adjectival compoundsLoan translations of complex adjectives may occur through the medium

of the adjective ʔiđaafa (ʔiđaafa ghayr ħaqiiqiyya):

mutaʕaddid al-ʔaŧraaf multilateralmutaʕaddid al-jinsiyyaat multinationalʕaalii al-mustawaa high-level



Regularized negative adjectival compounds

Loan translations of modifiers expressed in negative terms are oftenexpressed with the prefixation of the term ghayr ‘non; other than’ to an adjectiveor participle:

al-Zuruuf ghayr al-munaasiba the inappropriate conditionsşuʕuubaat-un ghayr-u mutawaqqaʕat-in unexpected difficultiesħasab-a ʔarqaam-in ghayr-i rasmiyyat-in according to unofficial figures

A further process to identify the internal coherence of a compound is toascertain if the second term can be conjoined to another term: for example,rawđat-u ʔaŧfaal-in wa-zuhuur-in (?) *‘garden of children (kindergarten) andflowers’. If the second term cannot logically be conjoined, that is an indicationthat the phrase functions as a lexical unit.

3.3. Blending and contractions (Arabic naht)

This involves parts of two (or more) words blending into one, sometimeswith truncation of the first component.

3.3.1. Fusing of word components

al-fawq-waaqiʕ iyya the supernaturalfaw-şawtiyy supersonicqab-milaadiyy before Christ (bc)maa-qab-taariixiyy prehistoricmimmaa from which

3.3.2. Formula-based verbs

These are verbs which have been coined based on the sequence of soundsin frequently used formulaic phrases. They tend to take the shape of Form Iquadriliterals:

basmala to say ‘bi-sm-i llaah-i’Hawqala to pronounce the formula: ‘laa Hawl-a wa-laa quwwat-a ʔillaa bi-

llaah-i’fanqala to say ‘fa-na-quul-u. . .’

3.4. Acronyms

These use the initials of a group or organization to form a word. Arabicdoes not usually create acronyms, but it may convert them from foreign languages

Lexical expansion through morphological processes 83


into Arabic in two ways, either as spelled-out words or as phrases denoting theforeign letters (or numbers):

yuuniskuu UNESCOʔuubik OPECsii ʔaay ʔiih CIAʔam ʔaay sitta MI-6

3.5. Criteria and diagnostics for determining compounds

Linguists search for principles and constraints to determine the status oflanguage elements. Determining whether compounds are morphological units orphrases can be done in three ways:

3.5.1. Orthography

This shows or does not show word-boundaries, and can be one criterionfor determining that an item is a lexical unit.

A compound word may be written as one word.

qaaʔimaqaam administrative officerʕarđħaal petition

3.5.2. Meaning

Meaning or semantic opacity is another criterion for determining the unityof a compound. The meaning of the compound may be non-compositional (opa-que). That is, the separate parts combine to constitute a meaning that is notdeterminable from their individual meanings, e.g., raʔsmaal ‘capital.’

3.5.3. Distributional

Distributional evidence can be used to determine if the sequence isconsidered a lexical unit (e.g., placement of definite article, demonstrative pro-nouns, -iyya, pluralization):

Placement of definite articleA compound may take the definite article:

al-laa-wujuud the non-existenceal-raʔsmaaliyya the capitalismal-şayd al-laa-qaanuuniyy illegal hunting



PluralizationA compound may pluralize as though it were one word, either by means

of a plural suffix at the end of the phrase or by means of a broken plural based onanalogy with similar singular structures:

maajaray-aat courses of eventsʕarđħaal-aat petitionsrasaamiil (forms of) capitalbarmaaʔiyy-aat amphibians

Derivational affixA compoundmay take a derivational affix, such as the yaaʔ of nisba or the

{-iyya} suffix denoting an abstract entity:

raʔsmaaliyy capitalistraʔsmaaliyya capitalism

PronounsA compound may take a pronoun suffix

maa-ʔadriyyat-ii my skepticismjawaaz safar-ii my passport

4. Lexical expansion through borrowing

Borrowing is either direct (taking a foreign word and Arabizing it in termsof pronunciation), or accomplished through loan-translation:

4.1. Borrowing

4.1.1. Nouns

Examples include:

talifiziyuun televisionintirnat internetbuuliis policeduktuur doctorbank bank

Lexical expansion through borrowing 85


4.1.2. Adjectives

Sometimes an adjective is borrowed without change; other times thesuffix -iyy replaces the adjectival suffix in the donor language:

biij beigemuuf mauveiliktruuniyy electronicmuusiiqiyy musical

4.1.3. Verbs

These borrowings are less common and need to be fitted creatively into anArabic verb-stem template, usually through quadriliteral Form I, or Form II andForm V triliteral stem templates.5

talfana to telephonetalvaza to televizebalshafa to Bolshevizetaʔamraka to be Americanized

4.2. Loan translation (calque)

These are words or expressions whose individual components are trans-lated literally into Arabic:

kiis hawaaʔ airbaglaa-markaziyya decentralizationnaad-in layliyy nightclubal-wujuudiyya existentialism

5. Lexical expansion through semantic shift

Sometimes a traditional Arabic word acquires a new, additional meaning,such as:

haatif invisible caller → telephoneshabka net → networkdharra speck, mote → atom



6. Summary

Arabic derivational systems combine both root/pattern morphologyand other highly productive morphological processes to create an extensivelexical base for MSA. Compared to root/pattern processes, the other forms oflexical derivation have been much less studied and yet raise some key ques-tions and concerns for morphological theory in general, especially the nature ofcompound or complex words and the criteria for determining their lexicalstatus, and the ability in Arabic to derive new lexemes from inflected wordstems.


(1) How would you express in Arabic “our passports?” or “their business-men?” What is happening morphologically with these expressions?Think of three new Arabic compound expressions and how they arepluralized, or how the possessive is expressed.

(2) The creation of new abstract nouns using the {-iyya} suffix seems toapply to all sorts of form classes or syntactic categories. Take a section ofan Arabic newspaper or a book chapter and find all the {-iyya} words in it.Do any of these surprise you? If you like, do a more extensive survey andwrite up your analysis in a short (five-page) paper.

(3) How would you classify these expressions in Arabic, one-word-stem ortwo (or more)? How can you tell? What are your criteria?

ghayr marghuub fii-hi undesirableʕadam taqdiir disrespectal-rabiiʕ al-ʕarabiyy the Arab springʕammaa about whichmaaward rosewaterlaa-silkiyy wirelessal-sharq al-ʔawsaŧ the Middle East

Further reading

Booij, Geert. 2005, 2007. The Grammar of Words: An introduction to morphology. Oxford:Oxford University Press. Especially chapter 4 on compounding.

Borer, Hagit. 2009. Afro-Asiatic, Semitic: Hebrew. In The Oxford Handbook ofCompounding, Rochelle Lieber and Pavel Štekauer, eds., 491–511. Oxford: OxfordUniversity Press.

Summary 87


Hijazi, Mahmoud Fahmi. 1978. Al-lugha l-ʕarabiyya ʕabr al-quruun. Especially pp. 61–118for examples of ishtiqaaq, tarkiib, and naHt.

Kossman, Maarten. 2013. Borrowing. In The Oxford Handbook of Arabic Linguistics, ed.Jonathan Owens, 349–368. Oxford: Oxford University Press.

Lieber, Rochelle and Pavel Štekauer. 2009. Introduction: Status and definition of com-pounding. In The Oxford Handbook of Compounding, Rochelle Lieber andPavel Štekauer, eds., 3–18. Oxford: Oxford University Press.

Ryding, Karin C. 2005. A Reference Grammar of Modern Standard Arabic. Cambridge:Cambridge University Press (especially pp. 90–92 and 254–275).

Notes

1. Although most borrowed nouns are pluralized using the sound feminine plural suffix/-aat/ (e.g., tiishirtaat ‘T-shirts’ kiibuurdaat ‘keyboards’), creative and even playfulbroken pluralizing of foreign terms by analogy with Arabic terms is very popular andproductive in vernacular Arabic, texted Arabic (“Arabizi”), and to some extent, mediaArabic, if and when the singular matches a typical Arabic singular noun template/patternwhich normally takes a broken or internal plural. Some examples include kaardiinaal(Roman Catholic ecclesiastical official)/karaadilah; fuldir ‘folder’/falaadir; filla ‘villa’/filal. In the case of such broken plurals, one might claim that the stems, althoughborrowed, are not actually solid, but permeable to inflection.

2. Many non-Arabic Middle Eastern place names also fall into the solid-stem category, likebaghdaad, ‘Baghdad,’ tuunis ‘Tunisia,’ and bayruut, ‘Beirut’ These names are notoriginally Arabic, but originate from other Middle Eastern languages, such as Aramaicor Persian. For more on these geographical names, see Ryding (2005: 96).

3. For a more extensive and detailed analysis of this derivational process, see Ryding (2005:261–269, 272–273).

4. The ʔiđaafa structure itself is a phrase, but I have noted it as a separate source ofcompounding because of its central importance for this type of morphology.

5. Vernacular Arabic is much more flexible than standard Arabic in borrowing or creatingverbs based on foreign expressions, e.g., kansal ‘to cancel,’sayyaf ‘to save,’ ‘kayyash’ ‘tocash.’ Note still, however, that the verb-stem template rule must be followed (quadri-literal verb or Form II verb).



8

Arabic inflectional morphology

1. Introduction to inflection

Inflectional morphology examines the nature and processes ofword-change within syntactic structures. It examines inflectional paradigms(conjugations, declensions) and the types of inflectional change realized onwords-in-use. That is, it examines the range of inflectional possibilities availableto particular word-stems (their paradigms) and it examines the nature of theirroles in context (their syntagmatic relations). Paradigms can be compared towardrobes of choices for particular words (options for dress), whereas syntag-matic relations can be compared to events which determine the wardrobeselection of particular words (a particular event requires a particular wardrobechoice – black tie, casual, come-as-you-are). Therefore, a word in context whichis filling a particular syntactic role bears a paradigm mark determined bothby the word’s inherent nature (an Arabic diptote, for example) and also by itscontextual relations (object of a preposition, for example). Here are threeinstances of a prepositional phrase whose noun object inflects for the genitivecase in different ways:

fii dimashq-ain Damascus

fii l-madrasat-iat the school

fii l-mustashfaaat the hospital

Every noun falls into a particular inflectional class or declension, whichallows or restricts its ability to exhibit the full range of inflectional distinctions.There are eight noun declensions in Arabic. See Appendix C for thesedeclensions.

89


2. Arabic inflection

Compared to English, words in Arabic are highly inflected.1 This waspartially illustrated in Chapter 5 where two words, maktab and ya-ktub-u wereanalyzed down to their most basic morphological components, maktab showingsix morphemes (four of which were inflectional), and ya-ktub-u showing eight(six of which were inflectional). Western linguistics recognizes inflectionalgrammatical categories such as number, gender, case, person, mood, tense, andvoice – all essential elements in marking word-function within syntax. Arabicgrammatical theory, however, designates that case and mood belong to a separatecategory which is determined by or governed by syntactic rules (ʕawaamil).

The difference between categories of number, gender, person and tense, on onehand, and case and mood on the other, are clear and significant. In the caseof nouns, for example, number and gender are conceived of as determined directlyby real-world information, (i.e., semantically) whereas mood and case are deter-mined by the syntactic function of the item within the clause structure; i.e., theyare purely intralinguistic features. (Ryding 1993: 175)

Bauer refers to these two types of inflection as “inherent” (semantic, extralinguis-tic) and “contextual” (syntactically determined). “Contextual inflection is the kindof inflection that is determined by the syntactic structure: agreement for person,gender/noun class and number, case-marking. Inherent inflection is the kind ofinflection that is not entirely determined by the syntax although it may have somesyntactic relevance” (Bauer 2003: 106).

3. Arabic morphosyntax

Within any phrase or clause in Arabic there are interactions betweenmorphology and syntax, networks of dependency relations that determine theshape or form of individual words. Two principles regulate these relations:

3.1. Agreement or concord (mutaabaqa)

Agreement or concord is where lexical items or words in a phrase orclause match or conform to each other, or reflect each other’s features, in order tomake sense. For example, a feminine dual noun will require a matching femininedual adjective (madiinataani kabiirataani ‘two big cities’); likewise, an Arabicverb with a masculine singular subject will inflect for masculine singular agree-ment (e.g., ħađar-a kariim-un ‘Karim came’).2 Agreement categories in Arabic

90 Arabic inflectional morphology


include: gender, number, definiteness, and case for nouns and adjectives, andinflection for gender, number, and person for verbs and pronouns.

3.2. Government (ʕamal)

Government is “a type of grammatical relationship between two or moreelements in a sentence, in which the choice of one element causes the selection ofa particular form of another element” (Richards and Schmidt 2010: 249).Traditional Arabic grammar identifies the “governing” element as the ʕaamil‘operator, governor, regent’ (plural: ʕawaamil). The ʕaamil is typically a verb,preposition, or particle that requires its ‘governed’ object to inflect in a particularway (e.g., transitive verbs will take a direct object in the accusative case; a particlesuch as lam requires its following verb to be in the jussive mood; a prepositionrequires its object to be in the genitive case). Sometimes an ʕaamil is not a concreteword but a principle or rule that applies in particular situations (such as, that thesubject of an equational sentence is in the nominative case). This latter type ofgoverning element is referred to in Arabic grammar as an ʕaamil maʕnawiyy, an‘abstract operator’ (as opposed to the overt ʕaamil, the ʕaamil lafZiyy). Inflectionalcategories determined by ‘governing’ elements include only case and mood, bothof which are classified under the Arabic technical term ʔiʕraab, or desinentialinflection.3

When speaking about Arabic “grammar,” most people who study and teachArabic are referring to the principle of governance or ʕamal; but agreement ormuŧaabaqa is equally significant and far more salient because it is usually overtlymarked, whereas government is marked only by desinential or word-final inflec-tion, often in the form of short vowels, which are invisible in ordinary runningwritten text.4

4. Arabic inflectional categories

Inflection for Arabic words includes the following eleven categories.Within each category are subcategories usually referred to as “morphologicalproperties” that are manifested on words in order to show the nature of thecategory. Each category applies to particular form classes, as noted through itsdistributional patterns. Usually, a number of these inflections apply at the sametime, such as tense, person, voice, mood, gender, and number for verbs, ordefiniteness, case, number, and gender for nouns. Certain of these categories areinherent and others contextual (determined by syntax through rules of agreement or

Arabic inflectional categories 91


government). some, such as gender, are inherent in nouns and contextual inadjectives.

Category 1. Tense/aspect (inherent)Properties: present, past, future (for tense); imperfect, perfect(for aspect)Distribution: verbs

Category 2. Person (inherent/contextual)Properties: first, second, thirdDistribution: verbs, personal pronouns

Category 3. Voice (inherent)Properties: active or passiveDistribution: verbs, participles

Category 4. Mood (contextual)Properties: indicative, subjunctive, jussive, imperativeDistribution: verbs

Category 5. Gender (inherent/contextual)Properties: masculine or feminineDistribution: nouns, adjectives, verbs, participles, pronouns

Category 6. Number (inherent/contextual)Properties: singular, dual, pluralDistribution: nouns, adjectives, verbs, participles, pronouns

Category 7. Case (contextual)Properties: nominative, genitive, accusativeDistribution: nouns, adjectives, participles, demonstrative and relativepronouns

Category 8. Definiteness: determiners (inherent/contextual)Properties: definite and indefiniteDistribution: nouns, adjectives, pronouns

Category 9. Comparison (inherent)Properties: positive, comparative, superlativeDistribution: adjectives

Category 10. Deixis (distance from speaker) (contextual)Properties: near, far, (in some cases) fartherDistribution: adverbs and demonstrative pronouns

Category 11. Humanness (inherent)Properties: human/non-human distinctionDistribution: nouns

The inflectional category of humanness is an inherent quality of nouns. It comesinto play in agreement situations when the noun head of a construction is in theplural. If the noun is plural and human, agreement will reflect plurality. If the noun



is plural and non-human (e.g., inanimate objects, animals, abstractions), agreementis deflected to feminine singular.

5. Inflection in Arabic: definitions and examples

As noted above, Arabic has syntactically relevant (or inherent) inflec-tional categories such as number and gender which are less bound to syntax butwhich trigger agreement processes; and it also has syntactically determined(or contextual) categories such as case and mood which are more tightly boundto syntax and even determined by syntactic rules. Inflectional categories orparadigms are characteristic of particular lexical classes. Each class has a rangeof paradigmatic values (“cells” in the paradigm chart) that are distinctive to thatclass. Nouns and adjectives fall into specific “declensions” that show case anddefiniteness, whereas verbs fall into “conjugations.” Verb conjugations in Arabicare extraordinarily regular and predictable; complexities arise, however, whenthe inflectional markers of conjugations encounter weak or defective lexicalroots, with resulting stem allomorphy. Inflectional markers in Arabic may takethe form of prefixes, suffixes, infixes, or circumfixes (also referred to as “trans-fixes” [Bauer 2003: 30–31]). Larcher maintains that the Arabic verb fascinateslinguists because the regularity of its inflection contrasts so starkly with thecomplexities of its derivation and stem variation.5

5.1. Verbs

These “conjugate,” showing six morphological distinctions: tense, per-son, gender, voice, number, and mood.6 Arabic distinction in tense is oftenportrayed as a difference in aspect (perfect and imperfect rather than past andpresent). The difference between these two usages reflects the way time is viewed,either as a linear stretch of points from past to future, or with reference tocompletion of an action (complete or incomplete). Soltan makes the proposalthat “the tense-aspect debate can actually be captured if the language is assumedto have both tense and aspect categories, but that tense is syntactically prominent incertain grammatical contexts, while aspect is prominent in others, with syntacticprominence yet to be defined” (2011: 245). I think that Soltan is on the right track,here, and that ambiguity about tense or aspect relates to speaker/hearer perspective.I would suggest that one could replace Soltan’s “syntactically prominent” with theconcept of “semantically prominent,” inasmuch as it is not so much the grammat-ical structures at play in the differences between tense and aspect interpretation as itis the meaning of such structures and utterance context.

Inflection in Arabic: definitions and examples 93


Verbal expression also includes compound tenses, where the verb kaana is used asan auxiliary with a main verb to precisely denote tense or aspect. These compoundverbs include: past progressive (kaana [past tense] + present tense main verb), pastperfect or pluperfect (kaana [past tense] + past tense main verb), future perfect(present or future tense of kaana + past tense of main verb), and contrary to factcondition (kaana [past tense] + future tense of main verb) (see Figure 4).

5.2. Nouns

These “decline,” showing distinctions for case and definiteness. They alsoinflect for number and (in some instances) gender. Most Arabic nouns have

Active

From I Sound root: AP: PP: VN: ‘to do; to make’

Perfect

Indicative Subjunctive ImperativeJussive

Imperfect Imperfect Imperfect Imperfect ImperfectPerfect

Active Active Active Active Passive Passive

Figure 4 A typical verb conjugation in all moods of the verbSource: From Ryding (2005)



inherent gender (feminine or masculine [although a few are both]), but nouns thatrefer to gendered beings such as male and female authors, engineers, chefs, orsurgeons; or that refer to animals, inflect to indicate biological gender.7

5.3. Adjectives

Adjectives also “decline,” showing the same range of case distinctions asnouns. they also inflect for number and gender. In addition, they inflect forcomparative and superlative.

5.4. Participles (deverbal adjectives)

These inflect as nouns do, but exhibit the additional feature of “voice” –i.e., they are either active or passive.

5.5. Pronouns

These divide into three classes:

5.5.1. Personal

Independent personal pronouns (ʔanaa, ʔanta, ʔanti, huwa, hiya, ʔantu-maa, humaa, naħnu, ʔantum, ʔantunna, hum, hunna) show inflection for number,gender, and person.8 The suffix personal pronouns (-ii/-nii, -ka, -ki, -hu, -haa,-kumaa, -humaa, -naa, -kum, -kunna, -hum, -hunna) realize either possessive func-tion (when suffixed to nouns) or object function (when suffixed to verbs).

5.5.2. Demonstrative

Demonstrative pronouns (haadaa, ‘this’; dhaalika ‘that’ and their variants)in Arabic inflect for number, gender, case (in the dual), and deixis (distance-relation).

5.5.3. Relative

Relative pronouns inflect for number, gender, and case (in the dual). Theyalso exhibit differences in definiteness, with one set (alladhii ‘who, which’ and itsvariants) marked for definiteness (the initial al-) and others, i.e., maa (‘what,whatever’) and man (‘who, whoever’) serving as indefinite relative pronouns.Adjectives and pronouns are “referential” rather than denotational in function.

Therefore they “agree” with nouns and reflect their inflectional categories.

Inflection in Arabic: definitions and examples 95


5.6. Locative adverbs (of time and space)

These may inflect for case under specific conditions:

(1) When the adverb is not followed by an object noun, it is inflected withđamma:

xaraj-tu min taħt-u I emerged from below.

(2) When the adverb is preceded by a preposition, it inflects with kasra:

xaraj-tu min taħt-i l-shajar-i. I emerged from under the tree.

(3) When the locative adverb is followed by a noun in the genitive or apronoun, it inflects with fatħa:

kutub-ii taħt-a l-shajar-i. My books are under the tree.kutub-ii taħt-a-hu. My books are under it.

6. Case, case relations, case theory

As a manifestation of particular importance in linguistic theory, case-marking, case relationships and “case theory” are key areas of analysis. AsLetourneau states, “Case is a concept with deep historical roots in Western andindigenous Arabic grammatical theory” (2006: 347). It is important to distinguishbetween the many uses of the term “case” in linguistics. Case-marking, that is,the overt labeling of nouns and adjectives according to grammatical rules ofagreement and government (i.e., the use of ʔiʕraab), is an area of Arabic grammarthat has received attention from Arabic grammarians since the inception ofindigenous grammatical analysis in the seventh century, and is to this day acentral focus of the teaching of Arabic grammar. On the other hand, caserelationships, in linguistic terms, refer not only the surface structure realizationof grammatical structure, but also the deeper and more abstract relations amongsentence elements that are semantic as well as syntactic. Case theory, especiallyas initiated and developed by Gruber ([1965]1976) and Fillmore (1968 and1977), is an area of theoretical semantics that embeds the concept of “case” inanalysis of predicate-argument structure. Elements of Fillmore’s case theory(“thematic roles”) were ultimately incorporated in generative grammar in theform of theta-rules or theta-structure, which provide case-type labels (e.g.,Agent, Beneficiary, Object) for sentence constituents that identify their relation-ships within the predication.9 Cases, their relationships, their meanings, and theirtheoretical status thus occupy central ground in linguistic theory, but, as yet,



few in-depth studies of Arabic case relationships have emerged, even thoughArabic in this respect has a great deal to offer linguistic theory in general.10 Keystudies of case include the classic work of Louis Hjelmslev ([1935] 1972), Blake1994, Anderson 2006, Butt 2006, and Malchukov and Spencer 2009.Surprisingly, Malchukov and Spencer, an expansive 900-page edited volumethat includes studies of case theories and case systems in a wide range oflanguages and language families, contains no contribution on Arabic.11

7. Key terms for Arabic inflectional morphology

case: “Case is a system of marking dependent nouns for the type ofrelationship they bear to their heads” (Blake 1994: 1). As Andersonpoints out, “the relations expressed by morphological case can beexpressed in other ways, notably by adpositions and position. ‘Case’refers to these common relations; and morphological case is only onekind of ‘case form’, one way of expressing ‘case relations’, or simply‘case’” (Anderson 2006: 2).

declension: a paradigm of case and definiteness inflectional realizationsfor nouns, adjectives, and participles. Standard Arabic has eight nom-inal declensions: triptote (three-way inflection), dual, sound feminineplural, sound masculine plural, diptote (two-way inflection), defective,uninflectable (for case, but showing definiteness), and invariable.12 InArabic, therefore, inflection for number (dual, plural) “can shift a nouninto a different inflectional class” (Ryding 2005: 168).

defective: refers to Arabic lexical roots whose final radical is eitherwaaw or yaaʔ.

desinential inflection: word-final inflection. Arabic case (on substan-tives) and mood (on verbs) are marked at the end of a word. This sort ofinflection is called “desinential inflection” (desinence = termination,ending, suffix). Case and mood inflections in particular are determinedby the role of the word in context, in a sentence. These roles areaffected by governing words, or “operators” – ʕawaamil – which canbe in the form of particular lexical items (such as ʔan + subjunctive orlam + jussive) or through the application of abstract rules (such as “Thesubject of a sentence is in the nominative case”).

diptote: a nominal word class or declension that is restricted to only twoovert case inflections when indefinite: /-u/ for nominative and /-a/ forgenitive and accusative. Diptotes do not take nunation. When they aredefinite, they inflect regularly, as triptotes.

Key terms for Arabic inflectional morphology 97


exponent: in morphology, the formatives or features that realize inflec-tional categories.13

gender: “A grammatical distinction in some languages that allowswords to be divided into categories such as masculine, feminine, orneuter on the basis of inflectional and agreement properties” (Richardsand Schmidt 2010: 240).14 Arabic has two genders: masculine andfeminine.

grammatical word: Aronoff defines a grammatical word as “a lexemein a particular syntactic context, where it [is] provided with morpho-syntactic features (like case and number) and with the morphophono-logical realization of these morphosyntactic features as bound forms.Grammatical words are members of the paradigm of a particularlexeme” (1994: 11). That is, they are instances of words with particular,syntactically relevant inflections.

inflectional class: a form class defined by the nature of its inflectionalparadigm. For example, a verb would be defined by the fact that it takesverb inflections (conjugations); a noun would be defined by the factthat it inflects for noun distinctions (declensions showing case anddefiniteness). That is, the type of paradigm that the word fits intodetermines its “inflectional class.”

number: “A grammatical category used for the analysis of word-classes displaying such contrasts as singular, plural, dual . . . andpaucal (few)” (Crystal 1997: 265). Arabic displays all these numbervariants.

paradigm: a fully inflected model of an example of a form class.Carstairs-McCarthy makes a distinction between the concept of para-digm in general (which he labels “paradigm1”) and a specific paradigmfor a given language (which he labels “paradigm2” ), as follows:Paradigm1: the set of combinations of morphosyntactic properties orfeatures (or the set of ‘cells’) realized by inflected forms of words (orlexemes) in a given word-class (or major category or lexeme-class) in agiven language; Paradigm2: the set of inflectional realizationsexpressing a paradigm1 for a given word (or lexeme) in a givenlanguage (Carstairs-McCarthy 1994: 739). The idea of “cells” withina paradigm refers to the organization of inflectional properties of aform-class into tabular form, each box (or “cell”) in the table represent-ing one of the properties by means of its particular exponents. Forexample, Arabic nouns have three cases (nominative, genitive, and



accusative) and two forms of definiteness (definite and indefinite).Therefore each noun paradigm has six cells:

Inflectional paradigm for bayt ‘house’Definite IndefiniteNominative: al-bayt-u bayt-u-nGenitive: al-bayt-i bayt-i-nAccusative: al-bayt-a bayt-a-n

Sometimes number (singular, dual, plural) is also indicated within an inflectionalparadigm, but Arabic noun inflection for number is often distinct, and falls intodifferent paradigm variants, or declensions (for example, diptote or dual).

stem: the base form of a word without inflections. Stem allomorph: avariant or alternant of the base or stem form of a word, conditioned bythe nature of an affix. For example, a doubled or geminate Form I past-tense verb has two forms, depending on whether the inflectional suffixstarts with a vowel or a consonant. Taking the verb radda, ‘to return; toreply’ for example, if the suffix starts with a vowel, then it is suffixed tothe form radd- (e.g., radd-at, ‘she replied’); if the suffix starts with aconsonant, it is affixed to the form radad- (e.g., radad-naa ‘wereplied’). Stem allomorphy characterizes many Arabic verbs andtheir derivatives, caused by the intersection of inflectional affixeswith phonological rules.

stem class: a particular class of words that exhibits similarity in form andwhich falls into a particular inflectional class (such as verbs in Semiticlanguages).

triptote: an Arabic noun or adjective that shows inflectional distinctionsfor all three cases: nominative, genitive and accusative (muʕrab).

7. Arabic inflectional affixes: (zaaʔida/ zawaaʔid) – ‘augments'

Certain consonant phonemes are used as exponents in the marking ofinflectional properties. Some phonemes (like taaʔ) have a wide range of func-tions; others have more limited marking functions. These consonants are notnormally used alone, but along with vowels as components of inflectionalpatterns or templates. These inflectional affixes are in many cases the sameconsonant phonemes as used for derivational morphology, but have differentfunctions and different meanings when used as inflectional markers. Note thatvowels are prevented by phonological rules from occurring alone as word-initialprefixes.

Arabic inflectional affixes 99


7.1. hamza

7.1.1. Word-initial hamza

This is used inflectionally as follows:

(1) In the first person present tense verb, e.g., ʔa-ktub-u ‘I write’; ʔu-ħibb-u‘I like’; ʔa-staʕmil-u ‘I use.’

(2) In certain plural patterns, e.g., ʔashjaar ‘trees’; ʔaqlaam ‘pencils’;ʔaşdiqaaʔ ‘friends.’

(3) In elative (comparative and superlative) inflections, e.g., ʔakbar ‘big-ger’; ʔahamm ‘more important.’

7.1.2. Word-medial inflectional hamza

This occurs in some noun/adjective plural patterns, such as faʕaaʔil: e.g.,qabaaʔil, jaraaʔid.

7.1.3. Word-final inflectional hamza

This occurs in noun plurals of the type ʔafʕilaaʔ and fuʕalaaʔ e.g.,wuzaraaʔʔ ‘ministers’; ʔaşdiqaaʔ ‘friends.’

It also occurs in color adjectives inflected for feminine gender, e.g., Hamraaʔ‘red f.’

7.2. taaʔ

taaʔ as an inflectional affix occurs as follows:

7.2.1. Word-initial taaʔ

This occurs in the second and third person prefixes of present tense/imperfect verbs: ta-drus-u ‘you m. study/ she studies’; ta-drus-aani ‘you twostudy’; ta-drus-uuna ‘you m. pl. study’; ta-drus-na ‘you f. pl. study.’

7.2.2. Word-final taaʔ

This occurs frequently as a component of gender and number inflections:

(1) In the form of taaʔ marbuuŧa, as the feminine gender inflectionon adjectives and certain nouns, e.g., şaghiira(t) ‘small f.’; ŧabiiba(t)‘doctor f.’



(2) In the sound feminine plural suffix {-aat}, e.g., intixaab-aat ‘elections’;majall-aat ‘magazines.’

(3) In the feminine singular third person past tense verb suffix -at, e.g.,katab-at ‘she wrote, takallam-at ‘she spoke.’

(4) In the first and second person singular past tense suffixes: {- tu}, {-ta},{-ti}, e.g., katab-tu ‘I wrote’; katab-ta ‘you m. wrote’; katab-ti ‘you f.wrote.’

(5) In the second person dual past tense suffix {-tumaa}, e.g., katab-tumaa‘you two (m. and f.) wrote.’

(6) In the third person feminine dual past tense suffix {-ataa}, e.g., katab-ataa ‘they two f. wrote.’

(7) In the second person plural past tense suffixes {-tum}, {-tunna}, e.g.,katab-tum ‘you m. pl. wrote’; katab-tunna ‘you f. pl. wrote.’

7.3. laam

Although laam is not an affix, it is a clitic in the form of the prefixeddefinite article {-l-} which attaches to nouns and adjectives. It therefore functionsas an inflectional marker of definiteness.

7.4. miim

As an inflectional affix, miim only occurs in past tense suffixes, as acomponent of the past tense second person dual suffix, {-tumaa} and the secondperson masculine plural suffix, {-tum}, e.g., katab-tumaa ‘you two wrote’; andkatab-tum ‘you m. pl. wrote.’

7.5. nuun

Inflectional nuun occurs word-initially and in word-final position

(1) As a prefix for the first person plural in the present tense/imperfect:{na-} or {nu-}: na-ktub-u ‘we write;’ nu-rattib-u ‘we arrange.’

(2) In word-final position, nuun forms a part of several suffixes:(a) Past tense/perfect verb suffix {-naa}: katab-naa ‘we wrote.’(b) On nouns and adjectives, the suffix {-n} occurs in the form of

nunation (tanwiin): e.g., kitaab-u-n, kitaab-i-n, kitaab-a-n.(c) On nouns, adjectives and verbs, the suffix {-uuna} is used to

indicate human masculine plural, e.g., mudarris-uuna ‘teachersm.’; kathiir-uuna ‘many m.’; ya-ktub-uuna ‘they m. write.’

(d) On nouns, adjectives, and verbs, nuun also occurs as part ofthe suffix /{-iina}, but this suffix has different distribution andmeanings. For nouns and adjectives, it indicates the oblique



(non-nominative) case of the sound masculine plural inflection, e.g.mudarris-iina ‘the teachers m.’; kathiir-iina ‘many m.’ As a verbsuffix, {-iina} is used to mark the second person feminine singular:ta-ktub-iina ‘you f. write.’

(e) On verbs, the suffix {-na} indicates feminine plural, e.g., ya-ktub-na‘they f. write’; ta-ktub-na ‘you f. pl. write,’ and in the past tense,katab-na ‘they f. wrote.’

(f) On verbs, nouns, and adjectives, nuun is a component of the dualmarkers: {-aani} and {-ayni}: ta-ktub-aani, ‘you two write’ and‘they f. dual write’; ya-ktub-aani, ‘they two m. dual write.’ Fornouns, dual pronouns, and adjectives, the dual suffix {-aani} alter-nates with /{-ayni} according to case restrictions, e.g., kitaab-aani‘two books nom.’ and kitaab-ayni ‘two books gen./acc.’; kabiir-aani‘big du. nom.’ for demonstrative and relative pronouns, the dualinflection is also marked by case, e.g., haadh-aani/ haadh-ayni‘these m. du.. nom./gen.- acc’; alladh-aani/alladh-ayni ‘which/who du. nom./gen.- acc’.

7.6. siin

The inflectional morpheme {sa-} is prefixed to present tense/imperfectverbs to specify future tense, e.g., sa-na-ktub-u ‘we will write.’ The proclitic sawfamay also be used in this way, e.g., sawfa na-ktub-u ‘we will write.’

7.7. yaaʔ

Inflectional consonantal yaaʔ occurs:

(1) As the third person masculine prefix in present tense/imperfective verbs({ya-} or {yu-} in the singular, dual, and plural, e.g., ya-ktub-u ‘hewrites’; ya-ktub-aani ‘they two write’; ya-ktub-uuna ‘they m. pl. write’;and ya-ktub-na ‘they f. pl. write.’ Similarly, yu-rattib-u ‘he arranges’; yu-rattib-aani ‘they two arrange’; yu-rattib-uuna ‘they (m.) arrange and yu-ratttib-na ‘they (f.) arrange.’

(2) As a component of the nominal dual oblique (genitive/accusative) suffix{-ayni}, e.g., kitaab-ayni, ‘two books.’15

Inflectional long-vowel yaaʔ occurs:

(1) In nouns as part of the sound masculine plural oblique (genitive/accusa-tive) suffix {-iina}, e.g., mudarris-iina ‘teachers m. gen./acc.’



(2) In verbs as a component of the present tense/imperfect second personfeminine singular suffix, e.g., ta-ktub-iina ‘you f. write.’

7.8 waaw

Consonantalwaaw is used as an inflectional component in noun plurals ofthe fawaaʕil type, e.g., ʕawaamil, shawaariʕ.Vocalic waaw is used inflectionally as follows:

(1) As a component of the sound masculine plural nominal suffix /-uuna/ ,e.g., mutarjim-uuna ‘translators.’

(2) As a verb suffix, in the present tense/imperfect second and third personsmasculine plural, e.g., ta-ktub-uuna ‘you m. pl. write’ and ya-ktub-uuna‘they m. write.’

(3) As the marker of the third person masculine plural on past tense verbs{-uu}, e.g., katab-uu ‘they wrote.’


(1) Fill out the following chart, noting which inflectional distinctions apply towhich form classes. What sort of patterns do they make? What conclu-sions can you derive about the nature of Arabic inflection?

Inflectional distinctions in Arabic:

verb noun adjective participle pronoun adverb

tensepersonvoicemoodgendernumbercasedef./indef.compar.deixishumanness

(2) Indicate the inflectional categories in the following words:

aljazaa’irusatataħaddathuunaalladhayniʔantumaayatakallamna



(3) Choose ten more Arabic words (of all lexical classes) from a short Arabictext, and analyze their inflectional affixes. Compare your list with yourclassmates. Is there anything that you disagree about?

(4) Within those ten words and their inflectional morphemes, break down themeanings of the affix morpheme components, e.g., for the suffix {-uuna},which part of the suffix indicates case? Which part indicates plural?Which part indicates ‘human’? Can other Arabic inflectional morphemesbe analyzed this way? What are the minimal meaningful components ofArabic inflectional affixes?

(5) In section 7 of this chapter (“inflectional affixes”), the most commoninflectional uses of phonemes have been listed. Can you think of others?

Further reading

Abdul-Raof, Hussein. 2006. Case roles. Encyclopedia of Arabic Language and Linguistics,vol. I, ed. Kees Versteegh, 343–347. Leiden: Brill.

Butt, Miriam. 2006. Theories of case. Cambridge: Cambridge University Press.Kouloughli, Djamel Eddine. 2007. Inflection. In Encyclopedia of Arabic Language and

Linguistics, vol. II, ed. Kees Versteegh, 345–354. Leiden: Brill.LeTourneau, Mark S. 2006. Case theory. In Encyclopedia of Arabic Language and

Linguistics, vol. I, ed. Kees Versteegh, 347–353. Leiden: Brill.Spencer, Andrew. 2001. Morphology. In The Handbook of Linguistics, eds. Mark Aronoff

and Janie Rees-Miller, 213–237. Oxford: Blackwell.Stump, Gregory. 1998. Inflection. In The Handbook of Morphology, eds. Andrew Spencer

and Arnold M. Zwicky, 13–43. Oxford: Blackwell.

Notes

1. “English [is] a language poor in inflection” (Carstairs-McCarthy 1994: 737).2. Verb-subject agreement in Arabic is complex, and depends to a large extent on word

order. This is dealt with in Chapter 11 on clause structure.3. Abbas Hasan, in his extensive Arabic reference grammar, Al-naħw al-waafii, defines

ʕaamil as “what supervenes on a word and thereby affects its ending by making itnominative/indicative, accusative/subjunctive, genitive, or jussive” (maa ya-dxul-uʕalaa l-kalimat-i fa-yu-ʔaththir-u fii $aaxir-i-haa bi-l-raf ʕ-i ʔaw-i l-naşb-i, ʔaw-i l-jarr-i ʔaw-i l-jazm-i) (Hasan 1987: 441).

4. Arabic theory posits a distinction between syntactically-governed inflection and othertypes of inflection. . . within the domain of inflectional theory, Arabic does not distin-guish at all between case and mood, or on the basis of the form-class category of the stem,but instead makes distinctions on the basis of the effect or effects of the “operating”syntactic element. That is, whether the inflection is realized on a noun or on a verb



is irrelevant. What is important is the nature of the inflectional marker itself, asdetermined by the governing element . . . The fact that the suffix markers for mostforms of nominative and indicative are syncretised into one ending, -u, and that the sameapplies for the accusative/subjunctive suffix, -a, gives us some idea of why and how theparallelism between case and mood emerged and was perceived. It was apparentlyestablished on purely formal grounds, by taking the shared phonological features of theinflectional exponents and adducing a unified classification based on their phonologicalidentity. In fact, in medieval Arabic grammatical theory the term for “imperfect indica-tive” verb is muđaariʕ, literally, “resembling,” or “resembler” because, it was said, theimperfect indicative “resembled” nouns in its ability to “inflect,” or change the voweltermination. This is due to the fact that the imperfect form of the verb is the only one thatchanges mood (indicative, subjunctive, jussive), and this feature causes it formally toresemble the change of case in nouns. (Ryding 1993: 176–177)

5. “Thanks to the simplicity of its inflection and the complexity of its derivation andmorphophonology, the Arabic verb continues to fascinate Western scholars of Arabic”(Larcher 2009: 645).

6. See Ryding (2005: 439–440) for discussion of this point.7. See Appendix C for the eight nominal and adjectival declensions.8. Except that there is no first person dual, or gender marking on the first person.9. “The notion of case employed in theories of syntax is an abstract notion which is

used to characterize the interaction between verbal lexical semantics, grammaticalrelations and word order. The overt realization of case must be dealt with bysome component of the theory, however, that component is often left under-specified” (Butt 2006: 11).

10. See the articles on case roles, case theory, and theta-theory in EALL (i.e., Abdul-Raof2006, LeTourneau 2006 and 2009) for an introduction to these concepts as they apply toArabic. Note also that Butt fortunately refers to “the Arabic tradition” as part of her“foundational perspectives” on theories of case (Butt 2006: 18–20). Hjelmslev notesthat in languages that have few cases (as does Arabic), then those cases tend to have veryabstract (rather than localist) meanings. “Dans un language tel que le grec qui est (selonla grammaire traditionelle) pauvre en cas et riche en prépositions, la signification d’uncas est très abstraite par rapport à celle d’une préposition” (1935 1972: 41).

11. It does contain one article on a Semitic language, Amharic. See Amberber (2009).12. For full illustrations of these eight declensions see Ryding (2005: 182–204).13. “The features which identify a morphosyntactic property may be referred to as its

exponents” (Matthews 1974: 144) (emphasis in original).14. “The classification of nouns into different genders is quite an intriguing phenomenon

because of its strong arbitrariness” (Booij 2007: 129).15. Sometimes this form of yaa’ is transliterated using an /i/ (e.g., kitaab-aini). In keeping

with the phonological rule against vowel combination or vowel adjacency in Arabic,however, I believe it is more accurate to indicate the diphthongal nature of the sequenceas /-ay-/, e.g., kitaab-ayni.



9

Syntactic analysis and Arabic

1. The study of syntax

Benmamoun provides a definition of syntax as follows: “Syntax is thestudy of phrasal and sentential patterns of natural language. It is the engine thatcombines the sound/gesture and meaning components of language. Syntax dealsprimarily with how words combine to form phrases and sentences, and the depend-encies that obtain between the constituents of the phrase or sentence” (2009: 391).Thus, the study of syntax deals with phrase structure and clause structure – the waythat words interrelate to form coherent, meaningful, and grammatically acceptablesentences. “A linguist . . . will try to characterize the principles that determine theformation of [Arabic] sentences. The goal will be to provide a systematic descrip-tion of [Arabic] sentence formation, the grammar of [Arabic]” (after Haegeman1994: 4).1 In order to undertake the study of syntax, it is necessary to make certaindistinctions between form and function of lexical items within sentences fordiscussing surface structure phenomena.A first step is to distinguish labels of linguistic “forms” or “form classes” (such

as noun, verb, adjective) from the labels of their linguistic functions in context(e.g., terms such as subject, object, predicate). This enables discussion of thenature of individual words (such as their derivation, meaning, or inflection)separately from the syntactic slots or functions that they fill when used in context,as syntactic constituents. At a more abstract level of analysis, constituency may beviewed from a number of angles that involve hierarchical relations, semanticrelations, and various theories of dependency.

1.1. Traditional Arabic syntax: ʕamal and mutaabaqa

The rules and structures of Arabic syntax have long been the object ofstudy within the Arabic grammatical tradition, under the rubric of the term naħw.The scope of naħw includes both the analysis of combinatorial sequences of wordsin context and the effect of syntactic regulations on the inflection of words. The twokey components of traditional Arabic syntactic analysis are the functions of

107


government (ʕamal) and agreement (muŧaabaqa), as discussed and defined inChapter 7. The rules of government (ʕamal) require that words in context obeythe requirements of syntactic operators (ʕawaamil) through the processes ofdesinential (word-final) inflection. Rules of agreement and anaphora require thatmodifiers, verbs, pronouns, and other referents to a particular noun head of aconstruction, must be in concord with, or match the nature of that noun inall respects: semantic (e.g., inflection for number and gender) and grammatical(e.g., inflection for case). The classic work in English on modern Arabic syntax isCantarino’s three-volume opus, The Syntax of Modern Arabic Prose, published inthe 1970s. Nothing matches it in terms of coverage of literary Arabic. However,Cantarino omits media Arabic from his work, and media Arabic is easily the mostfar-reaching and widely accessed form of Arabic today. It is also what manycontemporary students are studying and want to study.

2. Syntactic theory

In order to provide a linguistic overview of Arabic syntax, it is useful toreview some theoretical approaches to syntax and their fundamental assumptions.Choosing what to provide as an introduction to ways of analyzing syntax is,however, a daunting task. As one linguist puts it:

The available literature is vast. The consensus on any particular analysis,however, is minimal. It is therefore a challenge to illustrate the basic ideasand assumptions comprehensively without also introducing the complete formalmachinery and the various discussions which argue for or against a particularsolution. (Butt 2006: 46)

In western linguistics, especially of the American school, structural linguisticsapproaches using immediate constituent (IC) analysis were predominant until thelate 1960s.2 But with the emergence and development of the work of NoamChomsky in the 1960s and 1970s, the major theoretical paradigm became that ofgenerative grammar.3 The key difference between structural approaches to syntaxand generative ones is that structural linguistics focuses on the organization of‘surface’ structure, i.e., language as it is used, whereas generative syntax focuseson the cognition of language, i.e., language and mind.4 Probably the most salientfeature of the shift of linguists’ attention to “generative” syntax was the emergenceof the importance of abstract mental structures that underlie sentence structure andmeaning; that is, instead of examining only the surface structure of language, or itsovert structure, attention shifted to abstract levels of language called “deep struc-ture,” conceptual structures that operate to produce language.

108 Syntactic analysis and Arabic


The most important point of [Chomsky’s] position is this; the goals of linguistictheory are psychological. Language is a mental phenomenon, to be studied assuch, and theories of language are to be considered psychological theories. So theobject of study is the human mind, and it is the nature of the human mind asreflected in the acquisition and use of language that provides the central questionsof the field. (Green and Morgan 1996: 2)

With the success of the Chomskian approach to linguistic analysis, syntax emergedas the central component of general linguistic theory.5 Generative grammar calledattention to the fact that syntactic rules and operations operate at a very abstract andperhaps even autonomous level of cognition, as illustrated by Chomsky’s famousexample, ‘Colorless green ideas sleep furiously.’As one linguist observes, “syntaxworks (i.e., makes sentences seem right, somehow) independently of any reason-able context of use, and even in the absence of interpretable meaning itself”(Hall 2005: 157). The most widely applied and widely published research articlesand books on theoretical syntax often derive from the Chomskian school ofgenerative syntax, but that school itself has undergone many refinements andextensions since the emergence of Standard Theory in the 1970s and 1980s.Perhaps the most widely known is Government and Binding Theory, later versionsof which were termed Principles and Parameters Theory. Most recent work in thisvein is usually done through the prism of the Minimalist Program and post-Minimalist theory. These latter works are characterized, however, due to thesuccess and explicitness of the original theory, by the fact that they take a greatdeal of highly explicit technical terminology for granted (by using what onelinguist calls “baroque technical terms”).6

It is therefore difficult for those outside this disciplinary subfield to read andcomprehend its writings in all their detail and theoretical ramifications.7 Moreover,the study of morphology as well as semantics has in many ways moved beyond thelimits of syntactic theory, especially in terms of the study of case.8

2.1. Sentential syntax

The essentials of Arabic sentence structure are described here not from anyparticular formal theoretical viewpoint, but with regard to providing a generalframework for further study. This is done to provide options to the reader in viewof the fact that many publications on Arabic syntax published in the past twentyyears have employed a generative approach, which focuses on syntactic hierarchiesand relationships within the Principles and Parameters framework (Chomsky 1981),and the subsequent Minimalist framework (Chomsky 1995, 2000). Whereas this is apopular and pervasive approach to the analysis of Arabic syntax (especially in theUnited States), it is also undertaken with particular goals of exploring how Arabic

Syntactic theory 109


syntax relates to human cognitive language faculties in general, and ultimately, theissue of Universal Grammar (UG). Rather than this, a more theory-neutral approachseems called for in an introductory study of linguistics and Arabic. I will, therefore,in addition to discussing generative theory, incorporate elements of basic linguistictheory (BLT) (Dixon 2010a and 2010b), construction grammar (Goldberg 1992 and2006), case grammar (Fillmore 1968 and 1977) and predicate-argument structure(Goldberg, Levin and Rappoport, Pinker) in discussing the elements of Arabicsentential syntax.

2.2. Strengths and limitations of generative grammar

The most useful aspects of generative theory for descriptive purposes are(1) the rigor of its logical argumentation and (2) its development of abstractrelations and formalizations to explain grammatical regularities.9 The use of tree-diagrams (hierarchical structures with branching nodes) to illustrate relationshipswithin clauses and sentences has been a salient element of generative theory, as hasthe incorporation of case theory, which studies the nature and number of partic-ipants in any predication, and their thematic roles (or “theta” roles).10 This latterarea of syntactic theory interfaces with Arabic morphology to a great extent, andshould be considered “morphosyntax.” In fact, I think that to a great degree, Arabicsyntax is so deeply interwoven with inflectional morphological structure that thecentral features of Arabic syntactic theory fall under the category of morphosyntax.This is why agreement and government structures are key to understanding Arabicsyntactic dependencies, and why the issue of case relationships (overt and abstract)are of special interest.

2.3. Predicate/argument syntax and valence theory

In predicate/argument approaches to syntax, the predicate (usually a verb,but prepositions as well) is key to determining the structural roles of othercomponents of a phrase, clause, or sentence.11 “The predicate is the nucleus of aclause. The word – generally a verb – that is placed in the predicate slot willdetermine the number and type of arguments which the predicate takes. . . Themeaning of a [predicate] determines the kinds of noun which can fill a coreargument slot” (Dixon 2010a: 98).12 The key analogy made here is chemical,comparing the predicate to an atom surrounded by a specific number of electronswhich determine its ability to bond chemically with other atoms (the “valence” ofan atom). But a second analogy is also logical and mathematical, relating to thefield of predicate calculus, which is a way of stating how certain objects, or“arguments,” relate to a predicate, and of mapping those arguments to appropriatetruth values.13 There are therefore two analogies applicable to the constructions of



predicates, both of which share the image of a central force surrounded by andbounded by entities that lie within its sway and which are cast into ‘roles’ inexpressing syntactic argument functions.14

The concept of valence (or valency) deals with the number and nature ofsemantic roles that are associated with the meaning of a particular verb (mostoften, one to three roles); some roles are central, others peripheral. These roles havebeen labeled differently by different authors as “cases,” “arguments,” “theta roles,”and “functions.” Research based on argument structure designates those semanticroles as to their syntactic function (Agent, Patient/Object, Beneficiary/Recipient),or refers to them with semantically neutral labels such as X or Y.15 By “decom-posing” predicate meaning through analysis of the interaction between predicateand arguments, one can discover interrelated semantic and syntactic regularities, a“set of principles for relating semantic representations with facts about grammat-ical form and the structural organization of sentences” (Fillmore 1976: ix).16

Designations of cases or arguments vary, depending on authors’ preferences.Although case frames and function labels are not directly equatable to traditional orpre-theoretical grammatical terms (such as “subject” or “object”), some of thefollowing apply. For the traditional notion of indirect object, Fillmore 1968 used“Dative” and Chafe used “Beneficiary,” whereas Goldberg and others use “recip-ient.” For the traditional concept of direct object, Fillmore used “Object” or“Objective,” Blake and others have used “Patient,” and others use the term“Theme,” introduced by Gruber ([1965], 1976).Arabic is, from a surface-structure viewpoint, a nominative/accusative lan-

guage with the genitive as the third separately marked case in inflectionalparadigms. All case-type relations are therefore marked with one of thesethree cases. The dative case, for example, is not separately marked in Arabic;datives (Recipients) are marked either as accusative (in ditransitive structures),or genitive, as object of the benefactive/allative preposition li- prefixed to theRecipient argument.17

3. Topics and approaches in Arabic syntax

As previously noted, much of the recent work done in Arabic syntacticanalysis has been done from the generative standpoint, ranging from the Principlesand Parameters framework to Minimalist and post-Minimalist frameworks. Onestrength of these studies is that they have examined spoken (especially Egyptian,Lebanese, and Moroccan) as well as written Arabic, and have yielded insightsinto both.18

Topics and approaches in Arabic syntax 111


The most prominent issues that have received a great deal of attention in the lastfour years include lexical and functional categories, clause structure and VerbSubject (VS) order, agreement, particularly the correlation between the richness ofagreement and the position of the subject relative to the verb, verbless sentencesand copular constructions, negation, questions, relatives, the status of the ‘subject’in the Subject Verb (SV) order, Case, Tense/Aspect, and the structure of the nounphrase, with the so-called Construct State getting more scrutiny. (Benmamoun2009: 391–392)

From the standpoint of general syntactic knowledge, research on Arabic syntaxhas brought attention to its differences from and its commonalities with otherworld languages. “The picture that emerges is that the syntax of Arabic isnot . . . radically different from the syntax of other languages” (Benmamoun2009: 399–400).

4. Predication analysis, agreement, and government

Construction grammar, case grammar, frame semantics, and the predicatedecomposition approach focus on the central role of predication and the manner inwhich it determines the semantic/syntactic relationships within phrases and clauses(predicate/argument structure). These approaches have the advantage of dealingwith syntactic and semantic relationships by casting a wide and yet fine-gauge netaround predication structures and then being able to sort and assign sentencecomponents to their particular, context-determined roles. The incorporation ofsemantics into syntactic analysis works very well for Arabic in particular, wheremeaning-based relationships are often determined by the meaning potential of thelexical root combined with grammatical templates, and the valency-changingnature of the Arabic verb template system. The following chapters will discussArabic syntactic structures through illustration of agreement and government, thenature of decompositional procedures, case roles, and grammatical relations at thephrase level and the clause level.

5. Syntactic terminology

Terminology and shorthand notation of terms form at the same time oneof the key advantages and disadvantages of contemporary syntactic theory. Theshorthand notations in particular are advantageous in writing and illustratingformalizations of rules, but some are not intuitively comprehensible, and needto be learned as part of syntactic study. As noted earlier, a thorough grasp of



syntactic notation is usually taken for granted in writings within the generativeframework; other approaches are less opaque. Here are some terms that are usefulin the study of syntax.

ACC.: accusativeadj.: adjectiveadv.: adverbAgent: in theta theory, “the one who intentionally initiates the action

expressed by the predicate” (Haegeman 1994: 49)AGR.: agreementaux: auxiliary verb; kaan may be used as an auxiliary verb in Arabic

(e.g., kaan-at ta-drus-u, ‘she was studying’)Beneficiary: in case theory, the person or entity that receives the benefit

of the action. See also Recipient.C: constituent (in IC analysis)/complement or complementizer in gen-

erative syntaxcategorical specification: labeling of lexical or syntactic categories

(e.g., noun, verb, adjective, preposition)complement: object or object-clause; “a term used in the analysis of

grammatical function, to refer to a major constituent of sentence orclause structure, traditionally associated with ‘completing’ the actionspecified by the verb” (Crystal 1997: 75)

complementizer: a subordinating conjunction that joins clauses (such asthat or whether in English; or such as ʔinna and her sisters in Arabic),creating an embedded sentence or complement

CP: complementizer phraseDAT.: dativedet.: determinerDP: noun phrase (“In recent work it has been proposed that the head of

NP is not N but rather the determiner. NP is reinterpreted as DP”(Haegeman 1994: 99).

head: the main or central part of a phrase; the head of an NP is a noun, thehead of a PP is a preposition, etc.

I: inflection (also, INFL.): the category (in generative syntax) INFLcontains all verbal inflection, including tense, person, and number, aswell as agreement features (AGR.). “Inflection” is actually consideredthe head of a sentence (IP) because it controls the inflectional propertiesof the verb.19

IC: immediate constituent. In syntactic analysis, a grouping of wordsthat form a construction, such as a noun phrase, a relative clause, or

Syntactic terminology 113


prepositional phrase. Immediate constituents are usually reducible to“ultimate constituents”

IP: inflection phraselexical category: the form class of a particular word, e.g., noun (n.), verb

(v.), adjective (adj.), preposition (p.)20

NP: noun phraseO: object (in discussions of word order)Patient: in theta theory, “the person or thing undergoing the action

expressed by the predicate” (Haegeman 1994: 49)PP: prepositional phrasePS: phrase structureRecipient: (also Beneficiary): “person [or entity] that benefits from the

action expressed in the predicate” (Haegeman 1994: 50)S: sentence; subject (in discussions of word order, e.g., SVO)Spec: “specifier” (e.g., article, demonstrative pronoun, quantifier for an

NP; a qualifier for a VP)SVAA: subject–verb agreement asymmetrysyntactic categories: “the syntactic category to which a word belongs

determines its distribution, that is, in what contexts it can occur”(Haegeman 1994: 36). See note 20 to this chapter.

T: tensetheme: in discussions of theta theory, “the person or thing moved by the

action expressed by the predicate” (Haegeman 1994: 49), sometimesamalgamated with the term “Patient”

UG: universal grammarultimate constituents: the morphological units that constitute a phrase

or clause. Ultimate constituents are said to be “irreducible.”V: verbVP: verb phrasewh.: shorthand for “wh-word” that is, a question word that starts with the

sequence wh- but also includes “how.” It also refers to wh- relatives,that is, English relative pronouns such as “who” and “which”

XP: here “X” is a generalization that stands for any syntactic categorythat is a “head” in a phrase; thus XP might stand for NP, VP, or PP orother types of phrases




(1) The Arabic term naħw is often used as the equivalent of the term“syntax.” Look up a definition of naħw and see what it covers explicitlyin Arabic, and compare it to the English term, “syntax.”

(2) Syntax deals with word order and word groupings in sentences. In fact, itis sentence-based analysis (rather than discourse-based). Do you think thesentence (Arabic jumla) is the most central or most useful component ofthe study of grammar? If so, why? If not, what do you consider the mostcentral component of Arabic grammar?

(3) Some approaches to syntax consider it a system of sentence structureindependent of meaning (e.g., Chomsky’s “Colorless green ideas sleepfuriously”). Sibawayhi in chapter 6 (baab al-istiqaama min al-kalaamwa-l-ʔiħaala) (‘Chapter on correctness and deviation in speech’) of hiseighth-century work Al-Kitaab, gives the example “sawfa ʔashrab-umaaʔ-a l-baħr-i ʔams-i” (‘I shall drink sea-water yesterday’), calling it“al-muħaal al-kadhib” (‘impossible and false’) (1991: 26). DiscussSibawayhi’s example and compare it with Chomsky’s. How are theydifferent?

(4) Read the entire chapter 6 of Al-Kitaab (it is only about one page long),translate it into English, and discuss Sibawayhi’s categorization of sen-tences. What are his criteria for correctness? How does he distinguishbetween form, meaning, and distribution of constituents?

(5) Read the EALL article “Syntax” by Benmamoun and prepare a discussionof one of the five sections/topics that he deals with. What do you think arethe strengths and weaknesses of his analysis? How would you approach adescription of Arabic syntactic theory?

Further reading

Benmamoun, Elabbas. 2009. Syntax. In Encyclopedia of Arabic Language and Linguistics,vol. IV, ed. Kees Versteegh, 391–402. Leiden: Brill.

Benmamoun, Elabbas and Lina Choueiri, 2013. The syntax of Arabic from a generativeperspective. In The Oxford Handbook of Arabic Linguistics, ed Jonathan Owens,115–164. Oxford: Oxford University Press.

Fillmore, Charles. 1968. The case for case. In Universals in Linguistic Theory, eds.Emmon Bach and Robert T. Harms, 1–88. NewYork: Holt, Rinehart, andWinston, Inc.

Green, Georgia and Jerry L.Morgan. 1996. Practical Guide to Syntactic Analysis. Palo Alto,CA: CSLI Publications.

Haegeman, Liliane. 1994. Introduction to Government and Binding Theory. 2nd edn.Oxford: Blackwell. Introduction and Part 1.



Maalej, Zouhair. 2009. Valency. In Encyclopedia of Arabic Language and Linguistics,vol. IV, ed. Kees Versteegh, 624–627. Leiden: Brill.

Sibawayhi, Abu Bishr ʻAmr bn ʻUthman bin Qanbar. 1991. Kitaab Siibawayhi, ed. ʻAbdal-Salam Muhammad Harun. Beirut: Daar al-jiil. Chapter 6.

Notes

1. Haegeman refers to English in the original text, but this quote applies to Arabic (or anyother language) as well.

2. Immediate constituent (IC) is “a term used in grammatical analysis to refer to the majordivisions that can be made within a syntactic construction” (Crystal 1997: 190). The ideais to divide constituents into ever smaller components until the parts of a sentence werereduced to their smallest form (free or bound morphemes). In a sentence such as “The girljumped,” the immediate constituents would be “the girl” and “jumped.” Then each ofthese would be further broken down into “the+ girl”, and “jump + -ed.” The final, reducedform of a sentence that has undergone IC analysis is said to consist of its “ultimateconstituents” – the morphologically irreducible components that cannot be furtheranalyzed (such as the {-ed} morpheme in the previous sentence.

3. For a concise summary of major developments in twentieth-century linguistic theory seeFerguson (1992).

4. Another way to put this is to use Chomsky’s distinction between performance andcompetence. Competence is the internal, conscious or unconscious abstract model oflanguage in the speaker/hearer’s head (“I-language”). It underlies the ability to uselanguage. Performance is the actual use of language – how it is externalized and howhumans use rules and models in accomplishing real-world language tasks(“E-language”).

5. “Some syntacticians like to argue that their area of linguistics is at the very core of humanlanguage, since it is the only component of our mental grammars that directly interfaceswith neither sound nor meaning, both of which lie outside of language: phonologyconnects sound with the lexicon; the lexicon and morphology mediate between sound,syntactic patterns and meaning; and semantics connects the lexicon and syntax withmeanings in conceptual systems” (emphasis in original) (Hall 2005: 154).

6. Aronoff (1994: 2), where he also refers to the sometimes excessive coining of new terms,taxonomies, and abbreviations in generative grammar as its “terminological ebullience.”

7. “The technical terms in contemporary syntactic theory tend to be very unstable and short-lived; how linguist X defines a term in a certain paper may differ from the way linguist Yuses it in a different paper. . . We urge readers to use their wits to track down whatparticular linguists mean by the terms they use, and to keep in mind that it isn’t alwayspossible to tell exactly what a term is being used to refer to – sometimes writers fail to sayexactly what they mean by some term that figures crucially in their analysis. It is not anacceptable practice, but it sometimes happens anyway” (Green and Morgan 1996: x).

8. For example, Lexical-Functional Grammar (Butt 2009), Lexical Semantics (Lieber2004), Case Grammar (J. Anderson 2006 and 2009, and Spencer 2009).



9. Helpful introductions to generative theory include Radford (1988), Haegeman (1994),and Green andMorgan (1996). Aoun, Benmamoun, and Choueiri (2010) use generativetheory to analyze a number of key Arabic syntactic structures.

10. For an introduction to argument structure and theta roles see Haegeman (1994: 42–60);for a discussion of theta roles in Arabic see LeTourneau (2009).

11. “Le verbe est au centre du nœud verbal. . . Il est donc le régissant de toute la phraseverbale” (Tesnière 1959: 103).

12. In generative theory, it is said that the verb “theta-marks” its arguments, or clauseconstituents. It is also said to “theta-govern” its object, but not its subject (Haegeman1994: 87).

13. The foundation of the predicate-argument approach can be found in predicate calculusas applied to language structure by German logician and mathematician Gottlob Fregein his Grundgesetze der Arithmetik (1893/1903; 1962).

14. “Most theories of case today assume that predicates (verbs, nouns, prepositions, andalso adjectives) come with some kind of underlying specification as to their argumentstructure, that is, a specification as to the number and semantic type of participant rolesinvolved” (Butt 2009a: 33).

15. Labels of semantic roles are usually capitalized.16. In the predicate decomposition approach, “the verb’s meaning is represented using

members of a fixed set of primitive predicates together with constants – typically chosenfrom a limited set of semantic types” (Levin and Rappoport-Hovav 1998: 251).

17. The dative “case” is a widely studied subject in case analysis and in theoreticalapproaches that include aspects of semantic and syntactic roles typically indicated bycase-type relations (e.g., theta-roles, frame semantics, construction grammar, lexicalsemantics). See Ryding (2011) for a discussion of ditransitive relations in Arabic syntax.

18. For recent work in agreement systems, negation, and wh-questions, see especiallySoltan (2006, 2007, 2009, 2010, 2012).

19. Haegeman states: “We propose that INFL, a category of the zero level. . . is the head of S(sentence). If we assume that S is headed by INFL, it follows that S, like other phrasalcategories, such as VP, is endocentric; it is a projection of I, IP” (1994: 114). I find thisrole for “inflection” perplexing because it inserts a morphological category at thehighest level in the hierarchy of a syntactic structure and thus morphology is given acentral role in syntactic analysis. Although I acknowledge the centrality of morpho-syntax, especially in Arabic, the privileged nature of “inflection” here, as I understand it,only applies to verbs, not other syntactic categories such as nouns or adjectives. It is thusa restricted notion of inflection.

20. Lexical categories are sometimes referred to as “syntactic categories” and to someextent as “grammatical categories,” depending on the author or the system being used(in non-linguistic terms, they are usually referred to as “parts of speech”). In general,lexical categories refer to single words, rather than to phrases, whereas “grammaticalcategories” in generative grammar may refer to larger groupings such as sentences (S.),noun phrases (NP), or verb phrases (VP).



10

Arabic syntax I: phrase structure

1. Arabic phrase structure

Arabic syntactic study can be undertaken from several perspectives, asnoted in the previous chapter. Phrases and clauses are the two key sites of syntacticanalysis; phrases are organized groups of words that fill particular functions withinsentences, but which also have a certain integrity and rule-structure of their own.Phrases have no predication (for example, haadhihi l-şuurat-u ‘this picture’ oral-bayt-u l-ʔabyađ-u ‘the white house’). Clauses (or sentences) involve a predi-cation of some kind (for example, haadhihi hiya l-şuurat-u ‘This is the picture,’ oral-bayt-u ʔabyađ-u ‘The house is white’). This chapter focuses on Arabic phrasestructure; the following chapter will focus on clause structure.1

As noted earlier in the discussion of Arabic morphosyntax, the dominantprinciples of Arabic syntactic structure are agreement and government. Theseprevail in both phrase structure and clause structure, but in different ways. In thischapter I will first discuss agreement-based phrase structure and then government-based phrase structure.

2. Agreement-based phrase structures

Phrases consist of “heads”and other phrase components. The head of aphrase determines its syntactic category and its distribution within a sentence.2

Typical agreement-based Arabic phrase structures include the following.

2.1. Noun phrases (NPs)

2.1.1. N + N: nouns in apposition

Nouns in apposition co-identify and co-specify each other, each referringto the other and therefore acting as one syntactic unit. They are in a balanced

119


relationship and together constitute one common type of Arabic phrase structure.Normally, they agree with each other in case.3 For example:

al-malik-u ħusayn-un King Husseinfii l-waqt-i nafs-i-hi at the same timeal-ʔumm-u maryam-u Mother Mariam4

2.1.2. N + N (+N): Nouns in coordination

Nouns in coordination are linked by the coordinating conjunctions wa-‘and’ or ʔam ‘or, or ʔaw ‘or.’ Syntactically, they behave as a coordinated group,filling a single syntactic slot, such as Agent, Object, or Beneficiary. Each memberof the coordinated group carries the same case-marking. Unlike nouns in apposi-tion, each term of the coordination structure counts as an individual, so that nounphrases with nouns in coordination are usually counted as two or more in terms ofreference and agreement structures.

al-raʔiis-u ʔaw-i -l-waziir-u ‘the president or the minister’al-thaʕlab-u wa-l-ghuraab-u ‘the fox and the crow’kariim-un wa-rashiid-un wa-qaasim-un ħadar-uu l-muʔtamar-a.Karim, Rashid, and Qasim attended (3 pl.) the conference.

2.1.3. N+adj. (+adj.)

Arabic noun–adjective phrases require multiple agreement: in case, def-initeness, gender, and number.

al-hilaal-u l-xaSiib-u the Fertile Crescentthe-crescent (m.)-nom. the-fertile m. nom.

marduud-u-n ʔiijaabiyy-u-n a positive yieldyield (m.)-nom.-indef. positive-(m.)-nom. -indef.

madiinat-u-n faransiyyat-u-n kabiirat-u-n a big French citycity (f.)-nom.-indef. french-f -nom.-indef. big-f.-nom.-indef.

2.1.3.1. Deflected agreement The term ‘deflected’ agreement refers tothe Arabic use of feminine singular agreement with non-human nouns in the plural:5

fii l-ʔaʕwaam-i l-ʔaxiirat-i in recent yearsin the-years-gen. the-recent-f.-gen.al-dhiʔaab-u l-ramaadiyyat-u the gray wolvesthe-wolves-nom. the-gray-f.-nom.

120 Arabic syntax I: phrase structure


2.1.4. Agreement in DPs: Arabic determiner phrases6

Determiners may be of various types in Arabic: articles, personal pro-nouns, or demonstrative pronouns. The definite article does not exhibit variation ingender. Possessive personal pronouns agree semantically with their referents, notwith the noun to which they are attached. Demonstrative pronouns, however, mustexhibit agreement with their nouns in case, gender, and number. In terms ofdefiniteness, the semantics of demonstratives require that the noun be definite.7

haadhaa l-maŧʕam-u ‘this restaurant’this-m. the-restaurant (m.)-nom.haadhaani l-miftaaħ-aani ‘these two keys’these-two-m. the-two-keys-m.-nom.ʔuulaaʔika l-baaħith-uuna ‘those researchers’those the-researchers-m.-nom.

3. Government-based phrase structures

In some cases, phrases exhibit internal governing; that is, one part of thephrase governs another and causes it to inflect for a particular case, usuallygenitive. The two main categories in this respect are prepositional phrases andthe genitive ‘construct’ phrase, or ʔiđaafa.

3.1. Prepositional phrases (PP)

Prepositions and semi-prepositions (locative adverbs, Zuruuf) in Arabicrequire the genitive case on the following noun. If the object is a pronoun, it is insuffix form.preposition plus noun

fii l-jaamiʕat-i at the universityat the-university-gen.

‘semi-preposition’ plus noun8

baʕd-a l-ħaflat-i after the partyafter-acc. the-party-gen.

preposition plus pronoun

min-haa from herfrom-her

Government-based phrase structures 121


‘semi-preposition’ plus pronoun

fawq-a-naa above usabove-acc.-us

3.2. N + N genitive construct, ʔ idaafa

The ʔiđaafa, variously translated as the ‘genitive construct,’ or ‘annex-ation structure,’ is one of the most productive, distinctive, and frequently occurringphrase-types in Arabic, used to indicate an entire range of meaningful relationshipsbetween entities, from possession (maktab-u ʔabii ‘my father’s office’), to identity(madiinat-u baghdaad-a ‘the city of Baghdad’), to determination/quantification(kull-u l-mudarris-iina ‘all the teachers’), to measurement (nişf-u finjaan-in ‘half acup’), to composition (timthaal-u dhahab-in ‘a statue of gold’), to contents(finjaan-u qahwat-in ‘a cup of coffee’), and even to activities (e.g., wuşuul-ul-waziir-i, ‘the arrival of the minister’). The essential logical notion behind theNP-internal ‘possessive’ construct is that of connection. The primary syntacticfeature is that the first term (al-muđaaf ) “governs” the second (al-muđaaf ʔilay-hi)in the genitive case.9 The first term, as head of the construction, takes the caserequired by its syntactic function. Thus the two terms are tied together in a close-knit construction. Other morphosyntactic restrictions apply:

(1) The first term may carry neither the definite article, nor nunation; that is, itis unmarked for definiteness.

(2) The second term defines the first through its definiteness or indefiniteness.That is, it may be definite (by virtue of the definite article, or by virtue ofits being a proper noun) or indefinite, marked typically by nunation.

The two terms in this way impose mutual restrictions: the first term carries the case-marker of syntactic function; the second term carries the definiteness marking. Thestrength of the binding between these two nouns is such that they cannot beinterrupted by any other word, except for a demonstrative pronoun modifyingthe second term (nişf-u haadhaa l-finjaan-i ‘half of this cup’). Any modifiers mustfollow the entire ʔiđaafa structure (maktab-u ʔabii l-jadiid-u ‘my father’s newoffice’). Overlapping relations may occur when a complex relationship amongitems needs to be expressed, such as wuşuul-u malikat-i l-ʔurdun-i, ‘the arrival ofthe queen of Jordan.’ In such a case, all terms but the first are in the genitive case;all but the last are unmarked for definiteness.10



lawn-u l-baab-i the color of the doorcolor-nom. the-door-gen.zuʕamaaʔ-u qabaaʔil-a leaders of tribesleaders-nom. tribes-indef.-gen.kull-u l-banaat-i all the girlsall-nom. the-girls-gen.kiis-u fustuq-i-n a bag of nutsbag-nom. nuts-gen.-indef.

One aspect of construct structure which has been of interest to linguists is theability of the first term – if it is a verbal noun derived from a transitive verb – togovern a following object noun in the accusative:

qabl-a mughaadarat-i l-raʔiis-i l-ʕaaşimat-abefore-acc. leaving-gen. the-president-gen. the capital-acc.before the president’s leaving the capital

This kind of iđaafa NP is analyzed by some researchers as a reduced sentence,with the subject marked as genitive, the object as accusative, and the action (verbalnoun) open to taking whichever case is required by its function within a largerpredication.11

4. Summary

The two phrase types – agreement-based and government-based – formthe core of Arabic phrasal (non-predicational) syntax. Note that the identificationof the category of a phrase is not the same as identifying its function within syntax.For example, a prepositional phrase (PP) may function as an adjective phrase (AP),as in: “The student in her office is from Yemen.” Where “in her office” describesthe subject of the sentence. Or a PP may function as an adverbial expression as in“We were reading in her office,” where “in her office” describes the location of theaction. In linguistic description and argumentation it is essential to be able todistinguish form from function at all levels.


(1) Discuss the difference between phrases and clauses and give five exam-ples of each in Arabic.

(2) Examine the nouns-in-apposition examples listed above. Why are noneindefinite? Can indefinite appositions occur?

Summary 123


(3) Demonstrative pronouns in English include ‘this,’ ‘that,’ ‘these,’ and‘those.’ They are said to indicate “deixis.” Look up a definition of deixisand discuss how Arabic handles this feature.

(4) The semantic relations between the parts of the genitive construct aremany and complex. How many can find examples of and list?

Further reading

Badawi, El-Said, Michael G. Carter, and Adrian Gully. 2004. Modern Written Arabic: Acomprehensive grammar. London: Routledge (especially on annexation structures130–143).

Belnap, R. Kirk and Osama Shabaneh. 1992. Variable agreement and nonhuman plurals inclassical Arabic and modern standard Arabic. In Perspectives on Arabic LinguisticsVol. IV, eds. Ellen Broselow, Mushira Eid, and John McCarthy, 245–262.Amsterdam/Philadelphia: John Benjamins.

Hoyt, Frederick. 2008. Noun phrase. In Encyclopedia of Arabic Language and Linguistics,vol. III, ed. Kees Versteegh, 428–434. Leiden: Brill.

Procházka, Stephan. 2008. Prepositions. In Encyclopedia of Arabic Language andLinguistics, vol. III, Kees Versteegh, ed., 699–703. Leiden: Brill.

Ryding, Karin and Kees Versteegh, 2007. Iđaafa. In Encyclopedia of Arabic Language andLinguistics, vol. II, ed. Kees Versteegh, 294–298. Leiden: Brill.

Notes

1. Note that in generative grammar, there is a component called “verb phrase” (VP) whichcontains the predication. VP will be discussed in the next chapter.

2. Note also that in some types of grammatical analysis, a “phrase”may consist of only oneword. For example, a “noun phrase” (NP) may actually consist of only one noun. Theterm “phrase” is often redundant here, but is standard usage.

3. This is true except in the case of “accusative apposition” where a noun is apposed to apronoun and identifies it by means of a form of tamyiiz, or accusative of specification,e.g., naħnu l-ʕarab-a, “We, the Arabs.”

4. For a more extensive analysis of the apposition structure in Arabic see Ryding (2005:224–227).

5. See Belnap and Shabaneh (1992) for the use of the term “deflected,” and for furtherdiscussion of feminine singular agreement patterns. See also Belnap (1999, 2000).

6. DP (determiner phrase) is an alternative form of reference to NP, but NP is still used bymany authors.

7. Dixon refers to demonstratives as “the class of shifters with deictic reference to someperson (or some thing) other than speaker or addressee” (2010b: 224). On “shifters” seeDixon (2010a: 114), and especially Jakobson (1990: 386–392).



8. Semi-prepositions are locative adverbs that behave in ways very similar to prepositions,but which are more noun-like, and related to triliteral roots. When followed by a noun orpronoun they are marked for accusative case.

9. See Dixon (2010b: 267–271) for discussion of genitive-marking in NP-internalpossessives.

10. For more detailed examples of ʔiđaafa structures see Ryding (2005: 205–224). For ahistorical overview and analysis see Ryding and Versteegh (2007: 294–298). See alsoBadawi, Carter, and Gully on annexation (2004: 130–143). For a generative approach toconstruct state analysis, see Benmamoun (2006).

11. See Ouhalla (1997), “Genitive subjects and the VSO order” for an analysis of bothstandard Arabic and Berber NP structures.

Summary 125


11

Arabic syntax II: clause structure

1. Clauses in Arabic

Clauses are centered structurally and systematically around the predicate,and the predicative essence of a clause is what distinguishes it from a phrase. InArabic syntax, there are verbal sentences and verbless (equational) sentences, andpredicates may be of almost any lexical category: verbs (daras-naa l-kitaab-a ‘Westudied the book’), pronouns (haadhaa huwa ‘This is he’), prepositional phrases(al-kitaab-u fii l-maŧbax-i ‘The book is in the kitchen’), adjectives (al-bayt-ukabiir-un ‘The house is big’), or nouns (haaʔulaaʔi ŧullaab-un ‘These are stu-dents’).1 Thus although verbs are at the heart of most predications, because the verb‘to be’ in Arabic does not surface in the present tense indicative, other syntacticcategories may bear the predicate or copular function in equational sentences.Traditional Arabic grammars often classify sentence-types according to the first

word in the sentence (noun or verb – jumla ismiyya/ jumla fiʕliyya, ‘noun-sentence’/ ‘verb-sentence’), but the division is also viewed alternatively, accordingto whether or not the sentence contains an overt verb at all.2 Verbless sentences areconsidered a distinct linguistic category and usually referred to in English as“equational” sentences, with a basic predication distinction between the “topic”component (al-mubtadaʔ) and the “comment” component (al-xabar).

2. Verbal sentences/verbal clauses

Syntactic relations in Arabic verbal sentences may be characterized ascentering around the verbal predicate, which acts as the primary “governor” orʕaamil within the clause. The verb assigns theta-roles, marked overtly as Arabicaccusative case. In traditional Arabic syntactic analysis, and in some discussions ofargument structure, the subject NP of the verb is “governed” in the nominativecase. In generative grammar, subject NPs are seen as different from object argu-ments, as indirect and external to the VP. It is said that the verb “theta-marks” a

127


constituent directly if it assigns a role directly; “if a verb theta-marks an argumentcompositionally we call this indirect theta-marking” (Haegeman 1994: 72).3

Interestingly, in terms of contemporary Arabic syntactic theory, “the verbalclause seems to have elicited less theoretical interest than nominal clauses have,except with relation to agreement” (Hoyt 2009b: 653). Word order issues havedominated in the analysis of Arabic syntax, especially as they relate to subject-verbagreement, but as Hoyt notes, “In the Government-Binding and the Minimalisttraditions. . . the verbal clause has no independent theoretical status. . . Theseframeworks make extensive use of ‘null’ or unpronounced abstract structure, andas such, ‘word order’ as it is traditionally known does not correspond directly toconstituent order” (Hoyt 2009b: 657). This difference, albeit subtle, is important tounderstanding the aims and procedures of Arabic generative syntactic analysis. Interms of constituency and schematic relations, agreement features and governmentfeatures characterize Arabic clausal relations: agreement between subject and verb,and various kinds of hierarchal relations between the verbal predicate and itsarguments, including clausal arguments. The argument structures that surface inverbal clauses are factors of the semantic valence of the verb. In an earlier workI have summarized these relations as follows:

The valence of a verb or other predicate (such as a preposition) is expressed interms of the number of core arguments that the predicate requires. Thus a verbsuch as ‘give’ in English or ʔaʕŧaa in Arabic has a valence of three (Agent, Object,and Recipient), whereas a verb such as ‘buy’ or ishtaraa, has two core arguments(Agent and Object). Pinker proposes the term “thematic core” for the set of apredicate’s required arguments and defines it as follows: “a thematic core is aschematization of a type of event or relationship that lies at the core of themeanings of a class of possible verb” (1989: 73). Goldberg considers argumentstructure of central importance in relating semantics to syntax, stating that “argu-ment structure constructions are a special class of constructions that provides thebasic means of clausal expression in a language” (1995: 3). In some approaches toargument structure, such as Fillmore’s ‘case grammar’ (Fillmore 1968 and 1977),the different arguments are distinguished according to thematic role labels such as“Agent,” “Patient,” and “Beneficiary.” As Haspelmath notes, “Fillmore’s inten-tion was to highlight the importance of abstract semantic roles for languages likeEnglish that have (almost) no case distinctions” (2009: 507). Anderson states that“if we interpret the relations involved here [in dative and accusative relations] assemantic . . . in the case of the post verbal elements (at least), then their identi-fication is ensured by the semantic valency of the verb, which regulates the syntax.This . . . is the crucial insight of ‘case grammar.’” (Anderson 2006: 28)

Other approaches to predicate-argument structure, such as Pinker 1989 and Levinand Rappaport 1998, forgo the semantic labeling of arguments and differentiatethem only as X and Y. In a later work, Levin and Rappaport describe the verb’ssemantic core structure using the term “lexical semantic template” (1995: 24). In

128 Arabic syntax II: clause structure


their 1998 article on morphology and lexical semantics, Levin and RappaportHovav distinguish between “the lexical syntactic representation, often called ‘argu-ment structure,’ and the lexical semantic representation which . . . has come to beknown as ‘lexical conceptual structure’” (LCS) (1998, 2001: 249). Thus, a numberof alternative perspectives have been proposed regarding the nature of semanticcore arguments required by predicates, focused on the interrelationship between thesyntax and semantics of the clause, and on linking or mapping the semanticinformation to surface structure. (Ryding 2011: 288–289)4

Five syntactic issues are introduced here as examples of topics of interest toresearchers in Arabic syntactic theory: word order and subject-verb asymmetry,wh-movement, dative-movement, sentential complements, and case theory.

2.1. Word order

Verbal sentences in standard Arabic tend to have VSO word order (Verb-Subject-Object); but this is by no means a strict standard, and it varies dependingon context and discourse function. For example, newspaper headlines tend to beSVO, reflecting “the attention-getting function of the SVO word order” (Ryding2005: 67). Moreover, vernacular Arabic word order differs considerably fromstandard Arabic and is heavily influenced by discourse context. Current issues inword order studies (for both standard and colloquial Arabic) have centered aroundthe following factors, as described by Dahlgren: (1) foreground and background,(2) topicalization, (3) focusing, (4) topicality, (5) animacy, (6) aspect, (7) rhythm(2009: 731–734).According to Soltan

the study of clause structure and word order has figured as one major topic in thestudy of Arabic syntax. There have been three main questions in this regard:(i) what are the syntactic categories in the clausal hierarchy, e.g., is Arabic a tenselanguage, and if so, how is tense expressed? (ii) What are the dominance relationsbetween such categories on the hierarchy, e.g., where is Neg projected in theclausal hierarchy? (iii) How can this clausal hierarchy account for possible wordorders attested in Arabic dialects, e.g., the alternation between verb-initial . . . andnominal-initial . . . structures. (2011: 238)

Researchers have engaged in a rich debate about these topics, particularly as theyaffect both standard and colloquial Arabic.5

2.1.1. Subject-verb agreement asymetry

Certainly the topic of key interest in Arabic word order studies is subject-verb agreement asymmetry (SVAA). Agreement rules normally require that verbs

Verbal sentences/verbal clauses 129


reflect the number and gender of their subjects, but in Arabic, this rule only appliesfully when the verb follows the subject:

al-banaat-u daras-na fii l-maktabat-ithe-girls nom. studied 3f.pl. in the-library-gen.The girls studied in the library.

al-bint-aani daras-ataa fii l-maktabat-i.the-two-girls nom. studied 3f.du. in the-library gen.The two girls studied in the library.

If a plural or dual subject follows the verb, agreement is only partial; the verbagrees in gender only, not number:

daras-at al-banaat-u fii l-maktabat-istudied 3f. sing. the-girls-nom. in the-library-gen.The girls studied in the library.

daras-at al-bint-aani fii l-maktabat-istudied 3f. sing. the-two-girls-nom. in the-library-gen.The two girls studied in the library.

This agreement restriction applies solely to human subjects, because onlyhuman plurals are reflected in Arabic agreement morphology. If the subjects arenon-human dual or plural, their agreement features are feminine singular no matterwhat the word order:

al-Suquur-u ʔakal-at al-samakat-athe-hawks-nom. ate 3f. sing. the-fish-acc.The hawks ate the fish.

ʔakal-at al-Suquur-u al-samakat-aate 3f. sing. the-hawks-nom. the-fish-acc.The hawks ate the fish.

Accounting for the agreement asymmetry with human subjects has been a topicof extended discussion, especially in generative approaches to Arabic syntax. “Themajor challenge in this respect has always been how to account for the presence ofthe SVAA in SA [standard Arabic] given standard assumptions about agreement ingenerative syntax” (Soltan 2006: 241).6 In Chekili’s analysis of word-order issueshe notes that “Arabic raises the question of how to analyze the initial NP in SVO, assubject or topic,” and notes that this question “has given rise to a dual account ofsuch structures” (2009: 527). Ouhalla states that “a more appropriate way ofcharacterizing the situation in standard Arabic is not in terms of the notion ‘lackof agreement,’ but, rather, in terms of the notion ‘poor agreement’ (1997: 205). He



further refers to “the fact, attested in a large number of languages . . . that poorsubject agreement elements (characteristic of so-called impersonal forms) tend to beidentical with the third person singular agreement morpheme” (205). Soltan spec-ulates that “full agreement obtains in the SV orders because of the presence of apronominal subject . . . Partial agreement in the VS order could be viewed then as theresult of a default agreement on T(ense)” (2005: 203).7 The topic of Arabic SVAAhas generated substantial research and theoretical speculation, especially amonggenerative linguists, and remains an area of key interest for theoretical linguistics.

2.2. WH-movement

The study of interrogative structures through WH-movement is a sec-ond topic of research interest in Arabic linguistics from a generative viewpoint.8

Wh-words are question words (e.g., who, when, why) or relative pronouns (who,which). Chekili notes that certain transformational/generative models generated“even greater interest in ‘WH-constructions’ because they relied on such con-structions in developing a general theory of conditions on transformations”(2009: 524). According to Choueiri, “WH-movement plays a key role in thesyntax of long-distance dependencies. Typically, it is involved in the derivationof wh-interrogatives, but it is also involved in the formation of other construc-tions, such as topicalized constructions and relative clauses” (2009: 718). Inrecent articles, Soltan (2010 and 2011) addresses issues of scope and question-formation in Egyptian Arabic. Most wh-movement studies have focused onvernacular Arabic, since it is spoken discourse that most vividly contextualizesthe various kinds of question formation, their acceptability, their structure, andmeaning.

2.3. Case relationships and case theory

Case theory centers around the key role of the verb and its semanticvalence (the number of arguments that the verb takes). The relationships ofsentence or clause constituents to the verbal predication are characterized interms of their case roles, which include labels such as Agent, Benefactive/Recipient, Experiencer, Instrumental, Locative, and Object. These terms havebeen subject to discussion and dissatisfaction within syntactic theory, so thenumber of them and labels of them are still under scrutiny and evaluation.9 It isimportant to note that these “cases” do not correspond exactly to the case-markingsystem of Arabic (or any other language), nor are they intended to. They are labelsof semantic relations between the verb and its arguments, rather than syntacticcategories. Nonetheless, case theory and case-marking can be interrelated and usedproductively to discuss and analyze Arabic case structures and their functions.



Particularly important to this analysis is the study of Arabic verbal ʔawzaan, theForms of the verb, and their valency alternations. Amberber (2000) is a study ofAmharic verbs and valency encoding, but similar studies for Arabic have yet to bedeveloped.10 Ryding (2011) discusses dative structures and their underlying case-role relationships, as does Ryding-Lentzner (1977) and 1981. Abdul-Raof (2006)and Letourneau (2006) and 2009 describe case roles, case theory, and theta roles asthey apply to Arabic.

2.4. Dative structures in Arabic

Themorphology ofArabic verbswherein lexical root information interrelateswith morphosyntactic verbal template information yields a perspective for analysisof “key issues in syntactic theory through analysis of the formal semantics ofArabic lexical roots and their derivational modifications. . . Modern StandardArabic lexical items remain largely transparent in terms of their lexical structureand syntactic argument requirements. When derivational or syntactic modifica-tions yield ditransitive constructions, it is often possible to discover semanticreasons for particular syntactic constraints” (Ryding 2011: 283–284). In Ryding(2011) and Ryding-Lentzner (1981), Arabic ditransitive constructions and thedistinction between Arabic ‘to’- datives and ‘for’ – datives, are analyzed usingcase grammar, construction grammar, and lexical semantics to examine the com-position of dative semantics and their realizations in Arabic.

Ditransitive structures in Arabic include the double-object construction and theli-construction. The double-object constructions result from the underlying seman-tics of the verb, including those that are doubly transitive due to the lexical contentof the root and others that result in double transitivity through derivational mor-phology. Here are some categories of ditransitives in Arabic:

(1) The dative-alternation construction where the beneficiary argument shifts place,with preposition deletion, often based on the notion of “giving.”

(2) Causative constructions where a valency-changing derivation modifies the lexicalroot, e.g.

Form IV ʔaħđara to bring (Cause-to-come)Form IV ʔaŧʕama to feed (Cause-to-taste)11

(3) Verbs of permission or denial, e.g.

manaʕa to forbidmanaħa to grant



(4) Verbs of perception and cognition (afʕaal qalbiyya), e.g.

ʕadda to consider, deemiʕtabara to consider, deemwajada to find, deem

(5) Verbs of transformation (afʕaal al-taHwiil), e.g.

şayyara to convertittaxadha take, adopt (as)jaʕala to makeʕayyana to appointtawwaja to crown12

(Ryding 2011: 286–287)

For example, in the predication of “giving,” the option is to use either doubleaccusative, or the prepositional dative structure, li-with Beneficiary noun, shiftedto the second object position, a procedure that is referred to as “dative shift” or“dative movement:”

ʔaʕtay-tu l-bint-a l-kitaab-ai-gave the-girl-acc the book-acc.I gave the girl the book.ʔaʕtay-tu l-kitaab-a li-l-bint-ii-gave the-book-acc to-the-girl-gen

I gave the book to the girl.

With the predicate for “buying,” however, the ditransitive structure is notpossible in Arabic (although it is in English):

ishtaray-tu l-kitaab-a li-l-bint-ii-bought the-book-acc. for-the-girl-gen.I bought the book for the girl.

But not:

*ishtaray-tu l-bint-a l-kitaab-ai-bought the-girl-acc. the-book-acc.I bought the girl the book.

In Ryding 2011 I show through compositional analysis that variations on Arabicdative structures are clearly semantically motivated, and that there is “a majordifference in syntactic behavior between the to-dative and the for-dative in Arabic,even though they are represented in the surface structure by an identical item, li-”(295).



2.5. Complement clauses in Arabic

Complement clauses are predications which are subordinated withinanother sentence. In Maria Persson’s work on object complements in standardArabic, she defines a complement as “any element whose presence and form isdecided by the principal lexical item of the phrase or clause in which it occurs”(2002: 7). Complement clauses are often linked to main or matrix clauses throughthe use of subordinating conjunctions referred to as “complementizers.” Perssonconsiders a complementizer as “a morpheme (a word, particle, clitic, or affix)signalling that the preceding or following clause is a complement” (2002: 13). InArabic this would include ʔinna and her sisters as well as ʔan-plus-subjunctiveclauses. From a theoretical standpoint, Persson takes a “functional perspective” toanalyze complementation, rather than a generative standpoint.

Dixon (2010b) also provides a useful definition of complement clause:“A complement clause is a type of clause which fills an argument slot in thestructure of another clause” (370). Dixon’s analysis of the grammatical structureof clauses sets “three defining criteria for a complement clause.:” (1) “It has theinternal constituent structure of a clause [not a phrase], at least as far as corearguments are concerned,” (2010b: 375); (2) it “functions as a core argument of ahigher clause. In every language in which complement clauses occur they functionas O[Object] argument; there are often other possibilities as well” (2010b: 377);(3) “A complement clause must refer to a proposition, something involving at leastone participant who is involved in an activity or state” (2010b: 380). In addition,Dixon provides an analysis and summary of the semantics of matrix verbs, whichin Arabic determine the nature of the complementizer and complement clause.These semantic types include: (1) attention (see, hear, notice, smell, show (2010b:395), (2) thinking (think, consider, imagine, assume, suppose, know, understand(2010b: 396); (3) deciding (decide, resolve, plan, choose (2010b: 397); (4) liking(like, love, prefer, regret, fear, enjoy (2010b: 397); and (5) speaking (say, report,inform, state – among others) (2010b: 397–398).

Although covering the full range of variation in Arabic complement studies hereis not possible, I would like to provide an indication of what is theorized about suchstructures. First of all, the type of complementizer is selected by the semanticnature of the main verb. Kirk Belnap’s research on Arabic complementationstructures (1986) focused on the use and distribution of ʔan, ʔinna, ʔanna, andverbal noun complements in a corpus of about 17,000 words, classifying themaccording to Givón’s (1980) “hierarchical implicational scale of binding.”Whereas English may use the word ‘that’ to embed a complement in many differentrespects (“I wish that. . .; they told me that . . .; she thinks that . . .” ), Arabic



distinguishes among the complementizers (and the structures they govern) accord-ing to the meaning of the matrix verb. These complementizers in Arabic, the“sisters of ʔinna,” are subordinating conjunctions that are followed by a clausewhose subject is required to be in the accusative case. For example, after the verbqaal-a ‘to say,’ the complementizer ʔinna is required:

qaal-uu ʔinna l-siyaasiyy-iina ya-staxdim-uunasaid-they m. that the-politicians m. they-use-m.muşŧalaħaat-in diiniyyat-anterminologies-acc. religious-acc.They said that politicians use religious terms.

After verbs of reporting or relaying factual information, the particle ʔanna isused as complementizer:

ʔadrak-a ʔanna-hu nasiy-a l-kalimat-ahe-realized that-he he-forgot the-word-acc

He realized that he forgot the word.

After a matrix verb indicating an attitude or feeling toward the action in thecomplement clause, the complementizer ʔan is used, with subjunctive marking onthe verb:

tu-riid-u ʔan ta-ʕrif-a maadhaashe-wants-indic that she-know-subj what

sa-ya-ʕnii haadhaa l-salaam-uwill-he-mean this-m the-peace-nom

She wants to know what this peace will mean.

Matrix verb semantics, the choice of complementizer, and subsequent comple-ment structure have been areas of some interest for Arabic syntactic study. AsPersson states in her conclusion, “The CTPs [complement-taking predicates] in theMSA corpus have rather well-defined semantic fields, and . . . each class of CTPsis, to a remarkable extent, associated with specific complement patterns” (Persson:2002: 135).

3. Verbless predications

Equational or verbless sentences constitute one of the major categories ofArabic syntax, one that has attracted substantial attention from researchers insyntactic theory. The key feature of these sentences is that there is no surfacerepresentation of the present tense indicative verb ‘to be’, although it surfaces in

Verbless predications 135


negatives, in subjunctive mood, in jussive mood, and in the past and future tenses.Bahloul writes that “contextual triggers such as tense, aspect, modal, and moodmarkers” affect the appearance of the verb ‘to be’ (kaana) in copular sentences, andhe advocates a “model which places modality at the center of the sentencestructure” (2006: 510). Because verbless sentences often begin with a noun orNP, they are sometimes terminologically conflated with verbal sentences that startwith a nominal (jumal ismiyya ‘nominal sentences’). As Badawi, Carter, and Gullypoint out, Arabic has no separate term for “equational sentence,” rather, “it fallsunder ‘nominal sentence’” (2004: 307). It is helpful, however, to examine thespecific nature of verbless sentences.

Verbless predications in standard Arabic are of two types: one in which thereis no overt “copula morpheme,” and one where there is. The copula morpheme isrealized, when it occurs, as an Arabic subject pronoun.13

3.1. Predications with no copula pronoun

In Arabic, predication without a copula pronoun is standard for express-ing quality, quantity, location, identity, and other states of being in the present tenseindicative. The two parts of the sentence, the subject (al-mubtadaʔ) and thepredicate (al-xabar) form a balanced predication. The subject of such sentencesis usually definite. If the predicate is a modifier, it agrees with the subject in numberand gender, but is indefinite. Both parts of the basic equational sentence are in thenominative case. When the predicate is a prepositional phrase it normally followsthe subject, except when it indicates possession.

al-ŧariiq-u ŧawiil-u-nthe-road-nom. [is] long-m.-sing.-nom. -indef.The road is long.

ʔuxt-u-haa ʔustaadhat-u-nsister-nom.-her [is] a-professor-f.-nom.-indef.Her sister is a professor.

ʔax-uu-naa musaafir-u-nbrother-nom.-our [is] traveling-m.-sing.- indef.Our brother is traveling.

al-zuwwaar-u fii l-maktab-ithe-visitors-nom. [are] in the-office-gen.The visitors are in the office.

la-naa l-qudrat-uto-us [is] the-ability-nom.We have the ability.



3.2. Predications with copula pronoun

The copula pronoun is used when it is necessary to distinguish a phrase(e.g., ‘this book’) from a predication (‘this is a book’), or when the predicate of theequational sentence is definite.

haadhaa l-makaan-uthis-m. sing. the-place-nom.this place

haadhaa huwa l-makaan-uthis-m. sing. he [is] the-place-nom.This is the place.

haaʔulaaʔi l-masʔuul-uunathese the-officials-nom

these officials

haaʔulaaʔi hum-u l-masʔuul-uuna14

these they-masc. [are] the-officials-nom.These are the officials.

This predicative function of the pronoun has led one researcher to maintain that“evidence from Arabic suggests that the copula pronoun be analyzed as a predicateexpressing the relation of identity” (Eid 1991: 33), and that “pronouns function asanti-ambiguity devices to force a sentential, vs. a phrasal, interpretation of astructure” (Eid 1991: 42).15

4. Summary

Arabic clausal syntax is a vast and fertile field for linguistic study, nomatter which approach is taken or which theories applied. Generative theory hasshown that Arabic has an important role to play in contributing to the concept ofUniversal Grammar; valence theory has shown the importance of Arabic inextending the analysis of lexical and morphological composition of verb forms;and the special role of copular clauses in Arabic has brought attention to thecentrality of mood-marking in Arabic surface structure.

Study questions and discussion

(1) Find ten examples of subject-verb asymmetry in Arabic and discuss theseexamples with the other students in your class. What would be the ten bestexamples, overall, for use in teaching Arabic as a foreign language?

Summary 137


(2) Subordination or complementation is a complex topic in standardArabic. Aside from ʔinna and her sisters, list any other kinds of com-plementizers and their effect on the embedded clause. Be sure to citeexamples.

(3) Equational or verbless sentences may or may not have a copula mor-pheme. Find five examples of each and discuss them.

(4) What kind of insight can valence theory provide for Arabic syntaxor morphosyntax? Prepare a two-page paper discussing valence theoryand Arabic.

Further reading

Aoun, Joseph E., Elabbas Benmamoun and Lina Choueiri. 2010. The Syntax of Arabic.Cambridge: Cambridge University Press.

Benmamoun, Elabbas. 2000. The Feature Structure of Functional Categories: A compara-tive study of Arabic dialects. Oxford: Oxford University Press. Especially chapter 1 oncomparative Arabic syntax.

Chekili, Ferid. 2009. Transformational grammar. In Encyclopedia of Arabic Language andLinguistics, vol. IV, ed. Kees Versteegh, 520–528. Leiden: Brill.

Ryding, Karin C. 2011. Arabic datives, ditransitives, and the preposition li-. In In theShadow of Arabic: Festschrift for Ramzi Baalbaki, ed. Bilal Orfali, 283–298. Leiden:Brill.

Soltan, Osama. 2006. Standard Arabic subject-verb agreement asymmetry revisited in anAgree-based minimalist syntax. In Agreement Systems, ed. Cedric Boeckx, 239–264.Amsterdam: John Benjamins.

Notes

1. “Apart from a Verb Phrase (VP), the initial NP of the copular sentence may be followedby any other lexical category” (Bahloul 2006: 507).

2. Issues of clause classification have been significant factors in Arabic theoretical syntax.Hoyt, for example, reviews the implications of two definitions of “verbal clause,” onereferring to “V-initial” word order, and the other to “V-headed” clauses, making thedistinction between sentences (clauses) in which inflected verbs come first, and sentenceswhere the verb appears later in the sentence (2009: 653).

3. “The theta role assigned to the subject is assigned compositionally: it is determined bythe semantics of the verb and other VP constituents. Roughly, the verb assigns an objectrole first, the resulting verb-argument complex will assign a theta role to the subject. Thesubject argument is as if it were slotted in last” (Haegeman 1994: 71–72) (emphasis inoriginal).

4. For more on case roles and theta roles in Arabic, see LeTourneau (2006, 2009).



5. See for example, Aoun, Benmamoun, and Choueiri (2010); Chekili (2009: 523–524);Soltan (2006, 2007, 2009, 2010, 2011, and 2012).

6. For an examination of SVAA in both spoken and written Arabic, see Aoun,Benmamoun, and Choueiri (2010: 73–95) on sentential agreement. For summaries ofthe issues and helpful bibliographies, see Hoyt (2009a, 2009b).

7. See Soltan (2006, 2011) for further discussion of a minimalist approach to Arabicsyntactic derivations, especially as regards SVAA.

8. Wh-movement “is used to refer to a transformational rule which moves a wh-phrase(wh-XP) to initial position in the sentence” (Crystal 1997b: 418).

9. “Across theories there is a huge amount of dissatisfaction with these role labels” (Butt2009a: 33).

10. See Maalej (2009) for a discussion of valency as it applies to Arabic.11. For more on the semantics of ‘cause’ in Arabic, see Măcelaru (2006).12. Categories 4 and 5 include verbs which belong to the traditional ‘nawaasix’ category in

Arabic grammar, that is, verbs that shift one or more arguments in the VP to accusativecase. See Ryding (2005: 176–179) for further description.

13. The term “copula morpheme” is taken from Eid (1991).14. The /-u/ suffix on hum ‘they’ in this sentence is a helping vowel, not an inflectional

vowel.15. For more extensive discussion of copular sentences in Arabic, see Ryding (2005:

59–63); and especially Badawi, Carter, and Gully (2004: 307–344). Bahloul (2006)provides an excellent summary of copular structures in standard Arabic.

Summary 139

C:/ITOOLS/WMS/CUP-NEW/4458965/WORKINGFOLDER/RARA/9781107023314APX01.3D 141 [141–144] 23.10.20135:37PM

Appendix A: Fields of linguistics and Arabic

Because the main part of this book focuses on theoretical linguistics, thisappendix summarizes research interests and traditions of three other major fields oflinguistics: applied linguistics, sociolinguistics, and computational linguistics, asthey relate to Arabic.

1 . APPLIED LINGUISTICS AND ARABIC

The core of applied linguistics is the connection between theory andpractice. The field of applied linguistics is concerned with real-world issuesthat involve language, such as language teaching. It also includes disciplinessuch as lexicography (dictionary design and compilation), language and thelaw (forensic linguistics), interpretation and translation, second languageacquisition research, language testing, and language planning. As it appliesto Arabic at the current time, applied linguistics has been heavily weighted inthe direction of language teaching and learning, textbook and curriculardesign, proficiency testing, and teacher training. The practical needs andprofessional demands of teaching a greatly expanded number of studentshave necessitated a critical professional focus on language teaching resourcesand approaches.

There are no dedicated journals or other periodicals that specialize in Arabicapplied linguistics. Articles on Arabic do appear in applied linguistics journalsand foreign language teaching journals (Elkhafaifi 2005a and 2005b; Al-Batal2006; Ryding 1991; Alhawary 2009), in Al-Arabiyya, the journal of theAmerican Association of Teachers of Arabic, and in edited collections (e.g.,Al-Batal 1995, Elgibali 1996 and 2005, Wahba, Taha, and England Owens2013a 2006).

A key issue in applied Arabic linguistics is diglossia, the systematicdisparity and coexistence of spoken and written Arabic variants in theArabic speech community, and the effect of this disparity on language useand language teaching. The foundational work on this topic is Ferguson(1959a). Diglossia has since become a widely discussed topic and has devel-oped into theories of discourse continua, interpersonal discourse pragmatics,and many more fine-grained analyses. This topic can raise resistance among

141


some native speakers, who perceive the Arabic language as a unified wholeand who consider the study of spoken variants trivial, divisive, and distract-ing. This attitude has strongly privileged the study, analysis, and teaching ofwritten Arabic and discouraged the formal study and teaching of colloquialArabic discourse.

For many years, the sole topic of study considered legitimate in formallearning situations has been Classical Arabic (CA) or Modern Standard Arabic(MSA), the written language. Formal study of vernacular Arabic has beenneglected in academic curricula and in government-based training programs –even programs that claim to be teaching full communicative competence. Withpressing real-world needs for interactive skills, the profession has had to turnits attention to spoken Arabic, but there is as yet no consensus on the timing,amount, proportion, or varieties that would best serve English-speaking learn-ers. Recent articles by Choueiri (2009), Ryding (2006a, 2006b, 2008, 2009),and Younes (2009) deal explicitly with this issue. This area of study is inter-woven with deeply felt attitudes and ideologies having to do with the social,cultural, and religious values attached to written and spoken varieties ofArabic.

2 . SOCIOLINGUISTICS AND ARABIC

Sociolinguistics, the study of language use in social contexts, hasemerged as one of the central fields for the study not only of Arabic language,but of culture, nationalism, and identity in the Arabic speech community.Discourse of all kinds and at all levels provides the raw data for sociolinguisticanalysis (see Haeri 2003, for example), but much of the work thus far has beendone on political discourse (Dunne 2003, Bassiouney 2006, 2009) and other formsof public discourse (Suleiman 2003, 2004), especially as seen through variousforms of media (Eid 2007; Bassiouney 2010).

Studies of Arabic language in society and the relationship between the writtenand spoken variants have been done by Badawi 1985; Ferguson 1959a and 1959b;Parkinson 1991, 2003; Haeri 2003 and 2006; Hary 1996, Mejdell 2006, andRosenbaum 2008, among others. These studies have focused on discourse inboth formal and informal situations, on the different registers or levels of Arabicin use, on the mixing of language registers, and on attitudes toward language thatcharacterize the Arabic speech community. MSA borders on vernacular Arabic inmany situations and native speakers easily cross and re-cross those discourseborders as they see necessary and fit for rhetorical purposes within differentcontexts and different genres. Thus there is, especially in more formal spokenArabic situations, a tradition of code-mixing or hybridization that occurs on aregular basis. Code-mixing (the mingling of language levels and registers) hasbecome more and more commonplace and conventionalized in the broadcast

142 Appendix A


media and in relatively formal contexts, such as university lectures, classroomdiscussions, conferences, and official receptions, for example.1 Progress ofresearch in Arabic sociolinguistics is key to understanding how spoken and writtenArabic are calibrated in actual discourse situations, and in forming the foundationfor unearthing the inherent regularities in ordinary spoken Arabic discourse.Particularly in relationship to code-mixing, sociolinguistic research has begun todocument the contextualized practices of Arabic speakers and the cultural prag-matics of language use.2

3 . COMPUTATIONAL LINGUISTICS AND ARABIC

Computational linguistics refers to the use of computers in the analysis oflanguage, in data collection, and in machine translation. Arabic language computa-tional resources have developed slowly compared with other languages, but therehas been accelerated progress in the past few years. One of the most importantcontributions of the growing field of Arabic computational linguistics has been thecompilation of authentic corpora for linguistic analysis such as frequency countsand occurrences of lexical collocations. The Arabic corpus project at BrighamYoung University (BYU) currently provides a key source for researchers needingto examine language in context. According to the BYU digital humanities website,“ArabiCorpus is a free, untagged, 30-million-word corpus with a user-friendlyinterface. Maintained by Dil Parkinson, professor of Arabic, this corpus allowsusers to find larger structures and grammatical patterns through frequency analysis,regular expression searching, and other advanced interface features. TheArabiCorpus is a highly regarded tool for both researchers and advanced Arabicstudents, and can be found at http://arabicorpus.byu.edu” (Brigham YoungUniversity, 2013).

Other uses of corpora include syntactic and morphological parsing andcompilation of lexicons. With the advent of reliable computational resources,searching various corpora for particular structures and usages has been renderedfar easier and Arabic linguistics researchers can now process considerably moredata than before. Most of the corpora are in MSA, but the development of spokenArabic corpora is also well underway. Ditters (2006 and 2013) provides an over-view of Arabic computational linguistics, including an extensive bibliography. Therecent publication of A Frequency Dictionary of Arabic by Tim Buckwalter andDilworth Parkinson lists the most frequent 5,000 words based on “a corpus of 30million words of which 10 percent was made up of spontaneous (unscripted)speech data . . . [and] the remaining 90 percent of the corpus came from writtensources” (2011: 3). This new and important resource will undoubtedly improve thetargeting of key vocabulary in the development of Arabic language teachingmaterials and in classroom interaction, and is just one of the results of computa-tional linguistics research in Arabic.3

Fields of linguistics and Arabic 143

http://arabicorpus.byu.edu


Notes

1. See Mejdell (2006), for a close study of code-mixing in Egypt, as well as Bassiouney(2006). See Eid (2007) for a stylistic analysis of hybrid Arabic used in the media. See alsoAl-wer (2013) for a cogent current overview of Arabic sociolinguistics, and Suleiman(2013) for an astute analysis of diglossia and Arabic folk linguistics.

2. See Rosenbaum 2008 for a lively discussion of what is termed “fuşħaamiyya.”3. For a critical overview of contemporary Arabic lexicography, see Buckwalter and

Paokinton (2013).

144 Appendix A

C:/ITOOLS/WMS/CUP-NEW/4464129/WORKINGFOLDER/RARA/9781107023314APX02.3D 145 [145–148] 25.10.20139:18AM

Appendix B: Arabic transcription/transliteration/romanization

The process of changing Arabic script into a romanized (or latinized)equivalent would seem to be a straightforward or even trivial one, but it is not.Because it requires complete accuracy, “scientific transcription” is a painstakingeffort demanding high levels of phonological and morphological knowledge, andtransliteration errors and inconsistencies tend to creep into even the best-editedpublications. For a number of reasons, romanization (conversion from a non-roman alphabet to a roman one) from Arabic can be problematic, especiallybecause of the invisibility of short vowels in Arabic script, which are crucial toaccurate pronunciation. Other common areas of romanization problems include themisrepresentation of gemination, short versus long vowel notation, representationof epenthetic vowels, morpheme boundaries (if noted), case-marking, and wordboundaries.

Traditionally, a distinction is drawn between transcription and translitera-tion. I have long relied on Charles Ferguson’s definitions of these processes, andreproduce them here.

TRANSCRIPTION

(Transcription = phonemic conversion: e.g., ʕabd-u n-naaşir)

Transcription: The written representation of a language by symbols or spellingsother than those of the standard orthography of the language. If the language isnormally unwritten, any writing system devised is called a transcription unless itbecomes the accepted orthography. If a transcription is based exclusively on thesounds of the language, it is called a phonetic transcription. One important varietyof phonetic transcription is the phonemic transcription in which each symbolconsistently represents one phoneme . . . of the language being written.Transcriptions which are only partially phonetic (or phonemic) are also used forvarious reasons; they are usually based in part on grammatical or semanticconsiderations. (Ferguson 1962: 10)

Transcription, therefore, is primarily a way of representing language as it is spoken.For spoken languages that are normally not written down, such as Arabic dialects,

145


transcription systems may be used to represent the spoken sounds, and the tran-scription symbols might be Arabic, or phonetic, or romanized, depending on thereasons and audiences for the transcription.

TRANSLITERATION

(Transliteration = graphemic conversion, e.g., ʕabd-u l-naaşir)

Transliteration: The systematic use of the symbols of one writing system torepresent those of another, the ideal being a one-to-one correspondence so thatsomething written by means of the transliteration can be converted to the originalorthography and vice versa without ambiguity. The term is most often used to referto systems of using the Roman alphabet to represent various orientalalphabets. (Ferguson 1962: 10)

Thus transliteration would be the writing of romanized script equivalent to writtenArabic script, representing all orthographic elements. The primary problem here isthat Arabic script has both “shallow” and “deep” orthographies, i.e., it differs inhow it “portrays sound-to-symbol relationships” (Everson 2011: 260). Script thatincludes all short vowels and diacritics (such as shadda and waşla) is termed“shallow” – that is, easier to read, and is therefore used in teaching Arabic-speakingchildren how to read. Script that is “deep” lacks these features, assuming that adultreaders readily knowwhat they are. The fact that short vowels and diacritics are notrepresented in deep script does not mean that they do not exist; standard Arabicorthographic convention simply omits them and takes them as understood. This is aconceptual and cognitive processing problem for learners of Arabic as a foreignlanguage, and it also impacts the rendering of written Arabic into full translitera-tion.1 Short vowels and diacritics are invisible in normal Arabic script, but are ofcourse pronounced if the written text is read out loud.

Therefore, for Arabic, a hybrid system of transliteration/transcription – onethat takes into account pronunciation as well as orthography – has become thenorm for western publications that need to use transliterated Arabic.

GENERAL RULES

(1) When transliterating or transcribing, think in terms of mathematicalprecision.

(2) Transliterated terms from a foreign language should be in italics.(3) Every constituent of an Arabic word must be transliterated. This includes

short vowels, contrasts in vowel length, and the existence of geminateconsonants; these are all phonemic distinctions in Arabic and must beindicated.

(4) When you choose a particular system, you need to stick with it. You mustbe consistent.

146 Appendix B


(5) Widely used and accepted transliteration systems include those of theLibrary of Congress (LC), the Board of Geographic Names (BGN), andthe Encyclopedia of Islam (EI). The International Phonetic Alphabet(IPA) provides a standard for transcription and transliteration of conso-nants and vowels. A complete chart of the IPA system can be found inLaver (2001: 179).

Note

1. See Ryding (2013) on issues in teaching reading in Arabic as a foreign language.

Arabic transcription/transliteration/romanization 147


Appendix C: Arabic nominal declensions

Modern Standard Arabic nouns and adjectives fall into eight declen-sions.1 These categories of inflection are not usually treated separately asdeclensions in textbooks and grammars of Arabic, but I have found that thisoften leads to confusion for Arabic learners. I have therefore distinguishedamong all forms of nominal declension, including inflections for the suffixdeclensions of dual, sound feminine plural and sound masculine plural. Three-way inflection nominals (triptotes) are traditionally considered the base categorybecause they exhibit three distinctive case-markings for nominative, genitive,and accusative. All other categories have fewer case distinctions: some two,some only one. Moreover, sometimes there is a distinction between definite andindefinite inflection, and sometimes not (for example, the declensions for thedual and for the sound masculine plural do not exhibit distinctions indefiniteness).2

Arabic nominal declensions:Three-way inflection(1) three-way inflection (‘triptote’)

Two-way inflection(2) dual(3) sound feminine plural(4) sound masculine plural(5) diptote(6) defective

One-way inflection(7) indeclinable (for case, but marking definiteness), and(8) invariable

149


1. THREE-WAY INFLECTION: TRIPTOTE (MUʕRAB )

1.1 . s ingular masculine noun

‘saddle’ sarj

Definite Indefinite

Nominative: al-sarj-u sarj-u-nGenitive: al-sarj-i sarj-i-nAccusative: al-sarj-a sarj-a-n

1.2 . plural noun

‘saddles’ suruuj

Definite Indefinite

Nominative: al-suruuj-u suruuj-u-nGenitive: al-suruuj-i suruuj-i-nAccusative: al-suruuj-a suruuj-a-n

1.3 . feminine s ingular noun

‘ship’ safiina

Definite Indefinite

Nominative: al-safiinat-u safiinat-u-nGenitive: al-safiinat-i safiinat-i-nAccusative: al-safiinat-a safiinat-a-n

1.4 . plural noun

‘ships’ sufun

Definite Indefinite

Nominative al-sufun-u sufun-u-nGenitive: al-sufun-i sufun-i-nAccusative: al-sufun-a sufun-a-n

150 Appendix C


1.5 . masculine s ingular adjective

‘big’ kabiir

Definite Indefinite

Nominative: al-kabiir-u kabiir-u-nGenitive: al-kabiir-i kabiir-i-nAccusative: al-kabiir-a kabiir-a-n

1.6 . broken plural adjective

‘big’ kibaar

Definite Indefinite

Nominative: al-kibaar-u kibaar-u-nGenitive: al-kibaar-i kibaar-i-nAccusative: al-kibaar-a kibaar-a-n

2. TWO-WAY INFLECTION

2.1 . declens ion two:the dual

2.1.1. Masculine dual noun

‘two saddles’ sarj-aani

Definite Indefinite

Nominative: al-sarj-aani sarj-aaniGenitive: al-sarj-ayni sarj-ayniAccusative: al-sarj-ayni sarj-ayni

2.1.2. Feminine dual noun

‘two cities’ madiinat-aani

Definite Indefinite

Nominative: al-madiinat-aani madiinat-aaniGenitive: al-madiinat-ayni madiinat-ayniAccusative: al-madiinat-ayni madiinat-ayni

Arabic nominal declensions 151


2.1.3. Masculine dual adjective

‘big’ kabiir-aani

Definite Indefinite

Nominative: al-kabiir-aani kabiir-aaniGenitive: al-kabiir-ayni kabiir-ayniAccusative: al-kabiir-ayni kabiir-ayni

2.1.4. Feminine dual adjective

‘big’ kabiirat-aani

Definite Indefinite

Nominative: al-kabiirat-aani kabiirat-aaniGenitive: al-kabiirat-ayni kabiirat-ayniAccusative: al-kabiirat-ayni kabiirat-ayni

2.2 . declension three: the sound masculine plural

2.2.1. Sound masculine plural noun

‘assistants’ musaaʕid-uuna

Definite Indefinite

Nominative: al-musaaʕid-uuna musaaʕid-uunaGenitive: al-musaaʕid-iina musaaʕid-iinaAccusative: al-musaaʕid-iina musaaʕid-iina

2.2.2. Sound masculine plural adjective

‘many’ kathiir-uuna

Definite Indefinite

Nominative: al-kathiir-uuna kathiir-uunaGenitive: al-kathiir-iina kathiir-iinaAccusative: al-kathiir-iina kathiir-iina

152 Appendix C


2.3 . declension four: the sound feminine plural

2. 3.1. Sound feminine plural noun

‘differences’ ixtilaaf-aat

Definite Indefinite

Nominative: al-ixtilaaf-aat-u ixtilaaf-aat-u-nGenitive: al-ixtilaaf-aat-i ixtilaaf-aat-i-nAccusative: al-ixtilaaf-aat-i ixtilaaf-aat-i-n

2.3.2. Sound feminine plural adjective

‘Tunisian’ tuunisiyy-aat

Definite Indefinite

Nominative: al-tuunisiyy-aat-u tuunisiyy-aat-u-nGenitive: al-tuunisiyy-aat-i tuunisiyy-aat-i-nAccusative: al-tuunisiyy-aat-i tuunisiyy-aat-i-n

2.4 . declension five: diptote

2.4.1. Singular diptote noun

‘desert’ Saħraa$

Definite Indefinite

Nominative al-şaħraaʔ-u şaħraaʔ-uGenitive al-şaħraaʔ-i şaħraaʔ-aAccusative al-şaħraaʔ-a şaħraaʔ-a

2.4.2. Plural diptote noun

‘leaders’ zuʕamaaʔ

Definite Indefinite

Nominative al-zuʕamaaʔ-u zuʕamaaʔ-uGenitive al-zuʕamaaʔ-i zuʕamaaʔ-aAccusative al-zuʕamaaʔ-a zuʕamaaʔ-a



2.4.3. Singular masculine adjective

‘blue’ ʔazraq

Definite Indefinite

Nominative al-ʔazraq-u ʔazraq-uGenitive al-ʔazraq-i ʔazraq-aAccusative al-ʔazraq-a ʔazraq-a

2.4.4. Singular feminine adjective

‘blue’ zarqaaʔ

Definite Indefinite

Nominative al-zarqaaʔ-u zarqaaʔ-uGenitive al-zarqaaʔ-i zarqaaʔ-aAccusative al-zarqaaʔ-a zarqaaʔ-a

2.4.5. Plural diptote adjective

‘foreign’ ʔajaanib

Definite Indefinite

Nominative al-ʔajaanib-u ʔajaanib-uGenitive al-ʔajaanib-i ʔajaanib-aAccusative al-ʔajaanib-a ʔajaanib-a

2. 5 . declension s ix : defective

2.5.1. Singular defective noun

‘lawyer’ muħaamin

Definite Indefinite

Nominative: al-muħaamii muħaaminGenitive: al-muħaamii muħaaminAccusative: al-muħaamiy-a muħaamiy-an

154 Appendix C


2.5.2. Diptote defective plural

‘cafés’ maqaahin

Definite Indefinite

Nominative: al-maqaahii maqaahinGenitive: al-maqaahii maqaahinAccusative: al-maqaahiy-a maqaahiy-a

3. ONE-WAY INFLECTION

3.1 . declension seven: indeclinable nouns

3.1.1. Singular indeclinable noun

‘level’ mustawan

Definite Indefinite

Nominative: al-mustawaa mustawanGenitive: al-mustawaa mustawanAccusative: al-mustawaa mustawan

3.2 . declension eight: invariable

3.2.1. Invariable noun

‘anniversary’ dhikraa

Definite Indefinite

Nominative: al-dhikraa dhikraaGenitive: al-dhikraa dhikraaAccusative: al-dhikraa dhikraa

Notes

1. For detailed description of these declensions see Ryding (2005: 182–204).2. This may be at least partially due to the fact that in both inflections, the final consonant is

inflectional nuun (-aani and -uuna).


C:/ITOOLS/WMS/CUP-NEW/4464129/WORKINGFOLDER/RARA/9781107023314GSY.3D 157 [157–166] 25.10.2013 8:11AM

Glossary of technical linguistic terms

In this basic glossary I have defined some items myself, but more often Ihave used definitions taken from contemporary studies and reference works onlinguistics, as noted. Also note the morphology glossary in Chapter 5 and the list ofsyntactic terms in Chapter 9.

affix: “A letter or sound, or group of letters or sounds (= a morpheme),which is added to a word, and which changes the meaning or functionof the word” (Richards and Schmidt 2010: 17). This includes prefixes,circumfixes, infixes and suffixes. Another definition states that “anaffix is a grammatical element, belonging to a closed set, which canonly function as a component of a word” (Cruse 1986: 77).

Agent: in theta theory, “the one who intentionally initiates the actionexpressed by the predicate” (Haegeman 1994: 49). Also, in case theory“the case of the typically animate perceived instigator of the actionidentified by the verb” (Butt: 2006: 30).

allomorph: a variant of a morpheme that does not alter its basic identityor function, e.g., different forms for the English plural morpheme suchas: books, dogs, houses, oxen, children, sheep; or the contextual var-iation of the indefinite article a/an that depends on the initial sound ofthe following word. In Arabic also, the plural morpheme takes onvarious shapes: sound feminine plural (-aat), sound masculine plural(-uuna/-iina), and the many variations of broken or internal plurals.The laam of the definite article (al-) has a wide range of allomorphicshapes because of the fact that “sun letters” assimilate it, and change itsrealization (e.g., al-dhahab ‘gold’ is pronounced adh-dhahab).

allophone: an allophone is a contextually determined pronunciationvariant of a phoneme.

assimilation: a change or spread of phonetic feature values (such asvoicing or velarization) that makes segments more similar, or evenidentical.

157


aux.: auxiliary verb; kaan is used as an auxiliary verb in Arabic (e.g.,kaan-at ta-drus-u, ‘she was studying’).

Beneficiary: in case theory, the person or entity that receives the benefitof the action. See also recipient.

binyan: “A binyan (plural binyanim) is a verbal paradigm in a Semiticlanguage, involving root-and-pattern morphology” (Bauer 2003: 325).In Arabic, binyan corresponds to wazn, or “Form” of the verb.

blend: “A blend is a new lexeme formed from parts of two or more otherlexemes” (Bauer 2003: 325).

bound morpheme: a grammatical formative that cannot occur on its ownas a word (e.g., in Englishword parts such as -ish, -ment, un-, -ly, -s, -ed).In Arabic, a lexical root (jidhr) such as {k-t-b} or {d-r-s}, a word patternor template such as {ma__ __ a __ } for noun of place, or case-markerssuch as -u, -i and -a). Arabic abounds in bound morphemes.

case: “Case is a system of marking dependent nouns for the type ofrelationship they bear to their heads” (Blake 1994: 1).

categorical specification: labeling of lexical or syntactic categories(e.g., noun, verb, adjective, preposition).

circumfix: a letter or sound, or group of letters or sounds which is addedat both ends of a word (i.e., prefix and suffix together), and whichchanges the meaning or function of the word. “A circumfix is adiscontinuous affix which surrounds the base with which it occurs”(Bauer 2003: 325–326). In Arabic, the present tense inflectional mor-phemes “surround” the present-tense stem; e.g., ta-drus-na ‘you f. pl.study’ consists of the prefix ta-, the stem -drus-, and the suffix -na.

citation form: the basic word stem listed in a dictionary, glossary, orword list. For Arabic verbs, it is the third person masculine singularpast tense; for English it is the infinitive.

clause: “An expression that contains (minimally) a subject and a predi-cate” (Fagan 2009: 283).

clitic: “A clitic is an obligatorily bound morph which is intermediatebetween an affix and a word” (Bauer 2003: 326).

complement: object or object-clause; “a term used in the analysis ofgrammatical function, to refer to a major constituent of sentence orclause structure, traditionally associated with ‘completing’ the actionspecified by the verb” (Crystal 1997b: 75)

complementary distribution: “If two elements never occur in the samecontexts, but, instead, divide up some set of contexts between them,they are said to be in complementary distribution” (Bauer 2003: 329).

158 Glossary of technical linguistic terms


complementizer: a subordinating conjunction that joins clauses (such asthat or whether in English; or such as ʔinna and her sisters in Arabic),creating an embedded sentence or complement.

compound: “A lexeme formed by adjoining two or more lexemes”(Fagan 2009: 283).

concatenative morphology: the formation of words through combina-tion of elements into a linear sequence.

conjugation: “A class of verbs that show the same pattern of inflection”(Booij 2005, 2007: 310).

copula: a word that links a sentential subject to a complement. Often aform of the verb ‘to be,’ but in Arabic, may also take the form of apersonal pronoun.

declension: “A class of nouns or adjectives with the same inflectionalpattern” (Booij 2005, 2007: 311).

degree: “The morphological marking on adjectives of different degreesof presence of a property” (Booij 2005, 2007: 311). The positive,comparative, and superlative forms of adjectives reflect differentdegrees of a property.

derivational morphology: the creation of lexical items, word stems.determiner: “A functional category that serves as the specifier of noun

phrase” (Fagan 2009: 284). The definite article and demonstrativepronouns are two types of determiners in Arabic.

discontinuous morphology: splitting one morpheme by insertion ofanother unit, such as the interlocking of grammatical patterns andlexical roots in Arabic.

distinctive feature: in phonology, “a particular characteristic which dis-tinguished one distinctive sound unit of a language (see phoneme)from another or one group of sounds from another” (Richards andSchmidt 2010: 179).

epenthesis: insertion of a sound into a sequence in order to ease pronun-ciation and facilitate transition from one sound to the next; in Arabic,usually a vowel sound.

equational: in Arabic, referring to sentences without an overt form of theverb ‘to be.’

experiencer: in case theory, refers “to an entity or person affected by theaction or state expressed by the verb” (Crystal 1997b: 143); “the thematicrole of the entity that feels or perceives something” (Fagan 2009: 285).

form class: “A set of forms displaying similar or identical grammaticalfeatures” (Crystal 1997b: 155).

Glossary of technical linguistic terms 159


formalization: “It ought to be possible, in any formalization, for alinguistic analysis to be formally interpreted in logical or mathematicalterms, and a calculus developed” (Crystal 1997b: 156). That is, rulesare specified in precise and abstract ways so as to be concise, max-imally applicable, and elegant.

formative: any element entering into word formation, either derivationalor inflectional.

free morpheme: can function independently as a word (e.g., Arabic min‘from’).

free variation: in phonology, refers to the optional “substitutability ofone sound for another in a given environment, with no consequentchange in the word’s meaning” (Crystal 1997b: 158).

head: the main or central part of a phrase; the head of an NP is a noun, thehead of a PP is a preposition, etc. “Most constructions can be describedin terms of an obligatory member (the head) and an optional mem ber(the dependent)” (Blake 1994: 201).

infix: a letter or sound, or group of letters or sounds, which is addedinside a word or morpheme, and which changes its meaning orfunction.

inflectional morphology: the study of word variation in context (e.g.,number, gender, case, definiteness, tense, voice, and person are somecategories of inflectional morphology). Inflections are applied to wordstems.

instrumental: in case theory, “the case of the inanimate force or objectcausally involved in the action or state identified by the verb” (Butt2006: 30).

item and arrangement (IA) morphology: a “model of morphologicalstructure in which a complex word consists of a sequence of concaten-ated morphemes” (Booij 2005, 2007: 315).

item-and-process (IP) morphology: a “model of morphological struc-ture in which each complex word is the output of one or more morpho-logical processes such as affixation and internal modification” (Booij2005, 2007: 315). It “sets up one underlying form for alternatingallomorphs and derives the surface forms by applying feature-changingrules to the underlying form” (Bybee 1988: 119).

lexeme: “A lexeme is a (potential or actual) decontextualized vocabu-lary word, a member of a major lexical category: noun (N), verb (V)or adjective/adverb (A)” (Aronoff 1992b: 13). Another definition is“the abstract unit that stands for the common properties of all the



forms of a word” (Booij 2005, 2007: 316). Cruse refers to lexemesas “the items listed in the lexicon, or ‘ideal dictionary,’ of a lan-guage” (Cruse 1986: 49).1 Aronoff states, “a lexeme, at least in itsextrasyntactic state, is uninflected, both abstractly and concretely”(1994: 11).

Lexeme-Morpheme Base Morphology (LMBM): “Is a theory of mor-phology which claims that lexical morphemes, called Lexemes, andgrammatical morphemes,Morphemes, are radically different linguisticphenomena” (Beard and Volpe 2005: 189) (emphasis in original).Essentially, lexemes are word-stems of the major syntactic categories(nouns, verbs, adjectives), whereas grammatical morphemes such astense, person, plural, and so forth, are non-lexical.

lexical category: the form class of a particular word, e.g., noun (n), verb(v), adjective (a), preposition (p).

Lexical Phonology andMorphology (LPM): a theory of the interactionof morphology and phonology, especially as pertaining to issues ofcyclicity and sequencing of morphophonological processes.

lexicon: essentially, the set of all words and idioms known to a nativespeaker of a language, including information on a word’s syntacticcategory (sometimes called “lexical category”), its grammatical func-tions and patterning, and its meaning/s.

locative: in case theory, “The case which identifies the location or spatialorientation of the state or action identified by the verb” (Butt 2006: 30).

manner of articulation: refers to “the kind of articulatory process usedin a sound’s production” (Crystal 1997b: 232).

modality: “A semantic category that involves the expression of differentattitudes towards or degrees of commitment to a proposition” (Fagan2009: 288), such as hoping, wishing, liking, believing, commanding.

mood/mode/: “Modality distinctions that are marked by verbal inflec-tion” (Fagan 2009: 288). In Arabic, the moods of the verbs are indica-tive, subjunctive, jussive, and imperative.

morph: “A morph is a constituent element of a word-form. It is therealization of a morpheme (or sometimes more than one, see portman-teau morph)” (Bauer 2003: 334).

morpheme: a minimum unit of form having an independent lexical orgrammatical meaning. “The morpheme is an abstract unit realized . . .

by morphs, or . . . allomorphs” (Bauer 2003: 334–335). Some theoriesof morphology crucially distinguish between lexical and grammaticalmorphemes (e.g., lexeme-morpheme base morphology).



morphology: the study of word-structure and word-formation, espe-cially in terms of morphemes and morphological processes.

morphophonology: the study of the interaction/interface between mor-phology and phonology.

morphosyntax: the study of the interface between morphology andsyntax.

non-concatenative morphology: “morphology that makes use of otherprocesses than affixation or compounding to create new words or wordforms” (Booij 2005, 2007: 318). The root/pattern morphology ofArabic is an example of non-concatenative morphology because theroot morphemes and the pattern morphemes are discontinuous and arecombined through interlinking rather than linear affixation.

objective: in case theory: “the semantically most neutral case, the caseof the thing representable by a noun whose role in the action or stateidentified by the verb is identified by the semantic interpretation ofthe verb itself; conceivably the concept should be limited to thingswhich are affected by the action or state identified by the verb” (Butt2006: 30).

paradigm: “A set of forms, corresponding to some subset (defined interms of a particular morphological category) of the grammatical wordsfrom a single lexeme. Paradigms are frequently presented in tabularform” (Bauer 2003: 337). For example, in Arabic, the possible forms ofa word can be listed in a table consisting of “cells” that constitute therange of word-form options possible in the language. For example, atriptote or fully inflectable (muʕrab) noun paradigm would look likethis, showing both case and definiteness within six cells:

Word-stem xanjar ‘dagger’Case definite indefiniteNominative al-xanjar-u xanjar-u-nGenitive al-xanjar-i xanjar-i-nAccusative al-xanjar-a xanjar-a-n

paronymy: “The relationship between one word and another belongingto a different syntactic category and produced from the first by someprocess of derivation” (Cruse 1986: 130).

patient: “Semantic role of the participant of an action that undergoes thataction” (Booij 2005, 2007: 319).

pattern (Arabic): a pattern is a bound and in many cases, discontinuousmorpheme consisting of one or more vowels and slots for root pho-nemes (radicals), which either alone or in combination with one to



three derivational affixes, interlocks with a root to form a stem, andwhich generally has grammatical meaning.

person: “A morphosyntactic category that identifies the participants in asituation. A typical distinction is among first person (the speaker or agroup including the speaker), second person (the person or personsaddressed), and third person (anyone else)” (Fagan 2009: 289).

phoneme: a minimal distinctive unit in the sound system of a language.phonemics/ phonemic analysis: the study of the sound system of a

language.phonetics: “The science which studies the characteristics of human

sound-making, especially those sounds used in speech, and providesmethods for their description, classification and transcription” (Crystal1997b: 289).

phonology: a general term that includes phonemics and phonetics. The“establishment and description of the distinctive sound units of alanguages (phonemes) by means of distinctive features” (Richardsand Schmidt 2010: 435).

phonotactics: is the study of sound distribution patterns and distributionrestrictions within words.

portmanteau morph: incorporates two (or more) meanings in one mor-pheme, such as the dual suffix in Arabic:

-aani = number (dual) and case (nominative)-ayni = number (dual) and case (accusative/genitive)

prefix: a letter or sound, or group of letters or sounds which is added tothe beginning of a word, and which changes the meaning or function ofthe word.

recipient: in case theory, “the semantic role for the participant in anevent that receives something” (Booij 2005, 2007: 320). See alsobeneficiary.

root (Arabic): a root is a relatively invariable discontinuous boundmorpheme, represented by two to five phonemes, typically three con-sonants in a certain order, which interlocks with a pattern to form astem and which has lexical meaning. In this book, usually referred to asa “lexical root.” Aronoff defines a root in more vivid terms: “A root iswhat is left when all the morphological structure has been wrung out ofa form. This is the sense of the term in Semitic grammar” (1992b: 15).

stem or word stem: the base or bare form of a word without inflectionalaffixes. A lexeme may have more than one stem.



stem allomorphy: the variation of a stem resulting from interaction withinflectional or derivational morphology.

suffix: a letter or sound, or group of letters or sounds which is added tothe end of a word, and which changes the meaning or function of theword.

syntactic category: “The syntactic category to which a word belongsdetermines its distribution, that is, in what contexts it can occur”(Haegeman 1994: 36).

syntax: “Syntax concerns the ways in which words combine to formsentences and the rules which govern the formation of sentences,making some sentences possible and others not possible within aparticular language” (Richards and Schmidt 2010: 579).

thematic role: “The semantic role of a participant in an event” (Booij2005, 2007: 323). See also theta theory.

theta theory: in government-binding theory, “the component of thegrammar that regulates the assignment of thematic roles is calledtheta theory” (Haegeman 1994: 49) (“theta” is shorthand for “theme”).

transcription: “The written representation of a language by symbols orspellings other than those of the standard orthography of the language”(Ferguson 1962: 10).

transliteration: “the systematic use of the symbols of one writing sys-tem to represent those of another, ideally one-to-one correspondence soit can be converted to original (Ferguson 1962: 10).

universal grammar (UG): “A theory which claims to account for thegrammatical competence of every adult no matter what language he orshe speaks” (Richards and Schmidt 2010: 617).

valence/valency: refers to the number and nature of semantic roles(arguments) that are associated with the meaning of a particularpredicate.

WFR: word formation rule. A rule that applies in derivationalmorphology.

wh-movement: the movement of an interrogative word (e.g., who,when, why) to the first part of a sentence.

word-and-paradigm morphology: a “theoretical model of inflectionthat takes the lexeme and its paradigm of cells as the starting pointfor the analysis of inflectional systems” (Booij 2005, 2007: 324).“Word-and-paradigm is an approach to morphology which gives the-oretical centrality to the notion of the paradigm and which derives theword-forms representing lexemes by a complex series of ordered rules



which do not assume that the word-form will be easily analyzable intomorphs or that each morph will realize a single morpheme. It is alsoknown as a-morphous morphology” (Bauer 2003: 344).

Note

1. Cruse makes a distinction between lexemes and “lexical units,” especially as these termsare used in lexical semantics (1986: 49–50).


C:/ITOOLS/WMS/CUP-NEW/4464129/WORKINGFOLDER/RARA/9781107023314RFA.3D 167 [167–180] 25.10.201312:15PM

References

Abdel Haleem, Muhammad. 2009. Qur’an. In Encyclopedia of Arabic Language andLinguistics, vol. IV, ed. Kees Versteegh, 21–31. Leiden: Brill.

Abdo,DaudA. 1969.OnStress andArabicPhonology: A generative approach. Beirut:Khayats.Abdo, Daud and Salwa N. Hilu. 1968. Arabic Writing and Sound Systems (Aşwaat

al-ʕarabiyya wa-ħuruuf-u-haa). Beirut: Ras Beirut Bookshop.Abdul-Raof, Hussein. 2006. Case roles. In Encylopedia of Arabic Language and

Linguistics, vol. I, ed. Kess Versteegh, 343–347. Leiden: Brill.AIDA: Association Internationale de Dialectologie Arabe. Website: www.aida.or.atÅkesson, Joyce. 2001. Arabic Morphology and Phonology. Leiden: Brill2009a. The Complexity of the Irregular Verbal and Nominal Forms and the PhonologicalChanges in Arabic. Lund: Pallas Athena.

2009b. Şarf. In Encyclopedia of Arabic Language and Linguistics, vol. IV,Kees Versteegh, ed., 118–122. Leiden: Brill.

Al-Ani, Salman H. 1970. An acoustical and physiological investigation of the Arabic ./ع/Actes du Xe congrès international des linguistes. Bucharest: Éditions de l’académie dela république socialiste de Roumanie.

1993. The treatise of Ibn Sina: An original contribution to Arabic phonetics. InInvestigating Arabic: Linguistic, pedagogical and literary studies in honor of ErnestN. McCarus, eds. Raji M. Rammuny and Dilworth B. Parkinson, 137–153. Columbus,OH: Greydon Press.

2008. Phonetics. In Encyclopedia of Arabic Language and Linguistics, vol. III, ed.Kees Versteegh, 593–603. Leiden: Brill.

Al-Batal, Mahmoud, ed. 1995a. The Teaching of Arabic as a Foreign Language. Provo, UT:American Association of Teachers of Arabic.

1995b. Issues in the teaching of the productive skills in Arabic. In The Teaching of Arabicas a Foreign Language, ed. Mahmoud Al-Batal. Provo, UT: American Association ofTeachers of Arabic.

2006. Facing the crisis: Teaching and learning Arabic in the United States in the post-September 11 era. ADFL Bulletin 37 (2/3): 39–46.

Alhawary, Mohammad T. 2009. Arabic Second Language Acquisition of Morphosyntax.New Haven, CT: Yale.

Al-Wer, Enam. 2013. Sociolinguistics. In The Oxford Handbook of Arabic Linguistics, ed.Jonathan Owens, 241–263. Oxford: Oxford University Press.

Amberber, Mengistu. 2000. Valency-changing and valency-encoding devices in Amharic.In Changing Valency: Case Studies in Transitivity, eds. R.M. Dixon and AlexandraY. Aichenvald, 312–332. Cambridge: Cambridge University Press.

167

http://www.aida.or.at


2009. Differential case marking of arguments in Amharic. The Oxford Handbook of Case,Andrej Malchukov and Andrew Spencer, eds. 742–755. Oxford: Oxford UniversityPress.

Anderson, John M. 2006. Modern Grammars of Case: A retrospective. Oxford: OxfordUniversity Press.

2009. Case in localist case grammar. In The Oxford Handbook of Case, eds.Andrej Malchukov and Andrew Spencer, 121–135. Oxford: Oxford University Press.

Anderson, Stephen R. 1988. Inflection. In Theoretical Morphology, eds. Michael Hammondand Michael Noonan, 23–43. New York: Academic Press.

1992. A-Morphous Morphology. Cambridge: Cambridge University Press.Aoun, Joseph E., Elabbas Benmamoun and Lina Choueiri. 2010. The Syntax of Arabic.

Cambridge: Cambridge University Press.Aronoff, Mark. 1976. Word Formation in Generative Grammar. Cambridge, MA: MIT

Press.ed. 1992a. Morphology Now. Albany: State University of New York Press.1992b. Stems in Latin verbal morphology. InMorphology Now, ed. Mark Aronoff, 5–32.

Albany: State University of New York Press.1994.Morphology by Itself: Stems and inflectional classes. Cambridge, MA: MIT Press.

Aronoff, Mark and Janie Rees-Miller, eds. 2001. The Handbook of Linguistics. Oxford:Blackwell.

Baalbaki, Ramzi. 2013. Arabic linguistic tradition I: Naħw and şarf. In The OxfordHandbook of Arabic Linguistics, ed. Jonathan Owens, 92–114. Oxford: OxfordUniversity Press.

Bachra, Bernard M. 2001. The Phonological Structure of the Verbal Roots in Arabic andHebrew. Leiden: Brill.

Badawi, El-Said. 1985. Educated spoken Arabic: A problem in teaching Arabic as a foreignlanguage. In Scientific and Humanistic Dimensions of Language, ed. KurtR. Jankowsky, 15–22. Washington, DC: Georgetown University Press.

Badawi, El-Said, Michael G. Carter, and Adrian Gully. 2004. Modern Written Arabic:A comprehensive grammar. London: Routledge.

Badecker, William and Alfonso Caramazza. 1989. Lexical distinction between inflectionand derivation. Linguistic Inquiry 20(1): 108–116.

Bahloul, Maher. 1994. The Syntax and Semantics of Taxis, Aspect, Tense and Modality inStandard Arabic. Ithaca, NY: Cornell University Department of Modern Languagesand Linguistics.

2006. Copula. In Encyclopedia of Arabic Language and Linguistics, vol. I, ed.Kees Versteegh, 506–511. Leiden: Brill.

Bakalla, Muhammad Hasan. 2009. Tafxīm. Encyclopedia of Arabic Language andLinguistics, vol. IV, ed. Kees Versteegh, 421–424. Leiden: Brill.

Barkat, Melissa. 2009. Vowel backing. Encyclopedia of Arabic Language and Linguistics,vol. IV, ed. Kees Versteegh, 669–674. Leiden: Brill.

Bassiouney, Reem. 2006. Functions of Code-Switching in Egypt. Leiden: Brill.2009. Arabic Sociolinguistics. Edinburgh: Edinburgh University Press.ed. 2010. Arabic and the Media. Leiden: Brill.

Bassiouney, Reem and E. Graham Katz, eds. 2010. Arabic Language and Linguistics.Washington, DC: Georgetown University Press.

168 References


Bateson, Mary Catherine. 1967, 2003. Arabic Language Handbook. Washington, DC:Georgetown University Press.

Bauer, Laurie. 2003. Introducing Linguistic Morphology. Washington, DC: GeorgetownUniversity Press.

2007. The Linguistics Student’s Handbook. Oxford: Oxford University PressBeard, Robert and Mark Volpe. 2005. Lexeme-morpheme base morphology. In Handbook

of Word-Formation, eds. Pavol Štekauer, and Rochelle Lieber, 189–205. Dordrecht:Springer.

Beeston, A. F. L. 1970, 2006. The Arabic Language Today. Washington, DC: GeorgetownUniversity Press.

Belguedj, M. S. 1973. La démarche des premiers grammariens arabes dans le domaine de lasyntaxe. Arabica; 168–185.

Belnap, Kirk. 1986. Complementation in Modern Standard Arabic: A corpus-basedapproach. Unpublished master’s thesis, Brigham Young University.

1999. A new perspective on the history of Arabic variation in marking agreement withplural heads. Folia Linguistica 33(2): 169–185.

2000. Methodological pitfalls in investigating grammatical agreement variation and thehistory of Arabic. In Proceedings of the Third International Conference of AIDA,Association Internationale de Dialectolgie Arabe, 41–46. Malta: n.p.

Belnap, R. Kirk and Osama Shabaneh. 1992. Variable agreement and nonhuman plurals inclassical Arabic and modern standard Arabic. In Perspectives on Arabic Linguistics IV,eds. Ellen Broselow, Mushira Eid, and John McCarthy, 245–262. Amsterdam/Philadelphia: John Benjamins.

Benmamoun, Elabbas. 2000. The Feature Structure of Functional Categories: A compara-tive study of Arabic dialects. Oxford: Oxford University Press.

2006. Construct state. In Encyclopedia of Arabic Language and Linguistics, vol. I, ed.Kees Versteegh, 477–482. Leiden: Brill.

2009. Syntax. In Encyclopedia of Arabic Language and Linguistics, vol. IV, ed.Kees Versteegh, 391–402. Leiden: Brill

Benmamoun, Elabbas and Lina Choueiri. 2013. The syntax of Arabic from a generativeperspective. In The Oxford Handbook of Arabic Linguistics, ed. Jonathan Owens,115–164. Oxford: Oxford University Press.

Birkeland, Harris. 1954. Stress Patterns in Arabic. Avhandlinger utgitt av det norkevidenskaps-akademi i Oslo II. Hist.-Filos. Klasse. No. 3. Oslo: I Kommisjon HosJacob Dybwad.

Blake, Barry. 1994. Case. Cambridge: Cambridge University Press.Bloomfield, Leonard. 1933, 1984. Language. Chicago, IL: University of Chicago Press.Bohas, Georges. 1997. Matrices, Étymons, Racines: Éléments d’une théorie lexicologique

du vocabulaire arabe. Leuven: Peeters.Bohas, Georges, Jean-Patrick Guillaume, and Djamal Eddine Kouloughli. 1990. The Arabic

Linguistic Tradition. London and New York: Routledge.Bohas, Georges and Abderrahim Saguer. 2006. Sur un point de vue heuristique conce-

rnant l’homonymie dans le lexique de l’arabe. In Grammar as a Window onto ArabicHumanism, eds. Lutz Edzard and Janet Watson, 130–154. Wiesbaden: Harrassowitz.

2007. The explanation of homonymy in the lexicon of Arabic. In Approaches to ArabicLinguistics, eds. Everhard Ditters and Harald Motzki, 255–289. Leiden: Brill.

References 169


Booij, Geert. 1996. Review of Lexeme-morpheme base morphology by Robert Beard.Language 72(4): 812–815.

2005, 2007. The Grammar of Words: An introduction to morphology. Oxford: OxfordUniversity Press.

Borer, Hagit. 2009. Afro-Asiatic, Semitic: Hebrew. In The Oxford Handbookof Compounding, Rochelle Lieber and Pavel Štekauer, eds., 491–511. Oxford:Oxford University Press.

Brame, Michael K. 1971. Stress in Arabic and generative phonology. Foundationsof Language 7: 556–591.

Brigham Young University. 2013. Website: Digital Humanities at Brigham YoungUniversity. At digitalhumanities.byu.edu/project/arabicorpus/.

Broselow, Ellen. 2008. Phonology. In Encyclopedia of Arabic language and Linguistics,vol. III, ed. Kees Versteegh, 607–615. Leiden: Brill.

Broselow, Ellen, Marie Huffman, Su-i Chen, and Ruohmei Hsieh. 1995. The timingstructure of CVVC syllables. In Perspectives on Arabic Linguistics VII, ed.Mushira Eid, 119–138. Amsterdam: John Benjamins.

Buckwalter, Tim and Dilworth B. Parkinson. 2011. A Frequency Dictionary of Arabic: Corevocabulary for learners. London: Routledge.

Buckwalter, Tim and Dilworth B. Parkinson. 2013. Modern lexicography. In The OxfordHandbook of Arabic Linguistics, ed. Jonathan Owens, 539–560. Oxford: OxfordUniversity Press

Butt, Miriam. 2006. Theories of Case. Cambridge: Cambridge University Press.2009a. Modern approaches to case: An overview. In The Oxford Handbook of Case, eds.Andrej Malchukov and Andrew Spencer, 27–43. Oxford: Oxford University Press.

2009b. Case in lexical-functional grammar. In The Oxford Handbook of Case, eds.Andrej Malchukov, and Andrew Spencer, 59–71. Oxford: Oxford University Press.

Bybee, Joan. 1988. Morphology as lexical organization. In Theoretical Morphology, eds.Michael Hammond, and Michael Noonan, 119–141. New York: Academic Press.

Cantarino, Vicente. 1974–76. The Syntax of Modern Arabic Prose. 3 vols. Bloomington, IN:Indiana University Press.

Carstairs, Andrew. 1987. Allomorphy in Inflexion. London: Croom Helm.Carstairs-McCarthy, Andrew. 1988. Nonconcatenative inflection and paradigm economy.

In Theoretical Morphology, eds.Michael Hammond andMichael Noonan, 71–77. NewYork: Academic Press.

1992. Current Morphology. London: Routledge.1994. Inflection classes, gender, and the principle of contrast. Language 70(4): 737–788.2005. Basic terminology. In Handbook of Word-Formation, eds. Pavol Štekauer andRochelle Lieber, 5–23. Dordrecht: Springer.

Carter, M. C. 1972. Les origines de la grammaire arabe. Revue des Études Islamiques. 40:69–97.

1973. An Arab grammarian of the eighth century, A.D. A contribution to the historyof linguistics. Journal of the American Oriental Society 93: 146–57.

1981. Arab Linguistics. Amsterdam: John Benjamins.1985. “When did the Arabic word naħw first come to denote grammar?” in Language &

Communication 5(4): 265–272.Chafe, Wallace L. 1970. Meaning and the Structure of Language. Chicago, IL: University

of Chicago Press.

170 References

http://digitalhumanities.byu.edu/project/arabicorpus/


Chekili, Ferid. 2009. Transformational grammar. In Encyclopedia of Arabic Language andLinguistics, vol. IV, ed. Kees Versteegh, 520–528. Leiden: Brill.

Chomsky, Noam. 1965. Aspects of the Theory of Syntax. Cambridge: MIT Press.1970. Remarks on nominalization. In Readings in English Transformational Grammar,eds. R. Jacobs and P. Rosenbaum, 184–221. Waltham, MA: Ginn and Company.

1974. Morphophonemics of Modern Hebrew. New York: Routledge.1981. Lectures on Government and Binding. Dordrecht: Foris.1995. The Minimalist Program. Cambridge, MA: MIT Press.2000. Minimalist inquiries: The framework. In Step by Step: Essays on minimalist syntaxin honor of Howard Lasnik, eds. R. Martin, D. Michaels, and J. Uriagereka, 89–156.Cambridge, MA: MIT Press.

Chomsky, Noam and Morris Halle. 1968. The Sound Pattern of English. New York:Harper & Row.

Chouairi, Raja. 2009. Arabic to where? Some problems that ail the Arabic teaching industry.NECTFL Review 64: 35–48.

Choueiri, Lina. 2009. WH-movement. In Encyclopedia of Arabic Language andLinguistics, vol. IV, ed. Kees Versteegh, 718–722. Leiden: Brill.

Cohn, Abigail. 2001. Phonology. In The Handbook of Linguistics, eds. Mark Aronoff andJanie Rees-Miller, 180–212. Oxford: Blackwell.

Cruse, D. A. 1986. Lexical Semantics. Cambridge: Cambridge University Press.2001. The lexicon. In The Handbook of Linguistics, Mark Aronoff and Janie Rees-Miller,eds., 238–264. Oxford: Blackwell.

Crystal, David. 1997a. The Cambridge Encyclopedia of Language. 2nd edn. Cambridge:Cambridge University Press.

1997b. Dictionary of Linguistics and Phonetics. 4th edn. Oxford: Blackwell.Culler, Jonathan. 1976, 1986. Ferdinand de Saussure. Ithaca, NY: Cornell University Press.Dahlgren, Sven-Olof. 1998.Word Order in Arabic. Gothenburg, Sweden: Acta Universitatis

Gothoburgensis.2009. Word order. In Encyclopedia of Arabic Language and Linguistics, vol. IV, ed.Kees Versteegh, 725–736. Leiden: Brill.

Danesi, Marcel and Andrea Rocci. 2009. Global Linguistics: An Introduction. Berlin:Mouton de Gruyter.

Danks, Warwick. 2011. The Arabic Verb: Form and meaning in the vowel-lengtheningpatterns. Amsterdam: John Benjamins.

Davis, Stuart. 1993. Arabic pharyngealization and phonological features. In Perspectives onArabic Linguistics, vol. V, eds. Mushira Eid and Clive Holes, 149–162. Amsterdam:John Benjamins.

2009. Velarization. Encyclopedia of Arabic Language and Linguistics, vol. IV, ed.Kees Versteegh, 636–638. Leiden: Brill.

Ditters, Everhard. 2006. Computational linguistics. In Encyclopedia of Arabic Languageand Linguistics, vol. I, ed. Kees Versteegh, 455–465. Leiden: Brill.

2013. Issues in Arabic computational linguistics. In The Oxford Handbook of ArabicLinguistics, ed. Jonathan Owens, 312–240. Oxford: Oxford University Press.

Ditters, Everhard and Harald Motzki, eds. 2007. Approaches to Arabic Linguistics. Leiden:Brill.

Dixon, R.M.W. 2010a. Basic Linguistic Theory. Volume I: Methodology. Oxford: OxfordUniversity Press.

References 171


2010b. Basic Linguistic Theory. Volume II: Grammatical topics. Oxford: OxfordUniversity Press.

Dunne, Michele Durocher. 2003. Democracy in Contemporary Egyptian PoliticalDiscourse. Amsterdam: John Benjamins.

Durand, Jacques. 1990. Generative and Non-linear Phonology. London: Longman.Edzard, Lutz and Janet Watson, eds. 2006. Grammar as a Window onto Arabic Humanism.

Wiesbaden: Harrassowitz.Ehret, Christopher. 1989. The origin of third consonants in Semitic roots: An internal

reconstruction (applied to Arabic). Journal of Afroasiatic Languages 3: 109–202.Eid, Mushira. 1990. Arabic linguistics: The current scene. In Perspectives on Arabic

Linguistics, vol. I. ed. Mushira Eid, 3–37. Amsterdam/Philadelphia, PA:John Benjamins.

1991. Verbless sentences in Arabic and Hebrew. In Perspectives on Arabic Linguistics III,eds. Bernard Comrie and Mushira Eid, 31–61. Amsterdam/Philadelphia, PA: JohnBenjamins.

2007. Arabic on the media: Hybridity and styles. In Approaches to Arabic Linguistics,eds. Everhard Ditters and Harald Motzki. Leiden: Brill.

Elgibali, Alaa, ed. 1996. Understanding Arabic: Essays in contemporary linguistics inhonor of El-Said Badawi. Cairo: American University in Cairo Press.

ed. 2005. Investigating Arabic: Current parameters in analysis and learning. Leiden:Brill.

Elkhafaifi, Hussein. 2005a. Listening comprehension and anxiety in the Arabic languageclassroom. Modern Language Journal 89(2): 206–220.

2005b. The effect of prelistening activities on listening comprehension in Arabic learners.Foreign Language Annals 38(4): 505–513.

Everson, Michael E. 2011. Best practices in teaching logographic and non-Roman writingsystems to L2 learners. Annual Review of Applied Linguistics, 31: Topics in secondlanguage pedagogy, 249–274. Cambridge: Cambridge University Press.

Fagan, Sarah M.B. 2009. German: A linguistic introduction. Cambridge: CambridgeUniversity Press.

Fassi Fehri, Abdelkader. 1993. Issues in the Structure of Arabic Clauses and Words.Dordrecht: Kluwer.

Ferguson, Charles. 1956. The emphatic el in Arabic. Language 32: 486–52.1959a. Diglossia. Word 15: 325–340.1959b. The Arabic koinè. Language 35: 616–630.1962. Glossary of Terms Related to Languages of the Middle East. Washington,DC: Center for Applied Linguistics.

1992. Principles and parameters in language and society. In Georgetown UniversityRoundtable on Languages and Linguistics, ed. James E. Alatis, 275–285.Washington, DC: Georgetown University Press.

Fillmore, Charles. 1968. The case for case. In Universals in Linguistic Theory, eds.Emmon Bach and Robert T. Harms, 1–88. NewYork: Holt, Rinehart, andWinston, Inc.

1976. Foreword to Essentials of Grammar by Domenico Parisi and Francesco Antinucci,ix–x. New York: Academic Press.

1977. The case for case reopened. In Grammatical Relations, eds. Peter Cole and JerroldM. Saddock. 59–81. New York: Academic Press.

172 References


Fischer, Wolfdietrich. 1992. Arabic. International Encyclopedia of Linguistics, vol. I. NewYork & Oxford: Oxford University Press.

2002. A Grammar of Classical Arabic. 3rd rev. edn, trans. Jonathan Rogers. New Haven,CT and London: Yale University Press.

1997. Classical Arabic. In The Semitic Languages, ed. Robert Hetzron, 187–219.London: Routledge.

2002. A Grammar of Classical Arabic. 3rd rev. edn, trans. Jonathan Rogers. New Haven,CT and London: Yale University Press.

Fleisch, Henri. 1957. Esquisse d’un historique de la grammaire arabe. Arabica, 4: 1–221961, 1979. Traité de philologie arabe I et II. 2 vols. Beirut: Imprimerie Catholique 1961

(vol. I); Beirut: Dâr al-Machreq 1979 (vol. II).Frisch, Stefan A. 2008. Phonotactics. In Encyclopedia of Arabic language and Linguistics,

vol. III, ed. Kees Versteegh, 624–628.. Leiden: Brill.Gafos, Adamantios I. 2009. Stem. In Encyclopedia of Arabic language and Linguistics, vol.

IV, ed. Kees Versteegh, 338–344. Leiden: Brill.Gairdner, W.H. T. 1925. The Phonetics of Arabic. London: Oxford University Press.Givón, Talmy. 1980. The binding hierarchy and the typology of complements. Studies in

Language 4(3): 333–377.Goldberg, Adele E. 1992. Constructions: A construction grammar approach to argument

structure. Chicago, IL: University of Chicago Press.1995. A Construction Grammar Approach to Argument Structure. Chicago, IL: Universityof Chicago Press.

2006. Constructions at Work. Oxford: Oxford University Press.Gordon, Cyrus. 1970. The accidental invention of the phonemic alphabet. Journal of Near

Eastern Studies 29(3): 193–97.Gray, Louis. 1934. Introduction to Semitic Comparative Linguistics. New York: Columbia

University Press.Green, Georgia and Jerry L. Morgan.1996. Practical Guide to Syntactic Analysis. Palo Alto,

CA: CSLI Publications.Greenberg, Joseph. 1950. The patterning of root morphemes in Semitic. Word 6: 162–181.Gruber, Jeffrey S. [1965]1976. Lexical Structures in Syntax and Semantics. Amsterdam:

North-Holland.2001. Thematic relations in syntax. The Handbook of Contemporary Syntactic Theory,eds. Mark Baltin and Chris Collins, 257–298. Oxford: Blackwell.

Haegeman, Liliane. 1994. Introduction to Government and Binding Theory. 2nd edn.Oxford: Blackwell.

Haeri, Niloofar. 2003. Sacred Language, Ordinary People: Dilemmas of culture and politicsin Egypt. New York: Palgrave Macmillan.

2006. Culture and language. In Encyclopedia of Arabic Language and Linguistics, vol. I,ed. Kees Versteegh, 527–536. Leiden: Brill.

Hall, Christopher J. 2005. An Introduction to Language and Linguistics: Breaking thelanguage spell. London: Continuum.

Hammond,Michael andMichael Noonan. 1988. Morphology in the generative paradigm. InTheoretical Morphology, eds. Hammond, Michael and Michael Noonan, 69–90. NewYork: Academic Press.

Hardcastle, William J. and John Laver, eds. 1997. The Handbook of Phonetic Sciences,589–618. Oxford: Blackwell.

References 173


Hary, Benjamin. 1996. The importance of the language continuum in Arabic multiglossia. InUnderstanding Arabic, ed. Alaa Elgibali. Cairo: American University in Cairo Press.

Hasan, Abbas. 1987. Al-naħw al-waafii. (A complete grammar). 4 vols. Cairo: Daaral-maʕaarif.

Haspelmath, Martin. 2009. Terminology of case. The Oxford Handbook of Case, ed. byAndrej Malchukov and Andrew Spencer, 505–517. Oxford: Oxford University Press.

Heine, Bernd. 1997. Cognitive Foundations of Grammar. Oxford: Oxford University Press.Heine, Bernd and Tania Kuteva. 2007. The Genesis of Grammar. Oxford: Oxford University

Press.Hetzron, Robert, ed. 1997a. The Semitic Languages. Oxford: Routledge.

1997b. Preface. The Semitic Languages. Oxford: Routledge.Hijazi, Mahmoud Fahmi. 1978. Al-lugha l-ʕarabiyya ʕabr al-quruun (The Arabic Language

across the Centuries). Cairo: Daar al-thaqaafa li-l-ŧabaaʕa wa-l-nashr.Hjelmslev, Louis. [1935], 1972. La Catégorie des Cas. (‘The Category of Case’). Munich:

Wilhelm Fink Verlag.Holes, Clive. 2004. rev. edn. Modern Arabic: Structures, Functions and Varieties.

Washington, DC: Georgetown University Press.Howell, Mortimer Sloper. 1986 (repr.). A Grammar of the Classical Arabic Language.

4 vols. in 7. New Delhi: Gian Publishing.Hoyt, Frederick. 2008. Noun phrase. In Encyclopedia of Arabic Language and Linguistics,

vol. III, ed. Kees Versteegh, 428–434. Leiden: Brill.2009a. Verb phrase. In Encyclopedia of Arabic Language and Linguistics, vol. IV, ed.Kees Versteegh, 646–652. Leiden: Brill.

2009b. Verbal clause. In Encyclopedia of Arabic Language and Linguistics, vol. IV, ed.Kees Versteegh, 653–659. Leiden: Brill.

Hymes, Dell, ed. 1974. Studies in the History of Linguistics: Traditions and Paradigms.Bloomington, IN: Indiana University Press.

Jackendoff, Ray. S. 1972. Semantic Interpretation in Generative Grammar. Cambridge,MA: MIT Press.

Jakobson, Roman. 1990. On Language. Eds. Linda R. Waugh and Monique Monville-Burston. Cambridge, MA: Harvard University Press.

Jespersen, Otto. 1949. Language; Its Nature, Development, and Origin. New York:Macmillan.

Jesry, Maher. 2009. Syllable structure. In Encyclopedia of Arabic Language and Linguistics,vol. IV, ed. Kees Versteegh, 387–389. Leiden: Brill.

Katamba, Francis. 1993. Morphology. New York: Saint Martin’s Press.Kaye, Alan S. and Judith Rosenhouse. 1997. Arabic dialects and Maltese. In The Semitic

Languages, ed. Robert Hetzron, 263–311. Oxford: Routledge.Al-Khalil ibn Ahmad. 1970–1977. Kitāb al-ʻayn (The Book of ʻAyn’). 7 vol. Eds.

M. Makhzoumi and I. Samarrā’i. Baghdad: Dar al-rashid li-l-nashr.Kouloughli, Djamel Eddine. 2007. Inflection. In Encyclopedia of Arabic Language and

Linguistics, vol. II, ed. Kees Versteegh, 345–354. Leiden: Brill.Kossman, Maarten. 2013. Borrowing. In The Oxford Handbook of Arabic Linguistics, ed.

Jonathan Owens, 349–368. Oxford: Oxford University Press.Ladefoged, Peter. 1997. Linguistic phonetic descriptions. In The Handbook of Phonetic

Sciences, eds. William J. Hardcastle and John Laver, 589–618. Oxford: Blackwell.

174 References


Larcher, Pierre. 1998. La linguistique arabe d’hier à demain: tendences nouvelles de larecherche. Arabica 45(3): 409–429.

2009. verb. In Encyclopedia of Arabic Language and Linguistics, vol. IV, edKees versteegh, 638-645. Leiden: Brill.

2013. Arabic linguistic tradition II: Pragmatics. In The Oxford Handbook of ArabicLinguistics, ed. Jonathan Owens, 185–212. Oxford: Oxford University Press.

Larsen-Freeman, Diane and Lynne Cameron. 2008. Complex Systems and AppliedLinguistics. Oxford: Oxford University Press.

Laver, John. 2001. Linguistic phonetics. In The Handbook of Linguistics, eds. Mark Aronoffand Janie Rees-Miller, 150–179. Oxford: Blackwell.

LeComte, Gerard. 1968. Grammaire de l’arabe. Paris: Presses universitaires de France.Lehn, Walter and William R. Slager. 1983. A contrastive study of Egyptian Arabic and

American English: The segmental phonemes. In Second Language Learning:Contrastive analysis, error analysis, and related aspects, eds. BettyWallace Robinett and Jacquelyn Schachter, 32–40. Ann Arbor, MI: Universityof Michigan Press.

Lieber, Rochelle. 1996. Review of Combinatorial Morphology by John T. Stonham.Language 72(1): 130–132.

2004. Morphology and Lexical Semantics. Cambridge: Cambridge University Press.Lieber, Rochelle and Pavol Štekauer, eds. 2009a. The Oxford Handbook of Compounding.

Oxford: Oxford University Press.2009b. Introduction: Status and definition of compounding. In The Oxford Handbook ofCompounding, Rochelle Lieber and Pavel Štekauer, eds., 3–18. Oxford: OxfordUniversity Press.

LeTourneau, Mark. 2006. Case theory. In Encyclopedia of Arabic Language andLinguistics, vol. I, ed. Kees Versteegh, 347–353. Leiden: Brill.

2009. Theta roles. In Encyclopedia of Arabic Language and Linguistics, vol. IV, ed.Kees Versteegh, 487–494. Leiden: Brill.

Levin, Beth andMalka Rappaport-Hovav. 1995.Unaccusativity. Cambridge,MA:MIT Press.1998. Morphology and lexical semantics. In The Handbook of Morphology, ed. byAndrew Spencer and Arnold M. Zwicky, 248–271. Oxford: Blackwell.

Linguist List. 2012. Announcement: Interdisciplinary workshop on the grammatical word.October 26. Issue 23.4467.

Maalej, Zouhair. 2009. Valency. In Encyclopedia of Arabic Language and Linguistics, vol.IV, ed. Kees Versteegh, 624–627. Leiden: Brill.

McCarthy, John. 1982. Formal Problems in Semitic Phonology and Morphology.Bloomington, IN: Indiana University Linguistics Club.

2008. Morphology. In Encyclopedia of Arabic Language and Linguistics, vol. III, ed.Kees Versteegh, 297–307. Leiden: Brill.

McCarthy, John and Alan Prince. 1990a. Prosodic morphology and templatic morphology.In Perspectives on Arabic Linguistics, Vol. II, eds. Mushira Eid and John McCarthy,1–54. Amsterdam: John Benjamins.

1990b. Foot and word in prosodic morphology: The Arabic broken plural. NaturalLanguage and Linguistic Theory 8(2): 209–284.

Majdi, Basim. 1990. Word order and proper government in classical Arabic. In Perspectiveson Arabic Linguistics, vol. I, ed. Mushira Eid, 127–153. Amsterdam: John Benjamins.

References 175


Majdi, Basim and Millicent Winston. 1993. Gemination and antigemination in Iraqi. InPerspectives on Arabic Linguistics, Vol. V, eds. Mushira Eid and Clive Holes,163–191. Amsterdam: John Benjamins.

Makhzoumi, M. 1960. Al-Khalīl ibn Aħmad: aʕmāl-u-hu wa manhaj-u-hu al-zahraa(Al-Khalil ibn Ahmad: His works and his method). Baghdad: Maktabat al-zahaa’.

Malchukov, Andrej and Andrew Spencer, eds. The Oxford Handbook of Case. Oxford:Oxford University Press.

Măcelaru, Adrian. 2006. Causative. Encyclopedia of Arabic Language and Linguistics,vol. I, Kees Versteegh, ed., 355–360. Leiden: Brill.

Mejdell, Gunvor. 2006. Mixed Styles in Spoken Arabic in Egypt. Leiden: Brill.Matthews, P. H. 1974. Morphology: An introduction to the theory of word-structure.

Cambridge: Cambridge University Press.McCarus, Ernest and Raji Rammuny. 1974. “Programmed Course in Modern Literary

Arabic Phonology and Script.” Ann Arbor, MI: University of Michigan Departmentof Near Eastern Studies.

Mejdell, Gunvor. 2006.Mixed Styles in Spoken Arabic in Egypt: Somewhere between orderand chaos. Leiden: Brill.

Mitchell, T. F. 1990–1993. Pronouncing Arabic. 2 vols. Oxford: Clarendon.Mohammad, M.A. 1990. The problem of subject-verb agreement in Arabic: Towards a

solution. In Perspectives on Arabic Linguistics, vol. I, ed. Mushira Eid, 95–126.Amsterdam: John Benjamins.

Moore, John. 1990. Doubled verbs in modern standard Arabic. In Perspectives on ArabicLinguistics, vol. II, eds. Mushira Eid and John McCarthy, 55–93. Amsterdam: JohnBenjamins.

Moscati, Sabatino.1969. An Introduction to Comparative Grammar of the SemiticLanguages. Weisbaden: Harrassowitz.

Mrayati, M. 1987. Statistical studies of Arabic language roots. In Applied Arabic Linguisticsand Signal and Information Processing, ed. Raymond Descout, 97–103. Washington,DC: Hemisphere.

Al-Nadīm, Muħammad ibn Isħāq. 1950. Fihrist (Index). 2 vols. Ed. and trans.Bayard Dodge. New York: Columbia University Press.

Nydell, Margaret K. (Margaret Kleffner Omar). 1967. The verbal roots of modern writtenArabic: Classification and analysis. Master thesis, Department of Linguistics,Georgetown University.

Ouhalla, Jamal. 1997. Genitive subjects and the VSO order. In Studies on UniversalGrammar and Typological Variation, eds. T. Alan Hall and Artemis Alexiadou,197–218. Amsterdam: John Benjamins.

Owens, Jonathan. 1988. The Foundations of Grammar: An introduction to medieval Arabicgrammatical theory. Amsterdam: John Benjamins.

1990. Early Arabic Grammatical Theory: Heterogeneity and standardization.Amsterdam: John Benjamins.

1997. The Arabic grammatical tradition. In The Semitic Languages, ed. Robert Hetzron,46–58. Oxford: Routledge.

2006. A Linguistic History of Arabic. Oxford: Oxford University Press.2013a, ed. The Oxford Handbook of Arabic Linguistics. Oxford: Oxford University Press.2013b, History. In The Oxford Handbook of Arabic Linguistics, ed. J. Owens, 451–471.

Oxford: Oxford University Press.

176 References


Parisi, Domenico and Francesco Antinucci. 1976. Essentials of Grammar. New York:Academic Press.

Parkinson, Dilworth. 1991. Searching for modern fusha: Real-life formal Arabic. Al-Arabiyya 24: 31–64.

2003. Verbal features in oral fuşħa performances in Cairo. International Journal of theSociology of Language, 163: 27–41.

Parrish, Wayland Maxfield. 1932. Reading Aloud: A technique in the interpretationof literature. New York: Thomas Nelson.

Percival, W. Keith. 1976. The applicability of Kuhn’s paradigms to the history of linguistics.Language 52: 285–94.

Perlmutter, David. 1988. The split morphology hypothesis. In Theoretical Morphology,eds. Michael Hammond and Michael Noonan, 79–100. New York: Academic Press.Perspectives on Arabic Linguistics. 1990–2011. Vol. I-XXI. Amsterdam: JohnBenjamins.

Persson, Maria. 2002. Sentential Object Complements in Modern Standard Arabic.Stockholm: Almqvist & Wikesell.

Pierrehumbert, Janet. 1992. Dissimilarity in the Arabic verbal roots. In Proceedings of theNorth East Linguistic Society 23, University of Ottawa, vol.I., ed. Amy J. Schaefer,1–15. Amherst, MA: Graduate Linguistics Student Association.

Pinker, Steven. 1989. Learnability and Cognition: The acquisition of argument structure.Cambridge, MA: MIT Press.

Primus, Beatrice. 2009. Case, grammatical relations, and semantic roles. In The OxfordHandbook of Case, eds. Andrej Malchukov and Andrew Spencer, 261–275. Oxford:Oxford University Press.

Procházka, Stephan. 1993. Some remarks on the semantic function of the redupli-cated quadriliteral verb. The Arabist: Budapest studies in Arabic 6–7: Part I:97–103.

2008. Prepositions. In Encyclopedia of Arabic Language and Linguistics, vol. III,Kees Versteegh, ed., 699–703. Leiden: Brill.

Radford, Andrew. 1988. Transformational Grammar. A first course. Cambridge: CambridgeUniversity Press.

Rammuny, Raji M. and Dilworth B. Parkinson, eds. 1993. Investigating Arabic: Linguistic,pedagogical and literary studies in honor of Ernest N. McCarus. Columbus, OH:Greydon Press.

Ratcliffe, Robert. 2013. Morphology, In The Oxford Handbook of Arabic Linguistics, ed.Jonathan Owens, 71–91. Oxford: Oxford University Press.

Reichmuth, Philipp. 2009. Transcription. In Encyclopedia of Arabic Language andLinguistics, vol. IV, ed. Kees Versteegh, 515–520. Leiden: Brill.

Retsö, Jan. 2013. What is Arabic? In The Oxford Handbook of Arabic Linguistics, ed.Jonathan Owens, 433–450. Oxford: Oxford University Press.

Richards, Jack C. and Richard Schmidt. [1985] 2010. Longman Dictionary of LanguageTeaching and Applied Linguistics. Harlow, UK: Longman.

Robinett, Betty Wallace and Jacquelyn Schachter, eds. 1983. Second Language Learning:Contrastive analysis, error analysis, and related aspects. Ann Arbor, MI: University ofMichigan Press.

Rosenbaum, Gabriel M. 2000. Fuşħāmiyya. Alternating style in Egyptian prose. Zeitschriftfür Arabische Linguistik. 38: 68–87;

References 177


Ryding, Karin C. 1991. Proficiency despite diglossia: A new approach for Arabic. ModernLanguage Journal 75(2): 212–218.

1993. Case/mood syncretism in Arabic grammatical theory: Evidence for the split morphol-ogy hypothesis and the continuum hypothesis. In Investigating Arabic: Linguistic,pedagogical and literary studies in honor of Ernest N. McCarus, eds. RajiM. Rammuny and Dilworth B. Parkinson, 173–179. Columbus, OH: Greydon Press.

2005. A Reference Grammar of Modern Standard Arabic. Cambridge: CambridgeUniversity Press.

2006a. Teaching Arabic in the United States. In A Handbook for Arabic LanguageTeaching Professionals in the 21st Century, ed. Kassem Wahba, Zeinab Taha, andLiz England, 13–20. Mahwah, NJ: Lawrence Erlbaum.

2006b. Educated Arabic. Encyclopedia of Arabic Language and Linguistics, ed.Kees Versteegh, vol. I: 666–671. Leiden: Brill.

2007. Forum on Language Policy and the Politics of Language. ADFL Bulletin, 38(1–2): 61–62.

2008. New Alignments, New Discourses, Profession 2008, 214–218.2009. Educated spoken Arabic: A flexible Spoken Standard. In NECTFL Review 64:spring/summer 2009, 49–52.

2010a. Media Arabic as a regional standard. In Arabic and the Media, ed.Reem Bassiouney, 219–228. Leiden: Brill.

2010b. Postmodern models for Arabic: Answering old questions in new ways. ADFLBulletin 41(2): 69–71.

2011. Arabic datives, ditransitives, and the preposition li–. In In the Shadow of Arabic:Festschrift for Ramzi Baalbaki, ed. Bilal Orfali, 283–298. Leiden: Brill.

2013a. Arabic second language acquisition. In Handbook of Arabic Linguistics, ed.Jonathan Owens. Oxford: Oxford University Press.

2013b Teaching and Learning Arabic as a Foreign Languate: A guide for teachers.Washington, DC: Georgetown University Press.

Ryding, Karin and Kees Versteegh, 2007. IDaafa. In Encyclopedia of Arabic Language andLinguistics, vol. II, ed. Kees Versteegh, 294–298. Leiden: Brill.

Ryding-Lentzner, Karin. 1977. Semantic and syntactic aspects of Arabic prepositions.Dissertation, Georgetown University.

1981. Semantic motivation for Arabic dative-movement. Al-Arabiyya 14: 19–23.Sampson, Geoffrey, David Gil, and Peter Trudgill, eds. 2009. Language Complexity as an

Evolving Variable. Oxford: Oxford University Press.Sapir, Edward. 1921, 1949. Language. New York: Harcourt Brace & World.Sara, Solomon. 1991. Al-Khalil, the first Arab phonologist. International Journal of Islamic

and Arabic Studies 8(1): 1–57.de Saussure, Ferdinand. 1972 (Payot, Paris). 1983 copyright English translation. 2009

[1986]. Course in General Linguistics, ed. Charles Bally, Albert Sechehaye withAlbert Riedlinger. Trans. and ed. Roy Harris. Chicago, IL: Open Court.

Sawaie, Mohammed. 1993. Sociolinguistic factors and historical linguistic change: The caseof /q/ and its reflexes in Arabic dialects. In Investigating Arabic: Linguistic, pedagog-ical and literary studies in honor of Ernest N. McCarus, eds. Raji M. Rammuny andDilworth B. Parkinson, 15–39. Columbus, OH: Greydon Press.

Schramm, Gene M. 1962. An outline of classical Arabic verb structure. Language 38(4): 360–375.

178 References


Selkirk, Elizabeth O. 1982. The Syntax of Words. Cambridge, MA: MIT Press.Semaan, Khalil. 1968. Linguistics in the Middle Ages: Phonetic studies in early Islam.

Leiden: Brill.Shahin, Kimary N. 1996. Accessing pharyngeal place in Palestinian Arabic. In Perspectives

on Arabic Linguistics, Vol. IX, eds. Mushira Eid and Dilworth Parkinson., 131–150.Amsterdam: John Benjamins.

Shlonsky, Ur. 1997. Clause Structure and Word Order in Hebrew and Arabic. Oxford:Oxford University Press.

Shopen, Timothy, ed. Language Typology and Syntactic Description. 2nd ed, vol. III,Grammatical categories and the lexicon. Cambridge: Cambridge University Press.

Sibawayhi, Abu Bishr ʻAmr bn ʻUthman bin Qanbar. 1991. Kitaab Siibawayhi, ed.Abd al-Salam Muhammad Harun. Beirut: Daar al-jiil.

Smeaton, B. Hunter. 1956. Some problems in the description of Arabic.Word 12: 357–368.Soltan, Osama. 2006. Standard Arabic subject-verb agreement asymmetry revisited in an

Agree-based minimalist syntax. In Agreement Systems, ed. Cedric Boeckx, 239–264.Amsterdam: John Benjamins.

2007. On the individual-property contrast in free state possessive nominals in EgyptianArabic. In Perspectives on Arabic Linguistics, Vol. XX, ed. Mustafa Mughazy, 71–86.Amsterdam: John Benjamins.

2009. Agree in the functional domain: Evidence from the morphosyntax of positive andnegative imperatives in standard Arabic. In Proceedings of the 39th Annual Conferenceon African Linguistics, eds. Akinloye Ojo and Lioba Moshi, 86–99. Somerville, MA:Cascadilla Proceedings Project.

2010. On licensing wh-scope: wh-questions in Egyptian Arabic revisited. In ArabicLanguage and Linguistics, eds. Reem Bassiouney, E. Graham Katz, 99–114.Washington, DC: Georgetown University Press.

2011. On issues of Arabic syntax: an essay in syntactic argumentation. Review article ofThe Syntax of Arabic by Aoun, Banmamoun and Choueiri. Brill’s Annual of AfroasiaticLanguages and Linguistics 3: 236–280.

2012. Morphosyntactic effects of NPI-licensing in Cairene Egyptian Arabic: The puzzleof -š disappearance resolved. In Proceedings of the 29th West Coast Conference onformal Linguistics, eds. Jaehoon Choi et al., 241–249. Somerville, MA: CascadillaProceedings Project.

Spencer, Andrew. 1991. Morphological Theory: An introduction to word structure ingenerative grammar. Oxford: Blackwell.

1994. Review of Morphology by Itself by Mark Aronoff. Language 70(4): 811–817.2001. Morphology. In The Handbook of Linguistics, eds. Mark Aronoff and Janie Rees-Miller, 213–237. Oxford: Blackwell.

2009. Case as a morphological phenomenon. In The Oxford Handbook of Case, eds.Andrej Malchukov and Andrew Spencer, 185–199. Oxford: Oxford University Press.

Spencer, Andrew and ArnoldM. Zwicky, eds. 1998. The Handbook of Morphology. Oxford:Blackwell.

Štekauer, Pavol and Rochelle Lieber, eds. 2005. Handbook of Word-Formation. Dordrecht:Springer.

Stetkevych, Jaroslav. 1970, 2006. The Modern Arabic Literary Language. Washington, DC:Georgetown University Press.

References 179


Stump, Gregory T. 1998. Inflection. In The Handbook of Morphology, Andrew Spencer andArnold M. Zwicky, eds. 13–43. Oxford: Blackwell.

Suleiman, Yasir. 2003. The Arabic Language and National Identity: A study in ideology.Washington, DC: Georgetown University Press.

2004. AWar of Words: Language and conflict in the Middle East. Cambridge: CambridgeUniversity Press.

2011. Arabic, Self, and Identity. Cambridge: Cambridge University Press.Sword, Helen. 2012. Stylish Academic Writing. Cambridge, MA: Harvard University Press.2013. Arabic folk linguistics: Between mother tongue and native language. In The OxfordHandbook of Arabic Linguistics, ed. Jonathan Owens, 264–280. Oxford: OxfordUniversity Press.

Talmon, Rafael. 1982. Nahwiyyûn in Sîbawayhi’s kitâb. ZAL 8: 12–38.1985. Who was the first Arab grammarian? A new approach to an old problem. ZAL 15:128–145.

Tesnière, Lucien. 1959. Éléments de Syntaxe Structurale. Paris: KlincksieckVan Valin, Robert D., 2001. Functional linguistics. In The Handbook of Linguistics, eds.

Mark Aronaoff and Janie Rees-Miller, 319–336. Oxford: BlackwellVersteegh, C. H.M.(Kees). 1978. The Arabic terminology of syntactic position. Arabica

25(3): 261–81.1980b. The origin of the term ‘qiyās’ in Arabic grammar. Zeitschrift für Arabische

Linguistik 4: 7–30.1997. The Arabic Language. New York: Columbia University Press.ed. 2006–2009. Encyclopedia of Arabic Language and Linguistics. 4 vols. Leiden: Brill.

Wahba, Kassem, Zeinab Taha, and Liz England, eds., 2006. A Handbook for ArabicLanguage Teaching Professionals in the 21st Century. Mahwah, NJ: LawrenceErlbaum.

Wahba, Kassem. 2006. Arabic language use and the educated language user. In A Handbookfor Arabic Language Teaching Professionals in the 21st Century, eds. Kassem Wahba,Zeinab Taha and Liz England, 139–156. Mahwah, NJ: Lawrence Erlbaum.

Wright, Elizabeth K. 2000. Sound and meaning in medieval Arabic linguistic theory.Dissertation, Georgetown University.

Younes, Munther. 1993. Emphasis spread in two Arabic dialects. In Perspectives on ArabicLinguistics, vol. V, eds. Mushira Eid and Clive Holes, 119–145. Amsterdam: JohnBenjamins.

2009. The case for integration in the Arabic-as-a-foreign-language classroom. NECTFL,Review 64: 59–67.

Zemánek, Petr. 2006. Assimilation. In Encyclopedia of Arabic Language and Linguistics,vol. I, ed. Kees Versteegh, 204–206. Leiden: Brill.

2009. Root. In Encyclopedia of Arabic Language and Linguistics, vol. IV, ed.Kees Versteegh, 93–100. Leiden: Brill.

180 References

C:/ITOOLS/WMS/CUP-NEW/4466921/WORKINGFOLDER//9781107023314IND.3D 181 [181–188] 26.10.2013 10:14AM

Index

ʕaamil (pl. ʕawaamil) 90, 91, 108, 127lafziyy 91maʕnawiyy 91

accusative (see case, accusative)acoustic phonetics 13acronyms 83adjective 83, 95, 127adjective iđaafa 82adjective phrase 123adverbs 79, 80locative 96

affix 48, 157derivational 64ff.inflectional 99ff.

affricate 15Agent (case role) 111, 113, 120, 128, 131, 157agreement 43, 90ff., 110, 112, 113, 119, 128asymmetry 248:2categories 90default 131deflected 93, 120

allomorph 48, 157allophone 13, 18ff., 157ʕamal (syntactic government) 107, 108analogical modeling 56analogy (qiyaas) 56annexation structure 122antepenult 36, 37apposition 119ff.Arabicand Semitic 3as a foreign language viiiclassical 3, 6, 142

corpora 3grammarians 7grammatical tradition 7ff.media 3modern standard (MSA) vii, 6, 14, 142registers 3, 4, 142

ArabiCorpus 143, 298:18argument structure 128arguments (syntactic) 60, 110, 111, 128articulation 15articulatory phonetics 13aspect and tense 92, 93assimilation 18, 23, 157across word boundaries 26at a distance 25full 24hamza 24laam 18, 25, 59:1nuun 26partial 24, 26progressive 23reciprocal 24regressive 25waaw 24, 25

article, definite 121aux 113, 158ʔawzaan (sing. wazn) (see also verb, Forms) 56,

57, 132

base (morphological) 56basic linguistic theory (BLT) 23:12, 110Beneficiary (case role) 111, 113, 120, 128, 133, 158binyan 57, 158

181


blend, blending 110:10, 83, 158borrowing 48, 55, 85, 85ff., 85bound morpheme (see morpheme, bound)

C (constituent) 113calque 82, 86case 90, 92, 94, 96ff., 97, 158accusative 71, 123, 127, 128, 133, 135definition 97frames 111genitive 89, 111, 121, 122, 123grammar 110, 128markers 46marking 96, 149nominative 111, 127relations 96, 110roles 112, 132theory 96, 110, 131, 132

categorical specification 113, 158cell (paradigm) 93, 98, 99Chomsky, Noam 43, 108, 109, 114ff.circumfix 49, 65, 93, 158citation form 49, 158clause 158clitic 158code-mixing 142, 143coining 110:10, 142, 143collocations 143comparative 95comparison 92competence and performance 4complement 113, 129, 158clauses 134

complementary distribution 158complementizer 113, 134, 135, 321:compounding/compounds 110:10, 81ff., 159adjectival 82criteria 84diagnostics 84distributional evidence 84meaning 84negative adjectival 83one-word 81ff.orthography 84two-word 82ff.

concatenative morphology 49, 159concord 90conjugation 93, 159consonant clusters 27, 34initial 28medial 27ff.

consonant phonemes 15ff.consonants

co-occurrence restrictions 60homorganic 60

construction grammar 110, 112continuants 16contractions 83contrastive distribution 13coordinating conjunctions 120coordination 120ff.copula 159copula morpheme 136copula pronoun 136, 137copular constructions 112, 127corpus/corpora 3, 143, 143ff.CP (complementizer phrase) 113

dative 111, 113, 128, 132, 133alternation 132movement/shift 133prepositional 133

declension 89, 93, 97, 305:1decomposition

lexical 6morphological 46noun 46ff.predicate 6verb 46ff.

deep structure 108defective (root) 97defective noun (declension) 149, 154ff.definiteness 92, 94degree 159deixis 92, 95demonstrative pronouns (see pronoun,

demonstrative)derivation 90:8, 55derivational morphology 43, 55ff., 159de Saussure, Ferdinand 4desinence 47, 5:13desinential inflection 47, 97, 98, 108determiner 92, 113, 159determiner phrases 121diachronic analysis 4dialectology 6diglossia 141diptote 97, 149, 153ff.discontinuous morphology 49, 55, 159discourse 3distinctive feature 14, 55, 159ditransitive 59, 111, 132, 133

182 Index


double-object construction 132DP (determiner phrase) 113dual 102, 149, 151ff.

emphasis 19, 26spread (see also velarization) 19, 26

epenthesis 27, 145, 159equational sentence 127, 135, 136, 159etymology 57derivational 57

Experiencer (case role) 131, 159exponent 98, 99

Fillmore, Charles 128feature spread 26

flap 15form class 107, 159formal notation 29formalization 30, 56, 160formatives 49, 65, 65ff., 160Forms of the verb 56, 57formula-based verbs 83free morpheme (see morpheme, free)free variation 160fricative 14, 15function words 79fuşħaa 9fusional language type 45future tense 102

gemination 13, 73ff.gender 46, 90, 92, 93, 94, 98, 130generative grammar (see grammar, generative)genitive case (see case, genitive)genitive construct (see also iđaafa) 122glottal stop 17government (ʕamal) 43, 91, 110, 112grammar 2ff., 5, 91case 110, 112, 128definition 2descriptive/prescriptive 2, 4generative 5, 108, 109, 110ff., 127relational 5

grammatical categories 90grammatical word 89, 97, 98

ħadhf ‘deletion’ 28ff.

hamza (glottal stop) 17as derivational affix 66ff.as inflectional affix 100ff.

elidable (weak) 17, 28fixed (strong) 17, 66ff.

head 113, 119, 160human plural 130humanness 92

Ibn Jinni 8iđaafa 165:2ff , 82, 121, 122ff.indefinite marker 46indeclinable (nominal declension) 149, 155indefinite marker 46indicative 127, 135indirect object 111infix 49, 65, 93, 160immediate constituent (IC) 108INFL. 113inflection 43contextual 90, 91desinential 37, 47, 97, 98, 108inherent 90, 91

inflectional affixes (see affix, inflectional)inflectional categories 91ff.inflectional class 98inflectional morphology (see morphology,

inflectional)ʔinna and her sisters 134, 135Instrument (case role) 131, 160invariable noun (declension) 149, 155IPA (International Phonetic Alphabet) 147ʔiʕraab (desinential inflection) 91, 96ishtiqaaq 57

fiʕliyya 127ismiyya 127, 136Al-Khalil ibn Ahmad 29:9, 8,

23, 60Al-Kitaab 8Kitaab al-ʕayn 8, 60

laam as definiteness marker 101lateral 15lexeme 41, 45, 49, 55, 160lexical category 114, 161lexical–conceptual structure 129lexical expansion 159:2lexical root (see also root, lexical) 47, 112lexical–semantic template 129lexical semantics 129lexicalist hypothesis 43lexicography 8, 141lexicon 161

Index 183


linguisticsand grammar 2applied 3–4, 57, 141ff.Arabic 2, 6as science 1cognitive 3computational 3, 143ff.definitions 1descriptive/prescriptive 2history of 3sociolinguistics 3, 142ff.structural 4, 5, 108theoretical 3, 4ff., 5, 6

loan translation 82, 86loanwords 80Locative (case role) 131, 161

manner of articulation 14, 15, 161maşdar (see also verbal noun) 57matrix verbs 134, 135measure 57miim as affix 69ff., 101minimal pair 13modality 136, 161model root (f-ʕ-l) 56mood 98:17, 90, 92, 93, 136, 137, 161morpheme 49, 161morp 45, 45ff.bound 42, 48, 158definition 42, 49, 161discontinuous 55free 42, 49, 160

morphological modelsitem and arrangement 44, 160item and process 44, 160lexeme–morpheme base 45, 161lexical phonology and morphology 45, 161word and paradigm 44

morphological properties 91morphology 41ff., 109, 162analytical procedures 46autonomous 42, 43autosegmental 57boundaries 42, 45center, at the 42concatenative 42, 49, 159definition 49, 162derivational 6, 43, 49, 55, 159discontinuous 49, 159generative 42inflectional 49, 65, 89ff., 160

interfaces 44key role 44location 42non-concatenative 42, 50, 162prosodic 33split 44templatic 57terminology 48ff.

morphophonemics 23morphophonology 23, 30, 45, 50, 162morphosyntax 45, 50, 90ff., 162mubtadaʔ 127, 136muđaaf 82, 122muđaaf ilay-hi 122muŧaabaqa 91, 107, 108

naħt 83naħw 47, 107nasal 15negation 112nisba 72, 81noun decomposition 46ff.noun–adjective phrase 231:5noun phrase (NP) 114, 119ff.nouns 127

apposition 119coordination 120declensions 149ff.paradigm 50ff.

number 46, 91, 92, 93, 94, 98, 130nunation (tanwiin) 71, 101, 122nuun as affix 70ff., 101ff.

Object/Objective (case role) 110, 111, 114, 120,128, 131, 162

Obligatory Contour Principle (OCP) 60orthography 8, 56, 146

deep 146shallow 146

paradigm 50, 61, 89, 98, 99, 162paronym 56paronymy 56, 162participle 58, 65, 69, 95

active 58, 95passive 58, 95

Patient (case role) 111, 114, 128, 162pattern (morphological) 46, 47, 50, 56, 57, 99,

162penult 36, 37person 46, 90, 92, 93, 163

184 Index


philology 4comparative 4

phoneme 13, 163feature matrix 14ff.inventory 14

phonemic chart 15phonemic processes 17ff.gemination 17vowel lengthening 17

phonemics 13ff., 163phonetics 163acoustic 13articulatory 13

phonology 36:2, 163phonotactics 23, 163point of articulation 14, 15portmanteau morph 50, 163predicate 110predicate–argument structure 110, 112, 128predicate calculus 110predicate decomposition 112predication analysis 112prefix 50, 65, 93, 163prepositional phrases 89, 121, 127, 136pronouns 80, 95, 127copula 136, 137demonstrative 95, 121, 122personal 95relative 95suffix 47

prosodic morphology 33prosody 74

qalb ‘shift’ 29ff.qiyaas ‘analogy’ 56Qur’an 7

Recipient (case role) 111, 114, 131, 163register 142resonants 15retracted tongue root (RTR) 16role, semantic 60, 111romanization 145ff.root, lexical 46, 56, 60ff., 163biliteral 61ff.definition 50model 56as ordered set 60quadriliteral 57, 63, 83

acronymic 64borrowed 64

compound 64reduplicated 63sound 63

quinquiliteral 64triliteral 57, 61ff., 63

defective 63geminate 62hamzated 62hollow 63strong 62

root–pattern system 48, 55, 79

şarf 47and taşriif 47semantic shift 86semantics 109, 112semi-prepositions 121Semitic 44, 51, 56, 57, 65, 74semivowels 15Sibawayhi 8, 23siin as affix 102sociolinguistics 3, 142ff.sound feminine plural 149, 153ff.sound masculine plural 102, 103, 149, 152-st- as affix 71stem 51, 99, 163allomorphy 45, 51, 60, 93, 99, 164class 99solid 79ff.

stop 14, 15stress 33full-form pronunciation 34ff., 36ff.lexical 36ff.pause-form pronunciation 35, 37placement 33rules 36ff.shift 37

stricture 15subject 111subject–verb agreement 128subject–verb asymmetry (see SVAA)suffix 51, 65, 93, 164superlative 95suprasegmentals 36SVAA (subject–verb asymmetry) 114, 129, 130,

131syllable structure 33, 33ff.and stress 33coda 33constraints 34formalization 35

Index 185


syllable structure (cont.)margins 33nucleus 33onset 33types 34ff.

synchronic analysis 4syntactic categories 114, 164syntax 4, 107ff., 164autonomous 109centrality of 4, 109definition 107, 164generative 4ff., 5ff., 108, 108ff., 109, 130Government and Binding Theory 109, 128Minimalist Program 109, 111, 128Principles and Parameters Theory 109, 111Standard Theory 109technical terms 112ff.

synthetic language type 45

taaʔ as affix 67ff., 100ff.taaʔ marbuuŧa 19ff., 68tafxiim/tafkhiim “backing” 19templates 56, 57, 99tense 46, 90, 92, 93and aspect 93compound 94

terminologyinflectional morphology 97ff.morphological 48syntactic 112ff.

thematic roles (theta roles) 96, 127, 132, 164Theme 111, 114theta-marking 127compositional 128direct 127

theta-theory 164transcription 145, 145ff.transfix 93transliteration 145ff., 146ff., 164triptote 99, 150ff.

ultimate constituents 114universal grammar (UG) 110, 114, 137,

333:169:8, 17, 11:20, G11universals 5uvular 16

valence/valency 59, 110, 111, 112, 128, 164theory 110

variation theory 3velarization 19, 26

verbassociative 58causative 58, 132chart 122:19decomposition 46ff.derivation 56ditransitive 59, 132, 133estimative 58Forms (ʔawzaan) 56, 57, 132formula-based 83intensive 58intransitive 59reciprocal 58reflexive 6:9repeated action 58requestative 58resultative 58stem 58transitive 122:5

system 56templates 57, 59

verbal noun (maşdar) 57, 58, 69, 82, 123verbless predications 135ff.verbs 127

of permission or denial 132of perception and cognition 133of transformation 133

vernacular 3, 6, 142voice 46, 90, 92, 93voicing 13vowel/s 17ff.

chart 17deletion 23, 28, 29harmony 25helping 27insertion 27ff.length 17prosthesis 28shift 23, 29

VP (verb phrase) 114VS (Verb–Subject) order 112

waaw as affix 73ff., 103ff.WFR (word formation rule) 51, 55, 164WH- constructions 131WH- movement 131, 164WH-word 114word

boundaries 47ff.definition 41formation 42, 74

186 Index


grammatical 41, 97, 98order 6, 128, 129, 130structure 41

word-and-paradigm modelof morphology 44, 164

yaaʔ as affix 72ff., 102ff.yaaʔ of nisba 72, 80ff.

zawaaʔid (sing. zaaʔida) 65, 99(‘affix/augment’)

Index 187

arabic - a linguistic introduction

Documents

arabic morphology

linguistics of arabic

arabic phonotactics

arabic lexicon

arabic phonology

university of cambridge

arabic inectional morphology

arabic linguistics forover