You are on page 1of 498

TRANSLATING IDIOMATIC ENGLISH

PHRASAL VERBS INTO ARABIC

By

Ali Yunis Aldahesh

A thesis

presented to the

University of Western Sydney

In fulfillment of the requirements for the degree of

Doctor of Philosophy

February, 2008

© A Y Aldahesh February 2008


ii

DEDICATION
To my beloved kids: Jaafar, Ahmad and Ielaf
iii

ACKNOWLEDGEMENTS

I should like to express my sincere indebtedness to all those who, without their

generous assistance, this work would not exist. I should like in particular to thank my

chair supervisor Dr. Raymond Chakhachiro for his expert help, insightful comments,

and supportive encouragement, my co-supervisor Professor Stuart Campbell for his

invaluable feedback and his constructive criticism in various stages of completion of

this thesis, and my co-supervisor Dr Rosemary Suliman for going into a number of

issues of this study with me and bringing some great ideas to my attention.

I am profoundly grateful to my colleagues and my dear students who expressed their

interest and agreed to take part in the empirical part of the study. Their responses were

of optimal benefit and are deeply appreciated.

I also warmly thank my colleague Dr. Paul White for his proofreading of the thesis

and for the professional comments and corrections he made.

I owe special dept of gratitude to my colleague Dr. Abbas Brashi for many hours of

invaluable discussions about my plan and proposal from the very beginning of my

candidature.

Finally, I am particularly indebted to my lovely family for their support,

encouragement and patience throughout my long journey. They imparted me with the

most creative environment. Without them, it is true, I would not make it.
iv

STATEMENT OF AUTHENTICATION

The work presented in this thesis is, to the best of my knowledge and belief, original

except as acknowledged in the text. I herby declare that I have not submitted this

material, either in whole or in part, for a degree at this or any other institution.

………………………………………………..

Ali Yunis Aldahesh


v

TABLE OF CONTENTS

Page

LIST OF TABLES ............................................................................................... xiiiii

LIST OF FIGURES..................................................................................................xv

KEY TO TRANSLITERATION SYSTEM ............................................................ xvi

KEY TO ABBREVIATIONS................................................................................. xix

ABSTRACT............................................................................................................ xx

CHAPTER ONE: INTRODUCTION ........................................................................ 1

1.1 Historical background of PVs in English.......................................................... 2

1.2 Justification of the study .................................................................................. 7

1.3 Purpose of the study......................................................................................... 8

1.4 Layout of the study ........................................................................................ 11

1.5 Limitations of the study ................................................................................. 12

CHAPTER TWO: REVIEW OF RELEVANT LITERATURE................................ 14

2.1 Introduction ................................................................................................... 14

2.2 Idioms and idiomaticity in English and Arabic............................................... 14

2.2.1 Idioms and idiomaticity in English.......................................................... 15

2.2.2 Idioms and idiomaticity in Arabic ........................................................... 19

2.3 Definition of PVs in English .......................................................................... 23

2.4 Syntactic properties of EPVs.......................................................................... 27

2.4.1 Word class of EPVs ................................................................................ 27

2.4.1.1 Verbs ................................................................................................... 28

2.4.1.2 Particles ............................................................................................... 29


vi

2.4.1.2.1 Adverbs............................................................................................. 30

2.4.1.2.2 Prepositions....................................................................................... 33

2.4.2 Syntactic categories of EPVs................................................................... 37

2.4.2.1 Verb + adverb combinations................................................................. 39

2.4.2.1.1 Intransitive PVs................................................................................. 39

2.4.2.1.2 Transitive PVs................................................................................... 40

2.4.2.2 Verb + preposition combinations.......................................................... 41

2.4.2.2.1 Intransitive prepositional verbs.......................................................... 41

2.4.2.2.2 Transitive prepositional verbs............................................................ 41

2.4.2.3 Verb + adverb + preposition combinations ........................................... 42

2.4.2.3.1 Intransitive phrasal-prepositional verbs ............................................. 43

2.4.2.3.2 Transitive phrasal-prepositional verbs ............................................... 43

2.4.3 Word order of EPVs................................................................................ 44

2.5 Semantic properties of EPVs.......................................................................... 49

2.5.1 EPVs' criteria .......................................................................................... 50

2.5.2 Semantic categories of EPVs................................................................... 53

2.5.3 Cohesion of PVs' elements ...................................................................... 62

2.5.4 British and American dialects ................................................................. 63

2.5.5 Register variations of EPVs .................................................................... 65

2.6 PVs in Arabic ................................................................................................ 70

2.6.1 Lentzner.................................................................................................. 71

2.6.2 Abboud and McCarus ............................................................................. 78

2.6.3 Heliel...................................................................................................... 79

2.6.4 Alkhuli.................................................................................................... 82

2.6.5 Najiib...................................................................................................... 82
vii

2.6.6 Siinii, Hussein and Al-ddoush ................................................................. 82

2.6.7 Summary of section 2.6........................................................................... 86

2.7 Derivation of PVs in English and Arabic ....................................................... 90

2.7.1 Derivation in English .............................................................................. 91

2.7.2 Derivation in Arabic................................................................................ 94

2.7.3 Derivation of PVs in English................................................................... 99

2.7.3.1 Deriving PVs from adjectives..............................................................100

2.7.3.2 Deriving PVs from nouns....................................................................100

2.7.3.3 Deriving PVs from Latinate verbs .......................................................101

2.7.3.4 Deriving nouns from PVs....................................................................102

2.7.4 Productivity ...........................................................................................104

2.7.4.1 Productivity in English........................................................................105

2.7.4.2 Productivity in Arabic .........................................................................107

2.8 PVs in English lexicography .........................................................................110

2.8.1 PVs in general English-English dictionaries ...........................................111

2.8.2 PVs in specialized English-English dictionaries .....................................113

2.8.2.1 An Analysis and Dictionary of Two-Word Verbs in English

Illustrated in Sentences ...................................................................................113

2.8.2.2 Dictionary of English Phrasal Verbs and their Idioms..........................115

2.8.2.3 Time-Chambers Dictionary of Phrasal Verbs ......................................119

2.8.2.4 Longman Dictionary of Phrasal Verbs.................................................122

2.8.2.5 The Student's Dictionary of Phrasal Verbs...........................................125

2.8.2.6 Collins COBUILD Dictionary of Phrasal Verbs ..................................127

2.8.2.7 NTC's Dictionary of Phrasal Verbs and Other Idiomatic Verbal Phrases

.......................................................................................................................131
viii

2.8.2.8 Oxford Dictionary of phrasal verbs .....................................................133

2.8.2.9 Cambridge International Dictionary of Phrasal Verbs..........................137

2.8.2.10 Oxford Phrasal Verbs Dictionary for Learners of English..................141

2.8.2.11 Summary of section 2.8.2..................................................................143

2.9 PVs in bilingual English-Arabic dictionaries.................................................147

2.9.1 PVs in general English-Arabic dictionaries ............................................148

2.9.1.1 Al-Mawrid ..........................................................................................148

2.9.1.2 Al-Mughni Al-Akbar ..........................................................................149

2.9.1.3 The Oxford English-Arabic Dictionary of Current Usage ....................150

2.9.2 PVs in specialized English-Arabic dictionaries.......................................151

2.9.2.1 English Phrasal Verbs in Arabic..........................................................151

2.9.2.2 York Dictionary of Phrasal Verbs and Their Idioms ............................154

2.9.2.3 Summary of section 2.9.......................................................................157

2.10 Teaching IEPVs..........................................................................................158

2.10.1 Teaching IPVs to learners of English ...................................................158

2.10.1.1 Tom McArthur's method ...................................................................159

1.10.1.2 Colin Mortimer's method...................................................................161

2.10.1.3 Alan Cornell's method.......................................................................162

2.10.1.4 Richard Side's method.......................................................................165

1.10.1.5 Peter Dainty's method........................................................................168

2.10.1.6 Martin Shovel's method.....................................................................169

2.10.1.7 Malcolm Goodale's method...............................................................171

2.10.1.8 Berman and Kirstein's method...........................................................172

2.10.1.9 Peter Hannan's method ......................................................................173

2.10.1.10 Darwin and Gray's method ..............................................................176


ix

2.10.1.11 Joan Sawyer's method .....................................................................180

2.10.1.12 Ron Sheen's method ........................................................................183

2.10.1.13 Rosemary Sansome's Method ..........................................................185

2.10.1.14 John Flower's method......................................................................186

2.10.1.15 Summary of section 2.10.1 ..............................................................187

2.10.2 Teaching IEPVs to Arab students .........................................................189

2.11 PVs in translation studies ............................................................................196

2.11.1 Translating EPVs into other languages .................................................196

2.11.2 Translating English idioms into other languages...................................201

2.12 Conclusion..................................................................................................209

CHAPTER THREE: LINGUISTIC CONTRASTIVE ANALYSIS.........................211

3.1 Introduction ..................................................................................................211

3.2 Contrastive analysis ......................................................................................211

3.3 Contrastive analysis and translation ..............................................................212

3.4 Review of contrastive analysis approaches....................................................215

3.4.1 Juliane House.........................................................................................215

3.4.2 Carl James .............................................................................................224

3.4.2.1 Microlinguistics ..................................................................................224

3.4.2.2 Macrolinguistics..................................................................................225

3.4.3 Equivalence ...........................................................................................229

3.4.4 Theory of sense (Interpretive Approach to Translation)..........................233

3.4.5 Theory of speech acts.............................................................................236

3.4.6 Communicative Competence..................................................................241


x

3.5 A linguistic contrastive analysis model for the analysis and translation of IEPVs

into Arabic .........................................................................................................244

3.6 Conclusion....................................................................................................249

CHAPTER FOUR: RESEARCH METHODOLOGY .............................................251

4.1 Introduction ..................................................................................................251

4.2 Research design ............................................................................................251

4.2.1 Research questions.................................................................................251

4.3 Participants...................................................................................................253

4.3.1 Arabic professional translators ...............................................................253

4.3.2 Arabic translation students .....................................................................255

4. 4 Data collection.............................................................................................256

4.4.1 Instrument..............................................................................................256

4.4.2 Data collection procedures .....................................................................259

4.5 Data analysis ................................................................................................261

CHAPTER FIVE: RESULTS AND DISCUSSIONS OF THE TRANSLATION

TESTS ...................................................................................................................265

5.1 Introduction ..................................................................................................265

5.2 General overview of the results of the translation tests ..................................265

5.2.1 General overview of the results of the Arabic professional translators' group

...........................................................................................................................266

5.2.2 General overview of the results of the Arabic translation students' group ...269

5.2.3 Summary and comparison ......................................................................270

5.3 Subjects' performance in light of the research questions................................274


xi

5.3.1 Subjects' performance in light of the first research question....................275

5.3.1.1 Summary and comparison ...................................................................278

5.3.2 Subjects' performance in light of the second research question ...............280

5.3.3 Subjects' performance in light of the third research question ..................283

5.3.4 Subjects' performance in light of the fourth research question ................285

5.3.5 Subjects' performance in light of the fifth research question ...................288

5.3.6 Subjects' performance in light of the sixth research question ..................291

5.3.6.1 Overtly erroneous errors......................................................................292

5.3.6.1.1 Literal translation .............................................................................293

5.3.6.1.2 Mistranslating ..................................................................................296

5.3.6.1.3 Reducing the idioms to sense ...........................................................298

5.3.6.1.3 Breaching of the Arabic language system .........................................300

5.3.6.2 Covertly erroneous errors ....................................................................302

5.3.6.2.1 Wrong Arabic collocation ................................................................302

5.3.6.2.2 Shift of register ................................................................................305

5.3.6.2.3 Incorrect delivery of speech acts.......................................................308

5.3.6.2.4 Usage of paraphrasing ......................................................................311

5.3.6.2.5 Usage of Arabic colloquial and regional dialects ..............................313

5.3.7 Subjects' performance in light of the seventh research question ..............315

5.4 Conclusion................................................................................................324

CHAPTER SIX: CONCLUSION AND RECOMMENDATIONS..........................325

6.1 Conclusion....................................................................................................325

6.1.1 Summary of the study ............................................................................325

6.1.2 Summary of the findings ........................................................................331


xii

6.2 Recommendations ........................................................................................341

6.2.1 Recommendations for Arabic professional translators ............................341

6.2.2 Recommendations for Arabic lexicographers .........................................343

6.2.3 Recommendations for Arabic pedagogues..............................................346

6.3 Contributions made by this study ..................................................................350

6.4 Directions for further research ......................................................................351

BIBLIOGRAPHY ..................................................................................................353

APPENDICES .......................................................................................................368

APPENDIX 1:TRANSLATION TESTS ................................................................369

APPENDIX 2: SUGGESTED ARABIC FUNCTIONAL-PRAGMATIC

EQUIVALENTS ....................................................................................................393

APPENDIX 3: FULL DETAILS OF THE RESULTS OF THE TRANSLATION

TESTS ...................................................................................................................413

APPENDIX 4: AVERAGE OF THE STANDARD DEVAITION ..........................468

APPENDIX 5: RANKING THE ITEMS ACCORDING TO THEIR PERCENTAGES

OF THE CORRECT IN A DESENDING ORDER.................................................473


xiii

LIST OF TABLES

Table Page

2.1 Verb + particle combinations ………………………………………..……..……38

2.2 Types and subtypes of verb + particle combinations……….……………………44

2.3 The word order of EPVs ……………………………....……...………………… 46

2.4 Semantic categories of EPVs ………………………..………….……………… 55

4.1 Years of experience of the Arabic professional translators ………...…………. 254

4.2 Text types used in the translation tests and the items representing each type…. 257

4.3 Categories/subcategories covered in the translation tests and the items

representing each category ………...……………………………….……………... 259

4.4 Examples of satisfactory answers ………………………………………….…...262

4.5 Examples of mistakes in the surrounding words …………………………..…...263

5.1 Categories of IEPVs covered in the translation tests ………………………......266

5.2 Average of the correct answers of the Arabic professional translators' group .. 267

5.3 Standard deviation of the Arabic professional translators' group ..……………267

5.4 Ranking the categories and their subcategories-Arabic professional translators'

group ………………………………………………………………….………….…268

5.5 Average of the correct answers of the Arabic translation students' group ...…...269

5.6 Standard deviation of the Arabic translation students' group ………….….……269

5.7 Ranking the categories and their subcategories – Arabic translation students'

group…………………………………………………………..…………….………270

5.8 Differences of the averages of the correct answers ……….……………………271

5.9 Differences of the standard deviation ……………………….…………….……274

5.10 Summary of the performance of the Arabic professional translators' group..…275


xiv

5.11 Summary of results of the Arabic professional translators' group ….……...…276

5.12 Summary of the performance of the Arabic translation students' group .……..277

5.13 Summary of results of the Arabic translation students' group …….……...…..277

5.14 Differences in achieving functional-pragmatic equivalents …….….…….…...278

5.15 Differences of the average of the unanswered items ………...…………….….278

5.16 Differences in achieving functional-pragmatic equivalents of the Complex

idioms category ……….….…………….……….……..……….….………….……281

5.17 Differences in achieving functional-pragmatic equivalents of the Syntactic forms

(s) category ……….………………….………....………………….….……………284

5.18 Differences of achieving functional-pragmatic equivalents of the Derivational

forms category …….………....…….………....……….……….....……….………..287

5.19 Differences in achieving functional-pragmatic equivalents of the Productive PVs

category …….………....……….….………....……….……..………....…………...289

5.20 Examples of literal translations ….………....…….………....….……….....….293

5.21 Examples of mistranslations ….………....…….……….……....…….….…….297

5.22 Examples of idioms reduced to sense .………...………....……..…….....……299

5.23 Examples of breaches to the Arabic language system ……...………………...301

5.24 Examples of wrong Arabic collocations ……...………………...………...…..304

5.25 Examples of register shift ……...…………..……………....……….........……306

5.26 Examples of incorrect delivery of speech acts ……...…...…....……….…...…309

5.27 Examples of Paraphrasing …...…..………...…..………...…..……...…...……311

5.28 Examples of usage of Arabic colloquial and regional dialects ...…......………313

5.29 Performance and years of experience of the Arabic professional translators'

group...….…..…..…..…..…..….…..…..…………………………...…..………...…321

5.30 Performance and stages of study of the Arabic translation students' group ..... 322
xv

LIST OF FIGURES

3.1 A model for the analysis and translation of IEPVs into Arabic ………………….247
xvi

KEY TO TRANSLITERATION SYSTEM

Elements of the transliteration system of Arabic used in this study are eclectically

selected from systems adopted by Lentzner (1977); Beaugrande et al. (1994); Hatim

(1997); Chakhachiro (1997) and Campbell (1998).

Arabic consonants

Arabic letters Arabic letters name Transliteration

‫ء‬ hamza ?

‫أ‬ ?alif a

‫ب‬ baa? b

‫ت‬ taa? t

‫ث‬ thaa? th

‫ج‬ jiim j

‫ح‬ Haa? H

‫خ‬ khaa? kh

‫د‬ daal d

‫ذ‬ dhal dh

‫ر‬ raa? r

‫ز‬ zaay z

‫س‬ siin s

‫ش‬ shiin sh

‫ص‬ Saad S

‫ض‬ Daad D
xvii

‫ط‬ Taa? T

‫ظ‬ Zaa? Z

‫ع‬ 3ayn 3

‫غ‬ ghayn gh

‫ف‬ faa? f

‫ق‬ qaaf q

‫ك‬ kaaf k

‫ل‬ laam l

‫م‬ miim m

‫ن‬ nuun n

haa? h

‫و‬ waaw w

‫ي‬ yaa? y

Arabic short vowels

َ fatHa a

ِ kasra i

, Damma u

Arabic long vowels

‫َا‬ aa

‫و‬ uu

‫ي‬ ii
xviii

Arabic diphthongs

‫أي‬ ay

‫أو‬ aw

Other features

The Arabic definite article with sun letters is not elided.

The Arabic geminated consonants (with .ّ.. shadda) are doubled.

The coordination particles (‫ و‬wa and ‫ ف‬fa) are marked.

This system may not conform to some Arabic names occuring in this study, which

have their own previously establiehed transiliterations. For example: %&&'()*(+‫ ا‬,&-./

Munir Ba'albaki and 0-123)+ 4)*+‫ دار ا‬Dar El Ilm Lilmalayin.


xix

KEY TO ABBREVIATIONS

A.H = After Hijrah (Islamic Calander)

APV = Arabic Phrasal Verb

c = complex idioms

d = derivational forms of idiomatic English phrasal verbs

de = phrasal verbs derived from Adjectives without suffix –en

den = phrasal verbs derived from Adjectives with suffix –en

dnp = nouns derived from phrasal verbs

dpn = phrasal verbs derived from nouns

EFL = English as a foreign language

EPV = English Phrasal Verb

ESL = English as a second language

IAPV = Idiomatic Arabic Phrasal Verb

IEPV = Idiomatic English Phrasal Verb

L1 = First Language

L2 =Second Language

p = productive, newly coined idiomatic English phrasal verbs

PV = Phrasal Verb

s = syntactic forms of idiomatic English phrasal verbs

sva = syntactic form of verb + adverb

svap = syntactic form of verb + adverb + preposition

svp = syntactic form of verb + preposition


xx

ABSTRACT

This study concerns itself with a linguistic contrastive analysis of one particular

characteristic of grammar and vocabulary in both English and Arabic languages. That

is, combinations of proper verbs with adverbial and/or prepositional particle(s), which

are commonly known as: phrasal verbs (5(‫آ‬,3+‫ل ا‬7*89‫ )ا‬.

In addition, the thesis attempts to approve the fact that there are fundamental

similarities and dissimilarities between English and Arabic phrasal verbs. Such

similarities and dissimilarities are investigated at length through a theoretical

comparison of these combinations in both languages.

The main hypothesis of the study is that there are wide ranges of difficulties posed to

Arabic professional translators and Arabic translation students when translating

idiomatic English phrasal verbs into Arabic.

Exploring such difficulties is the principle aim of the study. The other aim is to

propose a number of recommendations for professional translators, lexicographers

and pedagogues. Such recommendations are based on a range of findings arrived at

from the empirical research carried out in the study.

A linguistic contrastive analysis model for the analysis and translation of idiomatic

English phrasal verbs into Arabic is devised in order to establish a yardstick by which

the translation quality of such idiomatic expressions can be analysed, compared,

evaluated and assessed.


xxi

Translation tests are conducted to identify types of errors and translational pitfalls

made by Arabic professional translators and Arabic translation students when

handling the most problematic and challenging idiomatic English phrasal verbs.

The results revealed that there are in fact lots of difficulties encountered by Arabic

professional translators and Arabic translation students when dealing with the

phenomenon of idiomatic English phrasal verbs. The most important aspect of such

difficulties is the failure to achieve functional-pragmatic equivalents of such verbs.

Overtly erroneous errors such as: literal translation, mistranslating, reducing idioms to

sense and breaching of the Arabic language system, along with covertly erroneous

errors, such as wrong Arabic collocation, shift of register, incorrect delivery of speech

acts, usage of paraphrasing and usage of Arabic colloquial and regional dialects, were

the major reasons behind the failure of delivering the appropriate functional-

pragmatic equivalents of the idiomatic English phrasal verbs listed in the translation

tests.
1

CHAPTER ONE
INTRODUCTION

Owing to their syntactic structures, idiomatic and non-idiomatic use and their complex

idioms, phrasal verbs (henceforth PVs) constitute one of the most difficult problems for

learners of the English language (Cowie & Mackin, 1993; Courtney, 1983; Taha, 1972;

Azzaro, 1992). They are, to use McArthur's (1975) words "a foreign learner's biggest

headache" (p. 6). At first glance they look deceptively easy to the non-native speaker,

but their significations can be fundamentally different from what one might expect

(Khalaili, 1979). English grammarians and linguists point out that to be fluent in

English you have to master using PVs ably. The type of PVs posing the learners with

particular difficulty are idiomatic English phrasal verbs (henceforth IEPVs) (Turton &

Manser, 1985).

In their Dictionary of English Phrasal Verbs and their Idioms, McArthur and Atkins

(1974) attempt to confine the problem of IEPVs in four points as follows:

a) A verb of this type may have a meaning which is simply the sum of its parts,
but may also have a meaning which bears little apparent relation to those parts.
b) The particle may indicate some kind of direction but may just as easily have a
meaning little related to direction.
c) The same particle can serve as a preposition or adverb and a student can easily
confuse these functions.
d) There are so many phrasal verbs in modern English (and the number is
constantly growing), they are so important in the spoken language, and they
have so many shades of meanings that the student may despair of ever
mastering this area of language. This situation has been aggravated by the lack
of a good description of the phrasal verb, and by a shortage of useful teaching
material (p. 5).
2

Yet, when it comes to translating IEPVs into Arabic, the problem far exceeds the four

points mentioned above due to the fact that there are tremendous dissimilarities in terms

of syntactic, semantic and pragmatic peculiarities of such combinations in English and

Arabic.

The present study is an investigation of the difficulties Arabic professional translators

and Arabic translation students encounter when translating IEPVs into Arabic.

1.1 Historical background of PVs in English

The difficulties of English PVs (henceforth EPVs) were first noted by Samuel Johnson in

1755 in the preface to his Dictionary of English Language (McArthur, 1989) in which he

wrote:

There is another kind of composition more frequent in our language than


perhaps in any other, from which arises to foreigners the greatest difficulty
[…] We modify the signification of many words by a particle subjoined; as to
come off, to escape by a fetch; to fall on, to attack; to fall off, to apostatize […]
with innumerable expressions of the same kind, of which some appear widely
irregular, being so far distant from the sense of the simple words, that no
sagacity will be able to trace the steps by which they arrived at the present use
(as cited in McArthur, 1989, p. 38).

What Johnson said about EPVs, McArthur (1989) indicates, "is still true, except that

nowadays they stand out more because there are even more of them about in the 20th than

in the 18th century" (p. 38), but Johnson had no name for the phenomenon, McArthur

elaborates, as there was no agreed academic name for it. Furthermore, McArthur (1989)

makes the point that by the 'composition' Johnson meant 'compound' which is widely used

nowadays in describing the co-occurrence of verbs with particles. Such a co-occurrence,


3

however, has been given many names by grammarians, linguists, pedagogues and

lexicographers.

In his essay English Idioms, Logan Pearsall Smith has first used the name of ‘Phrasal

verbs’ in print in 1925, following a suggestion from Henry Bradley (Dixon, 1982; Sroka,

1972). The same term has been employed by Jowett (1951), Mitchell (1958), Fairclough

(1965, as cited in Lindner, 1983), Heaton (1965), Bolinger (1971), Quirk and Greenbaum

(1973), Palmer (1974), Khalaila (1976), Turton and Manster (1985), Cornell (1985),

Thomas and Martinet (1986), McArthur (1989), Side (1990), Constant (1991), Azzaro

(1992), Dixon (1982), Cowie (1993), Heleil (1994), Holes (1994), Jacob (1995), Berman

and Kirstein (1996) Lindstormberg (1998), Sinclair et al. (1998), Darwin and Gray

(1999), Sawyer (2000), Sansome (2000) and Trush (2001) among others (cf. Lindner,

1983; McArthur, 1989).

Other names given to label the co-occurrence of verb with particle are 'Particle verbs'

(Crutchley, 2007), 'Verb particle constructions' (Morgan, 1997; Zughoul, 1979;

Lindner, 1983), 'Verb-adverb combination' (Kennedy, 1967; Fraser, 1976), 'Two-word

verbs' (Taha, 1960), 'Discontinuous verbs' (Live, 1965), 'Verb-preposition constructions'

(Aarts, 1989) and 'compound verbs' (Palmer, 1974) (cf. Dixon, 1982). But, Spears

(1993) considers the term 'phrasal verb' as a generic term and indicates that "verb +

particle collocation is more accurate" (p. vii).

Yet, the winning term is said to be the first one, that is, 'phrasal verb' which recently

quite commonly appears in English literature and linguistics works (McArthur, 1989;

Sroka, 1972).
4

Therefore, the term 'phrasal verb' being adopted in this study to refer to combinations of

verbs and particles owing to the fact that it has been widely used by the majority of

scholars in such domains as grammar, linguistics, lexicography, bibliography and

pedagogy.

Interestingly, 1965 was, according to Fraser (1974), a big year for EPVs, as four major

studies were published in that particular year by Fairclough, Fraser, Live, and Sroka.

The previous scholarly research into EPVs has been categorized by Azzaro (1992) into

three types depending upon the ways in which researchers approached the topic, being:

1. Some authors produced verb studies in which PVs were viewed as part of the
general English verbal system, without being given particular attention (cf. for
example Allen 1966; Kiparsky 1968: 30-37; Leech 1971; and Nehls 1978; 45-
62; etc.).
2. Others concentrated on the constituents of PVs without studying their
contextual behaviour: cf. for instance: Hills 1968 on the definition of EPVs'
constituent particles; Heaton 1969 and Wood 1975 on prepositions and their
idioms; Greenbaum 1970 on adjuncts; Bennet 1975 on spatial and temporal
uses of prepositions; and finally Fraser 1970 on idioms in a T-GG framework.
3. Other authors, on the other hand, tackled the study of EPVs proper. Kennedy
(1920) listed 900 verbal combinations, indicating their most productive
elements; Mitchell (1958) produced one of the cornerstones of all modern
studies on PVs […], together with Bolinger (1971); Fraser (1976) offers ample
details on the semantic and syntactic aspects of EPVs […]. The syntax of
EPVs is exhaustively discussed in Sroka (1972), Vestergaard (1977), De
Armond (1977), Dixon (1982) and also in Mathews (1984) and Radford
(1988) (pp. 40-41).

It is useful to mention here that in a paper entitled Phrasal verbs in English grammar

books before 1800, Hiltunen (1983) investigates "the treatment of combinations

involving a verb and adverbial or prepositional element in a selection of 16th, 17th, and

18th century grammars" (p. 376). He primarily concentrates on "finding out how the

idea of taking phrasal verbs as single units developed in the early treatises, rather than
5

on attempting to analyse the contents of adverbial and prepositional categories in detail,

or to evaluate the issues in terms of what we today know about phrasal verbs" (p. 377).

Hiltunen (1983) indicates that the "terms 'phrasal verb' and 'prepositional verb' are

comparatively recent and were not used by the early grammarians, but the history of these

constructions in English may be traced to the earliest periods of the language" (p. 376).

He elaborates that "Old English was still fairly close to its Germanic ancestors with its

'separable prefixes', words that appeared joined to the verb or detached from it, depending

primarily on various syntactic factors" (p. 376). He goes on to say that "[…] even in the

Old English period a tendency towards the establishment of the structural variant where

the phrasal element follows the verb, may be clearly discerned. In the early Middle

English period this pattern is already the predominant one" (p. 376). And by "the

beginning of the nineteenth century […] English grammarians had a fair conception of

many of the properties of the phrasal verbs, as well as of problems confronting them in a

structural description of these constructions" (p. 377).

Furthermore, Hiltunen surveys the grammar of New English starting from Bullokar

(1586) who "distinguished adverbs and prepositions, recognizing the fact that some

prepositions may be used adverbially" (p. 377), through Wallis (1652) and Poole (1655)

who accidentally noted "some such constructions in translating Latin prefixed verbs into

English" (p. 378), to Maittaire (1712) who surprisingly commented "on the

morphological, semantic, and also syntactic properties of the various constructions

involving prepositions and adverbs in greater detail than any of the other writers" (p.

382). Hiltunen concludes that

[…] in spite of the emphasis of traditional grammar on a word-for-word


analysis, grammarians gradually become aware of the possibility of looking
6

upon phrasal verbs as single units. This process begins subconsciously, phrasal
verbs creeping into the texts as translations of Latin examples or as descriptive
characterizations of the meaning of various 'prepositions'. Some of the
semantically more striking combinations prompt a comment (often in a
footnote) from the writer, and in this way the phrasal verb gradually finds its
way into English grammar (p. 384).

In addition, Hiltunen provides a number of reasons "why the grammatical recognition of

the phrasal verb was slow in coming", claiming that the grammar itself was the main

reason "because seeing the combination of a verb and a 'particle' as a unit involved

transcending the traditional boundaries between the parts of speech, and realizing that

words from different categories may melt into one another and form a new unit together"

(p. 384).

The other reason, according to him, was the effect of "Latin grammar, together with a

normative attitude towards language [...] All that did not directly fit into the Latin model

was often felt to be inferior or incorrect, something that ought to be resisted both in

theory and practice" (p. 384). "But indirectly" Hiltunen elaborates "the Latin background

also facilitated the acceptance of phrasal verbs as units by providing a point of

comparison for the English grammarians, and constantly reminding them how their own

language differed from Latin in this respect" (p. 385).

Along these lines, McArthur (1989) maintains that EPVs "have been widespread since at

least the Middle Ages" (p. 39). He provides the following two "typical antique usages

with go":

(1) in Wyclif's Bible of 1388, 'Thei that gon down in to the see in schippes', and
(2) in Shakespeare's Measure for Measure of 1603, 'So long, that nineteene
Zodiacks have gone round' (p. 39).

Moving on to our modern time, the phenomenon of PVs has been the focus of many

scholarly studies. Works in the last decade or so include: Flower (2000); Sansome
7

(2000); Sawyer (2000); Heliel (2000); Kaminska (2001); Zeller (2001); Jackendoff

(2002); Campoy (2002); Heine (2002); Villavicencio & Copestake (2003); Armstrong

(2004); Dehe (2005), Rottet (2005), Crutchley (2007) and Gardner & Davies (2007).

1.2 Justification of the study

In her article Plain English? A study of plain English vocabulary and international

audiences, Thrush (2001) narrates that "[i]n 1989, a China Airline flight, flying in zero

visibility, crashed into the side of a mountain shortly after takeoff. On the voice recorder,

the last words of the Chinese pilot to the co-pilot were "What does pull up mean?" (p.

289). This tragedy highlights the fact that the lack of a comprehension of IEPVs may lead

to catastrophes. Likewise, Palmer (1968) reports "the famous story of the foreigner in the

train who was told to look out. Instead of realizing that LOOK OUT was a phrasal verb

meaning TAKE CARE, he took it as a literal combination of LOOK+OUT and put his

head further out of the window with disastrous consequences" (p. 185) [Emphasis in

original].

The significance of IEPVs and the difficulty they pose to learners of English, translation

students, professional translators, professional interpreters and others have been

underlined by a great deal of researchers. Despite their important role in understanding

spoken and written English, they have not received enough attention from Arab

researchers. There has been no comprehensive research to date shedding light on how

IEPVs can be translated into Arabic or what sort of difficulties are encountered by Arabic

translators when handling them. Therefore, this study attempts to fill this gap by

investigating IEPVs, analyzing them, contrasting them with their Arabic counterparts,
8

highlighting the difficulties they constitute when translating them into Arabic and

suggesting a number of recommendations to be taken into account by professional

translators, lexicographers and translation teachers.

1.3 Purpose of the study

The principal aim of this study is to explore the difficulties encountered by Arabic

professional translators and Arabic translation students when translating IEPVs into

Arabic. The other objective is to propose a number of recommendations for Arabic

translators, Arabic lexicographers and Arabic pedagogues based on a range of findings

arrived at from the empirical research conducted in the study. More specifically, this

piece of research is trying to achieve the following purposes:

1. To explore the syntactic, semantic and pragmatic properties of the phenomenon of

PVs in both English and Arabic languages.

2. To investigate the similarities and dissimilarities of such a phenomenon between

both languages in the framework of linguistic contrastive analysis.

3. To device a workable eclectic linguistic model as a disciplined approach for the

analysis and translation of IEPVs into Arabic.

4. To look into difficulties of translating IEPVs into Arabic by investigating the

translational pitfalls made by Arabic professional translators and Arabic translation

students when translating IEPVs into Arabic. Such an investigation will be


9

conducted on the empirical data collected from both groups of the subjects by the

means of translation tests.

5. To suggest a number of practical, constructive and theoretically based

recommendations for Arabic translators, Arabic lexicographers and Arabic

pedagogues.

The main hypothesis of the study, however, is that there is a wide range of difficulties

posed to Arabic professional translators and Arabic translation students when

translating IEPVs into Arabic.

In order to achieve the set up goals of the study and to examine the established

hypothesis, two sets of research questions has been put forward. The first set was

intended to address the theoretical part of the thesis. The second set of research questions,

however, was meant to analyse the empirical data from pragmatic, semantic, syntactic

and stylistic perspectives. The research questions of the theoretical part of the study

being:

1) What is the notion of idioms and idiomaticity in both English and Arabic?

2) What are the syntactic, semantic, and pragmatic criteria of IEPVs?

3) Are there structures in Arabic similar to that of the IEPVs?


4) What are the similarities and dissimilarities between idiomatic PVs in English

and Arabic?

5) What are the difficulties of learning IEPVs by non-native speakers of English in

general and by Arab learners of English in particular?


10

6) To what extent have English and Arabic lexicographers been successful in

covering, explaining and providing the appropriate equivalents for IEPVs in

their general and specialized dictionaries? And what are the gaps that need to be

closed in this respect?

7) To what extent the methods and materials employed by English and Arabic

pedagogues can help in solving the learning, teaching and translating difficulties

of IEPVs? And what are the gaps that need to be bridged in this difficult area of

pedagogy?

8) How has the phenomenon of IEPVs been dealt with in translation studies?

The research questions of the empirical part of the study being:

1) To what extent Arabic professional translators and Arabic translation students

were successful in providing the Arabic functional-pragmatic equivalents of

IEPVs?

2) To what extent the issue of complex idioms poses a difficulty for Arabic

professional translators and Arabic translation students when translating

IEPVs into Arabic?

3) Do syntactic forms of IEPVs pose difficulty to Arabic professional translators

and Arabic translation students when dealing with IEPVs?

4) To what extent the derivational forms of IEPVs affect the process of

translating them into Arabic?

5) It is well known that English is very productive in coining IEPVs. Many of

them pop up on a regular basis. Such newly coined PVs are occasionally

emerging in both written and spoken modes of the English language where there
11

are no readily Arabic equivalents for them. Hence, how did the Arabic subjects

deal with such a phenomenon?

6) What are the types of translational errors made by Arabic professional

translators and Arabic translation students?

7) Is there any difference as to translation communicative competence between

the two groups of Arabic subjects: The professional Arabic translators and

Arabic translation students? If so, what are the major areas in which their

competence varies?

1.4 Layout of the study

The framework of this thesis comprises six Chapters as follows:

In Chapter One a brief historical background of the notion of EPVs in previous studies is

provided. Then the study's justification, purpose, layout, and limitations are outlined.

Chapter Two comprehensively reviews the relevant literature of the concept of PVs and

its peculiarities. It begins with exploring the vital notion of idioms and idiomaticity in

both English and Arabic languages. It goes on to address the key issues of the study,

they are: the definition of PVs in English, the syntactic and semantic properties of PVs

in English, PVs in Arabic, derivation of PVs, PVs in English lexicography, PVs in

bilingual English-Arabic dictionaries, teaching PVs, and PVs in translation studies.

Chapter Three constitutes a link between the theoretical part of the study (the literature

review) and the practical part of it (the experimental research). It establishes a theoretical
12

rationale for the empirical research by reviewing main and influential approaches of

contrastive analysis to see their appropriateness/inappropriateness to the topic at hand in

an attempt to formulate a practical and scientific model for the analysis and translation of

IEPVs into Arabic.

Chapter Four presents the methodology utilized in this study to conduct the

experimental research. It elaborates on the issues of research design, participants, data

collection and data analysis.

Chapter Five comprises two distinct sections. The first section outlines a general

overview of the results of the translation tests conducted by the two groups involved in

the present study (Arabic professional translators and Arabic translation students). The

second section concerns itself with analysing and discussing the subjects' performance in

light of the second set of research questions.

Chapter Six concludes by summarizing the findings of the study and proposing

recommendations for Arabic translators, Arabic lexicographers and Arabic pedagogues. It

also draws on the contributions made in the present study and indicates a number of

directions for further research.

1.5 Limitations of the study

This study will only concentrate on idiomatic EPVs and the difficulties they pose to

Arabic professional translators and Arabic translation students as a result of their

complex idiomatic meanings and complicated functional-pragmatic use. It will not,


13

however, cover non-idiomatic type of EPVs, owing to the fact that they have

straightforward meanings, and do not cause Arabic professional translators and Arabic

translation students any difficulty in rendering them into Arabic. More regarding the

differentiating between idiomatic and non-idiomatic types of PVs is given in detail in

the forthcoming Chapter, but for the purpose of immediate clarification the following

examples illustrate the excluded non-idiomatic type of PVs where both components of

the construction (verb and particle) keep their individual lexical meanings:

Agree with, alert to, arrange for, arrive in, and assist with.

In addition, despite the fact that PVs are generally used in spoken English more than in

written English, the study will exclude the employment of colloquial Arabic as

functional-pragmatic equivalents since it concerns itself with translation not interpreting

of IEPVs. However, the study will have a considerable impact on interpreting for the

reason that translation and interpreting are two sides of one coin. I also assume that it will

be of benefit to translation teachers, learners of English language and Arabic-English

lexicographers.

Lastly, given that the study is devoted to investigate the difficulties encountered by

Arabic professional translators and Arabic translation students when translating IEPVs, it

will not cover the issue of recognition of such idiomatic expressions. It is assumed that

professional translators and translation students, the subjects of this study, must have the

required English proficiency level that allows them to recognize IEPVs.


14

CHAPTER TWO
REVIEW OF RELEVANT LITERATURE

2.1 Introduction

In this Chapter the relevant literature will be comprehensively reviewed in an attempt

to address the first set of research questions. The crucial notion of idioms and

idiomaticity in both English and Arabic languages will be taken up as a point of

departure from which I will move on to address the key issues of the study, namely:

the definition of PVs in English, the syntactic and semantic properties of PVs in

English, PVs in Arabic, the derivation of PVs, PVs in English lexicography, PVs in

bilingual English-Arabic dictionaries, teaching PVs, and PVs in translation studies.

The primary purpose of this Chapter is to cast light on the main features of the

phenomenon of PVs in both English and Arabic, and to bring together what has been

already explored by linguists, lexicographers and pedagogues in order to set up a solid

theoretical ground upon which the following Chapters of the study can be established.

2.2 Idioms and idiomaticity in English and Arabic

Before proceeding to the definition of PVs and elaborating on their syntactic,

semantic and pragmatic properties, it is quite essential to consider, in a brief account,

the notion of idioms and idiomaticity in both the English and the Arabic languages.

The aim here is to shed light on such a critical notion and to take it as a platform from
15

which one can move on to explore the characteristics of the idiomatic nature of PVs in

both languages.

2.2.1 Idioms and idiomaticity in English

Idioms have been defined in rather various ways by linguists, grammarians,

lexicographers and pedagogues. To take just a few of such definitions, an idiom is

broadly defined as "[a] fixed expression whose meaning is not guessable from the

meaning of its parts" (Trask, 2000, p. 67), or it is "[a]n expression which functions as

a single unit [where its] meaning cannot be derived from the meaning of the

individual elements" (Daud, Dollah, & Zubir, 2003, p. 100). Fraser (1976) considers

an idiom as "a single constituent or series of constituents, whose semantic

interpretation is independent of the formatives which compose it" (p. v). Idioms,

according to Bolinger (1975), are "groups of words with set meanings that cannot be

calculated by adding up the separate meanings of the parts" (as cited in Lattey, 1986,

p. 219). Further, Baker (1992) rightly points out that idioms "are frozen patterns of

language which allow little or no variation in form and […] often carry meanings

which cannot be deduced from their individual components" (p. 63). Thus, with

idioms a user cannot normally change the order of the words in them, delete a word

from them, replace a word with another, or change their grammatical structure unless

he or she is intentionally making a joke or trying a play on words (Baker, 1992).

Therefore, Lattey (1986) points out that "as far as the form of idioms is concerned, we

have groups of words, and in terms of meanings, we can say that we are dealing with

new, not readily apparent meanings when we confront idioms" (p. 219).
16

Some of the fairly common examples include: let the cat out of the bag (meaning:

reveal a secret), buy a pig in a poke (meaning: commit oneself to a course of action

without investigating), a fish out of water (meaning: a person struggling in an

unfamiliar environment) and kick the bucket (meaning: die) (Trask, 2000).

In his attempt to coin a comprehensive definision, Ghazala (2003) defines idioms as

"special, metaphorical, fixed phrases whose meanings and forms are not negotiable"

(p. 204).

Additionally, he sums up the main features of idioms in five points as follows:

1. Idioms are all in all metaphorical and cannot be understood directly.


2. They should not be taken literally; in the sense that their meanings are not
the outcome of the individual meanings of their constituent words taken
collectively.
3. Their syntactic form is usually fixed and cannot be changed or described
as ungrammatical […].
4. Their meanings are also invariable.
5. […] They are mainly cultural and informal (p. 204).

Idioms, though, are considered as "vivid, verbal images which add life and verve to

speech and writing. Without them language would be very bland and unexciting"

(John & Smithback, 1991, preface). Hence, their importance in any language "cannot

be doubted [in view of the fact that their] ubiquity makes them anything but a

marginal phenomenon" (Wallace, 1968, p. 112).

Further, idiomaticity is, as Ghazala (2003) puts it, the idioms' "most special

component [which constitutes] their metaphorical aspect" (p. 204), or "the heart of the

matter of any idiomatic expression" (p. 208), and "the gist of any idiomatic phrase"

(p. 209). Idiomaticity, in the words of Weinreich (1969), is "important for this reason,
17

if for no other, that there is so much of it in every language" (as cited in Fernando,

1996, p. 1). Lastly, it is, according to Palmer (1974) "a lexical feature [i.e.],

something to be dealt with in the lexicon or dictionary rather than the grammar" (p.

213).

Yet, understanding idioms and using them properly entails a degree of proficiency

which is hard for the non-native speaker of a given language to acquire (Turton &

Manser, 1985). Therefore, Wallace (1981) makes the point that "[w]hen it comes to

understanding English it is these expressions which cause most difficulty to the

foreign learner" (p. 5).

Great deals of attempts have been made by English scholars to classify idiomatic

expressions. As a result, they have been categorized in many different ways, being: 1)

according to their 'grammatical type' such as verb-adverb idioms, or idioms that

function like a particular part of speech; 2) according to the 'concept or emotion

portrayed' for instance, the idiom tell someone a tall tale would be categorized under

LIE; and 3) according to the 'image', that is, the picture drawn by the idiom, for

example, a category BODY PARTS would include she lost her head (Lattey, 1986).

It must be stressed here that idioms involve many aspects of English language. They

may occur in such forms as: slang, proverbs, allusions, similes, dead metaphors,

social formulate, and collocations (Fernando, 1996).

Further, Ghazala (2003) puts idioms into five main types, being:

1. Full / pure idioms;


2. Semi-idioms;
18

3. Proverbs, popular sayings and semi-proverbial expressions;


4. Phrasal verbs;
5. Metaphorical catchphrases and popular expressions (p. 208).

What has to be confirmed at this stage is that PVs constitute an integral part of

English idiomatic expressions. They have been classified as one category of English

idiomatic expressions by many researchers other than the abovementioned Lattey

(1986) and Ghazala (2003) (e.g., Spears, 1987; Alexander, 1984: Urdang, 1979 as

cited in Bataineh & Bataineh, 2002, p. 40).

In his book The Verb-Particle Combination in English, Fraser (1976) emphasizes that

"[p]ractically every grammarian of English has noted and commented about idioms in

general. More specifically, almost all have noted the regularity with which certain

adverbials (particles) co-occur with certain verbs" (p. 63).

PVs' elements sacrifice their individual meanings and by the act of combination

assume a new meaning, as, for example, bear out (meaning: corroborate), come by

(meaning: acquire), get at (meaning: reach), make out (meaning: understand), own up

(meaning: confess), and put out (meaning: extinguish) (Kennedy, 1967). It is quite

evident that in such idiomatic PVs, the meanings of the separate parts tell us little or

nothing about the meaning of the whole. For instance, one may be quite familiar with

the meanings of the items pick and up as individual words, but such familiarity does

not help in understanding the idiomatic meaning of the PV pick up in such sentence

as: Business is picking up (Turton & Manser, 1985). The idiomaticity of EPVs will be

explored in detail in the forthcoming sections of this Chapter.


19

2.2.2 Idioms and idiomaticity in Arabic

The phenomenon of idiomatic expressions in the Arabic language has been

exclusively tackled in the Arabic rhetoric (‫ن‬7&-(+‫ ا‬4&)E) by many ancient and modern

Arabic writers such as Al-jaaHiZ (died 255 A.H.) in his book 0--(F+‫ن وا‬7-(+‫( ا‬rhetoric and

clarification) [my translation], Al-jurjaanii (died 471 A.H.) in his books 5&G2(+‫ار ا‬,&H‫أ‬

(secrets of rhetoric) [my translation] and ‫ز‬7&IEJ‫ ا‬K&LM‫( د‬Indications of the Miraculous)

[my translation], and Al-zamakhsharii (died 538 A.H.) in his books 5&&G2(+‫س ا‬7&&H‫أ‬

(Foundation of Rhetoric) [my translation] and ‫ف‬7ّN&'+‫( ا‬The Explorer) [my translation]

among others.

Bearing in mind that rhetoric (‫ن‬7-(+‫ ا‬4)E) being defined by Arabic linguists as "a science

by which the stating of a single meaning in different ways, with a clear indication to it

[the meaning] being known" [my translation] (Shakkour, 1992, p. 64). Yet, this

science falls into four rhetorical styles being: 1) ‫ز‬ ‫( ا‬figurative expression), which

means: Using a word in a meaning which is not its original one owing to a relation

(other than the similarity relation) between the two meanings with a presumption that

the original meaning is not intended (Sayyd Ahmad, 1993) [my translation]. To take

one example:

‫ء‬7FN&+‫ ا‬,&3OP‫( ا‬The winter [sic] poured down) (Shakkour, 1992) [my translation]. In this

example the original meaning of winter (‫ء‬7FN&+‫ )ا‬is not intended. What is meant by

winter here is the rain due to the fact that the rain pours down only in winter in the

Arab land; 2) ‫( ا‬simile), which is, following Shakkour (1992, p. 65), "An

indication of sharing of one meaning by two things" [my translation]. For instance:
20

‫ل‬7&3I+‫ ا‬%&8 5&.I+7‫ن آ‬7&.(+ (Lebanon is like a paradise in beauty) [my translation]. In this

sentence Lebanon and paradise share one meaning, that is, the beauty (p. 65); 3)

‫رة‬ ‫( أ‬metaphor), which means: Using a word in a meaning which is not its original

one owing to a similarity relation between the two meanings with a presumption that

the original meaning is not intended (Sayyd Ahmad, 1993) [my translation]. For

example:.‫ًا‬,&-./ ‫ح‬7(U&3+‫ ا‬SQ&R (The light smiled shining) (p. 125) [my translation]. In

this example the original meaning of light is not intended, it is rather borrowed here to

refer to a handsome person due to the similarity relation between the light and the

handsome person, i.e. shining; and 4) ‫( ا‬antonomasia), which means: Any word

that expresses a meaning which can be perceived literally and metaphorically at the

same time (Shakkour, 1992) [my translation]. For example: 5WXY&(/ 0Y&Z [&1 (Hassan's

hand is outstretched) (Sayyd Ahmad, 1993) [my translation]. In this example it is

possible to perceive the meaning of outstretched literally, i.e. it is not grasped, or

metaphorically, i.e. he is a generous person.

People in the Arab world use idiomatic expressions for two reasons, according to Abu

Sa'ad (1987 as cited in Bataineh & Bataineh, 2002), they are:

1) to beautify their language and distinguish it through such a stylistic


phenomenon […and] 2) to avoid mentioning a word that may cause
embarrassment or annoyance (p. 47).

Interestingly, Abu Sa'ad (1987 as cited in Bataineh & Bataineh, 2002) outlines the

structural and semantic characteristics of Arabic idioms as follows:

1. Idioms come in the following structural patterns:


a. The sentence which consists of two or more words;
Example: to put the cart before the horse [‫ن‬7UQ+‫م ا‬7/‫ أ‬5\,*+‫^] ا‬1]
b. Genitive constructions whose individual meanings are familiar,
while the result of their combination is unfamiliar;
Example: Noah's ark [‫ح‬X&&P 5.-_&&H] 'something that gathers many
objects or species'
21

the patience of Job [‫ب‬X1‫ أ‬,(`] 'real patience and tolerance'


c. Individual idiomatic words;
Example: He is an ear [‫ أذن‬X&‫' ]ه‬He tells of what he hears without
thinking'
2. They are influenced by certain linguistic phenomena such as:
a. synonymy [‫ادف‬,&F+‫]ا‬, where different structures express the same
meaning;
b. homonymy [017&&&(F+‫]ا‬, where one structure expresses different
meanings; and
c. antonymy [‫د‬7^&&&F+‫]ا‬, where one structure expresses opposite
meanings.
3. Idioms are related to proverbs and, thus, the more common the proverb the
greater its chance of being an idiom.
4. Idioms derive their figures from the environment, Arabic, like English, is
full of idiomatic expressions. Yet, there are more of them in Arabic
dialects than in Modern Standard Arabic. […] Many of the Arabic idioms
are easy to understand because their meanings are not that far from the
sum total of their respective components. But others, just as in English, are
difficult to understand, especially for non-native speakers of Arabic,
simply because their meanings are far from the sum of their components
(pp. 47-48).

Along these lines, Awwad (1990) indicates that what is said about English idioms

regarding their semantic and syntactic restrictions also applies to Arabic idioms. He

gives the following example : 0-&.Z %&_n\ ‫د‬7&E (he returned empty-handed) (literally: he

returned with the slippers of Hunain) where we cannot substitute K&`‫( و‬he arrived) or

‫ر‬7&H (he walked) for ‫د‬7&E (he returned) and keep the idiomaticity of the expression (p.

58).

Moreover, in his attempt to further compare English idioms with their Arabic

counterparts, Awwad (1990) makes the point that English idioms can be lexemic as in

(hammer and tong), phraseological as in (to fly off the handle) and proverbial as in

(don't wash your dirty linen in public). The lexemic idioms, however, can be verbal

(verb + particle) as in (break in), nominal as in (hot dog), adjectival as in (pepper and

salt), and adverbial as in (hammer and tong). By the same token, Arabic idioms can be

lexemic as in ‫ر‬7&&P‫ و‬4Q&&o (literally: fat and fire, meaning: completely opposites),
22

phraseological as in %&H‫ را‬/ %&.-E p&)E (literally: on my eye / head, meaning: with

pleasure), and proverbial as in K&`‫&[رب و‬+‫ ا‬p&)E ‫ر‬7&H 0&/ (literally: he who walks on the

road will get there, meaning: he who takes the first step will eventually achieve his

aims). ,Like English Arabic lexemic idioms can be verbal, nominal, adjectival, and

adverbial (p. 58). Yet, "Arabic verbal lexemic idioms do not occur with particles" (p.

58). Therefore, the Arabic equivalent for (he broke into the house) is x-(+‫ ا‬4QFw‫ ا‬or K&y‫د‬

‫ة‬X&.E x&-(+‫( ا‬meaning: he entered the house by force) (p. 58). Hence "Arabic verbal

lexemic idioms are made up of either the verb alone or the verb followed by an

adverbial nominal" (p. 58).

In their monolingual Arabic-Arabic dictionary 5-Z2{&`M‫ات ا‬,&-(*F)+ %w7-Y&+‫ ا‬4&I*3+‫( ا‬The

Contextual Dictionary of Idiomatic Expressions) [My translation] Siinii, Hussain and

Al-ddoush (1996) put together more than 2000 Arabic idiomatic expressions collected

from a wide range of ancient and modern Arabic literature, representing all the

aspects of such a phenomenon in the Arabic language. This book will be reviewed in

more detail in section six of this Chapter when the issue of PVs in Arabic will be

attended to.

To sum up so far, five fruitful insights can be arrived at as to the notion of idioms and

idiomaticity in both Arabic and English languages, they are: 1) generally speaking,

both languages rely on idiomatic expressions in all aspects of their spoken and written

modes; 2) idiomatic expressions in both languages are of a special nature and subject

to syntactic and semantic restrictions; 3) they are, in both languages, rather culture-

specific and their meanings are far from the sum of the meanings of their individual

components; 4) their semantic and syntactic complexities require a high proficiency


23

level on the part of non-native speakers to be able to understand and produce them

properly; and 5) unlike English which gives a clear prominence to IEPVs as one of

the most important types of idiomatic expressions, Arabic does not categorize them as

such. This is basically because they do not represent such a category in the Arabic

language. Therefore, since this study is devoted to translating IEPVs into Arabic,

more light will be shed, in the forthcoming sections, on this particular area in an

attempt to explore the gap that exists between the two languages which causes a great

deal of difficulties to Arabic translators, lexicographers, pedagogues and learners of

English.

2.3 Definition of PVs in English

In the previous section the notion of idioms and idiomaticity in general was described.

The present section, however, focuses particularly on the ways by which English

scholars have defined the phenomenon of PVs.

There is no doubt that defining EPVs is a fairly difficult task as the phenomenon is

still being debated among researchers in such domains as grammar, linguistics,

pedagogy and lexicography. Bolinger (1971) admittedly considers that "I do not

believe that a linguistic entity such as the phrasal verb can be confined within clear

bounds […] being or not being a phrasal verb is a matter of degree" (p. 6).

To start with, a variety of definitions produced by linguists, grammarians,

lexicographers and pedagogues will be outlined below. The aim is to highlight the
24

points they have in common, and to come up with a rather comprehensive definition

for the phenomenon of EPVs.

PVs are, following Live (1965), "a considerable group of basic verbs, each of which,

in certain of its occurrences, [is] closely linked with a particle-adverbial or

prepositional- in such a manner as to justify considering the two elements as

constituting one discontinuous verb" (p. 428). They are, in Heaton's words, "an

adverbial particle combines with a verb to form a collocation possessing a new

meaning. [Each] phrasal verb must be considered as a unit" (1965, preface).

PV, according to Bolinger (1971), is "a lexical unit in the strict sense of a nonadditive

compound or derivative, one that has a set of meaning which is not the sum of the

meaning of its parts" (p. xxi). Sroka (1972) maintains that "[…] the verb and particle,

or the verb and a group of particles, are said to constitute in this case a kind of integral

functional unit" (p. 14). Furthermore, the phenomenon has been defined in the

Dictionary of English Phrasal Verbs and their Idioms, by McArthur and Atkins

(1974) as "[…] combinations of simple, monosyllabic verbs (put, take, get, etc.) and

members of a set of particles (on, up, out, etc.)" (p. 5). Along these lines, McArthur

(1975) points out that "[a] phrasal verb is formed by combining a simple verb and one

of a number of particles. The result is called 'phrasal' because it looks like a phrase

rather than a single word. Although it looks like a phrase, it functions as a single

word. It is a unit" (p. 9). Similarly, in The Student's Dictionary of Phrasal Verbs,

Turton and Manser (1985) define PV as a "verb which consists of two or three

separate parts: come in, run away, look forward to, etc. With an idiomatic phrasal

verb, the meanings of the separate parts tell us little or nothing about the meaning of
25

the whole" (p. iv). By the same token, Dixon (1991) defines PVs as "a combination of

verb plus preposition(s) that has a meaning not inferable from the individual meanings

of verb and preposition(s), so that it must be regarded as an independent lexical item,

and accorded a dictionary entry of its own" (p. 274).

What is more, in Making Sense of Phrasal Verbs, Shovel (1992) defines PV as:

[…] a compound verb formed by one of the following combinations: (1) verb
and adverb; or (2) verb and preposition; or (3) verb with both adverb and
preposition […] Simple combinations like sit down and stand up cause the
learner fewer problems. The difficulties being when the combination is
'Idiomatic': that is, when the meaning of combination as a whole (i.e. the
phrasal verb) is different from the meanings of its separate parts (p. 5).

Likewise, Close (1992) makes the point that such "constructions are formed by the

verb be or a simple verb expressing physical action – come, go; put, take; give, get;

do, make; let, keep; bring, send; stand, fall, sit; turn; break, […] - followed by a

preposition or particle indicating direction or position" (p. 149).

Further, Cowie and Mackin (1993) indicate that "[t]he combination has to be

understood as one unit, meaning 'start suddenly or violently'. When a verb + particle

(or a verb + preposition) is a unit of meaning like this it is a PHRASAL VERB"

(p. xi) [Emphasis in original].

Crowley, Lynch, Siegel and Piau (1995), similarly observe that a PV is "a verb which

is made up of more than one element, usually including a verb and a PARTICLE, […]

The meaning of a phrasal verb is quite different from that of the sum of its parts"

(p. 334) [Emphasis in original].


26

In addition, Lindstormberg (1998) lists three criteria to define PVs, claiming that

"[n]on-literality, idiomaticity and paraphrasability-in-one-word are rather rough and

ready definition criteria for phrasal verbs" (p. 23). However, in The Grammar

Dictionary, Stern (2000) defines PV as "a word cluster that consists of a verb + one or

two adverbial particles" (p. 142). Likewise, in Collins Cobuild Dictionary of Phrasal

Verbs, Sinclair et al. (1998) define PVs as "[…] combinations of verbs with adverbial

or prepositional particles" (p. iv).

Another definition is given by Thrush (2001). He notes that a PV is "a verb and a

preposition (or two). The meaning is often idiomatic; that is, the meaning of the

phrasal verb cannot be derived by looking up the verb and the preposition separately

in a dictionary" (p. 292).

In a nutshell, the array of definitions outlined above have six points in common,

being: 1) an EPV is basically formed of two words: a simple, monosyllabic verb and

one of a number of particles; 2) the particle could be an adverbial or a prepositional;

3) some PVs are formed of three words: a verb and both an adverb and a preposition;

4) semantically speaking, there are two types of PVs: literal (non-idiomatic) and

metaphorical (idiomatic). The meaning of PV, in the former, can be easily deduced

from the individual meanings of verb and particle. In the latter, however, the meaning

is not inferable and utterly different from the total sum of the meanings of the separate

parts; 5) a PV constitutes one lexical unit and functions as a single word of one unit of

meaning; and 6) a PV can be paraphrased by one single word.


27

All in all, an IEPV is a combination of two or three items (a verb + a preposition, a

verb + an adverb, or a verb + an adverb + a preposition) which functions as a single

unit of meaning in the sense that its meaning cannot be deduced from the total sum of

the meanings of its separate elements. IEPVs are exemplified in what follows:

To carry out, to carry on, to turn up, to turn on, to turn off, to come across, to come

over, to come out, to write up, to slow down, to speed up, to throw out, to throw up,

to help out, to knock off, to knock down, to sort out, to give up, to give in, to give

away, to get away with, to black out, to tip off, to account for, to point out, to water

down, to take off, to take in, to look for, to put up with, etc.

2.4 Syntactic properties of EPVs

In this section I will closely look at the phenomenon of EPVs from the syntactic

standpoint, elaborating on such grammatical features as: word class, syntactic

categories, and word order. The key aim here is to draw a clear picture for such a

phenomenon that enables me, later on, to compare it with its Arabic counterpart so as

to pinpoint the underlying contrasts that present Arabic professional translators and

Arabic translation students with a great deal of challenge when handling IEPVs.

2.4.1 Word class of EPVs

There are two points that need to be noted here; the first one is the types of verb that

can be phrasalised, and the second is the types of particle that may co-occur with

these verbs:
28

2.4.1.1 Verbs

McArthur and Atkins (1974) maintain that there are at least six types of verb that can

be phrasalised. They list them as follows:

a) verbs of movement (usually monosyllabic and of Anglo-Saxon origin):


go, come, run, walk, hope, skip etc.
b) verbs of invitation and ordering etc.: invite, order, summon, let etc.
c) the so-called 'empty verbs' or verbs of indefinite meaning: get, put,
take, make, do etc.
d) verbs formed with or without the suffix –en, from simple monosyllabic
adjective: brighten, slacken, flatten, dry, cool etc.
e) verbs formed unchanged from simple, usually monosyllabic nouns
with such paraphrase patterns as:
chalk up= mark up with chalk
brick up= seal up with brick
f) a random scattering of two-syllable verbs of Latin origin, with which
some kind of direction or emphasis is required: contract (out),
measure (up), level (off) etc. (p. 6) [Emphasis in original].

However, some grammarians, for instance Dixon (1991), list these types of verb

under different names, such as: "MOTION (e.g., bring, carry), REST (e.g., sit, stand),

AFFECT (e.g., cut, kick, scrape), GIVE (e.g., give, get, have), MAKING (e.g., make,

let), or the grammatical verbs be and do" (p. 275) [Emphasis in original].

It is quite important to mention here that Quirk, Greenbaum, Leech, and Svartvik

(1985) make the point that there are "words which occur as verbs only when

combined with particles, [for example:] beaver in beaver away, egg in egg on, and

eke in eke out: she egged him on. *She egged (him)" (p. 1151 footnote) [Emphasis in

original].

Live (1965), also, indicates that there are some verbs "which never occur

independently, always being associated with some particles [e.g.,]: put, lay, set, step,
29

and others" (p. 432). Likewise, there are some "wedded pairs" where " the verb never

(or hardly ever) appears without its particular particle", such as: auction off, jot down,

tide over, cave in, dole out, balk at, cope with, trifle with, cater to, delve into, dote on

(p. 432).

Fraser (1976), on the other hand, emphasizes that "[s]tative verbs such as know, want,

see, hear, hop, resemble, etc. practically never combine with a particle. Hear out

appears to be an exception to this generalization" (p. 11) [Emphasis in original].

He also points out that although there are some verbs that can form a PV "with almost

every particle" such as Get, there are "some verbs that may co-occur with only one

particle and no others [e.g.,] book up, chicken out, fizzle out, fog up, jack up, shack up,

sober up, pan out, [and] jot down" (p. 9) [My emphasis].

2.4.1.2 Particles

Particles in EPVs are of two kinds, namely: prepositional and adverbial indicating

direction or position/location (Close, 1992).

It is interesting to note, in this respect, that most of the scholars who dealt with the

notion of PVs, start their studies by accounting for the distinction between the verb-

adverb combination and the verb-preposition combination (cf. Palmer, 1968;

Bolinger, 1971; Fraser, 1976; Dixon, 1982; Kaluza, 1984; Close, 1992; Beate Hampe,

1997 to cite only a few). Cowie (1993) points out that "[t]he distinction between verb

+ preposition and verb + adverb is central to any systematic treatment of phrasal


30

verbs" (p. 38). Along these lines Sroka (1972) observes that "[t]he problem of the

distinction between the adverb and the preposition in English […] constitutes one of

the central problems connected with the 'phrasal verbs'" (p. 15).

Thus, let us first have a close look at the differences between the two types of

particles, i.e. adverbs and prepositions; to be able to perceive the role they play when

combining with simple verbs to form PVs.

2.4.1.2.1 Adverbs

Adverbs in general, as defined by Aarts (1997), are modifiers of verbs, adjectives or

other adverbs as in:

a) Our colleague from Paris merrily marks student essays in his path.
b) The teachers are extremely unimpressed by his efforts.
c) Our new professor works very hard (p. 42) [Emphasis in original].

They are, on the other hand, of three types: place, time and manner as illustrated in

these examples given by (Azzaro, 1992):

(a) Let's stay here (place)


(b) We're going now (time)
(c) She speaks well (manner) (p.41) [My emphasis]

However, the majority of adverbs used in EPVs, as indicated by Azzaro (1992), are

locative (place) adverbs as in:

a) The furious farmer came after us.


b) He has quite a bit of money laid by.
c) He acted as a go-between. (p. 42) [My emphasis].

Moreover, Azzaro (1992) asserts that the only important exception to this rule being

on, as in:

He carried on telling the same old story (p. 42) [My emphasis].
31

Interestingly, McArthur (1989) lists the following 31 adverbial particles which are

typically occurred in EPVs:

Aback, about, ahead, along, apart, aside, around, away, back, backward(s),
beyond, by, down, downward(s), forth, forward(s), in, inward(s), off, on,
onward(s), out, outward(s), over, past, round, sideways, through, to and fro,
up, upward(s) (p. 40) [Emphasis in original].

Further, he nominates six particles of them as the commonest: "down, in, off, on, out,

up" (p. 40) [Emphasis in original].

Yet, Cowie and Mackin (1993) add to what is mentioned above the following 32

adverbial particles:

Aboard, above, abreast, abroad, across, adrift, after, aground, ahead, aloft,
alongside, aside, astray, before, behind, below, between, counter, downhill,
downstairs, home, indoors, in front, inside, near, on top, outside, overboard,
together, under, underground, upstairs, without (p. vii) [My emphasis].

In addition, McArthur (1989) makes the point that there are some words that may

function as adverbial particles such as: home, open and shut as in the following

examples given by him:

He forced the door open/forced open the door.


She hammered the nail home/hammered home the nail (p. 40) [Emphasis in
original].

By the same token, Bolinger (1971) gives the following two examples in which he

considers home as a particle:

He brought home the groceries.


He brought home the point in a convincing manner (p.16) [My emphasis].

However, unlike other adverbs, adverbial particles, as Heaton (1965) sketches them,

are "best regarded as helping to form a new verb, for they change, or add to, the

meaning of the verb, however slightly. [Therefore] the particle is really an integral
32

part of the phrasal verb, separable often in word order but nevertheless constituting a

single unit" (p. 45). Moreover, Heaton (1965) limits the function of most adverbial

particles to the following five ways:

(1) Many cause a verb to assume a new or subsidiary meaning. The


widow carried on as if nothing had happened. (continued) [...].
(2) Some particles assume a new or special meaning, with a verb but
do not change the normal meaning of that verb. […] Speak out.
(Speak loudly or candidly) As the verb retains its usual meaning
in such cases, it is often possible to deduce the meaning of the
whole collocation: e.g. look over = inspect; switch on = connect.
(3) Other particles provide stress emphasis, or a sense of completion.
Let me finish this off before I leave […].
(4) Others function in a similar way to ordinary adverbs, helping to
form a collocation which maintains a literal meaning. She got in
her car and drove away (or off) without speaking […].
(5) A number are similar in function to prepositions. Although they
are linked to the verb and conform to the same rules of word
order, etc. as other adverbial particles, a noun equivalent
following them is often understood (through not expressed). Take
your hat off (your head). She came down (the stair) (pp. 45-46)
[Emphasis in original].

Further, Fraser (1976) indicates that "there are numerous verb-particle combinations

[PVs] in which the verb, when occurring alone, has a radically different

interpretation" (p. 8). He gives the following sentence pairs to exemplify such verbs:

They usually box his ears as punishment


Please box up my gift.
They cracked the case with the hammer.
She really cracked up at my jokes (p. 8) [My emphasis].

Furthermore, Kennedy (1967) makes the point that in some cases "the object of the

[verb-adverb] combination is, or may be, of a very different character from that of the

simple verb". He provides the following illustrative instances:

argue a case but argue down an opponent,


burn a paper " burn off a field,
buy a house " buy off or out a person,
clean a room " clean out its contents,
dig a hole " dig out or up a plant
(pp. 26-27) [Emphasis in original].
33

What is more, there are some intransitive verbs which become transitive when

combining with particles (Fraser, 1976; Kennedy, 1967), as in:

He slept off the effects of the drinking.


She looked up the information.
(Fraser, 1976, p. 8) [Emphasis in original].

By contrast, there are a larger number of verbs which are ordinarily transitive become

intransitive when a particle is added (Fraser, 1976; Kennedy, 1967), for example:

They dug in and sustained the attack without loss of life.


She begged off at the last minute.
He caught on quickly.
The distraught couple split up (Fraser, 1976, p. 8) [Emphasis in original].

Likewise, Heaton (1965) maintains that PVs "offer a convenient means of making

intransitive use of transitive verbs by the addition of a particle" (p. 46).

In his elaborating on this point, Kennedy (1967) maintains that "[i]n many of these,

no doubt, a reflexive object is understood so that, strictly speaking, they are

intransitive in form only. So, for instance, black up and clean up imply "to black up or

to clean up oneself'" (p. 26).

2.4.1.2.2 Prepositions

A preposition, according to Quirk et al. (1985), "expresses a relation between two

entities, one being that represented by prepositional complement, the other by another

part of the sentence. The prepositional complement is characteristically a noun phrase,

a nominal wh- clause, or a nominal ing clause" (p 657). They are of five types, as
34

classified by Collins (1998) according to their (circumstantial) meaning: time, place,

manner, agency and recipience, as exemplified in what follows:

1 time after our match; during the exam.


2 place in the kitchen; against the wall.
3 manner with ease.
4 agency by the mechanic.
5 recipience to a friend (p. 32) [Emphasis in original].

Cowie and Mackin (1993) list the following 52 prepositions that are used to form

EPVs:

Aboard, about, above, across, after, against, ahead of, along, alongside,
among, around, as, as far as, astride, at, before, behind, below, beneath,
beside, between, beyond, by, down, for, from, in, in front of, inside, into, like,
near, of, off, on, onto, on top of, out of, outside, over, past, round, through, to,
toward(s), under, underneath, up, upon, with, within, without (p. vii) [My
emphasis].

However, despite the fact that adverbs and prepositions often share the same form,

they differ in that each of them has a different relationship to the rest of the sentence.

Adverbs are related only to the words that they modify, while prepositions connect

their objects with other words (Corless, 1979). For example:

I came across my classmate yesterday. (Adverb)

He came across the street to say hello. (Preposition)

Fraser (1976), however, indicates that prepositions "are syntactically more closely

associated with the noun phrase which follows them than with the verb which

precedes" (p. 2).

Similarly, Sroka (1972) makes the point that "the adverb is more closely connected

with the verb than the preposition is, and that the preposition is closely connected

with a noun or noun-equivalent" (p. 22).


35

Yet, there are many words used as either adverbs or prepositions. By comparing the

list of prepositional particles with the above list of adverbial particles one can come

up with the following list of particles of dual functions:

Aboard, about, above, across, after, along, alongside, around, before, behind, below,

between, beyond, by, down, in inside, near, off, on outside, over, past, through, to,

under, up, without.

It is worth mentioning that Bolinger (1971) introduces the term "Adpreps" to describe

such particles with dual functions, which "form the most typical phrasal verbs […

and] function now as adverbs, now as prepositions" (p. 23).

Such dual functions are illustrated in the following examples provided by Thomson

and Martinet (1986):

Peter is behind us. (preposition)


He is along behind. (adverb)
He runs up the stairs. (preposition)
He went up in the lift. (adverb) (p. 104) [Emphasis and bracketing in original].

Nevertheless, Jackendoff (2002) maintains that unlike most English particles back and

together "do not double as prepositions" (p. 70).

Such a matching in form prompts Palmer (1974) to argue that "the term 'particle' has

been used in order not to distinguish […] between preposition and adverb [as] a

striking characteristic of many [of them] is that they can function as either" (p. 214).

He goes beyond that to say "[i]t might be plausible to argue that English does not, in

fact, have two words classes adverb and preposition, but a single class 'particle' or,

perhaps, 'prepositional-adverb'. For there is considerable similarity in their function"


36

(p. 215). But, Azzaro (1992) declares that traditional grammarians "do distinguish

between adverb and preposition" (p. 43). Palmer (1968) indicates that "[w]hen there is

no noun at all following the verb we must, of course, have verb plus adverb, since by

definition a preposition will always be followed by a noun phrase" (p. 182). Thus, the

particles in the following examples are adverbs:

The tree blew down.


The injured man came to (p. 182) [Emphasis in original].

Morphologically speaking, Quirk et al. (1985) indicate that "[t]he words follow the

lexical verb in expressions like drink up, dispose of, and get away with are

morphologically invariable [they are] belong to two distinct but overlapping

categories, that of prepositions and that of spatial adverbs". Further they give them the

label "PARTICALES" as a "neutral designation" one (p. 1150) [Emphasis in original].

The fact that particles in EPVs are morphologically invariable has been ably

explained by Darwin and Gray (1999). They observe that:

This fact carries two major implications, the first being that all inflections are
of the verb proper. Thus, you're pulling me on and he makes up lies are
expected whereas *you're put me oning and * he make ups lies are reserved for
children, ESL learners, and slips of tongue. The second implication is that
morphologically variable words do not serve as particles (p. 69).

Last but not least, Azzaro (1992) limits the differences between these two classes of

lexemes (i.e. adverbs and prepositions) to one main difference which "lies in the

intransitiveness of the former and transitiveness of the latter" (p. 41). Further, he

introduces the term "adverbial prepositions", claiming that "even though modern

grammarians believe that some prepositions can be transitive and intransitive, so that

we may have cases of adverbial prepositions" (p. 41).


37

2.4.2 Syntactic categories of EPVs

Having known the difference between the classes of adverb and preposition, it is quite

essential here to cast a light on the major categories of verb + particle combinations.

There are three types of combinations:

1. Verb + adverb combination.

2. Verb + preposition combination.

3. Verb + adverb + preposition combination.

Basically, verb + adverb combination consists of a simple verb and an adverbial

particle as in:

1) Sorry, Australia, I've let you down

(The Sun-Herald, August 8, 2004, pp. 10-11) [My emphasis].

2) A post mortem has been carried out

(The Daily Telegraph, August 12, 2004, p. 9) [My emphasis].

3) We can't back down

(The Sunday Telegraph, April 11, 2004, p. 1) [My emphasis].

Whereas, verb + preposition combination consists of a simple verb and prepositional

particle followed by a prepositional object as in:

4) Prime Minister John Howard stepped into the furore

(The Daily Telegraph, August 12, 2004, p.1) [My emphasis].

5) The crossing was closed because militants were planning to tunnel under it

(The Sun-Herald, August 8, 2004, p. 51) [My emphasis].

6) Miss Universe win puts Jennifer over the moon

(The Sydney Morning Herald, June 3, 2004, p. 2) [My emphasis].


38

Finally, verb + adverb + preposition combination consists of a simple verb, adverbial

particle, and prepositional particle followed by a prepositional object. These

combinations are illustrated by Jacobs (1995) in what follows:

Cut down on "reduce"


Drop in on "visit casually"
Get away with "violate a rule without punishment"
Go back on "violate an agreement"
Go through with "finish, complete"
Keep up with "stay level with"
Look down on "despise"
Make up for "compensate"
Put up with "tolerate"
Run out of "have no more"
Run up against "meet as an obstacle"
Stand up for "defend" (p. 250) [Emphasis and bracketing in original]

The three categories of verb + particle combinations are summarized in Table 2.1

below:

Type Combination Example


1 Verb + adverb I will never give in.
2 Verb + preposition He looked at the mirror.
3 Verb + adverb + preposition She cannot get away with it.
Table 2.1 Verb + particle combinations

It is important to state here that scholars vary in which type of the above is a PV.

Some of them, like Quirk et al. (1985), consider the first type only as a PV, whereas

others, like Courtney (1983), consider the three types as PVs as far as they are

idiomatic. Such a question will be investigated in more details in the next section

when semantic properties of EPVs are accounted for.

Having generally known that, let us consider these three types of verb + particle

combinations one by one more carefully.


39

2.4.2.1 Verb + adverb combinations

Quirk et al. (1985) recognize this type only as "phrasal verbs" (pp. 1150-1161). It is,

according to them, of two types, namely: Intransitive PVs and transitive PVs.

2.4.2.1.1 Intransitive PVs

Intransitive PVs consist of a verb plus an adverbial particle. They do not require any

object. Quirk et al. (1985) cite the following illustrative examples:

The plane has just touched upon.


The plane has now taken off.
The prisoner finally broke down.
She turned up unexpectedly.
When will they give in?
The tank blew up (p. 1152) [Emphasis in original].

The adverbial particles in these PVs function like predication adjuncts, and usually

cannot be separated from their lexical verbs (Quirk et al., 1985). Registerwise, these

PVs are typically used informally (Quirk et al., 1985).

Moreover, Quirk et al. (1985) differentiate between this type of combinations and

what they call "FREE COMBINATIONS", saying that in the former "the meaning of

the combination manifestly cannot be predicted from the meanings of verb and

particle in isolation [but in the latter] the verb acts as a normal intransitive verb, and

the adverb has its own meaning" (p. 1152). They exemplified the former in: Give in

'surrender' and Catch on 'understand'. And they exemplified the latter in:

He walked past. [='past the object/place']


I waded across ['across the river/water/etc'] (p. 1152) [Emphasis and
bracketing in original]
40

2.4.2.1.2 Transitive PVs

Transitive PVs consist of a verb plus an adverb particle plus a direct object. They

require a direct object to complete their meaning. This type of PVs exemplified by

Quirk et al. (1985) as follows:

We still set up a new unit.


Shall I put away the dishes?
She's bringing up two children.
Someone turned on the light.
They have called off the strike (p. 1153) [Emphasis in original]

The adverb particle in such PVs "can either precede or follow the direct object: They

turned on the light. They turned the light on" (Quirk et al., 1985, pp. 1153-1154)

[Emphasis in original].

Like any transitive verb, this type of PVs "can normally be turned into passive

without stylistic awkwardness [as in]: Aunt Ada brought up Roy. Roy was brought up

by Aunt Ada" (Quirk et al., 1985, p. 1154) [Emphasis in original].

Additionally, Palmer (1968) indicates that "[v]erbs of this type [transitive PVs] are, of

course, very common especially with certain verbs like TAKE and PUT" [Emphasis

in original] (p. 187). He cites the following ilustrative example:

Put about a rumour.


Put back the clocks.
Put down a rebellion.
Put in an application.
Put out a pamphlet.
Put over an idea.
Put off a meeting (p. 187) [Emphasis in original].
41

2.4.2.2 Verb + preposition combinations

This type of a combination is called a 'prepositional verb'. Like the last one, it is of

two types as well, being: intransitive prepositional verbs, and transitive prepositional

verbs.

2.4.2.2.1 Intransitive prepositional verbs

This subtype of prepositional verbs "consists of a lexical verb followed by a

preposition with which it is semantically and/or syntactically associated. The

preposition […] precedes its complement" (Quirk et al., 1985, p. 1155). For example:

Look at these pictures.


I don't care for Jane's parties.
We must go into the problem.
Can you cope with the work? (Quirk et al., 1985, p. 1155) [Emphasis in
original].

Quirk et al. (1985) indicate that "[t]he noun phrase following the preposition in such

constructions is termed a PREPOSITIONAL OBJECT [that is, not a direct object]"

(p. 1156) [Emphasis in original]. And the passive is commonly possible as in:

The picture was looked at by many people (Quirk et al., 1985, p. 1156) [Emphasis in

original].

2.4.2.2.2 Transitive prepositional verbs

This subtype of prepositional verbs is made up of a lexical verb plus a preposition

"followed by two noun phrases, normally separated by the preposition: the former is

the direct object, the latter the prepositional object" (Quirk et al., 1985, p. 1158). For

example:
42

He invested his money in property.


The gang robbed her of her necklace.
This clothing will protect you from the worst weather.
May I remind you of our agreement? (Quirk et al., 1985, p. 1158) [Emphasis
in original].

It is worth mentioning to say that some grammarians (like Close, 1992) do not

differentiate between the direct object and the prepositional object. Therefore, they

consider all constructions of prepositional verb (verb + preposition) are transitive, that

is, they have to be followed by objects (Close, 1992, p. 149).

2.4.2.3 Verb + adverb + preposition combinations

This type of a combination is typically termed as "phrasal-prepositional verb"

(Mitchell, 1958; Palmer, 1968; Quirk et al., 1985), in view of the fact that "they

contain, in addition to the lexical verb, both an adverb and a preposition as particles"

(Quirk et al., 1985 p. 1160). Palmer (1968) maintains that "[w]here there are two

elements following the verb, the first will always be an adverb, and the second a

preposition" (p. 183). Whereas, Kennedy (1967) calls them "double combinations" (p.

32) due to the fact that the whole phrase functions as a single verb, and can be

translated by one simple verb.

Phrasal-prepositional verbs are, registerwise, "largely restricted to informal English"

(Quirk et al., 1985, p. 1160). Like the previously mentioned combinations, they are of

two types, being: intransitive phrasal-prepositional verbs, and transitive phrasal-

prepositional verbs.
43

2.4.2.3.1 Intransitive phrasal-prepositional verbs

This subtype of phrasal-prepositional verbs is followed by prepositional object rather

than a direct object. It is illustrated by Quirk et al. (1985) as follows:

We are all looking forward to your party on Saturday.


He had to put up with a lot of teasing at school.
Why don't you look in on Mrs. Johnson on your way back?
He thinks he can get away with everything (p. 1160) [Emphasis in original].

Quirk et al. (1985) make the point that "[t]he prepositional passive with such verbs is

not too common, and is liable to sound cumbersome" (p. 1160). However, they

indicate that sentences such as the following "are normal and acceptable" (p. 1160):

These tantrums could not be put up with any longer. ['tolerated']


The death penalty has been recently done away with. ['abolished']
Such problems must be squarely faced up to. ['confronted']
They were looked down on by their neighbours. ['despised']
(p. 1160) [Emphasis and bracketing in original]

2.4.2.3.2 Transitive phrasal-prepositional verbs

This subtype of phrasal-prepositional verbs is followed by a direct object. It is

exemplified by the following sentences given by Quirk et al. (1985):

Don't take it out on me! ['vent your anger']


The manager fobbed me off with a cheap camera. [especially British English]
We put our success down to hard work. ['attribute to']
I'll let you in on a secret (p. 1161) [Emphasis and bracketing in original]

Table 2.2 below gives an outline of the abovementioned three types of combinations

along with their subtypes:


44

Type Combination Intransitive Transitive


1 Verb + adverb Can you please carry Don’t let me down.
on?
2 Verb + preposition He is looking for a car She reminds me of our
to buy. meeting.
3 Verb + adverb + She cannot put up with I will get the price down
preposition you. on this.
Table 2.2 Types and subtypes of verb + particle combinations

It is necessary to point out, however, that Lindstromberg (1998, p. 23) rightly

criticizes the sub-classifying of EPVs into the abovementioned categories (i.e.,

prepositional verbs, phrasal verbs, and phrasal-prepositional verbs), saying that "I

have never found it useful to bear this terminology in mind when explaining meaning,

and so […] I use the term 'phrasal verb' to include all [the] three syntactic types" (p.

23).

2.4.3 Word order of EPVs

As it has been previously outlined, EPVs, like other ordinary verbs, can be either

transitive or intransitive (McArthur, 1975). There are some verbs, however, that can

be used both as transitive and intransitive verbs (Sinclair et al. 1998; Jespersen, 1976).

Along these lines, Dixon (1991) points out that "[t]ransitivity is a much more fluid

matter in English. There are, it is true, a number of verbs that are strictly transitive

[…] and a few that are strictly intransitive […]. But many verbs in English may be

used either transitively or intransitively" (p. 267). Such PVs have been illustrated by

Quirk and Greenbaum (1973) as follows:

Drink up quickly. [Intransitive PV]


Drink up your milk. [Transitive PV]
When will they give in? [Intransitive PV]
45

They gave in their resignation. [Transitive PV]


(pp. 347-348) [Emphasis in original. My bracketing].

Similarly, Palmer (1968) observes that "[s]ome of the combinations may be used

transitively as well as intransitively, e.g., BLOW UP and BLOW DOWN", for

example:

The house blew up.


They blew up the house.
The chimney pot blew down.
The wind blew the chimney pot down (p. 185) [Emphasis in original].

Further, Sinclair et al. (1998) indicate that some PVs can be ditransitive, that is, they

require two objects: a direct object and an indirect object as in:

The girl handed him back his card (p. xiv) [My emphasis].

Where the adverb back comes between the indirect object, which is the pronoun

(him), and the direct object, which is the noun group (his card).

It is necessary to indicate, however, that transitive PVs fall into two major types:

separable and non-separable PVs, or split alteration and non-split alteration as they

were termed by Sawyer (1999), or fused and separable as were named by McArthur

and Atkins (1974). Such a distinction typically depends upon the position of the

particle in the sentence (Bolinger, 1971; Quirk & Greenbaum, 1973; Fraser, 1976;

Jacobs, 1995; Gries, 2002). Particles may "occur on either side of the direct object

noun phrase as one of its defining characteristics" (Fraser, 1976, p. 16). Therefore, in

separable PVs the particle follows the object noun as in:

Can you turn the light off?

Whereas, in non-separable PVs the particle precedes the object noun as in:

I gave up smoking two years a go.


46

In other words, the former allows, and sometimes requires, "the particle to occur in a

slot that is not adjacent to its verb", but the latter does not allow such separation

(Jacobs, 1995, pp. 248-249). Normally, the particle, according to Quirk and

Greenbaum (1973), "tends to precede the object if the object is long or if the intention

is that the object should receive end-focus" (p. 248).

There is a crucial point that needs to be made here. When the direct object of PV is a

pronoun (e.g., him, her, it, them, etc.) the PV must be separable as in:

I picked it up.

He handed them in.

She let him down.

Lindner (1983) makes the point that "[t]he particle's position when the object is a

pronoun has been much commented on" (p. 5). Fraser (1976) emphasizes that "the

particle MUST move to the position following the direct object when the latter is a

pronoun; otherwise the particle movement is optional" (p. 16) [Emphasis in original].

Likewise, Chen (1986) indicates that "[i]f the direct object of the phrasal verb is a

personal pronoun, particle movement is obligatory" (p. 80).

The word order of IEPVs is summarized in Table 2.3 below:

Type Combination Separable Non-separable


1 Verb + adverb They closed their shop She will not give in.
down.
2 Verb + preposition - He looks after his parents.
3 Verb + adverb + I will get the price I'm looking forward to
preposition down on. meeting you.
Table 2.3 The word order of EPVs.
47

Beate Hampe (1997) indicates that "[t]here is a general agreement in the literature

that, in principle, the two types of multi-word verbs [PVs and prepositional verbs] can

be separated from each other, although […] this distinction is not always completely

clearcut" (p. 204). He, further, gives two reasons for unclearcut:

(i) There is considerable overlap between the two categories 'adverbial


particle' and 'preposition' […].
(ii) Some of the differences in syntactic behaviour disappear as soon as
lexicalization and idiomatization processes have set in and make the
separation of either adverbial particles or prepositional phrases from
the preceding verbs undesirable (p. 204) [Emphasis in original].

Yet, Fraser (1976) argues that "[i]t is this property of being able to appear after the

direct object which we claim to be sufficient to distinguish a verb-particle from a

verb-preposition combination" (p. 2).

Such a word order alteration has been termed as 'particle movement' (Fraser, 1976;

Chen, 1986), and 'particle placement' (Gries, 2002). It has been largely studied over

the last 100 years or so for two reasons:

i) to provide an adequate structural description of the two


possible constituent orders and
ii) to find the variables that determine native speakers' choices
governing the alteration (Gries 2002, p. 269).

Obviously, in spite of several decades of research in this matter, as Gries (2002)

indicates, "there is still no account of particle placement that tries to explain why

speakers choose one construction over the other in a particular discourse situation" (p.

272). The available literature does not tell us "which of the two word orders is more

common or acceptable with which degree of idiomaticity of the verb phrase" (Gries,

2002, p. 277). However, Gries (2002) believes that construction in which the particle
48

is positioned after the direct object is "the natural choice for a speaker who intends to

communicate a state of affairs where the spatial meaning is prominent" (p. 277).

To conclude this section, the following syntactic features can be deduced from the

abovementioned literature, which would help sketch a syntactic prototype image for

EPVs against which their Arabic counterparts may be contrasted:

1. There are certain types of verbs, mostly monosyllabic, that can be

phrasalised. The grammarians have listed them under different names.

2. There are a number of words that may function as verbs only when

combining with particles. On the other hand, there are a number of words,

other than adverbs and prepositions, that may function as particles.

3. Although there are some verbs that may co-occur with every particle, there

are other verbs which co-occur with one particle only.

4. Particles of EPVs are either adverbial or prepositional. The former is of

two types, namely: directional or locative. Bearing in mind that the

majority of particles used in EPVs are locative.

5. Particle in EPVs constitutes an integral part. It typically fused with the

verb with which it combines to form a combination of one semantic unit.

6. Particles of EPVs are morphologically invariable.

7. There are many words that may function as both adverbs and prepositions.

8. Adverb and preposition are identical in form, but they differ in function.

The former typically modifies the verb, while the latter connects what it

comes before them and what it comes after them in any given sentence.
49

9. There are three types of verb-particle combinations (verb + adverb, verb +

preposition, and verb + adverb + preposition) each of which can be

transitive and intransitive.

10. The transitive EPVs may become separable or non-separable depending

upon the particle position whether it is before or after the direct object.

11. When the direct object is pronoun the particle must follow it.

In the following section this syntactic prototype image of EPVs will be completed by

adding to it its integral half, i.e. the semantic properties of EPVs.

2.5 Semantic properties of EPVs

As it has been already mentioned, the phenomenon of PVs is a contentious one and

there are numerous problems regarding their syntactic and semantic explanation. Such

problems, according to Beate Hampe (1997), take place due to the fact that:

This category [PVs] is not a homogeneous group with all its members showing
similar semantic and syntactic properties. Phrasal verbs are 'situated' at the
border of syntax, lexicology and morphology – i.e. particular constructions
possess to varying degrees some of the properties of free syntagms, of
phraseological units as well as derivative word formations (pp. 207-208).

In the present section I will be concerned with the semantic properties of EPVs. As in

the previous section, the aim here is to sketch a clear picture of the phenomenon of

EPVs that allows me to contrast it with its Arabic counterpart, to identify the gaps that

cause Arabic professional translators and Arabic translation students a range of

difficulties when dealing with EPVs.


50

2.5.1 EPVs' criteria

To differentiate the proper PV combinations from the prepositional verb combinations

(verb + preposition), and "to narrow the class of PVs" (Azzaro, 1992, p. 44), scholars

have set out a number of criteria. Cowie (1993) considers that these criteria are related

to meaning rather than to grammar, whereas Dixon (1982) adopts the opposite view

and describes these criteria as "[a] number of explicit non-semantic criteria [which]

have been suggested to distinguish phrasal verbs from literal verb-preposition

combinations" (p. 3). Such criteria are outlined in what follows:

Substitutability by a single word (Live, 1965; Dixon, 1982; Cowie, 1993). For

instance: put off = postpone, look into = investigate, run away = escape, speed up =

accelerate, etc. This criterion was labeled by Bolinger (1971) as "replaceability by a

simple verb" (p. 6), and was described as "[t]he most general of all" (p. 6). Bolinger's

label has been adopted by many researchers such as Azzaro (1992, p. 44),

Lindstormberg (1998, p. 248), and Darwin and Gray (1999, p. 71).

Passivisation (Bolinger, 1971; Dixon, 1982; Azzaro, 1992; Darwin & Gray, 1999).

Transitive PVs can be passivized, to use Dixon's (1982) illustrative examples:

I was taken in (by his smooth talk).


We were put up for the night (by John and Mary) (p.7) [Emphasis and
bracketing in original].

Nominalisation (Fraser, 1976; Lindner, 1983; Azzaro, 1992). This criterion has been

termed by Darwin and Gray (1999) "formation of action nominals" (p. 71). It was

proposed by Lees in 1963 (Bolinger, 1971; Darwin & Gray, 1999). Bolinger (1971)

indicates that "[i]f transitive, the combination should yield an action nominal" (p. 8).
51

For example, "from he brought up the fact, one derives his bringing up of the fact"

(Darwin & Gray, 1999, p. 72) [Emphasis in original].

Gapping (Fraser, 1976; Dixon, 1982; Lindner, 1983), or "Double distribution"

(Azzaro, 1992, p.45), or "object movement" (Darwin & Gray, 1999, p.72). As it has

been explained (see 2.4.3 above), it means that the particle, in transitive combination,

can either precede the noun object or follow it (Bolinger, 1971, p. 10). For example:

He looked up his friend.


He looked his friend up (Bolinger's examples) [My emphasis].

Pronoun placement. In transitive combinations direct-object pronouns typically

precede the particle (Bolinger, 1971; Lindner, 1983; Darwin & Gray, 1999). For

example:

How did you find that out?


*How did you find out that? (Bolinger, 1971, p.11) [My emphasis].

Adverbial position (Mitchell, 1958), or "Adverb insertion" (Dixon 1982, p. 7;

Darwin & Gray, 1999, p. 73). This criterion also has been termed as "Insertion of

adverbial phrases" (Lindner, 1983, p. 12) and "adverb distribution" (Azzaro, 1992, p.

45). It means that Adverb cannot be placed between the verb and the particle, whether

the combination is transitive or intransitive, unless the latter is used in its literal

meaning (Bolinger, 1971). To use Mitchell's examples:

He turned suddenly off the road. [Literal]


*He turned suddenly off the light. [Non-literal] (pp. 11-12) [My emphasis and
bracketing].

Stress (Mitchell, 1958; Fraser, 1976; Bolinger, 1971; Lindner, 1983; Darwin & Gray,

1999). A particle of EPV can be stressed or, to use Bolinger's (1971, pp. 13-14) term,

"accented". For example:


52

He can't be taken IN at any price. [Stressed adverb]


It can't be taken in large doses. [Unstressed preposition]
(Mitchell, 1958, p. 104) [My emphasis].

Definite noun phrases (Bolinger, 1971; Azzaro, 1992; Darwin & Gray, 1999). "[i]f

the combination is transitive, the particle can precede a simple definite noun phrase (a

proper name or the plus a common noun) without taking it as its object" (Bolinger,

1971, p. 15) [Emphasis in original]. For examples:

I'm afraid to take on John in this contest.


You left out the caption.
Did you bring along the Joneses? (p. 15) [Emphasis in original].

Listing. (Bolinger, 1971, p. 17; Azzaro, 1992, p. 45; Darwin & Gray, 1999, p. 74).

This is not a criterion, it is rather a point suggested by Bolinger (1971) to define PVs.

He points out that "[p]hrasal verbs can be defined by simply listing them [the PVs]"

(p. 17). Nevertheless, he admits that the list of PVs will never be exhaustive for two

reasons. The first is that PVs are repeatedly being added to. The second is that they

are varying according to dialect (p. 17).

Nevertheless, it is quite vital to point out that these criteria have been subjected to

many criticisms even by the same abovementioned scholars who list them, illustrate

them and elaborate on their practicality. Dixon (1982), for instance, indicates that

"there are no clear-cut CRITERIA for distinguishing phrasal verbs from literal verb-

preposition constructions" (p. 9) [Emphasis in original]. In addition, he heavily

criticizes the criteria "Substitutability", "Passivisation" and "Adverb insertion" (pp. 4-

9). Bolinger (1971) follows suit and criticizes the criteria "Replaceability" and

"Passiveness" (pp. 6-7). Lindner (1983) considers that "these criteria are best thought
53

of as characteristic tendencies of each construction rather than exceptionalness

defining features" (p. 4).

This having been said, Trask (2000) neatly addresses the issue of differentiating EPVs

from prepositional verbs, or as he puts it "a sequence of a verb and a preposition" (pp.

101-102) by putting the differences in four points as follows, where call up is a PV,

whereas call on is merely a verb plus preposition:

1. The particle in a phrasal verb is stressed: They called up the teacher, but not
*They call on the teacher.
2. The particle of a phrasal verb can be moved to the end: They called the
teacher up, but not *They called the teacher on.
3. The simple verb of a phrasal verb may not be separated from its particle by
an adverb: *They called early up the teacher is not good, but The called early
on the teacher is fine.
4. The particle of a phrasal verb may not undergo PIED-PIPING1: *The
teacher up whom they called is no good, but The teacher on whom they called
is fine (pp. 101-102) [Emphasis in original].

Furthermore, Thomson and Martinet (1986) rationally maintain that one "need not to

try to decide whether the combination is verb + preposition or verb + adverb, but

should consider the expression as a whole" (p. 315).

2.5.2 Semantic categories of EPVs

In addition to classifying PVs on the ground of syntax, as it has been outlined in the

last section, English grammarians and linguists have classified them on the ground of

meaning, in view of the fact that they "vary in the extent to which the combination

preserves the individual meanings of verb and particle" (Quirk & Greenbaum, 1973,

p. 348). Palmer (1968), in this respect, indicates that

1
PIED-PIPING means moving a preposition to the front of its clause and replace it before its object,
e.g., To whom were you speaking? With what did they hit it? (Trask, 2000, p. 102)
54

Two Kinds of classification are appropriate. The forms may be distinguished


grammatically, first, in terms of the adverbial versus the prepositional nature
of the particles. Secondly, they may be distinguished in terms of idiom, some
of the combinations being idiomatic, others not. The two types of
classification must be kept distinct. It is often not clear whether the term
'phrasal verb' is defined in terms of the first, the second, or both (pp. 180-181).

Due to the fact that meaning of PVs ranges from literal to idiomatic (Gries, 2002), and

their semantic types, as Chen (1986) puts it, "vary from the most literal to the most

idiomatic" (p. 82), they can be semantically classified as follows:

1.Non-idiomatic, literal PVs, where both components of the construction retain

their individual lexical meanings, for example:

Bring the box in.

Take it out.

Put your hand up.

2. Semi-idiomatic PVs, where one component of the construction retains its lexical

meaning while the other one is less transparent, for example:

Drink your milk up.

Knock him out.

I will find out the truth.

3. Idiomatic PVs, where both components of the construction are not transparent,

and the whole meaning of the idiomatic PV cannot be gained from the total sum

of the lexical meanings of its individual parts, for example:


55

The market is closed due to the black out.

I will never give up.

He cannot put up with him anymore.

(cf. Quirk et al., 1985; Cowie, 1993; Heliel, 1994; Lindstormberg, 1998; Darwin &

Gray, 1999).

The semantic categories of EPVs are outlined in Table 2.4 below:

Type Semantic category Example


1 Non-idiomatic Adam has come back.
2 Semi-idiomatic She wrapped up her luggage.
3 Idiomatic The war broke out suddenly.
Table 2.4 Semantic categories of EPVs

Idiomatic and non-idiomatic PVs have been given many names by different scholars.

McArthur (1975), for instance terms them "literal and figurative" (p. 36), while

Sawyer (1999) calls them "Compositional (Literal) and Non-compositional

(Metaphorical)" (pp. 5-6).

By the same token, Dixon (1982) summarizes the variety of cases in which PVs may

occur, as follows:

In some cases the meaning of the phrasal verbs is similar to that of the
constituent simple verb, it being the preposition that is used in a non-norm
manner: thus eat up is plainly related to eat and slow down to slow, but in
these combinations up and down clearly do not refer to vertical displacement
away from or towards the centre of the earth […] In other cases the
preposition appears in its normal meaning but the verb takes on a non-central
sense, e.g., knock about/around (the world) […] Finally, the meaning of many
phrasal verbs does not relate directly to the normal meaning of either simple
verb or prepositions – examples include take off 'imitate' and put up with
'tolerate' (p. 1) [Emphasis in original].

Like PVs, prepositional verbs, according to Quirk and Greenbaum (1973),


56

[…] vary in their idiomaticity. Highly idiomatic combinations include go into


(a problem) 'investigate', come by (the book), 'obtain'. [In addition] Like
phrasal and prepositional verbs, [… phrasal-prepositional verbs] vary in their
idiomaticity. Some, like stay away from ('avoid'), are easily understood from
their individual elements, though often with figurative meaning, e.g: stand up
for ('support'). Others are fused combinations, and it is difficult or impossible
to assign meaning to any of the parts, e.g: put up with ('tolerate')" (p. 350)
[Emphasis in original].

Such a semantic classification of PVs, however, is arguable as the distinction between

these categories is "not always clear and the same combination may fall into more

than one category depending on the context in which it is used" (Swierzbin, 1996, p.

3). As a result, a close look at the available literature reveals that scholars have varied

in their approaches of covering the phenomenon of PVs as what to include/exclude in

their coverage. Some of them, (such as: Fraser, 1976; Dixon, 1982; Courtney, 1983;

Cowie, 1993), have limited their studies to the idiomatic type of PVs excluding the

other semantic categories. They observe that PVs are metaphorical/idiomatic. Dixon

(1982), for instance, prefers "to retain the term 'phrasal verb' for any combination of

verb and preposition(s) that does not have a literal meaning" (p. 2). Similarly, Cowie

(1993) emphasizes that he "follow[s] fairly common practice in applying the term

[PV] to idiomatic combinations, whether of verb + adverb or verb + preposition" (p.

39). On the contrary, other scholars have widened the scope of their research to

include all the semantic categories, such as Lindstormberg (1998) who goes for the

above classification, claiming that it is the correct view since "some literal verb-plus-

particle combinations tend to be stored in memory much like single verbs"( p. 143).

Along these lines, in her doctoral dissertation A lexico-Semantic Analysis of English

Verb Particle Constructios, Lindner (1983) classifies the scholars who dealt with the

phenomenon according to their ways of treatment to five different groups: 1) Scholars


57

who concentrate on both idiomatic and literal combinations of verb + adverb, and

exclude the verb + preposition combinations, such as (all of the following scholars are

cited in Lindner, 1983): Declerck (1976), Mitchell (1958), Fairclough (1965),

Bolinger (1971), Lipka (1972), Meyer (1971), Druisinga (1952), Wood (1955), Dogen

(1919), Draat (1921) and Lindner (1983); 2) Scholars who focus on both verb +

adverb combinations and "certain path" of verb + preposition combinations, such as:

Poutsma (1926), Taha (1964) and Jespersen (1961); 3) Scholars who "include the full

range of combinations", i.e. the literal and idiomatic types of both verb + adverb

combinations and verb + preposition combinations, such as: Kennedy (1967), Konish

(1958), Roberts (1958) and Sroka (1972); 4) Scholars who limit their studies to the

idiomatic type of the verb + adverb combinations, and exclude the verb + preposition

combinations, and the literal type of verb + adverb combinations, such as: Fraser

(1976) and Legum (1968); and 5) Scholars who cover verb + adverb combinations

and verb + preposition combinations "as long as both are idiomatic", such as: Jowtell

(1951), Potter (1965), Smith (1925), Live (1965) and Sweet (1955) (cf. Lindner, 1983,

pp. 2-4).

I have to make it clear here that for the purpose of this study I will adopt the view of

the scholars in number five above, (i.e. Jowtell, 1951; Potter, 1965; Smith, 1925;

Live, 1965; Sweet, 1955) and the view of Courtney (1983), Cowie (1993), and Cowie

and Mackin (1993) who consider PVs as combinations of a verb and an adverb, or a

verb and a preposition (or a verb with both an adverb and a preposition) as far as they

are idiomatic. Hence the idiomaticity will be my parameter in tracing such English

combinations and contrasting them with their Arabic counterparts. The reason for

such a focus is the fact that it is the idiomatic type of PVs which causes difficulties to
58

Arabic professional translators and Arabic translation students, rather than the non-

idiomatic/literal type, which has a straightforward meaning in Arabic (see Chapter

Five below).

Dixon (1982) asserts that the non-literal PVs "have essentially the same syntax as

literal co-occurrences of verb and preposition; the difference is almost entirely

semantic" (p. 2). Therefore PVs "are idiosyncratic only at the semantic level; their

syntactic behaviour can be predicted from general grammatical statement and rules

that apply to all combinations of verb and preposition, both literal and non-literal" (p.

2).

It needs to be noted though that the combinations of idiomatic PVs are not freely

formed; there are, rather, several "collocational restrictions" governing them (Palmer,

1974). In other words, we cannot substitute the particles of PVs for their opposites,

viz. we can look after someone, but we cannot look before someone, and we can put

up with something, but we cannot put down with it (p. 212). By the same token, we

cannot substitute the verbs of PVs for their opposites, viz. we can say 'I helped him

out', but we cannot say 'I aided him out', and we can say 'He yielded up all his

property', but we cannot say 'He abandoned up all his property' (p. 226). Palmer

(1968) has argued that "[a]part from their semantic unity and the collocational

restrictions on the occurrence of verb and the particle, there is nothing that will

establish which are phrasal verbs and which are not" (p. 185).

In their Oxford Dictionary of Phrasal Verbs Cowie and Mackin (1993) provide two

tests to distinguish idiomatic from non-idiomatic PVs; both of the tests have to do
59

with meaning rather than with syntax. The first test being "to ask whether one word

can be substituted for the whole phrase", and the second being "to ask whether the

second word can be deleted". For example, to know whether the PVs fall out, as used

in I was pleased with the way things had fallen out, an idiom or not, you need to apply

one of the tests. In fact the answer of the first one is: yes, we can substitute fall out for

one word, that is, 'happen' or 'occur', and the answer of the second test is: no, we

cannot delete the second part of fall out, as its form is fixed and cannot be broken up,

it is an idiomatic PVs (p. ix).

Moreover, the major function of PVs, following Yatskovich (1998), is "conceptual

categorization of reality in the speaker's mind" (p. 1). Hence, in addition to denoting

actions or states as "ordinary" verbs do, they specify their spatial, temporal or other

characteristics. The adverbial components impart PVs such ability to describe actions

or states more precisely, vividly and emotionally. "[b]y combining with these

elements, verbs of broader meaning are subjected to a regular and systematic

multiplication of their semantic functions" (pp. 1-2). Thus, adverbial particles'

function falls in two major types: 1) to pass on an additional aspective meaning to the

base verb, e.g., sit down; 2) to introduce a lexical modification to the verb's

fundamental semantics (p. 2). That is to say, PVs may have, according to McArthur

(1989):

(1) any of the meanings of the verb plus any of the meanings of the particle,
(2) any meanings that emerge from such a union for particular purposes in
particular contexts, and (3) the capacity to drift from 1 to 2 and back again, a
literal use carrying a figurative nuance and vice versa, especially in jokes (p.
40).

McArthur (1989) selects the PV get up to illustrate the range of possibilities as

follows:
60

This phrasal verb is intransitive in 'They got up', transitive in 'Get them up',
means from lower to higher in 'He got the child up on to the wall', means from
far to near in 'One of the other runners got up to him and passed him', means
accumulate under pressure in 'The engine got up steam', organize or make in
'He can get up the plot of a new film in no time at all', and put on the special
clothes in 'They got themselves up as pirates' (p. 40) [Emphasis in original].

The particle up, on the other hand, has a variety of meanings when it combines with

different verbs, to use McArthur's (1989) instances:

The smoke rose up (upward direction).


He swam up to the boat (approaching direction).
They used up all the oil (completion in the sense that nothing is left).
They tidied the room up (completion in the sense that something is done as
fully as possible).
Hurry up! (literal emphasis).
Shut up! (metaphoric emphasis).
Drink up (completive and emphatic) (p. 40) [Emphasis and bracketing in
original].

Furthermore, Lindstromberg (1998) sketches a class of PVs called "perfective phrasal

verbs" or perfectives. PVs of this class are "extremely common expressions each

consist of a main verb plus up, down, out, off, or through (e.g., wash up, calm down,

chill out, cool off, think through)" (p. 23) [Emphasis in original]. Such perfective PVs

"all have to do with the notions of completeness or thoroughness" (p. 24). For

example:

Type this letter up, please.


They've close down their business.
Mammoths died out a long time ago.
Let's finish this job off and go home.
I've read it through. Now let's talk about it.
(Lindstormberg's examples) [Emphasis in original].

The meanings of particles and the semantic contributions they make when combining

with verbs to produce PVs have been taken up by such researchers as Sinclair et al.

(1998) and Hannan (1998). They will be explored in more detail in the forthcoming

sections of this Chapter.


61

A few remarks need to be stated regarding the issue of polysemy, which means using

one PV in many different meanings. Some EPVs are said to be "highly polysemous"

(Villavicencio & Copestake, 2003, p. 358). Kennedy (1967) indicates that

[…] owing to the multiplication of meanings, the possibility of confusion or of


misunderstanding is greatly increased. A verb-combination that is capable of
fifteen different uses or phrases of meaning has become more or less a 'Jack of
all trades' and its capacity for good work on a given job is open to suspicion.
Even though it is not likely that one would be long in doubt as to the meaning
of blow up 'to inflate', 'to explode', 'to stop', 'to praise extravagantly' […] yet it
is not conductive to linguistic thrift but rather to a certain amount of verbosity,
if the phrase is not self-explanatory. If one holds up a person he may be doing
him a favor or a great wrong; if two actors make up they may resume a former
friendship or merely their places on the stage; if an object is all stuck up it may
be inclined to adhere or […] it may have been treated much worse (p. 43)
[Emphasis in original].

Moreover, a PV "may be polysemic in having both an idiomatic and a non-idiomatic

use [and] it may well be polysemic in having more than one idiomatic use" (Cornell,

1985, p. 270).

Live (1965) maintains that "[c]onfusion is further compounded by the obscuring of

the original metaphor; therefore non-native speakers may find these verbs

troublesome" (p. 430).

More detailed instances of this issue will be given in the forthcoming section of 'PVs

in English lexicography' as the phenomenon has been touched upon by many

lexicographers (Taha, 1972; McArthur & Atkins, 1974; Kirkpatrick & Davidson,

1982).
62

2.5.3 Cohesion of PVs' elements

Relevant to the purpose of this study is DeArmond's (1977) definition of cohesion.

Cohesion, according to him, "refers to the fact that certain verbs and a following

preposition receive a semantic interpretation which cannot be subdivided into two

semantic units, where one unit refers to the verb and the other to the preposition" (p.

20). Along these lines, Bolinger (1971) makes the point that PVs "show some special

degree of cohesion that sets them apart from the more freely composable

constructions like to fall headlong, to live at home, or to leave tomorrow" (pp. 3-4).

In the main, the closeness of the verb and the particle in PVs has been outlined by

Kennedy (1967, p. 9). He indicates that in such combinations as bear out

(corroborate), come by (acquire), get at (reach), hit it off (agree, be congenial), make

out (understand), own up (confess), put out (extinguish), stack up (fare), whack up

(share or divide), "the elements of the combination have almost or altogether

sacrificed their individual meanings and by the act of combination have assumed a

new meaning" (p. 9). While in such combinations as bake up (make a batch of),

blossom out (blossom in a showy manner), blot out (destroy), bottle up (enclose in a

bottle), button up (fasten with buttons), "the verb is modified in meaning by a certain

weakly adverbial function of the particle but does not entirely merge its verbal

personality in the combination [and the particle] loses much of its usual adverbial or

prepositional signification" (p. 9). Whereas in such combinations as brush off, brush

out, bubble over, burn down, cave in, fall down, flame up, hang up, leak out, rinse

out, and tack down "the usual value of verb and prepositional-adverb remain fairly

evident" (p. 9).


63

Further, Live (1965) provides evidence of the close cohesion of the verb and the

particle in these combinations, indicating that

[…] the two elements function as one verb is further demonstrated by the fact
that the combination readily occurs in conjunctional parallel with a single
verb, having a common object or joint membership in a series (I sent for and
received the goods. He was never heard from nor seen again. He was adopted,
cared for, brought up, and educated by the couple) (pp. 428-429) [Emphasis
in original].

Live (1965) provides another evidence of the close cohesion of the components of

PVs being:

[…] the juxtaposition of its two elements in other morphological


configurations, with the particle prefixed or suffixed: upstanding, ongoing,
incoming, outstretched, outspoken, downtrodden, broken-down (ruin), put-up
(job), paid-up (bills), sought-for (result), worn-out (clothing), shut-in (invalid)
(p. 429) [Emphasis in original].

2.5.4 British and American dialects

It is quite important to mention that EPVs differ according to the English dialects

(Bolinger, 1971; Dixon, 1982; McArthur, 1989; Lindstormberg, 1998; Villavicencio

& Copestake, 2003). This is particularly evident in the British and American dialects.

Palmer (1968) points out that "[a] rather interesting contrast across the Atlantic is the

use in Britain of FILL IN, and in America of FILL OUT. In Britain we fill in a form,

in America a form is filled out" (p. 187) [Emphasis in original]. McArthur (1989)

provides some more instances illustrating such a phenomenon. He indicates that the

particle on to is used as such in British English, whereas it is onto in American

English. The particle off of, which is typically regarded as standard in American

English, is considered non-standard in British English, The particle out, which is

typically followed by of in the standard usage of England as in: They looked out of the
64

window, need not be in Scotland and North American English as in: They looked out

the window. There is also some difference in the use of the particles about and

(a)round as Britons prefer to use the former as in: running about, while the North

Americans prefer to use the latter instead as in: running around. The two dialects

differ as well in terms of hyphenating nouns derived from PVs, such as breakdown

which commonly written solid in British English and hyphenated (break-down) in

North American English (McArthur, 1989, pp. 39-41).

Yet, Lindstormberg (1998) calls attention to the fact that such a difference is very

little and the vast majority of PVs have the same form and the same meaning in both

dialects. This seems, according to him, "to suggest that these meanings have derived

from common meanings of verbs and prepositions in non-idiomatic fashion" (p. 243).

Whereas Bolinger (1971) considers this phenomenon as one of the two reasons that

make the task of listing PVs difficult (see 2.5.1 above).

It should be noted in this connection that such differences in the English dialects

might become more serious when it comes to such sentences as knock me up which

means in British dialect (knock on my door) whereas when it is said by a woman in

America she may run the risk of being raped (Najiib, 2001, p. 37).

Incidentally, as we will see in section 2.8 of this Chapter, such an issue has been

given a considerable attention by lexicographers in such dictionaries as: Cambridge

international Dictionary of Phrasal verbs (1997); Oxford Phrasal verbs Dictionary

for Learners of English (2001); Longman Dictionary of Phrasal Verbs (1983); NTC's
65

Dictionary of Phrasal Verbs and Other Idiomatic Verbal Phrases (1993) among

others.

2.5.5 Register variations of EPVs

Although EPVs are commonly used informally in everyday spoken English (McArthur,

1975; Kennedy, 1967; Cornell, 1985; Sinclair et al., 1998; Turton & Manser, 1985;

McArthur, 1989; Villavicencio & Copestake, 2003), they are quite often used formally in

a variety of English written texts (Cowie & Mackin, 1993; Swierzbin, 1996). Thus,

Goodale (1994) maintains that "[i]t is misconception that phrasal verbs are mostly used in

spoken language. They can be found in many styles of writing, including highly formal

government reports" (p. iv). To take only a two examples of using them formally:

(i) Eight Iraqis were killed and 23 wounded when fresh clashes broke out

between US forces and the militia (The Sydney Morning Herald, June 3,

2004, p.11) [My emphasis].

(ii) Lexicographers need to have had some training in lexicography, if they are

to come up with good dictionaries (Crystal, 1997, p. 118) [My emphasis].

And to take just a few examples of using them informally:

(i) Let me get this straight - you had a job where you open fetes, cut ribbons,

sing the Anthem, drink champers, ride in the Rolls, get paid - and you stuffed

it up (Cartoon speech bubble, The Daily Telegraph, August 12, 2004, p. 41)

[My emphasis].
66

(ii) Greatly enfeebled by the strain of chronic scandal. The leader seems to be

slipping away, on long even responsive to criticism (Cartoon speech bubble,

The Australian, May 10, 2005, p. 22) [My emphasis].

(iii) Leave out the technical stuff because it will only confuse him (Cartoon

speech bubble, The Australian, May 10, 2005, p. 2) [My emphasis].

English speaking children, as McArthur (1975, p. 6) asserts, "learn them [EPVs]

before they learn other kinds of verbs: Get up! Go away! Drink up your milk! Put

your toys away darling! Shut up!" [Emphasis in original].

Kennedy (1967, p. 10) makes the point that "[…] correct usage [of PVs] is such an

intangible and varying thing that it is practically impossible to determine the social

status of each combination and usage" (p. 10). Some combinations, he elaborates,

would be accepted by all people such as ask for, bow down, cry out, go on, make off,

and point out, while "in some cases, very careful speakers might prefer to employ

single words of more highly specialized meaning, such as request, bow or keen,

exclaim, continue, depart, demonstrate" (p. 10) [Emphasis in original]. Other

combinations might be "justified by the technical or specialized use to which they are

generally put. So we can say call up by telephone, connect up with the assistance of

plumber or electrician, kick off at the beginning of a game, lay by corn at the last

plowing, make up for the stage" (p. 10) [Emphasis in original].

Nonetheless, Heliel (1994) makes the point that "[a]s the changing attitudes of users

of English toward levels of usage become more flexible, the combinations which were
67

once labeled 'slang', 'informal', or 'colloquial' are now considered 'neutral' and thus

have risen in social importance" (p. 142).

Furthermore, Turton and Manser (1985, p. viii) advise non-native speakers of English

to treat PVs with caution since many of them "could cause offence if used

inappropriately". Likewise, Sinclair et al. (1998) ask foreigners to be careful when

using the single-word equivalents of PVs as:

[…] in many cases phrasal verbs and their synonyms have different range of
use, meaning, or collocation, so that a single-word synonym cannot be
substituted appropriately for a phrasal verb. Single-word synonyms are often
much more formal in style than phrasal verbs, so that they seem out of place in
many contexts, and students using them run the risk of sounding pompous or
just unnatural (p. iv).

Similarly, Side (1990) indicates that "direct equivalents of phrasal verbs do not always

exist.'I'm done in' would be used in a different social context from 'I'm exhausted'" (p.

145) [Emphasis in original]. And "show off is show off, not to impress another with

one's prowess by preforming difficult yet completely unnecessary feats" (Darwin &

Gray, 1999, p. 66) [Emphasis in original].

Consequently, Cowie and Mackin (1993) make the point that PVs need to be

cautiously used not only in their correct grammatical patterns but also in their

appropriate contexts (p. xi).

Along these lines, Cornell (1985) indicates that quite a few of one-word or PV

equivalents can be deemed as alternatives for their PV. It all has to do with "the

degree of synonymity" since "synonymy is generally recognized as being a very

relative concept" (p. 274). To use some of Cornell's examples:


68

lie in does not merely mean "to stay in bed", but "to stay in bed beyond
one's normal time for getting up".
put up with: unlike tolerate it cannot be used in a positive manner (to
tolerate other people's opinions is not the same as to put with other
people's opinions).
run down is never constructive (unlike criticize) (pp. 274-275) [Emphasis
in original].

Kennedy (1967, p. 41) observes that although the combination of a verb and a particle

"may not be figurative, it is often more expressive than the simple loan-word. Blow

out, for instance, tells the average person more of the method employed than does

extinguish".

As we will notice in section 2.8 of this Chapter, most of the English lexicographers

have included the register variations of PVs in their dictionaries (cf. Cambridge

international Dictionary of Phrasal verbs (1997); Oxford Dictionary of phrasal verbs

(1993); Oxford Phrasal verbs Dictionary for Learners of English (2001); Longman

Dictionary of Phrasal Verbs (1983)) among others.

To sum up this section, the phenomenon of EPVs is, to a certain degree, a

controversial one. The literature at hand shows that a great deal of disagreements have

been arisen amongst the scholars who tackled the phenomenon. Such disagreements

are manifested in their contradictory views regarding such issues as the set of criteria

of phrasal verbness, what to include/exclude as PVs, semantic categories, and degree

of idiomaticity of each component of PVs. However, one can extract the following

semantic sketch as a prototype model of EPVs:

1. EPVs are of three semantic categories, namely: literal, semi-idiomatic, and

idiomatic PVs.

2. The meaning of EPV ranges from the most literal to the most idiomatic.
69

3. Both the literal and the idiomatic types have the same syntactic

characteristics. The distinction between them, however, is confined to their

semantic properties.

4. Unlike the non-idiomatic type, the idiomatic type of PVs has the virtue of

substitutability by one word synonym.

5. The combination of the components (verb and adverb) is not freely

formed. It is rather subjected to many collocational restrictions.

6. The particle modifies, to a greater or lesser degree, the meaning of the verb

it combines with.

7. There is a close cohesion between the verb and the particle that makes

them fuse together and sacrifice their individual meanings to produce one

semantic unit.

8. One EPV may carry more than one meaning. In other words, its meaning

may vary according to the contexts it is used in.

9. There are some varieties amongst English dialects with respect to using,

spelling, and hyphenating PVs.

10. Even though EPVs are typically used in an informal English, they are quite

often employed in more formal register.

It goes without saying that such a complicated picture of intricate semantic properties

of IEPVs has a great deal of impact on the process of learning and understanding

them by Arab English learners, and in turn on translating/interpreting them by Arabic

translators and interpreters.


70

2.6 PVs in Arabic

Having known the syntactic and semantic properties of EPVs, in this section I will

move on to investigate whether the phenomenon of PVs exist in Arabic. And if so, do

they fit the previously established syntactic and semantic criteria of IEPVs?

It is worth noting, at the beginning, that Arab scholars who dealt with the

phenomenon of EPVs have given them various labels in Arabic. Heleil (2000), for

instance, calls them 5&ّ1‫ر‬7(*+‫ل ا‬7&*89‫( ا‬literally: phrasal verbs), and Daud et al. (2003)

name EPV %&)*8 |&-‫آ‬,} (literally: verbal compound), whereas Najiib (2001) considers

EPVs as one part of 5&)ّ1~3+‫ل ا‬7&*89‫( ا‬literally: tailed verbs) which consist in addition to

PVs, prepositional verbs and phrasal-prepositional verbs.

As far as the Arabic language is concerned, there exist many constructions in which

verbs are followed by prepositions such as: 0/ K‫•آ‬1 (to eat from), p+‫ إ‬,€.1 (to look at),

0&E ‫& ّ[ث‬QF1 (to talk about), %&8 ,&ّ'_1 (to think of), p&)E ‚&ّ)*1 (to comment on), ‫ب‬
ِ |&F'1 (to

write with) etc. These constructions are fairly common, frequent and widely used in

both written and spoken Arabic. However, unlike English grammarians, Arabic

grammarians do not group such constructions under a specific heading. Therefore, the

main question here is can we consider them as PVs? And to what extent do they meet

the abovementioned syntactic and semantic criteria of IEPVs?

It is important to say that while there are some researchers (e.g., Kharama & Hajjaj

1989; Bataineh & Bataineh 2002) who have touched upon the issue of the existence of

PVs in Arabic without elaborating on their form and function criteria, there are, on the
71

other hand, quite a few who have given such an issue more attention and tried to

explore it in some more detail. In what follows, I will outline their contributions one

by one in order to sketch a clear picture of such constructions that will enable me to

contrast them with their English counterparts.

2.6.1 Lentzner

The leading study in this regard was conducted by Lentzner (1977) in his doctoral

dissertation entitled Semantic and Syntactic Aspects of Arabic Prepositions in which

he assigns a chapter to explore the verb-preposition structures in Arabic (pp. 155-

195). Given the importance of the profound insights considered in such a chapter, it

merits being summarized.

In his attempt to examine the relationships between verbs and prepositions in Arabic,

Lentzner claims, from the very beginning, that only the "true" Arabic prepositions

,&I+‫وف ا‬,&Z "are able to combine semantically with verb roots and to act as integral

parts of verb-preposition structures" (p. 19). By "true" Arabic prepositions, he means

prepositions such as 0&E ,0&/ ,p&+‫ إ‬,‫م‬2&+‫ ا‬,p&)E ,%&8 ,‫ء‬7&(+‫( ا‬baa?, fii, 3alaa, laam, ?ilaa, min,

3an) "which embody the most essential locative and directional notions" (p. 159).

Lentzner makes the point that such constructions in Arabic "exhibit characteristics

similar to both compound verbs [PVs] and tight verb-preposition constructions in

English" (p. 159). Further, he admits that it is difficult to compare such Arabic

structures with those in English mainly, due to the fact that unlike English

prepositions, Arabic prepositions always require an object of some sort and cannot

occur without one. This makes some testing procedures in English inapplicable to
72

Arabic. He also outlines the past efforts done by scholars, such as Wright, Cantarino,

Fleisch and others, at establishing systematicity between the classes of Arabic verbs

and the prepositions they combine with. He believes that such efforts were

unsuccessful, claiming that most attempts at categorization of Arabic verb-preposition

idioms "have tried to group verbs and prepositions together on the basis of (rather

vague) semantic cohesion" (p. 160), for example [cited from Cantarino], because 0&/

means "the casual point of departure" it is used with "verbs meaning to sell and to

give in marriage" or that because p&)E means 'to be on, over' it is employed with verbs

meaning "to cover, to include" and "domination, power" (p. 160). These analyses in

Lentzner's view:

[…] while often intuitively valid, are nonetheless restricted by two facts: first,
such statements are only generalities and cannot account for all instances of
verb-preposition occurrence. Second, they are not precise enough to be able to
predict which preposition will be used where (pp. 160-161).

He, accordingly, excludes such semantic analyses, and devotes his study to provide

"tentative categories of semantic structure for certain types of verb-preposition

idioms" (p. 161). These idioms have been extracted from a corpus, which is Halim

Barakat's novel entitled ",&Q(+‫ ا‬p&+‫ إ‬,L7&{+‫دة ا‬X&E", and categorized by prepositions. He

starts with the locative prepositions: p&)E ,%&8 ,‫ء‬7&(+‫ ا‬to consider their relations to verbs

they often combine with. He maintains that ‫ء‬7&&(+‫ ا‬and p&&)E "are by far the most

commonly used in conjunction with verbs" and ‫ء‬7&(+‫" ا‬is the most widely ranging

preposition of all" (p. 162).


73

‫ء‬7(+‫( ا‬baa?)

In addition to its function "as an integral part of certain verb-preposition idioms" (p.

162), ‫ء‬7(+‫" ا‬can act as a transitivizing particle which serves to convert intransitive verbs

of motion into transitive verbs of transport" (p. 162). Arabic grammarians refer to

such a function of ‫ء‬7&&(+‫ ا‬as 5&&1[*F+‫ء ا‬7&&\ (the transitivizing baa?) or K&&ƒ.+‫ء ا‬7&&\ (baa? of

transport), e.g., ‫ب‬


ِ K&ّƒ.} (to transport by), ‫ب‬
ِ ‫ب‬,&‫( ه‬to run off with) (p. 162). Moreover,

there are some intransitive verbs which may use ‫ء‬7&(+‫ ا‬in construction with a noun

phrase such as: ‫ب‬


ِ ]&3F„‫( ا‬to meet with), ‫ب‬
ِ ,&nF8‫( إ‬to boast about), ‫ب‬
ِ ,ّ &/ (to pass by) etc.

In such verbs ‫ء‬7(+‫ ا‬has two functions:

1) Specifying the subject, for example:

5-P7…+‫ ا‬5.Y+‫ب ا‬2{\ p+‫و‬9‫ ا‬5.Y+‫ب ا‬2W ]3F„‫ا‬

The first year-students met with the second year-students.

2) Bringing a second element into the predication, for example:

‫}~ة‬7H97\ ‫ب‬2{+‫] ا‬3F„‫ا‬

The students met with the professors.

In both of the above cases, ‫ء‬7(+‫" ا‬serves the purpose of specifying those who took parts

in the action" (pp. 165-168).

Furthermore, there exist some verbs that do not occur without ‫ء‬7&(+‫ ا‬such as: ‫ب‬
ِ 0&/† (to

believe in), ‫ب‬


ِ ‫ّع‬,(} (to donate), ‫ب‬
ِ ‫( „[ُر‬to be worthy of), ‫ب‬
ِ |ّZ‫( ر‬to welcome), ‫ب‬
ِ ‫ف‬,FE‫ا‬

(to confess), ‫ب‬


ِ ‚QF+‫( ا‬to join with), ‫ب‬
ِ SّY3} (to stick to), ‫ب‬
ِ 4ّ F‫( اه‬to be interested in), ‚ˆ‫و‬

‫ب‬
ِ (to trust), ‫ب‬
ِ KU}‫( ا‬to contact) etc. (pp. 170-171).
74

There are, however, some transitive verbs which use ‫ء‬7&(+‫" ا‬to mark noun phrases in

particular syntactic relations" (p. 171), but may occur without it in other contexts such

as: ‫ب‬
ِ 4&)Z (to dream about), ‫ب‬
ِ 4ّ)&H (to accept) and ‫ب‬
ِ ]3&H (to hear about) (p. 171). Also,

there are some verbs which use ‫ء‬7(+‫" ا‬to mark the direct object, but […] can be directly

transitive with no change in meaning" (p. 171) such as: ‫ب‬


ِ K&/‫( أ‬to hope for), ‫ب‬
ِ ‫ط‬7&Z‫( أ‬to

surround), ‫ب‬
ِ ‫( \[أ‬to begin), ‫ب‬
ِ ‰
ّ Z‫( أ‬to feel), ‫ب‬
ِ p/‫( ر‬to throw), ‫ب‬
ِ ‚Q+ (to overtake), ‫م‬ŠF+‫إ‬

‫ب‬
ِ (to pledge) etc. (p. 171). There are other verbs, "which are passive in either form or

meaning" (p. 172), and use ‫ء‬7&(+‫" ا‬to mark an underlying agent or instrument" (p. 172),

such as: ‫ب‬


ِ ~&yُ‫( أ‬to be influenced by), ‫ب‬
ِ |&IEُ‫( أ‬to admire), ‫ب‬
ِ ‹„X&8 (to be surprised at)

etc. (p. 172). Finally, there are some verbs which involve two noun phrases, one of

which is marked by ‫ء‬7&(+‫ ا‬such as: ‫ء‬7&(+‫ا‬+p&)E+ 4&'Z (to condemn someone to something),

‫ء‬7&(+‫ ا‬+ 4&H‫ا‬+ ‫ّف‬,&E (to introduce someone to someone), ‫ء‬7&(+‫ ا‬+ 4&H‫ ا‬+ ]&.w‫( أ‬to convince

someone of something), ‫ء‬7(+‫ ا‬+ 4H‫ ا‬+ [ّ /‫( أ‬to provide someone with something), 4H‫ ا‬+ ,/‫أ‬

‫ء‬7(+‫ ا‬+ (to command someone to do something) etc. (p172).

%8 (fii)

Lentzner claims that the occurrence of %8 with verbs "is not common" (p. 173), and he

pinpoints four different cases in which verbs combine with %8: Firstly, "verbs that take

[%8] optionally; can be directly transitive" (p. 173), e.g., %8 K/‫( أ‬to hope for). Secondly,

"verbs that take [%8] for one meaning, another preposition for other meaning" (p. 173),

e.g., %8 |G‫( ر‬to desire). Thirdly, "verbs that require [%&8]" (p. 173) such as: %&8 4‫ه‬7&H (to

participate in) and %&8 ‫ك‬,F&o‫( ا‬to share). Lastly, "Intransitive verbs that use [%&8] in

construction with a noun phrase" (p. 173), e.g., %&8 ,&ّ'8 (to think of) and %&8 ,ّ 3F&H‫( ا‬to

continue) (p. 173).


75

p)E (3alaa)

Lentzner observes that "[m]any intransitive verbs use [p&)E] in construction with a

noun phrase" (p. 175), such as: p&)E •&H‫( أ‬to regret), p&)E p&'\ (to cry over/for), p&)E ‫ف‬7&y

(to fear for), p&)E ,O&H (to watch over) in which p&)E "denotes a type of casual

relationship between the action depicted by the verb and the object of the preposition"

(p. 177), and embodies a concept of "FOR THE SAKE OF". The second function of

p&)E is illustrated in: p&)E ‫ر‬7&ˆ (to rebel against), p&)E SQ&R (to laugh at), p&)E |^&G (to

become angry of), p&)E |&ّ)Ž} (to overcome) which have to do with actions or feelings

directed "AGAINST" someone or something. Thus, p)E has two meanings (i.e. 'for the

sake of' and 'against') which are considered as "two sides of one coin", and the

difference in these meanings "lies not in the preposition itself, but in the verb with

which it is used" (pp. 176-180) [Emphasis in original].

There are some verbs that require p&)E, such as: p&)E ,&ّˆ‫( أ‬to influence), p&)E ‫ص‬,&Z (to

ensure), p&)E 4&'Z (to pass judgment on), p&)E ‫ل‬


ّ ‫( د‬to indicate), p&)E ,{-&H (to dominate),

p&)E K
ّ &W‫( أ‬to look out upon), p&)E ,U&Fw‫( ا‬to be limited to), p&)E 4&I‫( ه‬to attack) etc. (pp.

180-181). Whereas there are some transitive verbs that "may occur with [p&)E]", such

as: p)E Šّ‫( رآ‬to concentrate on), p&)E ‚ّ)E (to comment on), p&)E [&3FE‫( ا‬to depend on) etc.

(p. 181). Finally, there are some verbs which involve two noun-phrases, one of which

is marked with p)E such as: p)E + 4H‫ ا‬+ [E7H (to help someone to do something), + Kّ^8

p&)E + 4&H‫( ا‬to prefer something to something), p&)E + 4&H‫ ا‬+ p&)/‫( أ‬to dictate something to

someone) etc. (pp. 181-182).


76

Having considered the relations of Arabic locative prepositions to Arabic verbs,

Lentzner goes on to explore the relations of Arabic directional prepositions to Arabic

verbs. He makes the point that, like the locative prepositions, Arabic directional

prepositions ( 0&E ,0&/ ,p&+‫ إ‬,‫م‬2+‫ ) ا‬combine with verb roots to provide "lexical structures

with particular semantic content", however, they are less frequent than either ‫ء‬7&(+‫ ا‬or

p)E (p. 182).

‫م‬2+‫( ا‬laam)

This preposition is interchangeable with p&+‫ إ‬with no change in meaning. It may

combine with four types of verbs as follows: 1) Verbs which require it, such as: pŽ`‫أ‬

‫ل‬
ِ (to listen to), 2) "Transitive verbs that can optionally take it" (p. 183), such as: ]3FH‫ا‬

‫ل‬
ِ (to listen to). 3) "Verbs of permitting, where [‫م‬2&+‫ ]ا‬marks the person(s) (or thing(s))

to whom permission is granted" (p. 183), such as: ‫ل‬


ِ ‫ح‬7&}‫( أ‬to permit someone to do

something), and ‫ل‬


ِ •3&H (to permit someone to do something). 4) "Intransitive verbs

which use [‫م‬2&+‫ ]ا‬in construction with a noun-phrase" (p.183), such as: ‫ل‬
ِ ‫ى‬,&„ (to

happen to), ‫ل‬


ِ 4)YFH‫( ا‬to surrender to), ‫ل‬
ِ 4ƒFP‫( ا‬to avenge) etc. (pp. 182-183).

p+‫?( إ‬ilaa)

There are three types of verbs that may combine with p&+‫ إ‬: 1) verbs which require it,

such as: p+‫[ّق إ‬Z (to stare at), p+‫ إ‬0


ّ Z (to long for), p&+‫ج إ‬7&FZ‫( ا‬to need), p&+‫ إ‬pŽ&`‫( أ‬to listen

to), p&+‫( }{)ّ&] إ‬to look at), p&+‫&• إ‬I+ (to resort to), p&+‫ إ‬xU&P‫( أ‬to listen to ) etc.; 2) transitive

verbs which use p&+‫ إ‬optionally with no change in meaning, such as: p&+‫] إ‬3F&H‫( ا‬to listen

to), p&+‫ء إ‬7&H‫( أ‬to harm), p&+‫ق إ‬7F&o‫( ا‬to miss) and p&+‫ إ‬,&€P (to look at); and 3) intransitive
77

verbs which may use p&+‫ إ‬in construction with a noun-phrase, such as: p&+‫&[ّث إ‬Q} (to

speak to), p&+‫ر إ‬7o‫( أ‬to indicate to), p&+‫ن إ‬


ّ •&3W‫( ا‬to trust), p&+‫ إ‬4ّ ^&P‫( ا‬to join), and p&+‫(&‘ إ‬FP‫( ا‬to

notice). However, the function of p&+‫ إ‬in most of the above instances is "essentially

directional", and the relationship between the verb and the object of p+‫ إ‬can be termed

"TOWARDS" (pp. 185-187) [Emphasis in original].

0/ (min)

Similarly, there are three types of verbs that may combine with 0&/: 1) verbs which

require it, such as: 0&/ ’&ّ)n} (to get rid of), 0&/ ,n&H (to mock), 0&/ ‫ب‬,&Fw‫( ا‬to approach),

and 0&/ 4ƒF&P‫( ا‬to take revenge on); 2) "[t]ransitive verbs which may take [0&/]", such as

0/ ‫( زاد‬to increase), and 0/ Kّ)w (to reduce); and 3) "[i]ntransitive verbs which may take

[0&/] in construction with a noun-phrase", such as: 0&/ |&*} (to get tired of), 0&/ ‫ف‬7&y (to

be afraid of), 0/ p'o (to complain about), 0/ SQR (to laugh at), 0&/ |^G (to get angry

with), 0/ ,_P (to avoid) etc. (pp. 188-189).

0E (3an)

Likewise, 0&E may combine with three types of verbs: 1) verbs which require it, such

as: 0&&E p&&)ّn} (to abandon), 0&&E ,&&(ّE (to express), and 0&&E “&&n
ّ 3} (to produce); 2)

"[t]ransitive verbs which may take [0&E]", such as: 0&E •&Q\ (to search for), 0&E ‫ب‬7&„‫( أ‬to

answer), 0&E ]&8‫( دا‬to defend), 0&E ‫ب‬,&E‫( أ‬to express), 0&E 0&)E‫( أ‬to announce), and 0&E |&ّƒP

(to search for); and 3) "[i]ntransitive verbs which may use [0&E] in construction with a

noun-phrase", such as: 0E [*F\‫( ا‬to move away), 0&E •&)Fy‫( ا‬to differ from), 0&E K
ّ &w (to be

less than), 0&E ]&ّ.3} (to desist from), and 0&E •&ّwX} (to stop). With intransitive verbs, 0&E
78

has three kinds of functions: first, it embodies the semantic component "FROM" with

verbs of motion, as in: 7OL7w[&`‫ أ‬0E 5(+7{+‫*[ت ا‬F\‫( ا‬the student withdrew from her friends);

secondly, it functions "as a comparative predicate", as in: 7O}7ƒ1[&` 0&E 5&(+7{+‫ ا‬x&_)Fy‫( ا‬the

student differed from her friends); and thirdly, it extends "the condition predicted by

the verb to an action in which the subject is involved" as in: 5&H‫[را‬+‫ ا‬0&E |+7{+‫ّ• ا‬wX} (the

student stopped studying) [Emphasis in original] (pp. 191-194).

On the basis of the details outlined above, Lentzner (1977) labels every construction

of verb-preposition as an Arabic verb-preposition idiom, which is not accurate by any

means. He confuses the idiomatic verb-preposition constructions with those of non-

idiomatic ones, and this sort of confusion, in my view, is due partly to his approach.

Lentzner approaches the point ably from the prepositions rather than the verbs

standpoint, concentrating on the function of each preposition when it is combined

with certain types of verbs. In spite of giving the syntactic characteristics of the verbs,

he ignores the semantic features of them and their contribution when combined with

Arabic prepositions.

2.6.2 Abboud and McCarus

Some researchers, such as Abboud and McCarus (1968), have taken up the issue from

the verb standpoint. Abboud and McCarus (1968) observe that there exist two kinds

of verb-preposition constructions in Arabic; idiomatic and non-idiomatic. The first

one being the construction of verb-preposition in which the verb changes its basic

meaning, for example, the meaning of the verb •&Q\ which basically means (to

discuss), becomes (to look for) when combining with the preposition 0&E. Likewise,
79

the verb [&3FE‫( ا‬to authorize, to sanction) and the combination p&)E [&3FE‫( ا‬to depend on).

In such combinations, Abboud and McCarus elaborate, a verb-preposition idiom

constitutes a single unit of meaning "distinct from the verb alone or from other verb-

preposition idioms" (p. 352).

The second kind being construction of verb-preposition in which the verb retains its

basic meaning. For instance: the verb [E7H (to help someone) and the combination [E7H

p)E (to help someone in doing something) (p. 353).

2.6.3 Heliel

Another major contribution was made by Heliel (1994) in an article titled Verb-

Particle Combinations in English and Arabic: Problems for Arab Lexicographers and

Translators in which he approaches the issue from the transitivity 5&1[*F+‫ ا‬standpoint.

He starts with the traditional classification of verbs in the Arabic grammar where

they, like English verbs, can be transitive verbs 5&1[*F/ ‫ل‬7&*8‫ أ‬or intransitive verbs ‫ل‬7&*8‫أ‬

5&/‫ز‬M. The former verbs govern the accusative of a noun through themselves, they are

traditionally known as 7OY&_.\ 5&1[*F/ ‫ل‬7&*8‫( أ‬verbs which pass on to their objects through

themselves). The latter verbs govern a preposition with a noun in the genitive case

instead of accusative, they are traditionally known as ,„ ‫ف‬,Q\ 51[*F/ ‫ل‬7*8‫( أ‬verbs which

pass on to their objects through a preposition). There are some verbs, however, that

may be used in both ways with different meanings, for instance the verb 7&E‫( د‬to call),

which can be used transitively, as in: 7E‫( د‬he called him), or intransitively as in: ‘&+ 7E‫د‬

(he prayed for him) (p. 144).


80

Heliel, then, asks the same question arisen at the beginning of this section whether the

Arabic construction of verb + preposition constitutes "a 'phrasal verb', a 'prepositional

verb', or a different category" (p. 144). To answer this question, he gives the

following example: ‫&[اء‬Ž+‫ى \*&[ ا‬X&)Q+‫ ا‬0&E p.ŽF&H‫( ا‬he went without dessert after lunch).

Heliel asserts that the phrase ‫&[اء‬Ž+‫( \*&[ ا‬after lunch) is a genuine prepositional phrase.

But he proposes two ways of bracketing the remainder, verb phrase plus prepositional

phrase, or verb phrase plus noun phrase as follows:

[ ‫ى‬X)Q+‫ ا‬0E ] [ p.ŽFH‫ ] ا‬or [ ‫ى‬X)Q+‫ [ ] ا‬0E p.ŽFH‫] ا‬

Verb phrase + prepositional phrase / verb phrase + noun phrase

In order to identify a prepositional phrase Heliel conducts three tests:

a) fronting the prepositional phrase, e.g., p.ŽFH‫ى ا‬X)Q+‫ ا‬0E

b) questioning the prepositional phrase with a 'what' question, e.g.,

‫؟‬p.ŽFH‫ء ا‬%o ‫ أي‬0E

c) separating the prepositional phrase from the rest of the sentence by an

adjunct, e.g., ‫ى‬X)Q+‫ ا‬0E ~—.-Z p.ŽFH‫ا‬

He comes to the conclusion that the phrase ‫ى‬X)Q+‫ ا‬0E "is indeed one stable constituent,

and hence a prepositional phrase" (p. 145). Consequently, the combination of p.ŽF&H‫ا‬

0E (dispense with) along with similar combinations like: p)E ‫ف‬,o‫( أ‬supervise), ‫ب‬
ِ ˜&{\

(attack with violence) "may be structurally interpreted, as in English, to be

prepositional verbs, or alternatively to be transitive verbs followed by an 'oblique'

object related to the verb via a preposition" (p. 145). From a semantic standpoint,

Heliel argues that the preposition is essential to the meaning of the verb and is

invariable, so that changing the preposition leads to the alteration of the meaning of

the entire combination, e.g., 5&31,I+‫ ا‬0&E •N‫( آ‬expose the crime) versus 5&(-ƒQ+‫ ا‬p&)E •N&‫آ‬
81

(inspect the bag). In a similar manner, he elaborates that in the Arabic construction of

verb + preposition the verb often keeps most of its meaning and may be extended via

the preposition and perhaps the following noun, e.g., ‫†ة‬,&3+‫ ا‬%&8 ,&€P (looked in the

mirror) versus 5-^&ƒ+‫ ا‬%&8 ,&€P (looked into the case). Prepositions, on the other hands,

almost always retain a degree of their physical meaning, e.g., ‫ب‬


ِ ™(&}‫( ار‬to connect

with), 0&/ ‫ب‬,&‫( ه‬to escape from) etc. Yet both the preposition and the noun it governs

complete the meaning of the verb. This kind of link between the preposition and the

noun it governs and the verb is traditionally termed as ‚&ّ)*} (dependency) (p. 146).

Hunce, the verb is "a fair guide to the meaning of the combination" (p. 146). Heliel

explains that it is not difficult to understand the Arabic verb via the structure whether

it is used with or without a preposition, e.g., ‘ƒQ+ (to follow him) versus ‘\ ‚Q+ (to catch

up with him) (p. 146). This is mainly due to the fact that the verb in Arabic verb +

preposition constructions "is used in extended but rarely too idiomatic or unmotivated

ways" (p. 146), as in: ‫ن‬7U&G9‫ ا‬x+7/ (the branches swayed) [my translation] versus x&+7/

7Oƒ1[` p+‫ ا‬x.(+‫( ا‬the girl tended towards her friend) [my translation] where the verb x+7/

is literal in the first sentence and extended in the second since it is associated with the

preposition p+‫( إ‬to) (p. 146).

It is not unusual to see Arabic native speakers using Arabic prepositions wrongly.

Such a misuse of Arabic prepositions, according to Heliel (1994), "indicates the

centrality of the verb to whose meaning the preposition adds" (p. 146). He,

accordingly, concludes that "these Arabic combinations could be syntactically

considered verbs followed by specific prepositions and be classed as prepositional

verbs, not phrasal verbs" (p. 145).


82

2.6.4 Alkhuli

In his book Comparative Linguistics: English and Arabic, Alkhuli (1999) maintains

that there is only one type of PV in Arabic, that is, prepositional verbs, e.g., p&)E ‰&)„

%&H,'+‫( ا‬he sat on the chair), ™-&Q3+‫ ا‬%&8 ,&Q\‫( أ‬he sailed through the ocean), ,&/9‫ ا‬%&8 ‚&ّw‫د‬

(he inquired into the matter).

It is quite evident that the above examples used by Akhuli (1999) illustrate the

literal/non-idiomatic type of Arabic verb-adverb constructions.

2.6.5 Najiib

Along these lines, Najiib (2001) confirms that PVs exist in the Arabic language (p.

71). Further, he gives the following examples to illustrate them:

%8 |G‫( ر‬to desire, to crave for)


0E |G‫( ر‬to avoid, to dislike)
p+‫ل ا‬7/ (to like, to sympathize)
0E ‫ل‬7/ (to avoid, to dislike)
%8 ]w‫( و‬to fall down)
p)E ]w‫( و‬to come across, to find)
p+‫ ا‬,€P (to look at)
%8 ,€P (to consider) (p. 71) [My translation].

2.6.6 Siinii, Hussein and Al-ddoush

In their monolingual Arabic-Arabic dictionary 5-Z2{&`M‫ات ا‬,&-(*F)+ %w7-Y&+‫ ا‬4&I*3+‫( ا‬The

Contextual Dictionary of Idiomatic Expressions) [My translation], Siinii, Hussein and

Al-ddoush (1996) list around 2000 Arabic idiomatic expressions collected from a

wide variety of sources such as ancient and modern Arabic-Arabic dictionaries,


83

linguistic studies, arts and history literature, and proverbs. The compilers do not

mention Arabic PVs (henceforth APVs) as such, but the vast majority of the listed

Arabic idiomatic expressions contain constructions in which verbs are followed by

prepositions. To take only a few examples:

‘\7YZ %8 ~y‫( أ‬To take into his account) (p. 3) [My translation]

‫ر‬7(FEJ‫ ا‬0-*\ ~y‫( أ‬To take into consideration) (p. 5) [My translation]

‘()w %8 Š› Z (To be touched with) (p. 42) [My translation]

K€+‫ ا‬%8 ˜-*1 (To be away from people) (p. 143) [My translation]

‫ء‬7/ ,(o %8 ‫ق‬,G (To fail in facing even the easiest problem) (p. 94) [My translation]

[1 %8 ™ƒH‫( أ‬To be confused) (p. 10) [My translation]

‫ب‬,o‫‘ و‬-)E ,‫[ه‬+‫ ا‬K‫( أآ‬It is too old) (p. 15) [My translation]

0Y+‫ \‘ ا‬x/ّ[ƒ} (To grow older) (p. 33) [My translation]

‫ب‬7(H9‫ ا‬4O\ x*ّ{ƒ} (They got lost) (p. 33) [My translation]

0Y+‫ ا‬%8 0*W (to be old, to be advanced in years) (p. 80) [My translation]

What is so important about this book is the listing of a number of fixed and idiomatic

Arabic expressions in which verb and preposition fuse in one semantic unit to provide

a new idiomatic meaning different from their literal meanings. To cite just a few

instances:

1. p)E p}‫( أ‬p. 2) (To terminate or destroy). For example:

51,ƒ+‫ ا‬%8 MŠ./ 0-Y3y p)E ‫ان‬,-.+‫ ا‬x}‫( أ‬p. 2).

(The fire destroyed fifty houses in the village) [My translation]

2. 0&E ~&y‫( أ‬p. 6) (To learn from, to study under, to borrow, to transmit, to narrate). For

example:

0-ƒ\7Y+‫ ا‬0E 4‫ره‬7'8‫م أ‬X-+‫ء ا‬73)E 0/ ,-…‫~ آ‬y‫( أ‬p. 6)


84

(A lot of contemporary scientists have adopted their thoughts from their predecessors)

[My translation]

3. %8 ~y‫( أ‬p. 6) (To begin, to start). For example:

‫ب‬2{+‫] ا‬-3„ ‫ل‬X`‫}‘ \*[ و‬,R7Q/ ‫ء‬7ƒ+‫ إ‬%8 ‫ذ‬7FH9‫~ ا‬y‫( أ‬p. 6)

(The professor started delivering his lecture after the arrival of all the students) [My

translation]

4. 0E ])w‫( أ‬p. 15) (To give up). For example:

0-y[F+‫ ا‬0E “1,3+‫)] ا‬w‫( أ‬p. 15)

(The patient gave up smoking) [My translation]

5. 0E SY/‫( أ‬p. 18) (To refrain). For example:

‰3N+‫وب ا‬,G pFZ ,I_+‫ ا‬0/ ‫اب‬,N+‫م وا‬7*{+‫ ا‬0E 4L7U+‫ ا‬SY31 (p.18)

(The fasting man refrains from having food and drink from dawn to dusk) [My

translation]

6. p)E |QYP‫( ا‬p. 20) (To include or apply to). For example:

‫ء‬7.…FH‫ دون ا‬0-.W‫ا‬X3+‫] ا‬-3„ p)E [1[I+‫ار ا‬,ƒ+‫| ا‬QYP‫( إ‬p. 20)

(The new decision has applied to all citizens without exception) [My translation]

‫[د‬I+‫ب ا‬2{+‫ ا‬K‫ آ‬p)E 4ّ)*3+‫م ا‬2‫| آ‬QY.1 (p. 20)

(The teacher's order applies to all new students) [My translation]

7. p+‫{] إ‬ƒP‫( ا‬p. 21) (To devote oneself to, to dedicate oneself to). For example:

7O.ƒ}‫ أ‬pFZ 5-\,*+‫ ا‬5Ž)+‫ ا‬5H‫ درا‬p+‫[ إ‬3Z‫{] أ‬ƒP‫( ا‬p. 21)

(Ahmad devoted himself to study the Arabic language until he mastered it) [My

translation]

8. 0E ]{ƒP‫( ا‬p. 21) (To quit, to give up, to stop). For example:

2
َ 1XW 7./‫ ز‬7O}‫ر‬71‫ ز‬0E x*{ƒP‫ ا‬%F+‫ ا‬7OFƒ1[` ‫[ت ه[ى‬ƒF8‫( ا‬p. 21)

(Huda has missed her friend who stoped visiting her for a long time) [My translation]
85

9. ‫ ب‬p.\ (p. 27) (To consummate the marriage with). For example:

5-*/7I+‫ ا‬7OFH‫ درا‬x)3‫ \*[ أن أآ‬،‘F„‫و‬Š\ •‫([ ا‬E p.\ (p. 27)

(Abdullah consummated the marriage with his wife when she completed her

university study) [My translation]

10. %8 ,ّQ(} (p. 30) (To study thoroughly, to go deeply into, to be an authority or expert

in). For example:

4)*+‫ ا‬%8 ‫وا‬,ّQ(F1 4+7/ ،‰1‫[ر‬F+7\ 4‫~ه‬-/2F+ ‫ن‬XP‫•ذ‬1 M ‫ء‬7/[ƒ+‫ن ا‬7‫( آ‬p. 30)

(The ancestors did not allow their students to teach unless they are acquired through

knowledge) [My translation]

11. ‫( }*)ّ‚ ب‬p. 32) (To fall in love with, to be very fond of). For example:

,ŽU+‫~ ا‬./ ‘ّ3E x.(\ pF_+‫( }*)ّ‚ ا‬p. 32)

(The young man fell in love with his cousin since childhood) [My translation]

12. p+‫ إ‬p/‫( ر‬p. 61) (To intend to, to aim at). For example:

01[)(+‫ ا‬0-\ 5)*F_3+‫ة ا‬XI_+‫ ا‬5+‫ إزا‬p+‫‘ إ‬Q1,U} %8 %H7/X)\[+‫ ا‬p/‫( ر‬p. 61)

(The diplomat intended, by his statement, to eliminate the fabricated gap between the

two countries) [My translation]

13. p)E ,OH (p. 66) (To look after, to take care of). For example:

‫س‬73Q+‫[ وا‬I+‫ ا‬0/ ,-…'\ [)\ 717E‫ون ر‬žo p)E ,OY1 ,-_Y+‫ ا‬KŸ (p. 66)

(The ambassador kept looking after his fellow citizens' affairs with a lot of diligence

and enthusiasm) [My translation]

14. %8 ]ƒ1 (p. 144) (To consist of, to contain, to make up of). For example:

5Q_` 0-Y3y %8 ‫ل‬7ƒ3+‫] ا‬ƒ1 (p. 144)

(The article is made up of fifty pages) [My translation]


86

2.6.7 Summary of section 2.6

To sum up, I agree with Lentzner (1977) in his view that Arabic constructions of verb

+ preposition have characteristics of both PVs and prepositional verbs in English. I

believe that they are syntactically prepositional verbs but semantically PVs due to the

fact that they are idiomatic in nature. That is, each of which constitutes a single unit of

meaning which has nothing to do with the individual meanings of its components.

However, I disagree with him as to labeling all Arabic verb-preposition constructions

as verb–preposition idioms, owing to the fact that most of these constructions do not

satisfy the criteria of idioms set out by grammarians and linguists who define idioms

as "frozen patterns of language which allow little or no variation in form and […]

often carry meanings which cannot be deduced from their individual components"

(Baker, 1992, p. 63). I will apply the two tests, mentioned in the previous section,

introduced by Cowie and Mackin (1993) on some Arabic verb-preposition

constructions to see whether they are idioms or not. The first test is substituting the

whole phrase with one word. This begs the question: can we substitute a phrase like

p&)E [E7&H 'to help' with one word? The answer is: no, there is no one word for this

Arabic phrase, simply because it is made up of two lexical constituents each of which

has its own literal meaning, and the whole meaning of the phrase is so transparent that

it can be easily deduced from the total sum of the meanings of its components. The

second test is to delete the second word of the phrase. Again, can we delete the Arabic

preposition without changing the whole meaning of the phrase? The answer is: yes,

we can say [E7&H only to express the same meaning of the phrase 'to help'. Therefore,

this type of Arabic constructions is non-idiomatic/literal by any means.


87

I, also, disagree with Abboud and McCarus (1968) who divided these structures into

two types only, i.e. idiomatic and non- idiomatic. I believe that there is a third type in

between, that is, the semi-idiomatic type which is manifested in many Arabic verb +

preposition structures. To take just one example, p&)E [&3FE‫' ا‬to depend on', which cannot

be substituted for one word. However, unlike the case of the previous example, the

preposition in this phrase cannot be deleted without changing the whole meaning, this is

owing to the fact that the verb [&3FE‫( ا‬to authorize, to recognize) sacrifices its basic

meaning when combining with preposition, while the preposition retains its basic

meaning. For that reason this type of Arabic constructions, in my view, can be called

semi-idiomatic rather than idiomatic one.

With regard to Heliel's claim that these sorts of Arabic combinations are syntactically

prepositional verbs but not PVs, I believe that he opted for such a view because he

deems that EPVs mean verb + adverb only, i.e. not verb + preposition nor verb +

preposition + adverb, which explains why he studies them under the name of verb-

particle combinations and distinguished among them syntactically. As far as I am

concerned, as I have proposed before, I consider the three abovementioned types of

English combinations as PVs since they are idiomatic in nature. Idiomaticity is my

parameter in this study, therefore, I consider the idiomatic Arabic verb-preposition

constructions as idiomatic Arabic PVs (henceforth IAPVs). They, undoubtedly,

constitute constructions more or less similar to IEPVs in that each of which forms one

unit of meaning, and exhibits strong semantic relationships between prepositions and

the verbs with which they combine.


88

Consequently, in order to answer the question raised at the beginning of this section. I

strongly agree with the views of Lentzner (1977), Najiib (2001), Kharama and Hajjaj

(1989), Alkhuli (1999) and Bataineh and Bataineh (2002) that PVs do exist in the

Arabic language. PVs constitute a significant phenomenon in the English language;

however the Arabic language does not distinguish this phenomenon as such. It is quite

evident from the above literature that Arabic grammarians do not classify the Arabic

construction of verb + preposition under a specific heading.

As for the criteria, not all APVs can fit the syntactic and semantic criteria of IEPVs.

They have rather different usage and criteria. Some of them may fit the criteria of

their English counterparts, but differ in their usage. The similarities and dissimilarities

can be shaped as follows:

1. Unlike IEPVs which are of three types (i.e., verb + adverb, verb +

preposition, and verb + adverb + preposition), APVs are of one type only,

that is, verb + preposition.

2. The directional and locative 'true' prepositions, as Lentzner (1977) terms

them, can be used to form APVs, since adverbial particles are not used in

such Arabic constructions.

3. Like EPVs, Arabic verb + preposition structure can be idiomatic, semi-

idiomatic, and non-idiomatic.

4. Like IEPVs, the majority of APVs constitute integral parts of Arabic

idiomatic and fixed expressions.

5. Like polysamic IEPVs, APVs may occur in more than one meaning in

different contexts.
89

6. Like EPVs, meanings of APVs subject to change according to word or

words with which they may collocate.

7. Like the particle in IEPVs, in IAPVs the preposition modifies to a certain

extent the meaning of the verb it combines with. It is so essential that

changing it entails altering of the meaning of the whole structure.

8. Unlike EPVs, whose communicative function is mainly carried by

particles (Side, 1990), the communicative function of IAPVs is typically

carried by the verb in spite of the fact that the preposition modifies the

meaning of the verb it combines with.

9. Like EPVs, Arabic literal and idiomatic PVs share the same syntactic

properties. The distinction between them, however, is confined to their

semantic properties.

10. Like IEPVs, IAPVs can be substituted by a one word synonym.

11. Unlike EPVs which can occur without a syntactic object, APVs require a

syntactic object and cannot occur without it.

12. As far as register variations are concerned, unlike EPVs which are less

formal, less rhetorical and commonly used orally by everyone in everyday

contexts, APVs are far more formal and highly rhetorical. They are

typically used in formal settings such as literary works, religious sermons,

political speeches, academic contexts, etc. Most of them are confined to

the written mode and used only by educated people from a certain sector

of society and educational background. APVs, in short, are so rhetorical

that using them in an informal setting makes the speaker sound very odd

and unnatural.
90

The final point needs to be taken up here is the fact that Arabic makes "a sharp

distinction between written and spoken discourse" (Baker, 1992, p. 71). It "has two

varieties: a high written (formal) variety and a low spoken (mainly informal) variety"

(Al-Qinai, 2000, p. 501). Therefore, in rendering such colloquial expressions as EPVs

into written form of Arabic, Arabic translators would run the risk of shifting the

register from informal/colloquial to formal/standard expressions (see 5.3.6.2.2 below).

One of the strategies suggested to Arabic translators to tackle such a dilemma is,

according to Al-Qinai (2000), "to reduce the gap by steering a middle course between

the formal style of the high variety and the informal language of Colloquial Arabic"

(p. 501). In other words, translators are best advised to use Modern Standard Arabic

which lies half-way between the highly formal classic Arabic and the highly informal

colloquial Arabic.

2.7 Derivation of PVs in English and Arabic

One of the major aims of this study is to pinpoint the difficulties encountered by

Arabic translators when dealing with the notion of IEPVs. The phenomenon of

derivation of PVs constitutes, to a certain extent, one aspect of such difficulties. In the

previous sections the syntactic and semantic properties of idiomatic PVs in both

English and Arabic were explored in detail. In this section I will be considering the

ways by which EPVs can be derived from adjectives and nouns and vice versa. Yet,

before proceeding in so doing, a brief account of the notion of derivation in both the

English and Arabic languages will be outlined. The aim here is to present the

morphological, grammatical and semantic features of derivation upon which the

process and the scope of deriving PVs are based.


91

2.7.1 Derivation in English

The phenomenon of derivation in the English language has been of interest to a wide

range of grammarians, linguists and lexicographers (cf. Hurford & Heasley, 1983;

Crystal, 1997; Kaplan, 1995; Crowley, Lynch, Siegel & Piau, 1995; Trask, 2000). It

has been covered in the frameworks of word-formation, productivity, and in the

context of converting deep structure into surface structure in Transformational

Grammar (Kaplan, 1995; Crowley et al., 1995; Trask, 2000).

Derivation can be broadly defined as "[…] the process of forming new words

according to a (fairly) regular pattern on the basis of pre-existing words" (Hurford &

Heasley, 1983, p. 206), or a process of word-formation by which words are derived

from other words by adding affixes, such as deriving the words national, international

and internationally from the word nation (Trask, 2000, p. 40). The vast majority of

English vocabulary emerges by the process of forming new lexemes from old ones -

either by adding affixes to them, or combining them to create compounds (Crystal,

1997, p. 128). The new derived words, however, are typically of different "part of

speech"; a good example would be the English "manner" adverbs which are formed

from adjectives by adding the suffix –ly to them, as in the adjective rapid and the

manner adverb rapidly. Another example is making nouns from verbs by adding the

suffix –ment to them as in the verb excite and the noun excitement (Kaplan, 1995, p.

27).

Furthermore, English, following Crystal (1997), "does not have affixes in large

numbers – only about 50 common prefixes, somewhat fewer common suffixes, and no
92

clear instances of infixes. But these limited resources are used in a complex and

productive way" (p. 128).

There exist three kinds of affixes in the English language, namely:

1- Prefixes which are typically placed before the roots of words, such as: in-,

un-, de-, dis-, miss-, sub-, anti-, ex-, etc. (A more comprehensive list in

Crystal, 1997, p. 128).

2- Suffixes which are usually placed after the roots of words, such as: -tion, -

ship, -ness, - able, -ese, -like, -let, -ess, -ism, etc. (Crystal, 1997).

It is worth noting that unlike prefixes which rarely alter the words class, suffixes do

not only modify the meaning of the word to which they are attached. but also, in many

cases, change the word's grammatical status, for instance, the suffix –ify turns the

noun beauty into the verb beautify, and the suffix –ing turns the concrete noun farm

into the abstract one farming (Crystal, 1997, 128). Another point needs to be

mentioned here is that English does not allow using more than one prefix at a time,

but, however, it does allow using two or more suffixes as in: person-al-ity and norm-

al-is-er (Kharma & Hajjaj, 1989, p. 38).

3- Infixes which occur within the roots of words. It is necessary to point out

though that, unlike many languages including Arabic which make a great use

of infixes to express such notions as tense, number, or gender, the English

language has no system of infixes, but English speakers do occasionally create

words into which other forms have been inserted for the sake of swearing or

being emphatic, as in kangabloodyroo (Crystal, 1997, 128).


93

In their book The Design of Language, Crowley et al. (1995) approach the

phenomenon from the perspective of distinguishing between inflectional and

derivational affixes. They consider that words with inflectional affixes such as expect,

expects, expected and expecting are mere different forms for one lexeme (expect),

while words with derivational affixes such as expectant, expectancy, expectantly,

expectation and expectative are considered five different words, even though they are

all related to the form expect morphologically. In other words, inflectional affixes do

not create new lexemes, but they rather create other shapes of the same lexemes,

while derivational affixes do create new lexemes from other ones. Further, the full set

of forms of any lexeme that are inflectionally related are usually referred to as the

paradigm for that lexeme. Hence, the forms expect, expects, expected and expecting

are considered to be the paradigm for expect. Derived forms, however, have their own

paradigm as well. Therefore, the forms expectation and expectations are considered to

be the paradigm for expectation (pp. 257-258).

Further, the process of derivation, according to Hurford and Heasley (1983), is

actually not merely one process; it is rather of three simultaneous processes, as

follows:

1- A morphological process, by which the shape of a derived word is changed

by adding prefix or suffix as in: laugh – laughter, table – tabulate, bake –

bakery, honest – honesty, etc.

2- A syntactic process, by which the part of speech of a derived word is

changed, e.g., from verb to noun as in: teach – teacher, from adjective to

noun as in: red – redness, from adjective to verb as in: wide – widen, from

verb to adjective as in: avoid – avoidable, etc.


94

3- A semantic process, by which a new sense is produced, e.g., producing a

word denoting an act or an activity when deriving laughter from laugh;

producing a word denoting an agent when deriving teacher from teach; or

producing a word denoting a property when deriving redness from red, etc.

(pp. 206-207).

To be more precise, however, the process of derivation is not always involved in the

three abovementioned processes. There exist some cases in which derivation involves

two processes only, as in the case of 'zero derivation' where no morphological process

is involved, e.g., deriving cook (agent noun) from cook (transitive verb). This case

typically occurs when the same word is used in different part of speech. On the other

hand, there exist some cases in which no grammatical process is involved, as in

deriving larger from large where both words are adjectives, but have different forms

and distinct semantic properties (pp. 207- 209).

2.7.2 Derivation in Arabic

The notion of derivation in Arabic stems initially from the issue of inflection

(•1,U&F+‫ )ا‬which constitutes the cornerstone of morphology (‫ف‬,U&+‫ ا‬4&)E) in Arabic.

Inflection, as it has been defined by Al-jurjaanii (died 1413 A.H), is a "transferring of

the single root to a variety of patterns for intended meanings which cannot be

achieved without such patterns" (p. 61) [my translation]. Arabic scholars, however,

divide the verb morphologically into two types: defective ([&&/7„) and inflected

(‫ّف‬,U&F/). The defective verb is a verb that carries a meaning unrelated to time, hence

it is a frozen verb in the sense that it is confined to one particular pattern, e.g., ‰-&+ (it
95

is not), pY&E (it may be), 4َ &*ِP (how good), ‰


َ —&ِ\ (how bad). The inflected verb, on the

other hand, is a verb that carries a meaning related to time; therefore it is capable to

appear in different patterns depending on the time in which the action takes place. The

inflected verb is of two types: fully inflected which can produce the three verb

patterns (past: %&R73+‫ا‬, present: ‫رع‬7^&3+‫ ا‬and imperative: ,&/9‫)ا‬. And partially inflected

which can produce only two verbs patterns, either past and present such as: ‫د‬7&'1 – ‫د‬7&‫آ‬

(to be on the brink of), S&oX1 – S&o‫( أو‬to be about to),‫ال‬Š17&/ ¡‫زال‬7&/ (still, yet), – ‫ح‬,\7&/

‫ح‬,(17/ (to continue to be), S


ّ _.17/ – S
ّ _P‫ا‬7/ (not to cease doing) or present and imperative

such as ،‫[ع – دع‬1 (to leave) (Al-ghalaayiinii, 1986, pp. 55-64).

Derivation (‫ق‬7ƒF&oM‫ )ا‬in Arabic, in its simplest definition, is an extraction of one word

from another provided that there is suitability (5(&H7./) between them in meaning and

structure but not in form (?ibin Jinnii died 392 A.H as cited in Hammaad, 1983, p.17;

Al-jurjaanii, 1986, p. 27). For example, from an infinitive such as: 5&\7F‫( آ‬writing) we

can derive the past tense |


َ &َF‫( َآ‬he wrote), from which we can derive the present tense

|&F'َ1 (he writes) from which, in turn, we can derive the imperative |&F‫( اآ‬write).

Infinitive always constitutes the basic form from which all other derivative forms can

be derived (Al-ghalaayiinii, 1986, pp. 208-209).

Derivation in Arabic falls into three types:

1) The small derivation (,-ŽU&+‫ق ا‬7ƒF&oM‫ )ا‬where suitability (5(&H7./) has to be

between the two words in letters and letters' order such as: ‫ب‬
َ ,َ &َR (to hit)

and ‫ب‬,^+‫( ا‬hitting).


96

2) The big derivation (,-('+‫ق ا‬7ƒFoM‫ )ا‬where suitability between the two words

is in form and function but not in letters' order such as: ~َ &َ(„
َ (to attract) and

‫~ب‬I+‫( ا‬attraction).

3) The biggest derivation (,&(‫آ‬9‫ق ا‬7ƒF&oM‫ )ا‬where suitability between the two

words is in articulation as in: ‚


َ &َ*Pَ (to croak) and ‚&O.+‫( ا‬croaking) (Al-

jurjaanii, 1986, p. 87).

In his book 5&Ž)+‫&‘ ا‬ƒ8 (Philology), Waafii (1973, pp. 179-180) attempts to be more

general in outlining the notion of derivation, He argues that every triliteral root ( K&`‫أ‬

%&ˆ2ˆ) in Arabic correlates with a general meaning to which it has been assigned. Such

a general meaning can be captured in every word comprises the three sounds

occurring in the same order as the root from which the word has been generated. The

general meaning of 4&)*+‫( ا‬knowledge), for instance, correlates with the sounds of its

triliteral root, i.e. 0-&*+‫( ا‬3ayn), ‫م‬2&+‫( ا‬laam) and 4-&3+‫( ا‬miim), and it can be captured in

each word containing these three sounds occurring in the same order no matter what

sounds may be inserted before, after or through them. Hence, the general meaning of

4&)*+‫( ا‬knowledge) can be captured in the following derived words: 4َ &ِ)E


َ (he knew), 7&.3ِ)E
َ

(we knew), 4&)E‫( أ‬I know), 4&)*P (we know), 4َ &¢)E


َ (he taught), ‫ا‬X&3¢)E
َ (they taught), 4&£)*َ 1ُ (he

teaches), 4&£)*َ Pُ (we teach), 4&¢)*َ }َ (he learnt), ‫ا‬X&3¢)*َ }َ (they learnt), 4ِ+7&E (scholar), ‫م‬X&&)E

(sciences) etc. Waafii terms this sort of derivation as ‫م‬7&&*+‫ق ا‬7ƒF&&oJ‫( ا‬the general

derivation), and indicates that it is of two types:

1- Derivation from the names of substances (‫ن‬7&-E9‫ء ا‬73&H‫)أ‬, such as: |¢‫& َ~ه‬/ (gilded)

from |‫( ذه‬gold), “¢^_َ / (silver-plated) from 5ّ^8 (silver), etc.


97

2- Derivation of the synthetic infinitive (%E7.U&+‫[ر ا‬U&3+‫ )ا‬by adding to words ‫ء‬7&1

|Y&.+‫( ا‬relative yaa?) along with ‫ء‬7&F+‫( ا‬taa?) such as: 5&ّ-)‫ه‬7„ (state of ignorance)

from K&‫ه‬7„ (ignorant), 5&ّ-\X\‫ ر‬or 5&ّ-‫ه‬X+‫( إ‬godhood) from ‫ رب‬or ‘&+‫( إ‬god), 5&ّ-+X„‫ر‬

(manhood) from K„‫( ر‬man) etc.

Interestingly, these two types of derivation were not widely used in ancient Arabic,

but owing to the real need of them in the last forty years or so to express a tremendous

number of philosophical and scientific facts, Academy of the Arabic language ( ]&3I/

5-\,*+‫ ا‬5Ž)+‫ )ا‬allowed using them when necessary (pp. 179-180).

Moving on to the issue of Arabic derivative forms, In his book ‫ق‬7ƒF&oM‫ ا‬5+7&H‫( ر‬the

message of derivation) [my translation], Al-sarraaj (died 316 A.H.) imposes two

conditions for derivation to be accepted: 1) The two words have to share the three root

letters of the trilateral verb (َK*َ 8َ ), and 2) The two words have to share a particular type

of meaning (p. 20). Arabic grammarians have agreed on using the trilateral verb as

paradigm, and call the first radical of it ‫ء‬7_+‫( ا‬faa?), the second 0-*+‫( ا‬3ayn) and the third

‫م‬2+‫( ا‬laam) (Wright, 1981, p. 30).

The Arabic language plays with the trilateral root of the verb to make new words with

subtle changes to the meaning (Wightwick & Gaafar, 1998, p. 38).

In what follows I will enumerate the major eight derivative forms (paradigms) in

Arabic. There exist some other forms, but they are very rare and only used in poetry

and archaic texts (cf. Wright, 1981; Al-bustaanii, 1963; Wightwick & Gaafar, 1998):
98

1- َ َ (fa33ala) which is coined by doubling the second radical of the basic

verb, for example: ‫س‬


َ ‫( َد َر‬the basic verb meaning: to study), ‫س‬
َ ‫ر‬¢ ‫( َد‬to teach),

,َ Y
َ ‫( َآ‬to break) ,َ Y
¢ ‫( َآ‬to break in pieces).

2- ََ (faa3ala) which is coined by adding a long vowel aa after the first

radical, e.g., ‰
َ &َ)„
َ (the basic verb meaning: to sit down), ‰
َ +َ7&„ (to sit down

with someone) K
َ Fَ wَ (to kill), K
َ }َ 7w (to fight).

3- َ َ ‫?( َأ‬af3ala) which is formed by adding an ?alif before the first radical, e.g.,


َ &َ{Pَ (the basic verb meaning: to articulate) ‚
َ &َ{P‫( أ‬to make someone talk) ‫ى‬,&„

(to flow) ‫ى‬,„‫( أ‬to make flow).

4- َ َ َ (tafa33ala) which is coined by adding taa? on the front of the above first

form (َK*¢ 8َ ), e.g., ,َ ‫آ‬¢ ‫( َذ‬to remind someone) ,َ ‫آ‬¢ ~َ }َ (to remember), ‫ف‬
َ X¢ y
َ (to terrify)

‫ف‬
َ X¢ n
َ }َ (to be afraid).

5- َ َ َ (tafaa3ala) which is coined by adding taa? in front of the above second

form (َK&َE78), e.g., ‫ن‬


َ ‫ َو‬7&E (to help someone) ‫ن‬
َ ‫ َو‬7&*َ} (to cooperate with someone),

‫ع‬
َ ‫ َد‬7y (to deceive) ‫ع‬
َ ‫ َد‬7nَ} (to pretend to be deceived).

6- َ َ َ ‫?( إ‬infa3ala) which is coined by adding ?alif and nuun in front of the root

letters, e.g., ,َ Y
َ &َ‫( آ‬to break something) ,َ Y
َ &َ'P‫( إ‬to be broken), •
َ N
َ &َ‫( آ‬to uncover)


َ N
َ 'َ P‫( إ‬to be uncovered).

7- َ َ َ ‫?( إ‬ifta3ala) which is coined by adding ?alif with short vowel i in front of

the first radical and taa? after it, e.g., K


َ *َ 8َ (to do something) K
َ *َ Fَ 8‫( إ‬to fabricate

something), ‰
َ 3َ +َ (to touch) ‰
َ 3َ Fَ +‫( إ‬to seek for something).

8- ََ ‫?( إ‬istaf3ala) which is coined by adding ?alif with short vowel i, siin

and taa? in front of the first letter and placing a sukuun over the first root

letter, e.g., ‫ج‬


َ ,َ y
َ (he went out) ‫ج‬
َ ,َ nَF&H‫( إ‬he took out something) (Wright, 1981,

pp. 29-47 ;Wightwick & Gaafar, 1998, pp. 38-61).


99

The point to be made here is that derivation is not confined to verbs only; nouns can

be derived from verbs as well. There are twelve kinds of nonverbal derived from

Arabic verb, and Arabic linguists have agreed to call them ‫ت‬7ƒFN&3+‫( ا‬the derivatives).

They are: ‫[ر‬U3+‫( ا‬infinitive, verbal noun) ‫ّة‬,&3+‫ ا‬4H‫( ا‬noun that expresses the doing of an

action once) ‫ع‬X.+‫ ا‬4H‫( ا‬noun of kind or manner) ‫ن‬7'3+‫ ا‬4H‫( ا‬noun of place) ‫ن‬7/Š+‫ ا‬4H‫( ا‬noun

of time) 5&+¤‫ ا‬4&H‫( ا‬noun of instrument) K&E7_+‫ ا‬4&H‫( ا‬active participle) ‫ل‬X&*_3+‫ ا‬4&H‫( ا‬passive

participle) 5Oّ(N
¢ &3+‫ ا‬5_U&+‫( ا‬assimilate epithet) K-^&_F+‫ ا‬K&*8‫( أ‬competitive and superlative

adjectives) 5Ž+7(3+‫ ا‬5)…/‫( أ‬intensive paradigms) (Wright, 1981; Al-bustaanii, 1963).

It is important to stress that despite the complexity of the Arabic derivative forms,

they are of a great benefit for Arabic learners. For instance, one can guess the

meaning of any word if he recognizes the derivative form and knows another word

with the same root (p. 38). They are, on the other hand, not confined to the Arabic

language only; many of them occur in Hebrew and Aramaic (cf. Wright, 1981, pp. 31,

32, 34, 36, 38, 40, 41).

2.7.3 Derivation of PVs in English

Forming PVs from other word classes is quite noticeable phenomenon in the English

language. This is mainly due to the productive nature of these kinds of verbs which

can be emerged at any time, in any situation and within any context throughout the

stretch of the English language. In this regard there are two points that need to be

taken up. Firstly, the sources from which EPVs are derived and, secondly, the ways

by which nouns can be derived from PVs.


100

EPVs are derived from three sources, namely adjectives, nouns and Latinate verbs as

follows (McArthur, 1989, pp. 40-41):

2.7.3.1 Deriving PVs from adjectives

EPVs can be formed from the three types of adjectives:

a) Adjectives which can take the suffix –en as in: "fresh, freshen, freshen up

and flat, flatten, flatten down" (McArthur, 1989, p. 40) [Emphasis in original]

b) Adjectives which cannot take the suffix –en as in: "calm, calm down and

warm, warm up" (McArthur, 1989, pp. 40-41) [Emphasis in original]

c) Adjectives which can be both (with and without the suffix –en) as in:

"damp, dampen, dampen down and damp, damp down" (McArthur, 1989, p.

41) [Emphasis in original].

2.7.3.2 Deriving PVs from nouns

In this process "a phrasal verb containing a regular phrasal verb and a special noun is

telescoped into a new phrasal verb" (McArthur, 1989, p. 41) as in: wall in (close in

with a wall) and channel off (lead off by means of a channel) (McArthur, 1989, p. 41).

Other examples are: "button up, dish out, fog up, iron out, and sponge down"

(McArthur, 1989, p. 41) [Emphasis in original]. This process has been named by

Fraser (1976, pp. 22-24) as "noun verbalization" and the derived PV as "noun-particle
101

combination" where nouns may occur as verbs in combinations with such particles as

down, in, over, and out. He provides the following examples to illustrate the classes of

nouns and the particles that may occur with them:

a) Nouns occur with the particle down: batten, blot, button, cement, clamp,

glue, nail, past, pin, rivet, screw, staple, tack, and tape as in: He glued down

the loose edge of the painting.

b) Nouns occur with the particle in: box, fence, glass, pen, rope, screen, and

wall as in: He fenced in the porch.

c) Nouns occur with the particle over: board, brick, cement, glass, mortar, and

wall as in: The man bricked over the entrance way.

d) Nouns occur with the particle out: chalk, crayon, ink paint, pen, and pencil

as in: The clerk penciled out the entry.

Fraser (1976) also indicates that the majority of the abovementioned nouns never

occur as verbs except in these combinations, therefore, the combinations "will have to

be listed in the lexicon as individual verbal elements apart from the listing of the

verbs and/or nouns with which they are associated" (p. 24).

2.7.3.3 Deriving PVs from Latinate verbs

In this process Latinate verbs of two and three syllables are attached to particles for

emphasis or completion to form PVs. McArthur (1989) cites the following examples:

"contract out, divide off/up, level off, measure off/out, select out, [and] separate

off/out" (p. 41) [Emphasis in original]. PVs from this type are still being widely used
102

even though their usage is avoided by some researchers and regarded as pleonastic

(McArthur, 1989).

2.7.3.4 Deriving nouns from PVs

Nouns are typically formed from PVs. Such derived nouns "are becoming

increasingly common in modern English, in conversation, in newspapers and in

technical usage" (McArthur, 1975, p. 48).

Theses new forms have been labeled by Fraser (1976) as "Verb-Particle

Nominalizations" and exemplified in what follows:

The blast-off occurred at 4 o'clock.


The foulup was caused by a faulty gas line.
The stowaway was found hiding under the lifeboat.
Where were you during the cave-in? (p. 27) [Emphasis in original]
.

Moreover, they have been named by Cowie and Mackin (1993, p. xiii) as

"Nominalized forms" and given the following instances:

Make-up is formed from make up (to apply cosmetics to one's face).

Take-off is formed from take off (to leave the ground in an aircraft).

Telling-off is formed from tell off (to reproach or to reprimand).

McArthur (1975, pp. 48-52) demonstrates that there are two patterns of formation in

this regard. He terms them as: the major pattern and the minor pattern of formation.

The first pattern being the commonest one for deriving nouns from PVs in which the

PV remains unchanged as in:

to break down a breakdown

to round up a round-up [Emphasis in original]


103

The first element of the noun is stressed in the spoken mode. And the noun formed is

either written as one word or with a hyphen in the written mode. This "stress shifting

rule" of generating nouns from PVs has been also mentioned by Bolinger (1971) who

considers PVs as "probably the most prolific source of new nouns in English" (p. xiii).

The second pattern of deriving nouns from PV is the minor one wherein nouns are

derived by a process of inversion as in:

Disease broke out. The outbreak of disease alarmed us (McArthur 1975, p.

49) [Emphasis in original].

The noun here is derived from the PV by fronting (or prefixing) the particle. In

spoken mode stress should be placed on the first element of the derived noun, i.e. the

particle (McArthur, 1975).

It is necessary to point out, however, that there exist some nouns derived from PVs on

both patterns (i.e. the major and minor patterns) Consider, for example, the following

pairs cited by McArthur (1975):

Breakout and outbreak [Nouns derived from the PV break out]

Layout and outlay [Nouns derived from the PV lay out]

Lookout and outlook [Nouns derived from the PV look out]

Additionally, some of such nouns appear with the gerund form of the verb such as:

beating-up, bringing-up, dressing-down, falling-out, thinning-out, and chewing-out.

And some of them appear with the past participle form of the verb such as: left-over

(Fraser, 1976, p. 27).


104

It must be pointed out that each noun of the above pairs has its special use, for

instance the use of breakout differs from the use of outbreak in that the former usually

has something to do with people whereas the latter has something to do with diseases

and troubles (McArthur, 1975, p. 50).

2.7.4 Productivity

Productivity can be broadly defined as "[t]he degree to which a grammatical pattern

can be used freely to construct new instances" (Trask, 2000, p. 109). Or rather "[t]he

formation of new linguistic expressions based on the existing pattern of usage" (Daud

et al., 2003, p. 159). It is undeniable fact, however, that such a phenomenon is not

solely confined to the English language, it is rather a common factor among all

languages. Hence, the fact that the human language allows its users to communicate

in entirely productive way is considered one of the main features that differentiate the

human language from non-linguistic aspects of communications such as non-verbal

human communication, and perhaps all aspects of animal communication. There is in

fact no limit to the using of human language (Crowley et al., 1995). Therefore,

productivity in language means, following Crowley et al. (1995), "that once we have

learned a particular rule for combining two or more items together, we can combine

any like items in the same way to produce a novel utterance" (p. 262).
105

2.7.4.1 Productivity in English

English linguists, grammarians, and lexicographers have paid a considerable attention

to the phenomenon of productivity. It has been studied at length on the light of

derivation and word-formation. Since, as it has been outlined above, affixes are of two

types, namely, inflectional and derivational, the former is said to be more productive

than the latter, in that they can be used more freely to invent new words. Hurford and

Heasley (1983) maintains that "[i]t is doubtful whether any derivational process is

actually completely productive, but some are very productive and others hardly

productive at all" (p. 213).

Consequently, the English language scholars have divided affixes, according to their

productivity, into three categories: productive, non-productive, and semi-productive

(Kaplan, 1995; Crowley et al., 1995; Trask, 2000). Yet, for the purpose of the study, I

will not go into the details of theses types of productivity as my focus is on the

productivity of IEPVs in particular, in an attempt to explore the mechanism by which

such combinations can be typically produced.

It is not an exaggeration to say that thousands of IEPVs are being formed casually

when needed; they pop up on a regular basis with new special meanings (McArthur,

1975). They are undoubtedly "a highly productive category in English" (Heliel, 1994,

pp. 141-142), which "probably accounts for more new 'stereotypes' than any other

source" (Bolinger, 1971, p.xii). One when dealing with them, following Kennedy

(1967),

[…] is not dealing with a fixed category of English speech, but with a
changing, growing tendency in language which throws up over night, as
106

it were, new combinations, and new meanings, so that an absolute and


complete list would be impossible of realization (p. 5).

In his attempt to explore the mechanism of coining such "a floodgate of metaphor"

Bolinger (1971, p. xii) indicates that there is no need for the everyday inventor of a

PV to arrive at such elements as roots and affixes which have no reality for him. All

what he needs is a rough familiarity with other use of words like head and off to make

up of the PV head off , this is mainly due to the self-suggesting nature of this kind of

verbs when the occasion for them comes up.

Unlike people from non English speaking backgrounds, English native speakers have

no difficulty obtaining this kind of "a rough familiarity with other use" of both

elements of PV. They "have an understanding, albeit unconscious, of the meaning and

use of particles that allows them to create, almost at will, new phrasal verbs" (Darwin

& Gray, 1999, p. 66). This highly productive nature of EPVs complicates the problem

of learning them by learners of English who already have a great deal of difficulties

with their syntactic and semantic characteristics (p. 66).

It might be plausible to explore here one of the most controversial issues in this

connection, that is, the dilemma faced by lexicographers. There is no doubt that

people will keep generating newely coined PVs (along with other derived word

classes) unendingly. This will, in turn, make the task of capturing such new items by

lexicographers even harder. They would be as if they were, to use Hurford and

Heasley's (1983) words, "shooting at a moving target" (p. 205), because their

dictionaries will soon be out of date if they recorded only attested words "as new

words will have been coined and perhaps added to the everyday vocabulary of the
107

language" (p. 205). The question of lexicography will be investigated in more details

in the forthcoming sections. But what needs to be considered here is that such newely

coined, unrecorded, EPVs pose a great deal of challenge to Arabic translators as they,

in dealing with them, have to play the role of lexicographers, in addition to their basic

role, in tracing them, analysing them within their situational contexts, and providing

their appropriate Arabic equivalents. To give just a few examples of such newly

coined items: to sex up, to google around, to shop out, to party out, etc.

2.7.4.2 Productivity in Arabic

There is no dout that abovementioned Arabic derivative forms are "the major way in

which Arabic achieves its richness of vocabulary" (Wightwick & Gaafar, 1998, p.

38), and it is one of the most important mediums by which the Arabic language has

maintained its development and productivity (Hammaad, 1983, p. 19) [my

translation].

Although the phenomenon of derivation exhibits considerable complexity, it is not

uncommon to hear Arabic speakers making up new verbs from existing roots for the

sake of creating jokes or being poetic (Hammaad, 1983, p. 39) [my translation].

Thus, it is not unusual to hear such new Arabic words as: 5&.َ.(َ+ (to apply the Lebanese

sample of a sectarian civil war to another country), as in: ‫اق‬,&*+‫ ا‬5&..(+ (to put Iraq in a

sectarian civil war). And 5&‫َآ‬,/‫( أ‬to copy the American style), as in: ‫ك‬,/•&F1 K&„‫( ه&~ا ر‬this

man is copying the American style). Both words are derived by making use of the

Arabic derivative form K


َ )َ*َ8 (fa3lala).
108

What makes this issue so interesting is the fact that a wide range of Arabic speakers

nowadays, especially those who live in countries where English is the dominant

language, are constantly deriving new verbs from English words using the

abovementioned Arabic derivative forms. To take just a few examples:

• The verb ,َ N
َ &P‫ أ‬which is derived from the English verb (to insure) by

using the Arabic derivative form K


َ &َ*8‫أ‬, e.g., %}‫ر‬7ّ-&H ‫ت‬
ُ ,َN&P‫( أ‬I insured my

car).

• The verb ¥
َ .¢ &َ„ which is derived from the English verb (to change) by

using the Arabic derivative form K


َ *¢ 8َ , e.g., ‫ر‬7{ƒ+‫ ا‬¥ّ.„
َ (to change trains).

• The verb K
َYَ &.َ‫ آ‬which is derived from the English verb (to cancel) by

using the Arabic derivative form K


َ &َ*8‫أ‬, e.g., [&EX3+‫ ا‬x
ُ )Y&.‫( آ‬I cancelled the

appointment).

• The verb ¥
َY
¢ &َ/ which is derived from the English verb (to send a

message) by using the Arabic derivative form K


ّ *¢ 8َ , e.g., %.I£Y/َ (send me

a message).

In addition to derivation, the other way whereby the Arabic language maintains its

enrichment and productivity is metaphor (‫ز‬7&I3+‫)ا‬. Arabic relies heavily on metaphors

in producing new words, structures, and collocations with new semantic connotations.

To use Hammaad's (1983) examples: ‫ن‬7&I._+‫( أذن ا‬the ear of the cup), %&H,'+‫ ا‬K&„‫( ر‬the

leg of the chair), 5&„7„Š+‫&‚ ا‬.E (the neck of the bottle), ™N&3+‫ن ا‬7.&H‫( أ‬the teeth of the

comb), ‫ة‬,\9‫ ا‬0-E(the eye of the needle) [my translation].


109

Arab writers, poets and artists typically make use of such metaphorical expressions to

juice up their works or sometimes to avoid using taboo or being rude. They employ

words in a different way by widening/narrowing their semantic meanings according to

the requirements of contexts or thoughts they want to express. In other words, they

impart new life to such words, and inject the language with new items that can be

utilized later on in the everyday contexts (Hammaad, 1983) [my translation].

Let me now turn to the issue of the productivity of PVs in Arabic. As has been

previously shown, the only type of PVs that exists in the Arabic language is that of a

verb + a preposition one, where the both elements of the construction are fused

together to form one semantic unit (see 2.6 above). Such constructions, however, have

their productive nature. The metaphorical usage of theses verbs and prepositions plays

an important role in their productivity. Let me consider this issue by means of some

examples taken from Hammaad (1983):

‫ق‬7H p)E ‫ب‬,Q+‫ ا‬x/7w

(Literally: the war has stood on a leg, meaning, the war has erupted) [my translation]

7Ow7H 0E ‫ب‬,Q+‫ ا‬x_N‫آ‬

(Literally: the war has exhibited its leg, meaning, the war has broken up) [my

translation]

‘1~„7P 0E ,N+‫أ\[ى ا‬

(Literally: the evil has shown his molars, meaning, the evil has been revealed) [my

translation].

To recaptulate the current section, having given a brief account of the derivational

mechanisms in both English and Arabic, one can come up with a number of frultful
110

insights, being: 1) both The English and Arabic languages derive new words by

adding affixes to the roots of basic verbs; 2) unlike English, Arabic can derive new

words by adding/omitting sounds and short vowels to/from the roots of the basic

verbs; 3) unlike Arabic, English does not make use of infixes; 4) both languages allow

for the derivation of nouns and adjectives from verbs; 5) unlike English, Arabic

(among other languages such as Hebrew and Aramaic) has a fixed set of derivative

forms by which new derived items can be created; 6) unlike Arabic, English relies, in

derivation only, on sets of prefixes and suffixes; 7) both languages, like any other

languages, have the characteristic of productivity as a result of word-formation

systems; and 8) while EPVs constitute a highly productive category, their Arabic

counterparts (prepositional verbs) can only be productive when used as a part of

metaphorical expressions.

2.8 PVs in English lexicography

In this section attention will be paid to the ways by which English lexicologists have

treated the issue of PVs in their dictionaries. The aim is to explore the extent to which

such dictionaries have reached in covering, explaining, and providing equivalents to

such problematic items of the English language. The section will fall into two distinct

parts, the first will view the PVs in general English-English dictionaries, and the

second part will discuss the treatment of PVs in the specialized English-English

dictionaries.
111

2.8.1 PVs in general English-English dictionaries

It might be plausible to argue that the ways whereby general English-English

dictionaries approach the question of PVs really merit a careful investigation. Yet, it

is beyond the scope of this thesis to trace the phenomenon in all English lexicography.

Therefore, three main authoritative dictionaries will be looked at, they are: Webster's

Third new International Dictionary (1993), The Oxford English Dictionary (1989),

and The Macquarie Dictionary (2001). The reason for selecting these three

dictionaries is due to the fact that they represent the major three dialects of English:

the first one is a dictionary of American English, the second is a dictionary of British

English, and the third represents a dictionary of Australian English.

A close look at the three dictionaries at hand reveals that they do approach the issue of

PVs relatively well, but it also reveals that they are, by no means, comprehensive in

their coverage.

Gove (1993), in the preface of Webster's Third new International Dictionary,

indicates that verb plus adverb combinations, which "function like one-word verbs in

every way except for having a separate suffix" (p. 4a), have been recognized and

assigned separate entries (p. 4a). This is, to some extent, true and evident throughout

the dictionary. The PV make up, for instance, is accorded a separate entry and given

nine meanings as a transitive verb, and five meanings as an intransitive verb followed

by the noun makeup, which is derived from it, and, in turn, given six meanings (p.

1364). Other examples include back up, which is given three meanings followed by

the noun derived from it, backup (p. 160), and put off, which is given four different
112

meanings followed by the noun derived from it, put-off (p. 1850). On the other hand,

some PVs are not accorded separate entries, such as the PV account for, which is

listed under the entry of the verb account (p. 13). Other examples include the PVs

bitch up (p. 222), egg on (p. 726), switch on, and switch off (p. 2313) which are

tackled under the verbs bitch, egg, and switch respectively. As for the coverage,

however, many PVs are not covered in this dictionary such as log on, log off, sign on,

sign of, sign out and put up with.

Along these lines, The Oxford English Dictionary (1989) handles PVs in very much

the same manner where some of them are specified in separate entries such as back-

up, (vol. I, p. 870), bring about (vol. II, p. 555), bring forward (vol. II, p. 556), bitch

up (vol. II, p. 229), make-up (vol. LX, p. 249) and put up (vol. XII, p. 922). Others

listed under their root verbs such as account for (vol. I, p. 85), pop up (vol. XII, p.

395) and start up (vol. XVI, p. 815) while a number of PVs have been entirely

ignored such as bring along, bring together, dine in, chill out, and click on.

By the same token, The Macquarie Dictionary (2001) accords the majority of PVs

separate entries such as backup (p. 133), count down (p. 440), make up (p. 1156), pull

out (p. 1534), push up (p. 1540), put up (p. 1541), and shut down (p. 1748) etc. PVs,

in some cases, are tackled after the entries of their verbs as in account for (p. 12),

brighten up (p. 241), chill out (340), dine in (p. 536), switch on, and switch off (p.

1900) which are listed under the verbs account, brighten, chill, dine and switch in that

order. While such PVs as bring along, bring together, bitch up, sign in and sign off

have been completely disregarded.


113

2.8.2 PVs in specialized English-English dictionaries

Following Cowie (1993, p. 39), specialized monolingual English-English dictionaries

devoted to PVs "are relative newcomers to the market". In this subsection they will be

taken up in a historical order, that is, the oldest will be treated first. The aim here is to

trace the progress of handling such a phenomenon by dictionaries compilers and to

touch upon the points in common amongst them and the points in which they vary

from one to another. And, most importantly, to explore the gaps, if any, that need to

be bridged.

2.8.2.1 An Analysis and Dictionary of Two-Word Verbs in English


Illustrated in Sentences

Compiled by Abdul Karim Taha, and first published by Kuwait University in 1972.

This book, to the best of my knowledge and research, is the first monolingual English-

English dictionary entirely devoted to PVs in English literature. It has been

republished by Librairie du Liban as York Dictionary of Two-Word Verbs in English

in 1996.

The first part of this book (pp. 1-37) has been set out to investigate the structural and

syntactic properties of two-word verbs (PVs). The main focus has been to classify

them syntactically to transitive and intransitive verbs, and to account for their stress

pattern and word order, "and to point out the important signals which may be used to

distinguish the different meanings of a given two-word verb" (p. vii).


114

The Dictionary of Two-Word Verbs in English Illustrated in Sentences has occupied

the second part of the book (pp. 39-274), whereby PVs are listed alphabetically, each

of which is labeled either transitive or intransitive. Most of the intransitive PVs are

illustrated by one example each owing to the fact that they are non-separable, for

instance the PV burn down (to be destroyed by fire) is illustrated by the sentence The

house burned down (p. 67), and ride off (to depart by riding) by After saying good-by,

he rode off (p. 182).

The transitive PVs, on the other hand, are illustrated by three sentences each due to

the fact that they are separable and may appear in three different word order patterns.

For example the PV act out (to perform) is exemplified in the following three

sentences: They acted out a play; They acted a play out; and They acted it out (p. 39).

And the PV leave out (to omit) is given the following examples: He left out several

answers; He left several answers out; and He left them out (p. 136).

It is worth mentioning to say that the meanings of polysemous PVs, which have more

than one meaning, have been attended to, for example the PV take out is given the

following meanings: 1) to run; to follow; to pursue; 2) to escort; to accompany out; 3)

to extract; to remove from; 4) (in bridge) to bid higher than one's partner in another

suit; 5) to obtain official papers; 6) to enter (a subscription) (p. 238).

As far as register variations are concerned, two labels have been provided throughout

the dictionary, colloquial and slang. For instance the PVs bash up (p. 45), break off (p.

63), and dream up (p. 95) are considered as colloquial, while beard down (p. 49), doll

up (p. 93), and zero in on (p. 274) are considered as slang, where no attempt has been
115

made to account for other variations such as formal, humours, literary etc. as it has

been the case in the other specialized dictionaries. No attempt has been made as well

to account for such crucial features as nouns and adjectives which are derived from

PVs, idioms in which PVs constitute fundamental parts, common words which

typically collocate with each PV, and English dialects other than the American dialect

upon which the analysis and the dictionary are based.

Moreover, the dictionary is by no means comprehensive in that there are many PVs

which have been ignored such as bring along, bring together, juice up, do without,

bliss out, boot up, butt out, factor in, hack into, rock up, slag down, tough out, veg

out, type in, scroll down, and scan in among others.

2.8.2.2 Dictionary of English Phrasal Verbs and their Idioms

Edited by Tom McArthur and Beryl Atkins, and first published by Collins Clear-type

Press in 1974.

The compilers believe that PVs "are more consistent in their behaviour and far more

teachable than is generally supposed" (p. 5). Therefore, this dictionary has been

developed as a teaching material in the 1970-1972 classes of English for foreigners

carried out by the Department of Educational Studies, the University of Edinburgh.

Although the dictionary, according to its compilers, can be used as a general

reference, it is mainly meant to assist foreign learners of English, particularly those

who "have mastered the basic sentence patterns of the language and have an active

vocabulary of 3,000 words in plus" (p. 6). Such students have been advised to use the
116

dictionary in conjunction with a workbook entitled 'Using Phrasal Verbs' developed

by Tom McArthur in 1970-1971 as a teaching material as well, and published in

Collins 'Patterns of English' series (p. 6). However, the Dictionary of Phrasal Verbs

and their Idioms has some noticeable features, which I will now discuss.

In terms of the polysemy of PVs, unlike other scholars who argue that each sense of a

given PVs should be dealt with as a distinct verb, the compilers consider that

"differences of meaning occur within a continuum of meaning, beginning with the

simple verb + particle and ranging through to the more opaque idioms" (p. 6). Such

verbs, they elaborate, "are unique blends of syntax, semantics and idiom, and are dealt

with […] as so many themes and variations" (p. 6). To take just one instance:

The PV fix up is given five different meanings each of which is illustrated by at least

one example as follows: the first meaning is (arrange) where fix up is intransitive, as

in: We've fixed up to go out tonight. They have fixed up to visit us next month. I fixed

up to go a broad for a holiday. I have fixed up for a plumber to come tomorrow. The

second meaning is (fit or put up) as in: The carpenter fixed up the wall cupboards.

The third is (arrange) where fix up is transitive as in: I'll try to fix something up that

suits everyone. Let's fix it all up now. The forth is (provide an opportunity or work

for) as in: Can you fix him up? They fixed me up with this job. The fifth is

(accommodate) as in: They fixed him up in a small hotel. Can you fix her up for the

night? (p. 82).

As far as the coverage is concerned, the compilers admittedly indicate that "[n]o

claims are made to exhaustiveness either in the number of the verbs covered or in the
117

number of senses identified for any verb" (p. 6). Hence, the comparison with other

dictionaries reveals that many PVs have not been covered in this dictionary such as

bring along, bring together, drive through, trade in, chill out, juice up, and stand over

among others.

Some verbs are listed as 'special entries' and marked with asterisks. The compilers

justify such a distinction by claiming that "a large number of phrasal forms are simply

the grammatical operation of verbs of movement plus particle of direction" (pp. 6-7).

For example: "conduct + particle vt sep (convey by escort, with direction) [e.g.,] He

asked the guard to conduct us out. The receptionist conducted us in to meet the great

man" (p. 54) [Emphasis in original].

Further, in their attempt to account for every listed PV expansively, the compilers

provide four kinds of information in each entry about a given verb as follows:

1) A classification of the verb type, indicating whether it is a transitive (with a

direct object), or intransitive (without a direct object). And in the case of

transitive verbs, a distinction is made between separable and non-separable

(fused) PVs. For instance: "cash in vt sep (hand in for money) [e.g.,] She

needed money so she cashed in her shares. It's time to cash in those saving

bonds" (p. 41) [Emphasis in original] where vt stands for transitive and sep

stands for separable. Another example is "swear by vt fus (think highly of,

value) [e.g.,] He swears by the shop. I always swear by their products" (p.

223) [Emphasis in original] where fus stands for fused (non-separable).


118

2) The register variations of each PV are specified in order to show the field

and/or style of language in which a given PV may be employed. Special labels

are utilized to indicate the field such as:

Chem =chemistry; Cine =cinema; Fin =finance; Med =medicine; Mus =music,

etc. And some others to indicate the style such as:

Euph =euphemistic; Fam =familiar, informal; Lit =literal, basic; Pej

=pejorative; sl =slang; Vulg =vulgar.

3) A gloss is provided to each entry. McArthur and Atkins make the point that

"[t]he gloss for a particular entry is meant to be read together with the

illustrative sentences to show the area of meaning occupied by the verb" (p. 7).

Despite the fact that each gloss given to any PV in this dictionary is

considered by the compilers as a grammatical equivalent by which the PV can

be substituted, they admit that "[o]ccasionally the equivalent is not a perfect

match, because the verb may be unique in that sense" (p. 7), and in most cases

the Latinate verb equivalent does not carry "the same features of informality,

formality and vividness as the phrasal verb" (p. 7). As exemplified in the PV

book down in He booked us down for the next ship. This PV is glossed as (put

down in a book), which is not the perfect match, and given the Latinate verb

(register) as an equivalent which is in turn not as informal as the PV itself (p.

26).

4) One or more illustrative sentences. A random opening to page 142 shows two

distinct PVs, one (mark on), which is illustrated by one sentence, while the

other one (measure up), is exemplified in six sentences (p. 142).


119

Some PVs are given a reference to derived verbs, adjectives, nouns and idioms. For

instance, the PV be away, which has an idiomatic meaning, among other meanings, as

in: to be away with the fairies meaning to be slightly mad (p. 16). Similarly, the PV

come in, in to come in handy meaning to prove useful (p. 52).

In a nutshell, this dictionary has attracted a great deal of attention among researchers

and learners alike. Its list and classification of PVs are adopted to serve as the basis

for compiling four of the Collins bilingual dictionaries, namely: English to French,

English to German, English to Italian and English to Swahili (p. 9).

Such increasing attention, in my view, is because of its excellent presentation and

comprehensive treatment of the notion of PVs from a variety of perspectives. The

only weakness it has is the coverage of PVs. The number of PVs covered in this

dictionary is far less than the actual number of them presented in other dictionaries,

let alone the productivity nature of this type of combinations which makes them quite

difficult to be traced and captured.

2.8.2.3 Time-Chambers Dictionary of Phrasal Verbs

Edited by E. M. Kirkpatrick and G. Davidson, and published by Federal Publications

in 1982.

It is worth mentioning to say that this dictionary has been developed from the

Chambers Universal Learners' Dictionary which was published by W & R Chambers

Ltd Edinburgh (p. iv).


120

Time-Chambers Dictionary of Phrasal Verbs is intended for learners of EFL, and, as

it has been claimed by its editors, "[i]t will not only help them to understand phrasal

verbs but will give them the knowledge and confidence to use them" (p. ix).

Nevertheless, in compiling this dictionary, the editors, in my view, have relied heavily

on the abovementioned Dictionary of English Phrasal Verbs and their Idioms by

Tom McArthur and Beryl Atkins (1974) in that they utilized very much the same

classification and label system (cf. pp. vi-x). With very few exceptions, however, the

dictionary has its own distinct features, which I will now discuss.

In the definitions given to the listed PVs, only the figurative sense of each PV is

provided, whereas the literal meaning is omitted. The editors justify such an omission

by claiming that it was "[f]or reasons of space" (p. ix), and they refer learners who

want to know more about the literal meaning to the Chambers Universal Learners'

Dictionary (p. ix). For example the literal meaning of the PV get across (to cross

something) is disregarded and only the figurative sense (to be or make something

understood) is given (p. 39).

And, "[f]or reasons of space and convenience" as well, some pairs of PVs, such as:

drain away and drain off (p. 27); nose about and nose around (p. 77) are dealt with

together, that is, each pair is tackled in one entry.

On the contrary, the dictionary comprises not only PVs (simple verbs followed by

adverbs or prepositions) but also verb phrases (verbs + objects followed by

prepositions + objects) such as: apply to in: apply force to a door; and build on in: I've
121

built all my hopes on this book being published, which are, by no means, regarded as

PVs, even though they are included "because it was felt that they would be useful for

learners" (p. ix).

It should be noted that the issue of polysemy of PVs is treated by numbering the

definitions of each PV "to avoid confusion" (p. vi), for instance the PV knock off has

six different meanings (p. 64), while pick up has nine meanings (pp. 81-82).

It is important to mention, however, that the register variations of the majority of the

listed PVs are indicated by means of labels such as: formal; inf. =informal; sl. =slang

etc. To take but one instance, the PV figure out is labeled as (inf) since it is typically

used in informal situations, as in: I just can't figure it out (to understand) (p. 34),

whereas the PV lie with is labeled formal since it is usually employed in formal

contexts as in: The decision lies with you (to be the responsibility of) (p. 69).

In terms of classification, as in other dictionaries, PVs are given grammatical labels

such as: vt =transitive; vi =intransitive; sep =separable; fus =fused (non-separable) etc.

What is novel, in this dictionary, are the separable transitive PVs (typically labeled vt

sep) have been classified into three categories (p. vii):

1) Transitive PVs, which are obligatory separated, have been labeled as vt oblig

sep as in: push around (p. 86).

2) Transitive PVs, which are usually separated, have been labeled as vt usually

sep as in: get across (p. 39).

3) Transitive PVs, which may be separated, have been labeled as vt sep as in keep

up (p. 62).
122

2.8.2.4 Longman Dictionary of Phrasal Verbs

Compiled by Rosemary Courtney, and first published by Longman Group Limited in

1983.

Three types of PVs are included in this dictionary: 1) verb + adverb such as nail down

as in: We shall have to nail him down to his promise (to state his intention) (p. 397);

2) verb + preposition such as see over as in: may we see over the house/ I understand

that it is for sale (to visit and examine) (p. 543); and 3) verb + adverb + preposition

such as: shape up to as in: It will not be easy for you to change your whole way of

thinking, but I believe that you have the strength to shape up to it (to face it with

courage) (p. 563).

Since such combinations can be either idiomatic or non-idiomatic, the main focus of

the dictionary being on the former while the latter is not covered unless "the verbal

combination also has an idiomatic meaning, then both senses are included" (The

introductory section, not numbered). In order to distinguish between idiomatic

meanings and non-idiomatic ones, an asterix (*) is placed before each idiomatic

meaning.

Besides, idioms such as: let the cat out of the bag (to tell a secret) (p. 354), in which

PVs are employed, are included and printed in bold type.


123

Interestingly, there exist some combinations, which are by no means PVs, that are

included such as: lie low (verb + adjective) and kid oneself (verb + pronoun) (The

introductory section, not numbered).

One of the most important features of this dictionary is listing the nouns and

adjectives which are derived from PVs. The compiler terms them as "Related words"

and places them at the end of the listed PVs, for instance the noun blackout and the

adjective blacked-out are mentioned with the PV black out from which they are

derived (p. 39).

Equally important, the issue of polysemy is dealt with by giving a separate numbered

sense for each meaning. Hence, the majority of the senses are provided as follows:

1- Ordinary meanings

2- Idiomatic meanings, marked with a star *

3- Fixed idioms in which the phrasal verb is used (The introductory section, not

numbered).

Such meanings are explained in simple English by using words from a list of 2000

words cited in the Longman Dictionary of Contemporary English. Each meaning is

illustrated by one or more examples taken from newspapers and books. The examples

are written in a way that they not only help the user to understand the expression but

also to give information on how it is typically used (The introductory section, not

numbered).
124

As far as register variations of PVs are concerned, the compiler provides, in the

beginning of the book, a list of short forms used to indicate four types of notes: 1)

notes such as: AmE, AuE, CanE, and ScoE, to tell the user that the expression is

always used in American, Australian, British, Canadian, or Scottish English; 2) notes

such as: law, mil, and naut, to tell the user that the expression is only used in

particular profession, i.e. law, military, and nautical (used about ships and sailors); 3)

notes such as: humor, and dergo to tell the user that the expression is used to show the

attitude of the speaker, i.e. whether it is humorous or derogatory (showing dislike or

lack of respect); 4) notes such as: fml, infml, and sl, to tell the user about the "level of

use of the expression", i.e. whether it is formal, informal or slang (The introductory

section, not numbered).

Further, cross-references are employed to help the user find other expressions with

related meaning of the one at hand, for example: the PVs book in, book out, and check

out are listed after the definition and examples of the PV check in. (p. 74).

Finally, a great deal of grammatical information is provided by means of codes taken

from Longman Dictionary of Contemporary English. Such codes are listed and

explained at the front of the book. They consist of letters indicating the type of verb,

and numbers indicating what comes after the verb.

Last of all, this dictionary is regarded by Sjoholm (1995) as "the most complete one"

(p. 102). I totally agree with Sjoholm, since the comparison with other specialized

dictionaries reveals that the vast majority of PVs have been tackled in this dictionary.

Yet there exist some newly coined PVs which have not been covered, such as click
125

on, chill out, sex up, and google around. Such a skipping can be justified by the

productive nature of PVs which makes it almost impossible to list them all in one

book.

2.8.2.5 The Student's Dictionary of Phrasal Verbs

Edited by Nigel D. Turton and Martin H. Manser, and first published by Macmillan in

1985.

Intermediate and advanced students who, according to the editors, "want an easy-to-

use reference book that deals with this area of the language" (p. iii), are the intended

users of this dictionary, which is basically a development of two books written by the

late F. T. Wood, namely English Verbal Idioms and English Prepositional Idioms,

which are both described by Turton and Manser as "highly successful" (p. iii).

In their introduction to the dictionary, the editors have made a clear distinction

between idiomatic and non-idiomatic PVs, and made the point that they will exclude

the latter types of PVs "which simply combine the meanings of their parts" (p. iv).

Instead, the emphasis will be on the former "whose meanings and use are especially

difficult for the student" (p. iii). Since, however, there is no clear cut distiniction

between the idiomatic and non-idiomatic PVs and "these are merely labels for the

opposite ends of a scale" (p. iv), they decided to take the comprehensibility as a

"guide for selection in these cases" (p. iv), in that they included only the type of PV

whose "meaning is not obvious from its parts" (p. iv). Yet they introduced two
126

exceptions to this rule: "verbs used only in technical or specialized fields are not

included, nor are verbs which are no longer in common use" (p. iv).

Furthermore, the style-markers Formal and Informal are used occasionally to indicate

the register variations of the listed PVs, where formal indicates that such a PV is used

in "formal written English" (p. viii), such as bear down on/upon (p. 5). While

informal means that a PV is used in such situations as "between close friends or

express the speaker's anger or irritation" (p. viii), such as: barge in/into (p. 5). On the

other hand, students are advised to treat the latter with caution, "since many of these

verbs could cause offence if used inappropriately" (p. viii).

Further, in their endeavour to present the grammatical behaviour of the listed PVs, the

editors have devised a notation system "that shows at a glance how each verb is used

in a sentence" (p. iii). It is a simple guide, providing students "with a clear visual

impression of grammatical behaviour of a verb without involving him in a long

technical description" (p. iv). Two symbols are employed: "a filled circle and an

empty circle" (p. v). Each part of the PV is represented by a filled circle, and each

other word added by the user is represented by an empty circle. (p. v). For instance:

because there is no word that can be added after the PV went out, only two filled

circles are placed before it, representing its two parts, as in the sentence Suddenly all

the lights went out (p. iv), whereas three circles are placed before the PV burst into,

two of them are filled (representing the parts of the PV) and the third one is empty

(representing the added word) as in the sentence The trees burst into blossom (p. 20).
127

This dictionary, in my view, is far from being exhaustive in that the comparison with

other dictionaries indicates that many PVs, which have been covered by others, were

skipped by the editors of this dictionary such as: bitch up, bid up, brighten up, drive

through, chill out, juice up, factor in, slag off, type in, scan in, and scroll down among

others.

2.8.2.6 Collins COBUILD Dictionary of Phrasal Verbs

Edited by John Sinclair et al. and first published by Williams Collins Sons & Co Ltd

in 1989 and reprinted in 1990 (twice), 1991, 1993 (twice), 1994, 1995 (twice), 1996,

1997and 1998.

This dictionary contains more than 3000 combinations of verbs with adverbs or

prepositions, clarifying more than 5500 different meanings (p. v). Such combinations

are considered by the editors as "[…] the combinations which are in common use in

everyday modern English" (p. v). Further, they argue that they "[...] can make this

statement with confidence because the dictionary has been based on a detailed

computational study of the extensive Birmingham Collection of English Texts, now

part of The Bank of English" (p. v). In the foreword of the dictionary Sinclair et al.

list the 48 particles which are included along with the 38 verbs which "[…] are

especially problematic for students of English" (p. vi). All the combinations included

are made up of one verb from the first list and one or two particles from the second

list.
128

One important feature is that the types of combinations covered in this dictionary are

ranging from combinations with highly unpredictable meanings, such as: go off (to

explode), put off (to postpone), and turn down (to reject) to combinations with

reasonably predictable meanings, such as: spread out, link up, and slave away (p. v).

PVs are explained in "simple English". The explanations are designed in a way that

they tell the user about the meanings of PVs, as well as they show him/her how they

are typically used in terms of "what kinds of word collocate with it, what kind of thing

is usually mentioned as the subject or object of the verb, and what sort of sentence

structure it is used in" (p. viii).

In addition, an extra column is provided, to the right of the explanations, to highlight

the grammatical patterns of PVs. Such patterns "are mainly given in frequency order:

that is, the commonest pattern appears first" (p. xiii). The types of information given

in this extra column have something to do with the grammatical behaviour of PVs

such as: transitivity, word-classes of particles, positioning of particles and some

additional structures (pp. xiii-xiv).

Moreover, different meanings of PVs, which have more than one meaning, are

explained and "arranged in order of frequency, so that the commonest ones come

first" (p. viii), for example the PV shoot up is given the following three meanings

which are numbered in order of frequency: (1) to grow or increase very quickly; (2) to

move around in a place shooting a gun; and (3) to inject illegal drugs into oneself (pp.

330-331).
129

At the same time the register variations are clearly outlined throughout the dictionary.

The explanation of a given PV tells whether it is only used formally or informally,

and whether it is found in British or American English (p. viii). The PV yield up, for

instance, is labeled as "a formal expression" (p. 446), while sick up as "an informal

expression" (p. 335).

Since a lot of PVs have synonyms of single words, which are always much more

formal, synonyms and antonyms of PVs are mentioned right after the explanations

and examples. Other PVs may serve as synonyms or antonyms of the listed ones (p.

ix). For example, the verbs install, deposit, invest, and interject are given as more

formal synonyms for the PV put in, whereas the PVs pay in, chip in, and bring in are

given as synonyms which mean almost the same as put in (pp. 280-281).

Besides, nouns and adjectives which are derived from PVs are stated in the same

paragraph of the explanation of a given PV "[i]f they are closely linked with a

meaning of the phrasal verb […]. If there is not such a clear link, they appear in a

paragraph on their own" (p. ix). The noun offprint, for instance, is mentioned with the

PV print off to which it is closely related (p. 263). The noun turnover, on the other

hand, is specified in a separate paragraph away from its basic PV turn over (p. 423).

The most important feature, which makes this dictionary different from the others, is

the particle index provided at the end of it, where "the common meanings that

particles contribute to phrasal verb combinations" are explained, and PVs in which

such meanings may appear are listed (p. ix). In an attempt to account for the

productivity of PVs, the editors provide such an index maintaining that "phrasal verbs
130

are not just arbitrary combinations of verbs and particles. Instead, they fit into the

broad patterns of choice and selection in English" (p. 449). To take just one example,

the particle up is given the following twelve meanings each of which associated with

a group of PVs which share a given meaning: (1) movement and position; (2) increase

and intensification; (3) improvement and preparation; (4) fastening and restriction; (5)

approach; (6) disruption and damage; (7) completion and finishing; (8) rejection and

surrender; (9) happening and creation; (10) collection and togetherness; (11)

disclosure; and (12) separation (pp. 487-490).

In spite of the fact that this index has only concentrated on the meanings contributed

by the particles to the PVs and utterly ignored the meanings contributed by the simple

verbs which constitute the first element of each PV and can also be used in a

productive way, it is, in my view, a plausible remedy for the unavoidable

phenomenon of productivity of PVs which make them hard to be traced and taken up.

It is to be highlighted, however, that this dictionary, like the others, is by no means

comprehensive, since a lot of PVs have not been dealt with in it. Some of the omitted

PVs include: bid up, drive through, dine in, chill out, and juice up among others.
131

2.8.2.7 NTC's Dictionary of Phrasal Verbs and Other Idiomatic


Verbal Phrases

Compiled by Richard A. Spears, and published by National Textbook Company

(NTC) in 1993 and reprinted in 1996.

This dictionary is intended for both the new-to-English users and for fluent speakers

(p. vii). It has been presented as "a dictionary of form and meaning, and not a

dictionary of grammar or categorization" (p. viii). Hence, all types of collocations of

verb + particle are included, as well as both the idiomatic and non-idiomatic

sequences of them are covered.

There are 71 particles used throughout this dictionary some of which are prepositions,

and others are directional adverbs. The list of theses particles contains such words as

aboard, abroad, aground, alike, open, overbroad and still which are considered as

particles (p. xvi). On the other hand, there are 2796 verbs appearing in conjunction

with these particles, to form the 7634 combinations covered in this dictionary (p. vii).

One of its special features is the indication which is given to show whether the object

of a given PV is human, non-human or both, by employing the pronouns someone and

somebody. The compiler points out that such a kind of "information is vital to learners

of English, although it seems to come perfectly naturally to lifelong English speakers"

(p. viii).

The other important feature is the optional elements which are provided in

parentheses to refer to "the extended forms of the verb phrases with the frequently
132

omitted parts" (p. ix). For instance the PV see ahead is followed by the phrase (of

someone or something) in parentheses to indicate that such elements are optional parts

which can be added or omitted to/from the given PV (p. 643).

Furthermore, some of the listed PVs are explained by more than one definition where

the "additional definitions are usually given to show slight differences in meaning or

interpretation" (p. xi). The PV float (up) on, for example, is explained by the

following two definitions "to drift as if on the surface of something; to drift along

through the air" (p. 272).

In addition, the polysemous PVs are attended to by numbering the meanings of a

given PV with boldface numerals. The following three meanings, for instance, are

provided for the PV bring down "1. to move something from a higher place to a lower

place […]. 2. to lower something, such as prices, profits, taxes, etc. […]. 3. to defeat

or overcome something, such as an enemy, a government, etc," (p. 75).

However, each PV is illustrated by two or more of the "carefully written examples"

which are designed to "lead the user to the meaning and appropriate usage of each

expression" (p. viii). The following two examples, for instance, are provided to

illustrate the PV pull over: Betty pulled over to the side of the road and waited for the

traffic to thin; The police officer ordered her to pull over (p. 550).

Moreover, such labels as colloquial, formal, informal, jocular and slang are

introduced to indicate the register variations of the listed PVs. The PV conk off (to fall
133

asleep), for example, is labeled as slang (p. 151). While the PV itch for is labeled as

colloquial (p. 386).

Besides, nouns and adjectives derived from PVs have been accounted for, for example

the PV burn out is followed by the adjective burn-out and the noun burnout which are

derived from it and they are illustrated by the following two examples: Send the

burned-out part back to the factory; One burnout after another! These cheap motors

are not worth it. (p. 91).

Lastly, despite the fact that this dictionary comprises more than 12000 combinations

and is said to be "[t]he most comprehensive dictionary of its kind" (the back cover,

not numbered), it has, in fact, omitted a number of PVs such as: click on, dag out,

factor in, fang up, guck up, hack into, rock up, ramp up, scan in, sex up, and scroll

down among others; it may be, however, that some of theses PVs were coined after

the dictionary was compiled.

2.8.2.8 Oxford Dictionary of phrasal verbs

Edited by A. P. Cowie and R. Mackin, and first published by Oxford University Press

in 1993.

This dictionary is, following its editors, a revised and updated edition. The first

edition of it relied basically on "an analysis of works of fiction, history, biography

etc." (p. 422). This edition, however, has made use of "a variety of contemporary

sources" (p. 422). Data from three new sources has been added: 1) the collection
134

made by the co-author of the Oxford Dictionary of English idioms (Isabel McCaig) in

the late 1970s and 1980s; 2) the quotations illustrating PVs which are held at Oxford

for the updating of the Oxford English Dictionary; and 3) a computerized corpus of 30

million words which covers the period 1989 to 1992 and compiled at OUP as well (p.

422).

The dictionary is chiefly intended for students of English as "a practical and teaching

aid" (p. 428). Such prospective users are directed to the best way of using the

dictionary in systematically presented principles set out in the introductory section

whereby they are guided to where to find complex idioms, nominalized forms (nouns

derived from PVs), collocates of PVs, and synonyms (p. xi). Three sets of

complicated symbols are employed, one for grammar codes, the other for letters

introducing collocate lists, and the third for synonyms (pp. xvi-xvii).

From the very beginning of the dictionary, the areas of difficulty of PVs are

pinpointed and restricted to: grammar, idiomatic or non-idiomatic, and complex

idioms (pp. x-xi). By complex idioms the editors mean using PVs as part of idiomatic

expressions such as: put back in put the clock back (p. x).

The nature of idiomaticity, on the other hand, is chiefly discussed to exhibit the

criteria "used in deciding what to include in a dictionary which has 'idiomatic' as part

of its title" (p. 426). A decision is made to include the items that constitute units of

form and meaning (only idioms or semi-idioms). Therefore, the following types of

combination are considered:


135

1) Verb + particles (with no object following the verb) such as: (of a witness)

come forward; (of an aircraft) take off.

2) Verb + preposition (with no object following the verb) such as: come across

(an old friend); run into (difficulties).

3) Verb + particle + preposition (with no object following the verb) such as: face

up to (one's responsibilities); put up with (interruptions).

4) Verb + particle (with an object following the verb) such as: make (one's face)

up; take (a politician) off.

5) Verb + preposition (with an object following the verb) such as: hold

(someone's past failings) against (him or her); put (someone) off (driving).

6) Verb + particle + preposition (with an object following the verb) such as:

bring (someone) up against (a problem); take (one's anger) out on (someone)

(pp. 427-428).

Like the Collins COBUILD Dictionary of Phrasal Verbs, in Oxford Dictionary of

Phrasal Verbs a special attention is paid to the issue of collocation, i.e. collocates of a

PV or "the words which, in the judgment of native speakers, normally and naturally

combine with it to form sentences" (p. 429). As knowing accurately which noun or

adjective can collocate with a given PV is considered one of the difficulties facing the

English learners (p. xv). For example: the PV carry out typically collocates with the

noun investigation to make the commonly used sentence to carry out an investigation.

On the contrary, the PV carry on always collocates with the noun conversation to

make the sentence to carry on a conversation, but not vice versa (p. xv).
136

It is useful to mention here that the PVs with different meanings are given numbered

entries under one shared headphrase as in the PV pick up which has five different

meanings (to take hold of and raise (sth); to collect (sth); to collect (sth) as wages; to

take (sb) on board; and to rescue (sb) from the sea) each of which is numerated under

the same headphrase (pick up) (p. 439).

Formal, informal, slang and taboo are the style labels given to the listed PVs

throughout the dictionary to indicate their register variations. Such labels, the editors

maintain, "reflect various factors in the situations in which they are normally used" (p.

xix). The main three factors of theses are:

• the relationship between the speakers, or correspondents (remote or official, or


intimate and relaxed).
• whether one is speaking or writing (compare a spoken commentary on a
football match with a newspaper report of it).
• the level of seriousness, detachment, etc suggested or imposed by the occasion
(compare a speech at an official banquet with one given at a farewell party for
a personal friend) (p. 473).

Another noteworthy feature is that this dictionary relies mainly on the British English

as its entries "represent the usage of educated British speakers in the latter part of the

twentieth century" (p. xviii), and no attempt has been made by the editors to include

entries "which are solely, or largely, American" (p. xviii). However, there exist a few

entries marked (US) or (esp US) "which have a marginal status in British English" (p.

xviii) such as: run for, which is marked as (esp US), meaning "offer oneself as a

candidate for (office)" (p. xviii).

Finally, like the abovementioned dictionaries, this dictionary has skipped a number of

PVs such as: bid up, bliss out, boot up, butt out, chew out, chill out, click on, dine in,
137

drive through, fang in, guck up, hack into, pig out, rock in, scan in, juice up, sex up,

sign in, and scroll down among others.

2.8.2.9 Cambridge International Dictionary of Phrasal Verbs

Edited by Michael McCarthy et al. and first published by Cambridge University Press

in 1997.

This dictionary contains more than 4500 PVs presented in a quite logical and

accessible manner. Further, it exhibits several unique features which make it far more

distinct than the other dictionaries in this area.

In their endeavour to make this dictionary "truly international" (p. viii), the editors

have provided coverage of the main three dialects of the English language (p. xiii),

namely: British, American and Australian English by employing the following

regional labels:

1) British, placed after a PV which is only used in British English, such as:

clued-up (p. 49).

2) American, placed after a PV which is only used in American English, such

as: clued-in (p. 49).

3) Australian, placed after a PV which is only used in Australian English, such

as: get into (p. 114).

4) Mainly British, placed after a PV which is mainly used in British English,

such as: contract in (p. 60)


138

5) Mainly American, placed after a PV which is mainly used in American

English, such as: block in (p. 14).

It is evident that a great deal of information is presented with each PV "in a way

which is clear and helpful" (p. vii). Such information includes the inflection patterns

of each PV followed by ample details of syntactic properties given in "an explicit

fashion which does not require the user to decipher complex codes" (p. vii). Thus, in

addition to the provision of the grammatical cases of each PV, such as whether it is a

transitive or intransitive, or whether it is a separable or non-separable, the kind of

object used with each PV is specified. The label sb (somebody) is placed after the PV

if its object is human, while the label sth/sb (something/somebody) is introduced if the

object is human or non-human, whereas the label swh (somewhere) is used if the

object of the PV is a place (pp. viii-ix).

Interestingly, the editors are so comprehensive that they do not surpass such tiny

properties as: situations where PV is always used in the passive form, such as, be cut

out (p. 68). Or situations where PV is always used in the reflexive form, such as: fend

for (p. 95). And situations where PV is always used in continuous tenses, such as: die

for (p. 73).

Moving on to the issue of polysemy, the dictionary groups such different meanings

according to the syntactic patterns of the PV, as in: fill in which has three different

syntactic patterns, each of which carries a distinct meaning (p. 97). Each meaning of a

given PV is illustrated by example sentences taken from the Cambridge International

Corpus to show how it is "used in natural written and spoken English" (p. xi).
139

Besides, in order to make the definition of PVs easy to understand, they are "written

using words from a list of less than 2000 common words" (p. xi).

Most importantly, PVs occur as parts of fixed expressions (complex idioms) are also

clearly shown and well explained, such as: get along in getting along like a house on

fire (p. 110).

Furthermore, nouns and adjectives that are derived from PVs are stated after the PV

from which they are derived, such as: the PV drop off and the noun drop-off (p. 83).

Moreover, as a teaching hint for learners of English, the PVs, which are considered to

be quite common and useful, are highlighted as in: let down; let off; and let out (pp.

172-173).

What is more, register variations of most of the listed PVs are indicated by employing

the following register labels:

1) informal, placed after PVs which are typically "used with friends or family or

people you know in relaxed situations" (p.xiii) such as: stick up (p. 294) and

wash out (p. 334).

2) formal, placed after PVs which are typically "used in a serious or polite way"

(p. xiii), such as: issue from (p. 153) and join with (p. 155).

3) slang, placed after PVs which are typically "used in an informal or not very

polite way, often between members of a particular social group" (p. xiii), such

as: knock over (p. 164) and skin up (p. 272).


140

4) old-fashioned, placed after "phrasal verbs which are still used but sound old

fashioned" (p. xiii), such as: gad about/around (p. 109) and lay about (p. 168).

5) taboo, placed after "phrasal verbs which are likely to offend people and are

not used in formal situations" (p. xiii), such as: screw around (p. 254) and suck

off (p. 299).

6) humorous, placed after "phrasal verbs which are intended to make people

laugh" (p. xiii), such as: dragoon into (p. 79) and hark at (p. 139).

7) literary, placed after "phrasal verbs which are mainly used in literature" (p.

xiii), such as: thirst for (p. 313) and wheel around/round (p. 338).

The most valuable feature of this dictionary is the supplementary material attached to

it, where fifteen "theme panels" are provided in which "phrasal verbs shown in groups

according to their meaning" (p. vii). Such theme panels embrace the following topic

groups: Agreeing & Disagreeing; Computer; Crime; Emotions; Food & Drink; Giving

& Getting Information; Illness; Money; Reading, Writing & Studying; Relationships;

Speaking & conversation; Thinking & Considering; Travel; Weather; and Work (pp.

350-346). Each theme panel begins with a short text contains some PVs which are

commonly used in this particular field. Then, the contextual meanings of such PVs are

given. For example, the theme panel titled (Computer) starts with a short text in which

such PVs as switch on, log in, back up, and print out are used. Then, the contextual

meaning of each PV is individually provided. The meaning of switch on, for instance,

is given as "to turn on an electrical device by using a switch", while the meaning of

back up is "to make a copy of computer information so that you do not lose it" (p.

351).
141

Another valuable feature of this dictionary is the "photocopiable exercises" material

along with its answer key, which really makes the dictionary "a unique resource

which can be used not only for reference purposes but also as a valuable classroom or

self-study learning aid" (p. vii).

However, there exist some PVs which have been ignored in this dictionary such as:

knock off, bid up, click on, drive through, and dine in among others.

2.8.2.10 Oxford Phrasal Verbs Dictionary for Learners of English

First published by Oxford University Press in 2001.

This dictionary has benefited, to a great degree, from all the previous dictionaries by

adopting the best features of each one of them.

Like the Collins COBUILD Dictionary of Phrasal Verbs, which devotes an index to

account for the meanings contributed by each particle, the Oxford Phrasal Verbs

Dictionary for Learners of English has a section at the end of it entitled "Guide to the

particles" (pp. 349-371) whereby the meanings of each particle occurred in the

dictionary are outlined. The purpose of attaching such a guide is to help learners learn

and understand the ways by which PVs are formed and to help them understand the

newly produced PVs when they occur (p. 349).

This dictionary, like the Cambridge International Dictionary of Phrasal Verbs (1997),

provides learners with a "study pages" section along with photocopiable exercises to

help them practice and become more familiar with PVs (pp. 182-183).
142

There are, on the other hand, more than 6000 British and American PVs covered in

this dictionary illustrated by 10500 examples and explained by simple definitions

using a definition vocabulary of less than 3000 words (the back cover, not numbered).

Further, the dictionary covers not only the idiomatic type of PVs, but also it includes

such types as: 1) "[v]erbs which are always followed by a particular particle", e.g. rely

on, crop up, and abide by; 2) "[v]erbs that are followed by a particle in a particular

meaning", e.g. nod off, grow up, and walk out; 3) "[v]erbs with a particle, where the

particle adds to, but does not change, the basic meaning of the verb", e.g. spread out,

fade away, and slave away; 4) "[v]erbs plus particles where each has their normal

meaning", e.g. phone back, and pin up (pp. vi-vii).

Furthermore, idioms, in which PVs are incorporated, are highlighted in this

dictionary. For instance, screw up your courage (forcing yourself to get enough

courage to do something) (p. 249).

Moreover, nouns and adjectives which are derived from PVs are listed. The noun

overpass, for example, is listed with the PV pass over (p. 199), and the adjective plug-

in (as in: a plug-in kettle) is mentioned with the PV plug in (p. 210).

Like all the abovementioned dictionaries, the majority of PVs listed in this dictionary

are followed by labels indicating their register variations (formal, humorous, less

formal, literary, slang, etc.). For instance, the PV strike off is labeled as formal, while

chop off less formal (p. 290).


143

Likewise, information about grammatical patterns of PVs along with common

subjects and objects typically collocate with them is ably accounted for.

In addition, synonyms and antonyms are stated with the vast majority of the listed

PVs. For instance, the PVs end up and finish up are given as synonyms for the PV

land up (p. 163).

Finally, despite the large number of PVs covered in this dictionary, there are some of

them which have not been attended to such as: bid up, bliss out, dag out, rock up, and

slag down among others.

2.8.2.11 Summary of section 2.8.2

All in all, there are many features in common among the abovementioned dictionaries

in terms of handling the phenomenon of PVs. Such similarity is not unusual, Holes

(1994) indicates that:

In many, perhaps most cultures, dictionary making has traditionally been seen
as an accretive activity: you take the work of a predecessor and simply add the
new words and meanings you have discovered without changing the old ones,
which are regarded as 'original' or 'basic' meanings even if they have long
since dropped out of use. This is the way lexicography developed in both
English and Arabic (p. 167).

To recapitulate, the English-English dictionaries of PVs have a number of points in

common, being: 1) all of the dictionaries intended for English language learners, and

to be used as teaching aids. This is the reason behind the usage of simple and plain

English in defining PVs, and the employment of variety of symbols and labels in

verifying their register and grammatical patterns. Holes (1994) maintains that "[…]
144

the compilers' image of the typical user […] affects not only what they include and

how they put it in order but also how they format it with symbols, grammatical

terminology, pictures and so on" (pp. 161-162); 2) the register variations of PVs have

been accounted for unanimously, albeit mostly for the spoken mode; 3) grammatical

information has been presented for each PV; 4) a reference has been given to nouns

and adjectives derived from PVs in the majority of the dictionaries (An Analysis and

Dictionary of Two-Word Verbs in English, Time-Chambers Dictionary of Phrasal

Verbs and the Student's Dictionary of Phrasal Verbs are exceptions); and 5) the issue

of polysemy of PVs has been attended to in the all dictionaries (the Student's

Dictionary of Phrasal Verbs is an exception).

However, the abovementioned dictionaries vary in terms of the following features: 1)

the numbers of PVs covered in each dictionary. For instance the Collins COBUILD

Dictionary of Phrasal Verbs contains 3000 PVs, the Cambridge International

Dictionary of Phrasal Verbs consists of 4500 PVs, the Oxford Phrasal Verbs

Dictionary for Learners of English comprises 6000 PVs, and the NTC's Dictionary of

Phrasal Verbs and Other Idiomatic Verbal Phrases comprises 7634 PVs; 2) the issue

of collocation between PVs and the items typically appear with them have been

investigated only by three dictionaries namely the Collins COBUILD Dictionary of

Phrasal Verbs, the Oxford Dictionary of Phrasal Verbs, and the Oxford Phrasal

Verbs Dictionary for Learners of English; 3) the question of complex idioms,

whereby PV is part of fixed expressions and proverbs, has been accounted for only by

four dictionaries, they are: the Longman Dictionary of Phrasal Verbs, the Oxford

Dictionary of Phrasal Verbs, the Cambridge International Dictionary of Phrasal

Verbs, and the Oxford Phrasal Verbs Dictionary for Learners of English; 4) the issue
145

of synonyms and antonyms of PVs has only been tackled by the Collins COBUILD

Dictionary of Phrasal Verbs and the Oxford Phrasal Verbs Dictionary for Learners of

English; 5) some dictionaries have concentrated only on British English as in the

Oxford Dictionary of Phrasal Verbs, other on American English as in An Analysis and

Dictionary of Two-Word Verbs in English and the NTC's Dictionary of Phrasal Verbs

and Other Idiomatic Verbal Phrases, whereas others have included American and

Australian English such as: the Cambridge International Dictionary of Phrasal Verbs

and the Oxford Phrasal Verbs Dictionary for Learners of English; 6) the dictionaries

have differed in terms of what to cover. Some of them have devoted their books to the

idiomatic type of PVs in particular as in the Longman Dictionary of Phrasal Verbs

and the Student's Dictionary of Phrasal Verbs. Others have widened the scope of their

dictionaries to include the idiomatic and the semi-idiomatic types as in the Collins

COBUILD Dictionary of Phrasal Verbs and the Oxford Dictionary of Phrasal Verbs.

Others have included types which by no means considered as PVs as in the Time-

Chambers Dictionary of Phrasal Verbs and the Oxford Phrasal Verbs Dictionary for

Learners of English; 7) the essential issue of productivity of PVs has been almost

totally overlooked. The only exception is the index provided by the Collins

COBUILD Dictionary of Phrasal Verbs, in which the editors have paid a considerable

attention to such a crucial issue and attempted to provide a plausible solution for it.

Such an endeavour has been replicated in the Oxford Phrasal Verbs Dictionary for

Learners of English. Rooney (2004) points out that "[t]he most exciting challenge for

dictionary editors today is to keep up with the change in our language as new words

come in and linguistic norms and conventions change and develop in response to

technological and cultural innovation" (p xiii). Along these lines Leech (1974)

indicates that "dictionaries are open-ended, and continually being adapted to new
146

requirements by the addition of new lexical entries" (p. 202); 8) although the vast

majority of PVs have been tackled in all specialized dictionaries, there exist some PVs

which have been handled in some dictionaries and ignored in the others, such as bliss

out which has been covered only by the Collins COBUILD Dictionary of Phrasal

Verbs and the NTC's Dictionary of Phrasal Verbs and Other Idiomatic Verbal

Phrases, and factor in, hack into, and veg out which have been covered only by the

Cambridge International Dictionary of Phrasal Verbs and the Oxford Phrasal Verbs

Dictionary for Learners of English; lastly, and most importantly 9) there is a number

of PVs which have not been covered by any of the abovementioned dictionaries. Such

as dag out, fang up, google around, guck up, ramp in, shoo in, scan in, sex up, and

slag down among others. Such PVs constitute a real dilemma for translators as well as

English learners.

Having said that, there exist a number of dictionaries which are not entirely devoted

for PVs, but, however, PVs constitute essential parts of them. Some of such

dictionaries exclusively assigned to the phenomenon of idioms in the English

language, such as: English Verbal Idioms, by F. Wood (1964); English Prepositional

Idioms, by F. Wood (1967); Oxford Dictionary of Current Idiomatic English, by A. P.

Cowie and R Mackin (1975); Dictionary of English Idioms, by M. J. Wallace (1981);

and The BBI Dictionary of English Word Combinations, by M. Benson, E. Benson

and R. Ilson (1997).

Others are mere long studies devoted to the topic of prepositions and adverbial

particles, such as: Prepositions and Adverbial Particles, by J. B. Heaton (1965); and

Prepositions and Adverbial Particles, by L. A. Hill (1968).


147

It needs to be mentioned though that some of theses dictionaries constituted the basis

upon which some PVs dictionaries have been compiled. For instance, Turton and

Manser (1985) the compilers of The Student's Dictionary of Phrasal Verbs have

acknowledged that their dictionary "is a development of the highly successful English

Verbal Idioms and English Prepositional Idioms by the late F. T. Wood" (p. iii), and

the Oxford Dictionary of Phrasal Verbs has been developed from the Oxford

Dictionary of Current Idiomatic English edited by the same compilers in (1975).

It should be noted in this connection that there are a number of internet sites devoted

to the notion of PVs whereby some online dictionaries are included, to mention just a

few of them:

http://www.usingenglish.com/reference/phrasal-verbs/

http://www.eslcafe.com/pv/

http://www.phrasalverbdemon.com/

http://owl.english.purdue.edu/handouts/esl/eslphrasal.html

http://valenciaenglish.netfirms.com/phrasals00.htm

http://www.english-zone.com/index.php

http://www.super-memory.com/sml/colls/pv.htm

2.9 PVs in bilingual English-Arabic dictionaries

In the present section I will deal with the ways by which English-Arabic dictionaries

have tackled the question of PVs. The aim here is to ascertain the extent to which

such dictionaries have arrived in covering, explaining, and providing Arabic

equivalents to such challenging items of the English language. The section will be of
148

two distinct parts, the first one will look at PVs in general English-Arabic

dictionaries, and the second part will examine the treatment of PVs in specialized

English-Arabic dictionaries.

2.9.1 PVs in general English-Arabic dictionaries

In this subsection, three main authoritative general English-Arabic dictionaries will be

looked at. They are: Al-Mawrid, Al-Mughni Al-Akbar, and The Oxford English-Arabic

Dictionary of Current Usage.

2.9.1.1 Al-Mawrid

Compiled by Munir Ba'albaki, and first published in 1967 by Dar El Ilm Lilmalayin.

PVs are not specified in separate entries in this dictionary. They are rather casually

mentioned under their root verbs. The PV to iron out, for instance, is taken up under

its root verb iron (p. 481), and to use up under its basic verb use (p. 1019).

Nouns and adjectives derived from PVs, however, are accorded separate entries. Good

examples include the adjective cast-off (p. 158) and the noun close-up (p. 186).

Al-Mawrid is by no means comprehensive in covering PVs. The vast majority of PVs

have been utterly ignored. To mention just a few instances: bitch up, bring along,

bring together, brighten up, bump up, cast aside, chill out, do without, egg on, factor

in, pig out, and pop up among others.


149

Moreover, a number of the listed PVs are highly polysemous. Such PVs are not given

Arabic equivalents for all their meanings. Most of them, though, are glossed by one or

two meanings only. For example, the PV balled up is glossed only as ‫ّش‬XN&&/

(confused) (p. 85) whereas it has some other meanings as: "to change things so that

something is difficult to deal with" (Cambridge International Dictionary of Phrasal

Verbs, 1997, p. 7), "to make a ball of (a substance) […] to spoil (something)"

(Longman Dictionary of Phrasal Verbs, 1983, p. 15). The PV fly at is glossed as 4„7O1

•&.*\ (to attack violently) (p. 358), while it has another two meanings which are: "to

(cause to) travel by air (a certain height, cost, etc.)" (Longman Dictionary of Phrasal

Verbs, 1983, p. 199), "to suddenly speak to someone very angrily" (Cambridge

International Dictionary of Phrasal Verbs, 1997, p. 103).

2.9.1.2 Al-Mughni Al-Akbar

Compiled by Hasan S. Karmi, and first published in 1997 by Librairie Du Liban.

In the preface of this dictionary, Karmi points out that idiomatic usage of the

headwords, if they have any, listed in sub-entries under them. PVs, he elaborates "are

also listed after the main verb headword, together with the idiomatic usages" (p. xi).

Therefore, we find the PV to use up under the verb use (p. 1563), and under the verb

run we find the PVs to run about, to run across, to run against, to run away, to run

down, to run into, to run on, to run out, to run over, to run through, etc. (pp. 1185-

1186).
150

On the other hand, adjectives and nouns derived from PVs are given main and

separate entries such as the adjective worn-out (p. 1657), and the noun work-out (p.

1655).

As far as coverage is concerned, Al-Mughni Al-Akbar has taken up the question of

PVs in more detail than Al-Mawrid in terms of the number of the listed PVs and the

examples provided to illustrate them in contexts. Yet it is, in comparison with the

abovementioned English-English dictionaries of PVs, far from being comprehensive.

There are, for instance, many PVs which have been utterly skipped, such as: zoom in,

zoom out, win away, mock up, rock up, tough out, print out and chew out among

others.

2.9.1.3 The Oxford English-Arabic Dictionary of Current Usage

Compiled by N. S. Doniach, and first published in 1972 by Oxford University Press.

This dictionary has been described by Holes (1994) as "[…] the most commonly used

work" (p. 163). However, like in the previously mentioned two dictionaries, PVs are

not accorded separate entries in it. They are rather listed beneath their root verbs. The

PV bring up, for instance, is listed under its root verb bring (p. 157), and the PVs look

away, look back, look for, look over, and look through are listed under their basic verb

look (pp. 719-720).

On the other hand, nouns and adjectives derived from PVs are specified in separate

entries such as the noun pull-out (p. 1004), and the adjective lock-up (p. 715).
151

This dictionary, like the others, is by no means comprehensive in its coverage of PVs.

Thus, many of them are skipped, such as: bid up, dine in, juice up, pig out, and scan

in among others.

2.9.2 PVs in specialized English-Arabic dictionaries

It is very noticeable that the quality and quantity of bilingual English-Arabic

dictionaries dedicated to PVs are much lower than their counterparts of English-

English dictionaries of PVs. There are, to the best of my knowledge and research,

only two bilingual English-Arabic dictionaries of PVs. In this subsection they will be

accounted for in historical order. The aim here is to explore the ways by which the

compilers of these dictionaries have tackled the phenomenon of PVs, in order to

compare such ways with the ways by which the compilers of the abovementioned

monolingual English-English dictionaries of PVs have treated the same phenomenon.

In so doing, one can pinpoint the gaps in bilingual English-Arabic dictionaries of PVs

that need to be breached.

2.9.2.1 English Phrasal Verbs in Arabic

Compiled by Kamal Khalaili, and published by Hodder and Stoughton in 1979.

This book seems to be the first one of its kind, i.e. it is the first bilingual English-

Arabic dictionary which is entirely devoted to deal with the question of EPVs.
152

It must be pointed out that the book is intended for students in the Arabic-speaking

world (p. 5). It aims, according to its compiler, "to illustrate the different meanings in

context of a practical and representative selection of the most useful and widely used

phrasal verbs – both in spoken and written English" (p. 5). Therefore, it is divided into

forty-four sections, each of which is devoted to one particular basic verb. Under each

one of such verbs "the numerous combinations it can make with different prepositions

or particles" (p. 5) are listed. The section devoted to the basic verb catch, for instance,

contains the following combinations: catch on, catch out, and catch up (p. 33).

Like English-English dictionaries of PVs, this dictionary has provided grammatical

information about each listed PV to indicate whether it is transitive or intransitive,

and, in case it is transitive, whether it is separable or inseparable.

Furthermore, each PV is "given a definition in straightforward English" (p. 5). The

PV back up, for example, is defined as "give support to" (p. 11), and wear off as

"disappear; pass away" (p. 163). In addition, each definition is followed by two

examples "to fix the context" (p. 5). For instance, the PV hold with (to approve of; to

agree with) is given the following two examples: "We don’t hold with Communism

and all that it stands for. Do you hold with smoking in cinemas?" (p. 80).

Moreover, each listed PV is "translated into the appropriate Arabic equivalent" (p. 5).

For example, the Arabic equivalents 4َ &O8 ،َ‫& ّ[م‬ƒ} ،َK/7Eare given to the PVs: do by (p. 45),

get along (with) (p. 54), and make out (p. 104) respectively.
153

In addition, the polysemic PVs have been accounted for. The PV knock up, for

instance, is given the following three meanings: 1) rouse; awaken ¦&ƒ1‫ أ‬، pّQ&`, 2)

prepare quickly 5E,Y\ [E‫أ‬, 3) exhaust S&OP‫ أ‬، ‚&‫( أره‬p. 87). Whereas the PV pass away is

given the following two meanings: 1) die ‘&(QP p^&w ، ‫ت‬7&/, 2) disappear; vanish ، ‫زال‬

po2} (p. 107).

Another noticeable feature is the exercises provided at the end of each section and

their key which is attached at the back of the dictionary. The purpose of providing

such exercises, as the compiler puts it, is to "give student and teacher extra material to

practise the correct and appropriate usage of these phrases" (p. 5).

No claim, however, has been made by the compiler "to be comprehensive or

academically rigorous" (p. 5). Hence, the forty four basic verbs, covered in this book,

and the PVs produced by combining them with some particles or/and prepositions are

by no means exhaustive. Scores of PVs have been disregarded by skipping their basic

verbs. By skipping the verb fly, for example, such PVs as fly across, fly away, fly off,

fly out, fly over, and fly up, which are produced by combining it with some particles

or/and prepositions, are omitted as a result. And by dropping the basic verb zoom,

such PVs as zoom across, zoom along, zoom in, zoom off, zoom out, zoom over, and

zoom up are dropped consequently.

Despite the fact that this dictionary has much in common with the previously

mentioned English-English dictionaries of PVs in terms of the information provided

to the listed PVs, some crucial information has not been given. No attempt has been

made, for instance, to indicate the register variations of a given PV, and no attempt
154

has been made to account for nouns and adjectives derived from PVs, or the types of

words which typically collocate with them, or to give such information as synonyms

and antonyms of PVs and the complex idioms and fixed expressions in which PVs

constitute integral parts.

2.9.2.2 York Dictionary of Phrasal Verbs and Their Idioms

By Mohammad M. H. Heleil, first published by Librairie du Liban in 2000.

This book is an English-English-Arabic dictionary based on the previously mentioned

Dictionary of English Phrasal Verbs and their Idioms compiled by McArthur and

Atkins (1974).

In his endeavour to provide Arabic equivalents to the listed PVs, Heleil adopts the

work of McArthur and Atkins (1974) entirely from A to Z. That is, the list of PVs,

their definitions, glosses, classification, examples, and the special labels used to

account for the grammatical terms, the field and the style are also replicated. The only

change made by Heleil to the original dictionary, other than providing the Arabic

equivalents, is the division of the dictionary into 26 sections according to the English

alphabet.

It should be remembered that the original Dictionary of English Phrasal Verbs and

their Idioms has been the source of not only Heleil, but also many other

lexicographers and bilingual dictionaries compilers. McArthur and Atkins (1974)

point out that "[t]he list of verbs and the classification adopted [in their dictionary]
155

have also served as the basis for entering phrasal verbs and their translation

equivalents in Collins bilingual dictionaries (English to French, German, Italian and

Swahili)" (pp. 8-9).

Heleil, however, indicates, in the preface of the book, that his work is intended for

Arabic translators, claiming that it is a new of its kind in that it contains, in addition to

the monolingual English-English dictionary where the PVs are glossed in the source

language, the English-Arabic part which can help Arabic translators express the PVs

in the target language (Arabic). Heleil maintains that in so doing he creates a new

method in the field of lexicography. Such a method, according to him, concentrates

on: 1) providing the Arabic translator with the English text as it has been put by the

compilers of the dictionary; 2) providing the Arabic translator with an Arabic text

which can help him in translating the PVs by knowing their contextual equivalents

and their collocations; 3) providing the Arabic translator with a number of synonyms

to convey the shades of meaning; 4) Adding vowel signs to the Arabic text to help the

translator (Arabic native speaker or otherwise) read it correctly; 5) treating the PVs

which have never been tackled by bilingual Arabic-English dictionaries, or have been

partially translated by giving some of their meanings and ignoring the others; 6)

providing the collocations of some PVs which their collocations are not clearly stated

in the English text; 7) providing the Arabic equivalent which expresses the meaning

of a given EPV, and not necessarily constitutes the same grammatical class of it. Such

an equivalent may range between one word and an expression consists of more than

one word (The Arabic preface, not numbered) [my translation].


156

In translating the listed PVs, Heleil does not translate into Arabic the information

given in the English text. Hence, no attempt has been made by him to tell the Arabic

readership whether a given PV is transitive or intransitive, separable or fused (non-

separable), British or American, and formal or slang. He confines himself, in this

regard, to providing only the English text. The Arabic equivalent [‫ء‬7w[&`9‫ زار]ا‬, for

instance, is given to the PV visit with without mentioning that it is a transitive,

separable American PV (p. 297).

Moreover, the majority of the illustrative examples given in the English text are not

translated by Heleil. He gives only the Arabic equivalent of the PV associated, in

square brackets, with a word or some words which typically collocate with it. Most of

such collocations provided by Heleil are mentioned in the English examples. The PV

wet through, for example, is illustrated in the English text by the following two

examples: The rain has wet us through; He's wet through. And translated into Arabic

as: 7&/73} 7&.))\ [,&{3+‫ ]ا‬where the word (rain) typically collocates with wet through (p.

305).

Heleil sets up his own Arabic preface at the beginning of the book, and disregards the

introduction of the original dictionary where valuable information is included such as:

defining the phenomenon of PVs, classifying them, identifying the reasons behind the

difficulties posed by them, indicating their register variations, and outlining the

special features of the dictionary (cf. McArthur & Atkins, 1974, pp. 5-9). Such

ignorance has prevented the Arabic readership from understanding some special

features of the English text. A number of PVs, for instance, are listed as 'special

entries' in the English text and marked with asterisks. The compilers justify such a
157

distinction by claiming that "a large number of phrasal forms are simply the

grammatical operation of verbs of movement plus particle of direction" (pp. 6-7).

Heleil, on the other hand, does not explain to the Arabic readership in his preface

what the asterisks, which appear with some entries, mean.

With regard to coverage, the compilers of the original dictionary have made no claim

to be exhaustive, neither in the number of the listed PVs nor in the number of senses

given to each one of them (McArthur & Atkins, 1974). Therefore the previous

comparison between this dictionary and other specialized dictionaries has revealed

that many PVs have not been covered. As a result, Arabic translators, for whom the

York Dictionary of Phrasal Verbs and Their Idioms has been compiled, will be

confronted by a number of PVs listed in other English-English dictionaries and have

not been covered by English-Arabic dictionaries of PVs such as the one at hand.

2.9.2.3 Summary of section 2.9

To sum up, the comparison between bilingual English-Arabic dictionaries of PVs and

monolingual English-English dictionaries of PVs shows that there are a number of

similarities and dissimilarities between them, being: First, the former is far less than

the latter in both quantity and quality, i.e. in terms of the number of the covered PVs

and the information given to gloss, classify, and illustrate them. Second, like English-

English dictionaries of PVs English- Arabic dictionaries of PVs are designed to meet

the needs of students and learners of English. Third, like English-English dictionaries

of PVs, grammatical information has been given in English-Arabic dictionaries of

PVs with each PV to indicate whether it is transitive, intransitive, separable or non-


158

separable. Fourth, unlike the majority of English-English dictionaries of PVs, English

Phrasal Verbs in Arabic disregards fundamental information such as: the register

variations of PVs, derivation of nouns and adjectives from PVs, types of words

typically collocate with them, synonyms and antonyms of PVs and the complex

idioms and fixed expressions in which PVs constitute integral parts. Finally, like

English-English dictionaries of PVs, Arabic-English dictionaries of PVs are by no

means comprehensive in their coverage. As previously mentioned, there are a number

of PVs which have been skipped over, leaving the translators and learners of English

with no choice but to work them out individually and create Arabic equivalents for

them, which may or may not be accurate.

2.10 Teaching IEPVs

In the present section the question of teaching idiomatic English phrasal verbs to Arab

students will be investigated to identify the extent to which the methods and materials

employed by teachers can help in solving the learning, teaching and translating

difficulties. Yet, before going into this, the methods and materials used in teaching

phrasal verbs to learners of English in general will be explored. The aim is to compare

such methods and materials with those employed by Arab researchers in order to

highlight the gaps that need to be bridged in this difficult area of pedagogy.

2.10.1 Teaching IPVs to learners of English

A great deal of English literature has been devoted to account for the question of

teaching EPVs to non-native speakers who study English as a second or foreign


159

language. This stems from the fact that PVs, especially the idiomatic type of them,

constitute difficulty not only for learners of English but also for teachers, curriculum

designers and material writers in the fields of ESL and EFL alike. Heaton (1968)

makes the point that "[i]t has long been felt that this wide subject constitutes one of

the major areas of difficulty for students learning English as a second or foreign

language" (The preface, not numbered). Further, Cornell (1985), indicates that PVs

"have been 'discovered' as an important component in the curricula for English as a

foreign language" (p. 269).

In what follows a number of representative methods and materials will be

investigated.

2.10.1.1 Tom McArthur's method

In an article entitled Teaching English Phrasal Verbs, McArthur (1971) maintains that

PVs must be taught as units, and as the equivalent of single verbs (p. 71). Further, he

points out that the following points should be taken into account when planning a

course to teach them:

1. Phrasal verbs consist of a root verb and one or two particles.


2. They should not be confused with non-phrasal verbs which tend to take a
certain preposition, such as compromise (with), confess (to), etc.
3. They are both transitive and intransitive, and sometimes the same verb may
function in both ways […].
4. The total meaning of a phrasal verb is seldom simply the sum of its parts […].
5. Compilers of dictionaries have neglected the unitary nature of phrasal verbs
and therefore classed them under their root verbs […].
6. Some phrasal verbs allow free variation in the position of the particle, while
others do not (pp. 71-72).
160

Moreover, McArthur (1971) proposes five specimen exercises which "are not

intended to be exhaustive [but to] serve as an introduction to the phrasal verb" (p. 72).

Each exercise is illustrated by two examples followed by ten sentences, and a

prospective student is asked to make similar changes in the sentences. Exercise 1 is "a

specimen of how to approach freely varying phrasal verb" (p. 72); exercise 2 is a

"specimen of how problems arise with this type of verb when the object of the verb is

a pronoun" (p. 73); exercise 3 is a "specimen of how phrasal verbs may be exchanged

for single verbs" (p. 73); exercise 4 is a "specimen of how single verbs may be

exchanged for phrasal verbs" (p. 74).

McArthur (1971), however, admits that such specimen exercises "do not pursue the

matter as far as it should be taken [since they] ignore the considerable problem of how

a learner can begin to know which phrasal verbs can be divided and which cannot be

divided" (p. 75).

Such a problem, however, has been ably taken up by Tom McArthur himself in his

workbook Using Phrasal Verbs (1975), which is the fourth in the Collins' series of

Patterns of English. The material of this workbook, according to McArthur, "has been

developed out of linguistic research undertaken for Collins Bilingual Dictionaries" (p.

8), whereas its "teaching material has been developed in 1970 and 1971 English

Language Summer Schools run by the Edinburgh University Department of Education

Studies" (p. 8).


161

Further, the book is intended for intermediate and advanced learners of English who

are advised to study it in conjunction with the Dictionary of English Phrasal Verbs

and Their Idioms by McArthur and Atkins (1974).

This book deserves a special attention as it makes a valuable contribution to this field

of pedagogy. It "takes phrasal verbs as a single problem, and as a major part of

English vocabulary and word-formation" (p. 6). It guides the learners through its eight

units step by step in a very systematic manner. In addition to being relatively

comprehensive in covering the main syntactic and semantic features of PVs presented

in a simple language, it provides scores of guided exercises, examples, tables, and

illustrative diagrams.

1.10.1.2 Colin Mortimer's method

In his book Phrasal Verbs in Conversation, Mortimer (1979) approaches the topic

from a quite unique perspective by putting together 432 conversations, each of which

is devoted to one separate meaning of one particular IEPV. One of the most important

aims of this book is to "contextualize individual meanings of a large number of

phrasal verbs in such a way that these meanings will be remembered" (p. iii).

Students, however, are best advised to go through conversations repeatedly as "[t]he

more the conversations are used and discussed […], the more effectively will they fix

in the mind of the student the meaning and use of the particular verbs on which they

focus" (p. iv). Students are also advised to memorize the dialogue so as to give them,

later on, "an opportunity to 'free' the language in the dialogue and to use it in a

consonant, normalized situation arising from his own experience" (p. v).
162

Moreover, Mortimer (1979) gives drills after the conversations, each of which "quotes

from the conversation [it follows] one or two lines involving the use of the phrasal

verb that is featured" (p. v). He recommends that "[t]hese lines should be drilled first

in chorus, then in groups and then in pairs, for pronunciation practice, and to fix the

idiom" (p. v).

What is more, the book is associated with three tapes as "an extensive source of

listening and pronunciation practice" (p. vii), on which 124 selected dialogues are

recorded. Unlike the ways of other tapes recorded for teaching purposes, the

conversations in these tapes are spoken in normal speed with suitable hesitations,

repetitions and interruptions to get "a considerable gain in naturalness" (p. vii).

2.10.1.3 Alan Cornell's method

In his article Realistic goal in teaching and learning phrasal verbs, Cornell (1985)

suggests to assemble a "core" of PVs which "could be arrived at by native speakers

working through a collection of phrasal verbs" (p. 276). He, further, proposes the

following four criteria on the bases of which a selection can be made:

1. Idiomaticity. Is the meaning of the phrasal verb easily deducible from its
constituent parts? (In the case of polysemic combinations each meaning would
of course have to be considered separately)
2. Replaceability. Is there a one-word or already familiar phrasal verb equivalent
which the learner can readily use instead? Can the meaning of the phrasal verb
be easily and naturally paraphrased in already familiar words?
3. Restrictions. Is the phrasal verb subject to severe collocational restrictions
which would have to be learnt? Are there particular grammatical constraints
which have to be observed?
4. Frequency (and usefulness). Is the phrasal verb commonly used? (p. 276)
[Emphasis in original].
163

Cornell considers the last criterion (frequency) as the "overriding" one, and "presents

the greatest difficulties". This is mainly due to the fact that, unlike other aspect of the

English language, there is a shortage of frequency counts for EPVs. The perfect

solution to meet such a shortage is "a computer intelligent enough to scan a corpus

and recognize phrasal groupings and assign meanings to them" (p. 277). But such an

adequately intelligent computer has not been developed yet, therefore "the best

approach would appear to be to consult a sufficiently large group of native speakers

and see what frequency rating is assigned on average to each phrasal verb or phrasal

meaning" (p. 277).

Cornell points out that the core of PVs has to be of two lists, one for active mastery,

and the other for passive recognition. The former needs to receive "the extra practice

and attention necessary" (p. 276), it comprises the commonly used PVs such as: hung

up, put off, put up with, and show off, while the latter contains "phrasal verbs with

complicated restrictions" such as: drink up, go off, hold with, and shape up (pp. 276-

279).

Moreover, Cornell, in this article, ably addresses some didactic problems that PVs

raise. Due to the fact that there exist large quantities of PVs are used in everyday

spoken and written English, he makes the quantitative problem his starting point.

Cornell indicates that such a large number of PVs learners encounter constitutes a real

problem. In his attempt to water such a problem down, he confines it to the fully

idiomatic PVs, in the sense that the non-idiomatic PVs (which constitute the majority)

cause no real difficulty to learners owing to their transparent meanings. What

aggregates the problem, however, is the phenomenon of polysemy in that "[i]t is not
164

only the case that a particular verb + particle combination may be polysemic in having

both an idiomatic and a non-idiomatic use: in addition it may well be polysemic in

having more than one idiomatic use" (p. 270). Hence, Cornell raises the question of

"How many idiomatic phrasal verbs does an advanced learner know on average?" (p.

271). To answer this question he refers to the research he conducted in 1980 in which

he tested a group of his German ESL students "to establish their active knowledge of

selected idiomatic phrasal verbs" (p. 271) [Emphasis in original]. The result showed

"a widespread ignorance" of the 60 PVs tested. He concludes that "the learning of

phrasal verbs at school and university is generally not very successful" (p. 273). The

reason behind that, according to him, is the "limited contact with phrasal verbs", and

exposing students "to such a bookish form of the language" (p. 273).

The other didactic problem Cornell outlines is the interference between L1 and L2 (in

his case German and English) where PVs sound "illogical" for learners. As an

example, he cites: "why should one be laid up with illness when one is lying down?"

(p. 274).

He then investigates some of the semantic and collocational problems learners face

when dealing with PVs such as: the question of one-word equivalent. It is obvious

that some PVs have one-word equivalents, e.g., pull up which corresponds to stop,

and put up with to tolerate. Whereas other PVs have no such equivalents, they rather

have PVs equivalents, e.g., make up for which corresponds to compensate for, and put

in for to apply for. Or, otherwise, they have to be paraphrased (p. 274).
165

On the other hand, quite a few of one-word or PV equivalents can be deemed as

alternatives for their PVs. Such a problem is related to "the degree of synonymity"

since "synonymy is generally recognized as being a very relative concept" (p. 274).

To use some of Cornell's examples:

lie in does not merely mean "to stay in bed", but "to stay in bed beyond
one's normal time for getting up".
put up with: unlike tolerate it cannot be used in a positive manner (to
tolerate other people's opinions is not the same as to put with other
people's opinions) (p. 274) [Emphasis in original].

Finally, Cornell outlines the grammatical problems caused by the following syntactic

restrictions of PVs which typically "represent a considerable teaching and learning

load" (pp. 275-276). To cite just two of his illustrative examples:

come by cannot normally be used in the passive, unlike its equivalents


acquire and obtain.
do with can only be used with can or could in the sense of need: with could
it only has a potential sense and does not refer to the past; it cannot be used
in the passive (p. 275) [Emphasis in original].

2.10.1.4 Richard Side's method

In his paper Phrasal verbs: sorting them out, Side (1990) argues that the difficulties

PVs create for learners "are sometimes increased by the way in which phrasal verbs

are presented in course books or by teachers telling students that they will just have to

learn them by heart, thereby implying that there is no system" (p. 144). Thus, he

begins his article with criticizing the traditional treatment of PVs in course books in

which PVs are grouped according to the verb along with a definition and an example

for each one. Students, however, are advised to match the phrasal verb with its

definition and to learn them by heart (p. 144). Unfortunately, students in such cases,
166

stick to the Latinate definition given to them, and ignore the Anglo-Saxon PVs since

the Latinate verb is "easier to learn, particularly if it is related to a word in the

students' own language, and seems to make more sense" (p. 145). Another bad aspect

of the traditional approach is the random way by which teachers teach particles, Side

gives the following example:

A teacher recycling recently learned vocabulary is quite likely to ask 'Can


anyone give me a phrasal verb meaning arrive starting with turn?' Students
may then shout out the first particle which comes into their heads and this will
continue until one of them hit the jackpot with up (p. 145).

Such ways of treatment, according to Side, aggregate the students' negative attitude

towards PVs, who already dislike the issue of PVs for such reasons as their

idiomaticity, confusion, polysemy, register or appropriacy, grammatical conditions,

etc. What is more, the traditional approaches make the students see PVs as random

combinations of verbs and particles, which is completely incorrect. PVs are not so

random. There is, rather, a system behind forming them and a close look at the

function of particles shows the patterns underlying their combining with verbs (Side,

1990).

Newly coined PVs, according to Side, are not invented randomly. They are rather

"formed by analogy with existing phrasal verbs" (p. 146), and "it is possible to isolate

areas of meaning by finding the connections between them" (p. 146).

The particle, for Side, is "integral to the meaning of the phrasal verb and in some

cases carries more weight of meaning than the verb" (p. 146). That is, the

communicative function of the PV is mainly carried by the particle (Side, 1990).


167

In his attempt to pinpoint patterns underlying PVs, he takes up three particles (off, out,

and up) to illustrate how the system of forming PVs by these particles works. He, for

instance, gives the particle off five lexical meanings (indicating distance in time or

space, departure, removal, disconnection and separation) illustrated by the following

examples, to make the point that "[m]ost phrasal verbs with off fit into this pattern" (p.

148):

Strain off the liquid = removal, separation


The area was fenced off = separation (from surrounding area)
The plane took off = departure
I've been cut off = disconnection (telephone)
It's time to knock off = departure (from work)
Warn sb off = distance in space
The meeting was put off = distance in time
Come and see me off = departure, separation (p. 147) [Emphasis in original].

Some of PVs are ambiguous in nature. But, however, they could be understood by

analogy with other PVs from the same pattern, e. g., the PV ease off in: You should

ease off a bit "could be by analogy with taking one's foot off a car accelerator" (p.

148), and took off in: his business really took off "could be by analogy with an

aeroplane taking off" (p. 148).

What has to be noted here is that not all PVs with the particle off can fit easily in this

pattern. Some of them make Side (1990) admittedly declare that "[p]ersonally, I can

find no convincing place for these within the overall definition" (p. 148). Such

exceptional PVs are exemplified below:

He tried to buy me off


Stop showing off
You are always telling me off
I must dash off a letter (p. 148) [Emphasis in original].
168

Moreover, not all particles are as straightforward as off. The particle out is a good

example where one cannot formulate a single overall meaning for it (p. 148).

Therefore, "it is sometimes necessary to think laterally, metaphorically, or even

pictorially" to understand the system in which PVs work (p. 147).

In his endeavour to find out more patterns, Side quite often refers to his own

experience. For example, in outlining the highly idiomatic meaning of cough up, he

narrates a real story relating an incident that happened to him when he was a child: "if

I choked on my food, my father would thump me on the back and cheerfully cry

'Cough it up, it may be half a dollar!' "(p. 150). And, in explaining the PV hung up in:

She hung up (to put the phone down) Side indicates that it "at first seems strange until

one remembers what old fashioned telephones looked like" (p. 150). Consequently,

Side concludes that the traditional approach is inadequate "either in that it fails to

create learnable patterns, or in that it creates patterns of the wrong kind" (p. 150).

1.10.1.5 Peter Dainty's method

In his textbook Phrasal Verbs in Context, Dainty (1991) claims that "a new method

for learning phrasal verbs" (p. 5) is offered. The book is in three parts. The first part

contains "a specially written cartoon story in which 325 common phrasal verbs are

introduced in a tale of adventure, love, money, crime, honour and blue Rolls Royce"

(p. 5). Such a cartoon story is of fifteen chapters each of which is ended up with some

follow-up exercises and grammatical notes.


169

The second part of the book, on the other hand, is devoted to "an extended blank-

filling revision exercises based on the cartoon" (p. 5).

The third part is dedicated to the answers of the exercises along with a list of the 325

PVs used in the cartoon story. This textbook is associated with a tape on which the

whole story is recorded.

Interestingly, Dainty (1991) claims that if the learner memorizes part of the story by

heart and does the follow-up exercises, the 325 PVs can become a part of his

everyday language as he develops "a more natural and more instinctive command of

English" (p. 5).

2.10.1.6 Martin Shovel's method

In his book Making Sense of Phrasal Verbs, Shovel (1992) implements the

"illustrations and question-prompts" method. Throughout the twenty units of the

book, IEPVs are accounted for in chunks, that is, each unit "introduces and practises

six separate phrasal verbs" (p. 4). Each PV is exhibited through one or two lively

cartoon illustrations followed by a number of question-prompts, which "are designed

to focus the learner's attention and help him or her make an informed guess at the

meaning of the phrasal verb" (p. 4). To take only one example, in explaining the PV

take after, a cartoon picture with a man standing next to his son, who looks exactly as

same as his father, is presented along with the following question-prompts:

Do you think these two people are related?


What do you think their relationship is?
Do you think they look a like?
Make a sentence describing the way the small boy looks compared to his
father.
170

Think of another way of saying take after.


Now turn to page 95 [reference section] to check your answer.
(p. 10) [Emphasis in original].

This book, which is intended to increase the confidence of the students of English as a

second or foreign language at the intermediate level, is appropriate to be "used for

self-study, for pairwork, for conventional class or group teaching, and as a reference

book" (p. 4).

In addition, the presented PVs are listed alphabetically at the back of the book in a

dictionary-like reference section where each one of them is given the following:

- a list of words and phrases that can be used with [it]


- a clear definition
- a context sentence or sentences related to the introductory illustrations
- easy to read structural information showing the positioning of noun
phrases and pronouns (p. 4).

Surprisingly, unlike other scholars, Shovel (1992) avoids the employment of the

grammatical classifications of PVs, claiming that "such classifications are often more

complicated and difficult than the phrasal verbs are used to teach" (p. 4).

Lastly, each unit is ended up with a practice section where a variety of exercises are

included. Such exercises "are very controlled to begin with and then gradually lead to

free-production" (p. 4). Students are advised to study the PVs introduced in the unit

before doing the practice section.


171

2.10.1.7 Malcolm Goodale's method

In his workbook Collins COBUID Phrasal Verbs Workbook (which accompanies the

previously mentioned Collins COBUID Dictionary of Phrasal Verbs), Goodale (1994)

points out that "[t]hough the workbook can be used on its own, more benefit will be

gained by working closely with the Dictionary" (p. iv). He approaches the practice

"through the individual particles, as featured in the Particles Index of the Dictionary"

(p. iv).

Goodale (1994), also, makes it clear that owing to the fact that adverbial PVs are

almost always the most important type and the most difficult for learners of English to

understand, "prepositional phrasal verbs are not included in this workbook" (p. iv).

Hence, the adverbial particles only are accounted for in alphabetical order throughout

the ten units of the book. Each unit has an introduction in which the important

meanings of a given particle are provided along with a list of the PVs to be taken up

in the sections of that unit. Every section is devoted to one category of meaning.

Given that most PVs are polysemic and have "as many as 20 different meanings", it is

quite normal to see a phrasal verb appears in many different sections (p. iv).

Further, there is a section attached to each unit called "Other Meanings" which

includes PVs "which are too common to be excluded, but which do not clearly fit into

any particular category of meaning" (p. iv). As an example, the particle over is given

two sections each of which is assigned to one particular category of meaning

(Considering and Communicating, as in: look over, put over, talk over, and think over,
172

and Changing and Transferring, as in: change over, hand over, take over, and win

over), and a third section which is assigned to other meanings, which include: get over

with, pass over, run over, and smooth over (p. 87).

2.10.1.8 Berman and Kirstein's method

In their textbook Practical Idioms: Using Phrasal Verbs in Everyday Contexts,

Berman and Kirstein (1996) design the whole book as chunks of dialogues between

Pat, an instructor, and Lee, a talkative student.

Berman and Kirstein (1996) consider that the quickest way for learning PVs is "to

practice them by families" (p. xi), as long as "it is a psychological axiom that

learning related material is much easier than learning unrelated material" (p. xii)

[Emphasis in original]. They, as a result, suggest two families in which PVs may be

grouped: family A where PVs are listed alphabetically according to the verbs they

begin with, e.g., get about, get across, get around, get back, get on, get out, etc., and

family B where PVs are listed according to the particles, e.g., back out, get out, give

out, learn out, pass out, throw out, etc. (p. xii). Berman and Kirstein (1996) make the

point that "[f]amily A is more familiar arrangement-dictionary style […and] fine for

the purpose of reference" (p. xii). However, "[f]amily B makes far more sense as a

learning strategy" (p. xii). Therefore, they opt for treating PVs in this book by

particle, claiming that it is "much more likely to find similarities of meaning […]

among verb phrases [PVs] having the same particle than among verb phrases

beginning with the same verb" (p. xii).


173

2.10.1.9 Peter Hannan's method

In his paper Particles and gravity: phrasal verbs with 'Up' and 'Down' Hannan (1998)

employs the 'experientialism' approach which is a philosophical / linguistic approach

outlined and applied by George Lakoff and Mark Johnson in their studies Metaphors

we live by (1985), Women, fire and dangerous things: what categories tell us about

the mind (1987), and The body in the mind: the bodily basis of meaning, imagination

and reason (1987) (as cited in Hannan, 1998, p. 22).

Hannan (1998) attempts to explain how these two particles "have the uses they have",

and discuss a number of implications of the experientialism approach for teaching

PVs (p. 22).

The essential idea of this approach, as he puts it, is that "the meanings of our language

[English], and indeed our structure of thought, are built up from regular patterns of

bodily experience, extended into the other realms which we inhabit such as the

emotional, mental and social" (p. 22). Such an extension, he elaborates, "is principally

metaphorical, and what is preserved from the original physical pattern is the structure,

or relationship between the elements, or some important association, and not the

specific content" (p. 22). To illustrate this idea, Hannan (1998) provides two examples

as "standard examples of metaphors" (p. 22): The line of people snaked around the

corner, and The dawn of a new era. Where the line, in the first example, "does not

have scales or a forked tongue, but the winding form of the snake" (p. 22), and, in the

second example, "there is no literal sun or light, but a noticeable beginning, perhaps

where new knowledge is involved" (p. 22).


174

Moving on to the particles up and down, Hannan (1998) makes the point that by the

term 'phrasal verbs' he means not only non-literal meanings of verb-particle

combinations, but also the literal meanings of them, as part of his thesis is that "there

is a continuity and extension from the literal meanings of verb-particle combinations

to the metaphorical meanings" (p. 22). Hence, he deals with the literal meanings of up

and down as basic and central meanings for their metaphorical extension. Therefore,

he begins with the literal meanings of these particles in such PVs as go up, come up,

climb up, stand up, move up, go down, come down, climb down, sit/lie down, fall

down etc. where up and down literally "refer to our experience of gravity" (p. 23). He,

then, moves on to take up the graduation of the meanings of these two particles from

literal to metaphorical from different standpoints, being: 1) Quantity: This metaphor is

exemplified by our experience when we add objects to a pile the level of that pile goes

up, while when we take away objects its level goes down. Such an experience "leads

us to associate more with up, and less with down [and such an] association is extended

to non-physical things to produce a simple correspondence of 'up = more, down =

less' ", as in: turn up, and turn down (the volume / heat) (p. 23); 2) Size: This

metaphor is stemmed from the fact that physical size of anything around us reflects its

power, in that big size means strong and powerful whereas small size means the

contrary. Such an experience leads to associate up with big and powerful, and down

with small and weak, as in: "bring up (children), [and] bring down (cause someone's

fall from power / respect; lower the tone or moral level of a conversation, etc." (p.

24); 3) Body posture: This point is built up on the fact that "[o]ur physical posture is

obviously related to our activities and to our mental emotional state" (p. 24). In the

sense that "[w]hen we are standing and moving around, we are active [while] when

we are lying down we are inactive, and sometimes passive […] So there is a natural
175

association between 'up' and 'active' and 'down' and 'inactive' " (p. 25), as in: wake up,

get up, start up, open up, calm down, settle down, shut down, break down, etc. (p. 25).

Body posture, on the other hand, is related to our mental and emotional state where an

erect and open posture reflects bright, lively and cheerful states of mind, while a

bowed or slumped posture reflects dull, tired and sad states of mind. This fact makes

us to associate up with happiness etc., and down with sad etc., as in: cheer up, be up,

feel up, be down, feel down, let down, etc. (p. 25); 4) Perspective: Owing to the fact

that close objects seem bigger, in the human vision, than far ones, and when they

move closer they appear to go up in the visual field, one can associate up with

nearness, and down with distance, as in: come up, and go down (p. 26); and 5)

External environment: This point is stemmed from the fact that the "ground is home

[of human], and high places are less frequented and inherently dangerous" (p. 26).

Consequently, "[…] 'down' is associated with what is familiar, real, easily reached or

touched, known, and 'up' with the contrary" (p. 26), as in: bring up, come up with,

bring down, get down to, etc. (p. 26).

Hannan (1998) concludes "that usually literal meanings are basic and central, and that

metaphorical extensions can be understood and systematised with reference to the

central meaning" (p. 26). And he introduces some implications for teaching of PVs,

they include: 1) despite the fact that the systems of meaning accounted for in this

approach are not 100% logical, they are "comprehensible in terms of human

experience, generalisable, and often universal" (p. 26). The approach, as a result, is

"opposed to the type of superficial use of quantitative information about frequency of

use and collocations which simply says 'These are the common usages, Learn them' "

(p. 26); 2) lack of logic and sense of PVs in the eyes of students make them "respond
176

to phrasal verbs with various degrees of pain" (p. 27). Therefore, proving to students

"that there is a human logic, based on experiences which they can recognise, gives

them confidence that it is feasible to learn these things, and open doors to useful

methods of vocabulary storage and organization" (p. 27); 3) highlighting such

physical experiences makes the process of explaining PVs easier. For instance, to

explain come up and go down, the teacher may move towards students closer and

closer till they move their heads up to see him, and so on. The advantage of this

process is that "relatively abstract concepts are grounded in direct sensory experience

and so stick better" (p. 27); 4) it does not matter how to sequence PVs in a syllabus -

"same verb, various particles; same particle, various verbs; random verbs in context"

(p. 27) - what really matters is that "literal or near-literal meanings are generally

presented earlier than metaphorical ones" (p. 27); and, finally, that 5) all the patterns

outlined in this approach can be grasped easily; "[t]his can lay the foundation for a

positive and exploratory attitude to phrasal verbs in general" (p. 27).

2.10.1.10 Darwin and Gray's method

In their article Going After the Phrasal Verb: An Alternative Approach to

Classification, Darwin and Gray (1999) assert that "[i]n research and pedagogy,

approaches to the phrasal verb have been, and still are, rather arbitrary" (p. 66). The

reason for such an arbitrariness, according to them, is "[…] the understanding of the

phrasal verb, by both students and instructors, has not progressed as far as it might

have if a more systematic approach had been used" (p. 66). They, therefore, have

concerned themselves with providing such a systematic approach.


177

In their attempt to clarify the problem of the lack of progress in understanding of PVs,

Darwin and Gray (1999) attribute the problem to the following three reasons: 1) the

definitions provided for PVs by researchers produce conflicting results, and lead to

confusion for both students and instructors; 2) the frequency of the commonest and

more needed PVs has not been determined. As a result instructors, curriculum

designers, and researchers are left with no choice but to use their intuition, which may

or may not be correct; and 3) the method of grouping PVs according to the verb.

Although such a method may help learners understand the idiomatic nature of PVs, "it

does little to promote their use" (p. 67).

Further, they point out that in order to avoid ambiguity in classification procedure of

PVs "linguists must agree upon a definition, thereby requiring them to begin from the

same point" (p. 67). Consequently, they adopt the definition produced by Quirk et al.

(1985) as the standard whereby "[a] phrasal verb consists of a verb proper and a

morphologically invariable particle that function together as a single unit both

lexically and syntactically" (Darwin & Gray, 1999, pp. 76-77).

They criticize the nine traditional tests proposed by Bolinger (1971), (previously

outlined in section 2.5.1 of this Chapter) maintaining that they admit noteworthy

exceptions which cause "a problematic lack of agreement among those who study

phrasal verbs as to exactly which verb + particle combinations are or are not included

in the category" (p. 75). Such a disagreement "can seriously impair the learning of

phrasal verbs by ESL students, preventing the placement of verb + particle

combinations in a grammatical paradigm" (p. 75). Hence, they confirm the real need

for a more systematic classification that can "promote greater agreement among the
178

experts and better presentation of verb + particle combinations to the ESL learner" (p.

75).

Darwin and Gray's alternative approach is "to take the opposite stance" (p. 75). That is

to say, instead of "excluding a verb + particle combination from the phrasal verb

category until it is proven to belong, linguists should consider all verb + particle

combinations to be potential phrasal verbs until they can be proven otherwise" (pp.

75-76). In doing so, they explain, two advantages that can be accomplished: 1) a

degree of definiteness can be added; 2) a curriculum-based confusion students have

can be eliminated (p. 76).

Moreover, in their attempt to clarify their new approach, Darwin and Gray (1999) set

out seven tests focusing on semantics, phonology and syntax. In addition, they

indicate that there is no need to apply all these tests to all combinations; one test is

enough to divide up a combination (p. 77). The tests in brief are:

1. Particle repetition, e.g., *I looked up, up, up your name. [PV]

I looked up, up, up to the very highest point [Not PV]

2. Where questions, e.g., He ran up the rally. Where? Up the rally [Not PV]

I looked up the address. Where did you look? *Up the address. [PV]

3. Fronting, e.g., He made up a story. *Up he made a story. *Up a story he made

[PV]. Up the tree he went [Not PV]

4. Verb insertion, e.g., He pulled on the lever, but it was stuck. He pulled and

jerked on the lever, but it was stuck [Not PV].

I really messed up on my test. * I really messed and fouled up on my test. [PV]

5. Adverb insertion, e.g., * The mine caved quickly and forcefully in. [PV]
179

They crept slowly and silently down the hall. [Not PV]

6. Stress, e.g., she RAN UP a huge bill. [PV]

She RAN to the park. [Not PV]

7. Intonation units, e.g., *I passed / out in the doctor's office. [PV]

I hid / behind the door. [Not PV] (pp. 77-81) [Emphasis in original] [My

bracketing].

Interestingly, in their response to some critiques rose by some scholars (which will be

outlined later in this section), Darwin and Gray (2000) elaborate in more detail on

their approach. They warrant their choice of Quirk et al's definition by claiming that it

is "the most concise representation of definitions presented by others working on

phrasal verbs" (165), and expect that such a definition "would lead to agreement about

which verb + 'something' […] combinations to include in the category of phrasal

verb" (p. 165) in order to establish a list of PVs that ESL learners are more likely to

encounter. The definition consists of two parts: grammatical part where the verb +

particle combination functions as a simple verb; and lexical part where the

combination of verb + particle functions as "a single lexical item with a meaning

significantly different from that carried outside the combination" (p. 166). For

convenience, Darwin and Gray (2000) utilize the following features: V + X

combination (where X represents particle, adverb, and preposition), [+ G]

(representing a grammatical unity of the combination), [+ L] (representing the lexical

unity of the combination). Thus, the definition of Quirk et al, according to Darwin and

Gray (2000) "defines only an ideal, a phrasal-verb prototype" (p. 166) where a PV has

to be [+ G, +L] not [- G, -L], [+ G, - L] or [- G, + L] V + X combination. In so doing,

Darwin and Gray limit their list to only those "prototypical phrasal verbs" (p. 166).
180

Their new method is to exclude any combination that exhibits any negative feature [-

G] or [- L]. Any test of the seven tests proposed by them would be enough to

demonstrate inclusion or exclusion of any given combination.

Thus, the lists of PVs built up by utilizing this method overlook many combinations,

which exhibit the abovementioned negative features. Darwin and Gray (2000) justify

such an omission by claiming that it reduces the "conflict between definition and

example in the pedagogical tools produced" (p. 167).

Using freshman humanities textbooks as their corpus, Darwin and Gray utilize their

abovementioned tests in frequency count to develop the list of frequently occurring

PVs.

Nevertheless, Darwin and Gray's method has been heavily criticized by Joan Sawyer

(2000) and Ron Sheen (2000), who both agree with them on the question of choosing

the most frequent PVs list to be taught to ESL learners. However, they both reject the

method of teaching PVs proposed by Darwin and Gray, and instead each one has

proposed his own method as in what follows.

2.10.1.11 Joan Sawyer's method

In her reply to the article of Darwin and Gray (1999), Sawyer (2000) denies the ruling

out of the semantically transparent constructions and the concentration only on the

semantically opaque ones (those which function as single units). She considers that

scholars should open the membership of the class of PVs to include all types of
181

combinations of verb and morphologically invariable particles because the

semantically transparent combinations have the surface structure as the more

semantically opaque ones, can "lead students to understand the surface structure of

the combinations" (p. 152). This in turn can reduce "avoidance of these combinations

on the part of students and gives teachers a simple functioning of the semantically less

transparent combinations" (p. 152).

By adopting Fraser's (1976) view, Sawyer indicates that all PVs have a verb and a

morphologically invariable constituent, but this constituent could be a preposition, an

adverb or real particle (the element that forms a unit with the verb), as illustrated in

the following set of examples:

The cowboy shot up the hill. (preposition)


The cowboy shot up the bullet. (adverb)
The cowboy shot up the saloon. (real particle) (p. 153) [Emphasis and
pracketing in original]

Sawyer believes that knowing the type of the particle "not only helps predict which

combinations might be easily taught but also suggests the order in which they might

best be presented to students" (p. 155). The "[a]nalysis of their different function

offers teachers one way of considering which ones to teach and suggest a step-by-step

process for doing so" (p. 157). Therefore, she outlines her method of teaching PVs,

according to the particle type, in the form of three steps, which are outlined below.

Step 1: Do Not Teach Verb-Preposition Combinations as Phrasal Verbs. Verb-

Preposition Combinations, according to her, mere "standard verbs followed by PPs

[prepositional phrases]" and "they do not present a challenge for comprehension or

production of this group". In addition, ruling them out "from the class of phrasal verbs
182

leaves only the combinations that include adverbs and real particles requiring specific

instruction" (p. 155).

Step 2: Teach Verb-Adverb Combinations. Sawyer notes that the student knows

the meanings of both elements of this type of combinations since it is semantically

transparent and its elements retain their original meanings. Besides, like the case of

verb-real particle combinations, these combinations have "word-order alternation"

(i.e. they can be split/separable or non-split/non-separable). Teaching a word order of

such semantically transparent combinations gives learners confidence in that when

"they begin working on less transparent types, they no longer need to be concerned

with word order" (pp. 155-156).

Step 3: Teach Verb-Real Particle Combinations. In her endeavour to account for

this step, Sawyer refers to her work Verb-adverb and verb-particle constructions:

Teaching and acquisition (1999) in which she studies child language acquisition and

demonstrates that real particles have at least three common functions: 1) telicity or

completiveness, as an example, the particle up in: eat up and drink up whereby the

object "is consumed completely". Teachers are advised to teach each one of these real

particles by offering a number "of verbs with which the real particle has the telic

reading" to encourage the students "to see a pattern that they can use to decode new

combinations encountered" (p. 156); 2) real particles which do not add much

semantically to the verb, such as: clean up, lock up, wash up, act out, sort out, and

start out. Teachers as well are advised to "demonstrate each real particle that works

this way with a set of verbs to which it adds little semantically" and the students also

"would see that these are not isolated cases but show a pattern" (p. 156); and 3)
183

idiomatic combinations which are "the most difficult to organize into groups for

presentation; they must be presented in context" and have to be learned individually.

For example, give up, think up, and wear out (p. 156).

Sawyer (2000) concludes that "[t]he fact that real particles can be clustered into

groups by function […] makes teaching more efficient" (p. 157). Such a method "may

help students learn patterns for decoding new combinations and increasing their

vocabularies while reducing their avoidance of these combinations" (p. 157).

2.10.1.12 Ron Sheen's method

In his reply to the article of Darwin and Gray (1999), Sheen (2000) admits that they

"make a valuable contribution to teaching phrasal verbs […] in pointing out the

unreliability of choosing such items as curriculum content based on intuition and in

emphasisng the need to base such a selection on authentic frequency of use" (p. 160).

He considers "[s]uch a selection would result in a bank from which one might choose

a restricted list for active use and much longer one for passive" (p. 164).

Sheen, on the other hand, criticizes the approach of Darwin and Gray (1999), claiming

that it addresses the complexities of PVs without touching upon the major question:

"[w]hat is the best way to achieve familiarity with and fluency in the use of PVs,

which is the absolute essential to a mastery of English?" (p. 161). In his attempt to

address such an issue, Sheen proposes his method which is called "explicit-plus or

explicit-minus?". By 'explicit-plus' he means "should teachers devote time and effort

to enabling students to analyse PVs both syntactically and semantically?" (p. 161),
184

and by 'explicit-minus' he means "[s]hould teachers be content with the minimalist

approach […] which would teach the word-order problems with transitive PVs and

leave the rest to exposure, memorization, and practice?" (p. 161).

Reporting on his own experience as a teacher at university level, Sheen claims that he

applied both abovementioned approaches. He first adopted the explicit-plus approach

in which he devoted a great deal of time and effort to teaching syntactic and semantic

complexities discussed by Darwin and Gray (1999), and McArthur (1979).

Meanwhile, he spent some minimal time on classroom oral work encouraging

students to use PVs outside. He then, conducted some written and oral tests. The

results yielded a success in written proficiency, but such a success did not apply to

oral proficiency which was poor and not encouraging.

On the contrary, when Sheen decided to adopt the explicit-minus approach in which

he spent most of the time on oral activities, the results of oral proficiency were far

better. Such an empirical experience has convinced Sheen to argue for the explicit-

minus approach, which "allowed the students to reach a standard nearer to that of

Anglophones than did the explicit-plus approach" (pp. 161-163). Therefore, he

concludes that "teachers need to devote time and effort to activities encouraging

frequent and spontaneous use of PVs and not to the sort of analyses involved in

exploiting the classification system proposed by Darwin and Gray" (p.164).


185

2.10.1.13 Rosemary Sansome's Method

In her paper Applying lexical research to the teaching of phrasal verbs, Sansome

(2000) summarizes the insights obtained from research conducted in the Lexical

Research Unit, Leeds University 1980-1984. The research has taken up a large sub-

group of PVs, that is, combinations of verbs with collocates in which the latter

changes the meaning of the former "in a systematic way by subordinating it to a new

meaning introduced by the collocate" (p. 56). However, such meaning-changing

collocates, according to Sansome, include not only adverbial particles; adverb;

preposition; and prepositional phrases, but also adjectives and nouns (p. 61). The

following two examples provided by her to illustrate such collocates:

He tricked her into taking her medicine. Trick into means: "to get someone to

do something by tricking them" (p. 60).

She tricked him out of a fortune. Trick out of means: "to get something out of

someone by tricking them" (p. 60).

Sansome (2000) maintains that the issue of 'meaning-changing collocates' has been

outlined in the works of linguists who dealt with the phenomenon of PVs such as

Bolinger (1971) and Fraser (1976), but it has not been described systematically. In the

sense, both Bolinger and Fraser have approached the issue from a syntactic viewpoint

excluding verb-preposition combinations from their scope (p. 60).

Concentrating on analysing the meanings of PVs grouped into only one major

subsection of the semantic area CONTACT, that is, PHYSICAL CONTACT, the
186

research yields insights into the pattern underlying PVs belong to this subsection.

Such a pattern, as Sansome puts it, is:

Nearly all the meaning-changing collocates in the PHYSICAL CONTACT


area change the meaning of the verb according to the same patten: 'to___ by
___ ing' (e.g., pull apart 'to separate by pulling'; pull up 'to raise by pulling').
The verb-meaning is subordinated to a new verb-meaning introduced by the
collocate (pp. 60-61) [Emphasis in original].

Sansome, however, believes that such an insight has practical applications in the field

of EFL teaching. She herself, as an EFL teacher, has conducted a comprehension test

given to first-year undergraduate Japanese, Korean, and Taiwanese students to find

out "to what extent foreign learners of intermediate level and above have absorbed the

pattern underlying these verb-collocate combinations and whether they cause

comprehension problems" (pp. 61-62). The result showed that students have not

acquired the pattern underlying the tested combinations. This matter "has implications

not only for comprehension but also production; if students are not aware of any

underlying pattern, they are unlikely to be able to use verb combinations except in

cases where they have learnt the whole phrase" (p. 63).

2.10.1.14 John Flower's method

In his practice book Phrasal Verbs Organiser, Flower (2000) makes a noticeable

contribution to the field of teaching PVs. Unlike other scholars who prefer to treat

PVs either by particle or by verb, Flower treats more than 700 PVs in three different

ways, i.e. by particle, by verb, and by topic. He rightly makes the point that "[t]he

more different ways you meet these verbs, the more you will learn" (p. 3). In treating

PVs by topic, for instance, he groups them according to the field they are commonly
187

used in. In the section of "Technology and Computing", for example, he accounts for

such PVs as cut out, filter out, wire up, print out, back up etc. (p. 89). While in the

section of "Sport and Leisure" he takes up such PVs as worm up, ease up, play off,

stretch out, pass through, etc. (p. 92).

Moreover, Flower (2000) provides, other than the mini-dictionary of the used PVs, a

"Test Yourself" section (pp. 108-112), where five tests are set up to give the learners

an opportunity to examine themselves in what they have studied throughout the book.

In addition, he establishes a "Your Personal List" (pp. 135-144) section, where nine

well-organized blank tables are given to allow students to add their own PVs, or PVs

that they learn in class or come across while they read.

2.10.1.15 Summary of section 2.10.1

Having explored a number of methods proposed for teaching EPVs to non-native

speakers of English, one can come up with a number of insights, being: 1) pedagogues

vary in what to teach as PVs to foreign students. Scholars like Hannan (1998), Sawyer

(2000), and Sansome (2000) strongly believe that all types of verb-particle

combinations (literal and idiomatic/semantically transparent and semantically opaque)

have to be taught. They are driven by the reason that the former constitutes the central

and the basis upon which the meaning of the latter can be grasped and understood.

Other scholars, on the other hand, like Darwin and Gray (1999) consider that only the

idiomatic/semantically opaque type of PVs has to be taught as it is the prototypical; 2)

pedagogues, also, vary in how to teach EPVs to non-native learners of English. Such
188

scholars as Side (1990), Hannan (1998), Sawyer (2000), and Sansome (2000) are

totally convinced that PVs have to be taught by knowing the patterns underlying

them, in order to pinpoint the system and the logic by which they work. Others

believe that PVs are random combinations of verb and particles and they have to be

memorized by heart; 3) the issue of how to sequence EPVs in textbooks, as well,

constitutes a debatable point. Such scholars as Side (1990), Goodale (1994), Berman

and Kirstein (1996), Darwin and Gray (1999), Sawyer (2000) and Sansome (2000) are

quite persuaded that PVs must be dealt with by particles. On the contrary, others

prefer to tackle them by verbs. Interestingly, Flower (2000) believes that they have to

be presented in different ways, i.e. by particles, by verbs, and by topics; 4) scholars as

Cornell (1985), Darwin and Gray (1999) and Sheen (2000) call for frequency counts

of the EPVs, just like other aspects of the English language, to determine the most

common and needed ones, and in turn to avoid designing pedagogical tools according

to pedagogues' intuitions; 5) despite the fact that most of the specialized dictionaries

of PVs are developed as teaching materials to be utilized in classes of English as a

second or foreign language, there are some workbooks written to be studied in

conjunction with specialized dictionaries, such as Using Phrasal Verbs by McArthur

(1975) and Collins COBUILD Phrasal Verbs Workbook (1994), which are produced

in conjunction with the Dictionary of Phrasal verbs and their Idioms, and the Collins

COBUILD Dictionary of Phrasal verbs respectively; and finally 6) most of the

researchers agree on the necessity of teaching PVs in context owing to the fact that

presenting them in contexts enhances their learnability far much more than presenting

them as unrelated elements.


189

In short, the undeniable fact, in my view, is that PVs are not random combinations of

verbs and particles. There is, it is true, a pattern underlying each one of them. Though

these patterns vary in their degree of comprehensibility, they undoubtedly need to be

further investigated, and applied so as to provide learners of English in general, and

Arab learners of English in particular, with reliable pedagogical materials.

2.10.2 Teaching IEPVs to Arab students

As a result of the growing interest in the English language all over the world, and in

the Arab world in particular, as a means of communication in such vital domains as

business, transport, science and technology, teaching EFL has been boosted in the

majority of the Arab countries (Kharma & Hajjaj, 1989, p. 1). The Arab learner of

English spends approximately six to eight years learning the language "at a rate

ranging from four to eight 45-minute periods per week" (p. 1). In theory, this amount

of learning English should enable him/her to use the language perfectly. But,

unfortunately, this is not the case (pp. 1-2). Kharma and Hajjaj (1989) attribute such a

failure to a number of reasons, they include: 1) the majority of teachers are Arab

teachers of English. Their experience as trainees and teachers of the English language

is "limited to English-as-a-foreign-language teaching/learning situations, […with]

little, if any, genuine use of the language as a means of communication" (p. 2); 2) the

exposure of the Arab learners to the English language is "limited and lacks continuity

[…] it is exposure to formal/conscious instruction and learning with little, if any,

exposure to natural language" (p. 2); 3) the Arab learners' motivation of learning the

English language is very little in comparison with their motivation for acquiring a first

language (p. 2); 4) the Arab learners' attitude to EFL is a negative one, i.e. it is for
190

them "a 'school subject' rather than a means of communication" (p. 2); 5) the very low

pass mark indicates that "learners can proceed to further learning of the language

without having first mastered fully what they ought to have mastered in, say, a given

school year" (p. 2) and 6) the English language is taught to the Arab learner "years

after he [she] has already started learning formally, and acquiring informally, his [her]

mother tongue" (p. 2). In Addition, Kharma and Hajjaj (1989) introduce what they

call it the "idealisation of teaching materials" as a critical factor that leads to the

difficulty of teaching English to Arab students. By 'idealisation' they mean

The drill and practice exercises [utilized] are even more 'idealised' in the sense
that they do not represent communicative situations but exemplifications of
language structures. The result is that when learners have to use language at a
level higher than the sentence they fail to do so correctly, whether the task at
hand is composition writing or an extended monologue or dialogue (p. 8).

Along these lines, Holes (1994) summarizes the "Language-learning attitudes" of

Arab students "who have been through a government school education" in three

points as follows:

(a) perscriptivisim and dichotomization: language structure, phrases, and


words, whether in their native language or a foreign one, are classified as
'correct/incorrect', 'beautiful/ugly', 'classical' (= good)/'colloquial' (= bad),
'literary' (= good)/'slang' (= bad), etc. These black-and-white value
judgments are regarded as 'facts' regardless of evidence from actual usage.
(b) fixation on the lexicon: learning a language is viewed as essentially
learning lists of words; technical language is regarded as the same as
ordinary language, except for a lot of technical vocabulary (mainly
nouns).
(c) lexical equivalence between languages: it is believed that any given
lexical item in one language has a single 'correct' translation equivalent in
another; and that understanding and translating mainly involve learning
and using these unchanging, context-independent equivalences (p. 165)
[Emphasis in original].

In their book Errors in English among Arabic Speakers: Analysis and Remedy,

Kharma and Hajjaj (1989) investigate the kinds of difficulties Arab learners typically
191

encounter in studying English as a foreign language. The book is intended for "those

who will be or who are already engaged in teaching English to Arab students" (p. 1).

Thus, Kharma and Hajjaj (1989) attempt to "identify and organise those problems that

face Arab learners of English and to provide teachers with a description of them and

ways of solving them" (p.2).

Given that the question of EPVs constitutes one of such problems Arab learners

usually face, Kharma and Hajjaj (1989) assign a small section to briefly describe them

and suggest ways of teaching them to Arab students. They determine two main factors

as "Sources of difficulty" of PVs, as follows:

(a) sometimes the whole phrase in English is equivalent to a single verb in


Arabic (without any preposition) e.g.,:
be off ‚){.1
be out of his mind 0 ّ Iُ1
be over %OF.1
Adverbial particles are not used in Arabic in a similar fashion.
(b) At other times, the preposition used with the English verb is different from
its usual Arabic equivalent, e.g.,:
be after ‫ ل‬p*Y1
do without 0E %.ŽFY1
get rid of 0/ ’ّ)nF1
look for 0E •Q(1 (p. 70).

However, it is quite obvious, from the literature reviewed above, that the sources of

difficulty of EPVs to non-native speakers of English by far exceed the two sources

given by Kharma and Hajjaj. In fact, most, if not all, of the didactic problems of PVs

outlined by Cornell (1985) apply to Arab learners of EFL.

Moreover, Kharma and Hajjaj (1989) propose two suggestions for teaching EPVs to

Arab learners:
192

The teacher must make his [her] own list of those phrasal verbs that occur in
the textbooks, in oral discussions, dialogues, home-reading material, etc.
which the students have special difficulty with, and to try to deal with them in
different ways. For the teacher who is a non-native speaker of English and
who wants to ascertain the meaning and use of some of those phrasal verbs,
the Longman Dictionary of Contemporary English is a good source, whilst
Courtney's Longman Dictionary of Phrasal Verbs is a recent and very
comprehensive treatment of such verb (p. 72).

These suggestions are, in my view, neither practical nor recent in the sense that, as for

the first one, by advising the teachers to make their own list of PVs, Kharma and

Hajjaj give them a permission to rely on their own intuition in determining which PVs

may cause difficulty. Their intuition, it is true, may or may not be correct. Such an

approach, as previously mentioned, has been subject to a great deal of criticism by

pedagogues (cf. Cornell, 1985; Darwin & Gray, 1999) who proposed, instead, a 'core'

of PVs based on frequency counts.

As for the second suggestion, the two dictionaries recommended by Kharma and

Hajjaj are no longer the best and the most comprehensive ones. There are, as

previously outlined (see 2.8.2 above), many dictionaries of PVs now available which

are far much better and more comprehensive.

Surprisingly, Arabic books which are set up to teach Arab students the essence of the

English grammar and vocabulary have utterly overlooked, or slightly touched upon

the question of EPVs. To take just two examples, Taahir Al-bayyatii (1982) in his

book What Pupils Need in English, which is intended for beginners and advanced

Arab students of English, has totally ignored the issue of PVs in spite of its vital

importance to meet the needs of his intended readership. Other example is Ahmed

Mamdouh Al-Saghir's book English for all levels (1993) where PVs are assigned an

undersized section called "The Most Essential Phrasal Verbs" (pp. 324-329) in which
193

neither the syntactic nor the semantic nor the pragmatic properties of EPVs are

accounted for. It is rather a mere unrelated list of verbs followed by particles arranged

in alphabetical order and translated into Arabic.

The only workbook dedicated to EPVs, written in Arabic, to the best of my

knowledge and research, is Hind El-Nagar's book Phrasal Verbs (1998), which is

intended for Arab students at the intermediate level, and for those who prepare for the

first certificate (FCE) and (CPE) at Cambridge University (p. 6).

It is quite important to mention, however, that in her endeavour to present EPVs to

Arab learners of English in the best possible way, El-Nagar (1998) replicates the work

done by Malcolm Goodale (1994) and the method utilized by him in the previously

mentioned Collins COBUILD Phrasal Verbs Workbook without any

acknowledgement to Goodale's work or to the Collins COBUILD Dictionary of

Phrasal Verbs upon which the workbook has been based.

Hence, she sequences PVs, like him, by particles arranged in alphabetical order. Like

him as well, she organizes her workbook, in ten units. In the first nine units she takes

up the following particles one by one (away, back, down, in, off, on, out, over, and

up). She assigns the tenth unit to "other particles". Each unit deals with one specific

particle and subdivided into sections each of which is assigned to a particular category

of meaning of that particle illustrated by example sentences and translated into

Arabic. Unit five, for instance, is devoted to the particle off (0&E ‫&[ًا‬-*\) and divided into

four sections. The first section is assigned for the first category of meaning of the

particle off, which is "Leaving and beginning K-Z,+‫([ء وا‬+‫"ا‬, the second is for "Rejecting
194

and Preventing ].3+‫“ وا‬8,+‫"ا‬, the third is for "Stopping and canceling ‫ء‬7Ž+J‫ّ• وا‬wXF+‫ "ا‬and

the fourth is for the "Other meanings ‫ى‬,y‫ن أ‬7*/" (pp. 49-54). Each section contains the

PVs of a given category of meaning. For example, the third section of the particle off

embraces the PVs break off (•&wXF1) and call off (%&Ž)1). Each of which is followed by

some example sentences such as:

a- Let's break off for ten minutes. ‚L7w‫ د‬,NE ‫[ة‬3+ 5Z‫ا‬,)+ •wXF.+ ¡‫أ‬
b- He broke off and didn't start again. 7ً -P7ˆ ‫([أ‬1 4+‫• و‬wX} [ƒ+ ¡‫ب‬
c- She has broken off the engagement. 5({n+‫ ا‬xnY8 [ƒ+ ¡‫ج‬
d- They have broken off the negotiations. ‫ت‬7ˆ‫د‬7Q3+‫ا ا‬X_w‫[ أو‬ƒ+ ¡‫د‬
(pp. 53-54) [Emphasis in original].

As far as the exercises are concerned, unlike Goodale (1994) who provides a variety

of exercises, El-Nagar (1998) offers only one type of exercises, that is fill-in-the-

blank type, where a number of sentences are listed each of which has a gap needs to

be filled with a PV from the three alternatives provided.

Another drawback is, unlike Goodale who utilizes authentic example sentences taken

from COBUILD database, El-Nagar makes up most of her examples in such a way

that they sound unreal or, to use Kharma and Hajjaj's word, "idealised" which are far

from representing the communicative situations students need to learn. To provide

only a few examples:

He just kept on hitting the boy although I told him to stop (p. 60) [Emphasis
in original].
Even if you fail in the examination again you must try to keep on until you
pass (p. 60) [Emphasis in original].
The coat should keep out the cold (p. 71) [Emphasis in original].

Some of the given examples have grammatical mistakes, such as:

Nothing will be done until someone have [sic] the courage to speak out (p.
77) [Emphasis in original].
Time move [sic] on slowly (p. 61) [Emphasis in original].
195

Furthermore, there exist a number of translational pitfalls. For instance: The garden is

coming on nicely is translated as: K&-3„ ‫&[ّم‬ƒ} %&8 5&ƒ1[Q+‫( إن ا‬The garden is in progressing

nicely) (p. 61), and I was surprised at how well you all get on is literally mistranslated

as: 5&(-W 4'F&w2E %‫ ه‬4‫ آ‬0/ 7N‫[ه‬./ x.‫[ آ‬ƒ+ (I was surprised at how good your relationship is)

(p. 63), while I must brush up my French before I go to Paris is translated as: ‫\&[ أن‬M

%F-YP,8 0E ‫ر‬7(Ž+‫_“ ا‬P‫( أ‬I must dust off my French) (p. 135).

Having known all these facts, one can conclude that the methods of teaching EPVs to

Arab learners is, to a great extent, inadequate in comparison with the methods of

teaching them to learners of English from non-Arabic backgrounds. The lacking of

rich pedagogical materials and utilizing old fashioned teaching methods are the main

two reasons behind such inadequacy. For this problem to be solved two things need to

be done: first, owing to the fact that the phenomenon of PVs constitutes one of the

most important features of English grammar and vocabulary, and to the fact that

underestimating, overlooking or ignoring them can lead to a major difficulty in

learning, using and mastering the language, they must be included as an integral part

of the learning and training process, and of the interpreting and translating courses

undertaking by Arabs. Second, Arab teachers, curriculum designers and material

developers need to make use of up-to-date teaching methods of EPVs (some of which

are mentioned above) to provide Arab learners of English, interpreters and translators

with systematically presented patterns along with sufficiently rich pedagogical

materials that can help them overcome the difficulties EPVs cause, and in turn

enhance their ability to receive/understand and produce/translate EPVs into Arabic.


196

2.11 PVs in translation studies

This section will be devoted to address the essential issue of how the phenomenon of

IEPVs has been dealt with in translation studies, and, most importantly, how it has

been explored in translation theory.

Given that the phenomenon of PVs is regarded as one type of the English idiomatic

expressions, and constitutes an integral part of many idioms, proverbs and fixed

expressions, it has been investigated by linguists who studied the question of

translating English idioms into other languages.

The available relevant literature, in this regard, can be categorized into two distinct

types: studies dedicated to translating EPVs into other languages; and studies devoted

to translating English idioms as a whole into other languages. In what follows the two

types will be taken up respectively.

2.11.1 Translating EPVs into other languages

In his paper On the Semantic of Certain English Phrasal Verbs and Their Rendering

into Spanish, Bernstein (1974) looks at some EPVs from their semantic standpoint.

By focusing on idiomatic kind of EPVs whose meaning "cannot be rendered by a

literal translation into Spanish" (p. 59), he attempts to highlight the idea of "how

diverse the translations into Spanish may be when a phrasal verb family has many

members" (p. 59). Bernstein (1974) divides PVs into families according to their

proper verbs. He takes up two families as specimens, they are: 'come' plus
197

prepositions family, and 'get' plus prepositions family along with the Spanish

translation of each meaning of the combinations listed under each family. The

diversity of translation of such IEPVs into Spanish made him to indicate that "[t]here

is no overlap between Spanish and English" (p. 61). However, he suggests a "simpler

means" for translating the listed EPVs into Spanish. Such a means is applicable only

to certain cases where "Spanish families of compound verbs with an invariant root

and varying prefixes may be of use" (p. 61).

Due to the fact that Spanish and English languages share "a considerable etymological

debt to Latin" (p. 62), Bernstein (1974) claims that "it will often be convenient to

translate a cognate by a cognate" (p. 62).

Yet, the parallel pair of families of EPVs and Spanish cognates is not available in

EPVs families where numerous members are included, as in the 'get' plus prepositions

family (p. 62).

He concludes, however, that English and Spanish "do share a common SAE [Standard

Average European] characteristic" (p. 64), where "a vast array of actions and relations

may be generated by application of a relatively small number of prepositions/adverbs

to basic verbs" (p. 64). Also, a number of Latin prepositions "passed down to Spanish

through Vulgar Latin […such prepositions] form a productive sub-class that can

generate new verbs or create new verbal meanings by being prefixed to basic verbs

much as English […] may do with its phrases" (p. 64). Like English, he adds, which

"can multiply its verbal lexicon to suit needs as they arise" (p. 65), Spanish can do the

same but "to a much more limited extent" (p. 65).


198

In considering such a crucial issue, Heliel (1994), in his paper Verb-Particle

Combinations in English and Arabic: Problems for Arab Lexicographers and

Translators, enumerates a number of syntactic and semantic problems that Arabic

translators may encounter when dealing with the phenomenon of EPVs. These

problems occur as a result of the differences between English and Arabic with regard

to such a phenomenon, and include: 1) the Arabic translator finds it difficult to

envisage the effect of the particle on the meaning of the verb in an English idiomatic

combination. Whereas, unlike English, The Arabic verb is "a fair guide to the

meaning of the combination" (p. 146); 2) a lot of verbs in English verb-particle

combinations are employed idiomatically with certain particles, "which makes their

meanings unstable and indistinct" (p. 147); 3) some verbs, such as 'get', can combine

with almost every particle. Others, such as 'make' and 'do' "are more selective but still

quite versatile" (p. 147). Many combinations, such as 'pig out = overeat' are

productive; 4) a lot of idiomatic usages are exclusive to a single language, "where

they may sound natural to native speakers but strange to non-native speakers" (p.

147). This is appropriate to English verb-particle combinations where "the verb by

itself would have a radically separate interpretation" (p. 147) as in: the audience

cracked up at every joke versus the vendor cracked the coconut with a machete (p.

147); 5) an English verb may have a range of different meanings in various

combinations, which "may be wider and more idiomatic or even opaque in English

than in Arabic" (p. 147). Write off, for instance, has the following different meanings:

to send off a letter, to clear from the records, to judge to be a failure, and to reject as

beyond repair (p. 147); 6) English verb-particle combinations "in which a normally

intransitive verb takes a direct object puzzle the Arabic translator, who finds nothing

similar in Arabic" (p. 147) as in: the students laughed versus the students laughed off
199

their failing grades (p. 147); 7) some English verb-particle combinations are

employed in a specialized meaning in particular fields. Such a meaning "is usually

definite and its metaphorical motivation has faded, leaving it stylistically and

emotionally neutral, e.g. 'pull out' (military), 'blast off' (aerospace), 'lock down'

(computer science)" (p. 148); hence, the extent to which a register may affect the

meanings assigned to EPVs need to be assessed (p. 148); 8) some intransitive English

verb-particle combinations "may be hard to understand without further clues from the

text" (p. 148). For example, I don't know how the story got about, the opposition was

digging in, and they are still hanging around (p. 148); 9) while many English verb-

particle combinations can be employed both intransitively and transitively keeping the

same meaning, as in: they help out at the church sale, and they helped us out, some of

them can be employed both intransitively and transitively but in different meanings as

in: the engine cut out, and I cut out some photographs from magazines (p. 148); 10)

the meaning of transitive English verb-particle combination "may differ according to

whether the object is a person or a thing" (p. 148). For instance: take someone in, and

take something in (p. 148); 11) the verb + particle, in three-word combinations of

prepositional PVs, "may be much harder to render than the preposition" (p. 148) as in:

look up to (someone), put (someone) on to (a good idea), and come up against (a

problem) (p. 148); 12) the Arabic equivalents of English verb-particle combinations

vary according to their collocations with other words (p. 148). The Arabic equivalent

of the PV break off, for instance, is ]&{w if the following word is (negotiations),

whereas it is §Y&8 if the following word is (an engagement), and the Arabic equivalent

of the PV bring about is ‫&[ث‬Z‫ أ‬when the following word is (a change), while it is ‚&)y

when the following word is (an opportunity) (p. 148); 13) the figurative combinations

of verb + particle are "hard to translate when they are culture-bound" (p. 149). For
200

example, the literal use of lagged behind in the tired boy lagged behind the group,

versus the figurative use of it in prices are rising sharply while incomes are lagging

behind (p. 149).

Heliel (1994), however, concludes that "Arab[ic] translators are likely to encounter

problems finding Arabic equivalents that respect idiomatic meanings, syntactic

structures, lexical collocations and the specialized fields of discourse" (149). He

further says that such problems "should be treated in bilingual dictionaries and

translator training programmes [… and we] must work to develop both our linguistic

approaches for describing them and our strategies for teaching them" (p. 149). This is

the aim of the present thesis.

In the same manner, in his article Some Ways of Translating English Phrasal Verbs

into Russian, Yatskovich (1999) casts a light on "the essence of some semantic

correspondences in the English and Russian verbal systems" (p. 1). Since EPVs are

deemed to be idiomatic combinations of two elements (a verb and an adverbial

particle), and the latter element gives the whole combination the "ability to describe

actions or states more precisely, vividly and emotionally" (p. 1), he concentrates on

the semantic functions of the English adverbial particles in comparison with Russian

verbal prefixes, claiming that "[i]n addition to their function that is analogous to that

of English prefixes, Russian verbal prefixes resemble English adverbial particles in

their semantic functions, also indicating various qualities of actions and states" (p. 2).

Such an overlap, according to him, influences the process of translating EPVs into

Russian in the sense that "[…] the meaning of the English adverbial component of the
201

phrasal verb is mostly conveyed by using the Russian prefix that reflects the character

of the described action or state most accurately" (p. 2).

Yatskovich (1999), on the other hand, admits that "it seems almost impossible to

create a consistent rigid system of lexical correspondences between English adverbial

particles and Russian prefixes, without encountering numerous debatable problems"

(p. 2). One of such debatable problems, he elaborates, is the polysemic nature of

EPVs, which has to be always kept in the mind of translator when dealing with EPVs.

He, all in all, concludes that "[…] understanding of semantic correspondences in

English and Russian verbal systems can be quite a powerful tool in the translator's

arsenal" (p. 3).

2.11.2 Translating English idioms into other languages

Idiomatic expressions, as a whole, still constitute a serious challenge for translators in

spite of latest developments in the field of translation theory and application (Awwad,

1990). The difficulty of translating idioms, particularly from English into Arabic, has

been, to a certain degree, overlooked in translation studies. The vast majority of the

relevant literature has been limited to lexicology and applied linguistics, concentrating

on the translation of metaphor and metaphorical language, "[b]ut idioms as they are

traditionally known and classified have been yet to receive the due attention"

(Ghazala, 2003, p. 203). Consequently, "[…] the translation of English idioms into

Arabic has not been adequately investigated, which has resulted in a gap in the field

of translation that needs further study" (Bataineh &Bataineh, 2002, p. 34).


202

IEPVs, however, have been amongst the types of English idioms investigated by a

number of Arab researchers. Mohammad Awwad (1990), for instance, incorporates

IEPVs as one type of English idioms in his paper Equivalence and Translatability of

English and Arabic Idioms. He attributes the difficulty of translating English idioms

to two main reasons: 1) misinterpreting the intention of the original writer or speaker.

The Arabic expression ‫ب‬7&(+‫• ا‬F&8, for instance, has literal and idiomatic/metaphorical

meanings. Therefore it is translated into English literally as (he opened the door),

whereas it is translated idiomatically as (he established a precedent) (p. 58); and 2)

cultural differences. For example, in English if someone dies they say (he kicked the

bucket), while in Arabic they say 5&P7/9‫ ا‬4ّ)&H (he handed over/delivered what he was

entrusted with, which is a reference to the soul leaving the body) (p. 59).

Moreover, Awwad (1990) arrives at a theoretical framework for handling the

translatability of idioms. Such a framework is based on the fact that an idiom may fall

into one of the following four categories of correspondence between English and

Arabic idioms:

a- Idioms with no correspondence between expression[s] and functions, i.e.


expressions and functions are language specific.
b- Idioms with corresponding functions in both languages but with
completely different expressions.
c- Idioms with corresponding functions in both languages, but with slightly
different expressions.
d- Idioms with corresponding functions and expressions in both languages
(p. 66)

The first category is illustrated by the IEPVs turn in and turn down, which are

translated by Awwad (1990) as: ‫م‬X.)+ / ‫اش‬,_)+ ‫•وي‬1 (to go to bed / to sleep) and “8,&1 (to

refuse) respectively (p. 65). Translating this category of idioms, however, requires not

only "almost complete mastery of both SL [Source Language] and TL [Target


203

Language] linguistic system but also a deep understanding and awareness of the SL

and TL culture and way of life" (p. 63).

The second category, on the other hand, is illustrated by the English idioms to hold

the reins, and he was the scape-goat, which are translated into Arabic as ‫م‬7&/Š\ SY&31

‫ر‬X&/9‫( ا‬to hold the reins of the things), and ‫_&[اء‬+‫ن آ&(˜ ا‬7&‫( آ‬he was the ram of the

sacrifice) in that order (p. 62). What's more, translating this kind of idioms requires

"the translator to pay special attention to the areas of difference in expression between

SL and TL" (p. 62).

With regard to the third category, the translator "must either find the right idiom in

TL, or render a translation of the meaning of the idiom as best as he can" (p. 61). It is

exemplified by the IEPV trade in, which is translated as “17ƒ1 (to exchange something

for something else) (p. 61).

As far as the fourth category is concerned, where both expressions and functions

correspond, the resulting translation is typically correct and idiomatic in both SL and

TL provided that "the translator is a native speaker of one language and has native-

like competence in the other" (p. 59). This category is illustrated by the following

English idioms where PVs are included: turn over a new leaf, and to hold out the olive

branch, which are translated by Awwad (1990) into Arabic as: ‫[ة‬1[„ 5Q_` ‫([أ‬1 (to start a

new page), and ‫ن‬XF1Š+‫ ا‬0UG ]8,1 (to raise the branch of the olive tree) (p. 60).

Along the same lines, Baker (1992), in her attempt to account for the difficulties

translators encounter when translating idioms and fixed expressions, utilizes IEPVs,
204

as inherent idioms or part of idioms and fixed expressions, to illustrate such

difficulties. She uses, for instance, the IEPV go out with (to have a romantic or sexual

relationship with someone) and the IEPV take for in take someone for a ride (to

deceive or cheat someone in someway) to clarify the point that how misleading the

idioms can be when they carry both literal and idiomatic meanings and "seem

transparent because they offer a reasonable literal interpretation and their idiomatic

meanings are not necessarily signaled in the surrounding text" (p. 66). Such idioms,

according to Baker, can be easily manipulated by speakers and writers. This may

confuse a translator who is not familiar with such idioms and may make him/her

"easily accept the literal interpretation and miss the play on idiom" (p. 66).

Baker (1992), however, lists four difficulties involved in translating idioms and fixed

expressions as follows;

(a) An idiom or fixed expression may have no equivalent in the target


language. [They] may be cultural-specific […].
(b) [They] may have a similar counterpart in the target language, but
its context of use may be different: the two expressions may have
different connotations […].
(c) An idiom may be used in the source text in both its literal and
idiomatic senses at the same time […]. Unless the target-language
idiom corresponds to the source-language idiom both in form and
meaning, the play on idiom cannot be successfully reproduced in
the target text […].
(d) The very convention of using idioms in written discourse […] and
their frequency of use may be different in the source and target
languages […]. Using idioms in English is very much a matter of
style. Languages such as Arabic and Chinese which make a sharp
distinction between written and spoken discourse and where the
written mode is associated with a high level of formality tend, on
the whole, to avoid using idioms in written texts (pp. 68-71).

Moreover, Baker (1992) provides the following strategies for translating idioms:

1. [Translation by] [u]sing an idiom of similar meaning and form […].


2. [Translation by] [u]sing an idiom of similar meaning but dissimilar form
[…].
3. Translation by paraphrase […].
205

4. Translation by omission […].


5. [Translation by] compensation [which] means that one may either omit or
play down a feature such as idiomaticity at the point where it occurs in the
source text and introduce it elsewhere in the target text (pp. 74-78).

In the light of the above difficulties and the strategies proposed to deal with them,

Ruba Bataineh and Rula Bataineh (2002), in their article The difficulties Jordanian

Graduate Learners of English as a Second Language Face When Translating English

idioms into Arabic, investigate the problems graduate students of translation at two

Jordanian universities (Yarmouk University and the University of Jordan) encounter

when translating idiomatic English expressions into Arabic. They analyse

"translations produced by forty-five subjects of a forty-five-item test that consists of

sentences each of which contains an idiom" (p. 33). Bataineh and Bataineh (2002),

however, take IEPVs into their account when they set up their test sentences.

Therefore, in their attempt to examine the translatability of IEPVs along with the

other types of idiomatic English expressions, they incorporated seven sentences, each

of which contains an IEPV. Four of them were from the verb + particle category of

EPVs (act up, let out, do without, and lay up), while the other three were from the

verb + particle + preposition category (be in for, get up to, and add up to).

Bataineh and Bataineh (2002) come up with the following findings with regard to the

translatability of EPVs: precisely 60.6% of the examined subjects translated correctly

the PVs from the first category, whereas 56.7% of them were able to correctly

translate the PVs of the second category (p. 75). Despite the fact that PVs have come

at the end of the list of "Degree of Problematicity of Idiom Types" (pp. 75-76), which

is arranged in descending order, Bataineh and Bataineh (2002) outline the strategies

used by the subjects in translating these PVs. The subjects made use of four strategies
206

when translating PVs of the first category: paraphrase, literal translation, not

translating the sentences, and omission, while they made use of only the first three

strategies when translating PVs of the second category (pp. 66-68). Using these

strategies resulted in producing incorrect translation of the listed IEPVs by a number

of the examined subjects. Bataineh and Bataineh (2002) summarize the problems of

translating English idiomatic expressions that arise from the subjects' translations in

four points as follows:

1. the loss of some shades of meaning when omitting an idiom from the
sentence;
2. misunderstanding the meaning of the sentence through the incorrect
rendition of the idiom;
3. disrupting the coherence of a text when omitting or incorrectly translating
the idiom; and
4. unintelligibility through the use of literal translation (p. 77).

Also, they attribute such problems to the following six reasons:

1. some idioms are culture-bound;


2. some idioms may have counterparts similar in form but different in
meaning;
3. students may not be able to determine whether or not the expression at
hand is an idiom;
4. students lack the competence to translate the idiom into Arabic;
5. students lack the familiarity with the presence of idiomatic expressions;
and
6. students may be hindered by factors of carelessness and time pressure
(p. 77).

In his paper Idiomaticity Between Evasion and Invasion in Translation: Stylistic,

Aesthetic and Connotative Conditions, Ghazala (2003) identifies evasion and invasion

as the main two translation procedures of idiomaticity. By the former he means "the

elimination of the idiomaticity of the SL idiom when translating it into the TL with no

compensation of any kind" (p. 209), while by the latter he means "a translator's

deliberate use in the TL of an idiom that matches, if not supersedes, the original" (p.

217). Ghazala (2003) argues that both procedures are valid "on the right occasion for
207

the right reason(s)" (p. 209). Nonetheless, he goes strongly for invasion procedure "as

more creative, convincing and faithful procedure than the former [evasion] which is at

worst an escape from translating properly, and at best an inclination to practicality"

(p. 209).

Ghazala (2003) comes across the issue of translating EPVs, as an essential type of

English idiomatic expressions, into Arabic in his attempt to justify the adoption of the

evasion procedure. He provides two sub-procedures for evasion: 1) dissuasion from

idiomaticity; and 2) preference of insensible sense. The first sub-procedure, in turn, is

justified by the following three reasons: a) the translator's incompetence; b) zero

Language equivalence in TL; and c) avoidance of taboos (pp. 209-217). In elaborating

on the second reason (zero language equivalence in TL), Ghazala indicates that

"[t]here are many English idioms, including the majority of phrasal verbs, that have

no idiomatic equivalence in Arabic" (p. 211). In this case the Arabic translator is not

to blame for non-idiomatic Arabic translation of the idiomatic English expression,

still, "[…] he has to provide a satisfactory version of sense, with idiomaticity dropped

unwillingly, in which case a considerable amount of loss will take place with a little

chance to compensate" (p. 211).

Furthermore, Ghazala (2003) makes the point that although IEPVs have no

straightforward Arabic equivalents, "they all can be translated comfortably into their

precise literal sense, provided the translator understands them properly in their

English contexts before translating them into Arabic" (p. 213). He further emphasizes

that they should not be confused with prepositional verbs which, owing to the fact that
208

their verbs retain their common meanings, "can be understood and translated literally

and directly" (p. 312).

Ghazala (2003) concludes that the complexity of EPVs, which stems from the fact

that there are thousands of them, with tens of thousands of their different meanings in

existence, "may naturally make the task of translation extremely difficult so that a

non-idiomatic translation is often chosen in translation into Arabic, where such

phrasal combinations are infrequent" (p. 213).

To sum up, the phenomenon of IEPVs has been the focus of a number of translation

studies. The treatment of such a phenomenon has varied considerably from one

researcher to another depending upon the standpoint from which it has been

accounted for. Yet, one can infer a number of insights: firstly, translating IEPVs into

languages where there are a number of correspondences between them and the

English language, such as Spanish and Russian, being a task achieved relatively

easily. Such correspondences play a significant role, as a common ground, in

negotiating the idiomatic meaning of EPVs and, in turn, in finding the appropriate

equivalents to them. Secondly, translating IEPVs into Arabic, however, where there

exist no much correspondences neither syntactically nor semantically, is far more

complicated task and subject to numerous debatable difficulties. Thirdly, such

difficulties have constituted the basis upon which Arab researchers warrant the

employment of such strategies as evasion, omitting and paraphrasing IEPVs when

translating them into Arabic. Such strategies, as it will be manifested from the results

of this study, are not theoretically based and lack of a systematic approach that may

help tackle the difficulties encountered when translating IEPVs into Arabic. Last but
209

not least, I am strongly in favour of the fact (as confirmed by Awwad, 1999; Bataineh

& Bataineh, 2002; Ghazala, 2003) that the question of translating English idioms into

Arabic has not been comprehensively taken up. More insightful efforts are needed to

describe the difficulties encountered by Arabic translators when dealing with such a

problematic phenomenon, and more practical strategies are needed to be suggested,

examined and applied.

2.12 Conclusion

This Chapter was entirely devoted to address the first set of the research questions

outlined in the previous Chapter (see 1.3 above). The main purpose was to extensively

scrutinize the key points of the phenomenon of idiomatic PVs in both the English and

Arabic languages, and to highlight scholarly works which have previously been done

by a wide range of scholars in such domains as linguistics, lexicography and

pedagogy, to establish a theoretical platform from which one can proceed to

investigate the main question of this study - that is, the difficulties Arabic professional

translators and Arabic translation students encounter when translating IEPVs into

Arabic.

The enormous bulk of information outlined throughout the Chapter revealed that the

phenomenon of IEPVs has attracted the attention of many researchers in different

fields. It has been insightfully dealt with from a variety of perspectives. Grammarians,

semanticists, lexicographers and pedagogues found that structural oddity, semantic

ambiguity, high productivity, and the amount of challenge IEPVs constitute to a non-

native speaker necessitate a deep and careful investigation. Therefore, they devoted a
210

great deal of their scholarly efforts accounting for their syntactic features, semantic

properties, and other related peculiarities to agree upon some and disagree upon

others.

In the next Chapter a theoretical rationale for the experimental part of this study will

be established and a practical model for the analysis and translation of IEPVs into

Arabic will be devised.


211

CHAPTER THREE
LINGUISTIC CONTRASTIVE ANALYSIS

3.1 Introduction

As has been previously mentioned, the principal aim of this study is to investigate the

difficulties encountered by Arabic professional translators and Arabic translation students

when tackling the phenomenon of IEPVs, and to propose some recommendations based

on the results of the study.

In the last Chapter an ample and exhaustive literature review of the notion of PVs and its

peculiarities was provided in both the English and Arabic languages. This Chapter

constitutes a link between the theoretical part of the study (the literature review) and the

practical part of it (the forthcoming experimental research). In this Chapter I am going to

establish a theoretical rationale for my experiments by reviewing main and influential

approaches of contrastive analysis to see their appropriateness/inappropriateness to my

topic, so as to formulate a practical and scientific model for the analysis and translation of

IEPVs into Arabic.

3.2 Contrastive analysis

Contrastive analysis basically means "a linguistic study of two languages aiming to

identify differences between them in general selected areas" (Hoey & Houghton, 2001, p.

46). Or, as James (1980) puts it, "a linguistic enterprise aimed at producing inverted (i.e.
212

contrastive not comparative) two valued typologies […], founded on the assumption that

languages can be compared" (p. 3).

The discipline of contrastive analysis, following Hoey & Houghton (2001), is more or

less a modern one. It was developed in the United States of America during and after

World War Two, "in the context of second and foreign language teaching" (p. 46). It

became quite popular in Europe throughout the 1970s, when "several large contrastive

projects were set up, contrasting English with, for example, Polish and Finish among

others" (p. 46).

Fisiak (1981) indicates that since the forties the task of comparing two languages or more

to determine the differences and similarities between them has been termed contrastive

analysis or contrastive study.

In his book Contrastive Analysis, Carl James (1980) characterises contrastive analysis as

a form of interlingual study, and, as such, "has much in common with the study of

bilingualism" (p. 8). Further, modern contrastive analysis, according to him, "starts with

Lado's Linguistics across cultures (1957)" (p. 8) [Emphasis in original].

3.3 Contrastive analysis and translation

Hoey & Houghton (2001, p. 47) address the issue of the relevance of contrastive analysis

to translation. They argue that "[a]t a practical level, it [contrastive analysis] is probably

most useful in pointing out areas where direct translation of a term or phrase will not

convey accurately in the second language the intended meaning of the first" (p. 47). In
213

addition, contrastive analysis, at a global level, "leads the translator to look at broader

issues such as whether the structure of the discourse for a given text-type is the same in

both languages" (p. 47).

Furthermore, Hoey & Houghton (2001) indicate that there is a bidirectional relationship

between contrastive analysis and translation in that the latter may provide the data for the

former, while the former "may provide explanations of difficulties encountered in

translation" (p. 49). They exemplify the first relationship in the works of Gleason (1965),

Krzeszowski (1990), and James (1980), and illustrate the second relationship in the works

of Nida (1964), Beekman and Callow (1974), Yebra (1982), Enkvist (1978) and Baker

(1992).

Following Catford (1965) and James (1980), Chakhachiro (1997) indicates that

contrastive analysis

exercises an analysis on a pair of languages in order to: 1) detect any linguistic


changes in conveying certain messages (messages pertain here to genre and style
of texts), then suggest these changes as the translation equivalence […]; 2) find
the gap or 'imbalance', to use James' word, between both languages and attempt
to fill this gap in the target language (p. 107).

Chesterman (1998), however, argues that although translation theory and contrastive

analysis are "neighbouring disciplines, it is nevertheless often appears that theoretical

developments in one field are overlooked in the other, and that both would benefit from

each other's insights" (p. 6). Moreover, he indicates that "[b]oth translation theorists and

contrastivists have expanded their focus of attention towards each other, and some

scholars have openly sought to establish conceptual bridges between the two disciplines"

(pp. 27-28).
214

Hoey and Houghton (2001, p. 47) consider that a number of theoretical and practical

problems do occur in the course of application of contrastive analysis. Such problems are

"related to specific difficulties of identifying a common ground for comparison,

comparing descriptions of different languages, taking account of psycholinguistic and

sociocultural factors, and taking into account of extratextual and intertextual factors" (p.

47).

Further, Chakhachiro (1997) properly sums up the contribution of contrastive analysis to

translation theory as follows:

[Contrastive analysis] can provide translation theory with comprehensive,


flexible and credible strategies. It encompasses macro as well as micro features
of texts, it can accommodate translation across any pairs of languages and lastly,
it furnishes the theory of translation with a logical, systematic and practical
approach (pp. 106-107).

It is common knowledge that there are a variety of contrastive analysis models proposed

and applied to translation studies. The origin of the methodological diversity, according to

Chakhachiro (1997) is the adoption or adaptation of contrastive or comparative linguistic

models by modern translation theorists in their endeavour to provide strategies to analyse

a given pair of languages and, in turn, to propose recommendations for translation

equivalence.

Along these lines and in order to determine the linguistic contrastive analysis model

best relevant to the topic at hand, I am going to conduct a short review on the main and

influential approaches of contrastive analysis. The aim here is to provide a clear picture

as to which approach is relevant and which one is not.


215

3.4 Review of contrastive analysis approaches

Mason (2001) sketches the stands of thinking that influenced the perspective of

translation in three main stands: 1) the functionalist views of the British tradition in

linguistics; 2) the notion of communicative competence; and 3) the purpose (skopos)

theory. The first stand, Mason points out, is represented by the works of Firth, Catford,

Gregory, Halliday and others. The second was "developed originally by Dell Hymes in

response to Chomskyan view of language competence" (p. 29). The third is represented

by the works of Karl Buhler and Reiss and Vermeer (p. 29).

Campbell (1998), however, emphasizes that "[t]here is a dichotomy of standpoints in

the linguistic theories from which models of translation may be derived – the functional

standpoint that concerns the relationship among the writer, the text and the real world –

and the neo-Cartesian standpoint that is concerned with mental processes" (p. 158).

More about such stands of thinking and linguists standpoints is outlined in what

follows.

3.4.1 Juliane House

In her paper Translation Quality Assessment: Linguistic Description versus Social

Evaluation, House (2001) reports on the main three approaches to translation quality

evaluation, namely: 1) Mentalist Views; 2) Response-based Approaches (Behavioristic

Views, and Functionalistic, "Skopos"-Related Approach); and 3) Text and Discourse

Based Approaches (Literature-oriented Approaches: Descriptive Translation Studies,


216

Post-modernist and Deconstructionist Thinking, and Linguistically-oriented

Approaches). House (2001) makes the point that "translation is essentially an operation

in which the meaning of linguistic units is to be kept equivalent across languages

[therefore] one can distinguish at least three different views of meaning, each of which

leads to different conceptions of translation evaluation" (p. 243). Hence, before

proceeding to present her own approach, House (2001) elaborates briefly on each of the

abovementioned approaches, rejecting, as a result, the first two of them and going

strongly for the third. Her argument, for and against, was driven by the following facts:

1) the "Mentalists Views" are very much subjective and intuitive and their "relativising

stance, and especially the relativisation of 'content' and 'meaning' is particularly

inappropriate for the evaluation business of making argued statements about when, how

and why a translation is good" (p. 244); 2) the "Behavioristic Views", which are

"influenced by American structuralism and behaviorism, [and] most famously

associated with Nida's (1964) pioneering work", cannot be postulated as criteria for

translation evaluation since by no means the views can be measured (p. 244); 3) the

"Functionalistic, 'Skopos'-Related Approach", which was developed by Reiss and

Vermeer, "cannot be said to be an adequate theory when it comes to tackling the

evaluation of translation in its fundamental bidirectionality". This is owing to the fact

that "[...] any translation is simultaneously bound to its source text and to the

presuppositions and conditions governing its reception in the new environment", while

this theory gives rise to the purpose "Skopos" of a translation as a yardstick in the

process of evaluation (p. 245); 4) the "Literature-oriented Approaches: Descriptive

Translation Studies", as well, fail to provide criteria for judging merits and weaknesses.

As how can one judge "whether one text is a translation and another one not? And what

are the criteria for judging merits and weaknesses of a given 'translation text'?" (p. 246);
217

5) the "Post-modernist and Deconstructionist Thinking" approach is also rejected due to

the plausible wondering of "how one can ever differentiate between a translation and

any other text that may result from a textual operation which can no longer claim to be a

translation relationship with an original text" (p. 246); and 6) the "Linguistically-

oriented Approaches", which date back to the pioneering works of Catford (1965)2,

Reiss (1971)3, Wilss (1974)4 among others, and the scope of which has been widened

by such scholars as Baker (1992), Doherty (1993)5, Hatim and Mason (1997), Hickey

(1998)6 and others, are the most promising approaches, since they "take the relationship

between source and translation text seriously". The most promising ones are those

"which explicitly take account of the interconnectedness of context and text because the

inextricable link between language and the real world is both definitive in meaning and

in translation" (p. 247).

In line with the" Linguistically-oriented Approaches", House formulates her own

approach under the name of Functional-Pragmatic Model of Translation Evaluation,

which she developed more than 30 years ago and revised in 1981 and 1997. The model

is

[…] based on Hallidyan systematic-functional theory, but also draws


eclectically on Prague school ideas, speech act theory, pragmatics, discourse
analysis and corpus-based distinction between spoken and written language. It
provides for the analysis and comparison of an original and its translation on
three different levels: the levels of Language/Text, register (Field, Mode and
Tenor) and Genre" (p. 247).

House's main concern in this model is the functional-pragmatic equivalence, which is

"related to the preservation of 'meaning' across two different languages and cultures" (p.

2
Cited in House (2001)
3
Cited in House (2001)
4
Cited in House (2001)
5
Cited in House (2001)
6
Cited in House (2001)
218

247). Moreover, she asserts that the concept of functional-pragmatic equivalence has

been accepted in contrastive linguistics for a long time and it is "the type of equivalence

which is most appropriate for describing relations between original and translation. And

it is this type of equivalence which is used in the functional pragmatic model suggested

by House (1997)" (House, 2001, p. 247).

As for register, or "context of situation", House (2001) indicates that the broad notion of

"context of situation" has been broken down "into manageable parts, i.e., particular

features of the context of situation or 'situational dimension': for instance 'Field'

'Mode' and 'Tenor'" (p. 248) [Emphasis in original]. Where field "captures social

activity, subject matter or topic, including differentiations of degrees of generality,

specificity or 'granularity' in lexical items according to rubrics of specialized, general

and popular" (p. 248). Whereas tenor "refers to the nature of the participants, the

addresser and the addressees, and the relationship between them in terms of social

power and social distance, as well as degree of emotional charge" (p. 248). In addition,

tenor "captures 'social attitude', i.e., different styles (formal, consultative and informal)"

(p. 247). Mode "refers to both the channel – spoken or written […], and the degree to

which potential or real participation is allowed for between writer and reader" (p. 248).

'Text' and 'context of situation', House (2001) explaines "should not be viewed as

separate entities, rather the context of situation in which the text unfolds is encapsulated

in the text through a systematic relationship between the social environment on the one

hand and the functional organization of language on the other" (p. 248).
219

Moreover, the category of genre, which "enables one to refer any single textual

exemplar to the class of texts with which it shares a common purpose", is useful for the

process of analysis and evaluation because "although Register (Field, Tenor, Mode)

descriptions are useful for accessing the relationship between text and context, they are

basically limited to capturing individual features on the linguistic surface" (House,

2001, p. 248). That is, "[w]hile register captures the connection between texts and their

'microcontext', Genre connects texts with the 'macrocontext' of the linguistic and

cultural community in which texts are embedded" (p. 248).

Such a model, as it has been previously stated, is a revised version of House's model

proposed in her doctoral dissertation (1977) titled A Model for Translation Quality

Assessment. Given the importance of this model in both theory and practice of

translation, I am going to sum it up in what follows.

The essence of translation, according to House (1977), "lies in the preservation of

'meaning' across two different languages" (p. 25). Hence, she asserts that the concept of

'meaning' is of three aspects, namely: a semantic aspect, a pragmatic aspect and a

textual aspect. The semantic aspect of meaning is the referential one, which "consists of

relationship of reference or denotative, i.e. the relationship of linguistics units or

symbols to their referents in some possible world" (p. 25). This aspect of meaning is,

for House (1977), "(a) most readily accessible, and for which (b) equivalence in

translation can most easily be seen to be present or absent" (p. 26). The pragmatic

aspect of meaning, on the other hand, is manifested by the difference between

pragmatics and semantics, as pragmatics "relates to the correlation between linguistic

units and the user(s) of these units in a given communicative situation" (House, 1977, p.
220

27). It is this distinction which underlies the theory of speech acts introduced by Austin

(1962) and Searle (1969) where pragmatic meaning is "referred to as the illocutionary

force that an utterance is said to have, i.e. the particular use of an expression on a

specific occasion" (House, 1977, p. 27).

As for the third aspect of meaning, textual aspect, House (1977) indicates that since

"[t]ranslation is a textual phenomenon", the various ways of text constitution, which

account for the textual meaning, such as: occurrences of pro-forms, substitutions, co-

references, ellipses and anaphora, have to be kept equivalent in translation (pp. 28-29).

With the above three aspects of meaning in mind, House (1977) asserts that "an

adequate translation text is a semantically and pragmatically equivalent one [and] as a

first requirement for semantic-pragmatic equivalence we posit that the translation text

have a function equivalent to that of its source text" (p. 30) [Emphasis in original].

In order to sharpen her own notion of function, House (1977) reviews different views of

the function of language proposed by such scholars as Ogden and Richards (1946)7,

Karl Buhler (1965)8, Jakobson (1960)9 and Halliday (1970 a, 1970 b, 1971, 1973)10.

Having done this, she adopts Halliday's terms of function, i.e. ideational and

interpersonal, to refer to the traditional dichotomy of the two types of functions. The

first term, ideational function, refers to the referential functional component, or content-

oriented function, where, following Halliday, "language expresses content: the speaker's

vision of the external world as well as the experience of the internal world of his own

consciousness" (House, 1977, p. 34). The second term, interpersonal function, refers to

7
Cited in House (1977)
8
Cited in House (1977)
9
Cited in House (1977)
10
Cited in House (1977)
221

the non-referential functional component where "language serves as a means for

conveying the speaker's relationship with interlocutor(s), and for expressing social roles

including communication roles such as questioner and respondent" (House, 1977, p.

34).

House (1977) uses these two types of function for choosing and grouping her sample

texts and "for labeling the two components of the textual function discovered in the

individual texts" (p. 37). Under the ideational function category she groups the

following texts: scientific text, commercial text, journalistic article and tourist

information booklet, while under the interpersonal function category she groups the

following texts: religious sermon, political speech, moral anecdote and comedy

dialogue.

Function of a text, as defined by House (1977), is "the application […] or use which the

text has in the particular context of a situation. [Therefore,] in order to characterize the

function of a text precisely, we must analyze the text in detail" (p. 37).

In her endeavour to provide "explicit practical guidelines for a coherent analysis and

evaluation of a translation", House (1977) developed this model as an eclectic one "for

characterizing the linguistic–situational peculiarities of the source text, comparing

source and translation texts, and making objective statement about the relative match of

the two texts" (p. 2). The model is essentially based on such pragmatic theories of

language use as speech act theory, functional and contextual views of language, and text

linguistics confederations (p. 3). The basic requirement for equivalence of a given

source text and its translation text, according to House (1977), is that the latter "should
222

have a function – consisting of two functional components, the ideational and the

interpersonal – which is equivalent to ST's [Source Text] function, and that TT

[Translation Text] should employ equivalent pragmatic means for achieving that

function" (p. 244).

In order to determine the function of a text, House adapts the model of Crystal and

Davy (1969)11 "situational constraints", and breaks down the notion of situation into

eight manageable elements, calling them "situational dimensions". Such dimensions are

of two categories, namely: 1) dimension of language user, which consists of three

parameters: Geographical Origin, Social Class and Time; and 2) dimension of language

use, which consists of five parameters: Medium, Participation, Social Role

Relationship, Social Attitude and Province. House establishes some linguistic correlates

for these situational dimensions (pp. 37-50).

She, as well, employs the distinctions between different combinations of spoken and

written modes suggested by Gregory (1967)12. In such a distinction, the written mode

has three categories, namely: 1) to be spoken as if not written; 2) to be spoken; and 3)

not necessarily to be spoken. The last category has a subcategory which is "to be read as

if heard" (p. 43).

House's method of operation of the model starts with depicting a textual profile for

source text by analyising it according to the set of eight situational dimensions to

characterise its function. The resultant textual profile of source text is taken as a

yardstick against which translation text is measured to depict its own textual profile.

11
Cited in House (1977)
12
Cited in House (1977)
223

The two resultant textual profiles, source text and its translation text profiles, which

characterise their function, are compared to explore the matches and mismatches

between them, and to provide a statement of the relative match of the ideational and

interpersonal function.

Further, House (1977) makes a distinction between two types of mismatches or errors,

"covertly erroneous errors" and "overtly erroneous errors". By the former she means the

dimensional mismatches, i.e. the mismatch of any one of the situational dimensions,

and by the latter she means the non-dimensional mismatches which comprise "both

mismatches of the denotative meanings of ST [Source Text] and TT [Translation Text]

elements and breaches of the target language system" (p. 245).

It is quite useful, however, to end this subsection with the following remark made by

Peter Fawcett (2001):

One of the earliest applications of the concept of register to translation was


provided by House (1981), who showed how the two major text functions
(ideational: conveying ideas, and interpersonal: relating author, text and
reader) are supported by register parameters such as medium and social role
relationship, and how on this basis a translation can be judged not on semantic
match but the degree of register match or mismatch" (pp. 123-124). [Emphasis
in original].

It is plausible to declare here that there are many relevant elements in this model to the

topic of this study, namely, the notions of functional-pragmatic equivalence, context of

situation (situational dimensions), and speech acts. Relevant to my study, as well, is the

distinction made by House (1977, pp. 56-57) between the two types of errors, namely:

covertly erroneous errors and overtly erroneous errors.


224

3.4.2 Carl James

In his book Contrastive analysis, James (1980) distinguishes between two approaches

to contrastive analysis, namely: microlinguistics and macrolinguistics. The former,

which was adopted by a great deal of modern 20th century linguists and contrastivists,

aims at "the description of the linguistic code, without making reference to the uses to

which the code is put, or how messages carried by this code are modified by the context

in which they occur" (p. 27). The latter, however, which has been attracting an

increasing attention, aims at the "contextual determination of messages and their

interpretation" (p. 27). These two approaches have been ably outlined in length by

James (1980) as to how to make use of them to execute contrastive analysis. His

profound insights are summarised in what follows.

3.4.2.1 Microlinguistics

This approach, also named by James as 'code–oriented', can be conducted on the three

levels of language, i.e. phonology, lexis, and grammar, by employing the two principal

steps of contrastive analysis procedure, i.e. description and comparison respectively.

The main concern of linguists using this approach is the formal system of language,

which has been given different labels by different scholars. Saussure calls it 'langue',

Chomsky 'competence', others 'code' and James refers to it as microlinguistics or 'code–

linguistics' (p. 98). Language, in the view of such linguists, is "self–contained calculus,

a mechanism for the production of sentences" (p. 98). Yet, this view of language "has

given to linguistics the appearance of a discipline closely akin to mathematics or formal

logic, which are likewise concerned with abstract formal system" (p. 98). Moreover,
225

such linguists claim that in order to obtain the code underlying a language one has to

disregard many aspects of that language which are considered to be irrelevant or

complicating factors (p. 98). Such a disregarding process has been called by James,

following Lyons (1972)13, the "idealisation of data", which in turn can be conducted in

three distinct ways: 1) Regularisation, where such thing as false starts, hesitations,

backtracking, mixed constructions and others, which occur in spontaneous speech, have

to be "regularised out of the data for linguistic analysis" (p. 98). James quoted

Chomsky's (1965)14 attribution of these thing to "such grammatically irrelevant

conditions as memory limitations, distraction, shift of attention and interest" (p. 89); 2)

Standardisation, which is of two senses: "the selection of the Standard dialect for

description" and "the homogeneity of the data: since the task of linguistic description

would be complicated by having to cope with data taken from speakers with mixture of

regional or social backgrounds, informants are selected who speak the same, standard,

variety" (p. 99); and 3) Decontextualisation, which in turn can be done by two ways:

"either by [a sentence] being removed from the company of the sentences that precede

or follow it in a text (its context), or by being separated off from the real–world

situation in which it is used (its context of situation)" (p. 99).

3.4.2.2 Macrolinguistics

James (1980) indicates that this approach "represents a relatively new departure in 'pure'

and Applied Linguistics, and offers considerable scope for new work in CA

[Contrastive Analysis]" (p. 61). Unlike the goal of microlinguistics, which is "to specify

the universal and particular properties of human languages" (p. 100), the goal of

13
Cited in James (1980)
14
Cited in James (1980)
226

macrolinguistics is "to achieve a scientific understanding of how people communicate"

(Yngve, 1975 as cited in James, 1980, p. 100). Hence, "[a]ttention has shifted from the

code to a process: the process of communication" (p. 100).

James adopts six situational constraints proposed by Hymes (1974) "sensitivity to

which […] determines a speaker's communicative competence" (p. 101). Such

constraints are "socio–cultural variables that in part determine the form of successful

utterances [and] the ethnographer of speaking must refer to in characterizing any

particular speech event" (p. 100). They are: 1) Setting, i.e. time and place of speech; 2)

Participant, i.e. addressor, speaker, addressee and audience; 3) Purpose, i.e. persuasion,

command, advice, greeting, or even phatic communion; 4) Key, i.e. tone, manner or

spirit in which a speech act may be carried out; 5) Content, i.e. the topic; and 6)

Channel, i.e. speech and writing. James put the six variables in this simple formula:

"who says what to whom, where and when, how and why" [Emphasis in original] (pp.

100-101).

Further, James characterises macrolinguistics in three points, being:

i) A concern for communicative competence rather than for 'linguistic'


competence in Chomsky's sense.
ii) An attempt to describe linguistic events within their extra linguistic
settings.
iii) The research for units of linguistic organization larger than the singular
sentence (pp. 101-102).

Such a broadening of scope, according to James, is "aimed at, both 'vertically' in terms

of large linguistic units and horizontally, to incorporate socio–cultural linguistics" (p.

102). There are two ways by which such a broadening scope is achieved: 1) Text

analysis, which focuses "on the formal level and addresses the question of how
227

sentences are organized into larger, suprasentential units or texts; 2) Discourse analysis,

which is "the functional one, and looks at the ways in which people put language to

use" (p. 102) [Emphasis in original]. Text analysis, James elaborates, "starts with

linguistic forms and asks in which contexts they are appropriate", while discourse

analysis starts “with the outer frame of situations and working inwards to find the

formal linguistic correlates to the situational variables" (p. 102). Moreover, James links

this distinction to that sketched by Widdowson (1978)15 between usage and use, where

the former is related to form and the later to function (p. 102). Therefore, formal

devices such as: grammatical, lexical or intonational, which "signal the exact nature of

the relationships holding between successive sentences" in terms of cohesion and

coherence, are properties of text analysis, along with "Functional Sentence

Perspective", which means that such successive sentences must be informative, i.e.

presenting 'new' information, and relevant, i.e. associating this 'new' information with

the 'given' one (p. 109). In addition, James (1980) suggests three approaches by which

contrastive analysis might be conducted, they are: textual characterization, text

typology, and translated texts (pp. 113-118).

Discourse analysis, on the other hand, has very much to do with functionality of

language. It addresses the issue of use rather than usage or form in order to answer such

questions as: "what is the speaker (or writer) hoping to achieve? And what does he in

fact achieve, with this particular bit of language?" (p. 118).

In his attempt to account for the area of discourse analysis comprehensively, James

(1980) sheds light on it from different perspectives. He ably outlines the crucial notions

15
Cited in James (1980)
228

that may contribute in performing contrastive analysis on the ground of discourse

linguistics. In so doing, James draws on Austin's (1962)16 Speech Acts Theory, Grice's

(1967)17 Principles of cooperation, or Maxims of Conversation, Lakoff's (1973)18 Rules

of Politeness, and Laver's (1975)19 Phatic Communion among others.

In a nutshell, although, the macrolinguistics model is, following Chakhachiro (1997),

"highly theoretical and offers few practical examples for translation" (p. 104), it is,

however, an ambitious one "for translation problems that arise in any natural language

text type, particularly those texts that are ambiguous or have double meanings" (p. 104).

Given this, it is macrolinguistics rather than a microlinguistics notion which seems to be

more relevant to the issue of analysing and translating IEPVs into Arabic. This is

mainly due the fact that syntactic and lexical properties of IEPVs (their codes) by

themselves, as presented in the previous Chapter (see 2.4 above), are not enough to

account for their functional meaning and, in turn, determine their Arabic functional-

pragmatic equivalence. They need to be looked at within their communicative situations

and real-world context.

Before proceeding to set up a linguistic contrastive analysis model for the analysis and

translation of IEPVs into Arabic, it is quite significant to consider the crucial and

pertinent notions of equivalence, theory of sense, theory of speech acts and

communicative competence as understood and implemented by various authorities.

16
Cited in James (1980)
17
Cited in James (1980)
18
Cited in James (1980)
19
Cited in James (1980)
229

3.4.3 Equivalence

Equivalence constitutes "a central concept in translation theory, but it is also a

controversial one" (Kenny, 2001, p. 77). Theorists differ, Kenny (2001) elaborates, to a

great extent, as to how to approach the question of equivalence, in that some of them

(such as: Catford, 1965; Nida & Taber, 1969; Toury, 1980; Pym, 1992a 1995; Koller,

1995)20 "defined translation in terms of equivalence relations" (p. 77). Whereas others

(such as: Snell-Hornby, 1988; Gentzler, 1993)21 "reject the theoretical notion of

equivalence, claming it is either irrelevant […] or damaging […] to translation studies"

(p. 77). Other theorists (such as: Baker, 1992) "steer a middle course" and use

equivalence as "a useful category for describing translations" (Kenny, 2001, p. 77).

Moreover, House (2001, p. 247) points out that "[o]ver and above its role as a concept

constitutive of translation, 'equivalence' is the fundamental criterion of translation

quality". And translation for her is "[…] viewed as the recontextualization of a text in L1

by a semantically and pragmatically equivalent text in L2" (House, 2001, p. 247).

Equivalence has been defined as "the relationship between a source text (ST) and a target

text (TT) that allows the TT to be considered as a translation of the ST in the first place"

(Kenny, 2001, p. 77).

Equivalence in both translation theory and contrastive analysis, Chesterman (1998)

argues, has traditionally emerged from the notion of 'similarity' and gained "a wide

variety of interpretations in these two fields" (p. 16). In his attempt to examine such

20
Cited in Kenny (2001)
21
Cited in Kenny (2001)
230

interpretations, on the one hand, and assess the extent to which they may overlap on the

other, he explores the concept of equivalence in both translation theory and contrastive

analysis as in what follows:

Chesterman (1998) investigates three approaches to account for the concept of

equivalence in translation theory, namely: the equative view, the taxonomic view, and the

relativist view (pp. 18-27).

Following Kelly (1979)22 and Rener (1989)23, he considers the equative view as the oldest

approach, and it is "based on the original mathematical definition of equivalence,

denoting a reversible relation: A is B and B is A" (p. 18).

As for the second approach, i.e. the taxonomic view, Chesterman (1998) explores the

taxonomies of Nida (1964), Catford (1965) and Koller (1979) to illustrate the crux of this

approach, which is "equivalence is not a unitary concept but consists of several types.

Different types of equivalence are argued to be appropriate in the translation of different

kinds of texts. […] the concept of equivalence is argued to be context-sensitive" (p. 21).

Nida's taxonomy of equivalence, as briefly outlined by Chesterman (1998), shows that

equivalence is of two distinct types: formal equivalence and dynamic equivalence. "[t]he

binary division here is between the form-and-meaning of a message on the one hand, and

the effect of a message on the other" (p. 21). Translators are urged by Nida to "give a

higher priority to dynamic equivalence, so that the target text would have the same effect

on its readers as the source text did on the original readers" (p. 21). Further, by

22
Cited in Chesterrman (1998)
23
Cited in Chesterrman (1998)
231

emphasizing the factor of naturalness in the target text, Chesterman (1998) elaborates,

"Nida helped to shift theoretical attention away from texts-as-such to texts-as-people-use

them; semiotically, this meant a shift towards pragmatics, towards users and interpreters

of signs" (p. 21). Or, as Hatim and Mason (1990) put it, "Nida shifts attention away from

sterile debate of free versus literal towards the effects of different translation strategies"

(p. 7).

Catford's taxonomy of equivalence, on the other hand, is far much different. It is framed

by Chesterman (1998) as follows: "[w]e have three potential kinds of equivalence: formal

equivalence, which can only be approximate; semantic equivalence, which is theoretically

impossible; and situational equivalence, which is the basis for translation" (p. 22). Hatim

(2001) makes the point that "Equivalence [in Catford's taxonomy] is taken to be the base

on which source language (SL) textual material is replaced by target language (TL)

textual material" (p. 14) [Emphasis in original].

The third taxonomy of equivalence in translation theory is Koller's taxonomy, which

consists of five types. They are:

(i) Denotative equivalence (otherwise known as invariance of content,


semantic equivalence);
(ii) Connotative equivalence (including equivalence of style, register, and
frequency);
(iii) Text-normative equivalence (concerning text-type usage norms);
(iv) Pragmatic equivalence (receiver-oriented, equivalence of effect);
(v) Formal equivalence (including aesthetic and poetic features) (Chesterman,
1998, p. 23).

Yet, Chesterman (1998) criticizes the abovemantioned taxonomic approach, claiming that

some kinds of equivalence need to be appropriately defined. And, in terms of

dynamic/pragmatic equivalence of effect "it is not clear whether 'effect' can be defined and
232

measured at all, let alone how this might be done; nor is it clear whether we can determine

the recipients on whom some effect might be measured; or what the relation should be

between actual effect and another's intended effect; and so on" (p. 23) [Emphasis in

original].

Along these lines, Fawcett (2001) criticizes the theory of dynamic equivalence introduced

by Eugene Nida, claiming that it is "nothing less than a sociolinguistics of translation. By

focusing the translation process on the target-text receiver, who differs from the source-text

receiver in language, culture, world knowledge and text expectations in the same way that a

northern blue collar worker differs from a southern stockbroker compatriot" (p. 121).

Moving on to the third approach, i.e. the relativist view, Chesterman (1998) indicates that

the idea of identity assumption have been rejected altogether along with the concept of

equivalence in numerous recent contributions to translation theory. He elaborates, "[f]or

Snell-Hornby (1988), for instance, equivalence is no more than an illusion" (p. 24).

Away from taxonomies of equivalence in translation theory, Chesterman (1998)

investigates taxonomies of equivalence in contrastive analysis by elaborating on the seven

types proposed by Krzeszowski (1990)24, namely: statistical equivalence, translation

equivalence, system equivalence, semanto-syntactic equivalence, rule equivalence,

substantive equivalence, and pragmatic equivalence (pp. 31-35). Chesterman comes up

with the fact that there is a similarity between Krzeszowski's kinds of equivalence and

those of translation theory, and both fields (contrastive analysis and translation theory)

"are talking about the same phenomena in different words" (p. 37).

24
Cited in Chesterrman (1998)
233

In short, Chesterman (1998) indicates that "equivalence-as-identity [absolute equivalence]

is an exception, often […] impossible" (p. 57). Such a conclusion echoes that of Ivir

(1996) who considers equivalence as a "relative and not absolute, […] it emerges from

the context of situation as defined by the interplay of (many different factors) and has no

existence outside the context, and in particular it is not stipulated in advance by an

algorithm for the conversion of linguistic units of L1 into linguistic units of L2". (Ivir,

1996, p. 155 as cited in House, 2001, p. 247).

Relevant to the study at hand is the notion of functional-pragmatic equivalence adopted

by House (1977, 1981, 1997 and 2001), which has very much in common with Nida's

(1964) dynamic equivalence and Koller's (1979) fourth type of equivalence, i.e. pragmatic

equivalence.

3.4.4 Theory of sense (Interpretive Approach to Translation)

This theory was initially developed in the late 1960s by a number of researchers of the

Paris School who were studying "conference interpreting in real situations, with

particular emphasis on the mental and cognitive processes involved" (Salama-Carr,

2001, p. 112). Relying on their own experience in such fields as psychology,

neuropsychology and linguistics, the Paris School scholars concentrate on the process

of translation, and "particularly on the nature of meaning as sense – as oppose to

linguistic or verbal meaning – and the nature of ambiguities" (Salama-Carr, 2001, pp.

112-113) [Emphasis in original]. Danica Seleskovitch is considered as the leading

scholar of the Paris School. Other scholars, however, include M. Lederer, F. Herbulot,

J. Delisle and M. Pergnier (Salama-Carr, 2001). Being an experienced professional


234

conference interpreter, Seleskovitch (1977)25 developed the theory of sense and

distinguished between "linguistic meaning and non-verbal sense, where non-verbal

sense is defined in relation to a translation process which consists of three stages:

interpretation or exegesis of discourse, de-verbalization, and reformulation" (Salama-

Carr, 2001, p. 112) [Emphasis in original].

According to this theory, sense is composed of both implicitness and explicitness,

where the former means "what the writer or speaker intends to say or mean" (Salama-

Carr, 2001, p. 113), while the later means "what is actually said or written" (Salama-

Carr, 2001, p. 113). Having said that, the full understanding of sense "depends on the

existence of a sufficient level of shared knowledge between interlocutors, without

which the confrontation between text and cognitive structures does not lead to the

emergence of sense" (Salama-Carr, 2001, p. 113). It is crucial to say that cognitive

structures in this respect include both cognitive baggage and cognitive context, i.e. "real

word knowledge, […] and the knowledge acquired through the specific and immediate

reading of the text to be translated or interpreted" (Salama-Carr, 2001, p. 113).

Ambiguity, according to this theory, means "a direct result of lack of relevant cognitive

'complements' to verbal meaning" (Salama-Carr, 2001, p. 113). It is, as further

explained by Salama-Carr (2001):

The possibility of multiple interpretation [that] arises in situations in which


only the surface or verbal meaning of the text is available and the translator
does not have at his/her disposal all the cognitive elements and
complementary information needed to extract sense (p. 113).

25
Cited in Salama-Carr (2001)
235

A more detailed version of this approach was developed by the Canadian scholar Jean

Delsile (1980, 1988 and 1993)26, who "focuses on the intellectual process involved in

translation, the cognitive process of interlingual transfer, and stresses of the non-verbal

stage of conceptualization" (Salama-Carr, 2001, p. 113). Translation in Delsile's view is

"a heuristic process of intelligent discourse analysis" (Salama-Carr, 2001, p. 113). It

involves three stages, namely: comprehension, reformulation and verification (Salama-

Carr, 2001, p. 113). The first stage of translation, according to Delsile, "requires

decoding the source-text linguistic signs with reference to the language system […] and

defining the conceptual content of an utterance by drawing on the referential context in

which it is embedded" (Salama-Carr, 2001, p. 113). Whereas the second stage "involves

reverbalizing the concepts of the source utterance by means of the signifiers of another

language" (Salama-Carr, 2001, p. 113). While the third stage of translation "can be

described as a process of back-translation which allows the translator to apply a

qualitative analysis of selected solutions and equivalents. Its purpose is to confirm the

accuracy of the final translation" (Salama-Carr, 2001, pp. 113-114).

What makes this theory so relevant to the topic of translating IEPVs into Arabic is the

fact that it is the 'non-verbal sense', not the mere linguistic meaning, of any given IEPV

that has to be taken into account when rendering it into Arabic.

Targeting the 'non-verbal sense' will be considered as a key factor when analysing the

data, which will be collected from the subjects of this study to assess the quality of their

translation. This is driven by the fact that the abovementioned polysemous nature of

IEPVs along with their informal use in every day English renders their linguistic

26
Cited in Salama-Carr (2001)
236

meaning redundant. Non-verbal sense is, then, deemed appropriate for the discussion of

functional-pragmatic equivalence in focus for these highly productive, metaphorical and

complex expressions.

In order to achieve the required functional-pragmatic equivalence of any IEPV,

translators have to appreciate what a given speaker/writer intends to say rather than

what he or she is actually saying.

In analysing the data of this study, targeting the mere linguistic meaning by subjects

will be classified as incorrect translation as it will be considered, to use House's (1977)

terms, as a dimensional mismatch and in turn as a covertly erroneous error.

3.4.5 Theory of speech acts

This theory was first introduced by Austin (1962) in his book How to do Things with

Words and developed by Searle (1969) in his book Speech Acts: An Essay in the

Philosophy of Language. Austin (1975) defines speech acts as "[…] the total situation in

which the utterance is issued" (p. 52). He distinguishes between two types of utterances,

namely: constative and performative utterances. The constative utterances are statements

that may be judged true or false (p. 3), while the performative utterances, on the contrary,

cannot be judged true or false (pp. 4-7). He recapitulates that most utterances are

performative in nature, that is, "[…] the issuing of the utterances is the performing of an

action – it is not normally thought of as just saying something" (pp. 6-7), hence, "[…] in

saying something, we do something" (p. 91). In addition, Austin characterises three acts

of statements: 1) locutionary acts, which denote the act of doing something in the "full
237

normal sense" (p. 94). In other words, it is "roughly equivalent to uttering a certain

sentence with a certain sense and reference, which again is roughly equivalent to

‘meaning´ in the traditional sense" (p. 109); 2) illocutionary acts, which signify "[…]

performance of an act in saying something as opposed to performance of an act of saying

something" (pp. 99-100) [Emphasis in original], for instance "informing, ordering,

warning, undertaking, & c., i.e. utterances which have a certain (conventional) force" (p.

109); and 3) perlocutionary acts, which indicates "what we bring about or achieve by

saying something, such as convincing, persuading, deterring, and even, say, surprising or

misleading" (p. 109) [Emphasis in original].

Further, Hatim (2001) defines speech acts as "the acts we perform when, for example, we

make a complaint or a request, apologize or pay someone a compliment" (p. 179). Speech

acts, on the other hand, vary as to frequency from one culture to another, yet the

"common ones will include ask, refuse, praise, describe, excuse, [and] explain while rarer

ones are commiserate, condemn, [and] blaspheme" (James, 1980, p. 119) [Emphasis in

original].

Following Searle, Bell (1991) indicates that speech acts consist of two parts, being:

(a) Propositional content: the conceptual content; the nucleus; what the act is
about; what is referred to; the ideational macrofunction realized as a
preposition; the literal meaning (also locutionary act/meaning); the semantic
sense of the act.
(b) Illocutionary force: the communicative value the speaker intends the act to
have; the function it is intended to serve; the intentionality of the text.
Mirroring this there is, inevitably, the value the hearer puts on the act; the
perlocutionary force; part of the acceptability of the text (p. 174) [Emphasis in
original].

Thus, "a speech act is a combination of three simultaneous acts: a locutionary act,

which is the utterance itself, an illocutionary act, which is the communicative act, the
238

intended meaning behind the utterance, and the perlocutionary act, which is the reaction

the utterance produce on the listener" (Hale, 2004, p. 6).

Speech acts theory "was developed primarily for the analysis of spoken language"

(Fawcett, 2001, p. 124), and then became "the central focus of pragmatic analysis"

(Hale, 1996, p. 62). Hatim (2001) indicates that "[t]he pragmatic analysis of speech acts

sees all utterances in terms of the dual functions of 'stating' and 'doing things', of having

a meaning and a force" (p. 179). Along these lines, Austin's (1962) three aspects of

utterance 'locution', 'illocution', and 'perlocution' have been explained by Hatim as

'sense', 'force', and 'effect' respectively (p. 179).

The illocutionary force of an utterance, House (1977) elaborates, "is to be differentiated

from the propositional content, i.e. the semantic information that an utterance contains"

(p. 27) [Emphasis in original]. Such grammatical features as word order, mood of the

verb, stress, intonation, or the presence of performative verbs, may often indicate the

illocutionary force of an utterance. Yet "[i]n actual speech situation, it is, however, the

context which makes unambiguously clear what the illocutionary force of an utterance

is " (p. 27).

Taking pragmatic meaning or illocutionary force into consideration is of great

importance for translation, since it is concerned with instances of acts of speech (House,

1977, p. 27). House recapitulates this crucial point as follows:

In effect, translation operates not with sentences but with utterances, i.e. units
of discourse characterized by their use-value in communication. In translation,
it is always necessary to aim at equivalence of pragmatic meaning, if
necessary at the expense of semantic equivalence. Pragmatic meaning thus
overrides semantic meaning. We may therefore consider a translation to be
239

primarily a pragmatic reconstruction of its source text (p. 28) [Emphasis in


original].

Hence, translating an utterance literally means matching the locutionary act only, but not

the illocutionary and perlocutionay acts (Hale, 1996, p. 63).

The notion of speech acts is, to a great extent, relevant to the issue of translating IEPVs

into Arabic. The relevance of the notion is threefold: 1) the common use of IEPVs is

more prevalent in spoken English, which entails employing them in many different kinds

of speech acts. The fact remains, however, that written and spoken forms of IEPVs have

to be looked at within their communicative contexts in order to determine their intended

communicative values; 2) focusing on the locutionary/semantic meanings rather than the

illocutionary/pragmatic meanings when IEPVs are translated into Arabic would result in

providing unintended meanings of given utterances; and 3) Bearing in mind that many

IEPVs are polysemous in nature, in that they may occur in many different meanings

according to the contexts in which they are used (see 2.5.2 and 2.8.2 above).

It may well be useful to say, at this point, that the notion of 'conversational maxims',

also known as the 'co-operative principles', is quite pertinent to the theory of speech acts

(Baker 1992, p. 259), and in turn needs to be taken into consideration.

These maxims were first proposed by Grice (1967) (James 1980, p. 128; Bell 1991, p.

181; Baker 1992, p. 259). They are as follows:

1) Quantity: Be as informative as is required but no more than that – avoid


redundancy.
2) Quality: Say only what you believe to be true or what you have evidence for.
3) Relevance: Be to the point.
4) Manner: Be clear and succinct: avoid obscurity (James 1980, p. 128).
240

What makes these maxims so important and different from, say, grammatical rules,

according to James (1980), is that speakers almost always flout them and intend hearers

to notice flouts and draw conclusions. Thus, "[w]hen hearers notice these infringements

they continue to assume that the speaker is making infringements for a good reason.

These conclusions are referred to by Grice as conversational implicatures" (p. 128)

[Emphasis in original]. Hence, the concept of implicature, Fawcett (2001) writes, "is

based on the assumption that conversation is guided by a set of principles such as: be

polite, do not say more or less than you have to, and so on. When one of the principles

is violated, something is implied above and beyond the normal routines of

conversation" (p. 124).

The knowledge of conversational maxims and conversational implicatures, according to

Fawcett (2001), should constitute part of the translator's competence since "[…]

different languages will apply the principles in different ways in different situations" (p.

124).

What makes the theory of conversational maxims and conversational implicatures

relevant to the topic at hand is that they assist in the reception of idiomaticity of IEPVs.

Such a notion will be considered when the responses of the subjects of the study will be

analysed. In communication, different languages may flout or violate different maxims

according to their stylistic, idiomatic, cultural and linguistic norm. IEPVs have double

meanings, namely, literal/non idiomatic and metaphorical/idiomatic meaning. In other

words, they have direct and indirect meanings. Violation of maxims is about

indirectness, and translators have to appreciate the idiomatic/indirect meaning of a


241

given EPV if they are to achieve its functional-pragmatic equivalence. To take only one

example, the IEPV to sex up violates two maxims, namely, the maxim of quantity, as

the speaker/writer is not making their contribution as informative as required, and the

maxim of relevance. Hence this newly coined IEPV, which emerged in the 1990s in the

context of the war on Iraq and political corruption, implies a journalistic criticism

related to the fabrication of motivated evidence by some Western countries to attack

Iraq. Therefore, delivering its literal/direct meaning in translation will not convey the

intended implicature. Committing translational pitfalls of this kind will be seen as a

dimensional mismatch and in turn will be classified as a covertly erroneous error.

3.4.6 Communicative Competence

Developed originally by Hymes (1971)27, the notion of communicative competence "sees

the translator as a social being and considers his/her competence as a receiver and

producer of texts" (Mason, 2001, p. 31).

Hymes (1971) defines the communicative competence as: "the knowledge and ability

possessed by the translator which permits him/her to create communicative acts –

discourse – which are not only (and not necessarily) grammatical but … socially

appropriate" (p. 23 as cited in Bell, 1991, p. 42).

Such a knowledge has been divided by Johnson and Whitlock (1987) into five distinct

kinds, namely: "target language (TL) knowledge; text-type knowledge; source language

(SL) knowledge; subject area ('real-world') knowledge; and contrastive knowledge" (p.

137 as cited in Bell, 1991, p. 36).


27
Cited in Mason (2001)
242

Bell (1991, p. 36) adds to the above "the decoding skills of reading and encoding skills

of writing". Further, Bell (1991) makes the point that translators must know three

things:

(a) how propositions are structured (semantic knowledge), (b) how clauses can
be synthesized to carry propositional content and analysed to retrieve the
content embedded in them (syntactic knowledge), and (c) how the clause can
be realized as information-bearing text and the text decomposed into the
clause (pragmatic knowledge). Lack of knowledge or control in any of the
three cases would mean that the translator could not translate (pp. 36-37).

On the other hand, Canale (1983 as cited in Mason, 2001, p. 31) proposes four-part

classification for communicative competence: 1) Grammatical competence; 2)

Sociolinguistic competence; 3) Discourse competence; and 4) Strategic competence.

The first one means "knowledge of the rules of the code, including vocabulary and

word-formation pronunciation/spelling and sentence structure" (Bell, 1991, p. 41). It

entails, in the translator's case "passive command of one and active command of

another language system, in the sense of possessing the knowledge and skill required

to understand and express accurately the literal meaning of utterances" (Mason, 2001,

p. 31). The second classification means "knowledge of and ability to produce and

understand utterances appropriately in context" (Bell, 1991, p. 41). Discourse

competence, however, means "the translator's ability to perceive and produce

cohesive and coherent text in different genres and discourses" (Mason, 2001, p. 31)

(cf. Bell, 1991; Hatim & Mason, 1990). Whereas the strategic competence means "the

mastery of communication strategies which may be used to improve communication

or to compensate for breakdowns" (Bell, 1991, p. 41).

Bell (1991) asserts that "the translator must possess linguistic competence in both

languages and communicative competence in both cultures" (p. 42) [Italics in

original].
243

Using such proposals about translator competence has constituted the foundation

stone upon which Bell (1991) developed his model of translation (Campbell, 1998, p.

4).

Moreover, translation quality assessment approach, Campbell (1998) writes, can be

taken as a perspective from which the issue of translation competence can be

addressed. In spite of the fact that the textual product is the core and the key feature of

translation quality assessment and the individual translator is backgrounded, "[t]he

superficial relevance [of translation quality assessment] to translation competence is

that the quality of a translated text is a reflection of the translator's competence"

(Campbell, 1998, p. 8).

Along these lines, translation competence as it has been characterised by Bell (1991)

"is variable from individual to individual and is, in principle at least, measurable

against agreed objective criteria" (p. 14).

Given this and since my intention is to compare the performance of Arabic

professional translators with that of Arabic translation students, it is therefore

inevitable to include a parameter which can make such a comparison systematic,

practical and, most importantly, theoretically grounded.


244

3.5 A linguistic contrastive analysis model for the analysis and


translation of IEPVs into Arabic

Having outlined the major influential approaches of contrastive analysis along with

the pertinent notions, it is time now to set up a linguistic contrastive

analysis model for the analysis and translation of IEPVs into Arabic.

The aim of proposing the model is driven by three reasons: 1) to furnish a theoretical

foundation and set the ground for what is to follow; 2) to establish criteria upon which

the forthcoming translation tests can be devised; and 3) to create a yardstick by which

the quality of the translation of the subjects can be analysed, compared, measured and

evaluated.

Given the fact that "[…] models are complementary and often overlap and conflict"

(Hermans, 2001, p. 155), and the fact that any linguistic model has to be built on

existing models and theories (Chakhachiro, 1997, p. 113), in what follows, I will

develop a workable eclectic linguistic model based on the abovementioned models

and theories highlighting elements and parameters which are relevant to my topic.

To begin with, it is plausible to announce that there is, for the best of my knowledge

and research, no linguistic model that has been proposed so far for the analysis and

translation of IEPVs into Arabic. Given this, I am trying to devise a generic approach

in order to uncover those notions that may be used as a model for analysing and

translating IEPVs into Arabic. My contribution is to look scientifically into the

analysis and also into the translation of IEPVs into Arabic, and in turn to look into a

model that can be used for both. Yet, it is quite important to indicate here that due to

the fact that IEPVs, as it has been proven in Chapter Two, are highly productive in
245

nature and a tremendous number of them have been already established and stated in

general and specialized dictionaries, a qualitative rather than quantitative contrastive

analysis is carried out in this study.

I strongly agree with Fawcett (2001) on his claim that "[w]ord and phrase level

taxonomies, even where they are context-sensitive, are inadequate for dealing with all

the problems faced by translators" (p. 123). Hence, the scope of analysis has been

broadened by such researchers as Hatim and Mason (1990, 1997), Bell (1991) and

Baker (1992) to comprise "the TEXT LINGUISTICS level of register analysis (tenor,

mode, domain), DISCOURSE ANALYSIS (thematic structure, coherence, cohesion)

and PRAGMATIC analysis (speech acts, Gricean principles, language and text

functions)" (Fawcett, 2001, p.123) [Emphasis in original].

Discourse analysis, according to Hatim (2001), has been used in different ways to

indicate that it means different things to different people since it was first introduced

by Zelling Harris in 1952. But "the study of language beyond the level of the sentence

may in fact be just about the only thing that unites a broad array of disparate

approaches" (p. 67). Campbell (1998), however, indicates that text linguistics,

discourse analysis and the study of genre are, in fact, all the same in the sense that

they all mean organization of language above the level of sentence.

With this point in mind, my model for the analysis and translation of IEPVs into

Arabic will draw on the following pertinent and relevant parameters: context of

situation/register (field, mode and tenor); speech acts (locutionary and illocutionary

acts); covertly/overtly erroneous errors; functional-pragmatic equivalence and

translation communicative competence.


246

As a result of the above review of models and notions, one can come up with a

conclusion that most of the relevant parameters and pertinent elements to the topic at

hand are covered in House's A model for Translation Quality Assessment (1977, 1981,

1997, and 2001). House's model is quite comprehensive in the sense that it brings

together the most appropriate notions to my topic, i.e. macrolinguistics (text and

discourse analysis), context of situation (situational dimensions), speech acts,

pragmatics, overtly erroneous errors, covertly erroneous errors, and above all the

model's ultimate goal of the functional-pragmatic equivalence. However, it stops short

of including the pertinent notion of translation communicative competence.

Therefore, the model for the analysis and translation of IEPVs into Arabic will be

mainly based on House's model, but it will also draw on the theory of communicative

competence. The model is outlined in what follows:


247

Analysis and Translation Model

Context of Speech acts


situation/Register (locutionary,
(Mode-field- illocutionary and
tenor) perlocutionary
acts )

Functional-
pragmatic
equivalence

Covertly/overtly
erroneous errors

Translation
communicative
competence

Figure 3.1 A model for the analysis and translation of IEPVs into Arabic

My main concern in this model is to see whether or not the Arabic functional-pragmatic

equivalence of IEPVs is achieved. Hence, each utterance, as a linguistic unit embedded

in a certain type of contextual situation, will be scrutinized from two perspectives. Such

perspectives constitute the key parameters of the model, namely: context of

situation/register (mode, field and tenor) and speech acts (locutionary, illocutionary and

perlocutionary acts). As far as the first parameter (register) is concerned, IEPVs cannot

be understood, analysed and translated without taking into account their context of

situation or, to use House's term, situational dimensions. The issue of formality and
248

informality of IEPVs constitutes the cornerstone in this regard. As has been previously

pointed out in Chapter Two, IEPVs are mainly used informally in everyday spoken

English. But they also appear in formal documents and used formally in such contexts

as politics, academic, religion, legal, and literature (see 2.5.5 above). Therefore, such

register variations need to be kept in check when translating IEPVs into Arabic.

With regard to the second parameter (speech acts), as it has been outlined above (see

3.4.4), the polysemous nature of IEPVs and their typical use in everyday spoken

English require translators to target their illocutionary/pragmatic meanings rather than

locutionary/semantic meanings if they are to deliver their functional-pragmatic

equivalents.

As for, the parameter of overtly/covertly erroneous errors, introduced and implemented

by House (1977, 1981, 1997 and 2001), mismatches or errors made by the subjects of

the study will be divided into two types: overtly and covertly erroneous errors. Such a

process of division will help not only in pinpointing the reasons behind the translational

pitfalls of IEPVs but also in evaluating the subjects' translation communicative

competence from their performance, so as to highlight the areas in which they have a

shortage of knowledge and/or skill. Furthermore, I believe that, in order to deliver the

functional-pragmatic equivalents of IEPVs, translators have to pay a great deal of

attention to these two types of errors in order to avoid them. While covertly erroneous

errors are typically caused by the dimensional mismatches, i.e. the failure to take such

parameters as mode, field, tenor and illocutionary act into consideration when

translating IEPVs into Arabic, the idiomaticity of EPVs plays a significant role in
249

making the overtly erroneous errors. Such an idiomatic nature makes it hard for

translators to capture the denotative meanings of IEPVs.

Finally, with regard to the last component of the model, i.e. translation communicative

competence, by which I mean theoretical as well as practical knowledge of translators, I

will exploit this component as a final goal to compare the inputs of the subjects of the

study (Arabic professional translators and Arab students of translation). As it has been

diagrammatically placed in the above model, all the relevant parameters have to be

taken into account when examining the translation communicative competence of the

subject, since all of such parameters have something to do with the translator's

knowledge, ability and awareness in that missing or ignoring any parameter would

mean a given subject has no, or a shortage of, theoretical/practical knowledge, and

hence, a lack of translation communicative competence.

3.6 Conclusion

This Chapter was primarily intended to serve as a link between the previous theoretical

part of the thesis and the forthcoming practical part of it. In order to establish a theoretical

ground for the empirical research, a review of the main and influential approaches of

contrastive analysis was conducted. Then a scientific and workable model for analysis

and translation of IEPVs into Arabic was devised. The model is designed to be used as a

yardstick against which the adequacy and inadequacy of the translation of the subjects of

the study will be measured and evaluated.


250

The discrepancy in achieving the accurate functional-pragmatic equivalents and in the

level of translation communicative competence amongst Arabic professional translators

and Arabic translation students will be clearly manifested in the forthcoming

experimental part of the study. The next Chapter, however, will be devoted to the research

methodology of the study, where such points as research questions, research design, and

data collection will be attended to.


251

CHAPTER FOUR
RESEARCH METHODOLOGY

4.1 Introduction

As it has been previously mentioned, the primary objective of this study is to investigate

the difficulties encountered by Arabic professional translators and Arabic translation

students when tackling the phenomenon of IEPVs, and to propose a number of

recommendations based on the results of the study. In the present Chapter the

methodology employed in this study to conduct the investigation will be outlined by

elaborating on the research design, participants, data collection and data analysis.

4.2 Research design

4.2.1 Research questions

The previous theoretical part of the thesis was dedicated to answer the following

significant questions:

1) What is the notion of idioms and idiomaticity in both English and Arabic?

2) What are the syntactic, semantic, and pragmatic criteria of IEPVs?

3) Are there structures in Arabic similar to that of the IEPVs?

4) What are the similarities and dissimilarities between idiomatic PVs in English

and Arabic?

5) What are the difficulties of learning IEPVs by non-native speakers of English

in general and by Arab learners of English in particular?


252

6) To what extent have English and Arabic lexicographers been successful in

covering, explaining and providing the appropriate equivalents for IEPVs in

their general and specialized dictionaries? And what are the gaps that need to be

closed in this respect?

7) To what extent the methods and materials employed by English and Arabic

pedagogues can help in solving the learning, teaching and translating difficulties

of IEPVs? And what are the gaps that need to be bridged in this difficult area of

pedagogy?

8) How has the phenomenon of IEPVs been dealt with in translation studies?

In the following practical part of this study, however, translation tests were carried out to

pinpoint the translational errors made by Arabic professional translators and Arabic

translation students, and to identify the types of difficulties they encountered when

tackling the problematic features of IEPVs to suggest a range of recommendations for

Arabic professional translators, Arabic lexicographers, and Arabic pedagogues.

The translation tests were designed to address the following pertinent pragmatic, semantic

and syntactic research questions:

1) To what extent were Arabic professional translators and Arabic translation

students successful in providing the Arabic functional-pragmatic equivalents of

IEPVs?

2) To what extent does the issue of complex idioms pose a difficulty for

Arabic professional translators and Arabic translation students, when

translating IEPVs into Arabic?


253

3) Do syntactic forms of IEPVs pose difficulty to Arabic professional translators

and Arabic translation students when dealing with IEPVs?

4) To what extent do the derivational forms of IEPVs affect the process of

translating them into Arabic?

5) It is well known that English is a very productive in coining IEPVs. Many of

them pop up on a regular basis. Such newly coined PVs are occasionally

emerging in both written and spoken modes of the English language where there

are no readily Arabic equivalents for them. Hence, how did the Arabic subjects

deal with such a phenomenon?

6) What are the types of translational errors made by Arabic professional

translators and Arabic translation students?

7) Is there any difference as to translation communicative competence between

the two groups of Arabic subjects: The professional Arabic translators and

Arabic translation students? If so, what are the major areas in which their

competence varies?

4.3 Participants

The translation tests were distributed to two groups of participants, namely: Arabic

professional translators and Arabic translation students.

4.3.1 Arabic professional translators

Twelve Arabic professional translators responded to the translation tests. For the sake of

confidentiality, the participants were given code numbers (P1, P2, P3, etc.) to refer to
254

them throughout the study. Six (50%) of the professional subjects were females while the

other six (50%) were males. The professional subjects were aged between 33 and 67

years, and their period of living in Australia varied between 4 months and 36 years. Ten

(83.3%) of the professional subjects are accredited by NAATI (National Accreditation

Authority for Translators and Interpreters) level three, while the other two (16.6%) had

their accreditations from overseas (one from Iraq and the other one from Morocco). All

professional subjects (100%) participated in this study use Arabic as a language spoken at

home. All of them (100%) have academic qualifications in translation and/or interpreting,

two (16.6%) have BAs in translation and interpreting, two (16.6%) have Graduate

Diplomas in translation and interpreting, five (41.6%) have MAs in translation and

interpreting, one (8.3%) has an MA in translation and linguistics and two (16.6%) have

PhDs in translation and linguistics.

Arabic professional translators participated in this study varied in their years of

experience in the field of translation. Table 4.1 below outlines their experience:

Subjects P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12


code
number
Years 49 4 4 15 7 9 6 4 7 10 6 7
of
experience
Table 4.1 Years of experience of the Arabic professional translators

From the above Table and the academic qualifications details previously sketched out one

can say that Arabic professional translators participated in the study have the relevant

academic qualifications along with good translation practical experience. All of them are

well qualified, accredited, and have actively worked in the field of translation for at least
255

four years. In other words, they have the required theoretical knowledge and practical

experience to insure the validity of the data collected from them.

4.3.2 Arabic translation students

Twelve Arabic translation students in the College of Arts, School of Humanities and

Languages at the University of Western Sydney responded to the translation tests. Half of

them (50%) were undergraduate students doing the BA in translation and interpreting

programs (two (16.6%) in year one, two (16.6%) in year two and two (16.6%) in year

three). The other half (50%) of the Arabic translation students were postgraduate students.

Two (16.6%) of them were doing Graduate diploma in translation and interpreting, two

(16.6%) were doing MA in translation and interpreting and two (16.6%) were doing PhD

in translation and linguistics.

Six (50%) of the Arabic translation students subjects were females while the other six

(50%) were males.

The Arabic translation students were aged between 19 and 40 years, and their period of

living in Australia varied between 7 months and 17 years. Most of the Arabic translation

students (91.6%) that participated in this study use Arabic as a language spoken at home.

One (8.3%) subject only mixes Arabic with Turkish at home. All of the postgraduate

students (50%) have academic qualifications. Two (16.6%) have MA in translation and

interpreting. One (8.3%) has MA in applied linguistics, two (16.6%) have BA of Arts,

and one (8.3%) has Graduate certificate in translation and interpreting.


256

For the sake of confidentiality participants were given code numbers (S1, S2, S3, etc.) to

refer to them throughout the study.

The aim of targeting Arabic translation students is to experiment their knowledge and

ability of translating IEPVs into Arabic from their performance by analysing their

responses to the translation tests, and to compare such knowledge and ability with the

professional translators' knowledge and ability. Such a comparison will enable the present

researcher to explore the level of translation communicative competence each group has.

It will also cast light on types of difficulties encountered by each group when handling

IEPVs. The ultimate aim is to propose a number of recommendations to professional

translators, lexicographers and pedagogues based on the results of the empirical research.

4. 4 Data collection

4.4.1 Instrument

In order to achieve the principal objectives of this thesis, and to address the crucial

research questions outlined above, translation tests of one hundred items were designed as

an empirical instrument and distributed to the two abovementioned groups of Arabic

subjects. Each item of the translation tests contains an IEPV.

A variety of text types has been exploited in these translation tests. The purpose was to

challenge the subjects' abilities in terms of appreciating the register variations and

illocutionary acts when translating IEPVs into Arabic. Short contextual information was

given between square brackets after each item (see Appendix 1). The text types used in
257

the translation tests along with the items represent each text type are outlined in the

following Table:

Text types Items


Fiction novel - Conversation between friends 1,6,7,10, 18, 23, 47, 48, 59,
80, 91, 95, 98
Fiction novel – Conversation between school teachers 2, 97
Fiction novel - Conversation between classmates 5,13, 60, 77
Fiction novel - Conversation at a restaurant 94
Fiction novel - Conversation between a wife and husband 8, 31, 51, 66, 68, 96
Fiction novel - Conversation between fellow-workers 9,12, 14, 20, 24, 25, 55, 67,
70
Fiction novel - Conversation between a daughter and her 90
parents
Fiction novel - Conversation between parents 11, 26, 28, 65
Fiction novel - Conversation between a patient and a 85
doctor
Fiction novel - Conversation between neighbours 16, 21, 52
Fiction novel - Conversation between shoppers 84
Fiction novel - Conversation between two sisters 74
Fiction novel - Advice from parents to their children 72
Fiction novel - Advice from a school teacher to a student 71
Fiction novel - Conversation between townsfolk 56
Fiction novel - Conversation between commuters 29
Fiction novel - Advice from a factory manager to his 30
deputy
Fiction novel - Advice from friend to friend 32
Fiction novel - Announcement at a meeting 41
Fiction novel - Conversation between a father and his son 38
Fiction novel - Mother's advice to her daughter 36
Fiction novel - Recount on a job interview 22
Email - Advice from a businessman to his colleague 37
Biography 33, 58
Letter – union leader to members 34
Horoscope 35
News report 3, 39, 53, 75, 79, 89
News headline 42, 64, 93
Diary entry 4, 40, 54, 81, 83, 100
Medical brochure 99
An internet advertisement for mobile phone screen savers 92
Article on an Art exhibition 15
Article on stress 88
Art review - article 19
Police interview 86
A cross examination in court 76
Parent-teacher interview 17
258

An advertisement for a weight loss product 82


Advice from a doctor to parents in an article in a coffee 73
table magazine
Article on death penalty 69
Transcription of a public speech in the Opening of Law 63
Term Dinner
Background feature article - Conversation between 62
political figures
Article on Ash Wednesday 61
Question in a chat-room 43
Joke 50, 78
Notice on a shop 44
A note written on a flower stall 87
Advice in an article on well-being in a coffee table 57
magazine
Advice from a trade union official to a trader 49
Disclaimer - Medical brochure 46
Women’s magazine article 27

Table 4.2 Text types used in the translation tests and the items representing each type

As it can be noticed in the above Table, the data of the translation tests were taken from

many resources, they include: news articles, editorials, ads, court verdicts and judgments,

news satellite channels, internet web sites, textbooks, brochures, official letters,

magazine, jokes, among others.

Due to the complexity of the topic at hand, and for the sake of reliability and validity of

the empirical instrument, the items of the translation tests were grouped in a way that they

covered all the syntactic and semantic categories of IEPVs to see how each category is

dealt with by each groups of the subject. The categories are: complex idioms, syntactic

forms (verb + adverb, verb + preposition, and verb + adverb + preposition), derivational

forms (adjectives with suffix –en, adjectives without suffix –en, PVs derived from nouns

and nouns derived from PVs) and productive (newly coined) PVs. The categories and

subcategories of IEPVs covered in the translation tests and the items representing each

category/subcategory are presented in the following Table:


259

Categories Subcategories Items


Complex idioms 1-25
Syntactic forms
Verb + adverb 73, 74, 75, 77, 78, 79, 80,
81, 82, 84, 85, 86, 88, 89
and 90
Verb + preposition 76, 83 and 87
Verb + adverb + preposition 57- 72
Derivational forms
Adjectives with suffix –en 47- 51
Adjectives without suffix –en 52- 56
PVs derived from nouns 26-35
Nouns derived from PVs 36 – 46
Productive (newly 91-100
coined) PVs
Table 4.3 Categories/subcategories covered in the translation tests and the items representing each
category

Since the study is all about translation not interpreting, I advised my subjects to adhere to

acceptable written Arabic and avoid delivering colloquial equivalents for IEPVs itemised

in the translation tests (see Appendix 1).

4.4.2 Data collection procedures

As for Arabic professional translators, a list of their names, addresses and contact details

was made by referring to the translators' directory on the website of NAATI (National

Accreditation Authority for Translators and Interpreters). Initial phone contacts, then,

were made with them to brief them of the main aims of the study and to invite them to

take part in it. The contacted professionals were reassured that their names would not

appear in the study as they would be given code numbers for the sake of confidentiality.

They were also reassured that the information collected from them would be safely stored

at the University of Western Sydney. Such information would not be used by anyone
260

except the investigator and it will be destroyed after five years. In addition, they were

informed that they have the right to withdraw from the test at any time without having to

give the investigator any reason.

Interest and willingness were expressed by twelve Arabic professional translators. Copies

of the translation tests were posted to the interested subjects along with postage paid and

self-addressed envelopes to return the answered/unanswered materials.

With regard to Arabic translation students, subjects from the first group (undergraduate

students) were the researcher's students undertaking the Arabic Advanced Language and

Grammar Unit at the University of Western Sydney. They were approached directly by

the researcher in their class, given a brief account of the study and its main objectives and

invited to participate in it. All their questions were clearly answered. Interest and

willingness were expressed by six of them. Copies of the translation tests were handed to

the interested students along with postage paid and self-addressed envelopes to return the

answered/unanswered materials.

Subjects from the second group (postgraduate students) were the chair supervisor's

students undertaking the Graduate diploma in translation and interpreting, the MA in

translation and interpreting or translation and linguistics programs. The researcher with

the help of the chair supervisor arranged a visit to their class. A quick presentation was

given to highlight the main idea of the study and its primary goals. Interest and

willingness were expressed by four of them. Copies of the translation tests were

distributed to the interested postgraduate students along with postage paid and self-

addressed envelopes to return the answered/unanswered materials.


261

The other two subjects (PhD students) were the researcher's close friends. They were

approached directly by the researcher, given a brief account of the study and its main

aims and invited to take part in it. All their questions were manifestly answered. Interest

and willingness were expressed by both of them. Copies of the translation tests were

handed out along with postage paid and self-addressed envelopes to return the

answered/unanswered materials.

As was done with the Arabic professional translators, Arabic translation students were

reassured that they would be given code numbers for the sake of confidentiality so that

their names would not appear in the study. The information collected from them would be

safely stored at the University of Western Sydney and would not be used by anyone

except the researcher, who will destroy it after five years. They also were informed that

they have the right to withdraw from the test at any time without having to give the

researcher any reason.

4.5 Data analysis

In order to address the abovementioned research questions, the analysis and translation

model devised in Chapter Three (see 3.5 above) was utilised so as to produce my

suggested translations to items of the translation tests (see Appendix 2).

To ensure the validity and reliability of my suggested translations, the supervisory panel

of the thesis, who are specialists in the fields of translation, the Arabic language and

linguistics, have carefully checked them and gave me the appropriate feedback as to the

probability of the translations, based on the model discussed and adopted in Chapter
262

Three. However, allowance will be given for other possibilities of acceptable translations

if they meet the criteria adopted in the proposed model.

Such a model-based translation has been used as a yardstick to measure, analyse and

discuss the answers of the subjects participated in the study. Accordingly, the answers

were categorized as correct and incorrect answers. The incorrect ones (translational

errors) were divided into two types, namely: "overtly erroneous errors" and "covertly

erroneous errors". On the other hand, the correct answers were divided into two types,

being: "very good" and "satisfactory" answers. By "very good" I mean the correct

answers that meet all the criteria previously set up to achieve functional-pragmatic

equivalents of the listed IEPVs. Whereas, "satisfactory" means acceptable answers,

which may be semantically understood yet they have less currency. In other words, I set

up a model-based scale to determine the level of correctness of each answer. On the one

hand I have "very good" answers, which are current and extremely acceptable

translations of the listed IEPVs, and on the other hand I have "satisfactory" answers,

which are still acceptable and used but not as current as the previous ones. Illustrative

examples of satisfactory answers are listed in the following Table:

Item IEPVs Satisfactory answers


33 He was shopping around ‫ر‬X/9‫ازن ا‬X1 ‫ن‬7‫آ‬
35 Don't let anyone egg you on S*8[1 ‫[ًا‬Z‫}[ع أ‬M
40 while I had to make do with her 5)3*FY3+‫ ا‬7OY\23\ %_F‫ أن أآ‬%)E ‫ن‬7‫ آ‬73.-\
cast-offs
47 The party brightened up ‫ر‬X^Q+‫ ا‬¥OF\‫ا‬
88 because he doesn't tidy up ‫ء‬7-o9‫}| ا‬,1M ‘P9
91 I'm partied out ‫ت‬2_Q+‫ة ا‬,…‫ آ‬0/ x(*}
96 I'm really shopped out ‫ق‬XYF+‫ ا‬0/ 5ƒ‫ه‬,/ 2ً *8 7P‫أ‬
Table 4.4 Examples of satisfactory answers (for suggested Arabic functional-pragmatic equivalents
see Appendix 2)
263

Satisfactory answers also include answers which have been reduced to sense, and, as a

result, lost the required features of the idiomatic expressions. That is, they delivered

meanings but they stopped short from delivering the crucial factor of idiomaticity.

Examples and discussion of such answers will be given in the next Chapter (see

5.3.6.1.3 below).

I have to make a disclaimer here that I also classified as "satisfactory" the answers that

have delivered the correct translation of the listed IEPVs but contained some

grammatical, spelling and even semantic mistakes in the surrounding words i.e. in the

part of the sentence that does not contain the PV. Thus, I accepted them as correct

answers on the basis of the possibility as discussing spelling, grammatical and semantic

mistakes of the surrounding words is not my concern in this study. Examples of such

answers are listed in the following Table:

Item IEPVs Mistakes in the surrounding words


32 He won't cotton on ‘‫ أو ادراآ‬,/9‫ ا‬4O8 p)E ‫درًا‬7w 0'1 4+
54 My hopes of a better job damped ‫[ }([د‬w K/9‫ا‬
down
73 to use up sugar 51‫ار‬,Q+‫ات ا‬,*Y+‫ا ا‬Xw,Q-+
Table 4.5 Examples of mistakes in the surrounding words (for suggested Arabic functional-
pragmatic equivalents see Appendix 2)

It is worth mentioning that when a given subject provided more than one answer, I have

endeavoured to choose the correct and the most appropriate one in my data analysis.

Back translation, however, is provided when it is required.

A concluding statement will be made to account for the level of translation

communicative competence of each group based on the theory of communicative

competence (knowledge and ability) introduced by Hymes (1971) and improved by such

scholars as Bell (1991) and Campbell (1998). The aim of such a statement is to compare
264

the competence of the two groups judged by their performance (their answers) so as to

highlight the areas in which they have a shortage of knowledge and/or ability. This

statement will help in proposing recommendations in Chapter Six.

Finally, taking up the issue of translating IEPVs into Arabic from many perspectives

required me to deal with the data in a qualitative rather than in a quantitative manner. A

number of illustrative examples will be discussed to highlight the matches and

mismatches of functional-pragmatic equivalents of the listed IEPVs.

In the next Chapter, the results of the translation tests will be collected, classified,

analysed, interpreted and carefully compared against the parameters set up above to

explore the types of difficulties encountered by the Arabic subjects when tackling the

phenomenon of IEPVs, and the reasons behind such difficulties.


265

CHAPTER FIVE
RESULTS AND DISCUSSIONS OF THE TRANSLATION
TESTS

5.1 Introduction

This Chapter is of two distinct sections. The first section is devoted to present a general

overview of the results of the translation tests conducted by the two groups contributed in

the present study: the Arabic professional translators' group and the Arabic translation

students' group.

The second section of this Chapter, however, concerns itself with analysing and

discussing the subjects' performance in light of the second set of the research questions

previously outlined in Chapter One (see 1.3 above).

5.2 General overview of the results of the translation tests

The results of the translation tests of the two groups involved will be generally presented

in this section. As has been mentioned, the translation tests conducted by the two groups

participated in this study was designed to address the abovementioned seven research

questions. For the tests to be more reliable and valid, the questions were set up in a way

that they covered all the syntactic and semantic categories of EPVs, namely: complex

idioms (henceforth c), syntactic forms (henceforth s), derivational forms (henceforth d)

and productive (newly coined) EPVs (henceforth p). The syntactic forms (s) are of three

subcategories, being: verb + adverb (henceforth sva), verb + preposition (henceforth svp)
266

and verb + adverb + preposition (henceforth svap). The derivational forms (d), in

addition, are of four subcategories, they are: adjectives with suffix –en (henceforth den),

adjectives without suffix –en (henceforth de), PVs derived from nouns (henceforth dpn)

and nouns derived from PVs (henceforth dnp). The categories of IEPVs covered in the

translation tests are presented in the following Table:

Complex idioms (c)


Syntactic forms (s)
Verb + adverb (sva)
Verb + preposition (svp)
Verb + adverb + preposition (svap)
Derivational forms (d)
Adjectives with suffix –en (den)
Adjectives without suffix –en (de)
PVs derived from nouns (dpn)
Nouns derived from PVs (dnp)
Productive PVs (p)
Table 5.1 Categories of IEPVs covered in the translation tests

The general overview of results of the translation tests of the two groups involved is

presented below.

5.2.1 General overview of the results of the Arabic professional


translators' group

Table 5.2 below summarizes the average of the correct answers (i.e. both the very good

and the satisfactory answers) of the Arabic professional translators' group. For full details

see Appendix 4.
267

Total average 58%


c 50.66%
s 63.48%
sva 59.44%
svp 33.33%
svap 72.91%
d 65.32%
den 71.66%
de 75%
dpn 58.33%
dnp 64.39%
p 32.50%
Table 5.2 Average of the correct answers of the Arabic professional translators' group

As it is outlined in the above Table, the results of the translation tests showed that the

total average of the correct answers of the Arabic professional translators' group (58%)

was low, which clearly indicates the fact that the IEPVs listed in the translation tests

posed great deal of difficulties to this group of the subjects in that they failed to score a

higher average of correct answers.

In order to measure the amount of variation within the answers of participants, standard

deviation statistics have been carried out. Such a means of statistic casts more light on the

subjects' behavior and it will be a quite useful tool in elaborating on the translation

communicative competence in the forthcoming section 5.3.7.

Table 5.3 below summarizes the standard deviation of the Arabic professional translators'

group. For full details see Appendix 4.

Total standard deviation 0.275812


c 0.256748
s 0.281305
d 0.24216
p 0.27902
Table 5.3 Standard deviation of the Arabic professional translators' group
268

Interestingly, the results of the Arabic professional translators' group showed that the

total standard deviation (0.275812) demonstrates that the amount of variation within the

answers of the subjects of this group was a bit high or at least higher than that of the

other group as we will see later on.

Since the principal aim of this study is to investigate the difficulties encountered by

Arabic professional translators and Arabic translation students when tackling the

phenomenon of IEPVs, it is quite essential to determine the level of difficulty each

category of the IEPVs poses the Arabic professional translators and the Arabic

translation students participated in the study. In order to do so, I ranked the items

according to their percentages of the correct answers in a descending order (see

Appendix 5). The aim of conducting such a statistical method is not only to measure the

level of difficulty each category of the tests poses but also to investigate the reasons

behind such difficulties. And, eventually, this method aims to better falicitate

addressing the research questions in the forthcoming section.

Table 5.4 below outlines the ranking of the categories and subcategories according to

the averages of the correct answers of the Arabic professional translators' group:

Category Subcategory Average of correct


p 32.50%
c 50.66%
s 63.48%
svp 33.33%
sva 59.44%
svap 72.91%
d 65.32%
dpn 58.33%
dnp 64.39%
den 71.66
de 75%
Table 5.4 Ranking the categories and their subcategories - Arabic professional translators' group
269

5.2.2 General overview of the results of the Arabic translation


students' group

Table 5.5 below summarizes the average of the correct answers (i.e. both the very good

and the satisfactory answers) of the Arabic translation students' group. For full details see

Appendix 4.

Total average 48%


c 47%
s 52.20%
sva 60%
svp 27.77%
svap 49.47%
d 51.88%
den 56.66%
de 58.33%
dpn 51.66%
dnp 46.97%
p 24.16%
Table 5.5 Average of the correct answers of the Arabic translation students' group

The results of the translation tests outlined in the above Table showed that the total

average of the correct answers of the Arabic translation students' group (48%) was also

low.

Apart from the average of the correct answers, the statistics of standard deviation of the

Arabic translation students' group are presented in Table 5.3 below. For full details see

Appendix 4.

Total standard deviation 0.24421


c 0.208157
s 0.25891
d 0.198036
p 0.297737
Table 5.6 Standard deviation of the Arabic translation students' group
270

As it can be observed from the Table above, the amount of variation within the answers

of the subjects of this group was rather low, which demonstrates that they were more

consistent in their answers than the Arabic professional translators' group (see Table 5.3

above).

Table 5.7 below sketches the ranking of the categories and subcategories according to

the averages of the correct answers of the Arabic translation students' group:

Category Subcategory Average of correct


p 24.16%
c 47%
d 51.88%
dnp 46.97%
dpn 51.66%
den 56.66%
de 58. 33%
s 52.20%
svp 27.77%
svap 49.47%
sva 60%
Table 5.7 Ranking the categories and their subcategories – Arabic translation students' group

5.2.3 Summary and comparison

Before proceeding to the next section, it is quite fruitful to draw a conclusion here in

terms of the difficulties categories of IEPVs posed to the whole group of subjects

involved in this study (i.e. both the Arabic professional translators' group and the Arabic

translation students' group).

Table 5.8 below summarizes the differences of the averages of the correct answers (the

very good and the satisfactory answers) of both groups:


271

Category Subcategory Professionals Students Difference


Total average 58% 48% 10%
p 32.50% 24.16% 8.33%
c 50.66% 47% 3.67%
d 65.32% 51.88% 13.44%
dpn 58.33% 51.66% 6.67%
dnp 64.39% 49.97% 17.42%
den 71.66% 56.66% 15%
de 75% 58.33% 16.67%
s 63.48% 52.20% 11.29%
sva 59.44% 60% -0.55%
svap 72.91% 49.47% 23.44%
svp 33.33% 27.77% 5.55%
Table 5.8 Differences of the averages of the correct answers

From the abovementioned overview of the results and the differences of the averages of

the correct answers outlined in the Table above one can come up with the following

general findings:

1. Despite the differences between the two groups, the overall performance of the

subjects in general was poor. Such a poor performance demonstrates the fact that the

IEPVs listed in the translation tests posed a great deal of difficulty to both groups of

the subjects.

2. The overall performance of the Arabic professional translators' group, who scored an

average of 58%, is better than that of the Arabic translation students' group, who

scored an average of 48%, with a difference of 10% between the two groups (Table

5.8).

3. The (p) category constituted the most difficult category for both groups (Table 5.8).

4. The (c) category constituted the second hardest category for both groups (Table 5.8).
272

5. The (s) category constitutes the easiest category for the Arabic translation students'

group, while the (d) category represents the easiest category for the Arabic

professional translators' group (Table 5.8).

6. In spite of the abovementioned similarity in terms of ranking the levels of the

difficulty each category posed, the results revealed that the Arabic professional

translators' group performed much better in all of the categories and most of the

subcategories. (Table 5.8).

7. With the categories that are easier (s and d), the Arabic professional translators' group

performed much better than the Arabic translation students' group, while in the

categories that are more difficult (p and c) there was no major difference. This seems

to suggest that training and experience gained by professional translators improved

their performance only in the more teachable and learnable (s and d) categories, but

not in (p and c) categories where the sense of stylistic competence is required (Table

5.8).

8. The ranking of the subcategories of the (s) and (d) categories, according to their levels

of the difficulty, was not the same for the two groups of the study (Table 5.8).

9. Unlike the Arabic professional translators' group, who found the subcategory (dpn)

the most difficult subcategory of the (d) category (58.33%), the results of the Arabic

translation students' group showed that the subcategory (dnp) was the hardest

subcategory of the (d) category (46.97%) (Table 5.8).


273

10. The (de) subcategory constituted the easiest subcategory of the (d) category for both

groups involved. However, the Arabic professional translators' group, who scored an

average of 75%, did much better than the Arabic professional translators' group, who

scored an average of 58.33%, with a difference of 16.67% between the two groups

(Table 5.8).

11. While the (den) subcategory constituted the second easiest for the two groups, the

Arabic professional translators' group, who scored an average of 71.66%, performed

much better than the Arabic translation students' group, who scored an average of

56.66%, with a difference of 15% between the two groups (Table 5.8).

12. Surprisingly, with the (sva) subcategory, the Arabic translation students' group, who

scored an average of 60%, performed slightly better than the Arabic professional

translators' group, who scored an average of 59.44%, with a difference of -0.55%

between the two groups (Table 5.8).

13. With the (svap) subcategory the Arabic professional translators' group, who scored an

average of 72.91%, performed much better than the Arabic translation students' group

who scored an average of 49.47%, with a difference of 23.44% between the two

groups (Table 5.8).

14. The results showed that there were more variations within the answers of Arabic

professional translators' group than that of the Arabic translation students' group with

a difference of 0.031602 between the two groups in the standard derivations (see

Table 5.9 below).


274

15. With the (p) category the Arabic professional translators were more consistent in their

answers. Their standard deviations were lower than those of the Arabic translation

students with a difference of -0.01872 between the two groups (see Table 5.9 below).

16. With the (s) category there was no major difference between the two groups in terms

of the standard deviations. The difference was only 0.022395 between the two groups

(see Table 5.9 below).

17. With (c) and (d) categories the Arabic translation students were more consistent in

their answers than the Arabic professional translators, where the latter had higher

standard deviations than the former. With (c) category the difference was 0.048591,

and with (d) category the difference was 0.044124 (see Table 5.9 below).

Category Professionals Students Difference


Total 0.275812 0.24421 0.031602
c 0.256748 0.208157 0.048591
s 0.281305 0.25891 0.022395
d 0.24216 0.198036 0.044124
p 0.27902 0.297737 -0.01872
Table 5.9 Differences of the standard deviation

5.3 Subjects' performance in light of the research questions

Having given a general overview of the results of the translation tests, it is time now to

explore critical features of such results. In this section I will investigate the performance

of each group of the subjects in relation to the seven research question outlined above.

The aim here is to approach the data collected from the participants from several

different perspectives in an attempt to answer these pertinent research questions.


275

5.3.1 Subjects' performance in light of the first research question

In this subsection the following research question will be tackled:

To what extent were Arabic professional translators and Arabic translation

students successful in providing the Arabic functional-pragmatic equivalents of

IEPVs?

I will start with the performance of the Arabic professional translators' group. Table

5.10 below summarizes the performance of this group. For full details of the raw data

see Appendix 3.

Subject Correct Incorrect Unanswered


Code Very good Satisfactory Overtly Covertly
Number erroneous erroneous
error error
P1 50 14 26 10 0
P2 27 19 46 8 0
P3 32 12 48 8 0
P4 43 13 36 8 0
P5 36 15 36 13 0
P6 38 17 37 7 1
P7 44 15 33 7 1
P8 52 14 26 8 0
P9 40 18 36 6 0
P10 49 14 32 4 1
P11 63 14 18 5 0
P12 29 25 37 6 3
Table 5.10 Summary of the performance of the Arabic professional translators' group

As it has been indicated above, the results of the Arabic professional translators' group

showed that the total average of the correct answers (i.e. both the very good and the

satisfactory answers) of this group was 58% (see Table 5.2 above). Given the fact that

only the answers that labeled as "very good" are considered the perfect functional-
276

pragmatic equivalents of the listed IEPVs, the total average of the answers of the Arabic

professional translators' group that achieved the functional-pragmatic equivalents (very

good answers) was 43.35%. Such answers represented the right matches of the listed

IEPVs according to the criteria set up in the model devised in Chapter Three (see 3.5

above). The other answers, however, were either satisfactory (15.39%) which are not an

ideal translation, overtly erroneous errors (33%), covertly erroneous errors (8%), or

unanswered (0.58%) (see Table 5.11 below).

Category Very good Satisfactory Overtly Covertly unanswered


erroneous erroneous
errors errors
Total 43.35% 15.39% 33% 8% 0.58%
Average
p 19.17% 13.33% 56% 10% 1.67%
c 34% 17% 39% 10% 0%
s 49% 14% 29% 8% 0%
sva 52% 7.22% 32% 8.89% 0%
svap 51% 22% 24% 3% 1%
svp 27.77% 5.55% 39% 27.78% 0%
d 47.85% 17.47% 29% 5% 1%
de 48.33% 26.66% 22% 3% 0%
den 58% 13% 27% 2% 0%
dnp 53.03% 11.36% 30% 4.54% 2%
dpn 36.66% 21.66% 34% 7% 0%
Table 5.11 Summary of results of the Arabic professional translators' group

The results of Arabic translation students' group, on the other hand, showed that they

also varied considerably in achieving the functional-pragmatic equivalents of the listed

IEPVs. The highest percentage was 50% scored by subject number four (S4) and the

lowest percentage was 9% scored by subject number one (S1) (see Table 5.12 below.

For full details of the raw data see Appendix 3).


277

Subject Correct Incorrect Unanswered


Code
Number
Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
S1 9 14 73 4 0
S2 18 11 35 2 34
S3 41 24 32 2 1
S4 50 14 30 6 0
S5 39 19 40 2 0
S6 23 20 31 7 19
S7 27 21 43 6 3
S8 24 20 52 4 0
S9 20 15 49 4 12
S10 46 15 34 5 0
S11 27 14 47 12 0
S12 41 23 33 3 0
Table 5.12 Summary of the performance of the Arabic translation students' group

As previously outlined, the total average of the correct answers of the Arabic

translation students' group was 48% (see Table 5.4 above). Yet, the total average of the

answer of the Arabic translation students' group that achieved the functional-pragmatic

equivalents (very good answers) was only 32%. Such answers characterized the right

matches of the listed IEPVs in line with the criteria set up in the model devised in

Chapter Three (see 3.5 above). The other answers, however, were either satisfactory

(16%), overtly erroneous errors (40%), covertly erroneous errors (6%) or unanswered

(6%) (see Table 5.13 below).

Category Very good Satisfactory Overtly Covertly unanswered


erroneous erroneous
errors errors
Total 32% 16% 40% 6% 6%
Average
p 12% 12% 55% 5.83% 15%
c 19% 28% 48% 3% 2%
s 39% 13.23% 37% 6.62% 5%
sva 49% 11.11% 32% 5% 3%
svap 34% 16% 39.58% 5% 6.25%
svp 17% 11.11% 42% 25% 6%
278

d 36% 16% 38% 3% 7%


de 38% 18% 32% 8% 3%
den 37% 20% 31.66% 3.33% 8.33%
dnp 37.88% 9% 42.42% 1% 10%
dpn 32.50% 19.16% 39% 4% 5%
Table 5.13 Summary of results of the Arabic translation students' group

5.3.1.1 Summary and comparison

Table 5.14 below summarizes the differences in achieving functional-pragmatic

equivalents of both groups:

Category Percentage Percentage Difference


of of
functional- functional-
pragmatic pragmatic
equivalent / equivalent/
Professionals Students
Total 43.35% 32% 11.35%
Average
p 19.17% 12% 7.17%
c 34% 19% 15%
d 47.85% 36% 11.58%
s 49% 39% 10%
Table 5.14 Differences in achieving functional-pragmatic equivalents

Table 5.15 below summarizes the differences of the average of the unanswered items of

both groups:

Category Percentage Percentage Difference


of of
unanswered unanswered
items / items /
Professionals Students
Total 0.58% 6% -5.42%
Average
p 1.67% 15% -13.33%
c 0% 2% -2%
s 0% 5% -5%
sva 0% 3% -3%
279

svap 1% 6.25% -5.25%


svp 0% 6% -6%
d 1% 7% -6%
de 0% 3% -3%
den 0% 8.33% -8.33%
dnp 2% 10% -8%
dpn 0% 5% -5%
Table 5.15 Differences of the average of the unanswered items

From the facts outlined in the Tables above, one can deduce that the overall

achievement of functional-pragmatic equivalents of the listed IEPVs was quite

demanding for both groups who have scored low percentages of very good answers.

However, the Arabic professional translators' group performed better in achieving

functional-pragmatic equivalents (43.35%) than the Arabic translation students' group

(32%) with a difference of 11.35 between the two groups (Table 5.14). In addition, the

Arabic professional translators answered the vast majority of the translation tests in that

the total average of their unanswered questions was pretty low (0.58%), while the

Arabic translation students left 69 (6%) items unanswered, with a difference of -5.42%

between the two groups (Table 5.15).

Moreover, the Arabic professional translators' group made less overtly erroneous errors

(the total average was 33%. See Table 5.11 above) than the Arabic translation students

(the total average was 40%. See Table 5.13 above) with a difference of -7% between the

two groups. The former group, however, made slightly more covertly erroneous errors

(the total average was 8%. See Table 5.11 above) than the later group (the total average

was 6%. See Table 5.13 above) with a difference of 2% between the two groups. Such a

minor difference in the total average of the covertly erroneous errors can be justified by

the major difference in the total average of the unanswered items stated above. That is,
280

if we take the major difference in the total average of the unanswered items into our

consideration, the 2% difference of the average of the covertly erroneous errors will not

mean that the Arabic professional translators are less competent than the Arabic

translation students. They are in fact, to use Campbell's (1998) terms, risk takers, while

the Arabic translation students are more or less prudent.

Finally, the differences between the two groups in terms of achieving the functional-

pragmatic equivalents are not significant and this probably means that the training and

experience the professionals gained did not assist much in translating IEPVs into

Arabic. Given the difference between them and the students who presumably lack that

amount of training and experience was very small (11.35%).

5.3.2 Subjects' performance in light of the second research question

In this subsection the following research question will be dealt with:

To what extent does the issue of complex idioms pose a difficulty for Arabic

professional translators and Arabic translation students when translating

IEPVs into Arabic?

It is well known that tackling English idioms in general cause translators a great deal of

difficulties. Therefore, it is an inescapable fact that dealing with complex idioms, where

IEPVs constitute integral parts of such expressions, involves more difficulties due to the

doubly complex nature of such idioms. The question of complex idioms has been

tackled by a number of lexicographers in such dictionaries as: Longman Dictionary of


281

Phrasal Verbs, Oxford Dictionary of Phrasal Verbs, Cambridge International

Dictionary of Phrasal Verbs and the Oxford Phrasal Verbs Dictionary for Learners of

English (see 2.8.2 above).

The results of the present study revealed that complex idioms (c) category (represented

by items 1-25 of the translation tests) pose a large amount of difficulty to both Arabic

professional translators' group and Arabic translation students' group. This category

constituted the second hardest one for both groups (see Tables 5.4 and 5.7 above). The

percentage of functional-pragmatic equivalents (very good answers) of this category

scored by the Arabic professional translators' group was 34%, while that scored by the

Arabic translation students' was 19%, with a difference of 15% between the two groups

(see Table 5.14 above).

Table 5.16 below presents the differences in achieving functional-pragmatic equivalents

(very good answers) of the (c) category:

Question Category Percentage of Percentage of Difference


Number functional- functional-
pragmatic pragmatic
equivalent / equivalent/
Professionals Students
1 c 0% 0% 0%
2 c 50% 41.66% 8.34%
3 c 50% 8.33% 41.67%
4 c 16.66% 25% -8.34%
5 c 41.66% 25% 16.66%
6 c 16.66% 8.33% 8.33%
7 c 25% 25% 0%
8 c 83.33% 50% 33.33%
9 c 16.66% 8.33% 8.33%
10 c 33.33% 25% 8.33%
11 c 50% 16.66% 33.34%
12 c 41.66% 8.33% 33.33%
282

13 c 8.33% 0% 8.33%
14 c 41.66% 8.33% 33.33%
15 c 16.66% 25% -8.34%
16 c 16.66% 16.66% 0%
17 c 33.33% 8.33% 25.00%
18 c 75% 25% 50%
19 c 16.66% 25% -8.34%
20 c 58.33% 16.66% 41.67%
21 c 25% 33.33% -8.33%
22 c 25% 41.66% -16.66%
23 c 58.33% 0% 58.33%
24 c 25% 16.66% 8.34%
25 c 16.66% 25% -8.34%
Table 5.16 Differences in achieving functional-pragmatic equivalents of the Complex idioms
category

It must be pointed out that the items of this category, as it can be observed from the

above Table, varied in terms of their difficulties. Item one, for instance, constituted the

most difficult one for the subjects surveyed where no one was able to achieve its

functional-pragmatic equivalent. While item eight of this category was relatively easy

as ten (83.33%) of the twelve Arabic professional translators and six (50%) of the

Arabic translation students surveyed were able to answer it perfectly (see Table 5.16

above).

In spite of the fact that the performance of the Arabic professional translators' group in

general was slightly better than that of the Arabic translation students, there were six

items in which the performance of the latter was better than that of the former. And

there were two items in which the performance of both groups was equal (see table 5.16

above). This entails a haphazard strategy by both groups to dealing with IEPVs and

highlights the need for a systematic approach to translating these devices.

In addition, the Arabic professional translators' group abandoned only one item of the (c)

category, leaving it unanswered (the total average of the unanswered items was 0%),
283

whereas the Arabic translation students left six items unanswered where the total average

of the unanswered items was 2% with a difference of -2% between the two groups (see

Table 5.15 above).

Lastly, the Arabic professional translators performed much better than the Arabic

translation students with regard to the (c) category with a difference of 15% between the

two groups (see Table 5.14 above).

5.3.3 Subjects' performance in light of the third research question

In this subsection the following research question will be addressed:

Do syntactic forms of IEPVs pose difficulty to Arabic professional translators

and Arabic translation students when dealing with IEPVs?

As has been previously mentioned (see 2.4.2 above) IEPVs are of three syntactic forms,

namely: verb + adverb, verb + preposition and verb + adverb + preposition.

The results of the Arabic professional translators surveyed in this study showed that the

average of the correct answers (both the very good and the satisfactory answers) of the

items containing IEPVs of the verb + adverb form (sva) was 59.44%, while the average of

correct answers of the items containing IEPVs of the verb + preposition form (svp) was

33.33%. And the average of correct answers of the items containing IEPVs of the verb +

adverb + preposition form (svap) was 72.91%. Such percentages obviously indicate that

for the Arabic professional translators surveyed the (svp) form constitutes the most
284

difficult grammatical form, followed by the (sva) form and (svap) form (see Tables 5.2

and 5.4 above).

On the other hand, the results of Arabic translation students surveyed showed that the

average of the correct answers of the (sva) form was 60%, while the average of correct

answers of the (svp) form was 27.77% and the average of correct answers of the (svap)

form was 49. 47%. Therefore, for the Arabic translation students' group, the (svp) form

constitutes the most difficult grammatical form followed by the (svap) form and (sva)

form (see Tables 5.5 and 5.7 above).

Table 5.17 below summarizes the differences in achieving functional-pragmatic

equivalents (the very good answers) of the syntactic forms (s) category:

Category Average of Average of Difference


functional- functional-
pragmatic pragmatic
equivalent / equivalent/
Professionals Students
s 49% 39% 10%
svp 27.77% 17% 10.77%
svap 51% 34% 17%
sva 52% 49% 3%
Table 5.17 Differences in achieving functional-pragmatic equivalents of the syntactic forms (s)
category

The Table above indicates the following: firstly, even though the (s) category, as a whole,

constituted the easiest category for both groups (see Table 5.14 above), the subjects

encountered considerable difficulties when tackling the items containing such forms

judging from the low averages of the functional-pragmatic equivalents achieved (see

Table 5.17 above).


285

Secondly, the difference in achieving functional-pragmatic equivalents of the

subcategories was relatively large in the (svp) and (svap) subcategories, whilst it was

minimal in the (sva) subcategory (see Table 5.17 above).

Thirdly, the (svp) subcategory represented the most difficult one for both groups followed

by the (svap) subcategory, which constituted the second hardest one and the (sva)

subcategory, which was the easiest subcategory for both groups.

Fourthly, the Arabic professional translators' group left only one item of the (s) category

unanswered (the total average of the unanswered was 0%), whereas the Arabic translation

students left nineteen items unanswered, where the total average of the unanswered was

5% with a difference of -5% between the two groups (see Table 5.15 above).

Finally, overall, the Arabic professional translators perform better than the Arabic

translation students with regard to this category with a difference of 10% between the two

groups (see Table 5.14 above).

5.3.4 Subjects' performance in light of the fourth research question

The phenomenon of derivation in general and derivation of IEPVs in particular has been

exclusively accounted for in Chapter Two (see 2.7.3 above). In this subsection the

following research question will be dealt with:

To what extent do the derivational forms of IEPVs affect the process of

translating them into Arabic?


286

The results of the Arabic professional translators' group revealed that the average of the

correct answers (i.e. both the very good and the satisfactory answers) of the items

containing IEPVs derived from adjectives with suffix –en (den) was 71.66%. The

average of the correct answers of the items containing IEPVs derived from adjectives

without suffix -en (de) was 75%. The average of the correct answers of the items

containing IEPVs derived from nouns (dpn) was 58.33%. The average of the correct

answers of the items containing nouns derived from IEPVs (dnp) was 64.39%. Such

percentages indicate that for the Arabic professional translators surveyed the (dpn) form

constituted the most difficult derivational form, followed by the (dnp) from, (den) form

and (de) form.

On the other hand, the results of the Arabic translation students surveyed revealed that

the average of the correct answers of the (den) was 56.66%. The average of the correct

answers of the (de) was 52%. The average of the (dpn) was 51.66%. The average of the

correct answers of the (dnp) was 46.97%. Such percentages indicate that for the Arabic

translation students surveyed the (dnp) form constituted the most difficult derivational

form, followed by the (dpn) form, (den) form and (de) form.

Moving on to the functional-pragmatic equivalents (the very good answers) achieved by

each group in respect to this category, Table 5.18 below summarizes the differences in

achieving functional-pragmatic equivalents of the derivational forms category:


287

Category Average of Average of Difference


functional- functional-
pragmatic pragmatic
equivalent / equivalent/
Professionals Students
d 47.85% 36% 11.83%
de 48.33% 38% 10%
den 58% 37% 21%
dnp 53.03% 37.88% 15.157%
dpn 36.66% 32.50% 4.17%
Table 5.18 Differences of achieving functional-pragmatic equivalents of the Derivational forms
category

The results revealed that the (dpn) subcategory represented the most difficult one for both

the Arabic professional translators (with the average of 36.66%) and the Arabic

translation students (with the average of 32.50%) with the difference of 4.17% between

the two groups (see Table 5.18 above). Whereas the two groups differ in terms of the

easiest subcategory as it was the (den) for the Arabic professional translators 58% and the

(de) for the Arabic translation students 38% (see Table 5.18 above).

However, in spite of the previously stated fact that this category posed less difficulty than

(p) and (c) categories for both groups (see Table 5.14 above), it does not mean that it

caused no problems for the subjects surveyed. They seem to have encountered a great

deal of difficulties when translating the items containing the derivational forms of IEPVs

and this is particularly evident from the low averages of the functional-pragmatic

equivalents achieved (see Table 5.18 above). The main reason behind such difficulties, in

my view, is the differences in derivational mechanisms between the English and Arabic

languages (see Chapter Two, section 2.7 above) which made the derivational forms of

IEPVs odd and unusual for the Arabic subjects.


288

As for the unanswered items, the Arabic professional translators' group left only two

items of the (d) category unanswered (the total average of the unanswered was 1%),

whereas the Arabic translation students left sixteen items unanswered, where the total

average of the unanswered was 7% with a difference of -6% between the two groups

(see Table 5.15 above).

The Arabic professional translators clearly perform better than the Arabic translation

students concerning this category with a difference of 11.83% between the two groups

(Table 5.18 above).

5.3.5 Subjects' performance in light of the fifth research question

In this subsection the following research question will be tackled:

It is well known that English is a very productive in coining idiomatic PVs.

Many of them pop up on a regular basis. Such newly coined EPVs are

occasionally emerging in both written and spoken modes of the English

language where there are no readily Arabic equivalents for them. Hence, how

did the Arabic subjects deal with such a phenomenon?

The items of this category (number 91-100 in the translation tests. See Appendix 1)

were carefully selected to insure that they were not listed in any of the published

monolingual or bilingual dictionaries.


289

The results of this study revealed that the (p) category poses a great deal of difficulty to

both the Arabic professional translators' group and Arabic translation students' group. The

category constituted the most difficult category for both groups (see Tables 5.8 and 5.14

above). The results of the Arabic professional translators showed that the average of

correct answers (i.e. both very good and satisfactory answers) of this category was

32.50%. And the results of the Arabic translation students revealed that the average of the

correct answers was 24.61%. With a difference of 8.33% between the two groups (see

Table 5.8 above).

However, the percentage of functional-pragmatic equivalents (the very good answers)

of this category scored by the Arabic professional translators' group was 19.17%, while

that scored by the Arabic translation students was 12%, with a difference of 7.17%

between the two groups (see Table 5.14 above).

Table 5.19 below summarizes the differences in achieving functional-pragmatic

equivalents of the (p) category:

Question Category Percentage of Percentage of Difference


Number functional- functional-
pragmatic pragmatic
equivalent / equivalent/
Professionals Students
91 p 8.33% 0% 8.33%
92 p 50% 33.33% 16.67%
93 p 8.33% 0% 8.33%
94 p 25% 16.66% 8.34%
95 p 50% 50% 0%
96 p 25% 16.66% 8.34%
97 p 0% 0% 0%
98 p 0% 0% 0%
99 p 25% 0% 25%
100 p 0% 0% 0%
Table 5.19 Differences in achieving functional-pragmatic equivalents of the Productive PVs
category
290

It is crucial to indicate that the items of this category varied as to their difficulties. Items

97, 98 and 100, for instance, constituted the most difficult items for the subjects

surveyed where no one of them was able to achieve their functional-pragmatic

equivalents. While item 95 of this category was more or less easier as six (50%) of the

twelve Arabic professional translators and six (50%) of the Arabic translation students

surveyed were able to answer it perfectly (see Table 5.19 above).

The Arabic professional translators' group left only two items of the (p) category

unanswered (the total average of the unanswered items was 1.67%), whereas the Arabic

translation students left eighteen items unanswered, where the total average of the

unanswered items was 15% with a difference of -13.33% between the two groups (Table

5.15 above).

Moreover, despite the fact that the performance of the Arabic professional translators'

group in general was rather better than that of the Arabic translation students with

regards to this category with a difference of 7.17% between the two groups, the

performance of both groups was extremely poor (Table 5.19 above).

There are two reasons behind such a poor performance of the subjects. Firstly, the

previously mentioned fact (see Chapter Two, section 2.8 above) that the productive

nature of IEPVs makes it harder for lexicographers to have them listed into their

dictionaries. As a result, even the specialized dictionaries are far from being

comprehensive enough in their coverage of such verbs. The gaps left in coverage result

in missing a number of newly coined EPVs, leaving translators with no choice but to
291

work them out individually and intuitively in an attempt to create Arabic functional-

pragmatic equivalents for them, which may or may not be accurate.

Another reason, secondly, is the ineffective and old-fashioned teaching methods and

materials employed by Arab pedagogues in teaching IEPVs. In other words, there are

no practical teaching methods and materials that may assist the Arabic translators and

translation students to acquire the competence to tackle the phenomenon of newly

coined IEPVs. As has been argued in Chapter Two, EPVs are not random combinations

of verbs and particles. There are, however, patterns underlying them. Pedagogues such

as Side (1990), Hannan (1998), Sawyer (2000), and Sansome (2000) are totally

persuaded that EPVs have to be taught by knowing the patterns underlying them so as

to pinpoint the system and the logic by which they work. Arab pedagogues, regrettably,

still believe that EPVs are random combinations of verbs and particles, which must be

memorized by heart (see 2.10.2 above). As a result, they did not try to investigate the

patterns underling EPVs to provide Arabic translators and Arab learners of English with

reliable pedagogical materials and a systematic approach that can help overcome the

dilemma of translating the newly coined IEPVs.

5.3.6 Subjects' performance in light of the sixth research question

In this subsection the data collected from both groups of the subjects will be closely

looked into from one particular perspective, namely: the types of translational errors

made by them. The aim here is to address the following research question:
292

What are the types of translational errors made by Arabic professional

translators and Arabic translation students?

As has been earlier pointed out in Chapter Three (see 3.5 above), the model I devised

for analysis and translation of IEPVs into Arabic contains the parameter of

overtly/covertly erroneous errors, introduced and implemented by House (1977, 1981,

1997 and 2001). Accordingly, the mismatches or errors made by the two groups of the

subjects of the study were divided into two types: overtly erroneous errors and covertly

erroneous errors. Such a process of division well helped not only in determining the

reasons behind the translational pitfalls made by the subjects surveyed but also in

evaluating the subjects' translation communicative competence from their performance

(see 5.3.7 below). Moreover, I strongly believe that translators have to take these two

types of errors into consideration in order to avoid them when tackling the phenomenon

of IEPVs.

More details on the overtly erroneous errors and covertly erroneous errors made by both

the Arabic professional translators and the Arabic translation students are detailed

below.

5.3.6.1 Overtly erroneous errors

As presented in Tables 5.11 and 5.13 above, the total average of the overtly erroneous

errors made by the Arabic professional translators was 33% and that made by the

Arabic translation students was 40% with a difference of -7% between the two groups.
293

The idiomaticity of EPVs constitutes the main reason behind the overtly erroneous

errors made by the subjects involved. Such an idiomatic nature played a significant role

in hindering the subjects from obtaining the denotative meanings of the listed IEPVs.

The overtly erroneous errors made by the Arabic professional translators and the Arabic

translation students surveyed in this study were of four types, namely:

5.3.6.1.1 Literal translation

Literal translation or, as it is termed by Newmark (1988), "Translationese", is a word-for-

word translation of a text which "does not produce the appropriate sense" (p. 285).

IEPVs, as argued in Chapter Two (see 2.5.2 above) carry two types of meanings,

literal/non-idiomatic meanings and metaphorical/idiomatic meanings. Their idiomatic

nature (especially the examples utilized in the translation tests given to the subjects of this

study) makes them non-transparent and hard to be understood from their separate parts.

That is, their communicative meanings are utterly different from the total sum of the

meanings of their individual components. Some of the Arabic professional translators and

the Arabic translation students surveyed in this study perceived the listed IEPVs literally

and dealt with the two parts of the combination separately, rather than appreciating their

metaphorical/idiomatic meaning and looking at each combination as one semantic unit.

Illustrative examples of this type of translational errors from the two groups are presented

below:

Item IEPVs Literal translations


58 She always looks up to her parents 7O1[+‫ وا‬p+‫ وإ‬p)E9‫ ا‬p+‫ إ‬,€.} 7ً3L‫ دا‬%‫ه‬
84 to take in old ladies 0Y+‫ ا‬,-(‫ آ‬%8 ‫س‬7.+‫~وا ا‬y•1 ‫أن‬
47 The party brightened up K*Fo‫ء \ ا‬7R‫ \ أ‬K_Q+‫ ا‬¥َ ‫ه‬¢ X}
87 but they will grow on you in the end ,/9‫ ا‬,y† %8 S-)E X3.1 ‫ف‬XH
294

83 I ran into my wife %F„‫و‬Š\ x ُ /[{`‫ا‬


68 I'm tied up with something urgent ‫ريء‬7W ‫ء‬%N\ [ّ-ƒ/ 7P‫ريء\ أ‬7W ‫ء‬%N\ ™(},/ 7P‫أ‬
61 what are you giving up for Lent? ,-('+‫م ا‬XU)+ %{*} ‫ذا‬7/
18 will bring him down to earth. ‫رض‬9‫ ا‬p+‫ل إ‬Š.1 ‘)*I1
20 he'll come down on you like a ton of ‫ب‬X{+‫ ا‬0/ ‫ن‬7.W•‫ آ‬S-)E ‫ل‬Š.1 ‫ف‬XH
bricks
28 Why are the children all dolled up? ‫م‬7.`97‫ل آ‬7_W9‫• ا‬ƒ1 4+
6 You're going to drive me around the •{*.3+‫ل ا‬XZ %P‫د‬Xƒ} ‫ف‬XH
bend!
7 My eyes popped out %.-E 7Fw[Z x*Y}‫ي \ ا‬7.-E x*ƒ_P‫\ ا‬xQF_P‫ ا‬%PX-E
3 put back the clock ‫راء‬X)+ 5E7Y+‫^] ا‬1
4 I could have bitten my tongue off %P7Y+ x
ُ 3^w
22 he got down to brass tacks ‘)3*\ ‫م‬Xƒ1 ‫ن‬7‫ آ‬73.-\ ]w‫‘ و‬.'+‫و‬
86 will be taken down ~yž1 ‫ف‬XH
33 He was shopping around ‫ة‬,8XF3+‫ وا‬5R‫و‬,*3+‫ء ا‬7-o9‫ى ا‬,1 ‫ن‬7‫آ‬
79 I've cleaned up my act %8ّ,U} ‫ت‬ ُ ,ّOW [w
74 to build up skills S}‫را‬7O/ ‫ء‬7.(+
91 I'm partied out ,y† ‫ن‬7'/ %8 7ً„‫ر‬7y x ُ )_FZ‫ا‬
89 chilling out at the beach ‹W7N+‫ ا‬p)E ‫د‬,(+‫ا ا‬XP7*1
50 I want to fatten it up as fast as possible ‫ى‬XUƒ+‫ ا‬5E,Y+7\ 7ً .-3H ‫ ه~ا‬K*„‫[ أن أ‬1‫أر‬
93 Latham sexed up Iraq brief ‫اق‬,*+‫ ا‬%8 ‫ر‬7(y9‫] ا‬/7I1 4…1M
Table 5.20 Examples of literal translations (for the suggested Arabic functional-pragmatic
equivalents see Appendix 2)

In the first example (item 58) in the Table above, the IEPV look up was wrongly

literally translated as p)E9‫ ا‬p+‫ إ‬,€.}. Similarly, in the second example (item 84) the IEPV

take in was translated as %&8 ~&y•1 where both the verb and the particle were literally

translated into Arabic.

In some examples, however, the first components of the listed IEPVs (the verb) were

literally translated, while the second components (the adverb and/or the preposition)

were overlooked. The verb brightened in the third example (item 47), for instance, was

erroneously translated literally as K*Fo‫ء \ ا‬7R‫ \ أ‬¥


َ ‫ه‬¢ X} while the particle up was ignored. In

the same way, the first part of the IEPV sex up in the last example (item 93) was

literally translated as ]/7&I1 and the particle up was disregarded. The same is applicable

to the rest of the examples presented in that Table where either the two components of

the IEPVs were taken literally or only the first one while the other one left untranslated.
295

There are several apparent reasons for such distorted translations. Firstly, interference

of the L1 of the subjects (Arabic) with the L2 (English). That is to say, the syntactic

structure of IEPVs is totally odd for the Arabic subjects who are all Arabic native

speakers. As it has been explained in Chapter Two (see 2.6 above), unlike English,

Arabic does not allow proper verbs to collocate with adverbs. And the only type of PVs

in Arabic is that of verb + preposition structure. Therefore, the subjects considered the

literal meaning of the second parts of the listed IEPVs and either translated them as

prepositions or ignored them. Lindstormberg (1991) indicates that "wrong

understanding of preposition is the main source of difficulties that so-called phrasal

verb present to learners of English as a second or foreign language" (p. 47).

Secondly, the deceptive appearance of EPVs in general, Khalaili (1979) writes that they

"look deceptively easy to the foreigners at first sight, but their meanings can be

radically different from what one might expect" (p. 5). Such a deceptive appearance

may tempt translators to provide rushed literal translations without carefully taking into

account the idiomatic nature of this kind of verb.

Thirdly, great deals of IEPVs are not covered by bilingual English-Arabic general and

even specialized dictionaries. This is particularly due to the productive nature of IEPVs

which makes them hard to be captured and listed in dictionaries. This point implies that

the subjects understood or suspected the existence of the IEPVs, but failed to find the

appropriate resources that explain these verbs or provide Arabic equivalents, which

suggests a lack of translation skills.


296

Finally, the decotextualization way of dealing with the IEPVs covered in dictionaries,

especially the bilingual English-Arabic dictionaries which provide out of context,

decontextualized and idealised equivalents for IEPVs listed in them. Such a

decontextualization phenomenon appeared not only in the bilingual English-Arabic

dictionaries but also in the textbooks devoted to teach EPVs to Arab students (see

Chapter Two, section 2.10.2 above) which are, to some extent, to blame for producing

literal translations for IEPVs. Side (1990) makes the point that PVs ought to be learned

in contexts rather than in lists of unrelated words. He labels such an issue as

"contextualization" and elaborates that "single examples [of EPVs] should never be

taught in isolation", therefore, connections always have to "be made in order to

establish their context within the language, to show they are meaningfully idiomatic

rather than meaninglessly random" (pp. 150-151). Moreover, the majority of

researchers (see 2.10 above) advocate the necessity of teaching PVs in context as

presenting them in contexts enhances greatly their learnability and considered far better

than presenting them as unrelated elements.

Along these lines, Newmark (1988) attributes producing a literal translation to two

reasons, being: interference "if the TL [target language] is not the translator's language

of habitual use, or to automatic acceptance of dictionary meanings" (p. 285).

5.3.6.1.2 Mistranslating

Most IEPVs are polysemic, i.e. they have more than one meaning (see 2.5.2 above). To

use the words of McArthur and Atkins (1974), "they have so many shades of meanings"

(p. 5). Some of the Arabic professional translators surveyed in this study were not able

to appreciate such a polysemic nature of the IEPVs listed in the translation tests when
297

rendering them into Arabic. Examples of this type of translational errors from the two

groups are presented below:

Item IEPVs Mistranslations


17 but there are two other students ‘P7Y87.1 / ‫ذو‬7Q1
breathing down his neck
9 I'm fed up to my back teeth with K3*+‫ ا‬0/ xƒ‫ \ زه‬K3*+‫ ا‬0/ ,I^F/ 7P‫ا‬
work
11 goes over her head ‫ف‬X„‫™ أ‬L7Q\ ‫{[م‬U1
21 but he fell down on the job. ‘)3E %8 ,¢Uw [ƒ+
27 I used to wolf my food down so ‫ة‬,-(‫ آ‬5E,Y\ %/7*W ‫ول‬7.}‫ أ‬x
ُ .‫آ‬
quickly
36 It's going to be a walk-over ‫ًا‬,UP ‫ن‬X'1 ‫ف‬XH
37 It'll be a total wash-out. 7ً\‫ا‬,y ‫ن‬X'1 ‫ف‬XH
39 a political stand-off %H7-H ‫دل‬7*}
40 while I had to make do with her 7O}2^_\ %_F‫أن أآ‬
cast-offs
75 to work it out ‫ز‬7IPJ
90 The teacher told me off %.n¢\‫و‬
92 Funk up ‫زوّد \ أ\[ل‬
85 passed away •‫ ا‬53Z‫ ر‬p+‫ إ‬KƒFP‫ا‬
Table 5.21 Examples of mistranslations (for suggested Arabic functional-pragmatic equivalents see
Appendix 2)

In the abovementioned examples the IEPVs were mistranslated and their functional-

pragmatic equivalents were mismatched. In the first example (item 17) ‘P7Y&87.1 was the

incorrect translation for the idiomatic expression breathing down his neck because such

a translation did not deliver the complete meaning of this expression, while 5Y87./ ‘P7Y87.1

‫[ة‬1[&o is the correct one for it has the root echoing ‚&){3+‫ل ا‬X&*_3+‫ ا‬which not only covered

the required part of the meaning but also gives an air of idiomaticity to the Arabic

equivalent. The same argument applies to the second example (item 9) where ,I^&F/ 7P‫أ‬

K3*+‫ ا‬0/ was not the right mach of I'm fed up to my back teeth with work as semantically

and pragmatically there is something missing in this translation, that is, the sense of

exaggeration of the boredom. The Arabic fixed expression K-'+‫ ا‬%\ •_W can be considered

the correct translation where both the denotative meaning and the idiomaticity flavor
298

are delivered. The same argument is applicable to the rest of the examples outlined in

that Table.

The reasons behind producing such mistaken translations are threefold: First, as

mentioned above, the polysemic nature of the IEPVs makes it difficult for the subjects

to choose the appropriate meaning from the variety of shades of meanings given to each

PV. In order to determine the accurate sense, a translator has to appreciate the

situational context in which a given PV is used.

Second, what makes the task of choosing the appropriate meaning more difficult is the

fact that IEPVs are very much confusing in their structure. That is, one proper verb can

collocate with a number of particles to form a range of PVs with many different

meanings. On the other hand, one particle may co-occur with a number of proper verbs

to form a variety of PVs of diverse meanings.

Third, the treatment of the phenomenon of PVs in general and specialized dictionaries

is insufficient. That is to say, in their dealing with PVs, lexicographers either bypass a

great deal of them or provide inadequate definitions for the listed ones (see Chapter

Two, sections 2.8 and 2.9 above).

5.3.6.1.3 Reducing the idioms to sense

This occurred when the subjects may have understood the functional meaning of a

given IEPV in the source language (English) but did not deliver it correctly,

stylistically, into the target language (Arabic). They instead produced a translation in
299

which the denotative meaning is captured but at the expense of the idiomaticity of the

given EPV. Such a type of translational pitfalls was manifested in the subjects' answers

to the complex idioms (c) category. Given the double complexity of these items, their

Arabic equivalents have to be of an idiomatic nature. That is, they have to be translated

by employing APVs, proverbs or fixed expressions, which may require a bit of research

and thinking. Consequently, if there is a one to one Arabic idiom and the translator did

not use it, this means that s/he lacks adequate competence to deal with these kinds of

complex idioms.

Answers of this type of translational pitfalls were considered "satisfactory answers". To

be precise, they are correct as far as they deliver the denotative meanings of the IEPVs.

The reason behind mentioning them here, however, is the fact that the subjects

committing this kind of pitfall have failed to provide the perfect translation (functional-

pragmatic equivalence) of these items. Therefore, it is quite crucial to highlight such a

pitfall in order to raise translators' awareness of maintaining both meaning and style

when translating idiomatic expressions.

Illustrative examples of this type of translational pitfalls are listed in the following

Table:

Item IEPVs Reducing the idioms to sense

3 put back the clock p^/ 7/ [-*1


11 goes over her head 7—-o 4O_}M \ 7O},‫ ذاآ‬%8 ‚)*1M
12 I'll have to pull up my socks ‫اي‬Xw K‫] آ‬3IFH‫ ان ا‬%)E\ [OF„‫ ان ا‬%)E
23 but I don't know if it will ever get 5ƒ-ƒZ •(U-H ‫وع‬,N3+‫ن ه~ا ا‬7‫ إذا آ‬7/ ‫أدري‬M
off the ground.
25 to have it out with him S+~\ ‫م‬7-ƒ+‫[م ا‬E ‘./ |)W‫‘ وأ‬Z‫ر‬7`‫*‘ وأ‬/ 4ّ)'}‫أ‬
%)3*\ Ky[F+‫[م ا‬E\%*/
Table 5.22 Examples of idioms reduced to sense (for suggested Arabic functional-pragmatic
equivalents see Appendix 2)
300

The translation given to the first example (item 3) in the above Table has the denotative

meaning of the complex idiom at hand, but it lacks the idiomaticity required to achieve

its functional-pragmatic equivalent. To give an air of idiomaticity to this sentence, the

subject has to resort to the Arabic one-to-one equivalent of the item, which is ‫رب‬7ƒE [-*1

‫راء‬X+‫ ا‬p+‫ ا‬5E7Y+‫ ا‬.

In the second example (item 11) the Arabic fixed expressions ‫ام‬,&'+‫ور ا‬,&/ 7O.‫ \&~ه‬,&31 or

5&8Š+‫ ا‬%&8 ‫ش‬,W97&‫ آ‬seem to be the perfect functional-pragmatic equivalents, which deliver

not only the denotative meaning of the expression but also the idiomatic flavor of it. In

addition, the currency of such equivalents and the frequency of their usage in both

written and spoken Arabic meet our criterion of the middle ground between the classic

and the colloquial Arabic. Such a criterion, as it has been previously explained in

Chapter Two (see 2.6 above), is very much required in translating IEPVs due to the fact

that they are commonly used in spoken and informal written English. The same

argument is valid to the rest of the examples listed in the above Table.

5.3.6.1.3 Breaching of the Arabic language system

House (1977) indicates that overtly erroneous errors mean the non-dimensional

mismatches which include "both mismatches of the denotative meanings of ST and TT

elements and breaches of the target language system" (p. 245). Having covered the

mismatches of the denotative meanings of the IEPVs listed in the translation tests, it is

time to move on to the second kind of the overtly erroneous errors, i.e. the breaches of

the target language (Arabic) system.


301

The data collected from the subjects involved in the present study revealed that some of

them have, to a great extent, breached the Arabic language system by producing

grammatically and/or morphologically incorrect translations.

Examples of this type of translational errors from the two groups are presented in the

following Table:

Item IEPVs Breaches to the Arabic language system


8 he was falling for her 7O\ ‚ّ)*F-H ‘()w ‫ن‬
¢‫أ‬
18 will bring him down to earth ‘\‫ا‬X` ‘+ [-*1
2 you've let the cat out of the bag ‫ن‬¤‫ ا‬,Y+7\ x َ Q\‫أ‬
64 Net closes in on £50m robbers 0-w‫ر‬7Y+‫ ا‬p)E “(ƒ+‫ء ا‬7ƒ+‫ إ‬p)E ‫ا‬X8,o‫أ‬
96 I'm really shopped out ‫ّق‬XYF+‫ ا‬0/ S)Oُ/ 7P‫أ‬
Table 5.23 Examples of breaches to the Arabic language system (for suggested Arabic functional-
pragmatic equivalents see Appendix 2)

The first example in the above Table (item 8) illustrates a grammatically incorrect

translation. The sentence with the past continuous tense he was falling for her was

mistakenly translated into Arabic by a sentence of future tense 7O\ ‚ّ)*F-H ‘()w ‫ن‬
¢ ‫( أ‬literally:

that his heart will be attached to her). In the second example (item 18), the Arabic

translation provided is incorrect because the wrong usage of the Arabic preposition.

Arabic native speakers would more likely use ‘\‫ا‬X` p+‫[ ُ إ‬-*ُ1 .

The third example (item 2) illustrates a morphologically incorrect translation, where the

form of the past tense of the Arabic verb ‫ح‬


َ 7\ was erroneously spelt as ‫ح‬
َ 7\‫ أ‬by adding to it

the consonant ‫أ‬.

The rest of the examples (items 64 and 96) exemplify the wrong usage of the Arabic

verbs. The verb ‫ا‬X8,&o‫ أ‬in example 64 was used mistakenly because it delivers the wrong

meaning, which is to monitor, oversee, supervise etc, the correct Arabic verb which can
302

be used in such a context is p&)E ‫ا‬X8‫ر‬7&o or p&)E ‫ا‬X'&o‫( أو‬meaning: they are about to). The

same argument applies to S)Oُ/ 7P‫ أ‬in example 96 which has to be SO.ُ/ 7P‫ أ‬.

5.3.6.2 Covertly erroneous errors

The total average of the covertly erroneous errors made by the Arabic professional

translators was 8% (see Table 5.11 above), whereas the total average of the covertly

erroneous errors made by the Arabic translation students was 6% with a difference of

(2%) between the two groups in favour of students (see Table 5.13 above).

Covertly erroneous errors are typically caused by the dimensional mismatches, i.e. the

failure to take such parameters as field, mode, tenor and illocutionary act into

consideration when translating IEPVs into Arabic. The covertly erroneous errors of the

Arabic professional translators and Arabic translation students surveyed in this study

were of five types, namely:

5.3.6.2.1 Wrong Arabic collocation

In his study Arabic Collocations: Implications for Translation, Brashi (2005) elaborates

on semantic and distributional patterns of Arabic collocations. He indicates that Arabic

collocations can be categorized as unique collocations, metaphorical collocations,

idiomatic extensions of collocations, strong collocations, acceptable collocations, weak

collocations and unacceptable collocations. The first three categories, according to

Brashi (2005), represent patterns of meaning, whereas the last four categories represent

the range of frequency. What is concerned us here are the unacceptable Arabic
303

collocations, which are illustrated by Baker (1992) in the Arabic translation ‫ن‬XP7&ƒ+‫ ا‬,Y&‫آ‬

(literally: to break the law) provided to the English collocation break the law, as

opposed to the common Arabic collocation ‫ن‬XP7&ƒ+‫• ا‬+7&y (literally: to contradict the law)

(cf. Baker, 1992; Brashi, 2005).

As it has been pointed out (see 5.3.6.1.3 above), the subjects may have understood the

functional-pragmatic meanings of IEPVs in the source language (English) but stopped

short from delivering them correctly into the target language (Arabic).

Some IEPVs can be translated into Arabic by APVs, Arabic collocations, Arabic fixed

expressions, Arabic idioms, Arabic proverbs etc. The data collected revealed that most of

the very good answers (at the functional-pragmatic equivalents level) were based on

finding the right Arabic collocation (see Appendix 2).

Yet, translating IEPVs by IAPVs (one to one) was quite rare in the data collected. To give

illustrative examples, the IEPV to calm him down (item 52) was perfectly translated into

Arabic by one of the subjects as ‘&.E •&_y‫ أ‬, and the IEPV put away a big piece of meat

(item 82) was rightly translated into Arabic as p)E p}‫أ‬.

The main reason for the rarity of employing IAPVs when translating IEPVs is the fact

previously arrived at in Chapter Two (see 2.6 above), which is that APVs, unlike EPVs

which are typically used orally and colloquially by everyone in everyday contexts, are far

more formal and habitually used in a high register contexts due to their highly rhetorical

nature, which confines their use to educated people from a certain sector of society and

education.
304

Examples of translations of IEPVs by wrong Arabic collocations from the two groups are

presented below:

Item IEPVs Wrong Arabic collocations


73 to use up sugar ,'Y+‫ك ا‬2OFHM
98 dag out ‫ح‬,3+‫~ر وا‬O+‫ ا‬%8
33 He was shopping around “(.+‫[وس ا‬1 ‫ن‬7‫آ‬
5 cut her down to size 7‫[ه‬Z [.E 7O*^1
6 You're going to drive me around the bend! %PX.„ ,-…} S¢P‫إ‬
7 My eyes popped out 5N‫[ه‬+‫ ا‬%.}~y‫أ‬
14 has never stopped throwing her weight around 7O},{-H ‫اض‬,*FH‫ ا‬0E ‫• أ\[ًا‬wXF} 4+
74 to build up skills S}‫را‬7O/ ‫ّن‬X'} pFZ
85 passed away pOFP‫ا‬
87 but they will grow on you in the end ,/9‫ ا‬517OP %8 S\7IE‫ إ‬,-…} ‫ف‬XH
9 I'm fed up to my back teeth with work K3*+7\ %/[w ’3y‫ أ‬pFZ ‫رق‬7G
Table 5.24 Examples of wrong Arabic collocations (for suggested Arabic functional-pragmatic
equivalents see Appendix 2)

In the abovementioned examples, the IEPVs were translated by wrong Arabic

collocations, hence the mismatch was in their functional-pragmatic equivalents. In the

first example (item 73), the IEPV to use up sugar was translated into Arabic as ‫ك‬2OFHM

,'Y&+‫( ا‬literally: to consume sugar) which is wrong due to the fact that consuming sugar

could take place in a house or anywhere else other than by the human body. The correct

Arabic expression which can be used as a translation to this IEPV in such a context is

,'Y+‫ق ا‬,Q+ (to burn sugar).

In the second example (item 98), the IEPV dag out was erroneously translated into

Arabic as ‫ح‬,&3+‫&~ر وا‬O+‫ ا‬%&8 (literally: in prattle and joyfulness) these two words do not

collocate in the Arabic language. Arabic native speakers would more likely say ‫ح‬,&_+‫ا‬

‫ح‬,&3+‫( وا‬happiness and joyfulness). Therefore ‫ح‬,&31‫ح و‬,&_1 (to be happy and in joy) should

be used in this instance which is a derivative from the common Arabic collocation ‫ح‬,_+‫ا‬

‫ح‬,3+‫( وا‬happiness and joyfulness).


305

In the third example (item 33), the IEPV shopping around was inaccurately translated

into Arabic as “(&.+‫&[وس ا‬1 (literally: to step on pulse) Arabic native speakers do not say

“(&.+‫&[وس ا‬1 they instead say “(&.+‫ ا‬‰


ّ &I1 (literally: to feel pulse). The same argument is

applicable to the rest of the examples listed in the above Table.

5.3.6.2.2 Shift of register

It was manifestly explained in Chapter Two that IEPVs are typically employed in

different kinds of discourses (see 2.5.5 above). It was also explained that albeit they are

often used formally in a variety of English written texts, their common use is informal in

everyday spoken English. In order to achieve the functional-pragmatic equivalents of

IEPVs, translators have to consider such a formality/informality variation of register.

Translating informal IEPVs by formal Arabic expressions and vice versa is an

unacceptable register shift which may result in distortion of the intended meaning by

conveying the wrong message (cf. Hale, 1997& 2002).

In order to challenge the subjects' abilities in terms of appreciating the register variations

when rendering IEPVs into Arabic, an array of text types was used in the translation tests,

and short contextual information was provided between square brackets following each

item (see Appendix 1).

The collected data showed that in producing their Arabic translations to the listed IEPVs a

number of the subjects surveyed were fully aware of, and did consider the register

variations in terms of field, mode and tenor in which the listed IEPVs were used, whereas

others have not considered the text type when providing their Arabic translations. They,
306

instead, delivered translations in which the given register was either shifted from informal

to formal or vice versa.

Examples of IEPVs translated by incorrect register shift from the two groups are listed in

Table 5.25 below:

Item IEPVs Register shift


7 My eyes popped out x)‫ \ ذه‬xN‫ \ دُه‬xƒ*ُ`
68 I'm tied up with something urgent at the ‫ن‬¤‫ ا‬K„7E K3*\ S3O./ 7P‫أ‬
moment
100 we were slagged down 53OF\ 7.-/ُ‫ \ ر‬7.3„X‫ه‬
70 He flared up at me %)E 7ً(^G ‫ط‬7NFH‫ا‬
79 I've cleaned up my act ‘-)E x.‫ آ‬7ّ3E x*)w‫[ أ‬ƒ+
96 I'm really shopped out ‫ّق‬XYF+7\ 7ًE‫ ذر‬x
ُ ƒR [ƒ+ ،ً7ƒZ
Table 5.25 Examples of register shift (for suggested Arabic functional-pragmatic equivalents see
Appendix 2)

The Arabic equivalents given to the IEPV My eyes popped out in the first example

above (item 7) were x)‫ \ ذه‬xN‫ \ دُه‬xƒ*ُ` (I was shocked / surprised / astonished) wherein

the register shift from informal to formal is evident. The situational context in which

this IEPV was used was an informal everyday conversation between friends, while the

Arabic equivalents provided were highly formal Arabic expressions. Such a high

formality of the Arabic equivalents is manifested in the usage of the passive voice

which cannot be used in any informal everyday Arabic friendly conversation as it is less

common in Arabic than it is in English (El-Yassin, 1996; Wightwich & Gaafar, 1998).

The same argument is valid for the second and third examples above (items 68 and 100)

where highly formal Arabic expressions of a passive voice were given as equivalents to

the informal IEPVs.


307

Apart from employing the passive voice, making use of high formal Arabic collocations

and producing them as equivalents for informally used IEPVs is also deemed as an

unacceptable register shift. Such a drawback is manifested in items 70, 79 and 96

above, where 7ً (^&G ‫ط‬7N&FH‫( ا‬to burst with anger), 0&E x


ُ &*)w‫&[ أ‬ƒ+ (I have desisted from)

and 7ً&E‫ ذر‬x


ُ ƒ&R (I have no patience with) were give as equivalents to the everyday

colloquially used IEPVs He flared up, I've cleaned up and I'm shopped out respectively.

The main reason behind this kind of loss of register, as it has been argued in Chapter

Two, is the sharp distinction in the Arabic language between written and spoken

discourse (Baker, 1992), where two varieties (formal high written variety and informal

low spoken variety) are interchangeably used (Al-Qinai, 2000). Consequently, in

translating such everyday colloquial expressions as IEPVs into formal written Arabic,

Arabic translators may cause a shift of register from informal/colloquial to formal/high

written expressions. The best strategy suggested to Arabic translators to tackle such a

problem is reducing the gap between the informal/colloquial Arabic and the formal/high

written Arabic by steering a middle course between them (Al-Qinai, 2000). Hence,

Arabic translators are best advised to employ Modern Standard Arabic which lays half-

way between highly formal classic Arabic and highly informal colloquial Arabic.

Another reason for performing such a translational error by the Arabic subjects is the

fact that unlike the majority of English lexicographers who have included fundamental

information about register variations of EPVs in their dictionaries (cf. Cambridge

international Dictionary of Phrasal verbs (1997); Oxford Dictionary of Phrasal Verbs

(1993); Oxford Phrasal verbs Dictionary for Learners of English (2001); Longman

Dictionary of Phrasal Verbs (1983) among others), Arabic lexicographers disregard


308

such essential information when dealing with the phenomenon of EPVs and confine

themselves to providing either literal or intuitive idealised Arabic equivalents for them

(for illustrative examples see Chapter Two, sections 2.8 and 2.9 above).

5.3.6.2.3 Incorrect delivery of speech acts

I regarded "speech acts" as the second parameter of the model for the analysis and

translation of IEPVs into Arabic devised in Chapter Three. I also outlined that the

polysemic nature of IEPVs and their common use in everyday spoken English

necessitate translators to target their illocutionary/pragmatic meanings rather than

locutionary/semantic meanings if they are to achieve functional-pragmatic equivalents

(see Chapter Three, sections 3.4.4 and 3.5 above).

In the translation tests in this study, a number of Arabic professional translators and

Arabic translation students have failed to deliver the appropriate illocutionary meaning of

these verbs. Such a failure was as a result of misunderstanding the speech acts performed

in the utterances in which IEPVs were used. Since speech acts denote "[…] the total

situation in which the utterance is issued" (Austin, 1975, p. 52), such a misunderstanding

was due to not taking into account the "real-world situational factors" which are,

according to Mason (2001), "prime determinants of meaning and interpretation of

meaning" (p. 29). Along these lines Hatim and Mason (1997) assert that "[i]n any attempt

to examine the communicative nature of the translating task, a number of assumptions

will have to be made about texts, their users and the context in which they occur" (p. 14).

Therefore, pragmatically speaking, a translator "should consider the communicative force


309

of [any given] SL [Source Language] utterance which goes far beyond the propositional

meaning of ST [Source Text]" (Al-Qinai, 2000, p. 510).

Illustrative examples of this kind of translational pitfalls are presented in Table 5.26

below:

Item IEPVs Incorrect delivery of speech acts


76 We are not going into that again, S+~‫ آ‬‰-+‫ أ‬5-P7ˆ ‫ة‬,/ S+‫ ذ‬p+‫د إ‬X*P 0+
are we?
94 And go easy on the salt, please •)3+‫] ا‬R‫ و‬%8 7ًP‫و‬7OF/ 0‫ ًء آ‬7„‫ر‬
4 I could have bitten my tongue off %P7Y+ x
ُ 3^w
6 You're going to drive me around the •{*.3+‫ل ا‬XZ 5w7-Y+‫ ا‬p)E SoX} xP‫أ‬
bend!
91 I'm partied out ,y† ‫ن‬7'/ %8 7ً„‫ر‬7y x
ُ )_FZ‫ا‬
Table 5.26 Examples of incorrect delivery of speech acts (for suggested Arabic functional-
pragmatic equivalents see Appendix 2)

With regard to the first example in the Table above, most of the subjects involved in

this study have produced functionally/pragmatically inaccurate translations for such an

utterance (see Appendix 3). The Arabic translation of this utterance presented in the

Table above (S+~&‫ آ‬‰-&+‫ أ‬5&-P7ˆ ‫ة‬,&/ S&+‫ ذ‬p&+‫د إ‬X*P 0+) (literally: we will not return to that matter

again, are we?) is a striking example. In order to correctly translate such an utterance,

one has to embed it in its real-world context. This type of English structure (i.e. a

statement followed by a tag question) is habitually used in cross-examination in a court

context by prosecutors and solicitors to prompt defendants and/or witnesses with a

series of yes/no questions typically to discredit them (Hale, 2002 & 2004). The context

in which this utterance is embedded is "A cross-examination in court" (see Appendix1).

Pragmatically speaking, the utterance has a speech act of complaining, therefore, in

order to deliver its Arabic functional-pragmatic equivalent the illocutionary/pragmatic

meaning has to be taken into consideration rather than the locutionary/propositional

meaning such as the one in the above poor translation, and the performed speech act has
310

to be maintained. Hence, a possible functional-pragmatic equivalent for this utterance

could be:5&-P7ˆ ‫ة‬,&/ ‫ع‬X&RX3)+ ‫د‬X*.&H 7&.P‫ أ‬%&+ K&ƒ} M (literally: don’t say to me that we are going

back to the subject again) (see Appendix 2).

In the second example in the Table above (item 94), The Arabic translation produced

was also pragmatically incorrect. The situational context whereby this utterance took

place was an informal conversation at a restaurant between a waitress and a customer. It

is evident from the context that, due to the social distance between the two participants

of the conversation, a speech act of order was performed. The Arabic translation

presented above has failed to spell out such a speech act. In order to achieve the Arabic

functional-pragmatic equivalent of such an utterance, the speech act performed has to

be articulated (see Appendix 2 for suggested Arabic functional-pragmatic equivalents).

It is quite essential to indicate here that most of the examples of translational errors

categorized as (literal translation) (see 5.3.6.1.1 above) can well fit into the current type

of error, since providing the literal meanings of IEPVs not only delivers their distorted

denotative meanings but also mismatches the speech acts performed in the utterances in

which such PVs are embedded. Hale (1996) maintains that "[w]hen we translate an

utterance literally, we may be matching the locutionary act only, but not the other two

[illocutionary and perlocutionary acts]" (p. 63). The last three examples in the above

Table (items 4, 6 and 91) can be taken as illustrating instances whereby the subjects

produced literal translations of the utterances and failed not only in delivering their

locutionary act but also the speech acts of regret, complaint and exhaustion performed in

such utterances respectively.


311

Apart from the polysemic nature of IEPVs and their common informality, the major

reason behind making this type of translational errors, in my view, is that Arabic

equivalents given to IEPVs in bilingual English-Arabic dictionaries and in the available

textbooks on learning and translating EPVs are typically intuitive and not theoretically

based. Arabic lexicographers have totally overlooked such fundamental issues as

register variations, illocutionary acts, and functional-pragmatic equivalents. Arabic

translators, as a result, have no choice but to consult monolingual English-English

dictionaries to understand the pragmatic meaning of IEPVs and then to make up their

own Arabic equivalents, which can be functionally and pragmatically inaccurate.

5.3.6.2.4 Usage of paraphrasing

As I previously explained in Chapter Two (see 2.11 above), the translating by

paraphrasing strategy is a very common strategy and widely utilized by translators as one

of the attested strategies for dealing with the problem of non-equivalence (cf. Baker,

1992; Bataineh & Bataineh, 2002; Ghazala, 2003). However, this strategy has its

restrictive usage, especially in the case of translating IEPVs into Arabic, which have

Arabic equivalents that can deliver accurate functional-pragmatic equivalence.

Examples of this type of translational errors from the two groups are presented in Table

5.27 below:

Item IEPVs Paraphrasing


38 Where were you during the cave- pE‫[ا‬F1 ‫ء‬%o K‫ \[أ آ‬0-Z x.‫ آ‬01‫أ‬
in?
43 a cooling-off period K(w ,/97\ ,'_1 ‫‘ أو‬1‫’ رأ‬nN+‫ ا‬,ّ-Ž1 %'+ p{*ُ} ‫[ة‬/
%3H‫ ر‬K'N\ ‘\ ‫م‬7-ƒ+‫ا‬
312

60 she took up with Tim 4-} ]/ x„,y‫ و‬x8ّ,*}


10 she got so carried away she began ‫ء‬7'(+‫ ا‬%8 xW,nP‫ ا‬4ˆ x)*_P‫ وا‬7O_W‫ا‬XE %8 4ّ'QF} 4+
to cry
79 I've cleaned up my act %-E‫[ي أو و‬o‫ ر‬p+‫ت إ‬
ُ [E‫| و‬ŽN+‫ة ا‬7-Z 0E x ُ -)n} [ƒ+
91 I'm partied out %8 5(G,+‫[ي ا‬+ [*1 4+‫ة „[ًا و‬,-…‫ت آ‬2_Q+ x ُ „,y
[1Š3+‫ ا‬%8 ‫دة أو‬71Š+‫ا‬
Table 5.27 Examples of Paraphrasing (for suggested Arabic functional-pragmatic equivalents see
Appendix 2)

In their endeavour to translate the IEPVs at hand, the subjects glossed them in Arabic

instead of providing the appropriate Arabic functional-pragmatic equivalents for them.

In item 38 in the Table above, the noun derived from PV cave-in was rendered as K‫\[أ آ‬

pE‫&[ا‬F1 ‫ء‬%&o (literally: everything started collapsing) where there is a one-word Arabic

equivalent for it which is: ‫ر‬7-OPM‫( ا‬the collapsing).

Similarly, in item 43, the expression contains a noun derived from a PV a cooling-off

period which means a period to consider, was translated into Arabic as ,&ّ-Ž1 %'+ p{*ُ} ‫[ة‬/

%3&H‫ ر‬K'N\ ‘\ ‫م‬7-ƒ+‫ ا‬K(w ,/97\ ,'_1 ‫‘ أو‬1‫’ رأ‬nN+‫( ا‬literally: a period of time given to someone

in order to change his mind or think about the matter before carrying it out officially).

Such a long gloss is an obvious covertly erroneous error because it distorts the style of

the target language by employing a long sentence to express the meaning of one

expression. Suggested alternatives for such an expression are 5&)O/ \ ,&-'_} ‫ة‬,&F8 \ ‫ل‬7&O/‫ة إ‬,&F8

‫اء‬,N+‫ ا‬0E ‫[ول‬E (see Appendix 2).

Likewise, in item 60 the IEPV took up with was translated into Arabic as x&„,y‫ و‬x&8ّ,*}

]&/ (literally: she was introduced to and went out with) where there is a one-word

Arabic equivalent for it, which is:‫ت‬,&o7E (took him as a boyfriend). The same argument

is valid for the rest of the examples presented in Table 5.27 above.
313

5.3.6.2.5 Usage of Arabic colloquial and regional dialects

As it has been argued above (see shift of register 5.3.6.2.2 above), given the sharp

distinction between the two varieties of the Arabic language (formal high classical

written variety and informal low spoken variety), Arabic translators need to seek the

middle ground Arabic style by utilizing Modern Standard Arabic when tackling the

phenomenon of IEPVs. Therefore, producing highly classical Arabic expressions as

functional-pragmatic equivalents for IEPVs is unacceptable register shift, as well as

producing highly informal colloquial Arabic expressions or regional dialects as

functional-pragmatic equivalents in writing is unacceptable as well, except for in

limited text-types, for example newspaper commentary texts.

Despite the following note made at the beginning of the translation tests distributed to the

subjects: "Since the study is about translation only, please confine your answers to

Modern Standard Arabic and avoid colloquial equivalents for the idiomatic English

phrasal verbs itemized in the translation tests" (see Appendix 1), the data collected

revealed that a number of the subjects resorted to informal colloquial Arabic and regional

dialects when dealing with the listed IEPVs. Illustrative examples of this type of

translational error from the two groups are presented in Table 5.28 below:

Item IEPVs Arabic colloquial and regional


dialects
19 brought the house down 7-P[+‫ ا‬,ّY‫آ‬
20 he'll come down on you like a ton of SH‫ رأ‬,ّY'-H
bricks.
24 Bill must have got up on the wrong side ‫م‬X-+‫ر ا‬X'*/ K-\ ‫اج‬Š/ ‫ن‬
¢ ‫([و أ‬1
of the bed today
25 to have it out with him ‘*/ ‫ر‬X/9‫ ا‬%ّ_`9
30 You can't skirt round the matter ‫ع‬XRX3+‫ ا‬5_)_+ S.'31M
31 Did you notice the way he bossed his wife ,OŸ p)E 7ً (‫ن راآ‬7‫• آ‬-‫ آ‬x€ZM K‫ه‬
around ‘F„‫زو‬
314

66 to take it out on me %
¢ 8 Sƒ)y ˜_}
84 to take in old ladies ŠL7I*+‫ ا‬p)E SQ^1
92 Funk up 5RX3+‫ ا‬p)E S-)y
100 we were slagged down 7P‫ء‬ŠO}‫إ‬
77 and make up ‫ا‬XR,F}M ‫ذا‬73+
91 I'm partied out ‫ي‬,W7y x1,H
46 Handout 5/Š)/
78 Hard to figure out ‫روا‬ŠQُ1 ‫`*| أن‬
Table 5.28 Examples of usage of Arabic colloquial and regional dialects (for suggested Arabic
functional-pragmatic equivalents see Appendix 2)

In items 19 and 20, the high informal Arabic expressions 7-P[&+‫ ا‬,ّY&‫( آ‬literally: broke the

world into pieces) and S&H‫ رأ‬,ّY&'-H (literally: he'll break your head into pieces) were

given as Arabic equivalents to the IEPVs brought the house down and he'll come down

on you like a ton of bricks respectively. Despite the fact that such translations have

delivered the denotative and the communicative meanings of the English expressions at

hand, and they may well be used by interpreters who typically rely on the spoken mode,

they cannot, however be given as written Arabic translations due to the readers'

expectations.

The same argument is applicable to most of the examples presented in the above Table

(items 24, 25, 30, 31, 66, 84, 100 and 77).

In addition, in items 91, 46 and 78, the Arabic regional dialect expressions ‫ي‬,W7&y x1,&H

(I have had enough – Iraqi colloquial), 5/Š)/ (A printed material – Moroccan colloquial)

and ‫روا‬Š&Qُ1 ‫( `&*| أن‬hard to be guessed - Iraqi colloquial) were given as equivalents to

the IEPVs I'm partied out, Handout and Hard to figure out in that order. Again, even

though such Arabic translations have delivered the denotative and the communicative

meanings of the English expressions concerned, and they may well be used by

interpreters in the spoken mode, they cannot be provided as written Arabic translations
315

due to the fact that such Arabic expressions are both not lexicalized and not current in

that putting them in writing would also be considered odd and unacceptable by the

Arabic readership. Confining their usage to one Arabic country also makes them

unintelligible to Arabic native speakers of other Arabic countries.

5.3.7 Subjects' performance in light of the seventh research question

In the preceding sections I overviewed the results of the translation tests and discussed

the translation errors made by the two groups involved in the study. In this section I will

be considering the issue of translation communicative competence in relation to the

following last research question:

Is there any difference as to translation communicative competence between the

two groups of Arabic subjects: The professional Arabic translators and Arabic

translation students? If so, what are the major areas in which their competence

varies?

A statement will be made here to account for the level of competence of each group based

on the theory of communicative competence (knowledge and ability) discussed in

Chapter Three (see 3.4.5 above). The aim of such a statement is to compare the

translation communicative competence of the two groups judged by their performance

(their answers) in an attempt to pinpoint the areas in which they have a shortage of

knowledge and/or ability. This statement will be of a great benefit in suggesting

recommendations in the forthcoming Chapter.


316

In the model I devised for analysis and translation of IEPVs into Arabic (see 3.5 above),

I regarded translation communicative competence as the last component and I set it up

as a final stage to compare the inputs of the subjects of the study. Translation

communicative competence refers to theoretical and practical knowledge of translators.

As has been explained in Chapter Three (see 3.5 above), all the parameters of the model

(register, speech acts, functional-pragmatic equivalents and covertly/overtly erroneous

errors) have to be considered when evaluating the translation communicative

competence of the subjects, in view of the fact that all of such parameters relate to the

translators' knowledge and ability. Yet, paying no attention to any parameter would

mean a given subject has no, or a lack of, theoretical and/or practical knowledge.

It is quite evident from the above overview and discussion of the results of the

translation tests, that despite the differences between the two groups of the subjects

involved in this study, the overall performance of the subjects in general was poor,

which demonstrates the fact that the IEPVs listed in the translation tests posed

substantial difficulty to both groups of the subjects. This confirms the point argued in

Chapter Two (see 2.11 above), where I pointed out that due to the fact that the Arabic

language has few syntactic, semantic and pragmatic correspondences with the English

language in terms of the phenomenon of idiomatic PVs, the task of translating IEPVs

into Arabic is far more challenging and subject to a great deal of difficulties than

translating them into other languages.

Yet, the two groups vary in their level of translation communicative competence. Such

a variation is manifested in the fact that the overall performance of the Arabic

professional translators' group is better than that of the Arabic translation students'
317

group with a difference of 10% between the two groups in the average of the correct

answers, i.e. both the very good and the satisfactory answers (see Table 5.8 above) and

11.35% in the average of achieving functional-pragmatic equivalents that is, the very

good answers only (see Table 5.14 above). With this fact in mind, the differences

between the two groups in respect to achieving the functional-pragmatic equivalents

(which is the main concern of this study) are not distinct, which means that the years of

experience of the Arabic professional translators did not make a major difference when

translating IEPVs into Arabic. There were some items in which the performance of the

Arabic translation students' group was even better than that of the Arabic professional

translators' group. There were also other items in which the performance of both groups

was equal (see Table 5.16 above).

Bearing that in mind, all the Arabic professional translators who participated in this

study are academically qualified in the field of translation and/or interpreting. In

addition, they all have a good translation experience ranging from four to forty nine

years (see 4.3.1 above). They presumably have the required theoretical knowledge and

practical experience. On the other hand, the Arabic translation students involved in this

study have not, as yet, acquired such theoretical knowledge and practical experience

(see 4.3.2 above).

The variation in the level of translation communicative competence between the two

groups is also manifested in the fact that the Arabic professional translators answered

the vast majority of items of the translation tests, in that the total average of their

unanswered questions was very low (0.58%), while the Arabic translation students left

69 (6%) items unanswered, with a difference of -5.42% between the two groups (see
318

Table 5.15 above). Such a result demonstrates the fact that the Arabic professional

translators were more competent and have more sense of encouragement to take

initiative and risk-taking than the Arabic translation students, who opted for skipping

the items that were problematic to them.

Another aspect of variation in the level of translation communicative competence

between the two groups is observable from the fact that the Arabic professional

translators made less overtly erroneous errors than the Arabic translation students with a

difference of -7% between the two groups (see Tables 5.11 and 5.13 above). Such a

difference in performance indicates that the Arabic professional translators have been

more competent to the extent that they have avoided (more than the Arabic translation

students) being so close to the literal meanings of the listed idiomatic IEPVs, kept away

from being confused with their polysemic nature, and have not breached the Arabic

language system when providing Arabic equivalents.

The Arabic professional translators, however, made more covertly erroneous errors than

the Arabic translation students with a difference of 2% between the two groups (see

Tables 5.11 and 5.13above). As has been argued before (see 5.3.1.1 above), such a

minor difference does not mean that the Arabic professional translators are less

competent than the Arabic translation students. On the contrary, taking the major

difference in the total average of the unanswered items into account confirms the

abovementioned claim that the Arabic translation students were in fact more prudent,

while the Arabic professional translators were greatre risk-takers.


319

In addition, the results showed that there were more variations within the answers of

Arabic professional translators' group than that of the Arabic translation students' group

with a difference of 0.031602 between the two groups in the standard deviations (see

Table 5.9 above). Such a difference in the standard deviations confirms, again, the

abovementioned claim that the Arabic professional translators were freer and have more

sense of risk-taking than the Arabic translation students. The years of practical

experience and the higher level of academic qualifications the Arabic professional

translators have seem to be the factors for such ability.

Moreover, ranking the categories of IEPVs according to the average of achieving

functional-pragmatic equivalents revealed that in spite of the differences in the averages

of achieving the Arabic functional-pragmatic equivalents between the two groups, both

groups have experienced the same level of difficulties, as the results showed that the (p)

category represented the most difficult one for both groups with the difference of 7.17%

in favour of the professionals. The (c) category constituted the second hardest one for

both groups with the difference of 15%. The (d) category was the third hardest with a

difference of 11.58%. And the easiest category was the (s) category with the difference

of 10% between the two groups (see Table 5.14 above). It is quite obvious that the

Arabic professional translators outperformed the Arabic translation students in all

categories (see Table 5.14 above).

Even though the (d) and (s) categories, as a whole, constituted the easiest categories for

both groups (see Table 5.14 above), this does not mean that the derivational and

grammatical forms of IEPVs posed no problems to the subjects surveyed. The low
320

averages of the functional-pragmatic equivalents achieved indicate that the subjects

encountered difficulties when tackling the items comprising such forms.

The difference in the average of the correct answers (i.e. both the very good and the

satisfactory answers) of the categories that are easier (s and d) was rather large, whereas

there was no such difference in the categories considered difficult (p and c). This entails

that training and experience gained by professional translators enhanced their

performance only in the categories which are more teachable and learnable (s and d

categories), but such training and experience did not improve their performance in the

categories where the sense of stylistic competence is required (p and c categories) (see

Table 5.8 above).

Translating IEPVs, like translating any other kinds of idiomatic expressions, requires a

high level of competence not only in the source language (English) but also in the target

language (Arabic). In her attempt to elaborate on the required types of competence,

Nord (1992) lists them as follows:

[…] competence of text reception and analysis, research competence, transfer


competence, competence of text production, competence of translation quality
assessment, and, of course, linguistic and cultural competence both on the
source and the target side, which is the main prerequisite of translation activity (p.
47 as cited in Campbell, 1998, p. 5).

Having said that, in spite of the fact that both groups of the subjects involved in this

study have encountered the same problems as manifested in the types of translational

errors they made, the results of the present study reveal that the inadequate competence

of some subjects in correctly understanding the listed IEPVs caused them to make

"source text errors". Good examples of such kind of errors are found in the
321

abovementioned examples of literal translation, mistranslation, and incorrect delivery of

speech acts.

Other subjects did understand the listed IEPVs but they failed to put them in correct

Arabic, resulting in "target text errors". Good examples of this kind of error are found in

the abovementioned examples of shift of register, breaches to the Arabic language

system, wrong Arabic collocation, paraphrasing and employing Arabic colloquial and

regional dialects.

Before bringing this section to the end it seems appropriate here to compare the

performance of professionals among themselves and the performance of students

among themselves to see whether the years of experience, in the case of professionals,

and the stages of study, in the case of students, can make any difference as to translation

communicative competence of IEPVs. Table 5.29 below sketches the performance and

years of experience of the Arabic professional translators' group:

Subject Years Correct Incorrect Unanswered


Code Of Very Satisfactory Overtly Covertly
Number Experience good erroneous erroneous
error error
P1 49 Years 50% 14% 26% 10% 0%
P2 4 Years 27% 19% 46% 8% 0%
P3 4 Years 32% 12% 48% 8% 0%
P4 15 Years 43% 13% 36% 8% 0%
P5 7 Years 36% 15% 36% 13% 0%
P6 9 Years 38% 17% 37% 7% 1%
P7 6 Years 44% 15% 33% 7% 1%
P8 4 Years 52% 14% 26% 8% 0%
P9 7 Years 40% 18% 36% 6% 0%
P10 10 Years 49% 14% 32% 4% 1%
P11 6 Years 63% 14% 18% 5% 0%
P12 7 Years 29% 25% 37% 6% 3%
Table 5.29 Performance and years of experience of the Arabic professional translators' group
322

Interestingly, figures in the above Table reveal that the years of experience do not have

much effect on the performance of the professionals involved in this study. To take only

two instances, in achieving the functional-pragmatic equivalents of the listed IEPVs,

professionals with four and six years of experience were the best among their

colleagues. They performed a great deal better even than the professionals of fifteen and

forty nine years of experience. Turning our attention to the errors made by the subjects

of this group, the professional with fifteen years of experience made more overtly

erroneous errors than the professional with six years of experience. And the

professional with forty-nine years of experience made more covertly erroneous errors

than the professional with four years of experience. It seems that the reason behind such

differences in the performance, regardless the years of experience, is the lack of any

kind of theoretical fundamentals and systematic approach that can help these

professional translators to deal with this type of idiomatic constructions.

Table 5.30 below sketches the performance and the stages of study of the Arabic

translation students' group:

Subject Subject Correct Incorrect Unanswered


Code Stage Very Satisfactory Overtly Covertly
Number of Study good erroneous erroneous
error error
S1 Year 9% 14% 73% 4% 0%
S2 one 18% 11% 35% 2% 34%
S3 Year 41% 24% 32% 2% 1%
S4 two 50% 14% 30% 6% 0%
S5 Year 39% 19% 40% 2% 0%
S6 Three 23% 20% 31% 7% 19%
S7 Graduate 27% 21% 43% 6% 3%
S8 Diploma 24% 20% 52% 4% 0%
S9 MA 20% 15% 49% 4% 12%
S10 46% 15% 34% 5% 0%
S11 PhD 27% 14% 47% 12% 0%
S12 41% 23% 33% 3% 0%
Table 5.30 Performance and stages of study of the Arabic translation students' group
323

By the same token, figures in the above Table show that the stages of study do not have

much impact on the performance of the students participated in this study. To take only

two instances, in achieving the functional-pragmatic equivalents of the listed IEPVs,

year two students were the best among their colleagues. They performed much better

even than the postgraduate students. As for the errors made by the subjects of this

group, MA Students made much more overtly erroneous errors than year two students,

and a PhD student made more covertly erroneous errors than the year one student. Such

results tell us that there is an acute problem at all levels of study. There are two main

reasons behind such a problem. First, the abovementioned lack of theoretical

fundamentals and systematic approach that can help in dealing with this type of

idiomatic construction. Second, and more important, the inadequacy of the teaching

process, i.e. the lack of any teaching methodology for teachers that can provide students

with solid background knowledge to enable them to understand the syntactic, semantic

and pragmatic properties of this type of problematic verbs, and in turn to translate them

correctly into Arabic. Moreover, the question of recruiting students in the programs of

translating needs to be addressed here. Nida (2001) makes the point that "In fact,

relatively few students entering programs in translating have the necessary language

competence to begin translating. This is not the students' fault, but the fault of the

educational system" (p. 2).

All in all, the translation communicative competence of both groups, according to the

findings of the present study, was rather inadequate. Their knowledge of syntactic,

semantic and pragmatic characteristics of IEPVs was deficient, and their ability in

understanding them and translating them into Arabic was limited. However, the Arabic

translation students seem to have an acute need to build up their knowledge of the

syntactic, semantic and pragmatic properties of IEPVs in order to increase their own skills
324

and ability in appreciating the metaphoric and polysemic nature of IEPVs, and in seeking

communicative meanings to convey them, by taking into consideration the "real-word

situational factors" in which they are employed in order to achieve their Arabic

functional-pragmatic equivalents.

5.4 Conclusion

In this Chapter the results of the translation tests given to the subjects participated in

this study were carefully overviewed and comprehensively discussed.

A general overview of results of the translation tests of the two groups involved in this

study was first provided to present the average of their correct answers (i.e. both the very

good and the satisfactory answers). In the second section of this Chapter, the results of the

translation tests of the two groups were analysed and discussed in light of the seven

research questions, with more focus on achieving the Arabic functional-pragmatic

equivalents (very good answers) by the subjects.

A number of crucial findings were arrived at. Such findings highlighted the types of

difficulties Arabic professional translators and Arabic translation students encounter

when translating IEPVs into Arabic. The next Chapter will recapitulate these findings and

propose some recommendations for both groups as well as for lexicographers and

pedagogues.
325

CHAPTER SIX
CONCLUSION AND RECOMMENDATIONS

6.1 Conclusion

6.1.1 Summary of the study

The major objective of this study was to investigate the difficulties encountered by Arabic

professional translators and Arabic translation students when translating IEPVs into

Arabic. The other objective was to suggest a number of recommendations for Arabic

professional translators, Arabic lexicographers and Arabic pedagogues based on the

variety of findings arrived at from the empirical research conducted in the study.

The main hypothesis of the study was that there is a wide range of difficulties posed to

Arabic professional translators and Arabic translation students when translating IEPVs

into Arabic.

There were two sets of research questions put forward in this study. The first set was

presented to address the theoretical part of the thesis. The second set of research

questions, however, was presented to analyse the data collected from pragmatic, semantic

and syntactic standpoints.

This study started with an introductory Chapter where a historical background of the

notion of EPVs was provided followed by the justification, the purpose, the hypothesis,

the layout and the limitations of the study.


326

Chapter Two was dedicated to address the first set of the research questions. The Chapter

demonstrated that IEPV is a combination of two or three items (verb + adverb, verb +

preposition, and verb + adverb + preposition) which functions as a single unit of meaning

in that its meaning cannot be figured out from the total sum of the meanings of its

separate components. The syntactic properties of such combinations include such features

as: the basic verbs which can be phrasalised are mostly monosyllabic, and the particles

that typically combine with them are either adverbial or prepositional. Many of such

particles are of dual functions, i.e., they can be both adverbs and prepositions due to the

fact that they are identical in form but different in function. The abovementioned three

types of PV can be transitive or intransitive. The former may be separable or non-

separable depending upon the particle's position and movement.

The semantic properties comprise the following features: they are of three types, namely,

literal, semi-idiomatic, and idiomatic PVs, in that their meanings range from the most

literal to the most idiomatic. The idiomatic type can be substituted by one word synonym.

Furthermore, the particle plays an important role in modifying the meaning of the verb it

combines with, in the sense that they fuse together and sacrifice their basic meanings to

produce a new one semantic unit. Finally, IEPVs have the characteristic of polysemy, in

that any given idiomatic PV may occur in as many as ten, or more, different meanings

according to the contexts in which it is used.

Unlike the English language, which imparts a clear prominence to the phenomenon of

idiomatic PVs as one of the most essential features of English grammar and vocabulary,

the Arabic language does not classify similar structures. Arabic has one structure which

is, more or less, similar to IEPVs, that is, verb + preposition structure. Following a
327

number of researchers (Lentzner, 1977; Najiib, 2001; Kharama & Hajjaj, 1989; Alkhuli,

1999; and Bataineh & Bataineh, 2002), I considered the idiomatic type of such

constructions as idiomatic Arabic PVs due to the fact that the basic components of the

structure sacrifice their original meanings and merge together to form a single unit of new

meaning which has nothing to do with the meanings of the individual parts. However, not

all Arabic verb + preposition constructions are idiomatic, nor do all of them fit the

syntactic and semantic criteria of IEPVs. There are, in fact, several dissimilarities

between them in terms of syntactic and semantic properties, functionality, register and

pragmatic usage (see 2.6 above).

The review also addressed the issue of derivation of EPVs and the differences in

derivational mechanisms between the English and Arabic languages. Although both

languages have the characteristic of productivity as a result of word-formation systems,

IEPVs have the ability to be far more productive than their Arabic counterparts, which

can only be productive when used as integral parts of metaphorical and fixed expressions.

This is mainly because IEPVs are dominantly colloquial, hence are more easily produced

by English native speakers.

A comparison between the monolingual English-English dictionaries of PVs and bilingual

English-Arabic dictionaries of PVs showed that the bilingual English-Arabic dictionaries

of PVs are inferior in quality and quantity than the monolingual English-English

dictionaries of PVs. While there are more than ten monolingual English-English

dictionaries devoted to the phenomenon of EPVs, there are, to the best of my knowledge

and research, no more than two bilingual English-Arabic dictionaries dedicated to such a

phenomenon.
328

In addition, unlike the monolingual English-English dictionaries of PVs which are, to a

great extent, comprehensive in covering enormous amount of EPVs and providing

fundamental information about the syntactic, semantic and, collocational properties, as

well as register, derivation, and productivity of each listed PV, the bilingual English-

Arabic dictionaries of PVs paid no attention to most of such crucial information, leaving

Arabic translators and Arab learners of English with no choice but to find the meaning of

idiomatic PVs in English then to improvise on their Arabic equivalents, which may or

may not be accurate (see 2.9 above).

As far as teaching IEPVs to non-native speakers is concerned, a number of methods

proposed for teaching them have been investigated. Pedagogues varied in what to teach as

PVs and how to teach them. Some pedagogues believe that all types of EPVs (i.e.,

literal/semantically transparent, and idiomatic/semantically opaque) have to be taught

since the former constitutes the basic by which the latter can be understood. Others, on

the other hand, consider that only the idiomatic/semantically opaque type of PVs has to

be taught, as it is the prototypical one.

Likewise, they disagree upon how to teach EPVs to non-native speakers of English. Some

of them believe that they have to be taught by knowing the patterns underlying them, in

order to identify the system and logic by which they work. Others, however, consider

EPVs as random combinations of verbs and particles, and they have to be memorized by

heart.

By the same token, pedagogues differ in the question of how to sequence EPVs in

textbooks. Some are quite convinced that they must be dealt with by particles, while some
329

others prefer to handle them by verbs. Others recommend qualitative analysis by

conducting frequency counts of EPVs to determine the most common and needed ones in

order to be taught, and, in turn, to avoid depending upon pedagogues' intuitions in

designing pedagogical tools and materials.

Moreover, most pedagogues agree upon the issue of teaching EPVs in context rather than

as unrelated elements, due to the fact that such a manner enhances and facilitates their

learnbility.

Moreover, Chapter Two dealt with the issue of randomness of EPVs. There is very strong

evidence that EPVs are not random combinations. It was clearly demonstrated that there

is, in fact, a pattern underlying each one of them (see 2.10 above).

The comparison between the methods of teaching IEPVs utilized by Arab pedagogues

with those used by non-Arab pedagogues revealed that the methods of teaching IEPVs to

Arab learners of English is, to a great extent, inadequate. The two main reasons behind

such an inadequacy are the lack of rich pedagogical materials and the utilization of old-

fashioned teaching methods.

Furthermore, the question of how the phenomenon of IEPVs has been dealt with in

translation studies was, also, explored in Chapter Two. It was quite evident that unlike the

cases of such languages as Spanish and Russian, which have many correspondences

between them and the English language in terms of PVs (see 2.11.1 above), the Arabic

language which has few syntactic and semantic correspondences with English, hence the

task of translating IEPVs is far more challenging and subject to a great deal of difficulties.
330

Such difficulties prompted some Arab researchers to tentatively suggest the employment

of such strategies as evasion, omitting and paraphrasing IEPVs when translating them

into Arabic. According to my study, these strategies are proven to be haphazard and

cannot be replicated because they have no place in assessing, perceiving and producing

IEPVs by Arabic learners of English, Arabic translators and Arabic interpreters.

Chapter Two highlighted as well the fact that the question of translating English idioms in

general and IEPVs in particular into Arabic has not been adequately attended to in

translation studies. Scholarly efforts are needed to account for the challenges and

difficulties such expressions pose to Arabic translators. The literature review showed that

more practical approaches, as well, need to be established, tested and applied. That is

precisely what this study has accomplished by focusing on IEPVs, the most important

type of English idiomatic expressions. Hopefully, my efforts can contribute in narrowing

the existing gap in this particular area of translation studies.

A link between the theoretical part of the study (the literature review) and the practical

part of it (the experimental research) was established in Chapter Three in which I set up a

theoretical foundation for my experiments by reviewing the key and influential

approaches of contrastive analysis. I gauged the appropriateness/inappropriateness of

such approaches to my topic in an attempt to devise a practical and scientific model for

the analysis and translation of IEPVs into Arabic. Four most appropriate parameters were

selected to formulate my model, being: House's functional-pragmatic equivalence, which

is, as sketched by her (2001, p. 247), "the preservation of 'meaning' across two different

languages and cultures"; context of situation/register (field, mode and tenor), Austin's

(1962) speech acts (locutionary, illocutionary and perlocutionary acts), House's (1977)
331

overtly/covertly erroneous errors and Hymes's (1971) translation communicative

competence.

The research methodology employed in this study to conduct the experimental research

was outlined in Chapter Four where the research design, participants, data collection

and data analysis were elaborated on.

The results of the translation tests carried out by the two groups involved in the present

study were presented, analysed, interpreted and discussed in Chapter Five, where a

general overview of the results of the translation tests was given first followed by ample

analyses and discussion of the subjects' performance in light of the second set of the

research questions.

The results analysed and discussed supported the hypothesis outlined in the introductory

Chapter of this thesis and demonstrated that there were wide ranges of difficulties

encountered by Arabic professional translators and Arabic translation students when

translating IEPVs into Arabic.

6.1.2 Summary of the findings

The findings arrived at in this study are summarized in what follows:

1. Despite the differences between the two groups of subjects, both of them have

encountered a great deal of difficulties when translating IEPVs into Arabic. This is

quite evident from the overall performance of the subjects, which was poor, and from

the achievement of functional-pragmatic equivalents of the listed IEPVs which was so


332

hard that both groups have scored low percentages of (very good) answers. This

confirms the fact that since the Arabic language has a small number of syntactic,

semantic and pragmatic correspondences with English in terms of the phenomenon of

idiomatic PVs, the task of translating such verbs into Arabic is far more challenging

and more difficult than the task of translating them into other languages such as

Spanish and Russian, where many correspondences between them and the English

language exist.

2. The newly coined IEPVs, (p) category, where there are no readily Arabic equivalents

for them, constituted the most difficult one for both groups. I attributed that to the two

previously confirmed facts, being:

i) The productive nature of IEPVs prevents lexicographers from keeping up with

these and listing them in dictionaries. Consequently, there have been many gaps

in the coverage of IEPVs, even in specialized dictionaries. Such gaps resulted in

an absence of a number of newly coined PVs. A translator, in this case, is left

with no choice but to intuitively work them out one by one in order to produce

their Arabic functional-pragmatic equivalents, which may or may not be correct.

ii) The lack of effective teaching methods and materials which may help make

Arabic translators and translation students capable enough to deal with such

newly coined IEPVs. Unlike other pedagogues who suggested valuable methods

of teaching IEPVs based on the fact that these verbs are not mere random

combinations of verbs and particles and there are patterns underlying them,

Arab pedagogues nonetheless seem resigned to the fact that EPVs are random

combinations and for learners to master them they have to memorize them by
333

heart. Accordingly, there are no reliable Arabic pedagogical materials that can

help overcome the problem of the newly coined IEPVs.

3. The complex idioms, (c) category, constituted the second hardest category for both

groups. I attributed that to the double complexity nature of such idioms, which

requires a high level of competence on the part of translators, in both source and

target languages.

4. The derivative IEPVs, (d) category, were less difficult than categories (p) and (c) for

both groups. However, they caused a great deal of difficulties to the subjects and this

is particularly evident from the low averages of the functional-pragmatic equivalents

achieved. I attributed such difficulties to the differences in derivational mechanisms

between English and Arabic languages that make the derivational forms of IEPVs

sound abnormal and odd for Arabic subjects.

5. The items containing IEPVs that are derived from nouns, (dpn) subcategory,

represented the most difficult subcategory of the (d) category for both groups.

6. The syntactic forms of IEPVs, (s) category, constituted the easiest one for both

groups, even though the subjects encountered a number of difficulties when

attempting the items containing such forms. Such difficulties were quite evident from

the low averages of the functional-pragmatic equivalents achieved.

7. The (svp) subcategory represented the most difficult subcategory of the (s) category

for both groups.


334

8. The (svap) subcategory constituted the second hardest subcategory for both groups.

9. The easiest subcategory of the (s) category was the (sva) subcategory.

10. There were two types of translational errors made by Arabic professional translators

and Arabic translation students: the overtly erroneous errors and covertly erroneous

errors. The results of both groups showed that the overtly erroneous errors

outnumbered the covertly erroneous errors.

11. The Arabic professional translators made less overtly erroneous errors than the Arabic

translation students.

12. The overtly erroneous errors were mainly attributed to the idiomaticity of EPVs. This

played a significant role in hindering the subjects from obtaining the denotative

meanings of the listed IEPVs.

13. The overtly erroneous errors made by the subjects were of four types, namely: literal

translation, mistranslating, reducing the idioms to sense and breaching of the Arabic

language system.

14. Literal translation was manifested in the tendency of a number of the subjects to

perceive the listed IEPVs literally and deal with the two parts of the combination

separately rather than appreciating their metaphorical/idiomatic meaning and looking

at each combination as one semantic unit. I attributed committing such type of errors

to the following four reasons:


335

i) Interference of the subjects' L1 (Arabic) with their L2 (English). That is due to

the fact that the syntactic structure of IEPVs is quite odd for the Arabic

subjects. Unlike English, Arabic does not allow proper verbs to collocate with

adverbs. The only type of PVs in Arabic is that of verb + preposition structure

(see 2.6 above). Consequently, the subjects perceived the second parts of the

listed PVs literally and either translated them as prepositions or ignored them.

ii) The deceptive appearance of EPVs in general may catch the attention of

translators to provide prompt literal translations without carefully taking into

account the idiomaticity of this kind of verbs.

iii) Due to the productive nature of IEPVs which makes them hard to be captured

and listed in dictionaries, great deals of IEPVs are not covered by general and

even some specialized dictionaries.

iv) The decotextualization phenomenon in dealing with the IEPVs not only in

bilingual English-Arabic dictionaries, but also in textbooks devoted to

teaching EPVs to Arab students. Nevertheless, bilingual English-Arabic

dictionaries provide out of context, decontextualized and idealised equivalents

for IEPVs.

15. Mistranslating was apparent from the failure of a number of the subjects to

appreciate the polysemic nature of the listed IEPVs. I attributed producing such

mistaken translations to three reasons, they are:

i) The variety of shades of meanings given to each PV due to the polysemic nature

of the IEPVs which makes it hard for the subjects to choose the appropriate

meaning.
336

ii) The fact that the combinations of IEPVs are quite confusing makes the task of

choosing the appropriate meaning more difficult. That is, one proper verb can

collocate with a number of particles to form a range of PVs with many different

meanings, and one particle may co-occur with a number of proper verbs to form

a variety of PVs of diverse meanings.

iii) The inadequate treatment of the phenomenon of PVs in general and specialized

dictionaries. Lexicographers skip a large amount of PVs and provide insufficient

definitions for the listed ones.

16. Reducing the idioms to sense was noticeable when a number of the subjects seemed

to understand the functional meaning of the listed IEPVs in English but did not

deliver it correctly into Arabic. Instead, they produced a translation in which the

denotative meaning was achieved but at the expense of the idiomaticity of the given

EPVs. Such type of translational pitfalls was manifested in the subjects' answers of

the items of the complex idioms, (c) category, which have to be translated by Arabic

equivalents of an idiomatic nature i.e. employing Arabic PVs, Arabic proverbs,

Arabic fixed expressions etc. given the double complexity nature of such complex

idioms.

17. Breaching of the Arabic language system was manifested in the tendency of a

number of subjects to produce grammatically and/or morphologically incorrect

translations to the listed IEPVs. This is mainly owing to the lack of the required

linguistic competence.
337

18. The Arabic professional translators made more covertly erroneous errors than the

Arabic translation students.

19. The covertly erroneous errors made by the Arabic subjects were of five types,

namely: wrong Arabic collocation, shift of register, incorrect delivery of speech

acts, usage of paraphrasing and usage of Arabic colloquial and regional dialects.

20. Wrong Arabic collocation error was evident when a number of subjects seemed to

understand the functional-pragmatic meanings of the listed IEPVs in English but

stopped short from delivering them correctly into Arabic.

21. Some of the Arabic subjects surveyed in this study rightly made use of such

pragmatic/stylistic structures as Arabic collocations, Arabic fixed expressions,

Arabic idioms and Arabic proverbs when translating the listed IEPVs into Arabic,

while some others failed to do so.

22. The data collected revealed that most of the very good answers (functional-

pragmatic equivalents) were based on finding the right Arabic collocation (see

Appendix 2).

23. Translating IEPVs by its Arabic counterpart (IAPVs) was quite a rare phenomenon

in the data collected. I attributed such a rarity to the fact that Arabic PVs are more

formal than their English counterparts and, due to their highly rhetorical nature,

typically used in high register contexts by educated natives of a particular sector of

society and education.


338

24. Shift of register was apparent in the failure of a number of subjects to consider

formality/informality variations of the listed IEPVs. They instead translated

informal IEPVs by formal classical Arabic expressions or employing the passive

voice and vice versa, which resulted in distortion of the intended meanings by

conveying the wrong messages. Those types of errors were attributed to the

following three reasons:

i) When providing their Arabic translations, the Arabic subjects did not consider

the text types in which the listed IEPVs were embedded.

ii) The sharp distinction that exists in the Arabic language between written and

spoken discourse, where two varieties (formal high written variety and

informal low spoken variety) are interchangeably used.

iii) Arabic lexicographers, unlike the majority of the English lexicographers, did

not include enough information about such an essential issue as register

variations of EPVs in their bilingual English-Arabic dictionaries. They,

instead, confined themselves to providing either literal or intuitive idealised

and decontextualized Arabic equivalents for them.

25. Incorrect delivery of speech acts was manifested in the failure of a number of the

subjects to understand the speech acts performed in the utterances in which IEPVs

were used. I attributed committing such a type of errors to the following four reasons:

i) Not taking what is termed by Mason (2001) as the "real-world situational

factors" into account when dealing with the phenomenon of IEPVs.

ii) The polysemous nature of IEPVs and their typically informal use in everyday

spoken English.
339

iii) Most importantly, Arabic equivalents provided to IEPVs in bilingual English-

Arabic dictionaries and in the available textbooks on learning and translating

EPVs are intuitive and not theoretically based.

iv) The disregard of fundamental issues, such as register variations, illocutionary

acts, and functional-pragmatic equivalents by Arabic lexicographers left Arabic

translators with no choice but to consult monolingual English-English

dictionaries to understand the functional-pragmatic meaning of IEPVs in order

to be able to guess their Arabic equivalents, which are, in many cases,

functionally and pragmatically inaccurate.

26. Usage of paraphrasing was noticeable in the tendency of a number of subjects to gloss

the listed IEPVs by employing long Arabic sentences instead of providing their

appropriate Arabic functional-pragmatic equivalents.

27. The usage of Arabic colloquial and regional dialects was apparent in the failure of a

number of the subjects, despite the note I made at the beginning of the translation tests

distributed to them, to seek the abovementioned middle-ground Arabic style by

utilizing Modern Standard Arabic when tackling the phenomenon of IEPVs. They

instead produced highly informal colloquial Arabic expressions or regional dialects as

functional-pragmatic equivalents.

28. Despite the fact that translations by employing Arabic colloquial and regional dialects

may deliver the denotative and the communicative meanings of the IEPVs, and they

may well be used by interpreters, they cannot be accepted as written Arabic


340

translations as they are not lexicalized and not current and putting them in writing

sounds unacceptable.

29. The comparison between the two groups demonstrated the fact that the translation

communicative competence of both groups was relatively inadequate. There was a

shortage of knowledge of syntactic, semantic and pragmatic properties of IEPVs, and

a lack of ability in understanding them and translating them into Arabic. The

differences between the two groups regarding the main concern of this study i.e.

achieving the functional-pragmatic equivalents were marginal, which probably gives

the impression that the years of training and experience the Arabic professional

translators have did not help much in translating IEPVs into Arabic.

30. The comparison of the performance of professionals among themselves and the

performance of students among themselves confirmed the abovementioned fact and

revealed that neither the years of experience, in the case of professionals, nor the

stages of study, in the case of students, made any difference with regard to translation

communicative competence of IEPVs. Subjects of each group outperformed each

other regardless of the years of experience and the stages of study. Such a result was

attributed to lack of systematic approach and teaching process of IEPVs.

Nonetheless the Arabic professional translators were more competent and have more

sense of encouragement to take initiative and risk-taking, while the Arabic translation

students deem to require increased awareness of the syntactic, semantic and pragmatic

properties of IEPVs to boost their own ability in comprehending the idiomatic and

polysemic nature of EPVs, and their ability in targeting their communicative


341

meanings by appreciating situational contexts so as to achieve their Arabic functional-

pragmatic equivalents.

6.2 Recommendations

In light of the abovementioned findings, and owing to the diversity of the reasons to

which each piece of finding was attributed, the following pertinent recommendations

can be proposed:

6.2.1 Recommendations for Arabic professional translators

• When translating IEPVs into Arabic, Arabic translators ought to resist the

temptation of hastily translating them literally by taking up the components of

each PV separately. They have to look at each PV as one single, discrete unit of

meaning and consider its idiomatic tendency.

• Arabic translators should not adhere to the direct semantic meaning of IEPVs

(locutionary act/prepositional meaning) when translating them into Arabic. They,

however, have to take into account the situational context in which such verbs are

employed if they are to achieve their functional-pragmatic equivalence.

• In order to achieve the Arabic functional-pragmatic equivalence for IEPVs,

Arabic translators need, to a great extent, to appreciate the register of such

demanding verbs. Primarily, they must be conscious of the parameter of 'field' by


342

recognizing the type of the text in which a given PV is embedded to grasp the

field of the discourse and to see whether it is used in a specialised sense of a

particular field (cf. Heliel 1994). After that, they have to precisely consider the

parameters of 'mode' and 'tenor' to determine the medium by which the utterance

is delivered and the social distance of the interlocutors so as to convey the same

level of formality/informality and to avoid any shift of register as a result of

mismatches between the English and the Arabic languages in this regard. Indeed

such register variations need to be kept in check constantly by translators when

dealing with IEPVs.

• Achieving the functional-pragmatic equivalence of IEPVs also requires Arabic

translators to target their illocutionary acts/pragmatic meanings by accounting for

the speech acts performed in given utterances where IEPVs are used.

• Arabic translators should not reduce IEPVs to sense when translating them into

Arabic. That is, the idiomatic nature has to be kept equivalent. This can typically

be achieved by employing such Arabic idiomatic structures as Arabic proverbs,

Arabic collocations and Arabic fixed expressions.

• When translating IEPVs into Arabic, Arabic translators need to seek the middle

ground Arabic style by adhering to Modern Standard Arabic and avoiding the

usage of both classical and colloquial varieties of the Arabic language.


343

• The strategy of translating English idioms by paraphrasing is quite common

amongst researchers in the field of translation theory (cf. Baker, 1992; Bataineh &

Bataineh, 2002; Ghazala, 2003 to cite just a few). However, it is not

recommended for Arabic translators to typically opt for it when rendering IEPVs

into Arabic, as the findings of the present study demonstrated that resortring to

such an alternative would be very much at the expense of the idiomatic nature of

IEPVs, which has to be kept equivalent in order to achieve the appropriate Arabic

functional-pragmatic equivalence.

• Given the fact that bilingual English-Arabic dictionaries of PVs are far less in

both quantity and quality than the specialized English-English dictionaries of PVs

(see 2.9 above), Arabic translators are best recommended to consult specialized

English-English dictionaries first in order to obtain the contextual meaning and

the situational usage of a given IEPV, then they need to consult the specialized

English-Arabic dictionaries to grasp the Arabic equivalent if found. In case such

an Arabic equivalent is not found, Arabic translators are best advised to take the

parameters of the model devised in this study into their consideration when

coining the required functional-pragmatic equivalent of a given IEPV.

6.2.2 Recommendations for Arabic lexicographers

Arabic Lexicographers are recommended to take the following crucial points into

consideration when compiling bilingual English-Arabic dictionaries:


344

• First and foremost, there is a real need for more specialized bilingual English-

Arabic dictionaries to be compiled on IEPVs. There are, to the best of my

knowledge and research, only two bilingual English-Arabic dictionaries dedicated

to the phenomenon of EPVs, while there are more than ten monolingual English-

English dictionaries devoted to such a phenomenon. The former type of

dictionaries is far below; both in quality and quantity the latter type (see Chapter

Two above).

• Arabic lexicographers should always assign separate entries to IEPVs in their

general bilingual English-Arabic dictionaries, in the same way they are provided

in the monolingual English-English dictionaries, and list each one of them as one

entity rather than under the verb (first component) or the particle (second

component). Such a procedure makes IEPVs more prominent and easy to be

looked up by Arabic dictionaries users.

• Arabic lexicographers ought to put in a great deal of scholarly effort, when

compiling dictionaries to bridge the quantitative and qualitative gaps that exist in

their available dictionaries. This can only be done by accounting for such key

issues as derivational forms of IEPVs, complex idioms where they constitute a

fundamental part, syntactic and semantic properties and most importantly

pragmatic properties of IEPVs. The latter may be attended to by addressing the

question of register variations of IEPVs and by presenting them in a variety of

situational contexts to allow the Arabic readership to appreciate types of speech

acts preformed so as to grasp their appropriate illocutionary meanings.


345

• Arabic lexicographers need to keep up with the productive nature of IEPVs. In

order to do so, they are best advised to follow the Collins COBUILD Dictionary

and the Oxford Phrasal Verbs Dictionary for Learners of English in this respect

and assign separate appendices for exploring the English particles (prepositions

and adverbs) and the shades of meanings they impart when collocating with

proper verbs to form PVs. By doing so, Arabic lexicographers would provide their

readership with a sort of systematic approach to understand the way the IEPVs

work and the logic behind their collocations and, in turn, enable them to make an

informed decision rather than an arbitrary guessing when faced by newly coined

PVs.

• Another way of keeping up with the productive nature of IEPVs, and best

suggested for Arabic lexicographers, is to keep dictionaries up-to-date by

including any IEPV that pops up among English native speakers. This can be done

by many ways, such as utilizing up-to-dated English corpora (such as British

National Corpus, COBUILD The Bank of English and English Language Corpora

to cite only a few), or even by observing the English language used in the mass

media. Cowie (1993) makes the point that "[a] new edition of a dictionary has not

only to include phrasal verbs and meanings which have entered the language since

the first edition was published but must also take account of new sources of

information about the English language"( p. 40).

• Some English lexicographers provided workbooks to be studied in conjunction

with specialized dictionaries, such as Using Phrasal Verbs by McArthur (1975)

and Collins COBUILD Phrasal Verbs Workbook (1994), which are produced in
346

conjunction with the Dictionary of Phrasal Verbs and their Idioms, and the

Collins COBUILD Dictionary of Phrasal Verbs respectively (see 2.8 & 2.10

above). It would be of great benefit to Arabic learners of English, Arabic

translators and Arabic interpreters if Arabic lexicographers follow suit and

provide their readership with workbooks on IEPVs to be studied in conjunction

with their dictionaries. I believe that dictionary compilers are, more than any one

else, in a better position to write teaching materials on a given aspect of language

in which they are specialized.

6.2.3 Recommendations for Arabic pedagogues

It is undeniable fact that types of methods and materials utilized in teaching IEPVs to

Arabic students play a significant role not only in learning and perceiving them but

also in using and producing the right equivalents for them in Arabic. Along these lines

Darwin and Gray (1999) point out that:

[…] to alleviate the difficulties phrasal verbs present, ESL teachers and material
writers need to approach the problematic areas of phrasal-verb pedagogy
systematically, developing and presenting material in a manner that avoids
unnecessary confusion and loss of time for both student and instructor (p. 66).

Hence, Arabic pedagogues are recommended to pay a great deal of attention to the

following vital points:

• Owing to the findings arrived at in this dissertation, in which the poor

performance of both groups that were survyed was the rule rather than the

exception, IEPVs must have a prominent place in the overall learning and training
347

process, and they have to be included as an integral part of the interpreting and

translating courses undertaken by Arabic students.

• The findings of the present study revealed that the newly coined IEPVs, (p)

category were the most difficult for both groups involved. In order to enhance the

competence of Arabic professional translators and Arabic translation students, in

receiving/understanding and producing/translating the newly coined IEPVs into

Arabic, a systematic approach to account for their productive nature and to

explore a number of learnable patterns underling them is inevitable. Arabic

pedagogues should develop such patterns and present them in a systematic way so

as to provide good pedagogical materials that can help overcome the difficulties

caused by EPVs and, most importantly, to equip the learners with a learning

experience that can assist them in this life-long learning profession. English

pedagogues suggest that the patterns underlying EPVs have to be taught explicitly

so as to shortcut the learning process and make it more efficient (Sansome, 2000;

Sawyer, 2000; Side, 1990). McArthur and Atkins (1974) rightly believe that PVs

"are more consistent in their behaviour and far more teachable than is generally

supposed" (p. 5).

• Arabic pedagogues are best encouraged to spend more time and effort on teaching

the doubled idiomatic nature of the complex idioms, (c) category, which

according to the findings of this study constituted the second hardest category for

both groups.
348

• When teaching IEPVs to Arabic students, Arabic pedagogues need to put more

scholarly effort on their derivational mechanism and syntactic peculiarities,

especially the IEPVs derived from nouns, (dpn) subcategory, and the IEPVs of

verb + preposition constructions, (svp) subcategory, as the finding of this study

showed that they were the most difficult subcategories of (d) and (s) categories

respectively.

• Arabic pedagogues have to develop students' positive attitude toward IEPVs by

establishing a logic and sense of IEPVs when teaching them to Arabic students.

Hannan (1998) indicates that proving to students "that there is a human logic,

based on experiences which they can recognise, gives them confidence that it is

feasible to learn these things, and open doors to useful methods of vocabulary

storage and organization" (p. 27).

• In order to provide Arabic learners of English, Arabic translators and Arabic

interpreters with rich pedagogical materials that can help them overcome the

difficulties IEPVs cause, Arabic pedagogues, curriculum designers and material

developers need to consider and make use of the up-to-date teaching methods of

IEPVs (some of which are mentioned in Chapter Two above) and keep away from

the ineffective and fruitless old-fashioned methods in which IEPVs are deemed as

random and haphazard combinations of verbs and particles and there is no way for

students to master them but to memorize them by heart.

• The majority of experienced English pedagogues who proposed methods of

teaching EPVs to non-native learners of English are in favour of listing and


349

learning them according to their particles rather than their proper verbs. Arabic

pedagogues are best advised to follow suit, since it is the particles, more than the

verbs, which carry the main portion of the function of these combinations (see

2.10 above).

• Arabic pedagogues are required to keep in mind the previously stated point made

by McArthur (1971) in which he maintains that PVs must be taught as units, and

as the equivalent of single verbs. Prompting Arabic learners of English with such

a crucial principle would keep them away from tracing the literal meanings of

these constructions by taking each component of them on its own. IEPVs are

indeed constructions of one semantic unit and have to be taught as such.

• In order to avoid the intuitive decisions of material developers when choosing the

IEPVs to be taught, Arabic pedagogues are best recommended to take into

consideration the previously mentioned point made by Cornell (1985) who

suggests to assemble a "core" of EPVs to determine and choose the most frequent

and common IEPVs to be taught to non-native learners of English.

• Arabic pedagogues should consider the point raised by Side (1990) in which he

asserts that due to the analogous nature of EPVs "single examples should never be

taught in isolation". IEPVs have to be learned in contexts rather than in lists of

unrelated words, thus, connections always have to "be made in order to establish

their context within the language, to show they are meaningfully idiomatic rather

than meaninglessly random" (pp. 150-151).


350

• When teaching IEPVs to Arabic learners of English, Arabic translators and Arabic

interpreters, Arabic pedagogues have to utilize authentic example sentences

representing communicative situations taken from reliable, up-to-date and

comprehensive enough corpora, and avoid what is called by Kharma and Hajjaj

(1989) the "idealisation of teaching materials". Such bookish idealised examples

would do more harm than good and deepen the negative attitude of Arabic

learners toward IEPVs.

• Arabic students, Arabic translators and Arabic interpreters are best encouraged to

make their own list of IEPVs according to the particle in their vocabulary

notebooks. Such a method was suggested by some English pedagogues to non-

native learners of English as a way to mastering the perception and production of

problematic EPVs (cf. Side, 1990; Flower, 2000).

6.3 Contributions made by this study

The contributions made by this study to the knowledge base of linguistics and

translation are sevenfold: 1) it is the first academic thesis entirely devoted to

exploring the most challenging phenomenon of IEPVs on a linguistic contrastive

analysis framework, so as to describe and bring to light the correspondences and

differences of the characteristics of such a phenomenon between the English and

Arabic languages; 2) The study provides a generic practical linguistic model as a

disciplined approach for the analysis and translation of IEPVs into Arabic. The model

was applied as a yardstick by which the translation quality of such idiomatic

expressions was analysed, compared, and evaluated; 3) it presents a comprehensive


351

syntactic, semantic and pragmatic categorization of the problematic aspects of the

phenomenon of IEPVs. Such a categorization leads a greater perception of the

phenomenon and hopefully will constitute a platform from which prospective

researchers, practicing translators and interpreters, lexicographers and pedagogues can

move on to enhance their strategies of dealing with this phenomenon; 4) it is the first

of its kind to carry out a large scale experimental research into translating IEPVs into

Arabic. The considerable number of the items of the translation tests (100 questions)

and the enormous number of answers obtained (2400 answers) provided the study

with a reliable amount of data and enabled the present researcher to confidently

generalize the essential findings arrived at; 5) the study has the virtue of targeting not

only the professional translators but also the translation students and the virtue of

coming up with a number of fruitful insights as to the impact of the years of

experience and stages of study on translation communicative competence of the

participants; 6) the study contributes to reconciling the theory and the practice as to

translating IEPVs into Arabic, by providing vital insights on the theoretical gaps that

caused the abovementioned problems in practice. 7) it provides a number of

theoretically based recommendations for translators to help them tackle the

difficulties of translating IEPVs into Arabic, for lexicographers to close the existing

gaps and for pedagogues to modify their methods and materials and make them more

valid and practical.

6.4 Directions for further research

This dissertation mainly concentrated on translating IEPVs into Arabic. In spite of its

comprehensive coverage of the peculiarities of the phenomenon of IEPVs and the


352

difficulties they cause to Arabic professional translators and Arabic translation students,

it was limited to one language pair. It will be interesting to replicate and extend this

study to other language combinations. The workable model devised in this study for the

analysis and translation of IEPVs into Arabic can easily be employed in conducting

more parallel studies to analyse IEPVs and translate them into other languages.

Since IEPVs are typically used in everyday spoken English more than in formal written

English, they are more likely to pose a great deal of difficulties to interpreters when

rendering them to Arabic. It would also be important to replicate or extend this study to

investigate the difficulties encountered by Arabic (or any other language) interpreters

when facing such a phenomenon.

It goes without saying that the real need for a comprehensive, up-to-date, and

theoretically based bilingual English-Arabic dictionary of PVs is urgent. Such a very

much wanted dictionary can well be compiled in the light of the findings and

recommendations of this study which pave the way to Arabic-English lexicographers to

avoid the existing critical drawbacks if they are to achieve the objective of satisfying their

end users, that is, the learners of English and Arabic, as well as the English-Arabic

translators.
353

BIBLIOGRAPHY

Aarts, B. (1989). Verb-Preposition Constructions and Small Clauses in English J.


Linguistics (25), 277-290.

Aarts, B. (1997). English Syntax and Argumentation. London: University College.

Abboud, P. F., & McCarus, E. N. (1968). Elementary Modern Standard Arabic.


United Kingdom: Cambridge University Press.

Abdel-Hafiz, A.-S. (2003). Pragmatic and Linguistic Problems in the Translation of


Naguib Mahfouz's The Thief and the Dogs: A Case Study. Babel, 49 (3), 229-
252.

Al-bayyatii, T. (1982). What Pupils Need in English. Beirut: Al-NahDah Institution


for Cultural Services.

Al-bustaanii, F. A. (1963). munjid al-Tulaab (in Arabic). Beirut: al-maTba3atu al-


katholiikyyah.

Al-ghalaayiinii, M. (1986). Jaami3 al-dduruus al-3arabiyyah (in Arabic). Beirut: al-


maktabatu al-3aSriyyah.

Ali, A. S. M. (2004). A study of antonymous and synonymous couplings in Arabic


with reference to translation. Babel, 50 (4), 346-360.

Al-jaaHiZ, A. A. (1932). al-bayaan wa-altabyiin (in Arabic) Cairo: al-maTba3atu al-


raHmaanyyah.

Al-jurjaanii, A. A. (1991). ?asraar al-balaaghah (in Arabic). Beirut: Daar al-jiil.

Al-jurjaanii, A. A. (n.d.). dalaa?ilu al-?i3jaaz (in Arabic). Cairo: maTba3atu majallati


al-manaar.

Al-jurjaanii, A. M. (1969). kitaab al-ta3riifaat (in Arabic). Beirut. maktabat lubnan.

Alkhuli, M. A. (1999). Comparative Linguistics: English and Arabic. Jordan: Alfalah


House.

Al-Kufaishi, A. (2004). Translation as a learning and teaching strategy. Babel, 50 (1),


45-59.

Al-Qinai, J. (2000). Translation Quality Assessment: Strategies, Parameters and


Procedures. Meta, XLV (3), 497-519.

Al-raazii, M. (1981). mukhtaar al-SiHaaH (in Arabic). Beirut: Daar al-kitaab al-
3arabii.
354

Al-saghir, A. M. (1993). English for All Levels (Fifth ed.). Riyadh: Narjis Printing
Press.

Al-sarraaj, A. M. (1973). rissalat al-?ishtiqaaq (in Arabic). Damascus: maTba3at al-


3ilm.

Al-tha3aalibii, A. (n.d.). kitaab fiqih al-lughtia wa-asraar al-3arabiyyah (in Arabic).


Beirut: Daar maktabat al-Hayaat.

Al-zamakhsharii, A. J. (1948). al-keshshaaf (in Arabic). Cairo: Daar al-kitaab al-


3arabii.

Al-zamakhsharii, A. J. (1953). ?assasal-balaaghah (in Arabic). Ciaro: maTba3atu


?urfaand.

Armstrong, K. (2004). Sexing Up the Dossier: A Semantic Analysis of Phrasal Verbs


for Language Teachers. Language Awareness, 13(4), 213-224.

Austin, J. L. (1975). How to do things with words. Oxford: Clearndon Press.

Awwad, M. (1990). Equivalence and Translatability of English and Arabic Idioms.


Papersand Studies in Contrastive Linguistics (26), 57-67.

Azzaro, G. (1992). The Syntactic Learning of English Phrasal Verbs: Theory.


Rassegna Italiana di Linguistica Applicata, XXIV (1), 33-60.

Baker, M. (1992). In Other Words: A coursebook on translation. London and New


York: Routledge.
Barber, C. (1993). The English Language: A Historical Introduction. Cambridge:
Cambridge University Press.

Bataineh, R. F., & Bataineh, R. F. (2002). The Difficulties Jordanian Graduate


Learners of English as a Second language Face When Translating English
Idioms into Arabic. RASK, 16, 33-83.

Beate Hampe, J. (1997). Towards a solution of the phrasal verb puzzle:


Considerations on some scattered pieces. Lexicology, 3/2, 203-243.

Belkacemi, C. (1998). Prepositions with verbal functions in Arabic. Orientalia


Suecana (47), 13-20.

Bell, R. (1991). Translation and Translating: Theory and Practice. London and New
York: Longman.

Bennett, D. C. (1975). Spatial and Temporal Uses of English Prepositions: An Essay


in Stratificational Semantics: London: Longman Group Limited.

Berman, L. A., & Kirstein, L. (1996). Practical Idioms: Using Phrasal Verbs in
Everyday Contexts (Second ed.). Illinois: National Textbook Company (NTC).
355

Bernstein, J. S. (1974). On the Semantics of Certain English Phrasal Verbs and their
Rendering into Spanish. Bilingual Review/Revista Bilingue, 1 (1), 59-66.

Besserman, L. (1998). Chaucer's Multi-Word Verbs: A Historical Introduction and


Illustrative Sample. NOWELE (34), 99-153.

Bolinger, D. (1971). The Phrasal Verb in English. Cambridge and Massachusetts:


Harvard University Press.

Brashi, A. S. (2005). Arabic Collocations: Implications for Translation. Unpublished


doctoral dissertation, University of Western Sydney, Australia.

Brugman, C. M. (1983). Story of Over. Unpublished master's thesis, University of


California at Berkeley.

Campbell, S. (1998). Translation into the second language. London and New York:
Longman.

Campoy, M. C. (2002). Up and Over: Phrasal Verbs Expressing Superiority and


Quantity. Modern English Teacher, 11(1), 17-19.

Cantarino, V. (1974). Syntax of Modern Arabic Prose. Bloomington / London:


Indiana University Press.

Catford, J. (1974). A linguistic Theory of Translation: An Essay in Applied Linguistics


(Fourth ed.). London, New York and Toronto: Oxford University Press.

Chakhachiro, R. (1997). The Translation of Irony in Australian Political Commentary


Texts from English into Arabic. Unpublished doctoral dissertation, University
of Western Sydney, Sydney.

Chakhachiro, R. (2005). Revision for Quality. Perspectives: Studies in Translatology,


13 (3), 225-238.

Chen, P. (1986). Discourse and Particle Movement in English. Studies in Language,


10 (1), 79-95.

Chesterman, A. (1998). Contrastive Functional Analysis. Amsterdam/Philadelphia:


John Benjamins Publishing Company.

Close, R. A. (1992). A Teacher's Grammar: An Approach to the Central Problems of


English. London: Language Teaching Publications.

Collins, P. (1998). English Grammar. South Melbourne: Wesley Longman Australia


Pty Limited.

Corless, B. (1979). Formal English. Canberra: Australian National University Press.


Cornell, A. (1985). Realistic Goals in Teaching and Learning Phrasal Verbs. IRAL,
XXIII (4), 269-280.
356

Cowie, A. P. (1993). Getting to grips with phrasal verbs. English Today 36, 9 (4), 38-
41.

Crowley, T., Lynch, J., Siegel, J., & Piau, J. (1995). The Design of English: An
introduction to descriptive linguistics. Auckland: Longman Paul.

Crutchley, A. (2007). Comprehension of idiomatic verb + particle constructions in 6-


to 11- year-old children. First Language, 27 (3), 203-226.

Crystal, D. (1997). The Cambridge Encyclopedia of the English Language.


Cambridge: Cambridge University Press.

Dainty, P. (1991). Phrasal Verbs in Context. London and Basingstoke: Macmillan


Publishers Limited.

Darwin, C. M., & Gray, L. S. (1999). Going After the Phrasal Verb: An Alternative
Approach to Classification. TESOL QUARTERY, 33 (1), 65-83.

DeArmond, R. C. (1977). On Decide On and Similar Constructions in English.


glossa, 11 (1), 20-47.

Dehe, N. (2005). The Optimal Placement of up and ab-A Comparison. The Journal of
Comparative Germanic Linguistics, 8 (3), 185-224.

Dixon, R. M. W. (1982). The Grammar of English Phrasal Verbs. Australian Journal


of Linguistics, 2, 1-42.

Dixon, R. M. W. (1991). A New Approach to English Grammar, on Semantic


Principles. Oxford: Clarendon Press.

El-Nagar, H. Y. (1998). Phrasal Verbs. Cairo: Dar El-Talae for Publishing,


Distribution and Exporting.

El-Yasin, M. K. (1996). The Passive Voice: A Problem for the English-Arabic


Translator. babel, 42 (1), 18-26.

Emery, P. G. (1988). Collocation: A Problem in Arabic/English Translation?


Quinquereme, 11 (1), 178-184.

Emery, P. G. (1991). Collocation in Modern Standard Arabic. Journal of Arabic


Linguistics (23), 56-65.

Fawcett, P. (2001). Linguistic approaches. In M. Baker (Ed.), Routledge Encyclopedia


of Translation Studies. London and New York: Routledge.

Fernando, C. (1996). Idioms and Idiomaticity. Oxford: Oxford University Press.

Fisiak, J. (Ed.). (1981). Contrastive Linguistics and the Language Teacher. Oxford:
Pergamon Press.
357

Flower, J. (2000). Phrasal Verb Organiser (Seventh ed.). London: Language


Teaching Publications (LTP).

Fraser, B. (1974). The Phrasal Verb in English [Review of the book]. Language, 50
(3), 568-575.

Fraser, B. (1976). The Verb-Particle Combination in English. New York, San


Francisco and London: Academic Press.

Gardner, D. & Davies, M. (2007). Pointing Out Frequent Phrasal Verbs: A Corpus-
Based Analysis. TESOL QUARTERLY, 41 (2), 339-360.

Ghadessy, M. (1993). Register analysis: Theory and Practice. London and New
York: Printer Publishers.

Ghazala, H. (2003). Idiomaticity Between Evasion and Invasion in Translation:


Stylistic, Aesthetic and connotative Considerations. Babel, 49 (3), 203-227.

Goodale, M. (1994). Collins COBUILD Phrasal Verbs Workbook (Second ed.).


London: Harper Collins Publishers.

Gries, S. (2002). The influence of processing on syntactic variation: Particle


placement in English. In N. Dehe, R. Jackendoff, A. McIntyre & S. Urban
(Eds.), Verb-Particle Explorations Berlin and New York: Mourton de Gruyter.

Grundy, P. (1995). Doing Pragmatics. London: Edward Arnold, a division of Hodder


Headline PLC.

Hale, S. (1996). Pragmatic Considerations in Court Interpreting ARAL, 19 (1), 61-72.

Hale, S. (1997). The Treatment of Register Variation in Court Interpreting. The


Translator, 3 (1), 39-54.

Hale, S. (2002). How faithfully do court interpreters render the style of non-English
speaking witnesses' testimonies? A data-based study of Spanish-English
bilingual proceedings. Discourse Studies, 4 (1), 25-47.

Hale, S. B. (2004). The Discourse of Court Interpreting: Discourse practices of the


law, the witness and the interpreter. Amsterdam/Philadelphia: John Benjamins
Publishing Company.

Halliday, M., & Hasan, R. (1976). Cohesion in English. London and New York:
Longman.

Halliday, M., & Hasan, R. (1985). Language, context, and text: Aspects of language
in a social-semiotic perspective. Victoria: Deakin University.
Hammaad, A. (1983). 3awaamil al-taTawwri al-lughawii (in Arabic). Beirut: Daar al-
?andalus.
358

Hannan, P. (1998). Particles and Gravity: Phrasal Verbs with 'Up' and 'Down". MET,
7 (1), 21-27.

Hatim, B. (1997). Communication across Cultures: Translation Theory and


Contrastive Text Linguistics. UK: University of Exeter Press.

Hatim, B. (2001). Discourse analysis and translation. In M. Baker (Ed.), Routledge


Encyclopedia of Translation Studies. London and New York: Routledge.

Hatim, B. (2001). Pragmatics and translation. In M. Baker (Ed.), Routledge


Encyclopedia of Translation Studies. London and New York: Routledge.

Hatim, B. (2001). Teaching and Researching Translation. London: Longman.

Hatim, B. (2001). Text linguistics and translation. In M. Baker (Ed.), Routledge


Encyclopedia of Translation Studies. London and New York: Routledge.

Hatim, B., & Mason, I. (1990). Discourse and the Translation. London: Longman.

Hatim, B., & Mason, I. (1997). The Translator as Communicator. London: Routledge.

Heaton, J. B. (1965). Prepositions and Adverbial particles. London: Longman Group


Limited.

Heaton, J. B. (1968). Using Prepositions and Particles (Fourth ed.). London:


Longman Group Limited.

Heine, A. (2002). Functional Phrasal Verbs in a Learner's Dictionary Using Material


from Langenscheidts Grossworterbuch Deutsh als Fremdsprache.
Neuphilologische Mitteilungen, 103 (1), 51-62.

Heliel, M. H. (1994). Verb-Particle Combinations in English and Arabic: Problems


for Arab Lexicographers and Translators. In R. De Beaugrand, A. Shunnaq &
M. H. Heliel (Eds.), Language, Discourse and Translation in the West and
Middle East. Amsterdam/Philadelphia: John Benjamins Publishing Company.

Hermans, T. (2001). Models of translation. In M. Baker (Ed.), Routledge


Encyclopedia of Translation Studies. London and New York: Routledge.

Hill, L. A. (1968). Prepositions and adverbial Particles: An Interim Classification


Semantic, Structural, and Graded. London: Oxford University Press.

Hiltunen, R. (1983). Phrasal Verbs in English Grammar Books before 1800.


Neuphilologische Mitteilungen, 3 (Lxxxiv), 376-386.

Hoey, M., & Houghton, D. (2001). Contrastive analysis and translation. In M. Baker
(Ed.), Routledge Encyclopedia of Translation Studies. London and New York:
Routledge.
359

Holes, C. (1994). Designing English-Arabic Dictionaries. In R. De Beaugrand, A.


Shunnaq & M. Heliel (Eds.), Language, Discourse and Translation In the
West and Middle East (pp. 161-180). Amsterdam/Philadelphia: John
Benjamins Publishing Company.

Holes, C. (1995). Modern Arabic: Structures, Functions and Varieties. London and
New York: Longman.

Homeidi, M. A. (2004). Arabic translation across culture. Babel, 50 (1), 13-27.

Hopper, V. F., Gale, C., & Foot, R. C. (1990). Essentials of English (Fourth ed.). New
York, London, Toronto and Sydney: Barron's Educational Series, INC.

House, J. (1977). A Model for Translation Quality Assessment. Tubingen: TBL


Verlag Gunter Narr.

House, J. (1986). Acquiring Translational Competence in Interaction. In J. House &


S. Blum-Kulka (Eds.), Interlingual and Intercultural Communication.
Germany: Gunter Narr Verlag Tubingen.

House, J. (2001). Quality of translation. In M. Baker (Ed.), Routledge Encyclopedia of


Translation Studies. London and New York: Routledge.

House, J. (2001). Translation Quality Assessment: Linguistic Description versus


Social Evaluation. Meta, XLVI (2).

Huddleston, R. (1988). English grammar: an outline. Cambridge, New York, New


Rochelle, Melbourne and Sydney: Cambridge University Press.

Hurford, J., & Heasley, B. (1983). Semantics: a coursebook. Cambridge: Cambridge


University Press.

Hymes, D. H. (1972). On Communicative Competence. In J. B. Pride & J. Holmes


(Eds.), Sociolinguistics: Selected Readings. Victoria, Australia: Penguin
Books.

Ibrahim, Z., Aydelott, S., & Kassabgy, N. (Eds.). (2000). Diversity in Language:
Contrastive Studies in Arabic and English Theoretical and Applied
Linguistics. Cairo and New York: The American University in Cairo Press.

Jackendoff, R. (2002). English particle constructions, the lexicon, and the autonomy
of syntax. In N. Dehe, R. Jackendoff, A. McIntyre & S. Urban (Eds.), Verb-
Particle Explorations. Berlin and New York: Mourton de Gruyter.

Jacobs, R. A. (1995). English Syntax: A Grammar for English Language


Professionals. Oxford: Oxford University Press.
James, C. (1980). Contrastive Analysis. London: Longman.

Jespersen, O. (1976). Essential of English Grammar. London: Lowe and Brydone


Printers Limited.
360

John, & Smithback, C. Y. (1991). Fun with Idioms. Singapore: Federal Publications.

Jones, L. (1985). Use of English: Grammar practice activities for intermediate and
upper-intermediate students. Cambridge, London, New York, New Rochelle,
Melbourne and Sydney: Cambridge University Press.

Jong, A. L. a. N. d. (2002). German particle verbs and word formation. In Nicole


Dehe, Ray Jackendoff, Andrew McIntyre & S. Urban (Eds.), Verb-Particle
Explorations. Berlin and New York: Mourton de Gruyter.

Jowett, W. P. (1951). On Phrasal Verbs. English Language Teaching, 5, 152-157.

Kaluza, H. (1984). English Verbs with Prepositions and Partcles. IRAL, XXII (2), 109-
113.

Kaminska, P. (2001). Keywords for phrasal verbs. MET 10 (2), 59-61.

Kaplan, J. (1995). English Grammar Principles and Facts (Second ed.). Englewood
Cliffs, New Jersey: Prentice Hall.

Kennedy, A. G. (1967). The Modern English Verb-Adverb Combination. New York:


AMS Press Inc.

Kenny, D. (2001). Equivalence. In M. Baker (Ed.) Routledge Encyclopedia of


Translation Study. London and New York.

Kharma, N. (1983). A Contrasitve Analysis of the Use of Verb Forms in English and
Arabic. Heidelberg: Julius Groos Verlag.

Kharma, N., & Hajjaj, A. (1989). Errors in English Among Arabic Speakers: Analysis
and Remedy. London: Longman

Kroch, A. S. (1979). The Verb-Particle Combination in English [Review of the book].


Language 55 (1), 219-224.

Lattey, E. (1986). Pragmatic Classification of Idioms as an Aid for the Language


Learning. IRAL, XXIV (3), 217-233.

Leech, G. (1974). Semantics. Harmondsworth: Penguin Books.

Lentzner, K. R. (1977). Semantic and Syntactic Aspects of Arabic Prepositions.


Unpublished doctoral dissertation, Georgetown University, Washington, D. C.

Lindner, S. J. (1983). A Lexico-Semantic Analysis of English Verb Particle


Constructions With Out and Up. Unpublished doctoral dissertation, Indiana
University

Lindstromberg, S. (1991). (Re)teaching Prepositions. English Teaching Forum, XXIX


(2), 47-50.
361

Lindstromberg, S. (1998). English Prepositions Explained. Amsterdam/Philadelphia:


John Benjamins Publishing Company.

Live, A. H. (1965). The Discontinuous Verb in English. Word (21), 428-451.

Marton, W. (1972). Pedagogical Implications of Contrastive Analysis. Studia Anglica


Posnaniensia, 4, 115-125.

Mason, I. (2001). Communicative/functional approaches. In M. Baker (Ed.),


Routledge Encyclopedia of Translation Studies. London and New York:
Routledge.

McArthur, T. (1971). Teaching English Phrasal Verbs. Teaching, 43 (3), 71-75.

McArthur, T. (1975). Using Phrasal Verbs (Second ed.). London and Glasgow:
Collins.

McArthur, T. (1979). The strange cases of the English phrasal verb. Zielsprache
Englisch, 9 (3), 24-26.

McArthur, T. (1989). The long-neglected phrasal verbs. English Today, 5 (2), 38-44.

Mikio, K. (1997). Instructional Effects of Positive and Negative Evidence on


Prepositional/Phrasal Verbs. The IRLT Bulletin (11), 1-39.

Mitchell, T. F. (1958). Syntagmatic Relations in Linguistic Analysis. Translations of


The Philological Society, 101-118.

Moon, R. (1998). Fixed Expressions and Idioms in English. Oxford: Clarendon Press.

Morgan, P. S. (1997). Figuring out figure out: Metaphor and the semantics of the
English verb-particle construction. Cognitive Linguistics (8-4), 327-357.

Mortimer, C. (1979). Phrasal Verbs in Conversation (Seventh ed.). London:


Longman Group Limited.

Moutaouakil, A. (1989). Pragmatic Functions in a Functional Grammar of Arabic.


Dordrecht: Foris Publications.

Najiib, A. M. (2001). ?usus al-tarjamah (in Arabic) (Fourth ed.). Ciaro: maktabat
?ibn siinaa lilTiba3a wa-alnashr wa-lajwzii3 wa-altaSdiir.

Newby, M. (1987). The Structure of English: A Handbook of English Grammar.


Cambridge, New York, New Rochelle, Melbourne and Sydney: Cambridge
University Press.

Newmark, P. (1996). Introductory Survey. In R. Owens (Ed.), The Translator's


Handbook (Third ed.). London: Aslib, The Association for Information
Management.
362

Newmark, P. (1988). A Textbook of Translation. London: Longman.

Nida, E. (2001). Contexts in Translating. Amsterdam/Philadelphia: John Benjamins


Publishing Company.

Nord, C. (1991). Text Analysis in Translation: Theory, Methodology, and Didactic


Application of a Model for Translation-Oriented Text Analysis. Amsterdam -
Atlanta, GA: Rodopi B. V.

Palmer, F. R. (1968). A Linguistic Study of the English Verb. Coral Gables and
Florida: University of Miami Press.

Palmer, F. R. (1974). The English Verb (Second ed.). London: Longman.

Quirk, R., & Greenbaum, S. (1973). A University Grammar of English. London:


Longman.

Quirk, R., Greenbaum, S., Leech, G., & Svartvik, J. (1985). A Comprehensive
Grammar of the English Language. London and New York: Longman.

Ronowicz, E., Hehir, J., Kaimi, T., Kojima, K., & Lee, D. (2005). Translator's
frequent Lexic Store and Dictionary Use as Factors in SLT Comprehension
and Translation Speed-AComparative Study of Professional, Paraprofessiona
and Novice Transaltors. META, L (2), 580-596.

Rottet, K. J. (2005). Phrasal Verbs and English Influence in Welsh. Word, 56 (1), 39-
70.

Salama-Carr, M. (2001). Interpretive Approach. In M. Baker (Ed.), Routledge


Encyclopedia of Translation Studies. London and New York: Routledge.

Sansome, R. (2000). Applying lexical research to the teaching of phrasal verbs. IRAL,
38, 59-69.

Sawyer, J. H. (1999). Verb adverb and verb particle constructions: Their syntax and
acquisition. Unpublished doctoral dissertation, Boston University.

Sawyer, J. H. (2000). Comments on Clayton M. Darwin and Loretta S. Gray's "Going


After the Phrasal Verb: An Alternative Approach to Classification" A Reader
Reacts. TESOL QUARTERY, 34 (1), 151-159.

Sayyd Ahmad, R. A. (1993). fannu al-kitaabah. Beirut (in Arabic): Daar al-bayaan al-
3arabii.

Searle, J. R. (1969). Speech Acts: An Essay in The Philosophy of Language. Oxford:


Alden & Mowbray Ltd.

Shakkour, J. (1992). kitaab al-bayaan (in Arabic). Beirut: Daar al-fikr al-lubnanii.
363

Sheen, R. (2000). Comments on Clayton M. Darwin and Loretta S. Gray's "Going


After the Phrasal Verb: An Alternative Approach to Classification" Another
Reader Reacts. TESOL QUARTERY, 34 (1), 160-173.

Shovel, M. (1992). Making Sense of Phrasal Verbs. London: Prentice Hall


International (UK) Limited.

Side, R. (1990). Phrasal verbs: sorting them out. ELT Journal, 44 (2), 144-152.

Sjoholm, K. (1995). The Influence of Crosslinguistic, Semantic, and Input Factors on


the Acquisition of English Phrasal Verbs. Finland: Abo Akademi Printing
Press.

Snell-Hornby, M. (1995). Translation Studies: An Integrated Approach.


Amsterdam/Philadelphia: John Benjamins Publishing Company.

Sorvali, S. (1996). Translation Studies in a New Perspective. Frankfurt am Main,


Berlin, New York, Paris and Wien: Peter Lang.

Sroka, K. A. (1972). The Syntax of English Phrasal Verbs. The Hague and Paris:
Mouton.

Swierzbin, B. (1996). Word Order with Separable Phrasal Verbs. Unpublished


master's thesis, University of Minnesota.

Taha, A. (1960). The Structure of Two-Word Verbs in English. Language Learning,


X (3 & 4), 115-122.

Taymuur, M. (n.d). mushkilaatu al-lughati al-3arabyyah (in Arabic). Beirut: al-


maktabah al-3aSryyah.

Thomson, A. J., & Martinet, A. V. (1986). A Practical English Grammar (Fourth ed.).
Oxford: Oxford University Press.

Thrush, E. A. (2001). Pain English? A study of plain English vocabulary and


international audiences. Technical Communication, 48 (3), 289-296.

Vestergaard, T. (1977). Prepositional Phrases and Prepositional Verbs: A Study in


Grammatical Function. The Hague, Paris and New York: Mouton.

Villaviicencio, A., & Copestake, A. (2003). Verb-particle constructions in a


computational grammar of English. Paper presented at the 9th International
Conference on Head-Driven Phrase Structure Grammar (HPSG), South Korea.

Waafii, A. A. (1973). fiqh al-lughah (in Arabic). Ciaro: Daar al-nahDah.


Wallace, L. C. (1968). Idiomaticity as an Anomaly in the Chomskyan Paradigm.
Foundations of Language (4), 109-127.

Wightwick, J., & Gaafar, M. (1998). Arabic Verbs and Essentials of Grammar.
Illinois: Passport Books.
364

Wood, F. T. (1964). English Verbal Idioms. London and Basingstoke: Macmillan


Publishers Ltd

Wood, F. T. (1967). English Prepositional Idioms. London and Basingstoke:


Macmillan and Co Ltd.

Wright, W. (1981). A Grammar of the Arabic Language (Third ed.). Beirut: Librairie
Du Liban.

Yatskovich, I. (1999). Some Ways of Translating English Phrasal Verbs into Russian.
Translation Journal, 3 (3), 1-4. Retrieved April 12, 2002, from
http://accurapid.com/journal/09russ.htm

Zeller, J. (2001). Particle verbs and local domains. Amsterdam/Philadelphia: John


Benjamins Publishing Company.

Zughoul, M. (1979). Teaching English Prepositions. English Teaching Forum, 42 (3),


24-29.

DICTIONARIES

Ba'albaki, M. (1994). Al-Mawrid: A Modern English-Arabic Dictionary (Twenty


eighth ed.). Beirut: Dar El Ilm Lilmalayin.

Baalbaki, R. (2000). Al-Mawrid: A Modern Arabic-English Dictionary (Tirteenth ed.).


Beirut: Dar El Ilm Lilmalayin.

Benson, M., Benson, E., & Ilson, R. (1997). The BBI Dictionary of English Word
Combinations. Amsterdam/Philadelphia: John Benjamins Publishing
Company.

Butler, S. (Ed.). (1990). The Macquarie Dictionary of New Words. NSW: The
Macquarie Library.

Constant, M. C. (1991). English Phrasal Verbs in French. Hong Kong: Thomas


Nelson and Sons Ltd.

Courtney, R. (1983). Longman Dictionary of Phrasal Verbs. England: Longman


Group Limited.

Cowie, A. P., & Mackin, R. (1975). Oxford Dictionary of Current Idiomatic English.
London: Oxford University Press.

Cowie, A. P., & Mackin, R. (1993). Oxford Dictionary of Phrasal Verbs (Second ed.).
Oxford: Oxford University Press.
365

Daud, N., Dollah, N., & Zubir, B. (2003). Linguistics Dictionary: English-Arabic
Arabic-English. Kuala Lampur: A. S. Noordeen.

Delbridge, A., Bernard, J., Blair, D., Butler, S., Peter, P., & Yallop, C. (Eds.). (2001).
The Macquarie Dictionary (Federation ed.). NSW: The Macquarie Library.

Doniach, N. S. (1982). The Concise Oxford English-Arabic Dictionary. Oxford:


Oxford University Press.

Doniach, N. S. (Ed.). (1981). The Oxford English-Arabic Dictionary Of Current


Usage (Fourth ed.). Oxford: Oxford University Press.

Gove, P. B. (Ed.). (1993). Webster's Third New International Dictionary Springfield,


Massachusetts: Merriam-Webster INC, Publishers.

Heliel, M. H. (2000). York Dictionary of Phrasal Verbs and Their Idioms: English-
English-Arabic. Beirut: York Press-Librairie du Liban Publishers.

Karmi, H. (1997). Al-Mughni Al-Akbar. Beirut: Librairie du Liban.

Khalaili, K. (1979). English Phrasal Verbs in Arabic. London: Hodder and Stoughton.

Kirkpatrick, E. M., & Davidson, G. (Eds.). (1982). Time-Chambers Dictionary of


Phrasal Verbs. Singapore: Federal Publications (S) Pte Ltd.

McArthur, T., & Atkins, B. (1974). Dictionary of English Phrasal Verbs and their
Idioms. London and Glasgow: Collins.

McCarthy, M. (Ed.). (1997). Cambridge International Dictionary of Phrasal Verbs.


Cambridge: Cambridge University Press.

Mowafi, M. I., Kharma, N. N., Durayni, M. R., Fayad, S. H., Barbary, S., Hajjaj, S., et
al. (1985). A Dictionary of English Idioms: English-Arabic. Beirut: Librairie
du Liban.

Nasser, K. A. (1997). A Dictionary of American & Canadian Terms: English-Arabic.


Beirut: Librairie du Liban.

Oxford Phrasal Verbs Dictionary of Learners of English. (2001). Oxford: Oxford


University Press.

Oxford Wordpower Dictionary: English-English-Arabic (2000). (Third ed.). Oxford:


Oxford University Press.

Rooney, K. (Ed.). (2004). Macquarie International English Dictionary (Second ed.).


Sydney: Pan Macmillan Australia Pty Limited.
Siinii, M., Hussein, M., & Al-doush, S. A. (1996). al-mu3jam al-syaaqii lilta3biiraat
al-?iSTilaaHyyah (in Arabic). Beirut: maktabat lubnan nashiroun.
366

Simpson, J. A., & Weiner, E. S. C. (1989). The Oxford English Dictionary (Second
ed.). Oxford: Clarendon Press.

Sinclair, J. (Ed.). (1998). Collins COBUILD Dictionary of Phrasal Verbs. London and
Glasgow: Collins.
Spears, R. (1993). NTC's Dictionary of Phrasal Verbs and Other Idiomatic Verbal
Phrases. Illinois: National Textbook Company (NTC) Publishing Group.

Stern, G. (2000). The Grammar Dictionary: 250 terms explained, and illustrated with
real-life texts. Greenwood: R. I. C. Publications.

Taha, A. (1972). An Analysis and Dictionary of Two-Word Verbs in English


Illustrated in Sentences. Kuwait: Kuwait University.

Taha, A. (1996). York Dictionary of Two-Word Verbs in English. Beirut: York Press-
Librairie du Liban Publishers.

Trask, R. L. (2000). Dictionary of English Grammar. London: Penguin Group.

Turton, N., & Manser, M. (1985). The Student's Dictionary of Phrasal Verbs. London
and Basingstoke: Macmillan.

Wallace, M. J. (1981). Dictionary of English Idioms. Glasgow and London: Collins.

Wehr, H. (1980). A Dictionary of Modern Written Arabic: Arabic-English (Third


Printing ed.). Beirut and London Librairie Du Liban and Macdonald & Evans
LTD.

NEWSPAPERS

The Australian (2004, March 31). p. 1

The Australian. (2005, May 10). p. 2

The Australian. (2005, May 10). p. 22

The Daily Telegraph (2004, August 12). p. 1

The Daily Telegraph (2004, August 12). p. 9

The Daily Telegraph (2004, August 12). p. 14

The Daily Telegraph (2004, August 12). p. 23

The Sun Herald (2004, August 8). pp. 10-11

The Sunday Telegraph (2004, April 11). p. 1

The Sun-Herald (2004, August 8). p. 51


367

The Sydney Morning Herald (2004, June 3). p. 11

The Sydney Morning Herald (2004, June 3). p. 2

WEB SITES
http://a4esl.org/q/j/ck/fb-phrasalverbs.html

http://eslus.com/LESSONS/GRAMMAR/PHRASALS/Pv1.htm

http://eslus.com/LESSONS/GRAMMAR/PHRASALS/Pv1b.htm

http://eslus.com/LESSONS/GRAMMAR/PHRASALS/Pv1c.htm

http://eslus.com/LESSONS/GRAMMAR/PHRASALS/Pv1d.htm

http://homepage.powerup.com.au/~ozesl/quiz4.html

http://homepage.powerup.com.au/~ozesl/quiz5.html

http://homepage.powerup.com.au/~ozesl/quiz6.html

http://owl.english.purdue.edu/handouts/esl/eslphrasal.html

http://valenciaenglish.netfirms.com/phrasals00.htm

http://www.cse.unsw.edu.au/~billw/nlpdict.html#phrasalverb

http://www.english-zone.com/index.php

http://www.eslcafe.com/pv/

http://www.eslpartyland.com/quiz%20center/the%20big%20let%20down%202.htm

http://www.eslpartyland.com/quiz/get.htm

http://www.phrasalverbdemon.com/

http://www.sk.com.br/sk-twow.html

http://www.super-memory.com/sml/colls/pv.htm

http://www.usingenglish.com/reference/phrasal-verbs/
368

APPENDICES
369

APPENDIX 1
TRANSLATION TESTS

University of Westerm Sydney


College of Arts, School of Humanities and Languages
Locked Bag 1797
PENRITH SOUTH DC NSW 1797
___________________________________________________

Dear research participant,

Your are invited to participate in a research project conducted by Ali Yunis Aldahesh,

a PhD. student at the School of Humanities and Languages, College of Arts,

University of Western Sydney.

The present study attempts to explore the difficulties encountered by Arabic

translators and Arabic translation students when translating idiomatic English phrasal

verbs into Arabic, and to propose recommendations for professional translators,

translation students, translation teachers and lexicographers.

An idiomatic English phrasal verb can be broadly defined as a combination of a

simple verb with one of a range of particles (adverbial or prepositional). Such a

combination constitutes one unit of meaning, which is totally different from the

meanings of its components. Idiomatic English phrasal verbs, however, represent a

great deal of difficulty to English learners in general, and to Arabic translators and

Arabic translation students in particular due to their complexity and idiomaticity.


370

If you agree to participate, you will be given a translation test of one hundred English

sentences. You are required to translate into Arabic the underlined phrases only. The

test will take approximately one hour. If you decide not to participate, please return

the material in the postage paid envelope.

For the sake of confidentiality, the participants will be given code numbers to

replace their names, so that their real names will not appear in our records.

Furthermore, the information collected will be safely stored at the University of

Western Sydney. Only the researcher, Ali Yunis Aldahesh, will be permitted to use

the information. The information, however, will be destroyed after five years.

As a participant in this research, you may withdraw from the test at any time without

having to give us any reasons. There will be no penalty or disadvantage to participants

who decide to terminate or not participate in the research.

Your participation is greatly appreciated. You will receive a letter from the School of

Humanities and Languages acknowledging your valuable participation.

Note:
This study has been approved by the University of Western Sydney Human Research Ethics
Committee. If you have any complaints or reservations about the ethical conduct of this research, you
may contact the Ethics Committee through the Research Ethics Officers (tel.: 02 47 360883). Any
issues you raise will be treated in confidence and investigated fully, and you will be informed of the
outcome.

Yours sincerely,

Ali Yunis Aldahesh

Phone: 0402964764

Email: alialdahesh@hotmail.com
371

University of Westerm Sydney


College of Arts, School of Humanities and Languages
Locked Bag 1797
PENRITH SOUTH DC NSW 1797
___________________________________________________

Consent Form

I,………………………………………….., have read and understood the information

about the research, and any questions I have asked have been answered to my

satisfaction. I understand that my participation is voluntary and I agree to participate

in this research, knowing that I can withdraw at any time. I understand that the

information I provide will not be identified as coming from me in any way and that

the information I provide will be part of Ali Yunis Alahesh's thesis. I have been given

a copy of this form to keep.

Participant's Name: ………………………………………………….

Participant's Signature: ……………………… Date: ………………

Investigator's Name: Ali Yunis Aldahesh

Investigator's Signature: …………………….. Date: ………………

Note:
This study has been approved by the University of Western Sydney Human Research Ethics
Committee. If you have any complaints or reservations about the ethical conduct of this research, you
may contact the Ethics Committee through the Research Ethics Officers (tel.: 02 47 360883). Any
issues you raise will be treated in confidence and investigated fully, and you will be informed of the
outcome.
372

Part One
Please fill in the following form:

Please use a tick or a cross between the brackets, as appropriate.

• Name (optional):

• Gender:

Male ( )

Female ( )

• Age (optional):

• First language spoken at home:

• Period living in Australia:

• Qualification:

• Professional Translator:

Years of translation experience:

NAATI accreditation (if applicable):

Academic qualification in translation (if applicable):

• Translation Student:

Undergraduate ( )

Year one ( )

Year two ( )

Year three ( )

Postgraduate ( )

Graduate Diploma ( )

M.A ( )

PhD ( )
373

Part two:

Translation Tests

Note:

Since the study is about translation only, please confine your answers to Modern

Standard Arabic and avoid colloquial equivalents for the idiomatic English phrasal

verbs itemized in the translation tests.

Due to space limitation, short contextual information is provided between square

brackets following each item.

Please translate the underlined phrases only into Arabic

1. Don't keep talking about that letter. I know I shouldn't do it, but there's no need to

rub my nose in it. [Fiction novel - Conversation between friends]

….………………………………………………………………………………………

…….……………………………………………………………………………………

2. Father wasn't supposed to know about his birthday surprise – now you've let the cat

out of the bag. [Fiction novel - Conversation between two brothers]

….………………………………………………………………………………………

…….……………………………………………………………………………………

3. The 'Daily Mirror' says nobody can put back the clock and get rid of decimal

currency. [News report]

….………………………………………………………………………………………

…….……………………………………………………………………………………
374

4. When I realized that her husband was listening to what I was saying to her, I could

have bitten my tongue off. [Diary entry]

….………………………………………………………………………………………

…….……………………………………………………………………………………

5. Mary thinks too highly of herself. Someone should cut her down to size.

[Fiction novel - Conversation between classmates]

….………………………………………………………………………………………

….………………………………………………………………………………………

6. Slow down this instant, Percy! You're going to drive me around the bend!

[Fiction novel - Conversation between friends]

….………………………………………………………………………………………

….………………………………………………………………………………………

7. My eyes popped out when I saw Miriam. I hadn't seen her in years. She has aged a

lot since then. [Fiction novel - Conversation between friends]

….………………………………………………………………………………………

….………………………………………………………………………………………

8. The first time Max saw Edna he knew he was falling for her.

[Fiction novel - Conversation between a wife and husband]

….………………………………………………………………………………………

….………………………………………………………………………………………

9. I've been working for the last eight hours and now I'm fed up to my back teeth with

work. [Fiction novel - Conversation between fellow-workers]

….………………………………………………………………………………………

….………………………………………………………………………………………
375

10. When we gave Dorris a gift she got so carried away she began to cry. [Fiction

novel - Conversation between friends]

….………………………………………………………………………………………

….………………………………………………………………………………………

11. Poor Penelope. Mathematics has never been one of her favourite subjects. She

simply isn't able to comprehend it. She often complains that most of what her teacher

says goes over her head. [Fiction novel - Conversation between parents]

….………………………………………………………………………………………

….………………………………………………………………………………………

12. I'll have to pull up my socks if I'm going to finish my work today.

[Fiction novel - Conversation between fellow-workers]

….……………….………………………………………………………………………

….……….………………………………………………………………………………

13. I played when I should have studied. Now I'm stewing in my own juice trying to

pass my examinations. [Fiction novel - Conversation between classmates]

…...…...…………………………………………………………………………………

….....….…………………………………………………………………………………

14. She used to be so gentle but since being appointed supervisor Mrs. Wilson has

never stopped throwing her weight around in our office. [Fiction novel - Conversation

between fellow-workers]

…...….…….……………………………………………………………………………

…..….………..…………………………………………………………………………
376

15. George has been planning the exhibition for months but, because of the

workman's strike, the whole thing has fallen about his ears. [Article on an Art

exhibition]

…...…...…………………………………………………………………………………

…..………………………………………………………………………………………

16. If you think that I was responsible, then you are barking up the wrong tree: I am

completely innocent. [Fiction novel - Conversation between neighbours]

…...…...…………………………………………………………………………………

…….….…………………………………………………………………………………

17. Tom is top of the class, but there are two other students breathing down his neck.

[Parent-teacher interview]
…...….…..………………………………………………………………………………

…..………………………………………………………………………………………

18. He is not at all a practical person: perhaps having to earn his living will bring him

down to earth. [Fiction novel - Conversation between friends]

…...….…….……………………………………………………………………………

…..….…..………………………………………………………………………………

19. The young singer's performance brought the house down: the applause lasted for

about twenty minutes. [Art review - article]

…...….…….……………………………………………………………………………

…..…...…………………………………………………………………………………

20. If Mr. Brown finds out that you have been leaving the office early, he'll come

down on you like a ton of bricks. [Fiction novel - Conversation between fellow-

workers]

…...….………….………………………………………………………………………

…..….…..………………………………………………………………………………
377

21. He said he would have the whole house decorated by Christmas, but he fell down

on the job. [Movie script - Conversation between neighbours]

…...…...…………………………………………………………………………………

…….….…………………………………………………………………………………

22. We had a few minutes of polite conversation, but then he got down to brass tacks,

and asked me what salary I would want if I worked for him. [Fiction novel - Recount

on a job interview]

…...….…………….……………………………………………………………………

…..….…………..………………………………………………………………………

23. John has an idea for a new tennis club, but I don't know if it will ever get off the

ground. [Fiction novel - Conversation between friends]

…...….……….…………………………………………………………………………

……….…………………………………………………………………………………

24. Bill must have got up on the wrong side of the bed today: he has been very nasty

to me all day. [Fiction novel - Conversation between fellow-workers]

…...….…………………………………………………………………………………

…..….….………………………………………………………………………………

25. Mr. Brown keeps interfering with my work: I'm going to his office now to have it

out with him. [Fiction novel - Conversation between fellow-workers]

…...….…………………………………………………………………………………..

…..….………….………………………………………………………………………..

26. Jason is striving for his best. I expect he'll clock up quite a few successes in the

next year or two. [Fiction novel - Conversation between parents]

…...….………………………………………………………………………………….

…..………………………………………………………………………………………
378

27. I used to wolf my food down so quickly that I didn't have time to enjoy it.

[Women’s magazine article]

…...….…………..………………………………………………………………………

…...….………..…………………………………………………………………………

28. Why are the children all dolled up? Is someone important coming?

[Fiction novel - Conversation between parents]

…..….…….….…………………………………………………………………………

…...….…….……………………………………………………………………………

29. I was accustomed to the old train schedule. The new one has balled me up.

[Fiction novel - Conversation between commuters]

…...….…………………………………………………………………………………..

…..…...…..……………………………………………………………………………...

30. You can't skirt round the matter, you'll have to give the workers a satisfactory

answer. [Fiction novel - Advice from a factory manager to his deputy]

…...…..…………………………………………………………………………………

…..…...…………………………………………………………………………………

31. I'm shocked at Joseph's attitude. Did you notice the way he bossed his wife

around. [Fiction novel - Conversation between a wife and husband]

…...…...…………………………………………………………………………………

…..………………………………………………………………………………………

32. Don't bother asking Dean for support. He won't cotton on, he's too involved with

his own affairs. [Fiction novel - Advice from friend to friend]

…...….…………………………………………………………………………………..

…..…..…………………………………………………………………………………..
379

33. He was shopping around to see what support he could get for his proposals.

[Biography]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

34. I know you are so upset about that matter. Don’t bottle it up, speak out.

[Letter – union leader to members]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

35. Aquarius (Jan 21 - Feb 19)

The erratic behaviour of Uranus this week is likely to bring out the impish, impulsive

and unpredictable side of Aquarius. It's quite okay to walk on the wild side as long as

you recognise the limits beyond which it would be crazy to go. Don't let anyone egg

you on to make a complete fool of yourself. [Horoscope]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

36. Now, stop worrying about those exams. You've revised well, so relax. It's going to

be a walk-over. [Fiction novel - Mother's advice to her daughter]

….…………...…………………………………………………………………………..

….……….………...…………………………………………………………………….

37. If you launch a sales campaign now, when there's so little money about, it'll be a

total wash-out. [Email - Advice from a businessman to his colleague]

.….……………………………………………...………………………………………

….…..…….…………………………………………………..…………………………
380

38. I've been so worried about you. Where were you during the cave-in? Everyone has

been looking for you. [Fiction novel - Conversation between a father and his son]

….…………………………………………...…………………………………………

….……….………………………………..……………………………………………

39. Najaf is at the heart of a political stand-off.

US troops earlier drove through the centre of the holy city and used loud speakers to

order militiamen to leave and urge civilian to flee. [News report]

….…………..…………………….……………………………………………………

….……….……………………………..………………………………………………

40. My sister always had new clothes, while I had to make do with her cast-offs.

[Diary entry]

….……………...………………………………………………………………………

….………………………...……………………………………………………………

41. I want to propose a vote of thanks to Sergeant Wilson for an excellent mock-up of

a boat. [Fiction novel Announcement at a meeting]

….……….……………..………………………………………………………………

….……….………………………..……………………………………………………

42. Revised UN draft sets pullout date.

The United States would withdraw its troops from Iraq by the end of 2005 under a

new draft of a resolution before the United Nations Security Council. [News headline]

….………..………….…………………………………………………………………

…….……….…………………………………...………………………………………
381

43. I bought a house last week, and I requested three weeks as a cooling-off period.

[Question in a chat-room]

……….…………………………………...……………………………………………

……….…………………………………...……………………………………………

44. I went to the shop to buy some groceries. I found a sign on the display window

saying " The shop is closed due to the black out ". [Notice on a closed shop]

….……….…………………………………….………………………………………

….…………………………...…………………………………………………………

45. Managers concerned by a recent drop-off in sales. [News headline]

…..….……….…………………………………………………………………………

….……….……..………………………………………………………………………

46. This handout is a summary of Dr James' Method provided for information

purposes only without the assumption of a duty of care.

[Disclaimer - Medical brochure]

…..….……….…………………………………………………………………………

….……….……..………………………………………………………………………

47. The party brightened up when the pop group arrived. It would have been so boring

without them. [Fiction novel - Conversation between friends]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

48. You've had a long day today. Would you like to freshen up? The bathroom is

upstairs. [Fiction novel - Conversation between friends]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………
382

49. Don't put your prices too high or you'll frighten the customers off. [Advice from a

trade union official to a trader]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

50. Two smart, attractive, well-educated young law graduates, Sally and Edith, were

competing for a prestigious job. As part of the job interview, each was asked why she

wanted the job. Edith answered that she wanted to work for a firm with a reputation of

being concerned with truth and justice. When it was her turn, Sally simply opened her

purse, took out a rather thin wallet and laid it on the senior partner's desk. "I want to

fatten it up as fast as possible," she said. Sally got the job. [Joke]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

51. Jasmine cried out to her husband: Why fasten on me? It's not my fault.

[Fiction novel - Conversation between a wife and husband]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

52. I tried to calm him down after the bad news, but his emotions got the better of

him. [Fiction novel - Conversation between neighbours]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

53. They have just warmed up the same old plan, because it has always been

successful. [News report]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………
383

54. My hopes of a better job damped down after six years' waiting. [Diary entry]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

55. Harry thinks he is unique, but in fact he is just warming over ideas he's heard from

other people. [Fiction novel - Conversation between fellow-workers]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

56. The craftsman fined the wood down very gently. His boss told him to smoothen it

out as much as possible. [Fiction novel - Conversation between townsfolk]

…...….…………………………………………………………………………………

….……….……..………………………………………………………………………

57. It will not be easy for you to change your whole way of thinking, but I believe that

you have the strength to shape up to it.

[Advice in an article on well-being in a coffee table magazine]

….………………………………………………………………………………………

….………………………………………………………………………………………

58. Sandra knows that she can trust her parents and count on them to make her proud.

She always looks up to her parents. [Biography]

….………………………………………………………………………………………

….………………………………………………………………………………………

59. Michael is a well mannered young man. You shouldn’t look down on him just

because he left school at 16. [Fiction novel - Conversation between friends]

….………………………………………………………………………………………

….……………….…………...…………………………………………………………
384

60. I've never been able to come at Isabella since she took up with Tim.

[Fiction novel - Conversation between classmates]

….…………...…………………………………………………………………………

….……………...………………………………………………………………………

61. A popular question among Christians at this time of year is 'what are you giving

up for Lent?' [Article on Ash Wednesday]

….…………...…………………………………………………………………………

….…………….…..……………………………………………………………………

62. Bill and William became very successful in their political life since they made

away with their opponents. . [Background feature article - Conversation between

political figures]

...….……………………………………………………………………………………

...….……………………………………………………………………………………

63. Commencing with my very first speech as Chief Justice, on the day of my

swearing in, through several dozen other addresses and over two hundred speeches on

the occasion of the admission of legal practitioners, I have emphasised the

significance of the longevity of our institutions of parliamentary democracy and of the

rule of law. Many of you have had to put up with this more than once. You should

brace yourselves.

[Transcription of a public speech in the Opening of Law Term Dinner]

...….……………………………………………………………………………………

..….………………….…………………………………………………………………
385

64. The police are confident that the gang who carried out the robbery will be caught.

The Guardian therefore, has this headline: 'Net closes in on £50m robbers'.

[News headline]

...….……………………………………………………………………………………

..….…………………………………………………………………………………….

65. Robin is going all out to pass his final examination. He really wants to get his

Bachelor's degree. [Fiction novel - Conversation between parents]

...….……………………………………………………………………………………

..….…………………………………………………………………………………….

66. Vickie: Jack, can you do the dishes tonight? I'm very tired.

Jack: Can't we just put them in the dish-washer? I've had a very bad day, too! I just

want to kick back and relax.

Vickie: Hey! Just because you had a bad day today doesn't give you permission to

take it out on me! [Fiction novel - Conversation between a wife and husband]

...….……………………………………………………………………………………

..….……….……………………………………………………………………………

67. I have a tight schedule tomorrow. Can you fill in for me at the meeting?

[Fiction novel - Conversation between fellow-workers]

..….……………………………………………………………………………………

..….……………………………………………………………………………………

68. Priscilla: Rich, can you come downstairs for a minute right now?

Richard: I'm sorry I can't. I'm tied up with something urgent at the moment.

[Fiction novel - Conversation between a wife and husband]

...….……………………………………………………………………………………

…….……………………………………………………………………………………
386

69. The death penalty has been done away with in many European countries. However,

it's still being carried out in many other countries. [Article on death penalty]

...….………….…………………………………………………………………………

..….…………..…………………………………………………………………………

70. Judy: you look like you're fuming. What's going on?

Andy: The boss is being unreasonable! He flared up at me, making a mountain out of

molehill! [Fiction novel - Conversation between fellow-workers]

...….……………………………………………………………………………………

..….………………….…………………………………………………………………

71. I hear you're always getting into trouble. Keep out of what doesn't concern you.

Curiosity killed the cat.

[Fiction novel - Advice from a school teacher to a student]

…………………………………………………………………………………………

…………………………………………………………………………………………

72. If you want to get ahead in this world, you've to work, work and work.

[Fiction novel - Advice from parents to their children]

…………………………………………………………………………………………

…………………………………………………………………………………………

73. Kids have gone overweight from eating too much McDonalds. They need regular

exercise to use up sugar. [Advice from a doctor to parents in an article in a coffee

table magazine]

…………………………………………………………………………………………

…………………………………………………………………………………………
387

74. Sarah: I've managed to learn typing on the computer, but I still haven't got the

hang of it.

Sally: You have to practise a lot to build up skills.

[Fiction novel - Conversation between two sisters]

…………………………………………………………………………………………

…………………………………………………………………………………………

75. They finally realized that there is an urgent need for the health promotion

program. They will have a special meeting to work it out. [News report]

………………………………………………………………………………………….

…………………………………………………………………………………………

76. Accused: I wasn't at that market last night, and I don't know the person you are

talking about.

Prosecutor: We are not going into that again, are we? [A cross examination in court]

…………………………………………………………………………………………

…………………………………………………………………………………………

77. You are making a big fuss out of nothing. Why don't you two kiss and make up?

[Fiction novel - Conversation between classmates]

…………………………………………………………………………………………

…………………………………………………………………………………………

78. Men are like computers: Hard to figure out and never have enough memory.

[Joke]

…...….…………………………………………………………………………………..

….……….……..………………………………………………………………...………
388

79. Mr. Crutcher is working in a Starbucks coffee shop. He says that he is now a

reformed character. He says he's not back to his old tricks and uses a nice phrase 'I've

cleaned up my act.' [News report]

…………………………………………………………………………………………

…………………………………………………………………………………………

80. This is a nice tune and I think it'll catch on quickly.

[Fiction novel - Conversation between friends]

...….……………………………………………………………………………………..

..….………………………………………………………………………………………

81. My neighbour had a bad car accident yesterday. After the accident he blacked out

and couldn't remember what happened. [Diary entry]

...….……………………………………………………………………………………

..…..……………………………………………………………………………………

82. Five or six years ago I'd put away a big piece of meat most evenings. Now I just

eat a bit of chicken or fish. [An advertisement for a weight loss product]

...….………………..……………………………………………………………………

..…………………………………………………………………………………………

83. I was busy paying lots of bills this morning, and I ran into my wife at the post

office. [Diary entry]

..….……………………………………………………………………………………...

..….………………………………………………………………….…………………..
389

84. The salesman finds it easy to take in old ladies and persuade them to give him

their money. [Fiction novel - Conversation between shoppers]

...….……………………………………………………………………………………..

..….………….…………………………………………………………………………..

85. I had a bad pain in my stomach last night. It was severe to begin with, but it soon

passed away. [Fiction novel - Conversation between a patient and a doctor]

...….……………………………………………………………………………………..

..….……………………………………………………………………………………...

86. I must warn you that anything you may say will be taken down and used in court as

evidence against you. [Police interview]

...….……………………………………………………………………………………..

..….…………………..………………………………………………………………….

87. You may not like our flowers, but they will grow on you in the end.

[A note written on a flower stall]

...….……………………………………………………………………………………..

..….………………………….…………………………………………………………..

88. One woman says that her boss is horrible and he stresses her out. She says that she

deals with the stress by chanting. Another woman says her partner stresses her out

because he doesn't tidy up. And she copes with it by having a relaxing bath!

[Article on stress]

…...….…………………………………………………………………………………

…..….……………………………………………………………………………….…
390

89. It's spring break in Florida and students are chilling out at the beach and partying

late into the night. [News report]

…...….…………………………………………………………………………………

…..….………………………………………………………………………………….

90. One day, I went over to my Uncle Tony's house and didn't get a chance to do my

Maths homework. The teacher told me off for not doing it the day after.

[Fiction novel - Conversation between a daughter and her parents]

…...….…………………………………………………………………………………..

…..….……………………………………………………………………………….…..

91. Chris: you want to come out tonight?

Adam: I've had enough, I'm partied out. [Fiction novel - Conversation between

friends]

…...….…………………………………………………………………………………

…..….………………………………………………………………………………….

92. Funk up your mobile with cool screen savers [An internet advertisement for

mobile phone screen savers]

…...….…………………………………………………………………………………

…..….…………………………………………………………………………………

93. Latham sexed up Iraq brief.

Mark Latham's credibility as opposition leader faced its gravest test last night after

John Howard effectively accused him of misleading parliament over top-secret

intelligence briefings on Iraq. [News headline]

…...….…………………………………………………………………………………

…...….…………………………………………………………………………………
391

94. Waitress: Are you ready to order now?

Maria: Yes. I'll have some salad, roast beef, and mashed potatoes.

Waitress: How do you want the beef? Rare, medium, or well-done?

Maria: Well-done. And go easy on the salt, please.

[Fiction novel - Conversation at a restaurant]

…...….…………………………………………………………………………………..

…...….……………………………………………………………………………………

95. Bill: Where did you get this article from?

Jack: I googled around for three hours to get the site on which it was published.

[Fiction novel - Conversation between friends]

…...….…………………………………………………………………………………..

…...….……………………………………………………………………………………

96. I've spent four hours doing my late night shopping. I'm really shopped out.

[Fiction novel - Conversation between a wife and husband]

…...….…………………………………………………………………………………..

…...….……………………………………………………………………………………

97. How about we get the students of floral art from the Tech to fang up a few

arrangements using roses and jasmine? [Fiction novel – Conversation between school

teachers]

…..….…………………………………………………………………………………

…..….…………………………………………………………………………………
392

98. When I mentioned it, they all looked at me strangely and said 'What are we going to

do?' and I said we're going to talk, tell secrets, dance, dag out.

[Fiction novel - Conversation between friends]

…...….…………………………………………………………………………………

…...….…………………………………………………………………………………

99. Human beings run into trouble when arterial walls guck up with cholesterol and

blood can't easily flow. [Medical brochure]

…...….………………………………….………………………………………………

…..….……………………………….………………………………………………….

100. We had built up this cult following while being an independent band, but as soon

as we had a commercial record we were slagged down as cashing in on a theme.

[Diary entry]

…...….…………………………….……………………………………………………

…..….……………………………………..……………………………………………
393

APPENDIX 2
SUGGESTED ARABIC FUNCTIONAL-PRAGMATIC
EQUIVALENTS
IEPVs Suggested Arabic functional-
pragmatic equivalents
1 Don't keep talking about that letter. I 7O\ 7ً3L‫ دا‬%P,ّ-*} ‫ن‬9 ‫داع‬M

know I shouldn't do it, but there's no

need to rub my nose in it. [Fiction

novel - Conversation between friends]

2 Father wasn't supposed to know about ,Y+‫ ا‬x_N‫ \ آ‬,Y+‫ ا‬xQ^8 \ ,Y+‫ ا‬x-N8‫[ أ‬ƒ+

his birthday surprise – now you've let

the cat out of the bag. [Fiction novel -

Conversation between two brothers]

3 The 'Daily Mirror' says nobody can ‫راء‬X+‫ ا‬p+‫ ا‬5E7Y+‫رب ا‬7ƒE [-*1

put back the clock and get rid of

decimal currency. [News report]

4 When I realized that her husband was x-.3} \ ‘F)w 73\ X_}‫ أن أ‬K(w %P7Y+ ]َ {
ِ wُ X+ x-.3}

listening to what I was saying to her, I %.F*)F\‫رص وا‬9‫ ا‬xƒNP‫ ا‬X+

could have bitten my tongue off.

[Diary entry]

5 Mary thinks too highly of herself. %&&&8 7O*^&&&1 \ %&&&ƒ-ƒQ+‫ ا‬7&&&O3IZ 7&&&O1,1 \ 7&&&O3IQ1

Someone should cut her down to size. 7‫[ره‬w ‫ف‬,*} 7O)*I1 \ %ƒ-ƒQ+‫ ا‬7O3IZ

[Fiction novel - Conversation between

classmates]

6 Slow down this instant, Percy! You're %..IFH \ %\‫ا‬X` ,ّ-{FH \ %\‫ا‬X` [ƒ8•H

going to drive me around the bend!


394

[Fiction novel - Conversation between

friends]

7 My eyes popped out when I saw %


ّ .-E ‫ أ`[ّق‬4+

Miriam. I hadn't seen her in years. She

has aged a lot since then. [Fiction

novel - Conversation between friends]

8 The first time Max saw Edna he knew %&&8 ]&&w‫ \ و‬7&&O/‫ا‬,G ,-&&H‫&&] أ‬w‫ \ و‬7O‫آ‬7(&&o %&&8 ‚&&)E

he was falling for her. [Fiction novel - 7O(Z

Conversation between a wife and

husband]

9 I've been working for the last eight p&&&+‫ إ‬K)Y&&&F1 K&&&)3+‫ \ ا‬K&&&3*+‫ ا‬0&&&/ ‫&&&•م‬H 7&&&31‫ أ‬x3—&&&H

hours and now I'm fed up to my back x&&))/ \ K&&3*+‫&&‚ ا‬-W‫&&[ أ‬E‫ أ‬4&&+ \ K&&3*+‫ ا‬0&&/ %&&Z‫رو‬

teeth with work. 4&+ \ K&3*+‫ ا‬0&/ K&-'+‫ ا‬%&\ •&_W \ K&3*+‫ ا‬0&/ 517Ž)+

[Fiction novel - Conversation between [1[N&&+‫ ا‬K&&)3+7\ x(&&`‫ \ أ‬K&3*+‫ ا‬p&&)E ,(&&` %&&+ [&*1

fellow-workers] K3*+‫ ا‬0/

10 When we gave Dorris a gift she got so ‫ء‬7'(+7\ xNO„‫ أ‬7OP‫[ أ‬Z p+‫ًا ا‬,-…‫ت آ‬,ˆ•}

carried away she began to cry. [Fiction

novel - Conversation between friends]

11 Poor Penelope. Mathematics has never \ ‫ام‬,&'+‫ور ا‬,&/ 7O.‫ \&~ه‬,&31 \ 5&8Š+‫ ا‬%8 ‫ش‬,W97‫آ‬

been one of her favourite subjects. She 7O‫ى ادراآ‬XFY/ ‫ق‬X8 \ ‘3O8 7O-)E |*U1

simply isn't able to comprehend it. She

often complains that most of what her

teacher says goes over her head.

[Fiction novel - Conversation between

parents]
395

12 I'll have to pull up my socks if I'm %)E \ %*H‫ و‬%8 7/ K‫ آ‬K3E‫[ي \ أ‬E7H 0E ,ّ3o‫أ‬

going to finish my work today. ‫[ي‬O„ ‫رى‬7Uw ‫أن أ\~ل‬

[Fiction novel - Conversation between

fellow-workers]

13 I played when I should have studied. \ %&)3E ¥L7&FP K&ّ3Q}‫&[اي \ أ‬1 ‘&F)*8 7&/ 5I-FP %.„‫أ‬

Now I'm stewing in my own juice %+73‫ر اه‬7.\ ‫ي‬XF‫أآ‬

trying to pass my examinations.

[Fiction novel - Conversation between

classmates]

14 She used to be so gentle but since \ %‫ه‬7&.+‫ ا‬,&/¤‫ ا‬7OP•‫ّف وآ‬,UF+‫ ا‬0E ‫• ا\[ا‬wXF} 4+

being appointed supervisor Mrs. \ ٍ 7&(F\ 7O(U&./ ‫ل‬73*F&H‫ ا‬0&E\ ,('F\ ‫ف‬,UF+‫ ا‬0E

Wilson has never stopped throwing 5H,{Ž\ \ ‫ء‬2*FH7\

her weight around in our office.

[Fiction novel - Conversation between

fellow-workers]

15 George has been planning the ‫[ى‬H \ ‫ح‬71,+‫ أدراج ا‬x(‫{{‘ ذه‬y

exhibition for months but, because of

the workman's strike, the whole thing

has fallen about his ears. [Article on

an Art exhibition]

16 If you think that I was responsible, ‹{n/ xP•8 ‫ل \ اذن‬Xƒ} 7/ %*} M xP•8 ‫اذن‬

then you are barking up the wrong

tree: I am completely innocent.

[Fiction novel - Conversation between

neighbours]
396

17 Tom is top of the class, but there are \ ‫ ًة‬,&o7(/ ‘&_)y \ 7ً /73} ,ˆ‫ ا‬%8 0-(+7W ‫ك‬7.‫ ه‬0'+‫و‬

two other students breathing down his ‫ا‬XFY/ 0/ ‫ن‬7(1,w \ ‫ا‬XFY/ p+‫ ا‬2U1 ‫دان أن‬7'1

neck. [Parent-teacher interview]

18 He is not at all a practical person: \ ]&&&w‫ا‬X+‫ أرض ا‬p&&&+‫[ ا‬-*-&&&H\ ‘&&&3IZ p&&&+‫[ إ‬-*-&&&H

perhaps having to earn his living will ً 7-)3E ‘)*I-H \ 7-*w‫*)‘ وا‬I-H

bring him down to earth. [Fiction

novel - Conversation between friends]

19 The young singer's performance ]&&F/‫ب \ أ‬,&&{+‫ة ا‬XN&&P 0&&/ ‘\‫ا‬X&&` ‫ر‬X&&O3I+‫&&[ ا‬ƒ8‫أ‬

brought the house down: the applause |&&&O+‫ب \ أ‬7&&&IEJ‫ ا‬5&&&17G ‫ل‬7&&&P \ 5&&&17Ž)+ ‫ر‬X&&&O3I+‫ا‬

lasted for about twenty minutes. [Art ,-‫ه‬73I+‫• ا‬W‫ا‬XE

review - article]

20 If Mr. Brown finds out that you have \ §-\X&} 7&3ّ1‫ أ‬Snً\X-&H \ S-)E ‘(^G ‫م‬7„ |U-H

been leaving the office early, he'll ‫ة‬XYw K'\ 7ً _-.*} S-)E ‫ل‬7O.-H

come down on you like a ton of

bricks. [Fiction novel - Conversation

between fellow-workers]

21 He said he would have the whole S+‫ز ذ‬7IP‫ ا‬%8 KN8 ‘.'+‫ \ و‬S+‫ ذ‬%8 ‚_y‫‘ أ‬.'+‫و‬

house decorated by Christmas, but he

fell down on the job. [Movie script -

Conversation between neighbours]

22 We had a few minutes of polite 5&&{ƒ.+‫ ا‬p&&+‫ق ا‬,&&{}\ ‫ع‬X&&RX3+‫ `&&)| ا‬%&&8 K&&y‫د‬

conversation, but then he got down to [I+‫ ا‬%8 ‫ع \ \[أ‬XRX3+‫ ا‬%8 5/7O+‫ا‬

brass tacks, and asked me what salary

I would want if I worked for him.

[Fiction novel - Recount on a job


397

interview]

23 John has an idea for a new tennis club, \ •I.F&&H ‫ة‬,&&'_+‫ ا‬x&&P7‫ إذا آ‬7&&3-8 ‫ف‬,&&E‫ أ‬M %&&..'+‫و‬

but I don't know if it will ever get off ~-_.F+‫ ا‬Šّ-Z Ky[FH \ ‫ر‬X.+‫ى ا‬,FH

the ground. [Fiction novel -

Conversation between friends]

24 Bill must have got up on the wrong ‫(&[و أن‬1\ ‫م‬X-+‫ا ه~ا ا‬,'*F/ K-\ ‫اج‬Š/ ‫ن‬X'1 ‫\[ أن‬M

side of the bed today: he has been very ‫داوي‬XH ‘„‫ا‬Š/

nasty to me all day. [Fiction novel -

Conversation between fellow-

workers]

25 Mr. Brown keeps interfering with my ‫وف‬,&&Q+‫ ا‬p&&)E ‫ط‬7&&ƒ.+‫&&] ا‬R9 \ ‘&&*/ ,&&/9‫ ا‬4Y&&Z9

work: I'm going to his office now to 7O\7UP %8 ‫ر‬X/9‫] ا‬R9 \ ‘*/

have it out with him. [Fiction novel -

Conversation between fellow-

workers]

26 Jason is striving for his best. I expect ‫ز‬,Q-&&&&H \ ‫&&&&[ة‬1[E ‫ت‬7&&&&Z7IP ‚ƒQ-&&&&H ‘&&&&P‫&&&&] أ‬wX}‫أ‬

he'll clock up quite a few successes in ‫[ة‬1[E ‫ت‬7Z7IP

the next year or two. [Fiction novel -

Conversation between parents]

27 I used to wolf my food down so K&‫ّدت أن أأآ‬X&*} \ 5E,Y&\ %/7*W 4OF+‫[ت أن أ‬FE‫ا‬

quickly that I didn't have time to enjoy 5E,Y\

it. [Women’s magazine article]

28 Why are the children all dolled up? Is \ ‫ف؟‬7&8‫ ز‬p&+‫ن إ‬X&(‫ ذاه‬4OP•&‫ن وآ‬XƒP•&F/ ‫ل‬7_W9‫ ا‬4+

someone important coming? [Fiction ‫؟‬4O1[+ 7/ K^8‫ل أ‬7_W9‫ ا‬‰()1 ‫ذا‬73+

novel - Conversation between parents]


398

29 I was accustomed to the old train %P,¢-Z \ %.'\‫[ أر‬w [1[I+‫ر ا‬7{ƒ+‫„[ول ا‬

schedule. The new one has balled me

up. [Fiction novel - Conversation

between commuters]

30 You can't skirt round the matter, you'll ‫ع‬XRX3+‫| ا‬.IF}M \ ,/9‫ ا‬0/ ‫ّب‬,OF} M

have to give the workers a satisfactory

answer. [Fiction novel - Advice from a

factory manager to his deputy]

31 I'm shocked at Joseph's attitude. Did ‫‘؟‬F„‫ زو‬p)E 7O\ ,ّ/•F1 %F+‫ ا‬5ƒ1,{+‫ ا‬x€ZM K‫ه‬

you notice the way he bossed his wife ™ّ)YF\ ‘F„‫] زو‬/ K/7*F1 •-/ x€ZM K‫ه‬

around. [Fiction novel - Conversation

between a wife and husband]

32 Don't bother asking Dean for support. 4ّO_F1 0+ \ ,/9‫| ا‬EXFY1 0+

He won't cotton on, he's too involved

with his own affairs. [Fiction novel -

Advice from friend to friend]

33 He was shopping around to see what ]){FY&&&1 \ ‫ال‬X&&&Z9‫ّ&&&[ ا‬ƒ_F1 \ ‫ر‬X&&&/9‫ّى ا‬,&&QF1 ‫ن‬7&&&‫آ‬

support he could get for his proposals. “(.+‫ ا‬‰I1 \ ‫راء‬¤‫ا‬

[Biography]

34 I know you are so upset about that 7ً F(‫ آ‬4‫آ‬7_‫ \ آ‬XF('}M \ 4'\X)w %8 ,/9‫ا ا‬X3F'} M

matter. Don’t bottle it up, speak out.

[Letter – union leader to members]

35 ‫ّك‬Š_FY1 \ ‫ك‬,-…FY1 ‫[ًا‬Z‫ }[ع أ‬M


Aquarius (Jan 21 - Feb 19)

The erratic behaviour of Uranus this


399

week is likely to bring out the impish,

impulsive and unpredictable side of

Aquarius. It's quite okay to walk on

the wild side as long as you recognise

the limits beyond which it would be

crazy to go. Don't let anyone egg you

on to make a complete fool of

yourself. [Horoscope]

36 Now, stop worrying about those 2


ً O&&&H \ ‫ء‬7&&/ 5\,&&o 0&&/ KO&&H‫ن أ‬7&&QF/M‫ن ا‬X'-&&H

exams. You've revised well, so relax. 5+XOY+‫ ا‬pOF.3\ \ 517Ž)+

It's going to be a walk-over.

[Fiction novel - Mother's advice to her

daughter]

37 If you launch a sales campaign now, xwX)+ 5*-^/ ‫ن‬X'FH \ 7*1‫ ذر‬2N8 5)3Q+‫ ا‬KN_FH

when there's so little money about, it'll ‫ل‬73+‫وا‬

be a total wash-out. [Email - Advice

from a businessman to his colleague]

38 I've been so worried about you. Where ‫ر؟‬7-OPM‫ء ا‬7.ˆ‫ أ‬x.‫ آ‬01‫أ‬

were you during the cave-in? Everyone

has been looking for you.

[Fiction novel - Conversation between a

father and his son]

39 Najaf is at the heart of a political |&)w %&8 \ 5-H7-Y+‫ ا‬5_`7*+‫ ا‬0-E %8 •I.+‫ ا‬,(F*}

stand-off. 5-H7-Y+‫ ا‬5*\‫و‬Š+‫ا‬

US troops earlier drove through the


400

centre of the holy city and used loud

speakers to order militiamen to leave

and urge civilian to flee. [News report]

40 My sister always had new clothes, 7OY&&\2/ ‫ر}&&[اء‬7\ %&&_F‫ أن أآ‬%&&)E |&&I1 ‫ن‬7&&‫ آ‬7&&3.-\

while I had to make do with her cast- 7O+73H‫ \ أ‬5-+7(+‫ا‬

offs. [Diary entry]

41 I want to propose a vote of thanks to ‘*.` ‫~ي‬+‫ز ا‬7F33+‫رب ا‬7ƒ+‫ ا‬4ّYI/ |(Y\

Sergeant Wilson for an excellent

mock-up of a boat. [Fiction novel -

Announcement at a meeting]

42 Revised UN draft sets pullout date. ‫&&[ة‬QF3+‫ ا‬4&&/9‫ار ا‬,&&w ‫دة‬XY&&/ ‫ &[ّد‬Q} \ 0ّ-& *} \ ]^&&}

The United States would withdraw its 5&+ّ[*3+‫دة ا‬XY&3+‫’ ا‬.&} \ ‫ب‬7QYPª+ 7n1‫ر‬7} 5+[*3+‫ا‬

troops from Iraq by the end of 2005 ‫ب‬7QYPM‫§ ا‬1‫ر‬7} p)E ‫[ة‬QF3+‫ ا‬4/9‫ار ا‬,ƒ+

under a new draft of a resolution

before the United Nations Security

Council. [News headline]

43 I bought a house last week, and I ‫اء‬,N+‫ ا‬0E ‫[ول‬E 5)O/ \ ,-'_} ‫ة‬,F8 \ ‫ل‬7O/‫ة إ‬,F8

requested three weeks as a cooling-off

period. [Question in a chat-room]

44 I went to the shop to buy some %L7\,O'+‫ر ا‬7-F+‫ع ا‬7{ƒP‫(| ا‬Y\ ‚)Ž/ KQ3+‫ا‬

groceries. I found a sign on the display

window saying " The shop is closed

due to the black out ".

[Notice on a shop].

45 Managers concerned by a recent drop- %8 ,-y9‫ض ا‬7_nPM‫(| ا‬Y\ ‫ن‬Xƒ)w ‫[راء‬3+‫ا‬


401

off in sales. [News headline] ]„‫ا‬,F+‫ط \ ا‬X(O+‫ت \ ا‬7*-(3+‫ا‬

46 This handout is a summary of Dr ‫ر‬XF‫&&[آ‬+‫ ا‬5&&ƒ1,{+ ’&&n


ّ )/ 0&&E ‫رة‬7&&(E ‫ة‬,N&&.+‫ه&&~ ا‬

James' Method provided for ‰3-„

information purposes only without the

assumption of a duty of care.

[Disclaimer - Medical brochure]

47 The party brightened up when the pop 5)_Q+‫ ا‬%8 ‫ة‬7-Q+‫ ا‬x\‫د‬

group arrived. It would have been so

boring without them. [Fiction novel -

Conversation between friends]

48 You've had a long day today. Would |&G,} K&‫م؟ \ ه‬73QF&HM7\ SY_P ˜*.} ‫د أن‬X} K‫ه‬

you like to freshen up? The bathroom ‫ل‬7Y&&&&&&FGM‫ش؟ ا‬7&&&&&&*FPM‫ وا‬SW7N&&&&&&P [&&&&&&1[I} %&&&&&&8

is upstairs. [Fiction novel - ‫ش؟‬7*FPM‫وا‬

Conversation between friends]

49 Don't put your prices too high or you'll 0L7\Š+‫ ا‬S./ ,_-H M‫وإ‬

frighten the customers off. [Advice

from a trade union official to a trader]

50 Two smart, attractive, well-educated ‫&[ أن‬1‫ \ أر‬0&'31 7&/ ‫ع‬,&H•\ ‫د‬Xƒ.+7\ 7‫«ه‬/‫[ أن أ‬1‫أر‬

young law graduates, Sally and Edith, p&&&FZ 7&&&‫«ه‬/‫&&[ أن أ‬1‫ \ أر‬0&&&'31 7&&/ ‫ع‬,&&&H•\ %&&.FG‫أ‬

were competing for a prestigious job. 0'31 7/ ‫ع‬,H•\ 7‫ه‬,y†

As part of the job interview, each was

asked why she wanted the job. Edith

answered that she wanted to work for

a firm with a reputation of being

concerned with truth and justice.


402

When it was her turn, Sally simply

opened her purse, took out a rather

thin wallet and laid it on the senior

partner's desk. "I want to fatten it up

as fast as possible," she said. Sally got

the job. [Joke]

51 Jasmine cried out to her husband: %/X+ 0E •‫؟ آ‬%)E ‫م‬X)+‫ذا }^] ا‬73+

Why fasten on me? It's not my fault.

[Fiction novel - Conversation between

a wife and husband]

52 I tried to calm him down after the bad ‘.E •ّ_y‫\‘ \ أ‬7UE‫ أ‬0/ ‫ أن أه[ّئ‬x+‫و‬7Z

news, but his emotions got the better

of him. [Fiction novel - Conversation

between neighbours]

53 They have just warmed up the same ‫ا‬X&-Z‫&[ أ‬ƒ+ \ 5&31[ƒ+‫ ا‬5&{n+‫ ا‬‰_.&\ K3*+‫دوا ا‬7E‫[ أ‬ƒ+

old plan, because it has always been 531[ƒ+‫ ا‬5{n+‫ ا‬‰_P

successful. [News report]

54 My hopes of a better job damped p&)E ‫ل‬XU&Q+‫ ا‬%8 %+7/† ‫ \ }([ّدت‬x(y \ x+‫ء‬7^}

down after six years' waiting. ‫ح‬71,+‫ أدراج ا‬%+7/† x(‫ \ ذه‬K^8‫ أ‬K3E

[Diary entry]

55 Harry thinks he is unique, but in fact ‫ر‬7&&'89‫ ا‬‰(&&Fƒ1 \ ‫دد‬,&&1 \ [&&-*1 \ ,&&FI1 \ ‫ر‬,&&'1 ‘&&P‫ا‬

he is just warming over ideas he's ,-GM 01,y¤‫ ا‬0/ 7O*3H %F+‫ا‬

heard from other people. [Fiction

novel - Conversation between fellow-

workers]
403

56 The craftsman fined the wood down |Nn+‫] ا‬P7U+‫ ا‬4ّ*P \ ‫د‬,\ \ Kƒ`

very gently. His boss told him to

smoothen it out as much as possible.

[Fiction novel - Conversation between

townsfolk]

57 It will not be easy for you to change 4-3U}‫ و‬5E7IN\ ,/9‫ا„‘ ا‬X} ‫أن‬

your whole way of thinking, but I

believe that you have the strength to

shape up to it. [Advice in an article on

well-being in a coffee table magazine]

58 Sandra knows that she can trust her 7O1[&&+‫ وا‬,&&F(*} \ p&&)E9‫ ا‬7&&O)…/ 7&&3L‫ دا‬7&&3‫ ه‬7‫&&[اه‬+‫وا‬

parents and count on them to make her 7O+ ‫[وة‬w 7ً3L‫دا‬

proud. She always looks up to her

parents. [Biography]

59 Michael is a well mannered young ’ƒ.} \ ,ƒFQ}\‘1‫در‬Š}\ ‘-)*‫ى‬+7*F} ‫ أن‬%Ž(.1M

man. You shouldn’t look down on him ‫[ر‬w 0/ ™Q} \ ‫[ر‬w 0/

just because he left school at 16.

[Fiction novel - Conversation between

friends]

60 I've never been able to come at \ 4-&&&} ]&&&/ 5&&&w2E x&&&/7w‫ \ أ‬7ً ƒ1[&&&` 4-&&&} ‫&&&~ت‬n}‫ا‬

Isabella since she took up with Tim. 4-F\ x{(}‫ \ ار‬4-} ‫ت‬,o7E

[Fiction novel - Conversation between

classmates]

61 A popular question among Christians ‫م؟‬7-U+‫ة ا‬,F8 ‫ل‬2y ‫ن‬X*.F3-H ‫ء‬%o ‫ أي‬0E

at this time of year is 'what are you


404

giving up for Lent?'

[Article on Ash Wednesday]

62 Bill and William became very 4O/XUy 0/ 7Uّ)n}

successful in their political life since

they made away with their opponents.

[Background feature article -

Conversation between political

figures]

63 Commencing with my very first \ ,&/9‫ ه&~ا ا‬K&ّ3QF1 ‫ أن‬4'.&/ ,&-…'+‫ ا‬p)E 4ّFQ} [w‫و‬

speech as Chief Justice, on the day of ,/9‫ ه~ا ا‬p)E ,(U1 ‫أن‬

my swearing in, through several dozen

other addresses and over two hundred

speeches on the occasion of the

admission of legal practitioners, I have

emphasised the significance of the

longevity of our institutions of

parliamentary democracy and of the

rule of law. Many of you have had to

put up with this more than once. You

should brace yourselves.

[Transcription of a public speech in

the Opening of Law Term Dinner]

64 The police are confident that the gang ‫ﻴﻀﻴﻕ ﻋﻠﻰ ﺴﺎﺭﻗﻲ ﺍﻝﺨﻤﺴﺔ ﻤﻼﻴﻴﻥ ﺠﻴﻨﻴﻪ ﺍﻝﺨﻨﺎﻕ‬

who carried out the robbery will be


‫\ ﺍﻝﺩﺍﺌﺭﺓ ﺘﻀﻴﻕ \ ﺍﻝﻁﻭﻕ ﻴـﺸﺘﺩ \ ﻴـﺩ ﺍﻝﻌﺩﺍﻝـﺔ‬
caught. The Guardian therefore, has
405

this headline: 'Net closes in on £50m ‫ﺃﻭﺸﻜﺕ ﺃﻥ ﺘﻁﺒﻕ ﻋﻠـﻰ \ ﺒـﺩﺃ ﺍﻝﻌـﺩ ﺍﻝﺘﻨـﺎﺯﻝﻲ‬

robbers'. [News headline]


‫ ﻤﻼﻴﻴﻥ ﺠﻴﻨﻴﻪ‬5Y3n+‫ ا‬%w‫ر‬7Y\ ‫ك‬7Y/‫ﻝﻺ‬

65 Robin is going all out to pass his final \ %L7&&O.+‫&&‘ ا‬P7QF/‫ ا‬%&&8 ‫ح‬7&&I.)+ ‘*&&H‫ و‬K&&3*1 0&&\‫رو‬

examination. He really wants to get ‘E7{FY/ %87/ K3*1 \ ‫[ة‬O„ ‫رى‬7Uw ‫(~ل‬1

his Bachelor's degree. [Fiction novel -

Conversation between parents]

66 Vickie: Jack, can you do the dishes %\ S(^G 0E ‰_.} ‫أن‬

tonight? I'm very tired.

Jack: Can't we just put them in the

dish-washer? I've had a very bad day,

too! I just want to kick back and

relax.

Vickie: Hey! Just because you had a

bad day today doesn't give you

permission to take it out on me!

[Fiction novel - Conversation between

a wife and husband]

67 I have a tight schedule tomorrow. Can ‫ب‬X.} ‫ع؟ أن‬73F„M‫ ا‬%8 %)Q/ KQ} ‫ أن‬SP7'/7\ K‫ه‬

you fill in for me at the meeting? ‫ع؟‬73F„M‫ ا‬%8 %.E

[Fiction novel - Conversation between

fellow-workers]

68 Priscilla: Rich, can you come •)/ ,/•\ ‫ل‬XŽN/ ‫ن‬¤‫ ا‬7P‫أ‬

downstairs for a minute right now?

Richard: I'm sorry I can't. I'm tied up


406

with something urgent at the moment.

[Fiction novel - Conversation between

a wife and husband]

69 The death penalty has been done away ‫[ام‬EJ‫ ا‬5\XƒE ‫ء‬7Ž+‫ ا‬4} [ƒ+

with in many European countries.

However, it's still being carried out in

many other countries.

[Article on death penalty]

70 Judy: you look like you're fuming. %)E 7ً(R7G ,I_P‫ا‬

What's going on?

Andy: The boss is being

unreasonable! He flared up at me,

making a mountain out of molehill!

[Fiction novel - Conversation between

fellow-workers]

71 I hear you're always getting into S-.*1 M 73-8 Kّy[F} M

trouble. Keep out of what doesn't

concern you. Curiosity killed the cat.

[Fiction novel - Advice from a school

teacher to a student]

72 If you want to get ahead in this world, 4}‫ \ إذا أرد‬4+7*+‫ ه~ا ا‬%8 4'P•o X)*1 ‫ أن‬4F—o ‫إذا‬

you've to work, work and work. 7ً/[w ‫ا‬X^3} ‫ \ أن‬7ً /[w ,-Y+‫ا‬

[Fiction novel - Advice from parents

to their children]

73 Kids have gone overweight from ,'Y+‫ا ا‬Xw,Q-+


407

eating too much McDonalds. They

need regular exercise to use up sugar.

[Advice from a doctor to parents in an

article in a coffee table magazine]

74 Sarah: I've managed to learn typing on ‫رات‬7O/ ‫ب‬7YF‫آ‬M\ S}‫را‬7O/ 5-3.F+

the computer, but I still haven't got the

hang of it.

Sally: You have to practise a lot to

build up skills. [Fiction novel -

Conversation between two sisters]

75 They finally realized that there is an ¥/7P,(+‫[اد ا‬EJ \ [1[Q} \ ]RX+

urgent need for the health promotion

program. They will have a special

meeting to work it out. [News report]

76 Accused: I wasn't at that market last [1‫ أر‬M \ 5-P7ˆ ‫ة‬,/ ‫ع‬XRX3)+ ‫د‬X*.H 7.P‫ أ‬%+ Kƒ} M

night, and I don't know the person you ‫ى‬,y‫ة أ‬,/ ‫ع‬XRX3+‫ ا‬‰_} p+‫د إ‬X*P ‫أن‬

are talking about.

Prosecutor: We are not going into that

again, are we?

[A cross examination in court]

77 You are making a big fuss out of 7O1‫ر‬7I/ p+‫ ا‬7-3+‫[ا ا‬-*}‫؟\ و‬7Q+7UF}‫و‬

nothing. Why don't you two kiss and

make up? [Fiction novel - Conversation

between classmates]

78 Men are like computers: Hard to 0&&&&/ \ 4OF8,&&&&*/ |*U&&&&} \ 4&&&&O3O8 |*U&&&&+‫ ا‬0&&&&/
408

figure out and never have enough 4‫ه‬ŠŽ+ S8 |*U+‫ا‬

memory. [Joke]

79 Mr. Crutcher is working in a 7‫آ‬X)&H S)H‫ \ \[أت ا‬5.-N3+‫ت ا‬78,UF+‫ ا‬0E x_wX}

Starbucks coffee shop. He says that he 5Q_&&&` x&&&1XW \ ‫ء‬7^&&&-\ 5Q_&&&` ‫ \ \&&&[أت‬7\~&&&O/

is now a reformed character. He says 41Xƒ+‫‚ ا‬1,{+‫ ا‬p+‫ت إ‬


ُ [E \ %R73+‫ا‬

he's not back to his old tricks and uses

a nice phrase 'I've cleaned up my act.'

[News report]

80 This is a nice tune and I think it'll catch 7O+73*F&&H‫] ا‬-N&&-H \ 5E,Y&&\ ,OFN&&FH 7&&OP7\ [&&ƒFE‫ا‬

on quickly. [Fiction novel - 5E,Y\

Conversation between friends]

81 My neighbour had a bad car accident %EX+‫ ا‬0E ‫ب‬7G \ %EX+‫[ ا‬ƒ8 \ ‘-)E %3G‫أ‬

yesterday. After the accident he

blacked out and couldn't remember

what happened. [Diary entry]

82 Five or six years ago I'd put away a 4Q)+‫ ا‬0/ ‫ة‬,-(‫ آ‬5*{w p)E %}† \ 4OF+‫ أ‬x.‫آ‬

big piece of meat most evenings. Now

I just eat a bit of chicken or fish. [An

advertisement for a weight loss

product]

83 I was busy paying lots of bills this 58[U+7\ %F„‫ زو‬x)\7w \ %F„‫ زو‬x8‫د‬7`

morning, and I ran into my wife at the

post office [Diary entry]

84 The salesman finds it easy to take in ŠL7I*+‫[ع ا‬n1 ‫أن‬

old ladies and persuade them to give


409

him their money. [Fiction novel -

Conversation between shoppers]

85 I had a bad pain in my stomach last p_Fy‫ \ زال \ ا‬po2}

night. It was severe to begin with, but

it soon passed away. [Fiction novel -

Conversation between a patient and a

doctor]

86 I must warn you that anything you may ‘.1‫ }[و‬4F-H

say will be taken down and used in

court as evidence against you.

[Police interview]

87 You may not like our flowers, but they ‫ن‬X'F&H \ ‫ف‬7&{3+‫ ا‬5&17OP %8 S\7IE‫ل ا‬7.FH 7O.'+‫و‬

will grow on you in the end. [A note %&8 S&+ ‫وق‬,F&H \ ‫ف‬7&{3+‫ ا‬517OP %8 S\7IE‫™ ا‬Q/

written on a flower stall] ‫ف‬7{3+‫ ا‬517OP

88 One woman says that her boss is ‘&P9 \7&O_ّ)n1 %&F+‫ ا‬p&RX_)+ \ |&-},F+7\ ‘&/7-w ‫*[م‬+

horrible and he stresses her out. She ,O€3\ 4FO1 M \ ‫ي‬XRX8 ‘P9 \ 4ّ€./ ,-G

says that she deals with the stress by

chanting. Another woman says her

partner stresses her out because he

doesn't tidy up. And she copes with it

by having a relaxing bath!

[Article on stress]

89 It's spring break in Florida and \ ‹W7N&+‫ ا‬p&)E ‫ن‬X&Z,31 \‹W7N+‫ ا‬p)E ‫ن‬Xy,FY1

students are chilling out at the beach \ 4OY&&&_P‫ أ‬0&&&E ‫ن‬X&&&Z‫و‬,1 \ 5&&&Z‫ا‬,)+ 7ً& &Fw‫ن و‬X^& &ƒ1

and partying late into the night. 5Z‫ا‬,+‫ ا‬0/ ™Yƒ\ ‫ن‬X3*.1
410

[News report]

90 One day, I went over to my Uncle \ ‫[ّة‬N&&&&\ ‫&&&&[رّس‬3+‫ ا‬%&&&&.(ّP‫[ّة \ أ‬N&&&&\ 4&&&&ّ)*3+‫ ا‬%&&&&.ّn\‫و‬

Tony's house and didn't get a chance to %._ّ.E \ %P,„‫ز‬

do my Maths homework. The teacher

told me off for not doing it the day after.

[Fiction novel - Conversation between a

daughter and her parents]

91 Chris: you want to come out tonight? ‫ت‬2_Z %.-_'1 \ ,OY+‫ ا‬%Pّ[‫ \ ه‬,OY+‫ ا‬%.'OP‫أ‬

Adam: I've had enough, I'm partied

out. [Fiction novel - Conversation

between friends]

92 Funk up your mobile with cool screen ]&&&/ 7ً -&&&o73} ,&&&…‫ أآ‬S&&&_}7‫ ه‬K&&&*„‫& &[ّث \ ا‬Z \ 0&&&ّ1‫ز‬

savers [An internet advertisement for 5RX3+‫ا‬

mobile phone screen savers]

93 Latham sexed up Iraq brief. \ K-)^&&&F+‫ض ا‬,&&&Ž+ %&&&w‫ا‬,*+‫)&&&• ا‬3+‫ك ا‬,&&&(_1 4…&&&1M

Mark Latham's credibility as %w‫ا‬,*+‫)• ا‬3+‫ ا‬%8 |E2} 4…1M

opposition leader faced its gravest test

last night after John Howard

effectively accused him of misleading

parliament over top-secret intelligence

briefings on Iraq. [News headline]

94 Waitress: Are you ready to order now? ‫ ًء‬7„‫)• ر‬3+‫ ا‬0/ %))ّw \ S)^8 0/ •-_y •)/

Maria: Yes. I'll have some salad, roast

beef, and mashed potatoes.

Waitress: How do you want the


411

beef? Rare, medium, or well-done?

Maria: Well-done. And go easy on

the salt, please. [Fiction novel -

Conversation at a restaurant]

95 Bill: Where did you get this article ‫ت‬7E7H ‫ث‬2ˆ ‫[ة‬3+ xP,FP9‫ ا‬p)E KGXG %8 x…Q\

from?

Jack: I googled around for three hours

to get the site on which it was

published. [Fiction novel -

Conversation between friends]

96 I've spent four hours doing my late night \ ‫ّق‬XY&&&F+‫ ا‬0&&&/ ‫ي‬,&&&OŸ 4U&&&ƒP‫ّق \ ا‬XY&&&F+‫ ا‬%&&&.')‫أه‬

shopping. I'm really shopped out. ]ّ^(F+‫ة ا‬,…‫ آ‬0/ x(*}

[Fiction novel - Conversation between a

wife and husband]

97 How about we get the students of 5E,Y+‫ و„‘ ا‬p)E ‫ر‬X‫ه‬Š+‫ت ا‬7w7\ “*\ ‚-Y.F+

floral art from the Tech to fang up a

few arrangements using roses and

jasmine? [Fiction novel –

Conversation between school

teachers]

98 When I mentioned it, they all looked at ‫ح‬,3P‫ح و‬,YP \ [\,*P

me strangely and said 'What are we

going to do?' and I said we're going to

talk, tell secrets, dance, dag out. [Fiction

novel - Conversation between friends]


412

99 Human beings run into trouble when ‫ „&&&&&[ران‬p&&&&&)E ‫ول‬,Y&&&&&FY-+X'+‫ ا‬4‫اآ‬,&&&&&F1 7/[&&&&&.E

arterial walls guck up with cholesterol 5&&&&&&)ƒ…/ 0-1‫ا‬,N&&&&&&+‫• ا‬-U&&&&&&} 7/[&&&&&&.E \ 0-1‫ا‬,N&&&&&&+‫ا‬

and blood can't easily flow. ‫ول‬,FY-+X'+7\

[Medical brochure]

100 We had built up this cult following 7.F*3H ‘1XN}‫ و‬7P‫د‬7ƒFP‫ ا‬4} \ 5H,o 5)3Q+ 7.Rّ,*}

while being an independent band, but

as soon as we had a commercial

record we were slagged down as

cashing in on a theme. [Diary entry]


413

APPENDIX 3
FULL DETAILS OF THE RESULTS OF THE
TRANSLATION TESTS

Subject Number: P1

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 X
5 √
6 √
7 X
8 √
9 X
10 √
11 √
12 √
13 √
14 X
15 X
16 √
17 √
18 √
19 X
20 √
21 √
22 √
23 √
24 √
25 X
26 √
27 √
28 X
29 √
30 √
31 √
32 X
33 √
34 √
35 X
36 √
414

37 √
38 X
39 X
40 X
41 √
42 √
43 √
44 √
45 √
46 √
47 √
48 X
49 √
50 √
51 X
52 √
53 X
54 X
55 √
56 √
57 X
58 √
59 √
60 X
61 X
62 √
63 √
64 X
65 √
66 X
67 √
68 √
69 √
70 X
71 √
72 √
73 √
74 √
75 √
76 X
77 √
78 √
79 √
80 √
81 √
82 √
83 √
84 X
415

85 X
86 √
87 X
88 √
89 X
90 X
91 X
92 X
93 √
94 √
95 √
96 √
97 X
98 X
99 X
100 X
Very Good= 50 Satisfactory= 14 Overtly= 26 Covertly= 10 Unanswered= 0

Subject Number: P2

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 X
5 X
6 X
7 X
8 √
9 √
10 X
11 √
12 X
13 X
14 X
15 X
16 √
17 X
18 √
19 X
20 √
21 √
22 X
23 √
416

24 X
25 X
26 √
27 √
28 X
29 X
30 √
31 √
32 X
33 √
34 √
35 X
36 X
37 X
38 √
39 X
40 X
41 √
42 √
43 X
44 √
45 √
46 √
47 √
48 X
49 X
50 √
51 X
52 √
53 X
54 √
55 X
56 √
57 X
58 X
59 √
60 X
61 X
62 X
63 √
64 √
65 X
66 √
67 √
68 √
69 √
70 √
71 √
417

72 √
73 X
74 √
75 X
76 X
77 √
78 X
79 √
80 X
81 √
82 X
83 X
84 X
85 √
86 √
87 X
88 X
89 X
90 √
91 X
92 √
93 X
94 X
95 √
96 X
97 X
98 X
99 X
100 X
Very Good= 27 Satisfactory= 19 Overtly= 46 Covertly= 8 Unanswered= 0

Subject Number: P3

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 X
3 √
4 X
5 √
6 X
7 X
8 √
9 X
10 √
418

11 √
12 X
13 X
14 X
15 X
16 X
17 X
18 X
19 √
20 √
21 √
22 X
23 √
24 X
25 X
26 √
27 X
28 X
29 X
30 √
31 X
32 X
33 √
34 X
35 √
36 √
37 X
38 √
39 X
40 X
41 √
42 √
43 X
44 √
45 √
46 X
47 √
48 X
49 √
50 X
51 X
52 √
53 √
54 √
55 √
56 √
57 X
58 X
419

59 √
60 X
61 √
62 √
63 X
64 √
65 √
66 X
67 √
68 X
69 X
70 √
71 √
72 √
73 √
74 X
75 X
76 X
77 X
78 X
79 X
80 X
81 X
82 X
83 √
84 X
85 √
86 √
87 X
88 √
89 X
90 X
91 X
92 √
93 X
94 √
95 √
96 X
97 X
98 X
99 √
100 X
Very Good= 32 Satisfactory= 12 Overtly= 48 Covertly= 8 Unanswered= 0
420

Subject Number: P4
Q Correct Incorrect Unanswered
Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 X
5 X
6 X
7 √
8 √
9 X
10 √
11 X
12 √
13 √
14 √
15 X
16 X
17 X
18 √
19 √
20 √
21 X
22 X
23 √
24 X
25 X
26 √
27 √
28 √
29 √
30 X
31 X
32 X
33 X
34 X
35 √
36 √
37 √
38 X
39 X
40 X
41 X
421

42 √
43 √
44 √
45 √
46 √
47 X
48 √
49 √
50 √
51 √
52 √
53 √
54 √
55 X
56 X
57 X
58 √
59 √
60 √
61 √
62 X
63 √
64 √
65 √
66 X
67 √
68 √
69 √
70 √
71 √
72 √
73 √
74 X
75 X
76 X
77 X
78 √
79 √
80 X
81 √
82 X
83 X
84 X
85 √
86 √
87 X
88 √
89 √
422

90 √
91 X
92 X
93 √
94 √
95 X
96 √
97 X
98 X
99 X
100 X
Very Good= 43 Satisfactory= 13 Overtly= 36 Covertly= 8 Unanswered= 0

Subject Number: P5

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 X
5 √
6 X
7 X
8 √
9 X
10 X
11 √
12 √
13 X
14 √
15 X
16 √
17 X
18 √
19 X
20 X
21 X
22 X
23 √
24 √
25 √
26 √
27 X
28 X
423

29 √
30 √
31 X
32 √
33 √
34 X
35 √
36 X
37 X
38 X
39 X
40 X
41 √
42 √
43 X
44 X
45 √
46 X
47 X
48 √
49 X
50 √
51 X
52 √
53 X
54 √
55 X
56 √
57 X
58 X
59 √
60 X
61 √
62 √
63 X
64 √
65 √
66 X
67 √
68 √
69 √
70 √
71 √
72 √
73 √
74 √
75 X
76 X
424

77 √
78 X
79 √
80 √
81 √
82 X
83 √
84 X
85 √
86 √
87 √
88 √
89 √
90 √
91 X
92 X
93 X
94 √
95 X
96 X
97 X
98 X
99 X
100 X
Very Good= 36 Satisfactory= 15 Overtly= 36 Covertly= 13 Unanswered= 0

Subject Number: P6

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 X
5 √
6 X
7 X
8 √
9 √
10 √
11 √
12 √
13 X
14 √
15 X
16 X
425

17 √
18 X
19 X
20 √
21 X
22 X
23 √
24 X
25 √
26 √
27 X
28 X
29 √
30 √
31 √
32 X
33 X
34 X
35 X
36 √
37 X
38 X
39 √
40 X
41 ▬
42 √
43 √
44 √
45 √
46 √
47 √
48 √
49 √
50 √
51 √
52 √
53 √
54 √
55 √
56 √
57 X
58 X
59 √
60 √
61 X
62 X
63 √
64 X
426

65 X
66 X
67 √
68 √
69 √
70 √
71 √
72 √
73 √
74 √
75 X
76 X
77 √
78 √
79 X
80 X
81 √
82 X
83 √
84 √
85 √
86 √
87 X
88 X
89 X
90 X
91 √
92 √
93 X
94 √
95 X
96 X
97 X
98 X
99 X
100 X
Very Good= 38 Satisfactory= 17 Overtly= 37 Covertly= 7 Unanswered= 1

Subject Number: P7

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 X
3 √
4 X
427

5 X
6 X
7 √
8 √
9 X
10 √
11 X
12 X
13 X
14 √
15 X
16 √
17 X
18 √
19 √
20 √
21 X
22 √
23 ▬
24 X
25 X
26 √
27 X
28 √
29 √
30 √
31 √
32 X
33 X
34 √
35 X
36 √
37 X
38 X
39 X
40 √
41 √
42 √
43 √
44 √
45 √
46 X
47 √
48 √
49 √
50 √
51 X
52 √
428

53 X
54 √
55 X
56 √
57 X
58 √
59 √
60 √
61 √
62 √
63 √
64 √
65 √
66 √
67 √
68 √
69 √
70 √
71 √
72 √
73 X
74 √
75 X
76 X
77 √
78 √
79 X
80 X
81 √
82 X
83 X
84 √
85 √
86 √
87 X
88 √
89 √
90 X
91 X
92 √
93 X
94 √
95 √
96 √
97 X
98 X
99 √
100 X
429

Very Good= 44 Satisfactory= 15 Overtly= 33 Covertly= 7 Unanswered= 1


Subject Number: P8

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 X
4 X
5 √
6 √
7 X
8 √
9 X
10 √
11 √
12 √
13 X
14 X
15 X
16 √
17 X
18 √
19 X
20 X
21 √
22 √
23 √
24 √
25 X
26 √
27 √
28 X
29 X
30 √
31 √
32 √
33 X
34 √
35 X
36 X
37 √
38 √
39 √
40 √
41 √
430

42 √
43 √
44 √
45 √
46 √
47 √
48 √
49 √
50 √
51 √
52 √
53 √
54 √
55 X
56 √
57 X
58 √
59 √
60 √
61 √
62 √
63 √
64 √
65 √
66 √
67 √
68 √
69 √
70 X
71 √
72 √
73 X
74 √
75 X
76 X
77 √
78 √
79 X
80 √
81 X
82 √
83 √
84 √
85 √
86 √
87 X
88 √
89 X
431

90 X
91 X
92 √
93 √
94 √
95 √
96 X
97 X
98 X
99 X
100 X
Very Good= 52 Satisfactory= 14 Overtly= 26 Covertly= 8 Unanswered= 0

Subject Number: P9

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 X
5 √
6 X
7 X
8 √
9 X
10 X
11 √
12 √
13 X
14 √
15 X
16 √
17 X
18 √
19 X
20 √
21 X
22 √
23 √
24 √
25 √
26 √
27 √
28 X
432

29 √
30 √
31 X
32 X
33 √
34 √
35 √
36 X
37 √
38 √
39 √
40 √
41 √
42 √
43 X
44 √
45 √
46 √
47 √
48 √
49 √
50 √
51 X
52 √
53 √
54 √
55 X
56 X
57 X
58 X
59 √
60 X
61 √
62 √
63 √
64 X
65 X
66 √
67 √
68 √
69 √
70 X
71 √
72 √
73 √
74 √
75 X
76 X
433

77 √
78 √
79 X
80 X
81 √
82 √
83 X
84 √
85 √
86 √
87 X
88 X
89 √
90 X
91 X
92 X
93 √
94 X
95 X
96 X
97 X
98 X
99 X
100 X
Very Good= 40 Satisfactory= 18 Overtly= 36 Covertly= 6 Unanswered= 0

Subject Number: P10

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 X
3 X
4 √
5 √
6 X
7 √
8 X
9 X
10 √
11 X
12 √
13 X
14 √
15 √
16 √
434

17 √
18 √
19 X
20 X
21 X
22 √
23 √
24 X
25 X
26 √
27 √
28 X
29 √
30 √
31 √
32 √
33 X
34 √
35 X
36 √
37 X
38 ▬
39 X
40 √
41 √
42 √
43 X
44 √
45 √
46 X
47 √
48 √
49 √
50 √
51 √
52 √
53 X
54 √
55 √
56 √
57 X
58 X
59 √
60 √
61 √
62 √
63 √
64 √
435

65 √
66 X
67 √
68 √
69 √
70 √
71 √
72 √
73 √
74 X
75 √
76 √
77 √
78 √
79 √
80 X
81 √
82 X
83 √
84 √
85 √
86 √
87 X
88 X
89 X
90 X
91 X
92 √
93 √
94 √
95 √
96 √
97 X
98 X
99 X
100 X
Very Good= 49 Satisfactory= 14 Overtly= 32 Covertly= 4 Unanswered= 1

Subject Number: P11

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 √
436

5 √
6 √
7 X
8 √
9 √
10 √
11 √
12 X
13 √
14 √
15 √
16 √
17 √
18 √
19 X
20 √
21 √
22 √
23 √
24 X
25 X
26 √
27 √
28 √
29 √
30 √
31 √
32 X
33 X
34 √
35 √
36 √
37 X
38 X
39 √
40 X
41 √
42 √
43 √
44 √
45 √
46 √
47 √
48 √
49 √
50 √
51 √
52 √
437

53 √
54 √
55 √
56 √
57 X
58 √
59 √
60 √
61 √
62 √
63 √
64 √
65 √
66 √
67 √
68 √
69 √
70 √
71 √
72 X
73 √
74 √
75 X
76 √
77 √
78 √
79 X
80 √
81 √
82 X
83 √
84 X
85 √
86 √
87 X
88 √
89 √
90 X
91 √
92 √
93 X
94 √
95 √
96 √
97 X
98 X
99 √
100 X
438

Very Good= 63 Satisfactory= 14 Overtly= 18 Covertly= 5 Unanswered= 0


Subject Number: P12

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 X
3 √
4 √
5 √
6 X
7 X
8 √
9 X
10 √
11 X
12 X
13 X
14 X
15 X
16 √
17 X
18 √
19 √
20 X
21 √
22 X
23 √
24 √
25 √
26 √
27 X
28 X
29 √
30 X
31 X
32 X
33 √
34 √
35 X
36 √
37 X
38 √
39 X
40 √
41 √
439

42 √
43 X
44 √
45 √
46 √
47 X
48 √
49 X
50 X
51 √
52 √
53 √
54 √
55 √
56 X
57 X
58 X
59 √
60 √
61 √
62 √
63 √
64 ▬
65 √
66 √
67 √
68 √
69 √
70 X
71 √
72 √
73 X
74 √
75 X
76 X
77 √
78 √
79 X
80 √
81 X
82 X
83 √
84 √
85 √
86 √
87 √
88 √
89 √
440

90 X
91 ▬
92 X
93 X
94 X
95 X
96 √
97 X
98 X
99 X
100 ▬
Very Good= 29 Satisfactory= 25 Overtly= 37 Covertly= 6 Unanswered= 3

Subject Number: S1

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 X
3 X
4 X
5 X
6 X
7 X
8 X
9 X
10 √
11 X
12 √
13 X
14 X
15 X
16 X
17 X
18 X
19 X
20 X
21 X
22 X
23 √
24 X
25 X
26 X
27 X
28 X
441

29 X
30 X
31 X
32 X
33 √
34 √
35 √
36 X
37 X
38 X
39 X
40 X
41 X
42 X
43 X
44 √
45 √
46 X
47 X
48 X
49 X
50 √
51 X
52 √
53 X
54 X
55 X
56 √
57 X
58 X
59 √
60 X
61 X
62 X
63 X
64 X
65 X
66 X
67 √
68 X
69 X
70 X
71 X
72 √
73 X
74 √
75 X
76 X
442

77 √
78 X
79 √
80 X
81 X
82 X
83 X
84 X
85 X
86 √
87 X
88 √
89 √
90 √
91 X
92 X
93 X
94 √
95 √
96 X
97 X
98 X
99 X
100 X
Very Good= 9 Satisfactory= 14 Overtly= 73 Covertly= 4 Unanswered= 0

Subject Number: S2

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 X
3 X
4 X
5 X
6 ▬
7 X
8 √
9 X
10 X
11 √
12 X
13 √
14 √
15 √
16 √
443

17 ▬
18 X
19 √
20 X
21 X
22 ▬
23 X
24 ▬
25 X
26 √
27 √
28 ▬
29 ▬
30 X
31 X
32 ▬
33 X
34 √
35 X
36 X
37 ▬
38 ▬
39 ▬
40 ▬
41 ▬
42 ▬
43 ▬
44 ▬
45 ▬
46 ▬
47 ▬
48 ▬
49 √
50 ▬
51 √
52 √
53 √
54 X
55 X
56 ▬
57 X
58 X
59 √
60 X
61 ▬
62 ▬
63 √
64 ▬
444

65 √
66 ▬
67 √
68 X
69 √
70 X
71 √
72 X
73 √
74 √
75 √
76 √
77 X
78 √
79 ▬
80 ▬
81 X
82 √
83 ▬
84 X
85 √
86 X
87 ▬
88 √
89 X
90 X
91 X
92 ▬
93 ▬
94 X
95 √
96 X
97 ▬
98 ▬
99 X
100 ▬
Very Good= 18 Satisfactory= 11 Overtly= 35 Covertly= 2 Unanswered= 34

Subject Number: S3

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 X
3 √
4 √
445

5 √
6 √
7 √
8 √
9 X
10 √
11 √
12 √
13 √
14 √
15 X
16 X
17 √
18 √
19 √
20 √
21 √
22 √
23 √
24 X
25 √
26 √
27 X
28 √
29 X
30 √
31 √
32 X
33 X
34 √
35 X
36 X
37 √
38 √
39 X
40 √
41 X
42 √
43 X
44 √
45 √
46 √
47 X
48 √
49 X
50 √
51 √
52 √
446

53 X
54 √
55 √
56 √
57 X
58 X
59 √
60 √
61 X
62 X
63 √
64 √
65 √
66 X
67 √
68 √
69 √
70 √
71 √
72 √
73 √
74 √
75 √
76 √
77 √
78 √
79 √
80 X
81 √
82 X
83 √
84 √
85 √
86 √
87 X
88 √
89 √
90 √
91 X
92 X
93 ▬
94 X
95 √
96 X
97 X
98 X
99 X
100 X
447

Very Good= 41 Satisfactory= 24 Overtly= 32 Covertly= 2 Unanswered= 1


Subject Number: S4

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 X
5 √
6 √
7 X
8 X
9 √
10 √
11 X
12 √
13 √
14 √
15 X
16 √
17 X
18 √
19 √
20 √
21 √
22 √
23 √
24 √
25 √
26 X
27 √
28 √
29 X
30 √
31 X
32 √
33 X
34 √
35 √
36 X
37 √
38 X
39 √
40 X
41 √
448

42 √
43 X
44 √
45 √
46 √
47 √
48 √
49 X
50 X
51 √
52 √
53 X
54 X
55 √
56 √
57 X
58 X
59 √
60 √
61 X
62 X
63 √
64 √
65 √
66 X
67 √
68 X
69 √
70 √
71 √
72 X
73 √
74 √
75 √
76 X
77 √
78 √
79 √
80 √
81 √
82 √
83 √
84 X
85 √
86 √
87 X
88 √
89 √
449

90 X
91 √
92 √
93 X
94 √
95 √
96 X
97 X
98 X
99 X
100 X
Very Good= 50 Satisfactory= 14 Overtly= 30 Covertly= 6 Unanswered= 0

Subject Number: S5

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 X
5 √
6 X
7 √
8 √
9 X
10 X
11 X
12 X
13 X
14 X
15 √
16 √
17 X
18 X
19 X
20 X
21 √
22 √
23 X
24 √
25 X
26 √
27 √
28 √
450

29 X
30 √
31 √
32 X
33 √
34 √
35 √
36 √
37 √
38 X
39 X
40 √
41 X
42 √
43 X
44 √
45 √
46 X
47 X
48 √
49 √
50 X
51 √
52 √
53 X
54 √
55 X
56 √
57 X
58 X
59 √
60 √
61 X
62 √
63 √
64 X
65 √
66 X
67 √
68 X
69 X
70 √
71 √
72 √
73 X
74 √
75 X
76 X
451

77 √
78 √
79 X
80 √
81 √
82 √
83 √
84 √
85 √
86 √
87 √
88 √
89 √
90 X
91 √
92 √
93 √
94 √
95 √
96 √
97 X
98 X
99 X
100 X
Very Good= 39 Satisfactory= 19 Overtly= 40 Covertly= 2 Unanswered= 0

Subject Number: S6

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 X
3 √
4 ▬
5 √
6 ▬
7 √
8 √
9 X
10 √
11 √
12 √
13 √
14 X
15 √
16 √
452

17 X
18 √
19 √
20 X
21 √
22 √
23 X
24 X
25 X
26 √
27 √
28 X
29 ▬
30 X
31 X
32 X
33 ▬
34 X
35 √
36 √
37 X
38 X
39 √
40 √
41 X
42 ▬
43 √
44 ▬
45 √
46 ▬
47 √
48 ▬
49 X
50 √
51 √
52 √
53 X
54 √
55 √
56 ▬
57 ▬
58 √
59 X
60 ▬
61 X
62 √
63 ▬
64 ▬
453

65 √
66 X
67 √
68 X
69 ▬
70 X
71 √
72 √
73 √
74 X
75 X
76 √
77 √
78 √
79 X
80 X
81 √
82 X
83 X
84 ▬
85 √
86 X
87 X
88 √
89 √
90 X
91 X
92 ▬
93 ▬
94 X
95 √
96 X
97 ▬
98 X
99 X
100 ▬
Very Good= 23 Satisfactory= 20 Overtly= 31 Covertly= 7 Unanswered= 19

Subject Number: S7

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 √
454

5 X
6 √
7 √
8 √
9 X
10 √
11 √
12 X
13 X
14 √
15 X
16 X
17 X
18 X
19 √
20 √
21 X
22 √
23 √
24 √
25 X
26 √
27 X
28 X
29 ▬
30 √
31 X
32 X
33 X
34 X
35 √
36 √
37 X
38 X
39 X
40 √
41 X
42 √
43 X
44 √
45 X
46 √
47 ▬
48 √
49 X
50 √
51 X
52 √
455

53 √
54 X
55 √
56 X
57 X
58 √
59 √
60 √
61 X
62 X
63 √
64 X
65 √
66 X
67 X
68 √
69 X
70 X
71 √
72 X
73 √
74 X
75 X
76 X
77 √
78 √
79 X
80 √
81 √
82 √
83 √
84 √
85 √
86 X
87 X
88 √
89 X
90 X
91 √
92 X
93 √
94 √
95 √
96 X
97 X
98 X
99 X
100 ▬
456

Very Good= 27 Satisfactory= 21 Overtly= 43 Covertly= 6 Unanswered= 3


Subject Number: S8

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 X
3 √
4 X
5 √
6 X
7 √
8 √
9 X
10 √
11 X
12 X
13 X
14 √
15 X
16 X
17 √
18 X
19 √
20 X
21 √
22 X
23 X
24 X
25 √
26 √
27 √
28 √
29 √
30 X
31 X
32 X
33 √
34 √
35 X
36 X
37 √
38 X
39 X
40 X
41 X
457

42 √
43 X
44 X
45 √
46 X
47 √
48 X
49 √
50 √
51 X
52 √
53 X
54 √
55 √
56 X
57 X
58 X
59 √
60 √
61 X
62 X
63 X
64 X
65 √
66 X
67 X
68 √
69 √
70 √
71 √
72 √
73 X
74 √
75 X
76 X
77 √
78 √
79 √
80 √
81 √
82 X
83 X
84 X
85 √
86 X
87 X
88 √
89 X
458

90 √
91 X
92 √
93 X
94 X
95 √
96 X
97 X
98 X
99 X
100 X
Very Good= 24 Satisfactory= 20 Overtly= 52 Covertly= 4 Unanswered= 0

Subject Number: S9

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 X
4 X
5 X
6 X
7 √
8 √
9 X
10 X
11 √
12 X
13 X
14 X
15 √
16 √
17 √
18 X
19 X
20 √
21 X
22 X
23 X
24 X
25 X
26 √
27 X
28 √
459

29 √
30 √
31 X
32 X
33 X
34 √
35 X
36 X
37 X
38 X
39 X
40 X
41 X
42 √
43 X
44 √
45 √
46 √
47 X
48 √
49 X
50 √
51 X
52 √
53 √
54 √
55 X
56 X
57 ▬
58 X
59 √
60 X
61 X
62 X
63 ▬
64 ▬
65 √
66 X
67 √
68 √
69 √
70 √
71 √
72 √
73 √
74 √
75 X
76 X
460

77 √
78 √
79 X
80 X
81 X
82 X
83 X
84 ▬
85 X
86 X
87 X
88 X
89 X
90 ▬
91 X
92 √
93 ▬
94 ▬
95 X
96 ▬
97 ▬
98 ▬
99 ▬
100 ▬
Very Good= 20 Satisfactory= 15 Overtly= 49 Covertly= 4 Unanswered= 12

Subject Number: S10

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 X
5 √
6 √
7 √
8 √
9 X
10 √
11 √
12 √
13 X
14 √
15 X
16 X
461

17 X
18 √
19 √
20 √
21 X
22 X
23 √
24 X
25 X
26 √
27 √
28 √
29 √
30 √
31 √
32 √
33 √
34 X
35 X
36 X
37 √
38 √
39 X
40 X
41 √
42 √
43 √
44 √
45 √
46 √
47 √
48 √
49 X
50 √
51 √
52 √
53 X
54 X
55 X
56 √
57 X
58 X
59 √
60 X
61 √
62 √
63 √
64 X
462

65 X
66 X
67 √
68 X
69 √
70 √
71 √
72 √
73 X
74 √
75 X
76 X
77 √
78 √
79 √
80 √
81 √
82 √
83 √
84 √
85 √
86 √
87 X
88 √
89 √
90 X
91 X
92 √
93 X
94 √
95 √
96 X
97 X
98 X
99 X
100 X
Very Good= 46 Satisfactory= 15 Overtly= 34 Covertly= 5 Unanswered= 0

Subject Number: S11

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 X
3 X
4 √
463

5 √
6 X
7 √
8 √
9 X
10 X
11 √
12 √
13 X
14 X
15 X
16 √
17 X
18 X
19 X
20 X
21 √
22 √
23 X
24 X
25 X
26 X
27 √
28 X
29 √
30 √
31 X
32 X
33 X
34 X
35 X
36 √
37 √
38 X
39 X
40 X
41 X
42 X
43 X
44 √
45 √
46 √
47 X
48 √
49 √
50 √
51 X
52 √
464

53 √
54 X
55 √
56 √
57 X
58 X
59 √
60 √
61 X
62 X
63 √
64 X
65 X
66 √
67 X
68 √
69 X
70 √
71 √
72 √
73 √
74 √
75 X
76 X
77 √
78 √
79 X
80 X
81 √
82 X
83 X
84 X
85 X
86 √
87 X
88 X
89 √
90 √
91 X
92 X
93 X
94 X
95 √
96 X
97 X
98 X
99 X
100 X
465

Very Good= 27 Satisfactory= 14 Overtly= 47 Covertly= 12 Unanswered= 0


Subject Number: S12

Q Correct Incorrect Unanswered


Very good Satisfactory Overtly Covertly
erroneous erroneous
error error
1 X
2 √
3 √
4 X
5 √
6 √
7 √
8 √
9 X
10 X
11 √
12 √
13 X
14 √
15 X
16 √
17 X
18 X
19 √
20 √
21 √
22 √
23 X
24 X
25 √
26 √
27 √
28 √
29 √
30 √
31 √
32 √
33 X
34 √
35 √
36 √
37 √
38 √
39 X
40 √
41 √
466

42 √
43 √
44 √
45 √
46 √
47 √
48 √
49 √
50 √
51 √
52 √
53 X
54 X
55 X
56 √
57 X
58 X
59 √
60 X
61 √
62 √
63 √
64 X
65 X
66 √
67 √
68 √
69 √
70 √
71 √
72 √
73 √
74 X
75 X
76 √
77 √
78 √
79 X
80 √
81 X
82 X
83 X
84 √
85 √
86 √
87 X
88 X
89 √
467

90 X
91 X
92 X
93 X
94 √
95 √
96 √
97 X
98 X
99 X
100 X
Very Good= 41 Satisfactory= 23 Overtly= 33 Covertly= 3 Unanswered= 0
468

APPENDIX 4
AVERAGE OF THE STANDARD DEVAITION

Arabic professional translators' group

Question Category Total of Total of Unanswered Percentage Average Standard


number correct incorrect of the deviation
answers answers correct
1 c 0 12 0 0%
2 c 8 4 0 66.66%
3 c 10 2 0 83.33%
4 c 3 9 0 25%
5 c 9 3 0 75%
6 c 3 9 0 25%
7 c 3 9 0 25%
8 c 11 1 0 91.66%
9 c 3 9 0 25%
10 c 8 4 0 66.66%
11 c 8 4 0 66.66%
12 c 7 5 0 58.33%
13 c 3 9 0 25%
14 c 7 5 0 58.33%
15 c 2 10 0 16.66%
16 c 9 3 0 75%
17 c 4 8 0 33.33%
18 c 10 2 0 83.33%
19 c 4 8 0 33.33%
20 c 8 4 0 66.66%
21 c 6 6 0 50%
22 c 6 6 0 50%
23 c 11 0 1 91.66%
24 c 5 7 0 41.66%
25 c 4 8 0 33.33% 50.66% 0.256748
52 de 12 0 0 100%
53 de 7 5 0 58.33%
54 de 11 1 0 91.66%
55 de 6 6 0 50%
56 de 9 3 0 75% 75% 0.212453
47 den 8 4 0 66.66%
48 den 10 2 0 83.33%
49 den 9 3 0 75%
50 den 10 2 0 83.33%
51 den 6 6 0 50% 71.66% 0.139435
469

36 dnp 8 4 0 66.66%
37 dnp 4 8 0 33.33%
38 dnp 5 6 1 41.66%
39 dnp 4 8 0 33.33%
40 dnp 5 7 0 41.66%
41 dnp 10 1 1 83.33%
42 dnp 12 0 0 100%
43 dnp 6 6 0 50%
44 dnp 11 1 0 91.66%
45 dnp 12 0 0 100%
46 dnp 8 4 0 66.66% 64.39% 0.275136
26 dpn 12 0 0 100%
27 dpn 7 5 0 58.33%
28 dpn 3 9 0 25%
29 dpn 9 3 0 75%
30 dpn 10 2 0 83.33%
31 dpn 7 5 0 58.33%
32 dpn 3 9 0 25%
33 dpn 6 6 0 50%
34 dpn 8 4 0 66.66%
35 dpn 5 7 0 41.66% 58.33% 0.261125
65.32% 0.229834
91 p 2 9 1 16.66%
92 p 7 5 0 58.33%
93 p 5 7 0 41.66%
94 p 9 3 0 75%
95 p 7 5 0 58.33%
96 p 6 6 0 50%
97 p 0 12 0 0%
98 p 0 12 0 0%
99 p 3 9 0 25%
100 p 0 11 1 0% 32.50% 0.27902
73 sva 8 4 0 66.66%
74 sva 9 3 0 75%
75 sva 2 10 0 16.66%
77 sva 11 1 0 91.66%
78 sva 9 3 0 75%
79 sva 5 7 0 41.66%
80 sva 5 7 0 41.66%
81 sva 9 3 0 75%
82 sva 3 9 0 25%
84 sva 6 6 0 50%
85 sva 11 1 0 91.66%
86 sva 12 0 0 100%
88 sva 8 4 0 66.66%
89 sva 6 6 0 50%
90 sva 3 9 0 25% 59.44% 0.261381
57 svap 0 12 0 0%
470

58 svap 5 7 0 41.66%
59 svap 12 0 0 100%
60 svap 7 5 0 58.33%
61 svap 9 3 0 75%
62 svap 9 3 0 75%
63 svap 10 2 0 83.33%
64 svap 8 3 1 66.66%
65 svap 9 3 0 75%
66 svap 6 6 0 50%
67 svap 12 0 0 100%
68 svap 11 1 0 91.66%
69 svap 11 1 0 91.66%
70 svap 8 4 0 66.66%
71 svap 12 0 0 100%
72 svap 11 1 0 91.66% 72.91% 0.264398
76 svp 2 10 0 16.66%
83 svp 8 4 0 66.66%
87 svp 2 10 0 16.66% 33.33% 0.288675
63.48% 0.281305

Arabic translation students' group


Question Category Total of Total of Unanswered Percentage Average Standard
number correct incorrect of the devation
answers answers correct
1 c 0 12 0 0%
2 c 6 6 0 50%
3 c 8 4 0 66.66%
4 c 3 8 1 25%
5 c 8 4 0 66.66%
6 c 5 5 2 41.66%
7 c 9 3 0 75%
8 c 10 2 0 83.33%
9 c 1 11 0 8.33%
10 c 7 5 0 58.33%
11 c 8 4 0 66.66%
12 c 7 5 0 58.33%
13 c 4 8 0 33.33%
14 c 7 5 0 58.33%
15 c 4 8 0 33.33%
16 c 7 5 0 58.33%
17 c 3 8 1 25%
18 c 4 8 0 33.33%
19 c 8 4 0 66.66%
471

20 c 6 6 0 50%
21 c 7 5 0 58.33%
22 c 7 4 1 58.33%
23 c 5 7 0 41.66%
24 c 3 8 1 25%
25 c 4 8 0 33.33% 47% 0.208157
52 de 12 0 0 100%
53 de 5 7 0 41.66%
54 de 5 7 0 41.66%
55 de 6 6 0 50%
56 de 7 3 2 58.33% 58.33% 0.242979
47 den 5 5 2 41.66%
48 den 8 2 2 66.66%
49 den 5 7 0 41.66%
50 den 9 2 1 75%
51 den 7 5 0 58.33% 56.66% 0.149093
36 dnp 5 7 0 41.66%
37 dnp 7 4 1 58.33%
38 dnp 3 8 1 25%
39 dnp 2 9 1 16.66%
40 dnp 5 6 1 41.66%
41 dnp 3 8 1 25%
42 dnp 8 2 2 66.66%
43 dnp 3 8 1 25%
44 dnp 9 1 2 75%
45 dnp 10 1 1 83.33%
46 dnp 7 3 2 58.33% 46.97% 0.227523
26 dpn 9 3 0 75%
27 dpn 8 4 0 66.66%
28 dpn 7 4 1 58.33%
29 dpn 5 4 3 41.66%
30 dpn 8 4 0 66.66%
31 dpn 4 8 0 33.33%
32 dpn 3 8 1 25%
33 dpn 4 7 1 33.33%
34 dpn 8 4 0 66.66%
35 dpn 6 6 0 50% 51.66% 0.174793
51.88% 0.198036
91 p 3 9 0 25%
92 p 5 5 2 41.66%
93 p 2 6 4 16.66%
94 p 6 5 1 50%
95 p 11 1 0 91.66%
96 p 2 9 1 16.66%
97 p 0 9 3 0%
98 p 0 10 2 0%
99 p 0 11 1 0%
100 p 0 8 4 0% 24.16% 0.297737
472

73 sva 8 4 0 66.66%
74 sva 9 3 0 75%
75 sva 3 9 0 25%
77 sva 11 1 0 91.66%
78 sva 11 1 0 91.66%
79 sva 5 6 1 41.66%
80 sva 6 5 1 50%
81 sva 8 4 0 66.66%
82 sva 5 7 0 41.66%
84 sva 5 5 2 41.66%
85 sva 9 3 0 75%
86 sva 7 5 0 58.33%
88 sva 9 3 0 75%
89 sva 8 4 0 66.66%
90 sva 4 7 1 33.33% 60% 0.204606
57 svap 0 10 2 0%
58 svap 2 10 0 16.66%
59 svap 11 1 0 91.60%
60 svap 6 5 1 50%
61 svap 2 9 1 16.66%
62 svap 4 7 1 33.33%
63 svap 8 2 2 66.66%
64 svap 2 7 3 16.66%
65 svap 8 4 0 66.66%
66 svap 2 9 1 16.66%
67 svap 9 3 0 75%
68 svap 6 6 0 50%
69 svap 7 4 1 58.33%
70 svap 8 4 0 66.66%
71 svap 11 1 0 91.66%
72 svap 9 3 0 75% 49.47% 0.29255
76 svp 4 8 0 33.33%
83 svp 5 6 1 41.66%
87 svp 1 10 1 8.33% 27.77% 0.173459
52.20% 0.25891
473

APPENDIX 5
RANKING THE ITEMS ACCORDING TO THEIR
PERCENTAGES OF THE CORRECT IN A DESENDING
ORDER

Professional translators' group

Question number Total of Total of Unanswered Percentage of


correct incorrect the correct
answers answers
1 0 12 0 0%
57 0 12 0 0%
97 0 12 0 0%
98 0 12 0 0%
100 0 11 1 0%
15 2 10 0 16.6%
75 2 10 0 16.6%
76 2 10 0 16.6%
87 2 10 0 16.6%
91 2 9 1 16.6%
4 3 9 0 25%
6 3 9 0 25%
7 3 9 0 25%
9 3 9 0 25%
13 3 9 0 25%
28 3 9 0 25%
32 3 9 0 25%
82 3 9 0 25%
90 3 9 0 25%
99 3 9 0 25%
17 4 8 0 33.3%
19 4 8 0 33.3%
25 4 8 0 33.3%
37 4 8 0 33.3%
39 4 8 0 33.3%
24 5 7 0 41.6%
35 5 7 0 41.6%
38 5 6 1 41.6%
40 5 7 0 41.6%
58 5 7 0 41.6%
79 5 7 0 41.6%
80 5 7 0 41.6%
93 5 7 0 41.6%
21 6 6 0 50%
22 6 6 0 50%
474

33 6 6 0 50%
43 6 6 0 50%
51 6 6 0 50%
55 6 6 0 50%
66 6 6 0 50%
84 6 6 0 50%
89 6 6 0 50%
96 6 6 0 50%
12 7 5 0 58.3%
14 7 5 0 58.3%
27 7 5 0 58.3%
31 7 5 0 58.3%
53 7 5 0 58.3%
60 7 5 0 58.3%
92 7 5 0 58.3%
95 7 5 0 58.3%
2 8 4 0 66.6%
10 8 4 0 66.6%
11 8 4 0 66.6%
20 8 4 0 66.6%
34 8 4 0 66.6%
36 8 4 0 66.6%
46 8 4 0 66.6%
47 8 4 0 66.6%
64 8 3 1 66.6%
70 8 4 0 66.6%
73 8 4 0 66.6%
83 8 4 0 66.6%
88 8 4 0 66.6%
5 9 3 0 75%
16 9 3 0 75%
29 9 3 0 75%
49 9 3 0 75%
56 9 3 0 75%
61 9 3 0 75%
62 9 3 0 75%
65 9 3 0 75%
74 9 3 0 75%
78 9 3 0 75%
81 9 3 0 75%
94 9 3 0 75%
3 10 2 0 83.3%
18 10 2 0 83.3%
30 10 2 0 83.3%
41 10 1 1 83.3%
48 10 2 0 83.3%
50 10 2 0 83.3%
63 10 2 0 83.3%
475

8 11 1 0 91.6%
23 11 0 1 91.6%
44 11 1 0 91.6%
54 11 1 0 91.6%
68 11 1 0 91.6%
69 11 1 0 91.6%
72 11 1 0 91.6%
77 11 1 0 91.6%
85 11 1 0 91.6%
26 12 0 0 100%
42 12 0 0 100%
45 12 0 0 100%
52 12 0 0 100%
59 12 0 0 100%
67 12 0 0 100%
71 12 0 0 100%
86 12 0 0 100%

Translation students' group

Question number Total of Total of Unanswered Percentage of


correct incorrect the correct
answers answers
1 0 12 0 0%
57 0 10 2 0%
97 0 9 3 0%
98 0 10 2 0%
99 0 11 1 0%
100 0 8 4 0%
9 1 11 0 8.3%
87 1 10 1 8.3%
39 2 9 1 16.6%
58 2 10 0 16.6%
61 2 9 1 16.6%
64 2 7 3 16.6%
66 2 9 1 16.6%
93 2 6 4 16.6%
96 2 9 1 16.6%
4 3 8 1 25%
17 3 8 1 25%
24 3 8 1 25%
32 3 8 1 25%
38 3 8 1 25%
41 3 8 1 25%
43 3 8 1 25%
75 3 9 0 25%
476

91 3 9 0 25%
13 4 8 0 33.3%
15 4 8 0 33.3%
18 4 8 0 33.3%
25 4 8 0 33.3%
31 4 8 0 33.3%
33 4 7 1 33.3%
62 4 7 1 33.3%
76 4 8 0 33.3%
90 4 7 1 33.3%
6 5 5 2 41.6%
23 5 7 0 41.6%
29 5 4 3 41.6%
36 5 7 0 41.6%
40 5 6 1 41.6%
47 5 5 2 41.6%
49 5 7 0 41.6%
53 5 7 0 41.6%
54 5 7 0 41.6%
79 5 6 1 41.6%
82 5 7 0 41.6%
83 5 6 1 41.6%
84 5 5 2 41.6%
92 5 5 2 41.6%
2 6 6 0 50%
20 6 6 0 50%
35 6 6 0 50%
55 6 6 0 50%
60 6 5 1 50%
68 6 6 0 50%
80 6 5 1 50%
94 6 5 1 50%
10 7 5 0 58.3%
12 7 5 0 58.3%
14 7 5 0 58.3%
16 7 5 0 58.3%
21 7 5 0 58.3%
22 7 4 1 58.3%
28 7 4 1 58.3%
37 7 4 1 58.3%
46 7 3 2 58.3%
51 7 5 0 58.3%
56 7 3 2 58.3%
69 7 4 1 58.3%
86 7 5 0 58.3%
3 8 4 0 66.6%
5 8 4 0 66.6%
11 8 4 0 66.6%
477

19 8 4 0 66.6%
27 8 4 0 66.6%
30 8 4 0 66.6%
34 8 4 0 66.6%
42 8 2 2 66.6%
48 8 2 2 66.6%
63 8 2 2 66.6%
65 8 4 0 66.6%
70 8 4 0 66.6%
73 8 4 0 66.6%
81 8 4 0 66.6%
89 8 4 0 66.6%
7 9 3 0 75%
26 9 3 0 75%
44 9 1 2 75%
50 9 2 1 75%
67 9 3 0 75%
72 9 3 0 75%
74 9 3 0 75%
85 9 3 0 75%
88 9 3 0 75%
8 10 2 0 83.3%
45 10 1 1 83.3%
59 11 1 0 91.6%
71 11 1 0 91.6%
77 11 1 0 91.6%
78 11 1 0 91.6%
95 11 1 0 91.6%
52 12 0 0 100%

You might also like