Comparative Morphology of

Standard and Egyptian Arabic

Hassan A. H. Gadalla

Lincom Europa
2000

CONTENTS
Foreword............................................................................................ vii
0. Introduction.................................................................................... ix
0.1. The Arabic Language Situation...................................... ix
0.2. The Diglossic Situation in Egypt.................................... x
0.3. Purpose & Procedures of the Research........................... xii
0.4. Symbols & Abbreviations............................................... xv
1. Phonological & Morphological Basics.......................................... 1
1.1. Phonological Preliminaries..................................... 1
1.1.1 Consonant Systems........................................... 1
1.1.2. Vowel Systems................................................ 5
1.1.3. Syllable Structure & Stress.............................. 7
1.2. Phonological Alternations.............................................. 9
1.2.1. Epenthesis.................................................... 10
1.2.2. Elision...................................................... 14
1.2.3. Assimilation................................................. 16
1.2.4. Shortening........................................................ 19
1.2.5. Metathesis........................................................ 21
1.2.6. Glide Alternations........................................... 22
1.3. Morphological Preliminaries.......................................... 26
1.3.1. Word Classification......................................... 26
1.3.2. Word Formation.............................................. 28
1.4. Morphosyntactic Alternations........................................ 33
1.4.1. Pausal vs. Non-Pausal Forms.......................... 33
1.4.2. Nunation.......................................................... 35
1.5. Literature on Arabic Morphology................................... 36
1.6. Theory of Prosodic Morphology..................................... 40
2. Verbal Morphology........................................................................ 43
2.1. Verb Stems & Classes..................................................... 43
2.1.1. Triradical Verbs............................................... 43
2.1.1.1. Sound Verbs...................................... 44
2.1.1.2. Geminate Verbs................................ 51
2.1.1.3. Glottalized Verbs.............................. 55
2.1.1.4. Weak Verbs....................................... 58
2.1.2. Quadriradical Verbs......................................... 73
2.2. Inflection for Aspect & Mood.........................................76
2.3. Inflection for Voice......................................................... 85
2.4. Subjectival & Objectival Affixes.................................... 88
2.4.1. Subjectival Affixes.......................................... 89
2.4.2. Objectival Suffixes.......................................... 97
2.5. Verb Derivation & Transitivity...................................... 100
3. Nominal Morphology..................................................................... 106

3.1. Primary Nouns................................................................ 106

3.1.1. Triradical Nominal Stem Forms...................... 107
3.1.2. Quadriradical Nominal Stem Forms................ 114
3.1.3. A Prosodic Analysis of Nominal Stem Forms. 115
3. 2. Deverbal Nouns............................................................. 117
3.2.1. Verbal Nouns................................................... 118
3.2.2. Nouns of Exaggeration.................................... 125
3.2.3. Nouns of Place & Time....................................126
3.2.4. Nouns of Instrument........................................ 128
3.3. Definite vs. Indefinite Nouns.......................................... 129
3.4. Inflection for Case...........................................................130
3.5. Inflection for Gender....................................................... 136
3.6. Inflection for Number..................................................... 140
3.6.1. The Singular..................................................... 140
3.6.2. The Dual.......................................................... 140
3.6.3. The Plural......................................................... 143
3.6.3.1. Sound Masculine Plural.................... 145
3.6.3.2. Sound Feminine Plural...................... 146
3.6.3.3. Broken Plural.................................... 148
3.6.3.4. Pluralization of Nouns of Place/Time
& Instrument.................................... 155
3.7. Nouns & Suffixation....................................................... 156
3.8. The Diminutive Form......................................................157
4. Adjectival Morphology.................................................................. 159
4.1. Adjectival Stem Forms................................................... 159
4.2. Definite vs. Indefinite Adjectives................................... 163
4.3. Inflection for Case...........................................................165
4.4. Inflection for Gender....................................................... 168
4.5. Inflection for Number..................................................... 170
4.5.1. The Singular..................................................... 170
4.5.2. The Dual.......................................................... 170
4.5.3. The Plural......................................................... 171
4.5.3.1. Sound Masculine Plural.................... 172
4.5.3.2. Sound Feminine Plural...................... 173
4.5.3.3. Broken Plural.................................... 175
4.6. Inflection for Degree....................................................... 181
4.6.1. The Positive Degree......................................... 181
4.6.2. The Comparative Degree................................. 181
4.6.3. The Superlative Degree.................................... 184

4.7. Participles....................................................................... 186

4.7.1. The Active Participle....................................... 187
4.7.2. The Passive Participle...................................... 194
4.7.3. Participles Inflection & Suffixation................ 199
4.8. Relational Adjectives...................................................... 201
5. Closed-List Class Morphology...................................................... 206
5.1. Pronouns......................................................... 206
5.1.1. Personal Pronouns............................................ 206
5.1.2. Relative Pronouns............................................ 209
5.1.3. Demonstrative Pronouns.................................. 210
5.2. Prepositions..................................................................... 213
5.3. Adverbs........................................................................... 219
5.4. Interrogative & Responsive Particles..............................223
5.5. Negative Particles........................................................... 225
5.6. Possessive Particles........................................................ 230
6. Conclusions.................................................................................... 232
7. Notes.............................................................................................. 242
8. Bibliography.................................................................................. 245
9. Appendix........................................................................................ 254

!

$ ('
"
"
(22

And among His signs is the creation of the heavens and

the earth, and the variations in your languages and your
colors: Verily in that are signs for those who know.
(The Holy Quran 30: 22)

Foreword

It is my great pleasure to introduce this important monograph

devoted to a comparison of Cairene and Standard Arabic morphology,
written by Dr. Hassan Gadalla, currently of the Department of English at
Assiut University, Egypt. Dr. Gadalla's study originated as a doctoral
dissertation in Linguistics, written partly under my supervision at the
Department of Linguistics at the University of Pennsylvania, during a twoyear period while Dr. Gadalla was supported as a visiting scholar under
the joint-supervision scholarship program of Assiut University. While
completing the dissertation, Dr. Gadalla was also employed by the
Linguistic Data Consortium at Penn, leading an annotation team that
examined a large corpus of naturally occurring Arabic telephone
conversations and other specimens of modern spoken Arabic. The results
of this work were compiled as the LDC CallHome Egyptian Arabic
Lexicon (H. Gadalla et. al. (1998). This work represents the first electronic
phonological and morphological dictionary of Egyptian Arabic.
The present monograph is therefore based both on a variety of
previously published descriptive sources as well as the above-mentioned
electronic database, which revealed a number of previously unnoticed
linguistic phenomena of the Cairene dialect. It contains a systematic
comparison of the stem types of both languages and a descriptive
treatment of both languages phonologies using an easily interpreted rulebased formalism. It is shown that while Cairene Arabic can normally be
derived from the same underlying forms as Standard Arabic with the
addition of a suite of dialect-specific phonological rules, Cairene Arabic
also differs from Standard Arabic in the loss of certain stem types, in
certain sporadic phonological changes, and in the impoverishment of
inflectional categories. The historical development of these latter changes
all invite future research and pose an important agenda for scholars
studying the development of modern Arabic. Remarkably exhaustive in
its treatment of the categories and forms of both the standard and modern
colloquial Cairene varieties of Arabic, this work should serve as a useful
reference for linguistic researchers as well as teachers of modern Egyptian
Arabic phonology and morphology.

Rolf Noyer, PhD

Philadelphia, March 2000

Acknowledgments
First and foremost, I do thank Allah, the Almighty God, for all the
bounties He has showered upon me. Then, I thank Prof. Ramzy Radwan,
Dean of the Faculty of Languages and Translation, October 6 University,
who has given me all the valuable guidance I needed. I also thank Prof.
Ahmed Abdel-Hafiz, Vice-Dean of the Faculty of Arts, South Valley
University, for giving me many crucial insights throughout my work.
My research has also benefited greatly from my weekly meetings
with Prof. Rolf Noyer for two years at the University of Pennsylvania. I
am proud to have been his student and proud of his words in the first letter
to me: we will have much to learn from each other. Indeed it was he
who has taught me exactly what it is like to be a linguistics researcher. I
also appreciate the assistance of Prof. Tony Kroch, Chair of the Dept. of
Linguistics at the University of Pennsylvania who has been very helpful to
me. Finally, I thank all the members of my family, particularly my parents,
my wife and my children Gehad, Hamza and Sarah for their moral support
at every stage of this work.

Introduction
0.1. The Arabic Language Situation:
Arabic is the most widespread member of the Semitic group of
languages1. Two main varieties of this language can be distinguished in
the Arab world nowadays: Standard Arabic (SA), also called Modern
Standard Arabic (MSA) and Colloquial Arabic. The first variety is the
offspring of Classical Arabic, also called Quranic Arabic (e.g. by
Thackston 1984), which is now used in religious settings and the recitation
of the Holy Quran. Thus, Standard Arabic is considered the direct
descendant of the classical language, with modifications and
simplifications more suited to communication in a world quite different
from that of the Arab Golden Age in medieval times (Travis 1979: 6). It
has also been defined by Gaber (1986: 1) as the written form taught at
schools2. He goes on to say that in its spoken form it is the formal
speech of the educated people in public speeches, radio comments, news
broadcasts on radio and television. The written form of SA is relatively
uniform throughout the Arab world. The spoken form, on the other hand,
is more or less different from one Arab country to another since it is
affected by the local dialects.
Many labels have been given in the linguistic literature to the
Standard variety of Arabic. It has been named Literary Arabic (e.g. by
Becker 1964) although many of its manifestations are not related to
literature, as in the language of newspapers and magazines. Also, it has
been termed Written Arabic (e.g. by Beeston 1968) in spite of the fact
that it is frequently used as the means of spoken communication, as in
academic lectures and some radio and television programs. So, none of
these terms gives a well-defined description of this variety. The term
Standard Arabic will be employed in this book for three reasons. First, it
refers to all forms of Arabic stated above. Second, it covers the areas
missed by other terms. Third, the use of this term has become a longestablished tradition in modern linguistic studies. (See, for instance,
Cowan (1968), Malik (1976) and Abdel-Hafiz (1991) among many
others).
The second variety, Colloquial Arabic, has been defined by AlToma (1969: 3) as the actual language of everyday activities, mainly
spoken, though occasionally written. He adds that, it varies not only
from one Arab territory to another, but also from one area to another
within each territory. Hence, nearly every Arab country has its own
colloquial dialects that are more or less different from each other and

10

naturally from those of other countries. Several names have also been
given to this variety of Arabic, among which are Vernacular Arabic (e.g.
by Smith 1917) and Spoken Arabic (e.g. by Salib 1981). The term
Colloquial is chosen here because it is more common than all of the
other terms.
0.2. The Diglossic Situation in Egypt:
In Egypt, two main varieties of Arabic are commonly used:
Standard Arabic and Egyptian Colloquial Arabic. The former is the
language of reading and writing, while the latter is the language of daily
social intercourse. However, the latter appears in writing in certain
situations, e.g. some poets and playwrights adopt it in their works. Within
the Colloquial variety there are many vernacular dialects, such as the
Cairene, the Upper-Egyptian and other regional dialects.
For many reasons, the most prestigious dialect throughout the
country is Cairene Arabic (CrA). First, it is the language of the capital
where the government administration offices are located. Second, it is the
language of the cinema, theater and mass media. Third, it is spoken by a
great number of educated and cultured people. This prestige has led some
linguists studying Egyptian Arabic to concentrate on this dialect giving
generalizations on the dialects spoken in the whole country. (See, for
example, Gamal-Eldin (1967), Hanna (1967) and Omar (1976)). Needless
to say, this dialect does not cover the whole of Egypt. That is why some
researchers focused on the study of other regional dialects, such as Abu
Farag (1960) and Khalafallah (1969).
Some sociolinguists state that there are five levels of Arabic used
in Egypt. This was initiated by Badawi (1973) who posited these five
levels in his socio-linguistic analysis of contemporary Arabic in Egypt:
(a) fuSha al-turaa Classical Arabic of the heritage,
(b) fuSha al-9aSr Contemporary Classical Arabic,
(c) 9aamiyyat al-muaqqafiin Colloquial of the educated,
(d) 9aamiyyat al-mutanawwiriin Colloquial of the enlightened and
(e) 9aamiyyat al-?ummiyyiin Colloquial of the illiterate.
In his English paper (1985: 16), Badawi used different terms: (a) Classical
Arabic, (b) Modern Standard Arabic, (c) Educated Spoken Arabic, (d)
Semi-literate Spoken Arabic and (e) Illiterate Spoken Arabic.

11

In this book, I will recognize a division of these five levels into two
levels because, as Parkinson (1981: 24-5) comments, in Badawis schema
there would be a relatively sharp break between Standard and Colloquial
Arabic:
MSA and the Colloquials do share a large proportion of
their lexicon; there are, however, numerous very common
markers that immediately let the reader/hearer know which
variety is being used. These include certain verbal prefixes,
the negative construction, the demonstrative construction
and a lot of the most common words. ... With them, there is
a sharp break between MSA and Colloquial with very little
mixing.
In other words, there is a big difference between the Standard levels, on
the one hand, and the Colloquial levels, on the other, particularly in the
morphological domain. This will be detailed in Chapters Two through
Five.
The existence of two varieties of the same language in one society
is known in the linguistic literature as diglossia. This term has been
defined by Ferguson (1972: 242) as:
a relatively stable language situation in which, in addition
to the primary dialects of the language (which may include
a standard or regional standards), there is a very divergent,
highly codified (often grammatically more complex)
superposed variety, the vehicle of a large and respected
body of written literature and is used for most written and
formal spoken purposes but is not used by any sector of the
community for ordinary conversation.
In diglossic situations, the two varieties are sometimes called
high and low in terms of formality (Crystal 1985: 93). In our case, SA
is the high variety and EA is the low one. Evidently, the members of
diglossic communities are aware that their varieties or languages are
associated with particular domains. Thus, SA, on the one hand, is
restricted to the formal domain; it is taught at school and used in formal
settings. It is also the language of all printed materials such as literary
books, schoolbooks, newspapers, government publications and the like.
Consequently, it has greater social prestige. EA, on the other hand, is
confined to the informal domain; it is utilized by every member of the
community in Cairo as the major vehicle of communication at home, in
the market place, etc.

12

0.3. Purpose & Procedures of the Research:

As will be evident from the survey of literature in (1.5), most
researches, until now, have considered only certain aspects of Standard
Arabic or Egyptian Colloquial Arabic. Moreover, there is a great emphasis
on Arabic syntax in modern linguistic studies at the expense of phonology
and morphology; which has led Elgibali (1996: 10) to say:
A cursory review of current research reveals that syntax has
steadfastly and rapidly become the most dominant single
constituent of formal work in Arabic linguistics in general.
The situation is such that one hardly ever hears of ample or
systematic work being done involving other major elements
of linguistic analysis such as semantics, morphology, or
phonology.
The present book, in embracing both the Standard and Egyptian varieties
of the language, will attempt to elucidate their basic natural relationship
and to explain their differences in terms of morphological comparison.
Hence, this book aims to provide a comparative account of the
morphological aspects of SA and EA. It will focus on the similarities and
differences of the two varieties of Arabic in relation to their verbal,
nominal, adjectival, and closed-list class morphology.
So, the objective is to fill in a gap in Arabic studies, which has not
been adequately covered in previous works. Hopefully, there will be also
some pedagogical applications. This book is of great importance for
language teaching, since it serves as a guide for teachers of Arabic to
native and non-naive speakers. It can be used by course-designers for a
new approach to Arabic grammar based on modern linguistics. It can also
be helpful to teachers of foreign languages, particularly English, to
determine the degree of difficulty, due to Arabic interference, encountered
by Arab students when they are introduced to the basic morphological
phenomena of the foreign language(s). The book may also be beneficial
for non-native speakers when they start to learn Arabic, for it provides
them with an understanding of the morphological aspects of two varieties
of the language.
Moreover, this book offers material for contrastive and
comparative studies on Arabic. It is also significant for studies on
language problems related to diglossia and computer programs on the
Arabic language. Needless to say that this book will be useful to linguists
working on universal grammar who do not confine themselves to one
language but try to find common properties of all languages in the world.

13

This book is based on the comparative description of SA and EA.

It will not be confined to any particular school of thought, or to any
particular model proposed by a given school. Although, at some points,
resort will be made to the theory of Prosodic Morphology developed by
McCarthy & Prince (1986, 1988), yet there will be no exclusive
dependence on this theory. A brief sketch on this theory is provided in
(1.6). Thus, the framework adopted in the book is chiefly a descriptive
one, taking SA as the basis of description. The Standard feature will be
described first and the Colloquial counterpart will be shown afterwards. I
will concentrate on the morphological aspects because the morphological
component in Arabic overshadows other components in a way that makes
it quite fair to give it special consideration.
The Standard variety dealt with in this book is the language taught
in the Arabic language courses at Egyptian schools and adopted in the
literary pages of Egyptian newspapers. Since this conventional definition
of SA designates such sources, the data employed for the research is taken
from the grammar books used at schools as well as from the literary pages
of Al-Ahram newspaper.
On the other hand, the Colloquial variety described in this book
represents the dialect spoken in Cairo, the most prestigious dialect in
Egypt. Apart from my familiarity with this dialect, being myself a resident
of Cairo for a number of years, several informants from Cairo and
Egyptian Colloquial texts (both written and spoken) are intensively
consulted. In addition, I had the opportunity to dig into the depths of this
dialect by participating in the compilation of the LDC CallHome Egyptian
Arabic Lexicon, published as Gadalla et al. (1998) by the Linguistic Data
Consortium of the University of Pennsylvania, USA. This lexicon
represents the first electronic pronunciation dictionary of Egyptian
Colloquial Arabic, Cairene Dialect. It consists of 54,375 words. It contains
tab-separated information fields including orthographic representation in
both romanized and Arabic scripts, morphological structure,
pronunciation, stress, source and frequency information for each word.
The lexical entries found in the lexicon come primarily from the
transcripts of 100 ten-minute segments from 200 telephone calls initiated
from the USA and Canada and made to people in Cairo.
The book is divided into five chapters. The first is an introductory
chapter which sets the scene for the whole work. It presents a phonological
summary of the consonant and vowel systems of the two varieties as well
as their stress and syllable structure. Then, it analyses the important

14

phonological alternations involved in the formation of surface forms.

Moreover, it offers a comparison of the basic morphological and
morphosyntactic features of the two varieties. At the end, it surveys the
previous literature on the morphology of both varieties and explains the
theory of Prosodic Morphology.
Chapter Two deals with the verbal morphology component. It
presents a comparison between the two varieties in reference to the
triradical and quadriradical forms of verbs and the types of verb roots. In
addition, the inflection of verbs for aspect/mood and voice will be
contrasted in the two varieties. Subjectival and objectival affixes are then
illustrated. Verb derivation and transitivity will be dealt with at the end of
the chapter.
Chapter Three is related to the nominal morphology component. It
handles the primary nominal stems and investigates the nouns derived
from verbs: verbal nouns, nouns of place and time and nouns of
instrument. It involves a comparison between SA and the Colloquial
dialect spoken in Cairo in relation to the inflection of nouns for case,
gender, and number as well as a treatment of the suffixation of nouns.
The adjectival morphology component, in Chapter Four, provides a
comparison between SA and EA concerning the adjectival stems. Then,
the degrees of adjectives: positive, comparative and superlative, and the
inflection of adjectives for case, gender and number are discussed. In
addition, the derivation of participial forms and relational adjectives will
be compared in the two varieties.
Finally, the closed-list class morphology component, in Chapter
Five, is concerned with pronouns and prepositions in the two varieties.
Moreover, adverbs as well as interrogative and responsive particles in both
of them will be compared. Negative and possessive particles will also be
dealt with in this chapter.
0.4. List of Symbols & Abbreviations:
C
V
v
G
Adj
N
Part

Consonant
Vowel
Epenthetic Vowel
Glide
Adjective
Noun
Participle

SA
EA
sg
du
pl
pf
impf

Standard Arabic
Egyptian Arabic
singular
dual
plural
perfective
imperfective

15

Pro
1
2
3
intr
trans
caus
inch
hi
lo
rd
son
obstr
emph
non-emph
>
<

*
/ /
[ ]
( )
#

F-3-L

Pronoun
def
definite
First Person
fut
future
Second Person
Nom
Nominative
Third Person
Acc
Accusative
intransitive
Gen
Genitive
transitive
indic
indicative
causative
subj
subjunctive
inchoative
juss
jussive
high vowel
imper
imperative
low vowel
prog
progressive
round vowel
cons
consonantal
sonorant
cor
coronal
obstruent
hum
human
emphatic
m, masc
masculine
non-emphatic
f, fem
feminine
has the EA equivalent
has the SA equivalent
is changed into or becomes
ungrammatical word or structure
phonemic transcription
morphological forms, elements or transcription
elements found on the surface only, or in SA only
word boundary
morpheme boundary
syllable
phonological environment
First, Second and Third consonants of the root, i.e. = C1, C2
and C3, respectively
AB/YZ
A slash-dash notation in which, for instance, A
becomes B after Y and before Z.

16

Chapter One
Phonological & Morphological Basics
1.0. Introduction:
This chapter gives a phonological outline of Standard Arabic (SA)
and Egyptian Arabic (EA) and deals with some phonological alternations
in both of them (1.1 and 1.2). It handles two of the basic morphological
features in Arabic, namely word classification and word formation (1.3).
In addition, it presents an analysis of two important morphosyntactic
phenomena in Arabic, namely the use of pausal forms and nunation (1.4).
It offers a survey of the previous studies on the subject (1.5). Finally, it
gives a brief idea about the theory of prosodic morphology (1.6).
1.1. Phonological Preliminaries:
Understanding the morphology of Arabic, or any other language,
cannot be accomplished without an understanding of its phonology. So, in
this section the phonological preliminaries of the two varieties under
analysis are to be discussed. The consonantal systems of the two varieties
are to be compared in (1.1.1). Their vocalic systems are to be contrasted in
(1.1.2). The syllable structure and the stress patterns in the two varieties
are to be contrasted in (1.1.3).
1.1.1. Consonant Systems:
The orthographic alphabet of SA includes twenty-eight letters
which represent twenty-eight consonants, though three of them are also
used as vowels (See 1.1.2). EA, on the other hand, has no more than
twenty-six of these consonants. The consonants of both varieties are
represented in Table (1):
Since the SA interdentals // and // are non-existent in EA, they
are replaced in some words by the corresponding dental stops /t/ and /d/,
respectively, and in other words by the corresponding alveolar fricatives
/s/ and /z/, respectively:
SA
aman(-un)
ahab(-un)
aabit(-un)
akiyy(-un)

EA
taman
dahab
saabit
zaki

Gloss
price
gold
steady
intelligent

17

Fricative

Affricate
Flap
Voiced
Lateral
Voiced
Nasal
Voiced
m
Glide
Voiced
w
(+) = Found in SA only.

+
+

t
d
s
z

T
D
S
Z

k
g
x
g

Glottal

Pharyngeal

Uvular

Velar

Palatal

Voiceless
Voiced
Voiceless
Voiced

NonEmphatic
Emphatic

Stop

Interdental

Voicing

Labiodental

Manner

Bilabial

Table (1)
The Consonants of Both SA & EA
Place
DentoAlveolar

?
h
3

j
r
l
n
y

Hence, SA interdentals undergo change in EA along two parallel lines.

They become either dental plosives or sibilants. Gairdner (1925: 31)
explains this by proposing that the true spontaneous change was to
dental-plosives; the sibilants being probably the result of an attempt to
classicize, i.e. to imitate the interdentals of literary Arabic, on the part of
semi-educated people. It can also be said that the tendency to avoid
producing the interdentals in EA is due to the ease of production principle:
the dental plosives or sibilants are easier to produce than the interdentals.
(Cf. Hyman (1975: 16), for example, points out that children acquire /f/
before //).
The Standard dento-alveolar emphatic fricative /Z/ is used in SA
and EA as a substitute for the Classical Arabic interdental emphatic
fricative //. However, in EA, the SA /Z/ is replaced in some words by the
emphatic dento-alveolar /D/, e.g. SA /Zuhr(-un)/ noon has the EA form
/Duhr/ but SA /Zahar(-a)/ to appear has the EA form /Zahar/. This can
also be justified by assuming that the change to /D/ is the regular
spontaneous change and the use of /Z/ is a kind of borrowing from SA by
Egyptian people.
Moreover, the SA uvular stop /q/ has a marginal status in the
Cairene dialect. The latter deviates from the former in the use of the glottal
stop /?/ as a reflex of /q/ in most of the words having this phoneme. To

18

exemplify, /qadiim(-un)/ old is pronounced /?adiim/ and /fariiq(-un)/ a

team is articulated /farii?/. In Upper Egypt and some rural areas in Lower
Egypt, the voiced velar stop /g/ is used in the place of /q/. Thus, the above
words are pronounced /gadiim/ and /fariig/. However, in very few words,
/q/ is exceptionally retained in EA, e.g. /musiiqa/ music and /al-qaahir-a/
Cairo. In turn, the voiced velar stop /g/ is used in the Cairene dialect and
the Cairene pronunciation of SA as a reflex of the SA voiced palatal
affricate /j/ (whose IPA symbol is //). In Upper Egypt and other Arab
countries, the affricate is still employed for the pronunciation of this
phoneme in SA.
Both SA and EA make a phonological distinction between
emphatic consonants and non-emphatic ones. The first group includes
/T, D, S and Z/, while the second involves /t, d, s and z/. In addition, two
consonants occur both emphatically and non-emphatically: /l/ and /r/. To
explain the difference, Mitchell (1956: 6-7) refers to three distinguishing
features: First, for the emphatics the tongue must be broad and thick
filling the mouth; for the non-emphatics, on the other hand, the tongue is
narrow and thin. Second, the front of the tongue is very much lower in
the mouth for the emphatics. In other words, they are lowered coronals,
i.e. produced with the lower front part of the tongue. Finally, the lips are
held neutral or slightly rounded and protruded for the emphatics but they
are spread for the non-emphatics. Kenstowicz (1994: 42) indicates that the
gesture underlying the emphatic consonants is a secondary articulation
known as pharyngealization since the tongue root or body is retracted
during the production of these consonants. The emphatics are usually
transcribed by underdotting or capitalization. In this book, I am using
underdotting for /l/ and /r/ and capitalization for the other emphatics. The
following minimal pairs are illustrative contrastive examples that can be
found in both varieties (parenthesized elements are specifically Standard):
Emphatics
Tiin(-un) mud
Dall(-a) to lose his way
?aS-Siin(-u) China
Zahr(-un) back

Non-Emphatics
tiin(-un) figs
dall(-a) to direct
?s-siin(-u) a letter name
zahr(-un) flowers

The clearest Standard example of emphatic /l/ is the word for

God /allaah(-u)/, on condition that /l/ is not preceded by /i/, e.g. /?in
aa?-a llaah(-u)/ God-willing. If it is preceded by this vowel, it is plain,
as in /al-hamd-u li-llaah(-i)/ Praise be to God. In EA, the word for God
is pronounced in the same way as in SA. It has emphatic /l/ when it is not
preceded by /i/, as in /walla/ by God in contrast to /walla/ or. In both

19

varieties, /l/ is made emphatic in the neighborhood of the four emphatic

consonants /T, D, S and Z/, either preceding or following it. Vowels and
other consonants may intervene but this alternation is limited to the word
domain. Examples of preceding emphatics are: /Talaaq(-un) > Talaa?/
divorce, /Dalaal(-un)/ delusion, /Salaat(-un) > Sala/ prayer and
/Zulm(-un)/ injustice. Examples of following emphatics are /laTiif(-un)/
nice, /lanDan/ London, /luSuuS(-un)/ thieves and /laahaZ(-a)/ to
notice.
The second consonant that has emphatic and non-emphatic
versions is the voiced alveolar flap /r/. In SA, /r/ is made emphatic unless
it is followed by /i/. The rules3 for the emphaticization of /r/ in SA are:
(1) Emphaticization of /r/ in SA:
a. r r / i
b. r r / elsewhere
These rules indicate that in SA /r/ is non-emphatic before /i/ and emphatic
every where else. Examples of (1a) are /ribh(-un)/ profit and /fariiq(-un)/
a team. Examples of (1b) are /raml(-un)/ sand and /rub3(-un)/ a
quarter.
EA follows SA in regard to making /r/ non-emphatic before /i/.
However, it does not follow SA in relation to the adjacency of /a/. The
following rules can be proposed for the emphaticization of /r/ in EA:
(2) Emphaticization of /r/ in EA:
i

a. r r /

i C[ emph]
C[ + emph]
b. r r /

C[ + emph]

r a
c. r /

r a
These rules propose that in EA, /r/ is non-emphatic before /i/, and also
after /i/ provided that a non-emphatic consonant follows. It is emphatic
when followed or preceded by an emphatic consonant or /u/. And it is

20

either emphatic or non-emphatic in the neighborhood of /a/. EA examples

of rule (2a) are /rigiif/ a loaf and /irk-a/ a company. Examples of rule
(2b) are /mariiD/ a patient, /rukn/ a corner, /maTar/ rain, and /urb/
drinking. Rule (2c) reveals the accidental use of emphatic vs. nonemphatic versions of /r/ in the neighborhood of /a/. For example, /rabb/
Lord vs. /rama/ to throw and /marham/ an ointment vs. /ward-a/ a
rose. Further research is required to discover the exact rules governing the
use of emphatic vs. non-emphatic /r/ in EA. Since the distinction between
the emphatic and non-emphatic versions of /l/ and /r/ has no great
influence on the morphology of words, it will not be marked throughout
the book.
Regarding the distribution of consonants, it is noted that all
consonants in the two varieties may occur in the initial, medial or final
position, with the exception of the Colloquial /?/ which is rarely found in
the final position, unless it is originally /q/ in SA. All consonants may be
doubled and, in this case, will be considered geminate consonants.
As for consonant clustering, initial clusters are not found in either
SA or EA. However, all medial and final clusters are common in both
varieties. Concerning the number of consonants that may occur in a
cluster, both SA and EA allow no more than two consonants to occur
together. Geminate consonants do not occur initially but they do occur
medially and finally. And, according to Abdel-Massih (1979: 24), a
sequence of two different consonants may occur medially or finally.
1.1.2. Vowel Systems:
Six vowels are generally recognized in SA: three short /a/, /i/ and
/u/ and three long /aa/, /ii/ and /uu/. In orthography, short vowels are
represented by diacritics above or under the letter, while long vowels are
represented by the three letters /?alif/, /yaa?/ and /waaw/, respectively.
Long vowels are pronounced twice as long as their short counterparts. EA
exhibits all these six vowels and two additional long vowels /ee/ and /oo/
which are regarded in most cases as reflexes of the SA sequences /ay/ and
/aw/5, respectively. The vowels of both varieties are displayed in Table (2).
4

21

Table (2)
The Vowels of Both SA & EA
Short
Long
Front
High

Central

Back

Front

ii

uu

ee+

oo

Mid
Low

Central

Back

aa

(+) = Found in EA only.

That EA changes the SA sequences /ay/ and /aw/ into /ee/ and /oo/,
respectively, can be observed in the following examples:
SA
Sayf(-un)
Dayf-at(-un)
lawn(-un)
mawz-at(-un)

EA
Seef
Deef-a
loon
mooz-a

Gloss
summer
a guest (f)
a color
a banana

This change can be accounted for by the following rule:

(3) Monophthongization in EA:
ay
ee
/ when sharing syllable rime and stem-final.
aw
oo
The condition imposed on the rule indicates that not all SA /ay/ and /aw/
become EA /ee/ and /oo/. For instance, they are preserved without change
in two occasions. First, when the vowel is followed by a geminate glide, a
phenomenon that is linguistically called geminate inalterability. To
explain this phenomenon, Kenstowicz (1994: 410) says, geminates form
a tight bond that resists disruption by phonological rules. There are two
aspects to this integrity of geminates. First, geminates characteristically
repel insertion of an intervening segment. Second, geminates often escape
rules whose application would modify one half of the geminate while
leaving the other unchanged. Compare the following examples:
SA
mayyit(-un)
bayyaD(-a)

EA
mayyit
bayyaD

Gloss
dead
to whitewash, to paint

22

bawwaab(-un)
mawwat(-a)

bawwaab
mawwit

a doorman
to kill

Second, when the sequence of vowel and glide is in the stem-initial

syllable, as in these words:
SA
?awTaan(-un)
mawluud(-un)
?aymaan(-un)
Saydal-at(-un)

EA
?awTaan
mawluud
?aymaan
Saydal-a

Gloss
home countries
new born
oaths
pharmaceutical science

Concerning the distribution of vowels, both varieties under

investigation are identical in that no word begins with a vowel sound,
either short or long. In other words, /#V/ is impermissible in the language.
Also, short vowels occur stressed and unstressed, medially and finally.
However, EA differs from SA in that in the former long vowels occur only
stressed (Abdel-Massih 1979: 24) but in the latter they occur both stressed
and unstressed, e.g. /mafaatih(-u) > mafatih/ keys. Consequently,
although SA allows two or more long vowels in a phonological word, EA
does not permit more than one. This is an automatic consequence of the
fact that there is only one stressed vowel per word (disregarding secondary
stress). The rules responsible for the shortening of pre-final long vowels in
EA will be mentioned in (1.2.4). Nevertheless, vowel length has not been
leveled; it is still underlying in all forms in EA where it occurs in SA.
1.1.3. Syllable Structure & Stress:
An Arabic syllable may have three components: a peak, an onset
and a coda. Al-Ani & May (1973: 41) explain the nature of these
components:
The peak, or nucleus, is always the most prominent element
of the Arabic syllable. It must be composed of a vowel,
either long or short. The marginal elements, on the other
hand, contain only consonants, the onset always consisting
of a single consonant and the coda consisting of zero, one,
or two consonants.
Put differently, the following formula can be utilized to describe the
syllable structure in Arabic: C1V(V)C0-2. This can have the following
prosodic form:

23

(4)

V X (C)

where X = , C or V and (C) is extrasyllabic,

i.e. permitted only at the end of a word.
The distribution of vowels with consonants is basically similar in
SA and EA. The following five syllable patterns are found in both
varieties (cf. Mitchell 1956: 110):
1. CV: as in the first syllable in /fa-ras(-un)/ a horse,
2. CVV: as in the first syllable in /faa-ris(-un)/ a horseman,
3. CVC: as in the first syllable in /maf-huum(-un)/ understood,
4. CVVC: as in the last syllable in the stem of /ma3-luum(-un)/ known,
5. CVCC: as in the last syllable in the stem of /ka-tabt(-u)/ I wrote.
McCarthy (1985) reduces these five patterns to three quantities:
(i) Light: CV,
(ii) Heavy: CVV and CVC,
(iii) Superheavy or extraheavy: CVVC and CVCC.
The extraheavy syllables arise word-finally or before a vowel deletion site
(Kenstowicz 1994: 399).
Stress depends on the quantitative pattern of the whole word in
accordance with the following rules:
(i) In both SA and EA, the ultimate syllable is stressed if that syllable is
superheavy - i.e. terminates in a long vowel followed by a consonant
or in a short vowel followed by two consonants, as in these examples
(SA words are given in their pausal forms):
SA
manaadil
?abuh
fahmt6
gaslt6

EA
manadil
?abuh
fihmt
gaslt

Gloss
handkerchiefs
his father
I understood
I washed

24

(ii) In both varieties, the penultimate syllable is stressed in words whose

ultimate syllable is not superheavy (i.e. heavy or light), as in these
instances:
SA
mudrris
mktab
sa3-ah
mlik

EA
mudrris
mktab
sa3-a
mlik

Gloss
a teacher (m)
an office / a desk
an hour / a watch
a king

However, words of three and four syllables including a consonant

cluster between the antepenult and the penult are not stressed the same.
They are stressed before the cluster in SA and after the cluster in EA, as
in the following words (SA words are in the pausal forms):
SA
mdras-ah
mknas-ah
muhndis-ah

EA
madrs-a
makns-a
muhands-a

Gloss
a school
a broom
an engineer (f)

(iii)In both varieties, if there is an antepenultimate syllable, then it is

stressed when the ultimate syllable is not superheavy and both the
penultimate and antepenultimate syllables are light, as in these words:
SA
ktab-at
jar-ah
mlik-ah

EA
ktab-it
gar-a
mlik-a

Gloss
she wrote
a tree
a queen

1.2. Phonological Alternations:

Six phonological alternations are manifested in Arabic, both
Standard and Colloquial: epenthesis, elision, assimilation, shortening,
metathesis and glide alternation. These alternations may occur within
morphemes or at morpheme and word boundaries. An understanding of
these processes is essential to the study of the morphology of this
language. Therefore, a brief introduction to them is given below.
Phonological rules are given in this section and referred to by name
afterwards. Throughout the book, surface forms will be related to their
underlying representations and the phonological alternations involved will
be explained.

25

1.2.1. Epenthesis:
A group of three successive consonants is impossible in Arabic,
both Standard and Colloquial. Therefore, to prevent three consonants from
coming together, as when a word ending in two consonants is followed by
another word or a suffix beginning with a consonant, or vice versa, the
short high vowel /i/ is inserted at the juncture, i.e. at the end of the first
word. This vowel is called an anaptyctic vowel (Mitchell 1990: 96) and
is not heard as clearly as other vowels. The epenthesis rule in Arabic can
be represented as follows:
(5) C+CC
CC+C

CvCC
CCvC

The epenthesis occurring here is not really a condition on sequences of

consonants, but rather on the syllabification of consonants with one basic
difference between the two varieties: the epenthetic vowel occurs after the
first consonant in SA but after the second one in EA. So, it will be more
accurate to represent the epenthesis by these two rules (where the dot
represents the syllable juncture):
(6) Medial-Epenthesis in SA:
i / C C.C
(7) Medial-Epenthesis in EA:
i / C.C C
Examples of Medial-Epenthesis are given in (8) for SA and in (9) for EA:
(8) SA: a. /Darab-at + al-walad(-a)/ /Darab-ati l-walad(-a)/
She hit the boy.
b. /3an + al-bint(-i)/ /3a.ni l-bint(-i)/
about the girl.
c. /kam + as-saa3-at(-u)/ /ka.mi s-saa3-at(-u)/
What time is it?
(9) EA: a. /ba3d+bukra/ /ba3.di.bukra/
the day after tomorrow.
b. /?ult+lak/ /?ul.ti.lak/
I said to you.

26

There are special cases in which /u/ or /a/ is inserted. First, the
vowel /u/ is inserted before a pronoun or a pronominal suffix ending in
[-um] in both varieties and after the fractions 1/3 to 1/9 when in construct
with /miyya/ a hundred to form the hundreds 300 to 900 inclusive in EA.
Examples for the insertion of /u/ are given in (10) for SA and in (11) for
EA:
(10) SA: a. /?antum + al-mu3allim-uun/ /?antu.mu l.mu3allim-uun/
You are the teachers.
b. /wa 3alaykum + as-salaam/ /wa 3alayku.mu s.salaam/
And peace be upon you.
(11) EA: a. /dars + hum/ /dar.su.hum/ their lesson.
b. /uft + kum/ /uf.tu.kum/ I saw you (m).
c. /xums + miyya/ /xum.su.miyya/ five hundred.
Second, the vowel /a/ is inserted in SA after /min/ when followed by the
definite article [al-], as in (12); and in EA before the pronominal suffix
[-ha], as in (13):
(12) SA: /min + al-bayt(-i)/ /mi.na l.bayt(-i)/ from the house.
(13) EA: /ism + ha/ /?is.ma.ha/ her name.
One can account for the special cases of insertion of /u/ and /a/ by
proposing that the words or suffixes involved have a harmonic ghost or
latent vowel (Zoll 1996: 31) which is not phonetically present when
those words or suffixes are pronounced in isolation but appears in
connected speech. That harmonic ghost vowel is /u/ when preceded or
followed by /u/, to secure vowel harmony, as in suffixes containing /u/ in
both varieties and in the hundreds containing fractions with the form
[Fu3L] in EA. Otherwise, the harmonic ghost vowel is /a/, as after /min/ in
SA and before [-ha] in EA.
The idea of the ghost vowel can be applied to the /a/ in the SA
definite article [al-] which appears in initial positions and in isolation but
disappears in connected speech. When there is a battle between two ghosts
or between a ghost and the regular epenthetic vowel /i/, the first of them
wins, as in (8), (10) and (12) above.
To prevent an initial consonant cluster, both varieties require
word-initial epenthesis of the short high vowel /i/. This was argued for

27

the Colloquial variety by Broselow (1976: 20). Her rule for this epenthesis
can be reformulated as follows:
(14) Word-Initial Epenthesis:
i / # CC
Rule (14) applies in both varieties for imperatives with an initial consonant
cluster in the stem as well as to derived verbs with initial clusters, among
other cases. Examples of that rule are:
SA
(?i)rab
(?i)l3ab
(?i)nkasar(-a)
(?i)sta3mal(-a)

EA
(?i)rab
(?i)l3ab
(?i)nkasar
(?i)sta3mal

Gloss
drink!
play!
it broke
he used

These examples also represent the insertion of /?/ to protect the

phonotactic structure of words: no word should begin with a vowel (See
below).
When the consonant cluster is no longer initial, epenthesis does not
apply. Thus, SA derived verbs with epenthetic /i/ do not have that vowel
when the cluster is preceded by a vowel in the phrase domain, as in (15):
(15) SA: a. /al-fariiq-u + (i)nhazam/ /?al-fariiq-u nhazam/
The team was defeated.
b. /al-walad-u + (i)tagal/ /?al-walad-u tagal/
The boy worked.
Similarly, the epenthesis process does not apply in EA when the cluster is
no longer initial, e.g. when it is preceded by the negative marker [ma-], as
in (16):
(16) EA: a. /ma + (i)nhazam + / /ma-nhazam-/ It was not defeated.
b. /ma + (i)taal + / /ma-taal-/ He did not work.
In addition to the epenthesis of high vowels there are cases of
epenthesis of /?/. Traditional Arab grammarians (e.g. Al-Hammadi et al.
1977) distinguished between two kinds of glottal stop: deletable and nondeletable. This has led some modern linguists such as Mitchell (1990) to
assume the elision of initial /?/ of a word in isolation when it becomes
medial in the phrase domain. They give examples like those in (17):

28

(17) SA: /?az-zujaaj-u + ?inkasar/ /?az-zujaaj-u nkasar/

The glass was broken.
EA: /uf-t + ?abuuk/ /uf-t abuuk/
I saw your father.
However, careful investigation of the matter shows that Broselow (1976:
23) was right in proposing that there is epenthesis rather than elision of
glottal stop in contexts like those above. The argument for this is that if
speakers of Arabic must memorize for each morpheme whether it begins
with a deletable glottal stop or a non-deletable one, then it is required that
there be two varieties of glottal stop in speakers memories. This is an
unfortunate move because, in general, speakers memorize morphemes
(store them in memories) using only phonetic information. Since there is
only one glottal stop phonetically, the distinction in memory cannot be
between two varieties of the glottal stop. The most plausible alternative is
that morphemes whose glottal stop is said to be deleted are in fact stored
in memory without a glottal stop. For these, provided that no preceding
word can provide a syllable onset, a glottal stop is inserted in response to a
general prohibition on syllables lacking an onset. Since every syllable
requires an onset (See 1.1.3 above), any syllable lacking an onset will have
to acquire one, and the glottal stop is inserted to meet this requirement:
(18) Glottal Stop Insertion:
? / # V
So, the examples in (17) above will be more precisely represented as in
(19):
(19) SA: /inkasar(-a)/ /?inkasar(-a)/ when not preceded by C
EA: /abuuk/ /?abuuk/ when not preceded by C
On the other hand, morphemes with undeletable glottal stops have glottal
stops in their underlying (memorized) forms, as in the underlined words in
(20):
(20) SA: ?inna
?awlaad-a
r-rajul-i
?akiyaa?(-u)
verily
sons-Acc
the-man-Gen intelligent(Nom)
Verily, the sons of the man are intelligent.
EA: ya ?aanis-a
O, Miss!

29

The definite article, in particular, has no glottal stop in its

underlying form. Speakers memorize it as [al- > l-] then insert /?/ in some
contexts. Surface /?/ is inserted when there is no preceding consonant in
the relevant phrase domain to form an onset, as in (21) and (22):
(21) SA: a. /al-kitaab-u/
b. /ar-rajul-u/
(22) EA: a. /il-laban/
b. /i-ams/

/?al-kitaab(-u)/ the book

/?ar-rajul(-u)/ the man
/?il-laban/ the milk
/?i-ams/ the sun

1.2.2. Elision:
A remarkable difference between SA and EA is that only in the
latter are the short high vowels /i/ and /u/ elided in medial open syllables
in the word or phrase domain. This fact has been captured by a high vowel
deletion rule proposed by Broselow (1976: 20). This rule can be
reformulated as follows:
(23) High-Vowel Deletion (= Syncope) in EA:
V
+ hi

/ VC CV
long

stress
Mitchell (1956: 113) refers to two contexts involving the elision of
/i/ and /u/ in this Colloquial variety:
First, if a suffix beginning with a vowel is attached to a word of which the
ultimate syllable is of the type /CiC/ or /CuC/ and the penultimate syllable
is open, i.e. ending in a vowel, then the /i/ or /u/ of the ultimate syllable is
almost constantly elided. For example,
(24) EA: a. /kaatib/ a writer (m)
b. /yaaxud/ he takes

/katb-a/ a writer (f).

/yaxd-u/ he takes it (m).

Second, the vowels /i/ and /u/ occurring in a short and unstressed syllable
are elided when that syllable becomes medial in a phonological phrase and
the preceding word or prefix ends in a vowel. For instance,
(25) EA: a. /huSaan/ a horse /?abu hSaan/ the man with a horse.
b. /kitaab/ a book /da ktaab/ This is a book.

30

Another major difference between SA and EA is that a glottal stop

at the end of an SA word disappears in EA. This can be accounted for by
the rule in (26). Examples of this rule are provided below:
(26) Final /?/ Deletion in EA:
?/#
SA
samaa?(-un)
itaa?(-un)
mala?(-a)
waraa?(-a)

EA
sama
ita
mala
wara

Gloss
sky
winter
to fill
behind

The glottal stop is also lost in EA when it forms all or part of the coda of a
syllable. This leads to the following changes:
(27) a? aa
i? ii
u? uu
Such changes can be accounted for by the rule in (28) which indicates that
a glottal stop becomes similar to the preceding vowel if both come
together at the end of a syllable. Examples of the rule are given afterwards:
(28) Compensatory Lengthening in EA:
? Vi / Vi ]
SA
fa?r(-un)
a?r(-un)
bi?r(-un)

EA
faar
taar
biir

Gloss
a mouse
revenge
a well

An exception here is the noun /bu?r-at(-un) > bu?r-a/ focus which is

preserved in EA. It should also be asserted that the change of SA /q/ into
EA /?/ is ordered after rules affecting underlying /?/. Thus, Compensatory
Lengthening does not apply in words like /faqr(-un) > fa?r/ poverty.
There is one case in which this rule applies in SA: the initial-glottalized
verbs of Form IV (See 2.1.1.3).

31

1.2.3. Assimilation:
There are certain cases in which Arabic manifests complete
assimilation of juxtaposed consonants. The most famous example of
complete assimilation in SA is that of the lateral consonant /l/ of the
definite article, which becomes identical to the initial consonant of the
word if it is one of the so-called /amsiyy-at(-un)/ solar consonants7.
These are /t, d, T, D, , , s, z, , S, Z, r, l and n/. But there is no
assimilation in the case of the so-called /qamariyy-at(-un)/ lunar
consonants which are the remaining consonants. To account for this
division of consonants one must resort to Kenstowiczs (1994: 163)
classification of Arabic consonants below:
(29) a. labials
b. coronal sonorants
c. coronal stops
d. coronal fricatives
e. dorsals
f. gutturals

[f, b, m]
[l, r, n]
[t, d, T, D]
[, , s, z, S, Z, ]
[g, k, q]
[x, g, h, 3, h, ?]

By examination of the solar consonants stated above, it appears that they

are all [+coronal], i.e. produced with the front part of the tongue. The rule
responsible for the assimilation of /l/ in the definite article is called
l-Assimilation (Brame 1970: 20):
(30) l-Assimilation:
Ci
l [+def] Ci /

+ cor

This rule shows that the /l/ of the definite article is assimilated to the
following consonant if it is coronal. To see the difference between
assimilated and non-assimilated /l/, compare these examples in SA:
(31) a. /al+tilmii(-u)/ /?at-tilmii(-u)/ Not */?al-tilmii(-u)/ the pupil.
b. /al+kurat(-u)/ /?al-kurat(-u)/ Not */?ak-kurat(-u)/ the ball.
The same process occurs in EA. However, to the solar consonants stated
above, EA adds /g/ and very rarely /k/, in which cases the assimilation of
the /l/ is optional. This can be explained by assuming that EA adds the
[+dorsal] consonants which are produced with the body of the tongue to
the [+coronal] ones, and that /q/ is excluded because in the register in

32

which /q/ appears at all, the optional rule would never apply. Compare
(32a) with (32b) in EA:
(32) a. /il+gamal/

/?ig-gamal ~ ?il-gamal/ the camel.

b. /il+kursi/

/?ik-kursi ~ ?il-kursi/ the chair.

Another case of complete assimilation across word boundaries

occurring regularly in SA and EA is the assimilation of /n/ to a following
sonorant consonant. The rule for this process can be formulated as in (33):
(33) n-Assimilation:
Ci

n Ci / + son
+ cons
An SA example of this process was offered by Mitchell (1990: 90): the
particle min may be heard in the form mir before r, mil before l and mim
before m, as in mir rahmti llah from the mercy of God, mil lndun
[sic] from London, mimmaa (< min maa) from that which. Other
examples from SA are given in (34) and examples from EA are seen in
(35):
(34) SA: a. /3an + ramzii/ /3ar ramzii/ about Ramzy.
b. /3an + laylaa/ /3al laylaa/ about Laila.
c. /3an + muhammad/ /3am muhammad/ about Muhammad.
(35) EA: a. /min + ramsiis/ /mir ramsiis/ from Ramses.
b. /min + libnaan/ /mil libnaan / from Lebanon.
c. /min + maryam/ /mim maryam/ from Mariam.
A third case of complete assimilation across word boundaries is the
assimilation of an obstruent coronal consonant to a following consonant
that is identical in place of articulation to that consonant. This process can
be formulated in the form of a rule in (36) and examples from SA are
given in (37):

33

(36) Coronal-Assimilation:
C

+ obstr Ci /
+ cor

Ci

+ obstr
+ cor

(37) SA: a. /qaal-at + Taalib-at(-un)/ /qaal-aT Taalib-at(-un)/

A female student said.
b. /?ajaab-at + da3wat-ii/ /?ajaab-ad da3wat-ii/
She accepted my invitation.
Two cases of assimilation are present in EA only. The first
involves the change of the glottal stop /?/ to the high glide /y/ when it is
followed by /i/ and preceded by any vowel, i.e. the following change is
effected:
(38) V?i Vyi
This change can be represented by the rule in (39). Its examples are given
below:
(39) Intervocalic /?/ Assimilation in EA:
cons
cons
+ hi
+ hi
?
/V

rd
rd
SA
naa?im(-un)
3aa?im(-un)
Saa?im(-un)

EA
naayim
3aayim
Saayim

Gloss
sleeping
swimming
fasting

However, Brame (1970: 245) proposes a rule of Glottal Formation which

changes a glide into a glottal stop after /aa/ and before a vowel in SA as
follows:
(40) Glottal Formation in SA:
G ? / aa V
This means that the innovation in EA is simply in no longer having such a
rule.

34

1.2.4. Shortening:
In both varieties, a long vowel does not remain long in a closed
syllable, i.e. one ending in a consonant. However, in EA word-final
syllables ending in a single consonant represent an exception to be
discussed below. This is a consequence of a limited syllable template in
non-final positions. Long vowels are not permitted in closed syllables
(except word-finally in EA). The rule that captures this can be called
Closed-Syllable Shortening:
(41) Closed-Syllable Shortening:
(V)VV (V)V/ C]
This rule indicates that an extra-long vowel becomes a long one and a long
vowel becomes a short one in a closed syllable. The only difference
between the two varieties in this respect is that a final consonant in EA
does not count for syllable weight calculation but in SA it does. This can
be accounted for by a special Extrametricality Condition in EA, as
follows:
(42) Extrametricality Condition in EA:
Final C does not count for syllable weight calculation
The difference between the two varieties can be noted by comparing the
acceptability of these consonant-vowel sequences in both of them:
SA
*CVVC#
*CVVCC#
*CVVC.CV...

EA
CVVC#
*CVVCC#
*CVVC.CV...

Examples of Closed-Syllable Shortening are given in (43) for SA and in

(44) for EA. (The first change in the SA examples is produced by an
Identical-Consonant Metathesis rule that will be explained in 1.2.5):
(43) SA: a. /(?i)nsadad(-a) (?i)nsaadd(-a)/ (?i)nsadd(-a)/
to be blocked
b. /(?i)htajaj(-a) (?i)htaajj(-a) (?i)htajj(-a)/
to object
(44) EA: a. /?uul + li/

b. /kitaab + hum/

/?ul-li/ Say to me!

/kitab-hum/8 their book

35

A remarkable difference between the two varieties is that SA long

vowels are shortened in EA when occurring in final position in accordance
with the rule in (45). Its illustrative examples are presented below:
(45) Final-Vowel Shortening in EA:
VV V / #
SA
Sallaa
katab-uu
ya-mii

EA
Salla
katab-u
yi-mi

Gloss
he prayed
they wrote
he walks

However, the shortening rule does not apply before a suffix beginning
with a consonant because the vowel is no longer final. In other words, SA
long vowels are retained long in EA before suffixes beginning with a
consonant, as in the following instances:
SA
?iksirii-h
ramaa-haa
?ilguu-hum

EA
?iksirii-h
ramaa-ha
?ilguu-hum

Gloss
you (fsg) break it (m)
he threw it (f)
Cancel (2mpl) them!

Another difference between the two varieties is that in the case of

two EA long vowels (often resulting from morphological suffixation), the
first vowel is shortened. This is a result of the fact that only one long
vowel per word is permitted in EA whereas SA may permit more than one
long vowel per word. It is also a consequence of the shortening of
unstressed vowels since all unstressed vowels in EA should be short. This
can be accounted for by a rule that can be called Atonic Shortening in
EA. Compare the following examples:
(46) Atonic Shortening in EA:
VV V / when stressless
SA
miizan(-un)
Taabur(-un)
Tuufan(-un)
baa3u-h
Saadu-haa
kitaab-yni
xabbaaz-ina

EA
mizan
Tabur
Tufan
ba3u-h
Sadu-ha
kitab-en
xabbaz-in

Gloss
balance
a queue
flood
they sold it(m)
they caught it (f)
two books
bakers

36

In a few nouns with the nominal template [CayCVVC] the

Standard sequence /ay/ is replaced by /i/, which is the short counterpart of
/ee/, and in fewer instances by /a/ in the Colloquial variety:
SA
maydan(-un)
ayTan(-un)
rayhan(-un)
zaytun(-un)
laymun(-un)

EA
midan
iTan
rihan
zatun
lamun

Gloss
a square
a devil
myrtle
olives
lemon

The change of SA [Fay3aaL(-un)] into EA [Fi3aaL] can be accounted for

by proposing that the rule of Atonic Shortening applies after
Monophthongization in EA. Thus, the change of SA /maydan(-un)/ into
EA /midan/ runs as follows:
(47) maydan(-un) meedan midan.
1.2.5. Metathesis:
A very important phonological process which applies in both
varieties is the metathesis of identical consonants, which operates in
geminate roots when they are followed by a vowel. The rule of IdenticalConsonant Metathesis, given in (48), was first proposed by Brame (1970:
122):
(48) Identical-Consonant Metathesis:
CkVCkV VCkCkV
This rule shows that when two identical consonants are separated by a
vowel and followed by a vowel, we metathesize the first consonant and the
first vowel so that the consonants come together before the vowels. It is
apparent in the following examples:
Underlying
adad(-a)
adad(-a)
?aTbibaa?(-u)
?ahbibaa?(-u)

SA
madd(-a)
add(-a)
?aTibbaa?(-u)
?ahibbaa?(-u)

EA
madd
add
?aTibba(a?)
?ahibba(a?)

Gloss
to stretch
to pull
doctors
beloved ones

37

1.2.6. Glide Alternations:

The high glides /w/ and /y/ play an essential role in the phonology
of the Arabic language because they are susceptible to various
phonological processes. Many alternations occur when these glides are
juxtaposed with vowels. The underlying vowel and glide sequences and
their surface phonetic forms are first listed in (49). Then, a discussion of
how the surface forms are derived from their underlying counterparts in
terms of phonological rules is provided:
(49) Surface forms of underlying vowel and glide sequences:
- ayi ai aa
- ayu au aa
- aya aa
- awi ai aa
- awu au aa
- awa aa
- iyi ii
- iyu iu ii
- iyuu uu
- iya iya
- iwi ii
- iwu iu ii
- iwa iya
- Cwi SA Cii, EA Cwi ~ Cii
- Cyi SA Cii, EA Cyi ~ Cii
- Cwu Cuu
- Cyu Cii
- Cwa SA Caa, EA Cwa ~ Caa
- Cya SA Caa, EA Cya ~ Caa

- uyi ui ii
- uyu uu
- uya uwa
- uwi ui ii
- uwu uu
- uwa uwa
- iw. ii
- iy. ii
- uw. uu
- uy. ii
- aw. aw > oo
- ay. ay > ee

In the sequences /iya/ and /uwa/ the surface forms are identical to their
underlying representations. So, no phonological processes are involved
there, as in the SA verb /laqiy(-a)/ to meet (pf. indic) and /ya-d3uw(-a)/
to invite (impf. subj) where no change occurs. When the high vowels /i/
and /u/ are immediately followed by the low vowel /a/ a rule of Onset
Formation inserts a glide of the same nature of the first high vowel to
prevent hiatus, i.e. the occurrence of two consecutive vowels forming
separate syllables (Trask 1996: 170). The prosodic representation of this
phonological process is given in (50) and the rule is formulated in (51):

38

(50)

i
iya

a
uwa

(51) Onset Formation:

y i
/ a
w u
Thus, the transcriptional conventions of the high vowels can be listed as
follows:
/i/ in onset = y, e.g. /yadaw-iyy(-un)/ handmade
/i/ in nucleus = i, e.g. /birk-at(-un)/ a pool
/i/ in coda = y or i, e.g. /biyD(-un)/ = /biiD(-un)/ white (pl)
/u/ in onset = w, e.g. /walad(-un)/ a boy
/u/ in nucleus = u, e.g. /ukran/ Thanks!
/u/ in coda = w or u, e.g. /suwd(-un)/ = /suud(-un)/ black(pl).
For the sequences /aw./ and /ay./, the surface forms are identical to their
underlying forms in SA, but in EA the rule of Monophthongization
changes them into /oo./ and /ee./, respectively, in certain circumstances.
In most of the changes /VGV VV/, there is a process of glide
elision that can be formulated in the following rule (Brame 1970: 65):
(52) Glide Elision:
G / Vi Vj

if j = [+1o], then i = [+1o]

This rule indicates that we delete a glide when it occurs between two
vowels with one condition: if the second vowel is low, the first must also
be low. When the long vowel outputted from this rule is followed by a
consonant and a word boundary or two consonants, the rule of ClosedSyllable Shortening applies in SA to shorten that vowel.
The Glide Elision rule applies in both varieties and accounts for the
following alternations:

39

1- aGa aa:
ramay-a ramaa > rama to throw and
da3aw-a da3aa > da3a to invite.
2- iGi ii:
raamiy-in
raamiin
raamin
raami

daa3iw-in (Gen)
daa3iin
by Glide Elision
daa3in
by Closed-Syllable Shortening in SA
daa3i
by Final-V Shortening in EA

3- uGu uu:
e.g. SA /ya-d3uw-u ya-d3uu/ he invites.
4- iGu iu
5- aGu au
6- uGi ui
In the last three changes the output of the Glide Elision rule undergoes
assimilation processes that can be accounted for by these rules (The term
vocoid is used here to encompass both vowel and glide):
(53) u-to-i:
Cy

u i / i
y

(54) Perseverative Vocoid Assimilation:

cons
+ hi Vi / Vi

The rule u-to-i, which must be ordered before Perseverative Vocoid

Assimilation, indicates that /u/ becomes /i/ after a consonant followed by
/y/ and also before /i/ or y/. It accounts for these alternations:
1- ui ii:
buyi3-a
quwil-a
bui3-a
quil-a
by Glide Elision
bii3-a
qiil-a
by u-to-i
2- uy. ii:
buyD(-un) biyD(-un) = biiD(-un) white (pl)

40

3- Cyu Cii:
mabyuu3 SA mabii3 sold
Perseverative Vocoid Assimilation proposes that a high vowel or glide
becomes similar to its preceding vowel. It leads to these changes:
1- iu ii:
raamiy-un
raami-un
raami-in
raamin
raami

daa3iw-un
daa3i-un
daa3i-in
daa3in
daa3i

2- au aa:
ya-lqay-u
ya-lqa-u
ya-lqaa

yi-l?a

maqhaw-un (Nom)
maqha-un
by Glide Elision
maqhaan
by Perseverative Vocoid Assim.
maqhan
by Closed-Syllable Shortening in SA
maqha
by Final-V Shortening in EA

3- ai aa:
maqhaw-in (Gen)
maqha-in
maqhaan
maqhan
maqha

by Glide Elision
by Perseverative Vocoid Assim.
by Closed-Syllable Shortening in SA
by Final-V Shortening in EA

by Glide Elision
by Perseverative Vocoid Assimilation
by Closed-Syllable Shortening in SA
by Final-V Shortening in EA

4- iw. ii:
raDiw-tu raDii-tu > riDii-t I became satisfied
5- iy. ii:
hiyl-at hiil-at(-un) a trick
6- uw. uu:
Suwr-at Suur-at(-un) a picture
Now turning to the sequences /CGV CVV/, one can propose the
following assimilation rule:
(55) Anticipatory Vocoid Assimilation:

41

cons
+ hi Vi / C Vi]

This rule shows that a high vowel or glide preceded by a consonant is

assimilated to the following vowel in the same syllable. It takes care of
these alternations:
1- CGi SA Cii, EA CGi ~ Cii:
ya-syir-u
yu-hwil-u
ya-siir-u
yu-hiil-u

yi-hwil ~ yi-hiil

by Anticipatory Vocoid Assim.

in EA

In EA, both /yi-hwil/ and /yi-hiil/ are found, which suggests that the rule
of Anticipatory Vocoid Assimilation may be in the process of being lost in
this variety.
2- CGu Cuu:
ya-kwun-u
ya-kuun-u
yi-kuun

by Anticipatory Vocoid Assimilation

in EA

3- CGa SA Caa, EA CGa ~ Caa:

ma-syar-un ma-kwan-u
ma-saar-un
ma-kaan-un by Anticipatory Vocoid Assimilation
ma-saar
ma-kaan
in EA
In EA, the glide is preserved in some words, e.g. /(?i)sta-bwax/ to
consider silly and /(?i)sta-yax/ to pretend to be a sheik, which again
suggests that the rule of Anticipatory Vocoid Assimilation is about to be
lost.
1.3. Morphological Preliminaries:
In this section, two important aspects of Arabic morphology will
be handled: word classification (1.3.1) and word formation (1.3.2). These
two aspects are essential to the understanding of Arabic morphology in
general. The first is concerned with the part-of-speech system of the
language and the other is related to the processes involved in the formation
of words in the language.

42

1.3.1. Word Classification:

Arab grammarians, since the days of Sibawayh (d. 798), have
distinguished three word classes: nouns, verbs and particles. The class of
nouns includes substantives (e.g. /bayt(-un)/ a house), pronouns (e.g.
/?ana/ I), adjectives (e.g. /kariim(-un)/ generous), adverbs (e.g. /faqaT/
only) and numerals (e.g. /waahid(-un)/ one). Verbs are either perfective
(e.g. /jalas(-a)/ he sat) or imperfective (e.g. /ya-jlis(-u)/ he sits) (For
details, see 2.2). Particles comprise prepositions (e.g. /min/ from),
conjunctions (e.g. /wa/ and) and interjections (e.g. /?aah/ oh). Owens
(1988: 125) states that in justifying this classification Arab grammarians
drew on arguments from all levels of analysis -- phonological,
morphological [, syntactic] and semantic. However, since this book is
concerned with the morphological aspects, only the morphological criteria
will be discussed here.
According to Owens (1988: 126), SA nouns are morphologically
distinguished by the following criteria:
(i) occurrence with the definite article, e.g. /al-walad(-u)/ the boy,
(ii)occurrence with the indefinite suffix [-n] or what the Arabs call
/tanwiin/ nunation, e.g. /walad(-un)/ a boy9,
(iii) dualization, e.g. /walad-aani/ two boys,
(iv) pluralization, e.g. /?awlaad(-un)/ boys,
(v) diminutive formation, e.g. /wulayd(-un)/ a small boy, and
(vi) relational adjective formation, e.g. /balad-iyy(-un)/ native.
SA Verbs are characterized by two morphological criteria:
(i) occurrence with the pronominal suffixes: [-aa] dual, [-uu] plural,
[-u] I and the feminine marker [-at], e.g. /daxal-aa/ they (two)
entered, /daxal-uu/ they entered, /daxal-tu/ I entered and /daxal-at/
she entered, and
(ii) having a set of related morphological forms, e.g. /daxal(-a)/ he
entered, /ya-dxul(-u)/ he enters, /daaxil(-un)/ entering,
/madxal(-un)/ an entrance, etc.
Particles are defined as what have neither the characteristics of nouns nor
those of verbs (cf. Owens 1988: 128).
Thus, Arab grammarians have not distinguished adjectives from
nouns except in context. They have not shown any formal distinction
between the two classes. Moreover, they have grouped pronouns with

43

nouns because the former represent syntactic substitutes for the latter or
fill their slots in syntactic structures. Furthermore, they have oscillated
numerals between nouns and adjectives.
However, in modern linguistics, clear-cut distinctions between
word classes have become necessary. Hence, Aboul-Fetouh (1969: 35), for
instance, has recognized six form classes in Arabic: verbs, nouns,
adjectives, pronouns, adverbs and particles. Abdel-Malek (1972: 52) dealt
with these classes under four headings: verbs, nouns, adjectives and
closed-list classes. By closed-list classes he meant those to which no
new members can be added. This classification will be endorsed in this
book. Therefore, pronouns and adverbs will be discussed in the chapter on
closed-list classes, though some types of adverbs are not closed.
Nevertheless, it must be stressed that a word may have more than
one syntactic function. For example, the Standard form /al-miSriyy(-u)/ is
treated as a noun in /?al-miSriyy-u yu-hibb-u waTan-a-hu/ The Egyptian
likes his homeland, but as an adjective in /?a-a3b-u l-miSriyy-u
kariim-un/ The Egyptian people are generous. The Colloquial form
/kitiir/ is considered an adjective in /burtu?aan kitiir/ a lot of oranges, but
an adverb in /bi-yi-l3ab kitiir/ He plays a lot.
1.3.2. Word Formation:
Three basic elements are distinguished in the formation of an
Arabic word: a consonantal root, a vowel pattern and one or more affixes.
Upon those three segmental morphemes, an autosegmental or prosodic
template is imposed. The vowel pattern is also called a vocalism or vocalic
melody. The root and pattern system represents a basic morphological
characteristic in the structure of Arabic words. The vast majority of words
in SA and EA have a basic stem which consists of two interrelated parts: a
root consisting typically of three consonants and a pattern of vowels fitting
around the consonants of the root (cf. Al-Toma 1969: 31). If a derivational
affix is added to the stem it will be called a derived stem. That is why
Ghaly (1960: 9) defines a stem as a combination of one root + one pattern
+ the optional occurrence of one to three derivational affixes.
The root is a consonantal morpheme which is usually
discontinuous or nonsequential. ... any one member of this consonantal
form is a radical (Ghaly 1960: 15). It usually has a moderately
unchanging lexical meaning, e.g. [-r-b] is related to drinking and [l-3-b]
is associated with playing. This meaning serves as the basis of many
related meanings which are represented by the root consonants put in

44

different vowel patterns. The vocalism is a vocalic morpheme which may

consist of one or more vowels. If it has more than one vowel, it is, by
definition, a discontinuous morpheme, since the language does not permit
two vowels in succession. While the root carries the core lexical
meaning (Schmidt 1975: 108), the vowel pattern usually has the function
of determining the syntactic role of the stem, perfect or imperfect aspect of
verb, singular or plural number of noun, etc. For example, from the root
[k-t-b] one can have /katab(-a)/ he wrote vs. /ya-ktub(-u)/ he writes and
/kaatib(-un)/ a writer vs. /kuttaab(-un)/ writers.
This notion of interrelation between the root and the pattern led to
their being called discontinuous morphemes. This is a term which was
first stated in an article by Harris (1945) and expounded in a later book of
his (1958). It was also referred to by Hill (1958: 101): another class of
nonsequential morphemes are the dovetailed, discontinuous morphemes
characteristic of Semitic languages.
This root and pattern system is preserved in the Colloquial variety
with one basic difference: the vowel patterns retained in EA are not always
identical to those of SA. For instance, the SA form [FawL-at(-un)] is
changed to [Fool-a] in EA, as in /kawm-at(-un)/ a pile that is replaced by
/koom-a/ in the Colloquial variety. The /aw oo/ alternation was
discussed earlier in (1.1.2). The SA final /t/ is lost in EA.
The consonantal root is referred to by associating it with the
unmarked vowel pattern [-a-a-a], as in the perfect active third person
masculine singular form [Fa3aL(-a)] to do. Hence, /daxal(-a)/ he
entered is the form used for the lexical entry of [d-x-l] whose basic
meaning is related to entering. Translations of /daxal(-a)/, as the
unmarked verb, into English are often made in the infinitive form to
enter rather than he entered. Concerning the number of consonants that
may occur in a root, three is the unmarked number (Owens 1988: 95). So,
the overwhelming majority of roots is triradical, i.e. each root consists of
three consonants such as [q-t-l] to kill. Less frequent are the
quadriradical roots consisting of four consonants such as [b-3--r] to
scatter. The least frequent type of roots is the biradical roots which have
only two consonants, even though they are realized with three consonants
in EA, e.g. /?ab(-un) > ?abb/ a father and /dam(-un) > damm/ blood.
Those roots are restricted to a particular sector of the lexicon (mainly kin
terms and body parts) and they tend to be augmented in productive
morphology, e.g. /?abaw-aani > ?abaw-een/ two fathers and
/damaw-iyy(un)/ bloody (cf. Kenstowicz 1994: 297). Thus, it can be

45

proposed that biradicals are underlyingly triradicals with their final

consonants deleted in the surface form. That consonant reappears on the
surface in productive morphology. Nouns have three, four or five
consonants. Verbs usually contain three or four consonants.
The vocalic pattern is referred to by filling it in with the root
consonants of the verb whose meaning is the most general in reference to
activity [F-3-L] to do (Travis 1979: 30). So, the comparative adjective
can be named by the form [?aF3aL(-u)], the active participle by
[Faa3iL(-un)], the passive participle by [maF3uuL(-un)], etc. Certain
patterns are associated with certain meanings. To illustrate,
[Fa33aaL(-un)] characterizes a person by a constant occupation or
behavior, as in the words /haddaad(-un)/ a smith and /kaaab(-un) >
kaddaab/ a liar. Certain word classes are associated with certain vowel
patterns although some patterns belong to more than one word-class.
In the root and pattern system, distinction is made between basic
and non-basic sounds in a word (Owens 1988: 99-100). The basic sounds
are represented by [F-3-L] so that, for instance, the verb /daxal(-a)/ to
enter would have the form [Fa3aL(-a)], the noun /karam(-un)/
generosity the form [Fa3aL(-un)] and so on. In the case of words with
four or five consonants, the last basic representative character [L] is
reduplicated, e.g. /bulbul(-un)/ a nightingale is represented as
[Fu3L1uL2(-un)]. The non-basic sounds are represented by themselves. For
instance, /?a-xraj (-a)/ to cause to get out is [?a-F3aL(-a)] where the [?a-]
is a causative prefix. All Arabic sounds can function as basic sounds but
only some of them can function as non-basic sounds as well. These are /s ?
l t m n h w y uu ii aa/. It is difficult to gather these phonemes in one
natural class and there is no explanation yet for the choice of these
particular consonants to be non-basic, at least as far as I know.
The affixes used in Arabic word formation may be prefixes,
suffixes, or infixes. Some of these affixes are derivational and some are
inflectional. Derivational affixes are generally those that change the
grammatical class of a word, whereas inflectional affixes are those that
signal the grammatical relationship of a word in a sentence (Crystal 1985).
However, some derivational affixes do not change the word class of the
stem they are attached to, e.g. /?adxal(-a)/ to cause to enter is derived
from /daxal(-a)/ to enter and both of them are verbs. Derivational affixes
combine with roots and vocalisms to form derived stems. That is why they
are sometimes called stem affixes (Ghaly 1960: 15). Inflectional affixes,
on the other hand, occur with the stem, either primary or derived.

46

Derivational prefixes can be exemplified by the prefix [ma-] which

forms nouns of place and time on the pattern [ma-F3aL(-un) ~
ma-F3iL(-un)]. Derivational suffixes can be illustrated by the relational
ending [-i(y)] that forms relational adjectives. And derivational infixes can
be represented by the gemination of the second consonant in a verb to give
a causative/intensive meaning. For instance, the gemination in /kassar(-a)/
to smash is considered a derivational infix used to derive that verb from
its primary counterpart /kasar(-a)/ to break. Examples of inflectional
prefixes are the imperfective prefixes [?a-], [na-], [ya-] and [ta-] which
accompany the change of the verb stem from the perfect to the imperfect
aspect. Instances of inflectional suffixes are the case endings [-u(n)],
[-a(n)] and [-i(n)] which mark the cases of nouns.
Five methods are employed in the formation of Arabic words:
derivation, analogy, coinage, compounding and Arabicization. The first
method is called /al-?itiqaaq(-u)/ derivation. It involves the derivation of
one word (of any class) from another with a slight change in form and
meaning. Hence, the following derivations are possible:
(a) Nouns might be derived from verbs (e.g. /maktab(-un)/ an
office/desk is derived from /katab(-a)/ to write) or from adjectives
(e.g. /xuDr-at(-un)/ greenness is derived from /?axDar(-u)/ green).
(b) Verbs may be derived from all other word-classes; they may be
derived from other verbs (e.g. /?axrag(-a)/ to cause to get out is
derived from /xarag(-a)/ to go out), from adjectives (e.g.
/?ihmarr(-a)/ to become red is derived from /?ahmar(-u)/ red),
from particles (e.g. /sawwaf(-a)/ to put off is derived from the future
particle /sawfa/ will), or from nouns (e.g. /maSSar(-a)/ to
Egyptianize is derived from /miSr(-u) > maSr/ Egypt).
(c) Adjectives may be derived from verbs (e.g. /maksuur(-un)/
broken is derived from /kasar(-a)/ to break) or from nouns (e.g.
/ra?iis-iyy(-un)/ main is derived from /ra?iis(-un)/ a head/chief).
(d) Adverbs can be derived from adjectives or nouns in SA and from
nouns only in EA (e.g. SA /sarii3-an/ quickly is derived from
/sarii3(-un)/ quick and /bi-quwwat(-in)/ strongly is taken from
/quwwat(-un)/ strength. EA /bi-sur3-a/ quickly is derived from
/sur3-a/ speed).
The impossible conversions are nouns from particles or adverbs; particles
from any other word-class; adjectives from particles or adverbs; and
adverbs from verbs or particles. The discussion above illustrates that some
templates are derivationally related to others and some are derived from

47

the root. However, it cannot be assumed that a particular root has a certain
category since a root can be used in various categories.
The second method of word formation is /al-qiyaas(-u)/ analogy.
It means deriving new forms that parallel already existing ones. This
technique is utilized in the derivation of new vocabulary to meet the
requirements of science and technology. Travis (1979: 31) gave an
example of an Egyptian engineer who could derive 196 terms on the field
of metallurgy from the root [S-h-r] to melt, liquefy or fuse. This method
is not structurally different from the first one. The only difference is that
the words derived by analogy are not part of the conventional lexicon at
the time of first usage.
The third technique of word formation in Arabic is termed
/an-naht(-u)/ coinage. It refers to the formation of one word from the
sounds of a phrase or sentence to give the meaning of that phrase or
sentence. Examples are /basmal(-a)/ to utter the invocation: /bi-smi ^
llaahi ^ rahmaani ^ r-rahiim/ in the name of Allah, the Beneficent, the
Merciful and /hallal(-a)/ to say the words: /laa ?ilaah-a ?illa ^ llaah/
there is no god but Allah; to rejoice (cf. Travis 1979: 42). The only
formal criterion by which one can determine that a phrase has been
changed into a word is the phonology; the phonemes of the derived word
are the salient phonemes in the original phrase.
The fourth technique of word formation is /at-tarkiib(-u)/
compounding or periphrasis. It indicates the joining of two words
together to form a unit that is like a single word (Owens 1988: 123).
Examples are the expressions using kinship terms such as /?ibnu s-sabiil-i/
son of the road = a traveler and /?axawaat-u kaan-a/ sisters of the verb
/kaan(-a)/ to be. In the Colloquial periphrastic formations used to
designate one idea, words frequently become so closely connected that
they are treated as one word. For example, /?ibn-u ?aadam-a/ son of
Adam = man are fused into /bani?aadam/ and /xamsata 3aara/ fifteen
into /xamasTaaar/.
The final means of word formation is /at-ta3riib(-u)/
Arabicization. This refers to the assimilation of foreign words into
Arabic vocabulary. For instance, the English words democracy and
music have been Arabicized as /diimuqraTiyy-ah/ and /muusiiqaa/,
respectively.

48

1.4. Morphosyntactic Alternations:

In this section, two major morphosyntactic alternations will be
compared in the two varieties: the distinction between pausal and nonpausal forms and the use of nunation. These two alternations have a great
role to play in the morphology of the Arabic language and the two
varieties under discussion differ to a great extent in their adoption of these
alternations.
1.4.1. Pausal vs. Non-Pausal Forms:
A major difference between SA and EA is that the former
distinguishes between pausal and non-pausal forms of words while the
latter does not. Pausal forms are the forms of words in isolation or at the
end of an utterance. Non-pausal forms, on the other hand, are the forms of
words in context or connected speech. Final short vowels are dropped in
SA pausal forms and retained in contextual forms. As for nouns, most of
these short vowels constitute case endings in the singular, sound feminine
plural and broken plural. They also constitute parts of the case endings in
the dual and sound masculine plural. (The inflection of nouns for case,
gender and number will be discussed in 3.4-6). The difference between
contextual and pausal forms of nouns can be observed by contrasting the
SA sentences in (56-60a) with their counterparts in (56-60b), respectively:
(56) a. daxal-at
il-bint-u
enter.pf-3fsg
the-girl-Nom
The girl entered the class.
b. daxal-at
il-bint
enter.pf-3fsg
the-girl
The girl entered.

il-faSl
the-class

(57) a. ra?ay-tu
l-banaat-i
see.pf-1msg
the-girls-Acc
I saw the girls in the class.
b. ra?ay-tu
l-banaat
see.pf-1msg
the-girls
I saw the girls.

fi-1-faS1
in-the -class

(58) a. waDa3-tu
l-kutub-a
put.pf-1msg
the-books-Acc
I put the books on the shelf.
b. waDa3t-u
l-kutub
put.pf-1msg
the-books
I put the books.

3ala
on

r-raff
the-shelf

49

(59) a. xarag-a
1-walad-aani
get out-pf-3msg
the-boys-Nom.du
The two boys got out of the class.
b. xarag-a
1-walad-aan
get out.pf-3msg
the-boy-Nom.du
The two boys got out.

mina 1-faSl
from the-class

(60) a. sallam-tu
3ala 1-mu3allim-iina
fi-1-madras-ah
greet.pf-1msg (prep) the-teacher-Gen.mpl
in -the school
I greeted the teachers in the school.
b. sallam-tu
3ala
1-mu3allim-iin
greet.pf-1msg (prep)
the-teacher-Gen.mpl
I greeted the teachers.
Another difference between pausal and non-pausal forms in SA is
that final [-at] in non-pausal forms is changed to [-ah] in pausal forms.
Compare the word for school in (61a) and (61b):
(61) a. ?ahab-u
?ila
1-madras-at-i
impf.go-1msg
to
the-school-Gen
I go to school in the morning.
b. ?ahab-u
?ila
1-madras-ah
impf.go-1msg
to
the-school
I go to school.

Sabaah-an
in the morning

Turning to verbs, it can be observed that final short vowels used to

constitute the whole or part of mood suffixes are dropped in pausal forms
(e.g. 62-64b) and retained in non-pausal forms (e.g. 62-64a). (The
inflection of verbs for mood will be discussed in 2.2):
?an
?u-aakir-a
(62) a. naSah-a-nii
advise.pf.3msg-1msg to
impf-study.1sg-subj
He advised me to study my lessons.
b. naSah-a-nii
?an
?u-aakir
advise.pf.3msg-1msg to
impf-study
He advised me to study.

duruus-ii
lessons-1msg

(63) a. ?al-bint-aani
tu-aakir-aani
the-girl-Nom.du
impf-study-2du
The two girls are studying in the room.
b. ?al-bint-aani
tu-aakir-aan
the-girl-Nom.du
impf-study-2du

l-hujr-ah
the-room

fi
in

50

The two girls are studying.

(64) a. ?al-?awlaad-u
ya-13ab-uuna
the-boys-Nom
impf-play-3mpl
The boys are playing in the garden.
b. ?al-?awlaad-u
ya-l3ab-uun
the-boys-Nom
impf-play-3mpl
The boys are playing.

fi
in

l-hadiiq-ah
the-garden

1.4.2. Nunation:
Another morphosyntactic phenomenon that distinguishes EA from
SA is nunation. This is a feature which occurs only in SA. It is related to
the presence or absence of [-n] in the final position of nouns, usually after
case endings. A final [-n] is attached to short vowels used as case markers
in SA indefinite nouns. However, this [-n] is deleted when the nouns in
question are defined. As will be shown in (3.3), a noun can be made
definite by having a definite article or by forming the first part in a
construct phrase. Compare (65a) with (65b-c):
(65) a. haaihi
wardat-un
this (f)
a rose-Nom
This is a beautiful rose.
b. haaihi
1-wardat-u
this (f)
the- rose-Nom
This rose is beautiful.
c. haaihi
wardat-u
this(f)
rose-Nom
This is the boys rose.

jamiil-ah
beautiful-fsg
jamiil-ah
beautiful-fsg
1-walad
the-boy

There are very few other syntactically conditioned contexts in which [-n]
is omitted. For instance, the [-n] is dropped from nominative nouns after
the vocative particle /yaa / oh, as in /yaa muhammad-u/ Oh,
Muhammad; and from accusative nouns following the particle of absolute
negation /laa/ no, as in /laa akk(-a) / no doubt.
Nunation is not only used for singular nouns like that in (65) above
but also for sound feminine and broken plurals. Compare (66a) with
(66b-c) and (67a) with (67b-c), respectively:
(66) a. fi 1-faSl-i
Taalibaat-un
jamiil-aat(-un)
in the-classroom-Gen students(f)-Nom
beautiful-fpl(-Nom)
There are beautiful female students in the classroom.

51

b. daxal-at
iT-Taalibaat-u
1-faSl(-a)
enter.pf-3fsg
the-students(f)-Nom the-classroom(-Acc)
The female students entered the classroom.
l-faSl-i
jamiil-aat(-un)
c. Taalibaat-u
students(f)-Nom
the-classroom-Gen beautiful-fpl(-Nom)
The female students of the classroom are beautiful.
(67) a. fi
1-maktabat-i
kutub-un
kaiir-at(-un)
in
the-library-Gen
books-Nom many-fsg(-Nom)
There are many books in the library.
b. ?al-kutub-u
3ala
l-?arfuf(-i)
the-books-Nom
on
the-shelves(-Gen)
The books are on the shelves.
c. kutub-u
zayd-in
gadiid-at(-un)
books-Nom
Zayd-Gen
new-fsg(-Nom)
Zayds books are new.
What is interesting about the inflection of nouns in SA is that
nunation occurs with sound feminine plurals but not with duals or sound
masculine plurals. Though one can say /Taalib-aat(-un)/ female students
as in (66a) above, one cannot say */walad-ayn(-in)/ two boys (Gen) or
*/muslim-uun(-an)/ Muslims (Acc). It seems that it is a property of the
short vowel case suffixes [-u], [-a] and [-i] that they can receive nunation.
Long vowel case endings or those comprising a short vowel and a glide
cannot be nunated. However, although the [-n] in the dual and the sound
masculine plurals is not the [-n] of nunation, it is deleted in construct states
as will be shown in (3.6).
1.5. Literature on Arabic Morphology:
Arabic as a language has been under the analytic scope of Arab
grammarians for the past thirteen centuries. It has been the subject of
many studies since the time of early Arab grammarians, such as Al-Khalil
Ibn Ahmad (d. 791) and his student Sibawayh (d. 798), and up to the
present moment. In modern linguistics, several books and dissertations on
SA and EA were written in the United States of America and the European
Union as well as in the Arab world. Some of these studies were general
introductions, while others were concerned with specific areas of linguistic
research. For the sake of brevity, only the major morphological studies
that were conducted in English in the second half of the twentieth century
will be reviewed in this section.

52

For general statements on the morphology of SA, reference should

be made to the works of Bateson (1967) and Beeston (1970). The former
gave brief remarks on noun inflection, noun derivation and the number
system. It also dealt with verbal inflection, verbal derivation and the root
system. The latter investigated the participles, and the number and gender
of nouns. It also analyzed pronouns, demonstratives and adjectives.
Specific areas of SA morphology were the subjects of investigation
by some modern linguists. For instance, nominal SA morphology was
studied by Trager & Rice (1954) and Hamp (1959). The nominal
declensions of SA and their interesting interaction with the system of final
vowels which mark case relations were the subject of Rabin (1965).
Taking as a base form the unmarked imperative stem of the verb,
Schramm (1962) derived the entire system of Arabic verb stems through
synchronic, generative morphophonemic rules. This system is admirable in
its ability to account in a regular way for the behavior of the stems derived
from weak roots.
Travis (1979) provided a thorough investigation of Modern
Standard Arabic morphology using a transformational model. He analyzed
the Arabic derivational paradigm for verbs with its morphological forms.
He also dealt with the inflectional paradigm for verbs with its categories.
He dealt with the derivation of inflected forms in Arabic in particular and
presented some additions to the generative theory of morphology in
general.
McCarthys dissertation (1979) forms the basis of the
autosegmental approach to non-concatenative morphology. In that
dissertation he proposed a solution to the traditional problem of the root
and pattern morphological system of Semitic using the mechanisms of
autosegmental phonology. He illustrated his solution by an extensive
treatment of Classical Arabic verbal and nominal systems. He also dealt
with the representation of syllabic and accentual structure in Semitic
languages, particularly Hebrew, Classical Arabic and two modern Arabic
dialects, namely Cairene and Damascene.
McCarthy & Prince (1990a) offered an extended analysis of the
templatic morphology of SA. They argued that templatic constraints on
word structure should be characterized in prosodic terms - that is, in terms
of notions like minimal word, foot, syllable and mora. In particular, they
showed that basic, underived templates of Arabic must be analyzable in
prosodic terms, as required by the Prosodic Morphology Hypothesis.

53

McCarthy & Prince (1990b) proposed a theory of prosodic domain

circumscription, by means of which rules sensitive to the morphological
domain may be restricted to a prosodically characterized (sub-)domain in a
word or stem. They illustrated their theory primarily by an extensive
analysis of the Arabic broken plural. Taken together, both works of
McCarthy & Prince provide a comprehensive analysis of Arabic templatic
morphology within the prosodic theory.
Arabic verbal morphology was also investigated, but from a
syntactic point of view, by Al-harbi (1990). He proposed a syntactic
approach for analyzing the complex morphology of Arabic verbal
phenomena, which was capable of offering an explanation of various
constituents as well as simple verbal structures. That thesis dealt with the
agreement phenomena in Arabic verbal clauses, and the morphology and
syntax of perfective and imperfective verbal forms. It also investigated the
simple and complex verbal gerunds or /maSaadir/ as well as the participial
forms.
The morphological component of EA was dealt with in the context
of general linguistic approaches such as Mitchell (1956), Omar (1973) and
Gary & Gamal-Eldin (1982). Mitchells introduction was mainly written
for pedagogical purposes. Omars was an applied linguistic approach
which dealt with the acquisition of Egyptian Arabic as a native language.
Gary & Gamal-Eldin discussed the main morphological features of EA
inflectional and derivational morphology.
As early as 1959, an unpublished thesis by Gamal-Eldin gave a
detailed description of the morphophonemics of EA, the only such study
for the dialect. Within morphology, the verb and the noun in EA and their
classification on the basis of their canonical shapes as well their
inflectional and derivational categories were the subjects of two
unpublished dissertations by Helmy-Hassan (1960) and Ghaly (1960),
respectively. Both of them adopted the procedure of discontinuous
morphemes.
The morphology of EA was also deeply investigated by AboulFetouh (1959) and (1969). In the first work he dealt with the
morphological problem of the plural substantive forms. In the latter work,
he analyzed the morphological segmentals and suprasegmentals and
provided a classification of the stems of this dialect. He also dealt with the
morphology of verbs, nouns, pronouns, adjectives and adverbs. The
originality of his work lies in its departure from traditional treatises by
seeking to establish form classes on the basis of their inflectional

54

categories and in some cases their syntactic behavior. Members within

each class were then classified according to their stem alternations.
In the field of comparative studies between SA and the Colloquial
varieties, Ferguson (1972) presented an outline of the diglossic nature of
Arabic, and compared it to other contemporary language situations. He
attempted to identify the sociolinguistic characteristics of the phenomenon
of diglossia by studying the features common to Arabic and three other
languages which have a Standard-Colloquial language dichotomy; namely,
Greek, Swiss German and Haitian Creole.
Becker (1964) provided a comparative grammar of the verb
structures of Modern Literary Arabic and Lebanese Colloquial Arabic.
And in a very concise paper, Selim (1967) outlined some of the contrasts
between Classical Arabic and Egyptian Arabic. He offered very brief
remarks on the phonology, morphology and syntax of the two varieties.
The problem of diglossia in Iraq was discussed by Al-Toma
(1969). His work was a case study of this phenomenon based on the
comparative description of Classical and Iraqi Arabic. This comparison
involved the phonological, syntactic and lexical components. It also dealt
with the nominal system, the verbal system and the deverbal noun.
An historical linguistic study which showed the development of
Classical Arabic and Colloquial Cairene was conducted by Robertson
(1970). Her first chapter was a lengthy linguistic and historical
background about language study in general and the history of the Arabic
language in particular. Then, she focused on the conflict between Coptic
and Arabic after the Arab invasion of Egypt, showing the Coptic
influences on the Colloquial idiom spoken in Cairo. Finally, she presented
brief sketches of the phonetic, morphological and lexical differences
between Classical and Cairene Arabic. However, these sketches were in no
way comprehensive and, moreover, the author was mainly concerned with
the historical evolution rather than the present state of the language. In
other words, Robertsons dissertation is a diachronic rather than a
synchronic study.
Omar (1976) held a comparison between two of the colloquial
dialects of Arabic. Her book describes the major differences between the
Levantine (Palestinian and Lebanese) and the Egyptian (Cairene) dialects.
These differences are investigated at four levels: pronunciation
(phonology); words and affixes (morphology); phrases and sentences
(syntax); and vocabulary (lexicon).

55

At the end of this survey it has to be stated that my work in this

book has been aided by the existence of two important works on Arabic
phonology by Brame (1970) and Broselow (1976). The former was
devoted to discovering the deep representations of SA phonology and
postulating a set of phonological rules which adequately map the deep
representations into the phonetic representations. The latter was dedicated
to outlining the major phonological processes of EA and proposing new
approaches to the phenomenon of emphasis, the notion of the syllable and
the process of cliticization.
1.6. Theory of Prosodic Morphology:
In the Introduction, it was stated that reference would be made at
some points to the theory of Prosodic Morphology proposed by McCarthy
& Prince (1986, 1988). I will use this theory mainly to give a prosodic
analysis of singular nominal stems (3.1.3) and broken plural patterns
(3.6.3.3). Therefore, it will be useful to provide a brief note on this theory
in this section. Prosodic Morphology is an approach to shape-invariant
morphology which is essentially founded in the phonology of prosody.
The prosodic theory is guided by the Prosodic Morphology Hypothesis
which asserts that the templates of templatic morphology are defined in
terms of the authentic units of prosody: the mora, the syllable, the foot and
the phonological word. These prosodic units are arranged in a hierarchy of
exhaustive domination (McCarthy & Prince 1990a: 3):
(68) Prosodic Hierarchy:
Phonological Word
Foot
Syllable
Mora

W
F

The phonological word is roughly similar to the syntactic word; it is

typically the domain of stress assignment. The foot is a unit consisting of
at least one stressed syllable and usually an unstressed syllable too. The
mora is the unit with which syllable weight is measured. There is a
fundamental distinction between two types of syllables: heavy and light.
Heavy syllables are those containing a long vowel /CVV/ or are closed by
a final consonant /CVC/. Light syllables are those having a short vowel
/CV/. By definition, heavy syllables contain two moras, while light
syllables contain only one (McCarthy & Prince 1990a: 4).

56

In the moraic theory, a short vowel is counted as one mora and a

long vowel as two moras. An onset consonant is not counted as a mora,
simply because it does not carry syllable weight. A coda consonant is
normally counted as one mora, but in some languages it is not counted as
such if it is word-final, in which case it is never moraic but always
extrasyllabic as will be shown below.

The moraic theory gives certain means for distinguishing the

syllable types of a language. A syllable may contain one mora or two; a
monomoraic syllable is named light and a bimoraic one heavy. Both SA
and EA have those types of syllables in medial position: /CV/ and /CVX/
which can be represented as follows:

(69)

light syllable

heavy syllable

Peripheral elements - those at the left or right edge of a stem

or word - are considered extrametrical, i.e. not participating in the
overall prosody of a word. Extrametricality is conventionally marked
by parenthesis. Taking a simple word as /kitaab/ a book, it will
have the following prosodic representation:

57

(70)

t a a

(b)

58

Chapter Two
Verbal Morphology
2.0. Introduction:
This chapter is devoted to the morphology of verbs in SA and EA.
A brief idea about verb stem classes will be presented in (2.1). Then, the
inflection of verbs for aspect/mood and voice will be explained in (2.2)
and (2.3), respectively. After that, subjectival and objectival affixes will be
discussed in (2.4). Finally, a treatment of verb derivation and transitivity
will be included in (2.5).
2.1. Verb Stems & Classes:
Arabic verb stems can be divided into two classes according to
their morphological structure: primary and derived. Primary stems are the
simplest in the language, because they consist of two morphemes only: a
root and a vocalism. Derived stems are primary stems plus one or two
derivational affixes. According to Mahadin (1982: 136), primary and
derived stems are traditionally called unaugmented and augmented,
respectively: The verbal stem may be augmented by one or two
derivational morphemes, or it may be unaugmented. Traditional Arab
grammarians used models (?awzaan) to identify the augmented and the
unaugmented forms.
Moreover, Arabic verb stems have two forms in relation to the
aspect of the verb: perfect and imperfect. The first signifies completed
action, while the second denotes uncompleted action (For details, see 2.2).
Furthermore, the verbs of SA and EA are divided into two main classes
depending on the number of consonants in their roots: triradical verbs and
quadriradical ones. The two classes are further divided into subclasses
according to the type of consonants in their roots as will be shown below.
In the following subsections an attempt will be made to characterize the
features of these classes in both varieties under investigation.
2.1.1. Triradical Verbs:
There are ten forms of Arabic triradical verbs which are known in
the linguistic literature by their corresponding Roman numerals I-X. The
first is considered the primary form and the rest represent the various
forms of derived verbs. Schmidt (1975: 109) indicates the morphological
processes involved in this derivation: From the simplest underlying stem

59

shape, Form I, all the rest may be derived by consonant gemination, vowel
lengthening, prefixation, and infixation.
In relation to the type of consonants in their roots, the triradical
verbs are further subdivided into four classes (cf. Mahadin 1982: 144):
(i) sound verbs consisting of consonants other than /?/, /w/ or /y/;
they are also called strong or regular verbs,
(ii) geminate verbs in which the second and third radicals are the
same; they are also named doubled verbs,
(iii) glottalized verbs which have the glottal stop /?/ as a radical; they
are also termed hamazated verbs, and
(iv) weak verbs which have among their radicals /w/ or /y/ or both;
they are sometimes called defective verbs.
The above classes are found in both SA and EA. The similarities and
differences between those triradical verb classes will be indicated in the
following subsections.
2.1.1.1. Sound Verbs:
Sound verbs are those which contain any Arabic consonants other
than /?/, /w/ or /y/. Table (3) shows the ten perfect forms of sound verbs in
SA, while Table (4) displays their equivalents in EA.
Table (3)
Perfect Forms of Sound Verbs in SA
No Form
Example
Darab(-a) to hit10
a
I
kabur(-a) to grow up

Fa3 u L(-a)
hazin(-a) to become sad
i

II
Fa33aL(-a)
faDDal(-a) to prefer
III
Faa3aL(-a)
haarab(-a) to fight
IV
?aF3aL(-a)
?ahraj(-a) to embarrass
V
taFa33aL(-a)
taharrak(-a) to move
VI
taFaa3aL(-a)
tanaaqa(-a) to discuss
VII (?i)nFa3aL(-a)
(?i)nhazam(-a) to be defeated
VIII (?i)Fta3aL(-a)
(?i)jtahad(-a) to work hard
IX
(?i)F3aLL(-a)
(?i)hmarr(-a) to become red
X
(?i)staF3aL(-a) (?i)stagfar(-a) to ask God for forgiveness

60

Table (4)
Perfect Forms of Sound Verbs in EA
No Form
Example
Darab to hit
a a
I
F 3 L
kibir to grow up
i i
faDDal to prefer
a
II
Fa33 L
?addim to present
i
IIIa Faa3iL
haarib to fight
IIIb Foo3aL
soogar to lock up well
IV
?aF3aL
?ahrag to embarrass
(?i)tharrak to move
a
V
(?i)tFa33 L (?i)tgaddid to be renewed
i
VI
(?i)tFaa3iL
(?i)tnaa?i to discuss
VII (?i)nFa3aL
(?i)nhazam to be defeated
VIII (?i)Fta3aL
(?i)gtahad to work hard
IX
(?i)F3aLL
(?i)hmarr to become red
(?i)stagfar to ask God for forgiveness
a
X
(?i)staF3 L (?i)sta3gil to rush someone
i
Comparison of Tables (3) and (4) reveals that the ten forms of
sound verbs are retained in the Colloquial variety with few modifications,
as illustrated below:
(a) The two varieties are similar in using the same morphological processes
to derive Forms II-X from the primary Form I. In both of them, Form II is
derived by the gemination of the middle radical, Form III by lengthening
the first vowel, Form IV by prefixing [?a-], Form V by prefixing [ta- > t-]
and geminating the middle radical, Form VI by prefixing [ta- > t-] and
lengthening the first vowel, Form VII by prefixing [n-], Form VIII by
infixing [-t-] after the first radical, Form IX by geminating the last radical,
and Form X by prefixing [sta-] to Form I. In both varieties, where an initial
consonant cluster occurs an epenthetic /i/ is inserted for syllabic reasons by
Word-Initial Epenthesis and a glottal stop is inserted after pause by Glottal
Stop Insertion to prevent the occurrence of an initial vowel. This applies to
Forms VII-X in SA and Forms V-X in EA.
(b) There are no changes in consonants between SA and EA stem forms,
but there are some vowel differences. For instance, the final vowel (-a) in
SA patterns does not appear in EA because the latter has in general lost
final vowels if they are used for inflectional purposes in SA. The vowel

61

(-a) in this case is used as a subject marker of the third person masculine
singular. That is why it is lost in the Colloquial variety.
(c) In Form I, the first vowel is always /a/ in SA. On the other hand, EA
retains this vowel in some verbs (e.g. /kasar(-a) > kasar / to break) and
changes it to /i/ in some others (e.g. /harab(-a) > hirib/ to escape). Thus,
there are two versions of this form in the Colloquial variety: [Fa3aL] and
[Fi3iL]. The conditions governing this alternation are still unknown.
Moreover, EA does not have /u/ as a second vowel. It replaces the /u/ of
SA by /i/. In other words, the SA perfect stem [Fa3uL(-a)] is replaced by
[Fi3iL] in EA, e.g. /kabur(-a) > kibir/ to grow up. So, it is noted that EA
tends to harmonize the first vowel with the second vowel (cf. Malik 1976:
92).
(d) Each of the SA Forms II, V and X has two different EA counterparts
depending on the status of the pre-final or stem vowel. Although SA has
only /a/ as a pre-final vowel in these forms, EA has /a/ in certain verbs and
/i/ in other verbs.
(e) Form III has undergone an interesting change in EA as argued by Carter
(1996: 139): in addition to the familiar kaatib / yikaatib / mikatba there is
now a new stem with long first syllable -oo-. Other examples are /soora?/
to faint and /doohar/ to keep at. The vowel /oo/ cannot be related to
/aw/ by Monophthongization because it is not stem-final. Other examples
of /aw/ in non-stem-final position that do not become /oo/ are /mawrid/ a
source and /kawkab/ a planet. This fact suggests that the vowel /oo/ is
underlying in verbs of this novel form which do not have any etymological
counterparts in SA.
(f) The Colloquial Forms [(?i)tFa33aL ~ (?i)tFa33iL] and [(?i)tFaa3aL ~
(?i)tFaa3iL] are considered reflexes of the Standard Forms [taFa33aL(-a)]
and [taFaa3aL(-a)], respectively. Put differently, the SA derivational prefix
[ta-] is reanalyzed as [t-] in EA, trigging the addition of (?i) by regular
epenthesis rules.
Along with the differences in form stated above, there are some
differences in the frequency of usage. For example, although Form IV
occurs in EA, it is less frequent in usage than it is in SA. It is usually
replaced by Form II, e.g. /?adxal(-a)/ to bring in is often replaced by
/daxxal/ and sometimes by Form I, e.g. /?ab3ad(-a)/ to take away is
replaced by /ba3ad/. Two reasons have been given by Malik (1976: 171)
for this tendency of verbal Form IV to disappear in the Colloquial variety:
(a) the semantic significations of verbal Forms II and IV frequently
overlap even in SA and (b) people seem to encounter some difficulty in
pronouncing the glottal stop /?/ whose pronunciation involves some
tension in the larynx. Though the first reason seems plausible, the second
one is questionable because Egyptian people tend to change initial /q/ in

62

SA into /?/ and pronounce it without any difficulty. So, perhaps the real
issue here is a tendency of underlying /?/ to be lost in EA.
Semantically, the SA forms and their EA equivalents are
synonymous. This can be observed by investigating the meanings of the
various morphological forms (cf. Travis 1979: 15-16 and McGuirk 1986:
69-83):
Ia. [Fa3aL(-a) > Fa3aL ~ Fi3iL] non-stative transitive or intransitive.
Ib. [Fa3iL(-a) > Fi3iL]: temporary state intransitive.
Ic. [Fa3uL(-a) > Fi3iL]: permanent state intransitive.
II. [Fa33aL(-a) > Fa33aL ~ Fa33iL]: causative, intensive, reiterative or
estimative.
III. [Faa3aL(-a) > Faa3iL]: action done to or with a person. It denotes
effort
to do something to or for someone. It also signifies reciprocity with
regard to the
action indicated by Form I (Abdel-Malek 1972: 197).
IV. [?aF3aL(-a) > ?aF3aL]: causative or factitive.
V. [taFa33aL(-a) > (?i)tFa33aL ~ (?i)tFa33iL]: intransitive (reflexive or
inchoative) or passive of Form II.
VI. [taFaa3aL(-a) > (?i)tFaa3iL]: reciprocal of Form III.
VII. [(?i)nFa3aL(-a) > (?i)nFa3aL]: passive or inchoative of Form I.
VIII. [(?i)Fta3aL(-a) > (?i)Fta3aL]: reflexive of Form I. It denotes change
or development. It is also reflexive of Form IV (Moore 1979: 229).
IX. [(?i)F3aLL(-a) > (?i)F3aLL]: developing a color or a defect.
X. [(?i)staF3aL(-a) > (?i)staF3aL ~ (?i)staF3iL]: asking or doing
something for oneself. It is also causative of Form I. It has an
estimative
or desiderative prefix which frequently means to consider, to regard
(someone or something) as (Abdel-Malek 1972: 199).
Table (5) presents the imperfect forms of SA sound verbs, while
Table (6) introduces their EA analogues.

63

Table (5)
Imperfect Forms of SA Sound Verbs
No Form
Example
ya-Drib(-u)
to hit
i
I
ya-kbur(-u) to grow up

ya-F3 u L(-u) ya-hzan(-u) to become sad
a

II
yu-Fa33iL(-u)
yu-kassir(-u) to smash
III
yu-Faa3iL(-u)
yu-haarib(-u) to fight
IV
yu-F3iL(-u)
yu-hrij(-u) to embarrass
V
ya-taFa33aL(-u) ya-taharrak(-u) to move
VI
ya-taFaa3aL(-u) ya-tanaaqa(-u) to discuss
VII ya-nFa3iL(-u)
ya-nhazim(-u) to be defeated
VIII ya-Fta3iL(-u)
ya-gtahid(-u) to work hard
IX
ya-F3aLL(-u)
ya-hmarr(-u) to become red
X
ya-staF3iL(-u)
ya-stagfir(-u) to ask for forgiveness
Table (6)
Imperfect Forms of EA Sound Verbs
No Form
Example
yi-ktib to write
i
I
yi-skut to become silent

yi-F3 u L
yi-Drab to hit
a

yi-kassar to smash
a
II
yi-Fa33 L
yi-?addim to present
i
IIIa yi-Faa3iL
yi-haarib to fight
IIIb yi-Foo3aL
yi-soogar to lock up well
IV
yi-F3iL
yi-hrig to embarrass
yi-tharrak to move
a
V
yi-tFa33 L
yi-tgaddid to be renewed
i
VI
yi-tFaa3iL
yi-tnaa?i to discuss
VII yi-nFi3iL
yi-nhizim to be defeated
VIII yi-Fti3iL
yi-gtihid to work hard
IX
yi-F3aLL
yi-hmarr to become red
yi-stagfar
to ask for forgiveness
a
X
yi-staF3 L
yi-sta3gil to be in a hurry
i
By comparing Tables (3) through (6), it is noted that the SA vowel
/a/ in the imperfect prefix is changed into /i/ in EA (For details, see 2.2).
The tables also show that in both varieties under study the change of the

64

stem vowel (i.e. that before the last radical) of Form I between the perfect
and the imperfect is unsystematic. Therefore, it should be learned from a
reliable dictionary. However, a few generalizations can be offered:
(a) SA verbs of Form [Fa3aL(-a)] have two unpredictable stem vowels, /u/
or /i/, i.e. [ya-F3uL(-u) ~ ya-F3iL(-u)], e.g. /sakat(-a), ya-skut(-u)/ to be
silent but /hamal(-a), ya-hmil(-u)/ to carry. Exceptional here are the
verbs whose second or third radical is a guttural /3, h, ?, h, x or g/ which
tend to have /-a-/ in the imperfect base, e.g. /fatah(-a), ya-ftah(-u)/ to
open. This exception does not apply, however, in the verb /daxal(-a), yadxul(-u)/ to enter. Similarly, EA verbs of the perfect form [Fa3aL] have
three imperfect forms; [yi-F3aL ~ yi-F3iL ~ yi-F3uL], e.g. /Darab, yiDrab/ to hit, /katab, yi-ktib/ to write and /daxal, yi-dxul/ to enter.
(b) SA verbs of Form [Fa3uL(-a)] in the perfect have the form
[ya-F3uL(-u)] in the imperfect, e.g. /kabur(-a), ya-kbur(-u)/ to grow up.
And SA verbs of Form [Fa3iL(-a)] in the perfect have the imperfect [yaF3aL(-u)], e.g. /hazin(-a), ya-hzan(-u)/ to become sad. On the other hand,
EA verbs of the perfect form [Fi3iL] which is the equivalent of both
[Fa3uL(-a)] and [Fa3iL(-a)] have one imperfect form [yi-F3aL], e.g. /kibir,
yi-kbar/ to grow up and /hizin, yi-hzan/ to become sad.
To account for the changes in the stem vowel in SA, Brame (1970:
142) classifies verbs into three classes:
Class A: a u, e.g. /katab(-a), ya-ktub(-u)/ to write
a i, e.g. /nazal(-a), ya-nzil(-u)/ to descend
Class B: i a, e.g. /rakib(-a), ya-rkab(-u)/ to ride
Class C: u u, e.g. /kabur(-a), ya-kbur(-u)/ to grow up
Then, Brame (1970: 150) proposes a stem allomorphy rule which he calls
Ablaut:
(71) Ablaut:
V

hi [hi] / imperfect
C
where C = Class C-verbs
This rule indicates that the height of the perfect stem vowel is inverted in
the imperfect, except in Class C-verbs. In other words, when the perfect
stem vowel is /a/ which is a low vowel, it is changed in the imperfect into
/i/ or /u/ which are high vowels. Conversely, when the perfect stem vowel
is /i/ it is changed into /a/ in the imperfect. The exception is that when the
perfect stem vowel is /u/ it remains unchanged in the imperfect.

65

Contrasting Tables (5) and (6) shows that, apart from the vowel of
the imperfect prefix, the two varieties are identical in deleting the first
vowel of the perfect form (i.e. that which follows the first consonant of the
stem including the derivational prefixes) in the imperfect form of the
primary Form I. This can be accounted for by the rule of Vowel Elision
proposed by Brame (1970: 138):
(72) Vowel Elision:
V / V + C CV
This rule indicates that we delete the vowel that comes after the first
consonant of the stem in the imperfect of Form I. However, in other forms,
this vowel has different behaviors in EA. First, it is preserved, i.e. remains
identical to that of SA, in three forms: II, III and X, as in:
Form
II
III
X

SA
yu-qaddim(-u)
yu-saafir(-u)
ya-staslim(-u)

EA
yi-?addim
yi-saafir
yi-staslim

Gloss
to present
to travel
to surrender

Second, it is omitted in two forms: V and VI, as a direct consequence of

the reanalysis of the SA prefix [ta-] as just [t-] in EA, as in:
SA
EA
Gloss
Form
V
ya-ta3allam(-u)
yi-t3allim
to learn
"
ya-takallam(-u)
yi-tkallim
to speak
VI
ya-tafaaham(-u)
yi-tfaahim
to understand
"
ya-tamaaraD(-u)
yi-tmaariD
to feign illness
Finally, it is replaced by /i/ in the two forms VII and VIII in order to
secure vowel harmony with pre-final /i/. This harmony applies whenever
the /a/ is in a light syllable, as in:
Form
VII
"
VIII
"

SA
ya-nkasir(-u)
ya-nbasiT(-u)
ya-htariq(-u)
ya-ftaxir(-u)

EA
yi-nkisir
yi-nbisiT
yi-thiri?
yi-ftixir

Gloss
to break
to be happy
to burn
to boast

2.1.1.2. Geminate Verbs:

In these verbs, the second and third radicals form a geminate
consonant. The Standard surface form of the primary geminate verb is

66

[Fa33(-a)] whose Colloquial counterpart is [Fa33], as is evident in the

following examples:
SA
add(-a)
mall(-a)
jarr(-a)

EA
add
mall
garr

Gloss
to pull
to get bored
to draw

To explain the difference between the perfect forms of primary sound

verbs and those of primary geminate verbs, Wright (1967: 68) says:
When both the first and third radicals have vowels, the second radical
rejects its vowel, and unites with the third, so as to form a double letter.
Thus, one can propose that the Form [Fa33(-a)] is derived from underlying
[Fa3a3(-a)] by Identical-Consonant Metathesis and Closed-Syllable
Shortening as follows:
(73) Fa3a3(-a) Faa33(-a) Fa33(-a)
These rules apply in both varieties when the form is followed by a vowel
or word boundary.
The Colloquial variety deviates from the Standard variety in the
case of consonantal suffixation to the perfect form of the primary geminate
verb. The rules of Identical-Consonant Metathesis and Closed-Syllable
Shortening do not apply in SA because the second consonant is not
followed by a vowel. EA, on the other hand, applies those rules because it
inserts the vowel /ee/ between the stem and the consonantal suffixes. But
why is it /ee/ in particular that is inserted and not any other vowel? One
can propose that the /ee/ is used by analogy to the /ee/ at the end of finalweak verbs (See 2.1.1.4) which is derived from SA /ay/ by the regular rule
of Monophthongization. Compare the verb /add(-a)/ to pull in the two
varieties:
SA
adad-tu
adad-naa
adad-ta
adad-ti
adad-tum
adad-tunna

EA
add-ee-t
add-ee-na
add-ee-t
add-ee-ti
add-ee-tu
add-ee-tu

Gloss
I pulled
we pulled
you (msg) pulled
you (fsg) pulled
you (mpl) pulled
you (fpl) pulled

To compare the derived forms of geminate verb stems II-X in SA

and EA, one can look at Tables (7) and (8), respectively.

67

Table (7)
Perfect Forms of Geminate Verbs in SA
No Form
Example
I
Fa33(-a)
marr(-a) to pass
II
Fa33a3(-a)
harrar(-a) to liberate
IV ?aFa33(-a)
?amadd(-a) to provide
V
taFa33a3(-a)
taharrar(-a) to be liberated
VII (?i)nFa33(-a)
(?i)nsadd(-a) to be blocked
VIII (?i)Fta33(-a)
(?i)htajj(-a) to object
X
(?i)staFa33(-a) (?i)sta3add(-a) to be ready
Table (8)
Perfect Forms of Geminate Verbs in EA
No Form
Example
I
Fa33
marr to pass
harrar to liberate
a
II
Fa33 3
?allil to lessen
i
IV
?aFa33
?aSarr to insist
(?i)tharrar to be liberated
a
V
(?i)tFa33 3 (?i)tgaddid to be renewed
i
VII (?i)nFa33
(?i)nsadd to be blocked
VIII (?i)Fta33
(?i)htagg to object
X
(?i)staFa33
(?i)sta3add to be ready
The rule of Identical-Consonant Metathesis can be used to account for the
geminate consonants in the derived Forms IV and X. Form IV, for
instance, undergoes this derivation:
(74) ?aF3a3(-a) ?aFa33(-a) > ?aFa33
Identical-Consonant Metathesis and Closed-Syllable shortening can be
utilized to explain the derivation of Forms VII and VIII. For example,
Form VII is derived as follows:
(75) (?i)nFa3a3(-a) (?i)nFaa33(-a) (?i)nFa33(-a) > (?i)nFa33
Close investigation of Tables (7) and (8) shows that the perfect
forms of geminate verbs in EA are almost identical to their counterparts in
SA, except in deleting the inflectional (-a). Forms III, VI and IX are nonexistent in both varieties, though they are found in Classical Arabic. While

68

Form IV is commonly used in SA, the Colloquial variety very often uses
Form II in its place because both forms are semantically similar in
showing causativity, e.g. /?atamm(-a) > tammim/ to complete. We could
be witnessing the beginning of a change in which EA may eventually lose
Form IV as an independent type.
Turning to the imperfect forms of doubled verbs, it is observed that
except for the imperfect prefix, EA has kept the same morphological forms
of SA, as can be seen in Tables (9) and (10).
Table (9)
Imperfect Forms of Geminate Verbs in SA
No Form
Example
ya-mall(-u) to be bored
a
I
ya-murr(-u) to pass

ya-F u 33(-u)
ya-qill(-u) to become less
i

II
yu-Fa33i3(-u)
yu-harrir(-u) to liberate
IV yu-Fi33(-u)
yu-midd(-u) to provide
V
ya-taFa33a3(-u)
ya-taharrar(-u) to be liberated
VII ya-nFa33(-u)
ya-nsadd(-u) to be blocked
VIII ya-Fta33(-u)
ya-htajj(-u) to protest
X
ya-staFi33(-u)
ya-sta3idd(-u) to be ready
In his differentiation between sound and geminate verbs, Wright
(1967: 68) indicates that If the third radical has a vowel, but the first is
without one, the second radical throws back its vowel upon the first, and
then combines with the third, so as to form a double letter. In other
words, Arabic has a dispreference for two syllables with the same
consonant. So, a vowel between two identical consonants is either deleted,
as in the perfect forms accounted for by the rules of Identical-Consonant
Metathesis and ClosedTable (10)
Imperfect Forms of Geminate Verbs in EA
No Form
Example
yi-mall
to be bored
a
I
yi-murr to pass

yi-F u 33
yi-?ill to become less
i

yi-harrar to liberate
a
II
yi-Fa33 3
yi-?allil to lessen
i

69

IV

yi-Fi33

a
yi-tFa33 3
i
VII yi-nFa33
VIII yi-Fta33
X
yi-staFi33

yi-Sirr to insist
yi-tharrar to be liberated
yi-tgaddid to be renewed
yi-nsadd to be blocked
yi-htagg to protest
yi-sta3idd to get ready

Syllable Shortening, or metathesized so as to come before the identical

consonants, as can be interpreted by Identical-Consonant Metathesis alone.
This rule applies in the imperfect of Forms I, IV, VII, VIII and X. For
instance, Form X goes through this derivation:
(76) ya-staF3i3(-u) ya-staFi33(-u) > yi-staFi33
By analysis of Tables (9) and (10), it can be observed that the
Standard primary geminate verb may have one of three forms in the
imperfect aspect: [ya-Fa33(-u)], [ya-Fu33(-u)] or [ya-Fi33(-u)]. These
forms have been retained in the Colloquial variety with a tendency to
change the stem vowel in the first form into /i/, as in these verbs:
SA
EA
Gloss
ya-wadd(-u)
yi-widd
to like
ya-jurr(-u)
yi-gurr
to drag
ya-hinn(-u)
yi-hinn
to long, yearn
The derived geminate verbs show a similar development in EA like
that of the sound type regarding the vowel of the imperfect prefix and the
medial vowel. Compare the following:
SA
yu-lihh(-u)
ya-nDamm(-u)
ya-htall(-u)
ya-staqirr(-u)

EA
yi-lihh
yi-nDamm
yi-htall
yi-staqirr

Gloss
to insist
to join
to occupy
to settle down

2.1.1.3. Glottalized Verbs:

Glottalized verbs are trilateral verbs having the glottal stop /?/ as
one of their radicals. The glottal stop may occur in any position in the
Standard verb: initial, medial or final. This type of verb has become highly
restricted and in some cases non-existent in EA, in contrast to its high
frequency of occurrence in SA.

70

The initial-glottalized verb of Form I occurs in very few examples

in EA, as in these verbs (The last one only must preserve the /?/):
SA
?akal(-a)
?axa(-a)
?amar(-a)

EA
?akal/kal11
?axad/xad
?amar

Gloss
to eat
to take
to order

In the imperfect, the glottal stop in two of these three examples is replaced
by lengthening the vowel of the imperfect prefix in EA. This is
accomplished by Compensatory Lengthening:
SA
ya-?kul(-u)
ya-?xu(-u)
ya-?mur(-u)

EA
yaakul
yaaxud
yu?mur

Gloss
to eat
to take
to order

The derived forms of the initial-glottalized verb preserve their /?/

in EA, except for some verbs of Form X, as in the following verbs:
Form SA
EA
Gloss
II
?axxar(-a)
?axxar
to delay
"
?ajjal(-a)
?aggil
to postpone
IV
?aanas(-a)
?aanis
to cheer
"
?aaman(-a)
?aamin
to believe
V
ta?akkad(-a)
(?i)t?akkid
to be certain
X
(?i)sta?jar(-a)
(?i)sta?gar
to rent
"
(?i)sta?an(-a)
(?i)sta?zin
to ask for permission
"
(?i)sta?hal(-a)
(?i)staahil
to deserve
Form IV verbs /?aanas(-a) > ?aanis/ and /?aaman(-a) > ?aamin/ are derived
from their underlying forms /?a?nas(-a)/ and /?a?man(-a)/ by
Compensatory Lengthening. This is the only case in which this rule
applies in SA. The change of /(?i)sta?hal(-a) > (?i)staahil/ to deserve is
also accomplished by Compensatory Lengthening. A similar change
occurs in /(?i)sta?xar(-a) > (?i)staaxir/ to move out of the way.
The glottal stop is also preserved in the imperfect forms of these
verbs in EA, as follows:
Form
II
"
IV

SA
yu-?axxir(-u)
yu-?ajjil(-u)
yu-?nis(-u)

EA
yi-?axxar
yi-?aggil
yi-?aanis

Gloss
to delay
to postpone
to cheer12

71

"
V
X
"
"

yu-?min(-u)
ya-ta?akkad(-u)
ya-sta?jir(-u)
ya-sta?in(-u)
ya-sta?hil(-u)

yi-?aamin
yi-t?akkid
yi-sta?gir
yi-sta?zin
yi-staahil

12

to believe
to be certain
to rent
to ask for permission
to deserve

The employment of /?/ in medial and final positions is still more

restricted in EA. Thus, the following SA verbs are no longer used in EA:
/da?ab(-a)/ to continue, /sa?im(-a)/ to be bored or /Da?ul(-a)/ to
dwindle. There are very few examples of the Form I medial-glottalized
verb in the Colloquial variety:
SA
sa?al(-a)
ra?as(-a)
ra?af(-a)
ya?is(-a)

EA
sa?al
ra?as
ra?af
yi?is

Gloss
to ask
to head
to have mercy (on)
to lose hope

These verbs keep their /?/ in the Colloquial imperfect forms as well:
SA
ya-s?al(-u)
ya-r?as(-u)
ya-r?af(-u)
ya-y?as(-u)

EA
yi-s?al
yi-r?as
yi-r?af
yi-y?as

Gloss
to ask
to head
to have mercy
to lose hope

There are no derived forms of the medial-glottalized verb, except

for the Colloquial form [(?i)tFaa3iL] which is the counterpart of the
Standard form [taFaa3aL(-a)], where /3/ is a glottal stop, as in /tafaa?al(-a)
> (?i)tfaa?il/ to be optimistic and /taaa?am(-a) > (?i)taa?im/ to be
pessimistic. The imperfect forms of these verbs are /ya-tafaa?al(-u) >
yi-tfaa?il/ and /ya-taaa?am(-u) > yi-taa?im/, respectively.
The final-glottalized verb in SA, as a generalization, loses its /?/ in
EA by Final /?/ Deletion, as in /?ara, yi-?ra/ to read which is the reflex of
the SA verb /qara?(-a), ya-qra?(-u)/. The Colloquial variety also offers
derived forms of final-glottalized verbs that have lost their /?/:
SA
?aqra?(-a)
tabarra?(-a)
(?i)btada?(-a)

EA
?arra
(?i)tbarra
(?i)btada

Gloss
to make (s. o.) read
to disown
to begin

72

The glottal stop is still preserved in some other EA final-glottalized verbs

like:
SA
tahayya?(-a)
(?i)stahza?(-a)
?axTa?(-a)

EA
(?i)thayya?
(?i)stahza?
?axTa?

Gloss
to be prepared
to mock
to err

A morphophonemic dissimilarity between the two varieties should

be recorded here. The glottal stop at the end of final-glottalized SA verbs,
together with its preceding vowel, is replaced by the long vowel /ee/ in EA
before pronominal suffixes starting with a consonant, as in these
examples:
SA
qara?-tu
xabba?-ta
hanna?-tum
tabarra?-ti

EA
?aree-t
xabbee-t
hannee-tu
?itbarree-ti

Gloss
I read
you (msg) hid
you (mpl) congratulated
you (fsg) disowned

The change occurring here follows this direction:

(77) a? ay ee
There is a problem here: why does /a?/ become /ay/ and not /aa/ as
predicted by Compensatory Lengthening? I conceive of a mechanical
solution to this problem by proposing a special rule that changes /a?/ to
/ay/ in this particular context, as follows:
(78) a?-to-ay in EA:
a? ay / in final-glottalized verbs
However, another equally plausible explanation is to propose that the /?/ in
the verb stem is deleted by Final /?/ Deletion, then Egyptian people treat
the verb as a final-weak verb whose /ee/ is derived from /ay/ by
Monophthongization (See 2.1.1.4 below).
2.1.1.4. Weak Verbs:
Traditionally, verbs which have among their radicals one or both of
the glides /w/ and /y/ are considered /mu3tall-ah/ weak or defective.
There are various types of weak verbs depending on the position of the

73

glide: initial, medial or final; or a combination of two positions excluding

initial and medial. All these types are found in EA with a few
modifications.
Initial-weak verbs can be classified into two groups: w-verbs and
y-verbs. Verbs of the first group occur frequently in both SA and EA. The
following are just a few examples:
SA
wazan(-a)
waSal(-a)
waqa3(-a)
waqaf(-a)

EA
wazan
wiSil
wi?i3
wi?if

Gloss
to weigh
to arrive
to fall
to stop / stand up

In most cases, initial-weak w-verbs lose their /w/ in the formation of the
imperfect form in SA, whereas in EA the glide is preserved whenever
followed by a consonant other than /?/ and replaced by /u/ whenever
followed by /?/. The rule explaining the changes occurring in SA is given
in (79) which shows that we delete /w/ after a vowel and before a
consonant followed by /i/. It is called w-Occultation by Brame (1970:
173). The changes occurring in EA are shown in (80):
(79) w-Occultation in SA:
w / V Ci
(80) a. Vw? w? u?
b. VwC VwC
In other words, while SA has two forms for the imperfect of w-verbs:
[ya-3iL(-u)] and, less frequently, [ya-3aL(-u)], EA has three counterparts:
[yi-w3iL], [yi-w3aL] and [yu-3aL]:
SA
ya-zin(-u)
ya-Sil(-u)
ya-qa3(-u)
ya-qif(-u)

EA
yi-wzin
yi-wSal
yu-?a3
yu-?af

Gloss
to weigh
to arrive
to fall
to stop/stand up

The reason why the rule of w-Occultation applies to the SA verb

/ya-qa3(-u)/ to fall though there is no /i/ after /q/ is that this rule applies
before another rule which changes /i/ into /a/ in the vicinity of a laryngeal,
the /3/ in this case. This rule was stated by Brame (1970: 161):

74

(81) Laryngeal Assimilation in SA:

L
i a / / imperfect
L
where L stands for a laryngeal: /3, h, h or ?/.
The number of primary initial-weak y-verbs is extremely small in
both varieties. They behave like sound verbs and conform to their forms in
both varieties:
SA
EA
Gloss
ya?is(-a)
yi?is
to despair (pf)
yabis(-a)
yibis
to dry (pf)
ya-y?as(-u) yi-y?as
to despair (impf)
ya-ybas(-u) yi-ybas
to dry (impf)
The number of derived forms of initial-weak verbs is also small in
both varieties. A major distinction here is that Form VII is now regularly
formed from initial /w/ roots in EA, which is not possible in SA. Compare
the following derived verbs in SA and EA:
Form
II
"
III
IV
"
V
VI
VII
"
VIII
X

SA
wadda3(-a)
yabbas(-a)
waafaq(-a)
?awqa3(-a)
?awqaf(-a)
tawarraT(-a)
tawaajah(-a)

(?i)ttaSal(-a)
(?i)stawda3(-a)

EA
Gloss
wadda3
to bid farewell
yabbis
to make dry
waafi?
to agree
wa??a3to cause to fall
wa??af
to stop / make stand up
(?i)twarraT to be involved in difficulties
(?i)twaagih to face each other
(?i)nwazan
to be weighed
(?i)nwaga3
to feel pain
(?i)ttaSal
to contact
(?i)stawda3 to let, deposit

It is clear from these examples that EA is using Form II to replace Form

IV, another indication that the latter form is dying out in this variety. It can
also be observed that the initial glide is replaced by /t/ in Form VIII in
both varieties. This can be accounted for by a special rule as follows:
(82) w-to-t:
wt/t

in Form VIII

These are the imperfect counterparts of the derived verbs above:

75

Form
II
"
III
IV
"
V
VI
VII
"
VIII
X

SA
yu-waddi3 (-u)
yu-yabbis(-u)
yu-waafiq(-u)
yuuqi3(-u)
yuuqif(-u)
ya-tawarraT(-u)
ya-tawaajah(-u)

ya-ttaSil(-u)
ya-stawdi3(-u)

EA

Gloss
yi- wadda3
to bid farewell
yi-yabbis
to make dry
yi-waafi?
to agree
yi-wa??a3
to cause to fall
yi-wa??af
to stop / make stand up
yi-twarraT
to be involved in difficulties
yi-twaagih
to face each other
yi-nwizin
to be weighed
yi-nwigi3
to feel pain
yi-ttiSil
to contact
yi-stawdi3
to let, deposit

Medial-weak verbs are those which have surface /aa/ in some

forms and /w/ or /y/ in some others, in the place of the middle radical of
their roots. To see the difference between the two varieties, compare
Tables (11) and (12):
Table (11)
Perfect Forms of Medial-Weak Verbs in SA
No Form
Example
I
FaaL(-a)
gaab(-a) to be absent
Sawwar(-a) to photograph
ww
II
Fa aL(-a)
3ayyan(-a) to appoint
yy
haawal(-a) to try
w
III Faa aL(-a)
3aayan(-a) to inspect
y
IV ?aFaaL(-a)
?abaad(-a) to eradicate
taSawwar(-a) to imagine
ww
V
taFa aL(-a) taxayyal(-a) to imagine
yy
ta3aawan(-a) to cooperate
w
VI taFaa aL(-a)
ta3aaya(-a) to live with
y
VII (?i)nFaaL(-a)
(?i)mbaa3(-a) to be sold13
VIII (?i)FtaaL(-a)
(?i)xtaar(-a) to choose
(?i)swadd(-a) to become black
w
IX (?i)F aLL(-a) (?i)byaDD(-a) to become white
y
X
(?i)staFaaL(-a)
(?i)stafaad(-a) to benefit

76

Table (12)
Perfect Forms of Medial-Weak Verbs in EA
No Form
Example
Ia FaaL
gaab to be absent
dawa to bother
w
Ib Fa aL
xayal to distract
y
Sawwar to photograph
ww
IIa Fa aL
3ayyaT to weep
yy
kawwin to form
ww
IIb Fa iL
3ayyin to appoint
yy
haawil to try
w
III Faa iL
3aayin to inspect
y
IV ?aFaaL
?abaad to eradicate
(?i)tSawwar to imagine
ww
Va (?i)tFa aL (?i)txayyal to imagine
yy
(?i)tkawwin to be formed
ww
Vb (?i)tFa iL (?i)t3ayyin to be appointed
yy
(?i)t3aawin to cooperate
w
VI (?i)tFaa iL (?i)t3aayin to be inspected
y
VIIa (?i)nFaaL
(?i)mbaa3 to be sold
(?i)nxawat
to be pestered
w
VIIb (?i)nFa aL
(?i)ndayan to fall in debt
y
VIII (?i)FtaaL
(?i)xtaar to choose
(?i)swadd
to become black
w
IX (?i)F aLL
(?i)byaDD to become white
y
Xa (?i)staFaaL
(?i)stafaad to benefit
(?i)stabwax to consider silly
w
Xb (?i)staF aL (?i)stayax to pretend to be a sheik
y
(?i)stamwit to pretend to be dead
w
Xc (?i)satF iL
(?i)sta3yib to consider disgraceful
y
Comparison of Tables (11) and (12) shows that the medial-weak
verbs of SA and EA are basically similar in consonants. But there are
some differences in vowels. For instance, although the vowel before the
last radical is always /a/ in SA, it is changed to /i/ in two EA forms, III and

77

VI, and in some verbs of Forms II and V. Both tables reveal that forms
with /aa/ in the place of the middle radical are I, IV, VII, VIII and X.
Forms with /w/ or /y/ are II, III, V, VI and IX. These latter forms are
identical to their counterpart sound verb forms in Tables (3) and (4) above.
The reason why the medial glides are preserved in certain forms and
replaced by a long vowel in some others seems to follow the
generalization given by Thackston (1984: 215): The basic rule throughout
is as follows: any glide surrounded by short vowels is dropped along with
the following vowel, and the preceding vowel is lengthened in
compensation if possible, i.e. unless to do so would create a superheavy
syllable. Hence, the rules responsible for the change of a glide into a long
vowel in medial-weak verbs in both varieties are:
(83) a. CVGV CVV: Glide Elision, and
b. CGV CVV: Anticipatory Vocoid Assimilation.
Glide Elision applies to Forms I, VII and VIII, but it does not apply to
Forms II and V owing to geminate inalterability, nor does it apply to Form
III because the glide there is preceded by a long vowel. Anticipatory
Vocoid Assimilation, on the other hand, applies to Forms IV and X, but it
does not apply to Form IX because if it did the result would be a
superheavy syllable. In other words, [(?i)FGVLL(-a)] must not change to
*[(?i).FVVL.L(-a)] which contains a non-final superheavy syllable.
Therefore, it can be asserted that the requirement that /CGV/ become
/CVV/ is subordinate to the requirement that there be no non-final
superheavy syllables.
The primary form has its [VGV] replaced by a long vowel by Glide
Elision in both varieties, i.e. [FaaL(-a)], as in these examples:
Root
q-w-l
g-w-3
g-y-r
b-y-3

SA
qaal(-a)
jaa3(-a)
gaar(-a)
baa3(-a)

EA
?aal
gaa3
gaar
baa3

Gloss
to say
to be hungry
to be jealous
to sell

However, when a suffix beginning with a consonant is attached to a

primary medial-weak verb, the sequence [VGV] is changed to /u/ for
w-verbs and /i/ for y-verbs in both varieties. The reason behind this is that
if the sequence [CVGVC] underwent the /CVGV CVV/ rule, the result
would be a superheavy syllable. So, the right way to think of this is that
/CVwV/ and /CVyV/ are generally impossible, and some sort of repair

78

must take place to avoid these sequences. One repair involves /awa aa/
and /aya aa/, but this repair is impossible if the result is a superheavy
syllable, as in /kawan-tu *kaan-tu/ I was. Hence, an alternative repair
must be chosen, in this instance /awa u/ and /aya i/, as in these
verbs:
SA
qul-tum
ju3t-u
gir-ta
bi3-naa

EA
?ul-tu
gu3-t
gir-t
bi3-na

Gloss
you (mpl) said
I became hungry
you (msg) became jealous
we sold

Thackston (1984: 33) notes that /xaaf(-a)/ to be afraid, /naam(-a)/

to sleep and /maat(-a)/ to die are exceptions in that they emerge as
/xif-/, /nim-/ and /mit-/ when followed by C-initial suffixes. But this is
expected since they are /xawif(-a)/, /nawim(-a)/ and /mawit(-a)/
underlyingly, and the choice of /i/ as the second vocalism is possible. Note
that EA differs from SA in that it has /xuf-t/ and /mut-t/ which are the
expected forms if /xawif(-a)/ and /mawit(-a)/ have regularized to
/xawaf(-a)/ and /mawat(-a)/ underlyingly. The only exception that
persisted in EA is /nim-t/.
A phonological divergence between the two varieties is that the
long vowel /aa/ is shortened when the medial-weak verb is followed by a
long vowel in EA by Atonic Shortening, as in these examples (Subjectival
& Objectival Affixes are discussed in 2.4):
SA
qaal-uu lii
baa3-uu laka
(?i)xtaar-uu-haa

EA
?al-u-li
ba3-u-lak
(?i)xtar-u-ha

Gloss
They said to me.
They sold to you.
They chose her.

What really distinguishes EA from SA is a tendency for the glides

or weak consonants /w/ and /y/ to be more and more treated as strong
consonants in EA. This was hinted at by Carter (1996: 138): There are
now many instances where the semi-vowels w and y function as full
consonants, creating (or recreating?) new paradigms. As Tables (11) and
(12) reveal, the glides appear in Forms II, III, V, VI and IX in both
varieties. The innovation in EA is that these consonants are beginning to
appear in Forms I, VII and X as well. Other EA examples of verbs with
glides in the medial position are:

79

Form
I
"
"
VII
"
X
"
"
"

Example
hawag
rawa
ziwir
(?i)nhawal
(?i)nhawag
(?i)sta3wa?
(?i)stabya3
(?i)staxwin
(?i)staxyib

Gloss
to expose to need
to disturb
to choke
to develop a squint
to be exposed to need
to consider late
to act recklessly
to consider untrustworthy
to consider a failure

These examples reveal that EA has developed its own forms with /w/ and
/y/ which are non-existent in SA. To explain this I, propose that the rules
of Glide Elision (in Forms I and VII) and Anticipatory Vocoid
Assimilation (in Form X) are losing ground in EA. As a result, the
traditional hollow forms and the new strong forms now exist side by side
in this variety. If this is the correct analysis, one might expect that in the
future EA may also develop a new form IV, instead of or together with
[?aFaaL] by simply retaining the underlying form [?aFGaL] (if Form IV
survived at all, because there are indications that this form is beginning to
disappear).
Tables (13) and (14) offer the imperfect forms of SA and EA
medial-weak verbs, respectively:
Table (13)
Imperfect Forms of Medial-Weak Verbs in SA
No Form
Example
ya-naam(-u) to sleep
aa
I
ya-quul(-u) to say

ya-F uu L(-u)
ya-giib(-u) to be absent
ii

yu-Sawwir(-u) to photograph
ww
II yu-Fa iL(-u) yu-3ayyin(-u) to appoint
yy
yu-haawil(-u) to try
w
III yu-Faa iL(-u)
yu-3aayin(-u) to inspect
y
IV yu-FiiL(-u)
yu-biid(-u) to eradicate
ya-taSawwar(-u) to imagine
ww
V ya-taFa aL(-u) ya-taxayyal(-u) to imagine
yy

80

w
VI ya-taFaa aL(-u)
y
VII ya-nFaaL(-u)
VIII ya-FtaaL(-u)
w
IX ya-F aLL(-u)
y
X ya-staFiiL(-u)

No
Ia

Ib
IIa
IIb
III
IV
Va

ya-ta3aawan(-u) to cooperate
ya-ta3aaya(-u) to live together
ya-mbaa3(-u) to be sold
ya-xtaar(-u) to choose
ya-swadd(-u) to become black
ya-byaDD(-u) to become white
ya-stafiid(-u) to benefit

Table (14)
Imperfect Forms of Medial-Weak Verbs in EA
Form
Example
yi-naam to sleep
aa
yi-?uul to say

yi-F uu L
yi-giib to be absent
ii

yi-dwi to bother
w
yi-F iL
yi-xyil to distract
y
yi-Sawwar to photograph
ww
yi-Fa aL yi-3ayyaT to weep
yy
yi-kawwin to form
ww
yi-Fa iL
yi-3ayyin to appoint
yy
w
yi-Faa iL
y
yi-FiiL
ww
yi-tFa aL
yy

yi-haawil to try
yi-3aayin to inspect
yi-biid to eradicate
yi-tSawwar to imagine
yi-txayyal to imagine

ww
yi-tFa iL
yy
w
VI
yi-tFaa iL
y
VIIa yi-nFaaL
w
VIIb yi-nFi iL
y

yi-tkawwin to be formed
yi-t3ayyin to be appointed

VIII yi-FtaaL

yi-xtaar to choose

Vb

yi-t3aawin to cooperate
yi-t3aayin to get inspected
yi-mbaa3 to be sold
yi-nxiwit to be pestered
yi-ndiyin to fall into debt

81

IX
Xa
Xb

w
yi-F aLL
y
aa
yi-staF L
ii
w
yi-staF iL
y

yi-swadd to become black

yi-byaDD to become white
yi-stafaad to benefit
yi-stafiid to benefit
yi-stabwix to consider silly
yi-sta3yib to consider disgraceful

Comparing Tables (13) and (14), on the one hand, with Tables (11)
and (12), on the other, shows that both varieties under analysis are similar
in that the two rules responsible for the change of a medial glide into a
long vowel apply to the imperfect forms of medial-weak verbs. Glide
Elision applies to Forms I, VII and VIII, and Anticipatory Vocoid
Assimilation applies to Forms IV and X. The reasons why the former rule
does not apply to Forms II and V or the latter rule to Form IX were stated
in the discussion of the changes occurring in the perfect forms. An
interesting discrepancy between SA and EA is displayed in Forms I, VII
and X. Although Anticipatory Vocoid Assimilation always applies to all
verbs of Form I and X in SA, it applies to some verbs of these forms only
in EA, while other verbs keep their underlying forms [yi-FGiL] and
[yi-staFGiL]. Also, though Glide Elision applies to all imperfect verbs of
Form VII in SA, it applies to some verbs of this form only in EA. This
affirms our assumption that the scope of these rules is diminishing in EA.
The imperfect forms of the primary form of medial-weak verbs in
SA are: [ya-FaaL(-u)], [ya-FuuL(-u)] and [ya-FiiL(-u)], with the first one
as the least frequent. They are retained in EA, with the deletion of the
inflectional suffix and the change of the imperfect prefix to [yi-]:
Root
x-w-f
q-w-l
b-y-3
3-y-

SA
ya-xaaf(-u)
ya-quul(-u)
ya-bii3(-u)
ya-3ii(-u)

EA
yi-xaaf
yi-?uul
yi-bii3
yi-3ii

Gloss
to be afraid
to say
to sell
to live

It is apparent from the examples above that the imperfect forms

[ya-FaaL(-u) > yi-FaaL] and [ya-FuuL(-u) > yi-FuuL] are used in w-verbs.
On the other hand, the form [ya-FiiL(-u) > yi-FiiL] only is utilized in
y-verbs. Exceptionally, the imperfect forms of two EA verbs do not
conform to their counterparts in SA; /baat(-a)/ to stay overnight and
/baan(-a)/ to appear have the imperfect forms /ya-biit(-u)/ and

82

/ya-biin(-u)/ in the Standard variety, but /yi-baat/ and /yi-baan/ in the

Colloquial. Another exception is the verb for to come which has the form
/gaa?(-a), ya-gii?(-u)/ in SA but /ga or gih, yiigi/ in EA.
Final-weak verbs are those having a glide or a vowel in place of
the last radical of the root. The perfect forms of final-weak verbs can be
observed in the two varieties by comparing Tables (15) and (16).
Table (15)
Perfect Forms of Final-Weak Verbs in SA
No Form
Example
Ia
Fa3aa
ramaa to throw
Ib
Fa3iy(-a)
nasiy(-a) to forget
II
Fa33aa
Saffaa to strain
III
Faa3aa
naadaa to call
IV
?aF3aa
?agraa to entice
V
taFa33aa
tasammaa to be named
VI
taFaa3aa
tafaadaa to avoid
VII (?i)nFa3aa (?i)mbaraa to be sharpened
VIII (?i)Fta3aa
(?i)taraa to buy
X
(?i)staF3aa (?i)stagnaa to do without
Table (16)
Perfect Forms of Final-Weak Verbs in EA
No Form
Example
a a rama to throw
I
F 3 nisi to forget
i i
II
Fa33a
Saffa to strain
III
Faa3a
naada to call
IV
?aF3a
?agra to entice
V
(?i)tFa33a
(?i)tsamma to be named
VI
(?i)tFaa3a
(?i)tfaada to avoid
VII (?i)nFa3a
(?i)mbara to be sharpened
VIII (?i)Fta3a
(?i)tara to buy
X
(?i)staF3a
(?i)stagna to do without
Comparison of Tables (15) and (16) reveals that the glide /y/ at the
end of an SA verb stem form is deleted in the EA surface form, though it
is still found in the underlying form, as evidenced by its reappearance in
the verbal noun, e.g. /nisi/ to forget /nisyaan/ forgetfulness. The loss
of the surface /y/ then is a by-product of the general rules deleting /y/. The
comparison also shows that SA final /aa/ is shortened in EA as is regular.

83

Moreover, the tables reveal that Form IX of final-weak verbs is not found
in either variety, though it is found in one EA example: /(?i)hlaww/ to
become beautiful, which suggests that this form is beginning to emerge in
EA.
Final-weak verbs in SA have two patterns for the primary form:
[Fa3aa] whose final radical may be originally /w/ or /y/ and [Fa3iy(-a)]. In
EA, the first pattern becomes [Fa3a] by Final-Vowel Shortening in verbs
with original /w/ or /y/ and [Fi3i] in verbs with original /y/ only. The
second pattern is reduced to [Fa3a] in some verbs and [Fi3i] in some
others, as in these examples:
Root
d-3-w
r-m-y
m--y
l-q-y
r-D-w

SA
da3aa
ramaa
maaa
laqiy(-a)
raDiy(-a)

EA
da3a
rama
mii
la?a
riDi

Gloss
to invite
to throw
to walk
to meet/find
to be content

The first two examples reveal that [Fa3aG(-a)] becomes [Fa3aa] in

SA by Glide Elision, then becomes [Fa3a] by Final-Vowel Shortening in
EA. The change of SA [Fa3aa] which is underlyingly [Fa3ay(-a)] into EA
[Fi3i] is a morphological one since it involves a change of the vocalism
from /a/ to /i/, e.g. /maay(-a) maaa > miiy- mii/ to walk.
Again, the change of SA [Fa3iy(-a)] into EA [Fa3a] is a morphological
change since it involves the replacement of /i/ by /a/ for the stem vowel,
e.g. /laqiy(-a) > la?a/ to find. EA verbs with the perfect pattern [Fi3i] are
underlyingly [Fi3iy] with the raising of the first vowel of its SA
counterpart. It becomes [Fi3ii] by Perseverative Vocoid Assimilation and
then [Fi3i] by Final-Vowel Shortening, as in /raDiy(-a) > riDi/ to be
content. However, the raising of the first vowel from /a/ to /i/ in EA is a
morphological change.
Phonologically, EA differs from SA in the case of pronominal
suffixation to final-weak verbs. Although the sequences /aw/ and /ay/ are
employed before a pronoun starting with a consonant in SA, they are
replaced by the long vowel /ee/ in EA. This vowel is created from /ay/ by
Monophthongization and is then segmented off as an independent affix to
apply in other contexts such as consonantal suffixation to geminate verbs
(2.1.1.2) and final-glottalized verbs (2.1.1.3). (In the first example below
the /w/ is changed to /y/ before Monophthongization applies):
Form
I

SA
da3aw-tu

EA
da3ee-t

Gloss
I invited

84

II
III
IV
V

Sallay-naa
naaday-ti
?agray-tum
tahadday-ta

Sallee-na
nadee-ti
?agree-tu
(?i)thaddee-t

we prayed
you (fsg) called out
you (mpl) enticed
you (msg) challenged

However, if the final-weak verb ends in /i/ in the Colloquial variety, it is

simply re-lengthened (i.e. not shortened) before all suffixes except the
third person feminine singular marker:
SA
raDii-tu
saxii-ta
qawii-naa

EA
riDii-t
sixii-t
?iwii-na

Gloss
I was satisfied
you (msg) became generous
we became strong

In the case of suffixing the third person feminine singular marker [-it], the
underlying form in EA would be [Fi3iy-it], then the second /i/ is
syncopated in an open medial syllable, as in /maa-t > miy-it/ she
walked. Notice that the rule of Anticipatory Vocoid Assimilation does not
apply to the Colloquial form here, which suggests that Syncope should be
ordered after that rule. If we apply Syncope first, the result would be
/miiy-it miy-it *miiit/.
The imperfect forms of final-weak verbs in SA and EA are
contrasted in Tables (17) and (18).
Table (17)
Imperfect Forms of Final-Weak Verbs in SA
No Form
Example
ya-rjuu
to implore
uu
I
ya-rmii to throw

ya-F3 ii
ya-nsaa to forget
aa

II
yu-Fa33ii
yu-Saffii to strain
III
yu-Faa3ii
yu-naadii to call
IV
yu-F3ii
yu-grii to entice
V
ya-taFa33aa
ya-tasammaa to be named
VI
ya-taFaa3aa
ya-tafaadaa to avoid
VII ya-nFa3ii
ya-mbarii to be sharpened
VIII ya-Fta3ii
ya-tarii to buy
X
ya-staF3ii
ya-stagnii to do without

85

Table (18)
Imperfect Forms of Final-Weak Verbs in EA
No Form
Example
yi-rgu to implore
u
I
yi-rmi to throw

yi-F3 i
yi-nsa to forget
a

II
yi-Fa33i
yi-Saffi to strain
III
yi-Faa3i
yi-naadi to call
IV
yi-F3i
yi-gri to entice
V
yi-tFa33a
yi-tsamma to be named
VI
yi-tFaa3a
yi-tFaada to avoid
VII yi-nFi3i
yi-mbiri to be sharpened
VIII yi-Fti3i
yi-tiri to buy
IX
yi-F3aww
yi-hlaww to become sweet
X
yi-staF3i
yi-stagni to do without
Table (17) shows the effects of the Glide Elision rule: all the final
long vowels in final-weak verbs arise from this rule and subsequent FinalVowel Shortening in EA gives the final short vowels in these forms in
Table (18). Close investigation of the perfect forms (in Tables 15 and 16)
and the imperfect forms (in Tables 17 and 18) shows that final-weak verbs
have two consonants in their surface structure although they have three
consonants in their underlying structure. Traditionally, the third radical
can be determined from the imperfect form, because the SA final long
vowel (which is shortened in EA) is realized as a sequence of a short
vowel similar to the long vowel of the imperfect and a glide which is
cognate to that vowel. In other words, /uu/ becomes /uw/, /ii/ becomes /iy/
and /aa/ becomes /VG/ where /G/ is an unspecified glide (cf. Mahadin
1982: 222-3).
Returning to Tables (17) and (18), it is noted that final-weak verbs
have three patterns for the SA primary form in the imperfect: [ya-F3uu]
for verbs with /aa/ in the perfect whose original final radical is /w/; [yaF3ii] for verbs with /aa/ in the perfect whose original final radical is /y/;
and [ya-F3aa] for verbs whose perfect is [Fa3iy(-a)]. In EA, the first
pattern has two morphological counterparts [yi-F3i] and [yi-F3a], the
second pattern has one equivalent [yi-F3i] by Final-Vowel Shortening and
the last one has one analogue [yi-F3a] also by Final-Vowel Shortening. In
other words, a major difference between the two varieties is that when the
long vowel /uu/ occurs in the final position of the imperfect form of SA
verbs, it is replaced in EA by /i/ or /a/:

86

Root
d-3-w
3-l-w
S-f-w
r-m-y
m--y
l-q-y
r-D-w

SA
ya-d3uu
ya-3luu
ya-Sfuu
ya-rmii
ya-mii
ya-lqaa
ya-rDaa

EA
yi-d3i
yi-3la
yi-Sfa
yi-rmi
yi-mi
yi-l?a
yi-rDa

Gloss
to invite
to go up
to become pure
to throw
to walk
to meet/find
to be satisfied

The first three examples indicate that the innovation of EA consists in that
some verbs having final /w/ in their roots have imperfects with /a/ while
others have imperfects with /i/. Hence, the following development occurs
in EA:
(84) a. d-3-w:
b. 3-l-w:

yi-d3iw yi-d3iy yi-d3i

yi-3law yi-3laa yi-3la

This is in keeping with the fact that EA seems to have innovated contrasts
in the second vocalism where none are present in SA (See 2.1.1.1 above).
Some weak verbs include two glides. Hence, they are termed
doubly weak verbs (Mahadin 1982: 217). They are mainly final-weak
verbs that have another glide either in the initial or medial position. They
share the characteristics of the initial-weak and/or the final-weak verbs. In
the development from SA to EA, doubly weak verbs exemplify the same
changes as are expected from the combination of behaviors shown by the
singly weak verbs. For instance, the verb /waaa > waa/ to inform on
someone is derived from [waay(-a)] by Glide Elision in both varieties,
then by Final-Vowel Shortening in EA, as in (85a). The imperfect form is
/ya-ii/ by w-Occultation from /ya-wii/ in SA, as in (85b), but nothing
happens to the initial radical in EA /yi-wi/, as in (85c):
(85) a. waay-a (SA) waaa (EA) waa
b. ya-wiy-u ya-wiu ya-wii (SA) ya-ii
c. yi-wiy yi-wii (EA) yi-wi
Other examples of doubly-weak verbs are:
SA
EA
Root
w-f-y
wafaa
wafa/waffa
n-w-y
nawaa
nawa
q-w-y
qawiy(-a)
?iwi

Gloss
to fulfill
to intend
to become strong

87

h-y-y

hayiy(-a)

hiyi

The imperfect forms of these verbs are:

Root
SA
EA
w-f-y
ya-Fii
yuufi
n-w-y
ya-nwii
yi-nwi
q-w-y
ya-qwaa
yi-?wa
h-y-y
ya-hyaa
yi-hya

to survive

Gloss
to fulfill
to intend
to become strong
to survive

2.1.2. Quadriradical Verbs:

There are two main classes of quadriradical verbs in Arabic: sound
and reduplicated or doubled. The former is of four different radicals while
the latter has two radical consonants reduplicated. In the traditional
numbering system quadriradicals have four forms. Though Form III
[(?i)F3anL1aL2(-a)] is found in Classical Arabic, e.g. /(?i)slanTah(-a)/ to
be laid down flat (Thackston 1984: 171), it is no longer used in SA or EA.
As for the semantics of these forms, Form I is non-stative transitive or
intransitive (like Form I triradical), Form II is passive or intransitive of
Form I (like Form V triradical) and Form IV involves developing a state
(like Form IX triradical).
The number of possible forms for the sound quadriradical verb in
both SA and EA is limited to three, the first of which is the primary form
and the other two are the derived forms. Compare Tables (19) and (20)
below, taking into consideration that L1 and L2 are two different
consonants.
Table (19)
Perfect Forms of Sound Quadriradical Verbs in SA
No Form
Example
I
Fa3L1aL2(-a)
zaxraf(-a) to decorate
II taFa3L1aL2(-a)
taayTan(-a) to act like a devil
IV (?i)F3aL1aL2L2(-a)
(?i)Tma?ann(-a) to feel secure
Table (20)
Perfect Forms of Sound Quadriradical Verbs in EA
No Form
Example
zaxraf to decorate
a
I
Fa3L1 L2
targim to translate
i

88

a
(?i)tFa3L1 L2
i
IV (?i)F3aL1aL2L2

II

(?i)tayTan to act like a devil

(?i)ta3lil to flare up
(?i)Tma?ann to feel secure

By way of comparison, one can figure out the following discrepancies:

(i) The Colloquial forms drop the final (-a) of the perfect owing to the
general rule in EA to drop SA vowels used for inflection. Thus, since
the vowel (-a) at the end of the Standard forms is employed as an
inflectional marker of the third person masculine singular, it is omitted
in EA.
(ii)The Colloquial form [Fa3L1aL2 ~ Fa3L1iL2] has the pre-final /i/ as a
variant of /a/, while the Standard counterpart has only /a/ as a pre-final
vowel.
(iii)The second EA form [(?i)tFa3L1aL2] represents a reflex of the SA
form [taFa3L1aL2(-a)] as a result of the reanalysis of the prefix from SA
[ta-] to EA [t-] with (?i) being epenthetic.
Table (21) displays the imperfect forms of SA quadriradical verbs,
whereas Table (22) exhibits their EA equivalents:

No
I
II
IV

No
I
II
IV

Table (21)
Imperfect Forms of Sound Quadriradical Verbs in SA
Form
Example
yu-Fa3L1iL2(-u)
yu-zaxrif(-u) to decorate
ya-taFa3L1aL2(-u)
ya-taayTan(-u) to act like a devil
ya-F3aL1iL2L2(-u)
ya-Tma?inn(-u) to feel secure
Table (22)
Imperfect Forms of Sound Quadriradical Verbs in EA
Form
Example
yi-zaxraf to decorate
a
yi-Fa3L1 L2
yi-targim to translate
i
yi-tayTan to act like a devil
a
yi-tFa3L1 L2
yi-ta3lil to flare up
i
yi-F3aL1iL2L2
yi-Tma?inn to feel secure

Turning to the second class of quadriradical verbs, it is noted that

the doubled quadriradical verbs include one or two reduplicated
consonants. They have two forms in SA and three in EA, as shown in
Tables (23) and (24), respectively. All of these forms are basic; so they are

89

marked with the traditional quadriradical number I. But in this book they
are given lower case letters to indicate the difference in the position of
reduplicated consonant.

No
Ia
Ib
II

No
Ia
Ib
Ic
II

Table (23)
Perfect Forms of Doubled Quadriradical Verbs in SA
Form
Example
Fa3Fa3(-a)
zalzal(-a) to shake
Fa3FaL(-a)
samsar(-a) to act as a broker
taFa3Fa3(-a)
tazalzal(-a) to be shaken
Table (24)
Perfect Forms of Doubled Quadriradical Verbs in EA
Form
Example
SahSah to be come alert
a
Fa3F 3
zalzil to shake
i
samsar to act as a broker
a
Fa3F L
dardi to chat
i
za?TaT to be in high spirits
a
Fa3L1 L1
3aknin to disturb
i
(?i)tFa3Fi3
(?i)tzalzil to be shaken

Comparison of Tables (23) and (24) shows that quadriradical forms with
the reduplication of the third radical are only found in EA.
The imperfect forms of the reduplicated quadriradical verbs in SA
and EA are given in Tables (25) and (26), respectively.

No
Ia
Ib
II

Table (25)
Imperfect Forms of Doubled Quadriradical Verbs in SA
Form
Example
yu-Fa3Fi3(-u)
yu-zalzil(-u) to shake
yu-Fa3FiL(-u)
yu-samsir(-u) to act as a broker
ya-taFa3Fa3(-u)
ya-tazalzal(-u) to be shaken

No
Ia

Table (26)
Imperfect Forms of Doubled Quadriradical Verbs in EA
Form
Example
yi-Fa3Fi3
yi-zalzil to shake

90

Ib
Ic
II

a
yi-Fa3F L
i
a
yi-Fa3L1 L1
i
yi-tFa3Fi3

yi-samsar to act as a broker

yi-dardi to chat
yi-za?TaT to be in high spirits
yi-3aknin to disturb
yi-tzalzil to be shaken

2.2. Inflection for Aspect & Mood:

Both varieties under discussion express two aspects of the verb by
employing morphological devices: the perfect and the imperfect. The
former is used for finished or completed action. The latter denotes
uncompleted action (including habitual, on-going or future action)14.
Although some linguists (e.g. Eisele 1990: 173) claim that these
correspond to a distinction between past and non-past tense, others affirm
that there is no one-to-one correspondence between aspect and tense.
Thus, Radwan (1975: 30) asserts that:
Aspect and tense should be treated as two independent
categories.... Both terms are used to name two different
features of verbal patterns. The term Aspect covers the
semantic ranges of completion versus non-completion and
continuation versus non-continuation, whereas Tense
covers time reference.
Morphologically, the perfect form is obtained by the attachment of
suffixes only, whereas the imperfect form is obtained via the addition of
confixes, i.e. combinations of prefixes and suffixes (Yushmanov 1961:
28). Along with the inflection for aspect, Arabic verbs are inflected for
mood. SA verbs are conjugated for one mood in the perfect aspect - i.e.,
the indicative - and for three moods in the imperfect: indicative,
subjunctive and jussive; in addition to the imperative mood. The crossclassification of SA verbs for aspect and mood can be illustrated as
follows:
(86) Cross-Classification of Aspect & Mood in SA:
Indicative
Subjunctive Jussive
Perfect
+

Imperfect
+
+
+
EA, on the other hand, shows only one mood - i.e. the indicative, in both
perfect and imperfect aspect; aside from the imperative.

91

The conjugation for aspect/mood involves the attachment of

agreement markers to the verb stem. These markers are actually
subjectival affixes because they display the agreement of the verb with the
subject in person, gender and number.
Table (27) exhibits the perfect indicative markers used with the
verb /daxal(-a)/ to enter in both varieties:
Table (27)
The Perfect Indicative Markers in SA & EA
Referent
SA
EA
1sg (m & f)
daxal-tu
daxal-t
1pl (m & f)
daxal-naa
daxal-na
2msg
daxal-ta
daxal-t
2fsg
daxal-ti
daxal-ti
2du (m & f)
daxal-tumaa

2mpl
daxal-tum
daxal-tu15
2fpl
daxal-tunna

3msg
daxal-a
daxal-
3fsg
daxal-at
daxal-it
3mdu
daxal-aa

3fdu
daxal-ataa
daxal-u
3mpl
daxal-uu

3fpl
daxal-na

A noticeable morphophonemic difference between SA and EA is

that in the latter the rule of High-Vowel Deletion results in the dropping of
the vowel /i/ of the medial radical when the perfect suffix starts with a
vowel. This kind of vowel elision is common in the EA verb form [Fi3iL]
which is the counterpart of the two SA forms [Fa3iL(-a)] and [Fa3ul(-a)]
(cf. Malik 1976: 92):
SA
arib-atirb-it
arib-uu
kabur-at
kabur-uu

EA

Gloss
she drank
irb-u
they drank
kibr-it
she grew up
kibr-u
they grew up

The same morphophonemic feature involves High-Vowel Deletion

in the junction of the pronoun /?ana/ I and a verb of the form [Fi3iL] in
EA. For instance, SA /?anaa arib-tu/ I drank has the EA equivalent
/?ana rib-t/ with the initial vowel /i/ dropped. A similar example is /?anaa

92

kabir-tu > ?ana kbir-t/ I grew up. The reason behind this elision is that
the vowel is in an open medial syllable.
The imperfect form of the verb is obtained in the Standard variety
by adding one of four prefixes to the imperfect verb stem: [?a-, na-, ya- or
ta-] in conformity with the referent. The imperfect stem of Form I, for
example, has the template [-CCVC-]. The Colloquial variety, on the other
hand, has innovated in the formation of the imperfect. This can be
observed by comparing the imperfect markers in the two varieties, as
illustrated in Table (28).
Table (28)
The Imperfect Markers in SA & EA
Referent
SA
EA
Indicative
Subjunctive Jussive
Indicative
1sg (m & f) ?a-dxul-u
?a-dxul-a
?a-dxul-
(?)a-dxul-
1pl (m & f) na-dxul-u
na-dxul-a
na-dxul-
nu-dxul-
2msg
ta-dxul-u
ta-dxul-a
ta-dxul-
tu-dxul-
2fsg
ta-dxul-iina
ta-dxul-ii
ta-dxul-ii
tu-dxul-i
2du (m & f) ta-dxul-aani
ta-dxul-aa
ta-dxul-aa

2mpl
ta-dxul-uuna ta-dxul-uu ta-dxul-uu
tu-dxul-u
2fpl
ta-dxul-na
ta-dxul-na
ta-dxul-na

3msg
ya-dxul-u
ya-dxul-a
ya-dxul-
yu-dxul-
3fsg
ta-dxul-u
ta-dxul-a
ta-dxul-
tu-dxul-
3mdu
ya-dxul-aani ya-dxul-aa ya-dxul-aa

3fdu
ta-dxul-aani
ta-dxul-aa
ta-dxul-aa

3mpl
ya-dxul-uuna ya-dxul-uu ya-dxul-uu
yu-dxul-u
3fpl
ya-dxul-na
ya-dxul-na ya-dxul-na

The comparison of SA and EA imperfect markers reveals that the

SA dual and feminine plural markers have disappeared in EA. The
masculine plural markers, which can be considered the unmarked plural
markers, are used in their place. Moreover, the SA indicative suffix [-u]
has disappeared in EA. Disyllabic indicative suffixes have lost the second
syllable [-na] and have had their long vowels shortened by Final-Vowel
Shortening, i.e. [-iina] has been reduced to [-i] and [-uuna] to [-u].
A remarkable difference between the two varieties is that although
EA has preserved the SA imperfect prefixes, it has replaced the vowel in
them by /i/ with the exception of [(?)a-] where the vowel is retained.
However, in some verbs, particularly those whose stem vowel is /u/, the
imperfect vowel is changed to /u/, again with the exception of [(?)a-], as
follows:

93

SA
ya-dxul(-u)
na-rbuT(-u)
ta-xnuq(-u)
?a-ktub(-u)

EA
yu-dxul
nu-rbuT
tu-xnu?
?a-ktib

Gloss
he enters
we tie
she strangles
I write

When /u/ occurs in the imperfect prefix in SA, as in Forms II and

III, [Fa33aL(-a), yu-Fa33iL(-u)] and [Faa3aL(-a), yu-Faa3iL(-u)], it is
replaced by /i/ in EA, as in the following verbs:
SA
yu-?axxir(-u)
tu-?akkil(-u)
nu-3allim(-u)
yu-aakir(-u)
tu-haarib(-u)
nu-saafir(-u)

EA
yi-?axxar
ti-?akkil
ni-3allim
yi-zaakir
ti-haarib
ni-saafir

Gloss
he delays
she feeds
we teach
he studies
you fight
we travel

It should be stressed here that the imperfect prefixes of all verbs in

both varieties contain a short vowel. Exceptions in the Colloquial variety
are stated by Abdel-Malek (1972: 42):
There are only three verbs whose present tense forms have
long vowels in the prefix: those are the verbs /ga/ to
come, /axad/ to take and /akal/ to eat. The form /aagi/ I
come has /aa/ in the prefix; all the other present tense
forms of the verb /ga/ have /ii/ in the prefix. The present
tense forms of the verbs /axad/ and /akal/ have /aa/ in the
prefix if the stem begins with one consonant; if the stem
begins with two consonants, the vowel of the prefix is /a/.
Though the verb /ga/ is a real exception, the other two verbs are
exceptional only in having /a/ in the prefix rather than /i/. Then, we can
propose that the long vowel /aa/ at the beginning of the imperfect of
/?axad/ and /?akal/ is derived from /a?/ by Compensatory Lengthening, as
in (87). That the vowel of the prefix is /a/ in certain contexts can be
accounted for by the application of Shortening rules to the product of
Compensatory Lengthening, as in (88). (Subjectival & Objectival Affixes
will be discussed in 2.4):
(87) a. ya-?xud yaaxud he takes

94

b. na-?kul naakul we eat

(88) a. ya-?xud-u ya-?xd-u (by High-Vowel Deletion)
yaaxd-u (by Compensatory Lengthening yaxd-u (by ClosedSyllable Shortening) He takes it (m).
b. na-?kul-ha naakul-ha (by Compensatory Lengthening)
nakul-ha (by Atonic Shortening) We eat it (f).
Careful investigation of Table (28) reveals that the three SA moods
- indicative, subjunctive and jussive - are distinguished by means of
suffixes only, since the prefixes are the same in all moods. So, one can
safely propose that the prefixes there are just imperfect aspect markers
which do not express any mood. The table displays that the subjunctive is
formed by changing the [-u] of the indicative into [-a] and cutting off the
[-nV] of the two-syllable inflections, while the feminine plurals are not
changed. The jussive is formed by dropping the final [-u] of the indicative
if present, otherwise it is the same as the subjunctive.
Malik (1976: 357-8) refers to certain cases in which there is no
apparent change between the indicative and the subjunctive moods. One of
these cases is the final-weak verbs ending in /aa/, i.e. [ya-F3aa] which is
derived from [ya-F3aG(-u)] in the indicative by Glide Elision and
Perseverative Vocoid Assimilation and from [ya-F3aG(-a)] in the
subjunctive by Glide Elision alone, e.g. /ya-nsaa/ (both indic and subj) to
forget. However, verbs ending in /uu/ and /ii/ show a distinction between
indicative and subjunctive because, though Glide Elision applies in the
indicative, its environment is not present in the subjunctive. So, the
indicative form is [ya-F3uu] derived from [ya-F3uG(-u)] by Glide Elision,
while the subjunctive form is [ya-F3uG(-a)] because the condition on the
rule is not satisfied as /u/ is not similar in height to /a/, e.g. /ya-d3uu/
(indic) vs. /ya-d3uw(-a)/ (subj) to invite. Likewise, the indicative form
[ya-F3ii] is derived from [ya-F3iG(-u)] by Glide Elision and Perseverative
Vocoid Assimilation, while the subjunctive form is [ya-F3iG(-a)] again
because the condition on Glide Elision is not satisfied since /i/ is not as
low as /a/, e.g. /ya-rmii/ (indic) vs. /ya-rmiy(-a)/ (subj) to throw.
In addition, there is a case where there is no distinction between
the indicative and either the subjunctive or the jussive. This is the case of
all the imperfect forms of the feminine plural whether second or third
persons, e.g. /ta-l3ab-na/ you (fpl) are playing and /ya-l3ab-na/ they
(fpl) are playing.

95

In medial-weak or final-weak verbs having long vowels instead of

the middle or last radical, these long vowels are shortened in the jussive
mood. This shortening can be accounted for by the rule of Closed-Syllable
Shortening in medial-weak verbs, but it is not yet known why the
shortening occurs in final-weak verbs. Compare these examples in SA:
Indicative
ya-3uum(-u)
ya-bii3(-u)
ya-naam(-u)
ya-nsaa
ya-d3uu
ya-rmii

Jussive
ya-3um
ya-bi3
ya-nam
ya-nsa
ya-d3u
ya-rmi

Gloss
to swim
to sell
to sleep
to forget
to invite
to throw

Some linguists attempted to define the function of the perfect and

imperfect markers. For instance, Mahadin (1982: 138) states: One of the
differences between the perfect and the imperfect is that while in the
perfect suffixes mark person, number and gender, in the imperfect person
is marked by the prefixes and gender, number and mood is [sic] marked by
suffixes [italics mine]. Although this generalization seems definitive,
careful investigation of Table (28) shows that the last (italicized) part is
not quite correct. Mahadin says that gender and number are marked by
suffixes in the imperfect. However, gender is also marked by prefixes, e.g.
/ya-13ab(-u)/ he plays vs. /ta-l3ab(-u)/ she plays. Also, number is
marked by prefixes, e.g. /?a-13ab(-u)/ I play vs. /na-13ab(-u)/ we play.
So, Mahadins generalization might be reformulated as follows: in the
imperfect, person is marked by prefixes; gender and number are marked
by both prefixes and suffixes. (For a full discussion of the subjectival
affixes, see 2.4).
As stated at the beginning of this section, the perfect form
generally refers to past time or finished action. This can be noted in the
Standard example in (89a) and its Colloquial counterpart in (89b):
(89) a. katab-tu
d-dars-a
write.pf-1sg
the-lesson-Acc
I wrote the lesson yesterday.
b. katab-t
id-dars
write.pf-1sg
the-lesson
I wrote the lesson yesterday.

?ams
yesterday
imbaarih
yesterday

However, the perfect form may refer to the present or the future. It can be
used with the Standard present time adverbial /?al?aan/ now or its

96

Colloquial reflex /dilwa?ti/ to produce a meaning quite similar to the

present perfect in English (See Eisele 1990: 175). It may also refer to the
future if used in conditional sentences.
On the other hand, the imperfect form denotes the present (habitual
or progressive). Nevertheless, EA differs from SA in this respect. The
former precedes the imperfect prefixes by an additional prefix [bi-] to
signal progressive or habitual activities, as in (90):
ya-rab-u
(90) SA: ?al-walad-u
the-boy-Nom
impf-drink-3msg
The boy drinks / is drinking the milk.
EA: ?il-walad
bi-yi-rab
the-boy
impf-impf-drink
The boy drinks / is drinking the milk.

1-laban
the-milk
il-laban
the-milk

It can be stated here that for the first person singular, the prefix
[bi-] combines with the imperfect prefix [(?)a-] to form one morpheme
[ba-]. For instance, /bi-?a-ftah/ I open / am opening is reduced to
/ba-ftah/. This indicates that the glottal stop in [?a-] is not underlying in
EA, i.e. it is inserted in the surface structure by Glottal-Stop Insertion. If
the underlying form of the first singular prefix is [a-], the we can propose
that /bi-?a-ftah/ becomes /baa-ftah/ by Anticipatory Vocoid Assimilation,
then /ba-ftah/ by Closed-Syllable Shortening.
The imperfect form may also refer to the future. But what seems to
distinguish EA in this respect is the use of non-Standard markers. SA
future markers are the prefix [sa-] and the particle /sawfa/ will which are
preposed to the imperfect form. They are replaced in EA by [ha- or ha-]
which are also prefixed to the imperfect form. I follow Robertson (1970:
213) in proposing that such prefixes have probably developed from the
verb /raah(-a)/ to go. For the first person singular, [ha-] combines with
[?a-] to constitute one morpheme [ha-]. Compare the Standard examples in
(91-92a) with their Colloquial analogues in (91-92b), respectively:
(91) a. sa-?a-ktub-u
fut-impf-write-1sg
I will write the lesson.
b. ha-ktib
fut-write.1sg
I will write the lesson.

d-dars
the-lesson
id-dars
the-lesson

97

(92) a. sawfa
?aakul-u
will
impf.eat-1sg
I will eat the grapes.
il-3inab
b. haakul
fut.eat1sg
the-grapes
I will eat the grapes.

1-3inab
the-grapes

Comparison of SA /sa-?a-ktub-u/ with EA /ha-ktib/ confirms the fact that

although the glottal stop is still underlying in SA, it is no longer
underlying in EA. Thus, neither SA */sa-ktub-u/ nor EA */ha-?a-ktib/ is
acceptable.
Other tenses are indicated by employing the auxiliary verb
/kaan(-a)/ to be and/or the particle /qad/ may, might or did plus the
proper aspectual form of the verb. However, this is a syntactic issue which
is beyond the scope of the present book. (For a good discussion of the
syntactic expression of tenses, see Radwan (1975) and Kharma (1983)).
Related to the discussion of the aspect/mood conjugation is the
formation of the imperative. According to some linguists, the imperative
mood is derived in SA from the jussive. Schramm (1962: 364), for
instance, states a rule for this derivation:
The traditional rule of thumb for the derivation of the
imperative proceeds from the jussive: the first two
phonemes of the second persons of the jussive are
subtracted and if the residue begins with a consonant
cluster, /?/ plus a vowel is prefixed. Generally, if the stem
vowel of the residue is /u/, the prefixed vowel is /u/,
otherwise the prefixed vowel is /i/16.
There is of course a difference between a rule of thumb and an actual
grammatical rule. The former is a useful device for a learner, the latter is a
hypothesis about how native speakers unconsciously know what form to
produce. It is not yet known whether it is linguistically motivated to speak
of the imperative as being derived from the jussive, although they may
share similar properties. This can be evidenced by the fact that in EA there
is no jussive. Thus, one can propose, contrary to Schramms hypothesis,
that both the jussive in SA and the imperative in both varieties are derived
from the imperfect stem of the verb. Consider these examples:
SA
impf
ya-nzil(-u)

EA
imper
?i-nzil

impf
yi-nzil

Gloss
imper
?i-nzil

to descend

98

ya-dfa3(-u)
ya-Drib(-u)

?i-dfa3
?i-Drib

yi-dfa3
yi-Drab

?i-dfa3
?i-Drab

to pay
to hit

One can notice that where the imperfect stems are identical in both
varieties, the imperatives are also identical and where the imperfect stems
are different, the imperatives are different. This supports our view that the
imperative is derived from the imperfect. It is formed by cutting off the
prefix from the imperfect stem. Then, a high vowel and a glottal stop are
inserted for syllabication. The high vowel is chosen according to this
special rule:
(93) Imperative V u / C.Cu
i / elsewhere
In other words, the vowel of the imperative prefix is always /i/ unless the
stem vowel is /u/, in which case /u/ is chosen to secure vowel harmony. An
example which has /u/ in the imperfect stem in both varieties is the verb
/daxal(-a) ya-dxul(-u) > yu-dxul/ to enter. Compare the imperative
forms of this verb in both varieties:
Referent
2msg
2fsg
2du (m & f)
2mpl
2fpl

SA
?udxul
?udxul-ii
?udxul-aa
?udxul-uu
?udxul-na

EA
?udxul
?udxul-i

?udxul-u

The imperative prefix is not used in certain verbs, particularly

those where, after the omission of the imperfect prefix, the stem begins
with a vowelled consonant (Malik 1976: 381). These verbs include: (a)
geminate verbs, (b) medial-weak verbs, (c) sound verbs of Form II and (d)
sound verbs of Form III. Consider these examples:
SA
jurr
udd
qul
nam
bi3
3allim
saafir

EA
gurr
idd
?uul
naam
bii3
3allim
saafir

Gloss
pull!
pull!
say!
sleep!
sell!
teach!
travel!

99

The retention of the long vowels in the imperative form of medial-weak

verbs in EA represents a contrast with their reduced quantity in SA. This
can be explained by proposing that the Closed-Syllable Shortening rule
applies in SA but not in EA because in the latter the final consonant does
not count for syllable weight as argued by the Extrametricality Condition
in EA.
A distinctive Colloquial imperative prefix [ma-] is employed
before the imperfect form, particularly in invitations and suggestions, e.g.
/ma-taakul/ Go ahead, eat, /ma-tirab/ Go ahead, drink and /ma-tiigi/
Go ahead, come.
2.3. Inflection for Voice:
Arabic verbs are inflected for two voices: active and passive. The
active voice has the usual form of the verb. The passive voice is expressed
in SA by a set of vowel patterns: [u-i / u-a]. The choice of vowel pattern
depends upon the aspectual status of the verb form: perfect or imperfect. If
the verb form is perfect, the first pattern [u-i] is employed. For example,
/fatah(-a)/ he opened becomes /futih(-a)/ it was opened and /arib(-a)/
he drank becomes /urib(-a)/ it was drunk. In the case of imperfect verb
forms, the second pattern [u-a] is used. For instance, /ya-ftah(-u)/ he
opens is changed to [yu-ftah(-u)/ it is opened and /ya-rab(-u)/ he
drinks is changed to /yu-rab(-u)/ it is drunk. In medial-weak verbs, the
passive perfect has the surface form [FiiL(-a)] from underlying
[FuGiL(-a)] by Glide Elision and Anticipatory Vocoid Assimilation, e.g.
/qaal(-a)/ he said is rendered /qiil(-a)/ it was said. The passive perfect
has the surface form [yu-FaaL(-u)] from underlying [yu-FGaL(-u)] by
Anticipatory Vocoid Assimilation, e.g. /ya-bii3(-u) / he sells is rendered
/yu-baa3(-u)/ it is sold.
Table (29) provides the passive equivalents for all the active forms
of triradical verbs in SA.

No
I
II
III
IV

Table (29)
The Passive Forms of Triradical Verbs in SA
Perfect
Imperfect
Active
Passive
Active
Passive
Fa3aL(-a)
Fu3iL(-a)
ya-F3i/a/uL(-u) yu-F3aL(-u)
Fa33aL(-a)
Fu33iL(-a)
yu-Fa33iL(-u) yu-Fa33aL(-u)
Faa3aL(-a)
Fuu3iL(-a)
yu-Faa3iL(-u)
yu-Faa3aL(-u)
?aF3aL(-a)
?uF3iL(-a)
yu-F3iL(-u)
yu-F3aL(-u)

100

V
VI
VII
VIII
IX
X

taFa33aL(-a)
taFaa3aL(-a)
(?i)nFa3aL(-a)
(?i)Fta3aL(-a)
(?i)F3aLL(-a)
(?i)staF3aL(-a)

tuFu33iL(-a)
tuFuu3iL(-a)

?uFtu3iL(-a)

?ustuF3iL(-a)

ya-taFa33aL(-u)
ya-taFaa3aL(-u)
ya-nFa3iL(-u)
ya-Fta3iL(-u)
ya-F3aLL(-u)
ya-staF3iL(-u)

yu-taFa33aL(-u)
yu-taFaa3aL(-u)

yu-Fta3aL(-u)

yu-staF3aL(-u)

Table (29) reveals that Forms VII and IX do not have passive forms. This
is due to the fact that they are unaccusative verbs, i.e. verbs whose sole
argument is the theme or patient.
In relation to the passive in the Colloquial varieties, Al-Toma
(1969: 67) states that the passive form which results from an internal
vowel change in the verb has disappeared from modern dialects. This, of
course, applies to EA. This variety has abandoned the use of the vowel
patterns stated above in forming the passive. Alternatively, it has
introduced the new morphemes [t-] and, less frequently [n-] which are
prefixed to the perfect form and infixed in the imperfect (Omar 1976: 12).
An initial vowel and a glottal stop are inserted only in the perfect forms by
the regular phonological rules of Word-Initial Epenthesis and Glottal Stop
Insertion.
Close examination of the facts indicates that the morphemes [t-]
and [n-] are not new since they are employed in Forms V, VI and VII (See
Table 4). Hence, it can be proposed that the passive is losing ground in EA
and being replaced by those forms. In other words, EA normally uses
reflexive forms to indicate the passive voice. These forms are [(?i)tFa3aL,
(?i)tFa33aL ~ (?i)tFa33iL, (?i)tFaa3iL and (?i)nFa3aL] in the perfect and
[yi-tFi3iL, yi-tFa33aL ~ yi-tFa33iL, yi-tFaa3iL and yi-nFi3iL] in the
imperfect. This shows that [t-] is more frequent than [n-], as the former is
used with three forms, while the latter is used with one form. Also, the
former is used to form the equivalent of three SA passive forms: I, II and
III, whereas the latter is used to form the counterpart of one SA passive
form: I. Compare the following:
SA
urib(-a)
kussir(-a)
quubil(-a)
Durib(-a)
yu-rab(-u)
yu-kassar(-u)

EA
(?i)tarab
(?i)tkassar
(?i)t?aabil
(?i)nDarab
yi-tirib
yi-tkassar

Meaning
it was drunk
it was broken
he was met
he was hit
it is drunk
it is broken

101

yu-qaabal(-u)
yu-Drab(-u)

yi-t?aabil
yi-nDirib

he is met
he is hit

In certain Colloquial verbs, the passive verb formant [t-]

assimilates to the following consonant if it is a dental, a palatal or a velar
(cf. Robertson 1970: 212), as in the following verbs:
SA
dummir(-a)
ujji3(-a)
suriq(-a)
kunis(-a)
jumi3(-a)

EA
(?i)d-dammar
(?i)-agga3
(?i)s-sara?
(?i)k-kanas
(?i)g-gama3

Meaning
it was destroyed
he was encouraged
it was stolen
it was cleaned
it was collected.

It should be stressed at the end of this subsection that the

Colloquial passive markers [t-] and [n-] are also used to form intransitive
or inchoative verbs, as will be shown in (2.5). In other words, not all verbs
having these markers in their surface structure can be considered passive
transforms of active verbs. They can simply be verbs of Form V, VI or
VII, which are called pseudo-passive by Wise (1975: 43). She further
offers a semantic factor to differentiate between the two classes:
Whether the verb is interpreted as the passive proper or
what we shall call the pseudo-passive depends on whether
the process requires an external agent or not, e.g.:
54. il3asaakir itlammu fi ilmidaan
the soldiers (were) gathered in the square
can be interpreted as either, whereas in
55. ilfirawla itlammit
the strawberries were gathered
only the passive is possible.
This ambiguity problem is not found in SA where Forms V, VI and VII
can only be pseudo-passive and the passive has a different form involving
the vowel pattern [u....i/a]. Thus, one can find an EA verb that has two
different SA counterparts, one passive and the other pseudo-passive:
SA Passive
1- nubbih(-a)
he was alerted
2- nuuqi(-a)
he/it was discussed
3- kusir(-a)

SA Pseudo-Passive
tanabbah(-a)
he became alert
tanaaqa(-a)
he discussed with
(?i)nkasar(-a)

EA
(?i)tnabbah
(?i)tnaa?i
(?i)nkasar

102

it was broken

it broke

2.4. Subjectival & Objectival Affixes:

Arabic verbs have two types of affixes: subjectival affixes and
objectival ones. The former type is obligatory with verbs while the latter is
non-obligatory as indicated by Ghaly (1960: 92-3). He speaks of the
subject affixes saying, No verb stem can ever occur in the language
without including one member of this obligatory group and only one.
About object affixes he states: object pronominal suffixes...occur as a
non-obligatory group. When a member of the objectival affixes co-occurs
with a verb, it immediately follows the subjectival affixes. Another
distinction between subjectival and objectival affixes is that whereas the
former are confined to verbs, the latter can also be used with nouns and
some closed-list classes.
2.4.1. Subjectival Affixes:
In Arabic, verbs always have a subject as argued by Abdel-Malek
(1972: 45). He further explains this as follows:
the subject of a given verb is the noun or pronoun with
which the verb shows inflectional concord. If not explicit,
the subject is an implied pronoun; thus in the [Colloquial]
sentence /katabu ddars/ They wrote the lesson, the subject
of the verb is the implied pronoun /humma/ they.
In other words, Arabic verbs show agreement with their subjects, whether
explicit or implicit. Therefore, they are inflected for three categories of the
subject: person, gender and number. The agreement markers used vary
according to the aspect of the verb (See 2.2). The perfect form of verbs
takes only suffixes, while the imperfect form receives either prefixes only
or non-sequential affixes, i.e. including both prefixes and suffixes. To
illustrate, /arib-tu > irib-t/ I drank has the suffix [-tu > -t] because it is
in the perfect form, whereas /?a-rab-u > ?a-rab-/ I drink takes the
prefix [?a-] and the suffix [-u > -] since it is in the imperfect form. Table
(30) displays the markers used in both varieties to signify agreement
between verb and subject in person, gender and number in the perfect
aspect.
Table (30)
Subjectival Affixes with
Perfect Verb Forms in SA & EA

103

Referent
1sg (m/f)
1pl (m/f)
2msg
2fsg
2du (m/f)
2mpl
2fpl
3msg
3fsg
3mdu
3fdu
3mpl
3fpl

SA
-tu
-naa
-ta
-ti
-tumaa
-tum
-tunna
-a
-at
-aa
-ataa
-uu
-na

EA
-t
-naa ( -na#)
-t
-tii ( -ti#)

-tuu ( -tu#)

-
-it

-uu ( -u#)

It can be seen from Table (30) that the Standard variety has almost
twice the subject markers used with perfect forms that are found in EA:
while SA has thirteen markers, EA has only seven. According to AbdelHafiz (1994: 30), the Colloquial variety has been innovated by applying a
phonetic process of erosion to the Standard set. He follows Heine and
Reh (1984: 21) in defining erosion as a process by which the
phonological substance of a morpheme is reduced. Then, he states that
this phonetic process has leveled the first and second person singular
endings. The distinction between the two varieties can be seen by
contrasting (94) and (95):
(94) SA: a. arib-tu
drink.pf-1sg
I drank the milk.
b. arib-ta
drink.pf-2msg
You (msg) drank the milk.

1-laban
the-milk
1-laban
the-milk

(95) EA: a./b. irib-t

il-laban
drink.pf-1sg/2msg
the-milk
I or you (msg) drank the milk.
It is obvious from the examples above that while SA makes a distinction
between first and second person singular subjects, EA does not. It employs
a single suffix [-t] for both persons. However, it is still possible to argue
that there are two [-t] affixes, one [-t]+1 and the other [-t]+2 which have,

104

through erosion, come to be pronounced identically. That is to say,

perhaps we are dealing with homophony rather than identity.
It might also be noted from Table (30) that EA has eliminated the
number distinction between dual and plural. It employs the plural markers
to refer to dual subjects. This can be observed by comparing the Standard
sentences in (96) with their Colloquial counterparts in (97):
1-laban
(96) SA: a. arib-tumaa
drink.pf-2du
the-milk
You (two) drank the milk.
b. arib-aa
1-laban
drink.pf-3mdu
the-milk
They (two) drank the milk.
(97) EA: a. irib-tu
l-laban
drink.pf-2pl
the-milk
You (two) drank the milk.
b. irb-u
1-laban
drink.pf-3pl
the-milk
They (two) drank the milk.
Furthermore, EA has eliminated the gender distinction between the
second and third person plural subjects. The masculine endings have
replaced the feminine ones in the second and third person plural. To
observe the difference in the case of the second person plural, compare the
EA example in (99) with its SA counterparts in (98):
(98) SA: a. fahim-tumu
d-dars(-a)
understand.pf-2mpl
the-lesson(-Acc)
You (mpl) understood the lesson.
b. fahim-tunna
d-dars(-a)
understand.pf-2fpl
the-lesson(-Acc)
You (fpl) understood the lesson.
(99) EA: a./b. fihim-tu
d-dars
understand.pf-2pl
the-lesson
You (m/f pl) understood the lesson.
Again to note the difference in the third person plural markers,
contrast the EA sentence in (101) with its SA equivalents in (100):
(100) SA: a. kasar-uu

z-zujaaj(-a)

105

break.pf-3mpl
the-glass(-Acc)
They (m) broke the glass.
b. kasar-na
z-zujaaj(-a)
break.pf-3fpl
the-glass(-Acc)
They (f) broke the glass.
(101) EA: a./b. kasar-u
1-?izaaz
break.pf-3pl
the-glass
They (m/f) broke the glass.
A morphosyntactic difference between the two varieties is in the
use of the third person subject markers. While the two varieties are similar
when the subject precedes the verb, they differ when the verb precedes the
subject. The situation in SA has been described by Mohammad (1990: 95):
If the subject precedes the verb, the verb shows full
agreement with the subject in the features of person,
number, and gender. If, on the other hand, the subject
follows the verb, the verb shows some kind of
impoverished agreement: it has the features of third
person singular.
Thus, in SA the third person singular markers are used instead of
the dual and plural markers if the verb precedes the subject. In other
words, the dual and plural markers are only used if the verb follows the
subject or when the subject is absent. This can be accounted for by
proposing that when the subject follows the verb two steps are taken: (a)
copy the gender of the subject in the verb, then (b) apply the default
person and number values [+3] and [+sg] to the verb.
In EA, on the other hand, the plural marker [-u] is used for all dual
and plural subjects, simply because it prefers the word order SVO.
Compare the Standard and the Colloquial verbs in (102-104):
1-walad-aani
1-faS1
(102) SA: daxal-a
enter.pf-3msg
the-boy-Nom.du
the-classroom
The two boys entered the classroom.
EA: ?il-walad-een
daxal-u
1-faSl
the-boy-du
enter.pf-3mpl
the-classroom
The two boys entered the classroom.
(103) SA: qaTa3-a
cut.pf-3msg

r-rijaal-u
the-men-Nom

-ajar-ah
the-tree

106

The men cut the tree.

?aTa3-u
EA: ?ir-riggaala
the-men
cut.pf-3mpl
The men cut the tree.

-agar-a
the-tree

(104) SA: xaraj-at

il-banaat
go out.pf-3fsg
the-girls
The girls went out.
xarag-u
EA: ?il-banaat
the-girls
go out.pf-3mpl
The girls went out.
The subject markers used with imperfect verb forms are distinct
from those used with perfect ones. As discussed above, while the latter
take the form of suffixes, the former are discontinuous morphemes. Table
(31) indicates that the Egyptian subject markers are built upon their
Standard counterparts with a wide range of simplification.

Referent
1sg (m/f)
1pl (m/f)
2msg
2fsg
2du (m/f)
2mpl
2fpl
3msg
3fsg
3mdu
3fdu
3mpl
3fpl

Table (31)
Subjectival Affixes with
Imperfect Verb Forms in SA & EA
SA
EA
?a--u
?a--
na--u
ni--
ta--u
ti--
ta--ii(na)
ti--ii ( -i#)
ta--aa(ni)

ta--uu(na)
ti--uu ( -u#)
ta--na

ya--u
yi--
ta--u
ti--
ya--aa(ni)

ta--aa(ni)

ya--uu(na)
yi--uu ( -u#)
ya-na

It could be observed from Table (31) that EA has almost

eliminated the subject suffixes in the imperfect. Only two endings are
maintained: the second person feminine singular ending [-ii] and the
second or third person plural ending [-uu] (cf. Abdel-Hafiz 1994: 37-38).
The latter ending replaces the dual and plural markers of SA.
Consequently, the number distinction between dual and plural, as well as

107

the gender distinctions within second person plural and third person plural
are non-existent in EA. Another observation is that dual and feminine
plural subject markers are not found in the Colloquial variety. It uses the
reflexes of SA masculine plural markers to cover both dual and feminine
plural subjects.
The remark about the morphosyntactic distinction between the two
varieties that is mentioned at the end of the discussion of the perfect
subject markers applies to the imperfect markers too. While SA utilizes the
third person singular markers for dual and plural subjects when verbs
precede their subjects, EA uses the masculine plural marker for all dual
and plural subjects.
(105) SA: ta-13ab-u
impf.3fsg-play-indic
The two girls are playing.
EA: ?il-bint-een
the-girl-du
The two girls are playing.

1-bint-aan
the-girl-Nom.du
bi-yi-13ab-u
prog-impf-play-3mpl

(106) SA: ya-ktub-u

1-?awlaad-u
impf-write-3msg
the-boys-Nom
The boys are writing the lesson.
EA: ?il-?awlaad
bi-yi-ktib-u
the-boys
prog-impf-write-3mpl
The boys are writing the lesson.

d-dars
the-lesson
d-dars
the-lesson

In an attempt to define the content of each subjectival affix, we

have to know which categories are permitted, what co-occurrence
restrictions are imposed on features and what feature representation can be
given to each affix. Let [ptp] mean participant in speech act. So, [ptp]
means that neither the first person nor the second person is referred to.
Hence [ptp] means +third person by implication. The categories
permitted in SA are given in Table (32) and those permitted in EA are
presented in Table (33):
Table (32)
The Permitted Person/Number/Gender Categories in SA
1
2
ptp du
sg
pl fem
1sg
+
+

1pl
+
+

2msg
+
+

108

2fsg
2du
2mpl
2fpl
3msg
3fsg
3mdu
3fdu
3mpl
3fpl

+
+
+
+

+
+

+
+

+
+

+
+

Table (33)
The Permitted Person/Number/Gender Categories in EA
1
2
ptp du
sg
pl fem
1sg
+
+

1pl
+
+

2msg
+
+

2fsg
+
+
+

2pl
+
+

3pl
+

3fsg
+
+

The SA feature co-occurrence restrictions are:

*[+1 +2] (no inclusive first person)
*[+1 fem] (no distinction for gender in lst person)
*[+1 +du] (no dual in the first person)
*[+2 +du fem] (no distinction for gender in the 2nd person dual)
*[ptp +1] and *[ptp +2] and *[+sg +pl] and *[sg pl du]
(logical necessity).
The major difference between SA and EA is that the latter has developed
new co-occurrence restrictions on features:
(i) [du] is absent in verbs, as is the combination [sg pl]; in the process of
agreement a default rule applies to change [sg] into [+pl].
(ii)Any value for [+fem] is prohibited from co-occurring with [+pl], hence
/na/ cannot be used.
(iii)Since there is no dual in verbs then [sg] suffices to distinguish all
numbers (singular vs. plural) in the verbal agreement categories.

109

Now, the underlying forms of subjectival affixes in the two

varieties can be given the following feature representation:
Perfect Conjugation
Both Conjugations
Conjugation
-naa
[+1 +pl]
-tu > -t
[+1]
-tum > -tuu [+2 sg]
-ti > -tii
[+2 +fem]
-ta > -t
[+2]
-at > -it
[ptp pl +fem]
-a > (lost)
[ptp pl]
(-na
[+pl +fem])
(-aa
[+du])
-uu > -u [+pl]

Imperfect
na- > ni?a-

ta- > ti-

[+1 +pl]
[+1]

[+2]

-ii [+fem] /[+2] +

ta- > ti- [+fem]
(-a
[subjunctive])
(-
[jussive])
-u > (lost) elsewhere
ya- > yi- elsewhere
In each conjugation, the left column shows the affix, and the right column
shows the minimal feature representation each affix must have, i.e. the
minimal amount of information necessary for each affix; the rest can be
inferred from the whole system of affixes. For instance, it is not necessary
to say that [-tu- > -t] is singular because the grammar will attach this affix
only in situations where a more specific first person affix (i.e. [-naa] which
is [+1 +pl]) is not added. Because [-naa] is plural, it is inferred that the rest
are singular. Affixes in parenthesis occur only in SA owing to loss of
categories in EA. The affix [-ii] is a feminine marker that attaches to a
form in which the feature [+2] has already been expressed. The first part of
the SA feminine plural affix [-tunna] is not presented in the perfect
conjugation because it can be related to [-tum] by nasal assimilation, i.e.
[-tum + -na] [-tunna].
The spell-out procedure runs as follows: for each feature
combination, the grammar continually attaches the most specific affixes
from the list above until all features are discharged (Noyer 1997: xxiii),
i.e. expressed in some affix. Once a feature is discharged, it can no longer
be used to condition the addition of an affix. Also, the grammar attaches
affixes until there are no more affixes left to add. In the imperfect

110

conjugation, the attachment is limited to one prefix and one suffix, i.e. the
imperfect stem must have at most and at least one suffix, and at most and
at least one prefix (not counting object agreement clitics or the subject
agreement extensions /na/ and /ni/). For example, 3fdu = [ptp +du sg
pl +fem]. If we add [-at] it will discharge [ptp pl +fem]). Then if we
add [-aa] it will discharge [+du].
The order of affix attachment is the order shown in the feature
representation above. This ordering is predictable on the following
grounds:
(i) More specific affixes attach first. For instance, [-naa] which is [+1 +pl]
attaches before [-tu] which is [+1]. Hence, in the [1pl] only [-naa] may be
attached; by the time [-tu] can be attached, the feature [+1] has already
been discharged from a [+1 +pl sg] input.
(ii)
If two affixes do not have overlapping feature sets, the affix which
contains the highest feature on the feature hierarchy is attached first. The
feature hierarchy, according to Noyer (1997: xxii) is as in (107):
(107) person features > number features > gender features > other features
For example, [-tum] which is [+2 sg] is attached before [-ti] which is [+2
+fem]. They both tie on person features, since both have [+2], but [-tum]
has a number feature [sg] which places it above [-ti] which has only a
gender feature [+fem].
2.4.2. Objectival Suffixes:
Objectival suffixes, the second type of verbal affixes, are
summarized in Table (34). They comprise suffixes indicating objects of
the verbs to which they are attached. In this respect, they may be
considered a reduction of the Standard free object pronouns that will be
mentioned in (5.1). They may also signify the possession or relation of the
noun or particle to which they are suffixed. For illustration, [-kum] is a
direct object in /xada3-a-kum/ he deceived you (mpl), it shows
possession in /bayt-u-kum/ your (mpl) house and direction in /?ilay-kum/
to you (mpl). As for the grammatical function of these affixes, which are
also called bound personal pronouns, Abdel-Malek (1972: 141) points out
that when suffixed to an active participle or a verb, the bound forms of
personal pronouns are direct objects; when suffixed to a preposition, they
are objects of the preposition.

111

Referent
1sg (m/f)
1pl (m/f)
2msg
2fsg
2du (m/f)
2mpl
2fpl
3msg
3fsg
3du(m/f)
3mpl
3fpl

Table (34)
Objectival Suffixes in SA & EA
SA
EA
-ii,-nii
-ii,-nii/-yaa ( -i, -ni/-ya#)
-naa
-naa ( -na#)
-ka
-ak/-k
-ki
-ik/-kii ( -ki#)
-kumaa

-kum
-ku(m)
-kunna

u
-h
-uu/-h ( -u#)
i
-haa
-haa ( -ha#)

u
-h maa
i
u
-h m
-hum
i
u
-h nna
i

Table (34) indicates that the dual and feminine plural objectival
affixes have been eliminated in EA. This is a consequence of the more
general loss of the dual and feminine plural categories, as manifested in
the subjectival affixes (discussed above) and the free pronouns (discussed
in 5.1). Nevertheless, it must be noted that since the Colloquial variety
does not include dual objectival affixes, it expands the plural affixes to
refer to dual objects as well. Similarly, as it does not include any feminine
plural objectival affixes, it extends the masculine plural affixes to cover
feminine plural objects as well. In other words, masculine plural objectival
affixes are used for both dual and plural objects in EA.
It is also evident from Table (34) that SA has two forms of
objectival affixes in the third person masculine singular and the third
person plural to secure harmony with the preceding vowel. Thackston
(1984: 27) explains this as follows: The 3rd person enclitics, with the
exception of the 3rd fem. sing., harmonize with the vowel that precedes
immediately. When the immediately preceding vowel is u or a, the vowel
of the enclitic is u, when immediately preceded by i or ay, the vowel of the
enclitic is i. Since /i/ = /y/ in phonological content, being different only

112

with regard to syllabic position, this is part of a larger generalization - the

harmonic V is /i/ after /i/ and /u/ elsewhere. The Colloquial variety, on the
other hand, does not make that distinction. It has reduced the two forms
into one.
An interesting observation might be added here. For some referents
two sets of objectival suffixes are found in EA. This depends on whether
the words, to which they are added, end in a consonant or a vowel. The
first set (i.e. -ii, -ak, -ik and -uu) is used after consonants, while the second
(i.e. -yaa, -k, -kii and -h) is used after vowels. The choice here seems to
obey the principle that a C-initial allomorph is used after vowels and a
V-initial allomorph is used after consonants. Compare the following EA
examples:
After [+cons]
beet-i my house
Darabt-ak I hit you (msg)
Darabt-ik I hit you (fsg)
Darabt-u I hit him

After [cons]
ma3aa-ya with me
ramaa-k he threw you (msg)
ramaa-ki he threw you (fsg)
ramaa-h he threw him.

It might also be added that in both varieties the infix [-n-] is attached to the
objectival affix of the first person singular when that affix refers to a direct
object, e.g. /Darab-a-n-ii > Darab-n-i/ he hit me and /da3aa-nii >
da3aa-ni/ he invited me.
The attachment of objectival affixes to verbs causes one change in
SA, namely the lengthening of the vowel in the 2fsg [-ti]. This can be
accounted for by a special rule in SA with a form like that in (108):
(108) Subjectival [-ti] [-tii]/ C
But the attachment of objectival affixes causes some changes in the verb in
EA (cf. Malik 1976: 96). For example, the suffixation of [-ni] leads to the
use of long vowels in the preceding subjectival affixes in EA because the
rule of Final-Vowel Shortening does not apply to these vowels, being no
longer final. Compare these EA verbs before and after suffixation of the
objectival affixes:
Before Obj. Suff.
Darab-ti
Darab-tu
Darab-u

After Obj. Suff.

Darab-tii-ni
Darab-tuu-ni
Darab-uu-ni

Meaning
you (fsg) hit me
you (pl) hit me
they hit me

113

In addition, EA tends to obviate a sequence of three consonants by

inserting an extra vowel before the objectival affixes. With the objectival
suffix [-na] the extra vowel is /i/ by the regular rule of MedialEpenthesis. With the other objectival suffixes, the extra vowel is a ghost
vowel of the same nature of the vowel of the suffix to secure vowel
harmony (cf. Malik 1976: 110). Compare the following EA verbs before
and after objectival suffixation:
Before Obj. Suff.
sa?al-t
3azam-t
simi3-t
3irif-t
mana3-t

After Obj. Suff.

sa?al-ti-na
3azam-ti-ni
simi3-tu-ku
3irif-ta-ha
mana3-tu-hum

Meaning
you (msg) asked us
you (msg) invited me
I heard you (pl)
I knew her
I prevented them

2.5. Verb Derivation & Transitivity:

In both varieties under study, verbs can be derived from nouns,
adjectives or other verbs (cf. Gary & Gamal-Eldin 1982: 127). While the
first two derivational processes are limited in occurrence, the last one is
quite common and productive and will be the focus of discussion in most
of this section. Examples of the derivation of verbs from nouns are:
Noun
libnaan Lebanon
miSr Egypt
3askar army

SA Verb
labnan(-a)
maSSar(-a)
3askar(-a)

EA Verb
labnan
maSSar
3askar

Gloss
to make Lebanese
to Egyptianize
to camp

Examples of verbs derived from adjectives are:

Adjective
?ahmar(-u)
kabiir(-un)

SA Verb
hammar(-a)
?ihmarr(-a)
kabur(-a)
kabbar(-a)

EA Verb
hammar
?ihmarr
kibir
kabbar

Gloss
to make red
to become red
to become big
to make big

Before discussing the derivation of verbs from other verbs, a word

must be said about the transitivity of verbs in Arabic. Both varieties under
investigation distinguish between intransitive and transitive verbs, as in
(109) and (110), respectively:
(109) SA: naam-a
sleep.pf-3msg

1-walad
the-boy

114

The boy slept.

EA: ?il-walad
naam-
the-boy
sleep.pf-3msg
The boy slept.
il-?umm-u
(110) SA: ?a-naam-at
trans-sleep.pf-3fsg the-mother-Nom
The mother caused the boy to sleep.
EA: ?il-?umm
nayyam-it
the-mother
sleep.trans.pf-3fsg
The mother caused the boy to sleep.

1-walad
the-boy
i1-walad
the-boy

Moreover, there is a distinction between monotransitive and

ditransitive verbs in both varieties. In Gadalla (1992: 5), I explained what
Arab grammarians meant by this distinction: By definition, a
monotransitive verb is one that requires only one (direct) object. A
ditransitive verb is one that requires two objects (a direct object and an
indirect object). For instance, the verbs in (111) are monotransitive,
whereas those in (112) are ditransitive:
(111) SA: Darab-a
?ahmad-u
hit-pf-3msg
Ahmad-Nom
Ahmad hit the boy.
EA: ?ahmad
Darab-
Ahmad
hit.pf-3msg
Ahmad hit the boy.
(112) SA: ?a3Taa
?ahmad-u
give.pf-3msg
Ahmad-Nom
Ahmad gave the girl a rose.
EA: ?ahmad
?idda
Ahmad
give.pf.3msg
Ahmad gave the girl a rose.

1-walad
the-boy
il-walad
the-boy
l-bint-a
the-girl-Acc

ward-ah
(a) rose

l-bint
the-girl

ward-a
(a) rose

In both varieties, transitive verbs comprise two classes: causatives

and non-causatives. The former indicates that the subject causes a change
of state in the object, while the latter class does not (cf. Gadalla 1992: 6).
To illustrate, the verbs in (113) are causative, while those in (114) are noncausative:
(113) SA: qatal-a
l-qiTT-u
kill.pf-3msg
the-cat-Nom
The cat killed the mouse.
EA: ?il-?uTT
?atal

l-fa?r
the-mouse
il-faar

115

the-cat
kill.pf.3msg
The cat killed the mouse.

the-mouse

hind-u
(114) SA: qara?-at
read.pf-3fsg
Hind-Nom
Hind read the story.
EA: hind
?ar-it
Hind
read.pf-3fsg
Hind read the story.

l-qiSS-ah
the-story
il-qiSS-a
the-story

Morphologically, transitive verbs are related as causatives to

corresponding intransitive verbs and derived from them by a process of
affixation. Three affixes are used to form causatives from intransitives in
SA: gemination of the second consonant, the prefix [?a-] and the prefix
[sta-]; with the second as the most productive and the third the least
productive (Gadalla 1992: 62-66). In other words, three forms can be
identified as causatives in SA: Form II [Fa33aL(-a)], Form IV
[?a-F3aL(-a)] and Form X [sta-F3aL(-a)]. They are all derived from Form
I [Fa3aL(-a) ~ Fa3iL(-a) ~ Fa3uL(-a)]. To exemplify, the Standard
causative verbs in (115-117b) are derived from their intransitive
counterparts in (115-117a), respectively:
(115) a. saxin-a
become hot.pf-3msg
The water became hot.
b. saxxan-at
heat.caus.pf-3fsg
Hind heated the water.
(116) a. ariq-a
drown.pf-3msg
The boy drowned.
b. ?a-raq-a
caus-drown.pf-3msg
Ahmad drowned the boy.

l-maa?
the-water
hind-u
Hind-Nom

l-maa?
the-water

l-walad
the-boy
?ahmad-u
l-walad
Ahmad-Nom the-boy

(117) a. waqaf-a
r-rajul
stop.pf-3msg
the-man
The man stopped.
b. ?ista-wqaf-a
3umar-u
caus-stop.pf-3msg
Omar-Nom
Omar caused the man to stop.

r-rajul
the-man

116

EA, on the other hand, utilizes one causative marker to replace all
the Standard causative markers, namely the gemination of the second
radical. In other words, it substitutes Form II [Fa33a/iL] for the other two
causative Standard Forms IV [?a-F3aL(-a)] and X [sta-F3aL(-a)]. Hence,
the Standard causatives in (115-117b) have the Colloquial equivalents in
(118), respectively:
(118) a. hind
saxxan-it
il-mayya
Hind
heat.caus.pf-3fsg
the-water
Hind heated the water.
il-walad
b. ?ahmadgarra?
Ahmad caus.drown.pf.3msg
the-boy
Ahmad drowned the boy.
c. 3umar
wa??af
ir-raagil
Omar
caus.stop.pf.3msg
the-man
Omar caused the man to stop.
In addition to being derived from intransitive verbs, causative verbs
can also be derived, in both varieties, from transitive verbs which are noncausative (Gadalla 1992: 137), as in (119b) for SA and (120b) for EA:
(119) a. labis-a
l-walad-u
qamiiS-aa
put on.pf-3msg
the-boy-Nom
a shirt-Acc
The boy put on a shirt.
b. ?a-lbas-at
il-?umm-u
caus-put on.pf-3fsg the-mother-Nom
l-walad-a
qamiiS-aa
the-boy-Acc
a shirt-Acc
The mother caused the boy to put on a shirt.
(120) a. ?il-walad
libis
the-boy
put on pf.3msg
The boy put on a shirt.

?amiiS
a shirt

b. ?il-?umm
labbis-it
il-walad
the-mother
caus.put on.pf-3fsg the-boy
The mother caused the boy to put on a shirt.

?amiiS
a shirt

The intransitive counterparts of causative verbs are called

inchoative verbs. To clarify the relation between causative and
inchoative verbs, Mahmoud (1989: 7) indicates that inchoatives and
causatives form a transitive/intransitive dichotomy, where inchoatives

117

represent the intransitive counterparts and causatives the transitive. Three

affixes are used to form inchoatives in SA: the two prefixes [n-] and [ta-]
and the infix [-t-] inserted after the first radical, triggering the addition of
initial (?i) by Word-Initial Epenthesis and Glottal Stop Insertion. Thus, the
verbs in (121-123a) are causative, while those in (121-123b) are their
inchoative analogues:
(121) a. aqq-a
r-rajul-u
crack.pf-3msg
the-man-Nom
The man cracked the wood.
b. ?in-aqq-a
l-xaab
inch-crack.pf-3msg
the-wood
The wood cracked.

l-xaab
the-wood

(122) a. jammad-a
r-rajul-u
freeze.pf-3msg
the-man-Nom
The man froze the water.
b. ta-jammad-a
l-maa?
inch-freeze.pf-3msg
the-water
The water froze.

l-maa?
the-water

(123) a. mala?-at
fill.pf-3fsg
Hind filled the cup.
b. ?i-m-t-ala?-a
fill-inch.pf-3msg
The cup got filled.

hind-u
Hind-Nom

l-ka?s
the-cup

l-ka?s
the-cup

While the prefix [n-] is kept unchanged in EA, the prefix [ta-] and
the infix [-t-] are changed into [t-]. Compare the Egyptian structures in
(124-126) with their Standard counterparts in (121-123), respectively:
(124) a. ?ir-raagil
a??
the-man
crack.pf.3msg
The man cracked the wood.
b. ?il-xaab
in-a??
the-wood
inch-crack.pf.3msg
The wood cracked.
(125) a. ?ir-raagil
gammad
the-man
freeze.pf.3msg
The man froze the water.

il-xaab
the-wood

il-mayya
the-water

118

b. ?il-mayya
t-gammad-it
the-water
inch-freeze.pf-3fsg
The water froze.
(126) a. hind
mal-it
Hind
fill.pf-3fsg
Hind filled the cup.
b. ?il-kaas
it-mala
the-cupinch-fill.pf.3msg
The cup got filled.

il-kaas
the-cup

119

Chapter Three
Nominal Morphology
3.0. Introduction:
The morphology of nouns in SA and EA will be dealt with in this
chapter. Nouns will be divided into two classes: primary nouns which are
derived directly from the root and deverbal nouns which are derived from
verbs. The stem forms of primary nouns will be expounded in (3.1). Then,
the patterns of deverbal nouns will be discussed in (3.2). The difference
between definite and indefinite nouns will be indicated in (3.3). After that,
the inflection of nouns for case, gender and number will be shown in (3.4,
3.5 and 3.6), respectively. Moreover, the suffixation to nouns and the
formation of the diminutive form will be explained in (3.7) and (3.8),
respectively.
3.1. Primary Nouns:
Three morpheme-classes can be distinguished for the stem forms
of primary nouns in Arabic: roots, patterns and affixes. The primary nouns
of SA and EA can be grouped into two main classes according to the
number of consonants in their roots: triradical nouns and quadriradical
ones. The various types of triradical consonantal roots may be represented
as follows:
1. [F-3-L]: This is a sound root which consists of three different
consonants, e.g. /jamal(-un)/ a camel.
2. [F-33]: This is a doubled root which displays identical consonants in
the second and third positions, e.g. /hubb(-un)/ love.
3. [w/y-3-L]: This is an initial-weak root, i.e. having a glide (= weak
consonant) in the initial position, e.g. /walad(-un)/ a boy.
4. [F-w/y-L]: This is a medial-weak root, i.e. having a glide (= weak
consonant) in the medial position, e.g. /mawt(-un)/ death.
5. [F-3-w/y]: This is a final-weak root, i.e. having a glide (= weak
consonant) in the final position, e.g. /ra?y(-un)/ an opinion.
Quadriradical nouns have one basic root, i.e. [F-3-L1-L2] where L1 and L2
are two different consonants, as in /?arnab(-un)/ a rabbit.
The main vocalic patterns that occur in nominal stem forms are:

120

1. [-V--], as in /nahr(-un)/ a river.

2. [-V-V-], as in /haram(-un)/ a pyramid.
3. [-V-VV-], as in /kitaab(-un)/ a book.
4. [-VV-V-], as in /waalid(-un)/ a father.
Derivational affixes may be vocalic morphemes like the
lengthening of short vowels, consonantal morphemes like the gemination
of single consonants, or consonantal-vocalic morphemes, occurring as
prefixes, suffixes or infixes (cf. Ghaly 1960: 15):
1. [CV-], as in the prefix [ma-] in /maktab(-un)/ an office.
2. [-VVC], as in the suffix [-aan] in /gufraan(-un)/ forgiveness.
3. [-C-], as in the infix [-h-] in /bahhaar(-un)/ a sailor.
3.1.1. Triradical Nominal Stem Forms:
The majority of nominal stem forms are triradical; those which
have more than three consonants in their roots constitute a small portion of
Arabic nouns. Most of the triradical nominal stem forms of SA are
preserved in EA, i.e. they are identical in both varieties, as shown in Table
(35). Some other stem forms have undergone some regular phonological
changes in the Colloquial variety, as indicated in Table (36). Some SA
triradical nominal stem forms have two EA counterparts, one similar to
that of SA and one that has undergone a phonological change, as shown in
Table (37). Nominal stem forms derived from verbs will be dealt with in
(3.2).

No
1
2
3
4
5
6
7
8
9
10
11
12

Table (35)
Triradical Nominal Stem Forms Identical in SA & EA
Stem Form
Example
Gloss
Fa3L(-un)
nahr(-un)
a river
Fa3L-at(-un)
dast-at(-un)
a dozen
Fi3L(-un)
3ilm(-un)
a science
Fi3L-at(-un)
fikr-at(-un)
an idea
Fu3L(-un)
furn(-un)
an oven
Fu3L-at(-un)
furS-at(-un)
a chance
Fa3aL(-un)
haram(-un)
a pyramid
Fa3aL-at(-un)
barak-at(-un)
blessing
Fa3iL(-un)
malik(-un)
a king
Fa3iL-at(-un)
malik-at(-un)
a queen
Fu3uL(-un)
3unuq(-un)
a neck
Fa3aaL(-un)
salaam(-un)
peace

121

13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35

Fa3aaL-at(-un)
Fa3uuL(-un)
Fa3uuL-at(-un)
Fu3aaL(-un)
Fu3aaL-at(-un)
Fu3uuL(-un)
Fu3uuL-at(-un)
Fi3aaL-at(-un)
Faa3aL(-un)
Faa3iL(-un)
Fa33aaL(-un)
Fa33aaL-at(-un)
Fi33aaL-at(-un)
Fu33aaL-at(-un)
FaaL(-un)
FaaL-at(-un)
FiiL(-un)
FiiL-at(-un)
FuuL(-un)
FuuL-at(-un)
Fa3at(-un)
Fu3at(-un)
?uF3uuL(-un)

salaam-at(-un)
rasuul(-un)
3aruus-at(-un)
guraab(-un)
fukaah-at(-un)
muruur(-un)
hukuum-at(-un)
risaal-at(-un)
3aalam(-un)
Taabi3(-un)
bahhaar(-un)
dabbaas-at(-un)
Sinnaar-at(-un)
kurraas-at(-un)
baab(-un)
saa3-at(-un)
fiil(-un)
ziin-at(-un)
nuur(-un)
Suur-at(-un)
sanat(-un)
lugat(-un)
?usbuu3(-un)

safety
a messenger
a bride
a crow
humor
traffic
a government
a letter
world
a stamp
a sailor
a stapler
a hook
a notebook
a door
a watch
an elephant
decoration
a light
a picture
a year
a language
a week

The case marker (-un) is lacking in EA. Also, the final /t/ of SA is omitted
in EA by a regular morphological rule that can be called [-a ~ -t]
Suppletion in EA (For details, see 3.4):
(127) [-a ~ -t] Suppletion in EA:
-at -a / at the end of a phrase
-at -t / elsewhere
Table (35) illustrates that, except for final /t/ which is typically Standard,
the following templates are used for triradical nominal stem forms in both
varieties: [CVCC(a)], [CVCVC(a)], [CVCVVC(a)], [CVVCVC(a)],
[CVCCVVC(a)], [CVVC(a)], [CVCa] and [?uCCVVC]. An exceptional
noun in EA is /kubbaaya/ which has the SA equivalent /kuub(-un)/, though
the pattern [FuuL(-un)] is still preserved in the Colloquial variety.
Table (36)
SA Triradical Nominal Stem Forms
With Regular Phonological Changes in EA

122

No
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

Stem Form
SA
EA
FayL(-un)
FeeL
FayL-at(-un) FeeL-a
Fay3aaL(-un) Fi3aaL
FawL(-un)
FeeL
FawL-at(-un) FooL-a
Fa?L(-un)
FaaL
Fi?L(-un)
FiiL
Faa?(-un)
Fayya
Faa3uuL(-un) Fa3uuL
Fii3aaL(-un) Fi3aaL
Fuu3aaL(-un) Fu3aaL
Faa3iL-at(-un) Fa3L-a
Fu3Liyy(-un) Fu3Li
Fa3iyy(-un)
Fa3i
Fu3Laa
Fu3La
Fi3aa?(-un)
Fi3a
Fii3aa?(-un) Fii3a

SA
Sayf(-un)
layl-at(-un)
maydaan(-un)
awq(-un)
zawj-at(-un)
ra?s(-un)
bi?r(-un)
maa?(-un)
Saaruux(-un)
miizaan(-un)
duulaab(-un)
faakih-at(-un)
kursiyy(-un)
nabiyy(-un)
dunyaa
itaa?(-un)
miinaa?(-un)

Example
EA
Seef
leel-a
midaan
zoo?
zoog-a
raas
biir
mayya
Saruux
mizaan
dulaab
fakh-a
kursi
nabi
dunya
ita
miina

Gloss
summer
a night
a square
taste
a wife
a head
a well
water
a rocket
a scale
a cupboard
fruit
a chair
a prophet
world
winter
a harbor

Table (36) shows that SA triradical noun stem forms which have
undergone regular phonological changes in EA have one of the following
template groups: [CawC(a)], [CayC(a)], [CayCaC], [CayCaaC],
[CuCayC], [CV?C], [Caa?], [CVVCVVC], [CaaCiCa], [CVCCiy],
[CaCiy], [CuCCaa], [CVCaa?], [CVVCVV?]. The morphological rule [-a
~ -t] Suppletion in EA is responsible for the disappearance of SA /t/ at the
end of EA stem forms (2, 5 and 12). The phonological rules that govern
the changes shown in Table (36) are listed below:
(128) (i) Monopthongization applies to the stem forms (1-5)
(ii) Compensatory Lengthening is exemplified in (6-7)
(iii) Final /?/ Deletion is illustrated in stem forms (8, 16-17)
(iv) Atonic Shortening is apparent in stem forms (9-11)
(v) Final-Vowel Shortening occurs in stem forms (8, 15-17). It is
also represented in stem forms (13-14) since linguists consider /iy/
equal to /ii/ in phonological content. An evidence of this is the
change of /iy/ to /ii/, before a consonantal suffix in SA, e.g.
/nabiyy(-un)/ a prophet /nabii-na/ our prophet.
Some SA nominal stem forms have undergone two phonological
changes in EA. Those include stem forms (3, 16-17). In the stem form (3),
one of the two phonological changes does not appear on the surface. The

123

stem form [Fay3aaL(-un)] undergoes Monophthongization, producing the

form *[Fee3aaL] which then undergoes Atonic Shortening, resulting in
[Fi3aaL], as /i/ is the short counterpart of /ee/.
The two stem forms in (8) and (12) deserve special attention
because they undergo more than two phonological changes. First, the stem
form [Faa?(-un)], represented in the word /maa?(-un)/ water, undergoes
the deletion of final /?/, then the shortening of a final long vowel /aa/. This
leads to a form */ma/ which is impossible by the minimal stem
requirement (McCarthy & Prince 1990a: 17). The moraic representation of
this form is:
(129)

Because the form */ma/ contains only one mora, it is considered bad by
the minimal stem constraint for it is subminimal. So, the dialect resorts to
the addition of another mora to compensate for the lost mora in order to
satisfy the bimoraic minimum. It is quite reasonable to have a mora of the
same phonological nature of the deleted mora. The nearest phonological
equivalent to /a?/ is /aa/, as in the rule of Compensatory Lengthening. The
resulting form would be */maaa/ which exceeds the maximum length of a
vowel. One needs an onset for the final /a/. Adding the glide /y/ as an
onset results in the form /maaya/. But because /aya/ equals /aa/ as appears
in the regular change /baya3(-a) baa3(-a)/ to sell, the result would be
also bad. So, the dialect resorts to the replacement of the long vowel
before the glide by a short vowel and a glide, i.e. /aa ay/ ending with
/mayya/. Hence, the phonological processes involved in the change from
SA /maa?(-un)/ to EA /mayya/ are:
(130) a. Final /?/ Deletion,
b. Final-Vowel Shortening,
c. Compensation of lost mora,
d. Addition of an onset for /a/ which is /y/, and

124

e. Replacement of a long vowel /aa/ by a vowel and a glide /ay/.

These changes can be illustrated in (131a) and the moraic representation of
the final form is given in (131b).
(131). a. maa?(-un) *maa *ma *maaa *maaya mayya

b.

The other stem form that deserves special attention is

[Faa3iL-at(-un)]. This stem form undergoes the deletion of final /t/, then
the elision of /i/ in a medial open syllable, then the shortening of /aa/ in a
medial closed syllable because a /CVVC/ syllable occurs only wordfinally as stated in (1.1.3). So, that form becomes [Fa3L-a] in EA after one
morphological change and two phonological changes stated in (132).
Further examples of this phenomenon are listed below:
(132) a. Faa3iL-at Faa3iL-a (by [-a ~ -t] Suppletion in EA),
b. Faa3iL-a Faa3L-a (by High-Vowel Deletion in EA), and
c. Faa3L-a Fa3L-a (by Closed-Syllable Shortening in EA).
SA
jaami3-at(-un)
naaZir-at(-un)
aaniy-at(-un)

EA
gam3-a
naZr-a
sany-a

Gloss
a university
a headmistress
a second

However, in very few nouns, the Standard stem form [Faa3iL-at(-un)] is

kept the same in the Colloquial variety, except for dropping the final /t/, as
in /?aanis-at(-un) > ?aanis-a/ Miss used in addressing an unmarried lady,
and /Taalib-at(-un) > Taalib-a/ a student (f). These forms are exceptional

125

to the elision of /i/. The shortening of /aa/ then will not be necessary
because it is not in a medial closed syllable.

1
2
3
4
5
6
7
8
9

Table (37)
SA Triradical Nominal Stem Forms With Two EA Counterparts
Stem Form
Example
SA
EA
SA
EA
Gloss
Fi3aaL(-un)
Fi3aaL
hisaab(-un)
hisaab
an account
"
Fu3aal
hiSaan(-un)
huSaan
a horse
Fa3iiL(-un)
Fa3iiL
hadiid(-un)
hadiid
iron
"
Fi3iiL
ragiif(-un)
rigiif
a loaf
Fa3iiL-at(-un) Fa3iiL-a gariid-at(-un) gariid-a
a newspaper
"
Fi3iiL-a daqiiq-at(-un) di?ii?-a
a minute
Fa3iyy-at(-un) Fa3iyy-a Sabiyy-at(-un) Sabiyy-a a young girl
"
Fi3iyy-a hadiyy-at(-un) hidiyy-a
a present
Fu33aL(-un) Fu33aL sukkar(-un)
sukkar
sugar
"
Fi33iL
sullam(-un)
sillim
a ladder
Fu33aaL(-un) Fu33aaL duxxaan(-un) duxxaan smoke
"
Fi33aaL ubbaak(-un) ibbaak
a window
Fi33(-un)
Fi33
sinn(-un)
sinn
age
"
Fu33
qiTT(-un)
?uTT
a cat (m)
Fi33-at(-un)
Fi33-a
Sihh-at(-un) Sihh-a
health
"
Fu33-a
qiTT-at(-un) ?uTT-a
a cat (f)
Fa3aa?(-un)
Fa3a
samaa?(-un) sama
sky
"
Fi3a
masaa?(-un) misa
evening

An exceptional word here is the word for a train which has the SA form
[Fi3aaL(-un)], i.e. /qiTaar(-un)/ but the EA form [Fa3L], i.e. /?aTr/. Table
(37) indicates that SA triradical noun stem forms which have two EA
equivalents have one of the following templates (again excluding final
[-t]): [CVCVVC(a)], [CVCVyya], [CVCCVC], [CVCCVVC], [CVCC(a)]
and [CVVC]. The deletion of final /?/ at the end of (9) is also regular.
The existence of the two EA patterns for some SA nominal stem
forms may be explained by the process of lexical diffusion which was
defined by Trask (1996: 203) as:
The process by which a phonological change begins by
applying only to certain words and then spreads gradually
to other phonologically similar words. In some cases,
lexical diffusion stops at some point, leaving all remaining
words permanently unaffected; in other cases, the process

126

eventually goes to completion by affecting all remaining

words.
For example, one might propose that the raising of SA /a/ to /i/ began in
EA by applying to some words not beginning with a guttural. The guttural
consonants according to McCarthy (1991: 1) include the velar fricatives
/x/ and /g/, the pharyngeals /h/ and /3/ and the laryngeals /?/ and /h/. This
can be captured by the following proposed generalization:

(133) Vowel Raising in EA:

+ cons
ai/

guttural
The raising of /a/ to /i/ after non-guttural consonants can be seen in the
following words of the SA pattern [Fa3iiL(-un)]:
SA
ragiif(-un)
jamiil(-un)
ariiT(-un)
sariir(-un)
zamiil(-un)

EA
rigiif
gimiil
iriiT
siriir
zimiil

Gloss
a loaf
a favor
a tape
a bed
a colleague

The assumption that /a/ is not raised after gutturals can be supported by the
following examples:
SA
xabiir(-un)
gasiil(-un)
habiib(-un)
3aSiir(-un)
?amiir(-un)

EA
xabiir
gasiil
habiib
3aSiir
?amiir

Gloss
an expert
laundry
a lover
juice
a prince

There are two assumptions for the continuation of the raising process in
EA. It may have stopped or it may still be going on and may eventually
end up with applying to the gutturals as well. This latter assumption can be
supported by the existence of one word beginning with a guttural:
/hadiyy-at(-un) > hidiyy-a/ a present.

127

However, it should be stressed that at the present stage of the

language not all EA nouns of the [Fa3iiL] pattern start with a guttural.
This can be evidenced by the following examples:
SA
bariid(-un)
mariiD(-un)
waziir(-un)
daliil(-un)
safiir(-un)

EA
bariid
mariiD
waziir
daliil
safiir

Gloss
mail
a patient
a minister
an evidence, a guide
an ambassador

If the raising process is still going on, one might expect that it may apply
to words like those above. If it has stopped, those words will remain
unaffected.
3.1.2. Quadriradical Nominal Stem Forms:
All SA quadriradical nominal stem forms are retained in EA, as in
Table (38). (One can notice that L1 and L2 are two different consonants).
Table (38)
Quadriradical Nominal Stem Forms Identical in SA &EA
No Stem Form
Example
Gloss
1 Fa3L1aL2(-un)
kawkab(-un)
a planet
2 Fu3L1uL2(-un)
bulbul(-un)
a nightingale
3 Fi3L1aL2(-un)
dirham(-un)
a unit of currency
4 Fa3L1aL2-at(-un)
falsaf-at(-un)
philosophy
5 Fa3L1aL2-i(yy-un) 3askar-i(yy-un)
a soldier
6 Fu3L1aaL2(-un)
fustaan(-un)
a dress
7 Fi3L1aaL2(-un)
simsaar(-un)
a broker
8 Fa3L1iiL2(-un)
kabriit(-un)
matches
9 Fi3L1iiL2(-un)
?injiil(-un)
bible
10 Fu3L1uuL2(-un)
3uSfuur(-un)
a sparrow
Table (38) indicates that quadriradical nominal stem forms in both
varieties have the following templates: [CVCCVC], [CVCCVCa],
[CVCCVCiy] and [CVCCVVC].
3.1.3. A Prosodic Analysis of Nominal Stem Forms:
In order to characterize the nominal stem forms in terms of the
primitive units of prosody as proposed by McCarthy & Prince (1990a),

128

affixes are not included. Also, because the overall shape or canonical
pattern is the major concern, differences in vowel quality are disregarded.
So, the basic nominal stem forms dealt with in (3.1.1) and (3.1.2) can be
listed in (134a) and (134b), respectively:
(134) a. Triradical: CVCC, CVCVC, CVCVVC, CVVCVC and
CVVCVVC (> CVCVVC)
b. Quadriradical: CVCCVC and CVCCVVC
Nouns of the form [CVVC] are arguably derived from underlying
[CVGVC] by Glide Elision, those of the form [CVCat] are underlyingly
[CVCVG] with their final glide changed to /t/, and those of the forms
[CVCCVC] and [CVCCVVC] have a derivational infix of gemination.
The noun forms in (134) can be reclassified in prosodic terms according to
the number of moras they contain assuming that the final consonant is
extrasyllabic as follows:
Three Moras
Four Moras
CVCVVC
CVVCVVC
CVVCVC
CVCCVVC
CVCCVC
The prosodic representation of the two-mora forms with illustrative
examples is presented in (136). Forms of three moras are represented in
(137) and those of four moras are represented in (138):

(135) Two Moras

CVCC
CVCVC

(136) a. CVCC

b. CVCVC

()

()

V C C

V C V C

a h r

a r a m

(137) a. CVCVVC

b. CVVCVC
()

c. CVCCVC
()

()

129

C V

C V V C

C V V C V C

C V C C

V C

T a a b i 3

k a w k

a b

a a m

(138) a. CVVCVVC (SA only)

b. CVCCVVC

()

()

C VV C VV C

C V C C V VC

S a a r u u x

i m s a a r

The form in (138a) has three moras in SA, but it is reduced to two
in EA due to Atonic Shortening. Thus, in EA it will have a prosodic
representation quite similar to that in (137a) as follows:
(139) CVCVVC

C V
S

C
r

()

V V C
u

u x

Two remarkable observations arise from the prosodic representations of

the nominal stem forms: (i) none of them contains less than two moras and
(ii) all forms are disyllabic. To explain these two restrictions, McCarthy &
Prince (1990a) posited two constraints: the Minimal Word Constraint
and the Maximal Stem Constraint. The former explains the fact that no
nominal stem form contains less than two moras:
Since the word dominates the foot in the prosodic
hierarchy, the smallest word will be a single foot. We call a
word, stem, or other top-level category that exactly meets

130

this criterion MINIMAL. Since Arabic requires quantitative

trochaic stress feet, the minimal stem (and, therefore, word)
will be a single foot of this type, or two moras (McCarthy
& Prince 1990a: 17).
The Maximal Stem Constraint accounts for the fact that all stem forms
are disyllabic: Templates are maximally disyllabic (McCarthy & Prince
1990a: 25). Put differently, because all nominal stem forms are formed on
templates, no stem can exceed two intrametrical syllables. For further
details, see McCarthy and Prince (1990a).
3.2. Deverbal Nouns:
The term deverbal noun is linguistically used to refer to nouns
derived from verbs to distinguish them from denominal nouns which are
derived from other nouns and primary nouns which are derived directly
from the root. In this section, an attempt will be made to compare various
deverbal nouns including the verbal nouns (3.2.1), the nouns of
exaggeration (3.2.2), the nouns of place & time (3.2.3) and the nouns of
instrument (3.2.4).
3.2.1. Verbal Nouns:
A verbal noun is a noun that is derived from a verb to signify the
process expressed by that verb without reference to time. It is
approximately like the gerund in English. Syntactically, it can be used as a
noun, a verb or an absolute object (Haywood & Nahmad 1982). The
formation of the verbal noun differs from one verb to another, depending
on its form or class. There are more than forty verbal noun patterns for the
primary triradical verb. Certain patterns of this class are associated with
specific meanings and certain patterns are associated with specific verb
forms. However, it is not yet clear to what extent the choice of the verbal
noun pattern depends on the semantics of the verb and to what extent it
depends on its morphology, though some generalizations will be given
below. On the other hand, the verbal noun patterns of derived triradical as
well as quadriradical verbs can be predicted in most cases in a fairly
consistent manner.
Some of the Standard meaning-related patterns of primary
triradical verbs (Al-Hammadi et al. 1977) are listed below.
(i) [Fi3aaL(-un)] to signify refusal or abstaining, as in /ihaab(-un)/
going from /ahah(-a)/ to go, /qiyaam(-un)/ standing from

131

/qaam(-a)/ to stand and /giyaab(-un)/ absence from /gaab(-a)/ to be

absent
(ii)[Fi3aaL-at(-un)] to refer to an occupation, as in /ziraa3-at(-un)/
planting from /zara3(-a)/ to plant and /Sinaa3-at(-un)/
manufacturing from /Sana3(-a)/ to manufacture.
(iii)[Fa3aLaan(-un)] of continuous motion and disturbance, as in
/galayaan(-un)/ boiling from /galaa/ to boil and /dawaraan(-un)/
circulation from /daar(-a)/ to circulate.
(iv)[Fu3L-at(-un)] of color, as /humr-at(-un)/ redness from /?ahmar(-u)/
red /Sufr-at(-un)/ yellowness from /?aSfar(-u)/ yellow taking into
consideration that adjectives of color can be treated as stative verbs.
(v) [Fu3aaL(-un)] or [Fa3iiL(-un)] of sound, as /nubaah(-un)/ barking
from /nabah(-a)/ to bark and /Sahiil(-un)/ neighing from /Sahal(-a)/
to neigh.
(vi)[Fu3aaL(-un)] of sickness, as /zukaam(-un)/ cold from /zukim(-a)/ to
catch cold and /su3aal(-un)/ coughing from /sa3al(-a)/ to cough.
If a verbal noun does not denote any of the above meanings, it will
follow one of the following four patterns, which are the most common
among verbal noun patterns, according to the class of the verb from which
it is derived:
(i) [Fa3L(-un)] of transitive verbs of Forms [Fa3aL(-a)] and [Fa3iL(-a)], as
in /naSr(-un)/ victory from /naSar(-a)/ to make victorious and
/sam3(-un)/ hearing from /sami3(-a)/ to hear. This verbal noun
pattern has the variant [Fa33(-un)] in geminate verbs, e.g. /add(-un)/
pulling from /add(-a)/ to pull.
(ii)[Fa3aL(-un)] of intransitive verbs of the form [Fa3iL(-a)], as in
/Tarab(un)/ delight from /Tarib(-a)/ to be delighted and /marah(-un)/
joy from /marih(-a)/ to rejoice.
(iii)[Fu3uuL(-un)] of intransitive verbs of Form [Fa3aL(-a)], as in
/qu3uud(-un)/ sitting down from /qa3ad(-a)/ to sit down and
/sujuud(-un)/ prostration from /sajad(-a)/ to prostrate.
(iv)[Fu3uuL-at(-un)] and [Fa3aaL-at(-un)] from intransitives of Form
[Fa3uL(-a)], as in /Su3uub-at(-un)/ difficulty from /Sa3ub(-a)/ to
become difficult and /karaam-at(-un)/ nobility or dignity from
/karum(-a)/ to be noble. Since verbs of [Fa3uL(-a)] are usually stative,
one can propose that the [Fu3uuL-at(-un)] forms are usually abstract
nouns denoting a state. In this particular case semantics and
morphology concur in deciding the choice of the verbal noun pattern.
Most of the Standard patterns of primary triradical verbal nouns are
preserved in EA as indicated in Table (39). A few of these patterns are not

132

identical as shown in Table (40). The patterns of derived triradical verbal

nouns are contrasted in the two varieties in Tables (41) and (42). Those of
quadriradical verbal nouns in both of them are compared in Tables (43)
and (44).

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

Table (39)
Patterns of Primary Triradical
Verbal Nouns Identical in SA & EA
Pattern
Example
Gloss
Fa3L(-un)
Darb(-un)
striking
Fa3L-at(-un)
rahm-at(-un)
mercy
Fi3L(-un)
hifZ(-un)
preserving
Fi3L-at(-un)
xidm-at(-un)
service
Fu3L(-un)
ukr(-un)
thanking
Fu3L-at(-un)
ru?y-at(-un)
seeing
Fa3aL(-un)
karam(-un)
generosity
Fa3aL-at(-un)
afaq-at(-un)
compassion
Fa3aaL(-un)
fasaad(-un)
corruption
Fa3aaL-at(-un) Daxaam-at(-un) hugeness
Fi3aaL(-un)
hisaab(-un)
accounting
Fi3aaL-at(-un) kitaab-at(-un)
writing
Fu3aaL(-un)
su?aal(-un)
questioning
Fa3uuL(-un)
qabuul(-un)
agreeing
Fu3uuL(-un)
duxuul(-un)
entering
Fu3uuL-at(-un) buruud-at(-un)
coldness
Fa3iiL(-un)
rahiil(-un)
departure
Fi3Laan(-un)
nisyaan(-un)
forgetting
Fu3Laan(-un)
gufraan(-un)
forgiveness
Fa3aLaan(-un) xafaqaan(-un)
palpitating
maF3aL(-un)
maqtal(-un)
killing
maF3iL-at(-un) ma3rif-at(-un)
knowledge
maFa33-at(-un) mawadd-at(-un) friendliness
maFaaL(-un)
manaam(-un)
sleep/dream
maFiiL(-un)
maSiir(-un)
destination

Table (39) shows that the patterns of primary triradical verbal nouns have
the following templates: [CVCC-at], [CVCVC-at], [CVCVVC-at],
[CVCC-aan], [CVCVC-aan], [maCCVC-at], [maCVCC-at] and
[maCVVC]. What distinguishes EA from SA in this respect is that in the
former the pattern [Fa3aLaan] is more commonly used than in the latter. It
is used for the primary forms of geminate verbs and final-weak verbs
instead of or together with the pattern [Fa3L(-un)]. For instance, for the

133

verb /jarr(-a)/ to pull SA has one verbal noun pattern /jarr(-un)/ but EA
has two alternative nouns /garr/ or /gararaan/. Also, for the final-weak verb
/jaraa/ to run SA has the verbal noun /jary(-un)/ but EA has /gary/ or
/garayaan/.

No
1
2
3
4
5
6
7
8

Table (40)
SA Patterns of Primary Triradical
Verbal Nouns Phonologically Changed in EA
SA
EA
Gloss
Pattern
Example
Pattern
Example
Fa3Laa
da3waa
Fa3La
da3wa
call
Fi3aL(-un)
Sigar(-un)
Fu3L
Sugr
being small
Fa3iL(-un)
Dahik(-un)
Fi3L
Dihk
laughter
Fa3iLat(-un)
sariq-at(-un)
Fi3L-a
sir?-a
robbery
FawL(-un)
nawm(-un)
FooL
noom
sleeping
FayL-at(-un)
gayr-at(-un)
FiiL-a
giir-a
jealousy
Fa3aaLiy-at(-un) karaahiy-at(-un) Fa3aLiyy-a karahiyy-a hatred
Fi3aa
ginaa
Fi3a
gina
richness

Table (40) indicates that the SA verbal noun patterns of primary triradical
verbs that are phonologically changed in EA follow one of these
templates: [CaCCaa], [CVCVC-at], [CVCC-at], [CVCaaCiy-at] and
[CVCaa]. Some of the changes are phonologically regular and the others
are irregular:
(a)
(b)

(c)
(d)
(e)

Final-Vowel Shortening in EA can be used to account for the

changes in patterns (1 and 9).
High-Vowel Syncope and Raising can be used to explain the
changes in (3-4). Unfortunately, we have not yet adduced the exact
circumstances where raising occurs, so it is not obvious that this can
be called regular in the full sense of the word.
Monophthongization explains the change in (5).
Atonic Shortening occurs in pattern (7). However, the
gemination there is irregular.
The change in (2) is irregular. The change in (6) is also
irregular since we expect [ay > ee] normally.
Table (41)
Patterns of Derived Triradical Verbal Nouns in SA
No Verb Form
Verbal Noun
Example
II Fa33aL(-a)
taF3iiL(-un)
tahriib(-un) smuggling
taF3iL-at(-un)
tafriq-at(-un) dividing

134

tiF3aaL(-un)
tikraar(-un) repeating
III Faa3aL(-a)
mu-Faa3aL-at(-un) mu-3aamal-at(-un) treatment
Fi3aaL(-un)
xilaaf(-un) disagreement
IV ?aF3aL(-a)
?iF3aaL(-un)
?iDraab(-un) a strike
V taFa33aL(-a) taFa33uL(-un)
takabbur(-un) haughtiness
VI taFaa3aL(-a) taFaa3uL(-un)
tafaahum(-un) mutual
understanding
VII (?i)nFa3aL(-a) (?i)nFi3aaL(-un) (?i)nzi3aag(-un) disturbance,
VIII (?i)Fta3aL(-a) (?i)Fti3aaL(-un) (?i)htiraam(-un) respect
IX (?i)F3aLL(-a) (?i)F3iLaaL(-un) (?i)hmiraar(-un) turning red
X (?i)staF3aL(-a) (?i)stiF3aaL(-un) (?i)sti3maal(-un) using
Table (42)
Patterns of Derived Triradical Verbal Nouns in EA
No Verb Form Verbal Noun Example
IIa Fa33aL
taF3iiL
taksiir breaking
taF3iL-a
tafri?-a dividing
tiF3aaL
tikraar repeating
IIb Fa33iL
taF3iiL
ta?diim presenting
III Faa3iL
mu-Fa3L-a
mu-?abl-a meeting
Fi3aaL
xiSaam discord
IV ?aF3aL
?iF3aaL
?ihraag embarrassing
Va (?i)tFa33aL taFa33uL
tamarrud rebellion
taF3iiL
taksiir breaking
Vb (?i)tFa33iL taFa33uL
tagaddud renewal
VI (?i)tFaa3iL taFaa3uL
tafaahum mutual understanding
VII (?i)nFa3aL (?i)nFi3aaL
(?i)nbisaaT being happy
VIII (?i)Fta3aL (?i)Fti3aaL
(?i)ftikaar remembering
IX (?i)F3aLL
(?i)F3iLaaL
(?i)hmiraar turning red
Xa (?i)staF3aL (?i)stiF3aaL (?i)stigraab being surprised
Xb (?i)staF3iL (?i)stiF3aaL (?i)sti3gaal hurrying
Comparison of Tables (41) and (42) reveals that EA follows most
of the patterns of derived triradical verbal nouns used in the Standard
variety. There are two exceptions; the SA pattern [mu-Faa3aL-at(-un)] has
been reduced to [mu-Fa3L-a] in EA because it is underlyingly
[mu-Faa3iL-a] being derived from the verb [Faa3iL], then the verbal noun
form has its /i/ deleted by High-Vowel Deletion and its /aa/ shortened by
Closed-Syllable Shortening, leading to [mu-Fa3L-a]. For verbs of Form V,
EA tends to use the verbal noun pattern of Form II [taF3iiL] as an
alternative to [taFa33uL] which sounds purely Standard for Egyptian
speakers. Verb Forms II and III have more than one verbal noun pattern in

135

both varieties. They are listed in the order of their frequency of

occurrence. For instance, [mu-Faa3aL-at(-un)] is more frequent for verbs
of Form III than [Fi3aaL(-un)].
The following remarks should be observed in relation to the
formation of verbal nouns from derived non-sound verbs in both varieties:
(a) For geminate verbs of Form IV, the verbal noun pattern is
[?iF3aa3(-un)], in both varieties, e.g. /?itmaam(-un)/ completion. This
is expected because the second and third consonants in geminate verbs
are identical.
(b) For final-glottalized verbs, SA verbal nouns have their /?/ deleted in
EA, by Final /?/ Deletion as in /(?i)btidaa?(-un) > (?i)btida/ starting and
/(?i)xtifaa?(-un) > (?i)xtifa/ disappearance.
(c) For initial-weak verbs of Form IV, the underlying verbal noun pattern
[?iG3aaL(-un)] is changed into [?ii3aaL(-un)] by Perseverative Vocoid
Assimilation, as in /?iiqaan(-un)/ believing and /?iiraa(-un) > ?irs/
bequest. For those of Form X, the verbal noun is [(?i)stii3aaL(-un)]
which is derived from [(?i)stiG3aaL(-un)] by Perseverative Vocoid
Assimilation as well, e.g. /(?i)stiiqaan(-un)/ asking for affirmation and
/(?i)stiiqaaf(-un)/ stopping. Both of these examples sound SA but they
can be borrowed by educated Egyptian speakers.
(d) For medial-weak verbs of Form IV, the variant of the verbal noun
pattern [?iF3aaL(-un)] is [?iFaaL-at(-un)] from underlying [?iFGaaL(un)], as in /?i3aad-at(-un)/ repetition. Here, the [-at] is added to
compensate for the deleted glide. For those of Form VII, the verbal noun
has the pattern [(?i)nFiyaaL(-un)], e.g. /(?i)nsiyaaq(-un)/ submission.
For those of Form VIII, the verbal noun pattern is [(?i)FtiyaaL(-un)], e.g.
/(?i)xtiyaar(-un)/ choice. There is nothing exceptional in relation to the
verbal nouns of Forms VII and VIII y-verbs or w-verbs when we note
that medial w-verbs have their /w/ changed to /y/ after the high vowel /i/
by Perseverative Vocoid Assimilation. For those of Form X, the variant
of [(?i)stiF3aaL(-un)] is [(?i)stiFaaL-at(-un)] from underlying
[(?i)stiF3aaL(-un)], as in /(?i)stifaad-at(-un)/ benefiting. Here, as in
Form IV, the [-at] is added to compensate for the lost glide.
(e) For final-weak verbs of Form II, the verbal noun has the pattern
[taF3iy-at(-un)]. This is not irregular in y-verbs since the /y/ is used in
place of the third radical, as in /taniy-at(-un) > tasniy-a/ seconding. In
w-verbs, the /w/ is changed to /y/ after /i/ by Perseverative Vocoid
Assimilation. For those of Form III, the variant of the pattern [muFaa3aL-at(-un)] is [mu-Faa3aat(-un)] which is derived from [muFaa3aG-at(-un)] by Glide Elision. It is borrowed in EA for some verbs

136

and changed into [mi-Fa3iyy-a] for others, e.g. /mu-3aanat(-un) > mu3anah/ suffering but /mu-naadaat(-un) > mi-nadiyy-a/ calling. The
variant of the pattern [Fi3aaL(-un)] is [Fi3aa?(-un)] by the Glottal
Formation rule in SA. It is preserved as is or reduced to [Fi3a] by Final
/?/ Deletion and Final-Vowel Shortening in EA, e.g. /nidaa?(-un) >
nidaa? ~ nida/ calling. For those of Form IV, the variant o[?iF3aaL(un)] is [?iF3aa?(-un)] derived from underlying [?iF3aaG(-un)] by Glottal
Formation, e.g. /?igraa?(-un)/ enticement.
(f) For final-weak verbs of Form V, the verbal noun is [taFa33i(n)] which
is derived from underlying [taFa33uy(-un)] by u-to-i, Glide Elision and
Perseverative Vocoid Assimilation. The indefinite marker /n/ is deleted
in EA leading to [taFa33i], e.g. /ta?anni(n) > ta?anni/ deliberation. For
those of Form VI, the variant of [taFaa3uL(-un)] is [taFaa3i(n)], which is
derived from underlying [taFaa3uy(-un)] by u-to-i, Glide Elision and
Perseverative Vocoid Assimilation, e.g. /tafaadi(n) > tafaadi/ avoiding.
For Form VII, the verbal noun is [(?i)nFi3aa?(-un)] in both varieties, e.g.
/(?i)ntihaa?(-un)/ ending. For VIII, the verbal noun is [(?i)Fti3aa?(-un)]
which is preserved as is or reduced to [(?i)Fti3a] by Final /?/ Deletion
and Final-Vowel Shortening in EA, e.g. /(?i)ntihaa?(-un)/ ending but
/(?i)xtifaa?(-un) > (?i)xtifa/ disappearance. For X, the verbal noun is
[(?i)stiF3aa?(-un) > (?i)stiF3a/, e.g. /(?i)stinaa?(-un) > (?i)stisna/
exception.
Turning to the verbal noun patterns of quadriradical verbs, compare Tables
(43) and (44).
Table (43)
Verbal Noun patterns of Quadriradical Verbs in SA
No Verb Form
Verbal Noun
Example
I Fa3L1aL2(-a)
Fa3L1aL2-at(-un)
zaxraf-at(-un) decoration
"
Fi3L1aaL2(-un)
zilzaal(-un) earthquake
II taFa3L1aL2(-a)
taFa3L1uL2(-un)
tadahwur(-un)
deterioration
IV (?i)F3aL1aL2L2(-a) (?i)F3iL1L2aaL2(-un) (?i)Tmi?naan(-un)
peacefulness
Table (44)
Verbal Noun patterns of Quadriradical Verbs in EA
No Verb Form
Verbal Noun
Example
Ia Fa3L1aL2
Fa3L1aL2-a
laxbaT-a confusion
Ib Fa3L1iL2
Fa3L1aL2-a
falfal-a seasoning
"
Fi3L1aaL2
zilzaal earthquake
II (?i)tFa3L1aL2
Fa3L1aL2-a
margah-a swinging

137

IV (?i)F3aL1aL2L2

(?i)F3iL1L2aaL2

(?i)mi?zaaz shuddering
with disgust

It is evident from the comparison of Tables (43) and (44) that the
Colloquial variety has lost the Standard verbal noun pattern of
quadriradical verbs of Form II, [taFa3L1uL2]. Except when borrowing
from SA, it uses that of Form I [Fa3L1aL2-a] for some verbs of Form II.
For other verbs of this form it employs a new pattern, [tiFa3L1iiL2] which
was referred to by Carter (1996: 141): a new verbal noun pattern is
undeniably emerging, as seen in itfabrik / tifabriik, which has become
genuinely productive and is used with a wide range of stem types (true
quadrilateral, augmented, reduplicated).
3.2.2. Nouns of Exaggeration:
The noun of exaggeration is a nominal derived from a verb to refer
to the person who, as a general habit, performs some action; which implies
that it is done repetitively. It has five patterns in both varieties:
1. [Fa33aaL(-un)]: as in /kaaab(-un) > kaddaab/ a liar, lying
frequently. In both SA and EA, the pattern [Fa33aaL(-un)] is generally
used for occupational nouns, whereas the pattern [Faa3iL(-un)] is used
for their participial counterparts (cf. Radwan 1975: 43). For instance,
/xaddaam(-un)/ a servant vs. /xaadim(-un)/ has served,
/xabbaaz(-un)/ a baker vs. /xaabiz(-un)/ has baked and /jarraah(-un)
> garraah / a surgeon vs. /jaarih(-un) > gaarih / has wounded.
Contrary to their being changed to /oo/ and /ee/ in other situations, the
sequences /aw/ and /ay/ are preserved in EA in this pattern because the
glide in question forms a part of a geminate consonant and is thus
subject to geminate inalterability and also because Monophthongization
occurs only at stem-final positions. Other EA examples are /xawwaaf/
frightened so much, /gawwaal/ a boy scout, /bayyaa3/ a seller and
/ayyaal/ a porter.
2. [mi-F3aaL(-un)]: as in /mi-haar(-un) > mi-hzaar/ joking so much.
3. [Fa3uuL(-un)]: as in /akuur(-un)/ thanking so much and /gafuur(-un)/
forgiving so much; two epithets of God.
4. [Fa3iiL(-un)]: as in /samii3(-un)/ listener and /rahiim(-un)/ merciful.
5. [Fa3iL(-un)]: as in /hair(-un) > hazir/ cautious /marih(-un)/ merry
and /yaqiZ(-un)/ awake
Very few nouns of exaggeration are taken from derived triradical
verbs following the same patterns stated above. For instance, /baiir(-un)/
announcer of good news from /baar(-a)/ to announce good news,

138

/naiir(-un) > naziir/ warning from /?anar(-a)/ to warn and

/mi-qdaam(-un)/ courageous from /?aqdam(-a)/ to venture.
3.2.3. Nouns of Place & Time:
The nouns of place and time are nominals derived from a verb to
signify, respectively, the place and time of occurrence of the action
indicated by the verb. They follow one process of derivation and are only
distinguished by the context in which they occur. In both varieties, these
nouns are usually not derived from quadriradical verbs. They are formed
from the primary triradical verbs according to one of the following
patterns (cf. Al-Toma 1969):
1. [ma-F3aL(-un)]: This pattern is used for verbs having in the SA
imperfect /a/ or /u/ as a stem vowel. It is borrowed in EA even after the
stem vowel is changed to /i/:
SA Impf Verb
ya-3mal-u
ya-Sna3-u
ya-ktub-u
ya-dxul-u

N of Place/Time
ma-3mal(-un)
ma-Sna3(-un)
ma-ktab(-un)
ma-dxal(-un)

Gloss
a laboratory
a factory
an office
an entrance

An exception here is the noun of place /ma-xzan/ a warehouse which

is derived from a verb having /i/ as a pre-final imperfect vowel:
/ya-xzin(-u) > yi-xzin/ to store. In doubled verbs the variant of
[ma-F3aL(-un)] is [ma-Fa33(-un)] as a consequence of the general rule
of Identical-Consonant Metathesis, as in /ma-marr(-un)/ a passage and
/ma-hall(-un)/ a shop. For medial-weak verbs, nouns of place and
time have the pattern [ma-FaaL(-un)] derived from [ma-FwaL(-un)] in
w-verbs and [ma-FyaL(-un)] in y-verbs by Anticipatory Vocoid
Assimilation, e.g. /ma-kaan(-un)/ a place and /ma-Taar(-un)/ an
airport. For final-weak verbs, nouns of place and time have the pattern
[ma-F3a(n) > ma-F3a] from [ma-F3ay(-un)] by Glide Elision and
Perseverative Vocoid Assimilation, then by Closed-Syllable Shortening
in SA and Final-Vowel Shortening in EA, e.g. /ma-lha(n) > ma-lha/ a
playhouse.
2. [ma-F3aL-at(-un)]. This pattern, occurring often with verbs having /u/
in the SA imperfect stem, represents nouns taking the suffix [-at]. It is
also borrowed in EA, even after the stem vowel is changed to /i/. The
gender of the noun of place/time , i.e. whether it follows

139

[ma-F3aL(-un)] or [ma-F3aL-at(-un)], is not predictable (cf. Thackston

1984: 166):
SA Impf Verb
ya-drus(-u)
ya-ktub(-u)
ya-qbur(-u)
ya-hkum(-u)

N of Place/Time
ma-dras-at(-un)
ma-ktab-at(-un)
ma-qbar-at(-un)
ma-hkam-at(-un)

Gloss
a school
a library
a graveyard
a law court

For geminate verbs, the variant of [ma-F3aL-at(-un)] is

[ma-Fa33-at(-un)]
by
Identical-Consonant
Metathesis
from
[ma-F3a3-at(un)], e.g. /ma-haTT-at(-un)/ a station. For medial-weak
verbs, the noun of place is [ma-FaaL-at(-un)] from [ma-FGaL-at(-un)]
by Anticipatory Vocoid Assimilation, as in /ma-gaar-at(-un)/ a cave.
3. [ma-F3iL(-un)]: This pattern is employed for initial-weak verbs, and
sound verbs having /i/ as a stem vowel in the SA imperfect. It borrowed
as is in EA.
SA Impf Verb
ya-3id(-u)
ya-rid(-u)
ya-nzil(-u)
ya-jlis(-u)

N of Place/Time
ma-w3id(-un)
ma-wrid(-un)
ma-nzil(-un)
ma-jlis(-un)

Gloss
an appointment
a resource
a house
a council

An exception here is the noun of place /ma-sjid(-un) > masgid/ a

mosque which is derived from a verb having /u/ as an imperfect prefinal vowel, /ya-sjud(-u) > yi-sgud/ to prostrate. In some instances,
EA uses the pattern [ma-F3aL] for the SA nouns of place and time of
the pattern [ma-F3iL(-un)], e.g. /ma-wqif(-un) > ma-w?af/ a
stop/station. (One can say that the sequence /aw/ does not undergo
Monophthongization here because it is not in a stem-final syllable). In
some other instances, the Colloquial variety uses [ma-F3iL] for the
Standard pattern [ma-F3aL(-un)], e.g. /ma-rkab(-un) > ma-rkib/. The
choice of the vocalism in EA patterns seems to be arbitrary. In medialweak verbs, the pattern [ma-F3iL(-un)] has the variant [ma-FiiL(-un)]
from [ma-FGiL(-un)] by Anticipatory Vocoid Assimilation in both
varieties, as in /ma-Siir(-un)/ a destiny.
4. [ma-F3iL-at(-un)]: This also occurs with verbs with /i/ as a pre-final
vowel in the SA imperfect and is borrowed as is in EA. It is not known
which verbs of this class take [ma-F3iL(-un)] and which
[ma-F3iL-at(-un)]:

140

SA Impf Verb
ya-nzil(-u)
ya-nTiq(-u)

N of Place/Time
ma-nzil-at(-un)
ma-nTiq-at(-un)

Gloss
a position (rank)
a zone

Nouns of place and time that are taken from derived triradical
verbs have the form [mu-...aC] but they are less frequent in use than those
of the primary triradical verbs, particularly in EA:
SA Impf Verb
ya-jtami3(-u)
ya-stahill(-u)
ya-stawdi3(-u)
ya-stafii

N of Place/Time
mu-gtama3(-un)
mu-stahall(-un)
mu-stawda3(-un)
mu-stafa(n)

Gloss
a society
a starting date
a depository
a hospital

3.2.4. Nouns of Instrument:

A noun of instrument is one which denotes the instrument by
which the action shown by the verb is performed. Traditionally, there are
three Standard patterns for this type of nouns: [mi-F3aL(-un),
mi-F3aL-at(-un) and mi-F3aaL(-un)]. These patterns are also found in EA,
although this variety replaces the instrument prefix [mi-] in the first two
patterns by [ma-] and in some instances of the third pattern by [mu-].
Compare the following nouns of instrument in both varieties:
SA
mi-dfa3(-un)
mi-gzal(-un)
mi-Syad-at(-un)
mi-nqal-at(-un)
mi-hraa(-un)
mi-ftaah(-un)

EA
ma-dfa3
ma-gzal
ma-Syad-a
ma-n?al-a
mi-hraat
mu-ftaah

Gloss
a cannon
a spindle
a snare
a protractor
a plough
a key

For geminate verbs, the variant of the noun of instrument pattern

[mi-F3aL(-un) > ma-F3aL] is [mi-Fa33(-un) > ma-Fa33] by IdenticalConsonant Metathesis from [mi-F3a3(-un) > ma-F3a3], as in
/mi-qaSS(-un) > ma-?aSS/ a pair of scissors. For initial-weak verbs, its
variant is [mii3aaL(-un) > mi3aaL] from [mi-G3aaL(-un)] by
Perseverative Vocoid Assimilation in both varieties, then by Atonic
Shortening in EA, e.g. /miizaan(-un) > mizaan/ a balance. For final-weak
verbs the variant is [mi-F3aat(-un) > ma-F3a] from underlying
[mi-F3aG-at(-un)] by Glide Elision in both varieties, then by Final-Vowel
Shortening in EA, e.g. /mi-kwaat(-un) > ma-kwa/ an iron.

141

According to Al-Toma (1969: 75), the Egyptian Academy of

Arabic has recently sanctioned a fourth pattern [Fa33aaL-at(-un)] which is
now introduced in prescriptive grammar textbooks as a correct instrument
pattern in SA. Nouns like /gassaal-at(-un)/ a washing machine,
/barraay-at(-un)/ a pencil sharpener and /tallaag-at(-un)/ a refrigerator
are also frequently encountered in EA as well as in other Arabic dialects.
3.3. Definite vs. Indefinite Nouns:
Definiteness is considered one of the inflectional categories by
Aboul-Fetouh (1969: 24). It plays a significant role in the inflection of
nouns for another important category, namely case (See 3.4). Arabic nouns
are either definite or indefinite. The commonest device of defining a
Standard noun is the morphological marker [al-] called the definite article,
prefixed to the noun and assimilated to any initial solar, i.e. coronal,
consonant by the l-Assimilation rule. As argued in (1.2.1), a glottal stop is
inserted after pause by Glottal Stop Insertion. The Colloquial variety uses
a shortened form of this article, namely [l-]. Then an /i/ is inserted by
Word-Initial Epenthesis and a glottal stop is inserted after pause by Glottal
Stop Insertion. Examples of definite nouns in the two varieties are:
SA
(?)al-kitaab(-u)
(?)al-qamar(-u)
(?)aT-Taalib(-u)
(?)a-ams(-u)

EA
(?i)l-kitaab
(?i)l-?amar
(?i)T-Taalib
(?i)-ams

Gloss
the book
the moon
the student
the sun

Nouns also become definite when followed by a determining

complement (Bateson 1967: 10) which may be the second noun in a
construct phrase17, or a suffixed pronoun indicating possession. In this
case, they do not necessitate a definite article to be definite. Only the
second noun can carry the definite article even though both may be
definite (McGuirk 1986: 25). Examples from SA are: /kitaab-u
?ahmad(-a)/ Ahmads book, /kitaab-u 1-walad(-i)/ the boys book and
/kitaab-u-hu/ his book. This means of definiteness is not quite important
in EA, since case endings disappeared in this variety. Proper nouns are
also definite without an article in both varieties, e.g. /3ali/ Ali, /hasan/
Hassan and /3umar/ Omar.
3.4. Inflection for Case:
The term case has been defined by Crystal (1985: 42) as a
GRAMMATICAL CATEGORY used in the analysis of WORD classes to

142

identify the SYNTACTIC relationship between words in a SENTENCE.

Standard Arabic nouns inflect for three cases: nominative, accusative and
genitive.
It is generally agreed that the nominative case is used to mark the
following nouns:
(i) the subject of a verbal sentence,
(ii) the two parts of an equational sentence,
(iii) the subject noun of the verb /kaan(-a)/ to be and its sisters,18 and
(iv) the predicate noun of the particle /?inna/ indeed or verily and its
sisters.18
The accusative case is employed to signal the following nouns:
(i) the object of a transitive verb,
(ii) the subject noun of the particle /?inna/ verily and its sisters, and
(iii) the predicate noun of the verb /kaan(-a)/ to be and its sisters.
The genitive case marks the following nouns:
(i) the second noun in a possessive construct, and
(ii) the object of a preposition.
(iii) the object of a nominalization.
SA has a complex system of declension and case endings that may
seem unusual to the speakers of EA. Its complexity arises from the fact
that case endings are not always applicable to all nouns or to the same
nouns under different conditions. Thus, SA nouns are classified into two
classes according to the extent of their declension: Fully-declined and
semi-declined (Gaballa 1986: 25-6). Arab grammarians (e.g. Al-Hammadi
et al. 1977) call the first class /al-maSruuf(-u)/ the nunatable and the
second /al-mamnuu3-u min aS-Sarf(i)/ the non-nunatable. This
distinction is related to an Arabic phenomenon called /tanwiin(-un)/
nunation which refers to the presence of an [-n] at the end of a case
ending (See 1.4.2).
SA fully-declined nouns take two sets of case markers. The first
set, used for definite nouns, includes nominative [-u], accusative [-a] and
genitive [-i]. The second set, used for indefinite nouns, comprises
nominative [-un], accusative [-an] and genitive [-in]. In other words, the
presence or absence of the [-n] is determined by whether the noun in

143

question is [definite]. If the noun is indefinite, the [-n] final suffix is

used; otherwise the other alternative is employed.
The examples in (140 and 141) indicate the use of the case markers
with the noun /walad(-un)/ a boy in the Standard variety in its definite
and indefinite forms, respectively:
ward-ah
(140) a. rasam-a
1-walad-u
draw.pf-3msg
the-boy-Nom
(a) rose
The boy drew a rose.
b. ra?ay-tu
1-walad-a
fi-1-faSl
see.pf-1sg
the-boy-Acc
in-the-classroom
I saw the boy in the classroom.
c. sallam-tu
3ala
1-walad-i
fi-1-faSl
greet.pf-lsg (prep)
the-boy-Gen in-the-classroom
I greeted the boy in the classroom.
It should be stressed here that the short vowels which are used as case
endings in SA contextual forms (like those in (140) are dropped in pausal
forms, as was discussed in (1.4.1).
(141) a. rasam-a
walad-un
draw.pf-3msg
boy-Nom
A boy drew a rose.
b. ra?ay-tu
walad-an
see.pf-1sg
boy-Acc
I saw a boy in the classroom.
c. sallam-tu
3ala
walad-in
greet.pf-lsg (prep)
boy-Gen
I greeted a boy in the classroom.

ward-ah
(a) rose
fi-1-faSl
in-the-classroom
fi-1-faSl
in-the-classroom

The second class of SA nouns, semi-declined ones, is characterized

by the absence of nunation. Most of the nouns in this class take the three
endings in the definite form, i.e. [-u] in the nominative, [-a] in the
accusative and [-i] in the genitive. But they take two endings in the
indefinite, i.e. [-u] in the nominative and [-a] in both the accusative and the
genitive. That is why they are traditionally called diptotes, i.e. twocase nouns (Bateson 1967: 11). Among the nouns in this class are the
following major categories (cf. Al-Toma 1969: 34):
(a)

Broken plurals (See 3.6.3.3) with any of these patterns:

SA Pattern
Example
Gloss
Fa3aaL1iL2(-u)
qanaabil(-u)
bombs

144

Fa3aaL1iiL2(-u)
Fa3aa?iL(-u)
Fawaa3iL(-u)
?aFaa3iL(-u)
?aFaa3iiL(-u)
maFaa3iL(-u)
maFaa3iiL(-u)
Fu3aLaa?(-u)
?aF3iLaa?(-u)
Fa3aaLaa
Fa3Laa

3aSaafiir(-u)
qabaa?il(-u)
3awaaTif(-u)
?amaakin(-u)
?ahaadii(-u)
maSaani3(-u)
maSaabiih(-u)
u3araa?(-u)
?aqribaa?(-u)
yataamaa
marDaa

sparrows
tribes
emotions
places
talks
factories
lamps
poets
relatives
orphans
patients

McCarthy & Prince (1990a: 25-26) explain the reason why those
particular patterns are semi-declinable: what the diptotic broken
plurals all have in common is that they are stems with three syllables one greater than the maximum. By the maximum they refer to the
Maximum Stem Constraint which proposes that Templates are
maximally disyllabic. Though the last pattern [Fa3Laa] is two
syllables in the surface form, it has three syllables in the underlying
form [Fa3Lay(-u)] which undergoes Glide Elision and Perseverative
Vocoid Assimilation to produce [Fa3Laa]. Broken plurals of other
patterns are considered fully-declined and are treated like the singular
in relation to case.
(b)

Nouns with a final non-radical [-aa or -aa?], e.g. /buraa/ a

piece of good news and /Sahraa?(-u)/ a desert. If we postulate that
these suffixes are underlyingly [aGa] and [aGa?], respectively, the
nouns containing them will be three-syllable words and their being
diptotes can be accounted for by McCarthy & Princes generalization
above. Notice, for instance, that /buraa/ has the dual /burayaan/ and
/Sahraa?/ has the plural /Sahraawaat/, which indicates that they
contain underlying glides.
(c)
Proper nouns of foreign origin, or ending in [-at], non-radical [aan] or [-aa?], of the pattern [Fu3aL(-u)], verb-like names, names of
compound forms and most feminine proper names. Examples are:
/?idriis(-u)/, /xaliif-at(-u)/, /marwaan(-u)/, /najlaa?(-u)/, /3umar(-u)/,
/?ahmad(-u)/, /niyu yoork(-u)/ and /faaTim-at(-u)/. Proper nouns
ending in [-aa] and [-aa?] can be accounted for by the Maximal Stem
Constraint if we postulate that these suffixes are underlyingly
[-aGan] and [-aGa?]. It is difficult to prove this postulation though
Brame (1970: 431) asserts that in Arabic there do not exist long
vowels in underlying representations. Long vowels are simply the
result of the application of various phonological rules.

145

In contrast, EA has no case endings, nor does it use nunation in

forming indefinite nouns. The process which is responsible for the
disappearance of case endings in the Colloquial variety is known among
linguists as loss. Loss has been defined by Langacker (1977: 68) as the
disappearance from a language of some element or grammatical device.
Compare the Colloquial examples in (142 and 143) with their Standard
counterparts in (140 and 141), respectively:
rasam
(142) a. ?il-walad
the-boy
draw.pf
The boy drew a rose.
b. uf-t
il-walad
see.pf-1sg
the-boy
I saw the boy in the classroom.
c. sallam-t
3a
1-walad
greet.pf-lsg (prep) the-boy
I greeted the boy in the classroom.

ward-a
(a) rose
fi-1-faSl
in-the-classroom
fi-1-faSl
in-the-classroom

(143) a. walad
rasam
ward-a
(a) boy
draw.pf
(a) rose
A boy drew a rose.
fi-1-faSl
b. uf-t
walad
see.pf-1sg
(a) boy
in-the-classroom
I saw a boy in the classroom.
c. sallam-t
3ala
walad
fi-1-faSl
greet.pf-1sg (prep)
(a) boy
in-the-classroom
I greeted a boy in the classroom.
Concomitant with the loss of case inflections in EA was the introduction
of stricter word-order requirements. Whereas SA prefers VSO to SVO, EA
prefers SVO to VSO. That is why the subject of SA (140a and 141a) is
moved to the beginning of EA (142a and 143a).
At the end of the analysis of case endings a word must be said
about the so-called five nouns: /?ab(-un)/ father, /?ax(-un)/ brother,
/ham(-un)/ father-in-law, /fuu/ mouth and /uu/ possessor of. Table
(45) displays the various forms of these nouns in both varieties:
Table (45)
The Five Nouns in SA & EA
Citation Form
SA

EA

146

SA > EA
?ab(-un) > ?abu ~ ?abb
?ax(-un) > ?axu ~ ?axx
ham(-un) > hama
fuu > (lost)
uu > (lost)

Nom
?abuu?axuuhamuufuuuu-

Acc
?abaa?axaahamaafaaaa-

Gen
?abii?axiihamiifiiii-

?abuu?axuuhamaa

Table (45) shows that in SA the five nouns have three forms
depending on case; nominative with [-uu], accusative with [-aa] and
genitive with [-ii] but case distinction is lacking in EA. If we propose that
in SA the five nouns end in /w/ in their underlying representation, we can
see why these suffixes are long. An independent justification that
/?ab(-un)/, for instance, ends in /w/ is that this glide appears in the dual
and relational adjective forms, e.g. /?abaw-aani/ two fathers and
/?abawiyy(-un)/ parental. Thus, we can propose the following
derivations:
(144) a. ?abawu-ka ?abuu-ka (Nom)
b. ?abawa-ka ?abaa-ka (Acc)
c. ?abawi-ka ?abii-ka (Gen)
However, although the derivation in (b) can be accounted for by Glide
Elision and that in (c) by Glide Elision and Anticipatory Vocoid
Assimilation, the derivation in (a) is exceptional to the rule of
Perseverative Vocoid Assimilation as this rule requires /awu/ to become
/aa/.
In EA, /fuu-/ and /uu-/ are not used at all. For the remaining
nouns, the three case endings are reduced into one morpheme, namely
[-uu-] for /?abu/ and /?axu/ and [-aa-] for /hama/. So, instead of assuming
that the [-uu-] and [-aa-] are residues of case endings, it is more plausible
to consider them empty morphs and to assume that there are two forms
of the first two nouns in the EA lexicon, i.e. /?abu ~ ?abb/ and /?axu ~
?axx /, and one form /hama/. The vowel at the end of these nouns is
underlyingly long but is shortened by Final-Vowel Shortening in EA. This
rule does not apply before a suffix beginning with a consonant, i.e. the
vowel remains long in that case. Compare the Standard sentences with
their Colloquial equivalents in (145-147):
(145) SA: ?ab-uu-ka
fi-1-bayt
father-Nom-2msg in- the-house
Your father is in the house.

147

EA: ?abuu-k
fi-1-beet
father-2msg
in-the-house
Your father is in the house.
?ab-aa-ka
(146) SA: ra?ay-tu
see-pf-1sg father-Acc-2msg
I saw your father in the house.
EA: uf-t
?abuu-k
see.pf-1sg father-2msg
I saw your father in the house.

fi-1-bayt
in-the-house
fi-1-beet
in-the-house

3ala ?ab-ii-ka
fi-l-bayt
(147) SA: sallam-tu
greet-pf-1sg
(prep) father-Gen-2msg
in-the-house
I greeted your father in the house.
EA: sallam-t
3al-abuu-k
fi-1-beet
greet-pf-1sg
(prep)-father-2msg in-the-house
I greeted your father in the house.
Only singular case endings have been dealt with in this section.
Dual and plural case markers will be dealt with in (3.6) since they are
considered number markers as well.
3.5. Inflection for Gender:
A distinction is generally maintained between two genders in
Arabic nouns, as well as in personal, relative and demonstrative pronouns:
masculine and feminine. Nouns of feminine gender are often indicated by
a suffix, but there are exceptions either of feminine nouns lacking the
suffix or of masculine nouns having feminine suffixes. For instance, nouns
which semantically refer to females are feminine without feminine
markers, e.g. /bint(-un)/ a girl and /?uxt(-un)/ a sister, while nouns like
/xaliif-at(-un)/ a caliph and /hamz-at(-u)/ a mans name are masculine
though they have a feminine ending. In most cases, EA corresponds to the
Standard variety in making gender distinction.
There are two classes of feminine suffixes: [-at] with the pausal
form /-a(h)/ on the one hand; [-aa?] and [-aa] on the other. The suffix [-at]
is the most widely used marker to express feminine gender in both
varieties. With very few exceptions, it is considered the sign of a feminine
noun regardless of whether there exists a corresponding masculine form or
not (cf. Mitchell 1956: 19). Consider the following examples:
Masculine

SA Fem

EA Fem

Gloss

148

mudarris(-un)
muhandis(-un)

nahr(-un)
3alam(-un)

mudarris-at(-un)
muhandis-at(-un)
majall-at(-un)
daraj-at(-un)

mudarris-a
muhandis-a
magall-a
darag-aa

a teacher
an engineer
a magazine
degree
a river
a flag

This shows that in some cases there is a corresponding feminine singular

to the masculine singular but some feminine nouns have no corresponding
masculine forms. Likewise, some singular masculine nouns have no
corresponding feminine forms. In some instances, resort is made to other
lexical items, e.g. /?asad(-un) a lion has the feminine equivalent
/labu?-at(-un) > labw-a/ a lioness and /hiSaan(-un) > huSaan/ a horse
has the feminine counterpart /muhr-at(-un) > muhr-a/ a filly. The
existence of masculine nouns with final [-at] leads to the assumption that
[-at] is not a feminine affix but rather a class marker which is introduced
by the gender feature [+fem], as in the normal case, or by a diacritic
feature [+-at], as in the exceptional cases.
In SA, the morpheme [-at] has the allomorph [-ah] in the pausal
form of nouns. In other words, there are two allomorphs of the feminine
ending in SA: [-ah#] and [-at elsewhere]. Compare the contextual forms of
nouns in (148-149a) and their pausal analogues in (148-149b),
respectively:
fi-1-faSl
(148) a. ?al-mudarris-at-u
the-teacher-f-Nom
in- the-classroom
The teacher (f) is in the classroom.
b. ra?ay-tu
1-mudarris-ah
see.pf-1sg
the-teacher-f
I saw the teacher (f).
(149) a. haaihi
wardat-un
this (f)
rose-Nom
This is a pretty rose.
b. maa
?ajmal-a
what
make pretty.pf-3msg
What a pretty rose!

jamiil-ah
pretty-f
1-ward-ah
the-rose

In EA, this feminine marker presumably has two allomorphs: [-a#]

and [-t elsewhere]. This can be accounted for by the rule of [-a ~ -t]
Suppletion in EA stated in (3.1.1) above. The allomorph [-a] is used in

149

final positions, as in /madras-a/ a school and /ward-a/ a rose and the

allomorph [-t] is used in non-final positions, e.g. when the feminine noun
is followed by a defining genitive or a pronominal suffix. In this context,
the vowel /a/ preceding /t/ is either dropped or preserved. It is dropped
when it is preceded by a consonant, as in /madras-t il-walad/ the school of
the boy and /madras-t-u/ his school. The default epenthetic vowel /i/ is
inserted to break an ensuing long cluster of three consonants, as in
/madras-it 3ali/ the school of Ali, /ward-it il-walad/ the rose of the boy
and /ward-it-u/ his rose. The vowel /a/ is preserved in words containing
two light syllables, as in /luga-t il-3aSr/ the language of the age and
/luga-t-u/ its language. The reason /a/ is preserved here is that it is part of
the stem, as the minimal stem constraint requires any stem to have at least
two moras (McCarthy & Prince 1990a: 17). This can be captured by
comparing the moraic representations of the words /ward-a/ a rose and
/luga/ a language. The figure shows that the /a/ in /ward-a/ is a feminine
suffix but in /luga/ it is part of the stem.
The discussion above leads to the assumption that the allomorphs
of the feminine ending in EA are actually [-a#] and just [-t elsewhere].
Nothing is added before [-t] when it is preceded by a vowel or one
consonant and /i/ is inserted when a cluster of three consonants is likely to
occur.
(150) a.

b.

()

a r

d -a

g a

The second class of feminine suffixes in SA includes [-aa?] and

[-aa]. In the Colloquial variety, these two feminine markers are generally
replaced in the surface form by [-a]. The regular phonological rules
leading to this change are Final /?/ Deletion and Final-Vowel Shortening.
Compare the following words in both varieties:
SA
samaa?(-un)

EA
sama

Gloss
sky

150

Sahraa?(-un)
miinaa?(-un)
3adwaa
fatwaa

Sahra
miina
3adwa
fatwa

a desert
a port
an infection
a religious opinion

However, the final /?/ is retained in EA medial position, e.g. /mina?-een <
miinaa?-aani/ two ports, which suggests that the glottal stop is still there
in the underlying form of the Colloquial feminine marker. Also, nouns
ending in [-a < -aa] are dualized in EA by the use of /t/, e.g. /fatwit-een <
fatway-aani/ two religious opinions, which indicates that the ending [-aa]
has collapsed with [-at] in EA. Both are replaced by [-t ~ a#]. From the
discussion above, it can be observed that the Colloquial variety has offered
the following equivalents to the SA feminine markers:
SA
-at ~ -a(h)#
-aa?
-aa

EA
-t ~ -a#
-a? ~ -a#
-t ~ -a#

As stated earlier, there are certain exceptions to the morphological

rules formulated above. For instance, SA offers a significant number of
nouns taking either gender, e.g. /ruuh(-un)/ a soul, /suuq(-un)/ a
market, /kabid(-un)/ a liver and /Tariiq(-un)/ a road (cf. Al-Toma
1969: 36). EA uses only one gender for such nouns, some of which are
always masculine, e.g. /suu?/ a market; others are constantly feminine,
e.g. /rooh/ a soul.
Furthermore, the following exceptions in the Colloquial variety
were stated by Gary & Gamal-Eldin (1982: 75):
i. Not all nouns are marked for gender, although every noun has to
belong to a gender. Thus, /?alam/ a pen is masculine, but /widn/ an
ear is feminine.
ii. A noun may be feminine in the singular but masculine in the plural,
e.g. /burtu?aan-a hilw-a/ a sweet orange but /burt?aan hilw/ sweet
oranges. The opposite is also true, i.e. a noun may be masculine in the
singular but feminine in the plural, e.g. /?alam gidiid/ a new pen but
/?ilaam gidiid-a/ new pens.
At the end of this subsection, a word must be said about the
inflection of deverbal nouns for gender. Verbal nouns do not inflect for
gender in either variety. If a verbal noun form is inflected for gender, it
will be considered lexicalized as a simple noun. Nouns of place and time

151

as well as nouns of instrument are marked for gender in the singular. If

they are feminine, they carry the sign [-at > -a ~ -t] which is typical of
feminine nouns. For instance, /ma-xraj(-un) > ma-xrag/ an exit is
masculine, whereas /ma-mlak-at(-un) > ma-mlak-a/ a kingdom is
feminine and /mi-brad(-un) > ma-brad/ a file is masculine, but
/mi-rwah-at(-un) > ma-rwah-a/ a fan is feminine.
3.6. Inflection for Number:
In both varieties under discussion, the noun has three numbers:
singular, dual and plural. The singular is the unmarked form. The dual is
formed via suffixation. The plural is obtained either by suffixation or by
an internal change in the noun structure. Verbal nouns do not inflect for
number in either variety. If a verbal noun form is inflected for number, it
will be considered lexicalized as a simple noun. Nouns of place and time
as well as nouns of instrument are treated like primary nouns in the dual
and plural. However, the two varieties do not use the same markers for
number distinction as will be shown in the following subsections.
3.6.1. The Singular:
This is considered the unmarked number as stated above. All what
has been said in the previous sections applies to singular nouns. In the
following subsections an attempt will be made to summarize the
morphological devices used in the formation of the dual and the various
types plural in SA and EA.

3.6.2. The Dual:

In SA, the dual is formed by attaching to the singular a suffix
which may take one of two forms depending on the case of the noun in the
sentence: [-aa(ni)] in the nominative and [-ay(ni)] in the accusative and
genitive. The [-ni] part of the dual suffix is deleted before pronominal
suffixes as well as before the second noun in a construct phrase, as will be
shown later in this section. The vowel /-i/ of the SA dual suffixes is also
omitted by a regular rule which deletes final short vowels in the pausal
forms of nouns. Hence, the pausal forms of SA dual markers are [-aan]
and [-ayn] (See 1.4.1). EA, on the other hand, uses one form for the dual
suffix: [-een]. It appears that EA has maintained only SA [-ayn] which has
become [-een] by Monophthongization. It remains for future research to
discover why the accusative and genitive case ending of SA survives in
EA although it is usually the nominative that survives when paradigm

152

leveling takes place. The following sentences show how the word
/al-bint(-u)/ the girl is dualized in SA:
fi-1-madras-ah
(151) a. ?al-bint-aani
the-girl-Nom.du
at-the-school
The two girls are at school.
b. ra?ay-tu
1-bint-ayni
see.pf-1sg
the-girl-Acc.du
I saw the two girls at school.
c. sallam-tu
3ala 1-bint-ayni
greet.pf-1sg (prep) the-girl-Gen.du
I greeted the two girls at school.

fi-1-madras-ah
at-the-school
fi-1-madras-ah
at-the-school

That one suffix is used for dualization in EA can be observed by

comparing the Colloquial sentences in (152) with their Standard parallels
in (151) above, respectively:
(152) a. ?il-bint-een
fi-1-madras-a
the-girl-du
at-the-school
The two girls are at school.
b. uf-t
il-bint-een
see.pf-1sg
the-girl-du
I saw the two girls at school.
c. sallam-t
3a-1-bint-een
great.pf-1sg
(prep)-the-girl-du
I greeted the two girls at school.

fi-1-madras-a
at-the-school
fi-1-madras-a
at-the-school

As stated in the beginning of this section, the [-ni] part of the SA

dual marker is deleted when a dual noun represents the first element of
what is conventionally called a construct phrase and when it is succeeded
by a pronominal suffix. To exemplify, /qalam-aa-ni/ two pens has its
[-ni] deleted in /qalam-a 1-walad(i)/ the two pens of the boy (with /aa/
shortened by Closed-Syllable Shortening) and /qalam-aa-hu/ his two
pens. In EA, the construct state is used but without showing the
possessed-possessor relation. So, the counterparts of the examples above
are /?alam-een il-walad/ and /?alam-een-u/, respectively. An alternative in
EA is to separate the dual from the following noun or pronominal suffix
by the possessive particle /bituu3/ plural of /bitaa3/ belonging to and to
add the definite article [l-] to the possessed dual noun. Thus, for the
examples above, one finds /?il-?alam-in bituu3 il-walad/ (with /ee/
shortened into /i/ by Closed-Syllable Shortening) and /?il-?alam-een
bituu3-u/, respectively.

153

In SA, the allomorph [-at] of the feminine marker is used before

the dual marker. In the Colloquial variety, on the other hand, the
allomorph [-t] is employed. Examples are: /madras-at-ayn > madras-t-een/
two schools, /ward-at-ayn > ward-it-een/ two roses and /luga-t-ayn >
luga-t-een/ two languages.
As stated at the beginning of this section, nouns of place and time
as well as nouns of instrument are treated like primary nouns in relation to
the inflection for number. So, in the dual they have [-aan] in the
nominative and [-ayn] in the oblique in SA, while EA uses one marker
[-een] to form the dual. Examples of nouns of place and time in the dual
are:
Singular
ma-3mal(-un) a lab
ma-mlak-at(-un) a kingdom
ma-kaan(-un) a place

SA Dual
ma-3mal-aan
ma-3mal-ayn
ma-mlak-at-aan
ma-mlak-at-ayn
ma-kaan-aan
ma-kaan-ayn

EA Dual
ma-3mal-een
ma-mlak-t-een
ma-kan-een

Examples to illustrate the formation of the dual of nouns of instrument in

both varieties are:
Singular
mi-ftaah(-un) > mu-ftaah a key

SA Dual
mi-ftaah-aan
mi-ftaah-ayn
mi-naar(-un) > mu-naar a saw
mi-naar-aan
mi-naar-ayn
mi-rwah-at(-un) > ma-rwah-a a fan mi-rwahat-aan
mi-rwahat-ayn
mi-r?aat(-un)> mi-raaya a mirror mi-r?aat-aan
mi-r?aat-ayn

EA Dual
mu-ftah-een
mu-nar-een
ma-rwaht-een
mi-rayt-een

To conclude this subsection, one can say that EA has a general

tendency to simplify the process of dual formation. Because case
distinctions are lost in this variety, it adopts a single dual marker which
can be attached to any singular noun, in contrast to SA which uses two
different dual markers depending on the case of the singular form. The
Colloquial variety has another means of evading the morphological
process of dualization, particularly in the case of names of currencies and
weights and the requests for foods and drinks. This is by putting the word
/?itneen/ meaning two before singular nouns. Hence, one finds /?itneen

154

gineeh/ two pounds, /?itneen kiilu/ two kilos, /?itneen kabaab/ two
(dishes of) kebab and /?itneen aay/ two (cups of) tea.
3.6.3. The Plural:
Two types of plural, traditionally known as sound and broken,
exist in SA and EA. The first type, including masculine and feminine
plurals, is formed by attaching a suffix to the singular noun without
making any change in the noun itself, which is the reason of using the term
sound. The other type takes a number of patterns which involve an
internal change in the noun structure, triggering the use of the term
broken. In other words, sound plural is regular and predictable, being
formed through suffixation; while broken plural is irregular and
unpredictable, being formed through vowel pattern change. The question
that arises here is what criteria can help us to select a plural type for a
certain singular noun? To answer this question, we can postulate that a
noun has one of the following four features which are properties of the
lexeme rather than the singular form:
(153) [-at]
[broken]

[hum]
[masc]

We can also propose that:

(154) [-at] = [+-]
[masc] = [+fem]
Then, we can postulate that in the singular, a noun follows one of the
following feature-filling rules (Trask 1996: 141), i.e. which do not
change prespecified features:
(155) a. [+broken] [-at]
b. [+fem]
[+-at]
c. otherwise [-at]
In the plural, the following rules apply to supply plural extensions:
(156) a. [+broken] [+-], traditionally broken plural
+ hum

b. + masc [+-uuna], traditionally sound masculine plural

at

155

c. Otherwise [+-aat], traditionally sound feminine plural.

To apply the feature specification proposals stated above, consider the
examples in Table (46). In this table, the underlined feature values are
those which are unpredictable given the feature-filling rules postulated in
(155) above.

No
1
2
3
4
5
6
7
8
9
10

Table (46)
Examples of Plural Feature Specification in SA & EA
Noun
Sg. Feature
Pl. Feature
?arnab(-un) a rabbit
at, +broken
hum, +masc [+-] broken
rijl(-un) > rigl a leg
-at , +broken
hum, +fem [+-] broken
3umd-at(-un) a mayor
+-at, +broken
+hum, +masc [+-] broken
3aruus-at(-un) a bride
+-at, +broken
+hum, +fem [+-] broken
madras-at(-un) a school
+-at, +broken
hum, +fem [+-] broken
muhandis(-un) an engineer (m) -at, broken
+hum, +masc [+-uuna]
muhandis-at(-un) an engineer (f) +-at, broken
+hum, +fem [+-aat]
?umm(-un) a mother
-at, broken
+hum, +fem [+-aat]
xawaag-at(-un) a foreign man
+-at, broken
+hum, +masc [+-aat]
faaks(-un) > faks a fax
-at, broken
hum, +masc [+-aat]

Because the suffix [-aat] is the default plural marker, it has now
become the most popular marker for making the plural of loan words, such
as /tilifizyoon(-un) > tilifizyoon tilifizyoon-aat(-un) > tilifizyun-aat/
televisions and /faaks(-un) > faks faaks-aat(-un) > faks-aat/ faxes. A
semantic difference between sound masculine plural and sound feminine
plural is implied by the rules in (156) above. It has also been figured out
by Gary & Gamal-Eldin (1982: 73): All plural nouns ending in -iin are
[+human]. Plurals ending in -aat may be [+ or human]. This means that
the masculine sound plural applies to masculine human nouns, while the
feminine sound plural applies to human or non-human ones. This semantic

156

distinction holds true in both varieties. Thus, one finds /muhandis-uuna >
muhandis-iin/ engineers (mpl, +hum), but /muhandis-aat(-un) >
muhandis-aat/ engineers (fpl, +hum) and /tuffaah-aat(-un) > tuffah-aat/
apples (fpl, hum).
3.6.3.1. Sound Masculine Plural:
Two suffixes are employed in the formation of the sound
masculine plural in SA: [-uu(na)] in the nominative case and [-ii(na)] in
the accusative and genitive cases. The [-na] part of the plural suffix is
elided before pronominal suffixes and before the second noun in a
construct phrase as will be shown below. The final vowel is also deleted in
pausal forms (See 1.4.1). The following are illustrative SA examples:
fi-1-maSna3
(157) a. ?al-muhandis-uuna
the-engineer-Nom.mpl
at-the-factory
The engineers are at the factory.
b. ra?ay-tu
1-muhandis-iina
fi-1-maSna3
see.pf-1sg
the-engineer-Acc.mpl at-the-factory
I saw the engineers at the factory.
c. sallam-tu
3ala 1-muhandis-iina
fi-1-maSna3
greet.pf-1sg (prep) the-engineer-Gen.mpl at-the-factory
I greeted the engineers at the factory.
Turning to EA one finds that one suffix [-iin] is employed in the
formation of the sound masculine plural in all positions. In other words,
EA has generalized the non-nominative endings of SA in such a way that
they are now used with any singular noun. Again it is not yet known why
it is the oblique (non-nominative) case ending which has generalized in
EA. Compare the examples in (158), which are the Colloquial counterparts
of those in (157), respectively:
fi-1-maSna3
(158) a. ?il-muhandis-iin
the-engineer-mpl
at-the-factory
The engineers are at the factory.
b. uf-t
il-muhandis-iin
fi-1-maSna3
see.pf-1sg the-engineer-mpl
at-the-factory
I saw the engineers at the factory.
c. sallam-t
3a-1-muhandis-iin
greet.pf-1sg
(prep)-the engineer-mpl
I greeted the engineers at the factory.

fi-1-maSna3
at-the-factory

157

As stated in the beginning of this subsection, the [-na] final part of

the SA marker is deleted before the second noun of a construct phrase and
before a pronominal suffix. To illustrate, /muhandis-uuna/ engineers has
its [-na] omitted in /muhandis-u -arik-at(i)/ the engineers of the
company (with /uu/ shortened by Closed-Syllable Shortening) and
/muhandis-uu-ha/ its engineers. In EA, the construct state is expressed
without showing a morphological relation between the possessed and the
possessor. Hence, the counterparts of the phrases above are /muhandis-in19
i-irk-a and /muhandis-in-ha/. An alternative in the Colloquial variety is
that the possessed is separated from the possessor by the possessive
particle /bituu3/, plural of /bitaa3/ belonging to and the definite article is
prefixed to the possessed plural noun. Therefore, for the examples above
one finds /?il-muhandis-in bituu3 i-irk-a/ and /?il-muhandis-in bitu3-ha/,
respectively.
EA has a special sound plural marker which is not found in SA.
This is the suffix [-iyya] which is taken by a small number of professional
nouns ending with [-gi] or [-i] (Salib 1981: 171). Examples of the use of
this suffix are:
EA Singular
makwa-gi a laundryman
busTa-gi a postman
sufra-gi a butler
kumsaar-i a ticket collector
haraam-i a thief

EA Plural
makwag-iyya
busTag-iyya
sufrag-iyya
kumsar-iyya
haram-iyya

3.6.3.2. Sound Feminine Plural:

The suffix used in the formation of the sound feminine plural of
nouns in the Standard variety is [-aat]. This is followed in the non-pausal
forms by the case endings: [-u(n)] in the nominative case and [-i(n)] in
both the accusative and the genitive cases. The [-n] part of the case
endings is deleted before pronominal suffixes and before the second noun
in a construct phase as will be discussed below. The following are
illustrative SA examples:
(159) a. daxal-at
il-muhandis-aat-u
1-maSna3
enter.pf-3fsg the-engineer-fpl-Nom
the-factory
The engineers (f) entered the factory.
b. ra?ay-tu
1-muhandis-aat-i
fi-1-maSna3
see.pf-1sg
he-engineer-fpl-Acc
at-the-factory
I saw the engineers (f) at the factory.

158

c. sallam-tu
3ala 1-muhandis-aat-i
greet.pf-1sg (prep) the-engineer-fpl-Gen
I greeted the engineers (f) at the factory.

fi-1-maSna3
at-the-factory

In the pausal forms of SA nouns, the case endings represented in

the final vowels [-u] and [-i] following the feminine plural marker are
deleted, as was discussed in (1.4.1). The final [-n] of the sound feminine
plural case suffix in SA is used for indefinite nouns as was indicated in
(1.4.2).
On the other hand, EA employs the same suffix [-aat] to form the
sound feminine plural in all positions, but without any case ending, as in
the examples in (160) which are the Colloquial equivalents of those in
(159), respectively:
(160) a. daxal-it
il-muhandis-aat
enter.pf-3fsg the-engineer-fpl
The engineers (f) entered the factory.
b. uf-t
il-muhandis-aat
see.pf-1sg
the-engineer-fpl
I saw the engineers (f) at the factory.
c. sallam-t
3a-1-muhandis-aat
greet.pf-1sg (prep)-the-engineer-fpl
I greeted the engineers (f) at the factory.

il-maSna3
at-the-factory
fi-1-maSna3
at-the-factory
fi-1-maSna3
at-the-factory

As stated above, the [-n] of the SA case ending in sound feminine

plurals is omitted before the second element of a construct phrase and
before a pronominal suffix. For example, /muhandis-aat(-un)/ engineers
(f) has its [-n] omitted in /muhandis-aat-u miSr(-a)/ the engineers (f) of
Egypt and /muhandis-aat-u-ha/ its engineers (f). In EA, the construct
state is achieved without showing a morphological relation between the
possessed and the possessor. Thus, the equivalents of the phrases above
are: /muhandis-at19 maSr/ and /muhandis-at-ha/. An alternative in EA is
the use of /bituu9/ as a possessive particle: /?il-muhandis-at bituu3 maSr/
and /il-muhandis-at bitu3-ha/.
The use of [-aat] as a sound feminine suffix is basically identical in
SA and EA. However, the Standard variety has a somewhat complicated
system of forming the sound feminine plural of certain singular nouns, a
large number of which are of the pattern [FV3L-at(-un)]. In such nouns,
the formation of the plural often affects the original structure of the noun
through the infixation of a short vowel, identical to the first vowel, after
the second consonant, before attaching the suffix [-aat]. Hence, the plural

159

pattern will be [FV3VL-aat(-un)]. No account has yet been given for this
interesting alternation between [FV3L-] and [FV3VL-]. To illustrate,
/Darb-at(-un)/ a strike is pluralized as /Darab-aat(-un)/; /Zulm-at(-un)/
thick darkness as /Zulum-aat(-un)/ and /da3w-at(-un)/ an invitation as
/da3aw-aat(-un)/. The Colloquial variety, on the other hand, follows a
simplified pattern of sound feminine plural, by merely adding [-aat] to the
singular noun without any change in the structure of the noun. Examples
are /Darb-aat/ strikes and /da3w-aat/ invitations.
3.6.3.3. Broken Plural:
In contrast to the formation of the sound plural, the broken plural
involves an internal modification in the structure of the singular noun, as
has been pointed out above. The question that arises here is: To what
degree can we predict which broken plural type will be generated given a
particular singular template? The answer is that the degree of prediction is
very low as one singular pattern may have up to seven different plural
patterns. The diversity of the plural patterns can be seen by examining the
various possible patterns for each singular pattern, as in the Appendix.
That appendix also shows that there are no less than thirty different
patterns of Standard broken plural, most of which are also observed in EA
with a few modifications. However, some of those patterns are rarely used,
while others enjoy a high degree of frequency whether in SA or EA.
In order to solve the problem of diversity in the broken plural
patterns, I will give a prosodic analysis of those patterns in terms of
Prosodic Morphology. McCarthy & Prince (1990b: 213 ff.) divided
Wrights (1967) thirty-one patterns of broken plural into the four
categories in (161). They also proposed that templates with stem-initial
[?aCC-] have underlying /CaC-/ but they involve a [Ca-] metathesis.
Those templates are put between /.../ in (161) above. For instance,
[?aCCVVC] is originally /CVCVVC/. The patterns are classified
according to their prosodic structure: the forms in (161b) are all [CVCVC]
which is the typical quantitative trochee; [CVCC] is the only productive
monosyllabic canon in Arabic.
(161) McCarthy & Princes Categorization of Br. Pl. Patterns:
a. Iambic
CiCaaC
CuCuuC
CaCaaC
/CaCaaC/

b. Trochaic
CuCaC
CiCaC
CaCaC
CiCaC+at

c. Monosyllabic
CuCC
CiCC+at
CiCC+aan
CuCC+aan

d. Other
CuCCaC
CuCCaaC

160

CaCaaC+/ay/
CaCiiC
CuCuuC+at
CiCaaC+at
CawaaCiC
CaCaa?iC
CaCaaCiC
CaCaaCiiC

/CaCuC/
CaCC+/ay/
CuCuC
CaCC
CaCaC+at
CuCaC+at
CuCaC+aa?
/CaCiC/+at
/CaCiC/+aa?

The iambic pattern (161a) applies to five singular stem forms

[CVCC], [CVCVC], [CVCVVC+at], [CVVCVC+at] and [CVVCV(V)C].
The plural patterns of these stem forms have one invariant which is the
initial iambic sequence [CVCVV+]. This sequence is applied to the first
two moras of the singular stem form. Thus, a monosyllable like /ahr(-un)/
a month has /ah/ as the first two moras. Mapping them onto the iambic
template, while replacing the vowel melody with /u/ produces /uhuu/. In
the context of the base form, this gives /uhuur(-un)/. In a bimoraic
disyllable like /haram(-un)/ a pyramid, the first two moras are /hara/. The
mapping goes on as with /ahr(-un)/, yielding /haraa/ to satisfy the
template. The complete form in context will be /haraam/. Then, a
metathesis rule entails /?ahraam(-un)/.
For nouns of triconsonantal singulars with a long vowel, i.e. those
of the singular stem forms [CVCVVC+at] and [CVVCVC+at], a default
consonant /w/ is required. The position of such a consonant is determined
by the position of the long vowel. It is inserted in the syllable following
that of the long vowel. In other words, it is inserted in the second syllable
if the long vowel is in the first syllable and in the third syllable if the long
vowel is in the second syllable. The default consonant is realized as /?/
under certain phonological conditions. For instance, in /faakih+at(-un)/
fruit, the first two moras are /faa/. Applying the iambic template with /w/
insertion results in /fawaa/, then adding the residue of the singular stem
form leads to /fawaakih(-u)/. In /3agiib+at(-un)/ a wonder, the first two
moras are /3agi/. Applying the iambic template and inserting /w/ leads to
/3agaaw/. Then, adding the residue produces /3agaawib/ which has the
surface /3agaa?ib(-u)/ in SA, which is in turn changed to /3agaayib/ by
Intervocalic /?/ Assimilation in EA.
For a quadriradical stem form with a short vowel in the last
syllable like /bulbul(-un)/ a nightingale, the first two moras are /bul/. Of
the plural vowel melody /a-i/, the /a/ spreads over the template, resulting in
/balaa/. The residue of the word picks up the /i/ giving /bil/. Taken
together, they form /balaabil(-u)/. For a quadriradical stem form with a

161

long vowel in the last syllable like /finjaan(-un)/ a cup, the

moras are /fin/. Proceeding like /bulbul(-un)/, this templatic
emerges as /fanaa/ and the residue becomes /jiin/ absorbing the
plural melody /a-i/ to produce /fanaajiin(-u)/ in SA which
/fanagiin/ in EA by Atonic Shortening.

first two
segment
/i/ of the
becomes

The trochaic pattern (161b) is found in three lexical classes. First,

feminines [CiCC+at] and [CuCC+at] form plurals [CiCaC] and [CuCaC],
respectively. Examples: /3ilb+at(-un)/ a packet /3ilab(-un)/ and
/dawl+at(-un)/ a country /duwal(-un)/. Second, unsuffixed forms
[CVCVVC] with five distinct vocalizations take the trochaic pattern. The
choice among the various vocalizations depends on semantic grounds. For
example, /Tariiq(-un)/ a road Turuq(-un)/, /ahiid(-un)/ a martyr
/uhad+aa?(-u)/ and /silaah(-un)/ a weapon, /?aslih+at(-un)/ from
/salih+at(-un)/ by metathesis. Third, the lexicalized participles take the
trochaic plural [CaCaC+at or CuCaC+at] according to the phonology of
the root. Examples: /saahir(-un)/ a magician /sahar+at(-un)/ and
/qaaDiy(-un)/ whose surface form is /qaaDin/ a judge /quDay+at(-un)/
which becomes /quDaat(-un)/ by Glide Elision.
The monosyllabic plural pattern (161c) is usually suffixed with
[+at, +ay or +aan]. Its frequency is very low, indicating non-productivity
(McCarthy & Prince 1990b). The only common non-suffixed pattern
[CuCC] is limited to adjectives of color and physical defect. The singular
of these adjectives usually has the form [CaCaC] (surface [?aCCaC]). So,
the plural can be formed by imposing the /u/ vocalism on the stem form.
Examples are: /?aSfar(-u) yellow /Sufr(-un)/ and /?a3raj(-u)/ lame
/3urj(-un)/.
The other templates (161d), i.e. [CuCCaC and [CuCCaaC], arise
from lexicalized active participles [CaaCiC]. The medial consonant of the
singular is spread backwards to close the first syllable, usurping its second
mora. The spreading rule changes [CVVCVC] to [CVCCVC]. The vowel
melody of the singular is replaced by /u-a/, with the vowel of the second
syllable often lengthened. Example: /haakim(-un)/ a ruler
/hukkaam(-un)/.
What is important for the purposes of comparison is to show which
SA broken plural patterns are preserved the same in EA and which ones
have undergone phonological changes. Table (47) shows the patterns of
broken plural identical in SA and EA, along with examples that can be
found in the two varieties. Table (48) shows the SA broken plural patterns

162

which have two EA counterparts; one similar to that of SA and one that
has undergone a phonological change. It is not yet known whether these
changes are regular or accidental. Table (49) displays the phonological
changes that some SA broken plural patterns have undergone in EA.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

Table (47)
Broken Plural Patterns Identical in SA &EA
Pattern
Example
Gloss
Fu3uL(-un)
sufun(-un) > sufun
ships
Fu3aL(-un)
guraf(-un) > guraf
rooms
Fi3aL(-un)
minah(-un) > minah
grants
Fi3aL-at(-un)
dibab-at(-un) > dibab-a
bears
Fa3aL(-un)
ajar(-un) > agar
trees
Fa3aL-at(-un)
sahar-at(-un) > sahar-a
magicians
Fu3uuL(-un)
?usuud(-un) > ?usuud
lions
Fi3aaL(-un)
hibaal(-un) > hibaal
ropes
Fi3aaL-at(-un) hijaar-at(-un) > higaar-a stones
Fu3aat(-un)
quDaat(-un) > quDaah
judges
Fu3Laan(-un)
fursaan(-un) > fursaan
horsemen
Fu33aaL(-un)
hurraas(-un) > hurraas
guards
taFaa3iL(-u)
tajaarib(-u) > tagaarib
experiments
maFaa3iL(-u)
madaaris(-u) > madaaris schools
aF3uL(-un)
asTur(-un) > asTur
lines
aFaa3iL(-u)
akaabir(-u) > akaabir
great men
Fawaa3iL(-u)
3awaaSif(-u) > 3awaaSif storms
Fa3aaL1iL2(-u) sanaabil(-u) > sanaabil
ears of corn

Table (47) reveals that the broken plural patterns which are identical in
both varieties follow one of these templates: [CVCVC(a)], [CVCVVC(a)],
[CVCVVt], [CVCCaan], [CVCCVVC], [taCVVCVC], [maCVVCVC],
[aCCVC], [aCVVCVC] and [CVCVVCVC].
Table (48)
SA Broken Plural Patterns Having Two EA Counterparts
Pattern
Example
SA
EA
SA
EA
Gloss
1 Fa3iiL(-un)
Fa3iiL 3abiid(-un)
3abiid
slaves
"
Fi3iiL
hamiir(-un)
himiir
donkeys
2 ?aF3aaL(-un) ?aF3aaL ?aShaab(-un) ?aShaab
friends
"
?iF3aaL ?amaaT(-un) (?i)maaT combs
3 ?aF3iL-at(-un) ?aF3iL-a ?a3mid-at(-un) ?a3mid-a columns
"
?iF3iL-a ?argif-at(-un) (?i)rgif-a loaves

163

Table (48) shows that the SA broken plural patterns which have two EA
counterparts follow one of three templates: [CVCVVC], [?VCCVVC] and
[?VCCVC-at(-un)]. The existence of two EA counterparts for the SA
pattern [Fa3iiL(-un)] can be explained by the process of lexical diffusion
stated above for singular stem forms (3.1.1). The existence of
[(?i)F3aaL(-un)] as an alternate of [?aF3aaL(-un)] and of [(?i)F3iL-a] as a
variant of [?aF3iL-at(-un)] can be attributed to the loss of [?a-] in EA,
followed by the epenthesis of /i/ by Word-Initial Epenthesis and the
insertion of /?/ by Glottal Stop Insertion as follows:
(162) a. (SA) ?aCCVVC CCVVC iCCVVC ?iCCVVC (EA).
b. (SA) ?aCCVC-at CCVC-a iCCVC-a ?iCCVC-a (EA).

1
2
3
4
5
6
7
8
9
10
11
12
13
14

Table (49)
SA Broken Plural Patterns Phonologically Changed in EA
Pattern
Example
SA
EA
SA
EA
Gloss
Fa3Laa
Fa3La
marDaa
marDa
patients
Fa3aaLaa
Fa3aaLa Sahaaraa
Sahaara deserts
Fa3aaLii
Fa3aaLi karaasii
karaasi
chairs
?aFaa3iiL(-u)
?aFa3iiL ?asaaTiir(-u)
?asaTiir myths
Fawaa3iiL(-u) Fawa3iiL tawaabiit(-u)
tawabiit coffins
maFaa3iiL(-u) maFa3iiL mafaatiih(-u) mafatiih keys
Fa3aaL1iiL2(-u) Fa3aL1iiL2 3aSaafiir(-u)
3aSafiir sparrows
FiiLaan(-un)
FiLaan
niiraan(-un)
niraan
fires
Fu3aLaa?(-u)
Fu3aLa uhadaa?(-u)
uhada
martyrs
?aF3iLaa?(-u) ?aF3iLa ?anbiyaa?(-u) ?anbiya prophets
Fa3aa?iL(-u)
Fa3aayiL 3ajaa?ib(-u)
3agaayib wonders
Fa3aa?iL-at(-un) Fa3ayL-a malaa?ik-at(-un) malayk-a angels
?aFaa3iL-at(-un) ?aFa3L-a ?asaati-at(-un) ?asatz-a professors
Fu?uuL(-un)
FuuL
ru?uus(-un)
ruus
heads

Table (49) illustrates that the SA broken plural patterns which are
phonologically changed in EA have these templates: [CVCCaa],
[CVCVVCaa], [?aCVVC], [CVVC], [CVCVVCVVC], [maCVCVVC],
[CVVCaan], [CVCVCaa?], [?aCVCaa?], [CVCVV?VC(a)], and
[CV?VVC]. The phonological rules used to explain changes in the
singular nominal stem forms could also be used to explain the changes
occurring in the broken plural patterns. Patterns (1-3) in Table (49)
undergo Final-Vowel Shortening. Patterns (4-8) undergo Atonic
Shortening. Patterns (9-10) undergo Final /?/ Deletion, then Final-Vowel

164

Shortening. Pattern (11) [Fa3aa?il(-u) > Fa3aayiL] undergoes Intervocalic

/?/ Assimilation. Pattern (12) [Fa3aa?iL-at(-un) > Fa3ayL-a] undergoes
three phonological changes in addition to the deletion of final /t/, which is
typically morphological: the change of /?/ to /y/ by Intervocalic /?/
Assimilation, the elision of /i/ by High-Vowel Deletion, and the shortening
of the non-final long vowel by Closed-Syllable Shortening. So, the
changes occurring in that pattern are:
(163) a. Fa3aa?iL-at(-un) Fa3aa?iL-a Fa3aayiL-a
Fa3aayL-a Fa3ayL-a
b. malaa?ik-at(-un) malaa?ik-a malaayik-a
malaayk-a malayk-a
Pattern (13), [Fu?uul > FuuL] undergoes two phonological processes: the
elision of /?/ and the shortening of an extra-long vowel. Hence, the
changes occurring in that pattern are:
(164) a. Fu?uuL FuwuuL FuuL
b. ru?uus ruuus ruus
Of the patterns that are not mentioned in Tables (47-49), some are
specifically Standard, e.g. [Fu33aL(-un)] as in /rukka3(-un)/ kneeling
(pl), [Fu3aaLaa] as in /?usaaraa/ captives and [Fayaa3iL(-u)] as in
/qayaaSir(-u)/. One pattern is specifically Colloquial, namely [Fi33aaLa],
in one example: /riggaala/ men.
Both varieties under investigation have a category of certain nouns
which express plurality by means of referring to the species of objects
without being considered plural in the technical sense. These are termed
collective nouns by Al-Toma (1969: 44). He further divides such nouns
into two classes: /?ism-u 1-jins/ the noun of species and /?ismu l-jam3/
the noun of plural. The former can be singularized by the use of [-a(t)],
whereas the latter rejects [-a(t)] as a means of singular formation. To
exemplify, one could say that /nahl(-un)/ bees and /naxl(-un)/ palm
trees are nouns of species because they have the singulars /nahl-at(-un)/
and /naxl-at(-un)/, respectively; but this cannot be applied to /qawm(-un)/
a people and /jay(-un)/ an army which are nouns of plural.
There are various patterns for collective nouns of species in
Standard and Egyptian Arabic. They may refer to things or animals, as in
the following examples:

165

Pattern
Fa3L(-un)
Fa3aL(-un)
Fa3aaL(-un)
Fu33aaL(-un)
Fa33iil(-un)
Fi3L1iL2(-un)
Fa3L1iL2(-un)

Collective Noun
ward(-un) roses
balah(-un) dates
hamaam(-un) pigeons
tuffaah(-un) apples
baTTiix(-un) water melon
filfil(-un) pepper
narjis(-un) narcissus

Singular
ward-at(-un)
balah-at(-un)
hamaam-at(-un)
tuffaah-at(-un)
baTTiix-at(-un)
filfil-at(-un)
narjis-at(-un)

A further distinction has been drawn by Gary & Gamal-Eldin

(1982: 73) between a collective and a distributive plural. The former is
general in scope because it refers to the class of items as a whole, while
the latter is specific or countable and can be used after a numeral from
three to ten. In fact, the collective nouns can be understood as syntactically
and morphologically singular, but they have plurality only as part of their
semantics. The distributive plurals, on the other hand, serve as
morphological and syntactic plurals, that is they behave regularly. This
applies in both varieties under discussion. To observe this distinction, one
can compare the SA sentence in (165a) with that in (165b) and the EA
counterparts in (166a) and (166b), respectively:
faakih-at-un
(165) a. ?at-tuffaah-u
the-apples-Nom
(a) fruit-Nom
Apples are a delicious fruit.
b. ?akal-tu
xams-a
eat.pf-1sg
five-Acc
I ate five apples.
(166) a. ?it-tuffaah
fakh-a
the-apples
(a) fruit
Apples are a delicious fruit.
b. kal-t
xamas
eat.pf-1sg
five
I ate five apples.

laii-ah
delicious
tuffaah-aat
apples-fpl

laziiz-a
delicious
tuffah-aat
apples-fpl

In the examples above, /tuffaah(-un)/ is a collective plural for it refers to a

kind of fruit; but /tuffaah-aat(-un)/ is a distributive plural since it refers to
a number of this fruit. Some other examples that can be found in both
varieties are given below:
Singular
ward-at(-un) a rose
baSal-at(-un) an onion

Collective Plural
ward(-un)
baSal(-un)

Distributive plural
ward-aat(-un)
baSal-aat(-un)

166

jazar-at(-un) a carrot
mimi-at(-un) an apricot

jazar(-un)
mimi(-un)

jazar-aat(-un)
mimi-aat(-un)

3.6.3.4. Pluralization of Nouns of Place/Time & Instrument:

At the end of the discussion of the plural, a word must be said
about the plural formation of nouns of place/time and nouns of instrument.
First, a distinction is maintained between nouns of place/time taken from a
primary verb and those taken from derived verbs. For the former, the
broken plural - often of the form [ma-Faa3iL(-u)] - is used for masculine
or feminine nouns. For the latter, only the [-aat] default plural marker is
employed. This distinction applies in both varieties:
SA & EA Plural
Singular
ma-dxal(-un) an entrance
ma-daaxil(-u)
ma-nzil(-un) a house
ma-naazil(-u)
ma-hkam-at(-un) a court
ma-haakim(-u)
mu-?tamar(-un) a conference
mu-?tamar-aat(-un)
mu-jtama3(-un) a society
mu-jtama3-aat(-un)
mu-stafa(-n) a hospital
mu-stafay-aat(-un)
An exception here is the plural noun of place /ma-ktab-aat(-un)/ libraries
which has the [-aat] plural though its singular [ma-ktab-at(-un)] is derived
from a primary verb /katab(-a)/ to write. The reason behind this is that
the broken plural pattern [ma-Faa3iL(-u)] is already taken by the plural of
the other noun of place derived from the same verb, i.e. /ma-ktab(-un)/ an
office which has the plural form /ma-kaatib(-u)/. The same argument
holds true for /ma-nzil(-un)/ a house vs. /ma-nzil-at(-un)/ a
position/rank and /ma-kaan(-un)/ a place vs. /ma-kaan-at(-un)/ a
position/rank.
To form the plural of nouns of instrument, three plural patterns
(two of which are broken) are used in both varieties, depending on the
pattern of the singular: [ma-Faa3iL(-u)] if the singular pattern is
[mi-F3aL-at(-un) > ma-F3aL-a] and [ma-Faa3iiL(-u) > ma-Fa3iiL] if the
singular pattern is [mi-F3aaL (-un)] (cf. Thackston (1984: 167). The third
pattern will be mentioned below. One can now consider the following
examples of the first two patterns:
Singular
mi-dfa3(-un) > ma-dfa3 a cannon
mi-sTar-at(-un)> ma-sTar-a a ruler
mi-hraa(-un) > mi-hraat a plough
mi-Sbaah(-un) a lamp

SA Plural
ma-daafi3(-u)
ma-saaTir(-u)
ma-haarii(-u)
ma-Saabiih(-u)

EA Plural
ma-daafi3
ma-saaTir
ma-hariit
ma-Sabiih

167

An exception here is the noun of instrument /mi-qaSS(-un) > ma-?aSS/ a

pair of scissors which has the default plural marker [-aat] instead of the
broken plural, i.e. /mi-qaSS-aat(-un) > ma-?aSS-aat/.
The third plural pattern which employs the default plural marker
[-aat] is used for nouns of instrument which have the singular pattern
[Fa33aaL-at(-un)] in both varieties, e.g. /gassaal-at(-un)/ a washingmachine /gassaal-aat(-un) > gassal-aat/ and /allaaj-at(-un) > tallaag-a/
a refrigerator /allaaj-aat(-un) > tallag-aat/.
3.7. Nouns & Suffixation:
Both primary and deverbal nouns receive possessive suffixes,
which are identical in form to the objectival suffixes listed in Table (34).
Morphophonemic alternations may result from this suffixation in EA.
Examples of the suffixation to primary nouns are:
SA
baab-ii
qamiiS-ak
a3r-u-ki
?ustaa-u-hu

EA
baab-i
?amiiS-ak
a3r-ik
?ustaaz-u

Gloss
my door
your (msg) shirt
your (fsg) hair
his teacher

Examples of the suffixation to verbal nouns are:

SA
kitaabat-ii
tahammul-ak
(?i)ntiZaar-u-ki
tafkiir-u-hu

EA
kitabt-i
tahammul-ak
(?i)ntiZaar-ik
tafkiir-u

Gloss
my writing
your (msg) endurance
your (fsg) waiting
his thinking

Examples of the suffixation to nouns of place/time and nouns of

instrument are:
SA
ma-3mal-u-naa
ma-Sna3-u-hum
mi-ftaah-ii
allaaj-at-ak

EA
ma-3mal-na
ma-Sna3-hum
mu-ftaah-i
tallag-t-ak

3.8. The Diminutive Form:

Gloss
our laboratory
their factory
my key
your (msg) refrigerator

168

The diminutive is a nominal that is derived from a noun to give the

general meaning of a little or small instance of that noun. The general
rule for forming this nominal in SA is inserting the infix [-ay-] after the
second consonant and replacing the first vowel of the noun by /u/ if it is
not originally /u/ (cf. Al-Toma 1969: 46). All diminutive forms, according
to Fayez (1990: 27-30), must correspond to one of three basic patterns:
[Fu3ayL(-un)], [Fu3ay3iL(-un)] and [Fu3ay3iiL(-un)]. The first occurs
with triradical nouns, the second with quadriradical nouns and the last with
quadriradical nouns including long vowels before the last radical.
According to McCarthy & Prince (1990b), what is noticeable about all the
diminutive forms is that they follow the iambic template in their stems.
McCarthy & Prince (1990b: 246) also explain why the vocalization of the
templatic syllables in all of them is [u--ay]: an autosegmental suffix y fills
the last mora of the template. The diminutive melody /u-a-i/ then
associates one-to-one and left-to-right, as usual, leaving the /i/
unassociated. ... the /i/ of the melody associates to the residue whenever
possible. It can be added here that when the melody /i/ links to
quadriradicals with final bimoraic syllables, it occupies both vocalic
moras, i.e. it is lengthened. Below are some examples of the diminutive in
SA:
Noun
qalam(-un) a pen
hirr-at(-un) a cat
?arnab(-un) a rabbit
miftaah(-un) a key

SA diminutive
qulaym(-un) a small pen
hurayr-at(-un) a small cat
?uraynib(-un) a small rabbit
mufaytiih(-un) a small key

The Colloquial variety does not use any of the SA diminutive

patterns stated above. However, it uses another means of diminutive
formation of a type that might be called the hypocoristic diminutives, i.e.
nicknames, because it is generally used to address children fondly. It
involves geminating the second consonant, infixing [-uu-] before the final
consonant and suffixing [-a] after the final consonant. The vowel /uu/ is
infixed regardless of the type of vowel in the original noun. The
suffixation of [-a] is optional with masculine names and obligatory with
feminine names.
Noun
xaalid a boys name
Taari? a boys name
ams a girls name
?amiira a girls name

EA diminutive
xalluud(-a)
Tarruu?(-a)
ammuus-a
?ammuur-a

169

Chapter Four
Adjectival Morphology
4.0. Introduction:
The distinction between the class of nouns and that of adjectives is
controversial since the majority of the members of the first class may be
used as members of the second, and vice versa. One attempt to solve this
problem was made by Gamal-Eldin (1967: 20). He set up three classes: a
Noun class, an Adjective class and an Adjective-Noun class. Another
attempt to settle the controversy is to take the inflection for degree as a
parameter to distinguish adjectives from nouns: while adjectives are
inflected for degree, (either morphologically or syntactically) nouns are
not.
The morphology of adjectives in SA and EA is dealt with in this
chapter. Adjectival stem forms are presented in (4.1). The difference
between definite and indefinite adjectives is shown in (4.2). Then, the
inflection of adjectives for case, gender and number is indicated in (4.3),
(4.4) and (4.5), respectively. In addition, a treatment of the inflection for
degree is included in (4.6). Finally, the derivation of participial forms and
relational adjectives is discussed in (4.7) and (4.8), respectively.
4.1. Adjectival Stem Forms:
Numerous Standard adjectival stem forms are retained in the
Colloquial variety without any change, as shown in Table (50), and some
of these stem forms have undergone phonological changes, as indicated in
Table (51). Some of these changes are regular and a few are irregular as
will be shown below. Some SA adjectival stem forms have more than one
EA equivalent; one being similar to that of SA and the others having
undergone phonological changes, as displayed in Table (52). Three
adjectival stem forms are specifically Colloquial: [Fa33iiL], as in /hassiib/
proficient in book-keeping, [Fa33uuL], as in /dalluu3/ spoiled or
/habbuub/ much-loved, and [Fa3aL], e.g. /ala?/ rude.
Table (50) shows that adjectival stem forms which are identical in
SA and EA have one of six templates: [CVCC], [CVCVC], [CVCVVC],
[CVCCVVC], [CVCCaan], or [?aCCVC]. The last template is also used in
comparative forms (See 4.6.2).
Table (50)
Adjectival Stem Forms Identical in SA & EA

170

1
2
3
4
5
6
7
8
9
10
11
12

1
2
3
4
5
6
7
8
9
10
11
12
13
14

Stem Form
Fa3L(-un)
Fayy(-un)
Fu33(-un)
Fa3aL(-un)
Fa3aaL(-un)
Fa3uuL(-un)
Fu3aaL(-un)
Fa33aaL(-un)
Fi33iiL(-un)
Fu33uuL(-un)
Fa3Laan(-un)
?aF3aL(-u)

Example
sahl(-un) > sahl
hayy(-un) > hayy
hurr(-un) > hurr
baTal(-un) > baTal
jabaan(-un) > gabaan
Sabuur(-un) > Sabuur
ujaa3(-un) > ugaa3
Tammaa3(-un) > Tammaa3
sikkiir(-un) > sikkiir
qudduus(-un) > qudduus
farhaan(-un) > farhaan
?abyaD(-u) > ?abyaD

Gloss
easy
living
free
courageous
coward
patient
brave
greedy
alcoholic
holy
happy
white

Table (51)
SA Adjectival Stem Forms Phonologically Changed in EA
Stem Form
Example
SA
EA
SA
EA
Gloss
Fa3Laa?(-u) Fa3La
hamraa?(-u) hamra
red (f)
FayLaa?(-u) FeeLa
bayDaa?(-u) beeDa
white (f)
FawLaa?(-u) FooLa
sawdaa?(-u) sooda
black (f)
Faa?iL(-un)
Faayil
gaa?ib(-un)
gaayib
absent
Fa3?aan(-u)
Fa3yaan mal?aan(-u)
malyaan
full
Faa3i?(-un)
Faa3i
haadi?(-un)
haadi
calm
Fa3iyy(-un)
Fa3i
ganiyy(-un)
gani
rich
Faa3iL-at(-un) Fa3L-a maahir-at(-un) mahr-a
clever (f)
?aF3aa
?aF3a
?a3maa
?a3ma
blind
Faa3ii
Faa3i
baaqii
baa?i
remaining
FawLaan(-u) FaLaan jaw3aan(-u)
ga3aan
hungry
Fa3iL(-un)
Fi3iL
natin(-un)
nitin
stinking
Fu3w(-un)
Fi3w
hulw(-un)
hilw
sweet
Fu3yaan(-un) Fi3yaan 3uryaan(-un) 3iryaan
naked

As in nominal stem forms, a morphological rule is responsible for the loss

of /t/ at the end of EA adjectives ending in the suffix [-at] in SA. This is
clear in the adjectival stem form number (8). Table (51) shows that the
phonological alternations that might happen in adjectival stem forms are
governed by phonological rules similar to those governing phonological
changes in nominal stem forms, as shown in (167):
(167) (i) Monophthongization applies to the adjectival stem forms (2-3).
These are mostly adjectives of colors and bodily defects.

171

(ii) Intervocalic /?/ Assimilation is found in adjectival stem forms

(4-5).
(iii) Final /?/ Deletion is illustrated in adjectival stem forms (1-3
and 6).
(iv) Final-Vowel Shortening occurs in adjectival stem forms (1-3
and 9-10). It also applies to stem form (7) since /iy/ equals /ii/
in phonology.
(v) It cannot be explained why the changes in stem forms (11-14)
occurred. So, one can assume that those changes are
phonologically irregular.
Some SA adjectival stem forms undergo two phonological changes
in EA. These comprise stem forms (2-3). The adjectival stem form
[Faa3iL-at(-un)] follows the same changes occurring in its counterpart
nominal stem form (See 3.1.1): /t/ is lost in the suffix [-at]; /i/ is lost by
medial syncope, and the shortening of /a/ results from the ban on medial
superheavy syllables. Other examples are:
SA
maalih-at(-un)
Saabir-at(-un)
naajih-at(-un)
baarid-at(-un)
saabi3-at(-un)

2
3
4

EA
malh-a
Sabr-a
nagh-a
bard-a
sab3-a

Gloss
salty (f)
patient (f)
successful (f)
cold (f)
seventh (f)

Table (52)
SA Adjectival Stem Forms with More than One EA Equivalent
Stem Form
Example
SA
EA
SA
EA
Gloss
Fa3iiL(-un) Fa3iiL
rahiim(-un) rahiim
merciful
"
Fi3iiL
kabiir(-un) kibiir
big/old
"
Fu3ayyaL qaSiir(-un) ?uSayyar short
"
Fu3ayyiL qaliil(-un)
?ulayyil
few/little
Faa3iL(-un) Faa3iL
jaamid(-un) gaamid
solid
"
Fu3L
saaxin(-un) suxn
hot
Faa33(-un) Faa33
jaaff(-un)
gaaff
dry
"
Fa33
haarr(-un)
harr
hot
FayyiL(-un) FayyiL
jayyid(-un) gayyid
good
"
FayyaL Dayyiq(-un) dayya?
narrow

Table (52) shows that the SA adjectival stem form [Fa3iiL(-un)] is

retained in some Colloquial adjectives and phonologically changed in

172

some others. However, there are no definite rules concerning the

preservation of this stem form or its being changed into other stem forms.
But roughly speaking, the stem form [Fa3iiL(-un)] is preserved in
adjectives beginning with a guttural consonant (See 3.1.1) as in:
SA
xafiif(-un)
gariib(-un)
hadii(-un)
3ariiD(-un)
?amiin(-un)

EA
xafiif
gariib
hadiis
3ariiD
?amiin

Gloss
light
strange
modern
broad
honest

However, not all adjectives of the [Fa3iiL(-un)] pattern that are retained in
EA have initial gutturals, as in these adjectives:
SA
jamiil(-un)
kariim(-un)
naiiT(-un)
rahiim(-un)
sarii3(-un)
basiiT(-un)
abiih(-un)
faSiih(-un)
laTiif(-un)
Sahiih(-un)

EA
gamiil
kariim
naiiT
rahiim
sarii3
basiiT
abiih
faSiih
laTiif
Sahiih

Gloss
beautiful
generous
active
merciful
quick
simple
similar
eloquent
kind
correct

On the other hand, the stem form [Fa3iiL(-un)] is changed to [Fi3iiL] in

the Colloquial variety in a significant number of adjectives, as in these
instances:
SA
EA
Gloss
jadiid(-un)
gidiid
new
kabiir(-un)
kibiir
big/old
naZiif(-un)
niDiif
clean
raxiiS(-un)
rixiiS
cheap
samiin(-un)
simiin
fat
ba3iid(-un)
bi3iid
far
adiid(-un)
idiid
strong
aqiil(-un)
ti?iil
heavy
The existence of the two EA equivalents for the SA pattern [Fa3iiL(-un)],
as indicated above, can be attributed to the process of lexical diffusion

173

(See 3.1.1). If the process of raising /a/ to /i/ is still going on, one might
expect that EA adjectives of [Fa3iiL] will be changed into [Fi3iiL] in the
future. If that process has stopped, those adjectives will remain the same.
In adjectives denoting paucity or smallness (cf. Robertson 1970: 192), the
Standard stem form [Fa3iiL(-un)] is changed to [Fu3ayyaL], and in very
few instances to [Fu3ayyiL] in EA, as in the following examples;
SA
rafii3(-un)
qaSiir(-un)
Sagiir(-un)
qaliil(-un)
qariib(-un)

EA
rufayya3
?uSayyar
Sugayyar
?ulayyil
?urayyib

Gloss
thin
short
small/young
few/little
near

An exception here is the word /nahiif(-un)/ thin which is preserved

without change in EA.
4.2. Definite vs. Indefinite Adjectives:
Arabic adjectives agree with the nouns they modify in relation to
definiteness. So, there are two types of adjectives: definite and indefinite.
The same device of defining a noun is used to define an adjective in SA.
That is the definite article [al-] the. It is prefixed to the adjective and
assimilated to any solar consonant at the beginning of that adjective (See
1.4.2). EA uses the shortened form of the same morphological marker, i.e.
[l-]. Then /i/ is inserted by Word-Initial Epenthesis in EA and /?/ is
inserted by Glottal Stop Insertion in both varieties. Examples of definite
adjectives in both varieties are given below:
SA
EA
Gloss
(?)al-baxiil(-u)
(?i)l-baxiil
the stingy/miser
(?)al-ganiyy(-u)
(?i)l-gani
the rich
(?)aT-Tawiil(-u)
(?i)T-Tawiil
the tall
(?)as-sarii3(-u)
(?i)s-sarii3
the fast/speedy
For the adjectives that begin with a glottal stop /?/ in SA that is no longer
underlying in EA, leaving the vowel /a/ at the beginning, neither WordInitial Epenthesis nor Glottal Stop Insertion applies to the definite article
in EA, as in these examples:
SA
(?)al-?ahmar(-u)
(?)al-?asmar(-u)
(?)al-?a3raj(-u)

EA
l-ahmar
l-asmar
l-a3rag

Gloss
the red
the black
the lame

174

(?)al-?a3maa

l-a3ma

the blind

That Word-Initial Epenthesis and Glottal Stop Insertion do not apply to

adjectives with initial /a/ in EA is emphasized after nouns ending in a
single consonant, as in these Colloquial phrases:
(168) a. ?ik-kitaab
the-book
The red book.
b. ?ir-raagil
the-man
The lame man.

1-ahmar
the-red
1-a3rag
the-lame

Adjectives also become definite when they modify nouns followed

by a determining complement (See 3.3). However, though nouns in this
case do not carry a definite article, adjectives must carry this marker. SA
examples are: /kitaab-u ?ahmad-a 1-?ahmar/ Ahmads red book,
/kitaab-u-hu 1-?axDar/ his green book. The Colloquial analogues of
these examples are /kitaab ?ahmad l-ahmar/ and /kitaab-u 1-axDar/,
respectively. It has been argued by Ritter (1988) that in the construct state
the possessed noun occupies the syntactic position of the definite article
and hence pre-empts it from appearing. An adjective modifying the
possessed noun remains in situ:
(169) [D (l-) [?ahmad-a
[NP kitaab-u
[D kitaab-ui [?ahmad-a
[NP ti
____________________

1-ahmar]]]
1-ahmar]]].

4.3. Inflection for Case:

SA adjectives inflect for three cases: nominative, accusative and
genitive. They follow the nouns they modify in their inflection for these
three cases. Consequently, an adjective will be in the nominative case if it
modifies a nominative noun, in the accusative case if it modifies an
accusative noun and in the genitive case if it modifies a genitive noun.
Like nouns, SA adjectives are grouped into two classes according
to their extent of declension: fully-declined adjectives and semi-declined
ones. They are also called triptotes and diptotes, respectively.
Adjectives in the first class take two sets of case endings. The first set,
employed for definite adjectives, involves nominative [-u], accusative [-a]

175

and genitive [-i]. The second set, employed for indefinite adjectives,
includes nominative [-un], accusative [-an] and genitive [-in].
The use of the case endings with the adjective /naiiT(-un)/ active
is displayed in the following SA examples:
(170) a. daxal-a
1-walad-u
enter.pf-3msg
the-boy-Nom
1-bayt
n-naiiT-u
the-active-Nom
the-house
The active boy entered the house.
b. ra?ay-tu
1-walad-a
see.pf-1sg
the-boy-Acc
n-naiiT-a
fi-1-bayt
the-active-Acc
in-the-house
I saw the active boy in the house.
c. marar-tu
bi-1-walad-i
pass.pf-1sg
by-the-boy-Gen
fi-1-bayt
n-naiiT-i
the-active-Gen
in-the-house
I passed by the active boy in the house.
(171) a. daxal-a
walad-un
naiiT-un
il-bayt
enter.pf-3msg
(a) boy-Nom active-Nom the-house
An active boy entered the house.
b. ra?ay-tu
walad-an
naiiT-an
fi-1-bayt
see.pf-1sg
(a) boy-Acc active-Acc
in-the-house
I saw an active boy in the house.
c. marar-tu
bi-walad-in
naiiT-in
fi-1-bayt
pass.pf-1sg by-(a) boy-Gen
active-Gen
in-the-house
I passed by an active boy in the house.
The second class of SA adjectives, semi-declined ones, is
distinguished by the absence of nunation. Adjectives in this class have
three markers in the definite form, i.e. [-u] in the nominative, [-a] in the
accusative and [-i] in the genitive. However, they have only two markers
in the indefinite, i.e., [-u] in the nominative and [-a] in both the accusative
and the genitive. The following main categories are included in this class:
(i) Broken plurals (See 3.6.3) with any of these patterns:
SA Pattern
Fawaa3iL(-u)

Example
?awaa?il(-u)

Gloss
first

176

maFaa3iil(-u)
Fu3alaa?(-u)
?aF3ilaa?(-u)

masaakiin(-u)
buxalaa?(-u)
?aqiyaa?(-u)

poor
miserly
naughty

The reason why those particular patterns are diptotes is that they have
three syllables, i.e. they are exceeding the maximum (See 3.4).
(ii)

Masculine adjectives of the stem form [Fa3Laan(-u)] whose

feminine stem form is [Fa3Laa], as in /gaDbaan(-u) gaDbaa/
angry and /jaw3aan(-u) jaw3aa/ hungry. If an adjective is of
the stem form [Fa3Laan(-un)] whose feminine form is
[Fa3Laan-at(-un)], it will be fully declined, as in /farhaan(-un)
farhaan-at(-un)/ happy.
(iii)
Adjectives of the stem form [?aF3aL(-u)] and its feminine form
[Fa3Laa?(-u)], as in /?axDar(-u) xaDraaa?(-u)/ green and
/?a3raj(-u) 3arjaaa?(-u)/ lame.
Taking one example, the adjective /buxalaa?(-u)/ miserly
will have the following indefinite forms in the different cases:
(172) a. daxal-a
rijaal-un
buxalaa?-u
enter.pf-3msg men-Nom
miserly-Nom
(Some) miserly men entered the house.
b. ra?ay-tu
rijaal-an
buxalaa?-a
see.pf-1sg
men-Acc
miserly-Acc
I saw (some) miserly men in the house.
c. marar-tu
bi-rijaal-in
buxalaa?-a
pass.pf-1sg by-men-Gen miserly-Gen
I passed by (some) miserly men in the house.

l-bayt
the-house
fi-1-bayt
in-the-house
fi-1-bayt
in-the-house

By contrast, EA has no case markers for adjectives, nor does it use

nunation in forming indefinite adjectives. Compare the adjective /naiiT/
active in the Colloquial examples in (173 and 174) and their Standard
parallels in (170 and 171), respectively:
(173) a. ?il-walad
in-naiiT
daxal
the-boy
the-active
enter.pf
The active boy entered the house.
b. uf-t
il-walad
in-naiiT
see.pf-1sg the-boy
the-active
I saw the active boy in the house.
c. marr-eet
3a-1-walad
in-naiiT

il-beet
the-house
fi-1-beet
in-the-house
fi-1-beet

177

pass.pf-1sg by-the-boy
the-boy
I passed by the active boy in the house.
daxal
(174) a. walad
naiiT
(a) boy
active
enter.pf
An active boy entered the house.
b. uf-t
walad
naiiT
see.pf-1sg
(a) boy
active
I saw an active boy in the house.
c. marr-eet
3ala-walad
naiiT
pass.pf-1sg by-(a) boy
active
I passed by an active boy in the house.

in-the-house
il-beet
the-house
fi-1-beet
in-the-house
fi-1-beet
in-the-house

As in nouns, case markers are dropped in the SA pausal forms of

adjectives and preserved in contextual forms. EA, on the other hand, does
not make any distinction between pausal and contextual forms, nor does it
have any case markers. Compare the adjectives /nabiih(-un)/ intelligent
and /Tawiil(-un)/ long in the Standard sentence (175a) with their
Colloquial analogues in (175b):
(175) a. aakar-a
1-walad-u
n-nabiih-u
study.pf-3msg
the-boy-Nom
the-intelligent-Nom
d-dars-a
T-Tawiil
the-lesson-Acc
the-long
The intelligent boy studied the long lesson.
b. ?il-walad
in-nabiih
zaakir
the-boy
the-intelligent
study.pf
id-dars
iT-Tawiil
the-lesson
the-long
The intelligent boy studied the long lesson.
4.4. Inflection for Gender:
Adjectives agree with the nouns they modify in gender. Thus, two
genders of Arabic adjectives can be discerned: masculine and feminine. A
Few adjectives are genderless in the Colloquial variety. These adjectives
include /baladi/ native and /miiri/ military in addition to some color
adjectives such as /burtu?aani/ orange, /bunni/ brown and /ramaadi/
grey. Feminine adjectives are often marked by a suffix. EA corresponds
to the Standard variety in the use of suffixation to make adjectives
feminine.
Three feminine suffixes are generally recognized for adjectives in
the Standard variety: [-at], replaced by [-a(h)] in pausal forms, [-aa?] and

178

[-aa]. All are replaced by one suffix in EA, namely [-a]. The suffix [-at] or
[-a(h)] is the commonest means of forming feminine adjectives, as in these
examples:
Masc Adj
Tawiil(-un)
3aziiz(-un)
kariim(-un)
naiiT(-un)

Fem Adj
Tawiil-at(-un)
3aziiz-at(-un)
kariim-at(-un)
naiiT-at(-un)

Gloss
tall/long
dear
generous
active

In SA, the final /t/ of the suffix [-at] that is normally used for
feminine gender is dropped in the pausal form of adjectives leaving the
vowel preceding it or /h/ as final phonemes although it is preserved in
contextual forms. Compare the contextual forms of the adjectives
/naiiT-at(-un)/ active (f) and /gamiil-at(-un)/ beautiful (f) in (176177a) and their pausal equivalents in (176-177b), respectively:
(176) a. ?al-bint-u
n-naiiT-at-u
the-girl-Nom
the-active-f-Nom
The active girl is at school.
b. ra?ay-tu
1-bint-a
see.pf-1sg
the-girl-Acc
I saw the active girl.
(177) a. ?al-ward-at-u
1-jamiil-at-u
the-rose-f-Nom
the-beautiful-f-Nom
The beautiful rose is in the garden.
b. ra?ay-tu
1-ward-at-a
see.pf-1sg
the-rose-f-Acc
I saw the beautiful rose.

fi-1-madras-ah
at-the-school
n-naiiT-ah
the-active-f

fi-1-hadiiq-ah
in-the-garden
1-jamiil-ah
the-beautiful-f

In EA, on the other hand, the final [-t] of the singular feminine
suffix never appears in adjectives. This results from the fact that the
environment for this allomorph is never present (owing to the independent
lack of dual or possessed in adjectives).
The suffix [aa?] is frequently used in SA feminine adjectives which
express colors or bodily defects and whose masculine stem form is
[?aF3aL(-u)]. In the Colloquial variety, this feminine marker is generally
reduced to [-a] by Final /?/ Deletion and Final-Vowel Shortening.
Compare the following adjectives in both varieties:
Masc

SA Fem

EA Fem

Gloss

179

?aswad(-u)
?axDar(-u)
?a3raj(-u)
?a3maa

sawd-aa?(-u)
xaDr-aa?(-u)
3arj-aa?(-u)
3amy-aa?(-u)

sood-a
xaDr-a
3arg-a
3amy-a

black
green
lame
blind

The suffix [-aa] is used in SA only to make

comparative/superlative adjectives feminine. For instance, /?akbar(-u)/
older than has the feminine form /kubr-aa/ and /?aSgar(-u)/ younger
than has the feminine form /Sugr-aa/. In EA, the feminine comparative
form is dispensed with and the suffix [-a] is employed. For example,
/?al-bint-u l-kubr-aa > ?il-bint il-kibiir-a/ the older/old girl and
/?al-bint-u S-Sugr-aa > ?il-bint iS-Sugayyar-a/ the younger/young girl.
Some SA masculine adjectives of the stem form [Fa3Laan(-u)]
have the feminine form [Fa3Laa], as in /3aTaan(-u)/ thirsty (m) that
becomes /3aT-aa/ thirsty (f). The feminine stem form of such adjectives
in EA is formed by suffixing [-a] to the masculine form, as in /ta3baan/
tired (m) that becomes /ta3baan-a/ tired (f).
4.5. Inflection for Number:
SA adjectives are inflected for three numbers in agreement with the
nouns they modify: singular, dual and plural. EA adjectives, on the other
hand, are inflected for two numbers only: singular and plural or, more
precisely, singular and non-singular (i.e. one form including both dual and
plural).
4.5.1. The Singular:
In both varieties under discussion, the singular is considered the
unmarked form of adjectives. All what has been stated in the preceding
sections pertains to singular adjectives.
4.5.2. The Dual:
SA adjectives have two dual suffixes depending on the case of the
adjective: [-aan(i)] in the nominative and [-ayn(i)] in the accusative and
genitive, as in the dualization of the adjective /mahuur(-un)/ famous in
the sentences in (178a) and (178b), respectively:
(178) a. daxal-a
enter.pf-3msg
1-mahuur-aani

1-laa3ib-aani
the-player-Nom.du
1-mal3ab

180

the-famous-Nom.du
the-playground
The two famous players entered the playground.
b. ra?ay-tu
1-laa3ib-ayni
see.pf-1sg
the-player-Acc.du
fi-1-mal3ab
1-mahuur-ayni
the-famous-Acc.du
in-the-playground
I saw the two famous players in the playground.
c. naZar-tu
?ila
1-laa3ib-ayni
look.pf-1sg
at
the-player-Gen.du
1-mahuur-ayni
fi-1-mal3ab
the-famous-Gen.du
in-the playground
I looked at the two famous players in the playground.
In the pausal forms of SA adjectives, the final vowel /-i/ of the dual suffix
is deleted, as was explained in (1.4.1).
A very important difference between SA and EA springs from the
latters limited use of the dual in general. Indeed, it provides no dual form
for adjectives - nor does it for verbs (2.2) or pronouns (5.1). The Standard
dual form of adjectives is replaced by related plural forms to modify dual
nouns. Compare the Standard examples in (179-180a) with their
Colloquial counterparts in (179-180b) to see the neutrality of dual and
plural in Colloquial adjectives:
ba3iid-aan
(179) SA: ?al-balad-aani
the-town-Nom.du
far-Nom.du
The two towns are far.
EA:. ?il-balad-een
bu3aad
the-town-du
far.br.pl
The two towns are far.
naajih-at-aan
(180) SA: ?al-bint-aani
the-girl-Nom.du
successful-f-Nom.du
The two girls are successful.
EA:. ?il-bint-een
nagh-iin
the-girl-du
successful-pl
The two girls are successful.
An exception here is the use of the adjective /?asasiyyit-een/ essential
(du) to describe /hagt-een/ two things, but this can be considered an
inter-language borrowing from SA and a result of dialect mixture.

181

4.5.3. The plural:

Two kinds of plural adjectives are found in SA and EA: sound
plural and broken plural. However, while SA has two types of sound plural
adjectives, namely masculine and feminine, EA has only one type, i.e.
sound plural. As in the case of nouns, sound plurals are formed via
suffixation, whereas broken plural is formed by internal vowel pattern
change.
4.5.3.1. Sound Masculine Plural:
The sound masculine plural of SA adjectives is formed by
two suffixes depending on the case of the adjective: [-uun(a)]
nominative case and [-iin(a)] in the accusative and genitive
Consider the inflection of the adjective /mahuur(-un)/ famous
following examples:

one of
in the
cases.
in the

(181) a. daxal-a
1-laa3ib-uuna
enter.pf-3msg
the-player-Nom.mpl
l-mahuur-uuna
1-mal3ab
the-famous-Nom.mpl
the-playground
The famous players entered the playground.
b. ra?ay-tu
1-laa3ib-iina
see.pf-1sg
the-player-Acc.mpl
1-mahuur-iina
fi-1-mal3ab
the-famous-Acc.mpl
in-the-playground
I saw the famous players in the playground.
c. naZar-tu
?ila
1-laa3ib-iina
look.pf-1sg at
the-player-Gen.mpl
1-mahuur-iina
fi-1-mal3ab
the-famous-Gen.mpl
in-the-playground
I looked at the famous players in the playground.
In the pausal forms of SA adjectives, the final vowel [-a] of the sound
masculine plural ending is omitted, as was explained in (1.4.1).
In EA, on the other hand, one suffix [-iin] is generally used to form
the sound plural of adjectives in all positions, as is evident in the examples
in (182) which are the Colloquial analogues of the SA examples in (181),
respectively:
(182) a. ?il-la3ib-iin
the-player-mpl

il-mahur-iin
the-famous-pl

182

daxal-u
1-mal3ab
enter.pf-3pl
the-playground
The famous players entered the playground.
b. uf-t
il-la3ib-iin
see.pf-1sg
the-player-mpl
fi-1-mal3ab
il-mahur-iin
the-famous-pl
in-the-playground
I saw the famous players in the playground.
c. baSS-eet
3a-1-la3ib-iin
look.pf-1sg
at-the-player-mpl
I looked at the famous players.

il-mahur-iin
the-famous-pl

4.5.3.2. Sound Feminine Plural:

To form the sound feminine plural of SA adjectives, the ending
[-aat] is employed. To this ending, one of two case markers is added:
[-u(n)] in the nominative case and [-i(n)] in the accusative and genitive
cases. Observe the different declensions of the adjective /naagih-at(-un)/
successful (f) in the following SA sentences:
(183) a. ?axa-at
iT-Taalib-aat-u
take.pf-3fsg
the-student-fpl-Nom
n-naajih-aat-u
1-hadaaya
the-successful-fpl-Nom
the-presents
The successful students (f) took the presents.
b. ?a3Tay-tu
T-Taalib-aat-i
give.pf-1sg
the-student-fpl-Acc
n-naajih-aat-i
1-hadaaya
the-successful-fpl-Acc
the-presents
I gave the successful students (f) the presents.
c. qaddam-tu
1-hadaaya
li-T-Taalib-aat-i
give.pf-1sg the-presents
to-the-student-fpl-Gen
n-naajih-aat-i
fi-1-hafl-ah
the-successful-fpl-Gen
at-the-party
I gave the presents to the successful students (f) at the party.
In the pausal forms of SA adjectives, the final vowel /u/ or /i/ of the
feminine plural ending (i.e. the whole case marker) is deleted, as was
expounded in (1.4.1).

183

The final [-n] of the sound feminine plural ending in SA is

employed for indefinite adjectives, as in (184):
Taalib-aat-un
(184) a. fi-l-hafl-i
at-the-party-Gen
student-fpl-Nom
naajih-aat-un
kaiir-aat
successful-fpl-Nom
many-fpl
There are many successful students (f) at the party.
b. karram-tu
Taalib-aat-in
honour.pf-1sg
student-fpl-Acc
naajih-aat-in
fi-1-hafl
successful-fpl-Acc
at-the-party
I honored successful students (f) at the party.
c. marar-tu
bi-Taalib-aat-in
pass.pf-1sg
by-student-fpl-Gen
naajih-aat-in
kaiir-aat
successful-fpl-Gen
many-fpl
I passed by many successful students (f).
The sound feminine plural form of adjectives has disappeared in
EA. The sound masculine plural has been neutralized in such a way that it
covers both masculine and feminine. So, one can safely propose that the
suffix [-iin] is a sound plural suffix that is unmarked for gender in EA,
rather than just a masculine plural suffix. Compare the Colloquial
sentences in (185) with their Standard counterparts in (183) above:
(185) a. ?iT-Talib-aat
in-nagh-iin
the-student-fpl
the-successful-pl
xad-u
1-hadaaya
take.pf-3pl
the-presents
The successful students (f) took the presents.
b. 3aT-eet
iT-Talib-aat
give.pf-1sg
the-student-fpl
in-nagh-iin
il-hadaaya
the-successful-pl
the-presents
I gave the successful students (f) the presents.
c. ?addim-t
1-hadaaya
li-T-Talib-aat
introduce.pf-1sg
the-presents
to-the-student-fpl
in-nagh-iin
fi-1-hafla

184

the-successful-pl
in-the-party
I gave the presents to the successful students (f) at the party.
4.5.3.3. Broken Plural:
As has been pointed out in the beginning of section (4.5.3), the
broken plural of adjectives is formed by an internal change in the structure
of their singular forms. Most of the adjectives that have broken plural fall
into one of the following categories in both varieties:
(a)

Adjectives of the stem form [?aF3aL(-u)] have the plural form

[Fu3L(-un)] in both varieties, e.g. /?ahmar(-u)/ red /humr(-un)/.
Three exceptions here are the adjectives /?abyaD(-u)/ white which
has the plural form /biiD(-un)/, /?a3war(-u)/ one-eyed which has the
plural form /3uur(-un)/ and /?aswad(-u)/ black which has the plural
form /suud(-un)/. Their exceptional surface pluralization is due to
their having a glide in place of the medial radical in their underlying
representation. The first plural adjective undergoes the regular
phonological rules of u-to-i and Perseverative Vocoid Assimilation,
as displayed in (186a). The other two plural adjectives undergo
Perseverative Vocoid Assimilation only, as shown in (186b-c):
(186) a. buyD(-un) biyD(-un) biiD(-un)
b. suwd(-un) suud(-un)
c. 3uwr(-un) 3uur(-un)

(b)

Adjectives of the stem form [Fu3aaL(-un)] have the plural

pattern [Fu3Laan(-un)] in both varieties, e.g. /ujaa3(-un) > ugaa3/
brave /uj3aan(-un) > ug3aan/.
(c)
Adjectives of the stem form [Fa3iiL(-un) > Fi3iiL] have the
plural form [Fi3aaL(-un)] or [Fu3alaa?(-u)] in SA, which are changed
to [Fu3aaL] and [Fu3ala] in EA, e.g. /Tawiil(-un)/ tall /Tiwaal(un) > Tuwaal/, /kabiir(-un) > kibiir/ big /kibaar(-un) > kubaar/
and /?amiin(-un)/ honest /?umanaa?(-u) > ?umana/. The change
of SA [Fu3aLaa?(-u)] into EA [Fu3aLa] can be accounted for by
Final /?/ Deletion and Final-Vowel Shortening.
(d)
Adjectives of the stem form [Fa3iyy(-un)] have the plural
pattern [?aF3iyaa?(-u)] in SA, which is changed to [?aF3iya] by Final
/?/ Deletion and Final-Vowel Shortening, or [Fu3aay] in EA. The
first form is produced by Final /?/ Deletion and Final-Vowel
Shortening, e.g. /taqiyy(-un) > ta?i/ pious /?atqiyaa?(-u) >
?at?iya/ and /Tariyy(-un) > Tari/ soft /?aTriyaa?(-u) > Turaay/.

185

(e)

Adjectives of the stem form [Fa3aaL(-un)] have the plural

pattern [Fu3aLaa?(-u)] in SA, that is reduced to [Fu3aLa] by Final /?/
Deletion and Final-Vowel Shortening in EA, e.g. /jabaan(-un) >
gabaan/ cowardly /jubanaa?(-u) > gubana/.
Table (53) displays the patterns of broken plural adjectives
identical in SA and EA. Table (54) shows the phonological changes that
some SA broken plural patterns of adjectives have undergone in EA. Most
of those changes are governed by regular phonological rules, whereas a
few are phonologically irregular.

1
2
3
4
5
6
7

Table (53)
Patterns of Adjectival Broken
Plural Identical in SA & EA
Pattern
Example
Fu3L(-un)
humr(-un) > humr
Fu3Laan(-un) uj3aan(-un) > ug3aan
FiiL(-un)
biiD(-un) > biiD
FuuL(-un)
suud(-un) > suud
Fa3aa?iL(-u)
?awaa?il(-u) > ?awaa?il
(?)aF3aaL(-un) (?)ahraar(-un) > (?)ahraar
Fu33aaL(-un) uTTaar(-un) > uTTaar

Gloss
red
brave
white
black
first
free
clever

Table (53) reveals that the SA adjectival broken plural patterns

which are preserved without change in EA follow these templates:
[CVCC], [CVCCaan], [CVVC], [CVCVV?VC], [aCCVVC] and
[CVCCVVC]. Table (54) shows that the SA adjectival broken plural
patterns which are phonologically changed in EA have these templates:
[CVCVCaa?], [CVCVVC], [aCCVCaa?], [CVCVC], [CVCVV],
[maCVVCVVC], and [CVCVVCVV]. More specifically, stem forms (1
and 3) follow both Final /?/ Deletion and Final-Vowel Shortening. Stem
forms (5 and 7) undergo the rule of` Final-Vowel Shortening. Stem form
(6) is governed by the rule of Atonic Shortening. However, the changes
occurring in stem forms (2 and 4) are phonologically irregular or at least
the precise conditions in which these changes occur are not yet known.
Table (54)
SA Adjectival Broken Plural Patterns Phonologically Changed in EA
Pattern
Example
SA
EA
SA
EA
Gloss
1 Fu3aLaa?(-u) Fu3aLa
kuramaa?(-u) kurama
generous
"
Fu3aaL
Zurafaa?(-u)
Zuraaf
funny

186

2 Fi3aaL(-un)
3 (?)aF3iLaa?(-u)
"
"
4 Fu3uL(-un)
5 Fa3Laa
6 maFaa3iiL(-u)
7 Fa3aaLaa

Fu3aaL
(?)aF3iLa
Fa3aayiL
Fu3aaL
Fu3aaL
Fa3La
maFa3iiL
Fa3aaLa

kibaar(-un)
(?)agniyaa?(-u)
(?)aqribaa?(-u)
(?)aqiyaa?(-u)
judud(-un)
marDaa
masaakiin(-u)
kasaalaa

kubaar
(?)agniya
?araayib
u?aay
gudaad
marDa
masakiin
kasaala

big/old
rich
relative
naughty
new
sick
poor
lazy

A few adjectival broken plural patterns are specifically Colloquial,

such as [Fa33iiL-a], e.g. /hassiib-a/ clever in book-keeping (pl),
[Fa3aL1iiL2], e.g. /?anaziih/ showy (pl) and [Fi3Laan], e.g. /gid3aan/
brave. But it must be stressed that those patterns have very limited
frequency in the dialect.
A major similarity between SA and EA is that in both of them a
singular feminine adjective is used to modify a plural noun of non-human
reference, as in (187-188). However, EA extends this optionally to
feminine plural nouns of human reference as well. Compare (189a) with
(189b):
(187) a. qumSaan-un
shirts-Nom
New shirts.
b. ?umSaan
shirts
New shirts.

jadiid-ah
new-f

gidiid-a
new-f

(188) a. maktab-aat-un
library-fpl-Nom
Big libraries/bookshops.
b. maktab-aat
library-fpl
Big libraries/bookshops.

kabiir-ah
big-fsg

(189) a. banaat-un
girl.br.pl-Nom
Clever girls.
b. banaat
girl.br.pl
Clever girls.

aaTir-aat
clever-fpl

kibiir-a
big-fsg

aTr-a
clever-fsg

187

On the other hand, a remarkable difference in the pluralization

system between SA and EA is that the latter uses the sound masculine
plural or the broken plural forms of adjectives to modify sound feminine
plural nouns. In other words, the [-aat] feminine plural suffix is not used
for adjectives in the Colloquial variety. Compare the Standard structures in
(190-191a) with their Colloquial analogues in (190-191b):
(190) a. ?al-muhandis-aat-u
n-naajih-aat
the-engineer-fpl-Nom
the-successful-fpl
The successful engineers (f).
b. ?il-muhandis-aat
in-nagh-iin
the-engineer-fpl
the-successful-mpl
The successful engineers (f).
(191) a. ?al-ward-aat-u
1-jamiil-aat
the-rose-fpl-Nom
the-beautiful-fpl
The beautiful roses.
b. ?il-ward-aat
ig-gumaal
the-rose-fpl
the-beautiful-br.pl
The beautiful roses.
To make the agreement of plural nouns and adjectives clearer to
the reader, an illustrative chart, showing the agreement possibilities in SA
and EA for nouns of various classes, is provided in Table (55):

Table (55)
Plural Noun-Adjective Agreement in SA & EA
Pl Noun (SA)
SA Adj
EA Adj
muhandis-uuna
kibaar(-un)
kubaar
kabiir-uuna
*kibir-iin
+hum
naajih-uuna
nagh-iin
fem

broken
muhandis-aat(-un) *kabiir-at(-un)
kibiir-a
*kibaar(-un)
kubaar
+hum
kabiir-aat(-un)
*kibir-aat
+fem

*naajih-uuna
nagh-iin
broken
?awlaad(-un)
kibaar(-un)
kubaar
naajih-uuna
nagh-iin
+hum
fem

+broken
3araa?is(-u)
*kabiir-at(-un)
kibiir-a
*kibaar(-un)
kubaar
+hum
kabiir-aat(-un)
*kibir-aat
+fem

*naajih-uuna
nagh-iin
+broken

188

5 maktab-aat(-un)
hum
+fem

broken
6 buyuut(-un)
hum
fem

+broken
7 mudun(-un)
hum
+fem

+broken

kabiir-at(-un)
*kibaar(-un)
kabiir-aat(-un)

kibiir-a
kubaar
*kibir-aat

kabiir-at(-un)
kibaar(-un)

kibiir-a
kubaar

kabiir-at(-un)
kibaar(-un)

kibiir-a
kubaar

From Table (55), the feature specification rules in (192) can be

deducted to apply, in order, for the agreement between plural nouns and
their adjectives in SA, and those in (193) can be deducted for EA.
(Because the optional rules below do not necessarily apply, more than one
agreement pattern occurs):
(192) Agreement of Plural Nouns and Adjectives in SA:
+ fem
a.
[broken]
+ hum
+ fem

b. [hum] + sg
(optional)
broken
fem
c.
(optional)
[+broken]
broken
d. An adjective agrees with the noun in all remaining features.

(193) Agreement of Plural Nouns and Adjectives in EA:

a. [+fem] [+sg]
(optional)
+ fem
b. [hum]
(optional)

+ sg
c. no [+-aat] rule
d. [broken] cannot override an adjectives specification.

189

Careful investigation of Table (55) and the rules in (192-193)

reveals the following remarks:
First, the similarities between SA and EA are:
(i) A sound masculine plural adjective can modify a sound masculine
plural noun.
(ii)A singular feminine adjective modifies a feminine plural noun of nonhuman reference.
(iii)A broken plural adjective can be used to modify a broken plural noun
of non-human reference.
(iv)When a broken plural adjective exists, both varieties use it to modify a
sound masculine plural noun.
Second, the differences between SA and EA are:
(i) A sound feminine plural adjective can modify a sound feminine plural
noun in SA but not in EA.
(ii)A singular feminine adjective can modify a feminine plural noun of
human reference in EA but not in SA.
(iii)A broken plural adjective can modify a sound feminine plural noun of
human reference in EA but not in SA.
(iv)A sound masculine plural adjective can modify a feminine plural noun
of human reference in EA but not in SA.
Another major distinction between SA and EA is that whereas all
Standard adjectives are inflected for number and gender, a small number
of EA adjectives have the morphological anomaly of not accepting number
or gender affixes, though they may have semantic gender. These are
particularly relational adjectives denoting colors or relating things to a
group of people. Examples of the first type are: /bunn-i/ brown,
/ramaad-i/ grey and /burtu?aan-i/ orange. Examples of the second type
are /rigaal-i/ manly, /Subyaan-i/ boyish and /banaat-i/ girlish (cf.
Hussein 1973: 20).
4.6. Inflection for Degree:
Both Standard and Colloquial adjectives are inflected for degree.
Formally, according to Aboul-Fetouh (1969: 102), there are only two
degrees: positive (unmarked) and comparative (marked). The superlative
degree is expressed phrasally or by slot position. It might be added that
the superlative degree can be formed by affixing a plural pronoun to the
comparative form as will be shown in (4.6.3).

190

4.6.1. The Positive Degree:

The positive degree of adjectives includes the unmarked form of
these adjectives. For the various forms of adjectives in the positive degree,
reference should be made to the adjectival stem forms introduced in (4.1).
4.6.2. The Comparative Degree:
To derive the comparative form of adjectives from the positive
form, the pattern [?aF3aL(-u)] is employed in both varieties, with [?a-] as
an inflectional prefix. Examples of the derivation of comparatives are
given below (with their SA pronunciation):
Positive
Sa3b(-un) difficult
Da3iif(-un) weak
jabaan(-un) cowardly
ujaa3(-un) brave
mufiid(-un) useful
munaasib(-un) suitable

Comparative
?aS3ab(-u) more difficult
?aD3af(-u) weaker
?ajban(-u) more cowardly
?aja3(-u) braver
?afyad(-u) more useful
?ansab(-u) more suitable

As can be observed from the examples above, there are certain

morphophonemic alternations involved in the formation of the
comparative pattern [?aF3aL(-u)] from positive adjectives. These
alterations include the deletion of the first vowel of the stem form and the
replacement of the second vowel with /a/.
From adjectives with doubled roots, i.e. having identical second
and third radicals, the variant of [?aF3aL(-u)] is [?aFa33(-u)] by the
regular rule of Identical-Consonant Metathesis. From those with defective
roots, i.e. ending in a glide, the variant is [?aF3aa > ?aF3a] from [?aF3aGu] by Glide Elision and Perseverative Vocoid Assimilation in both
varieties, then by Final-Vowel Shortening in EA, as in the following
adjectives:
Positive
xafiif(-un) > xafiif
adiid(-un) > idiid
ganiyy(-un) > gani
3aali(n) > 3aali
hulw(-un) > hilw

SA Comp.
?axaff(-u)
?aadd(-u)
?agnaa
?a3laa
?ahlaa

EA Comp.
?axaff
?aadd
?agna
?a3la
?ahla

Gloss
lighter
stronger
richer
higher
sweeter

191

Syntactically, comparative adjectives are usually followed by the

preposition /min/ than. And, in both varieties, the comparative form is
invariable, i.e. employed with all numbers and genders, as in these
examples:
(194) a. ?al-walad-u ?akbar-u
min
the-boy-Nom older-Nom
than
The boy is older than his brother.
b. ?il-walad
?akbar
min
the-boy
older
than
The boy is older than his brother.

?ax-ii-h
brother-Gen-3msg
?axuu-h
brother-3msg

(195) a.?al-banaat-u
?akbar-u
min ?ixwati-hinn
the-girls-Nom
older-Nom
than brothers-3fpl
The girls are older than their brothers.
b.?il-banaat
?akbar
min ixwat-hum
the-girls
older
than brothers-3pl
The girls are older than their brothers.
In addition to the normal pattern above, an alternative syntactic
construction is used in the Colloquial variety to express comparison. It is
used in a low register and involves placing the preposition /3an/ after the
adjective which remains unchanged. Compare the Standard with the
Colloquial examples in (196):
(196) SA: ?ahmad-u
?akbar-u
Ahmad-Nom
older-Nom
Ahmad is older than Omar.
EA:. ?ahmad
kibiir
Ahmad
old
Ahmad is older than Omar.

min
than

3umar
Omar

3an
than

3umar
Omar

Some adjectives, particularly some of those in the form of active or

passive participles (See 4.7), cannot be inflected for comparison. Their
comparative is formed phrasally in SA by using a noun or a verbal noun
after a suitable comparative adjective like /?akar(-u)/ more, /?akbar(-u)/
bigger, /?aadd(-u)/ stronger, /?aqall(-u)/ less, etc. The noun or verbal
noun will be in the indefinite accusative case20. Such a noun is called the
accusative of respect/specification (Thackston 1984: 62) because it
indicates the basis of comparison or in what respect a thing is compared.
In EA, the two alternative constructions stated above are used without any

192

change in the adjectives. Compare the Standard examples in (197-198a)

with their Colloquial counterparts in (197-198b):
(197) a. ?al-burj-u
?akar-u rtifaa3-an
min al-3imaar-ah
the-tower-Nom more-Nom height-Acc
than the-building
The tower is higher than the building.
b. ?il-burg
murtafi3
3an
il-3imaar-a
the-tower
high
than the-building
The tower is higher than the building.
(198) a. ?ahmad-u
?aqall-u
htimaam-an
Ahmad-Nom less-Nom
interest-Acc
Ahmad is less interested than Omar.
b. ?ahmad
muhtammi
?a?alli
Ahmad
interested
less
Ahmad is less interested than Omar.

min
than

3umar
Omar

min
than

3umar
Omar

The comparative of adjectives of color and bodily defects which

have [?a-] in their singular masculine forms is formed phrasally in SA (e.g.
199a), whereas in EA there is a choice between the phrasal form (e.g.
199b) and the use of the singular masculine form (e.g. 199c):
(199) a. haaihi
1-ward-at-u
this.f.
the-rose-Nom
hmiraar-an
min
redness-Acc
than
This rose is redder than that (one).
b. ?il-ward-a
di
hamra
the-rose
this (f)
red (f)
This rose is redder than that one.
c. ?il-ward-a
di
?ahmar
the-rose
this (f)
redder
This rose is redder than that (one).

?aadd-u
stronger-Nom
tilk
that
?aktar min
more than

dikha
that

min
than

dukha
that

Certain adjectives in both varieties do not inflect for degree by any

means. These include numerical adjectives (e.g. /ar-raabi3(-u)/ the
fourth) and quantitative adjectives or quantifiers (e.g. /kul/ all and
/mu3Zam/ most).
4.6.3. The Superlative Degree:
In SA and EA, adjectives have the same form in both the
comparative and the superlative degrees. They are distinguished only by

193

affixation or by syntactic means. One of the morphological means of

forming the superlative degree is affixing a plural pronoun to the
comparative form, as in SA (200-201a) and EA (200-201b).
(200) a. haaihi
1-wardat-u
?ajmal-u-hunna
this (f)
the-rose-Nom
most beautiful-Nom-3fpl
This rose is the most beautiful of them.
b. ?il-ward-a
di
?agmal-hum
the-rose
this (f)
most beautiful-3pl
This rose is the most beautiful of them.
(201) a. haaa
1-walad-u
?akaa-hum
this (m)
the-boy-Nom
most intelligent-3mpl
This boy is the most intelligent of them.
b. ?il-walad
da
?azkaa-hum
the-boy
this (m)
most intelligent-3pl
This boy is the most intelligent of them.
Comparative adjectives are also made superlative by the
prefixation of the definite article to them:
huwa
1-?aTwal
(202) a. ?ahmad-u
Ahmad-Nom
he
the-tallest
Ahmad is the tallest.
b. ?ahmad
huwwa
1-aTwal
Ahmad
he
the-tallest
Ahmed is the tallest.
Syntactically, comparatives are turned into superlatives by being
followed by (a) an indefinite singular noun, (b) the indefinite pronoun
/waahid-at(-un) > wahd-a/ one or, (c) a definite plural noun. Examples of
superlative adjectives are SA (203-206a) and EA (203-206b):
?aTwal-u
walad/waahid(-in)
(203) a. ?ahmad-u
Ahmad-Nom
tallest-Nom boy/one(-Gen)
Ahmad is the tallest boy/one.
b. ?ahmad
?aTwal
walad/waahid
Ahmad
tallest
boy/one
Ahmad is the tallest boy/one.
(204) a. hind-u
?ajmal-u
bint/waahid-at(-in)
Hind-Nom
most beautiful-Nom girl/one-f(-Gen)
Hind is the most beautiful girl/one.
b. hind
?agmal
bint/wahd-a

194

Hind
most beautiful
girl/one-f
Hind is the most beautiful girl/one.
?aTwal-u
1-?awlaad(-i)
(205) a. ?ahmad-u
Ahmad-Nom
tallest-Nom the boy.br.pl(-Gen)
Ahmad is the tallest of the boys.
b. ?ahmad
?aTwal
il-wilaad
Ahmad-Nom
tallest
the boy.br.pl
Ahmad is the tallest of the boys.
(206) a. hind-u
?ajmal-u
1-banaat(-i)
Hind
most beautiful
the-girl.br.pl(-Gen)
Hind is the most beautiful of the girls.
b. hind
?agmal
il-banaat
Hind
most beautiful
the-girl.br.pl
Hind is the most beautiful of the girls.
4.7. Participles:
Arabic participles are morphological forms derived from verbs to
refer to the person or thing performing or undergoing the action expressed
by those verbs. Two participles are found in this language: the active
participle and the passive participle. Syntactically, the participles perform
a number of functions in both varieties: they may be used as nouns,
adjectives or tense forms (i.e. replacing verbs). To explain the complexity
of the situation, Mitchell (1956: 104) says:
The participle is nominal in form with no distinction in
person. It is possible, therefore, to consider katib
iggawab in the sentence huwwa katib iggawab as a
sequence of two nouns in the construct and translate he is
the writer of the letter. But the participle of many verbs
(e.g. ktab, yktib) may be said to refer to the state of
having performed the verbal action. In the appropriate
context the translation of huwwa katib iggawab is he
has written the letter.
Two reasons can be given here for the treatment of participles as
adjectives. First, they behave morphologically like adjectives, particularly
in their inflection for gender and number. They make their feminine, dual
and plural with regular adjectival endings (Thackston 1984). Second,
though Arabic participles can be used as nouns or verbs, their main
function is as adjectives. As for the SA participles, Thackston (1984: 41-2)
says: The active participle often functions, like the English present active

195

participle in -ing, as a verbal adjective for on-going action, or the durative

aspect. Speaking of the passive participle, he states that it is used purely
adjectivally, like the English past passive participle. The EA participles
are also treated as verbal adjectives by Badawi & Hinds (1986). The
discussion of the syntactic functions of participles is out of the scope of
the present work. They will be studied from a morphological perspective
only.
4.7.1. The Active Participle:
The active participle is a morphological form derived from a verb
to refer to the person or thing that performs the action expressed by the
verb. It has two patterns; one for the primary triradical form (i.e. Form I)
and the other for the derived triradical as well as the quadriradical forms.
The first pattern is characterized in SA by lengthening the vowel
after the first consonant of the perfect form of the verb, i.e. making it /aa/
and replacing the vowel before the last consonant by /i/ if it is not already
/i/. Hence, the normal active participle pattern is [Faa3iL(-un)] doer, as
in /aarib(-un)/ drinking which is the active participle of /arib(-a)/ to
drink. This holds true for EA in most cases. Thus, /gaalib/ beating is the
active participle of /galab/ to beat, /Taali3/ going up is that of /Tili3/ to
go up and /naazil/ going down is derived from /nizil/ to go down.
For geminate verbs the variant of [Faa3iL(-un)] is [Faa33(-un)] in
SA but [Faa3i3] in EA. The citation form in SA is usually followed by the
default vocalic nominative suffix [-un]. The changes occurring in the
transfer from underlying SA [Faa3i3(-un)] to surface [Faa33(-un)] can be
demonstrated by the following example:
(207) maadid(-un)
maaidd(-un)
maaadd(-un)
maadd(-un)

by Identical-Consonant Metathesis
by Perseverative Vocoid Assimilation
by Closed-Syllable Shortening

This shows that the rule of Identical-Consonant Metathesis applies to the

SA citation form because the stem form is followed by a vowel. However,
it does not apply to the EA citation form because the vocalic case suffix is
lost. Compare the following examples in citation form:
SA
aadd(-un)
haass(-un)

EA
aadid
haasis

Gloss
pulling
feeling

196

haadd(-un)
jaarr(-un)

haadid
gaarir

destroying
pulling

Before vocalic suffixes the SA active participle form of geminate

roots is similar to its citation form, i.e. [Faa33-]. It has its vowel shortened
by Closed-Syllable Shortening in EA, i.e. [Fa33-]:
SA
aadd-u-h
3aadd-u-ka
jaarr-u-ki

EA
add-u
3add-ak
garr-ik

Gloss
pulling it (m)
counting you (m)
pulling you (f)

These surface forms are derived from their underlying counterparts by

Identical-Consonant Metathesis, Perseverative vocoid Assimilation and
Closed-Syllable Shortening in both varieties. However, the last rule
applies twice in EA, as opposed to once in SA. This makes the SA forms
unusual insofar as they have a superheavy syllable in medial position. For
example, the derivation of /aad-uh > add-u/ runs as in (208):
(208) SA: aadid-uh aaidd-uh aaadd-uh aadd-uh
EA: aadid-u aaidd-u aaadd-u aadd-u add-u
Before consonantal suffixes, the underlying active participle form
of geminate roots is used in EA, i.e. [Fa3i3-] which is derived from
[Faa3i3-] by Atonic Shortening. On the other hand, active participles are
not followed by consonantal suffixes in SA. Compare these examples:
SA
jaarr-u-kum
aadd-u-haa
haadd-u-hum

EA
garir-kum
adid-ha
hadid-hum

Gloss
pulling you (mpl)
pulling it (f)
destroying them

From initial-glottalized verbs, the variant of [Faa3iL(-un)] is

[?aa3iL(-un)] in SA. In the EA verbs /?akal/ to eat and /?axad/ to take,
the active participle pattern is [waa3iL]. However, the glottal stop is
preserved in the active participle of one EA verb, /?amar/ to order.
SA
?aakil(-un)
?aaxi(-un)
?aamir(-un)

EA
waakil
waaxid
?aamir

Gloss
eating
taking
ordering

197

If we consider /?amar/ to order a result of dialect mixture in EA, we can

propose a generalization about the change of an initial glottal stop into /w/
in EA:
(209) ?-to-w in EA:
? w /# aa
For medial-glottalized verbs the SA active participle form is
[Faa?iL(-un)] which is rendered [FaayiL] by Intervocalic /?/ Assimilation
in EA, as in /Saa?im(-un) > Saayim/ fasting. Final-glottalized verbs have
a glottal stop at the end of their SA active participles, which is deleted by
Final /?/ Deletion in EA, e.g. /qaari?(-un) > ?aari/ reading.
From medial-weak verbs the SA variant of [Faa3iL(-un)] is
[Faa?iL(-un)]. The glottal stop is produced by the rule of Glottal
Formation in SA. Thus, the underlying form [FaaGiL(-un)] becomes
[Faa?iL(-un)] which is reduced by Intervocalic /?/ Assimilation into
[FaayiL] in the Colloquial variety. Compare these active participles:
SA
Saa?im(-un)
qaa?im(-un)
naa?im(-un)
qaa?il(-un)

EA
Saayim
?aayim
naayim
?aayil

Gloss
fasting
standing
sleeping
saying

From final-weak verbs the variant of [Faa3iL(-un)] in SA is

[Faa3i(n)] from [Faa3iy(-un)] in the nominative case by Glide Elision,
Perseverative Vocoid Assimilation and Closed-Syllable Shortening, and
from [Faa3iy(-in)] in the genitive case by Glide Elision and ClosedSyllable Shortening. In the accusative case no elision or assimilation
occurs, so the pattern is [Faa3iy(-an)]. These facts have been noted by
Thackston (1984: 46) as follows:
Only the accusative case actually shows its case ending.
This results from an internal collapse due to weakness:
*haadiyun haadin and *haadiyin haadin, where the
weak vowels u and i cannot maintain a weak consonant
between them; the strong vowel a does support a weak
consonant, so haadiyan and al-haadiya do not suffer
collapse.

198

Both SA forms are replaced by one in EA, namely [Faa3i], which is

derived from [Faa3iy Faa3ii] by Final-Vowel Shortening, as in the
following examples:
SA
EA
Gloss
maai(n)
maai
walking
maaiy(-an)
"
"
qaaDi(n)
?aaDi
judging
qaadiy(-an)
"
"
daa3i(n)
daa3i
inviting
daa3iy(-an)
"
"
baani(n)
baani
building
baaniy-an
"
"
The active participle of the derived triradical as well as the
quadriradical verbs is formed in SA from the imperfect form of the verb
by replacing the consonant of the imperfect prefix [yu-] by /m/ and
replacing the vowel before the last consonant by /i/ if it not already /i/. So,
the normal pattern would be in most instances [mu-...iC]. The pattern itself
has persisted in the Colloquial variety, but with [mi-] as a prefix in some
active participles instead of or alternating with [mu-]. Contrast Tables (56)
and (57).
Table (56)
Patterns of Derived Triradical Active Participles in SA
No Verb Form
Active Participle Example
II yu-Fa33iL-u mu-Fa33iL(-un) mu-jarrib(-un) experiencing
III yu-Faa3iL-u mu-Faa3iL(-un) mu-haasib(-un) accounting
IV yu-F3iL-u
mu-F3iL(-un)
mu-rif(-un) supervising
V ya-taFa33aL-u mu-taFa33iL(-un) mu-taharrik(-un) moving
VI ya-taFaa3aL-u mu-taFaa3iL(-un) mu-tanaaqi(-un) discussing
VII ya-nFa3iL-u mu-nFa3iL(-un) mu-nhazim(-un) defeated
VIII ya-Fta3iL-u
mu-Fta3iL(-un)
mu-jtahid(-un) working hard
IX ya-F3aLL-u
mu-F3aLL(-un)
mu-hmarr(-un) becoming red
X ya-staF3iL-u mu-staF3iL(-un) mu-stagfir(-un) asking God for
forgiveness

No
IIa
IIb
III

Table (57)
Patterns of Derived Triradical Active Participles in EA
Verb Form
Active Participle Example
yi-Fa33aL
mi-Fa33aL
mi-aggal employing
yi-Fa33iL
mi-Fa33iL
mi-?addim presenting
yi-Faa3iL
mi-Faa3iL
mi-saafir traveling

199

IV yi-F3iL
Va yi-tFa33aL
"
Vb yi-tFa33iL
VI yi-tFaa3iL
VII yi-nFi3iL
VIII yi-Fti3iL
"
IX yi-F3aLL
Xa yi-staF3aL

mu-F3iL
mi-tFa33aL
mu-taFa33iL
mi-tFa33iL
mi-tFaa3iL
mu-nFa3iL
mi-Fti3iL
mu-Fta3iL
mi-F3iLL
mi-staF3aL

"
Xb yi-staF3iL

mi-staF3iL
mistaF3iL

mu-rif supervisor
mi-tgarrab living abroad
mu-taxaSSiS specialized
mi-thammil enduring
mi-tnaa?i discussing
mu-nfa3il agitated
mi-htirim respecting
mu-gtahid working hard
mi-hmirr becoming red
mi-stagfar asking God for
forgiveness
mi-stamti3 enjoying
mi-sta3gil being in a hurry

Careful examination of Tables (56) and (57) reveals that only [mu] is used as an active participle prefix in SA but [mi-] is the usual active
participle prefix in EA, though [mu-] is borrowed in some forms. To
explain the difference between nouns beginning with [mu-] and those
beginning with [mi-], Radwan (1975: 42-3) distinguishes between
participial and occupational-noun forms in EA: The occupational noun is
characterized by the vowel /u/ in the first syllable, whereas the participial
form, expressing the action itself, is characterized by the vowel /i/ in the
first syllable. For example, /mu-darris/ a teacher but /mi-darris/ have
taught and /mu-haafiZ/ a governor but /mi-haafiz/ have kept.
Following Radwans argument, one can propose that when the SA
active participles are lexicalized as adjectives in EA, the adjectives are
marked with the vowel /u/ in the prefix and the participial forms are
marked with the vowel /i/ in the prefix. For instance, /mu-naasib/
suitable vs. /mi-naasib/ have suited and /mu-Sammim/ insistent vs.
/mi-Sammim/ have designed. Where one form with [mu-] is found in
EA, as well as in SA, it is considered a lexicalized adjective. Illustrative
examples are /mu-xtalif/ different, /mu-himm/ important and
/mu-haayid/ non-biased.
The problem that remains unsolved is that for some SA active
participles two participial forms, i.e. referring only to the action itself, are
found in EA. One can propose that the process of lexical diffusion (See
3.1.1) is responsible for that. To exemplify, for the following SA active
participles two EA counterparts are found in Gadalla et al. (1998):
SA
mu-3tabir(-un)

EA
mi-3tibir/mu-3tabir

Gloss
considering

200

mu-3tamid(-un)
mu-taxayyil(-un)
mu-ttafiq(-un)

mi-3timid/mu-3tamid
mi-txayyil/mu-taxayyil
mi-ttifi?/mu-ttafi?

depending on
imagining
agreeing

But it must be stated here that the doublets in EA are distinguished by the
level of formality. Forms with [mu-] are used by more educated people.
We can modify the lexical diffusion scenario to accommodate the
lexicalization notion by proposing that the prefix [mu-] became [mi-] at
some historical point in time and that participles which became
lexicalized (i.e. listed with a special form or meaning) before the [mu- >
mi-] change occurred retain their [mu-].
It can also be noticed that the active participles of Forms IV and
VII are identical in both varieties because educated speakers in Cairo
prefer the Standard forms of these participles. One might question the
activity of the active participle [mu-nFa3iL(-un)] because it is passive in
meaning, but Thackston (1984: 101) affirms that all patterns of Form VII
are active in form but middle or passive (medio-passive) in meaning.
Some remarks must be added about the derivation of active
participles from the derived forms of non-sound verbs:
(a) Doubled verbs of Form IV have the active participle [mu-Fi33(-un)]
from [mu-F3i3(-un)] by Identical-Consonant Metathesis in both varieties,
e.g. /mu-timm(-un)/ completing. For those of Form X, the pattern is
[mu-staFi33(-un)], e.g. /mu-sta3idd(-un)/ ready.
(b) Final-glottalized active participles in SA have their /?/ deleted in EA by
Final /?/ Deletion, e.g. /mu-tahanni?(-un) > mi-thanni/ rejoicing.
(c) For initial-weak verbs of Form IV, the active participle is
[muu3iL(-un)] which is derived from [mu-w3iL(-un)] in w-verbs by
Perseverative Vocoid Assimilation and from [mu-y3iL(-un)] in y-verbs by
a special rule that applies in initial-weak verbs only. This rule changes the
/y/ into /w/ after a prefixed /u/ as follows:
(210) y-to-w:
yw/u+
By this rule [mu-y3iL(-un)] becomes [mu-w3iL(-un)] which then
becomes [muu3iL(-un)] by Perseverative Vocoid Assimilation. An
example of an active participle derived from a w-verb is /muuqif(-un)/
stopping and of a y-verb is /muuqin(-un)/ believing.
(d) For medial-weak verbs of Form IV, the active participle is
[mu-FiiL(-un)] from underlying [mu-FGiL(-un)] by Anticipatory Vocoid
Assimilation, as in /mu-fiid(-un)/ useful. Those of Form VII have the

201

active participle [mu-nFaaL(-un)] from underlying [mu-nFaGiL(-un)] by

Glide Elision and Perseverative Vocoid Assimilation, e.g. /mu-nhaaz(-un)/
biased. Those of Form VIII have the SA active participle pattern [muFtaaL(-un)] from [mu-FtaGiL(-un)] by Glide Elision and Perseverative
Vocoid Assimilation. It has a [mi-FtaaL] counterpart in EA, e.g.
/mu-xtaar(-un) > mi-xtaar/ choosing. For Form X, the active participle is
[mu-staFiiL(-un)] from [mu-staFGiL(-un)] by Anticipatory Vocoid
Assimilation, e.g. /mu-staqiim(-un)/ straightforward.
(e) Final-weak verbs of derived triradicals have [-i(n)] at the end of SA
indefinite forms and a long /ii/ at the end of the SA definite forms which is
then shortened by Final-Vowel Shortening in EA. The surface form [i(n) ~
ii] is derived from [iG(-un) ~ iG(-u)] by Glide Elision and Perseverative
Vocoid Assimilation, then by Closed-Syllable Shortening in indefinite
forms, as in these examples:
V Form
II
III
IV
V
VI
VII
VIII
X

SA
mu-faDDi(n)
mu-raaDi(n)
mu-lqi(n)
mu-talaqqi(n)
mu-tanaasi(n)
mu-ntahi(n)
mu-tari(n)
mu-stakfi(n)

EA
mi-faDDi
mi-raaDi
mu-lqi
mi-tla??i
mi-tnaasi
mu-ntahi
mi-tiri
mi-stakfi

Gloss
emptying
satisfying
throwing
receiving
forgetting
ending
buying
having enough

(f) For quadriradical verbs of Form I, active participles have the form
[mu-Fa3L1iL2(-un)] which has two EA counterparts [mi-Fa3L1aL2 ~
mi-Fa3L1iL2], as in /mu-zaxrif(-un) > mi-zaxraf/ decorating and
/mu-tarjim(-un) > mi-targim/ translating.
4.7.2. The Passive Participle:
The passive participle is a morphological form which signifies the
person or thing that has undergone the action expressed by the verb. Arab
linguists (such as Al-Hammadi et al. 1977) define it as a nominal derived
from the passive verb to refer to what underwent the action. So, it can be
derived from transitive and unergative verbs but it cannot be derived from
unaccusative verbs since they cannot be passivized. Unergative verbs are
those verbs whose surface subject is the doer of the action, not the
experiencer of some action, e.g. /niim-a fi l-bayt-i/ There was sleeping in
the house; (literally) *It was slept in the house. This indicates that
passives of unergatives are grammatical in Arabic though they are not so
in English. Some passive participles are also used as simple nouns,

202

e.g./mawDuu3(-un)/ have been put ~ a topic and /ma?muur(-un)/ have

been ordered ~ a police chief.
The usual pattern for the passive participle of primary triradical
verbs in SA and EA is [maF3uuL(-un)], as in /maruub(-un)/ drunk
which is derived from /arib(-a)/ to drink. Following are some remarks
about the non-sound verbs:
(a) In geminate verbs, the passive participle in both varieties has the regular
underlying form [maF3uu3(-un)], e.g. /maduud(-un)/ pulled.
(b) Initial-glottalized verbs of Form I have the SA passive participle form
[ma?3uuL(-un)] in both varieties, e.g. /ma?muur(-un)/ ordered. However,
in some cases, EA uses the passive participle of Form VI to replace that of
Form I, i.e. [mi-t?aa3iL mi-ttaa3iL (by exceptional assimilation)], e.g.
/ma?kuul(-un) > mi-t?aakil mi-ttaakil/ eaten.
(c) In initial-weak verbs, the pattern becomes either [maw3uuL(-un)] or
[may3uuL(-un)]. EA corresponds to SA in the use of this pattern, and
retains the sequences /aw/ and /ay/ there because they are not stem-final,
as in /mawzuun(-un) > mawzuun/ weighed and /mawjuud(-un) >
mawguud/ present; /maysuur(-un) > maysuur/ easy and /maymuun(-un)
> maymuun/ fortunate.
(d) From medial-weak verbs the Standard variants of [maF3uuL(-un)] are
[maFuuL(-un)] which is derived from underlying [maFwuuL(-un)] by
Anticipatory Vocoid Assimilation and the shortening of an extra-long
vowel in the case of w-verbs and [maFiiL(-un)] which is derived from
underlying [maFyuuL(-un)] by u-to-i in the case of y-verbs. There is no
passive participle for medial-weak verbs of Form I in EA (Mitchell 1956:
101). Thus, both SA passive participle patterns of the primary active verb
Form I, i.e. [maFuuL(-un)] and [maFiiL(-un)] are replaced by [mi-nFaaL]
or [mi-tFaaL] which are the passive participles of the passive versions of
Form I, i.e. [(?i)nFaaL] and [(?i)tFaaL], respectively, as in the following
examples:
Root
S-w-n
q-w-l
b-y-3
k-y-d

SA
maSuun(-un)
maquul(-un)
mabii3(-un)
makiid(-un)

EA
mi-tSaan
mi-t?aal
mi-nbaa3
mi-tkaad

Gloss
preserved
said
sold
vexed/annoyed

(e) From final-weak verbs the SA variants of [maF3uuL(-un)] are

[maF3uww(-un)] for w-verbs, where the second part of the long vowel /uu/
is transcribed as /w/; and [maF3iyy(-un)] for y-verbs from [maF3uuy(-un)]

203

by u-to-i which renders the long vowel into /ii/ whose second part is
transcribed as /y/. They are shortened to [maF3u] and [maF3i],
respectively by Final-Vowel Shortening in Cairo Arabic, as in these
passive participles:
Root
d-3-w
b-n-y
k-w-y

SA
mad3uww(-un)
mabniyy(-un)
makwiyy(-un)

EA
mad3u
mabni
makwi

Gloss
invited
built
ironed

The glides at the end of passive participles are underlying in EA, as

evidenced by the fact that they appear on the surface in feminine and
plural forms, e.g. /mad3uww-a/ invited (f) and /mabniyy-iin/ built
(pl).
The formation of the passive participle of the derived triradical as
well as the quadriradical verbs is rarely identical in the two varieties. The
Standard pattern is formed by placing the prefix [mu-] before the active
perfect form of the verb, so that the resultant form will be [mu-...aC]. This
form has been replaced by [mi-...aC] in EA, though the Colloquial variety
still preserves the Standard pattern for a number of forms. Compare Tables
(58) and (59).
Table (58)
Patterns of Derived Triradical Passive Participles in SA
No Verb Form
Passive Participle Example
II Fa33aL(-a)
mu-Fa33aL(-un) mu-kayyaf(-un) conditioned
III Faa3aL(-a)
mu-Faa3aL(-un) mu-raaqab(-un) watched
IV ?aF3aL(-a)
mu-F3aL(-un)
mu-jbar(-un) obliged
V taFa33aL(-a) mu-taFa33aL(-un) mu-tawaqqa3(-un) expected
VI taFaa3aL(-a) mu-taFaa3aL(-un) mu-tanaaqa(-un) discussed
VII (?i)nFa3aL(-a)

VIII (?i)Fta3aL(-a) mu-Fta3aL(-un) mu-ftaraD(-un) presupposed

IX (?i)F3aLL(-a)

X (?i)staF3aL(-a) mu-staF3aL(-un) mu-staxdam(-un) used

Table (59)
Patterns of Derived Triradical Passive Participles in EA
No Verb Form
Passive Participle Example
IIa Fa33aL
mu-Fa33aL
mu-kayyaf conditioned
"
mi-Fa33aL
mi-3a??ad complicated
IIb Fa33iL
mi-Fa33iL
mi-lawwin colored

204

III Faa3iL
IV ?aF3aL
Va (?i)tFa33aL
Vb (?i)tFa33iL
VI (?i)tFaa3iL
VII (?i)nFa3aL
VIII (?i)Fta3aL
IX (?i)F3aLL
Xa (?i)staF3aL
Xb (?i)staF3iL

mu-Faa3aL
mu-F3aL
mi-tFa33aL
mi-tFa33iL
mi-tFaa3iL

mu-Fta3aL

mu-staF3aL
"

mu-Taarad chased
mu-gbar obliged
mi-tdammar destroyed
mi-tsaggil recorded
mi-tSaahib befriended

mu-htaram respected

mu-stab3ad taken away

mu-sta3mal used

Comparing Tables (58) and (59), these remarks can be observed:

(a)

Forms VII and IX do not have a passive participle in either

variety because they are unaccusative by nature.
(b)
Forms IV, VIII and X have identical passive participles in both
varieties.
(c)
The prefix [mu-] is the usual passive participle prefix in both
varieties. However, it is replaced by [mi-] in EA for some verbs of
Form II and for verbs with the passive prefix [(?i)t-], i.e. Forms V
and VI.
(d)
The pattern [mi-Fa33iL] is used for both the active and passive
participles in EA.
As far as the passive participles of non-sound verbs are concerned,
the following notes can be recorded:
(a) For geminate verbs of Form IV, the SA passive participle is [mu-Fa33(un)] which is derived from [mu-F3a3(-un)] by Identical-Consonant
Metathesis. EA uses the pattern of Form II, i.e. [mi-Fa33i3] e.g. /mutamm(-un) > mi-tammim/ completed. For those of Form X, the variant of
the passive participle form [mu-staF3aL(-un)] is [mu-staFa33(-un)] which
is also derived from underlying [mu-staF3a3(-un)] by Identical-Consonant
Metathesis, as in /mu-stamadd(-un)/ taken, derived from.
(b) For initial-weak verbs of Form IV, the passive participle has the SA
pattern [muu3aL(-un)] which is derived from [mu-w3aL-un)] in w-verbs
by Perseverative Vocoid Assimilation and from [mu-y3aL(-un)] in y-verbs
by the special rule of y-to-w and Perseverative Vocoid Assimilation. This
pattern is substituted by the pattern of Form II [mu-wa33aL] in EA, e.g.
/muura(-un) > mu-warras/ bequeathed and /muuqaZ(-un)/ awakened
(SA only).

205

(c) For medial-weak verbs of Form IV, the passive participle is [mu-FaaL(un)] which is derived from underlying [mu-FGaL(-un)] by Anticipatory
Vocoid Assimilation, as in SA /mu-maat(-un)/ killed and /mu-baa3(-un)/
sold. For those of Form VIII, the passive participle is [mu-FtaaL(-un)]
which is derived from underlying [mu-FtaGaL(-un)] by Glide Elision, e.g.
/mu-xtaar(-un)/ chosen. For X, the pattern is [mu-staFaaL(-un)], which is
derived from [mu-staFGaL(-un)] by Anticipatory Vocoid Assimilation,
e.g. /mu-staTaa3(-un)/ made within the range of ability.
(d) Final-weak verbs have passive participles ending in [-a(n)] derived
from [-ay(-un)] by Glide Elision, Perseverative Vocoid Assimilation and
Closed-Syllable Shortening in SA indefinite forms and a long /aa/ in
definite forms where Closed-Syllable Shortening is not applicable. In EA,
the active participles of final-weak verbs of Forms II, V and VI are
generalized to cover passive participles as well. The passive participle of
Form IV is borrowed in EA with the shortening of its final vowel by FinalVowel Shortening. In Forms III and VIII, the passive verb formant [t-] is
applied to the verb before the prefixation of [mi-]. The passive participle
of Form X has disappeared in EA:
Form
II
III
IV
V
VI
VIII
X

Underlying
mu-Saffay(-un)
mu-naaday(-un)
mu-3fay(-un)
mu-tawaffay(-un)
mu-tafaaday(-un)
mu-taray(-un)
mu-stasqay(-un)

SA
EA
mu-Saffa(n) mi-Saffi
mu-naada(n) mi-tnaadi
mu-3fa(n)
mu-3fa
mu-tawaffa(n) mi-twaffi
mu-tafaada(n) mi-tfaadi
mu-tara(n) mi-tiri
mu-stasqa(n)

Gloss
purified
called
exempted
dead
avoided
bought
asked for rain

(e) For quadriradical verbs SA uses the passive participle pattern

[mu-Fa3L1aL2(-un)] which has its [mu-] replaced by [mi-] in EA, as in
/mu-zaxraf(-un) > mi-zaxraf/ decorated. Thus, we can propose that the
active participle of quadriradical verbs is generalized in EA to cover
passive participles as well. EA sometimes uses the passive participle of
quadriradical verb Form II as an alternative to that of Form I, e.g. /mitzaxraf/ decorated and /mi-ta3lil/ flared up.
By close examination of Tables (56) through (59) and the remarks
about non-sound passive participles, one can arrive at the following
generalizations:
(a) Although SA maintains a distinction between the active and passive
participles of derived verbs by using /i/ as a pre-final vowel for the former

206

and /a/ as a pre-final vowel for the latter, EA does not make such a
distinction for verb Forms II, V, and VI.
(b) The use of active or passive participles starting with [mu-] rather than
[mi-] is a characteristic of the speech of educated speakers in Cairo.
(c) The presence of two different prefixes in certain EA participles can be
attributed to the process of lexical diffusion discussed in (3.1.1).
(d) The passive participle of Form IV is giving way to that of form II in
EA. This is notable in geminate and initial-weak verbs. The motivation
behind this is that both verb forms are similar in meaning; both are
causative and while Form IV starts with a glottal stop, Form II does not.
This is also an indication that Form IV is on its way to be lost in EA.
(e) Because verb Forms V and VI are considered the passive equivalents of
Forms II and III in EA, their passive participle forms are also used for verb
Forms II and III, keeping forms beginning with [mi-] to the active
participles.
(f) The passive participle of Form I quadriradical is giving way to that of
Form II quadriradical in EA as the latter is having a passive prefix [(?i)t-]
in the verb.
(g) All cases of passive participle substitution between SA and EA can be
summarized as follows:
SA
IV
V
VI
I (quadri)

EA
IV ~ II
II
III
II (quadri)

4.7.3. Participles Inflection & Suffixation:

Both active and passive participles can inflect for gender and
number21 in both varieties. In relation to gender, participles are treated like
regular adjectives, i.e. when they are feminine, they are mapped on the
[+-at] noun class in SA. This noun class marker is reduced to [-a] in EA
(See 4.4). The final /t/ never appears in EA participles, which asserts our
assumption that they are adjectives. Compare the following examples:
m
3aarif(-un)
mu-garrib(-un)
maktuub(-un)
mu-saggal(-un)

SA f
3aarif-at(-un)
mu-jarrib-at(-un)
maktuub-at(-un)
mu-sajjal-at(-un)

EA f
3arf-a
mi-garrab-a
maktuub-a
mu-saggal-a

Gloss
knowing
experiencing
written
recorded

207

The change of the SA feminine active participle form [Faa3iL-at(-un)] into

EA [Fa3L-a] can be accounted for by the rules of [-a ~ -t] Suppletion,
High-Vowel Deletion and Closed-Syllable Shortening.
As far as number is concerned, SA participles can have dual and
plural forms, whereas EA does not have dual forms. EA uses only singular
and non-singular (i.e. both dual and plural). In other words, participles are
similar to adjectives regarding the inflection for number. The same
suffixes employed to mark number for adjectives are used for participles.
The use of the dual in SA (vs. plural in EA) can be seen in these examples:
SA
daaxil-aan, daaxil-ayn
mukallim-aan, mukallim-ayn
marsuum-aan, marsuum-ayn
musta3amal-aan, musta3amal-ayn

EA
daxl-iin
mi-kallim-iin
marsum-iin
mu-sta3mal-iin

Gloss
entering (du)
speaking to (du)
drawn (du)
used (du)

The use of the plural participles in both varieties can be observed

in these instances:
SA
Taalib-uun, Taalib-iin
mukassir-uun, mu-kassir-iin
mawjuud-uun, mawjuud-iin
mustaxdam-uun, mu-staxdam-iin

EA
Talb-iin
mi-kassir-iin
mawgud-iin
mu-staxdam-iin

Gloss
requesting (pl)
breaking (pl)
present (pl)
employed (pl)

The change of SA [Faa3iL-] into EA [Fa3L-] before plural suffixes

can be accounted for in regular phonology. The short vowel /i/ is deleted
in an open medial syllable by High-Vowel Deletion and the long vowel
/aa/ is shortened by the rule of Atonic Shortening because it is unstressed;
it is the vowel of the plural suffix which is stressed and there should be
only one stressed syllable in each phonological word.
Non-lexicalized participles are treated like verbs with regard to
objectival suffixes. Both receive direct and indirect object suffixes. For a
list of objectival suffixes see Table (34). The reason that participles are
treated like verbs with respect to agreement is because syntactically they
are verbs, except that they lack some of the inflectional categories that
verbs normally have (e.g. they do not receive the same subjectival affixes).
The objectival suffixes are actually clitics, that is to say, these affixes
represent the argument in the sentence and are moved to the verb or
participle by a syntactic rule. In EA, some morphophonemic changes may
occur as a result of the suffixation. These changes include the shortening
of unstressed long vowels by the rule of Atonic Shortening in this variety.

208

The reason is that any following long vowel or closed syllable would
normally attract the stress, making the preceding long vowels syllable
unstressed.
SA
aarii-him
3aamil-un la-naa
faaDil-un la-haa
mu-?ajjirat-un la-hu
mu-ntabihat-un la-kum

EA
arii-hum
3amil-li-na
faDil-la-ha
mi-?aggar-aa-1-u
mi-ntibha-1-kum

Gloss
buying (msg) them
doing (msg) for us
remaining (msg) for her
hiring (fsg) to him
paying attention(fsg) to you

It is interesting that the preposition [li-] is added into the word in EA,
while this is not so in SA. The [li-] plus pronoun combination in the
Colloquial variety forms part of the same word, as in verbs, for the
purpose of stress assignment for example.
4.8. Relational Adjectives:
A relational adjective is one that is derived from a noun to relate a
person or thing to that noun. It can sometimes be used as a noun to refer to
that person or thing. The general rule for the formation of such an
adjective in both varieties is the attachment of one of three relational
suffixes to the noun: [-iyy], [-aaw-iyy] or [-aan-iyy] with the last one as
the least frequent. EA shortens these suffixes in final position, in which
case the final geminate consonant is lost due to a regular phonological
change in which a rule of Final-Vowel Shortening applies twice [-iyy = -iii
-ii -i]. It also uses a fourth distinctive Colloquial marker [-gi].
To see how the first ending, [-iyy > -i#] is used in forming
relational adjectives, one can look at the following adjectives in the two
varieties:
N
?asyuuT
faransaa
taariix
3ilm

SA Adj
?asyuuT-iyy(-un)
farans-iyy(-un)
taariix-iyy(-un)
3ilm-iyy(-un)

EA Adj
Gloss
asyuuT-i
from Assiut
farans-i
French
tariix-i
historical
3ilm-i
scientific

In the case of feminine nouns marked by [-at], this marker is not part of
the base, so it should not be found when a marker is added, as in these
adjectives:
N

SA Adj

EA Adj

Gloss

209

handas-at(-un) handas-iyy(-un)
haqiiq-at(-un) haqiiq-iyy(-un)
Tabii3-at(-un) Tabii3-iyy(-un)
ziraa3-at(-un) ziraa3-iyy(-un)
agricultural

handas-i
geometrical
ha?ii?-ireal
Tabii3-i
natural
ziraa3-i

The suffix [-aawiyy > -aawi#] is used in both varieties for forming
relational adjectives from feminine nouns ending in [-aa], [-aa?] or, rarely,
[-a(t)]:
N
qinaa
samaa?(-u)
Sahraa?(-u)
bayD-at(-un)

SA Adj
qinaaw-iyy(-un)
samaaw-iyy(-un)
Sahraaw-iyy(-un)
bayDaaw-iyy(-un)

EA Adj
?inaaw-i
samaaw-i
Sahraaw-i
bayDaaw-i

Gloss
from Qena
heavenly
desert-like
oval

One might propose that [-aaw] is part of the stem in the examples above,
as final /?/ and /t/ are replaced by /w/, but then it will be difficult to
account for the appearance of /w/ in the following examples:
N.
?al-garbiyy-at(-u)
?a-arqiyy-at(-u)
?abu-za3bal(-u)
?al-?ahlii

SA Adj
garabaaw-iyy(-un)
arqaaw-iyy(-un)
za3balaaw-iyy(-un)
?ahlaawiyy(-un)

EA Adj
Gloss
garabaaw-i
from Gharbiya
ar?aaw-i
from Sharqiya
za3balaaw-i from Abuzaabal
?ahlaaw-i
supporting
Al-Ahli
football team

The third relational marker that is used in both varieties is [-aaniyy > -aan-i#]. It is more common in EA than in SA. It is employed with
noun forms, adjective forms and some adverb forms (Ghaly 1960: 59), as
in the following examples:
Word
rabb(-un)
haqq(-un)
?asmar(-u)
?awwal(-u)
taht(u)
fawq(u)

SA Adj
rabbaan-iyy(-un)
haqqaan-iyy(-un)
?asmaraan-iyy(-un)
?awwalaan-iyy(-un)
tahtaan-iyy(-un)
fawqaan-iyy(-un)

EA Adj
rabbaan-i
ha??aan-i
?asmaraan-i
?awwalaan-i
tahtaan-i
fu?aan-i

Gloss
godly
righteous
darkish
first
lower
higher

The Colloquial variety has a distinctive relational marker which is

borrowed from Turkish, namely [-gi]:

210

EA N
sufra dining-room
?ahwa caf
makwa an iron
3arabiyya a cart

EA Rel. Adj
sufra-gi waiter
?ahwa-gi waiter
makwa-gi ironer
3arba-gi a cart-driver22

In both varieties, a feminine relational adjective ends in [-iyy-at(-un)]; the

[-yy] appears in EA because the environment causing shortening is no
longer found, i.e. affixation of [-a] prevents Final-Vowel Shortening from
applying, as in these examples:
msg (SA)
?ajnab-iyy(-un)
suudaan-iyy(-un)
Sahraaw-iyy(-un)
tahtaan-iyy(-un)

SA fsg
?ajnab-iyy-at(-un)
suudaan-iyy-at(-un)
Sahraaw-iyy-at(-un)
tahtaan-iyy-at(-un)

EA fsg
?agnab-iyy-a
sudan-iyy-a
Sahraw-iyy-a
tahtan-iyy-a

Gloss
foreign
Sudanese
desert-like
lower

When a relational adjective is pluralized, the SA form [-iyy] is preserved

as is in EA, again because affixation of the sound plural suffix prevents
Final-Vowel Shortening from applying, as follows:
SA pl
suur-iyy-uun
farans-iyy-uun
haqqaan-iyy-uun

EA pl
sur-iyy-iin
farans-iyy-iin
ha??an-iyy-iin

Gloss
Syrian
French
righteous

Some relational adjectives have a broken plural form in both varieties:

SA sg
?injliiziyy(-un)
3arabiyy(-un)
?iiTaaliyy(-un)
?amriikiyy(-un)

EA sg
?ingliizi
3arabi
?iTaali
?amriiki

Plural
Gloss
?ingliiz
English
3arab
Arab
Talyaan
Italian
?amriikaan >?amrikaan
American

To pluralize feminine relational adjectives, the feminine suffix [-aat] is

employed in SA but the masculine (or rather unmarked) suffix [-iin] is
used in EA:
fsg
lubnaaniyy-at(-un)
?urduniyy-at(-un)
rawhaaniyy-at(-un)

SA fpl
lubnaaniyy-aat(-un)
?urduniyy-aat(-un)
rawhaaniyy-aat(-un)

EA pl
lubnaniyy-iin
?urduniyy-iin
rawhaniyy-iin

Gloss
Lebanese
Jordanian
spiritual

211

One EA exception here is the adjective /maSriyy-aat/ Egyptian (fpl)

which occurs optionally with /maSriyy-iin/ as a plural of the feminine
adjective /maSriyy-a/, but this can be considered a direct borrowing from
SA as argued by McGuirk (1986: 22).
Hence, in both varieties relational adjectives agree in gender and
number with their modified nouns if those nouns have human reference,
except for the dual and feminine plural adjectives which are not found in
EA. Compare the SA examples in (211-212a) with their EA equivalents in
(211-212b):
(211) a. daxal-a
rajul-aani
enter.pf-3msg
man-Nom.du
Two French men entered.
b. ragl-een
faransiyy-iin
man-du
French-pl
Two French men entered.
(212) a. ra?ay-tu
rajul-ayni
see.pf-1sg
man-Acc.du
I saw two French men.
b. uf-t
ragl-een
see.pf-1sg
man-du
I saw two French men.

faransiyy-aan
French-Nom.du
daxal-u
enter.pf-3pl

faransiyy-ayn
French-Acc.du
faransiyy-iin
French-pl

However, if the modified nouns are of non-human reference, singular

feminine relational adjectives are employed to modify plural nouns, as in
(213-214a) for SA and (213-214b) for EA:
(213) a. ?aT-Turuq-u
z-ziraa3iyy-ah / *z-ziraa3iyy-uun
the-road.br.pl-Nom the-agricultural-fsg / *the-agricultural-mpl
The agricultural roads.
b. ?iT-Turu?
iz-ziraa3iyy-a / *iz-zira3iyy-iin
the-road.br.pl
the agricultural-fsg / *the-agricultural-pl
The agricultural roads.
(214) a. ?al-kulliyy-aat-u
n-naZariyy-ah / *n-naZariyy-aat
the college-fpl-Nom the-theoretical-fsg / *the-theoretical-fpl
The theoretical colleges.
b. ?il-kulliyy-aat
in-naZariyy-a / *in-naZariyy-iin
the college-fpl
the-theoretical-fsg /*the-theoretical-pl
The theoretical colleges.

212

In EA, there is a general tendency for relational adjectives of origin

to be invariable, i.e. to use their masculine singular form to modify nonhuman feminine singular nouns or broken plural ones. Compare the SA
phrases with their Colloquial equivalents in these examples:
(215) SA: sijjaad-at-un
?asyuuTiyy-ah
(a) carpet-f-Nom
from Assiut-fsg
A carpet from Assiut.
EA: siggaad-a
?asyuuT-i / *?asyuT-iyy-a
(a) carpet-f from Assiut-msg / *from Assiut-fsg
A carpet from Assiut.
(216) SA: qumSaan-un
?amriikiyy-ah
shirt.br.pl-Nom American-fsg
American shirts.
EA: ?umSaan
?amrikaan-i / *?amrikan-iyy-a
shirt.br.pl
American-msg / *American-fsg
American shirts.
Thus, one can propose a special rule for the agreement between nouns and
adjectives with the relational marker [-i] in EA:
(217) [hum] [-at].
This rule will prevent the addition of the feminine marker [-at] to
adjectives modifying non-human broken plural nouns.

213

Chapter Five
Closed-List Class Morphology
5.0. Introduction:
The morphology of some closed-list classes in SA and EA will be
dealt with in this chapter. A treatment of pronouns (personal, relative and
demonstrative) and prepositions will be handled in (5.1 and 5.2). After
that, adverbs will be discussed in (5.3). In addition, interrogative and
responsive particles will be included in (5.4). Finally, negative and
possessive particles will be compared in the two varieties in (5.5 and 5.6),
respectively. Reference will not be made to the particles which were
discussed in the preceding chapters, including the future particles /sa-,
sawfa > ha-, ha-/ and the Cairene progressive particle /bi-/.
5.1. Pronouns:
Arabic pronouns are words that are used instead of nouns; they fill
their slots in syntactic structures and inflect for their obligatory
morphological categories (i.e. case, gender and number). The Arabic
pronominal system comprises three main types of pronouns: personal,
relative and demonstrative. EA has greatly reduced the number of
pronominal forms employed in SA. This can be seen when the three types
of pronouns are examined in the two varieties.
5.1.1. Personal Pronouns:
Arabic personal pronouns are of two kinds: free and bound. The
bound personal pronouns were treated as subjectival and objectival affixes
in (2.4). The free personal pronouns are also of two types: subject and
object pronouns. Except for the dual pronouns and the feminine plural
pronouns, found only in SA, the Colloquial variety retains all other
personal pronouns. The personal pronouns which are usually used in
subject position are listed in Table (60).
Table (60) shows that the dual category has disappeared in EA.
Also, this variety does not comprise any feminine plural pronouns.
However, it should be asserted in this respect that since the Colloquial
variety does not have dual pronouns, it extends the plural pronouns to refer
to dual subjects (cf. Abdel-Hafiz 1994: 64). So, it is more plausible to
assume that the EA plural pronouns are not plural (in the sense of plural as
opposed to dual, as in SA), but are non-singular (i.e. both plural and dual).

214

Table (60)
Subject Personal Pronouns in SA & EA
Referent
SA
EA
1sg (m/f)
?ana
?ana or ?aana
1pl (m/f)
nahnu
(?i)hna
2msg
?anta
(?i)nta
2fsg
?anti
(?i)nti
2du (m/f)
?antumaa

2mpl
?antum
(?i)ntu
2fpl
?antunna

3msg
huwa
huwwa
3fsg
hiya
hiyya
3du (m/f)
humaa

3mpl
hum
humma
3fpl
hunna

Likewise, as the Colloquial variety does not include feminine plural

pronouns, it expands the masculine plural pronouns to cover feminine
plural subjects as well (cf. Abdel-Hafiz 1994: 64). Thus, it is better to
propose that the EA plural pronouns are actually genderless, like English
you and they. In terms of underlying forms, it might be proposed that
the difference between SA and EA is that for the first person plural and all
second person pronouns EA has lost the first syllable of the SA forms, i.e.
[na-] or [?a-]. Then, an initial /i/ is inserted by Word-Initial Epenthesis,
and a /?/ is inserted by Glottal Stop Insertion. On this analysis, the
differences between SA and EA will be as follows:
Referent
1pl
2msg
2fsg
2du
2mpl
2fpl

SA
nahnu
?anta
?anti
?antumaa
?antum
?antunna

EA
hna ?ihna
nta ?inta
nti ?inti

ntu ?intu

Phonologically, the Colloquial personal pronouns are somewhat

similar to those of the Standard variety. Three apparent phonological
differences distinguish EA from SA. In the first place, the Colloquial
variety inserts the first vowel /i/ in four forms after the deletion of the first
syllable having /a/, and replaces the final /u/ of /nahnu/ we with /a/. The
forms which retain their first vowel are /?ana/ I, /huwa/ he, /hiya/ she
and /hum/ they (m). Second, one form in EA displays a lack of

215

consonant correspondence as a result of the loss of final /m/ in /?antum/

you (mpl) having the reflex /?intu/. Finally, the Colloquial third person
singular and plural pronouns include a gemination of the second consonant
of their Standard counterparts. Thus, /huwa/ he, /hiya/ she and /hum/
they have become /huwwa/, /hiyya/ and /humma/, respectively.
The relationship between independent subject pronouns and the
subject agreement markers given in Table (30) has been discussed by
Abdel-Hafiz (1991: 181). He has proved that Arabic subject agreement
markers are diachronically developed from the independent subject
pronouns. The adequacy of this hypothesis is confirmed by the fact that
the agreement forms can co-occur with the elements (nouns or pronouns)
they refer to. In other words, the subjectival affixes shown in Table (30)
are derived by diachronic development from the free subject pronouns
presented in this section.
The object personal pronouns are no longer found in EA. They are
only present in the Standard variety, as displayed in Table (61).
Table (61)
Object Personal Pronouns in SA
Referent
Pronoun
1sg (m/f)
?iyyaa-ya
1pl (m/f)
?iyyaa-na
2msg
?iyyaa-ka
2fsg
?iyyaa-ki
2du (m/f)
?iyyaa-kumaa
2mpl
?iyyaa-kum
2fpl
?iyyaa-kunna
3msg
?iyyaa-hu
3fsg
?iyyaa-ha
3du (m/f)
?iyyaa-humaa
3mpl
?iyyaa-hum
3fpl
?iyyaa-hunna
There is a strong relation between the object personal pronouns and
the objectival affixes (discussed in 2.4). Abdel-Hafiz (1991: 180) has
proved that Arabic objectival affixes which he calls bound object
pronouns are phonologically derived from their equivalent free object
pronouns. He has shown the scenario they are developed as follows:
Having been desemanticized as object clitics, the independent object
pronouns were subject to a phonetic process (i.e. syllabic erosion) such

216

that their initial syllables are eroded; ?i-yaa-ka > -ka; ?i-yaa-kum > -kum,
etc. Thackston (1984: 122), on the other hand, proposes that /?iyyaa-/ is
a pronominal carrier, i.e. it is just a dummy stem used to support an
object clitic when the object clitic has nothing else to attach to. According
to this analysis the free object pronouns can be seen to consist of the
dummy stem plus the normal objectival affixes of SA. To characterize the
difference between SA and EA, we can then propose not that EA lacks the
free object pronouns (as was assumed at the beginning of the preceding
paragraph) but rather that it lacks the dummy stem to which the objectival
affixes attach.
5.1.2. Relative Pronouns:
The Standard system of relative pronouns is relatively complex due
to the existence of several forms depending on the number, gender and
case of the relativized nominal. This system is manifested in Table (62).

Gender
Masc
Fem

Table (62)
Relative Pronouns in SA
Singular
Dual
?alla-ii
?alla-aani [Nom]
?alla-ayni [Acc, Gen]
?alla-tii
?alla-taani [Nom]
?alla-tayni [Acc, Gen]

Plural
?alla-iina
?alla-atii/
?alla-a?i

Table (62) shows that in the Standard variety, a relative pronoun

involves two morphemes, the first of which is [?alla-]. EA, on the other
hand, has developed an extremely simplified system of relativization,
including only one morpheme /?illi/. So, the Colloquial system is not only
free from inflectional suffixes, but also from the number and gender
distinctions observed in other nominal and verbal forms.
To explain the reduction of the complex Standard relative
pronouns into one Colloquial form, Abdel-Hafiz (1994: 53-54) argues that
the second morpheme in all Standard relative pronouns is possibly felt to
be of no significance in communication. The number/gender information
conveyed by this morpheme is redundant since this information is
already contained in the relativized nominal itself. This explains why the
[loss of] number/gender information-carrying morphemes was able to
take place in EA. To account for the change of SA /?alla-/ to EA /?illi/,
one can propose that the change is from underlying SA [alla-] to EA [lli]

217

with /i/ epenthesized for syllabic reasons and /?/ inserted, in both varieties,
by Glottal Stop Insertion.
Other relative pronouns exist in SA, among which only /man/
who and /maa/ which are frequently used. The former is mainly
associated with human beings, whereas the latter is connected with other
objects (i.e. animals or things). Both are also replaced by [?illi] in EA.
5.1.3. Demonstrative Pronouns:
In general, demonstratives agree with the number and gender of the
object referred to in SA and EA. In both varieties, demonstratives can be
classified into two groups according to the distance of the object(s)
referred to: near demonstratives (Table (63)) and distant demonstratives
(Table (64)).
Table (63)
Near Demonstratives in SA & EA
Referent
SA
EA
3msg
haa-a
da(h)/dawwa(t)
3fsg
haa-ihi
di/diyya(t)
3mdu
haa-aani [Nom]

haa-ayni [Acc, Gen]

3fdu
haa-taani [Nom]
dool/doola(t)
haa-tayni [Acc, Gen]

3pl
haa-?ulaa?i23

Table (64)
Distant Demonstratives in SA & EA
Referent
SA
EA
3msg
aalika
dukha
3fsg
tilka
dikha
3pl
?ulaa?ika
dukham
In the Standard variety, near demonstratives are formed by two
morphemes, the first of which is the introductory element [haa-] that is
normally used for calling attention. EA has modified the system of near
demonstratives by eliminating the introductory morpheme [haa-] and by
reducing the number of these demonstratives. Consistent with the
neutralization of the dual and plural in verbs, adjectives and pronouns, the
dual demonstratives have been lost. The plural demonstrative /dool/ has
been extended to cover both dual and plural. So, it is more convenient to

218

call it non-singular. Robertson (1970: 198) shows how this particular

demonstrative developed from SA: The independent particle /a, i/
combined with the plural /?ulaa?i/ to form /dool/ by assimilation as
follows: [a?ulaa?i > a?ula > a?ul > daul > dool]. It should be noted
here that the third person plural demonstratives are genderless in both
varieties. It might also be observed that the morpheme [-a(t)] is not a
feminine marker in EA near demonstratives. Rather it is added optionally
to these demonstratives, and the /t/ is sometimes replaced by /k/ or /n/,
resulting in several forms for each. Singular forms with [a(t)] are derived
by the combination of the underlying demonstratives /da/ and /di/ with the
personal pronouns /huwwa/ and /hiyya/, respectively, as follows:
(218) a. da+huwwa da+uwwa daawwa dawwa
b. di+hiyya di+iyya diyya
The deletion of the /h/ at the beginning of the personal pronoun can be
accounted for by a special rule as follows:
(219) h-Deletion in EA:
h/VV

in near demonstratives

Then, /a+u/ in (218a) becomes /aa/ by Perseverative Vocoid Assimilation.

Finally, Closed-Syllable Shortening applies in both masculine and
feminine forms to shorten the first vowel. Another Colloquial feature of
innovation in both near and distant demonstratives is the replacement of
the interdental fricative // by the alveolar stop /d/. This is one of the
regular sound changes, as indicated in (1.1.1).
In SA, the introductory morpheme [haa-] may combine with the
third person personal pronouns to form a third set of demonstratives that
were termed deictic forms by Mitchell (1962: 58). They are used for
drawing attention or for emphasis. In EA, the morpheme [haa-] is reduced
to [?a-] and combined with the personal pronouns /huwwa/, /hiyya/ and
/humma/ or, more commonly, with the first syllable of them to form the
equivalent of these demonstratives. Compare the deictic demonstratives in
Table (65):

219

Referent
3msg
3fsg
3du(m/f)
3mpl
3fpl

Table (65)
Deictic Demonstratives in SA & EA
SA
EA
haa-huwa
?a-hu/?a-huwwa(t)
haa-hiya
?a-hi/?a-hiyya(t)
haa-humaa

haa-hum
?a-hum/?a-humma(t)
haa-hunna

It can be noticed here that EA /?a-hum/?a-humma(t)/ is both genderless

and non-singular.
According to El-Tagoury (1989: 43), these deictic forms agree in
number and gender with the demonstratives and with the following noun
heads. Compare the Standard examples in (220-223a) with their Colloquial
parallels in (220-223b):
(220) a. haa-huwa
aa
here-it(m)
this(m)
Here is my pen.
b. ?a-hu
da
here-it(m)
this(m)
Here is my pen.

qalam-ii
pen-1msg

(221) a. haa-hiya
ii
here-it(f)
this(f)
Here is my ball.
b. ?a-hi
di
here-it(f)
this(f)
Here is my ball.

kurat-ii
ball-1msg

(222) a. haa-hum
?ulaa?i
here-they(m) these
Here are my sons.
b. ?a-hum
dool
here-they
these
Here are my sons.

?awlaad-ii
sons-1msg

?alam-i
pen-1msg

kurt-i
ball-1msg

?awlaad-i
sons-1msg

(223) a. haa-hunna ?ulaa?i

banaat-ii
here-they(f) these
daughters-1msg
Here are my daughters
b. ?a-humdool
banaat-i

220

here-they
these
Here are my daughters.

daughters -1msg

5.2. Prepositions:
EA preserves most of the widely used SA prepositions in identical
or modified phonemic shapes (See Table 66). Those which are normally
lacking include the following: /?ilaa/ to, until, /hattaa/ until, up to /ka-,
mila/ as, like, and /munu/ since. These prepositions are replaced by
other prepositions denoting similar meanings. Other differences related to
the use or the phonemic shape of the prepositions are summarized below:
(i)

Prepositions ending in a long /aa/ in SA have that vowel

shortened before the definite article. This is due to ClosedSyllable Shortening; if the /l/ in the definite article must form
the coda of the preceding syllable, then the vowel must shorten
to avoid an over-long syllable, as in (224):
(224) a. /?ila 1-madrasat-i/ to the school
b. /hatta n-nihaayat-i/ up to the end
c. /3ala 1-maktab-i/ on the desk

(ii)

The preposition /bi-/ sometimes combines with the negative

particle /laa/ no to form /bi-laa/ without in SA which is rendered
/ba-la/ in EA, e.g. /bi-laa maa? > ba-la mayya/ without water. This
same preposition has merged with the verb /jaa?(-a)/ to come to
form one verb in EA. Thus, /jaa?-a bi > gaab/ he came with, he
brought.

1
2
3
4
5
6
7
8
9
10

SA
fii
libimin, minn3an, 3annma3a
3alaa, 3alay?ilaa, ?ilay3inda
ladaa

Table (66)
Prepositions in SA & EA
EA
Gloss
fi, f-, fii
in
li-, l-, lii
to, for
bi-, b-, bii
in, with, by
min, minn-, mifrom
3an, 3annfrom, about
ma3a(a), wayya
with
3ala, 3alee-, 3aon, over, against
ligaayit, lihadd
to, until
3and
with, at

with, at

221

11
12
13
14
15
16
17
18
19
20
21
22
23
24

gayra
kamila
fawqa
tahta
?amaama
quddaama
waraa?a
hatta
qabla
ba3da
qadra
bayna
hawla

25
26
27
28
29
30
31
32
33
34
35
36
37
38

hasba
bi-laa
janba
nahwa
qubaalat-a
daaxil-a
xaarij-a
Didda
3aksa
wa
munu
duuna, biduun
min ?ajl(i)
badal-an min

geer
kazayy
fooq
taht
?uSaad
?uddaam
wara
ligaayit, lihadd
?abl
ba3d
?add
been
hawaleen, hawalee-,
hawalayhasab
ba-la
gamb
nahyit, yammit
?ubaal
guwwa
barra
Didd
3aks
w(a)
min
min geer
3aaan
badal

other than
as, in the capacity of
like, similar to
over, on
under, below
in front of
in front of
behind
until, up to
before
after
as much as
between, among
around
according to
without
beside
towards
opposite
inside
outside
against
opposite of
by (in an oath)
since
without
for the sake of
instead of

(iii) Each of the prepositions /min/ from and /3an/ from, about has two
allomorphs in both varieties: a free one which has a single phoneme
at the end and is used before words or suffixes starting with
consonants, and a bound one which has a geminate phoneme at the
end and is used before pronominal suffixes beginning with vowels.
The bound form for /min/ is /minn-/, and that for /3an/ is /3ann-/.
Compare both prepositions before consonants and before vowels:

222

SA
min bayti-ka
3an bayti-ka
minn-ii
3ann-ii
min-ka
3an-ka

EA
min beet-ak
3an beet-ak
minn-ii
3ann-ii
minn-ak
3ann-ak

Gloss
from your house
about your house
from me
about me
from you
about you

The preposition /min/ sometimes loses its final phoneme in EA,

particularly before the definite article, e.g. /mi l-beet/ from the
house. The reason is that the preposition plus following complement
form a single domain for syllabification. In SA, the vowel /a/ in the
definite article acts as a linking vowel that breaks the three-consonant
cluster, i.e. /min al-bayt(-i)/. But in EA there is no linking vowel
available; so, one way to escape the three-consonants-in-a row
problem is to delete the /n/ by the Stray Erasure Convention which
says: erase segments and skeleton slots unless attached to higher
levels of structure (Trask 1996: 336).
(iv) The prepositions /?ilaa/ and /3alaa/ have two allomorphs each in SA:
/?ilaa vs. ?ilay-/ and /3alaa/ vs. /3alay-/. So, we can propose that
underlying forms are /?ilay/ and /3alay/ but before a word boundary
the /ay/ is changed to /aa/ by this special rule:
(225) ay-to-aa:
ay aa / #
The preposition /?ilaa/ is not found in EA. The allomorphs of /3alaa/
have their reflexes in EA: /3ala/ by Final-Vowel Shortening and
/3alee-/ by Monophthongization. In both varieties, the latter
allomorph occurs before some particles and all suffixes with the
exception of the first person singular suffix [-ya] to which /3alay-/ is
preposed. The fact that the alternant /3alay-/ appears before [-ya] in
EA is predictable from the Monophthongization rule: /ay/ becomes
/ee/ provided that /y/ is not wholly or partly in the syllable onset. In
/3alay-ya/ the /y/ is geminate. So, Monophthongization cannot apply
due to Geminate Inalterability. Compare the following:

223

SA
3alaa maktabii
3alay-naa
3alay-kum
3alay-ya

EA
3ala maktabi
3alee-na
3alee-kum
3alay-ya

Gloss
on my desk
on us
on you (mpl)
on me

However, EA often reduces the morpheme /3alaa/ to /3a/ before the

definite article: /3a l-maktab/ on the desk /3a k-kursi/ on the chair
and /3a r-ramla/ on the sand.
(v) Prepositions of the pattern [CVVC] have their vowel shortened by
Atonic Shortening in EA when they lose their major stress as a
result of the addition of a suffix which has a long vowel or is of the
pattern CV... (Aboul-Fetouh 1969: 112), as in /geer+na gir-na/
other than us and /been+hum bin-hum/ between them.
(vi) The SA prepositions /bi-, fii, li-/ have four variants each in EA. The
first is with a short /i/ and is used before words beginning with
consonants. The second variant is formed by the deletion of the final
vowel /i/ and is utilized before words starting with a vowel preceded
by an inserted /?/. The third variant is formed by lengthening the
vowel and is employed before pronominal suffixes beginning with a
consonant. The fourth variant is constituted by adding /y/ to the
vowel before the first person objectival suffix [-ya]. Compare the
following:
SA
bi-awk-at(-in)
fi--anT-at(-i)
li-l-walad(-i)
bi-?ahmar
fii ?ajaaz-ah
li-?abii-ka
bi-ka
fii-hi
la-naa

EA
bi-ook-a
fi--anT-a
li-l-walad
b-ahmar
f-agaaz-a
l-abuu-k
bii-k
fii-h
lii-na

Gloss
with a fork
in the bag
to the boy
with red
on holiday
for your father
with you
in it
for us

Thus, one can propose that in EA, these propositions are underlyingly
[bi-, fi- and li-]. The vowel there is deleted before words beginning
with vowels, and lengthened before consonantal suffixes. For
instance, /b-ahmar/ is underlyingly [bi+ahmar] which is changed to
/baahmar/ by Anticipatory Vocoid Assimilation, then to /b-ahmar/ by

224

Closed-Syllable Shortening. And /f-agaaz-a/ is underlyingly

[fi+agaaz-a] that is changed to /faagaaz-a] by Anticipatory Vocoid
Assimilation, then to /f-agaaz-a/ by Atonic Shortening. The
lengthening occurring before consonantal suffixes can be accounted
for by a special rule of the following form:
(226) Prepositional Lengthening in EA:
Vi / Vix] + C (where x is [bi-], [li-] or [fi-]
and C is in the beginning of a suffix).
This rule indicates that we lengthen the vowel in the prepositions
[bi-], [li-] and [fi-] before a consonantal suffix.
(vii) The preposition /ma3a/ has its final /a/ lengthened in EA before
suffixes beginning with consonants, e.g. /ma3-ii > ma3aa-ya/ with
me, /ma3a-ka > ma3aa-k/ with you and /ma3a-naa > ma3aa-na/
with us. Similarly, the Colloquial preposition /wara < waraa?a/
behind has its final vowel lengthened before consonantal suffixes,
e.g. /waraa?-ii > waraa-ya/ behind me, /waraa?a-ka > waraa-k/
behind you and /waraa?a-naa > waraa-na/ behind us. An
alternative approach is to propose that EA /ma3a/ and /wara/ are
underlyingly /ma3aa/ and /waraa/ and that they have their final /aa/
shortened by Final-Vowel Shortening.
(viii) In SA, no preposition can occur between a base and a pronominal
suffix. In EA, the only preposition that can come between a base and
an objectival suffix is [l-]. According to Abdel-Malek (1972: 85), this
preposition has four allomorphs in this position:
1- /-li-/ between a base ending in one consonant and [-na] us.
2- /-la-/ between a base ending in one consonant and [-ha] her.
3- /-lu-/ between a base ending in one consonant and [-kum] you
(pl)
or [-hum] them.
4- /-1-/ between any other base and any other pronominal suffix.
In this allomorphy, it seems clear that vowel harmony has a great role
to play. So, one assumption is that this preposition has two
allomorphs in EA: /lV/ (where V is a copy of the vowel of the suffix
or a harmonic vowel whose phonological quality is determined by
context) occurring before all consonant-initial suffixes, and /l/
occurring before vowel-initial suffixes. In the case of the suffix [-ha],
a plausible assumption is that [-a] is a ghost vowel associated with

225

this suffix and occupies the position of the V in the preposition /lV/.
In other cases, the vowel position is realized as a high vowel
harmonic with the vowel of the suffix (See 1.2.1).
The base occurring before the preposition /li-/ in such contexts can be
a verb, an active participle, a passive participle or an elative adjective.
In SA, the preposition /li-/ will be separated from the base and
prefixed to the pronominal suffix with one allomorph, /la-/. Compare
these examples:
SA
katab-a la-naa
kaatib-un la-haa
maktuub-un la-kum
mab3uu-un la-hum
?ahsan-u la-hu

EA
katab-li-na
katib-la-ha
maktub-lu-kum
mab3ut-lu-hum
?ahsan-l-u

Gloss
he wrote to us
writing to her
written for you
sent to them
better for him

(viii) The Colloquial preposition /hawaleen/ around has two bound

allomorphs before pronominal suffixes: /hawalay-/ before the first
person pronominal suffix [-ya], and /hawalee-/ before all other
pronominal suffixes (Abdel-Malek 1972: 82). So, one can propose
that the underlying form is [hawalayn] whose /ay/ is changed to /ee/
by Monophthongization and whose /n/ is deleted, when not
associated to any syllable, by the Stray Erasure Convention.
Compare these SA phrases with their EA analogues:
SA
hawla l-3aalam
hawl-ii
hawla-naa
hawla-hum

EA
hawaleen il-3aalam
hawalay-ya
hawalee-na
hawalee-hum

Gloss
around the world
around me
around us
around them

At the end of the discussion of prepositions, it can be asserted that

two prepositions may occur in succession in both varieties. The meaning
of the compound preposition will be composed of the combined meanings
of its two elements. Most possible combinations have /min/ from as a
first preposition and a preposition of place or time as a second, e.g. /min
ba3d/ from after and /min taht/ from under. The preposition /ragm/
instead of also combines with /bi-/ to form two compound prepositions:
/bi-ragm(-i)/ and /bi-r-ragm(-i) min/ both mean also in spite of. Other
combinations have /gayra > geer/ other than as a second preposition, /3an
gayr(-i) > 3an geer/ about other than.

226

5.3. Adverbs:
The existence of adverbs in Arabic is a controversial issue. For
instance, Haywood & Nahmad (1982: 426) assert that Arabic has no
adverbs, properly speaking, but this lack is hardly felt owing to the
inherent flexibility and expressiveness of the language. But many other
linguists, including Aboul-Fetouh (1969) and Salib (1981), assert that
there are adverbs in Arabic. Salib (1981: 285) defines adverbs in Arabic as
words which add circumstantial information to the sentences in which
they are used, that is, they indicate when, where, how, how often, etc., the
action, event, or state involved takes place.
The question of whether there are adverbs in Arabic or not is really
two different questions. First, there is the question of whether there is a
syntactic category of adverb; in other words, do the syntactic rules of the
language need to make reference to adverbs as opposed to nouns,
adjectives, etc. Second, is there a morphological category of adverb; in
other words, do the morphological and phonological rules of the language
treat adverbs as a special class for the purposes of word formation rules
and phonology? Haywood & Nahmad (1982) were speaking of SA,
whereas Aboul-Fetouh (1969) and Salib (1981) were speaking of EA. So,
perhaps the correct answer is that SA has the syntactic category of adverb,
but not the morphological one. As will be shown below, the
(morphological) form of nouns in the accusative case can have the
(syntactic) function of adverbs in SA. Nevertheless, the existence of
special forms in EA which do not have the form of nouns suggests that
perhaps adverb is becoming a morphological class in EA. To put the
matter in a nutshell, the syntactic category or function of adverb is found
in both varieties, whereas the morphological category or form of adverb is
found in EA only.
Whether adverbs represent a closed-list class or not is another
controversial issue. Though Abdel-Malek (1972) treats them as a closed
set, one might argue that since some types of adverbs can be derived from
adjectives or nouns (See 1.3.2), every time an adverb is formed a new
member is added to the class; which makes it difficult to classify adverbs
as a closed class. They are included in this chapter for organizational
purposes only.
There are a number of ways of expressing the function of adverbs
in Arabic. First, in SA the function of adverbs is achieved by nouns
expressing place, time or manner and placed in the accusative case, e.g.
/Sabaah-an/ in the morning, /bahr-an/ by sea, and /sarii3-an/ quickly.

227

Some of these adverbs are preserved in EA, as will be shown in Tables

(67-71). This can be accounted for by proposing that those adverbs are
lexicalized as such before being borrowed in the Colloquial variety. In
other words, Cairene people believe that the suffix is part of the adverb
and not an additional affix.
Second, in both varieties adverbs may be expressed by
prepositional phrases which are formed by a preposition followed by a
noun, e.g. /fii ?aS-Sabaah(-i) ~ fi-S-Sabaah(-i) > fi S-Subh/ in the
morning, /fii ?al-bahr(-i) ~ fi l-bahr(-i) > fi l-bahr/ in the sea and /bisur3-at(-in) > bi-sur3-a/ quickly.
Third, some SA adverbs are non-declinable nouns referring to
place or time and having the suffix [-u]. When these nouns end in [-a] they
are treated as prepositions (See Table 66). For instance, /tahtu/
downstairs is an adverb, but /tahta/ under is a preposition. That is why
Haywood & Nahmad (1982: 393) assert that Many so-called Particles or
Prepositions are really Adverbial Accusatives of nouns, at least in origin.
EA retains most of these adverbs after cutting of the suffix [-u]. (See
Tables 67-68).
Table (67)
Adverbs of Place in SA and EA
SA
EA
Gloss
1 quddaamu
?uddaam
in front of
2 fawqu
foo?
up, upstairs
3 tahtu
taht
down, downstairs
4 bi-d-daaxil
guwwa
inside
5 bi-l-xaarij
barra
outside
6 hunaa
hina
here
7 hunaaka
hinaak
there
8 amma

there
9 yamiin-an
3a-l-yimiin
on the right
10 imaal-an
3a--imaal
on the left
Table (67) shows that some SA adverbs of place are replaced by
suppletive counterparts in EA because they are no longer found in this
variety. These include /bi-d-daaxil/ inside and /bi-l-xaarij/ outside.
Some nouns are put in the accusative to function as adverbs of place in
SA. These are replaced by prepositional phrases in EA, e.g. /yamiin-an >
3a-l-yimiin/ on the right and /imaal-an > 3a--imaal/ on the left.

228

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

Table (68)
Adverbs of Time in SA & EA
SA
EA
Gloss
?awwal-an
?awwal-an
firstly
... 3aair-an
... 3aair-an
... tenthly
?axiir-an
?axiir-an
finally
daa?im-an
daym-an
always
haal-an
haal-an
soon
?ahyaan-an
?ahyaan-an
sometimes
?abad-an
?abad-an
never
Sabaah-an
Sabaah-an
in the morning
masaa?-an
masaa?-an
in the evening
3aadat-an
3aadat-an
usaully
mu?aqqat-an muwa??at-an temporarily
~ mu?aqqat-an
taarat-an
taara ~ taarat-an sometimes
yawm-an
fi yoom
one day
nahaar-an
bi-n-nahaar
by day
layl-an
bi-l-leel
by night
?amsi
?imbaarih
yesterday
gad-an
bukra
tomorrow
qariib-an
?urayyib
soon
mubakkir-an badri
early
muta?axxir-an waxri
late
?al-?aan-a
d-il-wa?t(i)
now
mubaaarat-an Tawwaali
directly
ba3du
lissa
just, still, yet
ba3du
ba3deen
later on
qablu
?abla
before
qad, laqad

already, perhaps

Table (68) shows that some SA accusative nouns functioning as adverbs of

time are unchanged in EA, e.g. /?ahyaan-an/ sometimes; some are
replaced by prepositional phrases, e.g. /layl-an > bi-l-leel/ by night and
some are replaced by suppletive forms. The SA adverb of time /?al-?aan-a/
now, which is actually the noun /?al-?aan/ the time in the accusative, is
replaced by another noun meaning the time, i.e. /?il-wa?t/, preceded by
the demonstrative particle /da/ this which loses its vowel, i.e. /d-il-wa?t/
this time.
Table (69)
Adverbs of Intensity in SA & EA

229

SA
1 tamaam-an

EA
tamaam
~ tamaam-an
2 muTlaq-an
muTlaq-an
3 fi3l-an
fi3l-an
4 xuSuuS-an
xuSuuS-an
5 qaliil-an
?ulayyil
6 kaiir-an
kitiir
7 jidd-an
?awi
8 ?ayDan
kamaan
9 faqaT ~ fahasb bass
10 qaTT-u
abad-an
11
xaaliS
12
yadoob(ak)
13
barDu, barDak

Gloss
completely
absolutely
truly
especially
little
much
very
also, more
only
never
ever, extremely
just, barely
also, nevertheless

Table (69) reveals that some SA adverbs of intensity are unchanged in EA

and some are replaced by suppletive forms. Furthermore, EA has
developed its own adverbs of intensity which are non-existent in SA.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

Table (70)
Adverbs of Manner in SA & EA
SA
EA
Gloss
3amaliyy-an
3amaliyy-an
practically
3amd-an
3amd-an
deliberately
muSaadafat-an Sudfa
accidentally
jamii3-an
gamii3-an
everyone
sawiyy-an
sawa
together
jiddiyy-an
bi-gadd, dugri
seriously
sarii3-an
bi-sur3a, ?awaam quickly
haakaa
kida
this way
mustaqiim-an dugri
straight on
bi-suhuulat(-in) bi-suhuula
easily
bi-Su3uubat(-in) bi-Su3uuba
hard
baTii?-an
3ala mahl+pro
slowly
wahd+pro
wahd+pro
alone
bi+mufrad+pro li+wahd+pro
alone
ma3-an

together

kuwayyis
well

Table (70) shows that some SA accusative nouns functioning as adverbs of

manner are preserved in EA, some are replaced by prepositional phrases

230

and some by suppletive forms. The adverb of intensity /ma3-an/ together

is found in SA only and the adverb /kuwayyis/ well is used in EA only.
Table (71)
Adverbs of Probability and Approximation in SA & EA
SA
EA
Gloss
1 rubbamaa
rubbama
perhaps, possibly
2 gaalib-an
gaalib-an
mostly
3 qaT3-an
qaT3-an
for sure, definitely
4 Tab3-an
Tab3-an
of course, naturally
5 taqriib-an
ta?riib-an
approximately
6 hawaali
hawaali
nearly
7 maal-an
masal-an
for example
8
?akiid
for sure, undoubtedly
9
yiigi
nearly
10
yiTla3
nearly
Table (71) reveals that most SA adverbs of probability are preserved in
EA. Moreover, the Colloquial variety uses the two verbs /yiigi/ to come
and /yiTla3/ to appear as adverbs of approximation, a practice which is
not followed in SA.
5.4. Interrogative & Responsive Particles:
There is a very significant divergence between SA and EA in the
use of interrogatives:
Table (72)
Interrogative Particles in SA & EA
SA
EA
Gloss
1 ?a-/hal
ya tara
I wonder
2 ?a-laa

I wonder ... not

3 ?a-lam

I wonder ... not

4 ?a-maa

I wonder ... not

5 ?am

I wonder ... or
6 ?ayyu
?anhu (msg)

7 ?ayyatu
?anhi (fsg)
which
8 ?ayyu/?ayyatu ?anhum (pl)

9 ?ayna
f-een
where
10 min ?ayna
min-een
from where
11 kam
kaam
how many
12 bi-kam
bi-kaam
how much

231

13
14
15
16
17
18
19
20

kayfa
maa/maaa
li-ma/li-maaa
mataa
man
li-man

?izzaay, ?izzayy
?ee(h)
l-ee(h)
?imta
miin
li-miin
?imi3na
maal

how
what
why
when
who
for who
why then
whats the matter with

The first six particles in the table above are yes/no interrogatives.
Though /ya tara/ equals /?a-, hal/ in meaning, it does not replace it in all
contexts. In many cases, the Standard form is replaced by the interrogative
intonation in EA. The forms (6-18) can be treated as indefinite pronouns
or pro-forms. Though the last three particles are not found in SA, they
have their origins there. For instance, /?imi3na/ why then has developed
from /?ayy-u ay?-in + ma3naa/ what thing + meaning, i.e. what is the
meaning of. After cutting off the case-endings /?ayy-u ay?-in/ becomes
/?ay ay?/ Then, the noun /ay?/ is reduced to just // as in the negative
particle. After that /?ay/ is changed by Monophthongization into /?ee/
which is then shortened by Closed-Syllable Shortening into /?i/. Finally,
the first vowel in /ma3naa/ is changed to /i/ to secure vowel harmony with
the vowel in /?i/ and the final vowel is shortened by Final-Vowel
Shortening producing /?imi3na/.
As shown in Table (72), some of the interrogatives are
monomorphemic, e.g. /kam > kaam/ how many, some are bimorphemic,
e.g. /li-ma > l-eeh/ why and some are compounds of two elements in
open transition (cf. Aboul-Fetouh 1969: 115), e.g. /bi-kam > bi-kaam/.
However, the interrogative for place is bimorphemic in EA although it is
monomorphemic in SA, i.e. /?ayna > f-een/. The Colloquial interrogative
consists of [fi-] in and [-een] which is a regular phonological
development of the SA particle [?ayna] where by Monophthongization.
A similar divergence is noted in the particles of response, as in
Table (73):
Table (73)
Responsive Particles in SA & EA
SA
EA
Gloss
1 na3am na3am
yes
2 ?ii
?aa/?aywa yes
3 laa
la??(a)
no
4 balaa

yes, certainly

232

5
6

?ajal
kallaa

yes
by no means

In addition to their use in reply to yes/no questions, the particles /na3am >
?aywa/ yes are used in reply to vocative particles.
5.5. Negative Particles:
Although SA has a complex syntactic system of negation, EA has a
simple morphological one. To negate a verb, SA employs a negative
particle followed by the imperfect form of the verb. This negative particle
differs in accord with the time reference of the verb. So, three negative
particles are commonly used: /lam/ for past time reference, /laa/ for
present time reference and /lan/ for future time reference. The use of these
particles is accompanied by a change in the mood of the verb. The jussive
is used after /lam/; the indicative after /laa/ and the subjunctive after /lan/.
For a detailed discussion of these moods, see (2.2). EA, on the other hand,
uses one discontinuous morpheme [ma--] which co-occurs with the verb.
The two elements of this morpheme are the assimilated forms of the SA
negation particle /maa/ not plus /ay?/ a thing originally meaning not
... a thing (cf. Robertson 1970: 36). According to Ghaly (1960: 99), the
first element [ma-] always immediately precedes any prefix co-occurring
with the stem, and the second element [-] follows any suffix attached to
it. In other words, [ma-] is always the first element of the verb
construction and [-] the last element. Now compare the Colloquial verbs
in (227-229b) with their Standard equivalents in (227-229a), respectively:
(227) a. lam ya-ktub-
id-dars
not impf.3msg-write-juss the-lesson
He did not write the lesson.
b. ma-katab-
id-dars
neg-write.pf.3msg-neg
the-lesson
He did not write the lesson.
(228) a. laa ya-ktub-u
not impf.3msg-write-indic
He does not write the lesson.
b. ma-yi-ktib-
neg-impf.3msg-write-neg
He does not write the lesson.
(229) a. lan
not

ya-ktub-a
impf.3msg-write-subj

d-dars(-a)
the-lesson(-Acc)
id-dars
the-lesson

d-dars(-a)
the-lesson(-Acc)

233

He will not write the lesson.

b. ma-ha-yi-ktib-
neg-fut-impf.3msg-write-neg
He will not write the lesson.

id-dars
the-lesson

Three regular phonological alternations are remarked here

concerning the negation of verbs in EA. First, the vowels in pronominal
suffixes are re-lengthened before the second element [-] (cf. Omar 1976:
11). This supports our view in (2.4) that the rule of Final-Vowel
Shortening does not apply to these vowels when they are no longer final,
being followed here by the consonant //. Compare the following examples
before and after [-] suffixation in EA:
Affirmative
Darab-tu
Darab-tik
Darab-tuku
Darab-ni

Negative
ma-Darab-tuu-
ma-Darab-tikii-
ma-Darab-tukuu-
ma-Darab-nii-

Meaning
I did not hit him
I did not hit you (f)
I did not hit you (pl)
He did not hit me

The word /ma-Darab-tikii-/ shows that the allomorph [-kii] of the second
person feminine singular objectival affix [-ik] is chosen here (See 2.4).
Second, if there is a cluster of two consonants before the suffix
[-], the vowel /i/ is added between the cluster and the suffix. This is the
normal location of the epenthetic vowel as suggested by the rule of
Medial-Epenthesis in EA, as in the following verbs:
Affirmative
?afal-t
3irif-t
daxal-t

Negative
ma-?afal-ti
ma-3rif-ti
ma-daxal-ti

Meaning
I did not close
I did not know
I did not enter

Third, where the addition of [-] creates a cluster of two consonants

there is a shift in stress (cf. McGuirk 1986: 49), e.g. /ftah/ he opened but
/ma-fath-/ he did not open.
The morpheme [ma-] replaces or alternates stylistically with
[ma--] after a word or phrase expressing an oath, a wish or a period of
time, as in the examples in (230), respectively. Aboul-Fetouh (1969: 114)
states that the occurrence of [-] with [ma-] in this environment is rare:
(230) a. w-alla ma-3raf
By God, I dont know.

234

b. ya-reet-u ma-raah
I wish he hadnt gone.
c. 3umr-i ma-Darab-t-u
I never beat him (pf).
d. rabbi-na ma-yi-hrim-na minnak
~ rabbi-na ma-yi-hrim-naa- minnak
May God not deprive us of you.
There is a particular case in which a free allomorph [mu] or [mi]
alternates with [ma--]. This is when the progressive prefix [bi-] or the
future prefix [ha-] also occurs with the verb. However, it has to be noted
that the discontinuous allomorph is more frequent with [bi-] and the free
allomorph is more frequent with [ha-]:
(231) a. ma-bi-yiktib- ~ mi bi-yiktib
He is not writing.
b. ma-b-ni-zra3- ~ mu bi-ni-zra3
We are not growing.
c. mi ha-yi-ktib ~ ma-ha-yi-ktib-
He will not write.
d. mu ha-ti-nDaf ~ ma-ha-ti-nDaf-
It will not be clean.
To negate the predication of existence, SA employs the particle of
absolute negation /laa/ no. This is replaced by the negative preposition
/ma-fii-/ in EA:
(232) a. laa rijaal-a
fi-l-madiin-at(-i)
no men-Acc
in the city(-Gen)
There are no men in the city.
b. ma-fii-
riggaala
fi-l-madiin-a
neg-in-neg
men
in the city
There are no men in the city.
To negate personal pronouns and particles, SA uses the negative
particle /maa/. This is replaced by the discontinuous morpheme [ma- -] in
EA. Compare the SA constructions in (233-236a) with their EA parallels
in (233-236b), respectively:
(233) a. maa huwa bi-aahib(-in)
He is not going.
b. ma-huwwaa- raayih
He is not going.

235

(234) a. maa nahn-u bi-laa3ibiin

We are not going to play.
b. ma-hnaa- la3biin
We are not going to play.
(235) a. maa ma3ii qalam(-un)
I dont have a pen.
b. ma3yaa- ?alam25
I dont have a pen.
(236) a. maa laka nuquud-un 3indii
I dont owe you any money.
b. ma-lik-i fluus 3andi
I dont owe you any money.
To negate nouns, adjectives and particles (including adverbs), SA
uses the negative verb /laysa/ not. This is replaced by the free
morphemes [mu] or [mi] in EA. Compare the SA examples in (237240a) with their EA analogues in (237-240b), respectively:
(237) a. laysa xaTTa-hu
b. mi xaTT-u
(238) a. laysa bi-?ahmar
b. mu ahmar
(239) a. laysa 3ala l-maktab
b. mu 3a-lmaktab
(240) a. laysa sarii3-an
b. mi bi-sur3a

It is not his handwriting.

It is not his handwriting.
It is not red.
It is not red.
It is not on the desk.
It is not on the desk.
Not quickly.
Not quickly.

To repeat negation, SA repeats the negative particle after the

connective /wa/ but EA uses only one repetitive morpheme [la--wala]
neither -- nor. Compare the SA constructions in (241-243a) with their
EA counterparts in (241-243b):
(241) a. lam ya-?kul
wa
not impf.3msg-eat and
He neither ate nor drank.
b. la-kal
wala
neither-eat.pf
nor
He neither ate nor drank.

lam
not

ya-rab(-)
impf.3msg-drink(-juss)

irib-
drink.pf-3msg

236

(242) a. lan ya-qra?-a

wa
not impf.3msg-read-subj and
He will neither read nor write.
b. la-ha-yi-?ra
neither-fut-impf.3msg-read
He will neither read nor write.
(243) a. laa ya-sma3-u
not impf.3msg-hear-indic
He neither hears nor sees.
b. la-b-yi-sma3
neither-prog-impf-hear
He neither hears nor sees.

lan ya-ktub(-a)
not impf.3msg-write(-subj)
wala
nor

ha-yi-ktib
fut-impf.3msg-write

wa
and

laa
not

yu-bSir(-u)
impf.3msg-see(-indic)

wala
nor

bi-y-oof
prog-impf-see

The morpheme /wala/ is used in EA to signal emphatic negation

(Abdel-Malek 1972: 207). Compare the following EA examples, where
simple negation is given in (244a) and emphatic negation is in (244b):
(244) a. mi haagi I will not come.
b. wala haagi I will not come at all.
To give a negative imperative, a construction formed by the
enclitic carrier /?iyyaa/ + an objectival suffix + ?an + the subjunctive form
of the verb is used in SA. This construction is replaced by the imperative
verb /?iw3a/ be careful + the imperfect form of the verb in EA. Compare
the warning statements in (245a) for SA and (245b) for EA:
(245) a. ?iyyaa-ka
?an
ta-rab-a
l-xamr
enclitic carrier-2msg that
impf.2msg-drink-subj the-wine
I warn you not to drink wine.
b. ?iw3a
ti-rab
l-xamr
be careful not to
impf.2msg-drink
the-wine
Be careful not to drink wine.
A negative morpheme which is used in EA only is /balaa/. If it is
followed by a verbal noun, it signals a strong negative command,
elsewhere it signals a mild negative suggestion or request, e.g. /balaa
kalaam/ Stop talking! and /balaa ni-l3ab in-naharda/ Lets not play
today!'
5.6. Possessive Particles:

237

In SA, possession is expressed phrasally by the construct phrase

(See 3.3). This is formed by placing the possessed before the noun naming
its possessor in the genitive case, e.g. /kitaab-u l-walad-i/ the book of the
boy. In EA, this structure is preserved with the deletion of the case
endings, i.e. /kitaab il-walad/.
EA has developed a possessive particle which is not employed in
SA, though it has its origin there: /bitaa3/ belonging to which probably
developed from the SA noun /mataa3(-un)/ property. For instance,
/?il-kitaab bitaa3 il-walad/ the book belonging to the boy. This is an
appositional phrase in which /bitaa3 il-walad/ is a construct phrase
meaning the property of the boy which is in apposition to /?il-kitaab/.
Thus, the word /bitaa3/ is actually a nominal which can stand alone or be
used in construct phrases to express possession. This can be supported by
the fact that it is treated like nouns in relation to gender and number. It has
the feminine form /bitaa3-a ~ bitaa3-it/ and the plural form /bituu3/, e.g.
/?il-ward-a bita3-t il-bint/ the rose belonging to the girl and /?il-li3ab
bituu3 il-wilaad/ the toys belonging to the kids.
The prepositions /ladaa/, /ma3a/ and /3inda/ at, with are
sometimes used to express possession in SA. Because the first one is not
used in EA, it is replaced by the other two. The preposition /3inda/ is
changed into /3and/ by Final-Vowel Deletion and exceptional first vowel
lowering. Thus, /laday-naa kutub-un kaiir-ah > 3and-ina kutub kitiir-a/.
We have many books. Other prepositions used for expressing possession
in both varieties are /fii/ in and /li-/: for:
la-hu
(246) a. ?ahmad-u
Ahmad-Nom prep-3msg
Ahmad has many friends.
b. ?ahmad
lii-h
Ahmad
prep-3msg
Ahmad has many friends.

?aSdiqaa?-u
friends-Nom
?aSdiqaa?
friends

kaiir-uun
many-mpl
kitiir
many

(247) a. ?al-madrasat-u
fii-haa
mal3ab-un
kabiir(-un)
the-school-Nom
prep-3fsg
playground big(-Nom)
The school has a big playground.
b. ?il-madrasa fii-ha
mal3ab
kibiir
the-school
prep-3fsg
playground big
The school has a big playground.

238

6. Conclusions
6.0. General Findings:
(i) The differences between SA and EA are more phonological than
morphological. Most of the basic morphological patterns of the former
variety are preserved in the latter, though some of those patterns are
phonologically changed. Also, EA has innovated some very few
patterns of its own.
(ii)The similarities between SA and EA morphologies are more than the
divergences between them. With the exception of certain cases of
reanalysis, if we assume the same underlying representation of both SA
and EA, the two varieties can be related by systematic and predictable
rules.
(iii)A major difference between SA and EA is the disappearance of
inflectional endings (case and mood suffixes) from the latter variety.
All final short vowels used as inflectional endings are lost and all final
long vowels are shortened.
(iv)Another difference between SA and EA stems from the latters limited
use of the dual, for it does not provide dual forms for verbs, adjectives
or pronouns.
(v) In EA there is no gender distinction in the plural of verbs, adjectives or
pronouns, though such a distinction is maintained in SA.
(vi)There is a great deal of suppletion between SA and EA in closed-list
classes, more than in all other parts of speech.
6.1. Phonological & Morphological Basics:
The following remarks can be noted in relation to the consonantal
system in SA and EA:
(i)
The consonantal system of EA includes all SA consonants with the
exception of the interdental consonants // and //. These have been
replaced by the dental stops /t/ and /d/, respectively, or the alveolar
fricatives /s/ and /z/, respectively. Though the SA alveolar fricative /Z/ is
also preserved in EA, it is replaced in some words by /D/.
(ii)
The use of the glottal stop /?/ in place of the SA uvular stop /q/
represents a hallmark of EA. However, the process of Standardization has
led to the use of /q/ in some isolated borrowings from SA.
(iii) SA and EA are identical in making a distinction between the
emphatic dento-alveolar consonants /T, D, S and Z/ and their nonemphatic counterparts /t, d, s and z/. Two other consonants have emphatic

239

and non-emphatic versions in both varieties: /l/ and /r/. However, the
environment of emphaticization of /r/ is different in EA from that of SA.
(iv)
EA deviates considerably from SA in the use of /?/ in medial and
final positions. While still retaining /?/ in a small number of examples, EA
tends to lose it either by lengthening the preceding vowel or supplying a
glide instead. The final /?/ has virtually disappeared except where Standard
borrowings are kept intact.
(v)
SA and EA are similar in relation to consonant clustering. Though
medial and final clusters are common in both of them, initial clusters are
non-existent in either of them. Both varieties are identical in not allowing
more than two consonants to occur in a cluster.
In the vocalic system, EA retains all SA vowels and has in addition
/ee/ and /oo/ which can be traced to SA /ay/ and /aw/ by a rule which we
called Monophthongization. The most significant difference which sets
EA apart from SA is the lack of vowel correspondence in related items or
forms. Some cases of discrepancy include the shortening of final long
vowels and stressless pre-final vowels, the raising of low /a/ to /i/ as in
perfect verbs and imperfect prefixes, and high vowel deletion.
As for stress, both varieties are identical in stressing:
- the ultimate syllable when it is superheavy,
- the penultimate syllable when the ultimate is not superheavy, and
- the antepenultimate syllable when the ultimate is not superheavy and
both the penult and antepenult are light.
However, words of three or four syllables with a consonant cluster are
stressed before the cluster in SA and after it in EA.
In the realm of phonological alternations, the following rules apply
in both varieties:
(1) Word-Initial Epenthesis:
i / # CC
(2) Glottal-stop Insertion:
?/#V
(3) 1-Assimilation:
Ci
l [+def] Ci /

+ cor
(4) n-Assimilation:

240

Ci

n Ci / + son
+ cons
(5) Coronal-Assimilation:
C
Ci

+ obstr Ci / + obstr
+ cor
+ cor
(6) Closed-Syllable Shortening:
(V)VV (V)V/ C]
(7) Identical-Consonant Metathesis:
CkVCkV VCkCkV
(8) Onset Formation:
y i
/ a
w u
(9) Glide Elision:
G / Vi Vj
if j = [+1o], then i = [+1o]
(10) u-to-i:
Cy

u i / i
y

(11) Perseverative Vocoid Assimilation:

cons
+ hi Vi / Vi +

(12) Anticipatory Vocoid Assimilation:

cons
+ hi Vi / C Vi]

The following rules apply in SA only:

(1) Medial-Epenthesis in SA:
i / C C.C
(2) Glottal Formation in SA:
G ? / aa V
The following rules occur in EA only:

241

(1) Medial-Epenthesis in EA:

i / C.C C
(2) High Vowel Deletion in EA:
V
+ hi

/ VC CV
long

stress
(3) Final /?/ Deletion in EA:
?/#
(4) Compensatory Lengthening in EA:
? Vi / Vi ]
(5) Intervocalic /?/ Assimilation:
cons
cons
+ hi
+ hi
?
/V

rd
rd
(6) Final-Vowel Shortening in EA:
VV V / #
(7) Atonic Shortening in EA:
VV V / when stressless
These are the results related to the morphological and
morphosyntactic features of SA and EA:
(i) The same word classes are found in SA and EA. Though traditional
Arab grammarians say that there are three word classes: noun, verb and
particle, modern linguists assert that there are also adjectives, pronouns
and adverbs.
(ii)In both varieties, a word is composed of three basic morphemes: a
consonantal root, a vocalic pattern and one or more affixes. Upon them
a prosodic template is imposed.
(iii)Five methods are utilized in the formation of words in both varieties:
derivation, analogy, coinage, compounding and Arabicization.
(iv)While SA differentiates between pausal and non-pausal forms, EA
does not.
(v) Whereas SA employs nunation in forming indefinite nouns, EA
does not.
6.2. Verbal Morphology:

242

These findings are related to the verbal stem forms in SA and EA:
(i) EA has retained all the patterns of SA perfect forms of sound verbs.
Both varieties use the same morphological processes to derive Forms II-X
from the primary Form I, with a tendency in EA to replace Form IV by
Form II in some verbs and by Form I in some others.
(ii)EA is characterized by the raising of the first vowel in some
[Fa3aL(-a)] verbs, with a tendency to harmonize the first vowel with the
second one.
(iii)Though SA has only /a/ as a stem vowel in Forms II, III, IV and X, EA
has /a/ in some verbs and /i/ in others.
(iv)The prefix [ta-] in the SA Forms V and VI is reanalyzed as [t-] in EA.
(e) The alternation of the stem vowel of Form I between the perfect and
imperfect is not systematic in either variety. The two varieties are also
identical in deleting the first vowel of the perfect in the imperfect of that
form.
(f) The perfect of geminate roots takes the form of a final-weak verb
before pronominal suffixes starting with a consonant in EA, though it
behaves regularly in SA.
(g) Glottalized verbs are highly restricted in EA and in some cases nonexistent, in contrast to their occurrence in SA in a fairly large number of
instances.
(h) Four salient features characterize EA in the use of weak verbs: (i) the
retention of /w/ in the imperfect of initial-weak w-verbs, (ii) the retention
of the glide in some instances of Forms I, VII and X medial-weak verbs,
(iii) the reanalysis of the SA prefix [ta-] in Form V final-weak verbs as
[sta-], and (iv) the change of final /uu/ in the imperfect forms of final-weak
verbs into /a/ or /i/.
(i) EA is characterized by its innovation of quadriradical forms with the
reduplication of the third radical which are lacking in SA.
The following results are concerned with the inflection of verbs for
aspect/mood and voice:
(i)
Though SA imperfect verbs are inflected for three moods:
indicative, subjunctive and jussive, EA shows only the indicative.
(ii)
Although EA has preserved the SA imperfect prefixes, it has
replaced the vowel in them by /i/ with the exception of [(?)a-] where the
vowel is retained.
(iii) EA differs from SA in that though in the latter the imperfect form
can express habitual or progressive actions, the former precedes the
imperfect prefixes by an additional prefix [bi-] to signal these actions.

243

(iv)
The SA future tense markers [sa-] and [sawfa] have been replaced
by [ha-] or [ha-] in EA.
(v)
The imperative forms are identical in SA and EA except in medialweak roots where Closed-Syllable Shortening applies in the former but not
in the latter.
(vi)
The SA passive form has left no vestiges in EA. The latter
employs, instead, reflexive patterns to indicate the passive voice,
extending them to verbs for which SA uses only the passive form.
These remarks are discovered from the study of verb affixation and
derivation in both varieties:
(i)

EA has almost half of the subject affixes used with perfect verbs that are
found in SA and it has almost eliminated those used with the imperfect.
The dummy stem [?iyyaa-] to which object affixes attach in SA has also
been lost in EA, except in intensive constructions.
(ii) In both varieties, verbs can be derived from nouns, adjectives or other
verbs.
(iii) Though SA uses three affixes to derive causatives from intransitives:
gemination of the second radical, the prefix [?a-] and the prefix [sta-], EA
uses gemination only.
(iv) SA uses three affixes to form inchoatives: the two prefixes [n-] and [ta-]
and the infix [-t-]. While the first affix is kept unchanged in EA, the other
two are changed into [t-].
6.3. Nominal Morphology:
The following results are related to primary and deverbal nominal
stem forms in SA and EA:
(i)

Most of the SA triradical nominal stem forms are preserved in EA.

Some other stem forms are phonologically changed and some have
two EA equivalents, one identical to that of SA and the other
phonologically changed.
(ii) All SA quadriradical nominal stem forms are retained in EA.
(iii) In both varieties, none of the prosodic representations of nominal stem
forms contains less than two moras and all stem templates are
disyllabic.
(iv) Most of the SA patterns of primary triradical verbal nouns are
preserved in EA and some are phonologically changed. The majority
of verbal noun patterns of derived triradical and quadriradical verbs
are preserved in EA.

244

(v) All SA patterns of nouns of exaggeration and nouns of place/time

are preserved in EA.
(vi) EA replaces the instrument prefix in the SA patterns [mi-F3aL(un)] and [miF3aL-at(-un)] by [ma-] and that in the pattern [miF3aaL(-un)] by [mu-].
These findings are concerned with the various inflections of nouns
as well as to diminutive formation:
(i)

Both SA and EA make a distinction between definite and indefinite

nouns. However, the SA definite article [al-] is reduced to just [1-] in EA.
Although nouns become definite in SA by being followed by a
determining complement, such a device is not important in EA. Proper
nouns are also definite in both varieties.
(ii)
Though SA nouns inflect for three cases, EA has no case endings nor
does it use nunation in the formation of indefinite nouns.
(iii)
EA corresponds to SA in making gender distinction in nouns but it has
reduced the three suffixes normally expressing feminine gender into two:
both SA [-at and -aa] are collapsed into one EA suffix [-t ~ -a#] and SA [aa?] is replaced by [-a? ~ -a#].
(iv)
Though SA uses two variants of the dual suffix depending on the case
of the noun: [-aani (Nom)] and [-ayni (Acc & Gen)], EA uses only one
variant for that purpose: [-een] derived from [-ayn] by
Monophthongization.
(v)
Though SA employs two allomorphs for the masculine plural [-uuna
(Nom)] and [-iina (Acc & Gen)], EA uses only one: [-iin].
(vi)
The suffix [-aat] is used to form the sound feminine plural in both
varieties, but it is followed by case endings in SA only.
(vii)
EA deviates from SA in two salient features related to the plural: (i)
its simplified system of sound feminine plurals for [FV3L-a] by adding [aat] without any vowel change in the stem of the singular as required in
SA in a number of cases, and (ii) its use of the suffix [-iyya] in forming the
plurals of certain nouns, primarily those ending in [-gi].
(viii)
In both varieties, it is not predictable which broken plural type is
generated given a specific singular pattern. Most SA broken plural patterns
are preserved in EA and some are phonologically changed.
(ix)
EA does not use any of the SA diminutive patterns but uses a special
hypocoristic form to address children fondly.
6.4. Adjectival Morphology:
These remarks are related to adjectival stem forms and their
various inflections in SA & EA:

245

(i) Some SA adjectival stem forms are retained in EA and some of them
have undergone phonological changes. Some SA adjectival stem forms
have more than one EA equivalent; one is identical to that of SA and
the others are phonologically changed.
(ii)In both varieties, adjectives agree with the nouns they modify in terms
of definiteness.
(iii)SA adjectives inflect for three cases: nominative, accusative and
genitive in accordance with the nouns they modify. EA, on the other
hand, does not have case markers at all.
(iv)In both varieties, adjectives inflect for gender in accordance with the
nouns they modify, though a few adjectives, particularly those of color,
are genderless in EA.
(v) EA has extended the ending [-a] normally expressing feminine gender
in adjectives to cover the three SA such endings: [-at], [-aa?] and [-aa].
(vi)Though SA adjectives are inflected for three numbers in agreement
with the nouns they modify: singular, dual and plural, EA adjectives are
inflected for two numbers only: singular and non-singular (i.e.
including both dual and plural).
(vii)In both varieties, plural adjectives are of two types: sound and broken.
However, though SA has two types of sound plural adjectives:
masculine and feminine, EA has only one, i.e. sound plural.
(viii)Both SA and EA adjectives are inflected for the comparative degree
by the pattern [?aF3aL(-u)] which is invariable with nouns of all
numbers and genders.
The following findings are concerned with the participial and
relational forms of adjectives in both varieties:
(i)

In both varieties, the active participle of primary sound verbs has the
form [Faa3iL] and that of derived verbs has the form [mu-...iC] with the
prefix changed to [mi-] in some EA active participles. We accounted for
that by proposing that the prefix [mu-] became [mi-] at some historical
point and that participles which became lexicalized before this change
retain their [mu-].
(ii)
The citation form of the active participle of geminate verbs is
[Faa33(-un)] in SA, but [Faa3i3] in EA due to the loss of case endings.
Before vocalic suffixes, the pattern is [Faa33-] in both varieties which has
its vowel shortened in EA by Closed-Syllable Shortening. Active
participles are not followed by consonantal suffixes in SA. In EA, on the
other hand, the form [Faa3i3], reduced to [Fa3i3-] by Atonic Shortening, is
used before consonantal suffixes.

246

(iii)

In initial-glottalized verbs, the SA active participle pattern is [?aa3iL]

whose glottal stop is changed /w/ by ?-to-w in EA. In medial-glottalized
verbs, the pattern is [Faa?iL] whose glottal stop is changed into /y/ by
Intervocalic /?/ Assimilation. Final-glottalized active participles have their
/?/ deleted in EA by Final /?/ Deletion.
(iv)
In medial-weak verbs, the active participle is [Faa?iL] whose glottal
stop is produced by Glottal Formation in SA and changed into /y/ by
Intervocalic /?/ Assimilation in EA. In final-weak verbs the SA pattern is
[Faa3i(n)] in the nominative and genitive, and [Faa3iy(-an)] in the
accusative. Both are reduced to [Faa3i] in EA.
(v)
Most of the SA participles of the derived forms of non-sound verbs
have undergone regular phonological changes in EA.
(vi)
The normal pattern of passive participles of primary triradical verbs is
[maF3uuL] in both varieties.
(vii)
The passive participles of medial-weak primary verbs in SA are
replaced by those of the passive forms of the primary verb in EA.
(viii)
The passive participle of SA derived verbs has the form [mu-...aC]
whose prefix is changed into [mi-] in the majority of EA passive
participles.
(ix)
Though SA maintains a distinction between the active and passive
participles of derived verbs by using the pre-final vowel /i/ for the former
and /a/ for the latter, EA does not make such a distinction in Forms II, V
and VI.
(x)
The passive participle of Form IV is losing ground to that of Form II
in EA. This is notable in geminate and initial-weak verbs. Similarly, the
passive participles of Form I quadriradical is losing ground to that of Form
II quadriradical in EA.
(xi)
SA forms the relational adjectives by the attachment of one of three
suffixes to the noun: [-iyy, - aaw-iyy or -aan-iyy]. EA uses the shortened
form of these suffixes in which [-iyy] becomes just [-i]. Another
characteristic feature of EA is the use of [-gi], borrowed from Turkish, as a
relative suffix.
6.5. Closed-List Class Morphology:
(i) EA has greatly reduced the number of personal pronouns in SA. Both dual
and feminine plural pronouns have disappeared in the Colloquial variety.
Most of the free pronouns which are still found have acquired modified
forms.
(ii)EA has developed an extremely simplified system of relative pronouns
reducing the various forms of SA to one form [?illi]. Also, it does not use
/man/ or /maa/ as relative pronouns.

247

(iii)Most of the demonstratives retained in EA are not identical with those

of SA though they may be phonologically developed from them.
(iv)Many SA prepositions and adverbs are virtually lacking in EA. Some
of those retained have modified forms.
(v) Many SA particles of interrogation, response and negation are nonexistent in EA.
Notes
Other members in the Semitic family of languages are Hebrew,
Aramaic, Amharic (or Ethiopian) and some extinct languages as Akkadian
(or Babylonian), Canaanite and Phoenician. More scientific details about
the genetic relation or origin of Arabic can be found in Thomason (1973).
1

Standard Arabic is taught at schools as a subject in Arabic

language classes, though the Colloquial dialect is used as the medium of
instruction in other subjects.
3

I am using the term rule in the technical sense of phonological

theory. It has been defined by Trask (1996: 313) as: Any statement
which, in some analysis, is intended to express a generalization about the
facts of the language being described. Rule in this sense does not entail
that there can be no exceptions.
4

My representation of vowels is phonological rather than phonetic.

Thus, the allophonic variations of vowels, e.g./ ~ a/, as in /b:t/ he
spent the night vs. /qa:l/ he said, are not taken into account.
5

Some linguists call the sequences of vowels and glides /ay/ and
/aw/ diphthongs. However, the issue of whether there are diphthongs in
Arabic or not is still unsettled.
6

The SA verbs /fahmt/ and /gaslt/ are jussive forms or pausal

forms of indicative or subjunctive moods (For more details about mood,
see 2.2).
7

The solar consonants are so called because they are similar to

the consonant // occurring at the beginning of the word /ams(-un)/ sun.
The lunar ones are thus named due to their similarity with the consonant
/q/ occurring at the beginning of the word for moon /qamar(-un)/.
8

Closed-Syllable Shortening in /kitab-hum/ can be considered a

result of the historical loss of the inflectional vowel from SA. The deletion
of /u/ forced /b/ to syllabify with the preceding syllable and forced the

248

vowel to shorten because long vowels are not permitted in closed syllables
except word-finally:
SA
k i t a a bu hum

EA
kitab hum

The suffix in /walad(-un)/ is a nominative case marker. Nouns

inflection for case will be discussed in (3.4).
10

The citation form of the verb in Arabic is the perfective third

person masculine singular form. Thus, for citation purposes, the form
/Darab(-a)/ is referred to as to hit although in a specific context it means
he hit.
11

The verbs /kal/ to eat and /xad/ to take are two of the very few
instances of biradical verbs in EA. They are used alternatively with
triradical forms with an initial glottal stop: /?akal/ and /?axad/.
12

Notice the dropping of the long vowel /aa/ in the imperfect forms
of the verbs to cheer and to believe in SA to the contrary of its
preservation in EA, perhaps to facilitate the pronunciation of the glottal
stop.
13

The /n/ in the verb /?inbaa3(-a)/ to be sold is changed into /m/

to assimilate to the following consonant /b/ in place of articulation.
14

Arab grammarians call the imperfect /al-muDaari3(-u)/ that

which is similar, i.e. to the noun in that it can be inflected for three moods
like the noun which can be inflected for three cases, unlike /al-maaDii/
which is uninflected (= Arabic /mabnii/). See Kharma (1983: 31) and
Bateson (1967: 25).
15

Very few speakers pronounce [-tu] as [-tum] but this

pronunciation is so rare that one can consider the presence of /m/ a mark
of Standardization or inter-language borrowing on the part of those people.
16

In forming the imperative of Form IV verbs, this vowel is /a/, e.g.

/yu-Slih/ (juss) to repair /?a-Slih/ (imper) Mend!, repair!.

249
17

The term construct phrase denotes two words juxtaposed in

what may be considered possessed-possessor relationship. It is also
called a nominal construct by Gamal-Eldin (1967: 35).
18

Sisters of /?inna/ verily are /?anna/ that, /ka?anna/ as if,

/laakinna/ but, yet, /layta/ if only, would that, /la3alla/ perhaps.
Sisters of /kaan(-a)/ to be are /Saar(-a)/ to become, /?aSbah(-a)/ to
become (in the morning), /?aDhaa/ to become (in the forenoon),
/?amsaa/ to become (in the evening), /Zall(-a)/ to remain, /baat(-a)/ to
become, do during the night, /laysa/ not to be, /maa zaal(-a)/ to
continue to be and /maa daam(-a)/ to continue to be.
19

The vowel in the suffixes [-iin] and [-aat] is shortened in these

examples because EA does not allow a long vowel in a pre-final position
due to Atonic Shortening, nor does it allow a long vowel in a closed
syllable.
20

This may indicate that the accusative is the default case in SA.

21

Gary & Gamal-Eldin (1982: 100-101) assume that participles

inflect for person as well. But it is not clear how these forms have such an
inflection.
22

In the change from /3arabiyya/ to /3arba-gi/ there is a syncope of

/a/ which is difficult to account for.
23

With non-human feminine plural, /haa-ihi/ is more widely used

than /haa-?ulaa?i/.
24

This sentence can be acceptable if it is taken to mean The pen of

this (man) is new.
25

Because the preposition /ma3a/ starts with /ma-/, the first part of
the morpheme [ma--] is often deleted.

250

Bibliography
Abboud, Peter Fouad, et al. (1975) Elementary Modern Standard Arabic.
2nd ed. Ann Arbor: University of Michigan Press.
Abdel-Hafiz, Ahmad Sukarno (1991) The Development of Agreement
Markers in Standard Arabic. Bulletin of the Faculty of Arts - Qena
1: 175-188.
------- (1994) The Relationship between Classical Arabic and Colloquial
Arabic: A Linguistic Study. Bulletin of the Faculty of Arts - Qena.
4: 10-98.
Abdel-Malek, Zaki N. (1972) The Closed-List Classes of Colloquial
Egyptian Arabic. The Hague: Mouton.
Abdel-Massih, Ernest T. (1979) An Introduction to Egyptian Arabic. Ann
Arbor: Center for Near Eastern and North African Studies. The
University of Michigan.
Abdel-Tawab, Ramadan (1983) at-taTawwuru l-lughawiy: maZaahiruh
wa 3ilaluh wa qawaaniinuh [Language Development: Its Features,
Causes and Rules]. Cairo: Al-Khangy.
------- (1985) al-madxalu ila 3ilmi il-lughati wa manaahiju il-bahi
l-lughawiy [An Approach to Linguistics and Methods of Linguistic
Research]. Cairo: Al-Khangy.
Abdou, K. M. (1965) The Construct Phrase in the Structure of Colloquial
Cairene Arabic. M.A. Thesis. Brown University.
Abd-Rabbo, Mohamed Z. (1990) Sound Plural and Broken Plural
Assignment in Classical Arabic. Perspectives on Arabic
Linguistics I. ed. by Mushira Eid. 55-93.
Aboul-Fetouh, Hilmi Mohamed (1959) The Plural Morpheme of Egyptian
Arabic Nouns. M.A. Thesis. Texas University.
------- (1969) A Morphological Study of Colloquial Egyptian Arabic. The
Hague: Mouton.
Abu Farag, M. A. (1960) A Grammatical Study of the Arabic Dialect of
Tahway (Minufiya Province). Ph.D. Dissertation. University of
London.
Abou-Seida, Abdelrahman M. (1971) Diglossia in Egyptian Arabic:
Prolegomena to a Pan-Arabic Socio-Linguistic Study. Ph.D.
Dissertation. The University of Texas at Austin.
Al-Ani, Salman Hassan & May, Darlene R. (1973) The Phonological
Structure of the Syllable in Arabic. The American Journal of
Arabic Studies 1. ed. by Sami A. Hanna. Leiden: E. J. Brill.
Al-Antaki, Muhammad (1972) al-muhiiT fi aSwaat al-3arabiyyah
wanahwihaa wa Sarfihaa [The Comprehensive in the Phonology,
Syntax and Morphology of Arabic]. Beirut: Dar Al-Shuruq.

251
c

Al-A Zami, Mu in Al-Din (1988) Arabic Morphology: A Study in the

System of Conjugation. Hyderabad, India: Hasan Publishers.
Al-Hammadi, Yusuf et al. (1977) al-qawaa3idu l-?asaasiyatu fi n-nahwi
wa S-Sarf. [Basic Rules in Syntax and Morphology]. Cairo: State
Press.
Al-harbi, Abdallah (1990) A Syntactic Approach to Arabic Verbal
Morphology. Ph.D. Dissertation. University of Essex. (U. K.).
Al-Toma, Salih J. (1969) The Problem of Diglossia in Arabic: A
Comparative Study of Classical and Iraqi Arabic. Cambridge,
Mass: Harvard University Press.
Al-Tonsi, Abbas (1982) Egyptian Colloquial Arabic: A Structure Review.
Cairo: American University in Cairo Press.
Ayoub, A. E. R. (1949) The Verbal Piece in the Egyptian Language (A
Morphological Study). M.A. Thesis. University of London.
Baba, Abd al-Jawad Husayn (1988) Al-Sarf Al-Arabi: Siyaghah jadiidah
[Arabic Morphology: A New Approach]. Alexandria: Mu?assasat
Shabaab Al-gaamicah.
Badawi, El-Said (1973) Mustawayaatu l-3arabiyyati 1-mu3aSirati fi MiSr
[ Levels of Contemporary Arabic in Egypt]. Cairo: Dar alMacaarif.
------- (1985) Educated Spoken Arabic: A Problem in Teaching Arabic as
a Foreign Language. Scientific and Humanistic Dimensions of
Language: Festschrift for Robert Lado. ed. by Kurt R. Jankowsky.
Amsterdam: Benjamins. 15-22.
------- & Hinds, Martin (1986) A Dictionary of Egyptian Arabic: ArabicEnglish. Beirut: Libraire du Liban.
Bakalla, Mohammed Hasan (1979) The Morphological and Phonological
Components of the Arabic Verb (Meccan Arabic). Beirut: Libraire
du Liban.
Bakir, Murtadha Jawad (1980) Aspects of Clause Structure in Arabic: A
Study in Word Order Variation in Literary Arabic. Ph.D.
Dissertation. Indiana University.
Bateson, Mary Catherine (1967) Arabic Language
Washington, DC: Center for Applied Linguistics.

Handbook.

Becker, Valerie (1964) A Transfer Grammar of the Verb Structures of

Modern Literary Arabic and Lebanese Colloquial Arabic. Ph.D.
Dissertation. Yale University.

252

Beeston, Alfred Felix Landon (1968) Written Arabic: An Approach to

Basic Structures. Cambridge: Cambridge University Press.
------- (1970) The Arabic Language Today. London: Hutchinson
University Library.
Belnap, R. Kirk (1991) Grammatical Agreement Variation in Cairene
Arabic. Ph.D. Dissertation. University of Pennsylvania.
Blau, Joshua (1974) Classical Arabic, Middle Arabic, Literary Standard,
Neo-Arabic, Judaeo-Arabic and Related Terms. Joshua Finkel
Festschrift. New York: Yeshivah-Universitah.
Block, Bernard & Trager, George (1942) Outline of Linguistic Analysis.
Linguistic Society of America.
Brame, Michael (1970) Arabic Phonology: Implications for Phonological
Theory and Historical Semitic. Ph.D. Dissertation. MIT,
Cambridge, Mass.
Broselow, Ellen (1976) The Phonology of Egyptian Arabic. Ph.D.
Dissertation University of Massachusetts, Amherst.
------- (1979) Cairene Arabic Syllable Structure. Linguistic Analysis 5:
345-82.
Bulos, Afif A. (1965) The Arabic Trilateral Verb: A Comparative Study of
Grammatical Concepts and Processes. Beirut: Khayats.
Carter, Michael (1981) Arabic Linguistics. Amsterdam: John Benjamins.
------- (1996) Signs of Change in Egyptian Arabic. Understanding
Arabic: Essays in Contemporary Arabic Linguistics in Honor of
El-Said Badawi. ed. by Alaa Elgibali. Cairo: American University
in Cairo Press.
Cowan, William George (1968) Notes Toward a Definition of Modern
Standard Arabic. Language Learning 18: 29-34.
Crystal, David (1985) A Dictionary of Linguistics and Phonetics. Oxford:
Basil Blackwell.
Czapkiewicz, Andrzej (1975) The Verb in Modern Arabic Dialects as an
Exponent of the Development Processes Occurring in them.
Wroctaw: Zaktad Narodowy.
Eid, Mushira (1988) Principles of Code-Switching between Standard and
Egyptian Arabic. Al-cArabiyya 21: 51-79.
Eisele, John C. (1988) The Syntax and Semantics of Tense, Aspect and
Time Reference in Cairene Arabic. Ph.D. Dissertation. University
of Chicago.
------- (1990) Time Reference, Tense and Formal Aspect in Cairene
Arabic. Perspectives on Arabic Linguistics I. ed. by Mushira Eid.
El-Bakry, Fathia Abdu (1990) Active Participle in Colloquial Egyptian
Arabic: A Functional Analysis. Ph.D. Dissertation. Georgetown
University.

253

El-Derwi, Ahmad Mohammad Ata (1965) Number in Cairene Colloquial

Arabic with Comparative Reference to English. M.A. Thesis.
Brown University.
El-Ghawabi, Samira Mahmoud (1965) The Pronominal System in
Colloquial Cairene Arabic. M.A. Thesis. Brown University.
Elgibali, Alaa (1996) Understanding Arabic: Essays in Contemporary
Arabic Linguistics in Honor of El-Said Badawi. Cairo: American
University in Cairo Press.
El-Hassan, Shaher (1977) Educated Spoken Arabic in Egypt and the
Levant. A Critical Review of Diglossia. Archivum Linguisticum
13: 112-38.
------- (1978) Variation in the Demonstrative System in Educated Spoken
Arabic. Archivum Linguisticum 14: 32-57.
El-Saidi, Abdel-Latif Mahmoud (1967) Verb Morphosemantics in Arabic
and English. M.A. Thesis. Brown University.
El-Sayid, Dawood Helmy Ahmad (1962) A Descriptive Analysis of the
Part-of-Speech System and the Grammatical Categories of
Egyptian Colloquial Arabic. Ph.D. Dissertation. Cornell
University.
El-Tagoury, Maha Yousry (1989) A Contrastive Study of the Structure of
Modification in English and Colloquial Egyptian Arabic. M.A.
Thesis. Al-Azhar University.
El-Zorkani, Fatma Abdel-Hamid (1965) A Contrastive Study of Egyptian
Cairene Arabic and English Verb Phrases. M.A. Thesis. Brown
University.
Erickson, Jon Laroy (1965) A Contrastive Study of the Verb System of
Old and Modern English and Classical and Cairene Arabic. Ph.D.
Dissertation. University of Texas.
Fayez, Essam Abdel-Aziz (1990) Sibawaihs Linguistic Analysis of the
Diminutive in Classical Arabic and Its Subsequent Developments.
Ph.D. Dissertation. Georgetown University.
Ferguson, Charles A. (1972) Diglossia. Language and Social Context.
ed. by P. Gioglioli. London: Penguin. 232-51.
------- (1989) Grammatical Agreement in Classical Arabic and the
Modern Dialects. Al-cArabiyya 22: 5-17.
Fouad, Aida Mohammed (1965) Morphological and Syntactic
Construction of Adjectives in Colloquial Cairene Arabic. M.A.
Thesis. Brown University.
Gaballa, Hassan (1986) Case-marking and Classical Arabic: A Semantic
and Historiographic Study. Ph.D. Dissertation. University of
Delaware.
Gaber, Abdel-Rahman Mohamed. (1986) Sounds of Arabic. Giza: New
Offset Printing Shop.

254

Gadalla, Hassan A. H. (1992) Causative-Transitivity in English and

Arabic: A Contrastive Study. M.A. Thesis. Assiut University.
------- et al. (1998) LDC CallHome Egyptian Arabic Lexicon. Philadelphia:
Linguistic Data Consortium, University of Pennsylvania.
Gairdner, William Henry (1925) The Phonetics of Arabic. New York:
Oxford University Press.
------- (1926) Egyptian Colloquial Arabic. London: Oxford University
Press.
Gamal-Eldin, Saad (1959) Morphophonemics of Colloquial Egyptian
Arabic. M.A. Thesis. Texas University.
------- (1967) A Syntactic Study of Colloquial Egyptian Arabic. The
Hague: Mouton.
Gary, Judith Olmsted & Gamal-Eldin, Saad (1982) Cairene Egyptian
Colloquial Arabic. Amsterdam: North Holland.
Ghaly, Muhammad Mahmoud (1960) Substantive Morphology of
Colloquial Egyptian Arabic. Ph.D. Dissertation. University of
Michigan.
Ghazeli, Salem (1977) Back Consonants and Backing Coarticulation in
Arabic. Ph.D. Dissertation. University of Texas at Austin.
Goldsmith, John (1976) Autosegmental Phonology. Ph.D. Dissertation,
MIT. Cambridge, Mass.
Halle, Morris (1995) Feature Geometry and Feature Spreading. Linguistic
Inquiry 26/1: 1-46.
Hamp, Eric P.(1959) The Personal Morphemes of Classical Arabic. SIL
14: 21-2.
Hanna, H. Morcos (1967) The Phrase Structure of Egyptian Colloquial
Arabic. The Hague: Mouton.
Harris, Zellig (1945) Discontinuous Morphemes. Language 21: 121127.
------- (1958) Methods in Structural Linguistics. Chicago: The University
of Chicago Press.
Hashim, Kamal M. Z. (1980) Sentential Complements in Egyptian
Colloquial Arabic. Ph.D. Dissertation. Georgetown University.
Hassan, Mohamed M. Aly (1966) Gender in Egyptian Colloquial Arabic
(Alexandrian Dialect). M.A. Thesis. Brown University.
Haywood, J. A. & Nahmad, H. M. (1982) A New Arabic Grammar of the
Written Language. Cambridge: Harvard University Press.
Heine, B. A. & Reh, M. (1984) Grammaticalization and Reanalysis in
African Languages. Hamburg: Buske.
Helmy-Hassan, Salah-Eldin (1960) Verb Morphology of Egyptian
Colloquial Arabic, Cairene Dialect. Ph.D. Dissertation. University
of Michigan.

255

Hill, Archibald (1958) Introduction to Linguistic Structures. New York:

Harcourt, Bruce & World.
Howell, M .S. (1986) A Grammar of the Classical Arabic Language.
Delhi: Gian Publishing House.
Hudson, Grover (1986) Arabic Root and Pattern Morphology Without
Tiers. Journal of Linguistics 22: 85-122.
Hussein, Mohamed Hamza (1973) Adjectives in Egyptian Colloquial
Arabic and in English. A Contrastive Study. M.A. Thesis. AlAzhar University.
Hyman, Larry (1975) Phonology: Theory and Analysis. New York: Holt,
Rinehart & Winston.
Kenstowicz, Michael (1994) Phonology in Generative Grammar.
Cambridge, Mass: Basil Blackwell.
Khalafallah, Abdelghany Abdellah (1969) A Descriptive Grammar of
Sacidi Colloquial Egyptian Arabic. The Hague: Mouton.
Kharma, Nayef (1983) A Contrastive Analysis of the Use of Verb Forms in
English and Arabic. Heidelberg: Julius Groos Verlug.
Langacker, R. (1977) Syntactic Reanalysis. Mechanisms of Syntactic
Change. ed. by C. Li. Texas Press.
Levy, Mary Mansnerus (1971) The Plural of the Noun in Modern Standard
Arabic. Ph.D. Dissertation. University of Michigan at Ann Arbor.
Mahadin, Radwan Salim (1982) The Morphophonemics of the Standard
Arabic Tri-Consonantal Verbs. Ph.D. Dissertation. University of
Pennsylvania.
Mahmoud, Abdelgawad Tawfik (1989) A Comparative Study of Middle
and Inchoative Alternations in Arabic and English. Ph.D.
Dissertation. University of Pittsburgh.
Malik, Sayed Hamza Ayantunde (1976) A Contrastive Study of the Verbal
Patterns in Standard Arabic and Spoken Egyptian Arabic. Ph.D.
Dissertation. University of Ibadan, Nigeria.
Matthews, Peter (1974) Morphology. Cambridge: Cambridge University
Press.
McCarthy, John (1985) Formal problems in Semitic Phonology and
Morphology. Ph.D. Dissertation. MIT. Cambridge, Mass.
------- (1991) The Phonetics and Phonology of Semitic Pharyngeals.
Papers in Laboratory Phonology III, Phonological Structure and
Phonetic Form. ed. by P. Keating. London: Cambridge University
Press.

256

McCarthy, John & Prince, Alan (1986) Prosodic Morphology. Ms.

University of Massachusetts Amherst and Brandeis University.
------- (1988) Quantitative Transfer in Reduplicative and Templatic
Morphology. Linguistics in the Morning Calm 2. ed. by Linguistic
Society of Korea. Seoul: Hanshin Publishing Co. 3-35.
------- (1990a) Prosodic Morphology and Templatic Morphology.
Perspectives on Arabic Linguistics II. ed. by Mushira Eid and John
McCarthy. Amsterdam: John Benjamins. 1-54.
------- (1990b) Foot and Word in Prosodic Morphology: The Arabic
Broken Plural. Natural Language and Linguistic Theory 8/2: 21084
McGuirk, Russell, H. (1986) Colloquial Arabic of Egypt. London:
Routledge.
McLoughlin, L. J. (1972) Towards a Definition of Modern Standard
Arabic. Archivum Linguisticum III: 57-73.
Mitchell, T. F. (1956) An Introduction to Egyptian Colloquial Arabic.
London: Oxford University Press.
------- (1962) Colloquial Arabic, The Living Language of Egypt. London:
The English Universities Press.
------- (1990) Pronouncing Arabic I. Oxford: Clarendon Press.
------- & El-Hassan, S. A.(1991) Modality, Mood and Aspect in Spoken
Arabic with Special Reference to Egypt and the Lavant. New York:
Kegan Paul.
Mohammad, M.A.(1990) The Problem of Subject-Verb Agreement in
Arabic: Towards a Solution. Perspectives on Arabic Linguistics I.
ed. by Mushira Eid. 95-125.
Moore, John (1979) Arabic Derivational Morphology and Lexical
Theory. Papers from the Regional Meeting, Chicago Linguistic
Society 15: 228-43.
Nasr, Raja Tawfik (1967) The Structure of Arabic, from Sound to
Sentence. Beirut: Libraire du Liban.
Noyer, Rolf (1997) Features, Positions And Affixes in Autonomous
Morphological Structure. New York: Garland Publishing, Inc.
Omar, Margret K. (1973) The Acquisition of Egyptian Arabic as a Native
Language. Washington, DC: Foreign Service Institute.
------- (1976) Levantine & Egyptian Arabic: Comparative Study.
Washington, DC: Foreign Service Institute.
Osman, Ali Ezzat (1968) Interrogative Form and Function in Cairene
Arabic. Ph.D. Dissertation. University of Leeds.
Osman, Mariam Hussein (1990) Bound Anaphora in Egyptian Arabic.
Perspectives on Arabic Linguistics I. ed. by Mushira Eid. 155-77.

257

Owens, Jonathan (1988) The Foundations of Grammar: An Introduction to

Medieval Arabic Grammatical Theory. Amsterdam, Philadelphia:
John Benjamins.
Parkinson, Dilworth B.(1981) VSO to SVO in Modern Standard Arabic:
A Study in Diglossia Syntax. Al-cArabiyya 14: 24-37.
Rabin, Chaim (1965) The Diptote Declension. Arabic and Islamic
Studies in Honor of Hamilton A. R. Gibb. Cambridge: Harvard
University Press. 547-62.
Radwan, Mohamed Ramzy (1975) A Semantico-Syntactic Study of the
Verbal Piece in Colloquial Egyptian Arabic. Ph.D. Dissertation.
University of London.
Ritter, Elizabeth (1988) A Head-Movement Approach to Construct-State
Noun Phrases. Linguistics 26: 909-29.
Robertson, Alice Marian (1970) Classical Arabic and Colloquial Cairene:
An Historical Linguistic Analysis. Ph.D. Dissertation. University
of Utah at Salt Lake City.
Saada, M. Hachemi (1966) The Journey from Classical to Contemporary
Arabic. Arab News (Winter, 66).
Salama, Ahmed Muhammad (1983) A Grammatical and Semantic Study
of the Verb in Written Arabic. Ph.D. Dissertation. London
University.
Salib, Maurice Boulos (1981) Spoken Arabic of Cairo. Cairo: American
University in Cairo Press.
Schmidt, Richard (1975) Sociolinguistic Variation in Spoken Egyptian
Arabic: A Reexamination of the Concept of Diglossia. Ph.D.
Dissertation Brown University.
Schramm, Gene M. (1962) An Outline of Classical Arabic Verb
Structure. Language 38: 360-75.
Schulz, David Eugene (1981) Diglossia and Variation in Formal Spoken
Arabic in Egypt. Ph.D. Dissertation. The University of Wisconsin
Madison.
Selim, George (1967) Some Contrasts between Classical Arabic and
Egyptian Arabic. Linguistic Studies in Memory of Richard Slade
Harrell. ed. by Don Graham Stuart. Washington, DC: Georgetown
University Press.
Smith, Percy (1917) A Plea for Literature in Vernacular Arabic. The
Muslim World 7: 333-43.
Suleiman, Saleh Mahmoud (1986) Jordanian Arabic between Diglossia
and Bilingualism. Amsterdam, Philadelphia: John Benjamins.
Talaat, Hala Muhammad (1987) The Verb Phrase in Egyptian Arabic.
Ph.D. Dissertation. Michigan University.

258

Thackston, W. M. (1984) An Introduction to Koranic Arabic. Cambridge,

Mass: Department of Near Eastern languages and Civilizations,
Harvard University.
Thatcher, G. W. (1993) Arabic Grammar of the Written Language. New
York: Hippocrene Books.
Thomason, Sarah (1973) Language Classification and Language
Structures. University of Pittsburgh Press.
Trager, George L. & Rice, Frank A. (1954) The Personal-Pronoun
System of Classical Arabic. Language 30: 224-9.
Trask, R. L. (1996) A Dictionary of Phonetics and Phonology. New York:
Routledge.
Travis, D. Ann (1979) Inflectional Affixation in Transformational
Grammar: Evidence from the Arabic Paradigm. Indiana: Indiana
University Linguistics Club.
Tritton, A. S. (1958) Teach Yourself Arabic. London: The English
Universities Press.
Versteegh, Kees (1984) Pidginization and Creolization: The Case of
Arabic. Amsterdam, Philadelphia: John Benjamins.
Weiss, Bernard (1976) A Theory of the Parts of Speech in Arabic. (Noun,
Verb, Particle): a Study of cilm al-waDc . Arabica 32: 23-36.
Wehr, Hans (1961) A Dictionary of Modern Written Arabic. ed. by J.
Milton Cowan. Cornell: Cornell University Press.
Wise, Hilary (1975) A Transformational Grammar of Spoken Egyptian
Arabic. Oxford: Basil Blackwell.
Wright, William (1967) A Grammar of the Arabic Language. 3rd ed.
Cambridge: Cambridge University Press.
Yushmanov, N. W. (1961) The Structure of the Arabic Language. Trans.
from Russian by Moshe Perlmann. Washington, DC: Center for
Applied Linguistics.
Ziadeh, Farhat Jacob (1957) An Introduction to Modern Arabic. Princeton:
Princeton University Press.
Zoll, Cheryl Cydney (1996) Parsing Below the Segment in a Constraint
Based Framework. Ph.D. Dissertation. University of California at
Berkeley.

259

Appendix
Possible Broken Plural Patterns for Singular Patterns in SA &EA
Sg. Pattern
Pl. Patterns
Example
Fa3L(-un)
aF3aaL(-un) nahr(-un) a river anhaar(-un)
Fu3uuL(-un) ahr(-un) a month uhuur(-un)
Fi3aaL(-un)
bahr(-un) a sea bihaar(-un)
aF3uL(-un)
nafs(-un) self anfus(-un)
Fawaa3iL(-u) ?amr(-un) an order ?awaamir(-u)
Fa3iiL(-un)
3abd(-un) a slave 3abiid(-un)
Fa3aali(n)
?ahl(-un) relatives ?ahaali(n)
Fa3Lat(-un) Fu3aL(-un)
dawlat(-un) a country duwal(-un)
Fi3L(-un)
Fu3uuL(-un) jild(-un) leather juluud(-un)
aF3aaL(-un) himl(-un) load ahmaal(-un)
Fi3Lat(-un) Fi3aL(-un)
3ilbat(-un) a packet 3ilab(-un)
aF3aaL(-un) fikrat(-un) an idea afkaar(-un)
Fu3L(-un)
aF3aaL(-un) burj(-un) a tower abraaj(-un)
Fu3uuL(-un) burj (-un) a tower buruuj(-un)
Fi3aaL(-un)
rumh(-un) an arrow rimaah(-un)
Fu3Lat(-un) Fu3aL(-un)
furSat(-un) a chance furaS(-un)
Fu33(-un)
Fi3a3at(-un) dubb(-un) a bear dibabat(-un)
FaaL(-un)
aFwaaL(-un) baab(-un) a door abwaab(-un)
FiiLaan(-un) jaar(-un) a neighbor jiiraan(-un)
FiiL(-un)
aFyaaL(-un) fiil(-un) an elephant afyaal(-un)
FuuL(-un)
aFwaaL(-un) nuur(-un) a light anwaar(-un)
FuuLat(-un) FuwaL(-un)
Suurat(-un) a picture Suwar(-un)
Fa3aL(-un) aF3aaL(-un) haram(-un) a pyramid ahraam(-un)
Fi3aaL(-un)
balad(-un) a town bilaad(-un)
Fi3aaLat(-un) hajar(-un) a stone hijaarat(-un)
Fu3uuL(-un) ?asad(-un) a lion ?usuud(-un)
Fa3aLat(-un) Fi3aaL(-un)
raqabat(-un) a neck riqaab(-un)
Fa3iL(-un)
Fu3uuL(-un) malik(-un) a king muluuk(-un)
Fu3uL(-un) aF3aaL(-un) 3unuq(-un) a neck a3naaq(-un)
Fu3aaL(-un) Fi3Laan(-un) guraab(-un) a crow girbaan(-un)
aF3iLat(-un) su?aal(-un) a question as?ilat(-un)

260

Appendix Cont.
rasuul(-un) a messenger rusul(-un)
xaruuf(-un) a sheep xirfaan(-un)
3aruusat(-un) a bride 3araa?is(-u)
makaan(-un) a place amaakin(-u)
kitaab(-un) a book kutub(-un)
silaah a weapon aslihat(-un)
risaalat(-un) a message rasaa?il(-u)
Tariiq(-un) a way Turuq(-un)
ahiid(-un) a martyr uhadaa?(-u)
nabiyy(-un) a prophet anbiyaa?(-u)
mariiD(-un) a patient marDaa
yatiim(-un) an orphan yataamaa
yatiim(-un) an orphan aytaam(-un)
hadii(-un) a talk ahaadii(-u)
safiinat(-un) a ship sufun(-un)
3ajiibat(-un) a wonder 3ajaa?ib(-u)
haakim(-un) a ruler hukkaam(-un)
Taabi3(-un) a stamp Tawaabi3(-u)
Saahib(-un) a friend aShaab(-un)
aahid(-un) a witness uhuud(-un)
3aalim(-un) a scientist 3ulamaa?(-u)
saahir(-un) a magician saharat(-un)
3aaSifat(-un) a storm 3awaaSif(-u)
kawkab(-un) a planet kawaakib(-u)
bulbul(-un) a nightingale balaabil(u)
Fu3L1uL2at(-un) Fa3aaL1iL2(-u) sunbulat(-un) an ear of corn
sanaabil(-u)
Fi3L1aL2(-un) Fa3aaL1iL2(-u) dirham(-un) currency unit
daraahim(-u)
Fu3L1aaL2(-un) Fa3aaL1iiL2(-u) sulTaan(-un) a ruler salaaTiin(-u)
Fa3uuL(-un)

Fu3uL(-un)
Fi3Laan(-un)
Fa3uuLat(-un) Fa3aa?iL(-u)
Fa3aaL(-un)
aFaa3iL(-u)
Fi3aaL(-un)
Fu3uL(-un)
aF3iLat(-un)
Fi3aaLat(-un) Fa3aa?iL(-u)
Fa3iiL(-un)
Fu3uL(-un)
Fu3aLaa?(-u)
aF3iLaa?(-u)
Fa3Laa
Fa3aaLaa
aF3aaL(-un)
aFaa3iiL(-u)
Fa3iiLat(-un) Fu3uL(-un)
Fa3aa?iL(-u)
Faa3iL(-un)
Fu33aaL(-un)
Fawaa3iL(-u)
aF3aaL(-un)
Fu3uuL(-un)
Fu3aLaa?(-u)
Fa3aLat(-un)
Faa3iLat(-un) Fawaa3iL(-u)
Fa3L1aL2(-un) Fa3aaL1iL2(-u)
Fu3L1uL2(-un) Fa3aaL1iL2(-u)