7188755 the Expletive Fiih in Palestinian Arabic

7/29/2019 7188755 the Expletive Fiih in Palestinian Arabic

1/56

1

The Syntax of Indefinite Subjects in Equative Sentences in Palestinian ArabicM. A. Mohammad

University of Florida

Abstract

In this paper I provide what I believe to be the most detaileddescription of the conditions under which equative sentences with indefinitesubjects are licensed in Palestinian Arabic. I provide a detailed description ofthe expletivefiih in this dialect. I show that in spite of its probable historicalderivation from the prepositionfii 'in,' I show that it is not showing any of theusual properties that are the trade-mark of prepositions in Arabic. Forexample, prepositions never impose a definiteness requirement on theircomplements. Fiih, by contrast, with one exception, never permits a definiteNP to occur after it. I also show that the internal composition offiih is as Itranscribed it, namely, that it consists offii and the third person masculinepronoun -h. I argue thatfiih is an expletive pronoun that is brought in as asubject to license an indefinite subject in an equative sentence. I study someinteresting subject-verb agreement facts involving kaan 'was' andfiih, andargue thatfiih can be involved in licensing agreement features on the verb.

1. Introduction

Descriptively speaking there are two major sentence types in

Palestinian Arabic (PA, henceforth): verbal and equative sentences. A verbal

sentence is one that contains an overt verb whereas an equative sentence is a

sentence that contains no verbs. This paper is primarily concerned with the

latter type. Specifically, this paper is an attempt of provide a detailed

description of the syntax of equative sentences with indefinite subjects .

To set the stage, here are some examples that are representative of the

equative type of sentences :

(a) The subject is a definite NP with an indefinite adjectival predicate:

(1) el-walad gaSiir


2/56

2

the-boy short

"The boy is short."

(b) The subject is a definite NP with a Prepositional Phrase as its predicate:

(2) el-bent b-d-ddar

the-girlin-the-house

"The girl is in the house."

(c) The subject is a definite NP with the predicate an adverb of place:

(3) ?ez-zalame hoon

the-man here

"The man is here."

(d) The subject is a definite NP with an indefinite NP as its predicate :

(4) ?el-mara daktoora

the-woman doctor

The woman is a doctor."

(e) The subject is a definite with an indefinite construct state NP as

predicate :

(5) el-bent ?uxt waziir

the-girlsister minister

"The girl is a minister's (as in member of the cabinet) sister."

(f) The subject is a definite NP with a definite construct state NP as

predicate :


3/56

3

(6) el-bent ?uxt l-waziir the-girlsister the-minister

"The girl is the minister's sister."

"the girl, the minister's sister"

As can be seen from glosses, (6) is ambiguous between a sentence and a NP

reading. To disambiguate it, PA resorts to the pronoun of separation strategy:

(7) el-bent hi ?uxt l-waziir

the-girlshe sister the-minister

"The girl is the minister's sister."

(7) has only a sentential reading.

(g) The subject is a definite NP with an active participle as its predicate :

(8) el-bent naayme

the-girl asleep'The girl is asleep.'

(h) The subject is a definite NP with a passive participle as the predicate :

(9) el-mara maHboose

the-woman imprisoned

"The woman is imprisoned."

Word order in all the examples above is rigid on the readings given in

their translations. As noted all the examples have a definite subject. As is


4/56

4

well-known in Arabic, a sentence with no verb is automatically interpreted as a

sentence in the present tense. In order to obtain a past tense sentence, PA uses

past tense form of the root kwn 'to be." Consider all the sentences above inthe past:

(10) el-walad kaan gaSiir

the-boy was short

"The boy was short."

(11) el-bent kaanat be-d-daat

the-girlwas in-the-house

"The girl was in the house."

(12) ez-zalame kaan hoon

the-man was here

"The man was here."

(13) el-mara kaanat daktoora

the-woman was doctor

The woman was a doctor."

(14) el-bent kaanat ?uxt waziir

the-girlwas sister minister

"The girl was a minister's (as in member of the cabinet) sister."

(15) el-bent kaanat ?uxt l-waziir

the-girlwas sister the-minister

"The girl was the minister's sister."

(16) el-bent kaanat naayme


5/56

5

the-girl was asleep

'The girl was asleep.'

(17) el-mara kaanat maHboosethe-woman was imprisoned

"The woman was imprisoned."

The interesting thing about the sentences with the past tense verb to be

is that word order is considerably free. In addition to S(ubject), V(erb)

P(redicate) word order, all of them allow three other word orders: VSP, VPS,

and PVS. They all disallow SPV and PSV. I.e., the constraint is that the

subject and the predicate cannot precede the verb at the same time.

2.0. Equative Sentences with Indefinite Subjects1:

1I must point the reader's attention to one important point regarding indefinite subjects in

equatives sentences. The restriction on indefinite subjects to be elaborated on in the text is

not absolute. If the indefinite subject is heavily focused, 'newsworthy,' then it can indeed

occur in a pre-predicate position in a simple equative sentence. Consider for example the

following Palestinian proverbs:

(i) gerd mwaalef wa-la ghazaal mxaalef

monkey affectionate and-not gazelle distant

"A monkey that is affectionate is better than a distant gazelle."

(ii) 9anze walaw Taarat

goat even flies.3sf

"(I still say that it is a goat, even it can fly."

The first proverb is used in reference to the unimportance of appearances. The second is

used when someone is so obstinate and would not budge, no matter what evidence you bring.

2I must point the reader's attention to one important point regarding indefinite subjects in

equatives sentences. The restriction on indefinite subjects to be elaborated on in the text is

not absolute. If the indefinite subject is heavily focused, 'newsworthy,' then it can indeed

occur in a pre-predicate position in a simple equative sentence. Consider for example the

following Palestinian proverbs:

(i) gerd mwaalef wa-la ghazaal mxaalef monkey affectionate and-not gazelle distant


6/56

6

PA imposes strict conditions on its equative sentences with indefinite

subjects. Consider all the examples of the previous section but with the subjectnow indefinite3:

(18)* walad gaSiir

boy short

"A boy is short."

(19)* bent be-d-daar

girl iin-the-house

"A girl is in the house."

(20)* zalame hoon

man here

"A man is here."

(21)* mara daktoorawomandoctor

"A woman is a doctor."

"A monkey that is affectionate is better than a distant gazelle."

(ii) 9anze walaw Taarat

goat even flies.3sf

"(I still say that it is a goat, even it can fly."

The first proverb is used in reference to the unimportance of appearances. The second is

used when someone is so obstinate and would not budge, no matter what evidence you bring.

3Many of the examples below are grammatical as NP's. The point is that none isgrammatical as a sentence.


7/56

7

(22)* bent ?uxt waziir

girl sister minister "A girl is a minister's (as in member of the cabinet) sister."

(23)* bent ?uxt l-waziir

girl sister the-minister

"A girl is the minister's sister."

(24)* bent naayme

the-girl asleep

"A girl is asleep"

(25)* mara maHboose

womanimprisoned

"A woman is imprisoned."

As can be seen from the paradigm above an indefinite subject inducesungrammaticality in equative sentences. When the predicate is a PP or an

adverbial, the sentences improve sharply if the indefinite subject is modified.

Consider:

(26)? bent Tawiilebe-d-daar

girl tall in-the-house

"A tall girl is in the house."

(27)? zalame gaSiir hoon

man short here


8/56

8

"A short man is here."

3.0. Possessive and Locative PP's

It is perhaps useful to compare PA with Modern Standard Arabic

(MSA). The same facts described above obtain in MSA with one major

difference, namely when the subject of an equative sentence is indefinite and

the predicate is a PP. In such cases MSA permits permutation between the

subject and the predicate. Consider4:

(28) rajul-un fi l-bayt-i MSA

man-NOM in the-house-GEN

"A boy is in the house."

(28) has, the perhaps preferred, (29) as an alternative:

(29) fi l-bayt-i rajul-un MSA

in the-house-GEN man-NOM"(There is ) a man is in the house."

This option of permuting the subject and the predicate in MSA is available only when

the predicate is a prepositional phrase. This permutation strategy is not available to the

majority of prepositions in PA. Consider:

4Some readers may be surprised as to why (28) is not starred. I concur with Fassi Fehri

(1993) and (1989) in considering (28) grammatical. This, as the reader may know,contradicts the Arab grammarians' position that (28) is ungrammatical; and that (28) is only

grammatical if the subject and the predicate are permuted as in (29). However, asMoutaouakil (1989) points out, (28) is grammatical only if it is interpreted as contrastive.


9/56

9

(30)(a)* walad be-d-daar

boy in-the-house

"(There is) a boy in the house."

(b)* be-d-daar walad

in-the-house boy


(31)(a)* ktaab 9ala T-Taawle

book on the-table

"(There is) a book on the table."

(b)* 9ala T-Taawle ktaab

on the-table book


(32)(a)? xarariif kthiire 9an falaSTiin

stories many about Palestine

"(There are) many stories about Palestine."

(b)* 9an falaSTiin xarariif kthiireabout Palestine stories many


(33)(a)* walad zay Hmad

boy as Ahmed

"(There is) a boy that looks like Ahmed."

(b)*zay Hmad walad

as Ahmedboy



10/56

10

(34)(a)* ktaab teHt T-Taawle

book under the-table"(There is) a book under the table."

(b)*teHt T-Taawle ktaab

under the-table book

"(There is) a book under the table."

(35)(a)* 9aSfoor foog l-HeeT

bird above the-roof

"(There is) a bird on the roof."

(b)* foog l-HeeT 9aSfoor

above the-roof bird


(36)(a)* kundara janb l-baabshoe beside the-door

"(There is) a shoe next to the door."

(b)* janb l-baab kundara

beside the-door shoe


(37)(a)* kundara wara l-baab

shoe behind the-door

"(There is) a shoe behind the door."


11/56

11

(b)*wara l-baab kundara

behind the-door shoe"(There is) a shoe behind the door."

(38)(a)* bent been l-waladeen

girl between the-two boys

"(There is) a girl between the two boys."

(b)*been l-waladeen bent

between the-two boys girl


The only prepositions that permit (in reality, require) this permutation

are: ?el-/la-, ma9, and 9ind. ?el- and la are bound morphemes, where la- is

used with non-pronominal lexical NPs and ?el- is used with pronominal clitics.

Consider:

(39)(a) la-mona ktaab

to-Mona book"Mona has a book."

(b)* ktaab la-mona

book to-Mona

"Mona has a book."

(40)(a) ?el-i ktaab

to-me book

"I have a book."


12/56

12

(b)* ktaab ?el-i

to-me to-me

"I have a book."

(41)(a) 9ind mona ktaab

with Mona book

"Mona has a book."

(b)* ktaab 9ind mona

book with Mona

"Mona has a book."

(42)(a) ma9 mona ktaab

with Mona book

"Mona has a book with her."

(b)* ktaab ma9 mona

book with Mona

"Mona has a book with her."

(39) means that Mona owns a book. (41) is ambiguous between two readings(i) Mona owns a book or (ii) Mona has a book, say in his house, without any

implication as to the ownership of the book. Let us refer to these two

interpretations as possessive and locative, respectively. (42) has only the

locative meaning where Mona has the book on her, irrespective of the fact

whether she owns the book or not. That is to say, the three sentences above

have some kind of a continuum: (39) has the possessive reading only, (41) the

locative and possessive, and (42) has only the locative meaning. Consider now

what happens if the NP ktaab 'a book' is made definite:

(43)(a) la-mona le-ktaab


13/56

13

to-Mona the-book

"Mona has the book."

(b) le-ktaab la-mona

book to-Mona

"Mona has a book."

(44)(a) 9ind mona le-ktaab

with Mona book


(b) le-ktaab 9ind mona

book with Mona


(45)(a) ma9 mona le-ktaab

with Mona book

"Mona has the book with her."

(b) le-ktaab ma9 monabook with Mona


Pragmatically speaking, the (b) sentences above are more neutral than the (a)

sentences. The (a) sentences are only acceptable on a contrastive reading

along the lines 'Mona has x but not y." As can be seen, these three prepositions

are in a class by themselves.

4.0. Negation and Equative Sentences with Indefinite Subjects


14/56

14

Interestingly, when some of the equative sentences with an indefinite

subject and a PP as a predicate are negated their grammatical status slightly

improves depending on whether the preposition is marginally capable ofamalgamating with negation or not. Consider:

(46)(a)* ktaab ma-9ala- T-Taawle

book neg-on-neg the-table

"(There is) not a book on the table."

(b)* ma-9ala- T-Taawle ktaab

neg-on-neg the-table book

"(There is) not a book on the table."

(47)(a)* xarariif kthiire ma-9an- falaSTiin

stories many neg-about-neg Palestine

"(There are) not many stories about Palestine."

(b)* ma-9an- falaSTiin xarariif kthiireneg-about-neg Palestine stories many

"(There are) not many stories about Palestine."

(48)(a)* walad ma-zayye- Hmad

boy neg-as-neg Ahmed

"(There is) not a boy that looks like Ahmed."

(b)*ma-zayye- Hmad walad

neg-as-neg Ahmedboy

"(There is) not a boy that looks like Ahmed."


15/56

15

(49)(a)* ktaab ma-teHte- T-Taawle

book neg-under-neg the-table"(There is) not a book under the table."

(b)? ma-teHte- T-Taawle ktaab

neg-under-neg the-table book

"(There is) not a book under the table."

(50)(a)* 9aSfoor ma-foog- el-HeeT

bird neg-above-negthe-roof

"(There is) not a bird on the roof."

(b)? ma-foog- el-HeeT 9aSfoor

neg-above-negthe-roof bird

"(There is) not a bird on the roof."

(51)(a)* kundara ma-janbe- l-baabshoe neg-beside-neg the-door

"(There is) not a shoe next to the door."

(b)? ma-janbe- l-baab kundara

neg-beside-neg the-door shoe

"(There is) not a shoe next to the door."

(52)(a)* kundara ma-waraa- l-baab

shoe neg-behind-neg the-door

"(There is) not a shoe behind the door."


16/56

16

(b)?ma-waraa- l-baab kundara

neg-behind-neg the-door shoe"(There is) not a shoe behind the door."

(53)(a)* bent ma-been- el-waladeen

girl neg-between-neg the-two boys

"(There is) not a girl between the two boys."

(b)?ma-been- el-waladeen bent

neg-between-neg the-two boys girl

"(There is) not a girl between the two boys."

The generalization that emerges from the above examples is: negation

improves the grammaticality status of an equative sentence with indefinite subject so

long as the subject is placed in a post-predicate position. In so far as the three

prepositions ?el-/la-, ma9, and 9end, only 9endis marginally capable of amalgamating

with negation. Consider:

(54)(a)* ma-la- mona ktaabneg-to-neg Mona book

"Mona has no book."

(b)* ktaab ma-la- mona

book neg-to-neg Mona

"Mona has no book."

(55)(a)? ma-9ende- mona ktaab

neg-with-neg Mona book

"Mona has no book."


17/56

17

(b)* ktaab ma-9ende- mona

book neg-with-neg Mona"Mona has no book."

(56)(a)* ma-ma9e- mona ktaab

neg-with-neg Mona book

"Mona has no book with her."

(b)* ktaab ma-ma9e- mona

book neg-with-neg Mona

"Mona has no book with her."

Negation with these prepositions marginally changed the grammaticality status of the

9endsentences. la- and ma9 do not amalgamate with negation. Hence, the

ungrammaticality of (54) and (56).

When a pronominal clitic is used, all the prepositions mentioned in the previous

paragraphs are capable of merging with negation. This always results in producing

grammatical sentences as long the prepositional phrase precedes the subject.

Consider:

(57)(a) ma-l-ii- ktaab

neg-to-me-negbook

"I have no book."

(b)* ktaab ma-?el-ii-

book neg-to-me-neg

"I have no book."

(58)(a) ma-9end-oo- ktaab


18/56

18

neg-with-him-neg book

"He has no book."

(b)* ktaab ma-9end-oo-

book neg-with-him-neg

"He has no book."

(59)(a) ma-ma9-haa- ktaab

neg-with-her-neg book

"She has no book with her."

(b)* ktaab ma-ma9-haa-

book neg-with-her-neg

"She has no book with her."

The exact same situation obtains with all the prepositions mentioned so far.

5.0. The Expletive fiih and Equative Sentences with Indefinite Subjects

In the previous paragraphs we have seen that whether a grammatical sentence

emerges when an indefinite subject is permuted with its predicate or not depends on

the preposition. We also saw that clitics play a critical role in the ability of all

prepositions to amalgamate with negation. The question is, then, how does PA rescue

the ungrammatical sentences?

When the subject of an equative sentence is indefinite PA resorts to the

use of an existential expletive:fiih. Thus, all the ungrammatical sentences


19/56

19

above become grammatical whenfiih appears. Further, thisfiih can co-occur

with any PP. Consider the following:

(60)(a) fiih bent men ?ameerka

there girl from America

"There is a girl from America."

(b) fiih men ?ameerka bent

there from America girl

"There is a girl from America."

(61)(a) fiih zalame hoon

there man here

"There is man here."

(b) fiih hoon zalame

there here man

"There is man here."

(62)(a) fiih walad be-d-daar PA

there boy in-the-house"(There is) a boy in the house."

(b) fiih be-d-daar walad PA

there in-the-house boy


(63)(a) fiih ktaab 9ala T-Taawle

there book on the-table



20/56

20

(b) fiih 9ala T-Taawle ktaab

there on the-table book


(64)(a) fiih xarariifkthiire 9an falaSTiin

there stories many about Palestine


(b) fiih 9an falaSTiin xarariif kthiire

there about Palestine stories many


(65)(a) fiih walad zay Hmad

there boy as Ahmed


(b) fiih zay Hmad walad

there as Ahmedboy"(There is) a boy that looks like Ahmed."

(66)(a) fiih ktaab teHt T-Taawle

there book under the-table


(b) fiih teHt T-Taawle ktaab

there under the-table book



21/56

21

(67)(a) fiih 9aSfoor foog l-HeeT

there bird above the-roof


(b) fiih foog l-HeeT 9aSfoor

there above the-roof bird


(68)(a) fiih kundara janb l-baab

there shoe beside the-door


(b) fiih janb l-baab kundara

there beside the-door shoe


(69)(a) fiih kundara wara l-baab

there shoe behind the-door "(There is) a shoe behind the door."

(b) fiih wara l-baab kundara

there behind the-door shoe

"(There is) a shoe behind the door."

(70)(a) fiih bent been l-waladeen

there girl between the-two boys



22/56

22

(b) fiih been l-waladeen bent

there between the-two boys girl


As the above facts show whenfiih is used word order between the

subject and the predicate is free.

6.0. Properties offiih

6.1. Fiih's Internal Composition

The careful reader would have noticed that I have transcribedfiih with

a pronominal clitic without justification. The question is: doesfiih contain the

third person masculine singular pronominal clitic as my transcription claims?

Two pieces of evidence can be presented in favor of my transcription.

First, recall that all the prepositions were able to amalgamate with

negation once they have a pronoun cliticized unto them. Recall also that none

of them produced a perfectly grammatical expression when they amalgamate

by themselves with negation. fiih produces perfectly grammatical sentences

when amalgamating with negation. This suggests that it is patterning withprepositions that do contain a clitic. I take this as an argument in favor of

treatingfiih as a combination of the prepositionfii 'in' and the third person

singular masculine clitic -h.

Second, on the assumption thatfiih derives historically form the

prepositionfii 'in', we can show that it contains a clitic when we consider the

behavior of the prepositionfii 'in' PA. The propositionfii in PA has two

variants, namely bi- andfi-. bi is used when it governs a non-pronominal

lexical NP whilefii is used with pronominal clitics. Consider:


23/56

23

(71) mona be-d-daar

Mona in-the-house

"Mona is in the house."

(72) mona fe-d-daar

Mona in-the-house

"Mona is in the house."

It is important to note the vowel [e] between [f} and [d] infeddaarand

between [b] and [d] in beddaaris short. It is even shorter in the former.

When these two prepositions have a clitic suffixed to them the vowels are

lengthened. Consider:

(73)(a) mona bii-haa

Mona in-it

"Mona is in it."

(b)*mona be-haa

Mona in-it

"Mona is in it."

(74)(a) mona fii-haa

Mona in-it

"Mona is in it."

(b)*mona fe-haa

Mona in-it

"Mona is in it."


24/56

24

Given that the vowel infiih is long, I conclude that it does indeed contain a

clitic.

6.2. Fiih and the Definiteness Effects

With one exception to be noted below,fiih never permits a definite

subject to follow it regardless of word order between the subject and the

predicate. Consider:

(75)(a)* fiih la-mona le-ktaab

there to-Mona the-book


(b)* fiih le-ktaab la-mona

there book to-Mona

"Mona has a book."

(76)(a)* fiih 9ind mona le-ktaab

there with Mona book"Mona has the book."

(b)* fiih le-ktaab 9ind mona

there the-book with Mona


(77)(a)* fiih ma9 mona le-ktaab

there with Mona book



25/56

25

(b)* fiih le-ktaab ma9 mona

there book with Mona"Mona has the book with her."

(78)(a)* fiih l-kundara janb l-baab

there the-shoe beside the-door

"There is the shoe next to the door."

(b)*fiih janb l-baab l-kundara

there beside the-door the-shoe

"There is the shoe next to the door."

(79)(a)* fiih l-kundara wara l-baab

there shoe behind the-door

"There is the shoe behind the door."

(b)*fiih wara l-baab l-kundara

there behind the-door the-shoe"(There is) the shoe behind the door."

(80)(a)* fiih l-bent been l-waladeen

there the-girl between the-two boys

"(There is) the girl between the two boys."

(b)*fiih been l-waladeen l-bent

there between the-two boys the-girl

"(There is) the girl between the two boys."


26/56

26

The only environment that I am aware of wherefiih may co-occur with

a definite subject is when the sentence is intended as a list of options which

may include only one item. Consider the following:

(81) fiih l-baaS we-t-taksi

there the-busand-the-cab

"There is the bus and the cab."

(81) gives a list in answer to a question, say, "How shall we go to Amman?"

(82) fiih Hmad w-mona w-rasmiyye

there Ahmedand-Mona and-Rasmiyye

"There is Ahmed, Mona, or Rasmiyye."

(82) is an appropriate answer to a question along the lines "Who shall we

send?"

(83) fiih ed-d-ktoor Hmad

there the-d-doctor Ahmed

"There is doctor Ahmed."

(82) is an appropriate response to a remark such as "I am not feeling well and I

do not know who to see?"

The examples (81), (82), and (83) seem to suggest that a definite NP

may followfiih as long as this NP is not part of a sentence. Thus, the

generalization is: iffiih is followed by a sentence, the subject of this sentence

must be indefinite.

6.3. Word Order and fiih


27/56

27

Recall that word order in equative sentences when the subject is

indefinite and when these sentences are grammatical is rigid. In this section we

look more closely at word order variation withfiih. Consider:

(84)(a) fiih bent be-d-daar

there girl in-the-house

"There is a girl in the house."

(b) fiih be-d-daar bent

there in-the-house girl


(c) be-d-daar fiih bent

in-the-house there girl


(d)* be-d-daar bent fiih

in-the-house girl here"There is a girl in the house."

(e)*bent be-d-daar fiih

girl in-the-house there


(f)* bent fiih be-d-daar

girl here in-the-house



28/56

28

The contrasts in (84) show that the only restriction on word order is that the

indefinite subject may never precedefiih.

7.0. Equative Sentences with Indefinite Subjects and Past Tense

As we noted earlier an equative sentence can be put in the past tense by

using the past tense form of the verb to be kwn. When the verb to be is used in

a sentence with an indefinite subject, the sentence behaves just like any

sentence with a verb. Consider first a simple sentence with ordinary main

verbs:

(85)(a) weSel Hmad

arrived.3sm Ahmed

"Ahmed arrived."

(b) eHmad weSel

Ahmed arrived

"Ahmed arrived."

(86)(a) weSel walad

arrived.3sm boy"A boy arrived."

(b)*walad weSel

boy arrived.3sm

"A boy arrived."

(85) shows that when the subject is definite, the subject may either precede or

follow the verb. (86), on the other hand, shows that when the subject is

indefinite, the subject may not precede the verb.


29/56

29

Consider now what happens when we use past tense kwn in equative

sentences with indefinite subjects:

(87)(a)* kaan walad be-d-daar

was boy in-the-house

"(There was) a boy in the house."

(b) kaan be-d-daar walad

was in-the-house boy

"(There was) a boy in the house."

(88)(a)* kaan ktaab 9ala T-Taawle

was book on the-table

"(There was) a book on the table."

(b) kaan 9ala T-Taawle ktaab

was on the-table book

"(There was) a book on the table."

(89)(a)* kaan xarariif kthiire 9an falaSTiinwas stories many about Palestine

"(There were) many stories about Palestine."

(b) kaan 9an falaSTiin xarariif kthiire

was about Palestine stories many

"(There were) many stories about Palestine."

(90)(a)* kaan walad zay Hmad

was boy as Ahmed

"(There was) a boy that looks like Ahmed."


30/56

30

(b)*kaan zay Hmad walad

was as Ahmedboy"(There was) a boy that looks like Ahmed."

(91)(a) kaan ktaab teHt T-Taawle

was book under the-table

"(There was) a book under the table."

(b) kaan teHt T-Taawle ktaab

was under the-table book

"(There was) a book under the table."

(92)(a)* kaan 9aSfoor foog l-HeeT

was bird above the-roof

"(There was) a bird on the roof."

(b) kaan foog l-HeeT 9aSfoor was above the-roof bird

"(There was) a bird on the roof."

(93)(a)* kaan kundara janb l-baab

was shoe beside the-door

"(There was) a shoe next to the door."

(b) kaan janb l-baab kundara

was beside the-door shoe

"(There was) a shoe next to the door."


31/56

31

(94)(a)* kaan kundara wara l-baab

was shoe behind the-door "(There was) a shoe behind the door."

(b) kaan wara l-baab kundara

was behindthe-door shoe

"(There was) a shoe behind the door."

(95)(a)* kaan bent been l-waladeen

was girl between the-two boys

"(There was) a girl between the two boys."

(b) kaan been l-waladeen bent

was between the-two boys girl

"(There was) a girl between the two boys."

(96)(a) kaan la-mona ktaabwas to-Mona book

"Mona had a book."

(b) kaan ktaab la-mona

was book to-Mona

"Mona had a book."

(97)(a) kaan ?el-i ktaab

was to-me book

"I had a book."


32/56

32

(b)* kaan ktaab ?el-i

was book to-me"I had a book."

(98)(a) kaan 9ind mona ktaab

was with Mona book

"Mona had a book."

(b)* kaan ktaab 9ind mona

was book with Mona

"Mona had a book."

(99)(a) kaan ma9 mona ktaab

was with Mona book

"Mona had a book with her."

(b)* kaan ktaab ma9 monawas book with Mona


The above sentences leave no doubt that when kaan is used, all

equative sentences with indefinite subject are grammatical, provided that the

indefinite subject follows the predicate. All the above examples are

grammatical iffiih is used instead ofkaan. I, therefore, conclude that only the

presence offiih permits an indefinite subject to precede its predicate.

7.1. fiih and kaan in Equative Sentences with Indefinite Subjects


33/56

33

In the previous section we established that when kaan is used, it

licenses an equative sentence with an indefinite subject provided that thesubject is in post-predicate position. No such constraint was imposed on fiih.

In this section we explore facts of PA when bothfiih and kaan are used in the

same sentence. Let us take sentences with ma9 as a representative, noting that

all equative sentences with indefinite subject and a prepositional predicate

behave exactly like ma9. Consider:

(100)(a) fiih kaan ma9 mona ktaab

there was with Mona book


(b) fiih kaan ktaab ma9 mona

there was book with Mona


(c) fiih ma9 mona kaan ktaabthere with Mona was book


(d)* fiih ma9 mona ktaab kaan

there with Mona book was


(e) fiih ktaab kaan ma9 mona

there book was with Mona

"There is a book which Mona had (or used to have)."


34/56

34

(f)* fiih ktaab ma9 mona kaan

there book with Mona was"Mona had a book with her."

To spare the reader of having to go all through the twenty four

logically possible word orders, let me just summarize the constraints:

(101)(a) The indefinite subject may never precedefiih regardless of any

other word order changes that may take place.

(b) There is a strong preference to keepfiih and kaan adjacent

(c)kaan andfiih can never be stranded, i.e., left as the final

elements in the sentence5.

(d) The order betweenfiih and kaan is free, i.e., either can

precede or follow the other.

(e)There is a strong preference forkaan to precedefiih.

(g) For a sentence to be interpreted in past tense, i.e., 'there

was,' kaan andfiih must be kept adjacent.

Perhaps a brief comment on (f) and (g) is in order. Consider thefollowing sentences:

(102) be-l-Hafle kaan fiih mara bet9allem taariiix

5The only environment in whichfiih can be stranded is in wh-questions with or without

kaan. Consider:

(i) uu fiih

what there

"What is up?'

(ii) uu kaan fiih

what was there"What was there? I.e., what happened?"


35/56

35

in-the-party was there womanteaches.3sf history

"At the party, there was a woman who teaches history."

(103) be-l-Hafle fiih kaan mara bet9allem taariiix

in-the-party there was womanteaches.3sf history

"At the party, there was a woman who teaches history."

(104) be-l-Hafle fiih mara kaanat bet9allem taariiix

in-the-party there woman was.3sf teaches.3sf history

"At the party, there was a woman who used to teach history."

"At the party, there is a woman who used to teach history."

The difference between (102)-(103), on the one hand, and (104), on the other,

is that, as their respective translations indicate, we can add 'but now she

teaches Arabic' only to (104) without making a contradictory statement.

Furthermore, only (104) allows the ambiguity of tense in the matrix clause. I

take these two points as evidence that in (102) and (103) what followsfiih and

kaan is a NP. Whereasfiih and kaan are followed by a sentence in (104).

7.2. The Syntactic and Categorial Identity offiih

Recall that we have already provided evidence thatfiih must contain a

pronominal clitic and it is not simply a preposition. In addition to the arguments

presented later we add the fact that it is showing a very unusual property which casts

serious doubt on its being a simple preposition. This property is that prepositions in

Arabic (in all its varieties) are never stranded, nor do they ever occur without an NP.

This alone forces us to conclude that it is not a preposition.

Since we say thatfiih contains the prepositionfii and a clitic, does it follow

that it is a PP? At first sight the evidence suggests that it is indeed a PP. Recall that


36/56

36

all prepositions, without any exceptions, must merge with negation when there is clitic

prefixed to them. Consider the following negative sentence:

(105) l-HeeT ma-foogoo- l-9aSfoor

the-wall neg-above-negthe-bird

"The roof, the bird is not on it."

When the subject precedes its predicate in (105), negation may not merge with the

preposition:

(106)(a)* l-HeeT l-9aSfoor ma-foog-oo-

the-wall the-bird neg-above-it-eg


(b) l-HeeT l-9aSfoor me foog-oo

the-wall the-bird neg above-it


The examples in (106), found in all prepositions in PA, is totally absent withfiih.

Recall thatfiih has an almost absolute requirement that it be not followed by a definite

subject. This is unique tofiih. No other preposition in PA imposes such arequirement. This, I believe, casts serious doubt that it is a PP, in spite of its historical

origin.

7.3. Is fiih a Verbal Category?

In PA there are four categories that are able to merge with negation: (i)

prepositions with clitics; (ii) verbs; (iii) pronouns; and (iv) the polarity itemHada 'one.'

We have dismissedfiih as prepositional. Can it then be verbal?

Compared with any other verb in PA, it shows the following differences and

similarities. Consider first the following negative sentences:


37/56

37

(107)(a) el-walad ma-kaan- bel9ab be-l-Hakoora

the-boy neg-was.3sm-neg play.3sm in-the-garden"The boy was not playing in the garden."

(b) ma-kaan- el-walad bel9ab be-l-Hakoora

neg-was.3sm-neg the-boy play.3sm in-the-garden

"The boy was not playing in the garden."

As can be seen from (108) below, when there are two verbs in a sentence as in (107),

only kaan may merge with negation:

(108)(a)* el-walad kaan ma-bel9abe- be-l-Hakoora

the-boy was.3sm neg-play.3sm-neg in-the-garden

"The boy was not playing in the garden."

(b)* kaan el-walad ma-bel9abe- be-l-Hakoora

was.3sm the-boy neg-play.3sm-neg in-the-garden"The boy was not playing in the garden."

When kaan andfiih co-occur in a negative sentence, only kaan may merge

with negation and kaan must precedefiih:

(109)(a)* ma-fiih- kaan ma9 mona ktaab

neg-there-neg was with Mona book

"Mona had no book with her."

(b) ma-kaan- fiih ktaab ma9 mona


38/56

38

neg-was-neg there book with Mona


Recall that when the indefinite subject occurs betweenfiih and kaan,

the sentence is interpreted as a relative clause. The correctness of that

conclusion is confirmed with negative sentences. Compare the following

sentence:

(110) fiih ktaab ma-kaan- ma9 mona

there book neg-was-neg with Mona

"There is a book that was not with Mona."

(110) indicates that kaan overridesfiih to merge with negation if they belong

to the same clause.

Another verbal property thatfiih seems to have is its ability to occur

immediately following a wh-word; a position usually reserved for verbs. Consider:

(111)(a) miin gara le-ktaab

who read the-book

"Who read the book?"

(b)* miin le-ktaab gara

who the-book read

"Who read the book?"

(112)(a) uu gara l-walad

what read the-boy

"What did the boy read?"

(b)* uu l-walad gara


39/56

39

what the-boy read

"What did the boy read?"

As (111) and (112) indicate, wh-movement in PA triggers verb fronting. The verb

must move to immediately follow the wh-word. Consider the following withfiih:

(113)(a) miin fiih be-d-daar

who there in-the-house

"Who is in the house?"

(a)* miin be-d-daar fiih

who in-the-house there

"Who is in the house?"

(114)(a) u u fii 9ind-ak

what there at-you

"What do you have?"

(a)* uu 9ind-ak fii

what at-you there"What do you have?"

(113) and (114) are interesting examples since their affirmative counterparts allow the

following orders:

(115)(a) fiih walad de-d-daar

there boy in-the-house

"There is a boy in the house?"

(b) fiih de-d-daar walad


40/56

40

there in-the-house boy


(a) de-d-daar fiih walad

in-the-house there boy


(116)(a) fii 9ind-ak ktaab

there at-you book

"You have a book."

(b) fii ktaab 9ind-ak

there book at-you

"You have a book."

(c) 9ind-ak fii ktaab

at-you there book

"You have a book."

However, in these sentences, when questioned,fiih must immediately follow the wh-

word. Thus, it seems that it also shares this property with verbs.

The above are the similarities betweenfiih and other verbs. Next, we turn our

attention to differences. Consider first the following sentences in the past:

(117)(a) el-walad kaan bel9ab be-l-Hakoora

the-boy was.3sm play.3sm in-the-garden

"The boy was playing in the garden."

(b) kaan el-walad bel9ab be-l-Hakoora

was.3sm the-boy play.3sm in-the-garden


41/56

41

"The boy was playing in the garden."

As the examples above show the verb kaan can precede or follow its subject. Theabsolute restriction here when there is an auxiliary and a main verb, the main verb can

never precede the auxiliary:

(118)(a)* el-walad bel9ab kaan be-l-Hakoora

the-boy play.3sm was in-the-garden

"The boys was playing in the garden."

(b)* bel9ab el-walad kaan be-l-Hakoora

play.3sm the-boy was in-the-garden

"The boys was not playing in the garden."

No matter which word order we choose, if the main verb precedes the auxiliary

the sentence is ungrammatical. Now, comparefiih when it is accompanied by the

auxiliary kaan with (118):

(119)(a) fiih kaan be-d-daar walad

there was in-the-house boy"There was a boy in the house."

(b) kaan fiih be-d-daar walad

was there in-the-house boy

"There was a boy in the house."

(c) fiih kaan walad be-d-daar

there was boy in-the-house



42/56

42

(d) kaan fiih walad be-d-daar

was there boy in-the-house


(e) fiih walad kaan be-d-daar

there boy was in-the-house


(f)* kaan walad fiih be-d-daar

was boy there in-the-house


The important point that can be deduced from the paradigm in (119): fiih can precede

or follow kaan. This property shows that iffiih is a verbal element, then it is unusual

since verbs cannot ever precede kaan.

Let us now look at wh-questions that contain the auxiliary and a main verb. It

is impossible for the main verb to precede kaan. Consider:

(120)(a) uu kaanat l-bent btegra

what was.3sf the-girl read.3sf "What was the girl reading?"

(b)* uu btegra kaanat l-bent

what read..3sf was.3sf the-girl

"Where was the girl playing?"

Compare (120) above with (121) below where the sentence contains the

auxiliary kaan, andfiih:

(121)(a) miin kaan fii 9ind-ak


43/56

43

who was there at-you

"Who was with you (at your place)?"

(b) miin fii kaan 9ind-ak

who there was at-you

"Who was with you (at your place)?"

In (119) and (121)fiih shows a dual character in that it can occur immediately after a

wh-word, and it can precede kaan even in wh-questions. I say dual, since only verbs

must occur after a wh-word, and only non-verbal categories can precede kaan. This

casts doubt on consideringfiih a verbal element.

In addition to the above, the following arguments can be mustered against

consideringfiih a verbal element. Recall that we have established thatfiih requires an

indefinite subject: a property that , to the best of knowledge, is not shared by any other

verb in PA. Another unusual property for a verbal element thatfiih has is the total

absence of any agreement features that are always shown on verbs. Furthermore, if

fiih is a verbal element, it means that it is a predicate. As a predicate, it is behaving in

a very unusual manner, namely that it must always precede its subject. Added to that is

the fact that iffiih is a verbal element, it is very hard, if at all possible, to identify a

subject forfiih.

7.4. What is fiih?

Above, I have argued against consideringfiih prepositional or verbal. In this

section I identify it.fiih seems to belong to an extremely closed class of elements

which do not seem to be verbal. Yet, they merge with negation. Consider the

following example:

(122) ma-Hadaa- be-d-daar

neg-one-neg in-the-house


44/56

44

"Noone is in the house."

There can be no doubt thatHada is a noun. It dictates its agreement features of thirdperson masculine singular on the verb when it the subject. Consider:

(123) ma-Hadaa- be9ref 9rarabi

neg-one-neg know.3sm Arabic


What is so interesting aboutHada is the fact that it is unique among all nouns

in PA in that it is the only noun that not only merges with negation but it can merge

with negation even when a verb is present as in (123) indicates. This merger is

optional: negation may merge with either the verb or with Hada6. Consider:

(124) ma-be9ref- Hada 9rarabi

know.3sm neg-one-neg Arabic


As we saw earlier,fiih, unlikeHada, cannot merge with negation ifkaan is

present. This is probably due to the fact thatHada is so intimately related withnegation:Hada owes its very existence to negation, so to speak.

Consider now howHada andfiih interact. Recall from footnote (4) that

Hada, independently offiih, can only occur in either a negative or an interrogative

6The absolute restriction on Hada is that it can never precede negation:(i)* Hada ma-be9ref- 9rarabi

know.3sm neg-one-neg Arabic"Noone is in the house."

(i) can be grammatical if given a yes-or-no question intonation contours. This is so sinceHada can only be licensed by either negation or interrogation. The non-polarity counterpart

of Hada is incidentally waaHad 'one.ms' or waHade 'one.fs.' Notice that there is no femininecounterpart for Hada.


45/56

45

sentence. When bothfiih andHada co-occur, there seems to be a strong preference

forfiih to precede. Consider:

(125) ma-fiih- Hada be-d-daar

neg-there-neg one in-the-house

"There is noone in the house."

While (125) is perfectly grammatical, (126) below is marginal:

(126)? ma-Hada- fiih be-d-daar

neg-one-neg there in-the-house


While (126) is marginal, if, however,fiih precedesHada and negation, then the

sentence is hopelessly ungrammatical:

(127)* fiih ma-Hadaa- be-d-daar

there neg-one-neg in-the-house


The same absolute constraint withfiih and kaan if negation is present:fiih may neverprecede bothHada and negation if negation hasfiih in its scope:

(128)* fiih ma-kaan- walad be-d-daar

there neg-was-neg boy in-the-house

"There is no boy in the house."

Compare (128) with (129) below wherefiih is outside the cope of negation:

(129) fiih walad ma-kaan- be-d-daar

there boy neg-was-neg in-the-house


46/56

46

"There is a boy who was not in the house."

Hada behaves in a way similar tofiih when it, kaan, and negation co-occur:kaan may not precede bothHada and negation:

(130)* kaan ma-Hadaa- be-d-daar

was neg-one-neg in-the-house

"There was noone is in the house."

The only grammatical versions of (130) are for eitherHada and negation to precede

kaan or forkaan and negation to precedeHada.

Ifkaan,fiih,Hada and negation co-occur, we have the following:

(131) ma-Hadaa- kaan fiih be-d-daar

neg-one-neg was there in-the-house

"There was noone in the house.

(132) ma-fiih- Hadaa kaan be-d-daar

neg-there-neg one was in-the-house"There was noone who was in the house."

(133) ma-kaan- fiih Hadaa be-d-daar

neg-was-neg there one in-the-house


(134) be-d-daar ma-kaan- fiih Hadaa

in-the-house neg-was-neg there one



47/56

47

These are the only grammatical word orders permitted.

Thusfiih andHada share the following properties:

(i) They both are able to merge with negation;(ii) they are both capable of preceding kaan; and, as we shall see below,

(iii) they both interact with verbal agreement.

Given, (i)-(iii) above, I find it plausible to conclude that fiih anHada belong to

the same syntactic category: they are both nominal.

8.0. fiih and Verbal Agreement7

In PA when a subject precedes its verb, the verb must necessarily agree with

the subject in all its features. When the verb follows the subject, it seems that the verb

agrees fully with the closest member of the subject NP. To clarify consider:

(135) ajat mona w-el-banaat

arrived.3sf Mona and-the-girls

"Mona and the girls arrived."

(136) ajen l-banaat w-mona

arrived.3pf the-girls and-Mona

"The girls and Mona arrived."

7For detailed discussions of agreement, both in terms of fact and theory, the reader is

referred to Aoun, Benmamoun, and Sportiche (1994), Bahloul and Harbert (1992),

Benmamoun (1992), Fassi Fehri (1993), Mohammad (1990, 1998), and van Gelderen

(1996).8

For detailed discussions of agreement, both in terms of fact and theory, the reader isreferred to Aoun, Benmamoun, and Sportiche (1994), Bahloul and Harbert (1992),

Benmamoun (1992), Fassi Fehri (1993), Mohammad (1990, 1998), and van Gelderen(1996).


48/56

48

(137)* aja mona w-el-banaat

arrived.3sm Mona and-the-girls"Mona and the girls arrived."

(138) l-banaat w-mona ajen

the-girls and-Mona arrived.3pf


(139)* l-banaat w-mona ajat

the-girls and-Mona arrived.3sf


(135) and (136) show that when the verb precedes it agrees fully with the left-most

conjunct of a conjoined subject. (137) shows that when the verb precedes agreement

cannot be suspended with the default third person singular masculine used. (139)

shows that when the subject precedes the verb, the verb must agree fully with features

associated with the matrix subject NP.

With this brief background in mind let us examine how agreement operates

whenfiih is present in a sentence. Consider:

(140) fiih kaan xams zlaam be-d-daar

there was.3sm five men in-the-room

"There were five men in the room."

(141) fiih kaan xams neswaan be-d-daar

there was.3sm five women in-the-room

"There were five women in the room."

(142) fiih kaan xams kutub be-d-daar


49/56

49

there was.3sm five books in-the-room

"There were five books in the room."

(143) fiih kaan xams bagaraat be-d-daar

there was.3sm five cows in-the-room


As can be seen from (140) through (143) the verb kaan need not agree at all with any

of the features associated with the following NP. The option of agreement, however,

is available:

(144) fiih kaan-u xams zlaam be-d-daar

there was-3pm five men in-the-room


(145) fiih kaan-en xams neswaan be-d-daar

there was.3pf five women in-the-room


(146) fiih kaan-u xams kutub be-d-daar

there was-3pm five books in-the-room


(147) fiih kaan-en xams bagaraat be-d-daar

there was.3pf five cows in-the-room



50/56

50

The same options are available ifkaan precedes bothfiih and the lower

subject:

(148) kaan fiih xams zlaam be-d-daar

was.3sm there five men in-the-room


(149) kaan fiih xams neswaan be-d-daar

there was.3sm five women in-the-room


(150) kaan fiih xams kutub be-d-daar

there was.3sm five books in-the-room


(151) kaan fiih xams bagaraat be-d-daar

there was.3sm five cows in-the-room


(152) kaan-u fiih xams zlaam be-d-daar

was-3pm there five men in-the-room"There were five men in the room."

(153) kaan-en fiih xams neswaan be-d-daar

was.3pf there five women in-the-room


(154) kaan-u fiih xams kutub be-d-daar

was-3pm there five books in-the-room



51/56

51

(155) kaan-en fiih xams bagaraat be-d-daar

was.3pf there five cows in-the-room


When the lower subject precedes kaan, there is only one option: the verb must

necessarily agree with the subject:

(156)(a) fiih xams zlaam kaan-u be-d-daar

there five men was-3pm in-the-room


(b)* fiih xams zlaam kaan be-d-daar

there five men was-3sm in-the-room


(157)(a) fiih xams neswaan kaan-en be-d-daar

there five women was.3pf in-the-room


(b)* fiih xams neswaan kaan be-d-daar

there five women was.3sm in-the-room"There were five women in the room."

The claim that I would like to put forward is that when fiih is present, there are two

subjects:fiih and the other NP in the sentence. As far as subject-verb agreement is

concerned, I assume that the verb has the option of either agreeing withfiih which is

third person singular masculine, or with the post verbal subject.

9.0. The Empty fiih


52/56

52

In this section I look at agreement facts that pattern with those described in the

immediately preceding section. The difference is in this section we look at sentence

wherefiih is not lexically present. Consider:

(158) kaan 9end Hmad sayyara

was.3sm with Ahmedcar

"Ahmed had a car."

(159) kaanat 9end Hmad sayyara

was.3sf with Ahmedcar

"Ahmed had a car."

(160) been-na w-been-ku kaan 9adaaweh

between-us and-between-you.pm was.3sm enmity.sf

"Between you and us there is enmity."

(161) been-na w-been-ku kaan-at9adaaweh

between-us and-between-you.pm was-3sf enmity.sf

"Between you and us there is enmity."

(162) kaan 9en-na xams zlaamwas.3sm with-usfive men

"We had five men (visiting us)."

(163) kaan-u 9en-na xams zlaam

was-3pm with-usfive men

"We had five men (visiting us)."

(164) kaan 9en-na xams neswaan

was.3sm with-usfive women

"We had five women (visiting us)."


53/56

53

(165) kaan-en 9en-na xams neswaan

was-3pf with-usfive women"We had five women (visiting us)."

The same set of facts obtains if we use other prepositions. As I mentioned

earlier, the facts here are identical to those withfiih. The one difference being

here is that the indefinite subject can never precede kaan. How do we explain

these agreement facts? I submit that there is an emptyfiih present in these

examples that is responsible for two things (i) the subject-verb agreement facts,

and (ii) the appearance of an indefinite subject.

The account above leaves one important question unanswered, namely,

why canfiih allow the indefinite subject to precede the verb or the PP, but its

empty counterpart cannot? I.e., why is the following contrast?

(166) fiih mara kaanat beteHki 9arabi

there womanwas.3sf speak.3sf Arabic

"There was a woman who spoke Arabic."

(167)* mara kaanat beteHki 9arabiwomanwas.3sf speak.3sf Arabic

"A woman spoke Arabic."

Recall that we have already claimed that when the indefinite subject

precedes its predicate withfiih, this subject is the head of a relative clause. If

this claim is correct, then (166) is actually an equative sentence. The latter

claim provides us with the answer to our question regarding the contrast

above.

There is a general constraint in PA: empty subjects cannot appear in

equative sentences unlike the so-called verbal sentences. Consider:


54/56

54

(168) ajat

arrived.3sf'She arrived."

(169) aju

arrived.3pm

"They arrived."

(170) ajen

arrived.3pf

"They arrived."

Compare the above with the following:

(171)* ta9baan9

tired.ms

"I am, he is, you.ms are, tired."

(172)* ta9baaneh

tired.fs

9I am aware of the fact, irrelevant in my opinion, that (171)-(173) are perfectly grammatical

as responses to the question "How is x?" The difference between (171)-(173) and sentences

that contain a verb, is (171)-(173) are ungrammatical unless they are immediate responses to

this question. That is to say, the fact that once the identity of the person in question is

identified, a few sentences down the discourse, one must repeat the lexical pronoun. Thus, it

ungrammatical to come back in the discourse with:

(i)* sa9iiid

happy.3sm"he is happy."


55/56

55

"She is, you.fs are, tired."

(173)* ta9baanaattired.fp

"You.ms are, they.fp are tired."

Thus the ungrammaticality (167) is due to the general constraint in

Palestinian syntax that bars empty pronouns in equative sentences10.

10. Conclusion

In this paper I have given what I believe to be the most detailed

description of the properties of the expletivefiih in PA (and to the best of

knowledge, in any other dialect of Arabic) to date. I have argued that, in spite,

of its appearance, fiih is not a PP. I have shown thatfiih and the negative

polarity itemHada occupy a class of their own in that they are the only two

nominals in PA that are able to merge with negation. I have also shown that

this expletive plays a crucial role in licensing indefinite subjects in equative

sentences.

References

Aoun, Joseph, Elabbas Benmamoun, Dominique Sportiche. 1994. Agreement,Word Order, and Conjunction in Some Varieties of Arabic. LinguisticInquiry 25:195-220.

Bahloul, Maher, and Wayne Harbert. 1992. , Agreement Asymmetries inArabic,Proceedings of the Eleventh West Coast Conference on

Formal Linguistics.

10Mohammad (1990) claims that the feature person is crucial in licensing an empty pronoun.This feature is always present in verbs but is always missing in adjectives and nouns.


56/56

Benmamoun, Elabbas. 1992. Functional Categories, Problems of Projection,Representation and Derivation. Ph. D. dissertation, University ofSouthern California .

Fassi Fehri, Abelkader. 1993. Issues in the Structure of Arabic Clauses andWords. Kluwer, Dordrecht.

Mohammad, Mohammad A. 1998. Word Order, Agreement, andPronominalization in Standard and Palestinian Arabic. ms., Universityof Florida.

Mohammad, Mohammad A. 1990. The Problem of Subject Verb Agreement:Towards a Solution. In Mushira Eid (ed.)Perspectives on ArabicLinguistics I, John Benjamins, Amsterdam & Philadelphia.

Moutaouakil, Ahmed. 1989.Pragmatic Functions in a Functional Grammarof Arabic. Foris, Dordrecht.van Gelderen, Elly. 1996. Parametrizing Agreement Features in Arabic, Bantu

Languages, and Varieties of English. Linguistics 34.

7188755 the Expletive Fiih in Palestinian Arabic

Documents