Free Genitive and Construct State in Eḥeiθymme

Free Genitive and Construct State in Eḥeiθymme

Auteur: Daniel Quigley

Date MS: 05-31-2020

Date FL: 09-01-2020

Numéro FL: FL-00006C-00

Citation: Quigley, Daniel. 2020. “Free Genitive and

Construct State in Eḥeiθymme.
FL-00006C-00, Fiat Lingua, . la toile. 01 September 2020.

droits d'auteur: © 2020 Daniel Quigley. Ce travail est sous licence

sous une attribution Creative Commons-
Non Commercial-Pas de Dérivés 3.0 Licence non transférée.

http://creativecommons.org/licenses/by-nc-nd/3.0/

Fiat Lingua est produit et maintenu par la Language Creation Society (LCS). Pour plus d'informations
à propos du LCS, visitez http://www.conlang.org/

Free Genitive and Construct State in Eh. eiθymme

Daniel Quigley

Abstrait

This paper is an overview of the Free Genitive and Construct State
in the constructed language Eh. eiθymme im Ajjad Eh. Deirymme Amran. Ce
is done with frequent reference to the Semitic languages, which exhibit
these methods for genitival relationships. This presentation follows: un
overview of the construction; adjacency, prosody, and definiteness of the
Construct State; a syntactic treatment of the Free Genitive and the Construct
State relative to the Semitic; semantic relationships of the elements of the
Construct State; a brief historical outline of the Construct State.

Keywords— constructed language, syntax, free genitive, construct state

Contenu

1 Free Genitive and Construct State

2 Salient Features of the Construct State

2.1 Adjacency .
.
2.2 Prosody .
2.3

.
.
(Dans)definiteness .

.
.

.
.

.

.
.
.

.
.
.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3 Structures of the Free Genitive and Construct State

3.1 Semitic FG and CS .
3.2 Eh. eiθymme FG and CS . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

4 Semantic Relationships of the Construct State
.
.
.
.
.
.
.
.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1 Substantive .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
4.2 Adjective .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
4.3 Partitive .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
4.4 Numerical
4.5 Prepositional Phrase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.

.
.
.
.

.
.
.
.

5 Historical Development of the Construct State

Les références

1

2

2
3
4
6

8
8
13

16
18
20
21
22
22

24

27

1 Free Genitive and Construct State

Eh. eiθymme handles genitive phrases by either of two methods: the Free Genitive
phrase (FG) and the Construct State phrase (CS). The FG is a Determiner Phrase (DP)
in which the genitive is introduced by the preposition im- “de” as a casemarker, et
le (dans)definite head noun precedes the genitive. The CS is a DP in which the head
noun assumes the construct state and is never marked for definiteness, and the most
embedded genitive element determines the definiteness of the whole DP. (1) shows the
construction of the FG and CS in Eh. eiθymme.

(1) Eh. eiθymme genitive constructions:

À. Free Genitive – relates the elements of the phrase explicitly with the prepo-
sition as casemarker im-; adjectives are adjuncts of their respective NPs.

J’ai. Form:

[Adj](dans)def + [N](dans)def + [im-] + [Adj]gen + [N]gen

b. Construct State – relates the elements of the phrase by passing one element
in the construct state and the other element in the non-obligatorily marked
génitif; adjectives are adjuncts of their respective NPs.

J’ai. Form:

Ii. Form:

[Adj]constr
[Adj]constr

(dans)def + [N]constr + [Adj]gen + [N]gen
(dans)def + [N]constr + [Adj](dans)def + [N](dans)def

This presentation is arranged as follows: section 2 describes the salient features
of the CS, its composition, and how to control its status as definite or indefinite; section
3 compares and contrasts the structures of the CS and FG as they are used in the Semitic
languages and Eh. eiθymme; section 4 describes the use of the CS in Eh. eiθymme, noting the
various semantic interpretations depending on what the elements of the CS are; section
5 gives a brief outline of the historical derivation of the CS in Eh. eiθymme, tracking both
its construction and phonological history.

2 Salient Features of the Construct State

The FG and CS strategies for establishing a genitive relationship between elements
are hallmarks of the Semitic languages1. Of particular interest to general syntacticians
studying the Semitic languages is the CS. The CS contains an overtly casemarked geni-
tive phrase following the head noun, which may be of any case required by the context.
The properties and behavior of the CS have been the subject of much investigation, et
for a detailed discussion, see Borer (1996). It is comparable to a genitive or attributive
relation, where the first noun (or adjective) is the head constituent and the second
noun is the attribute (Abu Shaqra 2007).

1. The CS is also used in Persian, where it is known as ezafe; voir, entre autres, Parsafar (2010).
The CS is found as well in Berber (El Hankari 2014). See Creissels (2009) for an investigation
into CS-like strategies in various African languages.

2

The salient properties of the Arabic CS nominal concerning morphosyntactic
features may be found discussed in Borer (1996, 1999), Benmamoun (2000), Almansour
(2012), and Versteegh (2014). The properties as they are relevant to the CS in Eh. eiθymme
are listed in (2):

(2)

1. The elements of the CS tend to be adjacent one to the other.
2. The CS constitutes a single prosodic unit.
3. The last element of the CS carries (dans)definite marking.

Morphologically, a noun in the construct state is marked by -n for words ending in
a vowel, and by -’n for words ending in a consonant. The apostrophe is an orthographic
convention, and itself carries no phonological value.

(3) Typical genitive relationships in Eh. eiθymme:

À. k¯en’n

feDi
rock.constr
girl
‘A rock of a girl’

b. k¯en’n

le-feDi
the.girl
rock.constr
‘The rock of the girl’

c. k¯en’n

i-feDi
gen.girl

rock.constr
‘A/the rock of a/the girl’
i-feDi
la-k¯an
je suis
gen.girl
de
the.rock
‘The rock of a/the girl’

d.

The CS in (3a), (3b), et (3c) is a phonological word, and the first element k¯an’n
is affected by the vowel harmony rules triggered by the indefinite noun feDi, the definite
noun le-feDi, and the genitive-marked i-feDi. Dans (3d), the trigger for vowel harmony is
blocked at the word boundary. En plus, while the first element may be inflected
for any of the nominative, accusatif, or genitive cases in Arabic, the second element is
inflected only for the genitive case (Borer 1996; Almansour 2012); the genitive case is
never obligatorily marked in the second element in Eh. eiTymme, un péché (3a) et (3b); À
genitive relationship is implied by the initial element being in the construct state.

2.1 Adjacency

Comme indiqué ci-dessus, a feature of the CS is that its elements tend to be adjacent one to

l'autre.

(4)

À. *n¯anumman
elk.constr
‘The child’s elk is sleeping’

ajjad
is.imp.3

i-Deyr
gen.child

adwan
sleep.imp.3

b. n¯enymmen
elk.constr
‘The child’s elk is sleeping’

i-Deyr
gen.child

ajjad
is.imp.3

adwan
sleep.imp.3

3

The ungrammaticality of (4a) is due to the copula interrupting the CS. Parce que
the CS forms an entire nominal unit, the insertion of the copula breaks that unit, et est
agrammatical.

As opposed to the modification rules found in the Semitic CS, the modification
of the CS by adjectives and modifiers in Eh. eiθymme remain intact with the noun they
are modifying. The ungrammaticality of (5a) is due to the element’s modifier being
dislocated from its position prior to the noun it is modifying.

(5)

À. *deir’n

i-h. ¯ıT
gen.ghost

deylyn
marsh.adj.constr

land.constr
‘The ghost’s marshy land’

b. deylyn

marsh.adj.constr
‘The ghost’s marshy land’

deir’n
land.constr

i-h. ¯ıT
gen.ghost

Les exemples dans (5) also illustrate a detail of the CS in Eh. eiθymme that is absent in
the Semitic languages: adjectives modifying the first element of the CS are also inflected
for the construct state.

The CS is strictly right-branching and head first. This leads to a linear nesting
of the appropriate elements, yielding a CS nominal of several elements marked in the
construct state.

(6)

À. k¯en’n

[menyn
river.constr

[eiD’n
father.constr

le-feDi]]
the.girl

b.

rock.constr
‘The rock of the river of the father of the girl’
[erjeD’n
king.constr
‘The friend of the writer’s king of the river, The writer’s friend’s king of the river’

[dein’n
friend.constr

le-enjel]]
the.writer

le-meny
the.river

2.2 Prosody

The elements of the CS form a nominal unit, and are phonologically considered a
mot. Borer showed that, via data from Modern Hebrew, the vowel of the first element
is reduced, and the main stress falls on the rightmost element of the CS (Borer 1988,
1996). In those same papers, he showed that word-level phonological processes take
place within the CS nominal, un péché (7).

(7) Phonological influence of the CS

Modern Hebrew (Borer 1988, 1996)

À. beit

mora
teacher

maison
‘a teacher’s house’

4

Sel
mora
b. bayit
teacher
de
maison
‘a teacher’s house’

In Eh. eiθymme, a similar phonological influence occurs. The elements of the CS
are a phonological word, and are subject to the vowel harmony morphophonology,
which in turn may trigger consonantal change. There exists three general rules of
sound change in Eh. eiθymme, shown in (8) with corresponding examples, one of which
involves a harmonization of the vowels in a word, one of which involves the mutation
of consonants, and one that turns semi-vowels into vowels.

(8)

1.

If a close, devant, unrounded vowel exists anywhere in the word, alors
the close, back vowel becomes the close, devant, rounded vowel, et le
ouvrir, back vowel becomes the mid, devant, unrounded vowel.

(8)

(cid:20)

tu
À (UN)

(cid:21)

(cid:21)

(cid:20) et
E

/ […J’ai…]

À. *la-lani

b.

the.mouse

‘the mouse’
le-leni
the.mouse
‘the mouse’

c. *i-nalu

d.

gen.moss
‘of moss’
i-nely
gen.moss
‘of moss’

2. The consonant moves to a fricative counterpart in the environment that

it follows a front vowel or a voiceless palatal fricative.

(9)

























(cid:26) (cid:2) Vfront
(cid:2) j (cid:3)

(cid:3)

/









p
b
t
d
k
g
h

f
v
J
D
X
g
ç

À. *ihlit
cadeau

b.

‘given; gifted; gift’
ih. liT
cadeau
‘given; gifted; gift’

c. *i-g¯al

d.

gen.joy
‘of joy’
i-G¯el
gen.joy
‘of joy’

3. A semi-vowel becomes its vowel counterpart in the environment that it

preceeds immediately a non-semi-vowel consonant.

5

(dix)

(cid:21)

(cid:20) j
w

(cid:20) J’ai
tu

(cid:21)

/

(cid:2) C (cid:3)

À. *la-majs

b.

the.leader

'le meneur'
le-meis
the.leader
'le meneur'

c. *la-tawr
the.boy
‘the boy’
la-taur
the.boy
‘the boy’

d.

The ungrammatical items in (8), and the ungrammaticality of (11a) et (11c), sont
due to the elements of those respective CS nominals ignoring the morphophonological
triggers from another element. If there is a trigger for sound change in any appropriate
lieu, then that sound must change.

(11)

À. *k¯an’n

b. k¯en’n

le-feDi
the.girl
rock.constr
‘The rock of the girl’
le-feDi
the.girl
rock.constr
‘The rock of the girl’
dein’n
friend.constr

c. *erjeD’n

king.constr
‘The friend of the writer’s king of the river, The writer’s friend’s king of the river’

la-anjal
the.writer

la-manu
the.river

d. erjeD’n

dein’n
friend.constr

le-enjel
the.writer

le-meny
the.river

king.constr
‘The friend of the writer’s king of the river, The writer’s friend’s king of the river’

2.3

(Dans)definiteness

The CS is a DP itself, and its definiteness (indicated by an article, the presence of
a genitive, or the element being a proper noun) or indefiniteness (typically indicated
by the lack of an article, or the presence of a genitive) are determined by the second
élément. This determination of the (dans)definiteness by the genitive element of the CS
is referred to asDefinite Spreading” (Borer 1999). Because only the second element
can carry the (dans)definiteness, comme on le voit dans (12a), (12b), et (12e), the CS nominal has its
(dans)definiteness determined, et (12c) et (12d) are ungrammatical.

(12)

À.

b.

¯ulh¯ak’n
la-abrad
spear.constr
the.warrior
‘The spear of the warrior’
¯ulh¯ak’n
abrad
warrior
spear.constr
‘A spear of a warrior’

6

la-abrad
the.warrior

c. *la- ¯ulh¯ak’n

d. *la- ¯ulh¯ak’n

the.spear.constr
‘The spear of the warrior’
abrad
warrior

the.spear.constr
‘The spear of a warrior’
D¯an
¯ulh¯ak’n
D¯an
spear.constr
‘The spear of D¯an’

e.

To make examples (12c) et (12d) grammatical, then a periphrasis by a preposi-
tion, c'est à dire., the FG, est requis, un péché (13). Because it is not explicit, a greater context is
required to determine the (dans)definiteness of the noun in the genitive; the form of the
genitive is identical for the indefinite and the definite state.

(13)

i-evreD
la- ¯ulh¯ak
the.spear
gen.warrior
‘The spear of a/the warrior’

je suis
de

(Dans)definiteness is respected by adjectives modifying the CS nominal. If the first
element is (dans)definite, then the adjective modifying it is likewise (dans)definite as deter-
mined by the second element, and it will show marking for being in the CS; if the second
element is (dans)definite, then the adjective modifying it is likewise (dans)definite.

An adjective modifying the first element appears in the NP of that element, avant

le nom, and agrees with the (dans)definiteness determined by the second element.

(14) Modifying the first element:

À. k¯en’n

feDi
girl
rock.constr
‘A rock of a girl’
¯elsyn
water.adj.constr
‘A watery rock of a girl’

b.

k¯en’n
rock.constr

feDi
girl

c. k¯en’n

le-feDi
the.girl

rock.constr
‘the rock of the girl’
le-¯elsyn
the.water.adj.constr
‘The watery rock of the girl’

d.

k¯en’n
rock.constr

le-feDi
the.girl

An adjective modifying the last element appears before the CS nominal, et

agrees with the (dans)definiteness determined by the second element.

(15) Modifying the second element:

7

À. k¯en’n

b. k¯en’n

feDi
girl

feDi
girl
rock.constr
‘A rock of a girl’
elj¯e
sad

rock.constr
‘A rock of a sad girl’
le-feDi
the.girl

rock.constr
‘the rock of the girl’
le-elj¯e
the.sad

rock.constr
‘The rock of the sad girl’

c. k¯en’n

d. k¯en’n

le-feDi
the.girl

For an adjective modifying the second element, if that second element respects

the presence of an overtly marked genitive, then the adjective will as well.

(16)

À. k¯en’n

i-feDi
gen.girl

rock.constr
‘The rock of the girl’
i-elj¯e
gen.sad

rock.constr
‘The rock of the sad girl’

b. k¯en’n

i-feDi
gen.girl

3 Structures of the Free Genitive and Construct State

The theory of syntax adopted here for presenting the structure of the FG and CS
is x-bar theory, in which inflection is a consequence of movement in the syntax tree of
its various constituents. For a reference to this framework, see Carnie (2012).

An alternative syntactic investigation might be in the framework of head-driven
phrase structure grammar, or HPSG; for such an analysis of the CS within such a
framework, see AlQurashi (2015). A presentation of the FG and CS of Eh. eiθymme by the
author within the framework of HPSG is forthcoming.

To have a full picture of the structures of the FG and the CS, this presentation
follows two parts. D'abord, an examination is given of the FG and CS in Modern Hebrew
and Syrain Arabic, as investigated and presented by Cowell (1964), Ritter (1991), Ben-
mamoun (2000), and Hagstrom (2001). This is followed up by a presentation of those
structures as it applies to Eh. eiθymme.

3.1 Semitic FG and CS

Benmamoun (2000) agrees with the structures of the CS as described in the analysis
by Ritter (1991), who also supplied the structure of the FG: the complement of D is,

8

instead of NP, a second non-lexical category, Num, whose head is responsible for the
number specification (Singulier et pluriel) of the noun phrase. The assigning of case is
accounted for by either the specifier of DP or by an inserted casemarker. (17) shows the
tree structures prior to movement for the FG in (17a) and CS in (17b).

(17) D-structure for FG and CS

(Ritter 1991; Benmamoun 2000)

À.

DP

D

NumP

Num

NP

AdjP

NP

DP

N(cid:48)

DPN

b.

DP

D

NumP

DP

Num(cid:48)

Num

NP

AdjP

NP

DP

N(cid:48)

DPN

The process of the Modern Hebrew FG is given in (18a) et (18b). N cannot
move to D, as it is occupied by the definite article ha-, so it must move to something
lower, call it Num. There are three assumptions in this framework: first, assume that
adjectives are NP-adjuncts (similar to adverbs as VP-adjuncts); deuxième, assume that
the genitive starts off in the SpecNP (subject of NP); third, assume that the presence
of the overt case marker Sel “de” is inserted where it is needed and not a preposition,
since there is no Dgen as in the CS to give the genitive relation, and it does not affect the
c-command relationship between the subject and object.

The derivation of the CS in Modern Hebrew is given in (18c) et (18d). Le
definite article ha- is forbidden, so there is something else occupying the head position
of DP, call it Dgen. This head is responsible for assigning the genitive to the subject
rightward to the adjacent DP. Since the subject appears before the adjective, it must
have moved there. De la même manière, the head noun appears before the subject, so it must have
moved to D.

(18) FG and CS Structure and Derivation

Modern Hebrew (Ritter 1991)

ha-menumeset
the.polite

À. ha-axila

the.eating
ha-uga
et
the.cake
acc

Sel
de

Dan
Dan

b.

DP

D

NumP

Ha-

Num

NP

‘Dan’s polite eating of the cake’

axilai

AdjP

NP

menumeset

DP

N(cid:48)

Sel Dan

N

DP

de

et ha-uga

9

Dan
Dan

ha-menumeset
the.polite

c. axilat

eating.constr
ha-uga
et
the.cake
acc

d.

DP

Dgen

NumP

axilaj

DP

Num(cid:48)

Dani

Num

NP

‘Dan’s polite eating of the cake’

tj

AdjP

NP

ha-menumeset

DP

N(cid:48)

de

N

DP

tj

et ha-uga

Summarily, if Dgen is used, then N moves through Num to D, and the subject DP
moves to the spec position of NumP where it receives case from the Dgen. inversement, si
a dissimilar D is used, then N moves only as far as Num, and the subject DP get case
from the inserted case marker Sel. To preserve the surface form, the AdjP remains in situ.
Dgen, cependant, is not inherently definite. To account for this, Ritter (1991) fait valoir que
the definite feature is acquired via Spec-head agreement with the DP at the derivational
niveau.

It should be pointed out that pronominal subjects of simple CSs are realized as
clitics on the head noun. The pronominal subjects, like any such full noun phrase
sujets, appear in the spec position of the NumP at the s-structure. Ritter suggests,
alors, that cliticization occurs here.

(19) Pronominal CS Structure and Derivation

Modern Hebrew (Ritter 1991)

À. axilat-o

eating.constr.his
ha-uga
et
the.cake
acc

ha-menumeset
the.polite

b.

DP

Dgen

NumP

axilaj

DP

Num(cid:48)

-oi

Num

NP

‘His polite eating of the cake’

tj

AdjP

NP

ha-menumeset

DP

N(cid:48)

de

N

DP

tj

et ha-uga

The FG does not necessarily preclude the CS, et vice versa; there exists a third
genitive method in the language which simultaneously employs both FG and CS. Ritter

dix

démontre “clitic doubled construct states” (doubled CS) arise when both methods
are employed in the same noun phrase: it has the subject of a FG and the head of a CS
(Ritter 1991). The doubled CSs never contain an initial determiner, as in the regular CS,
but they have the inserted casemarker before the subject as in the FG. En outre, là
exists a pronominal clitic suffixed to the head noun, which agrees with the subject.

The derivation of the doubled CS in Modern Hebrew is the following: the full
noun phrase subject receives genitive case in its d-structure position from the inserted
case marker, as in the FG, and N + Num moves to Dgen as in the CS. Assume that the
Dgen has case which it must assign, and the pronominal element, which is realized as a
clitic on the head of the DP, serves this purpose.

The pronominal clitic in the doubled CS is assumed to be generated as a pronom-
inal element in the spec position of NumP. Ici, as opposed to (19), the clitic is a
pleonastic pronoun which acquires its definiteness from the element in the spec position
of NP, with which it is co-indexed. The full noun phrase subject moves to the spec
position of NumP. The derivation of the doubled CS in Modern Hebrew is given in
(20).

(20) Doubled CS Structure and Derivation

Modern Hebrew (Ritter 1991)

À. axilat-o

eating.constr.his
Sel
et
acc
de

Dan
Dan

ha-uga
the.cake

ha-menumeset
the.polite

‘Dan’s polite eating of the cake’

b.

DP

Dgen

NumP

axilaj

DPi

Num(cid:48)

-o

Num

NP

tj

AdjP

NP

ha-menumeset

DPi

N(cid:48)

Sel Dan

N

DP

tj

et ha-uga

As opposed to being formed simultaneously, as in the doubled CS, the FG and
CS may freely combine, one after the other. This is seen in (21), where the dashed line
indicates the completed tree structure derivation from (18b).

(21) Combined FG and CS Structure and Derivation Modern Hebrew (Hagstrom

2001)

À.

tmunat
picture.constr

ha-yalda
the.girl

Sel
de

ha-mora
the.teacher

‘the teacher’s picture of the girl’

11

b.

DP

DP

DP

Dgen

NumP

Sel ha-mora

tmunatj

DP

Num(cid:48)

ha-yeldai

Num

NP

tj

DP

N(cid:48)

de

N

tj

Cowell (1964) describes coordination of leading elements of the CS in Syrian
Arabic. Assuming the structure of a Conjunction Phrase CoP (Johannessen 1998), tel
a CS has the following structure and derivation:

(22) CS with Leading Element Coordination

Syrian Arabic (Cowell 1964)

À.

Saw¯are
streets

u-h¯ar¯at
and.quarters

al-mad¯ıne
the.city

b.

DP

Dgen + CoP[N]

NumP

N

Co(cid:48)

DP

Num(cid:48)

‘the streets and quarters of the city’

Saw¯are

Co

N

al-mad¯ınei

Num

NP

tu-

h¯ar¯at

tCoP[N]

DP

N(cid:48)

de

N

DP

tCoP[N]

Often, cependant, such coordinations are avoided. Plutôt, the use of an anaphoric
pronoun is preferred, where the CS DP is in the spec position of the CoP[DP], et le
complement of C is a simple CS co-indexed with the subject:

(23) CS with Anaphoric Pronoun

Syrian Arabic (Cowell 1964)

À.

Saw¯are
streets

al-mad¯ıne
the.city

u-h¯ar¯atha
and.quarters.its

b.

CoP[DP]

DP

CoP(cid:48)

‘the streets of the city and its quarters’

Saw¯arej

DPk

Num(cid:48)

tu-

h¯ar¯atha

Dgen

NumP

Co

DPk

al-mad¯ınei

Num

NP

tj

DP

N(cid:48)

de

DPN

tj

12

3.2 Eh. eiθymme FG and CS

The FG and CS in Eh. eiθymme are derived slightly differently than in the Semitic
counterparts. Because nouns in Eh. eiTymme are not marked for number, but rely on
context and other modifiers to indicate such granularity, it is not wholly accurate to
describe the intermediate non-lexical category as a Number Phrase; number distinction
in Eh. eiθymme is typically non-explicit. Ainsi, in the context of Eh. eiθymme, the NumP
is relabeled as nP.

En plus, the location of the AdjP in Eh. eiθymme is relocated to a different
adjunct position, and the movement is phrasal instead of head. This respects the order
of constituents in the surface forms of the FG and CS, and is a result of the observation
that modifiers tend to remain adjacent to the nominal they are modifying, and carry the
(dans)definite information and state information as well. (24) shows the tree structures
prior to movement for the FG in (24a) and CS in (24b).

(24) D-structure for FG and CS

Eh. eiTymme

À.

DP

D

nP

est

NP

DP

NP

NP

DP

N(cid:48)

AdjP

N(cid:48)

N

b.

DP

Dgen

nP

DP

est(cid:48)

est

NP

DP

NP

NP

DP

N(cid:48)

AdjP

N(cid:48)

N

The derivations proceed analagously. If Dgen is used, then the NP moves through
n to D and receives marking for the construct state, and the subject DP moves to the
spec position of nP where it receives case from the Dgen. If a different D is used, then the
NP moves only as far as n, and the subject DP is assigned case from the inserted case
marker im-. As in the Semitic description, Dgen is not inherently definite; the definite
feature is acquired via Spec-head agreement with the DP at the derivational level (Ritter
1991).

Donc, the structures and derivations in Eh. eiθymme are given in (25), où

the FG is given by (25a) et (25b), and the CS is given by (25c) et (25d).

(25) FG and CS Structure and Derivation

Eh. eiTymme

13

À.

je suis
de

d¯an
Dan

la-SaSren
the.eating

le-iD
the.polite
le-mes¯ır
the.cake
‘Dan’s polite eating of the cake’

b.

DP

D

nP

la-

n + NPt

NP

N(cid:48)

DP

NP

c.

d.

SeSren’n
the.eating.constr

le-iD’n
the.polite.constr
le-mes¯ır
d¯en
Dan
the.cake
‘Dan’s polite eating of the cake’

DP

Dgen + NPt

nP

N(cid:48)

DP

est(cid:48)

AdjP

iD

N(cid:48)

N

D¯ani

n + tNP

NP

DP

NP

im D¯an

tNP

DP

AdjP

iD

N(cid:48)

N

SaSren

le-mes¯ır

SaSren

de

tNP

DP

le-mes¯ır

For simple CSs, pronominal subjects may be cliticized to the head noun as in the

Modern Hebrew. Full pronouns may be elements of the CS as well.

(26) FG and CS Structure and Derivation

Eh. eiTymme

À. m¯en

ih. liT
gen.Ih. liT

house.constr
‘the house of Ih. liθ, Ih. liθ’s house’

b.

DP

Dgen + NPt

nP

m¯a

DP

est(cid:48)

c. m¯an-¯a

house.constr.3
‘her house’

d.

DP

Dgen + NPt

nP

m¯a

DP

est(cid:48)

Ih. liTi

n + tNP

NP

-¯ai

n + tNP

NP

DP

NP

de

tNP

DP

DP

NP

de

tNP

DP

The mechanism of the doubled CS in Eh. eiθymme is similar to the Modern Hebrew.
The inserted case marker im- gives the genitive case to the subject, and the n + DP
moves to Dgen. The pleonastic pronominal element is generated in the spec position of
nP, which receives its case from Dgen, which realizes as a clitic on the head of DP. Le
clitic may level to the entire phrase, including any modifiers, for the effect of emphasis
or idiosyncrasy.

(27) Doubled CS Structure and Derivation

Eh. eiTymme

14

À.

SeSren’n-¯a
eating.constr.3

iD’n
polite.constr
le-mes¯ır
d¯an
the.cake
Dan

je suis
de

b.

DP

Dgen + NPt

nP

N(cid:48)

DPi

est(cid:48)

‘Dan’s polite eating of the cake’

AdjP

iD

N(cid:48)

N

-¯a

n + tNP

NP

DPi

NP

SaSren

im D¯an

tNP

DP

le-mes¯ır

Eh. eiθymme may combine a FG and CS. The resulting DP is always a FG, whereas
a combination which would result in a DP that is a CS is deliberately avoided. C'est,
when a FG and a CS is combined, the nominal elements of the FG may be DPs that are
each a CS, but a CS whose either element is a FG is unnatural.

(28) Combined FG and CS Structure and Derivation

Eh. eiθymme

À.

¯adn
grammaire
i-f¯el¯ım
gen.art

je suis
de

eiT’n
language.constr

b.

DP

D

nP

n + NPt

NP

¯adn

DP

D

DP

je suis

Dgen + NPt

nP

NP

tNP

DP

eiT

DP

est(cid:48)

f¯el¯ımi

n + tNP

NP

DP

NP

de

tNP

DP

‘a grammar of a constructed language’

15

c.

¯ıveT’n
painting.constr
i-eyDe
gen.teacher

i-feDi
gen.girl

je suis
de

‘the teacher’s painting of the girl’

d.

D

DP

n + DP

nP

NP

Dgen + NPt

nP

DP

NP

¯ıveT

DP

est(cid:48)

im auda

tDP

DP

feDii

n + tNP

NP

DP

NP

de

tNP

DP

Coordinating the leading element in the CS with a conjunction is similar to the
situation in Syrian Arabic. En plus, there is no preference for the coordinated
leading element or for an anaphoric pronominal in Eh. eiθymme.

(29) CS with Coordination

Eh. eiθymme

eD
et

p¯esyn
sections

À. eTyven
streets
‘the streets and sections, divisions, quar-
ters of the city’

i-Delein
gen.city

b.

DP

Dgen + CoP[NP]

nP

NP

Co(cid:48)

DP

est(cid:48)

c. eTyven
streets
‘the streets of the city and its sections’

p¯asu-¯a
sections.3

i-Delein
gen.city

annonce
et

d.

CoP[DP]

DP

Dgen + NP

nP

Co(cid:48)

Co

DP

atuva

DP

est(cid:48)

annonce

p¯asu-¯a

atuva

Co

N

Deleini

n + tCoP[NP]

NP

annonce

p¯asu

DP

NP

de

tCoP[NP]

DP

Deleini

n + tNP

NP

DP

NP

de

tNP

DP

4 Semantic Relationships of the Construct State

Cowell (1964) and Watson (1993) describe two ways in which the following
element may qualify the leading element in the CS: Identificatory and Classificatory,
described in (30):

(30)

À. Identificatory

J’ai.

If definite, the following element of the CS generally answers the ques-
tion “qui?” ou “whose?” applied to the leading element. Par exemple,
la-taur and i-θeyr in (30a-ii) et (30a-iii), respectivement, show which or
whose cat is referred to.
la-taur
the.boy

cat.constr

Ii. murS’n

16

‘the boy’s cat’

i-Teyr
gen.boy

iii. myrS’n

cat.constr
‘the boy’s cat’
b. Classificatory

J’ai. Whether definite or indefinite, the following element of the CS generally
answers the questionwhat kind of…?” applied to the leading element.
Dans (30b-ii) et (30b-iii), le-¯els and i-¯els indicate what kind of bird is
meant (that it is lexicalized asheronnotwithstanding).

Ii. ferin

le-¯els
bird.constr
the.water
‘the water bird, heron’

iii. ferin

i-¯els
gen.water

bird.constr
‘the water bird, heron’

Identification is fundamentally a function of defintieness, and classification is
fundamentally a function of indefintieness; but since the definiteness of the following
element is not always clear (especially if the element is in the explicitly marked genitive),
it is not possible to simply equate identificatory terms with definitizable ones. Beaucoup
CS phrases, when taken out of context, can be understood either as identificatory or
classificatory, an ambiguity of which may be seen in (31) et (32).

(31)

le-DemeiD
the.school

Identificatory
leyl’n
À.
child.constr
‘the child(ren) of the school’
leyl’n
child.constr
‘the child(ren) of the school’

i-DemeiD
gen.school

b.

(32) Classificatory

À.

b.

le-DemeiD
the.school

leyl’n
child.constr
‘the schoolchild(ren)'
leyl’n
child.constr
‘the schoolchild(ren)'

i-DemeiD
gen.school

Crucially, the grammatical difference that separates the identificatory CS and the
classificatory CS is that, in identificatory constructs, the genitive element, if it is definite,
can be pronominalized; whatever the following element refers to may subsequently
or alternatively be referred to by a pronoun, either affixed to the leading element or
sinon. This may be seen in (33). Classificatory CSs are not able to be pronominalized
in this manner.

17

(33) Pronominalization
i-Teyr
gen.boy

À. myrS’n

cat.constr
‘the boy’s cat’

¯ıme
son

b. myrS’n

cat.constr
‘his cat’
c. murS’n-¯a

cat.constr.3
‘his cat’

The semantic relationships between the elements of the CS are variable, and have
been studied and catalogued by various authors (Thackston 1994; Holes 2004; Ryding
2005; Abu Shaqra 2007). De même, there are various kinds of semantic relationships2
determined by the CS in Eh. eiθymme, which depend on the element in the construct
État: substantive, adjectif, partitive, and numerical (cardinal and ordinal). Quelques
prepositional phrases are rendered in the CS as well.

The CS and FG are more or less isomophormic one with the other. Some geni-
tive relationships that can be related by the CS and FG may be represented by some
periphrastic construction with a preposition; such will be noted, following the example
of Cowell (1964). The CS semantics in Eh. eiθymme are in many places dissimilar to their
equivalents in the Semitic languages.

4.1 Substantive

Ordinary noun constructs are used to express widely varied relationships of

meaning between leading and following elements.

(34)

Item from Collective (generally classificatory; periphrasis with ev- “de, part/piece
de”)

À.

b.

i-mesi
gen.bread

irseT’n
piece.constr
‘a piece of bread’
irseT
i-mesi
ev
gen.bread
de
morceau
‘a piece of bread’

c.

d.

i-eDDemywwe
gén.(world’s)personnes

¯ynm¯ej
map.constr
‘a map of the world’s people’
i-eDDemywwe
¯unm¯aj
carte
gén.(world’s)personnes
‘a map of the world’s people’

ev
de

(35) Kind Differentiation (classificatory; periphrasis with various prepositions)

2. As Cowell (1964) disclaims, “The categories of relationship given here are intended to
suggest the semantic scope of this construction, and are not meant to constitute a definitive
classification or kind of classification.

18

i-veir
gen.cattle

À. weryn

meat.constr
‘beef ’
b. waru
viande
‘beef ’

ev
de

i-veir
gen.cattle

c. beh. r’n

flower.constr
‘apple blossoms’

i-meDyim
gen.apple

d. bahr

i-meDyim
gen.apple

ev
flower
de
‘apple blossoms’

(36) Kind and Specific Name (classificatory; no periphrasis)

À. ferin

esse
Essa

deir’n
land.constr
bird.constr
‘a cormorant, bird of the land of Essa’
Seirymmen
raptor
‘a roc, raptor of the rocks’

i-x¯en
gen.rock

b.

c. warm’n

deir’n
land.constr

emmej
Emmej

berry.constr
‘a grape, berry of the land of Emmej’

d. seiT’n

deir’n
land.constr
tree.constr
‘a beech tree, tree of the land of N¯an’

n¯en
N¯an

(37) Kind and Individual Name (identificatory; no periphrasis)

À.

ten¯ır’n
lake.constr

imer
Imer

‘Imer’s Lake, Sea of Imer, The Sunken
Sea’

b. deir’n

land.constr

emren
Amran

‘The Land of Amran’

c.

d.

eh. Deneilymme
Eh. Deneilymme

the Wisdom of
le

the Wisdom of

¯end’n
wisdom.constr
‘The Morality,
Eh. Deneilymme,
(Spirit) of the City’
teyl’n
word.constr
‘the word “lac”'

i-Ten¯ır
gen.lake

There are some cases in which the individual name is originally composed of
an adjective modifying the noun. The adjective then became less strictly descriptive,
and more a name, and so assumed the CS formation, and losing the article in the
processus.

(38)

À.

b.

le-Ten¯ır
la-lalmenu
the.lake
the.harmful
‘the harmful lake, the wounding lake”
lelmenyn
harmful.constr
‘The Lalmenu Sea”

i-Ten¯ır
gen.lake

c.

d.

le-Deir
the.land

la-¯amu
the.airy
‘the airy lands, the open lands’
i-Deir
¯emyn
airy.constr
gen.land
‘The Airy Lands, the Wide Lands’

(39) Relation and Related Object (generally classificatory; periphrasis with u- “pour”)

À.

l¯ıw’n
name.constr

i-Teyr
boy

‘the boy’s name’

b.

le-l¯ıw
the.name

tu
de

i-Teyr
gen.boy

‘the boy’s name’

19

c. erjeD’n

i-DereiD
gen.kingdom

ruler.constr
‘the ruler of the kingdom’

d.

tu
la-arjad
the.ruler
de
‘the ruler of the kingdom’

i-DereiD
gen.kingdom

(40) Associated Object and its Association (generally identificatory; periphrasis with

tu- “pour”)

À. m¯en

i-eneD
scholar

house.constr
‘a scholar’s house’
i-eneD
tu
gen.scholar
pour
maison
‘a scholar’s house’

b. m¯a

c. Delein’n

i-eive
gen.sage

city.constr
‘the city of the sage’
le-Delein
i-eive
the.city
gen.sage
‘the city of the sage’

tu
pour

d.

4.2 Adjective

An adjective may form with nouns as CS nominals; such constructions are classi-
ficatory. En plus, unlike substantive constructions, leading term adjectives may be
marked for definiteness.

(41) Adjective (classificatory; no periphrasis)
À. With indefinite first element:

J’ai. eDrew’n

dull.constr

i-¯en
gen.eyes

‘unintelligent, an intelligent person’

Ii. DeSemlyn

bright.constr

i-¯en
gen.eyes

‘intelligent, an intelligent person’

iii.

inril’n
lacking.constr

i-s¯eT
gen.health

‘sickly, a sickly person’

b. With definite first element:

J’ai.

le-eDrew’n
the.dull.constr

i-¯en
gen.eyes

‘unintelligent, the intelligent person’

Ii.

le-DeSemlyn
the.bright.constr

i-¯en
gen.eyes

‘intelligent, the intelligent person’

iii.

le-inril’n
the.lacking.constr

i-s¯eT
gen.health

20

iv.

inril’n
lacking.constr
‘rude, a rude person’

i-xextem
gen.manners

v. erwen’n

i-¯em
gen.breath

light.constr
‘likeable, agréable, a likeable person’
i-¯em
gen.breath

heavy.constr
‘unlikeable, unpleasant, an unlikeable
personne'

vi. esdix’n

‘sickly, the sickly person’

iv.

le-inril’n
the.lacking.constr

i-xextem
gen.manners

‘rude, a rude person’

v.

le-erwen’n
the.light.constr

i-¯em
gen.breath

‘likeable, agréable, the likeable person’

vi.

le-esdix’n
the.heavy.constr

i-¯em
gen.breath

‘unlikeable, unpleasant, the unlike-
able person’

4.3 Partitive

In a CS where the first element is a partitive, then that partitive tends to be
subordinate to the element with which it is in construction in the CS. Typiquement, these
include nouns designating indefinite proportions and quantities.

Of the available items denoting partitivity, some are used as identificatory con-
structions, of which are generally definite, some are used as classificatory constructions,
which tend to generally be indefinite, and some that may be used as either.

(42)

À. Partitive; identificatory; generally definite

J’ai.
Ii.
iii.

ifrejmost
iGrem “repos, remainder
fal “entier”

b. Partitive; classificatory; generally indefinite

J’ai. g¯an “quelques, À”
Ii. w¯anua few, a little

c. Partitive; identificatory or classificatory; definite or indefinite

Sema pair
J’ai.
Ii. b¯ala single
iii. k¯eja part
iv. m¯el “tous”

(43)

À. Partitive; identificatory; generally definite

J’ai.

Ii.

i-Telm
gen.tree

ifrej’n
most.constr
‘most of the trees’
iGrem’n i-Deleinywwe
rest.constr gen.inhabitants

‘rest of the inhabitants (of the city)'

iii. fel’n i-veil

whole.constr gen.day

‘whole day’

b. Partitive; classificatory; generally indefinite

J’ai. g¯en’n

some.constr

i-vevreD
gen.songs

Ii. w¯enyn i-veil

few.constr gen.day

‘some songs’

‘a few days’

21

c. Partitive; identificatory or classificatory; definite or indefinite

J’ai.

Sem’n i-membel
pair.constr gen.shoe

iii. k¯ej’n i-eDyxen

part.constr gen.box

‘a pair of shoes’

Ii. b¯el’n i-veh. le

single.constr gen.flower

‘part of the box’

iv. m¯el’n i-jens

all.constr gen.river

‘a single flower’

‘all the rivers’

4.4 Numerical

Both cardinal and ordinal numbers may be built in the CS, and each may be either
definite or indefinite. As in the adjective CS, the cardinal and ordinal may be explicitly
marked for definiteness.

(44) Numerical CS (identificatory; no periphrasis)

À. With indefinite first element:

J’ai. myn’n

i-Deln
gen.mountain

one.constr
‘one mountain’
leyr’n
seven.constr
‘seven fish’

i-neyn
gen.fish

b. With definite first element:

Ii.

J’ai.

Ii.

iii. mynyn

i-veh. est
gen.dog

first.constr
‘first dog’
leyryn
seventh.constr
‘seventh gift’

i-ineT
gen.gift

iv.

iii.

iv.

le-myn’n
the.one.constr
‘the one mountain’
le-leyr’n
the.seven.constr
‘the seven fish’

i-¯en
gen.mountain

i-neyn
gen.fish

i-veh. est
gen.dog

le-mynyn
the.first.constr
‘the first dog’
le-leyryn
the.seventh.constr
‘the seventh gift’

i-ineT
gen.gift

4.5 Prepositional Phrase

In the framework treated by Svenonius (2007, 2010), Asbury (2008), and Saeed
(2014), it is argued that phrases and clauses, en particulier, Adpositional Phrases, avoir
complex rich internal structures. Such structures can be magnified, or broken down,
into several functional elements. The prepostions that occur in the CS in Eh. eiθymme
are specific internal constituents of the PP. Because leading elements are derived from
nominals, a specific subset of spatial prepositions are put into the construct state.

22

(45) Prepositional Phrase CS (identificatory or classificatory)

weDin
À. est
middle.constr
dans
‘in the middle of house’

i-m¯e
gen.house

Sen’n
b. Al
bottom.constr
à
‘at bottom of well’

i-DeD¯ım
gen.well

c. niw

desin
sous
outside.constr
‘under outside of arch’

i-xeis
gen.arch

d. bat

weDin
milieu

i-Delein
gen.city

à travers
‘through middle of city’

e. deis’n

i-Deleinywwe
gen.inhabitants

among.constr
‘among the people of the city’

If the object of the preposition is not phonologically explicit, then the preposition
which is not in the construct state may be inflected for person, apropriately agreeing
with that object of the preposition. This phenomena of inflection of prepositions is a
hallmark of the Celtic languages; see Hickey (1985), Roberts (2005), Borsley, Tallerman,
and Willis (2007), and Brennan (2008, 2009)

(46) Prepositional Phrase CS with inflected preposition (identificatory or classifica-

tory)
weDin
À. n-¯a
in.3
middle.constr
‘in the middle of it’

Sen’n
bottom.constr

b. al-¯a
at.3
‘at bottom of it’
desin
outside.constr

c. niw-¯e

d. bat-¯a

under.3
‘under outside of it’
weDin
milieu

through.3
‘through middle of it’
im-¯e
deis’n
among.constr
of.3
‘among it’

e.

(46e) shows the insertion of the dummy preposition im-, which is needed when
there is no overt object of the preposition, and the preposition in the construct state

23

cannot take that agreement marker. This is because agreement of prepositions is an
upward movement of the outermost preposition, to collect person information; le
inflection of the construct state stops that nominal from further movement, so some-
thing must collect personal information of the phonologically unrealized object of the
preposition. This agreement is realized with a dummy insertion of the prepostion im-.
A more detailed presentation of this process is forthcoming by the author.

5 Historical Development of the Construct State

A diachronic investigation into the CS in the Semitic languages is not readily
available to the author. Hasselbach (2013) provides an overview of the case system in
the Semitic languages, with particular attention to the historicity and derivation; elle
ne fait pas, cependant, venture into the historical derivation of the CS itself.

The derivation presented here for the Eh. eiθymme CS follows a direct lineage from
the proto-language through a single branch of the language family to the language as is
presented in this paper. Both the formulation of the CS and the derivation of the words
represented themselves therein will be given. This derivation tracks one thousand
eight-hundred years of developmental history.

The reconstructed proto-language did not have explicit marking for (dans)definiteness,

and nor did it have an explicit case marking for adnominal relationships. The marking
for the genitive, either definite or indefinite, was a later development born of concate-
nating preposition-like particles and deictic markers. The proto-language marked a
reletionship between nouns via an adnominal particle, EN, which linked two nominal
elements, and formed general relationships of possession, qualité, ou une description. C'est
from this particle that the construct state marking is derived. This marking survives
only in the construct state explicitly, and few traces of it remain elsewhere.

(47) KARAN ¯A
rock
’rock of girl’

EN
adnominal

FADAJA
girl

Proto-language

From the proto-language, for any derivation, a vowel loss occurred, dependant
on the presence and kind of affixation. Pour (48), the presence of the nominal suffixes ¯A
and A from (47) caused the loss of the second syllable.

(48) KARN ¯A
rock
’rock of girl’

EN
adnominal

FADJA
girl

Loss of vowel

If the word had a final long vowel, then that vowel length shifted from its initial

place to the preceding vowel sound.

24

(49) K ¯ARNA
rock
’rock of girl’

EN
adnominal

FADJA
girl

Shift of vowel length

The final short vowel A is lost in an open ultimate syllable.

(50) K ¯ARN

EN
adnominal

rock
’rock of girl’

FADJ
girl

Loss of final short vowel A

In all places, if the rhotic followed a long vowel, then that rhotic was deleted.
This loss of rhoticity, both in the diachronic context and in the synchronic context, est
inconsistent across even closely related branches of the language family. Rhoticity,
presence or lack thereof, is a shibboleth.

(51) K ¯AN
rock
’rock of girl’

EN
adnominal

FADJ
girl

Deletion of rhotic following a long vowel

The changing of the semi-vowel into a vowel in word-final contexts is widespread

diachronically and synchronically.

(52) K ¯AN
rock
’rock of girl’

EN
adnominal

FADI
girl

Final semi-vowel goes to vowel

In all phonological contexts, the E (a schwa, for the sake of diachronic derivation;
whether the sound was a schwa at all is unknown) was lost. In languages which kept
this sound, it became any of /E, e, À, A/, always unstressed; in the family branch to
which Eh. eiθymme belongs, the E sound was lost entirely. The adnominal particle then
affixed to the leading element.

(53) K ¯AN’N

rock.constr
’rock of girl’

FADI
girl

Loss of schwa vowel

À ce stade, the productive formation of genitival relationships was accom-
plished via periphrasis. There were two options available to formulate this relationship:
the adnominal particles IM or I, the former of which was bare if its complement was an
explicit nominal or inflected for person if the complement was a pronoun (which then
dropped), the latter of which was used with the distal deictic marker ¯IH. The explicit
use of a particle with the distal deictic marker simplified to I, which grammaticalized
as the (dans)definite genitive. This became highly productive, and began to be used with
all adnominal constructions, such as prepositions, the FG, and the CS.

25

(54) K ¯AN’N

rock.constr
’rock of girl’

je
de

FADI
girl

Introduction of explicit genitive

This new genitive marker, identical for both the definite and indefinite, attached
to the word in the same way that the definite article does, which encodes case informa-
tion.

(55) K ¯AN’N

rock.constr
’rock of girl’

I-FADI
gen.girl

Affixation of genitive

Enfin, vowel harmony and consonant mutations occured, yielding the form of

the CS in Eh. eiθymme.

(56) K ¯EN’N

rock.constr
’rock of girl’

I-FEDI
gen.girl

Appropriate sound change

The optional marking of an explicit genitive is not represented in the historical
derivation; it is an innovation or idiosyncratism of the latest stage in the language’s
evolution.

(57)

À. k¯en’n

i-feDi
gen.girl

rock.constr
‘the/a rock of the/a girl’

b. k¯en’n

le-feDi
the.girl

rock.constr
‘the rock of the girl’

26

Les références

Abu Shaqra, Faruq. 2007. Arabic : an essential grammar. Londres : Routledge. ISBN:

9780415415729.

Almansour, Abdulrahman H. 2012. “A phase-based approach to the construct
state.” Journal of King Saud UniversityLanguages and Translation 24 (1):
23–34. ISSN: 2210-8319. https : / / doi . org / https : / / doi . org / 10 . 1016 / j .
jksult.2010.10.001. http://www.sciencedirect.com/science/article/pii/
S2210831911000385.

AlQurashi, Issa S. 2015. “An Analysis of Simple and Construct-State Noun
Phrases in Modern Standard Arabic,” edited by Stefan Muller, 6–26. CSLI
Publications.

Asbury, Anne. 2008. “The Morphosyntax of Case and Adpositions.” PhD diss.,

Universiteit Utrecht.

Benmamoun, Elabbas. 2000. The feature structure of functional categories: A compar-

ative study of Arabic dialects. Presse universitaire d'Oxford.

Borer, Hagit. 1988. “On the morphological parallelism between compounds and
constructs.” In Yearbook of Morphology, edited by Geert Booij and Jaap van
Marle, 45–65. Dordrecht: Foris Publications.

. 1996. “The Construct in Review.” In Studies in Afroasiatic Grammar,
edited by Jacqueline Lecarme, Jean Lowenstamm, and Ur Shlonsky, 30–61.
La Haye: Hollands Academic Graphics. ISBN: 978-90-272-3709-5.

. 1999. “Deconstructing the Construct.” In Beyond Principles and Parameters:
Essays in Memory of Osvaldo Jaeggli, edited by Ian Johnson Kyle and Roberts,
43–89. Dordrecht: Springer Netherlands. ISBN: 978-94-011-4822-1. https :
//doi.org/10.1007/978-94-011-4822-1_3. https://doi.org/10.1007/978-
94-011-4822-1_3.

Borsley, Robert D., Maggie Tallerman, and David Willis. 2007. The Syntax of
Welsh. Cambridge Syntax Guides. la presse de l'Universite de Cambridge. https://doi.
org/10.1017/CBO9780511486227.

Brennan, Jonathan. 2008. “Irish Prepositions: Agreement and Impoverishment.”
In Proceedings of the 26th West Coast Conference on Formal Linguistics, edited by
Charles B. Chang and Hannah J. Haynie, 105–113. Cascadilla Proceedings
Projet. ISBN: 9781574734232.

27

Brennan, Jonathan. 2009. “Pronouns , Inflection , and Irish Prepositions.” In
Papers in Syntax, edited by Patricia Irwin and Violeta Vázquez Rojas Mal-
donado, vol. 2. Penn Linguistics Club.

Carnie, André. 2012. Syntaxe: A Generative Introduction. Wiley Blackwell. ISBN:

9780470655313.

Cowell, Mark W. 1964. “Annexion and Prepositions.” In A Reference Grammar of
Syrian Arabic, edited by Richard S. Harrell, 455–491. Georgetown University
Presse. ISBN: 9781626163652.

Creissels, Denis. 2009. “Construct Forms of Nouns in African Languages,” edited
by Peter K. Austin, Oliver Bond, Monik Charette, Nathan David, and Peter
Sells, 73–82. SOAS University of London.

El Hankari, Abdelhak. 2014. “The Construct State in Tarifit Berber.” Lingua
148:28–52. ISSN: 0024-3841. https://doi.org/https://doi.org/10.1016/j.
lingua.2014.05.008. http://www.sciencedirect.com/science/article/pii/
S0024384114001107.

Hagstrom, Paul. 2001. Lecture notes in Syntax: DP and NumP.

Hasselbach, Rebecca. 2013. Case in Semitic: Roles, Relations, and Reconstruction.

Presse universitaire d'Oxford. ISBN: 9780199671809.

Hickey, Raymonde. 1985. “Salient features of Irish syntax.” Lingua Posnaniensis

26:15–26.

Holes, Clive. 2004. Modern Arabic: Ouvrages, Functions, and Varieties. Georgetown

University Press. ISBN: 9781589010222.

Johannessen, Janne Bondi. 1998. “The Conjunction Phrase.” In Coordination,
108–174. Oxford Studies in Comparative Syntax. Presse universitaire d'Oxford.
ISBN: 9780198237723.

Parsafar, Parviz. 2010. “Syntax, Morphology, and Semantics of Ezafe.” Iranian

Studies 43 (5): 637–666. ISSN: 00210862.

Ritter, Elizabeth. 1991. “Two functional categories in noun phrases: Evidence
from Modern Hebrew.” In Syntax and semantics vol. 25: Perspectives on phrase
structure: Heads and Licensing, edited by S.D. Rothstein, 37–62. San Diego:
Academic Press. ISBN: 9780126061062.

Roberts, Ian G. 2005. Principles and parameters in a VSO language: A case study
in Welsh. Oxford Studies in Comparative Syntax. Presse universitaire d'Oxford.
ISBN: 9780195168211.

28

Ryding, Karin. 2005. A Reference Grammar of Modern Standard Arabic. ISBN:

9780521771511.

Saeed, Sameerah. 2014. “The Syntax and Semantics of Arabic PS.” Newcastle and

Northumbria Working Papers in Linguistics 20:44–66.

Svenonius, Pierre. 2007. “Projections of P.” CASTL, University of Tromso.

. 2010. “Spatial P in English.” Edited by Cinque and Rizzi, 127–160.

Thackston, Wheeler. 1994. An Introduction to Koranic and Classical Arabic: An
Elementary Grammar of the Language. IBEX Publishers. ISBN: 9780936347400.

Versteegh, Kees. 2014. Arabic Language. Presse universitaire d'Édimbourg. ISBN: 9780748645299.

Watson, Janet C.E. 1993. “Annexion and Prepositions.” In A syntax of San’ani

Arabic, 173–204. Wiesbaden : Harrassowitz. ISBN: 9783447033923.

29Free Genitive and Construct State in Eḥeiθymme image

Télécharger le PDF

(visité 1 fois, 1 visites aujourd'hui)