Content uploaded by Hanadi Azhari
Author content
All content in this area was uploaded by Hanadi Azhari on Oct 07, 2024
Content may be subject to copyright.
Actes du congrès annuel de l’Association canadienne de linguistique 2023.
Proceedings of the 2023 annual conference of the Canadian Linguistic Association.
© 2023 Hanadi Azhari
THE GRAMMATICATIZATION OF ALIENABLE POSSESSION IN
MAKKAN ARABIC
Hanadi Azhari
Umm Al-Qura University
1. Introduction
Many of the world’s languages grammaticalize a distinction between inalienable and
alienable possession. Inalienable possession is defined as an inherited, inseparable,
permanent semantic relationship between two nominals, the possessee and the possessor,
while alienable possession is a an acquired, separable, changing relationship between the
possessee and the possessor (see Nichols 1988, 1992; Heine 1997; Alexiadou 2003). Such
a dichotomy does not exist in Standard Arabic (henceforth SA) in which possession is
expressed by simple juxtaposition of the possessee and the possessor nominals, a structure
known as the Semitic construct state (see Benmamoun and Choueiri 2013). However, in
Makkan Arabic (henceforth MA), the Urban Hijazi Arabic spoken in the city of Makkah,
Saudi Arabia, an alternative analytic possessive construction has emerged with the
functional morpheme ħaɡ as a genitive exponent that occurs between the possessee and
the possessor nominals. This periphrastic possessive construction is specific to alienable
possession and is assumed to signify the emergence of the alienability split in MA. In this
paper, I provide an account for such a change within the framework of grammaticalization
and discuss the principles by which it has been motivated. Besides, I trace the pathway of
the change and demonstrate how it adheres to principles of grammaticalization, and thus
is a legitimate case of grammaticalization.
This paper proceeds as follows: §2 discusses the synthetic possessive construction
in Arabic and the analytic construction that has emerged in MA. The source of the
possessive morpheme ħaɡ and its grammaticalization as a genitive exponent is discussed
in §3. The motivating factors underlying the evolvement of the periphrastic possessive
construction and its limitation to alienable position are discussed in §4. The conclusion in
§5 summarizes the main sections of the paper and discusses the development from
synthetic to analytic constructions as a pervasive phenomenon in modern Arabic dialects.
2. Possession in Arabic
This section discusses possession in Standard and Makkan Arabic. The synthetic
possessive construction common in SA, the construct state, is referred to as the original
construction whereas the periphrastic, genitive exponent, possessive construction that has
evolved in MA is referred to as the emergent construction.
2
2.1 Possession in SA
In SA, the oldest well-documented stage of Arabic, possession is expressed by simple
juxtaposition of the possessee and the possessor nominals (Possessee + Possessor), as in
(1), a structure known as the Semitic construct state (Benmamoun and Choueiri 2013). In
such structures the possessee nominals are incompatible with the definite article al-, as
shown in (1) and (2).
(1) (*al)-kita:b Sami1
DEF-book Sami
‘Sami’s book’
The possessor may also be a pronoun. If so, it has to be an enclitic pronoun attached to
the possessee NP. As a morphologically complex language with a very rich templatic
morphology (see Altantawy et al. 2011, Bouamor et al. 2014), Arabic employs a system
of enclitic pronouns, classified by person (1st, 2nd, and 3rd ), number (singular, dual, and
plural), and gender (masculine vs. feminine), as markers of possession (representing the
possessor). Thus, the possessive structure in (1) could be expressed using an enclitic
pronoun (encliticization), as in (2) below:2
(2) (*al)-kita:b-u-h
DEF-book-NOM-3SG.M
‘his book’
2.2 Possession in MA
In MA, besides the construct state of SA, an alternative possessive construction has
emerged in which possession is expressed using the functional morpheme ħaɡ between
the possessee and the possessor.3 This functional morpheme might be followed by a
possessor NP, as in (3), or take an enclitic pronoun, as in (4).
(3) (al)-kita:b ħaɡ Sami
DEF-book POSS Sami
‘Sami’s book’
1 The following symbols and abbreviations are used in my data: ħ = voiceless pharyngeal fricative, ʕ =
voiced pharyngeal fricative, ɡ = voiced velar stop, ʒ = voiced palato-alveolar sibilant, ṣ = voiceless alveolar
fricative, DEF = definite article, NOM = nominative case marker, 1 = first person, 3 = third person, SG =
singular, PL = plural, M = masculine, F = feminine, POSS = possessive exponent.
2 Both structures, juxtaposition and encliticization, will be generally referred to as juxtaposition and/or
construct state constructions since the main point of the discussion is to distinguish them from the emergent
analytic construction.
3 In the present paper, the possessive marker ħaɡ is considered a grammaticalized possessive exponent
rather than a preposition meaning ‘of’ (cf. Bardeas 2009, Alrasheedi 2019).
3
(4) (al)-kita:b ħaɡɡ-i
DEF-book POSS-1SG
‘my book’
When the morpheme ħaɡ is inserted between the possessee and the possessor, the
resulting construction is no longer a construct state; hence, the possessee could be definite
or indefinite as shown in (3) and (4) above.
This periphrastic construction is not always used alternatively with the construct
state in MA since its usage is restricted to particular possessive relations, namely alienable
possession, while it is incompatible with inalienably possessed nouns. Inalienable
possession is defined as an unchanging, inherent, and/or permanent semantic relationship
between two nominals, the possessee and the possessor, while alienable possession is a
changing, acquired, context-dependent relationship between the possessee and the
possessor (see Nichols 1988, 1992; Heine 1997; Alexiadou 2003, among others cited in
this paper). The alienability split is grammaticalized in many of the world’s languages,
like for example, Tauya (MacDonald 1990), Tigrinya (Gebregziabher 2012), Blackfoot
(Ritter and Rosen 2014), and Mi'gmaq (McClay 2012), among many others (see Chapell
and McGregor 1989).
While no evidence of inalienable/alienable distinction appears in Standard Arabic,
the periphrastic possessive construction that has developed in MA clearly encodes the
alienability split.4 Cross-linguistically, kinship and body parts terms are prototypical
members of the class of inalienably possessed nouns in languages that display distinct
grammatical markings for inalienable vs. alienable possession (see Nichols 1992, Heine
1997, Alexiadou 2003, among many others). These two classes of nouns appear in
construct state structures but are incompatible with the periphrastic possessive
construction in MA, thus (5b) and (6b) are ungrammatical.
(5) a. ʕamm-i
uncle-1SG
‘my uncle’
b. *ʕam ħaɡɡ-i
uncle POSS-1SG
(int.) ‘my uncle’
(6) a. riʒl-i
foot-1SG
‘my foot’
b. *riʒil ħaɡɡ-at-i
foot POSS-F-1SG
(int.) ‘my foot’
4 Alshehri (2023) argues that the inalienable/alienable distinction is syntactic in SA.
4
Inalienability covers several semantic fields, which vary from one language to another.
That is, members of the inalienable category cannot be predicted but are specified based
on language-specific cultural conventions (Heine 1997, Chapell and McGregor 1989). A
considerable body of research has suggested that whenever the alienability split is
encoded in a language, inalienably possessed terms form a closed class with limited
membership while the alienable category is always an open class (Nichols 1988, Heine
1997). The noun ṣa:ħib ‘friend’ seems to pattern with inalienably possessed nouns in MA
since it is incompatible with the genitive exponent possessive construction, as shown in
(7b).
(7) a. ṣa:ħb-i
friend-1SG
‘my friend’
b. *ṣa:ħib ħaɡɡ-i
friend POSS-1SG
(int.) ‘my friend’
The fact that the periphrastic possessive construction is exclusive to alienable possession
is a clear indication for the emergence of the alienability split in MA.
The remainder of the paper focuses on the emergent possessive construction as a
case of grammaticalization in MA. The grammaticalization of the possessive marker ħaɡ
from a noun to a genitive exponent is discussed in the following section.
3. The grammaticalization of the possessive exponent ħaɡ
Grammaticalization has been defined as the process by which a lexical form receives a
grammatical meaning besides its conventional lexical meaning or as the development of
grammatical forms or constructions out of already existing ones (see Hopper 1996, Heine
1997, Heine and Kuteva 2004, Kuteva et al. 2019, among others cited in this paper). In
this section the grammaticalization of the possessive genitive exponent ħaɡ is discussed
with reference to Hopper’s (1991, 1996) principles of grammaticalization, namely (i)
layering, (ii) divergence, (iii) specialization, (iv) persistence, and (v) decategorialization,
along with Heine and Kuteva’s (2004) four main mechanisms of grammaticalization,
namely (i) extension, (ii) desemanticization, (iii) decategorialization, and (iv) erosion (see
also Kuteva et al. 2019).
The possessive genitive exponent ħaɡ, I assume, has been grammaticalized from the
source noun ħaɡ, which means ‘something that a person is legally or morally entitled to
have, get, or do’ (the equivalent to the English noun right).5 Unlike the source noun, the
functional morpheme ħaɡ occurs between two nominals, the possessee and the possessor,
5 Eifan (2017) assumes that the source form of the possessive morpheme ħaɡ is the Egyptian noun ħa:ɡa
‘something or anything’, and that this grammaticalization reflects the influence of Egyptian Arabic on the
Urban Hijazi Arabic, a proposal that the author of the present paper does not adopt.
5
which denotes the extension of its distribution to a new context that is distinct from the
context of its original counterpart, the noun ħaɡ. As mentioned previously, the possessive
exponent accepts encliticization. Moreover, insertion of the functional word ħaɡ between
the possessee and the possessor nominals gives way to the deletion of the possessee if it
could be inferred from the context. Therefore, the forms ħaɡ Sami ‘Sami’s’, and ħaɡɡ-i
‘mine’ are attested forms in MA and are used alternatively with the full form (possessee
ħaɡ possessor) and the construct state structure, which conforms to Hopper's (1991,
1996) principle of layering, the co-occurrence of multiple forms and/or structures in the
lexical, as well as the grammatical domain of a given language. Following are the
constructions that co-occur as attested possessive structures in MA (examples (1), (3),
and (4) above are repeated, respectively, as (8), (10), and (11) below).
(8) kita:b Sami
book Sami
‘Sami’s book’
(9) kita:b-i
book-1SG
‘my book’
(10) (al)-kita:b ħaɡ Sami
DEF-book POSS Sami
‘Sami’s book’
(11) (al)-kita:b ħaɡɡ-i
DEF-book POSS-1SG
‘my book’
(12) ħaɡ Sami
POSS Sami
‘Sami’s’
(13) ħaɡɡ-i
POSS-1SG
‘mine’
This co-occurrence is in compliance with the premise that language change is a gradual
process that does not take the form of an abrupt shift, but a gradual transition from using
one form to using another emergent form. Throughout such a diachronic transition, areas
appear in which the original form and the emergent one(s) co-exist and are equally
attested by the language speakers (Hopper 1991, 1996; Hopper and Traugott 2003; Heine
and Kuteva 2004; Kuteva et al. 2019).
As a grammaticalized functional morpheme, ħaɡ has undergone semantic bleaching
(desemanticization). That is, the sense of ‘something that a person is legally or morally
6
entitled to have, get, or do’ has been bleached and left behind its implied sense of
possession/property. As a result, the functional morpheme ħaɡ has acquired an abstract
grammatical meaning as a mere possessive marker, and thus become divergent from the
source form, the noun ħaɡ. Nonetheless, it is evident that the morpheme ħaɡ is not an
empty syntactic element; rather, it still has remnants of its original meaning in the sense
of ‘possession/property’. Consequently, the source noun ħaɡ and the possessive genitive
exponent ħaɡ have become polysemous in MA, so ħaɡɡ-i is ambiguous between ‘my
right’ (14a) and ‘mine’ (see (13) above, repeated as (14b) below).
(14) a. ħaɡɡ-i
right-1SG
‘my right’
b. ħaɡɡ-i
POSS-1SG
‘mine’
When contextualized, the two forms in (14) can be distinguished, and the intended
meaning can easily be discerned since the two forms have different distributions.
According to Hopper (1991), it can be anticipated that a form will be polysemous during
intermediate stages of grammaticalization, and that at least one of its meanings will
reflect previous dominant meaning. The resistance of ħaɡ to undergo complete semantic
bleaching or sudden emptying of meaning is in line with Hopper’s (1991, 1996) principle
of persistence that relates the meaning of the grammaticalized item to its source, the
lexical item out of which it has evolved.
The morpheme ħaɡ has become grammaticalized in a particular context, the analytic
possessive construction, but it retains its lexical status as a noun in other contexts. This is
consistent with the mechanism of extension, use in a new context (Heine and Kuteva
2004, Kuteva et al. 2019). It is worth noting here that while the possessive marker ħaɡ is
confined to possessive constructions, the noun ħaɡ has a wider distribution. However, the
noun ħaɡ seems to appear less frequently, and this observation conforms to Segalowitz
and Lane’s (2000) account that access to function words is faster than access to content
words, and thus function words tend to appear more frequently.
The choice of the noun ħaɡ to be grammaticalized as a possessive marker is not an
arbitrary one, but it has to do with the principle of specialization on the basis of textual
frequency and semantic generalizations (see Hopper 1991, Hopper and Traugott 2003).
As a lexical noun, ħaɡ appears frequently in expressions of possession and property. That
is, it is combined in discourse with property, ownership, and possession, as shown in the
following examples:
(15) al-kita:b ħaɡ li-Sami
DEF-book right for-Sami
‘it is Sami’s right to have, get, or own the book’
7
(16) al-kita:b min ħaɡ Sami
DEF-book of right Sami
‘it is Sami’s right to have, get, or own the book’
The frequent occurrence of the lexical morpheme ħaɡ ‘right’ in expressions of
possession, I assume, has gradually led to the generalization of its meaning to denote
‘property’, which is directly related to the concept of possession (see Heine 1997, Heine
and Kuteva 2004, Rubin 2004, Eifan 2017). Such generalization has set the noun ħaɡ
apart as a plausible candidate to undergo the process of grammaticalization and become a
functional possessive morpheme. This premise is consistent with the conception of
grammaticalization as a product of context-induced reinterpretation (see Heine 1997,
Heine and Kuteva 2004, Kuteva et al. 2019).
The possessive marker ħaɡ has lost its lexical category as a noun and become a
functional morpheme. That is, it has been decategorialized. Decategorialization refers to
the loss of morphological markers and syntactic properties of the lexical/or the source
category (see Heine 1997, Heine and Kuteva 2004, Kuteva et al. 2019). As a noun, ħaɡ
can (i) be pluralized (inflected for number), as in (17), and (ii) take the bound definite
article al-, as in (18). A combination of the two properties is exemplified in (19).
(17) ħuɡu:ɡ al-ʔinsa:n
rights DEF-human
‘the human rights’
(18) al-ħaɡ al-ʕa:m
DEF-right DEF-public
‘the public right’
(19) al-ħuɡu:ɡ al-ʕa:mma
DEF-rights DEF-public
‘the public rights’
However, when grammaticalized as a possessive morpheme, ħaɡ has lost these two
properties. Consequently, the possessive marker ħaɡ does not inflect for number and is
incompatible with the definite article al-, so (20) and (21) are ungrammatical.
(20)*(al)-kita:b al-ħaɡ Sami
DEF-book DEF-POSS Sami
(int.) ‘Sami’s book’
(21) al-kutub ħuɡu:ɡ Sami
DEF-books POSS.PL. Sami
(int.) ‘Sami’s books’
8
Nonetheless, grammaticalization does involve gains (Hopper and Traugott 2003, Kuteva
et al. 2019), and in the present case ħaɡ as a possessive morpheme is inflected for gender
and, thus, agrees with the possessee for gender, a property which does not exist for the
lexical source, the noun ħaɡ.
(22) al-kita:b ħaɡɡ-u
DEF-book POSS-3SG.M
‘his book’
(23) al-waraɡa ħaɡɡ-at-u
DEF-paper POSS-F-3SG.M
‘his paper’
In Arabic, all nouns carry grammatical gender as either masculine or feminine. Nouns
that refer to professions, like teacher, doctor, and engineer, have two distinct forms, one
masculine and one feminine. Other nouns are either masculine or feminine and the gender
category into which a noun falls is semantically arbitrary and is assigned by convention
(see Holes 2004, Ryding 2005). The abstract noun ħaɡ ‘right’ is masculine in Arabic, but
the possessive marker ħaɡ does not carry grammatical gender, and it agrees with the
possessee in this respect. Since the noun kita:b ‘book’ is masculine in Arabic, the form
ħaɡ appears with it in (22). However, waraɡa ‘paper’ is feminine in Arabic, so the
possessive morpheme ħaɡ agrees with it by taking the feminine suffix -at in (23).
To sum up, this section sheds light on the compliance of the target possessive
morpheme, ħaɡ, with the principles and mechanisms of grammaticalization as identified
by Hopper (1991, 1996), Heine (1997), Heine and Kuteva (2004), and Kuteva et al.
(2019). Dissimilar to its source counterpart, the genitive exponent ħaɡ occurs between
two nominals (extension), and in such a context it is no longer a contentive morpheme
but a functor (desemanticization). Forms such as ħaɡɡ-i ‘my right’ and al-kita:b ħaɡɡ-i
‘my book’ co-exist in MA (layering). And even after ħaɡ has come to be used as a
function word indicating possession, the noun ħaɡ continues to be used and each form
has its distinct distribution (divergence). From among a plethora of nouns denoting
property in Arabic, only ħaɡ has turned into a possessive morpheme (specialization).
Moreover, as a possessive morpheme ħaɡ still retains some vestiges of the original
meaning of the source noun in the sense of ‘property’ (persistence), yet it lost its syntactic
and morphological nominal characteristics (decategorialization). It is worth noting here
that no evidence of erosion has been observed in the grammaticalization of ħaɡ, and this
is justifiable given that while some processes of grammaticalization can be universally
observed, others, among which is erosion, could but need not occur (Heine and Reh
1984).
4. Alienability split in MA
This section provides an account for the inalienable/alienable distinction in MA within
the framework of grammaticalization. In §4.1, the grammaticalization of the periphrastic
9
possessive construction and its motivating factors are discussed. In §4.2, the
exclusiveness of the emergent possessive construction to alienable possession and its
incompatibility with alienable category are discussed with reference to the principles of
iconicity and economy.
4.1 The grammaticalization of alienability split in MA
This section discusses the emergence of the periphrastic alienable possessive construction
(possessee ħaɡ possessor) in MA in relation to the grammaticalization concepts of
expressivity, routinization, renewal, economy, and pragmatic inferencing (see Heine
1997, Hopper and Traugott 2003). Expressivity plays a central role in the account for the
alienability split in MA. The zero-marked possessive construction of mere juxtaposition
of the possessee and the possessor has reached a very advanced stage in its
grammaticalization to the extent that it has become routinized and worn out, and
consequently, lost its expressivity. Loss of expressivity could also be the result of several
diachronic forces that have triggered semantic bleaching effect (see Heine 1997).
Therefore, a need has arisen to specify and emphasize the possessive relation between the
two juxtaposed nominals, which induced renewal and led to the introduction of a more
explicit periphrastic construction with an explicit possessive morpheme between the
possessee and the possessor. According to Heine (1997), possessive markers are always
chosen from among already existing words in the language. Moreover, they are drawn
cross-linguistically from one of eight main sources, namely action, location, companion,
genitive, goal, source, topic, and equation. In the case of MA, I assume that the
possessive marker ħaɡ follows the ‘equation pattern’ as possession seems to be equated
with ‘right’ or ‘property’ in the sense “Y is X’s (property)” where X is the possessor and
Y is the possessed noun (Heine 1997: 47).
In accordance with the principle of economy, using already existing forms for new
purposes (Heine 1997), the possessive morpheme is selected from among already
existing words, as mentioned previously, yet this selection is not arbitrary but is based on
and influenced by pragmatic inferencing. Pragmatic inferences play a significant role in
semantic change in general, and in grammaticalization in particular (see Heine 1997,
Kuteva et al. 2019); however, for them to be effective, they must occur frequently
(Hopper and Traugott 2003). As mentioned in §3 and exemplified in (15) and (16) above,
the noun ħaɡ ‘right’ occurs very frequently in expressions of possession in Arabic which
resulted in a strong association between ħaɡ ‘right’ and ‘possession’. Gradually, ħaɡ has
become an integral part of the possessive construction as it has acquired possessive
denotations through the conventionalization of possessive implicatures. This periphrastic
construction is used for alienable possession in MA along with, yet more frequently than,
the construct state.
4.2 Why is the emergent construction incompatible with inalienable possession?
According to the principle of iconicity, syntactic structures reflect conceptual perception
(Haiman 2008). Thus, while it is normal for alienably possessed nouns to be separated
10
from the possessor by a functional morpheme, this is not the case with inalienably
possessed nouns that are supposed to be closer to the possessor as they express relational
or part-to-whole relation. According to Haiman (2008) and Haspelmath (2008a), the
inalienable/alienable dichotomy is a clear manifestation of diagrammatic iconicity. That
is, correspondence always appears between conceptual distance and syntactic distance.
Therefore, cross-linguistically the expression of alienable possession is more complex
and requires usage of extra morphemes between the possessee and the possessor, which
results in a greater linguistic distance. In inalienable possession, on the other hand, the
distance between the possessee and the possessor tends to be kept to the minimum
distance possible as a reflection of the speaker’s conceptualization of the close relation
(relational with kin terms and part-to-whole with body parts) between the two nominals.
As a result, inalienable possession usually displays a closer relationship between the
possessee and the possessor (McClay 2012). In MA, closeness of the possessee and the
possessor is indicated through the restriction on intervening morphemes between the two
nominals. Thus, the ungrammaticality of the periphrastic possessive constriction with
inalienably possessed nouns is a rational consequence of the effect of iconicity.
Another explanation for the incompatibility of the periphrastic possessive
construction with inalienable possession has to do with the principles of frequency and
economy. According to Nichols (1988), inalienably possessed nouns occur more
frequently in possessive constructions than do alienably possessed nouns. That is, it is
very common for kinship and body part terms to appear as possessed nouns. Thus,
inalienably possessed nouns are perceived of as being always possessed, but this is not the
case with alienably possessed nouns, which are perceived of as unpossessed nouns.
Consequently, the predictability of nominals of the inalienable category as possessee is
higher than that of nominals of the alienable category and upon appearing in a routinized
juxtaposed construction, alienably possessed nouns require an overt marker to specify
their possessive relation to the other nominal, the possessor. However, upon hearing a
typical inalienably possessed noun, speakers can predict that it will occur as a possessee
in a possessive construction which renders the overt marking relatively redundant (see
Haspelmath 2008b). As languages are efficient systems that activate the principle of
economy and place restrictions on redundancy, the ban on the overt possessive marker
ħaɡ in inalienable possessive constructions in MA is evidently justified.
5. Conclusion
This paper discusses the emergence of the alienability distinction in MA through the
grammaticalization of the possessive exponent ħaɡ out of the lexical word meaning
‘right’. The grammaticalization of ħaɡ is considered with reference to the principles and
mechanisms of grammaticalization, namely layering, divergence, specialization,
persistence, decategorialization, extension, and desemanticization, as specified by Hopper
(1991, 1996) and Kuteva et al. (2019), respectively. Moreover, the alienability split in MA
is explained considering the principles of expressivity, routinization, renewal, and
pragmatic inferencing, and the limitation of the analytic possessive construction to
alienable possession is justified based on iconicity and economy.
11
The cline of grammaticalization discussed here, the tendency to develop an analytic
possessive construction, is not specific to MA. It is a pervasive phenomenon in almost all
modern dialects of Arabic since its motivating factor, to specify the possessive relation,
applies equally to all varieties of Arabic (see Dahl and Koptjevskaja-Tamm 1998,
Boumans 2006, Brustad 2000, Alshehri 2023). The major difference between the various
dialects of Arabic lies in the source and nature of the possessive exponents which are
generally derived from words meaning ‘wealth’ or ‘property’ (see Brustad 2000, Rubin
2004, Soltan 2007, Stolz 2011, Al-Salman 2013, among others cited in this paper). It is
worth stating that the possessive exponent ħaɡ is used in Kuwaiti (Brustad 2000), Najdi
(Alrasheedi 2019), Hijazi (Eifan 2017), Emirati, Omani, Sudanese, and Yamani Arabic
(Jarad 2017) with some variation in its use from one dialect to another. Although some
studies have referred to the analytic possessive construction with the functional
morpheme ħaɡ, no study has thoroughly discussed the pathway of the grammaticalization
of this morpheme and the principles that have motivated its emergence as a possessive
genitive exponent indicating the alienability split in Arabic, which is the main
contribution of this paper.
References
Alexiadou, Artemis. 2003. Some notes on the structure of alienable and inalienable possessors. In From Np
to DP: Volume 2: The expression of possession in noun phrases, ed. Martine Coene and Yves
D’hulst, 167–188. Amsterdam/Philadelphia: John Benjamins.
Alrasheedi, Eisa. 2019. Possessive constructions in Najdi Arabic. Doctoral dissertation, Newcastle
University.
Al-Salman, Ibrahim. 2013. Jordanian pidgin Arabic. Master’s thesis, Yarmouk University.
Alshehri, Othman. 2023. The syntax and semantics of Arabic possession. Doctoral dissertation, The
University of Arizona.
Altantawy, Mohamed, Nizar Habash, and Owen Rambow. 2011. Fast yet rich morphological analysis.
In Proceedings of the 9th International Workshop on Finite State Methods and Natural Language
Processing, ed. Andreas Maletti and Matthieu Constant, 116–124. Blois: Association of
Computational Linguistics.
Bardeas, Suzanne. 2009. The syntax of the Arabic DP. Doctoral dissertation, University of York.
Benmamoun, Elabbas, and Lina Choueiri. 2013. The syntax of Arabic from a generative perspective. In The
Oxford handbook of Arabic linguistics, ed. Jonathan Owens, 115–164. Oxford: Oxford University
Press.
Bouamor, Houda, Nizar Habash, and Kemal Oflazer. 2014. A multidialectal parallel corpus of Arabic. In
Proceedings of the Ninth International Conference on Language Resources and Evaluation
(LREC'14), ed. Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn Loftsson, Bente
Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk, and Stelios Piperidis, 1240–1245.
Reykjavik: European Language Resources Association (ELRA).
Boumans, Louis. 2006. The attributive possessive in Moroccan Arabic spoken by young bilinguals in the
Netherlands and their peers in Morocco. Bilingualism: Language and Cognition 9: 213–231.
Brustad, Kristen E. 2000. The syntax of spoken Arabic: A comparative study of Moroccan, Egyptian, Syrian
and Kuwaiti dialects. Washington, D.C.: Georgetown University Press.
Chappell, Hilary, and William McGregor. 1989. Alienability, inalienability, and nominal classification. In
Proceedings of the 15th Annual Meeting of the Berkeley Linguistics Society, ed. Mary Niepokuj, 24–
36. California: Berkeley Linguistics Society.
12
Dahl, Östen, and Maria Koptjevskaja-Tamm. 1998. Alienability splits and the grammaticalization of
possessive constructions. In Papers from the 16th Scandinavian Conference of Linguistics, ed. Timo
Haukioja, 38–49. Turku: University of Turku.
Eifan, Emtenan. 2017. Grammaticalization in Urban Hijazi Arabic. Master’s thesis. University of
Manchester.
Gebregziabher, Keffyalew. 2012. The alienable-inalienable asymmetry: Evidence from Tigrinya.
In Selected proceedings of the 42nd Annual Conference on African Linguistics, ed. Michael R.
Marlo, Nikki B. Adams, Christopher R. Graan, Michelle Morrison, and Tristan M. Purvis, 161–182.
Somerville: Cascadilla Proceedings Project.
Haiman, John. 2008. In defence of iconicity. Cognitive Linguistics 19(1): 35–48.
Haspelmath, Martin. 2008a. Frequency vs. iconicity in explaining grammatical asymmetries. Cognitive
Linguistics 19(1): 1–33.
Haspelmath, Martin. 2008b. Syntactic universals and usage frequency: Alienable vs. inalienable possessive
constructions. Ms., Leipzig Spring School on Linguistic Diversity.
Heine, Bernd, and Mechthild Reh. 1984. Grammaticalization and Reanalysis in African Languages.
Hamburg: Helmut Buske.
Heine, Bernd. 1997. Possession: Cognitive sources, forces, and grammaticalization. Cambridge:
Cambridge University Press.
Heine, Bernd, and Tania Kuteva. 2004. World lexicon of grammaticalization. Cambridge: Cambridge
University Press.
Holes, Clive. 2004. Modern Arabic: Structures, functions, and varieties. Washington, D.C.: Georgetown
University Press.
Hopper, Paul, and Elizabeth Traugott. 2003. Grammaticalization. Cambridge: Cambridge University Press.
Hopper, Paul. 1996. Some recent trends in grammaticalization. Annual Review of Anthropology 25: 217–236.
Hopper, Paul. 1991. On some principles of grammaticalization. In Approaches to Grammaticalization:
Volume 1, ed. Elizabeth Traugott and Bernd Heine, 17–36. Amsterdam/Philadelphia: John
Benjamins.
Jarad, Najib. 2017. Grammaticalization in Emirati Arabic. Arabica 64: 742–760.
Kuteva, Tania, Bernd Heine, Bo Hong, Haiping Long, Heiko Narrog, and Seongha Rhee. 2019. World
lexicon of grammaticalization. Cambridge: Cambridge University Press.
MacDonald, Lorna. 1990. A Grammar of Tauya. Berlin/New York: Mouton de Gruyter.
McClay, Elise. 2012. Possession in Mi'gmaq. Master’s thesis, McGill University.
Nichols, Johanna. 1988. On alienable and inalienable possession. In In Honor of Mary Haas: From the
Haas Festival Conference on Native American Linguistics, ed. William Shipley, 557–610.
Berlin/New York: De Gruyter Mouton.
Nichols, Johanna. 1992. Linguistic diversity in space and time. Chicago: University of Chicago Press.
Ritter, Elizabeth, and Sara Rosen. 2014. Possessors as external arguments: Evidence from Blackfoot. In
Papers of the 42nd Algonquian Conference, ed. J. Randolph Valentine and Monica Macaulay.
Albany: SUNY Press.
Rubin, Aaron. 2004. Studies in Semitic grammaticalization. Doctoral dissertation, Harvard University.
Ryding, Karin C. 2005. A reference grammar of Modern Standard Arabic. Cambridge: Cambridge
University Press.
Segalowitz, Sidney J. and Korri C. Lane. 2000. Lexical access of function versus content words. Brain and
Language 75(3): 376–389.
Soltan, Usama. 2007. On the individual-property contrast in free state possessive nominals in Egyptian
Arabic. In Perspectives on Arabic linguistics XX, ed. Mustafa Mughazy, 71–86.
Amsterdam/Philadelphia: John Benjamins.
Stolz, Thomas. 2011. The possessive relative clause in Maltese. In Variation and change: The dynamics of
Maltese in space, time and society, ed. Sandro Caruana, Ray Fabri, and Thomas Stolz, 183–232.
Berlin: Akademie Verlag.
